Split (1)

train · 1.47k rows

filename stringlengths 13 34	markdown stringlengths 0 359k
10.3390_biom11010118	Article Ginsenoside Rg3 Prevents Oncogenic Long Noncoding RNA ATXN8OS from Inhibiting Tumor-Suppressive microRNA-424-5p in Breast Cancer Cells Heejoo Kim †, Hwee Won Ji †, Hyeon Woo Kim , Sung Hwan Yun, Jae Eun Park and Sun Jung Kim * Department of Life Science, Dongguk University-Seoul, Goyang 10326, Korea; [email protected] (H.K.); [email protected] (H.W.J.); [email protected] (H.W.K.); [email protected] (S.H.Y.); [email protected] (J.E.P.) * Correspondence: [email protected]; Tel.: +82-31-961-5129 † These authors contributed equally to this work. Abstract: Ginsenoside Rg3 exerts antiproliferation activity on cancer cells by regulating diverse noncoding RNAs. However, little is known about the role of long noncoding RNAs (lncRNAs) or their relationship with competitive endogenous RNA (ceRNA) in Rg3-treated cancer cells. Here, a lncRNA (ATXN8OS) was found to be downregulated via Rg3-mediated promoter hypermethylation in MCF-7 breast cancer cells. SiRNA-induced downregulation of ATXN8OS decreased cell proliferation but increased apoptosis, suggesting that the noncoding RNA possessed proproliferation activity. An in silico search for potential ATXN8OS-targeting microRNAs (miRs) identiﬁed a promising candidate (miR-424-5p) based on its high binding score. As expected, miR-424-5p suppressed proliferation and stimulated apoptosis of the MCF-7 cells. The in silico miR-target-gene prediction identiﬁed 200 potential target genes of miR-424-5p, which were subsequently narrowed down to seven that underwent hypermethylation at their promoter by Rg3. Among them, three genes (EYA1, DACH1, and CHRM3) were previously known oncogenes and were proven to be oppositely regulated by ATXN8OS and miR-424-5p. When taken together, Rg3 downregulated ATXN8OS that inhibited the tumor-suppressive miR-424-5p, leading to the downregulation of the oncogenic target genes. Keywords: ceRNA; CpG methylation; ginsenoside Rg3; long noncoding RNA; microRNA 1. Introduction Ginsenoside Rg3 is a steroidal saponin derivative that is abundant in heat-processed ginseng extract [1]. Rg3 possesses potent anticancer properties and is known to modulate diverse cellular events such as cell proliferation, immune response, autophagy, metastasis, and angiogenesis [2]. Rg3 activates proapoptotic proteins such as caspase-3 and Bax but suppresses antiapoptotic protein Bcl-2 [3]. In the process, NF-κB, which drives cell-cycle progression, is inhibited by blocking the phosphorylation of Akt and ERK kinases [4]. In MDA-MB-231 breast-cancer cells, Bcl-2 can be suppressed by destabilizing a mutant P53 with Rg3 [4]. In osteosarcoma cell lines, Rg3 inhibits migration and invasion by suppressing MMPs and the Wnt/β-catenin pathway, which are related to epithelial-mesenchymal transition (EMT) and angiogenesis [5]. Rg3-treated gastric cancer cells show a remarkably lower expression of HIF-1α and VEGF under hypoxia [6]. The SNAIL signaling axis is another key pathway regulated by Rg3 during metastasis, which regulates EGFR and ﬁbronectin in cancer stem cells [7]. Rg3 can inhibit cancer-cell growth by modulating epigenetic factors of oncogenes or tumor suppressors. A genome-wide methylation analysis identiﬁed over 250 genes with signiﬁcant changes in methylation level at speciﬁc CpG sites in Rg3-treated MCF-7 breast-cancer cells [8]. These genes were largely associated with cell-morphology-related pathways. Notably, NOX4 and KDM5A were hyper- and hypo-methylated on their pro- Citation: Kim, H.; Ji, H.W.; Kim, H.W.; Yun, S.H.; Park, J.E.; Kim, S.J. Ginsenoside Rg3 Prevents Oncogenic Long Noncoding RNA ATXN8OS from Inhibiting Tumor-Suppressive microRNA-424-5p in Breast Cancer Cells. Biomolecules 2021, 11, 118. https://doi.org/10.3390/biom 11010118 Received: 30 December 2020 Accepted: 14 January 2021 Published: 18 January 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Biomolecules 2021, 11, 118. https://doi.org/10.3390/biom11010118 https://www.mdpi.com/journal/biomolecules biomolecules(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Biomolecules 2021, 11, 118 2 of 13 moter regions, respectively, which led to gene dysregulation and increases in cell apopto- sis [8]. Other genes such as p53, Bcl-2, and EGF were affected by Rg3-mediated promoter methylation in the HepG2-hepatocarcinoma cell line [9]. Approximately a dozen (mi- croRNAs) miRs are known to be regulated by Rg3, many of which are involved in cancer malignancy, metastasis, or EMT [10,11]. For example, miR-145 comprises the DNMT3A- miR-145-FSCN1 axis in ovarian cancer, and its downregulation by Rg3 inhibits EMT [12]. Recently, miRs associated with the Warburg effect [13] and autophagy [14] were identiﬁed as Rg3 targets. Rg3 upregulated miR-519a-5p via reducing DNMT3A-mediated DNA methylation to inhibit an HIF-1α-stimulated Warburg effect in ovarian cancer [13]. MiR- 181b impaired the antiautophagy effect of Rg3-mediated tumor cytotoxicity by modulating the CREBRF/CREB3 signaling pathways in gallbladder cancer [14]. LncRNAs (i.e., noncoding RNAs larger than 200 nucleotides) are known to regulate a variety of genes, leading to tumor-development stimulation or suppression [15]; however, only a few lncRNAs have been identiﬁed as Rg3 targets. LncRNA-CASC2 is upregulated by Rg3, thereby activating PTEN signaling and suppressing drug-resistant pancreatic cancer cells [16]. Two tumor-related lncRNAs (RFX3-AS1 and STXBP5-AS1) have been identiﬁed in Rg3-treated MCF-7 cells, and their expression is controlled by promoter methylation [17]. Moreover, lncRNA CCAT1 induces Caco-2 colorectal-cancer-cell proliferation but is also downregulated by Rg3 [18]. A number of epigenetic factors have been found to act in conjunction to regulate the expression of speciﬁc target genes. Moreover, competitive endogenous RNA (ceRNA) sponges miR by sharing the same target gene recognition sequence [19]. For example, lncRNA H19 acts as a miR-340-3p sponge to promote epithelial-mesenchymal transition in breast-cancer cells [20], thereby disrupting the gene-suppression activity of miR. Although ginsenosides are known to regulate miRs and lncRNAs in cancer cells, few studies have characterized the role of ceRNA. In this study, a genome-wide methylation-array dataset was analyzed to identify lncRNAs that were epigenetically regulated by Rg3. Notably, the lncRNA ATXN8OS was found to be hypermethylated by Rg3 in MCF-7 breast-cancer cells. The effect of Rg3 on ATXN8OS expression was then examined, and the role of the lncRNA in cancer-cell growth was elucidated. A miR that interacts with ATXN8OS was examined to identify sponge-activity relationships between the two RNAs during miR-mediated gene regulation in the presence of Rg3. 2. Materials and Methods 2.1. Cell Culture Human mammary-gland-derived cell lines (MCF-10A, MCF-7, and MDA-MB-231) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). MCF-10A was cultured in MEGM (Lonza, Basel, Switzerland) with 100 ng/mL cholera toxin. MCF-7 and MDA-MB-231 were cultured in RPMI 1640 medium (Welgene, Seoul, Korea) supplemented with 10% fetal bovine serum (Capricorn Scientiﬁc, Ebsdorfergrund, Germany). All cells were supplemented with 2% penicillin/streptomycin (Capricorn Scientiﬁc) and cultured at 37 ◦C with 5% CO2 in a humidiﬁed incubator. 2.2. Rg3 Treatment and Transfection 5 × 104 cells were seeded in a 60 mm culture dish with 50% conﬂuence and cultured for 24 h before Rg3 treatment or transfection. The cells were then treated with 20 and 50 µM of Rg3 using a 20 mM Rg3 stock (LKT Labs, St. Paul, MN, USA) in 100% ethanol. For transfection, siRNA (Bioneer, Daejon, Korea), mimic miR (Bioneer), and inhibitor miR (Bioneer) were diluted to ﬁnal concentrations of 20 and 40 nM in Opti-MEM Medium (In- vitrogen, Carlsbad, CA, USA), mixed with 5 µL of Lipofectamine RNAiMAX (Invitrogen), and added to the cell culture. For Rg3 and RNA cotreatments, RNA was processed follow- ing the aforementioned transfection protocol, and, after 24 h, Rg3 was added. The cells were further cultured for 24 h and then harvested using 0.05% trypsin-EDTA (Gibco BRL, Carlsbad, CA, USA). Biomolecules 2021, 11, 118 3 of 13 2.3. Rg3-Quantitative Reverse-Transcription Polymerase Chain Reaction (qRT-PCR) Chromosomal DNA and total RNA were extracted from the 60 mm culture dishes using the ZR-Duet DNA/RNA MiniPrep kit (Zymo Research, Irvine, CA, USA) and eluted to 50 and 20 µL, respectively. MiR cDNA was synthesized from 1 µg of total RNA using a miScript II RT kit (Qiagen, Valencia, CA, USA) in 20 µL reactions. qRT-PCR was conducted with 3 µL cDNA per reaction using the miScript SYBR Green PCR kit (Qiagen) and miScript Primer Assay kit (Qiagen). mRNA cDNA was synthesized from 2 µg of total RNA using ReverTra Ace qPCR RT Master Mix (Toyobo, Osaka, Japan) in 10 µL reactions. PCR was then conducted from 1 µL cDNA using SYBR Fast qPCR Kit Master Mix (Kapa Biosystems, Wilmington, MA, USA). The expression of miR and mRNA samples was normalized to that of U6 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), respectively. PCR was performed with an ABI 7300 instrument (Applied Biosystems, Foster City, CA, USA), and the expression level was calculated following the 2−∆∆Ct method. Methylation-speciﬁc PCR was performed with bisulﬁte-treated DNA, and the methylation level was calculated by the 1/[1+2− (CTu−CTme)] × 100% method, as previously described [21]. PCR primers are listed in Supplementary Table S1. 2.4. Data Mining LncRNAs showing a signiﬁcant methylation change by Rg3 were retrieved after analyzing the methylation-array data of the NCBI GEO DataSet (GSE99505). LncBase Predicted v.2 (http://diana.imis.athena-innovation.gr/DianaTools) and StarBase v3.0 ( http://starbase.sysu.edu.cn/index.php) were used to identify miRs that potentially interact with ATXN8OS. MiR target genes were selected using ﬁve miR target-prediction programs: MicroT (www.microrna.gr/microT-v4), RNA22 (https://cm.jefferson.edu/rna22), Tar- getScan7 (http://www.targetscan.org/vert_72), miRWalk (http://http://mirwalk.umm. uni-heidelberg.de), and miRmap (https://mirmap.ezlab.org). 2.5. Cell Proliferation and Apoptosis Assay The effect of Rg3 and noncoding RNAs on cell growth was analyzed by a dye-based cell-proliferation assay as previously described [22]. Brieﬂy, 2 × 103 cells were seeded per well on a 96-well plate and cultured for 24 h. Afterward, the cells were treated with either Rg3 or noncoding RNA and cultured for up to six additional days. After an appropriate culture period, the cells were stained with WST-8 using the Cell Counting Kit-8 (CCK-8) (Enzo Biochem, New York, NY, USA) to measure cell density at OD450 using a spectropho- tometer. For the apoptosis analysis, 1 × 106 cells were seeded in a 60 mm plate, treated with Rg3 or transiently transfected with siRNA, and cultured for 24 h. After harvesting, 1 × 105 cells were suspended in a 1x binding buffer provided with the Annexin V-FITC Apoptosis Detection kit II (BD Bioscience, San Jose, CA, USA), then stained with FITC Annexin V(BD Bioscience) and PI (Sigma-Aldrich, St. Louis, MO, USA). Fluorescence was detected with a BD Accuri C6 ﬂow cytometer (BD Bioscience), and the data were analyzed with the BD Accuri C6 software (BD Bioscience). Cell-cycle analysis was performed using a ﬂow cytometer as previously described [23]. The cell-proliferation index was calculated using the following formula: proliferation index = (S+G2+M)/(G0/G1+S+G2+M) × 100 (%), where each letter represents the number of cells at each stage. 2.6. Western Blot Analysis Proteins were extracted from the harvested cells using ice-cold RIPA lysis buffer (Thermo Fisher Scientiﬁc, Waltham, MA, USA) with a 1% protease-inhibitor cocktail (Thermo Fisher Scientiﬁc). The proteins (15 µg) were then subjected to SDS-PAGE, blotted on a PVDF membrane (Sigma-Aldrich), and treated with primary antibodies overnight at 4 ◦C. The blot was then incubated with HRP-conjugated antirabbit IgG antibodies (1:1000, GTX213110-01; GeneTex, Irvine, CA, USA) for 2 h. The signals were visualized with the ECL reagent (Abfrontier, Seoul, Korea), quantiﬁed using the Image Lab software (Bio-Rad, Hercules, CA, USA), and normalized with β-actin. The antibodies used were anti-CHRM3 Biomolecules 2021, 11, 118 4 of 13 (1:1000, GTX111637; GeneTex), anti-DACH1 (1:1000, A303-556A-M; Bethyl, Montgomery, TX, USA), and anti-β-actin (1:1000, bs-0061R; Bioss, Woburn, MA, USA). 2.7. Statistical Analyses All experiments were independently conducted in triplicate, and the results were expressed as the mean ± SD. Statistical analyses were performed using the SPSS 23.0 software (SPSS, Chicago, IL, USA). T-tests, originally created by Two-tailed Student, were performed to analyze the qRT-PCR, Western blot, and apoptosis assay results. p-value < 0.05 was considered statistically signiﬁcant. 3. Results 3.1. Rg3 Induces Hypermethylation and Downregulation of ATXN8OS We previously performed a genome-wide methylation analysis of Rg3-treated MCF-7 breast-cancer cells [8]. In addition to 866,895 CpGs in protein-coding genes, the array covered 10,733 CpGs in noncoding RNAs. Six lncRNAs exhibited signiﬁcant methylation changes in the promoter (i.e., \|methylation level change (∆β)\| > 1.5 and \|methylation fold change\| > 1.4) (Figure 1A). Given that many lncRNAs have been linked to the development of various cancer types, our study focused on their regulatory mechanisms. ATXN8OS was selected for further study as it exhibited the highest methylation level change (∆β = 0.189). Although little is known about its role in cancer development and progression, previous studies indicate that ATXN8OS has oncogenic properties and therefore stimulates cancer- cell growth [24]. The induction of hypermethylation at the ATXN8OS promoter by Rg3 was veriﬁed via methylation-speciﬁc PCR in MCF-7 cells treated with 20 and 50 µM of Rg3. This experiment resulted in a similar methylation change (methylation-fold change = 1.4 and ∆β = 1.5) to that of the array-based analysis. Moreover, according to the qRT-PCR analysis, ATXN8OS was downregulated by up to 76% in the Rg3-treated MCF-7 cells (Figure 1B). As Rg3 is known to share a structural similarity with estrogen [25], regulation of ATXN8OS may be affected by the estrogen-receptor (ER) status. To test this, the effect was examined in an ER-negative breast-cancer cell line, MDA-MB-231, and in an ER-positive normal cell line, MCF-10A. The result showed that expression of ATXN8OS was less affected in MDA-MB-231 than in the other two cell lines (Supplementary Figure S1), possibly implying an ER dependence on Rg3 for ATXN8OS regulation. To address how ATXN8OS contributes to cancer-cell growth, its downregulation was induced using two siRNAs (siATXN8OS#1 and #2) in MCF-7, which targeted different sites of ATXN8OS (Supplementary Table S1, Supplementary Figure S2), after which cell proliferation and apoptosis were monitored. It was found that ATXN8OS siRNA sup- pressed cancer-cell growth by up to 18%, increased apoptosis by up to 5%, and decreased the cell-proliferation index from 36.7% to 21.5% (Figure 1C–F; Supplementary Figure S3). These results suggest that ATXN8OS promotes proliferation by stimulating the MCF-7 cancer-cell growth while also suppressing apoptosis. Biomolecules 2021, 11, 118 5 of 13 Figure 1. ATXN8OS with proproliferation activity in the MCF-7 cells was downregulated by Rg3 via promoter methylation. (A) ATXN8OS was among the six lncRNAs that exhibited significant changes in methylation level (\|∆β\| ≥ 0.15 and \|fold change\| ≥ 1.4), as demonstrated by the analysis of an Rg3-treated MCF-7-cell methylation array. (B) MCF-7 cells were treated with 20 and 50 µM of Rg3, and the methylation and expression of ATXN8OS were examined by methylation-specific PCR and qRT-PCR, respectively. (C) ATXN8OS was downregulated in MCF-7 using siRNA, and its effect on cell proliferation was examined in the presence of Rg3 using the CCK-8 assay. (D,E) The effect of ATXN8OS on apoptosis (D) and cell cycle (E) was monitored using flow cytometry. All experiments were performed in triplicate, and the values are presented as the mean ± SE. siNC, control siRNA (40 µM); siATXN8OS, ATXN8OS-specific siRNA (40 µM). * p < 0.05, p < 0.01, * p < 0.001. 3.2. ATXN8OS Stimulates Cancer-Cell Proliferation via Sponging miR-424-5p LncRNAs are known to often interact with and regulate miRs and act as ceRNA to modulate the expression of miR target genes. Therefore, our study sought to identify potential miRs for ATXN8OS. Three candidates were identiﬁed upon screening the LncBase Biomolecules 2021, 11, 118 6 of 13 and StarBase public databases, which offer potential partner miRs for lncRNAs (Figure 2A). MiR-424-5p was selected for further analysis as it showed the highest binding score. Rg3 treatment in MCF-7 cells induced the upregulation of the miR (Figure 2B). To see whether ATXN8OS could regulate miR-424-5p, the expression of the miR was quantiﬁed via qRT- PCR in MCF-7 cells treated with ATXN8OS-speciﬁc siRNA (siATXN8OS). Compared to the scrambled siRNAs, siATXN8OS signiﬁcantly increased the expression of miR-424-5p (Figure 2C). The expression of ATXN8OS was then examined after deregulating miR-424-5p using a mimic or an inhibitor RNA (Supplementary Figure S1). Interestingly, the miR-424- 5p mimic RNA downregulated ATXN8OS, whereas the inhibitor upregulated the lncRNA (Figure 2D). Figure 2. ATXN8OS and miR-424-5p sponge each other. (A) Three miRs that could potentially bind ATXN8OS were screened in silico using two miR-prediction databases (LncBase Predicted v.3 and StarBase). (B) miR-424-5p exhibited the highest binding score and was therefore examined to characterize its regulation by Rg3. MCF-7 cells were treated with Rg3, and the RNA expression was quantiﬁed by qRT-PCR. (C,D) The association between the ATXN8OS and miR-424-5p expression was monitored by examining the expression of each RNA after inhibiting ATXN8OS using siRNA (C) and overexpressing (40 µM) or inhibiting miR-424-5p (20 µM) (D). All experiments were performed in triplicate, and the values are presented as the mean ± SE. Testing was done using siNC, negative control siRNA (40 µM); siATXN8OS, ATXN8OS-speciﬁc siRNA (40 µM); mimic NC, negative control mimic for miR-424-5p (40 µM); and inhibitor NC, negative control inhibitor for miR-424-5p (20 µM). p < 0.01, * p < 0.001. Afterward, the effect of miR-424-5p on MCF-7 cell proliferation and apoptosis in the presence of Rg3 was examined after deregulating miR-424-5p in combination with Rg3. As shown in Figure 3A, cell growth was suppressed by 30% using the miR mimic alone, and further decreased by Rg3 exposure in a dose-dependent manner. The miR mimic Biomolecules 2021, 11, 118 7 of 13 increased apoptosis by 15% (Figure 3B). In contrast, the miR-424-5p inhibitor reversed the effect of the mimic RNA by increasing cell growth while decreasing apoptosis of MCF-7 (Figure 3C,D). Therefore, we concluded that Rg3 inhibited the proproliferation effect of the miR-424-5p inhibitor. Figure 3. MiR-424-5p inhibited MCF-7 cell proliferation. MiR-424-5p was deregulated in MCF-7 by transiently transfecting the cells with a mimic (A,B) or an inhibitor (C,D), after which cell proliferation and apoptosis were assessed with the CCK-8 assay and flow-cytometry analysis. Rg3 was coadministered with the mimic (40 µM) or inhibitor (20 µM) for the proliferation assay. Testing was done using mimic NC, negative control miR-424-5p mimic (40 µM) and inhibitor NC, negative control inhibitor for miR-424-5p (20 µM). All experiments were performed in triplicate, and the results are presented as the mean ± SE. Representative images are shown for flow-cytometry analysis. * p < 0.05, p < 0.01, * p < 0.001. 3.3. MiR-424-5p Target Genes are Regulated by ATXN8OS Given the regulatory effect of miRs on target genes, we sought to determine whether ATXN8OS also affects target-gene expression. Potential targets were ﬁrst identiﬁed using the ﬁve target-gene prediction algorithms described in the Materials and Methods, which rendered 200 candidate genes according to all ﬁve prediction tools (Figure 4A). To nar- row down the number of target genes, the pool was then further ﬁltered by applying genome-wide methylation-array data, which were obtained from the Rg3-treated MCF-7 Biomolecules 2021, 11, 118 8 of 13 cells (GSE99505). We aimed to identify target genes that were controlled by miR-424-5p and subject to promoter methylation by Rg3. Through this double-ﬁltering approach, seven genes were identiﬁed, satisfying both the target-gene prediction and the methylation crite- ria (\|∆β\| > 1.5) (Figure 4B). Speciﬁcally, our study focused on EYA1, CHRM3, and DACH1 because they had a target sequence for miR-424-5p (Figure 4C) and showed hypermethy- lation in the array data, suggesting that they were downregulated by Rg3. Additionally, these three genes had previously been reported to possess oncogenic properties in several cancer types [26,27], except DACH1, which functioned as either a tumor promoter [28] or suppressor [29] depending on the cancer type. Consistent with the hypermethylation status, EYA1, CHRM3, and DACH1 were downregulated by 39–95% by Rg3, as determined by our qRT-PCR assays (Figure 4D). ATXN8OS inhibition resulted in downregulation of all the target genes (Figure 4E). Moreover, the miR-424-5p mimic downregulated the three target genes, whereas an inhibitor upregulated them (Figure 4F,G). Figure 4. Regulation of miR-424-5p target genes by Rg3 and ATXN8OS. Potential miR-424-5p target genes were identified by analyzing five public databases (miRmap, miRWalk, TargetScan, MicroT, and RNA22) (A), after which they were compared with the methylation-array data of the Rg3-treated MCF-7 cells (GSE99505) (B). (C) Potential binding sequence of the target genes on miR-424-5p. The seed sequence is denoted in bold. (D–G) Effect of Rg3, ATXN8OS, and miR-424-5p on miR-424-5p target-gene expression. Gene expression was examined by qRT-PCR for samples treated with Rg3 (D), ATXN8OS-specific siRNA (40 µM) (E), miR-424-5p mimic (40 µM) (F), and a miR-424-5p inhibitor (20 µM) (G). Testing was done using siNC, control siRNA (40 µM); mimic NC, negative control mimic for miR-424-5p (40 µM); and inhibitor NC, negative control inhibitor for miR-424-5p (20 µM). All experiments were performed in triplicate, and the results are presented as the mean ± SE. p < 0.01, * p < 0.001. Biomolecules 2021, 11, 118 9 of 13 The protein expression of DACH1 and CHRM3 was then examined by Western blot analysis. DACH1 and CHRM3 protein-expression exhibited a similar proﬁle to that of the transcripts. Speciﬁcally, protein expression was downregulated by Rg3, siATXN8OS, and a miR-424-5p mimic RNA but upregulated by the miR-424-5p inhibitor (Figure 5 and Supplementary Figure S2). The EYA1 protein was barely detected in MCF-7 as in a previous study [30]. Therefore, further conﬁrmation of the effect of Rg3 and noncoding RNAs at the protein level was deemed unnecessary. Overall, Rg3 downregulated EYA1, DACH1, and CHRM3 via the Rg3/ATXN8OS/miR-424-5p axis, whereas ATXN8OS inhibited the miR to modulate the expression of the target gene (Figure 6). Figure 5. Effect of Rg3, ATXN8OS, and miR-424-5p on the target genes of miR-424-5p at the protein level. Western blot analysis of CHRM3 and DACH1 was performed after treating the MCF-7 cells with Rg3 (A) or deregulating ATXN8OS (40 µM siRNA) and miR-424-5p (40 µM for mimic and inhibitor) (B,C). Testing was done using siNC, control siRNA (40 µM); mimic NC, negative control mimic for miR-424-5p (40 µM); and inhibitor NC, negative control inhibitor for miR-424-5p (20 µM). The band intensity was measured with the Image Lab software and indicated by bar graphs. * p < 0.05, p < 0.01, * p < 0.001. Biomolecules 2021, 11, 118 10 of 13 Figure 6. Schematic of the Rg3/ATXN8OS/miR-424-5p axis regulation process. ATXN8OS downreg- ulates the tumor-suppressive miR-424-5p, which in turn activates oncogenic CHRM3 and DACH1, leading to cancer-cell proliferation. Rg3 blocks the oncogenic activity of ATXN8OS by inducing promoter hypermethylation. 4. Discussion Our study aimed to identify lncRNAs that are dysregulated in Rg3-treated cancer cells to elucidate the mechanisms by which they control cancer-cell proliferation, with a particu- lar focus on ceRNA-miR interaction. Most studies on ATXN8OS have so far examined the genetic expansion of CAG repeats. For instance, spinocerebellar ataxia type 8 (SCA8), an autosomal dominant neurodegenerative disease, is caused by CTA/CTG repeat expansion in the ATXN8OS gene [31]. In contrast, little is known about the role of ATXN8OS in tumor development. Recently, Deng et al. found that ATXN8OS stimulated the prolif- eration and migration of MCF-7 and MDA-MB-231 breast-cancer cells [24]. Speciﬁcally, the authors reported that ATXN8OS sequestered the tumor-suppressive miR-204. However, the mechanisms by which miR-204 is regulated by Rg3 remain to be determined. Our study revealed that the oncogenic ATXN8OS is epigenetically regulated by Rg3 via promoter methylation. A few other lncRNAs also showed methylation level changes: DOCK4-AS1, LINC00911, and RFX3-AS1 were hypermethylated, whereas STXBP5-AS1 and LINC01477 were hypomethylated. Notably, LINC00911 and RFX3-AS1 are known as oncogenes [17,32], whereas STXBP5-AS1 is known as a tumor suppressor [33]. These ﬁndings suggest that the tumor-suppressive activity of Rg3 could be attributed in part to its epigenetic regulation of tumor-related lncRNAs. However, the mechanisms by which ATXN8OS methylation is controlled by Rg3 remain to be determined. Moreover, although a close association was identiﬁed between gene methylation and expression levels, additional studies are required to determine whether inducing hypermethylation could drive gene downregulation. MiR-424-5p has been shown to reduce cell viability by modulating the PTEN/PI3K/AKT /mTOR pathway in breast-cancer cells [34], the MAPK pathway in ischemic stroke [35], and the Hippo-signaling pathway in thyroid cancer [36]. MiR-424-5p target genes have been Biomolecules 2021, 11, 118 11 of 13 identified in various cancer types, including PD-L1 [34], VEGFA [37], and ARK5 [38]. These target genes generally exert a protumor activity by promoting proliferation, migration, or angiogenesis in cancer cells. A few lncRNAs have been found to regulate miR-424-5p in various cancer cells, including LINC00922 in breast cancer [39], CDNK2B-AS1 in hepatocel- lular carcinoma [40], and XIST in neuroendocrine tumors [41]. In all the aforementioned cases, regulation of miR-424-5p by the corresponding lncRNA resulted in cell proliferation or cancer-progression alterations. Limited cases of ceRNA have been identiﬁed in ginsenosides. However, there are reports of an Rg3-regulated lncRNA H19 that sponges miR-324-5p to enhance PKM2 expres- sion by directly binding the miR [42]. In another study, Rg1 inhibited high glucose-induced mesenchymal activation by downregulating lncRNA RP11-982M15.8 but upregulating miR-2133 to decrease Zeb1 [43]. The current study suggests a novel ceRNA relationship between the Rg3-regulated ATXN8OS and miR-424-5p, which is supported by the follow- ing ﬁndings: First, the expression of miR-424-5p increased after ATXN8OS was inhibited and vice versa. The lncRNA-induced miR regulation may increase through binding sites with special sequences or paring topology, which would trigger miR degradation upon binding [44]. Second, the two noncoding RNAs had opposite effects on the target-gene expression and the MCF-7 cell growth. Nonetheless, the mechanical interaction between the two RNAs should be elucidated to conﬁrm the proposed ceRNA relationship. Our study had a few noteworthy limitations. Particularly, all of our ﬁndings were based on the analysis of a single lncRNA. Therefore, data on lncRNAs other than ATXN8OS should be obtained to comprehensively explore how Rg3-regulated lncRNAs affect cancer- cell survival or proliferation. Additionally, further studies on other lncRNAs identiﬁed herein such as RFX3-AS1, DOCK4-AS1, and STXBP5-AS1 could provide useful insights. 5. Conclusions ATXN8OS was identiﬁed as a lncRNA that can be downregulated via promoter hy- permethylation by Rg3 in MCF-7 cancer cells. Moreover, ATXN8OS was found to induce the proliferation of cancer cells and this was suppressed by Rg3. At the molecular level, ATXN8OS sponged a tumor-suppressive miR-424-5p, thereby activating key oncogenes such as EYA1, DACH1, and CHRM3, which could be suppressed by Rg3 treatment. There- fore, our ﬁndings suggest that Rg3 suppresses MCF-7 cancer-cell proliferation but increases apoptosis by modulating the ATXN8OS/miR-424-5p/target-gene axis. Supplementary Materials: The following are available online at https://www.mdpi.com/2218-273 X/11/1/118/s1. Table S1: PCR primers, siRNA, miR-mimic, and miR-inhibitor used in this study, Figure S1: Regulation of ATXN8OS and miR-424-5p by Rg3 in mammary gland-derived cell lines, Figure S2: Induction of deregulation of ATXN8OS and miR-424-5p in MCF-7, Figure S3: Effect of ATXN8OS on apoptosis, cell growth, and cell cycle, Figure S4: Uncropped Western blots. Author Contributions: Conceptualization, S.J.K.; methodology, H.K. and H.W.J.; validation, H.W.K., S.H.Y., and J.E.P.; data curation, H.K. and H.W.J.; writing—original draft preparation, H.K., H.W.J., and S.J.K.; writing—review and editing, H.K. and S.J.K.; funding acquisition, S.J.K. All authors have read and agreed to the published version of the manuscript. Funding: This study was supported by the Basic Science Research Program (NRF-2016R1D1A1B01009235) of the National Research Foundation of Korea funded by the Ministry of Education, Science, and Tech- nology. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: All data are contained within the article or supplementary material. Conﬂicts of Interest: The authors declare no conﬂict of interest. Biomolecules 2021, 11, 118 12 of 13 References 1. 2. 3. 4. Nakhjavani, M.; Hardingham, J.E.; Palethorpe, H.M.; Tomita, Y.; Smith, E.; Price, T.J.; Townsend, A.R. Ginsenoside Rg3: Potential molecular targets and therapeutic indication in metastatic breast cancer. Medicines 2019, 6, 17. [CrossRef] Sun, M.; Ye, Y.; Xiao, L.; Duan, X.; Zhang, Y.; Zhang, H. Anticancer effects of ginsenoside Rg3 (Review). Int. J. Mol. Med. 2017, 39, 507–518. [CrossRef] [PubMed] Kim, B.M.; Kim, D.H.; Park, J.H.; Na, H.K.; Surh, Y.J. Ginsenoside Rg3 Induces Apoptosis of Human Breast Cancer (MDA-MB-231) Cells. Eur. J. Cancer Prev. 2013, 18, 177–185. [CrossRef] [PubMed] Kim, B.-M.; Kim, D.-H.; Park, J.-H.; Surh, Y.-J.; Na, H.-K. Ginsenoside Rg3 inhibits constitutive activation of NF-κB signaling in human breast cancer (MDA-MB-231) cells: ERK and Akt as potential upstream targets. Eur. J. Cancer Prev. 2014, 19, 23–30. [CrossRef] [PubMed] 6. 5. Mao, X.; Jin, Y.; Feng, T.; Wang, H.; Liu, D.; Zhou, Z.; Yan, Q.; Yang, H.; Yang, J.; Yang, J.; et al. Ginsenoside Rg3 Inhibits the Growth of Osteosarcoma and Attenuates Metastasis through the Wnt/β-Catenin and EMT Signaling Pathway. Evid. Based Complement Altern. Med. 2020, 2020, 6065124. [CrossRef] [PubMed] Li, B.; Qu, G. Inhibition of the hypoxia-induced factor-1α and vascular endothelial growth factor expression through ginsenoside Rg3 in human gastric cancer cells. J. Cancer Res. Ther. 2019, 15, 1642–1646. [CrossRef] Phi, L.T.H.; Wijaya, Y.T.; Sari, I.N.; Kim, K.S.; Yang, Y.G.; Lee, M.W.; Kwon, H.Y. 20(R)-Ginsenoside Rg3 Inﬂuences Cancer Stem Cell Properties and the Epithelial-Mesenchymal Transition in Colorectal Cancer via the SNAIL Signaling Axis. Onco Targets Ther. 2019, 12, 10885–10895. [CrossRef] 7. 8. Ham, J.; Lee, S.; Lee, H.; Jeong, D.; Park, S.; Kim, S.J. Genome-Wide Methylation Analysis Identiﬁes NOX4 and KDM5A as Key Regulators in Inhibiting Breast Cancer Cell Proliferation by Ginsenoside Rg3. Am. J. Chin. Med. 2018, 46, 1333–1355. [CrossRef] Teng, S.; Wang, Y.; Li, P.; Liu, J.; Wei, A.; Wang, H.; Meng, X.; Pan, D.; Zhang, X. Effects of R type and S type ginsenoside Rg3 on DNA methylation in human hepatocarcinoma cells. Mol. Med. Rep. 2017, 15, 2029–2038. [CrossRef] 9. 10. Cheng, Z.; Xing, D. Ginsenoside Rg3 inhibits growth and epithelial-mesenchymal transition of human oral squamous carcinoma cells by down-regulating miR-221. Eur. J. Pharmacol. 2019, 853, 353–363. [CrossRef] 11. Lee, A.; Yun, E.; Chang, W.; Kim, J. Ginsenoside Rg3 protects against iE-DAP–induced endothelial-to-mesenchymal transition by regulating the miR-139-5p–NF-κB axis. J. Ginseng Res. 2020, 44, 300–307. [CrossRef] [PubMed] 12. Li, J.; Lu, J.; Ye, Z.; Han, X.; Zheng, X.; Hou, H.; Chen, W.; Li, X.; Zhao, L. 20(S)-Rg3 blocked epithelial-mesenchymal transition through DNMT3A/miR-145/FSCN1 in ovarian cancer. Oncotarget 2017, 8, 53375–53386. [CrossRef] [PubMed] 13. Lu, J.; Chen, H.; He, F.; You, Y.; Feng, Z.; Chen, W.; Li, X.; Zhao, L. Ginsenoside 20(S)-Rg3 upregulates HIF-1α-targeting miR-519a-5p to inhibit the Warburg effect in ovarian cancer cells. Clin. Exp. Pharmacol. Physiol. 2020, 47, 1455–1463. [CrossRef] [PubMed] 14. Wu, K.; Huang, J.; Xu, T.; Ye, Z.; Jin, F.; Li, N.; Lv, B. MicroRNA-181b blocks gensenoside Rg3-mediated tumor suppression of gallbladder carcinoma by promoting autophagy ﬂux via CREBRF/CREB3 pathway. Am. J. Transl. Res. 2019, 11, 5776–5787. Schmitt, A.M.; Chang, H.Y. Long Noncoding RNAs in Cancer Pathways. Cancer Cell. 2016, 29, 452–463. [CrossRef] [PubMed] 15. 16. Zou, J.; Su, H.; Zou, C.; Liang, X.; Fei, Z. Ginsenoside Rg3 suppresses the growth of gemcitabine-resistant pancreatic cancer cells by upregulating lncRNA-CASC2 and activating PTEN signaling. J. Biochem. Mol. Toxicol. 2020, 34, e22480. [CrossRef] 17. Ham, J.; Jeong, D.; Park, S.; Kim, H.W.; Kim, H.; Kim, S.J. Ginsenoside Rg3 and Korean Red Ginseng extract epigenetically regulate the tumor-related long noncoding RNAs RFX3-AS1 and STXBP5-AS1. J. Ginseng Res. 2019, 43, 625–634. [CrossRef] 18. Li, J.; Qi, Y. Ginsenoside Rg3 inhibits cell growth, migration and invasion in Caco-2 cells by downregulation of lncRNA CCAT1. Exp. Mol. Pathol. 2019, 106, 131–138. [CrossRef] 19. Tay, Y.; Rinn, J.; Pandolﬁ, P.P. The multilayered complexity of ceRNA crosstalk and competition. Nature 2014, 505, 344–352. [CrossRef] 20. Yan, L.; Yang, S.; Yue, C.X.; Wei, X.Y.; Peng, W.; Dong, Z.Y.; Xu, H.N.; Chen, S.L.; Wang, W.R.; Chen, C.J.; et al. Long noncoding RNA H19 acts as a miR-340-3p sponge to promote epithelial-mesenchymal transition by regulating YWHAZ expression in paclitaxel-resistant breast cancer cells. Environ. Toxicol. 2020, 35, 1015–1028. [CrossRef] 21. Kim, S.J.; Kelly, W.K.; Fu, A.; Haines, K.; Hoffman, A.; Zheng, T.; Zhu, Y. Genome-wide methylation analysis identiﬁes involvement of TNF-α mediated cancer pathways in prostate cancer. Cancer Lett. 2011, 302, 47–53. [CrossRef] [PubMed] 22. Lee, S.; Lee, H.; Bae, H.; Choi, E.H.; Kim, S.J. Epigenetic silencing of miR-19a-3p by cold atmospheric plasma contributes to proliferation inhibition of the MCF-7 breast cancer cell. Sci. Rep. 2016, 6, 30005. [CrossRef] [PubMed] 23. Kang, S.; Kim, B.; Kang, H.S.; Jeong, G.; Bae, H.; Lee, H.; Lee, S.; Kim, S.J. SCTR regulates cell cycle-related genes toward anti-proliferation in normal breast cells while having pro-proliferation activity in breast cancer cells. Int. J. Oncol. 2015, 47, 1923–1931. [CrossRef] [PubMed] 24. Deng, Z.; Cai, H.; Lin, L.; Zhu, L.; Wu, W.; Yang, S.; Cai, J.; Tan, J. lncRNA ATXN8OS promotes breast cancer by sequestering miR-204. Mol. Med. Rep. 2019, 20, 1057–1064. [CrossRef] 25. Tian, M.; Li, L.N.; Zheng, R.R.; Yang, L.; Wang, Z.T. Advances on hormone-like activity of Panax ginseng and ginsenosides. Chin. J. Nat. Med. 2020, 18, 526–535. [CrossRef] 26. Cai, S.; Cheng, X.; Liu, Y.; Lin, Z.; Zeng, W.; Yang, C.; Liu, L.; Chukwuebuka, O.A.; Li, W. EYA1 promotes tumor angiogenesis by activating the PI3K pathway in colorectal cancer. Exp. Cell Res. 2018, 367, 37–46. [CrossRef] Biomolecules 2021, 11, 118 13 of 13 27. Wang, N.; Yao, M.; Xu, J.; Quan, Y.; Zhang, K.; Yang, R.; Gao, W.Q. Autocrine Activation of CHRM3 Promotes Prostate Cancer Growth and Castration Resistance via CaM/CaMKK-Mediated Phosphorylation of Akt. Clin. Cancer Res. 2015, 21, 4676–4685. [CrossRef] 28. Hu, X.; Zhang, L.; Li, Y.; Ma, X.; Dai, W.; Gao, X.; Rao, X.; Fu, G.; Wang, R.; Pan, M.; et al. Organoid modelling identiﬁes that DACH1 functions as a tumour promoter in colorectal cancer by modulating BMP signalling. EBioMedicine 2020, 56, 102800. [CrossRef] 29. Xu, H.; Yu, S.; Yuan, X.; Xiong, J.; Kuang, D.; Pestell, R.G.; Wu, K. DACH1 suppresses breast cancer as a negative regulator of CD44. Sci. Rep. 2017, 7, 4361. [CrossRef] 30. Wu, K.; Li, Z.; Cai, S.; Tian, L.; Chen, K.; Wang, J.; Hu, J.; Sun, Y.; Li, X.; Ertel, A.; et al. EYA1 phosphatase function is essential to 31. drive breast cancer cell proliferation through cyclin D1. Cancer Res. 2013, 73, 4488–4499. [CrossRef] Samukawa, M.; Hirano, M.; Saigoh, K.; Kawai, S.; Hamada, Y.; Takahashi, D.; Nakamura, Y.; Kusunoki, S. PSP-Phenotype in SCA8: Case Report and Systemic Review. Cerebellum 2019, 18, 76–84. [CrossRef] [PubMed] 32. Li, X.X.; Wang, L.J.; Hou, J.; Liu, H.Y.; Wang, R.; Wang, C.; Xie, W.H. Identiﬁcation of Long Noncoding RNAs as Predictors of Survival in Triple-Negative Breast Cancer Based on Network Analysis. BioMed Res. Int. 2020, 2020, 8970340. [CrossRef] [PubMed] 33. Cen, D.; Huang, H.; Yang, L.; Guo, K.; Zhang, J. Long noncoding RNA STXBP5-AS1 inhibits cell proliferation, migration, and invasion through inhibiting the PI3K/AKT signaling pathway in gastric cancer cells. Onco Targets Ther. 2019, 12, 1929–1936. [CrossRef] 34. Dastmalchi, N.; Hosseinpourfeizi, M.A.; Khojasteh, S.M.B.; Baradaran, B.; Safaralizadeh, R. Tumor suppressive activity of miR-424-5p in breast cancer cells through targeting PD-L1 and modulating PTEN/PI3K/AKT/mTOR signaling pathway. Life Sci. 2020, 259, 118239. [CrossRef] [PubMed] 35. Xiang, Y.; Zhang, Y.; Xia, Y.; Zhao, H.; Liu, A.; Chen, Y. LncRNA MEG3 targeting miR-424-5p via MAPK signaling pathway mediates neuronal apoptosis in ischemic stroke. Aging (Albany N.Y.). 2020, 12, 3156–3174. [CrossRef] 36. Liu, X.; Fu, Y.; Zhang, G.; Zhang, D.; Liang, N.; Li, F.; Li, C.; Sui, C.; Jiang, J.; Lu, H.; et al. miR-424-5p Promotes Anoikis Resistance and Lung Metastasis by Inactivating Hippo Signaling in Thyroid Cancer. Mol. Ther. Oncolytics. 2019, 15, 248–260. [CrossRef] 37. Vimalraj, S.; Saravanan, S.; Raghunandhakumar, S.; Anuradha, D. Melatonin regulates tumor angiogenesis via miR-424- 5p/VEGFA signaling pathway in osteosarcoma. Life Sci. 2020, 256, 118011. [CrossRef] 38. Wu, J.; Yang, B.; Zhang, Y.; Feng, X.; He, B.; Xie, H.; Zhou, L.; Wu, J.; Zheng, S. miR-424-5p represses the metastasis and invasion of intrahepatic cholangiocarcinoma by targeting ARK5. Int. J. Biol. Sci. 2019, 15, 1591–1599. [CrossRef] 39. Yue, X.; Wang, Z. Long Intergenic Non-Coding RNA LINC00922 Aggravates the Malignant Phenotype of Breast Cancer by 40. Regulating the microRNA-424-5p/BDNF Axis. Cancer Manag. Res. 2020, 12, 7539–7552. [CrossRef] Shen, X.; Li, Y.; He, F.; Kong, J. LncRNA CDKN2B-AS1 Promotes Cell Viability, Migration, and Invasion of Hepatocellular Carcinoma via Sponging miR-424-5p. Cancer Manag. Res. 2020, 12, 6807–6819. [CrossRef] 41. Zhou, K.; Li, S.; Du, G.; Fan, Y.; Wu, P.; Sun, H.; Zhang, T. LncRNA XIST depletion prevents cancer progression in invasive pituitary neuroendocrine tumor by inhibiting bFGF via upregulation of microRNA-424-5p. Onco Targets Ther. 2019, 12, 7095–7109. [CrossRef] [PubMed] 42. Zheng, X.; Zhou, Y.; Chen, W.; Chen, L.; Lu, J.; He, F.; Li, X.; Zhao, L. Ginsenoside 20(S)-Rg3 Prevents PKM2-Targeting miR-324-5p from H19 Sponging to Antagonize the Warburg Effect in Ovarian Cancer Cells. Cell. Physiol. Biochem. 2018, 51, 1340–1353. [CrossRef] [PubMed] 43. Xue, L.P.; Fu, X.L.; Hu, M.; Zhang, L.W.; Li, Y.D.; Peng, Y.L.; Ding, P. Rg1 inhibits high glucose-induced mesenchymal activation and ﬁbrosis via regulating miR-2113/RP11-982M15.8/Zeb1 pathway. Biochem. Biophys. Res. Commun. 2018, 501, 827–832. [CrossRef] [PubMed] Figliuzzi, M.; Marinari, E.; De Martino, A. MicroRNAs as a selective channel of communication between competing RNAs: A steady-state theory. Biophys. J. 2013, 104, 1203–1213. [CrossRef] [PubMed] 44.
10.2196_19678	JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Original Paper The Internet Hospital Plus Drug Delivery Platform for Health Management During the COVID-19 Pandemic: Observational Study Liang Ding1, PhD; Qiuru She2, BS; Fengxian Chen2, BS; Zitong Chen2, BS; Meifang Jiang2, BS; Huasi Huang2, BS; Yujin Li2, MS; Chaofeng Liao2, BS 1Clinical Trial and Research Center, People’s Hospital of Baoan Shenzhen, Shenzhen, China 2Department of Pharmacy, People’s Hospital of Baoan Shenzhen, Shenzhen, China *these authors contributed equally Corresponding Author: Chaofeng Liao, BS Department of Pharmacy, People’s Hospital of Baoan Shenzhen 118 Longjing 2nd Road Shenzhen, 518101 China Phone: 86 18926480093 Email: [email protected] Abstract Background: Widespread access to the internet has boosted the emergence of online hospitals. A new outpatient service called “internet hospital plus drug delivery” (IHDD) has been developed in China, but little is known about this platform. Objective: The aim of this study is to investigate the characteristics, acceptance, and initial impact of IHDD during the outbreak of COVID-19 in a tertiary hospital in South China Methods: The total number of and detailed information on online prescriptions during the first 2 months after work resumption were obtained. Patients’ gender, age, residence, associated prescription department, time of prescription, payment, and drug delivery region were included in the analysis. Results: A total of 1380 prescriptions were picked up or delivered between March 2 and April 20, 2020. The largest group of patients were 36-59 years old (n=680, 49.3%), followed by the 18-35 years age category (n=573, 41.5%). In total, 39.4% (n=544) of the patients chose to get their medicine by self-pickup, while 60.6% (n=836) preferred to receive their medicine via drug delivery service. The top five online prescription departments were infectious diseases (n=572, 41.4%), nephrology (n=264, 19.1%), endocrinology (n=145, 10.5%), angiocardiopathy (n=107, 7.8%), and neurology (n=42, 3%). Of the 836 delivered prescriptions, 440 (52.6%) were sent to Guangdong Province (including 363 [43.4%] to Shenzhen), and 396 (47.4%) were sent to other provinces in China. Conclusions: The IHDD platform is efficient and convenient for various types of patients during the COVID-19 crisis. Although offline visits are essential for patients with severe conditions, IHDD can help to relieve pressure on hospitals by reducing an influx of patients with mild symptoms. Further efforts need to be made to improve the quality and acceptance of IHDD, as well as to regulate and standardize the management of this novel service. (J Med Internet Res 2020;22(8):e19678) doi: 10.2196/19678 KEYWORDS internet hospital; drug delivery; internet hospital plus drug delivery; IHDD; health management; COVID-19 http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Introduction Background As the third largest country in the world by area, China has 34 provincial regions, over 1.4 billion people, but only 10 million licensed physicians (2.2 for every 1000 people) in 2019, according to the National Bureau of Statistics of China [1]. Since the severe acute respiratory syndrome (SARS) epidemic in 2003, the Chinese government has been rebuilding the three-tier health care system. Today, the health care system in China consists of community health centers (CHCs), and secondary and tertiary hospitals in urban areas; and village clinics, township health centers (THCs), and county hospitals in rural areas. CHCs, village clinics, and THCs are considered core primary care providers and are expected to provide affordable first contact care while secondary and tertiary care facilities provide specialist referral services [2]. However, with no gatekeeping in the primary health care system, patients can freely choose their provider at any health facility. Although many disorders could be treated by primary care providers conveniently and at a relatively low price, many patients are unwilling to see these providers owing to their lack of confidence in the health professionals’ skills and the quality of health care provided. They tend to go to high-level hospitals even for mild symptoms, effectively overcrowding those hospitals [3]. On the other hand, skilled doctors are unwilling to work at the community level and in remote rural areas for financial and professional reasons. These two problems have led to countless transprovincial patients, resulting in numerous additional economic and time costs [4]. The rapid increase in internet users (from 22.7% to 59.6% of the population between 2008 and 2018) [5] offers the Chinese government a new alternative to address these health care problems. On October 25, 2014, the first officially approved “internet hospital” went online in Guangdong Province. In the beginning, the internet hospital consisted of four clinics operated by doctors from the Second People’s Hospital of Guangdong Province, an online platform operated by a medical technology company, and a network of medical consulting facilities based in clinics in rural villages, CHCs, and large pharmacy chain stores [4]. The inchoate platform usually required onsite equipment (computers, cameras, speakers, and cable network). Patients needed to go to a medical consultation facility near their home and meet through the internet with the doctor based in a top-level hospital in a big city. With the widespread adoption of smartphones and tablet computers, and the ever-increasing popularity of mobile internet communication, a mobile health (mHealth) care model was made accessible to the public. mHealth allows patients to access information, assessments, and treatments in a timely manner. In addition, it empowers doctors with another way to connect with their patients and to practice without geographical limitations [6]. Therefore, the extra costs of health care, such as those associated with travel, time, and doctor consultations, can be dramatically reduced. http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX During the outbreak of coronavirus disease (COVID-19) [7], the Chinese government adopted a series of administrative measures to stop the spread of the epidemic [8], including requiring domestic internet hospitals to vigorously carry out remote medical services [9]. Although the convenience and ubiquity of internet hospitals makes them a promising avenue through which to overcome geographical limitations between patients and doctors, there is still a “social distance” barrier between the patients and their prescription medicines. In order to solve this problem, many hospitals intend to cooperate with delivery companies to build a partnership for drug delivery [10]. This bundled approach could offer an omnichannel solution that can help people on their path to urgently needed health care and medicine during the epidemic. Objective To explore the advantages of the IHDD model for health management during public health emergencies, we analyzed the prescriptions of online outpatients at the People’s Hospital of Baoan Shenzhen in Shenzhen City, Guangdong. Data from the first 2 months after work resumption were collected to reveal the characteristics, acceptance, and initial impact of the new bundled approach. Methods Data Sources The total number of online prescriptions and detailed information on them were obtained automatically from the hospital information system (DTHealth, V7.0) of the People’s Hospital of Baoan Shenzhen. Data from March 2 to April 20, 2020 were collected. Patients’ gender, age, residence, associated prescription departments, time of prescription, payment, and drug delivery region were included in the analysis. GraphPad Prism 8.0.2 (GraphPad Software) was used to summarize and analyze the data. Ethics Statement Ethics approval was obtained from the medical research ethics committee of the People’s Hospital of Baoan Shenzhen before the start of the study. Results The Internet Hospital Workflow Figure 1 shows the online consultation workflow of the internet hospital. The patients chose a department and doctor by self-assessment using the hospital miniprogram. There was no prescription option for picture/text counseling patients. For the online clinic, a confirmation would be sent by text message once the online clinic appointment was made. Another message would be sent to the patient 3 minutes before the video counseling session began to remind them to open the hospital miniprogram in time. The doctor then initiated a video consultation with the patient and made an online prescription, if necessary, based on the diagnosis of the patient. At the payment step, the patient could choose self-pickup of medication at the hospital’s pharmacy or delivery to an assigned place. J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Figure 1. Online workflow of the internet hospital. Number and Payment Amount of Online Prescriptions A total of 1380 prescriptions were picked up or delivered between March 2 and April 20, 2020. The weekly number and these prescriptions are payment amount pertaining to summarized in Figure 2. There was an increase in the use of the online prescription service. The number and total payments of the 7th week significantly increased, by 11.3 and 4.8 times, respectively, compared with the first week. Figure 2. The number of and total payments pertaining to online prescriptions. Patient Characteristics There was no sex-based differences among the patients who received prescriptions (Table 1). The patients were divided into four groups according to their age: 1-17 years old, 18-35 years old, 36-59 years old, and ≥60 years old. The largest group of patients were 36-59 years old (n=680, 49.3%), followed by those who were 18-35 years old (n=573, 41.5%). In total, 65.7% (n=907) of the patients were local residents, 5.6% (n=77) were from Guangdong cities other than Shenzhen, 28.7% (n=396) were from other provinces in China. Less than half the patients (n=544, 39.4%) chose to receive their medicine by self-pickup, while 60.6% (n=836) preferred to get their medicine by drug delivery service. http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Table 1. Baseline characteristics of patients (N=1380). Characteristic Sex, n (%) Male Female Age (years) Median (range) Group, n (%) 1-17 18-35 36-59 ≥60 Residence, n (%) Local (Shenzhen City) Other cities in Guangdong Province Other provinces Access to medicine, n (%) Delivery Self-pickup Value 693 (50.2) 687 (49.8) 38 (1-93) 12 (0.9) 573 (41.5) 680 (49.3) 115 (8.3) 907 (65.7) 77 (5.6) 396 (28.7) 836 (60.6) 544 (39.4) Distribution of Online Prescriptions The top five online prescription departments were infectious diseases (n=572, 41.4%), nephrology (n=264, 19.1%), endocrinology (n=145, 10.5%), angiocardiopathy (n=107, 7.8%), and neurology (n=42, 3%). The majority of infectious disease and neurology patients chose drug delivery services, while most patients with other diagnoses preferred to pick up their medication (Table 2). Table 2. Delivery/self-pickup preference of online prescription patients (N=1380). Department Delivery (n=836), n (%) Self-pickup (n=544), n (%) Infectious disease (n=572) Nephrology (n=264) Endocrinology (n=145) Angiocardiopathy (n=107) Neurology (n=42) 551 (96.3) 86 (32.6) 48 (33.1) 37 (34.6) 25 (59.5) 21 (3.7) 178 (67.4) 97 (66.9) 70 (65.4) 17 (40.5) Drug Delivery Details of Online Prescriptions For the 836 delivered prescriptions, 440 (52.6%) were sent to Guangdong Province (including 363 [43.4%] to Shenzhen) and 396 (47.4%) were sent to other provinces in China. The top 10 provinces for out-of-province deliveries were Heilongjiang, Hubei, Guangxi, Shandong, Jiangsu, Hunan, Shanxi, Henan, Anhui, and Jiangxi (Table 3). Most of them are located in the northeast, eastern, and central parts of China (Figure 3 and Table 3). The top 10 delivered medicines are listed in Table 4. Most of the medicines were used to treat infectious or chronic disease, which was consistent with the distribution of online prescription. http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Table 3. Out-of-province delivery details based on the geographical regions of China. Region Northeast Eastern Central Southern Northwest Northern Southwest Top 10 provinces Heilongjiang Hubei Guangxi Shandong Jiangsu Hunan Shanxi Henan Anhui Jiangxi Figure 3. Regional distribution of prescription deliveries. Patients (n=396), n (%) 107 (27.0) 102 (25.8) 82 (20.7) 35 (8.8) 35 (8.8) 21 (5.3) 14 (3.5) 94 (23.7) 36 (9.1) 33 (8.3) 27 (6.8) 24 (6.1) 24 (6.1) 24 (6.1) 22 (5.6) 14 (3.5) 14 (3.5) http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Table 4. Top 10 delivered medicines. Name Entecavir Metoprolol succinate Entecavir Metformin hydrochloride Tenofovir disoproxil fumarate Atorvastatin calcium Nifedipine Atorvastatin calcium Mecobalamin Amlodipine besylate Discussion Formulation Dispersible tablet Sustained release tablet Dispersible tablet Tablet Tablet Tablet Controlled release tablet Tablet Tablet Tablet Manufacturer ChiaTai TianQing AstraZeneca AB Dawnrays Bristol-Myers Squibb Brilliant Pfizer Bayer JiaLin Desano Pfizer Principal Findings We conducted a pilot evaluation of the IHDD health care model in a tertiary hospital in Shenzhen during the first 2 months after work resumption. The unbalanced distribution of medical resources and the outbreak of COVID-19 [11,12] promoted the growth and exploration of more convenient internet-based medical practices [13], especially in the well-developed southern and southeastern parts of China, where people use the internet more often for medical purposes [14]. Despite the advantages of internet hospitals, access to medication remained an obstacle that may have discouraged people from using this platform. The traditional internet hospital required patients to go to the hospital or drugstore for medicines, which could cause more infections during an epidemic. The drug pickup process may increase risk of acute infectious disease, particularly for patients with suppressed immune systems or disabilities, which can then lead to severe health deterioration. On the other hand, out-of-city and out-of-province patients could have problems finding the exact prescription medications they need, as those medicines might be not available at their local hospitals and drugstores. In January 2019, the General Office of the State Council of the People's Republic of China the National Centralized Drug Purchasing (NCDP) and Using pilot program and selected 11 cities (including Shenzhen) in mainland China to carry out the “4+7” City-Drug-Volume-Based-Purchasing pilot project [15]. As the “frontier” of Chinese prescription medicine reform, prescription medicines that were made by the doctors of Shenzhen’s hospitals may be more affordable. Therefore, the development of IHDD could enable patients across the country to access online prescription medication in a secure and convenient way. implemented The Chinese government has encouraged internet hospitals to join the epidemic prevention and control efforts of the COVID-19 outbreak [16]. On March 15, 2020, the first professional standard, “Specification for Online Consultation Service for Infectious Disease Epidemic Situation” was published on the national group standard information platform of China, requiring that internet hospitals provide 24/7 online services in response to the epidemic [17]. The internet hospital http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX of the People’s Hospital of Baoan Shenzhen has been officially online since March 2, 2020. From opening to April 20, 2020, it saw a total of 8638 patients, an average of 176 per day, with 5877 in picture/text counseling and 2761 in online clinic video counseling (including 1381 that did not result in a prescription). Most of the picture/text consultations were prehospital services such as psychological counseling and medical education. The number and payment amounts of online prescriptions increased progressively from the first investigated week to the 7th one (Figure 2), which indicates increased acceptance of IHDD. The drop in prescription numbers during the 5th week might be the result of two factors: Tomb-Sweeping holiday and a lack of antiviral medicine. Most prescription patients were between the ages of 18 and 60 years, had no time for onsite visitations, and had greater access to new medical platforms. At present, health authorities and the government have warned older people that they are at a higher risk of more serious and possibly fatal illness associated with COVID-19. Moreover, the global recommendation for older populations includes social isolation, which involves staying at home and avoiding contact with other people for an extended period of time [18]. Our data show that only 8.3% of IHDD users were ≥60 years (Table 1). This may be due to differences in public acceptance. Older populations usually take more time to become familiarized with the operations of IHDD. The stay-at-home order constrained people from going outside, which increased difficulties associated with health management, especially chronic disease management. Medical professionals at hospitals with fever clinics are required to participate in COVID-19 prevention, control, and treatment, which has reduced their concentration on other diseases. In fact, the management of chronic disease has become a crucial issue in cities with large outbreaks of COVID-19 [19]. The largest number of internet hospital prescriptions came from the department of infectious diseases, which includes acute and chronic viral hepatitis, fatty liver, alcoholic hepatitis, drug-induced liver damage, autoimmune liver disease, and genetic and metabolic liver disease. The second largest group was from nephrology, followed by endocrinology. Patients with chronic liver disease, kidney disease, or diabetes could easily renew their prescriptions and receive their medicine by IHDD. J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al A notable finding was that most patients with an infectious disease chose to receive their medicine by delivery (96.3%), whereas most patients with other diseases selected self-pickup (Table 2). This can be explained by the need for special storage of some medicines (eg, recombinant human erythropoietin for kidney disease or insulin for diabetes). Established in 1984, the People’s Hospital of Baoan Shenzhen is also the Eighth People's Hospital of Shenzhen, the Shenzhen Baoan Affiliated Hospital of Southern Medical University, and the Second Affiliated Hospital of Shenzhen University. It was recognized as a Grade A Tertiary Hospital by the Guangdong Health Department in 2012. As an important source of health care providers in Shenzhen, the hospital holds a great reputation in both basic clinical practice and diverse clinical research and training. The IHDD platform enabled patients all over the country to obtain access to its health professionals and quality-assured medicine (Figure 3). The hospital even offered medical service to the patients in Wuhan, the epicenter of COVID-19. In fact, online prescription medicines that were delivered to Hubei Province accounted for the second largest number of all out-of-province deliveries (Table 3). The top 10 provinces for IHDD delivery were located in the northeast, eastern, and central parts of China. One of the reasons is that these areas are relatively economically developed regions, and their residents are highly educated, which ensures they have a better understanding of the benefits of IHDD. The top 10 delivered medicines were used for the treatment of hepatitis, hypertension, hyperlipidemia, diabetes, climacteric symptoms, etc (Table 4). This finding was consistent with the distribution of prescription departments. In fact, 7 of these departments were enrolled in the National Essential Drugs of China program. Within these 7, 3 belonged to the “4+7” NCDP catalog. The affordability and quality of these medications were guaranteed by the government. One of the concerns of IHDD is the safety and security of drug delivery. Therefore, delivery service companies with good reputations were chosen by the hospitals as partners. As the industry leader, SF Express is the first logistics compan y to cooperate with both pharmaceutical pro viders and hospitals. It has a long history of ambient temperature and cold chain medicine transport. Moreover, it offers real-time package tracking and zero-touch delivery. Once the patients place their order, a tracking number is sent by text message to their cell phone. The processing information of the medicine is updated and sent to patients automatically. Couriers place the medicines into the customer-assigned delivery lockers, which are usually near the patient’s residence, so the patient can access them using a random cipher. This process could effectively reduce viral transmission while simultaneously providing convenience for patients. At present, an advanced cooperation model is under exploration: the pharmacist-audited prescription will be sent to the manufacturer directly, and the medicines will be delivered reduce from manufacturer’s substantial pressure on hospital drugstores and cut transportation expenses. stock house. This will severe IHDD can help Future Prospects Although in-person visits are essential when the patients are experiencing to symptoms, relieve pressure on hospitals by reducing the influx of mild cases. To make better use of IHDD during and after the current epidemic, more effort is needed. Simple and clear instructions are necessary to improve its acceptance by older people. Financial support, like adding medical insurance to payment methods, can also promote adoption by the public. The new hospital-manufacturer-patient transport model should be further evaluated. Moreover, official regulations are required in terms of standardization of the operational process and management of IHDD. Conclusion The pandemic of COVID-19 has clearly entered a new stage with rapid spread to countries outside China, becoming a global threat. This once-in-a-century pandemic might permanently change people’s lifestyle, especially when it comes to health management. In our study, IHDD has been proven to be efficient and convenient for many types of patients during the crisis. The to reduce widespread use of person-to-person transmission as well as the infection risk of patients with chronic diseases or disability. this platform can help Acknowledgments This research was supported in part by the Key Laboratory of Emergency and Trauma (Hainan Medical University), the Ministry of Education (KLET-201908), and th eScience, Technology and Inno vation Commission of Shenzhen Municipality (731144920168). We gratefully acknowledge the help of the Information Technology Center, People’s Hospital of Baoan Shenzhen. We also thank Mr Zhiyong Zhang for assistance with data collection. Authors' Contributions LD, QS, and CL designed the study. QS, FC, and YL searched for relevant national and regional information. MJ and HH gathered data. ZC carried out the data analyses. LD wrote the first draft of the paper, with revisions from QS and CL. All authors contributed to revisions and approved the final version. Conflicts of Interest None declared. References http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al 1. 2. Public Health. National Bureau of Statistics of China. 2020. URL: http://data.stats.gov.cn/english/easyquery.htm?cn=C01 [accessed 2020-04-20] Yip WC, Hsiao W, Meng Q, Chen W, Sun X. Realignment of incentives for health-care providers in China. The Lancet 2010 Mar 27;375(9720):1120-1130. [doi: 10.1016/S0140-6736(10)60063-3] [Medline: 20346818] 3. Wu D, Hesketh T, Shu H, Lian W, Tang W, Tian J. Description of an online hospital platform, China. Bull World Health 4. Organ 2019 Aug 01;97(8):578-579 [FREE Full text] [doi: 10.2471/BLT.18.226936] [Medline: 31384077] Tu J, Wang C, Wu S. The internet hospital: an emerging innovation in China. The Lancet Global Health 2015 Aug;3(8):e445-e446 [FREE Full text] [doi: 10.1016/S2214-109X(15)00042-X] [Medline: 26187488] 5. Work dynamics. Ministry of Industry and Information Technology of the People's Republic of China. URL: http://www. miit.gov.cn/n1146290/n1146402/index.html [accessed 2020-04-20] 6. Whitehead L, Seaton P. The Effectiveness of Self-Management Mobile Phone and Tablet Apps in Long-term Condition 7. Management: A Systematic Review. J Med Internet Res 2016 May 16;18(5):e97 [FREE Full text] [doi: 10.2196/jmir.4883] [Medline: 27185295] Sohrabi C, Alsafi Z, O'Neill N, Khan M, Kerwan A, Al-Jabir A, et al. World Health Organization declares global emergency: A review of the 2019 novel coronavirus (COVID-19). Int J Surg 2020 Apr 26;76:71-76 [FREE Full text] [doi: 10.1016/j.ijsu.2020.02.034] [Medline: 32112977] 8. Wu Z, McGoogan JM. Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention. JAMA 2020 Feb 24;323(13):1239-1242. [doi: 10.1001/jama.2020.2648] [Medline: 32091533] Gong K, Xu Z, Cai Z, Chen Y, Wang Z. Internet Hospitals Help Prevent and Control the Epidemic of COVID-19 in China: Multicenter User Profiling Study. J Med Internet Res 2020 Apr 14;22(4):e18908 [FREE Full text] [doi: 10.2196/18908] [Medline: 32250962] 9. 10. Hong Z, Li N, Li D, Li J, Li B, Xiong W, et al. Telemedicine During the COVID-19 Pandemic: Experiences From Western China. J Med Internet Res 2020 May 08;22(5):e19577 [FREE Full text] [doi: 10.2196/19577] [Medline: 32349962] 11. Mahase E. WHO declares pandemic because of "alarming levels" of spread, severity, and inaction. BMJ 2020 Mar 12;368:m1036. [doi: 10.1136/bmj.m1036] [Medline: 32165426] 12. Bedford J, Enria D, Giesecke J, Heymann DL, Ihekweazu C, Kobinger G, et al. COVID-19: towards controlling of a pandemic. The Lancet 2020 Mar 17;395(10229):1015-1018. [doi: 10.1016/S0140-6736(20)30673-5] [Medline: 32197103] 13. Tanne JH, Hayasaki E, Zastrow M, Pulla P, Smith P, Rada AG. Covid-19: how doctors and healthcare systems are tackling coronavirus worldwide. BMJ 2020 Mar 18;368:m1090. [doi: 10.1136/bmj.m1090] [Medline: 32188598] 14. Xie X, Zhou W, Lin L, Fan S, Lin F, Wang L, et al. Internet Hospitals in China: Cross-Sectional Survey. J Med Internet Res 2017 Jul 04;19(7):e239 [FREE Full text] [doi: 10.2196/jmir.7854] [Medline: 28676472] 15. Tang M, He J, Chen M, Cong L, Xu Y, Yang Y, et al. "4+7" city drug volume-based purchasing and using pilot program in China and its impact. Drug Discov Ther 2019;13(6):365-369 [FREE Full text] [doi: 10.5582/ddt.2019.01093] [Medline: 31956236] 16. Notice on accomplishing online consultation services in the epidemic prevention and control (Medical Letter 2020 No. 112). National Health Commission of the People's Republic of China. 2020 Feb 7. URL: http://www.nhc.gov.cn/yzygj/ s7652m/202002/32c3e98988894fa18280e4543d2710c7.shtml [accessed 2020-04-20] Specification for online consultation service for infectious disease epidemic situation. Zhejiang Digital Economy Association. 2020 Mar 15. URL: http://www.ttbz.org.cn/StandardManage/Detail/33876/ [accessed 2020-04-20] 17. 18. Brooke J, Jackson D. Older people and COVID-19: Isolation, risk and ageism. J Clin Nurs 2020 Jul;29(13-14):2044-2046. 19. [doi: 10.1111/jocn.15274] [Medline: 32239784] Shahid Z, Kalayanamitra R, McClafferty B, Kepko D, Ramgobin D, Patel R, et al. COVID-19 and Older Adults: What We Know. J Am Geriatr Soc 2020 May;68(5):926-929 [FREE Full text] [doi: 10.1111/jgs.16472] [Medline: 32255507] Abbreviations CHC: community health center COVID-19: coronavirus disease IHDD: internet hospital plus drug delivery mHealth: mobile health NCDP: National Centralized Drug Purchasing SARS: severe acute respiratory syndrome THC: township health center http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ding et al Edited by G Eysenbach; submitted 29.04.20; peer-reviewed by H Liu, F Zhang, X Xie, B Foroutan; comments to author 12.06.20; revised version received 22.06.20; accepted 23.07.20; published 06.08.20 Please cite as: Ding L, She Q, Chen F, Chen Z, Jiang M, Huang H, Li Y, Liao C The Internet Hospital Plus Drug Delivery Platform for Health Management During the COVID-19 Pandemic: Observational Study J Med Internet Res 2020;22(8):e19678 URL: http://www.jmir.org/2020/8/e19678/ doi: 10.2196/19678 PMID: 32716892 ©Liang Ding, Qiuru She, Fengxian Chen, Zitong Chen, Meifang Jiang, Huasi Huang, Yujin Li, Chaofeng Liao. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 06.08.2020. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. http://www.jmir.org/2020/8/e19678/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 8 \| e19678 \| p. 9 (page number not for citation purposes)
10.2196_39259	JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al Original Paper The Influence of Paid Memberships on Physician Rating Websites With the Example of the German Portal Jameda: Descriptive Cross-sectional Study Friedrich Aaron David Armbruster; Dörthe Brüggmann, Dr med, Prof Dr; David Alexander Groneberg, Dr med, Prof Dr; Michael Bendels, Dr med, Prof Dr rer nat Institute of Occupational, Social and Environmental Medicine, Goethe University, Frankfurt, Germany Corresponding Author: Friedrich Aaron David Armbruster Institute of Occupational, Social and Environmental Medicine Goethe University Theodor-Stern-Kai 7 Frankfurt, 60590 Germany Phone: 49 6963016650 Email: [email protected] Abstract Background: The majority of Germans see a deficit in information availability for choosing a physician. An increasing number of people use physician rating websites and decide upon the information provided. In Germany, the most popular physician rating website is Jameda.de, which offers monthly paid membership plans. The platform operator states that paid memberships have no influence on the rating indicators or list placement. Objective: The goal of this study was to investigate whether a physician’s membership status might be related to his or her quantitative evaluation factors and to possibly quantify these effects. Methods: Physician profiles were retrieved through the search mask on Jameda.de website. Physicians from 8 disciplines in Germany’s 12 most populous cities were specified as search criteria. Data Analysis and visualization were done with Matlab. Significance testing was conducted using a single factor ANOVA test followed by a multiple comparison test (Tukey Test). For analysis, the profiles were grouped according to member status (nonpaying, Gold, and Platinum) and analyzed according to the target variables—physician rating score, individual patient’s ratings, number of evaluations, recommendation quota, number of colleague recommendations, and profile views. Results: A total of 21,837 nonpaying profiles, 2904 Gold, and 808 Platinum member profiles were acquired. Statistically significant differences were found between paying (Gold and Platinum) and nonpaying profiles in all parameters we examined. The distribution of patient reviews differed also by membership status. Paying profiles had more ratings, a better overall physician rating, a higher recommendation quota, and more colleague recommendations, and they were visited more frequently than nonpaying physicians’ profiles. Statistically significant differences were found in most evaluation parameters within the paid membership packages in the sample analyzed. Conclusions: Paid physician profiles could be interpreted to be optimized for decision-making criteria of potential patients. With our data, it is not possible to draw any conclusions of mechanisms that alter physicians’ ratings. Further research is needed to investigate the causes for the observed effects. (J Med Internet Res 2023;25:e39259) doi: 10.2196/39259 KEYWORDS physician rating websites; physician rating portals; paid influence; Germany https://www.jmir.org/2023/1/e39259 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al Introduction According to a representative survey commissioned by the Bertelsmann Foundation in Germany, more than 90% of respondents see a deficit in information availability on choosing a physician [1]. At the same time, more than half the population assumes strong differences in the quality of care among different physicians [2]. With a patient-generated rating system, around 25 physician rating websites in Germany are attempting to provide a platform for choosing a physician [3]. The influence of these rating portals on choosing a physician is significant; a survey by the Friedrich Alexander University of Erlangen Nuremberg indicated that 65.5% of physician rating portal users made their choice of physician on the basis of the information provided [4]. Thereby, the selection of a physician seems to be highly subjective. In a study by Carbonell and Brand [5], it was shown that comments and ratings from other users were more influential than facts, such as specialization or experience. Another study investigated the influence of ratings more specifically. In a simulation of a physician rating platform, the decision-making behavior of study participants was examined before and after they were presented with ratings from fictitious physicians. Participants were shown to be significantly more likely to choose profiles with many recommendations and ratings; negative ratings as well as low number of ratings had a deterrent effect on the participants [6]. Consequently, the reputation on physician rating websites also gains an economic aspect [7]. In the United States, cases have been reported in which physicians were engaging agencies to improve their own representation on a rating portal [8]. In Germany, Jameda.de is the most popular and well-known physician rating portal, with around 6 million monthly users and approximately 2 million patient ratings [9,10]. With 87.2%, Jameda lists the majority of physicians practicing in outpatient care [9]. Listed physicians obtain a profile on which patients can leave ratings in German school grades, ranging from 1 (best) to 6 (worst). Each patient’s rating includes at least some text and grades in 5 mandatory categories, from the mean values of which the overall grade of an individual rating is calculated. Ratings in a further 12 categories can be given voluntarily, and the physician can be recommended to other patients. The average value of these individual reviews is used to calculate the overall marks of the profiles. The algorithm for determining the number of colleague recommendation quotas or recommendations is not known; users are only presented with a simple numerical value. Ratings older than 4 years are archived and thus excluded from calculations. Reported ratings can either be suspended, deleted, or rereleased after review by Jameda. However, Jameda’s role as a neutral referral platform is not undisputed. Jameda, like other German review websites, offers monthly paid membership packages (Gold, Gold Pro, and Platinum) for physicians. Each of these packages provided a gradual expansion in functionality. Starting with Gold, it was possible to add a profile picture. Higher paying membership plans also included, for example, appointment allocation services https://www.jmir.org/2023/1/e39259 XSL•FO RenderX and the option for web-based consultations. Jameda contradicts claims [11] that a paid membership plan has a positive effect on ratings or list rankings [12]. So far, the influence of paid memberships has not been the focus of scientific investigation. The goal of this study was to investigate whether a physician’s membership status has an impact on his or her quantitative evaluation factors and to possibly quantify the key parameters—overall grade, grade distribution, number of evaluations, colleague recommendations, and profile views—were analyzed as a function of membership status (ie, nonpaying, Gold, and Platinum). these effects. Specifically, recommendation quota, Methods Data Acquisition Between January 31, 2020, and February 2, 2020, a total of 25,549 Jameda physician profiles were retrieved via the provided search mask on the jameda.de website. Regions and medical disciplines were selected to result as many paying members as possible for the smallest amount of search queries. Specifically, profiles with an overall score from the following 12 most populous cities in Germany were acquired: Berlin (n=5456, 21.4%), Hamburg (n=3526, 13.8%), Munich (n=3057, 12%), Cologne (n=2171, 8.5%), Frankfurt (n=1516, 5.9%), Stuttgart (n=1182, 4.6%), Düsseldorf (n=1481, 5.8%), Leipzig (n=1058, 4.1%), Dortmund (n=793, 3.1%), Essen (n=971, 3.8%), Bremen (n=986, 3.9%), and Dresden (n=1022, 4%). In addition, the query delivered 2330 (9.1%) profiles from the surrounding urban regions. Due to technical restrictions on the part of the provider, only a maximum of 90 profiles could be read out for each defined search term. To increase the sample size, the search term was specified by the respective city districts. In terms of content, these 8 disciplines were selected: “internal medicine and general medicine” (n=8032, 31.4%), dentistry (n=7744, 30.3%), gynecology (n=2519, 9.9%), orthopedics (n=2068, 8.1%), ophthalmology (n=1391, 5.4%), dermatology (n=1375, 5.4%), neurology (n=1063, 4.2%), and plastic or aesthetic surgery (n=385, 1.5%). The searches yielded 972 (3.8%) profiles that were primarily assigned to other specialties. These profiles were also retained. The result sorting was left at default (“relevance”). Profiles that had been acquired more than once could be identified by means of the unique Jameda-specific profile ID, and of these, only the most recent acquisition was included in the evaluation. Location and specialty assignment were extracted from the internet address. The membership status was extracted from the website source code. Since both Gold and Gold Pro were displayed as Gold, it was not possible for us to distinguish between them. Data Analysis For analysis, the profiles were grouped according to member status (nonpaying, Gold, and Platinum) and analyzed according to the target variables—physician rating score, individual patient’s ratings, number of evaluations, recommendation quota, J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al number of colleague recommendations, and profile views. Individual grades were taken from the grade report, and the total number of evaluations were calculated from the sum of the number of these individual ratings. Significance testing of group-specific means with SDs was performed using a single factor ANOVA test followed by a multiple comparison test (Tukey test). Data analysis and visualization were performed using Matlab (The MathWorks, Inc). Results A total of 21,837 (85.5%) nonpaying profiles, 2904 (11.4%) Gold, and 808 (3.2%) Platinum member profiles provided the data base for our analysis. The proportion of profiles with paid membership plans was thus 14.5% (n=3,712). Overall Physician Rating The mean overall physician grade was 1.68 (SD 0.92) for nonpaying profiles, 1.21 (SD 0.36) for Gold members, and 1.18 (SD 0.33) for Platinum members; the group-specific mean scores between nonpaying profiles and paid members differed highly significantly (ANOVA; P<.001). No statistical significance was found for the mean overall physician score in between Gold and Platinum members (Figure 1A). Figure 1. Analysis of physician-specific total and individual scores. (A) Mean physician rating score grouped by membership status. The higher the membership status, the significantly better the mean rating (mean and SD are represented; ANOVA; represents P<.001). (B) The group-specific distribution of the patient’s rating in semilogarithmic representation similarly documents a relative overrepresentation of grade 1 ratings among paying members. (C) The group-specific distribution of physician rating score in semilogarithmic representation documents an accentuated relative overrepresentation of grade 1 among paying members. The higher a member’s status (Platinum > Gold), the more pronounced this effect becomes. G: Gold; NP: nonpaying; P: Platinum. an accentuated documented The group-specific distribution of physician rating score relative (N=25,549) overrepresentation of grade 1 in paid members (Gold: 2489/2904, 85.7% and Platinum: 736/808, 91.1%) compared to nonmembers (12604/21837, 57.7%) and a relative underrepresentation of the remaining grades in comparison with nonmembers (Figure 1C). The higher a member’s status, the more pronounced this effect was across the entire grading scale: 3.3%( 725/21837) of nonpaying profiles and only 0.2% (5/2904) and 0.1%( 1/808) of Gold and Platinum members, respectively, https://www.jmir.org/2023/1/e39259 XSL•FO RenderX had an overall grade of 4 or worse. In our sample, no Platinum profile had a total grade of 4.5 or worse. Distribution of Patient’s Reviews There were 299,579 patient ratings distributed among 174,730 (58.3%) nonpaying members, 84,319 (28.1%) Gold members, and 40,530 (13.5%) Platinum members. The mean number of ratings per physician was 8.0 (SD 11.2) for nonpaying profiles, 29.0 (SD 36.7) for Gold members, and 50.2 (SD 54.6) for Platinum members; group-specific means differed highly significantly (ANOVA; P<.001; Figure 2A). J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al Figure 2. Analysis of nongrade criteria. The higher the membership status, the significantly higher the number of ratings. (A) Recommendation quota; (B) Colleague recommendations; (C) Number of profile views; (D) Mean and SD values for profile views; ANOVA; represents P<.001. A significant difference was found in between the paid memberships, but no significant different was found for the recommendation quota. G: Gold; NP: nonpaying; P: Platinum. The group-specific distribution of patient reviews documented a relative overrepresentation of grade 1 reviews among paying members (Gold: 80,482/84,319, 95.4% and Platinum: 39,319/40,530, 97%) compared to 80.5% (140,695/174,730) of the nonpaying members and a relative underrepresentation of the remaining graded reviews compared with nonmembers (Figure 1B). The underrepresentation accentuated with worse patient reviews and higher membership status. The proportion of individual scores of 4 or lower was 13.7% (23,975/174,730) in the nonpaying group and 2.2% (1885/84,319) and 1.4% (555/40,530) in the Gold and Platinum member groups, respectively. Recommendation Quota and Colleague Recommendations A total of 7228 colleague recommendations were found, which were subdivided into 3326 (46%), 2675 (37%), and 1227 (17%) recommendations for nonpaying members, Gold members, and Platinum members. The mean number of colleague recommendations was 0.15 (SD 0.59) for nonpaying profiles, 0.92 (SD 1.92) for Gold members, and 1.52 (SD 2.83) for Platinum members; the differences in means turned out to be highly significant (ANOVA; P<.001; Figure 2C). There were 21,475 (84.1%) physician profiles with a recommendation quota indicated; the range of values was between 0% and 100%. The mean recommendation quota was 72.5% (SD 25.2%) for nonpaying members and 90% (SD 13.5%) and 91.2% (SD 13.2%) for Gold and Platinum members, respectively (Figure 2B); the group-specific difference in mean values between nonmembers and paid members turned out to be highly significant (ANOVA; P<.001; Figure 2B). Visit Counts A total of 306,630,270 profile views subsumed into 214,955,367 (70.1%), 63,271,297 (20.6%), and 28,403,606 (9.3%) views for https://www.jmir.org/2023/1/e39259 XSL•FO RenderX nonpaying profiles, Gold members, and Platinum members. The range of values was between 45 and 518,691 calls per profile. The mean number of profile views was 0.98x104 (SD 1.3x104) for nonpaying profiles, 2.2x104 (SD 2.9x104) for Gold members, and 3.5x104 (SD 4.5x104) for Platinum members; the differences in mean values turned out to be highly significant between all groups (ANOVA; P<.001; Figure 2D). Discussion Overview In this analytical, descriptive cross-sectional study, the most important quantitative rating indicators of the Jameda platform were analyzed group-specifically according to nonpaying, Gold, and Platinum membership. For this purpose, a sample of 25,549 profiles were examined. This corresponds to a share of 16.2% among approximately 157,300 physicians practicing in outpatient care in Germany [13]. Rating Correlates With Membership Status Statistically significant differences were found between paying (Gold and Platinum) and nonpaying profiles in all parameters we examined. Except for the recommendation quotas and mean overall ratings, significant differences were also found within the paying profiles in the parameters, number of evaluations, and number of colleague recommendations (Figure 1A, Figure 2A, and Figure 2C). The higher a physician’s membership status was (nonpaying, Gold member, or Platinum member), the significantly better were the evaluation parameters. J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al in Implications The decision-making characteristics mentioned the introduction [6] seem to apply specially to paid profiles. Overall, paid profiles had significantly more ratings but fewer negative ratings than nonmembers. The recommendation quota was also significantly higher among paying members compared to nonpaying profiles. In addition, Jameda presents presumably valuable “colleague recommendations.” These were mainly dedicated to paid profiles (Figure 2C). The identity of these colleagues is not visible to the user. Why Do Paid Profiles Perform Better? This cannot be clarified beyond doubt with the data available to us, since we only performed a descriptive study. However, there was a conspicuous absence of critical individual reviews for paying profiles (Figure 1C). Assuming that physicians with a paid package are rated similarly to nonpaying physicians, an active mechanism seems possible. Physician rating websites have the ability to suspend or even delete ratings, thus eliminating them [14]. There are various reasons to initiate a deletion of a rating, but one way toward deletion is a report by the physician. The rating portal then decides on the outcome of the deletion process. From a previous study on another portal, we know that the reviews affected by deletion are mainly negative reviews [15]. This thesis is in line with recurring allegations, which accuse Jameda of deleting critical posts or requiring the submission of written proof of treatment [16]. The exact decision criteria for deletion or suspension are not comprehensible for the user of the portal. This could occur more frequently with paid profiles, as those physicians may consider their representation on Jameda to be important. For example, it has been observed that a professional social media account correlates with a high number of ratings [17]. Physicians generally attaching less importance to review portals might be less inclined to invest monthly in a paid membership or to take action against negative reviews. Whether nonpaying members are less likely than paying members to report critical reviews is not possible for us to test due to lack of data. Why Do Paid Profiles Have More Reviews? In general, the number of ratings on physician rating portals increased over time [18], but profiles with paid memberships have particularly high numbers of ratings. Here a self-reinforcing effect seems possible. Since patients look for physicians with many and good ratings [6] on physician rating websites, they might choose paying profiles more often and rate them afterwards. Another possible explanation is that part of the reviews could also be purchased. On the internet, several websites offer to create reviews for the Jameda portal [19,20]. An assessment of how many reviews have been created by marketing agencies and whether this is specific to paid profiles is not possible due to a lack of data. https://www.jmir.org/2023/1/e39259 XSL•FO RenderX The impact of a paid membership seems to be noticeable to the physician: Steinfort [21] concludes in an article in the German journal Gynecology and Obstetrics about physician rating portals, “a premium status on physician rating portals guarantees a high inrush of new, but also very flexible and transient patients” [21]. Business Model Thus, the business principles of some commercial physician rating websites seem to rely on questionable presumptions by the members. To be represented as favorably as possible, some physicians may think that merely paying a rating portal leads to a better standing. In this respect, on other physician rating websites, paying physicians can also partially hide negative ratings for the patient [22]. This raises the question of the extent to which the data presented to the user on physician rating websites can be trusted. Methodological Limitations A fundamental limitation of the study is that it is purely descriptive. No conclusions can be drawn about any (possibly active) mechanisms for changing rating indicators post acquisition of a premium package. Therefore, the authors do not claim to have identified fraud mechanisms of a platform in this study; they supplied descriptive data to gain more information. A further limitation of the study design is that the search queries were restricted to the 12 largest cities in Germany. Compared to the results from a study by Emmert and McLennan [18], significantly more reviews were found per profile than what Emmert and McLennan found in 2019, which might indicate a sampling bias [18]. Furthermore, we can only separate 3 of the 4 membership plans. It is not possible for us to differentiate between Gold and Gold Pro. An assessment of the rejected or deleted ratings is not possible due to a lack of data. Finally, we only investigated one physician rating website, so it is unclear whether this is true for other portals or even for non-German physician rating websites. Conclusions Overall, we can conclude that profiles of paid members seemed to be optimized for decision-making characteristics of potential patients in all evaluation parameters analyzed by us. With one exception, these effects increased with increasing pay status. High call rates of Gold and Platinum profiles confirm the increased patient interest. Therefore, the results seemingly contradict Jameda’s claim of being a neutral rating platform (Jameda’s quality promise mentions “We treat all physicians the same,” and “Ratings are not for sale”) [12]. Rather, Jameda fulfills the criteria of an advertising platform for paying physicians. In this context, the nonpaying profiles seem to serve as a contrast to the paying members and are thus necessary for the business model of this platform. Due to the anonymity of the ratings and nontransparency of some other parameters, a well-founded physician counterposition is prevented. More analyses of different physician review websites are needed as a next step toward systematization. J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al Conflicts of Interest None declared. References 1. 2. 3. 4. 5. 6. 7. Qualität der Ärzte: Patienten haben zu wenig Informationen. Aerzteblatt.de. 2018. URL: https://www.aerzteblatt.de/ nachrichten/94908/Qualitaet-der-Aerzte-Patienten-haben-zu-wenig-Informationen [accessed 2023-02-21] Etgeton S, Weigand M. Public Reporting über Arztpraxen Open-Data-Politik schafft mehr Transparenz für Patienten. Spotlight Gesundheit 07 May 18;3:2-8 [FREE Full text] [doi: 10.24945/mvf.06.20.1866-0533.2282] Rothenfluh F, Schulz PJ. Content, quality, and assessment tools of physician-rating websites in 12 countries: quantitative analysis. J Med Internet Res 2018 Jun 14;20(6):e212 [FREE Full text] [doi: 10.2196/jmir.9105] [Medline: 29903704] Emmert M, Meier F, Pisch F, Sander U. Physician choice making and characteristics associated with using physician-rating websites: cross-sectional study. J Med Internet Res 2013;15(8):e187 [FREE Full text] [doi: 10.2196/jmir.2702] [Medline: 23985220] Carbonell G, Brand M. Choosing a physician on social media: comments and ratings of users are more important than the qualification of a physician. Int J Hum-Comput Int 2017 Aug 03;34(2):117-128. [doi: 10.1080/10447318.2017.1330803] Carbonell G, Meshi D, Brand M. The use of recommendations on physician rating websites: the number of raters makes the difference when adjusting decisions. Health Commun 2019 Nov 17;34(13):1653-1662. [doi: 10.1080/10410236.2018.1517636] [Medline: 30222006] Bensnes S, Huitfeldt I. Rumor has it: How do patients respond to patient-generated physician ratings? J Health Econ 2021 Mar;76:102415 [FREE Full text] [doi: 10.1016/j.jhealeco.2020.102415] [Medline: 33422733] 8. Wang JV, Heitmiller K, Boen M, Saedi N. Fake online physician reviews in aesthetic dermatology: bioethical and professional obligations. Dermatol Surg 2020 Jun 23;47(5):748-749. [doi: 10.1097/dss.0000000000002516] 11. 10. 9. McLennan S, Strech D, Reimann S. Developments in the frequency of ratings and evaluation tendencies: a review of german physician rating websites. J Med Internet Res 2017 Aug 25;19(8):e299 [FREE Full text] [doi: 10.2196/jmir.6599] [Medline: 28842391] Jameda Factsheet. URL: https://web.archive.org/web/20200921100743/https://www.jameda.de/jameda/jameda/ jameda-Factsheet.pdf Jung M. Dem Treiben Grenzen gesetzt. Frankfurter Allgemeine Zeitung. 2018. URL: https://www.faz.net/aktuell/wirtschaft/ jameda-kommentar-dem-treiben-grenzen-gesetzt-15458879.html [accessed 2023-02-22] Jameda. 2021. URL: https://www.jameda.de/qualitaetssicherung/ [accessed 2023-02-21] 2018. Bundesärztekammer. URL: https://www.bundesaerztekammer.de/baek/ueber-uns/aerztestatistik/aerztestatistik-2018 [accessed 2023-12-01] 2022;. Jameda. URL: https://www.jameda.de/erfahrungen [accessed 2023-02-22] 14. 15. McLennan S. Rejected online feedback from a swiss physician rating website between 2008 and 2017: analysis of 2352 ratings. J Med Internet Res 2020 Aug 03;22(8):e18374 [FREE Full text] [doi: 10.2196/18374] [Medline: 32687479] 12. 13. 16. Trustpilot. URL: https://de.trustpilot.com/review/jameda.de [accessed 2023-12-01] 17. McCormick JR, Patel MS, Hodakowski AJ, Rea PM, Naik KP, Cohn MR, et al. Social media use by shoulder and elbow surgeons increases the number of ratings on physician review websites. J Shoulder Elbow Surg 2021 Dec;30(12):e713-e723. [doi: 10.1016/j.jse.2021.06.018] [Medline: 34343661] 18. Emmert M, McLennan S. One decade of online patient feedback: longitudinal analysis of data from a german physician 19. rating website. J Med Internet Res 2021 Jul 26;23(7):e24229 [FREE Full text] [doi: 10.2196/24229] [Medline: 34309579] Jameda Bewertungen kaufen. Fivestar Marketing. URL: https://fivestar-marketing.net/bewertungen/jameda- bewertungen-kaufen/ [accessed 2023-02-21] 20. Bewertungsfee. URL: https://www.bewertungsfee.com/rezensionen/5-jameda-bewertungen-kaufen/ [accessed 2023-02-20] 21. Enderer-Steinfort G. Schreckgespenst Arztportale. gynäkologie + geburtshilfe 2021 May 31;26(3):3. [doi: 10.1007/s15013-021-4086-1] 22. Mulgund P, Sharman R, Anand P, Shekhar S, Karadi P. Data quality issues with physician-rating websites: systematic review. J Med Internet Res 2020 Sep 28;22(9):e15916 [FREE Full text] [doi: 10.2196/15916] [Medline: 32986000] https://www.jmir.org/2023/1/e39259 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Armbruster et al Edited by R Kukafka; submitted 05.05.22; peer-reviewed by S McLennan, P Morakis; comments to author 03.06.22; revised version received 27.08.22; accepted 11.10.22; published 04.04.23 Please cite as: Armbruster FAD, Brüggmann D, Groneberg DA, Bendels M The Influence of Paid Memberships on Physician Rating Websites With the Example of the German Portal Jameda: Descriptive Cross-sectional Study J Med Internet Res 2023;25:e39259 URL: https://www.jmir.org/2023/1/e39259 doi: 10.2196/39259 PMID: 37014690 ©Friedrich Aaron David Armbruster, Dörthe Brüggmann, David Alexander Groneberg, Michael Bendels. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.04.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e39259 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e39259 \| p. 7 (page number not for citation purposes)
10.2196_45777	JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Original Paper Public Attitudes Toward Anxiety Disorder on Sina Weibo: Content Analysis Jianghong Zhu, MSc; Zepeng Li, PhD; Xiu Zhang, BEng; Zhenwen Zhang, MEng; Bin Hu, PhD Gansu Provincial Key Laboratory of Wearable Computing, School of Information Science and Engineering, Lanzhou University, Lanzhou, China Corresponding Author: Bin Hu, PhD Gansu Provincial Key Laboratory of Wearable Computing School of Information Science and Engineering Lanzhou University No 222 South Tianshui Road Lanzhou, 730000 China Phone: 86 17352120733 Email: [email protected] Abstract Background: Anxiety disorder has become a major clinical and public health problem, causing a significant economic burden worldwide. Public attitudes toward anxiety can impact the psychological state, help-seeking behavior, and social activities of people with anxiety disorder. Objective: The purpose of this study was to explore public attitudes toward anxiety disorders and the changing trends of these attitudes by analyzing the posts related to anxiety disorders on Sina Weibo, a Chinese social media platform that has about 582 million users, as well as the psycholinguistic and topical features in the text content of the posts. Methods: From April 2018 to March 2022, 325,807 Sina Weibo posts with the keyword “anxiety disorder” were collected and analyzed. First, we analyzed the changing trends in the number and total length of posts every month. Second, a Chinese Linguistic Psychological Text Analysis System (TextMind) was used to analyze the changing trends in the language features of the posts, in which 20 linguistic features were selected and presented. Third, a topic model (biterm topic model) was used for semantic content analysis to identify specific themes in Weibo users’ attitudes toward anxiety. Results: The changing trends in the number and the total length of posts indicated that anxiety-related posts significantly increased from April 2018 to March 2022 (R2=0.6512; P<.001 to R2=0.8133; P<.001, respectively) and were greatly impacted by the beginning of a new semester (spring/fall). The analysis of linguistic features showed that the frequency of the cognitive process (R2=0.1782; P=.003), perceptual process (R2=0.1435; P=.008), biological process (R2=0.3225; P<.001), and assent words (R2=0.4412; P<.001) increased significantly over time, while the frequency of the social process words (R2=0.2889; P<.001) decreased significantly, and public anxiety was greatly impacted by the COVID-19 pandemic. Feature correlation analysis showed that the frequencies of words related to work and family are almost negatively correlated with those of other psychological words. Semantic content analysis identified 5 common topical areas: discrimination and stigma, symptoms and physical health, treatment and support, work and social, and family and life. Our results showed that the occurrence probability of the topical area “discrimination and stigma” reached the highest value and averagely accounted for 26.66% in the 4-year period. The occurrence probability of the topical area “family and life” (R2=0.1888; P=.09) decreased over time, while that of the other 4 topical areas increased. Conclusions: The findings of our study indicate that public discrimination and stigma against anxiety disorder remain high, particularly in the aspects of self-denial and negative emotions. People with anxiety disorders should receive more social support to reduce the impact of discrimination and stigma. (J Med Internet Res 2023;25:e45777) doi: 10.2196/45777 KEYWORDS anxiety disorder; linguistic feature; topic model; public attitude; social media https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Introduction Anxiety disorder is a common mental disorder. Anxiety disorders involve repeated episodes of intense anxiety and sudden feelings of fear or terror, which peak in a few minutes (panic attacks). These feelings of anxiety and panic can impact the patient’s normal schooling, work, and life. Anxiety disorders are the most prevalent among adolescents, and according to the statistics of the World Health Organization, 3.6% of the adolescents aged 10-14 years and 4.6% of the adolescents aged 15-19 years worldwide experience anxiety disorders [1]. According to a report on the status of nutrition and chronic diseases in China, the prevalence of anxiety disorders in China reached 4.98% in 2020 [2]. In particular, with the spread of COVID-19, there is a substantial increase in the global cases of anxiety disorders. According to Santomauro et al [3], in a study of 204 countries and regions worldwide, the prevalence of global major depression increased by 26% in 2020. Compared to people without anxiety disorders, those with anxiety disorders may have unstable interpersonal relationships, poorer functions, and higher rates of work absenteeism, with significant economic costs and impacts on physical health [4-6]. Moreover, anxiety disorders are associated with a significantly increased mortality risk. Compared with the general population, people with anxiety disorders have 1.4 times increased risk of death from natural causes and 2.5 times increased risk of death from nonnatural causes [7]. The high prevalence, chronicity, and the associated excessive mortality led the World Health Organization to rank anxiety disorders as the ninth leading health-related cause of disability [8], causing a significant economic burden, accounting for 3% of the global burden of disease worldwide and costing about €74.4 billion in 30 European countries [9]. Several factors prevent people from seeking help for mental illness, including poor quality of services, low levels of health literacy in mental health, and stigma and discrimination. Formal mental health services are not available in many places. Even if they do, they are often unavailable or unaffordable [10]. People often choose to endure mental suffering without relief, rather than risk discrimination and exclusion to access mental health services. The proportion of people with anxiety disorders who received any form of treatment is estimated to be 27.6% worldwide, but only 9.8% received adequate treatment [11]. Shame and stigma are the major obstacles for patients with anxiety disorders to receive an early diagnosis and professional treatment [12-14]. Patients with depression or anxiety disorders (compared to those without psychiatric disorders) are twice as likely to have a stigma, and the association between depression and anxiety is even stronger [15]. In addition, negative public attitudes may adversely impact patients’ social functioning (eg, interpersonal relationships, learning, work abilities) [16]. Public attitudes toward anxiety disorder can impact the psychology and daily lives of patients with anxiety. Most current research on attitudes toward mental illness use the method of questionnaire surveys [14,15,17]. This is because a questionnaire is an effective way of collecting information in research; other potentially richer measures of attitude, such as qualitative https://www.jmir.org/2023/1/e45777 XSL•FO RenderX analysis, are often not feasible in large-scale studies due to labor costs [18]. However, the information collected by existing research methods is highly influenced by the subjective will of the investigators. In addition, the existing research methods are still limited in their ability to explore the structure of language expression, which would reflect people’s internal activities. To obtain information on the linguistic expressions of large groups of people, we turn our attention to the information available on social media. As an important medium of mass communication, social media contain a large amount of information that reflects people’s inner activities and emotional states. Users may often present their mental problems or illnesses anonymously through a wide variety of social media or online social health communities [19]. Such an online health community can be a network to express compassion by communicating with people who have similar symptoms [20]. In addition, users often try to obtain health information related to their symptoms on social media in an attempt to diagnose themselves [21,22]. Therefore, the posts on social media provide natural language data on people’s attitudes toward anxiety disorders. A large number of studies in recent years have shown that social media data can be used to better understand, identify, and describe mental disorders [23,24] (eg, data from Facebook, Twitter, Instagram, Sina Weibo platforms). Individuals with mental disorders show changes in language and behavior, such as greater negative emotions and heightened self-attentive focus [25-28]. There is a high degree of similarity between patients with different forms of mental distress. Moreover, social media data can be used to assess the anxiety level of users [29,30], identify anxiety disorders [19,31], and assess the anxiety level of the public [32,33], thus enabling the assessment of the anxiety level of a large population of individuals or groups. These assessment methods might make up for the shortcomings of the traditional approach using questionnaires such as the Self-rating Anxiety Scale because it is difficult to ask a large number of individuals to fill in those questionnaires to estimate their anxiety levels separately. In addition, some studies showed that citizens’ perceptions and attitudes toward events can be tracked and discovered by analyzing the content posted on social media platforms [34,35]. Health-related stigmatizing attitudes (eg, depression, suicide) can be identified by analyzing the linguistic features expressed in social media posts [36-39]. Differences in stigmatizing attitudes toward mental health issues can also be reflected in different patterns of language use [40]. Social media–based research on attitudes toward mental illness is conducted at 2 main levels. The first level of analysis is at the text feature level, using word frequency statistics and affective tendency analysis to infer attitudes. Li et al [37] accurately discovered depression discrimination on social media through linguistic analysis methods. Li et al [40] constructed a model to distinguish schizophrenia-related stigma from depression-related stigma by using some psycholinguistic features automatically extracted from each post. Shen and Rudzicz [19] separated posts on 4 different anxiety-related subreddits from posts on control subreddits with an accuracy of 98%. They used a combination of N-gram language modeling and Linguistic Inquiry and Word Count (LIWC) [41]. J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al The second level of analysis is topic-based analysis. Compared to text feature analysis, topic-based analysis is a higher and a more meaningful level of research. Topic models are divided into qualitative topic modeling methods and quantitative topic modeling methods. The qualitative topic analysis includes mainly content analysis methods (eg, NVivo software, QSR International), which is a systematic method to summarize the expressed content by inference from the text [42]. Reavley and Pilkington [43] used content analysis methods to find that tweets related to depression and schizophrenia were mostly supportive or neutral, and more than one-third of the tweets reflecting stigmatized attitudes toward depression were mocking or belittling patients. Lachmar et al [44] used the qualitative content analysis method to code tweets containing the tag #My depression looks like# on Twitter in May 2016. The content analysis revealed 7 themes: dysfunctional thoughts, lifestyle challenges, social struggles, hiding behind a mask, apathy and sadness, suicidal thoughts and behaviors, and seeking relief. Although the qualitative analysis method is effective, it usually requires researchers to have clinical expertise, which limits the wide use of the method. The main methods of quantitative topic analysis include Latent Dirichlet Allocation (LDA) and its improvement methods. LDA is a statistical model that uses a data-driven unsupervised machine learning process to discover the underlying semantic structure from a series of documents. The latent semantic structure consists of a set of related topics that are identified by the words of co-occurrence. The results of the analysis are presented as word sets of co-occurrence, whose common themes are inferred by the researcher [45,46]. Franz et al [47] used the LDA model to detect content in social media containing self-harming thoughts and self-harming behaviors. Liu and Shi [28] used the LDA model to find that there were 7 main topics discussed by depressed patients on the Sina Weibo platform: negative emotion fluctuation, disease treatment and somatic responses, sleep disorders, sense of worthlessness, suicidal extreme behavior, seeking emotional support, and interpersonal communication. Jo et al [48] used a Structural Topic Model similar to the LDA model [49] to analyze users’ anxiety and worry concerns by analyzing data from 13,148 questions and 29,040 answers related to COVID-19 on Naver, a social networking platform in South Korea, and by using a structured topic model and a method of analyzing the network of words. They found that the long-term situation resulted in a slight increase in the proportion of work-related topics and that people’s anxiety and worry are closely related to physical symptoms and methods of self-protection. Paul and Dredze [50] obtained 13 health-related topics in Twitter data by constructing the Ailment Topic Aspect Model. Sik et al [51] used both quantitative (LDA topic modeling) and qualitative (deep reading) methods to determine the optimal number of topics and their interpretation in depression forums. Previous studies on mental disorder attitudes on social media have described attitudes statically at the level of textual features and topic features. There are fewer studies on the dynamics of attitudes over time. Yu et al [38] studied public attitudes toward depression and the changes in these attitudes over time by combining the textual feature level and the topic level. The analysis of linguistic features showed that the frequency of use of emotion, positive emotion, anger, cognition (including the insight subcategory), and conjunctions increases significantly over time. The topic feature results suggested an upward trend in social support for people with depression over time, although there is a persistent stigma attached to depression. Using data from the Reddit platform, Low et al [52] analyzed trends from 90 text-derived features and found that many features including categories such as economic stress, isolation, and home increased significantly during COVID-19, while other categories such as motion decreased significantly. They also combined the LDA model and clustering algorithms and found a significant increase in health anxiety topics in midpandemic posts compared to those in prepandemic posts, with a large number of posts related to health anxiety topics expressing concerns with daily life at home, school, and work. Table 1 summarizes the research objectives and analytical methods of the existing work. For more research on the application of text analysis and natural language processing methods in health care, see [53,54]. In this study, we explored public attitudes toward anxiety disorders on social media and the changes in the attitudes over time at 2 levels: text features and topic features. Sina Weibo is the most popular social media platform in China, known as the Chinese version of Twitter. With 582 million monthly active users in 2022, Sina Weibo is a valuable source of information for studying attitudes toward anxiety. The purpose of this study was to use this rich data source to study public attitudes toward anxiety and how these attitudes change over time by analyzing the textual features and topic structure of publicly available posts on Chinese social media. https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Table 1. A summary of the research objectives and analytical methods described in studies to date. Study Li et al [37] Li et al [40] Objective(s) Analytical methods Discover depression discrimination on social media Text feature analysis Distinguish schizophrenia-related stigma from depression-related stigma Text feature analysis Reavley and Pilkington [43] Analyze the emotional polarity in tweets related to depression and schizophrenia Qualitative content analysis Lachmar et al [44] Reveal the themes on tweets containing the tag #My depression looks like# in May 2016 Qualitative content analysis Franz et al [47] Liu and Shi [28] Jo et al [48] Detect content in social media containing self-harming thoughts and self-harming behaviors Latent Dirichlet Allocation model (quantitative topic analysis) Reveal the topics talked about by depressed patients on the Sina Weibo platform Latent Dirichlet Allocation model Analyze users’ anxiety and worry concerns by analyzing data re- lated to COVID-19 on Naver platform Structural Topic Model and the network of words Paul and Dredze [50] Reveal the health-related topics in Twitter Ailment Topic Aspect Model Determine the optimal number of topics and their interpretation in depression forums Latent Dirichlet Allocation model and qualitative analysis Analyze health anxiety topics in midpandemic posts compared to those in prepandemic posts Text feature analysis, Latent Dirichlet Allocation model and clustering algorithm Analyze public attitudes toward depression and the changes in these attitudes over time Text feature analysis and Latent Dirichlet Allocation model Sik et al [51] Low et al [52] Yu et al [38] Methods Data Collection Sina Weibo is one of the most influential Chinese social media platforms. We use the application programming interface of Sina Weibo to obtain posts containing the keyword “焦虑症 (anxiety disorder),” which were posted from April 1, 2018, to March 31, 2022. We collected 403,132 Weibo posts. By a statistical method, we found that these posts were from 295,905 accounts, which means that about 5 in every 10,000 Sina Weibo users were engaged in conversations about anxiety disorders. To better reveal the textual features and topic structure of Weibo posts, before performing the analysis, we preprocessed the original posts according to the following steps: 1. We deleted the information posted automatically by organizations or Sina Weibo platforms. 2. We deleted users’ names mentioned by the symbol “@” in the main body of the posts. 3. We deleted English letters and emoticons. 4. The keywords “topic” and “super topic” of the posts contain the symbol “#” before and after the body, for example, “#super topic of anxiety#;” so, we used regular expressions to remove all the titles of topics and super topics that appeared in the posts. 5. We deleted advertising posts whose content recurred many times. 6. Finally, we deleted posts with very limited information, such as those containing only keywords. Through the above preprocessing steps, we finally obtained 325,807 posts with the keyword “anxiety disorder.” To analyze the trend of the number of posts over time, we divided all the posts by month. https://www.jmir.org/2023/1/e45777 XSL•FO RenderX Ethical Considerations All the data in this paper were obtained from Sina Weibo’s public data, which protects those who have private profiles from being subject to research studies. Hence, this analysis meets the standards to waive informed consent and similar guidelines [55]. Furthermore, we desensitized the data to protect the privacy of the users. Specifically, we removed all individual information related to the identity of the users. Methods of Analysis First, the preprocessed posts were divided by month to obtain time series data, and then we analyzed the changes in the number of posts related to anxiety disorders by using a linear regression method (R language; R Core Team and the R Foundation for Statistical Computing). Time series analysis can provide a reasonable mathematical model for the sample data to analyze the information and patterns contained in the sample time series. A time series usually has the following characteristics: 1. The positions of the data in the data series are determined by the temporal order, and there is a temporal correlation between them. 2. The values of different points in the data series are somewhat random; therefore, it is difficult to predict completely and accurately from historical observations. 3. The values with a relatively close time interval have a strong correlation, and this correlation can reflect the system evolution pattern. 4. Time series data usually tend to have some sort of trend and periodicity, such as seasonality. We grouped the text data of the posts by month, from April 1, 2018, to March 31, 2022, and we fit a linear regression by using the number of posts and the total length of posts per month as J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al a function of time. In addition, by analyzing these posts, we found that the number of posts containing the keyword “anxiety disorder” increased dramatically at the beginning of each new term. Approximately 10.67% (34,771/325,807) of the posts contained the keyword “new term begins.” To more clearly analyze the changing trend of the information contained in the posts with the keyword “anxiety disorder,” we selected a subset of the posts that contained no keyword “开学 (new term begins),” which included 291,036 posts (Table S1 in Multimedia Appendix 1). Second, we performed textual analysis on the linguistic features of all the posts. The Chinese psychoanalysis tool TextMind [56] was used to conduct statistical analysis on high-frequency words related to the psychological characteristics in posts. TextMind can automatically segment words, classify words, and calculate the frequency of each category of word, using the Chinese lexicon named C-LIWC [57]. C-LIWC is a lexicon containing 32 linguistic features, 32 psychological features (belonging to 6 major categories), and 38 other features such as punctuation marks. In this study, 5 major categories were selected to analyze the trends and relevance characteristics of each category, and further, 11 subcategories of these 5 categories and 4 additional categories were selected to analyze the fine-grained changes (See Table 2 for details). Table 2. Categories of the mental characteristics selected in the Linguistic Inquiry and Word Count dictionary. Category Social process Family Affective process Positive emotion Negative emotion Anxiety Anger Sadness Biological process Body Health Abbreviation Examples Social Family Affect PosEmo NegEmo Anx Anger Sad Bio Body Health invite, hear, instruct, community, interact, public, culture son, daughter, husband, parents, uncle, cousin, family serious, excessive, willing, rich, hope, promise, cope affectionate, loving, welcome, praise, glorious, interesting, kind resentful, heartless, failure, worry, trash, protest, abuse restless, impatient, insomnia, fright, panic, anxiety, nervous resentful, angry, enemy, fight, criticize, rage, agitated pitiful, disappointed, inferior, sorrowful, suffering, helpless, sad sick, fever, healing, tired, pain, numbness, vessels finger, skin, ear, perception, breath, eye, shirt live, care, insomnia, wound, surgery, health, scar Cognitive process CogMech according, evidence, generally, intend, notice, otherwise, however Insight Exclusive Perceptual process Feel (Others) I (Others) Assent (Others) Work (Others) Achievement Insight Excl Percept Feel I Assent Work Achieve exactly, seem, think, admit, notice, believe, feeling regardless, rather, if, otherwise, unless, suppose, however say, show, watch, listen, feel, touch feel, soft, comfort, fuzzy, sharp, smooth, touch I hah-hah, alright, okay, yes, indeed, sure, clear, good work, research, postgraduate, study, colleague, interview, unit success, accomplish, achieve, encourage, reward, plan, effect The Chinese psychological analysis system TextMind is a software system developed by the Computational Network Psychology Laboratory of the Institute of Psychology, Chinese Academy of Sciences, for linguistic analysis of Chinese text. TextMind provides users with a full set of analysis solutions from automatic word segmentation in simplified Chinese to linguistic psychological analysis; its lexicon, text, and symbol processing methods are specifically tailored to the simplified Chinese context, and its lexicon classification system is also compatible with LIWC. Finally, we used a topic model to analyze the topics of all text data on Sina Weibo for each month. The topic model can generate keywords in each topic and the occurrence probability of each topic appearing in the entire document according to the predetermined number of topics. The LDA and the biterm topic model (BTM) are the classic topic models. The BTM [58] uses biterm (word pairs) to model the text, which can better display the hidden topics in the article; so, the BTM is more suitable for short-text modeling and analysis. Therefore, we chose the BTM to analyze the topics discussed in anxiety-related posts. In the preliminary analysis, the number of topics in BTM was set to 30. Results Trend Analysis of the Number and Total Length of Sina Weibo Posts Figure 1 shows the changes in the number and total length of anxiety-related posts from April 2018 to March 2022, as well as the changes in the number of posts excluding the keyword https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al “new term begins.” The number of all anxiety-related posts fluctuated over time and showed a clear upward trend, especially after the outbreak of COVID-19. The upward trend was more clearly illustrated by the changes in the number of posts excluding the keyword “new term begins.” The number of posts increased significantly each year before the start of new terms. However, due to the epidemic, the start of the spring semester in 2020 was postponed to March or April; so, there was no obvious peak. The trend in the total length of anxiety-related posts was almost identical to the trend in the number of posts. Moreover, the increase in the total length of posts was even more remarkable in the early stages of the pandemic. Table 3 shows the results of linear regression. The estimate index in Table 3 shows that regardless of whether there were all anxiety-related posts or anxiety-related posts without the Figure 1. The trend in the number and total length of Sina Weibo posts. keyword “new term begins,” the changes over time showed a significant increasing trend. The R2 value of the anxiety-related posts without the keyword “new term begins” was 0.8671, indicating that the linear regression model had a better fit. In addition, we used the time series decomposition function [59] to obtain the long-term trend, seasonal trend, and random term of the number of anxiety-related posts (see Figure 2). Similar to the results in Table 3, the long-term trend presented an obvious upward trend. In the seasonal trend, there were 2 peaks every year in February and August, but the peak in August was higher. Besides, in China, February and August happen to be the beginning of a new semester (spring/fall) every year, and August is also the beginning of the new school year. Thus, we can conclude that the beginning of new semester has a significant impact on anxiety in Sina Weibo posts. Table 3. Linear regression results of the number of Sina Weibo posts. Results Number of posts Number of posts without the keyword “new term begins” Time Total length of posts Time Predictor Estimate SE Time 176.83 141.34 1.2762 19.08 8.158 0.090 t test (df) R 2 9.267 (46) 0.6512 17.32 (46) 0.8671 14.16 (46) 0.8133 P value <.001 <.001 <.001 https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Figure 2. Time series decomposition of the number of Sina Weibo posts. Analysis of Linguistic Features Analysis of the Features of Major Categories of Words As shown in Table 2, 5 major categories of psychological features were chosen: social process (Social), affective process (Affect), biological process (Bio), cognitive process (CogMech), and perceptual process (Percept). We analyzed the occurrence frequency of these 5 categories of words in anxiety-related posts from April 2018 to March 2022 (Table S2 in Multimedia Appendix 1); the results are shown in Figure 3. The words related to CogMech appeared most frequently, followed by those related to Affect, Bio, and Social. The frequency of words related to Percept was the lowest, and the change was not obvious during the 4-year period. The frequency of the words related to CogMech, Affect, and Bio increased during the early stage of COVID-19, from January 2020 to July 2020, while the frequency of the words related to Social decreased. Furthermore, Table 4 shows the linear regression results of the frequency of these 5 major categories of words. The words related to Social showed a downward trend, while the other 4 categories of words showed an upward trend, and the use of words related to CogMech and Bio had the most obvious increasing trend. Moreover, the frequency of these 5 categories of words fluctuated greatly over time; therefore, the R2 values were all small under the linear regression fitting model. Figure 3. Occurrence frequency of the 5 major linguistic feature words: Social, Affect, Bio, CogMech, and Percept. Affect: affective process; Bio: biological process; CogMech: cognitive process; Percept: perceptual process; Social: social process. https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Table 4. Linear regression results of the occurrence frequency of the 5 major linguistic feature words. Predictor Estimate SE t test (df) R 2 –8.907e-05 2.060e-05 –4.323 (46) 0.2889 1.224e-05 2.622e-05 0.467 (46) 0.0047 1.452e-04 3.102e-05 4.680 (46) 0.3225 <.001 1.967e-04 6.229e-05 3.158 (46) 0.1782 2.942e-05 1.060e-05 2.776 (46) 0.1435 .003 .008 P value .001 .64 Time Time Time Time Time Sociala Affectb Bioc CogMechd Percepte aSocial: social process. bAffect: affective process. cBio: biological process. dCogMech: cognitive process. ePercept: perceptual process. Analysis of the Features of Subcategories of Words We selected 11 subcategories out of the 5 major categories (see Table 2) and the other 4 subcategories (I, assent, work, and achieve) to analyze the changing patterns of the various categories of words (Table S2 in Multimedia Appendix 1) and their correlation at a fine-grained level. Table 5 shows the linear regression results of the 15 subcategories of the features. Table 5 demonstrates that the use of words related to negative emotion (NegEmo), anxiety, anger, sadness, body, health, exclusive, feel, I, and assent was on the rise, while the use of words related to family, positive emotion (PosEmo), insight, work, and achievement was on the decline. Figure 4 shows the correlation results among the 15 subcategories of features. As shown in the figure, the words related to work and family were almost negatively correlated with other features. Anxiety had the highest correlation with NegEmo, followed by assent with exclusive, NegEmo with health, and health with feel. The features with the highest negative correlation were work and health. In addition to NegEmo, there were insight, health, and feel—all of which were highly correlated with anxiety. Table 5. Linear regression results of the 15 subcategories of the 5 features. Predictor Estimate SE t test (df) Family Positive emotion Negative emotion Anxiety Anger Sadness Body Health Insight Exclusive Feel I Assent Work Achievement Time Time Time Time Time Time Time Time Time Time Time Time Time Time Time –2.252e-05 3.775e-06 –5.964 (46) –1.623e-05 1.088e-05 –1.491 (46) 3.776e-05 1.516e-05 2.491 (46) 1.612e-05 8.281e-06 1.947 (46) 3.611e-06 1.331e-06 2.712 (46) 7.044e-07 2.888e-06 0.244 (46) 8.533e-05 1.093e-05 7.809 (46) 5.264e-05 1.685e-05 3.123 (46) R 2 0.4361 0.0461 0.1189 0.0761 0.1378 0.0013 0.5700 0.1750 –4.903e-07 1.204e-05 –0.041 (46) 3.61e-05 7.328e-05 1.068e-05 6.861 (46) 1.281e-05 4.384e-06 2.922 (46) 4.865e-05 1.254e-05 3.880 (46) 7.493e-05 1.243e-05 6.027 (46) –1.938e-05 3.645e-05 –0.532 (46) –3.314e-05 8.021e-06 –4.132 (46) 0.5058 0.1566 0.2466 0.4412 0.0061 0.2707 P value <.001 .14 .02 .06 .009 .81 <.001 .003 .97 <.001 .005 <.001 <.001 .60 <.001 https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Figure 4. Correlation results among the 15 subcategories of features. Achieve: achievement; Anx: anxiety; Excl: exclusive; NegEmo: negative emotion; PosEmo: positive emotion; Sad: sadness. Topic Analysis of Sina Weibo Posts We performed topic modeling on 325,807 anxiety-related posts from April 2018 to March 2022 by month. Since the setting of the number of topics in the topic model will impact the effect of the model, we compared the results of the models with 20, 30, 50, and 100 topics, and finally, we selected the model with 30 topics for further analysis. In addition, we selected a subset of topics that were easy to infer and mark their meanings. Then, we divided the topics into 5 areas: (1) discrimination and stigma, (2) symptoms and physical health, (3) treatment and support, (4) work and social, and (5) family and life (Table S3 in Multimedia Appendix 1). Each area included at least 3 related topics. To make the changing trend of occurrence probability in each area clearer, we calculated the probability of occurrence in each month in each area first and then calculated the average probability of occurrence in each quarter in each area (Figure 5). The results showed that from April 2018 to March 2022, the topical area with the highest probability of occurrence was discrimination and stigma, followed by symptoms and physical health, family and life, and work and social; the lowest was treatment and support. Especially, in the early stage of COVID-19, the occurrence probability of the topical area “discrimination and stigma” reached the highest value, while the occurrence probability of the topical areas “symptoms and physical health” and “family and life” both decreased, reaching almost the lowest value in 4 years. Specifically, the topical area related to discrimination and stigma averagely accounted for 26.66% in the 4-year period and 37.75% in the early stage of COVID-19. Table 6 shows the linear regression results of the occurrence probability of the 5 topical areas. It can be seen that “symptoms and physical health” had a significantly positive trend, while “family and life” had a significantly negative trend. In general, combining the results of linguistic feature analysis and topic model, we observed the following results: first, the level of anxiety among Weibo users was significantly affected by the beginning of new terms, which may be due to the high proportion of teenage users. The topic model shows that “work and social” had obvious periodicity, with a high occurrence probability in the first and third quarters of each year and a low occurrence probability in the second and fourth quarters. According to the trend analysis of the number and total length of posts, this area was greatly impacted by topics related to “new term begins.” Second, the public was more anxious around the spring festival. Treatment and support had an obvious upward trend in 2019 and the first quarter of 2021, reaching the minimum in the third quarter. The upward trend was not obvious in early 2020 because of COVID-19, but it still had an upward trend. Third, in the early stage of COVID-19, the public paid more attention to discrimination and stigma, while “symptoms https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al and physical health” and “family and life” showed a significant downward trend. This is consistent with the analysis of the linguistic features in Table 4 that the frequency of CogMech, Affect, and Bio increased, while the frequency of Social decreased. It is worth noting that the public can view anxiety in a positive light, with the topic model showing that family and life is inversely correlated with discrimination and stigma by the changing trend. From the results of the linguistic feature analysis, work and family were negatively correlated with almost all other features. This suggests that people should focus more on family and life to reduce discrimination and stigma. Figure 5. Average quarterly probability of occurrence of the 5 topics in 325,807 anxiety-related posts from April 2018 to March 2022. Table 6. Linear regression results of the probability of occurrence of the topics. Discrimination and stigma Symptoms and physical health Treatment and support Work and social Family and life Discussion Predictor Time Time Time Time Time Estimate 4.00e-03 4.36e-03 7.65e-04 1.48e-03 SE 2.78e-03 1.90e-03 1.42e-03 2.18e-03 t test (df) 1.440 (14) 2.302 (14) 0.538 (14) 0.678 (14) R 2 0.1291 0.2746 0.02026 0.03176 –3.29e-03 1.82e-03 –1.805 (14) 0.1888 P value .17 .04 .60 .51 .09 Changes in Anxiety-Related Sina Weibo Content Over Time This study analyzes the changes in anxiety-related post content over time from 3 aspects: post quantity, text characteristics, and topic structure. The results of our study indicated the following: 1. Between April 2018 and March 2022, there was a clear increase in public interest in anxiety disorders, particularly in terms of symptoms and treatment. According to the trend analysis of the number and total length of posts, whether it is the total number of anxiety-related posts or the number of anxiety-related posts without the keyword “new term begins,” the number of posts increased over time. Moreover, symptoms and physical health showed a significantly positive trend according to the linear regression results of topic frequency in 5 fields. https://www.jmir.org/2023/1/e45777 XSL•FO RenderX 2. Public anxiety was greatly impacted by COVID-19, particularly in terms of work and social life. From the analysis of the frequency of anxiety-related posts for the 5 categories, namely, Social, Affect, Bio, CogMech, and Percept, words related to CogMech appeared most frequently, followed by words related to Affect, Bio, and Social. The frequency of words related to Percept was the lowest, and the changes were not obvious during the 4-year period. At the beginning of COVID-19, the linear regression results of the frequency of the 5 major categories of linguistic features showed a downward trend for Social-related words, while the remaining 4 categories showed an upward trend, with the most obvious increase in CogMech-related words. According to the linear regression results of the 15 subcategories, NegEmo, anxiety, anger, sadness, body, health, exclusive, feel, I, and assent showed an upward trend, while family, PosEmo, insight, work, and achievement showed a downward trend. J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al 3. The public perception of discrimination and stigma against anxiety disorders remained high, especially in terms of self-denial and negative emotions. According to the correlation results among the 15 subcategories in Figure 4, the correlation between anxiety and NegEmo was the highest. Moreover, the most frequent topic in the topic analysis was discrimination and stigma, which was the most discussed topic, especially in the early stage of COVID-19. Notably, the topic model also showed that the proportion of posts referring to family and life was lower than that of posts referring to discrimination and stigma. However, according to the average occurrence probability of anxiety-related posts in the topic model, the 2 topics were inversely correlated by trend analysis. In addition, there was a clear seasonal pattern in the topic of symptoms and physical health. Figure 5 shows that the occurrence probability of the topical area “symptoms and health” increased from autumn to winter, and Figure 3 shows that there was a significant increase in the number of the words of Affect and Bio from autumn to winter. This may indicate that people’s anxiety increases from autumn to winter and that the public pays more attention to their health in the winter. This phenomenon was similar to the change in the depression degree among social media users [60]. Previous studies have shown that compared to the control group, users with mental disorders prefer to use first-person pronouns [61]. In this study, we found that the frequency of first-person pronouns in anxiety-related posts remained high and tended to increase over time. This phenomenon suggests that anxiety-related users pay more attention to their CogMech. In addition, our study found a high frequency and a significant increase in the occurrence frequency of the words related to assent in anxiety-related posts. Assent words are usually positively correlated with users’ positive emotions [62,63]. However, in social media, some assent words indicate users’ disdain or helplessness to some extent. Limitations This study has 3 main limitations. First, social media users cannot represent the entire Chinese population. According to the 50th statistical report on China’s internet development [64], the number of internet users in China reached 1.051 billion by June 2022. Most of these users were adults aged 20-39 years. Concurrently, there were 362 million non–internet users, most of whom lived in rural areas without access to the internet (41.2% of non–internet users live in rural areas). Lack of internet usage skills, literacy restrictions, age factors, and inadequate equipment are the main reasons for non–internet users not to access the internet [64]. Therefore, our results cannot be generalized to non–social media users. The trends of attitude changing over time found in this study may be different from the real trend. Second, the data analyzed in this paper are only for posts containing the keyword anxiety disorder; therefore, some posts that do not contain the keyword but express attitudes toward anxiety disorders are ignored. Third, the results obtained by this simple filtering method are only the overall attitude of Sina Weibo users toward anxiety disorders, lacking fine-grained analysis and without considering the impact of changes in the number of Weibo users. Additionally, there are other social media platforms in China besides Sina Weibo, and the generalization of our findings should be considered cautiously. Conclusions and Recommendations In this study, the attitudes of Weibo users toward anxiety disorder and the changes in attitudes over time were explored by analyzing the linguistic features and topical structure of anxiety-related posts on Sina Weibo from April 2018 to March 2022. The results of this study showed that public discrimination and stigma against anxiety was high and that public anxiety was greatly impacted by the beginning of a new term and COVID-19. More specifically, the topics related to discrimination and stigma averagely accounted for 26.66% in the 4-year period and 37.75% in the early stage of COVID-19. Approximately 10.67% (34,771/325,807) of the posts contained the keyword “new term begins,” and the number increased dramatically at the beginning of each new term. Fortunately, public awareness of the symptoms and physical health aspects of anxiety disorders has increased significantly over time. Therefore, social media can be used to provide resources and assistance to patients with anxiety disorders and improve social media activities related to discrimination and stigma reduction. In the future, we can try to consider English words, emoticons, and other characters in posts to more accurately reveal public attitudes toward anxiety disorder. In addition, by adding more keywords in the process of data collection, more posts on anxiety topics can be acquired and analyzed, which may expand the scope of research and generate more meaningful results. Acknowledgments This work was supported in part by the Sci-Tech Innovation (STI) 2030-major projects (grant 2021ZD0202000), the National Natural Science Foundation of China (grants 62227807 and 62072219), and the Supercomputing Center of Lanzhou University. Conflicts of Interest None declared. Multimedia Appendix 1 Supplementary data. [XLSX File (Microsoft Excel File), 28 KB-Multimedia Appendix 1] References https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al 1. Mental health of adolescents. World Health Organization. URL: https://www.who.int/news-room/fact-sheets/detail/ 2. 3. 4. 5. 6. adolescent-mental-health [accessed 2022-10-25] Report on the Nutrition and Chronic Diseases Status of Chinese Residents (2020). Beijing: People's medical Publishing House; 2020. COVID-19 Mental Disorders Collaborators. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. Lancet 2021 Nov 06;398(10312):1700-1712 [FREE Full text] [doi: 10.1016/S0140-6736(21)02143-7] [Medline: 34634250] Plaisier I, Beekman A, de Graaf R, Smit J, van Dyck R, Penninx B. Work functioning in persons with depressive and anxiety disorders: the role of specific psychopathological characteristics. J Affect Disord 2010 Sep;125(1-3):198-206. [doi: 10.1016/j.jad.2010.01.072] [Medline: 20185180] Iancu SC, Batelaan NM, Zweekhorst MBM, Bunders JFG, Veltman DJ, Penninx BWJH, et al. Trajectories of functioning after remission from anxiety disorders: 2-year course and outcome predictors. Psychol Med 2014 Feb;44(3):593-605. [doi: 10.1017/S0033291713001050] [Medline: 23659543] Yang X, Fang Y, Chen H, Zhang T, Yin X, Man J, et al. Global, regional and national burden of anxiety disorders from 1990 to 2019: results from the Global Burden of Disease Study 2019. Epidemiol Psychiatr Sci 2021 May 06;30:e36 [FREE Full text] [doi: 10.1017/S2045796021000275] [Medline: 33955350] 7. Meier SM, Mattheisen M, Mors O, Mortensen PB, Laursen TM, Penninx BW. Increased mortality among people with 8. 9. anxiety disorders: total population study. Br J Psychiatry 2016 Sep;209(3):216-221 [FREE Full text] [doi: 10.1192/bjp.bp.115.171975] [Medline: 27388572] GBD 2016 DiseaseInjury IncidencePrevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017 Sep 16;390(10100):1211-1259 [FREE Full text] [doi: 10.1016/S0140-6736(17)32154-2] [Medline: 28919117] Gustavsson A, Svensson M, Jacobi F, Allgulander C, Alonso J, Beghi E, CDBE2010Study Group. Cost of disorders of the brain in Europe 2010. Eur Neuropsychopharmacol 2011 Oct;21(10):718-779. [doi: 10.1016/j.euroneuro.2011.08.008] [Medline: 21924589] 10. Ho KP, Hunt C, Li S. Patterns of help-seeking behavior for anxiety disorders among the Chinese speaking Australian community. Soc Psychiatry Psychiatr Epidemiol 2008 Nov;43(11):872-877. [doi: 10.1007/s00127-008-0387-0] [Medline: 18575788] 11. Alonso J, Liu Z, Evans-Lacko S, Sadikova E, Sampson N, Chatterji S, WHO World Mental Health Survey Collaborators. Treatment gap for anxiety disorders is global: Results of the World Mental Health Surveys in 21 countries. Depress Anxiety 2018 Mar;35(3):195-208 [FREE Full text] [doi: 10.1002/da.22711] [Medline: 29356216] 12. Rüsch N, Angermeyer MC, Corrigan PW. Mental illness stigma: concepts, consequences, and initiatives to reduce stigma. Eur Psychiatry 2005 Dec;20(8):529-539. [doi: 10.1016/j.eurpsy.2005.04.004] [Medline: 16171984] 13. Davies MR. The stigma of anxiety disorders. Int J Clin Pract 2000;54(1):44-47. [Medline: 10750260] 14. Schofield CA, Ponzini GT. The Skidmore Anxiety Stigma Scale (SASS): A covert and brief self-report measure. J Anxiety Disord 2020 Aug;74:102259 [FREE Full text] [doi: 10.1016/j.janxdis.2020.102259] [Medline: 32585425] 15. Alonso J, Buron A, Bruffaerts R, He Y, Posada-Villa J, Lepine J, World Mental Health Consortium. Association of perceived stigma and mood and anxiety disorders: results from the World Mental Health Surveys. Acta Psychiatr Scand 2008 Oct;118(4):305-314 [FREE Full text] [doi: 10.1111/j.1600-0447.2008.01241.x] [Medline: 18754833] 16. Anxiety disorders. National Institute of Mental Health. URL: https://www.nimh.nih.gov/health/topics/anxiety-disorders [accessed 2022-08-10] 17. Yousefi R, Ghorban A. The comparison of appraisals and attitudes toward illness in schizophrenia, schizoaffective and 18. 19. major depression disorder. J Clin Psycol 2009;1(3):1-12. [doi: 10.22075/JCP.2017.2010] Imel ZE, Steyvers M, Atkins DC. Computational psychotherapy research: scaling up the evaluation of patient-provider interactions. Psychotherapy (Chic) 2015 Mar;52(1):19-30 [FREE Full text] [doi: 10.1037/a0036841] [Medline: 24866972] Shen J, Rudzicz F. Detecting anxiety through Reddit. 2017 Presented at: Proceedings of the Fourth Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality; August; Vancouver, Canada p. 58-65. [doi: 10.18653/v1/w17-3107] 20. Wang Y, Kraut R, Levine J. To stay or leave? the relationship of emotional and informational support to commitment in online health support groups. 2012 Presented at: Proceedings of the ACM conference on Computer Supported Cooperative Work (CSCW'12); February; New York, USA p. 833-842. [doi: 10.1145/2145204.2145329] 21. Giles DC, Newbold J. Self- and other-diagnosis in user-led mental health online communities. Qual Health Res 2011 Mar;21(3):419-428. [doi: 10.1177/1049732310381388] [Medline: 20739589] 22. Oh YS, Jung H. The Relationships between Depression and Anxiety Disorder and Online Social Media for Healthcare. Am J Health Behav 2020 Jul 01;44(4):409-419. [doi: 10.5993/AJHB.44.4.4] [Medline: 32553023] 23. Ríssola EA, Aliannejadi M, Crestani F. Mental disorders on online social media through the lens of language and behaviour: Analysis and visualisation. Information Processing & Management 2022 May;59(3):102890. [doi: 10.1016/j.ipm.2022.102890] https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al 24. Uban A, Chulvi B, Rosso P. An emotion and cognitive based analysis of mental health disorders from social media data. Future Generation Computer Systems 2021 Nov;124:480-494. [doi: 10.1016/j.future.2021.05.032] 25. Lyons M, Aksayli ND, Brewer G. Mental distress and language use: Linguistic analysis of discussion forum posts. Computers in Human Behavior 2018 Oct;87:207-211. [doi: 10.1016/j.chb.2018.05.035] 26. De Choudhury M, Counts S, Horvitz E, Hoff A. Characterizing and predicting postpartum depression from shared Facebook data. 2014 Presented at: Proceedings of the 17th ACM conference on Computer Supported Cooperative Work & Social Computing (CSCW'14); February; New York, USA p. 626-638. [doi: 10.1145/2531602.2531675] 27. Guntuku SC, Yaden DB, Kern ML, Ungar LH, Eichstaedt JC. Detecting depression and mental illness on social media: an integrative review. Current Opinion in Behavioral Sciences 2017 Dec;18:43-49. [doi: 10.1016/j.cobeha.2017.07.005] 28. Liu J, Shi M. What Are the Characteristics of User Texts and Behaviors in Chinese Depression Posts? Int J Environ Res Public Health 2022 May 18;19(10):6129 [FREE Full text] [doi: 10.3390/ijerph19106129] [Medline: 35627666] Primack BA, Shensa A, Escobar-Viera CG, Barrett EL, Sidani JE, Colditz JB, et al. Use of multiple social media platforms and symptoms of depression and anxiety: A nationally-representative study among U.S. young adults. Computers in Human Behavior 2017 Apr;69:1-9. [doi: 10.1016/j.chb.2016.11.013] 29. 30. Youngmann B, Yom-Tov E. Anxiety and information seeking: evidence from large-scale mouse tracking. 2018 Apr Presented at: Proceedings of the 2018 World Wide Web Conference; April; Lyon, France p. 753-762. [doi: 10.1145/3178876.3186156] 31. Ahmed A, Aziz S, Toro CT, Alzubaidi M, Irshaidat S, Serhan HA, et al. Machine learning models to detect anxiety and depression through social media: A scoping review. Comput Methods Programs Biomed Update 2022;2:100066 [FREE Full text] [doi: 10.1016/j.cmpbup.2022.100066] [Medline: 36105318] 32. Ta N, Li K, Yang Y, Jiao F, Tang Z, Li G. Evaluating Public Anxiety for Topic-Based Communities in Social Networks. IEEE Trans Knowl Data Eng 2022 Mar 1;34(3):1191-1205. [doi: 10.1109/tkde.2020.2989759] 33. Yang Y, Ta N, Li K, Jiao F, Hu B, Li Z. Influential Factors on Collective Anxiety of Online Topic-Based Communities. Front Psychol 2021;12:740065 [FREE Full text] [doi: 10.3389/fpsyg.2021.740065] [Medline: 34675846] 34. Giunti G, Claes M, Dorronzoro Zubiete E, Rivera-Romero O, Gabarron E. Analysing Sentiment and Topics Related to Multiple Sclerosis on Twitter. Stud Health Technol Inform 2020 Jun 16;270:911-915. [doi: 10.3233/SHTI200294] [Medline: 32570514] 35. Bai H, Yu G. A Weibo-based approach to disaster informatics: incidents monitor in post-disaster situation via Weibo text negative sentiment analysis. Nat Hazards 2016 May 30;83(2):1177-1196. [doi: 10.1007/s11069-016-2370-5] 36. Li A, Huang X, Jiao D, O'Dea B, Zhu T, Christensen H. An analysis of stigma and suicide literacy in responses to suicides broadcast on social media. Asia Pac Psychiatry 2018 Mar;10(1):e12314. [doi: 10.1111/appy.12314] [Medline: 29383880] 37. Li A, Jiao D, Zhu T. Detecting depression stigma on social media: A linguistic analysis. J Affect Disord 2018 May;232:358-362. [doi: 10.1016/j.jad.2018.02.087] [Medline: 29510353] 38. Yu L, Jiang W, Ren Z, Xu S, Zhang L, Hu X. Corrigendum to "Detecting changes in attitudes toward depression on Chinese social media: A text analysis": [Journal of Affective Disorders 280 (2021) 354-363]. J Affect Disord 2021 Feb 15;281:994-995. [doi: 10.1016/j.jad.2020.12.038] [Medline: 33341282] 39. Li A, Jiao D, Zhu T. Stigmatizing Attitudes Across Cybersuicides and Offline Suicides: Content Analysis of Sina Weibo. J Med Internet Res 2022 Apr 08;24(4):e36489 [FREE Full text] [doi: 10.2196/36489] [Medline: 35394437] 40. Li A, Jiao D, Liu X, Zhu T. A Comparison of the Psycholinguistic Styles of Schizophrenia-Related Stigma and 41. Depression-Related Stigma on Social Media: Content Analysis. J Med Internet Res 2020 Apr 21;22(4):e16470 [FREE Full text] [doi: 10.2196/16470] [Medline: 32314969] Pennebaker J, Booth R, Boyd R, Francis M. Linguistic inquiry and word count: LIWC2015. URL: https://www. researchgate.net/publication/337731895_Linguistic_Inquiry_and_Word_Count_LIWC2015 [accessed 2023-03-22] 42. Krippendorff K. Content Analysis: An Introduction to Its Methodology. Thousand Oaks, CA: SAGE Publications; 2019. 43. Reavley NJ, Pilkington PD. Use of Twitter to monitor attitudes toward depression and schizophrenia: an exploratory study. PeerJ 2014;2:e647 [FREE Full text] [doi: 10.7717/peerj.647] [Medline: 25374786] 44. Lachmar EM, Wittenborn AK, Bogen KW, McCauley HL. #MyDepressionLooksLike: Examining Public Discourse About Depression on Twitter. JMIR Ment Health 2017 Oct 18;4(4):e43 [FREE Full text] [doi: 10.2196/mental.8141] [Medline: 29046270] 45. Lafferty J, Blei D. Correlated topic models. 2005 Presented at: Proceedings of the 18th International Conference on Neural Information Processing Systems (NIPS'05); December; Vancouver, British Columbia, Canada p. 147-154. [doi: 10.5555/2976248.2976267] 46. Griffiths TL, Steyvers M, Tenenbaum JB. Topics in semantic representation. Psychol Rev 2007 Apr;114(2):211-244. [doi: 47. 48. 10.1037/0033-295X.114.2.211] [Medline: 17500626] Franz PJ, Nook EC, Mair P, Nock MK. Using Topic Modeling to Detect and Describe Self-Injurious and Related Content on a Large-Scale Digital Platform. Suicide Life Threat Behav 2020 Feb;50(1):5-18. [doi: 10.1111/sltb.12569] [Medline: 31264733] Jo W, Lee J, Park J, Kim Y. Online Information Exchange and Anxiety Spread in the Early Stage of the Novel Coronavirus (COVID-19) Outbreak in South Korea: Structural Topic Model and Network Analysis. J Med Internet Res 2020 Jun 02;22(6):e19455 [FREE Full text] [doi: 10.2196/19455] [Medline: 32463367] https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al 49. Blei DM. Probabilistic topic models. Commun. ACM 2012 Apr 01;55(4):77-84. [doi: 10.1145/2133806.2133826] 50. Paul MJ, Dredze M. Discovering health topics in social media using topic models. PLoS One 2014;9(8):e103408 [FREE Full text] [doi: 10.1371/journal.pone.0103408] [Medline: 25084530] Sik D, Németh R, Katona E. Topic modelling online depression forums: beyond narratives of self-objectification and self-blaming. J Ment Health 2021 Sep 28:1-10. [doi: 10.1080/09638237.2021.1979493] [Medline: 34582309] 51. 52. Low DM, Rumker L, Talkar T, Torous J, Cecchi G, Ghosh SS. Natural Language Processing Reveals Vulnerable Mental Health Support Groups and Heightened Health Anxiety on Reddit During COVID-19: Observational Study. J Med Internet Res 2020 Oct 12;22(10):e22635 [FREE Full text] [doi: 10.2196/22635] [Medline: 32936777] 53. Elbattah M, Arnaud É, Gignon M, Dequen G. The role of text analytics in health care: a review of recent developments and applications. 2021 Presented at: Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies; February 11-13; Virtual p. 825-832. [doi: 10.5220/0010414508250832] 54. Le Glaz A, Haralambous Y, Kim-Dufor D, Lenca P, Billot R, Ryan TC, et al. Machine Learning and Natural Language Processing in Mental Health: Systematic Review. J Med Internet Res 2021 May 04;23(5):e15708 [FREE Full text] [doi: 10.2196/15708] [Medline: 33944788] 55. Whitehead LC. Methodological and ethical issues in Internet-mediated research in the field of health: an integrated review of the literature. Soc Sci Med 2007 Aug;65(4):782-791. [doi: 10.1016/j.socscimed.2007.03.005] [Medline: 17512105] 56. TextMind: A Chinese Language Psychological Analysis System. URL: http://ccpl.psych.ac.cn/textmind/ [accessed 2022-08-10] 57. Gao R, Hao B, Li H, Gao Y, Zhu T. Developing simplified Chinese psychological linguistic analysis dictionary for Microblog. 2013 Presented at: International Conference on Brain & Health Informatics (BHI'13); October; Maebashi, Japan p. 359-368. [doi: 10.1007/978-3-319-02753-1_36] 58. Yan X, Guo J, Lan Y, Cheng X. A biterm topic model for short texts. 2013 Presented at: Proceedings of the 22nd International Conference on World Wide Web; May; Rio de Janeiro, Brazil p. 1445-1456. [doi: 10.1145/2488388.2488514] 59. Robert H, David S. Time Series Analysis and Its Applications: With R Examples (3rd Edition). New York, NY: Springer; 60. 2011. Schwartz H, Eichstaedt J, Kern M, Park G, Sap M, Stillwell D, et al. Towards assessing changes in degree of depression through Facebook. 2014 Jul 27 Presented at: Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality; 2014; Baltimore, Maryland, USA p. 118-125. [doi: 10.3115/v1/w14-3214] 61. Bathina KC, Ten Thij M, Lorenzo-Luaces L, Rutter LA, Bollen J. Individuals with depression express more distorted thinking on social media. Nat Hum Behav 2021 Apr;5(4):458-466. [doi: 10.1038/s41562-021-01050-7] [Medline: 33574604] 62. Nguyen T. Interaction Involvement in Cross-Culture Computer-Mediated Communication: Examination of a Communication Process in Dyadic Instant Messaging Conversations. Ann Arbor, MI: Proquest LLC; 2013. 63. Verberne S, Batenburg A, Sanders R, van Eenbergen M, Das E, Lambooij MS. Analyzing Empowerment Processes Among Cancer Patients in an Online Community: A Text Mining Approach. JMIR Cancer 2019 Apr 17;5(1):e9887 [FREE Full text] [doi: 10.2196/cancer.9887] [Medline: 30994468] 64. The 50th statistical report on China's internet development. China Internet Network Information Center. URL: http://www. cnnic.net.cn/index.html [accessed 2022-10-20] Abbreviations Affect: affective process Bio: biological process BTM: biterm topic model CogMech: cognitive process LDA: Latent Dirichlet Allocation LIWC: Linguistic Inquiry and Word Count NegEmo: negative emotion Percept: perceptual process PosEmo: positive emotion Social: social process https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 14 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zhu et al Edited by A Mavragani; submitted 16.01.23; peer-reviewed by M Tummalacherla, S Shaikh, M Elbattah; comments to author 27.02.23; revised version received 07.03.23; accepted 08.03.23; published 04.04.23 Please cite as: Zhu J, Li Z, Zhang X, Zhang Z, Hu B Public Attitudes Toward Anxiety Disorder on Sina Weibo: Content Analysis J Med Internet Res 2023;25:e45777 URL: https://www.jmir.org/2023/1/e45777 doi: 10.2196/45777 PMID: 37014691 ©Jianghong Zhu, Zepeng Li, Xiu Zhang, Zhenwen Zhang, Bin Hu. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 04.04.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e45777 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e45777 \| p. 15 (page number not for citation purposes)
10.1371_journal.ppat.1011443	RESEARCH ARTICLE Regulation of PKR-dependent RNA translation inhibition by TRIM21 upon virus infection or other stress 1, Shun Liu1, Qing Feng1, Rilin Deng1, Jingjing Wang1, Xintao Wang1, Huiyi LiID Renyun Tian1, Yan Xu1, Shengwen Chen1, Qian Liu1, Luoling Wang1, Xinran Li1, Mengyu Wan1, Yousong Peng1, Songqing Tang1, Binbin Xue2, Haizhen ZhuID 1,2 1 Institute of Pathogen Biology and Immunology of College of Biology, Hunan Provincial Key Laboratory of Medical Virology, State Key Laboratory of Chemo/Biosensing and Chemometrics, Hunan University, Changsha, Hunan, China, 2 Key Laboratory of Tropical Translational Medicine of Ministry of Education, Department of Pathogen Biology and Immunology, Institute of Pathogen Biology and Immunology, School of Basic Medicine and Life Science, The University of Hong Kong Joint Laboratory of Tropical Infectious Diseases, The First Affiliated Hospital and The Second Affiliated Hospital of Hainan Medical University, Hainan Medical University, Hainan, China * [email protected] (BX); [email protected] (HZ) a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Abstract Citation: Li H, Liu S, Feng Q, Deng R, Wang J, Wang X, et al. (2023) Regulation of PKR- dependent RNA translation inhibition by TRIM21 upon virus infection or other stress. PLoS Pathog 19(6): e1011443. https://doi.org/10.1371/journal. ppat.1011443 Editor: Jacob S. Yount, The Ohio State University, UNITED STATES Received: December 21, 2022 Accepted: May 25, 2023 Published: June 16, 2023 Copyright: © 2023 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by grants from the National Natural Science Foundation of China (81730064, 82072269, 81571985 and 81902069), National Major Science and Technology Projects of China (2017ZX10202201-005 and 2009ZX10004- 312), Postdoctoral Research Foundation of China (2019M652760) and Hunan Provincial Innovation Foundation For Postgraduate (CX20210395). The The host always employs various ways to defend against viral infection and spread. How- ever, viruses have evolved their own effective strategies, such as inhibition of RNA transla- tion of the antiviral effectors, to destroy the host’s defense barriers. Protein synthesis, commonly controlled by the α-subunit of eukaryotic translation initiation factor 2 (eIF2α), is a basic cellular biological process among all species. In response to viral infection, in addition to inducing the transcription of antiviral cytokines by innate immunity, infected cells also inhibit the RNA translation of antiviral factors by activating the protein kinase R (PKR)-eIF2α signaling pathway. Regulation of innate immunity has been well studied; however, regula- tion of the PKR-eIF2α signaling pathway remains unclear. In this study, we found that the E3 ligase TRIM21 negatively regulates the PKR-eIF2α signaling pathway. Mechanistically, TRIM21 interacts with the PKR phosphatase PP1α and promotes K6-linked polyubiquitina- tion of PP1α. Ubiquitinated PP1α augments its interaction with PKR, causing PKR dephos- phorylation and subsequent translational inhibition release. Furthermore, TRIM21 can constitutively restrict viral infection by reversing PKR-dependent translational inhibition of various previously known and unknown antiviral factors. Our study highlights a previously undiscovered role of TRIM21 in regulating translation, which will provide new insights into the host antiviral response and novel targets for the treatment of translation-associated dis- eases in the clinic. Author summary Transcriptional induction of antiviral cytokines, especially interferons (IFNs), is a hall- mark of innate immunity upon viral infection. In addition, viral infection can inhibit PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 1 / 28 PLOS PATHOGENSfunders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors declare no competing interests. Regulation of PKR-dependent protein synthesis by TRIM21 upon virus infection RNA translation of the antiviral factors by activating the PKR-eIF2α axis. Regulation of the innate immune signaling pathway has been well studied. However, regulation of the PKR signaling pathway remains unclear. Our previous study demonstrated that TRIM21, an E3 ligase, positively regulates the innate immune response by catalyzing K27-linked ubiquitination of MAVS. However, whether TRIM21 regulates the PKR signaling pathway is unknown. In this study, we found that TRIM21 can negatively regulate PKR activation by pro- moting K6-linked ubiquitination of the PKR phosphatase PP1α, contributing to the enhancement of the PKR-PP1α interaction, which leads to dephosphorylation of PKR. Moreover, our study identified various previously known and unknown antiviral factors regulated by the TRIM21-PKR axis. Our study highlights TRIM21-mediated PKR-depen- dent RNA translation in the antiviral response, which may provide a target for the treat- ment of translation-associated diseases in the clinic. Introduction To survive and produce viral particles, viruses employ various strategies, such as shutdown of the protein synthesis of antiviral factors, to escape immune supervision by host cells. Protein synthesis is a basic biological process among all species, including three steps: initiation, elon- gation and termination. Appropriate regulation of RNA translation is essential for the mainte- nance of protein homeostasis, and disruption of RNA translation by extrinsic or intrinsic stresses is extremely prone to causing severe organ damage, which contributes to tumorigene- sis and inflammation [1]. In response to extracellular or intracellular alterations, eukaryotic cells commonly employ programs that conserve energy and facilitate the reprogramming of gene expression by inhibiting RNA translation to adapt to alterations in the cellular environ- ment [2]. However, this protective mechanism of the host can be also beneficial to the survival of the virus because of the inhibition of antiviral effector RNA translation upon viral infection. The central mechanism regulating protein synthesis in cells upon stimulation involves the phosphorylation of the translation initiation factor eIF2α [3]. There are four kinases, the dsRNA-dependent protein kinase known as protein kinase R (PKR), general control nonre- pressed 2 (GCN2), PKR-like endoplasmic reticulum kinase (PERK) and heme-regulated eIF2α kinase (HRI), that are responsible for eIF2α phosphorylation in eukaryotic cells [4]. Among them, as one of the pattern recognition receptors (PRRs) involved in innate immunity, PKR is distinct for its essential role in response to viral infection or inflammasome-mediated innate immunity [5]. Viral infection triggers PKR-dependent protein synthesis inhibition. PKR is a serine-threo- nine kinase comprised of two conserved double-stranded RNA binding motifs (dsRBMs) in its N-terminal domain and a C-terminal kinase domain. In the cytoplasm, by detecting dsRNA via its N-terminal dsRBM, PKR dimerizes and rapidly phosphorylates itself at residues Thr446 and Thr451 and then activates eIF2α by phosphorylation, resulting in protein synthesis inhibi- tion [6,7]. Accumulated evidence have demonstrated that PKR-mediated inhibition of transla- tion initiation plays crucial roles in human physiological and pathological processes, such as the regulation of cell proliferation, differentiation, apoptosis, viral infection, cancer and inflammation [8–11]. However, the precise regulatory mechanism remains unclear. In addi- tion to activating the PKR-eIF2α signaling pathway, dsRNA can also trigger innate immunity activation by binding to the PRRs in the cytosol, such as retinoic acid-inducible gene I (RIG-I) and melanoma differentiation associated protein 5 (MDA5). RIG-I- or MDA5-mediated PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 2 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection innate immune signaling accounts for the activation of the transcription factors IRF3 and NF- κB, subsequently resulting in the induction of the transcription of multiple antiviral cytokines, particularly IFNs [12,13]. It is well known that regulation of the RIG-I- or MDA5-mediated signaling pathway has been extensively studied. However, regulation of the PKR signaling pathway remains unclear. TRIM21 is a member of the tripartite motif (TRIM) superfamily, which plays important roles in diverse biological and pathophysiological processes [14]. Our previous study showed that TRIM21, by catalyzing the K27-linked ubiquitination of mitochondrial antiviral signaling protein (MAVS), augments RIG-I/MDA5-mediated transcription of IFNs upon viral infection [15]. However, whether TRIM21 is able to restrict viral infection via regulation of the protein synthesis of antiviral effectors remains elusive. In this study, we found that TRIM21 negatively regulates PKR-dependent translational shutdown upon viral infection or thapsigargin (TG) treatment. Knockout or knockdown of TRIM21 augments virus- or TG-induced PKR phos- phorylation. Mechanistically, viral infection or TG treatment promotes TRIM21 binding to the PKR phosphatase PP1α, which leads to the K6-linked polyubiquitination of PP1α. Ubiqui- tinated PP1α augments its interaction with PKR, resulting in dephosphorylation of PKR and subsequent eIF2α inactivation, resulting in the release of protein synthesis inhibition. More- over, in addition to its antiviral function through promotion of the transcription of IFNs, pro- teomics analysis revealed that TRIM21 is able to restrict viral infection by reversing PKR- mediated inhibition of the protein synthesis of previously known and unknown intrinsic anti- viral genes in cells. Collectively, our data highlight the essential role of TRIM21 in regulating RNA translation, which may provide new strategies for the treatment of translation initiation- associated diseases. Results TRIM21 inhibits the activation of PKR and promotes RNA translation To ascertain whether TRIM21 affects PKR signaling pathway activation, we used CRISPR– Cas9 gene editing technology to establish TRIM21-deficient A549 cell lines (sg-TRIM21) (S1A Fig). We assessed the effect of TRIM21 in the regulation of the PKR signaling pathway by employing a classic agonist of PKR, poly (I:C), in these cells. By transfection of poly (I:C), the phosphorylation levels of PKR and eIF2α in TRIM21-deficient cells were higher than those in wild-type cells, suggesting a negative regulatory role of TRIM21 in PKR signaling pathway activation (Fig 1A). To rule out the possibility of an off-target effect of TRIM21 sgRNA, we used shRNA specifically targeting TRIM21 to knockdown the expression of TRIM21 in A549 cells (S1B Fig). Similarly, knockdown of TRIM21 enhanced poly (I:C)-induced phosphoryla- tion of PKR and eIF2α (S1C Fig). Next, we examined the effects of TRIM21 on PKR signaling activation upon viral infection. In line with previous studies, vesicular stomatitis virus (VSV) or Sendai virus (SeV) infection constitutively activated PKR and eIF2α, and TRIM21 defi- ciency enhanced their activation (Fig 1B and 1C). Moreover, we obtained similar results in TRIM21-silenced cells upon VSV or SeV infection (S1D and S1E Fig). To exclude the specific- ity of the cell lines, we performed the above experiments in HLCZ01 cells, a liver cell line, which supports the entire life cycle of hepatitis B virus (HBV) and hepatitis C virus (HCV) [16]. Consistently, the levels of p-PKR and p-eIF2α were also increased in HLCZ01 cells with TRIM21 knockdown upon poly (I:C) treatment or viral infection (S1F–S1H Fig), demonstrat- ing that TRIM21 can repress dsRNA- or virus-induced PKR signaling pathway activation. To further investigate the negative regulatory role of TRIM21 in PKR signaling pathway activa- tion, we used different stimuli. Because dsRNAs produced during viral infection can activate PKR [17], we tried to test the function of TRIM21 during infection with a DNA virus, herpes PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 3 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 1. TRIM21 inhibits the activation of PKR. (A-C) Wild-type (sg-vector) or TRIM21-knocked out (sg-TRIM21) A549 cells were transfected with poly(I:C) for 12 h (A), or infected with VSV (MOI = 0.2 or 0.5) for 6 h (B) or infected with SeV (MOI = 0.2 or 0.5) for 12 h (C). The indicated proteins were detected by western blot. β-actin was used as an internal control. (D) A549 cells were transfected with p3xFlag-CMV- vector or p3xFlag-CMV-TRIM21 for 48 h, then infected with HSV (MOI = 0.1 or 0.5) for 6 h. The indicated proteins were detected by western blot. β-actin was used as an internal control. (E-F) A549 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh- TRIM21) for 48 h, then stimulated by thapsigargin (TG) (10 μM or 20 μM) for 12 h (E), or TG (20 μM) for 8 h or 12 h (F). The indicated proteins were detected by western blot. GAPDH was used as a control. (G) Puromycin incorporation assays demonstrated the cellular protein synthesis in mock- or VSV-infected wild-type A549 cells (sg-vector) or TRIM21-deficient A549 cells (sg-TRIM21). The cells were pretreated with puromycin (10 μg/mL) for 1 h. The indicated proteins were detected by western blot. GAPDH was used as a control. (H) Puromycin incorporation assays of the cellular protein synthesis in TRIM21-deficient A549 cells (sg-TRIM21) transfected with p3xFlag-CMV-vector or p3xFlag-CMV-TRIM21 for 48 h, then infected by VSV (MOI = 0.5) for 6 h. GAPDH was used as an internal control. (I) Immunoblot analysis of puromycin and TRIM21 in wild-type A549 cells (sg-vector) or TRIM21-deficient A549 cells (sg-TRIM21) stimulated with TG (10 μM or 20 μM) for 12 h. GAPDH was used as an internal control. (J-K) Puromycin incorporation assays of the cellular protein synthesis in PKR-deficient A549 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 4 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection cells (sg-PKR) transfected with p3xFlag-CMV-vector or p3xFlag-CMV-TRIM21 for 48 h, then infected by VSV (MOI = 0.5) for 6 h (J), or infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then stimulated by TG (10 μM) for 12 h (K), GAPDH was used as an internal control. The relative ratios of p-PKR or p-eIF2α in (A-F) and the relative ratios of the puro signals in (G-K) were quantified by densiometric analysis, which were normalized to the value in the control group. Experiments were independently repeated two or three times with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, p<0.01, *p<0.001, NS, no significance difference. https://doi.org/10.1371/journal.ppat.1011443.g001 simplex virus-1 (HSV-1). Notably, HSV-1 infection substantially activated PKR, while this activation was inhibited by TRIM21 overexpression in the cells (Fig 1D), suggesting that TRIM21 inhibits DNA virus-induced PKR signaling pathway activation. Moreover, TRIM21 also repressed another PKR agonist, ER stress activator thapsigargin (TG)-induced PKR acti- vation (Fig 1E and 1F) [18]. These data demonstrated that TRIM21 negatively regulates PKR signaling pathway activation. PKR activation commonly leads to global protein synthesis inhi- bition, consistent with its inhibition of PKR activation, TRIM21 reversed the global translation inhibition caused by viral infection or TG treatment (Fig 1G–1I). However, the phenomenon was lost upon PKR deficiency (Figs 1J, 1K and S1I), indicating that TRIM21 can release PKR- dependent protein synthesis shutdown. Collectively, these data demonstrated that TRIM21 can inhibit PKR signaling pathway activation and reverse PKR activation-mediated protein synthesis inhibition. The ubiquitin ligase activity of TRIM21 is required for TRIM21-mediated inhibition of PKR activation To explore the mechanism of TRIM21-mediated inhibition of the PKR signaling pathway, we first examined the interaction between TRIM21 and the signaling molecules in the pathway. Notably, TRIM21 interacted with PKR but not eIF2α (Fig 2A and 2B), and the endogenous interaction between them was further confirmed in A549 cells, in which their interaction was enhanced upon viral infection or TG treatment (Fig 2C and 2D), indicating that TRIM21 may inhibit PKR activation by targeting PKR. TRIM21 is an E3 ligase with enzymatic activity within its RING finger domain, which is essential for its function in multiple biological and pathological processes. To investigate whether the E3 ligase activity of TRIM21 is involved in the regulation of PKR activation, we cotransfected pFlag-tagged PKR and hemagglutinin (HA)-tagged ubiquitin (HA-ub) together with pV5-tagged TRIM21 into HEK293T cells and found that TRIM21 has no effect on PKR ubiquitination, indicating that TRIM21 does not directly ubiquitinate PKR (Fig 2E). However, the E3 ligase-inactive mutant (TRIM21-C16S), in which Cys16 was replaced with Ser16, lost the ability to reduce VSV-induced PKR phos- phorylation (Fig 2F). To confirm the essential role of the E3 ligase activity in PKR inactivation, we constructed a TRIM21 mutant deleting RING domain (TRIM21-D-R), and found that the phosphorylation level of PKR in the cells expressing the full-length TRIM21 (Flag- TRIM21-FL) was lower than that in the cells transfecting Flag-vector upon VSV infection, however, the phenomenon was rescued by transfection with Flag-TRIM21-D-R (Fig 2G). Moreover, the E3 ligase activity deficiency also lost the ability to inactive PKR (Fig 2H), sug- gesting that the E3 ligase activity of TRIM21 is required for PKR inactivation. Collectively, all of the data demonstrated that TRIM21 inhibits PKR activation by targeting PKR and that the E3 ligase activity of TRIM21 is crucial for its inactivation. TRIM21 represses PKR activation by targeting PP1α PKR is a serine–threonine kinase comprised of a kinase domain and two dsRNA binding domains that regulate its activity. Upon engagement with dsRNA in the cytoplasm, PKR PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 5 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 2. The ubiquitin ligase activity of TRIM21 is required for TRIM21-mediated inhibition of PKR activation. (A) Immunoprecipitation analysis the interaction between PKR and TRIM21. HEK293T cells were transfected with p3xFlag-CMV-PKR and pcDNA3.1a-TRIM21 for 48 h. The anti-Flag antibody was used to perform immunoprecipitation and detect Flag-PKR, the anti-V5 antibody was used to detect V5-TRIM21. (B) HEK293T cells were co-transfected with p3xFlag-CMV-TRIM21 and pcDNA3.1a-eIF2α for 48 h. The anti-Flag antibody was used to perform immunoprecipitation and detect Flag-PKR. The anti-V5 antibody was used to detect V5-eIF2α. β-actin was used as an internal control. (C-D) Endogenous PKR and TRIM21 interaction was analyzed by IP in A549 cells with VSV (MOI = 0.2 or 0.5) infection for 6 h (C) or with TG (10 μM) treatment for 12 h (D). The anti-TRIM21 antibody was used to perform immunoprecipitation. The indicated proteins were detected by the corresponding antibodies. GAPDH (C) or β-actin (D) was used as an internal control. (E) Immunoprecipitation analysis of the ubiquitination of PKR. HEK293T cells were co-transfected p3xFlag-CMV-PKR with pcDNA3.1a-vector or pcDNA3.1a- TRIM21 as well as HA-ub for 48 h. The anti-Flag antibody was used to perform immunoprecipitation and detect Flag-PKR. The anti-V5 antibody was used to detect V5-TRIM21. GAPDH was used as an internal control. (F) HLCZ01 cells were infected with lentivirus-Flag-TRIM21-WT or lentivirus-Flag- TRIM21-C16S for 48 h, then infected with VSV (MOI = 0.2) for 6 h. The indicated proteins were detected by western blot. β-actin was used as an internal control. (G) TRIM21-deficient A549 cells (A549-sg-TRIM21) were transfected with p3xFlag-CMV-TRIM21 or p3xFlag-CMV-TRIM21-D-R for 48h, then infected with VSV (MOI = 0.5) for 6 h. The indicated proteins were detected by western blot. GAPDH was used as an internal control. (H) TRIM21-deficient A549 cells (A549-sg-TRIM21) were infected with lentivirus-Flag-TRIM21-WT or lentivirus-Flag-TRIM21-C16S for 48 h, then stimulated with TG (20 μM) for 12 h. The indicated proteins were detected by western blot. β-actin was used as an internal control. The relative interaction ratios of PKR and TRIM21 in (C) were quantified by densiometric analysis, which were normalized to the value in the control group. The relative ratios of p-PKR or p-eIF2α in (F-H) were quantified by densiometric analysis, which were normalized to the value in the control group. Experiments were independently repeated two or three times with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, *P<0.01, NS, no significance difference. https://doi.org/10.1371/journal.ppat.1011443.g002 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 6 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection undergoes homodimerization and subsequent rapid autophosphorylation at Thr446 and Thr451, increasing its catalytic activity. Therefore, there are three steps in PKR activation: dsRNA detection, homodimerization and autophosphorylation. To investigate how TRIM21 represses PKR activation, we examined whether TRIM21 affects the processes of PKR activa- tion. First, we performed RNA immunoprecipitation (RIP) analysis to explore whether TRIM21 affects PKR binding to viral RNA in HEK293T cells after infection with VSV. Consis- tently, PKR constitutively bound to viral RNA, and TRIM21 knockdown did not abolish their interaction (Fig 3A), suggesting that TRIM21 has no effect on the detection of dsRNA by PKR. To test whether TRIM21 affects PKR homodimerization, we delivered pFlag-tagged PKR and pMyc-tagged PKR together with different doses of pV5-tagged TRIM21 into HEK293T cells. Notably, overexpression of TRIM21 did not abolish the interaction between Flag-tagged PKR and Myc-tagged PKR (Fig 3B), suggesting that TRIM21 has no effect on PKR homodimeriza- tion, which implies that TRIM21 may impair PKR phosphorylation. Given that PKR phos- phorylates itself and TRIM21 has no effect on PKR dimerization, which is indispensable for the kinase activity of PKR, we speculated that TRIM21 may target the phosphatases of PKR to inactive PKR. To confirm our hypothesis, we focused on a phosphatase, serine/threonine pro- tein phosphatase type 1 alpha (PP1α), the key phosphatase of PKR [19]. Indeed, PP1α inter- acted with PKR as well as TRIM21 (Fig 3C), the endogenous interaction between PP1α and TRIM21 was also evidenced, and their interaction was enhanced upon VSV infection or TG stimulation (Fig 3D and 3E). Moreover, an in vitro pull-down assay with purified recombinant proteins demonstrated a direct interaction between TRIM21 and PP1α (Fig 3F). All the data indicated that TRIM21 may target PP1α for PKR inactivation. TRIM21 promotes PP1α polyubiquitination Since TRIM21 targets PP1α and TRIM21 inhibits PKR activation in a manner dependent on its ubiquitin ligase activity, we investigated whether TRIM21 can ubiquitinate PP1α. We deliv- ered pTRIM21-WT or pTRIM21-C16S together with pFlag-tagged PP1α and pHA-ub into HEK293T cells and found increased ubiquitination of PP1α in TRIM21-WT-transfected cells but not in TRIM21-C16S-transfected cells (Fig 4A), suggesting that TRIM21 promotes the ubi- quitination of PP1α. Different types of polyubiquitin linkages have distinct functions. We cotransfected pFlag-tagged PP1α into HEK293T cells with individual ubiquitin mutants (K6O, K11O, K27O, K29O, K33O, K48O or K63O), each of which contained only one lysine residue available for modification. Apparently, TRIM21 specifically promoted K6-linked ubiquitina- tion of PP1α; however, the K6-linked ubiquitination of PP1α disappeared by inactivation of the ubiquitination ligase activity of TRIM21 (Fig 4B and 4C). Moreover, as a negative control, neither TRIM21-WT nor TRIM21-C16S promoted K11-linked ubiquitination of PP1α (Fig 4C). Furthermore, TRIM21-mediated K6-linked ubiquitination of PP1α was enhanced under conditions of viral infection or TG treatment (Fig 4D). These data demonstrated that TRIM21 catalyzes K6-linked ubiquitination of PP1α. Next, we examined which lysine residue of PP1α is ubiquitinated by TRIM21. PP1α was divided into two fragments, the N-terminus and C-ter- minus, and we delivered the N-terminus or C-terminus of PP1α together with TRIM21-WT or TRIM21-C16S as well as HA-ub into HEK293T cells (Fig 4E). Co-IP assays revealed that the N-terminus of PP1α is ubiquitinated by TRIM21 (Fig 4F), demonstrating that TRIM21 ubiqui- tinates PP1α at the N-terminus. There are 5 lysine residues in the N-terminus of PP1α, and we generated the mutant PP1α-K0, in which all of the lysine residues in the N-terminus of PP1α were replaced with arginine. Then, we reintroduced individual lysine residues into PP1α-K0 to generate single lysine mutants (Fig 4G). Co-IP assay results showed that the K60 mutant of PP1α is ubiquitinated by TRIM21 (Fig 4H), suggesting that TRIM21 may catalyze the PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 7 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 3. TRIM21 represses PKR activation by targeting on PP1α. (A) HEK293T cells pre-infected with lentivirus-sh-vector (sh-vector) or lentivirus- sh-TRIM21 (sh-TRIM21) for 24 h were transfected with p3xFlag-CMV-PKR for additional 48 h, then infected with VSV (MOI = 0.2) for 6 h. RIP assay was performed to test the VSV RNA binding to PKR. (B) Immunoprecipitation detecting PKR dimerization. HEK293T cells were co-transfected p3xFlag-CMV-PKR with pCMV-N-Myc-PKR or pcDNA3.1a-vector or different doses of pcDNA3.1a-TRIM21 (1 μg or 2 μg) for 48 h. The anti-Flag tag antibody was used to perform the immunoprecipitation and detect Flag-PKR, the anti-Myc tag antibody was used to detect Myc-PKR and the anti- V5 tag antibody was used to detect V5-TRIM21. GAPDH was used as an internal control. The PKR dimerization effected by TRIM21 was quantified by densiometric analysis which were normalized to the value of PKR dimerization in TRIM21 non-transfection. The data shown are mean ± SD. P values were determined by Student’s t-test. NS, no significance difference. (C) IP analysis of the interaction of PP1α and TRIM21 as well as PP1α and PKR in HEK293T cells by co-transfection with p3xFlag-CMV-PKR and pCMV-N-Myc-PP1α or p3xFlag-CMV-TRIM21 and pCMV-N-Myc-PP1α for 48 h. β-actin was used as an internal control. (D-E) Endogenous PP1α and TRIM21 interaction was analyzed by immunoprecipitation in A549 cells infected with VSV (MOI = 0.2) for 6 h (D) or stimulated with TG (10 μM or 20 μM) for 12 h (E). The amounts of PP1α bound TRIM21 were quantified by densiometric analysis which were normalized to the value in the control group. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, *P<0.01. The TRIM21 antibody was used to perform immunoprecipitation and the indicated proteins were PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 8 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection detected by the corresponding antibodies. β-actin was used as an internal control. (F) In vitro immunoprecipitation analysis of the interaction between GST-TRIM21 protein and Flag-PP1α protein purified from bacteria. Experiments were independently repeated two or three times with similar results. https://doi.org/10.1371/journal.ppat.1011443.g003 K6-linked ubiquitination of PP1α on Lys60. To confirm our conclusion, we constructed a PP1α mutant with replacement of the lysine residue with an arginine residue at K60 (PP1α- K60R). Consistently, TRIM21 no longer ubiquitinated PP1α by K60 mutation (Fig 4I). Collec- tively, these data demonstrated that TRIM21 ubiquitinates PP1α at Lys60. TRIM21 inhibits the activation of the PKR signaling pathway by catalyzing the K6-linked ubiquitination of PP1α To determine whether TRIM21 represses PKR activation via TRIM21-mediated K6-linked ubiquitination of PP1α, we knocked down the expression of TRIM21 in PP1α-silenced A549 cells. Consistent with our data, TRIM21 knockdown substantially enhanced PKR activation, while this phenomenon was lost by additional knockdown of PP1α (Fig 5A). Consequently, virus- or TG-induced global protein synthesis inhibition reversed by TRIM21 was also blocked by PP1α knockdown (Figs 5B and S2A). To confirm this conclusion, we constructed a mutant with phosphatase activity inactivation of PP1α (PP1α-H248K) and delivered it into PP1α- silenced A549 cells. Consistently, TRIM21 lost the ability to inactivate PKR by PP1α-H248K introduction upon VSV infection or TG stimulation (S2B and S2C Fig). These data demon- strated that TRIM21 inhibits PKR activation via PP1α. Next, we explored whether K6-linked ubiquitination of PP1α is essential for TRIM21-mediated PKR inactivation. We reintroduced wild-type PP1α (PP1α-WT) or K60-mutant PP1α (PP1α-K60R) together with pFlag-tagged TRIM21 into PP1α-silenced cells to detect the phosphorylation of PKR under viral infection or TG treatment. The phosphorylation of PKR was inhibited, and the inhibition of RNA trans- lation was reversed by TRIM21 in PP1α-WT-transfected cells; however, this phenomenon dis- appeared in PP1α-K60R-transfected cells after VSV infection or TG treatment (Fig 5C–5E), demonstrating that TRIM21-mediated ubiquitination of PP1α is crucial for PKR inactivation. Collectively, these data suggested that TRIM21 impairs PKR activation by promoting the K6-linked ubiquitination of PP1α. TRIM21-mediated K6-linked ubiquitination enhances the PKR-PP1α interaction To determine how TRIM21-mediated polyubiquitination of PP1α represses PKR activation, we first investigated whether TRIM21 affects the stability of the PP1α protein. TRIM21 did not alter the protein level of PP1α with or without VSV infection or TG treatment (S2D–S2F Fig), suggesting that TRIM21 does not affect the stability of the PP1α protein. Previous studies have demonstrated that K6-linked ubiquitination can affect protein–protein interactions [20,21], and therefore we speculated that TRIM21-mediated K6-linked ubiquitination may play a role in the PKR-PP1α interaction. To confirm our hypothesis, we performed a co-IP assay to exam- ine the effect of the PKR-PP1α interaction by TRIM21. Although TRIM21 had no effect on the interaction of PKR and PP1α in the resting state, TRIM21 knockdown significantly reduced their interaction upon VSV infection or TG stimulation (Fig 5F and 5G). These data indicated that TRIM21 can enhance the interaction between PKR and PP1α. Next, we examined whether the enhancement of the PKR-PP1α interaction is dependent on TRIM21-mediated K6-linked ubiquitination of PP1α. Immunoprecipitation showed that the interaction between PKR and PP1α is increased in TRIM21-WT-reconstituted sg-TRIM21 cells but not in TRIM21-C16S- PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 9 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 4. TRIM21 promotes the polyubiquitination of PP1α. (A) IP analysis of the ubiquitination of PP1α in HEK293T cells by co-transfecting p3xFlag-CMV-PP1α with pcDNA3.1a-vector or pcDNA3.1a-TRIM21-WT or pcDNA3.1a-TRIM21-C16S as well as HA-ub for 48 h. The anti- Flag antibody was used to perform immunoprecipitation and detect Flag-PP1α, the anti-V5 antibody was used to detect V5-TRIM21-WT/C16S and the anti-HA antibody was used to detect HA-ub. GAPDH was used as an internal control. (B) Ubiquitination analysis ubiquitination of PP1α. HEK293T cells were co-transfected with p3xFlag-CMV-PP1α and pcDNA3.1a-TRIM21 as well as the indicated HA-tagged ubiquitin mutants for 48 h. The anti-Flag antibody was used to perform immunoprecipitation and detect Flag-PP1α, the anti-V5 antibody was used to detect V5-TRIM21 and the anti-HA antibody was used to detect HA-ub. GAPDH was as an internal control. (C) Ubiquitination analysis the K6-linked ubiquitination of PP1α. HEK293T cells were co-transfected p3xFlag-CMV-PP1α with pcDNA3.1a-vector or pcDNA3.1a- TRIM21-WT or pcDNA3.1a-TRIM21-C16S as well as the indicated HA-tagged ubiquitin mutants for 48 h. Immunoprecipitation was performed with anti-Flag antibody. Ubiquitination was detected by anti-HA antibody and GAPDH was used as an internal control. (D) Ubiquitination analysis the K6-linked ubiquitination of PP1α under stress. HEK293T cells pre-infected with lentivirus-sh-vector or lentivirus-sh-TRIM21 for 12 h were co-transfected p3xFlag-CMV-PP1α with HA-ub-WT or HA-ub-K6O for 36 h, then infected with VSV (MOI = 0.2) for 6 h or treated with TG (10 μM) for 12 h. Immunoprecipitation was performed with anti-Flag antibody. Ubiquitination was detected by anti-HA antibody, TRIM21 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 10 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection was detected by anti-TRIM21 antibody and GAPDH was used an internal control. (E) A schematic diagram of PP1α truncations. (F) Ubiquitination of C-terminus and N-terminus of PP1α in HEK293T cells by co-transfecting p3xFlag-CMV-PP1α-C-EGFP or p3xFlag- CMV-PP1α-N-EGFP with pcDNA3.1a-vector or pcDNA3.1a-TRIM21-WT or pcDNA3.1a-TRIM21-C16S as well as HA-ub for 48 h. Immunoprecipitation was performed with anti-Flag antibody. Ubiquitination was detected by anti-HA antibody, V5 was detected by anti-V5 antibody and GAPDH was used an internal control. (G-H) TRIM21 promotes the polyubiquitination of PP1α on Lys60. Mutants of only one lysine residue retained within N-terminal PP1α (G). HEK293T cells were co-transfected pcDNA3.1a-vector or pcDNA3.1a-TRIM21 with or the mutants of p3xFlag-CMV-PP1α-N for 48 h. Ubiquitination and immunoblotting were performed with the antibodies as described in (F) (H). (I) TRIM21 promotes K6-linked ubiquitination of PP1α on Lys60. HEK293T cells were co-transfected p3xFlag-CMV-PP1α-WT or p3xFlag- CMV-PP1α-K60R with pcDNA3.1a-vector or pcDNA3.1a-TRIM21-WT or pcDNA3.1a-TRIM21-C16S as well as HA-ub-K6O for 48 h. Ubiquitination and immunoblotting assays were performed with the antibodies as described in (F). Experiments were independently repeated two or three times with similar results. https://doi.org/10.1371/journal.ppat.1011443.g004 reconstituted sg-TRIM21 cells upon VSV infection (Fig 5H). Moreover, the interaction between PP1α and PKR was abolished by PP1α-K60R (Fig 5I), suggesting that TRIM21-me- diated ubiquitination of PP1α is essential for the enhancement of the PKR-PP1α interaction. Collectively, extracellular stimuli, such as viral infection or TG treatment, promote the interac- tion between TRIM21 and PP1α, which results in K6-linked ubiquitination of PP1α. Ubiquiti- nated PP1α enhances its interaction with PKR, resulting in PKR dephosphorylation and PKR inactivation, which subsequently inhibits eIF2α activation, leading to the release of PKR acti- vation-mediated protein synthesis shutdown (Fig 5J). TRIM21 restricts viral infection by inhibiting PKR activation Several studies have demonstrated that PKR activation is essential for viral escape by inhibiting the RNA translation of antiviral effectors [10,22]. Given the negative regulatory role of TRIM21 in PKR activation, we speculated that TRIM21 has the ability to restrict viral infection by inhibiting PKR-mediated translational shutdown. To confirm our hypothesis, we con- structed IFNAR1-deficient cell lines in both A549 cells (A549-sg-IFNAR1) and HLCZ01 cells (HLCZ01-sg-IFNAR1), as TRIM21 is able to inhibit viral infection by promoting IFN produc- tion (S3A Fig). Notably, the levels of phosphorylated PKR and eIF2α were repressed by TRIM21 in the cells upon VSV infection (Fig 6A and 6B), and silencing TRIM21 increased viral replication and production in these cells (Fig 6C and 6D). The antiviral role of TRIM21 was also evidenced in A549-sg-IFNAR1 cells upon SeV infection (S3B Fig). To exclude the pos- sible role of the type III IFN-mediated antiviral response by viral infection, we knocked down STAT2 to abolish the type III IFN signaling pathway in A549-sg-IFNAR1 cells (S3C Fig). Nota- bly, TRIM21 deficiency still increased the replication of VSV in A549-sg-IFNAR1 cells with STAT2 knockdown (Fig 6E). Furthermore, we investigated IRF3-silenced cells, in which IFN production was abolished upon viral infection. Silencing TRIM21 augmented viral replication and PKR activation upon VSV infection in these cells (Figs 6F and 6G and S3D). Similarly, the replication of VSV and SeV (Figs 6H and S3E), as well as the phosphorylation of PKR and eIF2α, was also increased by TRIM21 deficiency (Fig 6I) in Huh7.5 cells, in which one of the main pattern recognition receptors of RNA viruses, RIG-I, is deficient. These data demon- strated that TRIM21 restricts viral infection in an IFN-independent manner. Next, we examined whether this IFN-independent antiviral role of TRIM21 restricts viral infection by inhibiting PKR activation. PKR/IFNAR1 double-deficient A549 cell lines were constructed (S3F Fig), and the antiviral function of TRIM21 was lost by PKR knockout (Fig 6J). Likewise, PKR-deficient Huh7.5 cells with TRIM21 knockdown showed similar replication of VSV and SeV compared to that of the control cells (Figs 6K, S3G and S3H). Moreover, TRIM21 also lost the ability to restrict viral infection by overexpression of TRIM21 in PKR/ TRIM21 double-deficient Huh7.5 cells (S3I and S3J Fig), suggesting that TRIM21 restricts viral infection via PKR. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 11 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 5. TRIM21 inhibits PKR activation by augmenting PKR-PP1α interaction. (A) A549 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus- sh- PP1α (sh-PP1α) for 12 h, and infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected by VSV (MOI = 0.2 or 0.5) for 6 h. The indicated proteins were detected by western blot by using the corresponding antibodies. GAPDH was used as an internal control. (B) Puromycin incorporation assays demonstrated the cellular protein synthesis in wild-type A549 cells (sh-vector) or PP1α-silenced A549 cells (sh-PP1α). Cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected by VSV (MOI = 0.2) for 6 h or stimulated with TG PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 12 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection (10 μM) for 12 h. The indicated proteins were detected by western blot by using the corresponding antibodies. GAPDH was used as a control. (C-D) PP1α- silenced A549 cells (sh-PP1α) were infected with lentivirus-Flag-PP1α-WT or lentivirus-Flag-PP1α-K60R for 48 h, then infected with VSV (MOI = 0.5) for 6 h (C) or stimulated with TG (10 μM) for 12 h (D). The indicated proteins were detected by western blot by using the corresponding antibodies. β-actin was used as an internal control. (E) Puromycin incorporation assays demonstrated the cellular protein synthesis in PP1α-silenced A549 cells (sh-PP1α). Cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 12 h, and infected with lentivirus-Flag-PP1α-WT or Lentivirus-Flag-PP1α-K60R for 48 h, then infected by VSV (MOI = 0.2) for 6 h or stimulated with TG (10 μM) for 12 h. GAPDH was used as a control. (F-G) TRIM21 blocks PKR-PP1α interaction. A549 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected by VSV (MOI = 0.5) for 6 h. The interaction of PP1α and PKR was analyzed by IP and western blot (F). The wild-type A549 cells (sg-vector) or TRIM21-deficient A549 cells (sg-TRIM21) were stimulated with TG (10 μM) for 12 h, The interaction of PP1α and PKR was analyzed by IP and western blot (G). GAPDH (F) or β-actin (G) was used as an internal control. (H) TRIM21-deficient A549 cells (sg-TRIM21) were infected with lentivirus-Flag-TRIM21-WT or lentivirus-Flag-TRIM21-C16S for 48 h, then infected by VSV (MOI = 0.5) for 6 h. The interaction of PP1α and PKR was analyzed by IP and western blot. β-actin was used as an internal control. (I) PP1α- silenced A549 cells (sh-PP1α) were infected with lentivirus-Flag-PP1α-WT or lentivirus-Flag-PP1α-K60R for 48 h. The interaction of PP1α and PKR was analyzed by IP and western blot. β-actin was used as an internal control. (J) Schematic model of TRIM21 regulating the translation initiation. Upon stress stimulation, TRIM21 augments the polyubiquitination of PP1α, enhancing the PKR-PP1α interaction, leading to PKR inactivation and subsequent release of protein synthesis inhibition. The relative ratios of p-PKR or p-eIF2α in (A, C and D), the relative ratios of puro signal in (B, E) and the amounts of PP1α bound PKR in (F-H) were quantified by densiometric analysis, which were normalized to the value in the control group. Experiments were independently repeated two or three times with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, P<0.01, p<0.001, NS, no significance difference. https://doi.org/10.1371/journal.ppat.1011443.g005 TRIM21 restricts viral infection by releasing PKR-mediated inhibition of the RNA translation of intrinsic antiviral genes A previous study reported that PKR-mediated inhibition of the RNA translation of IFN-stimu- lated genes (ISGs) contributes to restoring HCV replication by IFN treatment [22]. Thus, we examined whether TRIM21 can inhibit HCV replication by inhibiting PKR activation. There- fore, we performed an investigation in HCV-infected Huh7.5 cells treated with IFN. Similarly, in HCV-infected Huh7.5 cells, TRIM21 knockdown augmented IFN-mediated PKR activation and inhibited the RNA translation of ISGs, such as ISG15 and STAT2 (S4A Fig). Consistently, the levels of HCV RNA, NS3 protein and viral particles in the supernatant were upregulated by TRIM21 knockdown (S4B–S4D Fig). The replication of H77-S, HCV genotype 1a, was also inhibited by TRIM21 in response to IFN-α (S4E Fig). These data indicated that TRIM21 restricts HCV infection by reversing the PKR-mediated inhibition of ISG protein synthesis. Given its beneficial role in protein synthesis of ISGs, we speculated that the IFN-indepen- dent antiviral role of TRIM21 in IFNAR1-deficient cells is due to promote protein synthesis of the intrinsic antiviral genes in the host. To confirm our hypothesis, we first investigated whether TRIM21 affects global protein synthesis in IFNAR1-deficient cells by assessing puro- labeled nascent peptides. Consistent with the results above that TRIM21 inhibits PKR signal- ing pathway activation, VSV-induced nascent protein synthesis inhibition was reversed by overexpression of TRIM21 in IFNAR1/TRIM21 double-deficient cells (S4F–S4H Fig), indicat- ing that TRIM21 can release virus-induced protein synthesis inhibition in IFNAR1-deficient cells, which implies that TRIM21 may have a role in reversing the inhibition of RNA transla- tion of antiviral effectors. Thus, to identify the specific TRIM21-PKR-regulated antiviral effec- tors, we performed proteomic analysis in IFNAR1-deficient cells and IFNAR1/TRIM21 double-deficient cells upon VSV infection (Fig 7A). The proteomics assay showed that the pro- tein abundances of 53 genes with significant differences were downregulated in IFNAR1-defi- cient cells with TRIM21 depletion relative to those of IFNAR1-deficient cells (Fig 7A). Notably, these genes included some known antiviral effectors, such as proteasome 20S subunit beta 9 (PSMB9), nuclear factor kappa B subunit 2 (NFKB2), OUT deubiquitinase 4 (OTUD4), methyl-CpG binding protein 2 (MECP2), H2A.X variant histone (H2AX), nuclear receptor coactivator 7 (NCOA7), superoxide dismutase 2 (SOD-2), S100 calcium binding protein A2 (S100A2), raftlin, lipid raft linker 1 (RFTN1), S100 calcium binding protein A2 (S100A4), fibroblast growth factor 2 (FGF2), tolloid like 2 (TLL2), and TNF-induced protein 2 (TNFAIP2) (marked in yellow) [23–35], in addition to multiple genes with no previously PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 13 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 6. TRIM21 restricts viral infection by inhibiting the activation of PKR signaling pathway. (A-B) IFNAR1-deficient A549 cells (A) or IFNAR1- deficient HLCZ01 (B) cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. The indicated proteins were detected by western blot. GAPDH (A) or β-actin (B) was used as an internal control. (C) RT-qPCR analysis of the VSV RNA or the mRNA levels of TRIM21 in IFNAR1-deficient A549 cells (sg-IFNAR1) infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. (D) Plaque assay analysis of the VSV titers in IFNAR1- PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 14 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection deficient HLCZ01 cells infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.5) for 6 h. (E) IFNAR1/STAT2 double-deficient A549 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. RT-qPCR analysis of the VSV RNA or the mRNA level of TRIM21. (F) HLCZ01-sh-IRF3 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. The indicated proteins were detected by western blot. (G) RT-qPCR analysis of the VSV RNA or the mRNA level of TRIM21 in IRF3-silenced HLCZ01 cells (HLCZ01-sh- IRF3) infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.5) for 6 h. (H) Huh7.5 cells infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. RT- qPCR analysis of the level of VSV RNA or TRIM21 mRNA. (I) Huh7.5 cells infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh- TRIM21) for 48 h, then infected with VSV (MOI = 0.5) for 6 h. The indicated proteins were detected by western blot. (J) IFNAR1/PKR double-deficient A549 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2) for 6 h. The levels of VSV RNA were examined by RT-qPCR. (K) Wild-type (sg-vector) or PKR-deficient (sg-PKR) Huh7.5 cells were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2) for 6 h. The levels of VSV RNA were examined by RT- qPCR. Experiments were independently repeated two or three time with similar results. The relative ratios of p-PKR or p-eIF2α in (A, B, F and I) were quantified by densiometric analysis, which were normalized to the value in the control group with two or three independent repeats. The mRNA data in (C-E, G-H, J-K) from two or three independent experiments. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, p<0.01, p<0.001. NS, no significance difference. ND, not detected. https://doi.org/10.1371/journal.ppat.1011443.g006 recognized antiviral function (marked in red and green). Among the genes with unknown antiviral function, we focused on nine genes (marked in blue) that were significantly downre- gulated in TRIM21 knockout cells. A549 cells were infected with lentivirus expressing shRNAs targeting IgG Fc-binding protein (FCGBP), interleukin-1 receptor-associated kinase-like 2 (IRAK2), proteasome subunit beta type-9 (PSMB9), reactive oxygen species modulator 1 (ROMO1), latexin (LXN), Jade family PHD finger 1 (JADE1), ribonucleotide reductase regula- tory TP53 inducible subunit M2B (RRM2B), non-SMC condensin II complex subunit H2 (NCAPH2), or capping actin protein (CAPG) and then infected with VSV. Notably, knock- down of these genes enhanced the replication of VSV (Fig 7B). Similar results were obtained in VSV-infected hepatocytes and Huh7 cells (Fig 7C). Of note, these genes could constitutively restrict HCV replication in HCV-infected Huh7.5 cells (Fig 7D). However, FCGBP had no effect on the replication of VSV or HCV in hepatocytes, which demonstrated that the function of FCGBP in hepatocytes is distinctive. These data supported the hypothesis that reversing the PKR-mediated inhibition of antiviral gene protein synthesis by TRIM21 is sufficient for viral clearance. TRIM21 restricts viral infection through IRF1-dependent and IRF1-independent mechanisms It has been reported that basal expression of interferon regulatory factor 1 (IRF1) drives hepa- tocyte resistance to multiple RNA viruses by maintaining constitutive transcription of antiviral effectors, while the level of IRF1 protein is tightly regulated by PKR [10,26,36]. Given that TRIM21 suppresses PKR activation, we investigated whether TRIM21 can resist viral infection by promoting the protein synthesis of IRF1. Notably, silencing TRIM21 reduced IRF1 protein levels in IFNAR1-deficient HLCZ01 cells infected with VSV (Fig 7E), suggesting that TRIM21 can increase the abundance of IRF1 protein. Similarly, the levels of IRF1 protein were reduced in TRIM21-silenced Huh7.5 cells with VSV infection; however, the protein levels of IRF1 were restored by additional knockdown of PKR (Fig 7F), indicating that TRIM21 promotes the pro- tein synthesis of IRF1 by negatively regulating PKR activation. Next, we examined whether the antiviral role of TRIM21 is dependent on IRF1 in Huh7 cells. Compared with the results of sin- gle knockdown of the upregulated genes (Fig 7C), several genes, such as JADE1, PSMB9, LXN and ROMO1, lost their antiviral ability after IRF1 ablation (Fig 7G), indicating an IRF1-depen- dent antiviral role of TRIM21. However, IRF1 had no effect on the antiviral function of NCAPH2, ZNF574, RRM2B and IRAK2 (marked in red), indicating an IRF1-independent anti- viral role of TRIM21 (Fig 7G). These data indicated that some genes are regulated by both PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 15 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Fig 7. TRIM21 resists viral infection through inhibiting PKR activation-mediated global translation shutdown. (A) Proteome analysis of the reduced genes in IFNAR1-deficient A549 cells (sg-IFNAR1) and IFNAR1/TRIM21 double-deficient cells (sg-IFNAR1/ TRIM21). The cells were infected with VSV(MOI = 0.2) for 6 h. Known antiviral effectors were marked with yellow and the effectors with antiviral function verified in our study were marked with blue. (B-C) RT-qPCR analysis of VSV RNA or the mRNA levels of the indicated genes in A549 cells (B) or Huh7 cells (C) infected with lentivirus expressing shRNAs targeting the indicated genes for 48 h, then infected with VSV (MOI = 0.2) for 6 h. (D) Huh7.5 cells were infected with Lentivirus expressing shRNAs targeting the indicated genes for 12 h, then infected with HCV (MOI = 0.1) for 72 h. HCV RNA level was analyzed by RT-qPCR. (E) IFNAR1-deficient HLCZ01 cells (HLCZ01-sg-IFNAR1) were infected with lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. IRF1 protein was analyzed by western blot. β-actin was detected as control. The relative amounts of IRF1 proteins were quantified by densiometric analysis, which were normalized to the value in the control group. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05; *p<0.01. (F) Huh7.5 cells were infected with lentivirus- sh-vector (sh-vector) or lentivirus-sh-PKR (sh-PKR), or lentivirus-sh-vector (sh-vector) or lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. IRF1 protein was analyzed by western blot. β-actin was detected as control. The PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 16 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection relative amounts of IRF1 proteins were quantified by densiometric analysis, which were normalized to the value in the control group. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, NS, no significant difference. (G) Huh7 cells were infected with Lentivirus expressing shRNAs targeting the indicated genes for 48 h, then infected with VSV (MOI = 0.2) for 6 h. The levels of VSV RNA were examined by RT-qPCR. Experiments were independently repeated two or three times with similar results. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05; p<0.01, *p<0.001, NS, no significant difference. https://doi.org/10.1371/journal.ppat.1011443.g007 TRIM21 and IRF1, contributing to the IRF1-dependent antiviral response; however, some genes, such as NCAPH2, ZNF574, RRM2B and IRAK2, are regulated by TRIM21 rather than IRF1, causing an IRF1-independent antiviral response and even enhancing the antiviral effi- ciency of IRF1. Collectively, these data revealed that TRIM21 restricts viral infection in IRF1-- dependent and IRF1-independent manners. Discussion In response to stimulation, cellular protein homeostasis is tightly regulated by various strategies in all steps of RNA translation, including initiation, elongation, and termination, to make cells rapidly adapt to the changed extracellular or intracellular environment [2]. When translation is completed, nascent proteins are always modified in various manners, such as by ubiquitination by TRIM proteins, to confer the specific function of proteins or regulate protein fate. The crucial role of TRIM proteins in posttranslational modification has been widely investigated [14]; however, the role of TRIM proteins in regulating protein synthesis initiation is not clear. In this study, we found that a TRIM protein, TRIM21, positively regulates protein synthesis by inhibiting the PKR-eIF2α signaling pathway in a ubiquitination-dependent man- ner, providing new evidence for the role of ubiquitination in stress-controlled protein homeostasis. As the first step of protein synthesis, translation initiation is typically controlled by eIF2α, which is phosphorylated by four cellular kinases, PKR, GCN2, PERK and HRI, upon stress stimulation [3]. Among them, PKR is distinct for its activation by direct detection of the dsRNA upon viral infection. In mammalian cells, at least two signaling pathways are initiated upon detection of cytosolic dsRNA. One is the RLR signaling pathway, and the other is the PKR signaling pathway [23]. The TRIM family regulates RLR-mediated transcription of antivi- ral cytokines for a proper and efficient antiviral response has been widely studied and well clar- ified [37–39]. However, the role of the TRIM family in the regulation of antiviral effector translation remains poorly studied. In the present study, we demonstrate that the E3 ligase TRIM21 has the ability to promote protein synthesis of antiviral effectors by inactivating the PKR signaling pathway, which results in an antiviral response upon virus infection, which pro- vides new insight into the TRIM family in the regulation of antiviral invasion. TRIM21 is a member of the TRIM superfamily, which has been reported to be involved in diverse biological processes and implicated in various diseases [14]. Our previous study showed an important role of TRIM21 in the virus-triggered RLR signaling pathway [15]. Whether TRIM21 regulates the PKR signaling pathway remains elusive. In the present study, we found that TRIM21 negatively regulates PKR activation by promoting K6-linked ubiquiti- nation of the PKR phosphatase PP1α. Several lines of evidence strongly support our conclu- sion. First, knocking out TRIM21 augments stress-induced PKR activation and subsequent release of protein synthesis shutdown. Second, although TRIM21 has no effect on PKR ubiqui- tination, E3 ligase activity is essential for TRIM21-mediated inactivation of PKR, suggesting indirect regulation of PKR by TRIM21. Third, TRIM21 specifically abolishes PKR phosphory- lation but not dsRNA detection or dimerization of PKR, indicating that the phosphatases of PKR may be regulated by TRIM21. Fourth, PP1α, the phosphatase of PKR, is responsible for PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 17 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection TRIM21-mediated inhibition of PKR activation. TRIM21 promotes its interaction with PP1α and catalyzes the K6-linked ubiquitination of PP1α at Lys60 under stress. Fifth, silencing PP1α or K60 mutation of PP1α abolishes the inhibition of PKR phosphorylation by TRIM21. Collec- tively, our findings demonstrate that the TRIM21-PP1α axis is a newly discovered program for regulating stress-induced PKR-mediated protein synthesis. Moreover, we found an IFN-inde- pendent antiviral function of TRIM21. By performing proteomics analysis, several previously known and unknown antiviral effectors regulated by the TRIM21-PKR axis have been identi- fied and proven to protect against viral infection, which broadens our understanding of antivi- ral genes and strengthens the evidence that the antiviral function of TRIM21 is achieved by reversing PKR-mediated translational shutdown. As one of the classic posttranslational modifications (PTMs), ubiquitination profoundly affects fundamental physiological processes, such as cell proliferation, differentiation, cell death, and protein stability and structure, in all species [39,40]. E3 ligases are vital components in this process, directly interacting with and catalyzing different ubiquitin linkages of their spe- cific substrates, which can occur through K6-, K11-, K27-, K29-, K33-, K48- and K63-linked ubiquitination [41]. Several types of ubiquitin linkages catalyzed by TRIM21 have been reported, and different ubiquitin linkages play distinct roles in host defense against pathogenic invasion [42]. For instance, TRIM21-mediated K48-linked ubiquitination is tightly related to proteasome-dependent degradation, which results in the degradation of viral proteins in virus-infected cells [43–45]. However, K48-linked ubiquitination of DDX41, one of the sensors detecting dsDNA, results in the inhibition of the innate immune response triggered by DNA virus [46]. K63-linked ubiquitination catalyzed by TRIM21 augments the activation of the intracellular antibody-triggered innate immune response [47]. K27-linked ubiquitination pro- motes RNA virus-induced MAVS activation and the innate immune response to viral infection [15]. Importantly, in this study, we find that TRIM21-catalyzed K6-linked ubiquitination of PP1α restricts viral infection by promoting the protein synthesis of intrinsic antiviral effectors. Thus, our findings indicate a newly discovered type of ubiquitin linkage mediated by TRIM21 that can participate in the host antiviral response with a distinct mechanism. PKR activation suppresses the synthesis of interferon-stimulated genes (ISGs) upon viral infection [10,22]. Emerging evidence demonstrate the negative role of PKR-triggered protein inhibition in the host defense against viral infection. For example, one study reported that NLRX1 is a positive regulator of the antiviral response by limiting PKR-mediated translational shutoff of the IRF1 protein, which results in inhibition of the production of IRF1-dependent antiviral factors [48]. Similarly, our data show an IRF1-dependent antiviral efficiency regulated by TRIM21. TRIM21 promotes IRF1 protein synthesis by inhibiting PKR activation upon viral infection. In addition to TRIM21’s positive role in innate immunity, we define the antiviral role of TRIM21 as both augmenting the transcription of IFN production and facilitating trans- lation of the IRF1 protein. In addition to its role in the innate antiviral response, IRF1 also plays a role in adaptive immunity and tumor suppression. Given that, we speculate that TRIM21 may have a role in restricting not only viral infection but also tumorigenesis via adap- tive immunity, which needs further study. In addition, we also found an IRF1-independent antiviral function mediated by TRIM21 by a proteomics assay. Some previously unknown antiviral factors were discovered, all of which are directly regulated by TRIM21. Together with these findings, we demonstrate that an IFN-independent antiviral role of TRIM21 restricts viral infection by reversing PKR-mediated RNA translation inhibition of antiviral factors. Col- lectively, the results of our study highlight a newly discovered biological role of TRIM21, and these data provide new evidence of PKR-mediated translational arrest in host resistance to virus infection. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 18 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Materials and methods Cell culture and reagents The HLCZ01 cell line, a hepatoma cell line supporting the entire life cycle of HCV and HBV, was previously established in our laboratory. HEK293T cells were purchased from Boster, Huh7.5 cells were kindly provided by Charles M. Rice (Rockefeller University, New York), and Huh7 and A549 cells were obtained from the American Type Culture Collection. HLCZ01 cells were cultured in collagen-coated tissue culture plates containing Dulbecco’s modified Eagle medium (DMEM)–F-12 medium supplemented with 10% (vol/vol) fetal bovine serum (FBS) (Gibco), 40 ng/ml dexamethasone (Sigma), insulin-transferrin-selenium (ITS) (Lonza), and penicillin–streptomycin (Thermo Fisher Scientific). Other cells were propagated in DMEM supplemented with 10% FBS, nonessential amino acid solution (Thermo Fisher Scien- tific), and penicillin–streptomycin. Cell transfection with plasmids was conducted using Via- Fect Transfection Reagent (Promega) or Lipofectamine 2000 (Thermo Fisher Scientific) in Opti-MEM medium (Thermo Fisher Scientific). For stable transfection or infection, monoclo- nal cells were screened using the antibiotic Puro (Gibco). Virus The pJFH1 plasmid was a gift from Takaji Wakita (National Institute of Infectious Diseases, Tokyo, Japan). The linearized DNA from the pJFH1 plasmid was purified and used as the tem- plate for in vitro transcription with a MEGAscript kit (Ambion, Austin, TX). In vitro-tran- scribed genomic JFH1 RNA was delivered into Huh7.5 cells by electroporation. The transfected cells were cultured for the indicated periods. The cells were passaged every 3 to 5 days, while the corresponding supernatants were collected and filtered with a 0.45 μm filter device. The viral titers are presented as focus-forming units (FFUs) per milliliter, determined as the average number of NS5A-positive foci detected in Huh7.5 cells. VSV were kindly shared by Jianguo Wu (Jinan University, Guangzhhou, China), and SeV was kindly shared by Xingyi Ge (Hunan University, Changsha, China). Antibodies Antibodies against the following proteins were obtained from commercial sources and used for immunoblots: anti-flag-tag (F3165, Sigma), anti-V5-tag (R960-25, Thermo Fisher Scien- tific), anti-puromycin (MABE343, Merck Millipore), anti-GAPDH (MAB374, Merck Milli- pore), anti-phospho-eIF2α Ser51 (3398, CST), anti-TRIM21 (92043, CST), anti-PKR (12297, CST), anti-Myc tag (2276, CST), anti-HA tag (ab236632, Abcam), anti-TRIM21 (ab91423, Abcam), anti-phospho-PKR T451 (ab81303, Abcam), and anti-IRF1 (8478, CST). Anti-inter- feron alpha/beta receptor 1 (ab45172, Abcam), anti-STAT2 (72604, CST), anti-PP1α (ab137512, Abcam), anti-ISG15 (ab285367, Abcam), anti-GST mouse monoclonal antibody (HT601-01, Transgen), β-actin (A5541, Sigma), and goat anti-mouse IgG (HRP-linked) (AP124P, Merck Millipore) were also used. The following antibodies were obtained from com- mercial sources and used for immunofluorescence staining: anti-puromycin (MABE343, Merck Millipore) and donkey anti-rabbit IgG (H + L) highly cross-adsorbed secondary anti- body conjugated with Alexa Fluor 594 (A-21207, Thermo Fisher Scientific). The mouse mono- clonal anti-HCV core antibody was a gift from Chen Liu. Plasmid construction TRIM21, PP1α, and PKR cDNAs were synthesized from total cellular RNA isolated from HLCZ01 cells by standard reverse transcription-PCR (RT–PCR). Subsequently, they were PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 19 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Table 1. Primers for the construction of plasmids. GST-TRIM21 (F) GST-TRIM21 (R) PP1α (F) PP1α-C (F) PP1α (R) PP1α-N (R) GST-PP1α (F) GST-PP1α (R) PKR (F) PKR (R) Myc- PKR (F) Myc- PKR (R) eIF2α (F) eIF2α (R) https://doi.org/10.1371/journal.ppat.1011443.t001 5’- CGGAATTCCCATGGCTTCAGCAGCACGCTTG-3’ 5’- ATAAGAATGCGGCCGCTCAATAGTCAGTGGATCC-3’ 5’-GGGGTACCATGTCCGACAGCGAGAAGCTC-3’ 5’-GGGGTACCATGAGGGGCAAGCAGTCCTTG-3’ 5’-GCTCTAGATTTCTTGGCTTTGGCGGAATT-3’ 5’-GCTCTAGAGTCCACATAGTCCCCCAGAAA-3’ 5’-CGGAATTCCCATGTCCGACAGCGAGAAGCTC-3’ 5’- ATAAGAATGCGGCCGCTCAATAGTCAGTGGATCC-3’ 5’-GGGGTACCATGGCTGGTGATCTTTCA-3’ 5’-GCTCTAGAACATGTGTGTCGTTAATTC-3’ 5’-CCCAAGCTTATGGCTGGTGATCTTTCA-3’ 5’-GGGGTACCACATGTGTGTCGTTCATT-3’ 5’-GGGGTACCATGCCGGGTCTAAGTTGTAGA-3’ 5’-GCTCTAGAATCTTCAGCTTTGGCTTCCAT-3’ cloned into the pcDNA3.1a vector, p3FLAG-CMV vector pCMV-N-Myc or pGEX4T2 vector. Multiple domains of TRIM21 and PP1α were amplified from the templates of full-length TRIM21 and PP1α, which were then cloned into p3xFLAG-CMV. The primers for amplifying these genes are listed in Table 1. The pHA-ub (K6, K11, K27, K29 and K33) plasmids were kindly shared by Hongbing Shu (Wuhan University). The pHA-ub (WT, K48 and K63) plas- mids were kindly provided by Zhengfan Jiang (Peking University). Lentivirus production and generation of stable cell lines HEK293T cells plated in 10 cm dishes were transfected with 8 μg packaging plasmid psPAX2, 2.7 μg envelope plasmid pMD2G and 8 μg target plasmid encoding shRNA or lentiCRISPRv2 encoding sgRNA using Lipofectamine 2000 (Thermo Fisher Scientific). Lentivirus superna- tants were collected at 36 h, 48 h, 56 h and 72 h post-transfection and clarified by filtration through 0.45 μm syringe filters. HLCZ01, A549 or Huh7.5 cells were plated in 6-well plates prior to being transduced with 500 μL per well of lentivirus supernatant. Transduced cells were selected for by the addition of 2 μg/mL puromycin at 72 h post-transduction. Loss of tar- get protein expression was confirmed by Western blot. The target sequences used for shRNA gene silencing and lentiCRISPRv2 encoding sgRNA plasmids are listed in Table 2. Real-time PCR assay Total cellular RNA was extracted by TRIzol reagent (Invitrogen) according to the manufactur- er’s protocol. The Superscript III first-strand synthesis kit for reverse transcription of RNA was purchased from Invitrogen. After DNase (Promega) treatment, the extracted RNA was used as the template for reverse transcription-PCR. Real-time PCR was performed as described previously [15]. GAPDH was used as the internal control. The primers used for real-time PCR are listed in Table 3. Western blotting Cells were washed with PBS and lysed with RIPA buffer (Thermo Fisher Scientific) supple- mented with protease inhibitor cocktail (Thermo Fisher Scientific). The lysates were incubated on ice for 30 min and centrifuged at 16000 g for 15 min at 4˚C. Proteins were resolved on SDS–PAGE gels and transferred to polyvinylidene difluoride (PVDF) membranes (Merck PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 20 / 28 PLOS PATHOGENSTable 2. Short hairpin RNAs used for knockdowns and single guide RNAs used for CRISPR/Cas9. Regulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Target gene sh-FCGBP sh-IRAK2 sh-PSMB9 sh-ROMO1 sh-LXN sh-JADE1 sh-RRM2B sh-CAPG sh-NCAPH2 sh-TRIM21 sh-PP1α sh-IRF3 sh-STAT2 sh-IFR1 sh-PKR sg-TRIM21 sg-IFNAR1 sg-PKR https://doi.org/10.1371/journal.ppat.1011443.t002 Target sequence 5’-GGGCTGTGTGGCAACTATAAT-3’ 5’-CCCACTTCGTCTGATTCAAAG-3’ 5’-CATCTACCTGGTCACTATTAC-3’ 5’-ATGGGCTTCGTGATGGGTTGC-3’ 5’-CCAGAAGTCAACTTCACATTT-3’ 5’-GCCTGAGGAAGTAGTGGATTT-3’ 5’-GCGATGGATAGCAGATAGAAA-3’ 5’-GCTGATATCTGATGACTGCTT-3’ 5’-TACAGTAAGAAGGTGGAATAC-3’ 5’-GGCATGGTCTCCTTCTACAAC-3’ 5’-GTGCAAGAGACGCTACAAC-3’ 5’-GCCAACCTGGAAGAGGAAT-3’ 5’-TGTCTTCTGCTTCCGATATAA-3’ 5’-CGTGTGGATCTTGCCACAT-3’ 5’-GCTGAACTTCTTCATGTATGT-3’ 5’-GGAGCCTGTGAGCATCGAGTG-3’ 5’-GCTGCGGACAACACCCA-3’ 5’-GATGGAAGAGAATTTCCAGA-3’ Millipore). The PVDF membranes were then blocked with 5% skim milk and sequentially incubated with primary and secondary antibodies. The bound antibodies were detected using SuperSignal West Pico chemiluminescent substrate (Pierce, Rockford, IL). IP and immunoblotting Cells were washed with PBS and lysed with IP Lysis Buffer (Thermo Fisher Scientific) supple- mented with protease inhibitor cocktail. The lysates were incubated on ice for 30 min and cen- trifuged at 16000 g for 15 min at 4˚C. The lysates were diluted to a concentration of 2 μg/μL with PBS before IP. The lysates (200 μg) were immunoprecipitated with the indicated antibod- ies. The immunocomplex was captured by adding Protein G Agarose (Merck Millipore, Darm- stadt, Germany). The protein binding to the beads was boiled in 2 x Laemmli sample buffer (Bio–Rad, Hercules) and was then subjected to SDS–PAGE. RNA immunoprecipitation The cells were trypsinized to detach them, and the supernatant was discarded. The cells were washed by gently resuspending them in 1 ml PBS and pelleted by centrifugation at 3,000 g for 1 min. Then, the cells were resuspended in 1 mL 1% formaldehyde (diluted in PBS) and allowed to stand for 10 min at room temperature. The cells were pelleted by centrifugation at 3,000 g for 1 min, resuspended in 1 mL 0.25 M glycine solution (diluted in PBS) and again allowed to stand for 10 min at room temperature. The cells were pelleted and washed with 500 μL PBS, and then the cells were lysed with RIPA buffer supplemented with a protease inhibitor cocktail and an RNase inhibitor on ice for 30 min. The subsequent immunoprecipita- tion was processed as described above. Protein–RNA complexes binding to beads were eluted in PBS at 70˚C for 45 min. The eluted material was lysed in ice-cold TRIzol reagent for RT– PCR. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 21 / 28 PLOS PATHOGENSTable 3. Primers for Real-time PCR. Regulation of PKR-dependent protein synthesis by TRIM21 upon virus infection GAPDH (F) GAPDH (R) VSV (F) VSV (R) SeV (F) SeV (R) HCV (F) HCV (R) H77-S (F) H77-S (R) TRIM21 (F) TRIM21 (R) PSMB9 (F) PSMB9 (R) IRAK2 (F) IRAK2 (R) FCGBP (F) FCGBP (R) ROMO1 (F) ROMO1 (R) LXN (F) LXN (R) JADE1 (F) JADE1 (R) RRM2B (F) RRM2B (R) CAPG (F) CAPG (R) NCAPH2 (F) NCAPH2 (R) https://doi.org/10.1371/journal.ppat.1011443.t003 5’-AATGGGCAGCCGTTAGGAAA-3’ 5’-GCGCCCAATACGACCAAATC-3’ 5’-CAAGTCAAAATGCCCAAGAGTCACA—3’ 5’-TTTCCTTGCATTGTTCTACAGATGG-3’ 5’-TGTTATCGGATTCCTCGACGCAGTC-3 5’-TACTCTCCTCACCTGATCGATTATC-3 5’-TCTGCGGAACCGGTGAGTA-3 5’- TCAGGCAGTACCACAAGGC-3 5’-TCACTGCTTATGCCCAGCAA-3 5’-GCTCCGTGGTAAACAGTCCA-3 5’-CCCCTCTAACCCTCTGTCCA-3’ 5’-GGGGAAAAGAGGCAGGGTTT-3’ 5’-GTGGATGCAGCATATAAGCC-3’ 5’-AGTGACCAGGTAGATGACAC-3’ 5’-CGCGTATCTGCCAGAGGATT-3’ 5’-AACCGGGCTTCGGTTGTTAT-3’ 5’-GTGTCTGCATCCCTGTCCAA-3’ 5’-GACAGGACACAGAGACCACG-3’ 5’-TTCGACCGTGTCAAAATGGG-3’ 5’-GCCACTCTGCATCATGGTTT-3’ 5’-AAACAAGCCAGCATGGAGGATA-3’ 5’-TCAGCTGTGCAGTTCACCTT-3’ 5’-TGGGTTCTCGATCTGTAGCG-3’ 5’-ACAGCAGGCAGCTGATCCAA-3’ 5’-ATGTTATTCGCCGCGGTCAG-3’ 5’-TGAAGATGATCTCCCGGCCT-3’ 5’-GGAAGGTGGTGTGGAGTCAG-3’ 5’-ACCAGGCGAAGATGTTCTGG-3’ 5’-GGGGGCAGCAGATGACTTT-3’ 5’-TCCTCGTAGCTCAGGGACAT-3’ Immunofluorescence staining Cells were seeded into a confocal dish and fixed with 4% paraformaldehyde for 15 min at room temperature. The cells were washed with PBS, permeabilized for 15 min with 0.2% Tri- ton X-100 in PBS for 10 min, blocked with goat serum (Thermo Fisher Scientific) for 30 min at room temperature, and sequentially incubated with primary and fluorescence-labeled sec- ondary antibodies (Invitrogen) (diluted in PBS to 1:500) at room temperature for 2 h. Nuclei were counterstained with DAPI (Vector Laboratories, Burlingame) for 5 min. Images were captured using a TI-E + A1 SI confocal microscope (Nikon). Plaque assay The supernatants were diluted into multiple concentrate gradients with medium without FBS and transduced into VERO cells. After 1 h, the cells were placed in complete medium. Two hours later, 1% agarose media was added on top of the virus-infected Vero cells. After 1 day, cells with plaques were fixed with 4% paraformaldehyde for 30 min at room temperature and stained with 0.1% crystal violet. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 22 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Puromycin incorporation assay Cells were pulsed with puromycin (10 μg/mL) for 1 h and then fixed with 4% paraformalde- hyde, followed by staining with anti-puromycin and DAPI. Cells were imaged by confocal microscopy. Alternatively, cells were pulsed with puromycin (10 μg/mL) for 1 h, lysed, and puromycin-labeled proteins were identified by immunoblot analysis. Statistical analysis All results are presented as the means and standard deviations (SDs). Comparisons between two groups were performed by using Student’s t test. Supporting information S1 Fig. TRIM21 inhibits the activation of PKR. (A) Immunoblot analysis of TRIM21 in wild-type A549 cells (sg-vector) and TRIM21-deficient A549 cells (sg-TRIM21). GAPDH was detected as an internal control. (B) TRIM21 was assessed by immunoblot in A549 cells infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 72 h. GAPDH was detected as an internal control. (C) A549 cells were infected with Lentivirus-sh- vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then transfected with poly (I: C) (1 μg or 2 μg) for 12 h. The indicated proteins were detected by western blot. β-actin was used as an internal control. (D-E) A549 cells were infected with Lentivirus-sh-vector (sh-vec- tor) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected by VSV (MOI = 0.2 or 0.5) for 6 h (D) or SeV (MOI = 0.2 or 0.5) for 12 h (E). The indicated proteins were detected by western blot. β-actin was used as an internal control. (F) HLCZ01 cells were infected with Len- tivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then transfected with poly (I: C) (1 μg or 2 μg) for 12 h. The indicated proteins were detected by western blot. β-actin was used as an internal control. (G-H) HLCZ01 cells were infected with Lentivirus-sh- vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected by VSV (MOI = 0.2 or 0.5) for 6 h (G) or SeV (MOI = 0.2 or 0.5) for 12 h (H). The indicated proteins were detected by western blot. β-actin was used as an internal control. (I) Immunoblot analysis of PKR in wild-type A549 cells (sg-vector) and PKR-deficient A549 cells (sg-PKR). GAPDH was detected as an internal control. The relative ratios of p-PKR or p-eIF2α in (C-H) were quantified by densiometric analysis, which were normalized to the value in the control group. Experiments were independently repeated two or three time with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, p<0.01, p<0.001. (TIF) S2 Fig. TRIM21 inhibits PKR signaling pathway activation via PP1α. (A) Immunoblot anal- ysis of PP1α protein in wild-type A549 cells (sh-vector) or PP1α-silenced A549 cells (sh- PP1α). β-actin was detected as an internal control. (B-C) PP1α-silenced A549 cells (sh-PP1α) pre-infected with Lentivirus-Flag-PP1α-WT or Lentivirus-Flag-PP1α-H248K for 24 h were infected with Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h (B) or stimulated with TG (10 μM) for 12 h (C). The indicated pro- teins were detected by western blot and β-actin was used as an internal control. (D-E) A549 cells were infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh- TRIM21) for 48 h, then infected with VSV (MOI = 0.2) for indicated times (D), or VSV (MOI = 0.2 or 0.5) for 6 h (E). The indicated proteins were analyzed by western blot. GAPDH was used as an internal control. (F) A549 cells were infected with Lentivirus-sh-vector (sh-vec- tor) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then stimulated with TG (10 μM or PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 23 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection 20 μM) for 12 h. The indicated proteins were analyzed by western blot. GAPDH was used as an internal control. The relative ratios of p-PKR in (B-C) were quantified by densiometric analysis, which were normalized to the value in the control group. Experiments were indepen- dently repeated two or three time with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, *p<0.01, NS, no significance difference. (TIF) S3 Fig. TRIM21 restricts viral infection by inhibiting PKR signaling pathway activation. (A) Immunoblot analysis of IFNAR1 in wild-type A549 cells or HLCZ01 cells (sg-vector) or IFNAR1-deficient A549 or HLCZ01 cells (sg-IFNAR1). GAPDH was detected as an internal control. (B) IFNAR1-deficient A549 cells (sg-IFNAR1) were infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with SeV (MOI = 0.5) for 12 h. The RNA level of SeV was analyzed by RT- qPCR. (C) Immunoblot anal- ysis of the STAT2 protein in STAT2-wild-type A549-sg-IFNAR1 cells (sh-vector) or STAT2-si- lenced A549-sg-IFNAR1 cells (sh-STAT2). β-actin was detected as an internal control. (D) Immunoblot analysis of IRF3 protein in wild-type HLCZ01 cells (sh-vector) and IRF3-silenced HLCZ01 cells (sh-IRF3). β-actin was used as an internal control. (E) Huh7.5 cells were infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with SeV (MOI = 0.5) for 12 h. The RNA level of SeV was analyzed by RT-qPCR. (F) Immunoblot analysis of the protein levels of PKR and IFNAR1 in wild-type A549 cells (sg-vec- tor) and PKR/IFNAR1 double-deficient A549 cells (sg-PKR/IFNAR1). β-actin was used as an internal control. (G) Immunoblot analysis of the expression of PKR in wild-type Huh7.5 cells (sg-vector) and PKR-deficient Huh7.5 cells (sg-PKR). GAPDH was used as an internal control. (H) Wild-type (sg-vector) or PKR-deficient Huh7.5 cells (sg-PKR) were infected with Lentivi- rus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, then infected with SeV (MOI = 0.2) for 12 h. The RNA level of SeV was analyzed by RT-qPCR. (I) Immunoblot analysis of the expression of PKR and TRIM21 in wild-type Huh7.5 cells (sg-vector) and PKR/ TRIM21 double-deficient Huh7.5 cells (sg-PKR/TRIM21). β-actin was used as an internal con- trol. (J) PKR/TRIM21 double-deficient Huh7.5 cells were infected with Lentivirus-Flag-vector or Lentivirus-Flag-TRIM21 for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h, or SeV (MOI = 0.2 or 0.5) for 12 h. The viral RNA levels were examined by RT-qPCR. Experiments were independently repeated two or three time with similar results, and the data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, p<0.01, p<0.001, NS, no significance difference. ND, not detected. (TIF) S4 Fig. TRIM21 restricts HCV replication in response to IFN. (A-D) Huh7.5 cells per- infected with HCV (JFH-1 strain) (MOI = 0.1) for 24 h were infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, followed by IFN-α (100 U/ mL or 500 U/mL) treatment for 24 h. The indicated proteins were examined by western blot. β-actin was used as an internal control (A). The levels of HCV RNA were analyzed by RT-qPCR (B). NS3 protein was analyzed by western blot (C). The levels of HCV particles in the supernatant were analyzed by immunofluorescence (D). The relative ratios of p-PKR or p-eIF2α in (A) and the relative amounts of HCV NS3 proteins in (C) were quantified by densiometric analysis, which were normalized to the value in the control group with two or three repeats. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05, P<0.01, P<0.001. (E) Huh7.5 cells pre-infected with HCV (H77-S strain) (MOI = 0.1) for 24 h were infected with Lentivirus-sh-vector (sh-vector) or Lentivirus-sh-TRIM21 (sh-TRIM21) for 48 h, followed by IFN-α (500 U/mL) treatment for 24 h. Viral RNA was quantified by real-time PCR. (F) Immunoblot analysis of the protein levels of TRIM21 or IFNAR1 in wild-type A549 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 24 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection cells, TRIM21-deficient A549 cells, IFNAR1-deficient A549 cells or IFNAR1/TRIM21 double- deficient A549 cells. GAPDH was used as an internal control. (G) IFNAR1/TRIM21 double- deficient A549 cells (sg-IFNAR1/TRIM21) were infected with Lentivirus-Flag-vector or Lenti- virus-Flag-TRIM21 for 48 h, then infected with VSV (MOI = 0.2 or 0.5) for 6 h. Puromycin incorporation assays of the cellular protein synthesis. GAPDH was used as an internal control. The relative ratios of puro signals were quantified by densiometric analysis, which were nor- malized to the value in the control group with two or three repeats. The data shown are mean ± SD. P values were determined by Student’s t-test. p<0.05. (H) IFNAR1/TRIM21 dou- ble-deficient A549 cells were infected with Lentivirus-Flag-vector or Lentivirus-Flag-TRIM21 for 48 h, then infected with VSV (MOI = 0.2) for 6 h. Cells were pulse-labeled with puromycin (10 μg/mL) for 1 h prior to fixation. Immunofluorescence staining for puromycin in the treated cells. The nuclei were stained with DAPI. Experiments were independently repeated two or three time with similar results. Student’s two-sided t test, and the data are represented as mean ± SD. p<0.05 versus the control, p<0.01 versus the control, **p<0.001 versus the control. (TIF) S1 File. Raw data for proteome analysis. (XLSX) S2 File. Raw data for western blot. (PDF) Acknowledgments We thank Charles M. Rice for the Huh7.5 cells, Takaji Wakita (National Institute of Infectious Diseases, Tokyo, Japan) for pJFH1 plasmid, Stanley Lemon (University of North Carolina, Chapel Hill, NC) for pH 77-S plasmid, Jianguo Wu (Jinan University, Guangzhhou, China) for VSV, and Xingyi Ge (Hunan University, Changsha, China) for SeV. We also thank Chen Liu (Yale University, New Haven, USA), Hongbing Shu (Wuhan University, Wuhan, China) and Zhengfan Jiang (Peking University, Beijing, China) for kindly sharing research materials. Author Contributions Conceptualization: Huiyi Li, Binbin Xue, Haizhen Zhu. Data curation: Huiyi Li, Binbin Xue. Formal analysis: Huiyi Li, Rilin Deng, Jingjing Wang, Binbin Xue. Funding acquisition: Huiyi Li, Binbin Xue, Haizhen Zhu. Investigation: Huiyi Li, Shun Liu, Qing Feng, Jingjing Wang, Xinran Li, Mengyu Wan. Methodology: Rilin Deng, Yousong Peng, Songqing Tang, Binbin Xue. Project administration: Binbin Xue, Haizhen Zhu. Resources: Huiyi Li, Shun Liu, Qing Feng, Xintao Wang. Supervision: Binbin Xue, Haizhen Zhu. Validation: Huiyi Li, Renyun Tian, Yan Xu, Shengwen Chen, Qian Liu, Luoling Wang. Visualization: Huiyi Li, Binbin Xue. Writing – original draft: Huiyi Li, Binbin Xue. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 25 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection Writing – review & editing: Huiyi Li, Binbin Xue, Haizhen Zhu. References 1. Wek RC. Role of eIF2alpha Kinases in Translational Control and Adaptation to Cellular Stress. Cold Spring Harb Perspect Biol. 2018; 10(7). Epub 2018/02/15. https://doi.org/10.1101/cshperspect. a032870 PMID: 29440070; PubMed Central PMCID: PMC6028073. 2. Talloczy Z, Jiang W, Virgin HWt, Leib DA, Scheuner D, Kaufman RJ, et al. Regulation of starvation- and virus-induced autophagy by the eIF2alpha kinase signaling pathway. Proc Natl Acad Sci U S A. 2002; 99(1):190–5. Epub 2002/01/05. https://doi.org/10.1073/pnas.012485299 PMID: 11756670; PubMed Central PMCID: PMC117537. 3. Pakos-Zebrucka K, Koryga I, Mnich K, Ljujic M, Samali A, Gorman AM. The integrated stress response. EMBO Rep. 2016; 17(10):1374–95. Epub 2016/09/16. https://doi.org/10.15252/embr.201642195 PMID: 27629041; PubMed Central PMCID: PMC5048378. 4. Hotamisligil GS. Endoplasmic reticulum stress and the inflammatory basis of metabolic disease. Cell. 2010; 140(6):900–17. Epub 2010/03/23. https://doi.org/10.1016/j.cell.2010.02.034 PMID: 20303879; PubMed Central PMCID: PMC2887297. 5. Chen YG, Hur S. Cellular origins of dsRNA, their recognition and consequences. Nat Rev Mol Cell Biol. 2022; 23(4):286–301. Epub 2021/11/25. https://doi.org/10.1038/s41580-021-00430-1 PMID: 34815573; PubMed Central PMCID: PMC8969093. 6. Dauber B, Wolff T. Activation of the Antiviral Kinase PKR and Viral Countermeasures. Viruses. 2009; 1 (3):523–44. Epub 2009/12/01. https://doi.org/10.3390/v1030523 PMID: 21994559; PubMed Central PMCID: PMC3185532. 7. Dalet A, Gatti E, Pierre P. Integration of PKR-dependent translation inhibition with innate immunity is required for a coordinated anti-viral response. FEBS Lett. 2015; 589(14):1539–45. Epub 2015/05/17. https://doi.org/10.1016/j.febslet.2015.05.006 PMID: 25979169. 8. Qiao H, Jiang T, Mu P, Chen X, Wen X, Hu Z, et al. Cell fate determined by the activation balance between PKR and SPHK1. Cell Death Differ. 2021; 28(1):401–18. Epub 2020/08/18. https://doi.org/10. 1038/s41418-020-00608-8 PMID: 32801355; PubMed Central PMCID: PMC7852545. 9. Hsu LC, Park JM, Zhang K, Luo JL, Maeda S, Kaufman RJ, et al. The protein kinase PKR is required for macrophage apoptosis after activation of Toll-like receptor 4. Nature. 2004; 428(6980):341–5. Epub 2004/03/19. https://doi.org/10.1038/nature02405 PMID: 15029200. 10. 11. Feng H, Lenarcic EM, Yamane D, Wauthier E, Mo J, Guo H, et al. NLRX1 promotes immediate IRF1- directed antiviral responses by limiting dsRNA-activated translational inhibition mediated by PKR. Nat Immunol. 2017; 18(12):1299–309. Epub 2017/10/03. https://doi.org/10.1038/ni.3853 PMID: 28967880; PubMed Central PMCID: PMC5690873. Lu B, Nakamura T, Inouye K, Li J, Tang Y, Lundback P, et al. Novel role of PKR in inflammasome acti- vation and HMGB1 release. Nature. 2012; 488(7413):670–4. Epub 2012/07/18. https://doi.org/10.1038/ nature11290 PMID: 22801494; PubMed Central PMCID: PMC4163918. 12. Wu J, Chen ZJ. Innate immune sensing and signaling of cytosolic nucleic acids. Annu Rev Immunol. 2014; 32:461–88. Epub 2014/03/25. https://doi.org/10.1146/annurev-immunol-032713-120156 PMID: 24655297. 13. Liu S, Cai X, Wu J, Cong Q, Chen X, Li T, et al. Phosphorylation of innate immune adaptor proteins MAVS, STING, and TRIF induces IRF3 activation. Science. 2015; 347(6227):aaa2630. Epub 2015/02/ 01. https://doi.org/10.1126/science.aaa2630 PMID: 25636800. 14. Wang L, Ning S. TRIMming Type I Interferon-Mediated Innate Immune Response in Antiviral and Anti- tumor Defense. Viruses. 2021; 13(2). Epub 2021/03/07. https://doi.org/10.3390/v13020279 PMID: 33670221; PubMed Central PMCID: PMC7916971. 15. Xue B, Li H, Guo M, Wang J, Xu Y, Zou X, et al. TRIM21 Promotes Innate Immune Response to RNA Viral Infection through Lys27-Linked Polyubiquitination of MAVS. J Virol. 2018; 92(14). Epub 2018/05/ 11. https://doi.org/10.1128/JVI.00321-18 PMID: 29743353; PubMed Central PMCID: PMC6026736. 16. Yang D, Zuo C, Wang X, Meng X, Xue B, Liu N, et al. Complete replication of hepatitis B virus and hepa- titis C virus in a newly developed hepatoma cell line. Proc Natl Acad Sci U S A. 2014; 111(13):E1264– 73. Epub 2014/03/13. https://doi.org/10.1073/pnas.1320071111 PMID: 24616513; PubMed Central PMCID: PMC3977290. 17. Hu S, Sun H, Yin L, Li J, Mei S, Xu F, et al. PKR-dependent cytosolic cGAS foci are necessary for intra- cellular DNA sensing. Sci Signal. 2019; 12(609). Epub 2019/11/28. https://doi.org/10.1126/scisignal. aav7934 PMID: 31772125. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 26 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection 18. Sharma NR, Majerciak V, Kruhlak MJ, Zheng ZM. KSHV inhibits stress granule formation by viral ORF57 blocking PKR activation. PLoS Pathog. 2017; 13(10):e1006677. Epub 2017/10/31. https://doi. org/10.1371/journal.ppat.1006677 PMID: 29084250; PubMed Central PMCID: PMC5679657. 19. Kim Y, Park J, Kim S, Kim M, Kang MG, Kwak C, et al. PKR Senses Nuclear and Mitochondrial Signals by Interacting with Endogenous Double-Stranded RNAs. Mol Cell. 2018; 71(6):1051–63 e6. Epub 2018/ 09/04. https://doi.org/10.1016/j.molcel.2018.07.029 PMID: 30174290. 20. Hussain M, Mohammed A, Saifi S, Khan A, Kaur E, Priya S, et al. MITOL-dependent ubiquitylation neg- atively regulates the entry of PolgammaA into mitochondria. PLoS Biol. 2021; 19(3):e3001139. Epub 2021/03/04. https://doi.org/10.1371/journal.pbio.3001139 PMID: 33657094; PubMed Central PMCID: PMC7959396. 21. Durcan TM, Tang MY, Perusse JR, Dashti EA, Aguileta MA, McLelland GL, et al. USP8 regulates mito- phagy by removing K6-linked ubiquitin conjugates from parkin. EMBO J. 2014; 33(21):2473–91. Epub 2014/09/14. https://doi.org/10.15252/embj.201489729 PMID: 25216678; PubMed Central PMCID: PMC4283406. 22. Garaigorta U, Chisari FV. Hepatitis C virus blocks interferon effector function by inducing protein kinase R phosphorylation. Cell Host Microbe. 2009; 6(6):513–22. Epub 2009/12/17. https://doi.org/10.1016/j. chom.2009.11.004 PMID: 20006840; PubMed Central PMCID: PMC2905238. 23. Yamane D, Feng H, Rivera-Serrano EE, Selitsky SR, Hirai-Yuki A, Das A, et al. Basal expression of interferon regulatory factor 1 drives intrinsic hepatocyte resistance to multiple RNA viruses. Nat Micro- biol. 2019; 4(7):1096–104. Epub 2019/04/17. https://doi.org/10.1038/s41564-019-0425-6 PMID: 30988429; PubMed Central PMCID: PMC6588457. 24. 25. Lu X, Chen Q, Liu H, Zhang X. Interplay Between Non-Canonical NF-kappaB Signaling and Hepatitis B Virus Infection. Front Immunol. 2021; 12:730684. Epub 2021/10/19. https://doi.org/10.3389/fimmu. 2021.730684 PMID: 34659217; PubMed Central PMCID: PMC8511458. Liuyu T, Yu K, Ye L, Zhang Z, Zhang M, Ren Y, et al. Induction of OTUD4 by viral infection promotes antiviral responses through deubiquitinating and stabilizing MAVS. Cell Res. 2019; 29(1):67–79. Epub 2018/11/10. https://doi.org/10.1038/s41422-018-0107-6 PMID: 30410068; PubMed Central PMCID: PMC6318273. 26. Wang J, Li H, Xue B, Deng R, Huang X, Xu Y, et al. IRF1 Promotes the Innate Immune Response to Viral Infection by Enhancing the Activation of IRF3. J Virol. 2020; 94(22). Epub 2020/09/04. https://doi. org/10.1128/JVI.01231-20 PMID: 32878885; PubMed Central PMCID: PMC7592201. 27. Jha HC A JM, Saha A, Banerjee S, Lu J, Robertson ES. Epstein-Barr virus essential antigen EBNA3C attenuates H2AX expression. J Virol. 2014; 88(7):3776–88. Epub 2014/01/17. https://doi.org/10.1128/ JVI.03568-13 PMID: 24429368; PubMed Central PMCID: PMC3993541. 28. Doyle T, Moncorge O, Bonaventure B, Pollpeter D, Lussignol M, Tauziet M, et al. The interferon-induc- ible isoform of NCOA7 inhibits endosome-mediated viral entry. Nat Microbiol. 2018; 3(12):1369–76. Epub 2018/11/28. https://doi.org/10.1038/s41564-018-0273-9 PMID: 30478388; PubMed Central PMCID: PMC6329445. 29. Wang W, Jin Y, Zeng N, Ruan Q, Qian F. SOD2 Facilitates the Antiviral Innate Immune Response by Scavenging Reactive Oxygen Species. Viral Immunol. 2017; 30(8):582–9. Epub 2017/06/03. https:// doi.org/10.1089/vim.2017.0043 PMID: 28574756. 30. Koga R, Kubota M, Hashiguchi T, Yanagi Y, Ohno S. Annexin A2 Mediates the Localization of Measles Virus Matrix Protein at the Plasma Membrane. J Virol. 2018; 92(10). Epub 2018/03/02. https://doi.org/ 10.1128/JVI.00181-18 PMID: 29491166; PubMed Central PMCID: PMC5923071. 31. Tatematsu M, Nishikawa F, Seya T, Matsumoto M. Toll-like receptor 3 recognizes incomplete stem structures in single-stranded viral RNA. Nat Commun. 2013; 4:1833. Epub 2013/05/16. https://doi.org/ 10.1038/ncomms2857 PMID: 23673618. 32. Yang W, Wu YH, Liu SQ, Sheng ZY, Zhen ZD, Gao RQ, et al. S100A4+ macrophages facilitate zika virus invasion and persistence in the seminiferous tubules via interferon-gamma mediation. PLoS Pathog. 2020; 16(12):e1009019. Epub 2020/12/15. https://doi.org/10.1371/journal.ppat.1009019 PMID: 33315931; PubMed Central PMCID: PMC7769614. 33. Wang K, Lai C, Li T, Wang C, Wang W, Ni B, et al. Basic fibroblast growth factor protects against influ- enza A virus-induced acute lung injury by recruiting neutrophils. J Mol Cell Biol. 2018; 10(6):573–85. Epub 2017/11/10. https://doi.org/10.1093/jmcb/mjx047 PMID: 29121325. 34. Galanina N, Goodman AM, Cohen PR, Frampton GM, Kurzrock R. Successful Treatment of HIV-Asso- ciated Kaposi Sarcoma with Immune Checkpoint Blockade. Cancer Immunol Res. 2018; 6(10):1129– 35. Epub 2018/09/09. https://doi.org/10.1158/2326-6066.CIR-18-0121 PMID: 30194084; PubMed Cen- tral PMCID: PMC6248326. 35. Chevrier N, Mertins P, Artyomov MN, Shalek AK, Iannacone M, Ciaccio MF, et al. Systematic discovery of TLR signaling components delineates viral-sensing circuits. Cell. 2011; 147(4):853–67. Epub 2011/ PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 27 / 28 PLOS PATHOGENSRegulation of PKR-dependent protein synthesis by TRIM21 upon virus infection 11/15. https://doi.org/10.1016/j.cell.2011.10.022 PMID: 22078882; PubMed Central PMCID: PMC3809888. 36. Dalet A, Arguello RJ, Combes A, Spinelli L, Jaeger S, Fallet M, et al. Protein synthesis inhibition and GADD34 control IFN-beta heterogeneous expression in response to dsRNA. EMBO J. 2017; 36 (6):761–82. Epub 2017/01/20. https://doi.org/10.15252/embj.201695000 PMID: 28100675; PubMed Central PMCID: PMC5350567. 37. Cao X. Self-regulation and cross-regulation of pattern-recognition receptor signalling in health and dis- ease. Nat Rev Immunol. 2016; 16(1):35–50. Epub 2015/12/30. https://doi.org/10.1038/nri.2015.8 PMID: 26711677. 38. Rehwinkel J, Gack MU. RIG-I-like receptors: their regulation and roles in RNA sensing. Nat Rev Immu- nol. 2020; 20(9):537–51. Epub 2020/03/24. https://doi.org/10.1038/s41577-020-0288-3 PMID: 32203325; PubMed Central PMCID: PMC7094958. 39. van Wijk SJ, Fulda S, Dikic I, Heilemann M. Visualizing ubiquitination in mammalian cells. EMBO Rep. 2019; 20(2). Epub 2019/01/23. https://doi.org/10.15252/embr.201846520 PMID: 30665942; PubMed Central PMCID: PMC6362358. 40. Berndsen CE, Wolberger C. New insights into ubiquitin E3 ligase mechanism. Nat Struct Mol Biol. 2014; 21(4):301–7. Epub 2014/04/05. https://doi.org/10.1038/nsmb.2780 PMID: 24699078. 41. Cruz Walma DA, Chen Z, Bullock AN, Yamada KM. Ubiquitin ligases: guardians of mammalian develop- ment. Nat Rev Mol Cell Biol. 2022. Epub 2022/01/27. https://doi.org/10.1038/s41580-021-00448-5 PMID: 35079164. 42. Jones EL, Laidlaw SM, Dustin LB. TRIM21/Ro52—Roles in Innate Immunity and Autoimmune Disease. Front Immunol. 2021; 12:738473. Epub 2021/09/24. https://doi.org/10.3389/fimmu.2021.738473 PMID: 34552597; PubMed Central PMCID: PMC8450407. 43. Mu T, Zhao X, Zhu Y, Fan H, Tang H. The E3 Ubiquitin Ligase TRIM21 Promotes HBV DNA Polymer- ase Degradation. Viruses. 2020; 12(3). Epub 2020/04/05. https://doi.org/10.3390/v12030346 PMID: 32245233; PubMed Central PMCID: PMC7150939. 44. Song Y, Li M, Wang Y, Zhang H, Wei L, Xu W. E3 ubiquitin ligase TRIM21 restricts hepatitis B virus rep- lication by targeting HBx for proteasomal degradation. Antiviral Res. 2021; 192:105107. Epub 2021/06/ 08. https://doi.org/10.1016/j.antiviral.2021.105107 PMID: 34097931. 45. Hauler F, Mallery DL, McEwan WA, Bidgood SR, James LC. AAA ATPase p97/VCP is essential for TRIM21-mediated virus neutralization. Proc Natl Acad Sci U S A. 2012; 109(48):19733–8. Epub 2012/ 10/24. https://doi.org/10.1073/pnas.1210659109 PMID: 23091005; PubMed Central PMCID: PMC3511720. 46. Zhang Z, Yuan B, Bao M, Lu N, Kim T, Liu YJ. The helicase DDX41 senses intracellular DNA mediated by the adaptor STING in dendritic cells. Nat Immunol. 2011; 12(10):959–65. Epub 2011/09/06. https:// doi.org/10.1038/ni.2091 PMID: 21892174; PubMed Central PMCID: PMC3671854. 47. McEwan WA, Tam JC, Watkinson RE, Bidgood SR, Mallery DL, James LC. Intracellular antibody- bound pathogens stimulate immune signaling via the Fc receptor TRIM21. Nat Immunol. 2013; 14 (4):327–36. Epub 2013/03/05. https://doi.org/10.1038/ni.2548 PMID: 23455675; PubMed Central PMCID: PMC3672961. 48. Darini C, Ghaddar N, Chabot C, Assaker G, Sabri S, Wang S, et al. An integrated stress response via PKR suppresses HER2+ cancers and improves trastuzumab therapy. Nat Commun. 2019; 10(1):2139. Epub 2019/05/16. https://doi.org/10.1038/s41467-019-10138-8 PMID: 31086176; PubMed Central PMCID: PMC6513990. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011443 June 16, 2023 28 / 28 PLOS PATHOGENS
10.1590_acb382323	Original Article https://doi.org/10.1590/acb382323 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke Jianwen Jia1 , Xiaochao Tian2 , Jinzhao He3 , Guozhong Ma3 , Weiliang He3,4* 1. Capital Medical University – Beijing Chaoyang Hospital – Department of Neurosurgery – Beijing, China. 2. Second Hospital of Hebei Medical University – Department of Cardiology – Hebei, China. 3. Heyuan People’s Hospital – Guangdong Provincial People’s Hospital Heyuan Hospital – Department of Neurology – Guangdong, China. 4. Heyuan People’s Hospital – Heyuan Key Laboratory of Molecular Diagnosis & Disease Prevention and Treatment – Doctors Station of Guangdong province – Guangdong, China. ABSTRACT Purpose: Motor function is restored by axonal sprouting in ischemic stroke. Mitochondria play a crucial role in axonal sprouting. Taurine (TAU) is known to protect the brain against experimental stroke, but its role in axonal sprouting and the underlying mechanism are unclear. Methods: We evaluated the motor function of stroke mice using the rotarod test on days 7, 14, and 28. Immunocytochemistry with biotinylated dextran amine was used to detect axonal sprouting. We observed neurite outgrowth and cell apoptosis in cortical neurons under oxygen and glucose deprivation (OGD), respectively. Furthermore, we evaluated the mitochondrial function, adenosine triphosphate (ATP), mitochondrial DNA (mtDNA), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PCG-1α), transcription factor A of mitochondria (TFAM), protein patched homolog 1 (PTCH1), and cellular myelocytomatosis oncogene (c-Myc). Results: TAU recovered the motor function and promoted axonal sprouting in ischemic mice. TAU restored the neuritogenesis ability of cortical neurons and reduced OGD-induced cell apoptosis. TAU also reduced reactive oxygen species, stabilized mitochondrial membrane potential, enhanced ATP and mtDNA content, increased the levels of PGC-1α, and TFAM, and restored the impaired levels of PTCH1, and c-Myc. Furthermore, these TAU-related effects could be blocked using an Shh inhibitor (cyclopamine). Conclusion: Taurine promoted axonal sprouting via Shh-mediated mitochondrial improvement in ischemic stroke. Key words: Stroke. Taurine. Mitochondria. Introduction Stroke is among the primary causes of disability and death worldwide1. In the past decade, new research has been focused on brain plasticity including axonal sprouting that is widely recognized in the functional recovery of an injured brain2. External interventions have been increasingly reported to accelerate repair processes3. Therefore, targeting the promotion of axonal sprouting is a more promising therapeutic strategy for ischemic stroke. Axonal sprouting requires adenosine triphosphate (ATP) to power fundamental developmental processes4. Mitochondria are among the important organelles and produce most of the required ATP5. They have been reported to be involved in brain Corresponding author: [email protected] Received: Jan 30, 2023 \| Accepted: Apr 18, 2023 Research performed at Department of Neurology, Heyuan People’s Hospital, Guangdong Provincial People’s Hospital Heyuan Hospital, Guangdong, China. Acta Cir Bras. V38 . e382323 . 2023 ACTA CIRÚRGICA BRASILEIRATaurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke plasticity after brain injury6. Several recent studies have established the effects of mitochondria in determining axonal sprouting7-9. Therefore, it may be ideal to identify agents promoting axon regeneration by protecting mitochondria. Taurine (TAU) is a rich amino acid stored in mammalian brain tissues10. It has been considered a target in neurological diseases causing traumatic brain injury, including Alzheimer’s disease, Parkinson’s disease, and ischemic stroke11-15. Studies have indicated that TAU plays a role in neurogenesis16,17. TAU has been reported to protect the brain from experimental stroke by preserving mitochondrial function18. Moreover, it exerts potentially protective effects against hepatic encephalopathy, hyperammonemia-induced mitochondrial dysfunction, and energy crisis19. However, whether TAU modulates mitochondrial function during axonal sprouting under stroke conditions remains unclear. Thus, we aimed to assess the role of TAU in axonal sprouting against cerebral ischemic injury, clarify the function of mitochondria in TAU-induced axonal sprouting, and further determine the underlying potential molecular mechanism. Methods This study was supported by grants from the scientific research start-up funds of Heyuan People’s Hospital, the High- level Talent Foundation of Hebei Province in 2021 (No. A202002004), and the National Natural Science Foundation of China (No. 81801312). The experimental protocols were approved by the Committee of Ethics on Laboratory Animals of Heyuan People’s Hospital and performed according to the recommendations from the Guide for the Care and Use of Laboratory Animals of the National Institute of Health. Experimental protocols The experiment was randomized into two parts (Fig. 1). Under part 1, we studied the effect of TAU on stroke mice. The mice were randomly divided into three groups: • Sham: 0.9% saline; • Model: 0.9% saline; • Model + TAU: 50 mg/kg dissolved in 0.9% saline. In the model + TAU group, TAU (1 mL/kg) was intravenously injected 30 min after the stroke for seven consecutive days (D). After the stroke, the rotarod test was performed on D7, D14, and D28. On D7, we injected biotinylated dextran amine (BDA), which was used as an anterograde neuronal tracer, into the contralateral hemispheric somatosensory cortex. On D14, we detected the BDA-labeled axonal density using immunocytochemistry (ICC). The expressions of mitochondrial DNA (mtDNA), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PCG-1α), and transcription factor A of mitochondria (TFAM) were measured on D14 after stroke. The role of TAU in primary cortical neurons under oxygen and glucose deprivation (OGD) and the potential underlying mechanism were determined in part 2. The neurons were divided into the following groups: • Control; • OGD; • OGD + TAU. The levels of neurite outgrowth, cell apoptosis, mitochondrial function, ATP, mtDNA, PCG-1α, TFAM, PTCH1, and c-Myc were detected in different groups. To understand the effect of the Shh pathway, a specific inhibitor of the Shh pathway, cyclopamine (CPM, 2.5 μM in 0.1% dimethyl sulfoxide [DMSO]), was added to the medium 30 min before TAU treatment. 2 Acta Cir Bras. V38 . e382323 . 2023 Jianwen Jia et al. (a) Experiment 1 Animal Experiment Sham Model Model+TAU (b) Experiment 2 Primary Cortical Neuron Experiment 8 mice from each group were prepared for Rota-Rod tests at D7, 14 and 28; Biotinylated dextran amine (BDA), an anterograde neuronal tracer, was prepared for axons regeneration. At D7, BDA was stereotaxically injected into the somatosensorimotor cortex in the contralesional hemisphere. At D14, mice were sacrificed by transcardiac perfusion of immunocytochemistry; 6 Brains from each group were prepared for detecting mtDNA, PCG-1a and TFAM at D14 Control OGD OGD+TAU OGD+TAU+CPM Primary cortical neurons in different groups were collected for detecting neurite ooutgrowth, cell apoptosis, ROS, MMP, ATP, mtDNA, PCG-1a, TFAM, PTCH and cMyc expression. Figure 1 – Schematic of experimental design. In (a) experiment 1, animals were divided into three groups: sham (0.9% saline), model (0.9% saline), and model + taurine (TAU) (50 mg/kg dissolved in 0.9% saline). In the model + TAU group, TAU (1 mL/kg) was intravenously injected 30 min after the stroke for seven consecutive days. After the stroke, the rotarod test was performed on D7, D14, and D28. On D7 after stroke, biotinylated dextran amine (BDA), an anterograde neuronal tracer, was stereotaxically injected into the somatic sensorimotor cortex in the contralesional hemisphere. The mice were sacriﬁced by transcardiac perfusion on D14 after TAU treatment. Furthermore, we evaluated mitochondrial function, ATP, mtDNA, the levels of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PCG-1α), and transcription factor A of mitochondria (TFAM). In (b) experiment 2, primary cortical neurons were divided into three groups: control, oxygen and glucose deprivation (OGD), and OGD + TAU groups. We evaluated cell viability and apoptosis, neurite outgrowth, mitochondrial function, ATP, mtDNA, PCG-1α, TFAM, protein patched homolog 1 (PTCH1), and c-Myc levels of different groups. To detect the effect of the Shh pathway, a specific inhibitor of the Shh pathway, cyclopamine (CPM, 2.5 μM in 0.1% dimethyl sulfoxide [DMSO]), was added to the medium 30 min before TAU treatment. Focal cerebral cortical ischemia model According to a previous report, we preceded focal cerebral cortical ischemia in C57BL/6 mice20. Briefly, C57BL/6 mice were anesthetized and permanently ligated in the right common carotid artery. Subsequently, the right middle cerebral arteries of the mice were exposed along with the coagulated striatal branch. The abovementioned procedures were performed in all groups. However, coagulation of the striatal branch was excluded in the sham group. Rotarod test The rotarod test was performed on D7, D14, and D28 to assess the motor function. Based on our previous report3, we placed the mice on a rotating rotarod cylinder. The speed of the cylinder accelerated from 4 to 40 rpm in less than 300 s. The riding time of the mouse on the cylinder was recorded three times. The mean value was considered for statistical analysis. Each mouse was trained before the tests and rested for 15 min after each test. Biotinylated dextran amine injection After the administration of anesthesia, 10% BDA was injected into the somatosensorimotor cortex of the contralesional hemisphere (two sites/brain, 1 μL/site) of the mice using a stereotactic device with nano injection pump (0.15 μL/min). The coordinates of the two injection sites are as follows: • 2 mm posterior to the bregma and 1.5 mm lateral to the midline; • 1.1 mm posterior to the bregma and 1.5 mm lateral to the midline. Based on a previous study, the needle (1 μL) was kept in situ for 10 min after injection21. Immunocytochemistry analysis The brains of mice were blocked with 4% paraformaldehyde. They were then cryoprotected with 30% sucrose and sliced into 20 μm. The three coronal sections, including the somatosensory motor cortex and infarct cavity (+0.74, -0.46, Acta Cir Bras. V38 . e382323 . 2023 3 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke and -1.58 mm from the bregma), were selected for ICC. The sections were fixed with 5% normal goat serum. After adding horseradish peroxidase (1:1,000), the sections were kept overnight at 4 °C. DyLight 633-conjugated goat anti-horseradish peroxidase (1:500, Jackson Immuno Research Labs, United States of America) was added to the cultured sections for 2 h. The nucleus was stained using Hoechst 33342 at a final concentration of 5 μg/mL. The images were obtained using an upright fluorescence microscope (Olympus, Japan). The quantification of mtDNA and nuclear DNA The DNeasy Blood and Tissue kit (Qiagen, Hermantown, MD, United States of America) was used to extract total DNA. Real-time polymerase chain reaction (RT-PCR) (Applied Biosystems, United States of America) was performed in the presence of SYBR Green I (CWBIO, China) to detect the copy number of the mtDNA. Compared with nuclear DNA (rRNA 18S), the relative copy number of mtDNA was detected. The primers used for mtDNA and rRNA 18S quantification are as follows22: mtDNA: sense, 5’-GCCCCAGATATAGCATTCCC-3’; anti-sense, 5’-GTTCATCCTGTTCCTGCTCC-3’; rRNA 18S: sense, 5’-TAGAGGGACAAGTGGCGTTC-3’; anti-sense, 5’-CGCTGAGCCAGTCAGTGT-3’. Quantitative real-time polymerase chain reaction Trizol reagent (Invitrogen, United States of America) was used to extract total RNA, and the first-strand cDNA synthesis kit (Fermentas International Inc, Canada) was used to reverse-transcribe the RNA into cDNA. In the presence of SYBR Green I (CWBIO, China), the cDNA was amplified using the RT-PCR system (Applied Biosystems company, USA). The 2–∆∆CT method was performed to calculate relative PCR products with the mouse glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene as the control. The related gene sequences for quantitative RT-PCR (RT-qPCR) are as follows23,24: PCG-1α: Forward, 5’-CACCAA ACCCACAGAAAACAG-3’; Reverse, 5’-GGGTCAGAGGAAGAGATAAAGTTG-3’; TFAM: Forward, 5’-CACCCAGATGCAAAACTTTCAG-3’; Reverse, 5’-CTGCTCTTTATACTTGCTCACAG-3’; PTCH1: Forward, 5’-ACCCGCCAGAAGATAGGAGA-3’; Reverse: 5’-GGAGTGCTGAGTCCAGGTGT-3’; c-Myc: Forward, 5’-ATCAAGAGGCCACAGCAAAC-3’; Reverse, 5’-TTGGCAGCTGGATAGTCCTT-3’; GAPDH: Forward, 5’-GACATCATACTTGGCAGG-3’; Reverse, 5’-CTCGTGGAGTCTACTGGT-3’. Primary cortical neuron culture and oxygen and glucose deprivation The cerebral cortex tissues from the embryonic brains (D18) of C57BL/6 were removed. The dispersed tissues were dissociated into a single-cell mixture using Hibernate-E (Sigma, United States of America). Cells were plated onto culture dishes (BD Biosciences, United States of America) coated with poly-L-lysine and grown in neurobasal medium with 2% B-27 (Invitrogen, United States of America) and 0.5 mM glutamine (Life Technologies, United States of America) at 37 °C in a 5% CO2 incubator. OGD was used to initiate ischemia. We used the same incubator in combination with a hypoxic workstation containing a gas mixture of 0.1% O2, 94.9% N2, and 5% CO2. To detect the effect of the Shh pathway, a specific inhibitor of the Shh pathway, CPM (2.5 μM in 0.1% DMSO), was added to the medium 30 min before TAU treatment. 4 Acta Cir Bras. V38 . e382323 . 2023 Jianwen Jia et al. Neurite outgrowth and apoptosis To measure the length of neurite outgrowth of the cortical neurons, cells were stained with mouse monoclonal anti-β III tubulin antibody (Tuj-1, 1:500, Sigma, United States of America) using the immunofluorescence method. Tuj1-positive cells were captured at ×40 objective with an upright fluorescence microscope (Olympus, Tokyo, Japan). The longest neurite length of Tuj-1-positive cell (60/group) was measured using the Image-J analysis system. The apoptosis of neurons was detected by staining the cells with Hoechst 33342 at the final concentration of 5 μg/mL. Cells were observed immediately using a fluorescence microscope. Additionally, apoptotic cells were also detected using an Annexin V-FITC/PI apoptosis detection kit according to the manufacturer’s protocols. Apoptotic and necrotic cells were quantified using a FACSCalibur cytometer (Becton Dickinson, Franklin Lakes, New Jersey, United States of America). The rate of apoptosis was determined using the Cell Quest software. Determination of reactive oxygen species The intracellular reactive oxygen species (ROS) level was detected using 2, 7-Dichloroﬂuorescin diacetate (DCFH-DA, Beyotime, China), as previously described3. Briefly, after the treatment, cells were incubated with DCFH-DA at 37 °C for 30 min and then washed with phosphate buffered saline. The fluorescence intensity was then immediately detected using a luminescence spectrometer (excitation: 488 nm and emission: 525 nm, LS50B, PerkinElmer, United States of America). Data are presented as percent of control. Determination of mitochondrial membrane potential The JC-1 assay kit (Beyotime, China) was used to detect the changes in mitochondrial membrane potential (MMP). In brief, cells were collected and stained with JC-1 staining solution (5 μg/mL) for 20 min at 37 °C and rinsed with JC-1 staining buffer twice. The fluorescence intensity of JC-1 was detected using a multimode reader (Tecan, Switzerland). Data were presented as a percentage of control. In addition, fluorescence images of MMP were obtained using a fluorescence microscope. Adenosine triphosphate detection The ATP levels were detected using an ATP assay kit (Beyotime, China) according to the manufacturer’s protocols. Luminance intensity was measured using a monochromatic microplate reader (Tecan, Switzerland). Data were presented as percent of control. The ATP content was determined as a percentage of untreated cells (control). Statistical analysis Quantitative data were presented as mean ± standard error of measurement (SEM). Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) 17.0 software. One-way analysis of variance (ANOVA) was performed for statistical analysis, and then Turkey’s post hoc test or Student t-test was used. A P < 0.05 was considered statistically significant. Results Taurine improves recovery of motor function in stroke To observe the effect of TAU on the motor function of stroke, the rotarod test was performed. Compared with the sham group, the average riding time of the model group on D7, D14, and D28 was shorter (P < 0.05). However, mice were alive for more time in the TAU-treated groups compared with those in the model groups (P < 0.05), indicating that TAU could improve the impairments in the motor function induced by ischemic stroke (Fig. 2). Acta Cir Bras. V38 . e382323 . 2023 5 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke (a) ) s ( e m i t g n i d i r e g a r e v A 200 150 100 50 0 (b) ) s ( e m i t g n i d i r e g a r e v A 200 160 120 80 40 0 # Sham Model Model+TAU # * 14d 28d D14 * # * 7d * D7 # # D28 * # Sham Model Model+TAU Sham Model Model+TAU Sham Model Model+TAU TAU: taurine; P < 0.05 vs. the sham group; #P < 0.05 vs. the model group. Figure 2 – Motor recovery in the different groups in stroke mice. Data revealed that DL-3-n-butylphthalide (NBP) significantly improved motor recovery. (a and b). Rotarod test on D7, D14, and D28 after stroke. The results are shown as the mean ± standard error of the mean (n = 8). Taurine promotes axonal sprouting in ischemic stroke We further explored the potential mechanism of TAU on axonal sprouting in stroke. As shown in Fig. 3, a few BDA- labeled axons could be observed in the sham group of the ipsilesional cortex. The axonal density of BDA-labeled axons was greater in the model group than in the sham group. However, the highest density was observed in the model + TAU group. Interestingly, the same trend was observed in the contralesional cortex. These results suggested that TAU could promote the growth of axonal branches in the ipsilateral cortex, as well as the contralateral cortex. (a) Sham Model Model+TAU l a n o i s e l i s p i l a n o t s e l a r t n o C (b) ipsllesional cortex (c) contralesional cortex ) % ( a e r A e v i t c a e R A D B 50 40 30 20 10 0 P<0.001 # P<0.001 Sham Model Model+TAU ) % ( a e r A e v i t c a e R A D B 80 60 40 20 0 P<0.001 # P<0.001 * Sham Model Model+TAU P < 0.05 vs. the sham group; #P < 0.05 vs. the model group. Figure 3 – Effect of taurine (TAU) on axonal sprouting in ischemic stroke. (a) Representative images of the biotinylated dextran amine (BDA)- labeled axons in the contralesional cortex and the ipsilesional cortex; scale = 50 μm. (b) The density of BDA-labeled axons in the ipsilesional cortex. (c) The density of BDA-labeled axons in the contralesional cortex. The results are shown as the mean ± standard error of the mean (n = 6). Taurine restores mitochondrial-related factors of the brain after an ischemic stroke To explore the underlying mechanisms, we focused on the mitochondria, and the expression of mitochondrial biogenesis markers was estimated. The mtDNA content significantly decreased in the model group, but increased after TAU treatment. 6 Acta Cir Bras. V38 . e382323 . 2023 Additionally, TAU significantly increased the expression of PGC-1α, and TFAM after stroke (Fig. 4). The data showed that TAU could preserve the mitochondria-related factors after an ischemic stroke. Jianwen Jia et al. (a) t n e t n o c A N D t m e v i t a l e R ) t n o C f o d o f ( l 1.2 0.8 0.4 0.0 (b) P<0.001 # P<0.001 Sham Model Model+TAU n o i s s e r p x e A N R t m e v i t a l e R ) H D P A G / a 1 - G C P ( 1.2 0.8 0.4 0.0 (c) n o i s s e r p x e A N R m e v i t a l e R P<0.025 # P<0.001 * Sham Model Model+TAU 1.2 ) H D P A G M A F T ( / 0.8 0.4 P<0.026 # P<0.001 * 0.0 Sham Model Model+TAU P < 0.05 vs. sham group; #P < 0.05 vs. model group. Figure 4 – Effect of taurine (TAU) on mitochondria-related factors after an ischemic stroke. (a) The mitochondrial DNA (mtDNA) content was measured in different groups. (b) The mRNA level of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PCG-1α) was measured by quantitative real-time polymerase chain reaction (RT-qPCR). (c) The mRNA level of transcription factor A of TFAM was measured by RT-qPCR. Results are expressed as the mean ± standard error of the mean (n = 6). Taurine restores the neuritogenesis ability of cortical neurons induced by oxygen and glucose deprivation We further explored the effect of TAU on the neurite outgrowth of cortical neurons under OGD. As shown in Figs. 5a and 5b, compared with the control group, the length of neurite outgrowth in the OGD group showed remarkable shortening. However, in the presence of different concentrations of TAU (10/20 mM), the length of neurite outgrowth signiﬁcantly recovered. These results indicated that 20 mM TAU could significantly recover neuronal neuritogenesis ability after an ischemic injury, and this concentration was used for subsequent experiments. Additionally, the apoptotic rate was determined. As shown in Figs. 5c and 5d, the apoptotic rates in the control, OGD, and OGD + TAU groups were 11.39 ± 1.41%, 51.08 ± 5.80%, and 38.27 ± 5.21%, respectively. The Annexin V-FITC/PI apoptosis detection results also indicated that TAU reduced apoptosis induced by OGD (Fig. 5e). All these results suggested that TAU could protect cortical neurons against ischemic injury. (a) Cont OGD OGD+5mMTAU OGD+10mMTAU OGD+20mMTAU (c) Cont OGD OGD+5mMTAU (b) / h t w o r g t u o e t i r u e M ) m μ ( l l e c 60 40 20 0 # # * Cont OGD OGD+5 m TAU OGD+10 m TAU OGD+20 m TAU (d) ) % ( e t a r s i s o t p o p a l l e C 60 40 20 0 10μm Hoechst 33342 P<0.001 * P<0.002 # Cont OGD OGD+TAU (e) ) % ( e t a r s i s o t p o p a l l e C 40 30 20 10 0 Annexin V-FITC/PI P<0.001 * P<0.004 # Cont OGD OGD+TAU Figure 5 – Effect of taurine (TAU) on neurite outgrowth and apoptosis in primary cortical neurons under oxygen and glucose deprivation (OGD). (a) Neurons were stained with the anti-β-III-tubulin (Tuj-1) antibody. Representative images of Tuj-1-positive neurons in different conditions. Scale = 10 μm. (b) Quantitative analysis of the length of the longest neurite. Values are expressed as the mean ± standard error of the mean (SEM) (n = 60); P < 0.05 vs. control group; #P < 0.05 vs. OGD group. (c and d) The apoptotic rate was determined by Hoechst 33342 staining. (e) The apoptotic rate was determined by using the Annexin V-FITC/PI apoptosis detection kit. Values are expressed as the mean ± SEM (n = 6); P < 0.05 vs. control group; #P < 0.05 vs. OGD group. Acta Cir Bras. V38 . e382323 . 2023 7 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke Taurine preserves mitochondria-related factors in cortical neurons under oxygen and glucose deprivation We investigated the potential mechanism of TAU-induced protection. Firstly, mitochondrial function was examined. Neurons under OGD exhibited ROS accumulation and dissipated MMP. In neurons incubated with TAU, ROS accumulation was reduced (Fig. 6a), and MMP was significantly restored (Fig. 6b). Secondly, ATP levels were assessed. TAU significantly increased the ATP levels, which were reduced compared to those in the OGD group (Fig. 6c). Lastly, under OGD, the mtDNA content significantly decreased, but increased after TAU treatment (Fig. 6d). Additionally, TAU significantly upregulated the expression of PGC-1α and TFAM after neurons were under OGD (Figs. 6e and 6f). The above data indicated that TAU preserved the mitochondria-related factors in cortical neurons under OGD. P<0.001 * P<0.013 # (a) S O R e v i t a l e R ) t n o C f o % ( 180 120 60 0 (c) ) t n o C f o % ( P T A e v i t a l e R 120 90 60 30 0 Cont OGD OGD+TAU P<0.001 # P<0.001 * Cont OGD OGD+TAU Green Red Merge (b) t n o C D G O U A T + D G O e n a r b n e m a i r d n o h c o t i M ) t n o C f o % ( l a i t n e t o p 120 80 40 0 P<0.001 # P<0.001 * Cont OGD OGD+TAU (d) 120 ) t n o C f o % ( 80 40 0 P<0.001 # P<0.001 * Cont OGD OGD+TAU (e) n o i s s e r p x e A N R m e v i t a l e R ) H D P A G / a 1 - G C P ( 1.2 0.8 0.4 0.0 P<0.003 # P<0.001 * Cont OGD OGD+TAU (f) n o i s s e r p x e A N R m e v i t a l e R ) H D P A G M A F T ( / 1.2 0.8 0.4 0.0 P<0.006 # P<0.001 * Cont OGD OGD+TAU P < 0.05 vs. control group; #P < 0.05 vs. OGD group. Figure 6 – Effect of taurine (TAU) on mitochondria-related factors in primary cortical neurons under oxygen and glucose deprivation (OGD). (a) Generation of reactive oxygen species (ROS) in different groups. (b) Mitochondrial membrane potential (MMP) was determined by JC-1 staining under an upright fluorescence microscope. (c) Cellular adenosine triphosphate (ATP) concentrations were assessed in different groups. (d) The mtDNA content was measured in different groups. (e and f) The mRNA levels of PCG-1α and transcription factor A of TFAM were measured by quantitative real-time polymerase chain reaction (RT-qPCR). Results are expressed as the mean ± standard error of the mean (n = 6). The Shh pathway is involved in the protective effect of taurine on the mitochondria in cortical neurons under oxygen and glucose deprivation In our previous study, we found that Shh exerted a neuroprotective effect and caused neurite outgrowth of cortical neurons by preventing mitochondrial dysfunction25. To further study the underlying mechanism, we explored whether the Shh pathway was involved in this process. Firstly, we administrated CPM to observe the mitochondrial functions. Compared with the TAU group, CPM significantly boosted ROS (Fig. 7a), reduced MMP (Fig. 7b), and decreased ATP levels (Fig. 7c). Secondly, we determined the effect of CPM on mitochondrial biogenesis in cortical neurons under OGD. Compared with the TAU group, CPM partly decreased the mtDNA content (Fig. 7d) and reduced the levels of PGC-1α and TFAM (Figs. 7e and 7f). Lastly, we tested the effect of TAU on the Shh pathway and CPM treatment by analyzing the expression of the Shh pathway genes such as PTCH1 and c-Myc. 8 Acta Cir Bras. V38 . e382323 . 2023 CPM partly reduced the levels of PTCH1 and c-Myc that TAU restored in neurons injured by OGD (Fig. 7 g-h). Collectively, these results indicated that TAU improved the mitochondrial function in cortical neurons under OGD possibly through the Shh pathway. Jianwen Jia et al. (b) e n a r b m e m a i r d n o h c o t i M ) t n o C f o % ( l a i t n e t o p (f) n o i s s e r p x e A N R m e v i t a l e R ) H D P A G M A F T ( / P<0.001 #P<0.026 &P<0.028 P<0.026 # (a) 180 120 60 S O R e v i t a l e R ) t n o C f o % ( (e) n o i s s e r p x e A N R m e v i t a l e R ) H D P A G / a 1 - G C P ( 0 Cont OGD OGD+TAU+CPM OGD+TAU 1.2 0.8 0.4 0.0 Cont P<0.004 # #P<0.027 &P<0.028 P<0.003 * OGD OGD+TAU+CPM OGD+TAU 120 80 40 P<0.001 # #P<0.008 &P<0.001 P<0.001 * (c) 120 P T A e v i t a l e R ) t n o C f o % ( 90 60 30 P<0.001 # #P<0.001 &P<0.038 P<0.001 * (d) 120 S O R e v i t a l e R ) t n o C f o % ( 80 40 OGD OGD+TAU+CPM OGD+TAU P<0.001 # #P<0.021 &P<0.002 P<0.002 * 0 Cont 10 ) H D P A G / c y M - c ( 8 6 4 2 (h) n o i s s e r p x e A N R m e v i t a l e R 0 Cont OGD OGD+TAU+CPM OGD+TAU (g) 0 Cont 2.0 1.5 1.0 0.5 ) H D P A G / 1 H C T P ( n o i s s e r p x e A N R m e v i t a l e R 1.2 0.8 0.4 0.0 Cont P<0.001 # #P<0.021 &P<0.015 P<0.001 * OGD OGD+TAU+CPM OGD+TAU 0.0 Cont OGD OGD+TAU+CPM OGD+TAU 0 Cont OGD OGD+TAU+CPM OGD+TAU P<0.001 # #P<0.001 &P<0.033 P<0.001 * OGD OGD+TAU+CPM OGD+TAU P<0.001 * #P<0.002 &P<0.006 P<0.001 # *P < 0.05 vs. control group; #P < 0.05 vs. OGD group; &P < 0.05 vs. OGD + TAU group. Figure 7 – The Shh pathway involving in the neuroprotective effect of taurine (TAU) on mitochondria-related factors in cortical neurons under oxygen and glucose deprivation (OGD). Effect of cyclopamine (CPM), a specific inhibitor of the Shh pathway, on the TAU-induced improvement of mitochondria were assessed by measuring (a) reactive oxygen species (ROS) levels, (b) mitochondrial membrane potential (MMP), (c) adenosine triphosphate (ATP) levels, (d) mitochondrial DNA (mtDNA) content, (e) the mRNA levels of proliferator-activated receptor gamma coactivator 1-alpha (PCG-1α), (f) transcription factor A of mitochondria (TFAM), (g) the mRNA levels of protein patched homolog 1 (PTCH1), and (h) cellular myelocytomatosis oncogene (c-Myc). Data are presented as mean ± standard error of the mean (n = 6). Discussion In the present study, we determined that TAU improved motor function recovery and restored neurogenesis in ischemic stroke. This possibly occurred via improvements in mitochondrial function. Furthermore, we investigated that the Shh pathway exerted an important role in these effects. Our study findings highlighted the novel viewpoint that TAU promoted axonal sprouting by improving Shh-mediated mitochondrial function in cerebral ischemic stroke. It is important to recover motor function among disabled patients with stroke21. A study reported that brain plasticity, including axonal spouting, is positively associated with motor function recovery26. Therefore, it is imperative to identify a novel pharmaceutical for promoting axonal sprouting in stroke. TAU is the main intracellular free-amino acid present in most animal tissues (including the brain). It can pass through the blood–brain barrier and exerts a protective role after brain ischemia18. Indeed, several studies have reported that TAU plays a protective role in stroke27-30. A study reported that TAU can be used to experimentally treat neuronal damage31. Schurr et al.32 suggested that TAU pretreatment can restore synaptic function in rat hippocampal slices. Furthermore, TAU increased the survival rate of newborn neurons and improved neurogenesis Acta Cir Bras. V38 . e382323 . 2023 9 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke in adults16. However, whether TAU affects axonal sprouting in ischemic stroke remains unclear. Our results suggested that TAU revived motor function and promoted axonal sprouting after stroke. The mechanism by which TAU improves axon sprouting has been further studied. The mitochondria, including axon remodeling, may exert a crucial effect on controlling neuroplasticity. Impaired mitochondria may impair neural plasticity after stroke6. Recently, a study reported that neurological diseases might be related to mitochondrial biogenesis33. Another study reported that, after cerebral ischemia, biological damage might adversely affect mitochondrial function in vitro34. Preservation of mtDNA copy number may alleviate damage to mitochondrial biogenesis35. We observed that loss of mtDNA induced by insult injury was restored after TAU treatment. In addition, PGC-1α could upregulate the expression of genes related to mitochondrial biology or increase their transcription activity36. TFAM regulated the copy number of mtDNA37. We measured the levels of TFAM and PGC-1α and observed that TAU could restore ischemic injury-induced increases in their levels. Taken together, the data suggested that TAU-induced axonal spouting after stroke was partially mediated by enhanced mitochondrial function. We also investigated the potential molecular mechanisms underlying mitochondrial function in TAU-stimulated axonal spouting. A study reported that the Shh pathway is involved in axonal spouting38. Shh overexpression increased neurogenesis in the dentate gyrus of nonischemic rats, but its blockade using CPM abolished cerebrolysin-induced neurogenesis in adult rats39,40. In cortical neurons, Shh may have improved neurite outgrowth by preventing mitochondrial dysfunction25,41. Furthermore, in mice, TAU promoted the proliferation and differentiation of cochlear neural stem cells via the Shh pathway42. However, whether Shh mediates the effect of TAU on the mitochondria in axonal remodeling remains unknown. Therefore, we hypothesized that TAU improves mitochondrial function in the axon via the Shh pathway. Interestingly, inhibition of the Shh pathway with CPM exacerbated the levels of ROS, MMP, ATP, mtDNA, PCG-1α, TFAM, PTCH1, and c-Myc, which were improved by TAU in neurons under OGD. Taken together, these data suggested that TAU-induced mitochondrial improvements in neurons partially occurred via Shh pathway. Our study has some limitations. First, additional studies are warranted to elucidate the detailed mechanisms by which TAU in axonal spouting regulates the mitochondria to prevent focal cerebral ischemia; this is beneficial in understanding TAU modulation. Furthermore, the additional mechanisms underlying the role of TAU in axonal spouting after ischemia stroke, such as the PI3K/Akt, MAPK/Erk, and other pathways, need to be explored further. Conclusions Our study results suggested that TAU promoted axonal spouting after stroke via mitochondrial regulation. This protective effect was partially mediated by the Shh pathway. Our findings highlighted that TAU might contribute to the therapeutic intervention of axonal remodeling after ischemic stroke-targeted mitochondrial and Shh pathways. Authors’ contribution Design of the study: He W; Acquisition and analysis of data: Jia J, Tian X, He J, and Ma G; Manuscript writing: Jia J, Tian X, and He W; Critical revision: Jia J, Tian X, and He W; Final approval the version to be published: Jia J, Tian X, He J, Ma G, and He W. About the Authors Jianwen Jia and Weiliang He are PhD. Xiaochao Tian, Jinzhao He and Guozhong Ma are Medical Doctor. Conflict of interest Nothing to declare. 10 Acta Cir Bras. V38 . e382323 . 2023 Preprint statement This manuscript was previously a preprint on a third-party platform41 and was not published elsewhere. https://doi. org/10.21203/rs.3.rs-417526/v1 Jianwen Jia et al. Data availability statement All generated data were presented in this study. Funding Scientific research start-up funds of Heyuan People’s Hospital High-level Talent Foundation of Hebei Province in 2021 Grant No. A202002004 National Natural Science Foundation of China Grant No. 81801312 References 1. Powers WJ, Rabinstein AA, Ackerson T, Adeoye OM, Bambakidis NC, Becker K, Biller J, Brown M, Demaerschalk BM, Hoh B, Jauch EC, Kidwell CS, Leslie-Mazwi TM, Ovbiagele B, Scott PA, Sheth KN, Southerland AM, Summers DV, Tirschwell DL, and on behalf of the American Heart Association Stroke Council. 2018 Guidelines for the early management of patients with acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2018;49(3):e46-e110. https://doi.org/10.1161/ STR.0000000000000158 2. Liao LY, Lau BW, Sánchez-Vidaña DI, Gao Q. Exogenous neural stem cell transplantation for cerebral ischemia. Neural Regen Res. 2019;14(7):1129-37. https://doi.org/10.4103/1673-5374.251188 3. He W, Wang H, Zhao C, Tian X, Li L, Wang H. Role of liraglutide in brain repair promotion through Sirt1-mediated mitochondrial improvement in stroke. J Cell Physiol. 2020;235(3):2986-3001. https://doi.org/10.1002/jcp.29204 4. Lin MY, Sheng ZH. Regulation of mitochondrial transport in neurons. Exp Cell Res. 2015;334(1):35-44. https://doi. org/10.1016/j.yexcr.2015.01.004 5. Todorova V, Blokland A. Mitochondria and synaptic plasticity in the mature and aging nervous system. Curr Neuropharmacol. 2017;15(1):166-73. https://doi.org/10.2174/1570159x14666160414111821 6. Cheng A, Hou Y, Mattson MP. Mitochondria and neuroplasticity. ASN Neuro. 2010;2(5):e00045. https://doi. org/10.1042/AN20100019 7. Sainath R, Ketschek A, Grandi L, Gallo G. CSPGs inhibit axon branching by impairing mitochondria-dependent regulation of actin dynamics and axonal translation. Dev Neurobiol. 2017;77(4):454-73. https://doi.org/10.1002/dneu.22420 8. Tao K, Matsuki N, Koyama R. AMP-activated protein kinase mediates activity-dependent axon branching by recruiting mitochondria to axon. Dev Neurobiol. 2014;74(6):557-73. https://doi.org/10.1002/dneu.22149 9. Spillane M, Ketschek A, Merianda TT, Twiss JL, Gallo G. Mitochondria coordinate sites of axon branching through localized intra-axonal protein synthesis. Cell Rep. 2013;5(6):1564-75. https://doi.org/10.1016/j.celrep.2013.11.022 Acta Cir Bras. V38 . e382323 . 2023 11 Taurine promotes axonal sprouting via Shh-mediated mitochondrial improvement in stroke 10. Brosnan JT, Brosnan ME. The sulfur-containing amino acids: an overview. J Nutr. 2006;136(6):1636S-40S. https://doi. org/10.1093/jn/136.6.1636S 11. Che Y, Hou L, Sun F, Zhang C, Liu X, Pia F, Zhang D, Li H, Wang Q. Taurine protects dopaminergic neurons in a mouse Parkinson’s disease model through inhibition of microglial M1 polarization. Cell Death Dis. 2018;9:435. https://doi.org/10.1038/s41419-018-0468-2 12. Jang H, Lee S, Choi SL, Kim HY, Baek S, Kim Y. Taurine directly binds to oligomeric amyloid-β and recovers cognitive deficits in Alzheimer model mice. Adv Exp Med Biol. 2017;75(Pt 1):233-41. https://doi.org/10.1007/978-94-024- 1079-2_21 13. Tadros MG, Khalifa AE, Abdel-Naim AB, Arafa HM. Neuroprotective effect of taurine in 3-nitropropionic acid- induced experimental animal model of Huntington’s disease phenotype. Pharmacol. Biochem Behav. 2005;82(3):574- 82. https://doi.org/10.1016/j.pbb.2005.10.018 14. Su Y, Fan W, Ma Z, Wen X, Wang W, Wu Q, Huang H. Taurine improves functional and histological outcomes and reduces inflammation in traumatic brain injury. Neuroscience. 2014;266:56-65. https://doi.org/10.1016/j. neuroscience.2014.02.006 15. Menzie J, Prentic H, Wu JY. Neuroprotective mechanisms of taurine against ischemic stroke. BrainSci. 2013;3(2):877- 907. https://doi.org/10.3390/brainsci3020877 16. Gebara E, Udry F, Sultan S, Toni N. Taurine increases hippocampal neurogenesis in aging mice. Stem Cell Res. 2015;14(3):369-79. https://doi.org/10.1016/j.scr.2015.04.001 17. Shivaraj MC, Marcy G, Low G, Ryu JR, Zhao X, Rosales FJ, Goh E.L.K. Taurine induces proliferation of neural stem cells and synapse development in the developing mouse brain. PLoS One. 2012;7(8):e42935. https://doi.org/10.1371/ journal.pone.0042935 18. Sun M, Gu Y, Zhao Y, Xu C. Protective functions of taurine against experimental stroke through depressing mitochondria- mediated cell death in rats. Amino Acids. 2011;40(5):1419-29. https://doi.org/10.1007/s00726-010-0751-8 19. Jamshidzadeh A, Heidari R, Abasvali M, Zarei M, Ommati MM, Abdoli N, Khodaei F, Yeganeh Y, Jafari F, Zarei A, Latifpour Z, Mardani E, Azarpira N, Asadi B, Najibi A. Taurine treatment preserves brain and liver mitochondrial function in a rat model of fulminant hepatic failure and hyperammonemia. Biomed Pharmacother. 2017;86:514-20. https://doi.org/10.1016/j.biopha.2016.11.095 20. Piao CS, Gonzalez-Toledo ME, Xue YQ, Duan WM, Terao S, Granger DN, Kelley RE, Zhao LR. The role of stem cell factor and granulocyte-colony stimulating factor in brain repair during chronic stroke. J Cereb Blood Flow Metab. 2009;29(4):759-70. https://doi.org/10.1038/jcbfm.2008.168 21. Cui L, Duchamp NS, Boston DJ, Ren X, Zhang X, Hu H, Zhao LR. NF-κB is involved in brain repair by stem cell factor and granulocyte-colony stimulating factor in chronic stroke. Exp Neurol. 2015;263:17-27. https://doi.org/10.1016/j. expneurol.2014.08.026 22. Jabir MS, Hopkins L, Ritchie ND, Ullah I, Bayes HK, Li D, Tourlomousis P, Lupton A, Puleston D, Simon AK, Bryant C, Evans, TJ. Mitochondrial damage contributes to Pseudomonas aeruginosa activation of the inflammasome and is downregulated by autophagy. Autophagy. 2015;11(1):166-82. https://doi.org/10.4161/15548627.2014.981915 23. Song MY, Jung HW, Kang SY, Park YK. Atractylenolide III enhances energy metabolism by increasing the SIRT- 1 and PGC1α expression with AMPK phosphorylation in C2C12 mouse skeletal muscle cells. Biol Pharm Bull. 2017;40(3):339-44. https://doi.org/10.1248/bpb.b16-00853 24. Cai D, Yu J, Qiu J, He B, Chen Z, Yan M, Liu Q. Dynamic changes of Sonic Hedgehog signaling pathway in gastric mucosa of rats with MNNG-induced gastric precancerous lesions. J Cell Physiol. 2019;234(7):10827-34. https://doi. org/10.1002/jcp.27908 25. He W, Cui L, Zhang C, Zhang X, He J, Xie Y, Chen Y. Sonic hedgehog promotes neurite outgrowth of cortical neurons under oxidative stress: involving of mitochondria and energy metabolism. Exp Cell Res. 2017;350(1):83-90. https:// doi.org/10.1016/j.yexcr.2016.11.008 12 Acta Cir Bras. V38 . e382323 . 2023 Jianwen Jia et al. 26. Sharma N, Cohen LG. Recovery of motor function after stroke. Dev Psychobiol. 2012;54(3):254-62. https://doi. org/10.1002/dev.20508 27. Prentice H, Gharibani PM, Ma Z, Alexandrescu A, Genova R, Chen PC, Modi J, Menzie J, Pan C, Tao R, Wu JY. Neuroprotective functions through inhibition of ER stress by taurine or taurine combination treatments in a rat stroke model. Adv Exp Med Biol. 2017;975(Pt 1):193-205. https://doi.org/10.1007/978-94-024-1079-2_17 28. Han Z, Gao LY, Lin YH, Chang L, Wu HY, Luo CX, Zhu DY. Neuroprotection of taurine against reactive oxygen species is associated with inhibiting NADPH oxidases. Eur J Pharmacol. 2016;777:129-35. https://doi.org/10.1016/j.ejphar.2016.03.006 29. Zhu XY, Ma PS, Wu W, Zhou R, Hao YJ, Niu Y, Sun T, Li YX, Yu JQ. Neuroprotective actions of taurine on hypoxic-ischemic brain damage in neonatal rats. Brain Res Bull. 2016;124:295-305. https://doi.org/10.1016/j.brainresbull.2016.06.010 30. Zhang Z, Yu R, Cao L. Neuroprotection of taurine through inhibition of 12/15 lipoxygenase pathway in cerebral ischemia of rats. Neurol Res. 2017;39(5):453-8. https://doi.org/10.1080/01616412.2017.1297906 31. Cheong SH, Lee DS. Taurine chloramine prevents neuronal HT22 cell damage through Nrf2-related heme oxygenase-1. Adv Exp Med Biol. 2017;975(Pt 1):145-57. https://doi.org/10.1007/978-94-024-1079-2_13 32. Schurr A, Tseng MT, West CA, Rigor BM. Taurine improves the recovery of neuronal function following cerebral hypoxia: an in vitro study. Life Sci. 1987;40(21):2059-66. https://doi.org/10.1016/0024-3205(87)90098-1 33. Uittenbogaard M, Chiaramello A. Mitochondrial biogenesis: a therapeutic target for neurodevelopmental disorders and neurodegenerative diseases. Curr Pharm Des. 2014;20(35):5574-93. https://doi.org/10.2174/1381612820666140305224906 34. Wang L, Chen M, Yuan L, Xiang Y, Zhen, R, Zhu S. 14,15-EET promotes mitochondrial biogenesis and protects cortical neurons against oxygen/glucose deprivation-induced apoptosis. Biochem Biophys Res Commun. 2014;450(1):604-9. https://doi.org/10.1016/j.bbrc.2014.06.022 35. Tian X, He W, Yang R, Liu Y. Dl-3-n-butylphthalide protects the heart against ischemic injury and H9c2 cardiomyoblasts against oxidative stress: involvement of mitochondrial function and biogenesis. J Biomed Sci. 2017;24(1):38. https:// doi.org/10.1186/s12929-017-0345-9 36. Raefsky SM, Mattson MP. Adaptive responses of neuronal mitochondria to bioenergetic challenges: roles in neuroplasticity and disease resistance. Free Radic Biol Med. 2017;102:203-16. https://doi.org/10.1016/j.freeradbiomed.2016.11.045 37. Kukat C, Davies KM, Wurm CA, Spåhr H, Bonekamp NA, Kühl I, Joos F, Polosa PL, Park CB, Posse V, Falkenberg M, Jakobs S, Kühlbrandt W, Larsson NG. Cross-strand binding of TFAM to a single mtDNA molecule forms the mitochondrial nucleoid. Proc NatlAcad Sci USA. 2015;112(36):11288-93. https://doi.org/10.1073/pnas.1512131112 38. Machold R, Hayashi S, Rutlin M, Muzumdar MD, Nery S, Corbin JG, Gritli-Linde A, Dellovade T, Porter JA, Rubin LL, Dudek H, McMahon AP, Fishell G. Sonic hedgehog is required for progenitor cell maintenance in telencephalic stem cell niches. Neuron. 2003;39(6):937-50. https://doi.org/10.1016/s0896-6273(03)00561-0 39. Lai K, Kaspar BK, Gage FH, Schaffer DV. Sonic hedgehog regulates adult neural progenitor proliferation in vitro and in vivo. Nat Neurosci. 2003;6(1):21-7. https://doi.org/10.1038/nn983 40. Zhang L, Chopp M, Meier DH, Winter S, Wang L, Szalad A, Lu M, Wei M, Cui Y, Zhang ZG. Sonic hedgehog signaling pathway mediates cerebrolysin-improved neurological function after stroke. Stroke. 2013;44(7):1965-72. https://doi. org/10.1161/STROKEAHA.111.000831 41. He W, Cui L, Zhang C, Zhang X, He J, Xie Y. Sonic hedgehog promotes neurite outgrowth of primary cortical neurons through upregulating BDNF expression. Neurochem Res. 2016;41(4):687-95. https://doi.org/10.1007/s11064-015-1736-5 42. Huang X, Wu W, Hu P, Wang Q. Taurine enhances mouse cochlear neural stem cells proliferation and differentiation to sprial gangli through activating sonic hedgehog signaling pathway. Organogenesis. 2018;14(3):147-57. https://doi. org/10.1080/15476278.2018.1477462 Acta Cir Bras. V38 . e382323 . 2023 13
10.2188_jea.je20190337	Journal of Epidemiology Original Article J Epidemiol 2021;31(10):545-553 Cultural Engagement and Incidence of Cognitive Impairment: A 6-year Longitudinal Follow-up of the Japan Gerontological Evaluation Study (JAGES) Akiho Sugita1, Ling Ling2, Taishi Tsuji3,4, Katsunori Kondo4,5, and Ichiro Kawachi6 1Faculty of Medicine, Chiba University, Chiba, Japan 2Graduate School of Medical and Pharmaceutical Sciences, Chiba University, Chiba, Japan 3Faculty of Health and Sport Sciences, University of Tsukuba, Tokyo, Japan 4Center for Preventive Medical Sciences, Chiba University, Chiba, Japan 5Center for Gerontology and Social Science, National Center for Geriatrics and Gerontology, Aichi, Japan 6Department of Social and Behavioral Sciences, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, United States Received December 20, 2019; accepted July 27, 2020; released online September 19, 2020 ABSTRACT Background: Active engagement in intellectually enriching activities reportedly lowers the risk of cognitive decline; however, few studies have examined this association, including engagement in traditional cultural activities. This study aimed to elucidate the types of cultural engagement associated with lower risk of cognitive impairment. Methods: We examined the association between cultural engagement and cognitive impairment using Cox proportional hazards models in a cohort of 44,985 participants (20,772 males and 24,213 females) aged 65 years or older of the Japan Gerontological Evaluation Study from 2010 to 2016. Intellectual activities (eg, reading books, magazines, and=or newspapers), creative activities (eg, crafts and painting), and traditional cultural activities (eg, poetry composition [haiku], calligraphy, and tea ceremony=ﬂower arrangement) were included among cultural engagement activities. Results: Over a follow-up period of 6 years, incident cognitive disability was observed in 4,198 respondents (9.3%). After adjusting for potential confounders, such as depression and social support, intellectual activities were protectively associated with the risk of cognitive impairment (hazard ratio [HR] for those who read and stated that reading was their hobby, 0.75; 95% conﬁdence interval [CI] 0.66–0.85 and HR for those who read but did not consider reading a hobby, 0.72; 95% CI, 0.65–0.80). Engagement in creative activities was also signiﬁcantly correlated with lower risk of cognitive impairment (crafts: HR 0.71; 95% CI, 0.62–0.81 and painting: HR 0.80; 95% CI, 0.66–0.96). The association between traditional cultural activities and the risk of cognitive impairment was not statistically signiﬁcant. Conclusions: Engagement in intellectual and creative activities may be associated with reduced risk of dementia. Key words: prevention; dementia; cultural activity; reading; hobby Copyright © 2020 Akiho Sugita et al. This is an open access article distributed under the terms of Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. INTRODUCTION Societies experiencing rapid population aging are grappling with the parallel rise in cases of dementia. There has been a rapid growth in the number of people with dementia not only in high-income countries but also low- and middle-income countries.1 In Japan, the situation is even more pressing. The estimated number of people aged 65 or older stood at 35.89 million as of October 2019, accounting for 28.4% of the nation’s total population,2 with both ﬁgures hitting record highs. The share of the older population was the highest among 263 countries and regions in the world.3 According to a government report, in 2012, one in seven older adults aged 65 and above had dementia, with this ﬁgure estimated to reach one in ﬁve people in 2025.4 of Current dementia treatment pharmacological (eg, cholinesterase inhibitors and N-methyl D-aspartate receptor antagonists) is only able to address the relief of symptoms or prevent the progression of dementia; the damage to brain cells is permanent. Therefore, the only viable population approach to dementia is to focus on prevention. Speciﬁcally, modiﬁable factors, such as cognitive training, exercise, and strong social support may be associated with increased brain and cognitive reserve.1 In observational studies, engagement in mentally challenging activities has been suggested to be protectively associated with dementia risk.5–12 The eﬀects of leisure activities (eg, reading, radio and TV, gardening, puzzles, social visits, going to the theater, cinema, and museums)5–11 and computer use11,12 have been evaluated. Address for correspondence. Katsunori Kondo, Center for Preventive Medical Sciences, Chiba University, 1-8-1 Inohana, Chuo Ward, Chiba City, Chiba 260- 8670, Japan (e-mail: [email protected]). DOI https://doi.org/10.2188/jea.JE20190337 HOMEPAGE http://jeaweb.jp/english/journal/index.html 545 Cultural Engagement and Incidence of Cognitive Impairment Figure 1. Participants ﬂow for analytic sample. However, such studies have not been without limitations. For example, these studies did not control for major confounders for cultural engagement, such as depression,6–8 or visual or hearing impairment,6–9,11 as well as receiving=giving social support.5–12 Additionally, some of the “intellectual activities” and=or “cultural activities” examined in previous studies included physical activities (eg, going to the theater, cinema, and museums),5,6 playing games, or simple tasks.7 Since existing studies have shown associations between such activities and dementia or cognitive impairment,9,10,13 we focused on the concept of cultural engagement in this study. As for participation in uniquely Japanese forms of cultural engagement, including poetry composition (haiku), calligraphy, tea ceremony=ﬂower arrangement, the inﬂuences on the risk of dementia remain unclear. Japan has achieved the world’s longest healthy life expectancy and life expectancy for both sexes in 2013.14 Engagement in traditional cultural activities is possibly one of the reasons why Japanese older people can maintain their health. In a cohort of 27 patients with neurocognitive disorders, an intervention based on a ﬂower arrangement program improved dysfunctions in visuospatial memory and recognition in patients.15 However, no study has examined the long-term association between engagement in Japanese cultural activities and the risk of cognitive deterioration among Japanese older adults living almost independent lives at baseline. In this study, we hypothesized that there is an association between cultural engagement (including uniquely Japanese forms of cultural activities) and cognitive impairment. We focused on three speciﬁc aspects: intellectual activities (eg, reading books, magazines, and=or newspapers), creative activities (eg, crafts and painting), and traditional cultural activities (eg, poetry composi- tion [haiku], calligraphy, and tea ceremony=ﬂower arrangement). We sought to investigate whether forms of cultural engagement are associated with lower risk of cognitive impairment, as these associations have not been fully identiﬁed yet. 546 j J Epidemiol 2021;31(10):545-553 METHODS Study sample The Japan Gerontological Evaluation Study (JAGES) is a nationwide cohort study established in 2010 to examine pro- spectively the predictors of healthy aging.16 A total of 95,827 community-dwelling people aged 65 or older in 13 municipalities within Japan were mailed our baseline questionnaire from August 2010 to January 2012. Of the 62,426 individuals who responded to the invitation (response rate 65.1%), 5,739 respondents were removed owing to missing=invalid information about gender and=or age. After excluding an additional 1,444 respondents who did not agree to the use of their data and those with invalid ID numbers, 55,243 participants remained available for analyses. The questionnaire survey inquired about respondents’ personal characteristics, health status, and health habits. As shown in Figure 1, of the 55,243 eligible participants from the baseline survey, we excluded 2,544 who were not independent in their activities of daily living (ADL). We also removed 6,641 respondents who failed to answer the section of the survey asking about their engagement in cultural activities. In order to address the possibility of reverse causality (ie, low engagement in cultural activities being a marker for pre-existing cognitive decline), we performed analyses excluding a further 1,073 respondents whose follow-up periods were ≤1 year. Finally, our longitudinal sample was n = 44,985 (male: n = 20,772, female: n = 24,213). Table 1 summarizes the character- istics of the ﬁnal analytic sample. Males made up 46.2% of the sample; the mean age was 73.6 (standard deviation, 5.81) years. For females, the mean age was 73.9 (standard deviation, 5.97) years. During 242,934 person-years of follow-up (mean 1,971 and maximum 2,328 days), cognitive impairment developed in 4,198 cases (9.3%) (eTable 1). The overall incidence rate was 17.28 per 1,000 person-years. For males, during 110,947 person-years of Sugita A, et al. Table 1. Characteristics of respondents by gender (baseline survey in 2010) Characteristics Level of cognitive impairment All (n = 44,985) Male (n = 20,772) Female (n = 24,213) n % n % n % 0,1 2–7 Total Age, years 65–69 70–74 75–79 79–84 ≥85 Total Educational attainment, years <6 6–9 10–12 ≥13 Missing data Total Equivalent income (time-invariant variable), Japanese yen <2 million 2–4 million ≥4 million Missing data Total Marital status Married Single Missing data Total Employment status Never worked Stopped working Currently working Missing data Total Hypertension No Yes Missing data Total Diabetes No Yes Missing data Total Obesity No Yes Missing data Total Hearing impairment No Yes Missing data Total Visual impairment No Yes Missing data Total 40,787 4,198 44,985 12,928 13,902 10,171 5,541 2,443 44,985 931 20,450 14,932 7,717 955 44,985 18,185 14,686 4,304 7,810 44,985 32,179 11,872 934 44,985 5,137 24,639 9,713 5,496 44,985 16,322 17,899 10,764 44,985 28,684 5,537 10,764 44,985 32,620 1,601 10,764 44,985 31,053 3,168 10,764 44,985 28,136 6,085 10,764 44,985 90.7 9.3 100.0 28.7 30.9 22.6 12.3 5.4 100.0 2.1 45.5 33.2 17.2 2.1 100.0 40.4 32.6 9.6 17.4 100.0 71.5 26.4 2.1 100.0 11.4 54.8 21.6 12.2 100.0 36.3 39.8 23.9 100.0 63.8 12.3 23.9 100.0 72.5 3.6 23.9 100.0 69.0 7.0 23.9 100.0 62.5 13.5 23.9 100.0 18,951 1,821 20,772 6,112 6,409 4,671 2,575 1,005 20,772 284 8,847 6,634 4,676 331 20,772 8,494 7,697 2,170 2,411 20,772 17,911 2,493 368 20,772 821 12,692 5,886 1,373 20,772 7,602 7,956 5,214 20,772 12,447 3,111 5,214 20,772 14,928 630 5,214 20,772 14,023 1,535 5,214 20,772 13,108 2,450 5,214 20,772 91.2 8.8 100.0 29.4 30.9 22.5 12.4 4.8 100.0 1.4 42.6 31.9 22.5 1.6 100.0 40.9 37.1 10.4 11.6 100.0 86.2 12 1.8 100.0 4.0 61.1 28.3 6.6 100.0 36.6 38.3 25.1 100.0 59.9 15 25.1 100.0 71.9 3 25.1 100.0 67.5 7.4 25.1 100.0 63.1 11.8 25.1 100.0 21,836 2,377 24,213 6,816 7,493 5,500 2,966 1,438 24,213 647 11,603 8,298 3,041 624 24,213 9,691 6,989 2,134 5,399 24,213 14,268 9,379 566 24,213 4,316 11,947 3,827 4,123 24,213 8,720 9,943 5,550 24,213 16,237 2,426 5,550 24,213 17,692 971 5,550 24,213 17,030 1,633 5,550 24,213 15,028 3,635 5,550 24,213 90.2 9.8 100.0 28.2 30.9 22.7 12.2 5.9 100.0 2.7 47.9 34.3 12.6 2.6 100.0 40 28.9 8.8 22.3 100.0 58.9 38.7 2.3 100.0 17.8 49.3 15.8 17 100.0 36 41.1 22.9 100.0 67.1 10 22.9 100.0 73.1 4 22.9 100.0 70.3 6.7 22.9 100.0 62.1 15 22.9 100.0 P-value <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Continued on next page: J Epidemiol 2021;31(10):545-553 j 547 Cultural Engagement and Incidence of Cognitive Impairment All (n = 44,985) Male (n = 20,772) Female (n = 24,213) n % n % n % 27,423 1,490 15,328 744 44,985 25,377 12,253 4,779 2,576 44,985 9,176 14,927 19,705 1,177 44,985 2,901 16,915 24,590 579 44,985 27,867 7,831 2,468 6,819 44,985 2,393 41,212 1,380 44,985 2,855 40,387 1,743 44,985 1,985 41,838 1,162 44,985 4,759 37,932 2,294 44,985 10,146 8,739 7,593 10,350 6,371 1,786 44,985 61.0 3.3 34.1 1.7 100.0 56.4 27.2 10.6 5.7 100.0 20.4 33.2 43.8 2.6 100.0 6.4 37.6 54.7 1.3 100.0 61.9 17.4 5.5 15.2 100.0 5.3 91.6 3.1 100.0 6.3 89.8 3.9 100.0 4.4 93.0 2.6 100.0 10.6 84.3 5.1 100.0 22.6 19.4 16.9 23.0 14.2 4.0 100.0 7,579 1,258 11,741 194 20,772 5,203 11,062 4,020 487 20,772 1,161 6,767 12,679 165 20,772 1,161 6,767 12,679 165 20,772 13,346 3,802 1,205 2,419 20,772 1,594 18,580 598 20,772 1,637 18,468 667 20,772 859 19,506 407 20,772 2,062 17,948 762 20,772 6,194 4,209 3,262 3,842 2,639 626 20,772 36.5 6.1 56.5 0.9 100.0 25 53.3 19.4 2.3 100.0 5.6 32.6 61 0.8 100.0 5.6 32.6 61 0.8 100.0 64.2 18.3 5.8 11.6 100.0 7.7 89.4 2.9 100.0 7.9 88.9 3.2 100.0 4.1 93.9 2 100.0 9.9 86.4 3.7 100.0 29.8 20.3 15.7 18.5 12.7 3 100.0 19,844 232 3,587 550 24,213 20,174 1,191 759 2,089 24,213 8,015 8,160 7,026 1,012 24,213 1,740 10,148 11,911 414 24,213 14,521 4,029 1,263 4,400 24,213 799 22,632 782 24,213 1,218 21,919 1,076 24,213 1,126 22,332 755 24,213 2,697 19,984 1,532 24,213 3,952 4,530 4,331 6,508 3,732 1,160 24,213 82 1 14.8 2.3 100.0 83.3 4.9 3.1 8.6 100.0 33.1 33.7 29 4.2 100.0 7.2 41.9 49.2 1.7 100.0 60 16.6 5.2 18.2 100.0 3.3 93.5 3.2 100.0 5 90.5 4.4 100.0 4.7 92.2 3.1 100.0 11.1 82.5 6.3 100.0 16.3 18.7 17.9 26.9 15.4 4.8 100.0 Continued: Characteristics Drinking habit Never drank Stopped drinking Current drinker Missing data Total Smoking habit Never smoked Stopped smoking Current smoker Missing data Total Walking time, minutes <30 30–59 ≥60 Missing data Total Frequency of going out Rarely About once a week Almost daily Missing data Total Depression symptoms (GDS-15 points) ≤4 5–9 ≥10 Missing data Total Receiving emotional support No Yes Missing data Total Providing emotional support No Yes Missing data Total Receiving instrumental support No Yes Missing data Total Providing instrumental support No Yes Missing data Total Frequency of meeting friends Never Once or twice a month About once a week 2–3 times a week Almost daily Missing data Total GDS, Geriatric Depression Scale. +Total may not become 100.0% due to rounding oﬀ. 548 j J Epidemiol 2021;31(10):545-553 P-value <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Sugita A, et al. follow up (mean, 1,950 days), cognitive impairment developed in 1,821 cases (1.6%). The overall incidence rate was 16.4 per 1,000 person-years. For females, during 131,988 person-years of follow up (mean, 1,990 days), cognitive impairment developed in 2,377 cases (1.8%). The overall incidence rate was 18.0 per 1,000 person-years. Outcome variable Our primary outcome was cognitive impairment. Participants in our study were linked to Japan’s Long-Term Care Insurance (LTCI) registry, which includes a standardized in-home assess- ment of cognitive disability.17 Registration in the national LTCI scheme is mandatory, and each applicant requesting long-term care is assessed for eligibility to receive services (eg, home help) by a team of trained investigators dispatched from the certiﬁcation committee in each municipality. During the home visit, each individual is assessed with regard to their ADL and instrumental ADL, cognitive functioning (eg, short-term memory, orientation, and communication), as well as mental and behavioral disorders (eg, delusions of persecution and confab- ulation) using a standardized protocol. Following the assessment, the applicants are classiﬁed into one of 8 levels (0: Independent to 7: Needs constant treatment in a specialized medical facility) according to the severity of their cognitive disability status. The resulting index of cognitive disability is strongly correlated with the Mini-Mental State Examination (Spearman’s rank correlation r = −0.73, P < 0.001)18 and level 1 of the cognitive decline scale has been demonstrated to correspond with a 0.5 point rating on the Clinical Dementia Rating scale (speciﬁcity and sensitivity respectively).19 The initial certiﬁcation is valid for 6 0.88, months, after which periodic re-assessments are conducted every 12 months. In the present study, we deﬁned our outcome as being certiﬁed as level 2 or higher (a state in which a subject at least manifests some symptoms, behaviors, or communication diﬃculties that might hinder daily activities).20,21 Explanatory variable First, participants were asked “Do you have any hobbies, or are you taking any lessons?” If they answered “yes,” they were asked to choose all activities they were engaged in from among the 25 choices mentioned on the questionnaire. Among all the activities included in the questionnaire, we deﬁned cultural engagement as activities that did not involve physical activities (eg, golf, ground golf, gate ball, and walking=jogging), nor playing games, such as mahjong or interacting with the PC. Thus, among intellectual-cultural or cognitive leisure activities described in previous studies,5–12,22 reading books, magazines, and=or newspapers; crafts; and painting were selected as representing forms of cultural engagement. Traditional Japanese cultural activities, including poetry composition (haiku), callig- raphy, and tea ceremony=ﬂower arrangement, also met our criteria. As for reading, we divided respondents into three groups based on the results of the survey on their hobbies and their answers to two questions: “Do you read newspapers?” and “Do you read books or magazines?” The three groups were as follows: 1) those who read books, magazines, and=or newspapers and stated that reading was a hobby; 2) those who read books, magazines, and= or newspapers but stated that reading was not a hobby; and 3) those who did not read (control group). Covariates and mediators Following previous reports, we included basic demographic information, including age (65–69, 70–74, 75–79, 80–84, or ≥85 years),19 educational level (<6 years, 6–9 years, 10–12 years, or ≥13 years),1 household equivalized income (low: <2,000,000 yen, middle: 2,000,000–3,999,999 yen, or high: ≥4,000,000 yen),23 marital status (married, widowed=divorced, or unmarried), and employment status (employed, not working [never been retired]).24 We also included other potentially employed or modiﬁable risk factors for dementia that could inﬂuence par- ticipation in cultural activities (ie, variables that could confound the association between our exposure and outcome), such as hypertension, diabetes, obesity, hearing impairment, smoking habit, physical inactivity, and depression.1 We evaluated physical activity in terms of hours of walking per day and frequency of going out. Depressive symptoms were measured by the Geriatric Depression Scale-15 (GDS-15), with mild depression deﬁned as a score of ≥5 points and severe depression as ≥10 points.25 We included drinking habit as an additional covariate,26 as well as visual impairment because it can aﬀect reading habits. We also assessed variables that could potentially mediate the association between cultural engagement and cognitive impair- ment. Social support is a potential mediator (ie, engagement in cultural activities increases social interaction with other people), thereby raising the probability of receiving=giving social support to others. Following previous JAGES cohort studies, social support was assessed in terms of receiving or providing four variables: emotional support, and receiving or providing instrumental support.20,25,27 We assessed emotional support by asking “Do you have someone who listens to your concerns and complaints?” and “Do you listen to someone’s concerns and complaints?” We measured instrumental support by asking “Do you have someone who looks after you when you are sick and conﬁned to the bed for a few days?” and “Do you look after someone when he=she is sick and conﬁned to the bed for a few days?” We also included frequency of contact with friends (never, once or twice a month, about once a week, two to three times a week, or almost daily). Statistical analysis We calculated descriptive statistics for all variables and conﬁrmed gender diﬀerences through the chi-square test. A Cox propor- tional hazards model was employed to determine the association between cultural engagement variables and incident cognitive impairment. The interaction term between gender and cultural engagement was not statistically signiﬁcant. Therefore, we performed the analyses without gender stratiﬁcation. The cultural engagement variables were added separately. In model 1, we income, statistically adjusted for age, education, equivalent marital status, hypertension, diabetes, obesity, hearing impairment, visual impairment, drinking habit, smoking habit, hours of walking per day, frequency of going out, and depression. In model 2, we added social support and network as potential mediators of the association between cultural engagement and cognitive impairment. employment status, In the analyses, we excluded respondents who developed cognitive impairment within 1 year of the baseline questionnaire to exclude reverse causality. The signiﬁcance level was set at P < 0.05. Statistical analyses were performed using IBM® SPSS® Statistics V25 (IBM Corp, Armonk, NY, USA). J Epidemiol 2021;31(10):545-553 j 549 Cultural Engagement and Incidence of Cognitive Impairment Table 2. Types of cultural engagement Cultural Engagement All (n = 44,985) Male (n = 20,772) Female (n = 24,213) n % n % n % Intellectual Activities Reading books, magazines, and=or newspapers Don’t read Read but it is not a hobby Read as a hobby Creative Activities Crafts Not a hobby Hobby Painting Not a hobby Hobby Traditional Cultural Activities Poetry composition (haiku) Not a hobby Hobby Calligraphy Not a hobby Hobby Tea ceremony=ﬂower arrangement Not a hobby Hobby 2,407 35,413 7,165 41,220 3,765 43,159 1,826 43,861 1,124 43,312 1,673 43,788 1,197 5.4 78.7 15.9 91.6 8.4 95.9 4.1 97.5 2.5 96.3 3.7 97.3 2.7 743 16,432 3,597 20,492 280 20,175 597 20,356 416 20,267 505 20,697 75 3.6 79.1 17.3 98.7 1.3 97.1 2.9 98.0 2.0 97.6 2.4 99.6 0.4 1,664 18,981 3,568 20,728 3,485 22,984 1,229 23,505 708 23,045 1,168 23,091 1,122 6.9 78.4 14.7 85.6 14.4 94.9 5.1 97.1 2.9 95.2 4.8 95.4 4.6 P-value <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Ethical considerations The JAGES protocol was reviewed and approved by the Ethics Committee on Research of Human Subjects at Nihon Fukushi University (approval No. 1005) and the Ethics Committee at the Chiba University Faculty of Medicine (approval No. 2493). RESULTS Gender diﬀerences in cultural engagement The chi-square test revealed the existence of gender diﬀerences in each category (Table 1 and Table 2). Male and female respondents reported diﬀerent activity proﬁles (Table 2). Males in intellectual activities reported slightly higher engagement (eg, reading books, magazines, and=or newspapers) compared to females, regardless of whether reading was a hobby. However, females showed higher involvement in other forms of cultural engagement, including creative activities (eg, crafts and painting) and traditional cultural activities (eg, poetry composition [haiku], calligraphy, and tea ceremony=ﬂower arrangement). Cultural engagement in relation to cognitive impair- ment The results from models 1 and 2 of the Cox proportional hazards analyses are depicted in Table 3, including HRs and 95% CIs for the outcomes and covariates. Model 2 includes the 15 variables from model 1 and an additional 5 variables related to social support and network. Intellectual activities (eg, reading books, magazines, and=or newspapers) were signiﬁcantly related to reduced risk of cognitive impairment after adjustment for all covariates. In model 2, for those who read and stated that reading was their hobby, the HR was 0.75 (95% CI, 0.66–0.85). Those who read but did not consider reading a hobby showed a similar trend in the risk of cognitive impairment (HR 0.72; 95% CI, 0.65–0.80). Figure 2 550 j J Epidemiol 2021;31(10):545-553 Table 3. Risk of cognitive impairment according to frequency of participation in cultural engagement at baseline (exclud- ing respondents whose follow-up periods were :1 year, n = 44,985) Cultural Engagement Model 1 Model 2 Hazard Ratio for Cognitive Impairment (95% CI) Intellectual Activities Reading books, magazines, and=or newspapers Don’t read Read as a hobby Read but it is not a hobby 1.00 0.72 (0.63–0.82) 0.71 (0.64–0.78) 1.00 0.75 (0.66–0.85) 0.72 (0.65–0.80) Creative Activities Crafts Not a hobby Hobby Painting Not a hobby Hobby Traditional Cultural Activities Poetry composition (haiku) Not a hobby Hobby Calligraphy Not a hobby Hobby Tea ceremony=ﬂower arrangement Not a hobby Hobby 1.00 0.69 (0.61–0.79) 1.00 0.71 (0.62–0.81) 1.00 0.78 (0.65–0.94) 1.00 0.80 (0.66–0.96) 1.00 0.93 (0.78–1.12) 1.00 0.97 (0.81–1.17) 1.00 0.91 (0.77–1.08) 1.00 0.93 (0.79–1.10) 1.00 0.96 (0.78–1.18) 1.00 1.00 (0.81–1.22) CI, conﬁdence interval. +Model 1 was adjusted for age, education, equivalent income, marital status, employment status, hypertension, diabetes, obesity, hearing impairment, visual impairment, drinking habit, smoking habit, hours of walking per day, frequency of going out, and depression; model 2 includes the variables in model 1, receiving or providing emotional support, receiving or providing instrumental support, and frequency of meeting friends. Sugita A, et al. P Figure 2. Kaplan-Meier curves for the cumulative risk of developing cognitive impairment according to whether participants read books, magazines, and/or newspapers, and whether or not reading was a hobby. depicts the Kaplan-Meier curves for the cumulative risk of developing cognitive impairment according to whether partic- ipants read books, magazines, and=or newspapers, and whether or not reading was a hobby. Involvement in creative activities such as crafts (HR 0.71; 95% CI, 0.62–0.81) and painting (HR 0.80; 95% CI, 0.66–0.96) was also related to a signiﬁcant decline in the risk of cognitive impairment. Writing short poems (HR 0.93; 95% CI, 0.78–1.12), calligraphy (HR 0.91; 95% CI, 0.77–1.08), and tea ceremonies= ﬂower arrangement (HR 0.96; 95% CI, 0.78–1.18) were associated with reduced HRs of cognitive impairment in model 1; however, none reached statistical signiﬁcance. Engagement in Japanese traditional cultural activities was not signiﬁcantly associated with the risk of cognitive impairment. The development of cognitive impairment according to the number of cultural engagement forms at baseline (eTable 2) and the risk of cognitive impairment according to the number of cultural engagement forms at baseline (eTable 3) are shown in the supplemental materials. DISCUSSION Of the three types of cultural engagement we examined, in a large sample of Japanese older adults, intellectual activities (eg, reading books, magazines, and=or newspapers) and creative activities (eg, crafts and painting) were signiﬁcantly protectively associated in traditional with cognitive impairment, while involvement cultural activities was not signiﬁcantly associated with the risk of cognitive impairment. In the present study, we showed that being engaged in reading books, magazines, and=or newspapers was associated with a lower risk of subsequent cognitive impairment. Participation in crafts or painting was also related to a decreased risk of cognitive impairment. These ﬁndings were robust even after adjustment for potential confounding variables, such as age, educational level, health status, depressive symptoms, physical exercise level, and social capital. This study, conducted using population-based data, there may be potential beneﬁts of cultural suggests engagement activities at the population level. that Our ﬁndings are consistent with previous reports showing the eﬃcacy of reading in the context of cognitive impairment prevention. Although many researchers have grouped reading with intellectual-cultural or cognitive leisure activities, some studies have directly clariﬁed the relationship between reading and dementia. Geda et al reading books was associated with decreased odds of mild cognitive impairment.11 Verghese et al reported that those who read several times per week had a decreased risk of dementia compared to those who read once per week or less frequently.10 reported that Our results suggested that respondents who were engaged in reading had a lower risk of cognitive impairment than those who do not, whether or not reading was a hobby. Reading itself may enhance brain stimulation. However, this cannot be conclusively stated, as the JAGES did not inquire about the frequency or purposes of reading. We showed that those engaged in creative activities (eg, crafts and painting) had a signiﬁcantly lower risk of cognitive impair- ment. Crafts such as knitting, quilting, and pottery are reportedly signiﬁcantly associated with decreased risk of mild cognitive impairment11 or dementia.9 According to a recent survey on healthy aging in Korea,28 an eight-week program combining physical and recreational activity and art, including crafts (making cards, clay pendants, pressed ﬂowers, mandala mobiles, and eco-bags), had positive eﬀects on cognitive function evaluated with the Mini- Mental State Exam-Korean. In our study, arts and crafts programs were particularly favored by females, which is consistent with previous reports,28,29 although the Japanese word for “crafts” on our questionnaire covered knitting, sewing, beadwork, and quilting. Painting has been classiﬁed as producing art,22 or an expressive activity.30 Crafts can be categorized as such because they require creativity. It is reported that a 12-week combined program that included painting helped patients with mild Alzheimer’s disease preserve their global cognitive function and improved their performance on attention tasks.30 Engagement in coloring or painting positively aﬀected behavioral symptoms in patients with dementia in a nursing home.29 Our results suggest that painting and=or making hand-drawn postcards can also help prevent dementia in healthy older people. J Epidemiol 2021;31(10):545-553 j 551 Cultural Engagement and Incidence of Cognitive Impairment It has been reported that a daily routine of 25 minutes of passive ﬁnger exercises incorporating several movements led to improved overall ADL in older adults with dementia.31 Finger activities through crafts or painting might have had a positive eﬀect in helping prevent cognitive impairment. To the best of our knowledge, this is the ﬁrst study focusing on the eﬀects of Japanese traditional cultural engagement on the risk of cognitive impairment. Disappointingly, no signiﬁcant associ- ations were found. Japanese care facilities for older adults often incorporate cultural pursuits for the rehabilitation and maintenance of functional and cognitive capacity.32 We expected engagement in haiku composition, calligraphy, and tea ceremony=ﬂower arrangement to stimulate cognitive functions, as these activities involve mental discipline and training. Although we investigated a large sample of Japanese older people, our analysis revealed no signiﬁcant association between engagement in Japanese traditional cultural activities and cognitive impairment. In theory, traditional cultural activities are enjoyed on special occasions, such as New Year’s, and may often require a signiﬁcant investment of time, money, in terms of cost of lessons, and basic instruments. Therefore, although a considerable number of participants reported engaging in popular Japanese cultural they might not practice these activities regularly. activities, Routine or daily engagement in traditional cultural activities may delay cognitive deterioration. Males reported slightly higher engagement in intellectual activities (eg, reading books, magazines, and=or newspapers) compared to females, regardless of whether reading was a hobby. Females showed higher involvement in creative activities (eg, crafts and painting), and traditional cultural activities (eg, poetry composition [haiku], calligraphy, and tea ceremony=ﬂower arrangement). Therefore, even though the HRs for cognitive impairment are indistinguishable between male and female participants, it is suggested that males beneﬁt more than females from intellectual activities at a population level, whereas females may beneﬁt more than males from creative activities and traditional cultural activities. Gender diﬀerences were described in a prospective study of Swedish twins, which showed that greater participation in intellectual-cultural activities (eg, reading, listening to the radio, or watching television, social visits, and cultural activities, such as going to the theater and cinema) was associated with a lower risk of Alzheimer’s disease in females but not males.8 Recent studies have shown the association between greater participation in leisure activities and a decreased risk of dementia.5,9,10 In the present study, those engaged in at least one cultural activity showed a lower risk of cognitive impairment than those engaged in none. However, the risk of cognitive impairment was almost the same between those engaged in at least one cultural activity and those engaged in two or more (eTable 3). Although the mechanisms that mediate between cultural activities and cognitive impairment remain unclear, acquired hippocampal neurogenesis can be cited to explain our results. Garthe et al demonstrated that mice living in a stimulus-rich, cognitively challenging environment demonstrated improved water maze learning and that they beneﬁted to the extent relevant to adult hippocampal neurogenesis.33 This concept of acquired hippocampal neurogenesis can be supported by clinical or epidemiological studies involving humans.13,22 Cultural engage- ment could also help to preserve cognitive function by promoting social interactions (which have been independently shown to 552 j J Epidemiol 2021;31(10):545-553 prevent the onset of dementia). However, comparing the results of model 1 and model 2, the HRs were almost the same; therefore, we could not ﬁnd evidence of mediation by social support or frequency of meeting friends. Of course, reading is a solitary activity, and we did not expect to see mediation by social support. However, crafts and painting are often performed in the context of social participation. Nevertheless, the HRs did not change after adjusting for social support and network in model 2. Hence, we did not ﬁnd evidence of mediation by social support, leaving the possibility that cultural engagement may enhance brain stimula- tion directly, thereby helping to preserve cognitive function. Despite the importance of the ﬁndings, several limitations of our study must be noted. First, as we deﬁned cognitive impairment or behavioral= impairment based on functional communication diﬃculties resulting from dementia symptoms, we did not classify the types of dementia, such as Alzheimer type, vascular dementia, or other treatable dementia. Second, since our questionnaire consisted of yes=no questions about hobbies and reading habits, we considered neither time commitment nor the frequency of cultural engagement. Finally, we examined one cohort, and veriﬁcation in other cohorts, such as those including other racial or ethnic groups, is required. Conclusion In conclusion, the present analyses demonstrate that certain types of cultural engagement could provide opportunities to prevent dementia in older adults. Intellectual activities, such as reading, were protectively associated with cognitive impairment. It is also possible that creative activities, such as crafts and painting, are related to a reduced risk of cognitive impairment. In the future, there is a need for experimental and prospective longitudinal studies with longer durations and involving detailed assessment of the frequency and contents of cultural engagement and cognitive impairment to demonstrate the mechanisms underlying the results reported here. Increasing opportunities for community participation in these forms of cultural engage- ment through the establishment of clubs and circles may be eﬀective in preventing dementia. ACKNOWLEDGMENTS This study used data from the JAGES, conducted by the Nihon Fukushi University Center for Well-Being and Society. We are extremely grateful to all study participants for the use of their personal data. We would like to express our deepest gratitude to everyone who participated and cooperated in the survey. Authors’ contributions: Data Curation, SA and LL. Formal analysis, SA and LL. Writing-original draft, SA. Writing-review & editing, TT, KK and KI. Supervision, TT, KK and KI. All authors read and approved the ﬁnal version of the manuscript. Conﬂicts of interest: Dr Kondo reports grants from The Ministry of Health, Labour and welfare, grants from Japan Agency for Medical Research and Development (AMED), grants from Japan Science and Technology Agency (JST), grants from Japan Society for the Promotion of Science (JSPS), grants from National Center for Geriatrics and Gerontology, during the conduct of the study; grants from Ryobi Systems co., ltd., grants from RESOL SEIMEI NO MORI, grants from NEC Corporation, outside the submitted work. Funding: This study used data from JAGES (the Japan Gerontological Evaluation Study). This study was supported by Sugita A, et al. (H28-Choju-Ippan-002), and Development JSPS (Japan Society for the Promotion of Science) KAKENHI Grant Number JP15H01972, Health Labour Sciences Research Japan Agency for Medical Grant (JP17dk0110017, Research (AMED) JP18le0110009, JP18ls0110002, JP18dk0110027, JP19dk0110034, JP19dk0110037), Open Innovation Platform with Enterprises, Research Institute and Academia (OPERA, JPMJOP1831) from the Japan Science and Technology (JST). APPENDIX A. SUPPLEMENTARY DATA Supplementary data related to this article can be found at https:== doi.org=10.2188=jea.JE20190337. REFERENCES 1. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. Lancet. 2017;390:2673–2734. 2. Statistics Bureau of Japan. Population estimates by age (ﬁve-year groups) and sex. http:==www.stat.go.jp=english=data=jinsui=2019np= index.html#a15k01-a Accessed 20.07.02. 3. The World Bank. Population ages 65 and above (% of total pop- ulation). https:==data.worldbank.org=indicator=SP.POP.65UP.TO.ZS? most_recent_year_desc=true&view=map= 2019 Accessed 19.09.23. 4. Cabinet Oﬃce. The estimated number of older people suﬀering from dementia (in Japanese). https:==www8.cao.go.jp=kourei=whitepaper= w-2017=html=gaiyou=s1_2_3.html Accessed 20.06.30. 5. Fancourt D, Steptoe A, Cadar D. Cultural engagement and cognitive reserve: museum attendance and dementia incidence over a 10-year period. Br J Psychiatry. 2018;213:661–663. 6. Hassing LB. Gender diﬀerences in the association between leisure activity in adulthood and cognitive function in old age: a prospective longitudinal population-based study. J Gerontol B Psychol Sci Soc Sci. 2020;75:11–20. 7. Lam LC, Ong PA, Dikot Y, et al. Intellectual and physical activities, but not social activities, are associated with better global cognition: a multi-site evaluation of the cognition and lifestyle activity study for seniors in Asia (CLASSA). Age Ageing. 2015;44:835–840. 8. Crowe M, Andel R, Pedersen NL, Johansson B, Gatz M. Does participation in leisure activities lead to reduced risk of Alzheimer’s disease? A prospective study of Swedish twins. J Gerontol B Psychol Sci Soc Sci. 2003;58:P249–P255. 9. Hughes TF, Chang CC, Vander Bilt J, Ganguli M. Engagement in reading and hobbies and risk of incident dementia: the MoVIES project. Am J Alzheimers Dis Other Demen. 2010;25:432–438. 10. Verghese J, Lipton RB, Katz MJ, et al. Leisure activities and the risk of dementia in the elderly. N Engl J Med. 2003;348:2508–2516. 11. Geda YE, Topazian HM, Roberts LA, et al. Engaging in cognitive activities, aging, and mild cognitive impairment: a population-based study. J Neuropsychiatry Clin Neurosci. 2011;23:149–154. 12. Almeida OP, Yeap BB, Alfonso H, Hankey GJ, Flicker L, Norman PE. Older men who use computers have lower risk of dementia. PLoS One. 2012;7:e44239. 13. Tan ZS, Spartano NL, Beiser AS, et al. Physical activity, brain volume, and dementia risk: the Framingham study. J Gerontol A Biol Sci Med Sci. 2017;72:789–795. 14. Tokudome S, Hashimoto S, Igata A. Life expectancy and healthy life expectancy of Japan: the fastest graying society in the world. BMC Res Notes. 2016;9:482. 15. Mochizuki-Kawai H, Kotani I, Mochizuki S, Yamakawa Y. Structured ﬂoral arrangement program beneﬁts in patients with neurocognitive disorder. Front Psychol. 2018;9:1328. 16. Kondo K, Rosenberg M; World Health Organization. Advancing universal health coverage through knowledge translation for healthy ageing: lessons learnt from the Japan Gerontological Evaluation Study. https:==apps.who.int=iris=handle=10665=279010. License: CC BY-NC-SA 3.0 IGO Accessed 20.07.13. 17. Tamiya N, Noguchi H, Nishi A, et al. Population ageing and wellbeing: lessons from Japan’s long-term care insurance policy. Lancet. 2011;378:1183–1192. 18. Hisano S. Kaitei Hasegawa shiki Kan’i Chinou Hyouka Scale (HDS- R), Mini-Mental State Examination (MMSE) to Syōgai Rōjin no Nichijyō Seikatsu Jiritsu do no Kanren ni tsuite [The relationship between Revised Hasegawa Dementia Scale (HDS-R), Mini-Mental State Examination (MMSE) and Bed-fast Scale, Dementia Scale]. Jpn J Geriatr Psychiatry. 2009;20:883–891 (in Japanese). 19. Meguro K, Tanaka N, Kasai M, et al. Prevalence of dementia and dementing diseases in the old-old population in Japan: the Kurihara Project. Implications for long-term care insurance data. Psychogeri- atrics. 2012;12:226–234. 20. Saito T, Murata C, Saito M, Takeda T, Kondo K. Inﬂuence of social relationship domains and their combinations on incident dementia: a prospective cohort study. J Epidemiol Community Health. 2018; 72:7–12. 21. Murata C, Takeda T, Suzuki K, Kondo K. Positive aﬀect and incident dementia among the old. J Epidemiol Res. 2016;2:118–124. 22. Sumowski JF, Rocca MA, Leavitt VM, et al. Searching for the neural basis of reserve against memory decline: intellectual enrichment linked to larger hippocampal volume in multiple sclerosis. Eur J Neurol. 2016;23:39–44. 23. Takasugi T, Tsuji T, Nagamine Y, Miyaguni Y, Kondo K. Socio- economic status and dementia onset among older Japanese: a 6-year prospective cohort study from the Japan Gerontological Evaluation Study. Int J Geriatr Psychiatry. 2019;34:1642–1650. 24. Sommerlad A, Sabia S, Singh-Manoux A, Lewis G, Livingston G. Association of social contact with dementia and cognition: 28-year follow-up of the Whitehall II cohort study. PLoS Med. 2019;16: e1002862. 25. Haseda M, Kondo N, Ashida T, Tani Y, Takagi D, Kondo K. Community social capital, built environment, and income-based inequality in depressive symptoms among older people in Japan: an ecological study from the JAGES Project. J Epidemiol. 2018;28:108– 116. 26. Bellou V, Belbasis L, Tzoulaki I, Middleton LT, Ioannidis JPA, Evangelou E. Systematic evaluation of the associations between environmental risk factors and dementia: an umbrella review of sys- tematic reviews and meta-analyses. Alzheimers Dement. 2017;13: 406–418. 27. Koga C, Hanazato M, Tsuji T, Suzuki N, Kondo K. Elder abuse and social capital in older adults: the Japan Gerontological Evaluation Study. Gerontology. 2020;66:149–159. 28. Kim D. The eﬀects of a combined physical activity, recreation, and art and craft program on ADL, cognition, and depression in the elderly. J Phys Ther Sci. 2017;29:744–747. 29. Cohen-Mansﬁeld J, Marx MS, Dakheel-Ali M, Thein K. The use and utility of speciﬁc nonpharmacological interventions for behavioral symptoms in dementia: an exploratory study. Am J Geriatr Psychiatry. 2015;23:160–170. 30. Viola LF, Nunes PV, Yassuda MS, et al. Eﬀects of a multi- disciplinary cognitive rehabilitation program for patients with mild Alzheimer’s disease. Clinics (Sao Paulo). 2011;66:1395–1400. 31. Liu B, Chen X, Li Y, Liu H, Guo S, Yu P. Eﬀect of passive ﬁnger exercises on grip strength and the ability to perform activities of daily living for older people with dementia: a 12-week randomized controlled trial. Clin Interv Aging. 2018;13:2169–2177. 32. Annear MJ, Otani J, Sun J. Experiences of Japanese aged care: the pursuit of optimal health and cultural engagement. Age Ageing. 2016;45:753–756. 33. Garthe A, Roeder I, Kempermann G. Mice in an enriched learn more ﬂexibly because of adult hippocampal environment neurogenesis. Hippocampus. 2016;26:261–271. J Epidemiol 2021;31(10):545-553 j 553
10.2196_13220	JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Original Paper Brief Motivational Interviewing Delivered by Clinician or Computer to Reduce Sexual Risk Behaviors in Adolescents: Acceptability Study Taraneh Shafii1, MPH, MD; Samantha K Benson2, MPH; Diane M Morrison3, PhD 1Division of Adolescent Medicine, Department of Pediatrics, University of Washington School of Medicine, Seattle, WA, United States 2Harborview Medical Center, University of Washington School of Medicine, Seattle, WA, United States 3School of Social Work, University of Washington, Seattle, WA, United States Corresponding Author: Taraneh Shafii, MPH, MD Division of Adolescent Medicine Department of Pediatrics University of Washington School of Medicine CSB-200 PO Box 5371 Seattle, WA, 98145 United States Phone: 1 206 987 2028 Fax: 1 206 987 3959 Email: [email protected] Abstract Background: Clinicians are expected to screen their adolescent patients for an increasing number of health behaviors and intervene when they uncover risky behaviors, yet, the clinic time allotted to screen, intervene, and provide resources is insufficient. Brief motivational interviewing (MI) offers succinct behavior change counseling; however, for implementation, clinicians need training, skill, and time. Computerized screening and counseling adjuvants may help clinicians increase their scope of behavioral screening, especially with sensitive topics such as sexual health, and provide risk-reduction interventions without consuming provider time during visits. Objective: The objectives of this study were to (1) understand the extent to which health care providers use brief MI for sexual health discussions with adolescent patients and (2) assess the acceptability of incorporating a brief MI-based intervention to reduce sexual risk behaviors into their clinical practice delivered by either themselves or a computer. Methods: At a national medical conference, surveys were administered to clinicians who provide sexual health care to adolescents. They were asked about their current use of MI for sexual risk behavior discussions and their willingness to implement computerized sexual health screening and computerized sexual risk behavior interventions into their clinical practice. Results: The large majority (87.6%, 170/194) of clinicians already used MI with their patients with less than half (72/148, 48.6%) reporting they had been formally trained in MI. Despite all (195/195, 100.0%) clinicians feeling very or completely comfortable discussing sexual risk behaviors with their patients, the large majority (160/195, 82.1%) reported it would be useful, very useful, or extremely useful for a computerized program to do it all: screen their patients, generate risk profiles, and provide the risk-reduction counseling rather than doing it themselves. Conclusions: In this study, most clinicians used some form of brief MI or client-centered counseling when discussing sexual risk behaviors with adolescents and are very comfortable doing so. However, the large majority would prefer to implement computerized sexual health screening, risk assessment, and sexual risk behavior interventions into their clinical care of adolescents. (J Med Internet Res 2019;21(7):e13220) doi: 10.2196/13220 KEYWORDS sexual health; risk behaviors; adolescent; healthcare providers; computer-assisted diagnosis; teen health; preventive care http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Introduction Background Clinician sexual health discussions with adolescents remain suboptimal in real-world clinical practice [1-4]. Health care providers continue to search for optimal ways to communicate with adolescents about sexual health and risk behaviors. Some behavioral interventions have been shown to increase the knowledge of risk-reduction strategies (eg, condom and birth control use and negotiating safe sex with partners) and decrease self-reported unprotected sex; however, these interventions were tested in nonclinical settings and with specific populations of adolescents. Such interventions have yet to be tested or implemented in real-world outpatient settings and delivered by clinicians [5-9]. Barriers to Screening Even experienced clinicians in busy practices may not have the time to engage adolescents in discussions, which are needed to build rapport and uncover risk behaviors. Adolescents may have concerns about talking face-to-face with clinicians about sex or may not be granted enough time for confidential conversations during their visit [10-12]. Brief motivational interviewing (MI) has gained popularity as a means to engage adolescents in behavior change [13-20]; however, there are barriers to clinicians in adopting MI. It takes time to be trained and become proficient in MI, and effectively using MI requires already precious clinic visit time [21-22]. Computer-Assisted Screening Computerized screening with brief MI may serve to alleviate the time burden for health care providers and any discomfort in discussing sensitive health topics for both the clinician and patient. Computer screening improves adolescents’ perceptions of medical visits [23-25]. The literature provides evidence that adolescents may prefer computerized sexual health screening to face-to-face interviews. A study of adolescents seeking care in a pediatric emergency department tested computerized sexual health screening and found that it was acceptable to adolescents, preferable to in-person interviews, and feasible for providers to implement in the emergency department [26]. A personal digital assistant screening tool that screened for several risk behaviors, including unprotected sex, was tested in primary care clinics before adolescent well visits and resulted in higher patient ratings for visit satisfaction, perceived confidentiality, and feeling listened to carefully [23]. Computerized Interventions Incorporating sexual behavior risk-reduction interventions into the computerized screening session takes these interventions one step further. Such interventions may be interactive and provide personalized feedback to the adolescent. Only a few computerized sexual health interventions for adolescents have been tested in real-world clinic settings, and these did not assess clinician acceptability of integrating the interventions into clinical practice [27-30]. Existing provider acceptability studies of computerized health screening and interventions are of adult patient populations, have small sample sizes, and may not include sexual health as a risk behavior [31-33]. We are not aware of any large studies assessing clinician willingness to be trained in brief MI for promoting adolescent sexual health. We were likewise unable to identify any studies assessing provider acceptability of incorporating computerized sexual health screening and interventions into visits with their adolescent patients. The objectives of this study were (1) to understand the extent to which health care providers use MI for sexual health with adolescent patients and (2) to assess the acceptability of incorporating a brief MI-based intervention to reduce sexual risk behaviors into their clinical practice delivered by either themselves or a computer. Methods Recruitment In March 2009, we administered a 28-item survey to clinicians at a national medical conference, with attendees representing a wide geographic range in the United States. For the purposes of this study, clinicians were asked about sexually transmitted infection (STI) testing and positive STI diagnoses in the past 3 months to characterize their patient population and practice experience. The inclusion criteria were clinicians practicing in the United States who provided sexual and reproductive health care to adolescents. The exclusion criteria were not providing such care to adolescents or being in training. A total of 365 surveys were initially distributed and 18 were omitted from the final denominator (n=347) for the following reasons: the attendee returned the survey blank (n=8); the survey was lost by the participant and a replacement survey was provided (n=8); or the survey was defective because of printing error and was replaced with a corrected survey (n=2). Of the 347 surveys distributed, 81.8% (284/347) were completed. The University of Washington Human Subjects Division approved this study. Sample for Analysis Of the 284 completed surveys, an additional 88 were ineligible because the attendee was in training (n=45), practicing outside the United States (n=10), not providing sexual and reproductive healthcare or not in practice or nonclinical (n=30), or did not indicate their degree or level of training (n=3). A total of 196 clinicians qualified for the study as they provided such care to adolescents, including the diagnosis and management of STIs and unintended pregnancy; identified themselves as medical doctor (MD)/doctor of osteopathy (DO), physician assistant (PA), or nurse practitioner (NP) and were not currently in training (Figure 1). STATA 11.0 by StataCorp LLC, was used for data analysis. T tests and chi-square tests were used to evaluate associations. http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Figure 1. Flow diagram of sample for analysis. Results Demographic and Clinical Practice Characteristics Physicians (MDs and DOs) comprised the majority of the sample (181/196, 92.3%), with (13/196, 6.6%) NPs and (2/196, 1.0%) PAs. In addition, 65.3% (128/196) of them were female. The most common practice type was academic (121/196, 61.7%) with a comparable proportion in private practice (19/196, 9.6%) and public health/community clinics (18/196, 9.2%). Over half (118/195, 60.1%) of the clinicians provided care only to adolescents (defined as ages 11 to 21 years). The number of years providers reported being in clinical practice ranged from less than 1 year to greater than 30 years and was evenly distributed across decades of practice. Through the use of Likert scales, we elicited information about clinical practice and patient population characteristics. The large majority (175/196, 89.3%) of clinicians provided care to 10 or more patients in a typical week. In the past 3 months, the majority (160/195, 82.1%) of clinicians tested 10 or more of their patients for STIs (Neisseria gonorrhoeae, Chlamydia trachomatis, trichomonas, genital warts, syphilis, and HIV) with the vast majority (178/196, 90.8%) of clinicians reporting 1 or more adolescents testing positive for an STI in the same time period (Table 1). Current Use of Motivational Interviewing in Clinical Practice All clinicians reported that they felt at least very comfortable and a large majority felt completely comfortable (195/195, 100.0%; missing date, n=1) discussing sexual risk behaviors (eg, inconsistent or lack of condom use/hormonal birth control; multiple partners/concurrency; HIV/STI; unintended pregnancy) with their patients. Although all except 3 clinicians reported feeling that they were at least somewhat effective in changing sexual risk behaviors of their adolescent patients, only 22.1% (43/195) felt very or completely effective. Many clinicians reported they saw themselves as more effective at changing patient behaviors than other clinicians. The vast majority of clinicians were familiar with MI defined in the survey as “… a directive, client-centered counseling style for eliciting behavior change by helping to explore and resolve ambivalence” [34]. The large majority of clinicians (170/194, 87.6%) already used MI with their patients. These 170 clinicians were asked if they were formally trained in MI and of 148 (missing data=22) respondents, less than half, 48.6% (72/148), reported formal training. Clinicians reported having used MI for many health topics, including obesity (93%), smoking (90%), alcohol (82%), substance abuse (87%), and sexual health (96%). Only half (52%) of the clinicians said they used it to discuss injury prevention (bike helmets/seat belts). Most clinicians (140/170, 82.4%) employ MI greater than half the time when discussing sexual health with their patients and feel they are more effective in communicating with their patient when they use MI compared with when they do not (Table 2). Motivational Interviewing Acceptability and Feasibility in Practice The vast majority of clinicians would be willing to use MI or another type of client-centered counseling technique in their practice, if effective at reducing sexual risk behaviors in adolescents. Although 93% of them were willing to attend training for such an intervention, they preferred the length of training be limited to less than 1 day. Clinicians are willing to spend a maximum of 10 min per patient to deliver the intervention. Approximately half the clinicians (103/195, 52.8%) were willing to have follow-up contact with their patients as part of the sexual health intervention. However, 21.0% (41/195) of them would only do so if reimbursed. Clinicians were willing to do at least 1 monthly follow-up with their patients lasting less than 10 min per encounter. Preferred modes of follow-up in order of preference were telephone, email, text message, and social media (Table 3). http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Table 1. Demographic characteristics of clinicians (N=196). Demographic characteristic Statistics, n (%) Gender Female Male Clinician type Medical doctor Nurse practitioner/physician assistant Years in clinical practice (N=193), n (%) ≤10 11-20 >20 Patient type (N=195), n (%) Adolescents only (aged 11-21 years) Children and adolescents (aged 0-21 years) All ages (0 through adulthood) Adolescents and adults (aged 11 years through adult) Practice type (N=195), n (%) Academic Private Community/public health Other Adolescent patients per week <10 10-29 ≥30 STIa tests on patients per 3 months (N=195), n (%) <10 10-29 ≥30 Positive STIa tests per 3 months <10 10-29 ≥30 128 (65.3) 68 (34.7) 181 (92.4) 15 (7.6) 67 (34.7) 64 (33.2) 62 (32.1) 118 (60.5) 30 (15.4) 19 (9.7) 28 (14.4) 121 (62.0) 20 (10.3) 23 (11.8) 31 (15.9) 21 (10.7) 87 (44.4) 88 (44.9) 35 (17.9) 59 (30.2) 101 (51.9) 123 (62.7) 55 (28.1) 18 (9.2) aSTI: sexually transmitted infection, including chlamydia, gonorrhea, trichomonas, herpes, genital warts, syphilis, and HIV. http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Table 2. Clinician perspective of sexual risk behaviors and motivational interviewing (MI; N=196). Clinician perspectives Statistics, n (%) Comfort talking about sexual risk behaviorsa (N=195), n (%) Not comfortable Somewhat comfortable Comfortable Very comfortable Completely comfortable Clinician effectiveness in changing sexual risk behaviors (N=194), n (%) Not effective Somewhat effective Effective Very effective Completely effective Personal effectiveness in changing behavior (N=195), n (%) Not effective Somewhat effective Effective Very effective Completely effective Use of MI with patient Yes No Formally trained in MI (N=170)a,b (n=148), n (%) Yes No Types of behavioral issues addressed Sexual risk behavior Obesity Smoking cigarettes Drinking alcohol Substance abuse Injury prevention (bike helmets/seat belts) Frequency of use of MI with patients for sexual risk behavior Never 25% of time 50% of time 75% of time Almost always Provider effectiveness in changing behavior when using MI versus when not (N=163), n (%) Much less effective Somewhat less effective No difference http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX 0 (0) 0 (0) 0 (0) 27 (13.8) 168 (86.2) 8 (4.1) 89 (45.9) 68 (35.0) 29 (15.0) 0 (0) 3 (1.5) 76 (39.0) 73 (37.5) 42 (21.5) 1 (0.5) 170 (87.6) 24 (12.4) 72 (48.6) 76 (51.4) 163 (96.4) 155 (92.8) 147 (89.6) 133 (82.1) 140 (87.0) 83 (52.2) 2 (1.2) 28 (16.5) 41 (24.1) 53 (31.2) 46 (27.0) 1 (0.6) 4 (2.4) 22 (13.5) J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Clinician perspectives More effective Much more effective aNo difference by number practice years or frequency sexually transmitted infection testing. bRemaining survey questions only asked if ever used motivational interviewing, n=170. Table 3. Clinician perspective of self-delivered motivational interviewing (MI). Clinician perspective Willing to attend MI training (N=186), n (%) Yes No Maximum length training (N=177), n (%) ≤2 hours Half day 1 day ≥2 days Maximum length of MI session with patient (N=192), n (%) ≤5 min 5 min 10 min 15 min ≥20 min Feasible for clinician follow-up with patient (N=195), n (%) Yes No Only if reimbursed Maximum length of MI follow-up session (N=152), n (%) ≤5 min 10 min 15 min ≥20 min Maximum number of monthly follow-up contacts (N=158), n (%) 1 2 3 4-5 6 Follow-up method willing to use, n/N (%) Phone call Text message Email Social media Statistics, n (%) 120 (73.6) 16 (9.8) Statistics 174 (93.6) 12 (6.4) 22 (12.4) 58 (32.8) 54 (30.5) 43 (24.3) 16 (8.3) 62 (32.3) 67 (34.9) 30 (15.6) 17 (8.9) 103 (52.8) 51 (26.2) 41 (21.0) 99 (65.1) 31 (20.4) 12 (7.9) 10 (6.6) 30 (19.0) 39 (24.7) 23 (14.6) 29 (18.3) 37 (23.4) 122/149 (81.9) 90/144 (62.5) 124/149 (83.2) 30/139 (21.6) http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Computer-Delivered Risk-Reduction Acceptability The large majority of clinicians (165/192, 85.9%) found it more feasible for a computer to provide the sexual risk behavior screening intervention to their patients rather than themselves and would use a computer-generated sexual risk profile printout to facilitate discussion with their patients. The large majority of clinicians (160/195, 82.1%) also thought it would be useful, very useful, or extremely useful for the computer do it all: screen their patient, generate their sexual risk profile, and provide the risk-reduction counseling itself, requiring the provider to review only the findings with their patients afterward. Preference for Computerized Risk Screening and Risk-Reduction Counseling No associations were found when comparing the number of years in clinical practice and comfort discussing sexual risk behaviors with adolescents; being trained in MI; or preferring computerized sexual risk screening and risk-reduction counseling. There was also no association between preference for computerized risk screening and counseling by clinician gender, type of practice, number of patients seen per week, and number of patients tested or testing positive for an STI in the past 3 months (Table 4). Table 4. Clinician perspective of motivational interviewing (MI) and computer-delivered risk reduction for sexual health (N=196). Clinician perspective Statistics, n (%) If MI sexual behavior risk reduction effectively delivered via clinician would it be feasible for you to do yourself?a,b Yes No 183 (95.8) 8 (4.2) If sexual behavior risk reduction effectively delivered via computer would that be more feasible for you than doing it yourself?b,c Yes No 165 (85.9) 27 (14.1) Likeliness to use computer printout of sexual risk behavior profile to facilitate risk-reduction counseling Not likely Somewhat likely Likely Very likely Extremely likely 8 (4.1) 34 (17.3) 59 (30.1) 68 (34.7) 27 (13.8) How useful would it be for you if computer generated a printout of sexual risk behavior profile AND provided risk-reduction counseling requiring you to do nothing further OR to simply review the findings with your adolescent patients?b,d Not useful Somewhat useful Useful Very useful Extremely useful 7 (3.6) 28 (14.4) 53 (27.2) 59 (30.3) 48 (24.6) an=191. bNo difference by number of years in clinical practice or frequency of sexually transmitted infection testing. cn=192. dn=195. Discussion Principal Findings In a survey of clinicians who provide sexual health care to adolescents from varied geographic regions around the United States and a wide range of clinical experience, the vast majority reported being very comfortable discussing sexual health with their adolescent patients. The majority of clinicians reported using MI for sexual health counseling with their patients, although less than half of these reported formally training in MI. Surprisingly, this sample of clinicians espousing such comfort with adolescent sexual health discussions reported that http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX it would be preferable for a computer to do it all: screen for sexual risk behaviors and provide their patients with risk-reduction counseling. Comparison With Previous Work There has been increasing focus on sexual health screening and MI in medical school curricula over the past 2 decades [35-40]. Other studies have found younger clinicians to be more comfortable discussing sexual health and using MI for behavior change as compared with older providers [41,3,1]. However, in this survey, providers with more than 30 years of clinical experience were just as likely as those with less than 10 years of clinical experience to report comfort in talking about sexual J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al health and using MI with patients. The similarity in comfort across respondents with different practice longevity could be because of most clinicians in the study primarily taking care of adolescents and so were comfortable with the population. Also most clinicians were in academic practice, and may be early adopters of evolving clinical practice approaches over the years. Most providers felt they were at least somewhat effective at influencing the sexual behavior of their patients. This sentiment echoes a qualitative study, with physicians reporting they had influence in the choice of contraception with their female patients [42]. In our study, most respondents considered themselves more effective when using MI than when not, and the majority of them considered themselves more effective at encouraging patients’ behavior change compared with other clinicians. Although there were no studies found in the literature that addressed providers’ perceptions of their effectiveness with MI, there are existing studies that demonstrate clinician use of MI for behavior change to be efficacious in changing health risk behaviors [13-18]. Although respondents were very comfortable discussing sex with adolescents and even felt they were effective at eliciting behavior change, they considered it more feasible for a computer to administer the screening and counseling rather than doing so themselves. To our knowledge, this finding is novel in the literature. For this population of providers, preferring a computerized approach to sexual health risk-reduction counseling may reflect time limitations for patient visits rather than reticence to discuss sexual health. Most providers were willing to be trained in an MI sexual risk behavior intervention that includes at least 1 follow-up session; however, 20% of providers indicated they would only do a follow-up session with patients if they were reimbursed, which may also reflect increasing pressures on clinicians for productivity. Limitations A limitation of this study is that we did not define MI in detail or what is required for training and proficiency in true MI. In the survey, we defined MI as “… a directive, client-centered counseling style for eliciting behavior change by helping clients explore and resolve ambivalence” [34]. It is possible that participants have different definitions for and experience in the use of MI, which may have biased the responses to questions about the use of and training in MI. In addition, the proposed computerized screening and intervention was theoretical, so clinicians were not providing feedback on a tangible product for which they may have different opinions. As most clinicians practiced in academic settings, the findings may not be generalizable to clinicians in other practice types. The decision was made to focus on clinicians practicing in the United States to account for the large variation worldwide in attitudes toward adolescent sexual and reproductive health and clinical practice. The authors acknowledge that this was a missed opportunity to learn about international clinician practice. MI has gained increasing popularity over the past decade since this study. Brief MI is used for many different health behaviors, and we anticipate an even higher acceptability by medical providers. However, the issue of lack of provider time with patients has also escalated over the past decade. These data are relevant as providers have not yet found an answer and continue to strategize on how they can provide comprehensive health care in the limited minutes they have for adolescent patient visits. Such an intervention as presented in this study is a possible solution. Conclusions Clinicians are increasingly pressed for time when providing care to patients and researchers and practitioners have not yet found the most effective way to consistently discuss sexual health with adolescents or promote healthy sexual behaviors. Computerized interventions, which incorporate both behavioral screening and risk-reduction counseling, may provide solutions to both issues. The development of computerized health interventions is a rapidly growing field and further research is needed to create and test such interventions in real-world clinical practice. Acknowledgments The authors would like to acknowledge the Funding Agency: Eunice Kennedy Shriver National Institute of Child Health and Human Development NICHD Grant No: 5K23HD052621, eGC1 Number: A66661; Project Title: Brief Clinician Intervention for High-Risk Behaviors in Adolescents, Grant Principal Investigator: TS; Project Period Dates: 7/1/07-5/31/13. All of those who have contributed significantly to this study are listed. This study was presented as a poster: TS, DMM, MRG, and KKH. “Acceptability to health care providers of using motivational interviewing delivered via clinician or computer in clinical practice to reduce sexual risk behaviors in adolescents,” at the 2013 Annual Meeting of the Society for Adolescent Health and Medicine, Atlanta, Georgia March 2013. Conflicts of Interest None declared. References 1. Alexander SC, Fortenberry JD, Pollak KI, Bravender T, Davis JK, Ostbye T, et al. Sexuality talk during adolescent health maintenance visits. JAMA Pediatr 2014 Feb;168(2):163-169 [FREE Full text] [doi: 10.1001/jamapediatrics.2013.4338] [Medline: 24378686] http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al 2. Marcell AV, Bell DL, Lindberg LD, Takruri A. Prevalence of sexually transmitted infection/human immunodeficiency 3. 4. 5. 6. 7. 8. virus counseling services received by teen males, 1995-2002. J Adolesc Health 2010 Jun;46(6):553-559 [FREE Full text] [doi: 10.1016/j.jadohealth.2009.12.002] [Medline: 20472212] Kelts EA, Allan MJ, Klein JD. Where are we on teen sex?: Delivery of reproductive health services to adolescents by family physicians. Fam Med 2001 May;33(5):376-381. [Medline: 11355649] Blum RW, Beuhring T, Wunderlich M, Resnick MD. Don't ask, they won't tell: the quality of adolescent health screening in five practice settings. Am J Public Health 1996 Dec;86(12):1767-1772. [Medline: 9003135] Jemmott JB, Jemmott LS, Fong GT, Morales KH. Effectiveness of an HIV/STD risk-reduction intervention for adolescents when implemented by community-based organizations: a cluster-randomized controlled trial. Am J Public Health 2010 Apr;100(4):720-726. [doi: 10.2105/AJPH.2008.140657] [Medline: 20167903] Jemmott JB, Jemmott LS, O'Leary A, Ngwane Z, Icard LD, Bellamy SL, et al. School-based randomized controlled trial of an HIV/STD risk-reduction intervention for South African adolescents. Arch Pediatr Adolesc Med 2010 Oct;164(10):923-929 [FREE Full text] [doi: 10.1001/archpediatrics.2010.176] [Medline: 20921349] DiClemente RJ, Wingood GM, Harrington KF, Lang DL, Davies SL, Hook EW, et al. Efficacy of an HIV prevention intervention for African American adolescent girls: a randomized controlled trial. J Am Med Assoc 2004 Jul 14;292(2):171-179. [doi: 10.1001/jama.292.2.171] [Medline: 15249566] Heeren GA, Jemmott JB, Ngwane Z, Mandeya A, Tyler JC. A randomized controlled pilot study of an HIV risk-reduction intervention for sub-Saharan African university students. AIDS Behav 2013 Mar;17(3):1105-1115 [FREE Full text] [doi: 10.1007/s10461-011-0129-2] [Medline: 22246515] 9. Morrison-Beedy D, Jones SH, Xia Y, Tu X, Crean HF, Carey MP. Reducing sexual risk behavior in adolescent girls: results from a randomized controlled trial. J Adolesc Health 2013 Mar;52(3):314-321 [FREE Full text] [doi: 10.1016/j.jadohealth.2012.07.005] [Medline: 23299011] 10. McKee M, Fletcher J, Schechter CB. Predictors of timely initiation of gynecologic care among urban adolescent girls. J Adolesc Health 2006 Aug;39(2):183-191. [doi: 10.1016/j.jadohealth.2005.11.022] [Medline: 16857529] 11. Hoopes AJ, Benson SK, Howard HB, Morrison DM, Ko LK, Shafii T. Adolescent perspectives on patient-provider sexual health communication: a qualitative study. J Prim Care Community Health 2017 Oct;8(4):332-337 [FREE Full text] [doi: 10.1177/2150131917730210] [Medline: 28929860] Fairbrother G, Scheinmann R, Osthimer B, Dutton MJ, Newell KA, Fuld J, et al. Factors that influence adolescent reports of counseling by physicians on risky behavior. J Adolesc Health 2005 Dec;37(6):467-476. [doi: 10.1016/j.jadohealth.2004.11.001] [Medline: 16310124] 12. 13. Gayes LA, Steele RG. A meta-analysis of motivational interviewing interventions for pediatric health behavior change. J Consult Clin Psychol 2014 Jun;82(3):521-535. [doi: 10.1037/a0035917] [Medline: 24547922] 14. Brown RA, Abrantes AM, Minami H, Prince MA, Bloom EL, Apodaca TR, Hunt JI. Motivational Interviewing to Reduce 15. Substance Use in Adolescents with Psychiatric Comorbidity. J Subst Abuse Treat 2015;59:20-29. [Medline: 26362000] Stanger C, Ryan SR, Delhey LM, Thrailkill K, Li Z, Budney AJ. A multicomponent motivational intervention to improve adherence among adolescents with poorly controlled type 1 diabetes: a pilot study. J Pediatr Psychol 2013 Jul;38(6):629-637 [FREE Full text] [doi: 10.1093/jpepsy/jst032] [Medline: 23699750] 16. Gourlan M, Sarrazin P, Trouilloud D. Motivational interviewing as a way to promote physical activity in obese adolescents: a randomised-controlled trial using self-determination theory as an explanatory framework. Psychol Health 2013 Nov;28(11):1265-1286. [doi: 10.1080/08870446.2013.800518] [Medline: 23756082] 17. Walton M, Chermack ST, Shope JT, Bingham CR, Zimmerman MA, Blow FC, et al. Effects of a brief intervention for reducing violence and alcohol misuse among adolescents: a randomized controlled trial. J Am Med Assoc 2010 Aug 4;304(5):527-535 [FREE Full text] [doi: 10.1001/jama.2010.1066] [Medline: 20682932] 18. Cunningham RM, Chermack ST, Zimmerman MA, Shope JT, Bingham CR, Blow FC, et al. Brief motivational interviewing intervention for peer violence and alcohol use in teens: one-year follow-up. Pediatrics 2012 Jun;129(6):1083-1090 [FREE Full text] [doi: 10.1542/peds.2011-3419] [Medline: 22614776] 19. Gold MA, Tzilos GK, Stein LA, Anderson BJ, Stein MD, Ryan CM, et al. A randomized controlled trial to compare computer-assisted motivational intervention with didactic educational counseling to reduce unprotected sex in female adolescents. J Pediatr Adolesc Gynecol 2016 Feb;29(1):26-32 [FREE Full text] [doi: 10.1016/j.jpag.2015.06.001] [Medline: 26514957] 20. Barnet B, Liu J, DeVoe M, Duggan AK, Gold MA, Pecukonis E. Motivational intervention to reduce rapid subsequent births to adolescent mothers: a community-based randomized trial. Ann Fam Med 2009;7(5):436-445 [FREE Full text] [doi: 10.1370/afm.1014] [Medline: 19752472] 21. McKee MD, Rubin SE, Campos G, O'Sullivan LF. Challenges of providing confidential care to adolescents in urban primary care: clinician perspectives. Ann Fam Med 2011;9(1):37-43 [FREE Full text] [doi: 10.1370/afm.1186] [Medline: 21242559] 22. Emmons KM, Rollnick S. Motivational interviewing in health care settings. Opportunities and limitations. Am J Prev Med 2001 Jan;20(1):68-74. [Medline: 11137778] http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al 23. Olson AL, Gaffney CA, Hedberg VA, Gladstone GR. Use of inexpensive technology to enhance adolescent health screening and counseling. Arch Pediatr Adolesc Med 2009 Feb;163(2):172-177. [doi: 10.1001/archpediatrics.2008.533] [Medline: 19188650] 24. Brown JD, Wissow LS. Discussion of sensitive health topics with youth during primary care visits: relationship to youth perceptions of care. J Adolesc Health 2009 Jan;44(1):48-54 [FREE Full text] [doi: 10.1016/j.jadohealth.2008.06.018] [Medline: 19101458] 25. Zieve GG, Richardson LP, Katzman K, Spielvogle H, Whitehouse S, McCarty CA. Adolescents' perspectives on personalized e-feedback in the context of health risk behavior screening for primary care: qualitative study. J Med Internet Res 2017 Dec 20;19(7):e261 [FREE Full text] [doi: 10.2196/jmir.7474] [Medline: 28729236] 27. 26. Goyal MK, Shea JA, Hayes KL, Badolato G, Chamberlain JM, Zaoutis T, et al. Development of a sexual health screening tool for adolescent emergency department patients. Acad Emerg Med 2016 Dec;23(7):809-815 [FREE Full text] [doi: 10.1111/acem.12994] [Medline: 27126128] Peipert J, Redding CA, Blume J, Allsworth JE, Iannuccillo K, Lozowski F, et al. Design of a stage-matched intervention trial to increase dual method contraceptive use (Project PROTECT). Contemp Clin Trials 2007 Sep;28(5):626-637. [doi: 10.1016/j.cct.2007.01.012] [Medline: 17374567] Peipert JF, Redding CA, Blume JD, Allsworth JE, Matteson KA, Lozowski F, et al. Tailored intervention to increase dual-contraceptive method use: a randomized trial to reduce unintended pregnancies and sexually transmitted infections. Am J Obstet Gynecol 2008 Jun;198(6):630.e1-630.e8 [FREE Full text] [doi: 10.1016/j.ajog.2008.01.038] [Medline: 18395692] 28. 29. Downs JS, Murray PJ, de Bruin WB, Penrose J, Palmgren C, Fischhoff B. Interactive video behavioral intervention to reduce adolescent females' STD risk: a randomized controlled trial. Soc Sci Med 2004 Oct;59(8):1561-1572. [doi: 10.1016/j.socscimed.2004.01.032] [Medline: 15279915] 30. DiClemente RJ, Wingood GM, Rose ES, Sales JM, Lang DL, Caliendo AM, et al. Efficacy of sexually transmitted disease/human immunodeficiency virus sexual risk-reduction intervention for african american adolescent females seeking sexual health services: a randomized controlled trial. Arch Pediatr Adolesc Med 2009 Dec;163(12):1112-1121. [doi: 10.1001/archpediatrics.2009.205] [Medline: 19996048] 31. Mackenzie SL, Kurth AE, Spielberg F, Severynen A, Malotte CK, St Lawrence J, et al. Patient and staff perspectives on the use of a computer counseling tool for HIV and sexually transmitted infection risk reduction. J Adolesc Health 2007 Jun;40(6):572.e9-572.16. [doi: 10.1016/j.jadohealth.2007.01.013] [Medline: 17531766] Paul CL, Carey M, Yoong SL, D'Este C, Makeham M, Henskens F. Access to chronic disease care in general practice: the acceptability of implementing systematic waiting-room screening using computer-based patient-reported risk status. Br J Gen Pract 2013 Sep;63(614):e620-e626 [FREE Full text] [doi: 10.3399/bjgp13X671605] [Medline: 23998842] 33. Ahmad F, Skinner HA, Stewart DE, Levinson W. Perspectives of family physicians on computer-assisted health-risk 32. assessments. J Med Internet Res 2010 May 7;12(2):e12 [FREE Full text] [doi: 10.2196/jmir.1260] [Medline: 20457555] 34. Rollnick S, Miller WR. What is motivational interviewing? Behav Cogn Psychother 2009 Jun 16;23(4):325-334. [doi: 10.1017/S135246580001643X] 35. Coleman E. Sexual health education in medical school: a comprehensive curriculum. Virtual Mentor 2014 Nov 1;16(11):903-908. [doi: 10.1001/virtualmentor.2014.16.11.medu1-1411] [Medline: 25397650] 36. Bayer CR, Eckstrand KL, Knudson G, Koehler J, Leibowitz S, Tsai P, et al. Sexual health competencies for undergraduate medical education in North America. J Sex Med 2017 Dec;14(4):535-540. [doi: 10.1016/j.jsxm.2017.01.017] [Medline: 28202322] 37. Galletly C, Lechuga J, Layde JB, Pinkerton S. Sexual health curricula in U.S. medical schools: current educational objectives. 38. Acad Psychiatry 2010;34(5):333-338. [doi: 10.1176/appi.ap.34.5.333] [Medline: 20833900] Poirier MK, Clark MM, Cerhan JH, Pruthi S, Geda YE, Dale LC. Teaching motivational interviewing to first-year medical students to improve counseling skills in health behavior change. Mayo Clin Proc 2004 Mar;79(3):327-331. [doi: 10.4065/79.3.327] [Medline: 15008606] 39. Haeseler F, Fortin AH, Pfeiffer C, Walters C, Martino S. Assessment of a motivational interviewing curriculum for year 3 medical students using a standardized patient case. Patient Educ Couns 2011 Jul;84(1):27-30 [FREE Full text] [doi: 10.1016/j.pec.2010.10.029] [Medline: 21123019] 40. White LL, Gazewood JD, Mounsey AL. Teaching students behavior change skills: description and assessment of a new Motivational interviewing curriculum. Med Teach 2007 May;29(4):e67-e71. [doi: 10.1080/01421590601032443] [Medline: 17786734] 41. Henry-Reid LM, O'Connor KG, Klein JD, Cooper E, Flynn P, Futterman DC. Current pediatrician practices in identifying high-risk behaviors of adolescents. Pediatrics 2010;125(4):e741-e747 [FREE Full text] 42. Henderson JT, Raine T, Schalet A, Blum M, Harper CC. "I wouldn't be this firm if I didn't care": preventive clinical counseling for reproductive health. Patient Educ Couns 2011 Feb;82(2):254-259 [FREE Full text] [doi: 10.1016/j.pec.2010.05.015] [Medline: 20558024] http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Shafii et al Abbreviations DO: doctor of osteopathy MD: medical doctor MI: motivational interviewing NP: nurse practitioner PA: physician assistant STI: sexually transmitted infection Edited by G Eysenbach; submitted 21.12.18; peer-reviewed by J Torres, A Aventin; comments to author 14.02.19; revised version received 05.04.19; accepted 23.04.19; published 10.07.19 Please cite as: Shafii T, Benson SK, Morrison DM Brief Motivational Interviewing Delivered by Clinician or Computer to Reduce Sexual Risk Behaviors in Adolescents: Acceptability Study J Med Internet Res 2019;21(7):e13220 URL: http://www.jmir.org/2019/7/e13220/ doi: 10.2196/13220 PMID: ©Taraneh Shafii, Samantha K Benson, Diane M Morrison. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 10.07.2019. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. http://www.jmir.org/2019/7/e13220/ XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 7 \| e13220 \| p. 11 (page number not for citation purposes)
10.2196_14112	JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al Original Paper Promoting Social Connection and Deepening Relations Among Older Adults: Design and Qualitative Evaluation of Media Parcels Isabela Zaine1, BSc, MSc, PhD; David Mark Frohlich2, BA, PhD; Kamila Rios Da Hora Rodrigues3, BSc, MSc, PhD; Bruna Carolina Rodrigues Cunha1,4, BSc, MSc, PhD; Alex Fernando Orlando1, BSc, MSc; Leonardo Fernandes Scalco1, BSc; Maria Da Graça Campos Pimentel1, BSc, MSc, PhD 1University of São Paulo, Institute of Mathematics and Computer Sciences, São Carlos - SP, Brazil 2University of Surrey, Department of Music and Media, Guildford, United Kingdom 3Federal University of São Carlos, Department of Computer Sciences, São Carlos, Brazil 4Federal Institute of Education, Science and Technology of São Paulo, Capivari, Brazil all authors contributed equally Corresponding Author: Isabela Zaine, BSc, MSc, PhD University of São Paulo Institute of Mathematics and Computer Sciences Avenida Trabalhador São Carlense, 400 - Parque Arnold Schimidt São Carlos - SP, 13566-590 Brazil Phone: 55 163373 9700 Email: [email protected] Abstract Background: Being socially connected is related to well-being, and one way of avoiding social isolation is to deepen existing relationships. Even though existing relationships can be reinforced by regular and meaningful communication, state-of-the-art communication technologies alone do not increase the quality of social connections. Thus, there is a need for the involvement of a trained human facilitator in a network of older adults, preferably for a short period, to promote the deepening of their relationships. Objective: This study aimed to evaluate the hypothesis that a human-facilitated, media-sharing social networking system can improve social connection in a small group of older people, who are more vulnerable to social isolation than most, and deepen their relationships over a period of a few weeks. Methods: We conducted the design and evaluation of Media Parcels, a novel human-facilitated social networking system. Media Parcels is based on the metaphor of a facilitator collecting and delivering parcels in the physical mail. Extending the metaphor, the system supports a facilitator in designing time-based dialogue requesting parcels from participants that bring out their memories and feelings, in collecting the parcels, wrapping them in annotations that communicate the corresponding requests, and delivering the wrapped parcel to a target person. Qualitative evaluation was carried out in two trials with a group of three people each, one with family members (children and father; aged 55, 56, and 82 years old) and the other with a group of friends (aged 72, 72, and 74 years old), over two weeks. In each trial, data were collected in three interviews (pre-, mid-, and posttrial) and via system logging. Results: Collected data indicate positive social effects for deepening and developing relationships. The parcel metaphor was easily understood and the computational system was readily adopted. Preferences with regard to media production or consumption varied among participants. In the family group, children preferred receiving media parcels (because of their sentimental value) to producing them, whereas the father enjoyed both. In the friendship group, preferences varied: one friend enjoyed both producing and receiving, while the other two preferred one over the other. In general, participants reported a preference for the production of items of a certain type depending on the associated content. Apart from having a strong engagement with the system, participants reported feeling closer to each other than usual. Conclusions: For both groups, Media Parcels was effective in promoting media sharing and social connections, resulting in the deepening of existing relationships. Its design informs researchers who are attempting to promote social connection in older adults. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al (J Med Internet Res 2019;21(10):e14112) doi: 10.2196/14112 KEYWORDS social interaction; interpersonal relations; communication; intervention; experience sampling; mobile apps; photography; video-audio media; elderly Introduction Background The concept of connectedness is multidimensional and there are multiple definitions and types of connectedness, including social connectedness, that focus on interpersonal relationships. Social connectedness, as defined by Lee and Robbins [1,2], includes feelings of being in a close relationship with the social world, which is critical to one’s sense of belonging and is based on proximal and distal relationships. As social interactions may be transient, feelings of being connected to others may change. Thus, social connectedness may also correspond to short-term experiences of belonging [3]. Having regular and meaningful contact with others usually increases social connectedness, but this type of contact may be hindering for older people because of mobility and geographic constraints. The global population aged 60 years old or above is growing at a rate of about 3% per year, which is faster than all younger age groups. At present, in Europe 25% of the population is already aged above 60 years old, and in Latin America they account for 12% of the population [4]. This shift in the world’s demographics has led to a need for action across multiple sectors to enable older people to age well and remain a resource to their families, communities, and economies. Thus, successful ageing and healthy ageing have been topics of interest in some of the most recent studies conducted among the older population. The World Health Organization [5] defines successful ageing as the process of developing and maintaining the functional ability that enables well-being in older age. By including well-being in this definition, it goes beyond physical health as it includes domains such as happiness, satisfaction, fulfillment, and feelings of belongingness. Older people often emphasize the role of social integration and well-being associated with successful ageing [6]. As they age, they wish to maintain their identities and social roles, have relationships, remain autonomous, feel safe, feel like they still have the potential for personal growth, and be able to enjoy life [7-9]. Older adults are more vulnerable to social isolation than the rest of the population. Typically, their social networks shrink with age because of bereavement, relocation, and reduced mobility [10]. Moreover, older adults who are socially isolated have been shown to be more depressed and disabled, be in poorer health, and report lower levels of well-being than those who are socially connected [11]. Thus, it is important that they are able to maintain and deepen social connections with their remaining family and friends and to establish new ones. Maintaining regular social connections can be challenging for older people living alone, who may struggle to travel to meet distant contacts in person or attend in-person social events. Web-based contact seems to be an obvious solution, particularly https://www.jmir.org/2019/10/e14112 XSL•FO RenderX given the rise in internet access via mobile devices for adults aged 65 years or older in the United Kingdom [12], Brazil [13], and the United States [14], among others. Such technology can support older adults [15] who prefer traditional one-to-one channels of communication [16,17]. Social connectedness can be stimulated by experiences that remind people of their social relationships. For example, looking at a photograph or listening to a song may remind them of a loved one and their relationship with them. Thus, feelings of social connectedness do not necessarily require physical proximity. Based on these insights, our hypothesis is that a human-facilitated, media-sharing social networking system can improve social connection in older people and deepen their relationships upon deployment over a few weeks. To evaluate our hypothesis, we designed a new type of social networking system called Media Parcels to address social connection in older people and deepen their relationships over a few weeks with the mediation of a human facilitator. Media collection and sharing, orchestrated by the facilitator, lies at the heart of the system, as this activity is meant to stimulate user reflection about current relationships, both strong and weak. In the Media Parcels system, parcels of media are solicited by the facilitator for later consumption by specific recipients, with the types of media requested designed to encourage reciprocal intimacy and self-disclosure between an older person and selected social contacts. Both the intervention duration and the group size are important in the proposed design. One of the main reasons is that a large group size would imply developing a large number of relationships, which is difficult to achieve in a short period and challenging for a facilitator to coordinate. Second, a duration of a few weeks allows the involvement of the facilitator to be short term, and finally, this would allow the study to take advantage of the temporary novelty effect observed with new technology adoption. In the current design of the system, media collection and distribution are done by a human facilitator among 3 users. In future designs, the facilitator might be supported or eventually replaced by an algorithm designed to allow larger networks. Social Networking and Older People Much research exists on the value of sharing digital media and other physical memorabilia for maintaining relationships but not the honesty and intent [18-21]. Here, we will concentrate on more recent broad social media services and studies of media sharing through social networking systems, especially those targeted at the older population. Facebook use is now part of daily life for many people around the globe, and a huge number of posts made by friends are delivered continuously to users. An early study on self-disclosure on Facebook by Park et al [22] with 317 student J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al participants showed that the number of and positivity of posts is associated with the intimacy of the relationship, but not the honesty and intent. In a later study with 243 participants, Orben and Dunbar [23] investigated how passive consumption of personal disclosures affects the development of relationships. Passive consumption occurs when a person reads the posts of others without directly interacting with them. Their results suggest that reading high intimacy, self-disclosures increases relationship closeness, triggering similar relationship formation in real-life interactions. to improve intergenerational When focused especially on the elderly population, the use of social media is slightly different to other groups. Rebelo [24] conducted an exploratory study in Portugal with 4 older adults to understand the motivations and interests of the elderly population when using social networks (in particular Facebook). The researcher found that, for this particular group, motivation was mainly related to solitude, belonging, reunions, and relationships. willingness Regarding shared content, the elderly stated that they like discussing memories but were concerned about privacy, and they thought that most of the published content on the social network site was inadequate or uninteresting. Chakraborty et al [25] studied the privacy-preserving behaviors of older Facebook users in direct contrast to their self-disclosure. They analyzed the profile information and privacy settings of 134 Facebook users aged above 55 years, together with 50 of their friends, and they observed that older adults hide or share information on their profiles depending on what information their contacts share. They tend to be more conservative about information sharing, in line with the findings of Lindley et al [16] and Pedell et al [17]. Sinclair and Grieve [26], in turn, investigated whether older adults could derive social connectedness from Facebook and whether the levels of social connectedness were similar to those seen in younger samples. The analysis revealed that Facebook social connectedness emerged as a separate factor to offline social connectedness, with correlations between the factors indicating that they were distinct constructs. The participants reported levels of Facebook-derived social connectedness similar to those seen in younger samples. About the social effects and benefits caused by the networks, Quinn [27] examined the engagement among novice social media users, aged 65 years and older, in 4 cognitive domains: attention, processing speed, working memory, and inhibitory control. Results reported include the improvement of intervention participants in inhibitory control. Quinn argued that the findings demonstrated that the benefits of social media use at older ages extended beyond mere social engagement and into other domains of everyday well-being. Chen and Schlz [28] conducted a systematic review to verify the effect of information communication technology on elderly social isolation. Their findings suggested that although the technology lessened social isolation, the technology alone does not guarantee quality of communication among older adults. Furthermore, when communication is not reciprocal, technology use may increase social isolation [29]. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX Several novel media sharing systems have been designed for the older population, with differing levels of effects on relationships. A number of these systems use situated displays in the home to share materials at a distance, such as photographs, text messages and broadcast media [30-33]. For instance, Garattini et al [34] designed a system to promote opportunistic social interaction among elderly people. Using a tablet, the system broadcast news of different topics and presents functionalities to enable group conversations through voice and text. The researchers conducted a 10-week study with 19 elders and some of their friends and family members. The results suggested that although the broadcasts encouraged social interactions, the quality of the engagement was limited by the absence of an approach to share personal information to help users become familiar with one another. Waycott et al [31] developed a tool to facilitate message and media sharing and conducted a study in which caregivers used the tool to communicate with their elderly clients. Their results showed that photographs with captions were able to increase and enhance communication, and were well-suited to promote psychosocial care. Similarly, Abrahão et al [35] conducted a mobile digital storytelling study in a care home for older people. The creation of the stories was facilitated by either a formal or informal carer and focused mainly on the resident, capturing aspects of everyday life such as visits, social events, therapy sessions, and health reports. The results showed that the technology stimulated expressivity and creativity in the resident, as well as richer conversations between the resident and other people. Cornejo et al [32] developed a system geared toward enhancing older people’s offline interactions with their family. The results emphasized how information shared on social media could provide conversational context for the elderly, prevent isolation, and increase offline conversations. Most studies report positive benefits for maintaining or making new relationships through lightweight multimedia messages. Given the established relationship between the level of self-disclosure and relationship closeness both online and offline, there is an opportunity to explicitly encourage media exchange related to the existing relationship. By encouraging self-disclosure, we mean to encourage participants to talk about feelings and emotions, to talk about topics they do not regularly talk about (eg, intimate things), and to talk about themselves and their relationship toward each other so that using an app might be easier than saying it face-to-face. Methods Media Parcels Design Media Parcels is a novel social networking system designed to promote facilitated media exchange between users. The interaction underlying Media Parcels is based on the metaphor of parcel delivery in the physical mail. First, a facilitator, upon specific requests to participants, collects media and wraps them in text commentary, bringing out their memories and meaning. Next, the facilitator passes the wrapped media parcel to a target person, who in turn unwraps them. J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al In the general case, Media Parcels supports facilitators in orchestrating interactions among any number of users by intervening in 2 steps: in phase 1 by requesting media parcels from a user and, in phase 2 by sending the parcels to any number of (other) users. The particular case of deploying Media Parcels for the balanced interaction among 3 users (P1, P2, and P3) orchestrated by a facilitator is illustrated in Figure 1. For the same scenario, Table 1 details the request of parcels of media (from-to). As currently designed, Media Parcels relies on a facilitator in charge of managing the interaction for the duration of the intervention. In a nonfacilitated approach, Media Parcels could, in principle, pass parcels of media between multiple providers and recipients of media, at various times, ad infinitum. In contrast to most current social media systems that rely on the spontaneous posting of media for feedback, the Media Parcels approach is based on directed requests for media items which are then shared within a group. For the deployment of Media Parcels, we used the Experience Sampling and Programmed Intervention Method (ESPIM) and an associated platform that support specialists (eg, health professionals) in communicating with their users at particular times of the day via a mobile app [36,37]. ESPIM is inspired by the experience sampling method for collecting self-reports in psychology proposed by Csikszentmihalyi et al [38], combined with the concept of programmed instruction [39]. The ESPIM software platform was designed to support health care and learning interventions in their natural environments throughout the day [36]. Such interventions can comprise open-ended and multiple choice questions, media requests, and deliveries. While creating the interventions, the professional defines the time-based moments in which they should be prompted on the user’s device [40]. Figure 1. Media Parcels exchange among three persons (P1-P3) in a two-phase interaction orchestrated by a Facilitator, who “wraps” the parcel by adding commentaries. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al Table 1. Operations for Media Parcels exchange among persons P1, P2, and P3 in request-wrap-post interaction orchestrated by a facilitator. Phase and facilitator (role) P1 P2 P3 Phase 1 Media parcel (Request) Media parcel (Receive) Wrapping Media parcel (Wrap) Phase 2 Media parcel (Post) Media parcel (Post) aNot applicable. from (P1) to (P2) from (P2) to (P1) from (P3) to (P1) from (P1) to (P3) from (P2) to (P3) from (P3) to (P2) —a — — from (P2) to (P1) from (P1) to (P2) from (P1) to (P3) from (P3) to (P1) from (P3) to (P2) from (P2) to (P3) Approach Toward Social Connection and Deepening Relations Aiming at supporting a facilitator who promotes social connection and deepening of relationships in a small network, comprising older persons and their social contacts, content exchanged via Media Parcels should encourage self-disclosure in the scope of the social relations involved while ensuring both intimacy and privacy. Facilitator The role of the facilitator includes: (1) designing media requests associated with feelings that encourage self-disclosure and reflection over a specific social contact; (2) wrapping together the media elements received and the associated feelings, without editing them; and (3) mediating the interaction between a pair of users by sharing the content produced between both ends of the targeted relationship. For the purposes of this study, such a role demands professional skills, so in the case of the two trials described here the facilitator had a background in clinical psychology. Media Requests Requests for media were deliberately designed to be thought provoking and vehicles for self-reflection and disclosure. For example, the following requests and questions were used, among others, to elicit responses across a variety of media forms, personalized by name: • Take a picture of an object that is special to you and explain why (image and text). • Record a short audio clip of yourself singing a snippet of a song you like and explain why you like it (audio and text). • Record a short audio to X, saying what he or she means to Media requests were inspired by methods used in design and behavioral psychotherapy. We used two strategies for the design of media requests, one concerning the type of media, and the other, the media content. Different types of media may produce different emotional responses [41], so we designed media requests to be balanced across image, audio, text, and video to provide participants with all types of media once, and regarding media content, the media requests were designed to probe participants’ relationships and themselves. Behavioral psychological interventions make use of asking questions, making requests, and giving instructions or suggestions for clients to carry out specific or generic actions outside the therapeutic setting [42]. Asking questions about happenings or feelings serves to gather information and also promotes self-observation, self-reflection, and self-knowledge. In our study, the media requests and questions about participants’ feelings were designed to encourage all participants to reflect on their relationships and express their feelings toward one another by producing media with emotional content. Furthermore, specifically for the older participants, the questions and requests also encouraged reminiscence of precious moments and self-disclosure. The questions and requests might also be referred to as relationship probes, as they have the character of cultural probe questions for self-report except they are focused on personal traits and relationships rather than culture [43]. Cultural probes is a technique used to inspire ideas in a design process. Typically, the probes are small packages that can include any sort of artifact along with evocative tasks that allow participants to keep record of specific events, feelings, or interactions. It serves as a means of gathering inspirational data about people’s lives, values, and thoughts. you (audio). Media Wrapping and Sharing • What do you like to do with X? Why? (text). • Send me a picture of a present X has sent you in the past. What was the occasion? How did you feel about it? (image and text). • Record a short video of you saying something X often says (video). https://www.jmir.org/2019/10/e14112 XSL•FO RenderX For each media item collected, before sending the item to the target person the facilitator included a text comment to expose the request that created it. An example is given in Figure 2, with the facilitator requesting for one person to record a piece of audio for another person saying what she means to you. When sharing that media, the facilitator includes “Your dad recorded this message to you”. As indicated by this example, no editing relative to the media collected took place. J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al Figure 2. Typical media request and delivery screens. Left: media request in phase 1 to the older adult. Right: media delivered to another participant in phase 2. Older Person’s Social Contacts Social contacts of an older person may be inter- or intragenerational, such as younger family members (their children or grandchildren) for intergenerational contacts, and the same generation friends or family members for intragenerational contacts. According to Lindley et al [16], the dynamics of family and friend relationships are very different, especially in terms of expected social support. Considering the two different types of social contacts, we designed two trials devoted to each type of relationship: one older person with two younger family members, and three same-generation older individuals that were friends. For practical reasons related to our international collaboration, we conducted the first trial with family participants in the United Kingdom and the second trial with friendship participants in Brazil. The system and methods for both trials were largely the same and are reported in study 1 for the UK context, with variations outlined in study 2 for the Brazilian context. The studies reported in this paper were designed so that phases 1 and 2 lasted around one week each. This amounted to a fixed period intervention aimed at increasing social connection between these partners. We designed a media collection phase (phase 1) and a media sharing phase (phase 2), so that participants could clearly discriminate between producing and receiving personal media content. In both studies, participants were aware of the two distinct phases and that the produced media in phase 1 would not be immediately shared with the other participants. They were also informed that the human facilitator, who was a clinical psychologist, would be in charge of sharing the produced media and its associated feelings from phase 1 with the other participants on phase 2, without editing the content. Participants were also aware that, once the facilitator mediated the media sharing, this person would have access to the content of the media and its associated feelings. In addition, the participants were informed that, after phase 2, the facilitator would not mediate their interaction anymore; thus, if they wished to talk about the media content received, they should use their conventional communication channels. Study 1: Family Trial This study focused on supporting an older person through media exchanges with two younger family members. The exchange of media parcels was asymmetric because they were not collected and delivered between the younger participants. Figure 3 shows the network configuration for study 1. An older adult who lives alone exchanged media (text, audio, image, and video) with 2 other social contacts. The facilitator (the first author) orchestrated the exchange in 2 phases. In phase 1, lasting a week, media relating to each pairwise relationship with the older person was collected by the facilitator at regular intervals. In phase 2, also lasting a week, that media was distributed to the reciprocal partner in each pair at regular intervals. As the older person is linked with 2 people, while those people are only linked to one and not each other, we collected and distributed twice as much content from the older person as from the reciprocal partners, as shown by the arrows in Figure 3. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al Figure 3. Network configuration and media flow in the Media Parcels system for study 1. Media A typical media collection and sharing screen is shown in Figure 2, as previously presented. Requests for media were deliberately designed to be thought provoking and vehicles for self-reflection and disclosure. For example, the following requests and questions were used, among others, to elicit responses across a variety of media forms, personalized by name: • Take a picture of an object that is special to you and explain why (image and text). • Record a short audio clip of yourself singing a snippet of a song you like and explain why you like it (audio and text). • Record a short audio to X, saying what he or she means to you (audio). • What do you like to do with X? Why? (text). • Send me a picture of a present X has sent you in the past. What was the occasion? How did you feel about it? (image and text). • Record a short video of you saying something X often says (video). Participants Recruitment requirements for the older person were as follows: (1) aged 60 years or older; (2) lives alone; (3) is based in the Guildford area, United Kingdom; and (4) is able to nominate at least two other social contacts who are able to participate. All names were changed to preserve participants’ identities. We formally broadcasted the search for participants for 6 weeks through the University of Surrey academic email to the existing network of research groups and personal contacts, on the research group website, and by partnering with Age UK Surrey and the University of the Third Age, Guildford. The participants recruited were resultant of personal contact. Participants were 1 older adult, Paul (aged 82 years), and 2 of his children: Karen (aged 55 years) and Charles (aged 56 years). Paul chose them because he considered them relevant people for social support in his life. Paul is an electrical and electronic https://www.jmir.org/2019/10/e14112 XSL•FO RenderX engineer and still works part-time, mentoring engineers at a company. He did not have any cognitive impairment according to the Mini-Mental State Examination to measure cognitive health [44]. He was also living alone for the past 6 months and feeling more isolated since his wife passed away. Charles, his son, lives close to him. His daughter, Karen, lives in Kenya, Africa, but was visiting her father for the first week of the trial. The field data collection was done from late April 2018 to late May 2018. Pretrial Interview A pretrial semistructured interview was conducted face-to-face with each participant. With the older person, we enquired about his relationship with his two children. To access feelings of a specific social connection at the individual level between Paul and his children, we designed a Relationship Semantic Differential Scale (RSDS, Multimedia Appendix 1) with 16 pairs of contrasting attributes related to social relationships. Participants were instructed to express their agreement with the attributes on a 7-point scale. As typical of semantic differential scales, the closer the participant responds to one of the attributes the more they agree with it [45]. Paul was asked to rate his relationship with each child separately. Finally, the Media Parcels mobile app was downloaded and installed on a dedicated tablet. Paul was given this tablet and trained in how to use it. This took a total of about 1 hour and 30 min. Each of Paul’s children also answered a similar semistructured interview enquiring about his or her relationship with their father and responded to the RSDS. The Media Parcels mobile app was downloaded to their existing smartphones and they were trained in how to use it. This took about 30 min per person. Trial Period Following the pretrial interview, the system was activated on all 3 devices and proceeded to collect media parcels from participants over 7 consecutive days in the first week of the trial. These were analyzed and selected by the facilitator for J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al redistribution over the following 8 consecutive days of the second week for the social contacts and 9 consecutive days for the older person. The amount of media parcels received in the second week was based on the amount of relevant media parcels produced by participants in the first week; therefore, it varied among participants. The older person, Paul, received 3 collection messages or 1 delivery message from the system each day, whereas his children Karen and Charles received only 1 collection or 1 delivery message a day. Notifications of each message sounded and appeared on the receiving devices, but users were able to defer answering or reading them until later. Users could also choose to decline answering any request. Once a media delivery had been seen, users could not directly respond with feedback to the person who sent the media, as is usual on most social media platforms. It was a design decision not to support interactional closure to encourage participants to contact each other offline or using their existing preferred communication channels. This decision was based on the fact that new technology solutions on communication mostly force users to abandon their usual channels to adopt new ones. This can be aversive to those people who take more time learning to use a novel technology, which is the case for most of the older population [46]. Midtrial Interview At the end of the first week (phase 1) of the trial, we conducted a midtrial interview. Participants once again each responded to the RSDS and were interviewed about the feelings generated by producing materials about their relationship. Participants were also briefed about the delivery of phase 2 of the study. Posttrial Interview After phase 2, we conducted final semistructured interviews with all participants, separately, about their relationships, connectedness toward one another, and experience using the system. They then responded to the RSDS. The meetings lasted an average of 30 min for each social contact and 1 hour for the older person. In this last interview, they also evaluated their experience with the ESPIM mobile app by answering a User Experience Questionnaire (UEQ) [47] and a System Usability Scale (SUS) [48]. All interviews were conducted in the participants’ native language, English. Study 2: Friendship Trial To test out the value of the system for deepening relationships with same generation friends, we recruited 3 older people with different levels of acquaintance, as described below. We decided to treat all these participants equally in the trial by using symmetrical media sharing between all three, where P1, P2, and P3 are three elderly ladies. This means the scripted dialogue between participants and the facilitator implies that media were collected in phase 1 from each participant about their other 2 partners and distributed to those partners in phase 2. Participants Recruitment requirements were as follows: (1) aged 60 years or older; (2) lives alone; (3) is based in São Carlos, Brazil; and (4) is able to nominate at least two other same-generation social contacts to also participate. The research was broadcast in a digital literacy course for older adults. In total, three elderly ladies volunteered. They described themselves as friends or acquaintances. Ronda is a 72-year-old retired teacher who was divorced and had a son. Linda is also aged 72 years and is a retired administrative assistant) who was never married and has had a boyfriend for the past 8 years. Finally, Irene is 74 and a retired laboratory assistant. All participants were Brazilian. They had some experience dealing with smartphones and tablets. None of them had any cognitive impairments. The network configuration and media flow in study 2 is presented in Figure 4. Both Ronda and Linda were more connected to Irene, who acted as a friendship mediator between the two others. Ronda and Linda were friends for about 10 years and they both described being close to each other. Linda and Irene have been friends for over 30 years. They worked together, and before their retirement they were very close. Irene said their relationship started to go cold after a while, especially after Linda started dating. Ronda and Linda consider themselves to be friends, they have known each other for 10 years, though they do not call or arrange to meet in person as this is always mediated by Irene. In addition, most of the time Irene is the one who initiates social contact with both friends. None of them described themselves as socially isolated, but Ronda expressed interest in deepening her relationship with both her friends, Linda and Irene would like to revive their friendship that had become distant over the years, and Linda would also like to become closer to Ronda. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al Figure 4. Network configuration and media flow in the Media Parcels system for study 2. Procedure The procedure was identical to study 1: (1) pretrial interview; (2) media collection; (3) midtrial interview; (4) media sharing; and (5) posttrial interview. In this case, the interviews were conducted in Portuguese. The total length of the data collection, from pre- to posttrial, was 20 days. Media were shared symmetrically as described above. The media requests were specially tailored to suit the particularities of the three participants and to deepen their relations. As before, requests were to share information about themselves (personal requests) and about their relationships with each other (relationship requests). In phase 1, a total of 14 requests were designed for each participant: 7 personal and 7 about the relationships. The requests involved the same media types as in study 1 (text, audio, video, and image) as well as text commenting about their feelings toward what they had produced. Among the images requested, there were photos of people (friends participating in the study), objects (presents they received from each other, special objects, and objects that reminded them of each other), and places (favorite spot in the house). The audio requests were to share a memory from childhood, talk about a miracle in their lives, and record a song snippet. The video requests were to record an ongoing activity, talk about a good day they spent with the friends, and record a video message to the other friends. The text requests were to write about what they would like to do more with the friends and to write about their favorite singers, television shows, and activities. A total of two requests were sent out each day, one in the morning and another in the afternoon for 7 consecutive days. As in study 1, participants could respond to the requests immediately after they received the sound reminder or any other time during the day. In phase 2, we shared the most compelling media via the system, which, in the case of this study, were the media and associated feelings with more emotional and personal content. All participants received 1 media parcel per day for 9 consecutive days. Results Study 1: Family Trial User Experience, Engagement, and Patterns of Interaction The results of the user experience and usability rating scales were very positive, with the SUS results indicating an average score of 77.5 points. This is about 10 points higher than average, and a B+ on a scale ranging from A to F. The UEQ scores were at excellent or above average across all dimensions. Overall, participants were able to use the app without any issue other than difficulty in identifying the status of video recording when the Android video app is used, an issue not under the control of the designers. Such positive results, which relate to the attention given to guidelines during the design of the mobile app [40], contrast with the poor accessibility of many apps when used by older people [49,50]. Looking ahead, the experience and usability scores were very similar for study 2 (average SUS=79.17 and slightly higher UEQ scores). Therefore, the rest of the paper concentrates on the process of using the system and its value for relationships. In phase 1, the media collection phase, all participants responded to more than half the media requests. Paul responded to 76% (16/21) of media requests on a daily basis. He interacted with the app immediately after the sound reminder in about 38% (8/21) of cases, and preferred to use the app when he had time to engage in the requests: https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al I heard the alarm, but if I was busy at that moment, I just ignored it (...). I knew I had to do the tasks, so I tried to come back to it when I had time. [Paul] On average, he spent 5 min per interaction, although he spent more time on the requests that required him to retrieve materials, such as pictures, written material, and objects. Karen responded to 100% (7/7) of media requests. She interacted with the app on two separate days and preferred to respond to most of the requests in a row. It took her an average of 3 min per interaction. She never responded to the requests immediately after the sound reminder: I preferred to do it all at once, in a day I had time. [Karen] Charles presented the lowest rate of engagement, 55% (5/9). In the intermediate interview, he said he did not interact with the app as much because he had a busy week. He spent 10 min on average in each request and also preferred to interact with the system when he had time, despite the sound reminders. In phase 2, the media sharing phase, the engagement with the app was higher. Paul and Karen interacted with all of the media shared and Charles did not view only one of the parcels of media, interacting with 87% of them (7/8). The pattern of interaction with the app also changed in phase 2. This time, Karen and Charles interacted with the app almost daily and usually before the sound alert was triggered as a reminder. This suggests an expectation to receive the media produced by the father, which was confirmed in the posttrial interviews in which they reported being curious about what they would receive that day. Type of Media and Media Content The requests were effective in leading to self-disclosures revealing user’s personalities, memories, and relationships. For example, a response to a request for a picture and explanation of a special object by the older person is shown in Figure 5. The story of this object is touching and something Paul may not have posted spontaneously on a social media site. On the basis of their experiences producing and receiving media, we asked participants about their favorite types of media and media content in each phase of the study. For Paul, the most important feature was the content of the media and not the type itself. Thus, when producing media, his least favorite was typing (text) “It was boring, I am slow (...) but not impossible.” Regarding the type of media received, he said that one of his favorites was a song snippet from his daughter singing (“You are my sunshine”). He said that this media inspired him to get back to learning the Ukulele, so that he could play that song for her in the future. Charles enjoyed more producing videos because he thought it was an opportunity for him to show something he was proud of. He also reported that he would have enjoyed recording longer videos. He enjoyed receiving the photos and videos best because they were visual. As for Karen, she enjoyed producing all types of media and was pleased that she had the opportunity to do a little bit of everything. She thought that the different types of media made the tasks more enjoyable and less predictable. When she received the media, she enjoyed text, audio, and video more than photos: The photos were not surprising, they were not new to me (...) the others were an expression of dad in that moment. [Karen] Figure 5. Examples of media collected in phase 1 (image at the left and text at the right). Producing Media Versus Receiving Media Both Karen and Charles preferred receiving media rather than producing them. They felt good being told good things about themselves by their father and because it was a surprise to see what their father had to say to them. Also, they considered this phase more effortless. Paul also reported that the second week https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al was easier, but he enjoyed both producing and receiving the media: I enjoyed them (...) it was a good surprise (...) and I was proud of them [Karen and Charles] for participating (...). It was like aconversation. [Paul] Feelings of Specific Social Connectedness All participants reported feeling closer to each other and contacting each other more than usual during the Media Parcels trial. This is reflected in the relationship (RSDS) scores taken at the beginning, middle, and end of the trial shown in Multimedia Appendix 2. Scores generally rose in the trial for Paul and Karen. Interestingly, their rise in scores was greatest in the middle of the trial, after media collection but before delivery. This suggests an effect of feeling closer to someone when encouraged to think about them through a media collection task. Charles’ scores remained about the same, perhaps because he found the exercise harder than the other two in terms of expressing his feelings (see Multimedia Appendix 2). Paul reported that he felt that his contact and relationship with his kids changed during the study: We ended up being more thoughtful, because of the questions asked. We have become closer and honest to each other. [Paul] He reported that producing and receiving the media and talking about the feelings associated with them made him think about being more effective and also about how to get things done in general: I realized I was letting things slide. You stimulated my attitude (...) because I had to do the activities it also encouraged me to do other things I needed to do. [Paul] He also thought that doing the activities made him think about his kids much more than usual, and he contacted them more than usual: It made me think more about emotions and feelings (...) we usually let it go and don’t make time for ourselves. I was happy to do that (...) I was relieved to have a reason to do that. [Paul] Charles reported that, for him, talking about feelings was not easy and as the activities required that he felt as though they were not simple. However, he considered that he thought and contacted his father more than usual during the study (“I did think of him more when I was doing the tasks”) and reported that he felt closer to his father when doing the tasks and when receiving the materials from him. He also pointed out that this type of intervention would have been even more relevant if they were living further apart. Karen considered that the activities made her think about her father in a different way and remember good moments from the past: It made me think about my dad in a different way. [Karen] About receiving the media, she reported: https://www.jmir.org/2019/10/e14112 XSL•FO RenderX It was nice to see how much I was meaning to Dad (...) he was appreciating things. I felt loved and valued. You got a more intense relationship snapshot from us. It was interesting to see how we were interacting (...). I really enjoyed the way everything was handled. It made me feel closer to dad. It was something that brought me, my dad, and my brother together. [Karen] Study 2: Friendship Trial Engagement and Patterns of Interaction In phase 1, there was a high level of engagement with the media requests for all participants. Ronda and Irene responded to 100% (14/14) of the media requests, and Linda to 71% (10/14). Linda did not respond to any of the video requests because of difficulties using the video recording tool in the tablet, despite being trained to do so twice by the researchers. All participants interacted with the system on a daily basis, except for 1 request that Linda completed on the following day. All participants preferred to interact with the system when they had time, rather than immediately after a request or parcel had arrived. Irene and Linda only responded to the requests immediately after the sound notification in 7% (1/14) of the times, and Ronda in 28% (4/14) of the times. On average, Ronda spent 4 min 23 seconds per interaction, Irene spent 3 min 51 seconds, and Linda spent 5 min 42 seconds. The replies that took the longest were those involving existing materials, such as pictures and objects. In phase 2, the media sharing phase, the engagement of interaction with the app was 100% (9/9) for all participants. They interacted with the received media parcels on a daily basis. Regarding the sound reminder, Ronda and Irene interacted with the app in the same proportion before, after, or immediately after the reminder 33% (3/9) of the time. Linda interacted with the media parcels immediately after the sound alert 55% (5/9) of the time and before or after it 22% (2/9) of the time. This again reflects increased interest and anticipation in receiving media. Type of Media and Media Content As in study 1, the media parcels elicited from users were often more intimate than social media posts or even than things they might have said to each other face-to-face. In fact, some had the character of greeting cards that allowed people to express deeper feelings than what they are comfortable saying in person. For example, a request to record a message about her friendship with Ronda caused Irene to say the following: She is a very kind friend, very caring, very dedicated. I think that regarding my friends I am privileged. Ronda is someone that I like very much. [Irene] When we asked Ronda to say how she felt about this to us, she wrote: I appreciate the message very much and it made me emotional. I am what I am! I prioritize my friends because they are the shoulders that I lean on. Listening to the audio from Irene I feel safe and cared J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al for. I know I can count on her friendship forever. Thank you, my friend. [Ronda] her. The media she most enjoyed producing was to find a snippet of a song to record: Regarding their favorite types of media, participants varied. While producing the media, Ronda liked pictures, writing, videos, and audios in that order. Her favorite tasks involved pictures because she retrieved them from old materials that reminded her of stories from the past that she had forgotten. One of her favorite tasks in the media collection phase was to talk about her favorites (singer, television show, and hobbies): Because of that I started to listen to music that I loved but had forgotten. [Ronda] I did not want to sing, but I enjoyed looking for a song that I liked to record a bit of it. [Linda] About receiving the media, she said that she appreciated all of them, irrespective of type. Producing Media Versus Receiving Media Feelings about preferring the producing or the receiving of the media parcels were different across participants. Irene preferred to receive the media: About the received media parcels, she reported that she did not have a favorite type and that the most important aspect was the content. She appreciated the messages she received from the friends: It is more fun (...) you get to know more about the other person and their feelings (...) I felt emotional and even cried over some materials [messages from the friends]. [Irene] I was moved and thought how true friendships are important in our lives. Who has friends has everything! [Ronda] She picked out 2 favorite media parcels: (1) the friends’ childhood stories (“They reminded me of my own childhood. We usually do not share old stories like that”); and (2) their personal profiles. This was a composite parcel made by the researchers. The format was a magazine cover with each participant’s photo and information that they had provided about themselves, in addition to the information provided from the friends to each other. She said that she really related to her own profile and was moved by it. She also thought that the profiles were a good snapshot of each of the friends. Receiving the personal histories about the friends also made her realize that they had similar life experiences. Although she also enjoyed going through old photos because they reminded her of good times. Ronda reported that she enjoyed them both: I loved them both (...) one phase completes the other (...). If you produce something and do not get anything back, you do not get closure. Producing the materials made me rethink my life, good times and difficult ones (...) and how I have overcome the difficulties (...). I visualized myself back in time, it was as if I was reliving it. I really enjoyed it. It is an interesting approach to carry out with my other friends as well. [Ronda] Similar to Ronda, Irene also enjoyed producing the photos: About the media sharing phase, she reported: It was cool. I looked for old, forgotten photos from my friends but also family. It was good to revisit these materials. [Irene] Then her order of preference was audio, video, and text: I do not like so much to show in pictures, so I did not like very much to show my face in the videos. But the audio was fine. I also don’t like writing (...) I have never been good with the written word. [Irene] Her favorite activity was to record an audio about her childhood: It made me remember my childhood, the people from my childhood (...) it was a good time of my life. [Irene] While receiving media, she said that she enjoyed everything and that the content was more important than the format. She noted that she especially enjoyed receiving the videos and pictures and that her favorite media parcel was an audio she received from Ronda: I felt truly touched and even cried (...). I never expect compliments from other people and I don't even know if I am all that (...) I immediately picked up the phone and called her to say how much she means to me [Irene] For Linda, she enjoyed producing the types of media in the following order: text, photos, audio, and video. Although she likes to watch videos, she had difficulties using the device’s camera to record videos by herself, which was frustrating for I always expected the messages as a surprise (...) the photos and audios moved me. [Ronda] As for Linda, she enjoyed producing the media parcels because she had the opportunity to revisit bits of her life and moments with the friends: It touched me (...) I got in touch with my emotions, with personal stuff. It was a bit of work having to look for old materials, but it felt good to revive the friendship (...) it was getting cold [Linda] As in study 1, receiving the media in general produced more spontaneous and emotional responses about their feelings. Feelings of Specific Social Connectedness During the study, all participants reported feeling closer to each other, thinking about each other, and getting in touch more than usual. All participants commented that during the study, they were encouraged to talk about personal things and talk about how they felt about their friendship in a way that they would not naturally do. Irene said: Sometimes it is hard for me to express certain feelings (...) this helped me to do that. [Irene] Ronda reported: These talking points do not come to our conversations by chance. The sequence of questions brought memories (...) and the opportunity to express https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al ourselves in a different way (...) the questions helped me to express my feelings. [Ronda] Although she had a stronger bond with Irene, she reported that she got more involved with Linda as well. Irene reported that she felt closer to the friends in the media collection phase but even more in the media sharing phase. Irene was moved by the messages she received from both friends, and she decided to call them both: I called Linda and I said, “I do not say it enough but I love you (...).” I also called Ronda to say that she is really important in my life and that I don’t know what I would do without her. [Irene] After this call, Linda also called Irene once, and Irene called both friends to schedule a night out including 3 other friends. In this occasion, Linda spoke about a misunderstanding she had with Irene 10 years earlier that she still felt hurt about. Irene could not go to Linda’s birthday, so she sent a gift through a friend. Linda did not appreciate it because she would rather have Irene come over in person to bring the present. Thus, she returned the gift and never talked about it again until this moment. They said that this was a good opportunity to set things straight and get closer to each other again, and it ended in a group hug. However, Irene felt a little hurt that Linda was still mad after such a long time. Ronda also commented that she and Irene will try to include Linda more in their social gatherings: Because she has a boyfriend, we do not invite her a lot to go out with us (...) but she really enjoys our company and we will try to include her more. [Ronda] These reported feelings of closeness are reflected in a systematic rise in their RSDS scores, as shown in Multimedia Appendix 3, at the pretrial, midtrial, and posttrial periods. The only drop was between the mid- and posttrial periods for Irene’s feelings of closeness with Linda. These went down because of the conflict over the birthday present reported above. This shows that developing relationships is a complex business. It may result in relationship work that raises contradictory feelings of closeness and distance within the same relationship pair. Discussion Principal Findings We designed a novel communication system called Media Parcels that makes time-based requests for media and distributes them within a minimal network of just a few social contacts. The social networking system we designed aimed to support older people in deepening their social connections and relationships while still respecting their privacy concerns. Broadly, our results indicated positive social effects for both deepening and developing of relationships. The elderly participants perceived personal and more intimate social connection as important to their lives, corroborating with other studies that present social integration with family, friends, or community as crucial to well-being in older age [6-9]. Our participants also appreciated the computer-mediated system as a means to encourage self-disclosure and initiate conversations about feelings toward one another, with more deep and https://www.jmir.org/2019/10/e14112 XSL•FO RenderX meaningful content. This result supports the findings of Lindley et al [51] that explored the attitudes of older adults to keeping in touch with people who are important to them. The authors found that the older adults wanted to be in touch, and that staying connected was worthy of dedication. Moreover, the participants most valued the communication that is personally expressed, which requires a level of intensity that contrasts with the lightweight interaction that is increasingly afforded by new technologies. This was precisely fostered in our study by the Media Parcels system, in which social interaction topics were designed to encourage meaningful reflections and expressions of feelings. The importance of fostering relevant, meaningful social interactions is highlighted in studies by Carstensen et al [52,53], which suggest that older people are more motivated to derive emotional meaning from life and establish intimacy with other people, presenting a preference to invest in relationships that are emotionally rewarding and significant to them. Testing the Media Parcels system on a trio of family members and in a separate trio of friends exposed values common to both networks as well as values exclusive to each social group. It also revealed new opportunities for computer-mediated communication. Family Connections The different family members showed differing levels of engagement with the system across media collection and delivery phases. The older person Paul was most enthusiastic overall, and his daughter Karen was a close second. However, his son Charles started out as a skeptic who found it difficult at first to respond to media requests about his father. All three became more enthusiastic in the second phase of the study when they received media parcels relating to Paul. Karen summed up their experience about the system by saying that it brought them all together in a way that would not have happened spontaneously. This is an important observation from the daughter, as the system was configured to focus on her father and his relationship with family following the death of his wife 6 months earlier. It shows that even bilateral media sharing of this kind stimulated reflections on family as a whole and discussion outside the system that affected all relationships. Thus, the system fostered a shift in the content of the communication between the older participant and his children, from basically discussing family affairs and obligations to talking about feelings toward one another and interests. This is particularly relevant as the complex nature of family ties, that includes feelings of both togetherness and responsibilities, may produce negative feelings with greater degrees of obligation and formality associated with familial relationships when compared with friendship [54]. Interestingly, participants told us they rarely discussed the media parcels themselves, in case they ruined the surprise of pending deliveries during the trial. The parcels were also about feelings that participants found hard to express or discuss in person. Friendship Relations Media sharing was made trilateral in the friendship trial, with three older people of differing closeness to each other. Again, we found various levels of engagement and different effects of the system across participants, but a positive endorsement from J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al all of them in the end. In fact, Ronda wrote to us after the trial saying how much she missed the rhythm of the parcel requests and deliveries and its new kind of connection to her friends. There was greater interest in media collection between friends than in the family trial, and all the participants perceived their friendship relations as precious and important to their well-being. In the specific case of the friendship between Ronda and Irene, they even spent time with each other more frequently than with family members and often offered emotional and health-related support to each other, because of either physical distance from close relatives or not desiring to be a source of concern to them. This points out the importance of having friendships later in life and is aligned to other studies that discuss the benefits of having peer relationships on the well-being of older people [55,56]. However, some points of contention were raised by the deliveries. This was most dramatically illustrated in the story of the rejected present between Irene and Linda, which raised a forgotten issue for Irene but resolved it for Linda. This example shows not only the power of the system and particular media requests to access facets of a relationship but also the dangers of doing that sometimes for relationship closeness. Facilitation A total of 2 types of facilitation seemed to be going on in the trials: (1) the media parcels themselves, as designed, were facilitating reflection and communication between parties and triggering further conversation outside the system; and (2) the human facilitator in the loop, with a background in clinical psychology, authored requests and selected responses to maximize positive effects on relationships. Concerning the media parcels themselves, we found that participants could express feelings to each other that were hard to communicate face-to-face, as in the use of greetings cards. In this respect, media requests were effective relationship probes, revealing aspects of a relationship to other partners as well as to us. Finding and selecting media to share was also motivating, especially for Ronda and Linda in the friendship trial. This caused them to retrieve forgotten images and other materials and remember the good times in the light of intervening events (see also Frohlich et al [57]). These findings are similar to those in a recent study of media sharing to facilitate young people’s conversations with relatives having dementia. The young people were encouraged to find media relating to the person with dementia through a system called Ticket to Talk and use these media as a kind of conversational playlist to stimulate conversation [58]. As for the human facilitator in the loop, human facilitation amplified the effect of media exchanges by personalizing them to the participants. It also continued outside the system, as the facilitator also conducted the interviews and became a sounding board for the participants’ reflections on their relationships. In fact, the facilitator, the first author who is a trained clinical psychologist, realized that it could be a powerful tool for counseling. This raises issues for the future of the approach in terms of the levels of facilitation involved. If the intervention lasts longer than a few weeks, how personal should the media be to deepen relationships between particular people? And how https://www.jmir.org/2019/10/e14112 XSL•FO RenderX important is it to have a professional facilitator designing and monitoring media? Regarding the first issue, if a system such as Media Parcels should be used for longer periods, then the content of the media requests should be balanced with deep personal content and lightweight content, so that the media exchange does not become burdensome. With respect to the second issue, in our study, the facilitator had a key role in designing the media requests and collecting and distributing the media produced by the participants. It is not uncommon to have human facilitators mediating technology or Web-based social interaction for vulnerable populations. For example, in Abrahão et al [35], the facilitators had a central role in helping the creation of digital storytelling by a care home elderly resident with dementia. Another example is the Scrapbook Circles network that is designed to allow disabled users to post content to friends and family through facilitators, if they wish. Similarly, the Media Parcels system could be easily scaled up to dozens of users by expanding the number of facilitators. However, personalized interventions such as the ones reported in this paper have a limit on scaling; a human facilitator dedicated to the task may be able to monitor and mediate from 3 to 4 dozens of groups weekly. In contrast, if personalization is not a requirement, the system could be scaled up to hundreds and thousands of users using predetermined templates and algorithms, such as those used in recommender systems toward suggesting books and songs. Also, by indicating users’ relationships in the system, it is feasible to create algorithms that automatize content distribution, so received media can be redirected without the need of facilitators. The ESPIM system used to deploy Media Parcels already allows the use of predefined templates and automatic media distribution. The figure of the human facilitator could also be removed by adding in the system functionality of direct communication between participants over a predefined set of conversational topics and media sharing requests. Future research could explore both options to reveal more about the role of media in deepening relationships at a distance, especially for the older population. Communication The communication dynamics of the Media Parcels system was unusual: (1) it was facilitated by a human who administered a time-based dialogue; (2) the pace of the facilitator-participant dialogue was slow, ranging from 1-3 times a day; (3) the explicit separation of the collection and delivery phases allowed us to assess the effects of each phase; (4) the distinct phases introduced delay into the communication, resulting in media exchange that was neither wholly asynchronous nor wholly synchronous; and (5) the introduction of a delay into the communication resulted in anticipation of media deliveries in which participants met the parcels as they arrived in real time. Such communication dynamics could be adapted within the current system design. For example, the time-based dialogs could make use of different media requests between different configurations of people within a wider network, interleaved more closely in time with their delivery. Our findings from J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 14 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al study 1 support the evolution of an intervention approach that could be offered professionally to individuals as a kind of relationship therapy. Moreover, the findings from study 2 support a self-paced long-term approach of unfacilitated media sharing within small groups. To make this sustainable, message requests might come from friends themselves. In this case, the figure of the human facilitator could be removed by adding to the system features that include direct communication between participants, supported by algorithms processing an ever-growing context-based (eg, time and location) list of conversational topics and media sharing requests. Conclusions The Media Parcels system was an effective approach to promote media sharing of emotional content for the elderly population that participated in this research. Although the system was tested in only 2 trios of users, it was suitable to promote communication and deepen social relations between participants from different generations (as in study 1) or from the same generation (as in study 2). It can be expected that similar results could be generalizable to the elderly population with characteristics comparable with our sample. In addition, the participants reported to feel motivated to produce and receive personal media content throughout the study. It is possible, though, that the interest in the use of the system could diminish over time because the system is not novel anymore or producing the media is too burdensome. Nevertheless, the social and personal nature of the media exchange could be motivating to keep users engaged in communication, especially if they feel socially isolated. Moreover, it could be easier for users to keep using the system for longer periods of time if the media requests for deep personal content are balanced with lightweight content and if the users are able to create their own threads of conversation. In fact, the use of a system such as Media Parcels is not necessarily intended to be of long-term. The length of use can be determined, for example, by a health care professional facilitator, focusing on shorter interactions between users according to therapeutic goals. The Media Parcels design presents a novel solution for including older adults in social media sharing by introducing the concept of intimate directed continuous slow media sharing, which is different from the existing online communities. Our trial showed that the parcel metaphor applied to media content was easily understood by the older population, and the supporting computational system was easy enough to be quickly adopted. As far as future study is concerned, the authors plan to conduct research expanding the number of participants in a group, targeting specifically older people that are classified as lonely or socially disconnected. Acknowledgments IZ was supported by São Paulo Research Foundation FAPESP (Grant Numbers 2017/09549-6 and 2015/18117-7). This study was also supported by funding from FAPESP (Grant Numbers 2016/00351-6 and 2016/50489-4), Conselho Nacional de Desenvolvimento Científico e Tecnológico (Grant Number 312058/2015-2), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Brazil—Finance Code001. Conflicts of Interest None declared. Multimedia Appendix 1 Relationship semantic differential scale. [PDF File (Adobe PDF File), 39 KB-Multimedia Appendix 1] Multimedia Appendix 2 Participants’ evaluation of their relationship toward one another according to RSDS scores. [PNG File , 8 KB-Multimedia Appendix 2] Multimedia Appendix 3 Participants’ evaluation of their relationship toward one another according to RSDS scores. [PNG File , 20 KB-Multimedia Appendix 3] References 1. 2. 3. Lee RM, Robbins SB. The relationship between social connectedness and anxiety, self-esteem, and social identity. J Couns Psychol 1998;45(3):338-345 [FREE Full text] [doi: 10.1037/0022-0167.45.3.338] Lee R, Steven B, Robbins SB. Understanding social connectedness in college women and men. J Couns Dev 2000;78(4):484-491. [doi: 10.1002/j.1556-6676.2000.tb01932.x] van Bel DT, Smolders KC, IJsselsteijn WA, de Kort Y. Social connectedness: concept and measurement. Intel Environ 2009;2:67-74. [doi: 10.3233/978-1-60750-034-6-67] https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 15 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al 4. Population Division - United Nations Population Division. 2017. World Population Prospects: Key Findings and Advance Tables URL: https://population.un.org/wpp/Publications/Files/WPP2017_KeyFindings.pdf [accessed 2019-08-15] [WebCite Cache ID 76i21AAN5] 5. World Health Organization. 2015. World Report on Ageing and Health URL: https://apps.who.int/iris/bitstream/handle/ 6. 10665/186463/9789240694811_eng.pdf [accessed 2019-08-15] [WebCite Cache ID 76i2Jr0WS] Bowling A, Dieppe P. What is successful ageing and who should define it? Br Med J 2005 Dec 24;331(7531):1548-1551 [FREE Full text] [doi: 10.1136/bmj.331.7531.1548] [Medline: 16373748] 7. Ward E, Barnes M, Gahagan B. The University of Brighton. 2012. Well-Being in Old Age: Findings From Participatory 8. Research URL: https://research.brighton.ac.uk/en/publications/well-being-in-old-age-findings-from-participatory-research [accessed 2019-08-15] [WebCite Cache ID 76i1Iq7qe] Grewal I, Lewis J, Flynn T, Brown J, Bond J, Coast J. Developing attributes for a generic quality of life measure for older people: preferences or capabilities? Soc Sci Med 2006 Apr;62(8):1891-1901. [doi: 10.1016/j.socscimed.2005.08.023] [Medline: 16168542] 9. McLaughlin SJ, Jette AM, Connell CM. An examination of healthy aging across a conceptual continuum: prevalence estimates, demographic patterns, and validity. J Gerontol A Biol Sci Med Sci 2012 Jun;67(7):783-789 [FREE Full text] [doi: 10.1093/gerona/glr234] [Medline: 22367432] 10. Cornwell B, Laumann EO, Schumm LP. The social connectedness of older adults: a national profile*. Am Sociol Rev 2008;73(2):185-203 [FREE Full text] [doi: 10.1177/000312240807300201] [Medline: 19018292] 11. World Health Organization. 2002. Active Ageing: A Policy Framework URL: https://apps.who.int/iris/bitstream/handle/ 10665/67215/WHO_NMH_NPH_02.8.pdf [accessed 2019-08-15] [WebCite Cache ID 76i2B1S9i] 13. 12. Office for National Statistics. 2018. Internet Access – Households and Individuals, Great Britain: 2018 URL: https://www. ons.gov.uk/peoplepopulationandcommunity/householdcharacteristics/homeinternetandsocialmediausage/bulletins/ internetaccesshouseholdsandindividuals/2018 [accessed 2019-08-15] IBGE \| Agência de Notícias. 2017. Continuous PNAD - ICT 2017: Internet Reaches Three of Every Four Households in the Country URL: https://agenciadenoticias.ibge.gov.br/en/agencia-press-room/2185-news-agency/releases-en/ 23453-continuous-pnad-ict-2017-internet-reaches-three-in-every-four-households-in-the-country [accessed 2019-08-15] Pew Research Center. 2019. Mobile Fact Sheet URL: https://www.pewinternet.org/fact-sheet/mobile/ [accessed 2019-08-15] Pimentel MG, Cunha BC, Antonelli HL, Rodrigues SS, Neto OJ, Rocha AC, et al. Enhancing Older Adults Connectivity by Introducing Mobile Devices Communication Tools. In: Proceedings of the 7th International Conference on Software Development and Technologies for Enhancing Accessibility and Fighting Info-Exclusion. 2016 Presented at: DSAI'16; December 1-3, 2016; Vila Real, Portugal p. 353-361 URL: https://dl.acm.org/citation.cfm?id=3019994 [doi: 10.1145/3019943.3019994] 14. 15. 16. Lindley SE, Harper R, Sellen A. Designing for Elders: Exploring the Complexity of Relationships in Later Life. In: Proceedings of the 22nd British HCI Group Annual Conference on People and Computers: Culture, Creativity, Interaction. 2008 Presented at: BCS-HCI'08; September 1-5, 2008; Liverpool, United Kingdom p. 77-86. [doi: 10.1145/1531514.1531525] Pedell S, Vetere F, Kulik L, Ozanne E, Gruner A. Social Isolation of Older People: The Role of Domestic Technologies. In: Proceedings of the 22nd Conference of the Computer-Human Interaction Special Interest Group of Australia on Computer-Human Interaction. 2010 Presented at: OZCHI'10; November 22-26, 2010; Brisbane, Australia p. 164-167. [doi: 10.1145/1952222.1952255] van Dijck J. Mediated Memories in the Digital Age. First Edition. Palo Alto, California: Stanford University Press; 2007. Sarvas R, Frohlich DM. From Snapshots to Social Media - The Changing Picture of Domestic Photography. First Edition. London: Springer; 2011. 17. 18. 19. 20. Dib L, Petrelli D, Whittaker S. Sonic Souvenirs: Exploring the Paradoxes of Recorded Sound for Family Remembering. In: Proceedings of the 2010 ACM Conference on Computer Supported Cooperative Work. 2010 Presented at: CSCW'10; February 6-10, 2010; Savannah, Georgia, USA p. 391-400. [doi: 10.1145/1718918.1718985] 21. Kirk DS, Sellen A. On human remains: values and practice in the home archiving of cherished objects. ACM Trans 22. Comput-Hum Interact 2010 Jul 1;17(3):10-53. [doi: 10.1145/1806923.1806924] Park N, Jin B, Jin SA. Effects of self-disclosure on relational intimacy in Facebook. Comput Human Behav 2011 Sep;27(5):1974-1983. [doi: 10.1016/j.chb.2011.05.004] 23. Orben AC, Dunbar RI. Social media and relationship development: the effect of valence and intimacy of posts. Comput Human Behav 2017 Aug;73:489-498. [doi: 10.1016/j.chb.2017.04.006] 24. Rebelo C. The use of the internet and Facebook by the elders in Portugal: an exploratory study. Observatorio 2015;9(3):129-153 [FREE Full text] 25. Chakraborty R, Vishik C, Rao HR. Privacy preserving actions of older adults on social media: exploring the behavior of opting out of information sharing. Decis Support Syst 2013 Nov;55(4):948-956. [doi: 10.1016/j.dss.2013.01.004] Sinclair TJ, Grieve R. Facebook as a source of social connectedness in older adults. Comput Human Behav 2017 Jan;66(5):363-369. [doi: 10.1016/j.chb.2016.10.003] 26. 27. Quinn K. Cognitive effects of social media use: a case of older adults. Soc Media Soc 2018 Jul 19;4(3):205630511878720. [doi: 10.1177/2056305118787203] https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 16 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al 28. Chen YR, Schulz PJ. The effect of information communication technology interventions on reducing social isolation in the elderly: a systematic review. J Med Internet Res 2016 Jan 28;18(1):e18 [FREE Full text] [doi: 10.2196/jmir.4596] [Medline: 26822073] 29. Karimi A, Neustaedter C. From High Connectivity to Social Isolation: Communication Practices of Older Adults in the Digital Age. In: Proceedings of the ACM 2012 Conference on Computer Supported Cooperative Work Companion. 2012 Presented at: CSCW'12; February 11-15, 2012; Seattle, Washington, USA p. 127-130 URL: http://doi.acm.org/10.1145/ 2141512.2141559 [doi: 10.1145/2141512.2141559] 30. Taipale S, Farinosi M. The Big Meaning of Small Messages: The Use of WhatsApp in Intergenerational Family Communication. In: Proceedings of the International Conference on Human Aspects of IT for the Aged Population. 2018 Presented at: ITAP'18; July 15–20, 2018; Las Vegas, NV, USA p. 532-546 URL: https://link.springer.com/chapter/10.1007/ 978-3-319-92034-4_40 [doi: 10.1007/978-3-319-92034-4_40] 31. Waycott J, Davis H, Vetere F, Morgans A, Gruner A, Ozanne E, et al. Captioned Photographs in Psychosocial Aged Care: Relationship Building and Boundary Work. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2014 Presented at: CHI'14; April 26-May 1, 2014; Toronto, Ontario, Canada p. 4167-4176 URL: https://dl.acm.org/ citation.cfm?id=2557290 [doi: 10.1145/2556288.2557290] 32. Cornejo R, Tentori M, Favela J. Enriching in-person encounters through social media: a study on family connectedness for the elderly. Int J Hum-Comput St 2013 Sep;71(9):889-899. [doi: 10.1016/j.ijhcs.2013.04.001] 33. Zargham S, Ćalić J, Frohlich DM. 4streams: An Ambient Photo Sharing Application for Extended Families. In: Proceedings of the 2015 British HCI Conference. 2015 Presented at: British HCI'15; July 13-17, 2015; Lincoln, Lincolnshire, United Kingdom p. 165-174 URL: https://dl.acm.org/citation.cfm?id=2783589 [doi: 10.1145/2783446.2783589] 34. Garattini C, Wherton J, Prendergast D. Linking the lonely: an exploration of a communication technology designed to support social interaction among older adults. Universal Access Inf Soc 2011 Jun 7;11(2):211-222. [doi: 10.1007/s10209-011-0235-y] 35. Abrahão AR, da Silva PF, Frohlich DM, Chrysanthaki T, Gratão A, Castro P. Mobile Digital Storytelling in a Brazilian Care Home. In: Proceedings of the International Conference on Human Aspects of IT for the Aged Population. 2018 Presented at: ITAP'18; July 15-20, 2018; Las Vegas, NV, USA p. 403-421. [doi: 10.1007/978-3-319-92034-4_31] 36. Zaine I, Rodrigues KR, Cunha BC, Viel CC, Orlando AF, Neto OJ, et al. ESPIM: An Ubiquitous Data Collection and Programmed Intervention System using ESM and Mobile Devices. In: Proceedings of the 22nd Brazilian Symposium on Multimedia and the Web. 2016 Presented at: Webmedia'16; November 8-11, 2016; Teresina, Piauí State, Brazil p. 13-14 URL: https://dl.acm.org/citation.cfm?id=2988222 [doi: 10.1145/2976796.2988222] 37. Viel CC, Rodrigues KR, Zaine I, Cunha BC, Scalco LF, Pimentel MG. Personalized Ubiquitous Data Collection and Intervention as Interactive Multimedia Documents. In: Proceedings of the 2017 ACM Symposium on Document Engineering. 2017 Presented at: DocEng'17; September 4-7, 2017; Valletta, Malta p. 223-226 URL: https://dl.acm.org/citation. cfm?id=3121046 [doi: 10.1145/3103010.3121046] 38. Csikszentmihalyi M, Larson R. Validity and reliability of the experience-sampling method. J Nerv Ment Dis 1987 Sep;175(9):526-536. [doi: 10.1097/00005053-198709000-00004] [Medline: 3655778] Skinner BF. Programmed instruction revisited. Phi Delta Kappan 1986;68(2):103-110 [FREE Full text] 39. 40. Cunha BC, Rodrigues KH, Zaine I, Scalco LF, Viel CC, Pimentel MG. Web-Based Authoring of Multimedia Intervention Programs for Mobile Devices: A Case Study on Elderly Digital Literacy. In: Proceedings of the 34th ACM/SIGAPP Symposium on Applied Computing. 2019 Presented at: SAC'19; April 8-12, 2019; Limassol, Cyprus p. 484-491. [doi: 10.1145/3297280.3297325] Schlochtermeier LH, Pehrs C, Kappelhoff H, Kuchinke L, Jacobs AM. Emotion processing in different media types: realism, complexity, and immersion. J Syst Integr Neurosci 2015;1(2):41-47. [doi: 10.15761/JSIN.1000109] 41. 42. Dougher MJ, Hayes SC. Clinical Behavior Analysis. Oakland, California: Context Press; 2000. 43. Gaver WW, Dunne A, Pacenti E. Design: cultural probes. Interactions 1999;6(1):21-29. [doi: 10.1145/291224.291235] 44. Folstein MF, Folstein SE, McHugh PR. 'Mini-mental state'. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975 Nov;12(3):189-198. [doi: 10.1016/0022-3956(75)90026-6] [Medline: 1202204] 45. Osgood CE, Suci GJ, Tannenbaum P. The Measurement of Meaning. First Edition. Champaign, USA: University of Illinois Press; 1957. 46. Neves BB, Amaro F. Too old for technology? How the elderly of Lisbon use and perceive ICT. J Community Inform 2012;8(1):189-194 [FREE Full text] 47. Laugwitz B, Held T, Schrepp M. Construction and Evaluation of a User Experience Questionnaire. In: Proceedings of the Symposium of the Austrian HCI and Usability Engineering Group. 2008 Presented at: USAB'08; November 20-21, 2008; Graz, Austria p. 63-76. [doi: 10.1007/978-3-540-89350-9_6] 48. Brooke J. SUS: a quick and dirty usability scale. In: Jordan PW, Thomas B, Weerdmeester BA, McClelland IL, editors. Usability Evaluation In Industry. London: Taylor & Francis; 1996:189-194. 49. Díaz-Bossini JM, Moreno L. Accessibility to mobile interfaces for older people. Procedia Comput Sci 2014;27:57-66 [FREE Full text] [doi: 10.1016/j.procs.2014.02.008] https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 17 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Zaine et al 50. Petrovčič A, Taipale S, Rogelj A, Dolničar V. Design of mobile phones for older adults: an empirical analysis of design guidelines and checklists for feature phones and smartphones. Int J Hum-Comput Int 2017 Sep 5;34(3):251-264. [doi: 10.1080/10447318.2017.1345142] 51. Lindley SE, Harper R, Sellen A. Desiring to Be in Touch in a Changing Communications Landscape: Attitudes of Older Adults. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 2009 Presented at: CHI'09; April 4-9, 2009; Boston, MA, USA p. 1493-1702. [doi: 10.1145/1518701.1518962] 52. Carstensen LL, Fung HH, Charles ST. Socioemotional selectivity theory and the regulation of emotion in the second half of life. Motiv Emot 2003;27(2):103-123 [FREE Full text] [doi: 10.1023/A:1024569803230] 53. Carstensen LL, Gross JJ, Fung HH. The social context of emotional experience. Annu Rev Gerontol Geriatr 1998;17(1):325-352 [FREE Full text] [doi: 10.1891/0198-8794.17.1.325] 54. Rook KS. Reciprocity of social exchange and social satisfaction among older women. J Pers Soc Psychol 1987;52(1):145-154 [FREE Full text] [doi: 10.1037/0022-3514.52.1.145] 55. Adams RG, Blieszner R. Aging well with friends and family. Am Behav Sci 2016 Jul 27;39(2):209-224 [FREE Full text] 56. 57. [doi: 10.1177/0002764295039002008] Fiori KL, Antonucci TC, Cortina KS. Social network typologies and mental health among older adults. J Gerontol B Psychol Sci Soc Sci 2006 Jan 1;61(1):P25-P32. [doi: 10.1093/geronb/61.1.P25] Frohlich DM, Wall S, Kiddle G. Rediscovery of forgotten images in domestic photo collections. Pers Ubiquitous Comput 2012 Nov 8;17(4):729-740 [FREE Full text] [doi: 10.1007/s00779-012-0612-4] 58. Welsh D, Morrissey K, Foley S, McNaney R, Salis C, McCarthy J, et al. Ticket to Talk: Supporting Conversation Between Young People and People with Dementia Through Digital Media. In: Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems. 2018 Presented at: CHI'18; April 21-26, 2018; Montreal QC, Canada. [doi: 10.1145/3173574.3173949] Abbreviations ESPIM: Experience Sampling and Programmed Intervention Method RSDS: Relationship Semantic Differential Scale SUS: System Usability Scale UEQ: User Experience Questionnaire Edited by G Eysenbach; submitted 29.03.19; peer-reviewed by J Favela, W Zhang, M Khazaee-Pool; comments to author 18.05.19; revised version received 12.07.19; accepted 21.07.19; published 03.10.19 Please cite as: Zaine I, Frohlich DM, Rodrigues KRDH, Cunha BCR, Orlando AF, Scalco LF, Pimentel MDGC Promoting Social Connection and Deepening Relations Among Older Adults: Design and Qualitative Evaluation of Media Parcels J Med Internet Res 2019;21(10):e14112 URL: https://www.jmir.org/2019/10/e14112 doi: 10.2196/14112 PMID: 31584001 ©Isabela Zaine, David Mark Frohlich, Kamila Rios Da Hora Rodrigues, Bruna Carolina Rodrigues Cunha, Alex Fernando Orlando, Leonardo Fernandes Scalco, Maria Da Graça Campos Pimentel. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 03.10.2019 This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2019/10/e14112 XSL•FO RenderX J Med Internet Res 2019 \| vol. 21 \| iss. 10 \| e14112 \| p. 18 (page number not for citation purposes)
10.1371_journal.pwat.0000137	RESEARCH ARTICLE Disparities in disruptions to public drinking water services in Texas communities during Winter Storm Uri 2021 Brianna Tomko1, Christine L. Nittrouer2, Xavier Sanchez-VilaID 3, Audrey H. SawyerID 1* 1 School of Earth Sciences, The Ohio State University, Columbus, OH, United States of America, 2 Department of Management, Rawls College of Business, Texas Tech University, Lubbock, TX, United States of America, 3 Department of Civil and Environmental Engineering, Universitat Politècnica de Catalunya, Barcelona, Spain * [email protected] Abstract Winter Storm Uri of February 2021 left millions of United States residents without access to reliable, clean domestic water during the COVID19 pandemic. In the state of Texas, over 17 million people served by public drinking water systems were placed under boil water adviso- ries for periods ranging from one day to more than one month. We performed a geospatial analysis that combined public boil water advisory data for Texas with demographic informa- tion from the 2010 United States Census to understand the affected public water systems and the populations they served. We also issued a cross-sectional survey to account for people’s lived experiences. Geospatial analysis shows that the duration of boil water adviso- ries depended partly on the size of the public water system. Large, urban public water sys- tems issued advisories of intermediate length (5–7 days) and served racially diverse communities of moderate income. Small, mostly rural public water systems issued some of the longest advisories (20 days or more). Many of these systems served disproportionately White communities of lower income, but some served predominantly non-White, Hispanic, and Latino communities. In survey data, “first-generation” participants (whose parents were not college-educated) were more likely to be placed under boil water advisories, pointing to disparate impacts by socioeconomic group. The survey also revealed large communication gaps between public water utilities and individuals: more than half of all respondents were unsure or confused about whether they were issued a boil water advisory. Our study rein- forces the need to improve resilience in public water services for large, diverse, urban com- munities and small, rural communities in the United States and to provide a clear and efficient channel for emergency communications between public water service utilities and the communities they serve. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Tomko B, Nittrouer CL, Sanchez-Vila X, Sawyer AH (2023) Disparities in disruptions to public drinking water services in Texas communities during Winter Storm Uri 2021. PLOS Water 2(6): e0000137. https://doi.org/10.1371/ journal.pwat.0000137 Editor: Majid Shafiee-Jood, University of Virginia, UNITED STATES Received: December 16, 2022 Accepted: May 8, 2023 Published: June 21, 2023 Copyright: © 2023 Tomko et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: Data have been uploaded to Zenodo 10.5281/zenodo.7447637. Funding: The authors received no specific funding for this work. 1. Introduction Competing interests: The authors have declared that no competing interests exist. Water is an essential resource that is unevenly distributed. For at least one month each year, two-thirds of the world’s population experiences conditions of severe water scarcity [1]. As the PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 1 / 16 Drinking water access during winter Storm Uri climate changes, water scarcity will persist for many communities [2] and expand to new ones. Extreme weather events such as droughts and floods are also expected to increase in frequency and severity in the coming decades, creating further disruptions to water supply and accessibil- ity. Droughts have triggered partial shutoffs of municipal water services in locations around the world. A prominent example is Cape Town, South Africa’s municipal water crisis in 2018, when the city narrowly averted a total shutdown of municipal water services due to drought conditions and water management decisions [3]. Droughts place heavy pressure on surface water resources and, to a lesser extent, on groundwater resources, which are inherently more insulated against drought. Floods damage infrastructure and test the limits of water treatment technologies. Severe weather such as winter storms bring freezing temperatures that damage pipes and can also disrupt the power supply to water treatment facilities. In the United States, 97% of the population has access to improved water [4], but supply was disrupted across Central and Gulf Coast states during Winter Storm Uri (February, 2021). The state of Texas experienced severe, but not unprecedented, cold temperatures in the teens and single digits in Fahrenheit (around -10 to -15˚C) [5]. Due to infrastructure damage, the storm left millions without power and water for days [6, 7]. The storm was estimated to have caused over 200 deaths and created about $100 billion in financial losses in Texas [8]. Over two out of three (69%) Texans lost electrical power at some point during the storm. Almost half (49%) reported losing access to running water, on average, for 52 hours [9]. Those with uninterrupted access to running water still reported that their water was unpotable for an aver- age of 40 hours during the week of the storm (for example, they were issued a boil water advi- sory). More than half (56%) of those who lost potable water considered the loss to be extremely serious or very serious. Nearly half (45%) also experienced difficulties finding bot- tled water, rating this impact as very serious or extremely serious. For comparison, loss of cell phone service, difficulties obtaining food, and illness or injury to immediate family were all rated less serious in terms of impact [9]. Loss of domestic water disproportionately affects the health of vulnerable populations such as children, older adults, and low-income individuals [7]. The combination of winter weather, which made it difficult to use public transportation, and the ongoing COVID-19 pandemic hindered access to bottled water supplies, particularly for older adults and low-income individuals [7]. The loss of clean, domestic water also made it harder for households to follow World Health Organization (WHO) and United Nations Chil- dren’s Fund (UNICEF) guidelines for handwashing, a key strategy to reduce the spread of the virus [10]. Amidst these factors, it is important to understand the effects of Winter Storm Uri on the public supply of drinking water. Winter Storm Uri was the largest known boil water event in U.S. history [11] and reflects the resilience of public water services in the region. Here, resil- ience is defined as “the ability to plan for, absorb, recover from, or more successfully adapt to actual or potential adverse effects” [12]. Existing analyses of this historic boil water event point to knock-on effects from power outages [6]. In a survey of large, mostly urban public water utilities, 85% lost power, impacting their ability to produce clean water [13]. Though many had backup generators, not all were operable due to low fuel supplies or cold temperatures. Additionally, water losses from burst pipes and leaks caused pressures to drop in many distri- bution systems below the regulatory minimum, triggering the issuance of boil water advisories to protect customers from pathogens that tend to infiltrate under low pressure [13]. By under- standing which public systems were most affected and the contributing factors, new strategies can be implemented to increase the resilience of public drinking water systems under future weather extremes. It is also important to identify communities that may be disproportionately vulnerable to disruption of services in order to prioritize equitable responses [14]. Across urban areas of the PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 2 / 16 PLOS WATERDrinking water access during winter Storm Uri United States, consistent disparities in piped water access have been linked to unpredictable housing conditions and racialized wealth gaps [15]. In peri-urban and ex-urban areas, munici- pal underbounding has excluded low-income neighborhoods and people of color from public water services altogether [16, 17]. Further, studies demonstrate that Black and Latino individu- als, and generally those at lower income levels, have less access to clean water [18] and indoor plumbing [19]. In the specific case of Winter Storm Uri, multiple studies have identified dis- parities in utility outages across racial, ethnic, and income groups. For example, Nejat et al. [20] showed that communities with a great proportion of non-Hispanic White residents, single family homes, and greater income experienced a smaller share of lingering power outages after the storm. Grineski et al. [21] revealed through a survey that Black participants were more likely to experience longer water outages. By analyzing 311 calls for the Houston area, Lee et al. [22] showed that burst pipes were more severe for low-income and racial minority groups. Glazer et al. [11] examined power and water outages by county and compared them with an index of social vulnerability. Although there was no clear correlation between the length of boil water advisories and social vulnerability at the county level, they observed that some of the most impacted counties had greater percentages of non-English speakers and minority resi- dents, apartment complexes, and mobile homes. They therefore suggested the need for a more granular analysis on a census tract level. Here, we sought to understand the finer-scale impacts to Texas public water systems and identify groups that were most affected through two related studies: 1) a geospatial analysis of community public water systems that issued boil water advisories, and 2) a cross-sectional sur- vey of individuals residing in Texas during February 2021. In the geospatial analysis, we used principal component analysis to test whether there was any relation between the length of the advisory (a measure of the recovery time for safe drinking water services after the storm) and various factors describing the public drinking water system. These factors included size or location of the public water system, severity of weather, and demographics at the system level (derived from mapping public census data onto each water system’s service area). We also asked how the advisories impacted specific demographic groups at the individual level through the cross-sectional survey using both an analysis of covariance and a multivariate analysis of covariance on human subjects data. Drawing on the integrated findings from these two stud- ies, we were also able to examine individual awareness of boil water advisories and explore communication gaps between public water providers and customers. 2. Method and materials 2.1. Geospatial analysis To gather data on boil water advisories issued by public water systems, a request for informa- tion was placed with the Texas Commission of Environmental Quality (TCEQ) under the Texas Public Information Act. We limited the request to community public water systems, defined as those that have the potential to serve at least fifteen residential connections or twenty-five residents on a year-round basis [23], and excluded other public water systems such as schools, hospitals, and seasonal communities. Most of the Texas population (roughly 27 mil- lion people, or 93%) is served by community public water systems [24] and therefore is repre- sented in the geospatial component of this study. Roughly 5% of Texans depend on private well water [25] and are not included in the geospatial analysis. TCEQ provided a list of 2,080 community public water systems that reported issuing a boil water advisory related to Winter Storm Uri. The list includes the name of the public water sys- tem, a unique identifier code, the county, the issue date of the boil water notice, the rescind date of the boil water notice, the population served, and the number of connections served. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 3 / 16 PLOS WATERDrinking water access during winter Storm Uri Retail service areas for community public water systems were obtained from the Texas Water Development Board’s (TWDB) water service boundary viewer in November of 2021 [26]. It is worth noting that as of 2022, Texas is one of only 24 states with geospatial data products for community water system service area boundaries [27]. The Texas dataset includes 4,572 out of 4,641 community public water systems [28]. Fourteen of the 2,080 public water systems that issued boil water advisories did not have a service area polygon, so they were excluded from the analysis. The remaining 2,066 records were screened for completeness and to ensure that the dates of boil water advisories were consistent with Winter Storm Uri. A small number of records [28] were removed from the analysis because they were incomplete, or the reported advisory was not conclusively connected to Winter Storm Uri. The excluded records fit at least one of these exclusion criteria: 1) the reported advisory was issued and rescinded prior to Win- ter Storm Uri; 2) the reported advisory was issued before the storm hit, and local minimum temperatures never fell below freezing; 3) no rescind date was provided. In total, 2,038 public water systems were retained for analysis. To relate information on boil water advisories to weather, daily climate summaries were retrieved from the National Oceanic and Atmospheric Administration (NOAA) from Febru- ary 1, 2021 to February 28, 2021 for 360 weather stations in the state of Texas [29]. Some of the longest boil water advisories were not lifted until March, but the month of February fully encompassed the climatological phenomenon of Winter Storm Uri; thus, we restrict our inves- tigation of the climatological phenomenon (for example, how long temperatures stayed below freezing) to February. A point feature class shapefile was created in ArcGIS for the weather sta- tions with attributes containing the minimum and maximum recorded temperatures for each day in February. We also calculated the sum of the number of days in February that the maxi- mum or minimum daily temperature was below freezing. Some stations had missing data for maximum and minimum temperatures on select days. No attempt was made to interpolate missing data because it is possible that data gaps are temperature-dependent or biased towards frozen temperatures. Our estimation of the number of frozen days is therefore conservative, meaning that the number of days below freezing may be underestimated, and the minimum daily recorded temperature may be overestimated. Weather data from the nearest station was attributed to each public water system using a spatial join with the nearest neighbor in ArcGIS. Information on urban and rural households, population, race, and housing tenure were obtained for the state of Texas from the 2010 decennial United States Census using the R pack- age tidycensus [30]. We opted for the 2010 census instead of the more recent 2020 census because the results of the 2020 decennial census and American Community Survey were impacted by the COVID-19 pandemic, and income data were only available as experimental estimates [24]. We acknowledge that Texas has experienced substantial growth and demo- graphic change since 2010, which introduces additional uncertainty to our analysis. Medium income and its margin of error (MOE) were taken from the 2010 American Community Sur- vey. The U.S. Census Bureau organizes data based on spatial hierarchy, ranging from states down to blocks, the smallest measurement scale. Income data are not available at the block level, but they are at the next largest block group level. Initial analyses showed that block group-level calculations compared well with block-level calculations for public water systems [31]. We, therefore, chose to analyze demographic information for block groups because there is simplicity and advantage to working at one consistent spatial level for demographic and income data. In ArcGIS Pro, areas were calculated for both block groups and public water systems. Over- lapping areas between the census block groups and public water systems were then used to compute aerially-weighted average demographics for each public water system. Further infor- mation is provided in Section 1 in S1 Text. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 4 / 16 PLOS WATERDrinking water access during winter Storm Uri To explore relationships between public water system characteristics, we performed infer- ential statistical analyses, including Pearson correlation coefficients (which provide a measure of linear correlation between two variables) and principal component analysis (PCA) using MATLAB. The goal of PCA is to reduce the dimensionality of the data set by finding the com- bination of variables that best explain the total variance [32]. Variables that displayed strong positive skewness were logarithmically transformed (specifically, number of advisory days, ser- vice area, homes served, and median income). All variables were then scaled to have a mean of 0 and a standard deviation of 1. We chose 15 variables to be included in the final matrix of cor- relation coefficients, detailed in the Results. These variables were selected to represent a range of conditions (meteorological: extent and duration of freezing temperatures; geographic: lati- tude and longitude, degree of urbanization; scale of the system: size of service area and number of homes served; and demographics: race, ethnicity, homeownership, and income), with the goals of understanding which public water systems took the longest to recover, who was affected, and for how long. We explored different combinations of variables (e.g., minimum of daily minimum temperature versus minimum of daily maximum temperature; population served versus homes served; fraction of White individuals versus fraction of White families) and found negligible differences; duplicated variables were removed. Last, the dataset was sub- jected to PCA to test whether there were any underlying patterns in the public water systems that issued short or long boil water advisories. We removed the length of the boil water advi- sory as a variable from the analysis, so that public water systems were only described by geo- graphic, meteorological, and demographic variables. We also chose to eliminate elevation, a variable strongly correlated with longitude and latitude, which was found in preliminary analy- sis to have a negligible effect on the amount of variance explained by the first two components in the PCA analysis, leaving a total of 13 remaining variables for the final statistical analysis. 2.2 Survey We cross-sectionally surveyed 407 people who lived in Texas during Winter Storm Uri. We received IRB approval (#IRB2021-882) from Texas Tech University to conduct this study. Par- ticipants were asked to indicate their consent to participate in the first question on the survey, and we only collected data from individuals who indicated their consent on this question. Par- ticipants were permitted to skip any question in the survey they did not feel comfortable answering without penalty. Participants were recruited from January 2022—April 2022. We retained only those participants without missing data who could also be located and thus con- nected to public water system service areas in the geospatial analysis (N = 289). Of these, N = 23 people in our data set were not on public water systems at all, so we excluded these individuals from our analyses regarding public water systems. Thus, our final sample consists of N = 266 participants. Importantly, these individuals self-identify as 56% (149) female, 43% (114) male, 1% (2) non-binary, and .5% (1) preferred not to say. Regarding race and ethnicity, these individuals self-identify as 71% (188) White and Non-Hispanic, 21% (57) Latino or His- panic, 3% (9) Asian or Pacific Islander, 4% (10) Black and Non-Hispanic, and 1% (2) who chose to write-in their responses (see Section 3 in S1 Text for details regarding how these cate- gories were combined and how identities differ slightly from categories in the United States Census). Our sample is on average 22.88 years old (ranging from 18 to 80 years). We associated participants with their public water system (if they lived in one) by asking them to provide a zip code and street address where they were living during the storm, or alter- natively, two nearby cross-streets, if they were uncomfortable listing their street address. This information was provided by N = 266 participants. Using a Google Sheets plug-in called Geo- Code, we generated longitude and latitude for each of these participants (Fig B in S1 Text) and PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 5 / 16 PLOS WATERDrinking water access during winter Storm Uri performed a spatial join in ArcGIS to the feature class of public water systems. Because some participants provided nearby cross-streets rather than exact addresses, we experimented with including a 500-m buffer, which only affected the spatial join for 4 participants (thus, we did not use this buffer). We asked participants a series of thirteen questions related to their access to clean water during Winter Storm Uri, their experiences with boil water advisories, electricity and Wi-Fi outages, and burst pipes or water damage due to the storm. Finally, we asked them a series of demographic questions related to their gender, race and ethnicity, age, income, and familial education levels (all survey questions are available in Section 3 in S1 Text). We defined “first- generation” status as any participant whose parents had not completed college (of note, all par- ticipants had received at least some college training themselves). First-generation status tends to be positively correlated with families who are also low income [33, 34]. To control for the differences across these variables’ scales, we z-transformed each variable before analysis. More information is provided in Section 3 in S1 Text. To explore how Participant Ethnicity, Income, and First-Generation Status related to their experiences with Water Access, we conducted first (1) an analysis of covariance (ANCOVA) on a binary variable indicating whether participants were issued a boil water advisory, derived from the linked geospatial data; and second (2) a multivariate analysis of covariance (MAN- COVA) on the thirteen survey items; we also controlled for two variables across both analyses from the geospatial study: the fraction of rural households (System-Level Rural Housing) as a measure of urban development in the participant’s area and the total number of households (System-Level Total Housing) as a measure of the scale of the public water system that served the participant. 3. Results 3.1. Analysis of public water systems Most boil water advisories were issued on February 17, approximately one day after freezing weather descended on the state of Texas (Fig 1A). Advisories were lifted anywhere from 1 to 36 days later, though below-freezing weather only lasted a maximum of 12 days (Fig 1B). The mean length of an advisory was 7.96 days (d), with an standard deviation of 3.57 d. Meanwhile, the mean length of below-freezing weather was 3.83 d, and the standard deviation was 3.26 d (Fig 1B). Some of the coldest weather occurred in the northwestern regions of Texas farther from the coast (Fig 1D), and consequently at higher latitude and longitude (for example, r = 0.64, p < 0.01 for latitude and frozen days in Fig 2). In contrast, the duration of advisories was scattered and showed no clear spatial trends with latitude or longitude (Fig 1C). As further evidence, global Moran’s I (a measure of spatial autocorrelation calculated here with a binary, nearest-neighbors weighting system) was 0.977 and 0.988 for minimum recorded temperature and number of freezing days, respectively, indicating smoothly varying weather conditions. Moran’s I was only 0.139 for the length of the advisory, indicating a more random distribution. Indeed, the length of the boil water advisory was weakly (linearly) correlated with all the individual weather and sociodemographic factors analyzed, according to the values of bivariate Pearson correlation coefficients (Fig 2). Public water systems that served a smaller number of homes had a weak tendency to issue longer advisories (r = -0.25, p < 0.01). Those public water systems that served a higher proportion of families who owned their homes outright also had a weak tendency to issue longer advisories (r = 0.19, p < 0.01). It is important to note that at the scale of public water systems, greater rates of homeownership were consistent with lower median income (r = -0.44, p < 0.01) and more rural service areas (r = 0.24, p < 0.01), meaning PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 6 / 16 PLOS WATERDrinking water access during winter Storm Uri Fig 1. Patterns in the length of boil water advisories differ from patterns in cold weather. (A) Histograms showing when advisories were issued and lifted (total number of samples, N = 2,038). (B) Histograms showing length of advisories and length of time the maximum daily temperature was below freezing in February. (C) Map of boil water advisories duration. (D) Map of freezing weather duration (number of days in February when the maximum daily temperature was below freezing). Map base layer and technical documentation available from U.S. Census 2010 TIGER/Line shapefiles for Texas. https://doi.org/10.1371/journal.pwat.0000137.g001 that homeownership rates are not an indicator of wealth when aggregated by public water sys- tem and compared across urban and rural areas. In fact, public water systems with greater median income tended to have a greater fraction of mortgaged homes (r = 0.76, p < .01) and be more urban (r = 0.24, p < .01). Within public water systems, the fraction of White families and Black or African American families was strongly negatively correlated (r = -0.86, p < 0.01), and a weak negative correla- tion was also evident between White families and families of Hispanic or Latino ethnicity (r = -0.38, p < 0.01). This outcome is not forced by having compositional variables that sum to one, as race and ethnicity are independent and overlapping categories in the census data. Also, families could identify with additional races that include American Indian or Alaska Native, Asian, and Native Hawaiian or Other Pacific Islander. The public water systems that served greater proportions of White families tended to be more rural (less urban, r = -0.50, p < 0.01), whereas the public water systems that served greater proportions of Black or African American families and Hispanic or Latino families were mostly urban (r = 0.32 and r = 0.33, respectively, with p < 0.01 in both cases). The bivariate correlations in Fig 2 do not provide a comprehensive vision of the dataset. For this reason, we developed a multivariate interpretation by means of PCA. We found that approximately half (48%) of the geographic, meteorological, and demographic variability among public water systems was explained by only the first two principal components (Fig 3B). PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 7 / 16 PLOS WATERDrinking water access during winter Storm Uri Fig 2. Correlation matrix for public water systems that issued an advisory in response to Winter Storm Uri. https://doi.org/10.1371/journal.pwat.0000137.g002 PCA resulted in geographic and weather-related variables being projected onto the first and third quadrants in the plot of components 1 and 2 (Fig 3B). For example, public water systems at greater latitude that experienced more freezing days in February of 2021 project toward the first quadrant. Meanwhile, variables that describe the scale and demographics of the service area projected more or less orthogonally. For example, public water systems that served greater number of homes and were located in more urban areas are projected toward the second quad- rant. These public water systems were also associated with a greater proportion of rented homes and lower proportion of White families. The four public water systems that served the greatest number of customers (all >1 million, as reported to TCEQ by the providers) clustered in the second quadrant and all experienced advisories of intermediate length (within mean plus one standard deviation). These include the City of Houston, San Antonio Water System, City of Fort Worth, and City of Austin Water & Wastewater (Fig 3B). In contrast, the public water systems associated with some of the longest advisories tended to cluster in the fourth quadrant. Specifically, 14 of the 15 systems with the longest advisories (all 20 days or more) were similar in terms of their small populations and service areas and their tendency to serve PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 8 / 16 PLOS WATERDrinking water access during winter Storm Uri Fig 3. Principal component analysis of public water systems. A) Percent of variance among public water systems explained as a function of the number of components. B) Projection of public water system data on a graph of principal components 1 and 2. Each point represents a public water system that issued a boil water advisory, colored according to length of the advisory. Large diamonds indicate the 15 public water systems with the longest boil water advisories. Large circles show the 4 public water systems that serve the largest populations. Vectors show the projection of the geographic, meteorological, and demographic variables involved in the analysis. https://doi.org/10.1371/journal.pwat.0000137.g003 more rural, homeowning families (Table A in S1 Text). The one system that did not cluster with the 14 others was a small, rural system that served a large proportion (51%) of Black or African American families (Table A in S1 Text). Interestingly, public water systems with very short boil water advisories (1–2 days) did not cluster strongly according to the first two princi- pal components (Fig 3A). In summary (Table 1), the geospatial analysis suggests that no one factor resoundingly explains recovery times for public water systems, but large, urban systems consistently issued advisories of intermediate length affecting large, diverse communities. The longest recovery times were experienced by small, rural systems of variable community demographics (Table 1). 3.2. Individual experiences and awareness Eighteen percent (N = 48) of the 266 participants who were on public water systems stated that they were under a boil water advisory during the storm. Meanwhile, 37% (N = 98) were not sure if they were under a boil water advisory, and 45% (N = 120) stated they were not under a boil water advisory. Interestingly, 53% (N = 29 people) who said they were under a boil water advisory actually were not, based on their locations at the time of the storm; 10% (N = 11) of those who were not sure actually were; and 5% (N = 6) of those who said they were not under a boil water advisory actually were. This finding highlights gaps in the way advisories were communicated to the public (Fig C and Table B in S1 Text). These gaps did not appear to vary strongly across racial and ethnic groups (Fig C in S1 Text). A majority of participants (69%; N = 183) reported that they did not lose access to water where they were living, while 31% (N = 76) did lose access to some degree. Specifically, 16% (N = 43) lost access to running water altogether (no water flowed when they turned on their tap); 12% (N = 33) experienced some change in water pressure. A majority (91%; N = 243) stated that they did not notice any visible changes in the color, taste, or smell of their water. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 9 / 16 PLOS WATERTable 1. Summary of key findings from geospatial analysis, the survey tool, and their relationships. Sample How extensive were boil water advisories, and what was the impact to surveyed individuals? System-Level Geospatial Analysis 2,038 community public drinking water systems 44% of the 4,572 community public water systems with mapped service areas in Texas issued boil water advisories. The mean length of an advisory was 7.96 days (d), and the standard deviation was 3.57 d. What communities were affected? There was no one clear demographic variable that explained the length of boil water advisories, but longer advisories tended to be issued by smaller public water systems (serving fewer homes). Drinking water access during winter Storm Uri Individual-Level Survey 266 individuals Boil water advisories were issued to 14% of participants based on their locations (for comparison, 18% said they were issued an advisory in the survey); 31% experienced loss of water pressure, 9% noticed changes in color, taste, or smell of their water, 18% experienced burst pipes, and 16% experienced water damage. ANCOVA results: Although White, non-Hispanic individuals were under confirmed boil water advisories significantly more often than Black/Hispanic/Biracial individuals, the difference was driven by the proportion of first-generation college participants across ethnic categories who were more likely to be under boil water advisories. The 4 public water systems that served the greatest populations (all >1 million, as reported to TCEQ by the providers) experienced advisories of intermediate length (within 1 standard deviation of the mean). These urban systems served relatively greater proportions of home renters and were more racially and ethnically diverse. MANCOVA results: White, non-Hispanic individuals also identified that they were under confirmed boil water advisories significantly more often that Black/Hispanic/Biracial individuals, but these differences (~5%) must be considered in light of communication gaps between water utilities and the public. The 15 pubic water systems with the longest advisories (> 20 days) were similar in terms of their small populations and service areas. Most served more rural, White, homeowning families, but 3 of the 15 worst served an above-average proportion of non-White or Hispanic and Latino families. How well were boil water advisories communicated? There was broad public confusion: 37% of the sample was not sure if they were placed under a boil water advisory; 53% of people who said they were issued a boil water advisory (18%) actually were not, based on their locations during the storm; 5% of those who said they were not issued a boil water advisory (45%) actually were, based on their locations. What are some of the limitations of the tool? Not all public water systems may have reported boil water advisories to the TCEQ; Age of census data; Uncertainties of calculating system demographics from aerially weighted census data. Snowball sampling approach, which favored college students and specific regions; Small numbers; Uncertainties in participants’ geographic locations; Delayed survey dissemination. https://doi.org/10.1371/journal.pwat.0000137.t001 Further, most participants did not experience burst pipes in their homes (82%; N = 217) or water damage (84%; N = 224) (Fig D in S1 Text). The detailed omnibus and univariate test sta- tistics, F-statistics, and effect sizes are reported in Tables C and D of S1 Text, so we summarize only high-level findings below. First, using the ANCOVA, we observed main effects of race and ethnicity (p = .01) and first-generation status (p = .04) on boil water advisories that were issued to participants (as determined from combining public water system and participant datasets). Although White individuals (MW = 0.21, SEW = 0.04) were under confirmed boil water advisories significantly more often than Black/Hispanic/Biracial (MBLB = 0.16, SEBLB = 0.05) participants, these effects were driven by the experiences of White first-generation participants (MFG = 0.25, SEFG = 0.05). Participants who identified as White first-generation were significantly more likely to be under boil water advisories than non-first-generation participants (MNFG = 0.14, SENFG = 0.03) (Table C of S1 Text). Income effects were less clear, but first-generation status may be a better indicator of socioeconomic status than income in this survey, given that many participants were college students whose income responses may have been shaped by multidimensional factors. Second, using the MANCOVA, we observed that race and ethnicity had a significant main effect on boil water advisories that were experienced by participants (as determined from par- ticipant responses alone), such that White, non-Hispanic participants were significantly more likely (p = .03) to report being under a boil water advisory (MW = 2.12, SEW = 0.11) than PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 10 / 16 PLOS WATERDrinking water access during winter Storm Uri Black/Hispanic/Biracial participants (MBLB = 2.35, SEBLB = 0.14) (Table C of S1 Text). The dif- ference was approximately 5%. Both the ANCOVA and MANCOVA analyses hold across pub- lic water system characteristics (the number of households that the participants’ water systems served, and the fraction of rural households, which we control for in both analyses). In summary (Table 1), we find that (1) across race and ethnicities, 42% to 49% of partici- pants were incorrect regarding their actual boil water advisory status, which points to a gaping opportunity to improve public health communications during extreme weather events and emergencies. Additionally, (2) when we examined the influence of various socioeconomic fac- tors on these experiences, we found that although White, non-Hispanic participants were more likely to report being under boil water advisories than Black, Hispanic and Latino, and Biracial participants, these results were driven by the experiences of first-generation partici- pants within the White-identifying group (ANCOVA results). Finally, our (3) MANCOVA results replicate the race and ethnicity effect we observed, and again point to communication gaps between drinking water utilities and the public across racial and ethnic groups (Table 1). 4. Discussion Considering all results in a holistic way (Table 1), we found that large, urban public water sys- tems and small, rural ones had different recovery response times to Winter Storm Uri. Fur- thermore, first-generation participants, who may come from more socioeconomically disadvantaged backgrounds, were more likely to be issued boil water advisories across race, ethnicity, and rural or urban communities. Advisories were not communicated effectively, which disadvantaged all racial and ethnic groups. Below, we consider factors behind these trends and implications for water security and future disaster recovery. 4.1. Response times across big, urban and small, rural systems Winter Storm Uri impacted water access for large urban and small rural communities differ- ently, revealing two scales of vulnerability in public water services. A small number of large water providers serve a majority of the Texas population and issued boil water advisories that left mostly urban residents without reliable drinking water for 5–7 days. Meanwhile, a very small number of mostly rural public water systems issued boil water advisories that lasted weeks and had acute effects on small portions of the Texas population. Of the 15 public water systems with the longest boil water advisories, 11 served exclusively rural communities (frac- tion of rural households = 1). All but one served fewer than 1,000 residents (M = 415 and SD = 398). The cross-sectional survey showed similar effects: the total number of households served and the fraction of those households being rural in a participant’s public water system had a significant impact on whether that participant was issued a boil water advisory (Table C of S1 Text). The scale or size of public water systems can influence resilience to severe weather in differ- ent ways. Large (typically urban) providers have more available resources for responding to power loss and infrastructure damage during extreme weather, but large treatment plants also require more power to operate and expertise to troubleshoot or maintain under extreme sce- narios [13, 35]. Large systems also have longer distribution networks or more places where pipes can burst, requiring more time and resources to identify and repair damage. As a result, the largest systems are highly vulnerable to extreme weather events. Importantly, this small number of large public systems impact the greatest share of the population, not only in Texas but all across the United States. Just 8% of the approximately 52,000 community water systems in the United States serve 82% of the population [36]. Therefore, investments in upgrades to large public water systems can yield big returns–particularly for urban communities. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 11 / 16 PLOS WATERDrinking water access during winter Storm Uri In comparison, smaller systems range widely in their resilience to extreme weather events because of uncertainties in the human and financial resources available to them [35]. In the current study, smaller systems (serving less than 1,000 individuals) displayed a wide range in the lengths of their boil water advisories (range of 1–36 days), consistent with Glazer et al. [11], who showed that smaller public systems tended to take longer to recover. Of the 15 sys- tems with the longest recovery times, many were already struggling to meet federal drinking water standards during typical weather conditions (they had multiple violations to the stan- dards before and after Winter Storm Uri). Most (12) of these 15 systems used groundwater as their water source, which often requires little treatment prior to distribution [37], making it likely that these public water systems had little to no infrastructure to oversee, but also few staff to address emergencies, leaving their customers water insecure in the face of extreme weather. In general terms, many rural communities have limited access to resources to make repairs during a winter storm [13]. Some of the rural systems with the longest advisories in this study also served residents in unconventional housing such as mobile homes, consistent with other studies that have noted unreliable water access in mobile home communities [11, 38]. In total, Winter Storm Uri revealed weaknesses in both water policy and management across urban and rural areas. The U.S. Federal Energy Regulatory Commission (FERC) and industry stakeholders had previously identified critical infrastructure to winterize in order to mitigate the effects of future winter storm events, but the recommendations were generally not implemented [11]. Policy reform and funding are thus imperative to incentivize weatheriza- tion and emergency preparedness. Under Texas Senate Bill 3, which passed in response to Winter Storm Uri, public water utilities were required to have an alternate power source for emergencies and to establish Emergency Preparedness Plans. In a follow-up survey of large water utilities conducted one year after the storm, most had either established backup power systems or were taking steps to do so [13]. However, 90% of these relatively well-resourced utilities still cited economics as a limit to further action. With the influence of climate change and urbanization, many large public water systems therefore remain vulnerable to extreme weather, which creates water insecurities for growing populations [39]. 4.2. Water service inequalities Given the large number of public water systems that issued boil water advisories (Fig 3), it is perhaps unsurprising that impacts were felt across racial and ethnic groups in both our sys- tem-level and individual analyses (Table 1), though public water systems are organized around communities that have been shaped by legacies of discrimination (including redlining and gentrification). Glazer et al. [11] also observed no clear relationship between duration of boil water notices and a social vulnerability index at the county level. We note, however, several limitations in our finer-scale analysis that introduced additional uncertainties (Table 1). For example, calculating an aerially weighted average set of demographics for a public water sys- tem assumes that housing density within the service area is uniform. Differences may also be masked by reducing demographic and income statistics to average values for entire communi- ties (for example, two public water systems might have very similar average incomes but very different income distributions). It is important to note that other studies have revealed clear racial and ethnic disparities in water supply and outage factors during Winter Storm Uri. Lee et al. [22] showed that more 311 calls related to burst pipes were placed by low-income and minority groups, and a survey by Grineski et al. [21] showed that Black participants experienced longer water outages. Burst pipes and loss of water pressure are not necessarily distributed evenly throughout individual PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 12 / 16 PLOS WATERDrinking water access during winter Storm Uri public water systems, unlike boil water advisories, allowing for more disparate impacts to neighborhoods based on race, ethnicity, and income demographics. Power outages were not equitably distributed across racial and ethnic identities either [20]. Although power outages had knock-on effects on water providers, they were only one of many factors that led providers to issue advisories [6, 40]. The importance of the first-generation status in our cross-sectional survey data may point to underlying socioeconomic disparities in how boil water advisories were issued. We specifi- cally found that first-generation, White participants were more often under boil water adviso- ries than non-first-generation, White participants. Our method of recruiting participants using snowball sampling via our own networks (e.g., research and conference contacts) and a human subjects research participant pool of college-aged students within a large, public uni- versity in Texas likely influenced the types of individuals within each income bracket and masked income effects. First-generation status may therefore be the best measure of socioeco- nomic status in our survey. Future research could explore income and education effects across even a wider community sample. Lastly, this study underscores other issues with disparity in public water services across racial and ethnic groups, particularly a tendency for small, rural systems in Texas tend to serve predominantly White communities (Fig 2), consistent with studies from elsewhere in the U.S. [19, 41]. For example, a study from North Carolina showed that only 15% of small public water systems served an above-average proportion of non-White or Hispanic and Latino fami- lies (>26.96%) [42]. Yet, in our study, 4 out of the 15 (25%) small systems with the longest advisories served an above-average proportion of non-White families; three served mostly His- panic and Latino families, and one served mostly Black or African American families. Our ANCOVA analysis (Table C of S1 Text), similarly shows that the longest recovery times in rural public water systems in Texas were not consistently concentrated in predominantly White communities. To dismantle inequalities in public water system services and improve resilience for all communities, there should be more investment in vulnerable geographic areas, including low- income non-White communities [20], and steps should be taken to de-centralize and diversify water supply systems. Additionally, identifying urban and rural communities that are under- served by public water systems, and extending those services is important for ensuring equita- ble water access [42]. 4.3. Public awareness gaps This study revealed wholesale communication gaps between water utilities and surveyed par- ticipants. Tiedmann et al. [13] Castellanos et al. [43] also highlighted gaps and inconsistencies in the way utilities communicated with the public. In a post-storm survey of large public water utilities, roughly 80% cited communication issues with the public as an important complicat- ing factor during the storm; yet, one year later, few of these utilities had taken steps to improve communications [13]. A number of strategies have been suggested to improve communica- tions, particularly to younger ages and minoritized groups. For example, public utilities could leverage social media more frequently in their communications [13] and translate messages to languages other than English [43]. Communication gaps can be exacerbated for communities that rely less on traditional forms of media communication or where there are more non- Native English speakers [44]. The large communication gap documented in this study has important public health consequences and reinforces the need for diverse and tailored com- munication strategies to reach diverse customer populations. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 13 / 16 PLOS WATERDrinking water access during winter Storm Uri 5. Conclusions The factors that affected public water supplies in Winter Storm Uri were complex, but the severity of temperatures was not clearly correlated to the length of the advisory. Instead, the size or scale of the public water system and its urban or rural location were most important. Smaller systems faced some of the longest boil water advisories lasting for multiple weeks, while a large portion of the Texas population living in urban areas served by large public water systems was placed under advisories lasting less than a week. Additionally, cross-sectional sur- vey data suggest that first-generation, White individuals were more likely to be issued a boil water advisory than non-first-generation, White individuals. These findings highlight differ- ences in the resilience of public water services to communities of varying size, urban or rural location, and socioeconomic status that affect water security for Texas residents. In the wake of Winter Storm Uri, a clear need exists to help public water service providers prepare for more extreme weather events in a changing climate. Survey data also revealed massive com- munication gaps between public water service providers and customers, indicating a need for new communication strategies. Supporting information S1 Text. Additional information on public water system demographics, survey methods, and results. (PDF) Acknowledgments We thank Michele Risko and Patrick Kading at the Texas Commission on Environmental Quality for their assistance with data acquisition. We also thank two anonymous reviewers for their helpful suggestions. This study has IRB (2021–882) approval. All datasets corresponding to this analysis are freely available at https://doi.org/10.5281/zenodo.7447637. Author Contributions Conceptualization: Christine L. Nittrouer, Audrey H. Sawyer. Data curation: Brianna Tomko, Christine L. Nittrouer, Audrey H. Sawyer. Formal analysis: Brianna Tomko, Christine L. Nittrouer, Audrey H. Sawyer. Investigation: Brianna Tomko, Christine L. Nittrouer. Methodology: Brianna Tomko, Christine L. Nittrouer, Xavier Sanchez-Vila. Project administration: Audrey H. Sawyer. Resources: Christine L. Nittrouer, Audrey H. Sawyer. Supervision: Audrey H. Sawyer. Validation: Brianna Tomko, Christine L. Nittrouer. Visualization: Brianna Tomko, Christine L. Nittrouer, Audrey H. Sawyer. Writing – original draft: Brianna Tomko. Writing – review & editing: Christine L. Nittrouer, Xavier Sanchez-Vila, Audrey H. Sawyer. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 14 / 16 PLOS WATERDrinking water access during winter Storm Uri References 1. Mekonnen MM, Hoekstra AY. Four billion people facing severe water scarcity. 2016. Sci Adv 2: e1500323–e1500323. https://doi.org/10.1126/sciadv.1500323 PMID: 26933676 2. Brown TC, Mahat V, Ramirez JA. Adaptation to future water shortages in the United States caused by population growth and climate change. Earth’s Future. 2019 Mar; 7(3):219–34. 3. Calverley CM, Walther SC. Drought, water management, and social equity: Analyzing Cape Town, South Africa’s water crisis. Frontiers in Water. 2022 Sep 7; 4. 4. United Nations Children’s Fund (UNICEF), World Health Organization (WHO) [Internet]. 2021. Avail- able from: https://data.unicef.org/resources/progress-on-household-drinking-water-sanitation-and- hygiene-2000-2020/ Licence: CC BY-NC-SA 3.0 IGO 5. Doss-Gollin J, Farnham DJ, Lall U, Modi V. How unprecedented was the February 2021 Texas cold snap?. Environmental Research Letters. 2021 Jun 8; 16(6):064056. 6. Busby JW, Baker K, Bazilian MD, Gilbert AQ, Grubert E, Rai V, et al. Cascading risks: Understanding the 2021 winter blackout in Texas. Energy Research & Social Science. 2021; 77:102106. 7. Cardinal C, Ratnapradipa D, Scarbrough A, Robins A, Boes K. Extreme Winter Storms: Environmental Impacts of Public Utility Policies on Vulnerable Populations. Journal of Environmental Health. 2022; 84:12–19. 8. Donald J. Winter Storm Uri 2021. The Economic Impact of the Storm. Texas Comptroller of Public Accounts [Internet]. 2021. Available at: https://comptroller.texas.gov/economy/fiscal-notes/2021/oct/ winter-storm-impact.php. 9. Hobby School of Public Affairs The Winter Storm of 2021 [Internet]. 2021. Available from: https://uh. edu/hobby/winter2021/storm.pdf 10. Hannah DM, Lynch I, Mao F, Miller JD, Young SL, Krause S. Water and sanitation for all in a pandemic. Nature Sustainability. 2020 Oct; 3(10):773–5. 11. Glazer YR, Tremaine DM, Banner JL, Cook M, Mace RE, Nielsen-Gammon J, et al. Winter storm Uri: a test of Texas’ water infrastructure and water resource resilience to extreme winter weather events. Jour- nal of Extreme Events. 2021 Dec 31; 8(04):2150022. 12. National Research Council. Disaster Resilience: A National Imperative. Washington, DC: The National Academies Press; 2012. 13. Tiedmann HR, Spearing LA, Castellanos S, Stephens KK, Sela L, Faust KM. Tracking the Post-Disaster Evolution of Water Infrastructure Resilience: A Study of the 2021 Texas Winter Storm. Sustainable Cit- ies and Society. 2023 Jan 21:104417. 14. Mullin M. The effects of drinking water service fragmentation on drought-related water security. Sci- ence. 2020 Apr 17; 368(6488):274–7. https://doi.org/10.1126/science.aba7353 PMID: 32299948 15. Meehan K, Jurjevich JR, Chun NM, Sherrill J. Geographies of insecure water access and the housing– water nexus in US cities. Proceedings of the National Academy of Sciences. 2020 Nov 17; 117 (46):28700–7. https://doi.org/10.1073/pnas.2007361117 PMID: 33139547 16. Seltenrich N. Unwell: the public health implications of unregulated drinking water. Environmental health perspectives. 2017 Nov 1; 125(11):114001. https://doi.org/10.1289/EHP2470 PMID: 29095690 17. Aiken CS. Race as a factor in municipal underbounding. Annals of the Association of American Geogra- phers. 1987 Dec 1; 77(4):564–79. 18. Patel AI, Schmidt LA. Water access in the United States: health disparities abound and solutions are urgently needed. American journal of public health. 2017 Sep; 107(9):1354–6. https://doi.org/10.2105/ AJPH.2017.303972 PMID: 28787195 19. US Water Alliance. Closing the water access gap in the United States: A national plan [Internet]. 2019. Available from: https://uswateralliance.org/sites/uswateralliance.org/files/publications/Closing%20the %20Water%20Access%20Gap%20in%20the%20United%20States_DIGITAL.pdf 20. Nejat A, Solitare L, Pettitt E, Mohsenian-Rad H. Equitable community resilience: the case of winter storm Uri in Texas. International Journal of Disaster Risk Reduction. 2022 Jul 1; 77:103070. 21. Grineski SE, Collins TW, Chakraborty J, Goodwin E, Aun J, Ramos KD. Social disparities in the duration of power and piped water outages in Texas after Winter Storm Uri. American Journal of Public Health. 2023 Jan; 113(1):30–4. https://doi.org/10.2105/AJPH.2022.307110 PMID: 36356281 22. 23. Lee CC, Maron M, Mostafavi A. Community-scale big data reveals disparate impacts of the Texas win- ter storm of 2021 and its managed power outage. Humanities and Social Sciences Communications. 2022 Sep 24; 9(1):1–2. https://doi.org/10.1057/s41599-022-01353-8 PMID: 36187845 TCEQ. Compliance Notebook for Transient Noncommunity Public Water System [Internet]. Texas Commission on Environmental Quality; 2021 Jul 1. Available from: https://www.tceq.texas.gov/ downloads/assistance/water/pdws/tnc/rg-549.pdf PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 15 / 16 PLOS WATERDrinking water access during winter Storm Uri 24. U.S. EPA. FACTOIDS: Drinking Water and Ground Water Statistics for 2007 [Internet]. U.S. Environ- mental Protection Agency; 2008. Available https://nepis.epa.gov/Exe/ZyPDF.cgi/P100N2VG.PDF? Dockey=P100N2VG.PDF 25. Reedy RC, Scanlon BR. Assessment of Arsenic in Groundwater and Water Supply Systems in Texas [Internet]. Bureau of Economic Geology Jackson School of Geosciences, University of Texas at Austin; 2018. Available from: https://www.beg.utexas.edu/files/content/beg/research/water/TCEQ%20Arsenic %20Report%20Final%206%20Mar%202019.pdf 26. Texas Water Development Board. Water Service Boundary Viewer [Internet]. 2022. Available from: https://www3.twdb.texas.gov/apps/WaterServiceBoundaries 27. McDonald YJ, Anderson KM, Caballero MD, Ding KJ, Fisher DH, Morkel CP, et al. A systematic review of geospatial representation of United States community water systems. AWWA Water Science. 2022 Jan; 4(1):e1266. 28. TCEQ. Public Drinking Water Program 2020 Annual Compliance Report [Internet]. Texas Commission on Environmental Quality; 2021. Available from: https://www.tceq.texas.gov/downloads/drinking-water/ epa-acr-2020.pdf 29. National Centers for Environmental Information (NCEI). Climate Data Online: Dataset Discovery [Inter- net]. 2021. Available from: https://www.ncdc.noaa.gov/cdo-web/datasets 30. Walker KE. A reproducible framework for visualizing demographic distance profiles in US metropolitan areas. Spatial Demography. 2018; 6(3):207–33. https://doi.org/10.1007/s40980-018-0042-7 31. U.S. Government Accountability Office. Decennial Census: Bureau should assess significant data col- lection challenges as it undertakes planning for 2030: Decennial Census: Bureau Should Assess Signifi- cant Data Collection Challenges as It Undertakes Planning for 2030. U.S. Government Accountability Office. 2021. Available from: https://www.gao.gov/products/gao-21-365 32. Tomko B. Public drinking water access in Texas communities during Winter Storm Uri 2021. B.S. The- sis, The Ohio State University. 2022. Available from: http://hdl.handle.net/1811/101414 33. Wang X, Kammerer CM, Anderson S, Lu J, Feingold E. A comparison of principal component analysis and factor analysis strategies for uncovering pleiotropic factors. Genetic Epidemiology: The Official Publication of the International Genetic Epidemiology Society. 2009 May; 33(4):325–31. https://doi.org/ 10.1002/gepi.20384 PMID: 19048641 34. Snell TP. First-generation students, social class, and literacy. Academe. 2008 Jul 1; 94(4):28–31. 35. DeRosa E, Dolby N. “I don’t think the university knows me.”: Institutional culture and lower-income, first- generation college students. InterActions: UCLA Journal of Education and Information Studies. 2014; 10(2). 36. Cutter SL, Ash KD, Emrich CT. Urban–rural differences in disaster resilience. Annals of the American Association of Geographers. 2016; 106(6):1236–52. 37. United Nations. The United Nations World Water Development Report 2022: Groundwater: Making the invisible visible. UNESCO Paris, 225 pp. [Internet]. 38. Pierce G, Jimenez S. Unreliable water access in U.S. mobile homes: Evidence from the American Housing Survey. Housing Policy Debate. 2015; 25(4):739–53. https://doi.org/10.1080/10511482.2014. 999815 39. Donner W, Rodrı´guez H. Disaster risk and vulnerability: The role and impact of population and society [Internet]. 2011. Available from: https://www.prb.org/resources/disaster-risk/ 40. Reed PM, Hadjimichael A, Moss RH, Brelsford C, Burleyson CD, Cohen S, et al. Multisector dynamics: Advancing the science of complex adaptive human-Earth systems. Earth’s Future. 2022 Mar; 10(3): e2021EF002621. 41. Meehan K, Jepson W, Harris LM, Wutich A, Beresford M, Fencl A, et al. Exposing the myths of house- hold water insecurity in the global north: A critical review. Wiley Interdisciplinary Reviews: Water. 2020 Nov; 7(6):e1486. 42. Leker HG, MacDonald Gibson J. Relationship between race and community water and sewer service in North Carolina, USA. PLoS One. 2018 Mar 21; 13(3):e0193225. https://doi.org/10.1371/journal.pone. 0193225 PMID: 29561859 43. Castellanos S, Potts J, Tiedmann H, Alverson S, Glazer YR, Robison A, et al. A synthesis and review of exacerbated inequities from the February 2021 winter storm (Uri) in Texas and the risks moving for- ward. Progress in Energy. 2023 Jan 9; 5(1):012003. 44. Elder JP, Ayala GX, Parra-Medina D, Talavera GA. Health Communication in the Latino Community: Issues and Approaches. Annual Review of Public Health. 2009; 30(1):227–51. https://doi.org/10.1146/ annurev.publhealth.031308.100300 PMID: 19296776 PLOS Water \| https://doi.org/10.1371/journal.pwat.0000137 June 21, 2023 16 / 16 PLOS WATER
10.3390_biom13060949	Article YES1 Kinase Mediates the Membrane Removal of Rescued F508del-CFTR in Airway Cells by Promoting MAPK Pathway Activation via SHC1 Patrícia Barros 1,2, Ana M. Matos 1,2 , Paulo Matos 1,2,,† and Peter Jordan 1,2,† 1 Departamento de Genética Humana, Instituto Nacional de Saúde Doutor Ricardo Jorge, 1649-016 Lisboa, Portugal; [email protected] (P.B.); [email protected] (A.M.M.); [email protected] (P.J.) BioISI—Biosystems & Integrative Sciences Institute, Faculty of Sciences, University of Lisboa, 1749-016 Lisboa, Portugal 2 Correspondence: [email protected] † These authors contributed equally to this work. Abstract: Recent developments in CFTR modulator drugs have had a signiﬁcant transformational effect on the treatment of individuals with Cystic Fibrosis (CF) who carry the most frequent F508del- CFTR mutation in at least one allele. However, the clinical effects of these revolutionary drugs remain limited by their inability to fully restore the plasma membrane (PM) stability of the rescued mutant channels. Here, we shed new light on the molecular mechanisms behind the reduced half-life of rescued F508del-CFTR at the PM of airway cells. We describe that YES1 protein kinase is enriched in F508del-CFTR protein PM complexes, and that its interaction with rescued channels is mediated and dependent on the adaptor protein YAP1. Moreover, we show that interference with this complex, either by depletion of one of these components or inhibiting YES1 activity, is sufﬁcient to signiﬁcantly improve the abundance and stability of modulator-rescued F508del-CFTR at the surface of airway cells. In addition, we found that this effect was mediated by a decreased phosphorylation of the scaffold protein SHC1, a key regulator of MAPK pathway activity. In fact, we showed that depletion of SHC1 or inhibition of MAPK pathway signaling was sufﬁcient to improve rescued F508del-CFTR surface levels, whereas an ectopic increase in pathway activation downstream of SHC1, through the use of a constitutively active H-RAS protein, abrogated the stabilizing effect of YES1 inhibition on rescued F508del-CFTR. Taken together, our ﬁndings not only provide new mechanistic insights into the regulation of modulator-rescued F508del-CFTR membrane stability, but also open exciting new avenues to be further explored in CF research and treatment. Keywords: Cystic Fibrosis; F508del-CFTR; plasma membrane half-life; MAPK pathway 1. Introduction The Cystic Fibrosis transmembrane conductance regulator (CFTR) protein is a member of the ATP-binding cassette transporter superfamily that functions as a cAMP-activated chloride and bicarbonate ion channel at the apical membrane of epithelial cells [1]. Mu- tations in the CFTR gene lead to Cystic Fibrosis (CF), a lethal disease characterized by pancreatic insufﬁciency, increased salt concentration in sweat, male infertility, and progres- sive lung disease [2]. Among the many CFTR pathogenic mutations identiﬁed to date, the deletion of phenylalanine 508 (F508del) is by far the most prevalent, occurring in up to 85% of CF individuals [3]. F508del causes CFTR to misfold, leading to the retention of most of the mutant protein at the endoplasmic reticulum (ER) and its premature degradation by the ER quality control machinery (ERQC) associated with proteasomes [4]. This causes a drastic reduction in the number of mutant channels reaching the plasma membrane (PM). In addition, the mutant protein exhibits a considerable gating defect due to an abnormal Citation: Barros, P.; Matos, A.M.; Matos, P.; Jordan, P. YES1 Kinase Mediates the Membrane Removal of Rescued F508del-CFTR in Airway Cells by Promoting MAPK Pathway Activation via SHC1. Biomolecules 2023, 13, 949. https://doi.org/ 10.3390/biom13060949 Academic Editor: Viswanathan Natarajan Received: 4 May 2023 Revised: 26 May 2023 Accepted: 1 June 2023 Published: 6 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Biomolecules 2023, 13, 949. https://doi.org/10.3390/biom13060949 https://www.mdpi.com/journal/biomolecules biomoleculesBiomolecules 2023, 13, 949 2 of 17 persistence in the closed state [5]. These combined defects result in a residual level of chloride secretion and epithelial ﬂuid transport in the airway, leading to the production of a thick mucus that promotes airway obstruction, chronic bacterial colonization and inﬂammation, severe lung disease, and ultimately, respiratory failure [6]. In recent decades, the recognition that F508del-CFTR molecular defects could be tar- geted by chemical compounds has led to considerable efforts to develop and approve effective drugs, now termed “CFTR modulators”, to treat CF disease [7]. The most success- ful thus far, Kaftrio (trade name in Europe; Trikafta in the U.S.), combines three modulators: two “correctors” to improve F508del-CFTR folding (tezacaftor and elexacaftor (also known as VX-661 and VX-445, respectively)), and another to improve F508del-CFTR gating (iva- caftor (also known as VX-770), termed “potentiator”) [8]. Kaftrio has been considered groundbreaking in the CF clinical practice, with signiﬁcant clinical beneﬁt in patients 12 years and older who have at least one allele with the F508del mutation [9]. However, Kaftrio is still unable to fully restore F508del-CFTR function [3]. Even in vitro, the combination of modulators only rescues F508del-CFTR to approximately 60% of wild-type (wt) CFTR function in non-CF cells [10,11]. A major issue with modulator-rescued F508del-CFTR (rF508del-CFTR) is the considerably reduced half-life of the protein at the cell surface [10,12–14]. Although corrector drugs enable a signiﬁcant amount of mutant protein to evade the ERQC, the rescued channels fail to completely escape the peripheral protein quality control (PPQC) machinery. The PPQC removes defective proteins from the PM, promoting their lysosomal degradation [13]. In the PM, wt-CFTR is internalized slowly via clathrin-mediated endocytosis (CME), but most of the internalized protein is rapidly recycled back to the PM [12,13,15]. In contrast, rF508del-CFTR is internalized by CME at a much faster rate [15–17], which is facilitated by the presence of chaperones and co-chaperones, and most of the internalized protein is sent for degradation in the lysosomal compartment [12,14]. We and others have shown that slower internalization from the PM depends on the anchoring of wt-CFTR to the actin cytoskeleton [14,15,18–20]. This is enabled by the binding of CFTR’s C-terminal postsynaptic density 95, disks large, zonula occludens-1 (PDZ)-binding motif to the second PDZ domain of the Na+/H+ exchanger regulatory factor (NHERF1) [14,20,21]. NHERF1 has two tandem N-terminal PDZ domains. The ﬁrst domain (PDZ1) can readily interact with CFTR during its trafﬁcking from the trans-Golgi to the PM [5]. However, the second PDZ domain (PDZ2) is blocked from binding to CFTR by an intramolecular NHERF1 interaction with its C-terminal Ezrin binding domain (EBD) [21], which is only released upon binding to active Ezrin at the PM [14,22]. Ezrin binding to NHERF1 EBD domain then allows NHERF1 PDZ2 to bind CFTR, resulting in its anchoring to the actin cytoskeleton via Ezrin and prolonged apical PM localization and efﬁcient activation of wt-CFTR [14,18,19,23]. Although rF508del-CFTR can also bind to the NHERF1 PDZ1 domain, it fails to recruit Ezrin and switch to NHERF1(cid:48)s PDZ2, thus lacking anchorage to actin ﬁlaments unless coaxed by active Ezrin overexpression [14,19]. In order to understand the underlying molecular mechanism, we recently used mass spectrometry (MS) to analyze rF508del-CFTR-containing complexes selectively immunoprecipitated from the PM of airway cells [24]. We found these PM-located complexes were enriched in Calpain-1, a calcium-dependent protease that has Ezrin among its substrates. Calpain-1 prevented Ezrin recruitment to rF508del-CFTR but not to wt-CFTR, and its depletion or chemical inhibition signiﬁcantly increased the functional abundance and stability of the rescued mutant channels at the cell surface [24]. However, our MS analysis identiﬁed other candidate proteins as potential interactors with rF508del-CFTR at the PM, suggesting additional mechanisms regulating the channel’s decreased stability at the cell surface. Among these proteins was the Yamaguchi sarcoma viral oncogene homolog 1 (YES1) [24], a nonreceptor tyrosine kinase of the Src family, whose dysregulation has been associated with multiple cancer signaling pathways [25]. Here, we describe that YES1 associates at the PM with rF508del-CFTR, but not with wt-CFTR. This is accomplished through the NHERF1-binding adaptor protein YAP1 (YES-associated protein 1), which phosphorylates Biomolecules 2023, 13, 949 3 of 17 the Src homology 2 domain-containing transforming protein 1 (SHC1). SHC1 is an adaptor protein that serves as the molecular link between CFTR internalization and the activation of the mitogen-activated protein kinase (MAPK) pathway. 2. Materials and Methods 2.1. Cell Culture, Treatment and Transfection CFBE41o- cells were generated by Adv Bioscience LLC and were stably transduced with lentivirus encoding mCherry-Flag-tagged, wt-, or F508del-CFTR under the Tet- ON promoter, as previously described [26]. A clone of CFBE cells stably expressing nontagged F508del-CFTR [27] was selected to stably co-express the halide sensor YFP- F46L/H148Q/I152L (HS-YFP; kindly provided by P. Haggie, University of California, San Francisco, CA, USA, School of Medicine) [14]. All cell lines were cultured at 37 ◦C with 5% CO2 and regularly checked for the absence of mycoplasm infection. They were maintained in minimum essential medium supplemented with GlutaMAX, Earle’s salts, 10% (v/v) heat-inactivated fetal bovine serum (FBS), and 2 µg/mL puromycin (all from Thermo Fisher Scientiﬁc, Waltham, MA, USA). CFBE mCherry-Flag-CFTR cells were additionally selected with 10 µg/mL blasticidin (Invivogen, San Diego, CA, USA), whereas the cells expressing HS-YFP were selected with 400 µg/mL hygromycin B (Thermo Fisher Scientiﬁc). In addition, CFTR construct expression in mCherry-Flag-tagged cells was induced with 1 µg/mL doxycycline (Dox; Sigma-Aldrich, Saint Louis, MO, USA). The cells were reverse transfected as described in [28] with Lipofectamine 2000 (Thermo Fisher Scientiﬁc, Waltham, MA, USA) in 35-mm or 60-mm dishes with 200 pmol or 400 pmol of the indicated siRNAs, respectively, and analyzed at the indicated time points. An siRNA oligonucleotide against luciferase (siCtrl, 5(cid:48)-CGUACGCGGAAUACUUCGA) from Euroﬁns Genomics was used as a mock control, and the siRNAs used to deplete YES1 (sc-29860), YAP1 (sc-38637) or SHC1 (sc-29480) were commercial mixes (each composed of three dif- ferent oligonucleotides) from Santa Cruz Biotechnology (Dallas, TX, USA). For ectopic expression of the constitutively active mutant construct pRK5-Myc-H-RAS-V12 [29], a total amount of 6 µg of DNA plasmid (supplemented with empty vector when required) was transfected, with a ratio of 1:3 (µg/µL) of DNA:Lipofectamine per 60-mm dish. Stock solutions of VX-661 (10 mM, Achemblock, Hayward, CA, USA), SU6656 (10 mM, Sigma-Aldrich), P505-15 (10 mM, Selleckchem, Houston, TX, USA), Selumetinib (10 mM, Santa Cruz Biotechnology, Dallas, TX, USA), Forskolin (Fsk, 10 mM, Sigma-Aldrich, Saint Louis, MO, USA), Genistein (Gen, 50 mM, Sigma-Aldrich), or CFTR-speciﬁc inhibitor 172 (inh172, 10 mM, Santa Cruz Biotechnology, Dallas, TX, USA) were prepared at least 1000-fold in DMSO. This ensured that the DMSO concentration during cell treatment did not exceed 0.1% (v/v). 2.2. Protein Thermal Destabilization Assay (Thermal Shift Assay—TS) CFBE cells expressing either mCherry-Flag-tagged (previously induced with Dox) or untagged F508del-CFTR were incubated for 48 h at 30 ◦C with VX-661 (5 µM, Achemblock, Hayward, CA, USA). These conditions attenuated deﬁcient protein folding and were previously optimized to achieve rF508del-CFTR PM levels close to those of wt-CFTR [14,24]. The cells were next transferred to 37 ◦C for 3 h, which restored misfolding and destabilized the rF508del-CFTR at the PM, as described in [24,30]. Cells were then placed on ice, washed three times with ice-cold PBS-CM (PBS (pH 8.0), containing 0.9 mM CaCl2 and 0.5 mM MgCl2), and left for 5 min in cold PBS-CM. The cells were then analyzed using confocal immunoﬂuorescence, biotinylation of surface proteins, or immunoblotting, as indicated below. 2.3. Immunoblotting and Immunoﬂuorescence The samples were analyzed using immunoblotting as previously described [24,30,31]. The antibodies used for Western blot (WB) were as follows: mouse anti-CFTR clone Biomolecules 2023, 13, 949 4 of 17 596 (obtained through the UNC CFTR antibody distribution program sponsored by Cystic Fibrosis Foundation Therapeutics—CFFT, Bethesda, MD, USA); mouse anti-α-Tubulin clone B-5-1-2 (T5168), mouse anti-Myc clone 9E10 (M5546), mouse anti-phospho-ERK 1/2 (M8159) and rabbit anti-ERK1/2 (M5670) from Sigma-Aldrich; rabbit anti-YES (#65890), and rabbit anti-YAP (#14074) from Cell Signaling; mouse anti-YES (sc-46674), mouse anti- phospho SHC (sc-81519) and mouse anti-SHC (sc-967) from Santa Cruz Biotechnology (Dallas, TX, USA); and rabbit anti-Glut (Ab652) from Abcam. The primary antibodies were detected using secondary, peroxidase-conjugated antibodies (Bio-Rad, Hercules, CA, USA) followed by chemiluminescence detection. For densitometric analysis of WB bands, x-ray ﬁlms from at least three independent experiments were digitalized, and the images were analyzed using ImageJ 1.53q software (NIH, Bethesda, MD, USA). For the immunoﬂuorescence analysis, mCherry-Flag-F508del-CFTR CFBE cells were grown on glass coverslips (10 × 10 mm), transfected, induced with Dox (1 µg/mL), and treated as indicated. Next, the cells were rinsed on ice with cold PBS-CM (PBS (pH 8.0), containing 0.9 mM CaCl2 and 0.5 mM MgCl2) three times and incubated with anti-Flag M2 Ab (F3165, Sigma-Aldrich, Saint Louis, MO, USA) in PBS-CM + 1% (w/v) BSA (Sigma- Aldrich, Saint Louis, MO, USA) for 90 min at 4 ◦C without permeabilization. Afterwards, the cells were washed for 5 min with ice-cold PBS-CM under soft agitation for three times and incubated with anti-mouse Alexa Fluor 488 secondary Ab (Thermo Fisher Scientiﬁc, Waltham, MA, USA) in PBS-CM + 1% (w/v) BSA for 30 min at 4 ◦C. Then, the cells were washed three times with ice-cold PBS-CM and ﬁxed with 4% formaldehyde for 20 min, followed by three washes with PBS + 0.01% Triton X-100. Finally, the cells were stained with 4(cid:48),6-diamidino-2-phenylindole (DAPI) and mounted on microscope slides with Vectashield (Vector Laboratories, Newark, CA, USA). Images were acquired on a Leica TCS-SPE confocal microscope. 2.4. Biotinylation of Cell Surface Proteins and Internalization Protocol CFBE cells expressing untagged F508del-CFTR were treated and incubated at the indicated conditions. Afterwards, the cells were placed on ice, rinsed three times with ice-cold PBS-CM (PBS (pH 8.0), containing 0.9 mM CaCl2 and 0.5 mM MgCl2), and left for 5 min in ice-cold PBS-CM to ensure the arrest of endocytic trafﬁc. To label cell surface proteins, the cells were incubated for 30 min with 0.5 mg/mL of EZ-Link Sulfo-NHS-SS- Biotin (Santa Cruz Biotechnology, Dallas, TX, USA) in PBS-CM. Subsequently, the cells were washed once with PBS-CM and once with ice-cold Tris-Q (100 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.9 mM CaCl2, 0.5 mM MgCl2, 10 mM glycine, 1% (w/v) BSA), then left for 15 min on ice with Tris-Q to quench the reaction. Then, the cells were washed three times with cold PBS-CM and underwent one of the two following steps: (1) Cell surface protein analysis: The cells were lysed in a pull-down (PD) buffer (50 mM Tris-HCl (pH 7.5), 100 mM NaCl, 10% (v/v) glycerol, 1% (v/v) Nonidet P-40, 0.1% (v/v) SDS, protease inhibitor cocktail (Sigma-Aldrich, Saint Louis, MO, USA), phosphatase inhibitor cocktail C (Santa Cruz Biotechnology, Dallas, TX, USA)). The cell lysates were cleared at 10,000× g for 5 min at 4 ◦C, and an aliquot was removed while the remaining supernatant was added to streptavidin-agarose beads (Sigma-Aldrich, Saint Louis, MO, USA) that were previously blocked for 1 h with 2% (w/v) milk in PD-buffer to represent the lysate input. The lysate and beads were rotated for 1 h at 4 ◦C, centrifuged for 1 min at 3000× g, and washed ﬁve times with an excess of wash buffer (100 mM Tris-HCl (pH 7.5), 300 mM NaCl, 1% (v/v) Triton X-100). The captured proteins were eluted in 2× Laemmli buffer with 100 mM of DTT and analyzed using immunoblotting, as described above. Internalization studies: Biotin-labeled cells were incubated for 3 h at 37 ◦C with warm media containing the indicated treatments. Next, the cells were replaced on ice, rinsed with ice-cold PBS-CM, and left for 5 min to block endocytosis. After two ice-cold PBS-CM washes, the cells were incubated for 30 min (2 × 15 min) with ice-cold stripping buffer (60 mM glutathione, 90 mM NaCl, 0.9 mM CaCl2, 0.5 mM MgCl2, (2) Biomolecules 2023, 13, 949 5 of 17 90 mM NaOH, 10% (v/v) FBS). The cells were then washed three times with ice-cold PBS-CM and processed as in step 1. 2.5. Co-Immunoprecipitation of Membrane CFTR-Associated Complexes mCherry-Flag-wt and mCherry-Flag-F508del-CFTR CFBE cells were used to co-immun- oprecipitate membrane CFTR-associated complexes, as described in [24]. Brieﬂy, after Dox-induced CFTR expression (1 µg/mL, Sigma-Aldrich, Saint Louis, MO, USA), wt-CFTR cells were incubated for 48 h at 37 ◦C, whereas F508del-CFTR were incubated for 48 h at 30 ◦C with VX-661-mediated pharmacological correction (5 µM, Achemblock, Hayward, CA, USA). The cells were then placed on ice and incubated under mild agitation for 2 h with anti-Flag M2 Ab (F3165, Sigma-Aldrich, Saint Louis, MO, USA) or a non-speciﬁc IgG (anti-HA, H6908, Sigma-Aldrich, Saint Louis, MO, USA) in PBS-CM (PBS (pH 8.0), containing 0.9 mM CaCl2 and 0.5 mM MgCl2), washed thrice with PBS-CM, and lysed with lysis buffer (50 mM Tris-HCl (pH 7.5), 2 mm MgCl2, 100 mm NaCl, 10% (v/v) glycerol, 1% (v/v) Nonidet P-40, 0.01% (v/v) SDS, protease inhibitor cocktail (Sigma-Aldrich, Saint Louis, MO, USA)). The cell lysates were cleared using centrifugation, an input control aliquot collected and the remaining supernatant was cleared using streptavidin–agarose beads (Sigma-Aldrich, Saint Louis, MO, USA). The protein complexes were captured with protein G magnetic beads (Thermo Fisher Scientiﬁc, Waltham, MA, USA), washed ﬁve times with wash buffer (50 mM Tris-HCl (pH 7.5), 2 mm MgCl2, 200 mm NaCl, 1% (v/v) Nonidet P-40, protease inhibitor cocktail (Sigma-Aldrich, Saint Louis, MO, USA)), eluted in 2× Laemmli buffer with 100 mM DTT, and analyzed through immunoblotting, as described above, with the indicated antibodies. 2.6. Co-Immunoprecipitation of Membrane Coaxed rF508del-CFTR Untagged F508del-CFTR-expressing CFBE cells were reverse transfected as indicated and incubated for 48 h at 30 ◦C with VX-661 (5 µM, Achemblock, Hayward, CA, USA). The cells were then placed on ice, washed three times with ice-cold PBS-CM (PBS (pH 8.0), containing 0.9 mM CaCl2 and 0.5 mM MgCl2), and lysed with lysis buffer (50 mM Tris-HCl (pH 7.5), 2 mm MgCl2, 100 mm NaCl, 1.5% (v/v) Nonidet P-40, protease inhibitor cocktail (Sigma-Aldrich, Saint Louis, MO, USA)). The cell lysates were cleared using centrifugation at 10,000× g for 5 min at 4 ◦C, an input control aliquot corresponding to 1/10 of the volume from each condition was removed, and the remaining supernatant was cleared using streptavidin-agarose beads (Sigma-Aldrich) for 1 h with agitation at 4 ◦C. Next, the cleared lysates were incubated overnight with 500 ng of control Ab (anti-HA, H6908, Sigma-Aldrich, Saint Louis, MO, USA) or anti-YES (#65890, Cell Signaling) at 4 ◦C with agitation. Then, protein-antibody complexes were incubated with protein G magnetic beads (Thermo Fisher Scientiﬁc, Waltham, MA, USA) for 1 h at 4 ◦C with agitation, and after ﬁve washes with wash buffer (50 mM Tris-HCl (pH 7.5), 2 mm MgCl2, 200 mm NaCl, 1% (v/v) Nonidet P-40, protease inhibitor cocktail (Sigma-Aldrich, Saint Louis, MO, USA)), the captured proteins were eluted in 2× Laemmli buffer with 100 mM DTT and analyzed using immunoblotting, as described above, with the indicated antibodies. 2.7. CFTR Functional Assay by Halide-Sensitive YFP (HS-YFP) CFTR activity was determined using the above-mentioned F508del-CFTR and HS- YFP-expressing CFBE cells, as described in [14,30]. Brieﬂy, the cells were treated and/or transfected as indicated, then washed with PBS and incubated for 30 min in PBS-containing compounds for CFTR stimulation/inhibition (5 µM Fsk, 50 µM Gen, or 25 µM inh172). The HS-YFP ﬂuorescence decay in cells was then analyzed by recording ﬂuorescence continuously (500 ms/point) for 10 s (baseline) and then for 40 s after the rapid (<1 s) apical addition of isomolar PBS in which 137 mM Cl− was replaced by I− (PBSI, ﬁnal NaI concentration in the well: 100 mM). After background subtraction, the cell ﬂuorescence recordings were normalized for the initial average value measured before the addition of I−. The initial rate of ﬂuorescence decay (QR), an indicator of the rate of halide transport Biomolecules 2023, 13, 949 6 of 17 by CFTR [27], was derived by ﬁtting the curves to an exponential decay function using GraphPad 5.0. 2.8. Statistical Analysis Quantitative results are shown as means ± SEM of at least three replicate observations. We used Student’s t tests to compare paired sets of data and one-way ANOVAs followed by Tukey’s posttests for multiple data sets. Differences were considered signiﬁcant when p < 0.05. 3. Results 3.1. Interaction with YES1 Decreases rF508del-CFTR Functional Permanence at the PM In our previous work, we determined that tyrosine kinase YES1 was part of the protein complexes interacting with corrector-rescued F508del-CFTR at the PM of bronchial epithe- lial airway cells using MS [24]. For this study, we used CFBE41o- cells stably expressing either mCherry-wt-CFTR or mCherry-F508del-CFTR with an extracellular Flag-tag, allow- ing us to selectively co-immunoprecipitate CFTR-containing protein complexes from the PM of intact cells. As shown in Figure 1A, YES1 was detected in PM-derived co-precipitates from VX-661-rescued (5 µM for 48 h) F508del-CFTR (rF508del-CFTR). However, it was absent from equivalent precipitates captured from the PM of mCherry-wt-CFTR cells. For functional validation, we ﬁrst used the Flag-tag to immunostain the CFTR at the PM of intact cells after depleting the endogenous YES1 levels by 80% using validated commercial siRNAs (Figure 1B). We observed a clear increase in the abundance of rF508del protein at the cell surface without any noticeable change in the overall CFTR protein abundance, as assessed using the intracellular mCherry tag (Figure 1C). Furthermore, a similar increase in cell surface rF508del protein was also apparent following a 3 h treatment with 10 µM SU6656, a chemical YES1 kinase inhibitor (Figure 1C). Finally, we tested for changes in CFTR-mediated ion transport using CFBE cells co-expressing untagged F508del-CFTR and the halide sensor YFP-F46L/H148Q/I152L (HS-YFP) [24]. We observed an over 2-fold increase in the rate of CFTR-mediated ion transport when YES1 levels were either de- pleted through RNA interference, or when its kinase activity was inhibited with SU6656 in VX-661-treated cells, consistent with the observed increased immunostaining signal of rF508del-CFTR at the cell surface (Figure 1D,E). 3.2. Inhibition of YES1 Impairs Thermal Destabilization and Internalization of PM-Bound rF508del- CFTR We next investigated whether the increase in the PM abundance of rF508del-CFTR upon YES1 downregulation reﬂected increased retention of the protein at the airway cell’s surface. For this, we employed the thermal shift (TS) assays described in [24,30], where rF508del-CFTR at the cell surface was ﬁrst coaxed to accumulate through thermal stabilization of its protein folding by incubating the cells at 30 ◦C. Subsequently, the cells were placed at 37 ◦C for 3 h to induce rF508del-CFTR thermal destabilization, leading to its internalization [24,30]. These conditions allowed us to assess the effect of YES inhibitors on the channel’s stability at the cell surface in comparison with the mock treatment. In two complementary assays (see Figure 2A), the cell surface rF508del-CFTR levels were assessed using protein biotinylation. While one assay detected rF508del-CFTR remaining at the PM after TS, another isolated the biotinylated rF508del-CFTR protein that had been internalized upon TS both in the absence and presence of YES1 inhibitors SU6656 (10 µM) or P505-15 (1 µM) (Figure 2B,C). We observed that YES1 inhibition signiﬁcantly prevented rF508del-CFTR internalization upon thermal destabilization, allowing most of the rescued protein to remain at the PM (Figure 2B, quantiﬁed in Figure 2C). Biomolecules 2023, 13, 949 7 of 17 Figure 1. Interaction with YES1 decreases the PM abundance and function of rF508del-CFTR in airway cells. (A) Intact CFBE cells expressing extracellularly Flag-tagged mCherry-wt-CFTR (Wt) or mCherry-F508del-CFTR (r∆F, after rescue to the PM by 48 h treatment with 5 µM VX-661), were labelled with either an anti-Flag antibody or a non-speciﬁc IgG prior to non-denaturing lysis, and antibody-bound CFTR complexes were precipitated using protein G-coupled magnetic beads. Input protein levels were adjusted so that equivalent amounts of wt- and rF508del-CFTR were precipitated. Both input lysates and co-precipitates were analyzed using WB to assess the levels of CFTR and YES1. Tubulin (known to not interact with CFTR at the PM [24]) was used as an additional co-precipitation control. (B) Efﬁciency of siRNA-mediated depletion in CFBE cells expressing mCherry-F508del-CFTR, rescued as in (A), transfected with either a mock siRNA (siCtrl) or a commercial triple oligonucleotide mix against YES1 (siYES1). Representative WBs of CFTR, YES1, and tubulin (used as loading control) are shown (left panels), as well as the quantiﬁcations as means ± SEM from four independent experiments (right panel). (C) Immunoﬂuorescence images of CFBE cells expressing Flag-tagged mCherry-F508del-CFTR rescued to the PM by 48 h treatment with 5 µM VX-661. The cells were transfected as in (B) and treated for 3 h with either vehicle (DMSO) or 10 µM of YES1 inhibitor SU6656. rF508del-CFTR at the surface of intact cells was immunolabelled on ice using an anti-Flag antibody followed by an Alexa 488-conjugated secondary antibody. Cells were then ﬁxed, and the nuclei were stained with DAPI. Confocal images of the cells’ surface, showing surface CFTR staining in green (left panels) and a merged overlay image of surface (green) and total CFTR signals (mCherry, red) along with the stained nuclei (DAPI, blue) are shown in the right panels. White scale bars represent 10 µm. (D) Representative traces of ion transport activity measured through iodide-induced HS-YFP sensor ﬂuorescence decay of untagged F508del-CFTR CFBE cells treated with 5 µM VX-661 for 48 h. The cells were transfected and treated as in (C) and co-treated with or without 25 µM of inh172 15 min prior to stimulation for 30 min in PBS with Fsk (5 µM) and Gen (10 µM), in the presence or absence of inh172. This was followed by continuous ﬂuorescence recording and the addition of I- (represented by the black arrow, ﬁnal concentration 100 mM). (E) Quantiﬁcation of HS-YFP ﬂuorescence quenching rates (QR) of at least ﬁve independent assays for each condition, calculated by ﬁtting the iodide assay results to exponential decay curves. The means ± SEM are shown. ns—not signiﬁcant, p < 0.01, and * p < 0.001 between conditions indicated by the horizontal lines. Biomolecules 2023, 13, x FOR PEER REVIEW 7 of 18 Figure 1. Interaction with YES1 decreases the PM abundance and function of rF508del-CFTR in air-way cells. (A) Intact CFBE cells expressing extracellularly Flag-tagged mCherry-wt-CFTR (Wt) or mCherry-F508del-CFTR (rΔF, after rescue to the PM by 48 h treatment with 5 µM VX-661), were labelled with either an anti-Flag antibody or a non-speciﬁc IgG prior to non-denaturing lysis, and antibody-bound CFTR complexes were precipitated using protein G-coupled magnetic beads. Input protein levels were adjusted so that equivalent amounts of wt- and rF508del-CFTR were precipi-tated. Both input lysates and co-precipitates were analyzed using WB to assess the levels of CFTR and YES1. Tubulin (known to not interact with CFTR at the PM [24]) was used as an additional co-precipitation control. (B) Eﬃciency of siRNA-mediated depletion in CFBE cells expressing mCherry-F508del-CFTR, rescued as in (A), transfected with either a mock siRNA (siCtrl) or a com-mercial triple oligonucleotide mix against YES1 (siYES1). Representative WBs of CFTR, YES1, and tubulin (used as loading control) are shown (left panels), as well as the quantiﬁcations as means ± SEM from four independent experiments (right panel). (C) Immunoﬂuorescence images of CFBE cells expressing Flag-tagged mCherry-F508del-CFTR rescued to the PM by 48 h treatment with 5 µM VX-661. The cells were transfected as in (B) and treated for 3 h with either vehicle (DMSO) or 10 µM of YES1 inhibitor SU6656. rF508del-CFTR at the surface of intact cells was immunolabelled on ice using an anti-Flag antibody followed by an Alexa 488-conjugated secondary antibody. Cells were then ﬁxed, and the nuclei were stained with DAPI. Confocal images of the cells’ surface, show-ing surface CFTR staining in green (left panels) and a merged overlay image of surface (green) and total CFTR signals (mCherry, red) along with the stained nuclei (DAPI, blue) are shown in the right panels. White scale bars represent 10 µm. (D) Representative traces of ion transport activity meas-ured through iodide-induced HS-YFP sensor ﬂuorescence decay of untagged F508del-CFTR CFBE cells treated with 5 µM VX-661 for 48 h. The cells were transfected and treated as in (C) and co-treated with or without 25 µM of inh172 15 min prior to stimulation for 30 min in PBS with Fsk (5 µM) and Gen (10 µM), in the presence or absence of inh172. This was followed by continuous ﬂuo-rescence recording and the addition of I- (represented by the black arrow, ﬁnal concentration 100 mM). (E) Quantiﬁcation of HS-YFP ﬂuorescence quenching rates (QR) of at least ﬁve independent assays for each condition, calculated by ﬁtting the iodide assay results to exponential decay curves. The means ± SEM are shown. ns—not signiﬁcant, p < 0.01, and * p < 0.001 between conditions indicated by the horizontal lines. 3.2. Inhibition of YES1 Impairs Thermal Destabilization and Internalization of PM-Bound rF508del-CFTR Biomolecules 2023, 13, 949 8 of 17 Figure 2. YES1 inhibition increases rF508del-CFTR retention at the PM upon thermal destabilization. (A) Diagram depicting the parallel cell surface protein biotinylation assays used to assess rF508del- CFTR thermal stability and internalization. Replicate dishes of CFBE cells expressing F508del-CFTR were incubated for 48 h at 30 ◦C in the presence of 5 µM of VX-661. One of the replicates was directly placed on ice, then the surface proteins were labeled with sulfo-NHS-SS-biotin, lysed, and the surface-labeled proteins were captured using streptavidin beads. These precipitates represented the input amount of CFTR at the PM without any thermal destabilization (DMSO 30 ◦C). A second set of replicates were moved to 37 ◦C in the absence (DMSO) or presence of YES1 inhibitors (10 µM SU6656 or 1 µM P505-15). Three hours later, surface proteins were labelled and isolated, as described above. This second set revealed the amount of CFTR remaining at the PM after thermal destabilization (TS) and inhibitor treatment. A third set of replicates was ﬁrst labeled with biotin, then placed at 37 ◦C for 3 h in the presence or absence of YES inhibitors. These samples were then returned to ice, and the labeled CFTR remaining at the cell surface was stripped from biotin with glutathione prior to lysis and isolation of the labeled proteins that entered the cells. This third set represents the amount of CFTR internalized from the surface upon TS and inhibitor treatment. (B) Analysis of input lysates and biotin-labelled fractions obtained as described in (A). WBs representative of ﬁve independent assays, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut- 1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. (C) Quantiﬁcation of CFTR abundance in the biotinylated fraction (mean ± SEM) in (B) after normalization to Glut-1 levels and to the respective controls. * p < 0.001 relative to DMSO (30 ◦C), ## p < 0.01 and ### p < 0.001, both relative to DMSO in the internalized set. 3.3. The Interaction between YES1 and rF508del-CFTR Is Mediated by YAP1 Although we previously identiﬁed YES1 as a constituent of the rF508del-CFTR pro- tein complex at the PM [24], when applying the STRING protein–protein interaction (PPI) network algorithm (https://string-db.org, accessed on 3 May 2021), it became clear that the strength of the annotated evidence for a direct interaction between YES1 and the NHERF1/Ezrin/CFTR complex was weaker than that of other complex components (Figure 3A). However, using the STRING algorithm network expansion function, we iden- Biomolecules 2023, 13, x FOR PEER REVIEW 8 of 18 We next investigated whether the increase in the PM abundance of rF508del-CFTR upon YES1 downregulation reﬂected increased retention of the protein at the airway cell’s surface. For this, we employed the thermal shift (TS) assays described in [24,30], where rF508del-CFTR at the cell surface was ﬁrst coaxed to accumulate through thermal stabili-zation of its protein folding by incubating the cells at 30 °C. Subsequently, the cells were placed at 37 °C for 3 h to induce rF508del-CFTR thermal destabilization, leading to its internalization [24,30]. These conditions allowed us to assess the eﬀect of YES inhibitors on the channel’s stability at the cell surface in comparison with the mock treatment. In two complementary assays (see Figure 2A), the cell surface rF508del-CFTR levels were assessed using protein biotinylation. While one assay detected rF508del-CFTR remaining at the PM after TS, another isolated the biotinylated rF508del-CFTR protein that had been internalized upon TS both in the absence and presence of YES1 inhibitors SU6656 (10 µM) or P505-15 (1 µM) (Figure 2B,C). We observed that YES1 inhibition signiﬁcantly prevented rF508del-CFTR internalization upon thermal destabilization, allowing most of the rescued protein to remain at the PM (Figure 2B, quantiﬁed in Figure 2C). Figure 2. YES1 inhibition increases rF508del-CFTR retention at the PM upon thermal destabilization. (A) Diagram depicting the parallel cell surface protein biotinylation assays used to assess rF508del-CFTR thermal stability and internalization. Replicate dishes of CFBE cells expressing F508del-CFTR were incubated for 48 h at 30 °C in the presence of 5 µM of VX-661. One of the replicates was directly placed on ice, then the surface proteins were labeled with sulfo-NHS-SS-biotin, lysed, and the sur-face-labeled proteins were captured using streptavidin beads. These precipitates represented the input amount of CFTR at the PM without any thermal destabilization (DMSO 30 °C). A second set of replicates were moved to 37 °C in the absence (DMSO) or presence of YES1 inhibitors (10 µM SU6656 or 1 µM P505-15). Three hours later, surface proteins were labelled and isolated, as described above. This second set revealed the amount of CFTR remaining at the PM after thermal destabiliza-tion (TS) and inhibitor treatment. A third set of replicates was ﬁrst labeled with biotin, then placed at 37 °C for 3 h in the presence or absence of YES inhibitors. These samples were then returned to ice, and the labeled CFTR remaining at the cell surface was stripped from biotin with glutathione prior to lysis and isolation of the labeled proteins that entered the cells. This third set represents the amount of CFTR internalized from the surface upon TS and inhibitor treatment. (B) Analysis of Biomolecules 2023, 13, 949 9 of 17 tiﬁed YES-associated protein 1 (YAP1), as a strong interactor of YES1 with documented evidence of association with NHERF1 (Figure 3B). YAP1 is a transcription co-regulator in the Hippo signaling pathway, which is involved in cell proliferation, apoptosis, and various stress responses [32]. However, YAP1 was ﬁrst identiﬁed as an adaptor protein that associates with the SH3 domain of YES1 kinase, modulating its activity and PM localization through a direct PDZ-mediated interaction with NHERF1 [33,34]. We therefore probed CFTR-containing complexes immunoprecipitated from the PM of mCherry-Flag-rF508del- and mCherry-Flag-wt-CFTR-expressing CFBE cells with anti-YAP1 antibodies. As observed for YES1 in Figure 1A, we determined that YAP1 co-precipitated only with rF508del-CFTR (Figure 3C). In order to understand the role of YAP1 in this complex, we tested its require- ment for the YES1/F508del-CFTR protein complex. For this, we took CFBE cells stably expressing untagged F508del-CFTR [31] and coaxed most of the channel to the PM through treatment with VX-661 for 48 h at 30 ◦C. Then, we immunoprecipitated YES1 and were able to conﬁrm the presence of untagged rF508del-CFTR in the co-precipitate (Figure 3D). This assay was then repeated using cells previously depleted of endogenous YAP1 expression. Depletion of YAP1 in these cells reached ~70% (p < 0.01) and was sufﬁcient to prevent rF508del-CFTR from co-precipitating with YES1 (Figure 3D), indicating that YAP1 mediated the interaction. Further supporting this observation, the siRNA-mediated downregulation of YAP1 signiﬁcantly increased the thermal stability of rF508del-CFTR at the PM to levels similar to those achieved through YES1 depletion (Figure 3E, quantiﬁed in Figure 3F). Figure 3. Cont. Biomolecules 2023, 13, x FOR PEER REVIEW 10 of 18 Figure 3. YAP1 is required for the binding of YES1 to rF508del-CFTR complexes at the PM. (A) STRING-based analysis (https://string-db.org/, accessed on 19 October 2021) of the strength of an-notated evidence on the interaction between YES1 and the CFTR/NHERF1 (SLC9A3R1)/Ezrin (EZR) membrane anchoring complex (the thickness of the grey lines is proportional to the degree of con-ﬁdence for the interaction between the two proteins they connect, extrapolated from text mining, experimental, and database-collected evidence). (B) A STRING-generated expanded interaction net-work, extended (green nodes) around the core complex in (A) (red nodes). (C) CFTR-containing complexes were immunoprecipitated from the PM of intact CFBE cells expressing extracellularly Biomolecules 2023, 13, 949 10 of 17 Figure 3. YAP1 is required for the binding of YES1 to rF508del-CFTR complexes at the PM. (A) STRING- based analysis (https://string-db.org/, accessed on 19 October 2021) of the strength of annotated evidence on the interaction between YES1 and the CFTR/NHERF1 (SLC9A3R1)/Ezrin (EZR) mem- brane anchoring complex (the thickness of the grey lines is proportional to the degree of conﬁdence for the interaction between the two proteins they connect, extrapolated from text mining, experi- mental, and database-collected evidence). (B) A STRING-generated expanded interaction network, extended (green nodes) around the core complex in (A) (red nodes). (C) CFTR-containing complexes were immunoprecipitated from the PM of intact CFBE cells expressing extracellularly Flag-tagged mCherry-wt-CFTR (Wt) or mCherry-F508del-CFTR, as described for Figure 1A. Both input lysates and co-precipitates were analyzed using WB to assess the levels of CFTR and YAP1. Tubulin was used as an additional co-precipitation control. (D) CFBE cells expressing untagged F508del-CFTR transfected with either mock siRNA (siCtrl) or a commercial siRNA mix targeting YAP1 (siYAP1), were incubated with 5 µM of VX-661 for 48 h at 30 ◦C to coax most of the mutant channel to the PM. The cells were then lysed in non-denaturing conditions and YES1 immunoprecipitated with a speciﬁc antibody (a non-speciﬁc IgG was used as a control) from whole cell lysates (WCL). Both input lysates and co-precipitates were analyzed using WB to assess the levels of precipitated YES1 and co-precipitated rF508del-CFTR. (E) Thermal shift (TS) assay, as described in Figure 2, to assess the thermal stability of untagged rF508del-CFTR in CFBE cells transfected with mock siRNA (siCtrl) or one of two commercial siRNA mixes targeting either YAP1 (siYAP1) or YES1 (siYES1). WBs repre- sentative of at least four independent assays, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut-1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. (F) Quantiﬁcation of CFTR abundance in the biotinylated fraction (mean ± SEM) in (E) after normalization to Glut-1 levels and to siCtrl (30 ◦C). * p < 0.001 relative to siCtrl (30 ◦C), ## p < 0.01 and ### p < 0.001, both relative to siCtrl (TS). 3.4. YES1 Participates in the Removal of rF508del-CFTR from the PM via the MEK/ERK1/2 MAPK Pathway So far, our data suggested that a YES1/YAP1 interaction with F508del-CFTR promotes its internalization from the PM. It was previously described that the activation of the Biomolecules 2023, 13, x FOR PEER REVIEW 10 of 18 Figure 3. YAP1 is required for the binding of YES1 to rF508del-CFTR complexes at the PM. (A) STRING-based analysis (https://string-db.org/, accessed on 19 October 2021) of the strength of an-notated evidence on the interaction between YES1 and the CFTR/NHERF1 (SLC9A3R1)/Ezrin (EZR) membrane anchoring complex (the thickness of the grey lines is proportional to the degree of con-ﬁdence for the interaction between the two proteins they connect, extrapolated from text mining, experimental, and database-collected evidence). (B) A STRING-generated expanded interaction net-work, extended (green nodes) around the core complex in (A) (red nodes). (C) CFTR-containing complexes were immunoprecipitated from the PM of intact CFBE cells expressing extracellularly Biomolecules 2023, 13, 949 11 of 17 mitogen-activated protein kinase (MEK)/extracellular signal-regulated kinase 1/2 (ERK 1/2) mitogen-activated protein kinase (MAPK) pathway, known to regulate cell prolifera- tion and survival, also plays a key role in triggering the internalization of wt-CFTR from the PM of human airway cells [35]. To investigate whether the same mechanism could be involved in the effect of the YES1/YAP1 complex, we proceeded to block MAPK signaling in CFBE cells using the MEK-selective inhibitor selumetinib (10 µM for 3 h) and determined the effect on the thermal destabilization of rF508del-CFTR at the PM. Selumetinib treatment produced a signiﬁcant, over 80% (p < 0.001) decrease in MAPK activity, as measured by the phosphorylation levels of ERK1/2 at Thr202 and Tyr204 (Figure 4A), the downstream effector kinase in this pathway [36]. Importantly, we observed that the inhibition of MAPK signaling produced a signiﬁcant retention of rF508del-CFTR at the PM upon thermal desta- bilization (Figure 4A, quantiﬁed in Figure 4B), comparable to the extent induced by YES1 inhibition or YAP1 depletion (see Figure 3E,F). To determine if the two events were connected, we inhibited YES1 activity with SU6656, as before, in cells expressing a constitutively active H-RAS mutant (H-RAS-V12) that was able to fully stimulate MAPK signaling, bypassing the need for receptor activation at the PM [37]. This experiment showed that while YES1 inhibition in mock transfected cells led to a signiﬁcant retention of rF508del-CFTR at the PM after thermal destabiliza- tion, the effect was abrogated in cells expressing H-RAS-V12 (Figure 4C, quantiﬁed in Figure 4D). Noteworthy, SU6656 treatment was sufﬁcient to reduce ERK1/2 phosphory- lation by over 70% (p < 0.001) in mock transfected CFBE cells, but not in cells expressing H-RAS-V12 (Figure 4C). These data indicate that YES1 interacts with the MAPK pathway activity upstream of RAS, and that the effect of YES1 on rF508del-CFTR PM retention is mechanistically connected with MAPK pathway activity. Figure 4. Cont. Biomolecules 2023, 13, x FOR PEER REVIEW 12 of 18 Figure 4. The MAPK pathway participates in YES1-mediated internalization of rF508del-CFTR at the PM. The thermal shift (TS) assay, as described in Figure 2, was used to assess the thermal stability of untagged rF508del-CFTR in CFBE cells. (A) The cells were treated for 3 h with either vehicle (DMSO) or 10 µM of selumetinib, or (C) transfected with empty vector or Myc-H-RAS-V12 (HRAS) and then treated for 3 h with vehicle (DMSO) or 10 µM of SU6656, as indicated. WBs representative of input and cell surface fractions from four independent assays, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut-1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. H-RAS V12 was detected using an anti-Myc antibody, and an anti-phosphorylated ERK1/2 antibody was used to monitor MAPK Biomolecules 2023, 13, 949 12 of 17 Figure 4. The MAPK pathway participates in YES1-mediated internalization of rF508del-CFTR at the PM. The thermal shift (TS) assay, as described in Figure 2, was used to assess the thermal stability of untagged rF508del-CFTR in CFBE cells. (A) The cells were treated for 3 h with either vehicle (DMSO) or 10 µM of selumetinib, or (C) transfected with empty vector or Myc-H-RAS-V12 (HRAS) and then treated for 3 h with vehicle (DMSO) or 10 µM of SU6656, as indicated. WBs representative of input and cell surface fractions from four independent assays, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut-1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. H-RAS V12 was detected using an anti-Myc antibody, and an anti-phosphorylated ERK1/2 antibody was used to monitor MAPK path- way activity, which was quantiﬁed and shown below the respective blot lanes. (B,D) Corresponding quantiﬁcation of CFTR abundance in the biotinylated fractions (mean ± SEM) from four independent assays after normalization to Glut-1 levels and DMSO (30 ◦C). * p < 0.05, p < 0.01, and * p < 0.001, relative to DMSO (30 ◦C) in (B), and as indicated by the horizontal lines in (C); ## p < 0.01, relative to DMSO (TS) in (B). 3.5. Phosphorylation of SHC1 by YES1 Links rF508del-CFTR Internalization to MAPK Pathway Activation In order to identify a direct link between YES1 activity and MAPK activation, we inves- tigated SHC-transforming protein 1 (SHC1), which has been reported to be phosphorylated by YES1 at Tyr239 and Tyr240 [38,39]. Phosphorylated SHC1 acts as an adaptor protein that improves the signal transduction between stimulated PM receptors and RAS proteins, leading to the stimulation of the MAPK cascade [38–40]. While we could not reliably detect SHC1 co-precipitating with YES1 in our experimental settings (possibly reﬂecting a tran- sient enzyme–substrate interaction), we could nevertheless detect a signiﬁcant (p < 0.01) decrease in SHC1 Tyr239/240 phosphorylation upon inhibition of YES1 activity after 3 h treatment with either SU6656 (10 µM) or P505-15 (1 µM) (Figure 5A). Moreover, an over 70% depletion of endogenous SHC1 levels (p < 0.01) in CFBE cells led to a signiﬁcant (p < 0.01) Biomolecules 2023, 13, x FOR PEER REVIEW 12 of 18 Figure 4. The MAPK pathway participates in YES1-mediated internalization of rF508del-CFTR at the PM. The thermal shift (TS) assay, as described in Figure 2, was used to assess the thermal stability of untagged rF508del-CFTR in CFBE cells. (A) The cells were treated for 3 h with either vehicle (DMSO) or 10 µM of selumetinib, or (C) transfected with empty vector or Myc-H-RAS-V12 (HRAS) and then treated for 3 h with vehicle (DMSO) or 10 µM of SU6656, as indicated. WBs representative of input and cell surface fractions from four independent assays, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut-1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. H-RAS V12 was detected using an anti-Myc antibody, and an anti-phosphorylated ERK1/2 antibody was used to monitor MAPK Biomolecules 2023, 13, 949 13 of 17 retention of rF508del-CFTR at the PM to levels comparable to those observed after YES1 or YAP1 downregulation upon thermal destabilization (Figure 5B, quantiﬁed in Figure 5C). Figure 5. SHC1 phosphorylation by YES1 mediates rF508del-CFTR internalization via MAPK path- way signaling. (A) Effects of YES1 inhibitors. Lysates from F508del-CFTR expressing CFBE cells were incubated with 5 µM of VX-661 for 48 h at 30 ◦C, treated with either vehicle (DMSO), SU6656 (10 µM), or P505-15 (1 µM) for 3 h at 37 ◦C, then were analyzed using WB. Immunoblots representa- tive of three independent experiments, probed with antibodies against the indicated proteins, are shown. p-SHC1 indicates the level of SHC1 phosphorylation at Tyr239/240 in the different conditions, and an anti-phosphorylated ERK1/2 antibody (p-ERK1/2) was used to monitor MAPK pathway activity (both show quantiﬁed band intensities below their respective blots). (B) Thermal shift (TS) assay described in Figure 2 to assess how much rF508del-CFTR remained at the PM after thermal destabilization in CFBE cells transfected either with a mock siRNA (siCtrl) or a siRNA against SHC1 (siSHC1). WBs representative of input and cell surface fractions, probed with antibodies against the indicated proteins, are shown. Glucose transporter 1 (Glut-1) and tubulin were used as controls for the equivalence and purity of the biotinylated fractions, respectively. Quantiﬁcations of SCH1 depletion efﬁciency and ERK1/2 phosphorylation levels are shown below their respective blots. (C) Corresponding quantiﬁcation of CFTR abundance in the biotinylated fractions (mean ± SEM) from three independent assays after normalization to Glut-1 levels and to siCtrl (30 ◦C). *** p < 0.001 relative to siCtrl (30 ◦C), ## p < 0.01 relative to siCtrl (TS). Taken together, these results suggest that SHC1 phosphorylation by YES1 mediates rF508del-CFTR internalization via MAPK signaling. Consistently, depletion of SHC1 in these cells also reduced ERK1/2 phosphorylation (p < 0.01) to levels comparable to those of YES1 inhibition (Figure 5A,B). 4. Discussion Despite the signiﬁcant progress that CFTR modulators have brought to CF therapy, their clinical effects remain limited by their inability to fully restore rF508del-CFTR stability at the PM [3,10,11,14]. Hence, recent efforts have been made to better characterize the Biomolecules 2023, 13, x FOR PEER REVIEW 13 of 18 pathway activity, which was quantiﬁed and shown below the respective blot lanes. (B,D) Corre-sponding quantiﬁcation of CFTR abundance in the biotinylated fractions (mean ± SEM) from four independent assays after normalization to Glut-1 levels and DMSO (30 °C). * p < 0.05, p < 0.01, and * p < 0.001, relative to DMSO (30 °C) in (B), and as indicated by the horizontal lines in (C); ## p < 0.01, relative to DMSO (TS) in (B). 3.5. Phosphorylation of SHC1 by YES1 Links rF508del-CFTR Internalization to MAPK Pathway Activation In order to identify a direct link between YES1 activity and MAPK activation, we investigated SHC-transforming protein 1 (SHC1), which has been reported to be phos-phorylated by YES1 at Tyr239 and Tyr240 [38,39]. Phosphorylated SHC1 acts as an adap-tor protein that improves the signal transduction between stimulated PM receptors and RAS proteins, leading to the stimulation of the MAPK cascade [38–40]. While we could not reliably detect SHC1 co-precipitating with YES1 in our experimental settings (possibly reﬂecting a transient enzyme–substrate interaction), we could nevertheless detect a signif-icant (p < 0.01) decrease in SHC1 Tyr239/240 phosphorylation upon inhibition of YES1 activity after 3 h treatment with either SU6656 (10 µM) or P505-15 (1 µM) (Figure 5A). Moreover, an over 70% depletion of endogenous SHC1 levels (p < 0.01) in CFBE cells led to a signiﬁcant (p < 0.01) retention of rF508del-CFTR at the PM to levels comparable to those observed after YES1 or YAP1 downregulation upon thermal destabilization (Figure 5B, quantiﬁed in 5C). Taken together, these results suggest that SHC1 phosphorylation by YES1 mediates rF508del-CFTR internalization via MAPK signaling. Consistently, depletion of SHC1 in these cells also reduced ERK1/2 phosphorylation (p < 0.01) to levels comparable to those of YES1 inhibition (Figure 5A,B). Figure 5. SHC1 phosphorylation by YES1 mediates rF508del-CFTR internalization via MAPK path-way signaling. (A) Eﬀects of YES1 inhibitors. Lysates from F508del-CFTR expressing CFBE cells were incubated with 5 µM of VX-661 for 48 h at 30 °C, treated with either vehicle (DMSO), SU6656 Biomolecules 2023, 13, 949 14 of 17 molecular mechanisms of CFTR proteostasis in order to identify targetable molecules that can modulate these processes and enhance the effectiveness of CFTR modulator therapy [3,41]. In previous work, we characterized a novel pathway that regulates wt-CFTR retention at the PM: when phosphorylated by spleen tyrosine kinase (SYK) at its Tyr512 residue, wt-CFTR is removed from the PM [42,43]. The effect was found to be mediated by adaptor protein SHC1, which recognizes Tyr512-phosphorylated CFTR through its phosphotyrosine-binding domain and links CFTR internalization to the activation of the MAPK pathway [35,43,44] (see Figure 6A). In contrast, rF508del-CFTR internalization did not respond to SYK modulation, likely due to the misfolding caused by the Phe508 deletion, which makes the kinase fail to recognize and phosphorylate the nearby Tyr512 residue [42,43]. However, our ﬁndings presented here uncovered an alternative mechanism that links rF508del-CFTR to SHC1 and MAPK pathway-associated internalization. We showed that rF508del-CFTR (but not wt-CFTR) at the PM binds to YES1 kinase through the adaptor protein YAP1. YAP1 has been described to bridge the interaction of YES1 with NHERF1 through a strong interaction between YAP1’s C-terminus and NHERF1’s second PDZ domain, even in the absence of Ezrin [34]. This is consistent with our previous ﬁnding that most rF508del-CFTR at the PM remains bound to NHERF1’s ﬁrst PDZ [14]. This results from the interaction of rF508del-CFTR with the protease Calpain-1 at the PM [24], which prevents Ezrin recruitment and consequently blocks rF508del-CFTR from switching to NHERF1’s PDZ2 [14,24] (see Figure 6B). Figure 6. Proposed model for SHC1-mediated removal of CFTR from the PM through activation of the MAPK pathway. (A) Phosphorylation of PM-anchored wt-CFTR at Tyr512 by SYK kinase leads to its recognition and binding by the adaptor protein SHC1. This links CFTR internalization to the activation of the MAPK pathway downstream of receptor tyrosine kinases. (B) F508del-CFTR pharmacologically rescued to the PM is not phosphorylated by SYK, but its deﬁcient anchoring to the actin cytoskeleton allows its interaction with the YES1 kinase via the adaptor protein YAP1 and the scaffold protein NHERF1. SHC1 is a substrate for YES1 at the PM, and its phosphorylation by YES1 increases its afﬁnity to membrane receptors in the vicinity, enhancing their activation of the MAPK pathway. This could contribute to the much faster internalization rate of rF508del-CFTR compared to the wild-type protein. RTK—receptor tyrosine kinases; EB—Ezrin binding domain; 1-2—NHERF1’s PDZ1 and PDZ2. In addition, we demonstrated that rF508del-CFTR-bound YES1 can also phosphorylate SHC1. Consistently, interfering with either component of this complex or inhibiting YES1 activity resulted in decreased SHC1 phosphorylation and a signiﬁcant delay in rF508del- CFTR internalization. We also showed that overexpression of a constitutively active mutant RAS protein (H-RAS-V12) could bypass SHC1-mediated activation of the MAPK pathway and lead to rF508del-CFTR internalization. MAPK pathway activation occurs downstream of receptor tyrosine kinases (RTK) in response to mitogenic stimuli [45,46]. In response to RTK activation, a complex composed by the GRB2 scaffold protein and the RAS guanine Biomolecules 2023, 13, x FOR PEER REVIEW 15 of 18 complexes at the PM may further contribute to the rapid internalization of the rescued mutant channels by improving MAPK signaling in their vicinity (Figure 6). Determining the precise mechanism by which MAPK activation promotes CFTR internalization will require further investigation. Figure 6. Proposed model for SHC1-mediated removal of CFTR from the PM through activation of the MAPK pathway. (A) Phosphorylation of PM-anchored wt-CFTR at Tyr512 by SYK kinase leads to its recognition and binding by the adaptor protein SHC1. This links CFTR internalization to the activation of the MAPK pathway downstream of receptor tyrosine kinases. (B) F508del-CFTR phar-macologically rescued to the PM is not phosphorylated by SYK, but its deﬁcient anchoring to the actin cytoskeleton allows its interaction with the YES1 kinase via the adaptor protein YAP1 and the scaﬀold protein NHERF1. SHC1 is a substrate for YES1 at the PM, and its phosphorylation by YES1 increases its aﬃnity to membrane receptors in the vicinity, enhancing their activation of the MAPK pathway. This could contribute to the much faster internalization rate of rF508del-CFTR compared to the wild-type protein. RTK—receptor tyrosine kinases; EB—Ezrin binding domain; 1-2—NHERF1’s PDZ1 and PDZ2. 5. Conclusions Our ﬁndings provide new, important insights into the mechanisms regulating the accelerated internalization of rF508del-CFTR in airway cells. Moreover, the involvement of YES1 kinase and the MAPK pathway in removing the rescued channels from the PM opens new avenues for future CF research. In this study, we demonstrated that treatment with the MEK inhibitor selumetinib signiﬁcantly improved the stability and retention of VX-661-rescued F508del-CFTR at the PM. Similar results were observed upon treatment with the YES1 inhibitors SU6656 and P505-15. Having identiﬁed these new pathways, it will now be important to validate these ﬁndings in patient-derived materials. It will also be important to determine the extent to which these pathways hinder the eﬀect of recently clinically explored VX-661/VX-445 additive corrector combination and whether the YES1/MAPK-mediated PM removal mechanism can be attenuated by any of the many new CFTR modulators in clinical trials. Moreover, while several drugs are currently avail-able to inhibit both the MAPK pathway and the activity of Src family kinases, such as YES1, [25,47] the continuous inhibition of these pathways may have deleterious side ef-fects [48]. Therefore, translation of our ﬁndings into a CF therapeutic context will require additional research in order to develop ways to safely target these pathways to enhance CFTR modulator therapy. Notwithstanding, in our view, our data bring exciting new av-enues for further exploration in both CF research and treatment. Author Contributions: Conceptualization, P.M. and P.J.; formal analysis, P.B. and P.M.; funding acquisition, P.M. and P.J.; investigation, P.B. and A.M.M.; writing—original draft, P.B. and P.M.; writing—review and editing, P.J. All authors have read and agreed to the published version of the manuscript. Biomolecules 2023, 13, 949 15 of 17 exchange factor SOS1 (which induces RAS activation) is recruited to the activated RTK, either directly or indirectly via adaptors such as SHC1 [45] (see Figure 6). SHC1 contains an N-terminal PTB domain and a C-terminal SH2 domain, which are both able to bind phosphorylated tyrosine residues on other proteins [46]. These ﬂank a central proline-rich region that also contains the tyrosine sites for SHC1 phosphorylation [39]. SHC1 phospho- rylation at Tyr239/240 by YES1 greatly increased its afﬁnity to GRB2/SOS1, boosting RAS activation and MAPK signaling [38,39,46]. Thus, in contrast to wt-CFTR, the phosphoryla- tion of SHC1 by YES1 upon its recruitment to rF508del-CFTR complexes at the PM may further contribute to the rapid internalization of the rescued mutant channels by improving MAPK signaling in their vicinity (Figure 6). Determining the precise mechanism by which MAPK activation promotes CFTR internalization will require further investigation. 5. Conclusions Our ﬁndings provide new, important insights into the mechanisms regulating the accelerated internalization of rF508del-CFTR in airway cells. Moreover, the involvement of YES1 kinase and the MAPK pathway in removing the rescued channels from the PM opens new avenues for future CF research. In this study, we demonstrated that treatment with the MEK inhibitor selumetinib signiﬁcantly improved the stability and retention of VX-661-rescued F508del-CFTR at the PM. Similar results were observed upon treatment with the YES1 inhibitors SU6656 and P505-15. Having identiﬁed these new pathways, it will now be important to validate these ﬁndings in patient-derived materials. It will also be important to determine the extent to which these pathways hinder the effect of recently clinically explored VX-661/VX-445 additive corrector combination and whether the YES1/MAPK-mediated PM removal mechanism can be attenuated by any of the many new CFTR modulators in clinical trials. Moreover, while several drugs are currently available to inhibit both the MAPK pathway and the activity of Src family kinases, such as YES1, [25,47] the continuous inhibition of these pathways may have deleterious side effects [48]. Therefore, translation of our ﬁndings into a CF therapeutic context will require additional research in order to develop ways to safely target these pathways to enhance CFTR modulator therapy. Notwithstanding, in our view, our data bring exciting new avenues for further exploration in both CF research and treatment. Author Contributions: Conceptualization, P.M. and P.J.; formal analysis, P.B. and P.M.; funding acquisition, P.M. and P.J.; investigation, P.B. and A.M.M.; writing—original draft, P.B. and P.M.; writing—review and editing, P.J. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the Grant PTDC/BIA-CEL/28408/2017 (to PJ and PM) and Center Grant UID/MULTI/04046/2019 to BioISI from the Portuguese Fundação para a Ciência e a Tecnologia. Institutional Review Board Statement: Not applicable since the study did not involve humans or animals. Informed Consent Statement: Not applicable since the study did not involve human subjects. Data Availability Statement: This study did not report any data not shown in the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest. References Riordan, J.R. CFTR Function and Prospects for Therapy. Annu. Rev. Biochem. 2008, 77, 701–726. [CrossRef] [PubMed] 1. 2. Myer, H.; Chupita, S.; Jnah, A. Cystic Fibrosis: Back to the Basics. Neonatal Netw. 2023, 42, 23–30. [CrossRef] [PubMed] 3. Brusa, I.; Sondo, E.; Falchi, F.; Pedemonte, N.; Roberti, M.; Cavalli, A. Proteostasis Regulators in Cystic Fibrosis: Current Development and Future Perspectives. J. Med. Chem. 2022, 65, 5212–5243. [CrossRef] Thibodeau, P.H.; Richardson, J.M.; Wang, W.; Millen, L.; Watson, J.; Mendoza, J.L.; Du, K.; Fischman, S.; Senderowitz, H.; Lukacs, G.L.; et al. The Cystic Fibrosis-Causing Mutation DeltaF508 Affects Multiple Steps in Cystic Fibrosis Transmembrane Conductance Regulator Biogenesis. J. Biol. Chem. 2010, 285, 35825–35835. [CrossRef] [PubMed] 4. Biomolecules 2023, 13, 949 16 of 17 5. 6. 7. Guggino, W.B.; Stanton, B.A. New Insights into Cystic Fibrosis: Molecular Switches That Regulate CFTR. Nat. Rev. Mol. Cell. Biol. 2006, 7, 426–436. [CrossRef] Elborn, J.S. Cystic Fibrosis. Lancet 2016, 388, 2519–2531. [CrossRef] [PubMed] Lopes-Pacheco, M.; Pedemonte, N.; Veit, G. Discovery of CFTR Modulators for the Treatment of Cystic Fibrosis. Expert. Opin. Drug. Discov. 2021, 16, 897–913. [CrossRef] 8. Heijerman, H.G.M.; McKone, E.F.; Downey, D.G.; Van Braeckel, E.; Rowe, S.M.; Tullis, E.; Mall, M.A.; Welter, J.J.; Ramsey, B.W.; McKee, C.M.; et al. Efﬁcacy and Safety of the Elexacaftor plus Tezacaftor plus Ivacaftor Combination Regimen in People with Cystic Fibrosis Homozygous for the F508del Mutation: A Double-Blind, Randomised, Phase 3 Trial. Lancet 2019, 394, 1940–1948. [CrossRef] Purkayastha, D.; Agtarap, K.; Wong, K.; Pereira, O.; Co, J.; Pakhale, S.; Kanji, S. Drug-Drug Interactions with CFTR Modulator Therapy in Cystic Fibrosis: Focus on Trikafta®/Kaftrio®. J. Cyst. Fibros. Off. J. Eur. Cyst. Fibros. Soc. 2023; in press. [CrossRef] 9. 10. Capurro, V.; Tomati, V.; Sondo, E.; Renda, M.; Borrelli, A.; Pastorino, C.; Guidone, D.; Venturini, A.; Giraudo, A.; Mandrup Bertozzi, S.; et al. Partial Rescue of F508del-CFTR Stability and Trafﬁcking Defects by Double Corrector Treatment. Int. J. Mol. Sci. 2021, 22, 5262. [CrossRef] 11. Veit, G.; Roldan, A.; Hancock, M.A.; Da Fonte, D.F.; Xu, H.; Hussein, M.; Frenkiel, S.; Matouk, E.; Velkov, T.; Lukacs, G.L. Allosteric Folding Correction of F508del and Rare CFTR Mutants by Elexacaftor-Tezacaftor-Ivacaftor (Trikafta) Combination. JCI Insight 2020, 5, e139983. [CrossRef] [PubMed] 12. Okiyoneda, T.; Barrière, H.; Bagdány, M.; Rabeh, W.M.; Du, K.; Höhfeld, J.; Young, J.C.; Lukacs, G.L. Peripheral Protein Quality 13. Control Removes Unfolded CFTR from the Plasma Membrane. Science 2010, 329, 805–810. [CrossRef] [PubMed] Fukuda, R.; Okiyoneda, T. Peripheral Protein Quality Control as a Novel Drug Target for CFTR Stabilizer. Front. Pharmacol. 2018, 9, 1100. [CrossRef] [PubMed] 15. 14. Loureiro, C.A.; Matos, A.M.; Dias-Alves, Â.; Pereira, J.F.; Uliyakina, I.; Barros, P.; Amaral, M.D.; Matos, P. A Molecular Switch in the Scaffold NHERF1 Enables Misfolded CFTR to Evade the Peripheral Quality Control Checkpoint. Sci. Signal. 2015, 8, ra48. [CrossRef] Swiatecka-Urban, A.; Brown, A.; Moreau-Marquis, S.; Renuka, J.; Coutermarsh, B.; Barnaby, R.; Karlson, K.H.; Flotte, T.R.; Fukuda, M.; Langford, G.M.; et al. The Short Apical Membrane Half-Life of Rescued {Delta}F508-Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) Results from Accelerated Endocytosis of {Delta}F508-CFTR in Polarized Human Airway Epithelial Cells. J. Biol. Chem. 2005, 280, 36762–36772. [CrossRef] Sharma, M.; Pampinella, F.; Nemes, C.; Benharouga, M.; So, J.; Du, K.; Bache, K.G.; Papsin, B.; Zerangue, N.; Stenmark, H.; et al. Misfolding Diverts CFTR from Recycling to Degradation: Quality Control at Early Endosomes. J. Cell. Biol. 2004, 164, 923–933. [CrossRef] 16. 18. 17. Varga, K.; Goldstein, R.F.; Jurkuvenaite, A.; Chen, L.; Matalon, S.; Sorscher, E.J.; Bebok, Z.; Collawn, J.F. Enhanced Cell-Surface Stability of Rescued DeltaF508 Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) by Pharmacological Chaperones. Biochem. J. 2008, 410, 555–564. [CrossRef] Favia, M.; Guerra, L.; Fanelli, T.; Cardone, R.A.; Monterisi, S.; Di Sole, F.; Castellani, S.; Chen, M.; Seidler, U.; Reshkin, S.J.; et al. Na+/H+ Exchanger Regulatory Factor 1 Overexpression-Dependent Increase of Cytoskeleton Organization Is Fundamental in the Rescue of F508del Cystic Fibrosis Transmembrane Conductance Regulator in Human Airway CFBE41o- Cells. Mol. Biol. Cell. 2010, 21, 73–86. [CrossRef] 19. Arora, K.; Moon, C.; Zhang, W.; Yarlagadda, S.; Penmatsa, H.; Ren, A.; Sinha, C.; Naren, A.P. Stabilizing Rescued Surface-Localized ∆F508 CFTR by Potentiation of Its Interaction with Na(+)/H(+) Exchanger Regulatory Factor 1. Biochemistry 2014, 53, 4169–4179. [CrossRef] 20. Haggie, P.M.; Kim, J.K.; Lukacs, G.L.; Verkman, A.S. Tracking of Quantum Dot-Labeled CFTR Shows near Immobilization by C-Terminal PDZ Interactions. Mol. Biol. Cell. 2006, 17, 4937–4945. [CrossRef] 21. Li, J.; Dai, Z.; Jana, D.; Callaway, D.J.; Bu, Z. Ezrin Controls the Macromolecular Complexes Formed between an Adapter Protein Na+/H+ Exchanger Regulatory Factor and the Cystic Fibrosis Transmembrane Conductance Regulator. J. Biol. Chem. 2005, 280, 37634–37643. [CrossRef] [PubMed] 22. Morales, F.C.; Takahashi, Y.; Momin, S.; Adams, H.; Chen, X.; Georgescu, M.-M. NHERF1/EBP50 Head-to-Tail Intramolecular Interaction Masks Association with PDZ Domain Ligands. Mol. Cell. Biol. 2007, 27, 2527–2537. [CrossRef] [PubMed] 23. Castellani, S.; Guerra, L.; Favia, M.; Di Gioia, S.; Casavola, V.; Conese, M. NHERF1 and CFTR Restore Tight Junction Organisation and Function in Cystic Fibrosis Airway Epithelial Cells: Role of Ezrin and the RhoA/ROCK Pathway. Lab. Investig. J. Tech. Methods Pathol. 2012, 92, 1527–1540. [CrossRef] [PubMed] 24. Matos, A.M.; Pinto, F.R.; Barros, P.; Amaral, M.D.; Pepperkok, R.; Matos, P. Inhibition of Calpain 1 Restores Plasma Membrane Stability to Pharmacologically Rescued Phe508del-CFTR Variant. J. Biol. Chem. 2019, 294, 13396–13410. [CrossRef] [PubMed] 25. Garmendia, I.; Redin, E.; Montuenga, L.M.; Calvo, A. YES1: A Novel Therapeutic Target and Biomarker in Cancer. Mol. Cancer Ther. 2022, 21, 1371–1380. [CrossRef] 26. Almaça, J.; Dahimène, S.; Appel, N.; Conrad, C.; Kunzelmann, K.; Pepperkok, R.; Amaral, M.D. Functional Genomics Assays to Study CFTR Trafﬁc and ENaC Function. Methods Mol. Biol. 2011, 742, 249–264. [CrossRef] 27. Galietta, L.J.; Haggie, P.M.; Verkman, A.S. Green Fluorescent Protein-Based Halide Indicators with Improved Chloride and Iodide Afﬁnities. FEBS Lett. 2001, 499, 220–224. [CrossRef] Biomolecules 2023, 13, 949 17 of 17 28. Botelho, H.M.; Uliyakina, I.; Awatade, N.T.; Proença, M.C.; Tischer, C.; Sirianant, L.; Kunzelmann, K.; Pepperkok, R.; Amaral, M.D. Protein Trafﬁc Disorders: An Effective High-Throughput Fluorescence Microscopy Pipeline for Drug Discovery. Sci. Rep. 2015, 5, 9038. [CrossRef] 29. Matos, P.; Oliveira, C.; Velho, S.; Gonçalves, V.; da Costa, L.T.; Moyer, M.P.; Seruca, R.; Jordan, P. B-Raf(V600E) Cooperates with Alternative Spliced Rac1b to Sustain Colorectal Cancer Cell Survival. Gastroenterology 2008, 135, 899–906. [CrossRef] 30. Loureiro, C.A.; Santos, J.D.; Matos, A.M.; Jordan, P.; Matos, P.; Farinha, C.M.; Pinto, F.R. Network Biology Identiﬁes Novel Regulators of CFTR Trafﬁcking and Membrane Stability. Front. Pharmacol. 2019, 10, 619. [CrossRef] 31. Matos, A.M.; Gomes-Duarte, A.; Faria, M.; Barros, P.; Jordan, P.; Amaral, M.D.; Matos, P. Prolonged Co-Treatment with HGF Sustains Epithelial Integrity and Improves Pharmacological Rescue of Phe508del-CFTR. Sci. Rep. 2018, 8, 13026. [CrossRef] [PubMed] 32. Badouel, C.; Garg, A.; McNeill, H. Herding Hippos: Regulating Growth in Flies and Man. Curr. Opin. Cell. Biol. 2009, 21, 837–843. 33. [CrossRef] [PubMed] Sudol, M. Yes-Associated Protein (YAP65) Is a Proline-Rich Phosphoprotein That Binds to the SH3 Domain of the Yes Proto- Oncogene Product. Oncogene 1994, 9, 2145–2152. [PubMed] 34. Mohler, P.J.; Kreda, S.M.; Boucher, R.C.; Sudol, M.; Stutts, M.J.; Milgram, S.L. Yes-Associated Protein 65 Localizes P62(c-Yes) to the Apical Compartment of Airway Epithelia by Association with EBP50. J. Cell. Biol. 1999, 147, 879–890. [CrossRef] [PubMed] 35. Xu, X.; Balsiger, R.; Tyrrell, J.; Boyaka, P.N.; Tarran, R.; Cormet-Boyaka, E. Cigarette Smoke Exposure Reveals a Novel Role for the MEK/ERK1/2 MAPK Pathway in Regulation of CFTR. Biochim. Biophys. Acta 2015, 1850, 1224–1232. [CrossRef] 36. Osrodek, M.; Wozniak, M. Targeting Genome Stability in Melanoma—A New Approach to an Old Field. Int. J. Mol. Sci. 2021, 22, 3485. [CrossRef] 37. Matos, P.; Jordan, P. Expression of Rac1b Stimulates NF-KappaB-Mediated Cell Survival and G1/S Progression. Exp. Cell. Res. 2005, 305, 292–299. [CrossRef] 38. Dubois, F.; Leroy, C.; Simon, V.; Benistant, C.; Roche, S. YES Oncogenic Activity Is Speciﬁed by Its SH4 Domain and Regulates 39. RAS/MAPK Signaling in Colon Carcinoma Cells. Am. J. Cancer Res. 2015, 5, 1972–1987. van der Geer, P.; Wiley, S.; Gish, G.D.; Pawson, T. The Shc Adaptor Protein Is Highly Phosphorylated at Conserved, Twin Tyrosine Residues (Y239/240) That Mediate Protein-Protein Interactions. Curr. Biol. 1996, 6, 1435–1444. [CrossRef] 40. Mushtaq, U.; Bashir, M.; Nabi, S.; Khanday, F.A. Epidermal Growth Factor Receptor and Integrins Meet Redox Signaling through 41. P66shc and Rac1. Cytokine 2021, 146, 155625. [CrossRef] Farinha, C.M.; Gentzsch, M. Revisiting CFTR Interactions: Old Partners and New Players. Int. J. Mol. Sci. 2021, 22, 13196. [CrossRef] [PubMed] 42. Mendes, A.I.; Matos, P.; Moniz, S.; Luz, S.; Amaral, M.D.; Farinha, C.M.; Jordan, P. Antagonistic Regulation of Cystic Fibrosis Transmembrane Conductance Regulator Cell Surface Expression by Protein Kinases WNK4 and Spleen Tyrosine Kinase. Mol. Cell. Biol. 2011, 31, 4076–4086. [CrossRef] [PubMed] 43. Loureiro, C.A.; Pinto, F.R.; Barros, P.; Matos, P.; Jordan, P. A SYK/SHC1 Pathway Regulates the Amount of CFTR in the Plasma Membrane. Cell. Mol. Life Sci. 2020, 77, 4997–5015. [CrossRef] 44. Wellmerling, J.; Rayner, R.E.; Chang, S.-W.; Kairis, E.L.; Kim, S.H.; Sharma, A.; Boyaka, P.N.; Cormet-Boyaka, E. Targeting the EGFR-ERK Axis Using the Compatible Solute Ectoine to Stabilize CFTR Mutant F508del. FASEB J. Off. Publ. Fed. Am. Soc. Exp. Biol. 2022, 36, e22270. [CrossRef] 45. Kolch, W.; Berta, D.; Rosta, E. Dynamic Regulation of RAS and RAS Signaling. Biochem. J. 2023, 480, 1–23. [CrossRef] [PubMed] 46. Zheng, Y.; Zhang, C.; Croucher, D.R.; Soliman, M.A.; St-Denis, N.; Pasculescu, A.; Taylor, L.; Tate, S.A.; Hardy, W.R.; Colwill, K.; et al. Temporal Regulation of EGF Signalling Networks by the Scaffold Protein Shc1. Nature 2013, 499, 166–171. [CrossRef] Fu, L.; Chen, S.; He, G.; Chen, Y.; Liu, B. Targeting Extracellular Signal-Regulated Protein Kinase 1/2 (ERK1/2) in Cancer: An Update on Pharmacological Small-Molecule Inhibitors. J. Med. Chem. 2022, 65, 13561–13573. [CrossRef] 47. 48. Viganò, M.; La Milia, M.; Grassini, M.V.; Pugliese, N.; De Giorgio, M.; Fagiuoli, S. Hepatotoxicity of Small Molecule Protein Kinase Inhibitors for Cancer. Cancers 2023, 15, 1766. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.1371_journal.pwat.0000075	RESEARCH ARTICLE Community perceptions and practices on quality and safety of drinking water in Mbarara city, south western Uganda Abaasa N. CatherineID Frederick Byarugaba1, Imelda K. Tamwesigire1 1, Savino Ayesiga1, Godfrey Zari Rukundo1, Julius B. Lejju2, 1 Faculty of Medicine, Mbarara University of Science and Technology, Mbarara, Uganda, 2 Faculty of Science, Mbarara University of Science and Technology, Mbarara, Uganda a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 [email protected] Abstract OPEN ACCESS Citation: Catherine AN, Ayesiga S, Rukundo GZ, Lejju JB, Byarugaba F, Tamwesigire IK (2023) Community perceptions and practices on quality and safety of drinking water in Mbarara city, south western Uganda. PLOS Water 2(5): e0000075. https://doi.org/10.1371/journal.pwat.0000075 Editor: Eugene Appiah-Effah, Kwame Nkrumah University of Science and Technology, GHANA Received: October 31, 2022 Accepted: April 29, 2023 Published: May 30, 2023 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.pwat.0000075 Copyright: © 2023 Catherine et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: To ensure confidentiality of participants’ information as agreed up on during Ethical approval and Consent Process, qualitative interview transcripts’ file is only Availability of clean drinking water is a universal human right. The quality of water differs across communities. When the quality is good, community members are the primary benefi- ciaries but they are also the first ones to experience the consequences of deteriorating qual- ity of water. In most communities, the inhabitants are able to tell if their drinking water is safe and of quality basing on organoleptic properties. The community perceptions and practices about safety and quality of drinking water are informed by their attitudes and levels of knowl- edge about water quality. This study aimed to assess community perceptions and practices on quality and safety of drinking water in Mbarara city, south western Uganda. A qualitative study was conducted between May and July 2022. Six focus group discussions among com- munity members and four Key informant interviews with stakeholders in the water service were conducted. Data was analysed basing on predetermined themes of: 1) perceived qual- ity of water 2) perceived factors associated with water quality 3) practices related to water quality and 4) perceived solutions for improving water safety and quality. Drinking water safety and quality in Mbarara city is perceived as not good, dirty, salty and limited in supply and the water sources are shared with animals. The poor quality of drinking water is due to poor waste disposal, poor treatment, poor maintenance of systems, flooding, political inter- ference, deficiency in city planning, increase in population growth and water hyacinth. Sensi- tizing the communities, community participation, proper water treatment and surveillance and monitoring are solutions to ensuring provision, use and maintenance of safe and quality drinking water in Mbarara city. Introduction Safe and quality water is defined as one free from any harmful chemicals and pathogenic agents such as Coliform bacteria, Escherichia coli and coliphages that affect its palatability as well as human livelihood [1, 2]. According to the United Nations Sustainable Development Goal 6, “water sustains life, but safe clean drinking water defines civilization” [3]. Drinking PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 1 / 20 visible to the direct research team or through Mbarara University of Science and Technology Research Ethics Committee, P.O. Box 1410 Mbarara, Tel: +256-48-543-3795, Fax: +256-48- 542-0782, E-mail: [email protected], [email protected] since they are not publically available. Funding: The corresponding author who happens to be the PI for this study CNA received research support as part of Faculty Research Support from Faculty of Medicine, Mbarara University of Science and Technology (MUST). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: NWSC, National Water and Sewerage Cooperation; NEMA, National Environment Management Authority; UBOS, Uganda Bureau of Statistics; WASH, Water, sanitation, and hygiene. Quality and safety of drinking water; A qualitative investigation water should be adequate, reliable, clean, accessible and acceptable to communities as a human right to live healthy lives [4]. There is evidence that steps aimed at providing safe water services have fairly increased, but its safety is uncertain [5]. Most households in low and mid- dle income countries lack sufficient safe and quality freshwater (physical scarcity). In other countries there is abundant freshwater of unknown quality hence communities supplement improved water supplies with unimproved water or multiple sources that may not be safe for human consumption because it is expensive to ensure adequate supply of safe and quality water [6, 7]. Safe drinking water” is water from an “improved water source,” that include household connections, public standpipes, boreholes, protected dug wells, protected springs and rainwater collections and it should not represent any significant risk to health over a life- time of consumption [8, 9]. Assessment of water quality is subjective and is based on beliefs, cognition, geographic location of the water source, socio economic characteristics, the experi- ence of the user and information provided by local media [10]. Community members are the primary beneficiaries of quality and safe water and they are the first to experience the consequences of the deteriorating water quality when known or sus- pected to be unsafe for human consumption due to regulatory problems and lack of support [11]. They evaluate the safety and quality of drinking water using its organoleptic properties like taste, smell, colour and clarity as well as presence of litter and sanitary conditions around the drinking water source [12]. These perceptions are however useful and help to complement scientific measurements hence supporting water management policies [13]. Community par- ticipation in water and sanitation is one of the prominent global indicators used to assess the achievement of water-related sustainable developmental goals [12]. In addition, public accept- ability of drinking water is one of the world health organization guidelines for drinking water quality [14]. Communities’ perception of the quality of drinking water is informed by health risk percep- tion, perceived control, past experience, trust on water service provider, influence of imper- sonal and interpersonal information like media and peers, contextual factors, colour, taste of the water, appearance and demographic variables [15, 16]. However, these perceptions are influenced by sociocultural, sociodemographic, and personal experiences and are shaped by service satisfaction, confidence in local and national authorities, selection of the water source as well as beliefs in human control of environment issues and formal and informal flow of information [17]. Community perceptions and practices of health risks related to drinking water are associated with drinking water, the persistence of these health problems and the level of awareness of the problem. Perception of the need for quality water drives the need to prac- tice activities at the source, during transportation or storage and handling practices that will ensure the safety and quality of water as well as the use and storage of this God given resource [18]. Water quality concerns like water scarcity, lack of awareness and knowledge of ’safe’ col- lection, handling and storage of water, inadequate sanitation services and/or unhygienic prac- tices exist in communities. In addition; water quality attributes like taste, colour, smell, litter and presence of feacal matter in and around the water source as well as education, age, number of years a person has lived in the community, presence of visible aspects of water pollution and water source catchment area encroachment influence community members’ perception on the quality and safety of drinking water [19]. These water quality concerns when left unresolved for long may lead to community perceptions of health risk and prompt community practices that may be dangerous to their health like use of chemicals or using alternative sources of water such as unprotected open wells, use of unprotected buckets left outside on the ground [20]. Through a dialogue with government, drinking water service providers, and community members’ perceptions of the quality of drinking water and associated health benefits and risks inform community practices to maintain water quality [11]. Understanding community beliefs PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 2 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation and behaviours is critical for water resource management, monitoring, and creating drinking water quality standards [11]. Anecdotal evidence reveals that the quality and safety of drinking water in Mbarara city do not meet World Health Organization criteria for drinking water quality. Thus, the purpose of this study was to explore community members’ perceptions and practices about the quality and safety of drinking water. Results of this study provides commu- nity members’ perceptions, practices and perceived solutions to improve and maintain safe and quality drinking water in Mbarara City, South Western Uganda. Materials and methods Ethics statement Administrative clearance was obtained from District, city, parish, National water and sewerage cooperation and Ministry of water, lands and Environment authorities. The protocol was reviewed and approved by Mbarara University of Science and Technology Institutional Review Committee (MUST-2021-39), and National Council of Science and Technology (HS1469ES). Permission was obtained from the district, local council leaders and household heads especially for water harvest tanks before commencement of data collection. Written informed consent was obtained from every participant before participating in the interviews and discussions. Study area This study was conducted in Mbarara city, south western Uganda. Mbarara city is the com- mercial and administrative capital of Mbarara district in south western Uganda. Mbarara city is located 270 kilometres, by road, southwest of the capital city, Kampala. Mbarara district lies between coordinates 00 36 48 S, 30 39 30 E and covers an area of 1,778.4 square kilometres. It has a population of 91867 [21]. Mbarara city receives an average annual rainfall of 1200 mm with two rainy seasons during the months of September-December and February-May. Tem- perature ranges between 17˚C to 30˚C, humidity of 80–90%. The topography is a mixture of fairly rolling and sharp hills and mountains, shallow valleys and flat land. Mbarara city is pro- vided, operated and maintained with safe water supply technologies and sanitation facilities to all communities of the city. Mbarara district recorded an increase in access to safe and clean water from 45% in 2000 to about 63% in the villages and 65% for the municipality in 2007. The safe water coverage is 65.9% in the rural areas and 95.7% in the urban, while accessibility to safe water lies between 29% and 95% [22]. Participant recruitment and description This study was a cross sectional study employing qualitative techniques. Purposive sampling was employed to recruit participants for the key informant interviews and focus group discus- sions. Four (4) key informants were recruited from the District water office, National Environ- ment Management Authority (NEMA), Ministry of Water, Lands and environment and National water and sewerage cooperation (NWSC) based their knowledge, expertise and expe- rience with water safety and quality in Mbarara city. The key informants were water quality control managers and policy makers. Eighty- four (84) community members from six (6) vil- lages/cells (Kaburangiire, Nyarubanga, Rubiri, Lugazi, Katebe and Katukuru) of Mbarara city were recruited for focus group discussions (FGDs). FGDs participants were residents of the selected villages who were consumers of the water from various water sources. Evidence has shown that people who reside and work near water sources are more likely to be concerned about the quality and safety of drinking water [23]. Both male and females across the different age groups that met the inclusion criteria of being community members in the six selected PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 3 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation villages or water service providers in Mbarara city irrespective of their gender and socio eco- nomic status were recruited to gain a diversity of perceptive and variability with in the com- munity. We did not collect information on the years of residency. We assumed that persons who had lived in the neighborhood for a longer period of time were better knowledgeable about the safety and quality of water in Mbarara. The participants in the Focus Group Discus- sions were chosen by the local chairman of the village. The six focus group discussions were constituted by senior inhabitants of the neighborhood who owned homes and had lived in those homes with their families for a long time, also known as "abataka," which literally means "permanent residents of the village." Data collection Local leaders as gatekeepers to the community were used to recruit FGD participants and fix dates and time for the discussions. Written informed consent was obtained from all research participants. The consent forms, focus group discussion and Key Informant interview guides were translated into Runyankore-Rukiga the language spoken and well understood in the study area. Discussion / Interview guides were used to collect data from key informants and FGDs between May and June 2022. The key study questions included: 1. What is your perception of the quality and safety of drinking water from sources in your community. 2. In your own view, who/what do you think is responsible for the quality and safety of drinking water you have described? 4. At family and community level, what has been done to ensure that drinking water in your community is of quality and is safe? 5. At family, community level, district/ as stakehold- ers what you done to ensure quality and safety of drinking water from drinking water sources in the community? These questions were elaborated on with more probing questions. AC con- ducted the interviews together with NP as a note taker and AT did the recording of the inter- views and discussion. Each focus group was comprised of 14 participants. Extra effort was made to ensure an equal number of males and females constituted the focus group discussions. The interviews and focus group discussions with the participants were conducted at a pri- vate location at the convenience of the different participants at the time agreed upon with the study team. The interviews were recorded with a Sony audio recorder and field notes were taken. Participants were not paid for participating in the study but time was compensated as was stipulated in the consent form. The interviews lasted between 60 and 90 minutes. The interviews were transcribed and those in Runyankole-Rukiga translated into English and back translated to Runyankole-Rukiga to ensure that what was recorded in Runyankole-Rukiga is what was captured in the English version of the transcript. Interview data was supplemented with field notes captured during the different interview and discussion sessions. One inter- view/ focus group discussions was conducted per day. Data analysis and interpretations Data analysis started with listening to the audio recordings alongside the field notes at the end of each day’s interview/ discussion session. They were transcribed sequentially on the daily basis by CA, NP and OJ which helped in giving a deeper insight into the inquiry during the data collection process in line with the study objectives. The data was transcribed by Research Assistant [24] and checked by CA and OJ. Data analysis was done through different stages of familiarization with data and dual coding was employed. CA, OJ and ACD independently read through the transcripts and identified emerging themes and manually identified correspond- ing quotes by highlighting them with different colors per theme. Data management from inter- views and focus group discussions were analyzed differently and merged in one codebook by incorporating data from audio recordings, verbatim notes and nonverbal observations during PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 4 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation the interview and discussion processes. A codebook with sections for parent themes, sub themes, description and illustrative quotes was developed from emerging themes. Using the four predetermined themes, indicative thematic analysis was done by analyzing statements from participants, identifying commonalities and developing sub themes. The same data was entered into Atlas Ti 7.5. Using the themes, each transcript was re-analyzed to reveal the best corresponding quotes. The same process was done for key informant interviews and focus group discussions data. Results Findings from this study reveal the community perceptions and practices of community mem- bers and stakeholders on the quality of drinking water in Mbarara city, south western Uganda. The results are from Four (4) key informant interviews and six Focus Group Discussions from stakeholders and community members in Mbarara city. A total of 28 males and 56 females constituted the interviews and discussions. Participant quotes are presented to support the findings. Four themes were identified, Community perceptions on the quality and safety of drinking water, factors responsible for the quality of drinking water, community perceived solution for safe and quality drinking water and community practices for safe and quality drinking water as well as several subthemes as shown in Table 1. Community member’s perception on the quality and safety of drinking water from sources in their community We explored community members’ perceptions on the quality and safety of drinking water in selected villages in Mbarara city. We asked the community members about their perception Table 1. Themes and subthemes. THEME Community perceptions on the quality and safety of drinking water Community practices for safe and quality drinking water Community perceived factors responsible for poor quality of drinking water Community Perceived solution for safe and quality water https://doi.org/10.1371/journal.pwat.0000075.t001 SUBTHEME Not good Dirty Salty Water rationing Legislation/permits Quality control Security Water treatment Alternative sources Sorting/recycling plastics Poor waste disposal Poor water treatment Poor maintenance of systems Political interference Population growth Water hyacinth Sensitization Behavioural Change Community participation, water committees, catchment management committees, Water Treatment Surveillance and Monitoring PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 5 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation on the quality and safety of their drinking water from the different drinking water sources in their villages. They gave a wide range of perceptions regarding the different drinking water sources. Generally, they indicated that the drinking water from different drinking water sources is not good, dirty and tasted salty. They believe that the quality is generally lacking since it is contaminated with human waste, cow dung, always changes in colour, it sediments on settling and the water sources are communally used and shared in some cases shared with animals. The supply on taps is unpredictable while boreholes suffer serious mechanical prob- lems and are usually out of use and for some long period of time. To be honest, our water is bad since we share it with animals, dumping cow dung in it, and we also use it for cooking and bathing and the problem is that we do not know which water is fit for consumption and how it looks like because we think good water is pure white, which you may discover is not actually good, so we do not know which one is good and we cannot avoid water from the well even taps are few and used by a small number of people; most of us go to the well (FGD II). We have a borehole, but the water cannot be used for cooking or bathing. Nothing good comes from that borehole, unless you wish to mix the soil for making bricks. Even when you wash your hands, they turn black. So in Kaburangire, we have a water problem, and even the rust they mentioned is there, so that is the type of water we have, and the borehole is now not oper- ational (FGD III). Similarly, a participant was noted to have said: The safety and quality of drinking water is not all that bad because we follow world health drinking water standard regulations, but at times you find that the sources are contaminated when you are not aware, and we can face a challenge because this water that passes under- ground can be contaminated with anything, let us talk about boreholes, even a tap. Knowing that taps are sometimes passing through sewerage areas, someone can cut it and start getting water back flow (KII IV). In addition, to the poor quality of drinking water, the participants reported that the supply is not constant It comes and goes, and when it returns, it is different from what we had before it disappeared, and even when you put it in your mouth, you can feel the difference, and later, when it settles, a lot of sediments appear, which raises the concern on the quality (FGD IV). Community member’s perceived factors responsible for the quality and safety of water in their community Communities attribute the poor quality and safety of drinking water from sources in their communities to the growing population in the city. Mbarara’s population is growing rapidly because, as a new city, there is scramble for resources. There are a lot of issues and of course as national water, we cannot meet the demands at the moment and people are opting for other sources (KII I). Because of the rising population, garbage is deposited everywhere, rendering our water unfit for human consumption (FGD IV). PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 6 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation There is a lot of garbage volume and residents live in congested homesteads with no provi- sion for waste disposal. They resort to throwing their waste in River Rwizi at night. Waste management is poor, and everyone dumps wherever they please; laws are in place, but they are not followed; personnel exist to execute laws, but when you try to do what is right, they will claim that you are interfering with voters (KII II). Sometimes you see someone on a boda boda (motorcycle) and you think that he is carrying luggage, and for those of us who walk at night, he reaches the bridge and stops for a while and dumps garbage into the river, thus we live by God’s grace. For example, someone may have garbage in the house and when she wakes up in the morning, she goes with it and throws it in the river because she has nowhere to put it, but those ditches will undoubtedly help us, so in your research, you should coordinate with the government to put emphasis on landlords dig- ging up ditches for their tenants (FGD I). The toilet facilities are few and unhygienic making residents to resort to other options like open defecation, using polythene bags and plastic bottles for their excreta which is dumped /poured in the garbage bins or trenches. Some landlords open their toilets to empty directly in River Rwizi. We have seen that pollution is from domestic wastes, people opening their latrines anyhow, so everything ends up in the system, which means you cannot rule out the drainage system, so we simply urge people to connect to national water because we know it is safe (KII I). Participants believe that national water and sewerage cooperation is not treating the water they supply for use or does not treat the water properly. They believe that sometimes they use poor quality chemicals which remain as residues in the water as it is supplied. You go to get water and find that it is really River Rwizi water that is very brown in color and I don’t know what causes that because I believe they treat it and how come it is dirty as if it was not treated and even after boiling it and putting it down to settle, you find those brown things on the bottom so that water is not good and those who drink it without boiling it will get sick (FGD VI). Other participants noted that drinking water is treated before it is availed for use but how- ever attributed the poor quality of water accessed to the location of the consumer. People who live in valleys have a higher chance of receiving dirty water because that is where settlement takes place and everything settles where there is a valley and you find that people who are in valleys have issues with water so the people at the end have a higher chance of receiving dirty water but we always put mitigation measures, for example, we encourage regu- lar flashing of our systems. There are planned sessions every three months sometimes we con- duct unplanned system flashing, which is done after getting complaints (KII III). Participants believe that the safety and quality of drinking water is affected by poor mainte- nance of water systems. Once the water gets to the individual consumers, the overhead tanks are dirty since they are never or rarely cleaned. National water and sewerage cooperation is using old dysfunctional water systems (pipes) that have never been changed from the time they were installed and sometimes lack enough chemicals for treatment. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 7 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation They are also not cleaning their tanks on a regular basis. Sincerely speaking, people do not wash their tanks for 3–4 years, the tank is just there, there are lizards, bird droppings, mon- keys playing on it, and people are unaware of its effect, but people keep saying that national water gives us bad water because people are not sensitized, this is because our duty ends at your tap beyond your tap, it is your responsibility (KII III). We have ancient pipes. I believe these pipes are 72 years old, making them incredibly old, and you can detect water pollution from those old pipes since they wear out over time. We might not recognize if the problem has occurred, but with this next project, we are replacing all of them (KII III). Participants reported a deficiency in city planning. Most buildings are not built according to city authorities’ plan. Factories and industries are constructed in water catchment areas with no permits and a provision for their waste disposal hence ending up in dumping their waste inappropriately that ends up in water catchment areas and water sources. These illegal developments are hard to regulate and monitor because of political interference. We will not evict a factory near the river because there are so many industries along this river here, and when we look at the analysis we have been doing, we find that some samples taken at night have a different water quality than those taken during the day because we suspect that a lot of things are dumped there at night (KII III). We no longer listen to technical experts; instead, we listen to politicians, which is driving our people to regress to the early 1960s. Let us prevent political involvement, and since there is a political hand somewhere, people construct factories anywhere even in water catchment areas, making it difficult for us to intervene. Hence, we consider our integrity, I believe there is much we can avoid (KII II). River Rwizi which is the main source of drinking water in Mbarara city has been covered by water hyacinth. The weed has covered the biggest part of the river, it has led to reduction in water volume, it traps garbage deposited in the river and makes pumping of water for treat- ment hard and costly. This weed in the River Rwizi called water hyacinth, it collects and traps polythene bags and bot- tles, and their contents slowly leak into the water, thus I believe they are to blame for the polluted water we drink these days, and you may find that some individuals use it the way it is (FGD I). Community member’s practices for safe and quality drinking water Water rationing is employed where some areas receive water at night while others during the day to make sure that at least all communities have water per day. Communities are encour- aged to have overhead tanks to ensure a continuous supply of water. We practice water rationing, in which we decide to give water to one zone during the day and another during the night. We make sure everyone has access to water. We encourage people to get overhead tanks because residents in Mbarara get water via direct lines, so if there is no water on a certain day, overhead tanks might be of help (KII III). Participants revealed that there are laws in place to ensure water catchment areas are pro- tected and not encroached on for human activities. Any developments along water catchment PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 8 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation areas must be evaluated for impact assessment and must receive permission inform of permits that clearly stipulates what activity is going to take place and for how long and their waste dis- posal and environment protection and conservation plan. We have legislation in place, such as the National Environmental Act, which serves as a foun- dation for all environmental concerns, including the preservation of all water resources. We have national wetland, river bank, and lake shores management in place for the preservation of water sources, and as part of our mandate, we aim to engage communities and stakeholders in the conservation of water sources, and the battle is still ongoing (KII IV). Surface water abstraction permit, ground water abstraction permit, water discharge permit, construction permit, there are quite a few and for surface water permit, the main idea is that issuing a permit is to ensure that water is available for all not just some because they all need water so if we allow individuals, and you know individuals are selfish by nature, one can decide to take all the water excluding others so one needs to tell our department based on what they want (KII I). Participants revealed that National water and Sewerage Corporation engages in both inter- nal and external quality control measures to ensure that the drinking water supplied for human consumption is safe and of quality. We sample together and then discuss the results. They also audit, and now we are going to audit so that if one group is not speaking the truth, another group will. I believe that with that level of transparency, we can perform those system checks and, at the end of the day, compare the data (KII III). Community members have put in place security controls around water sources, the water sources are faced to stop animals from drinking from sources meant to supply drinking water for human consumption, they encourage community members not to send young children to fetch water hence minimising defecating and swimming in drinking water sources. We have attempted to secure the water in our wetland so that when children go to get water from there, they do not defecate in the area surrounding our water and that cows that go to our water source do not go close the water that we fetch for drinking, which is what we do with our wetland. We make sure that when someone fetches, she/he ensures that the tap is properly closed and that children do not go there to play on the borehole/tap so that we can protect it (FGD III). Water service providers ensure that drinking water supplied for use by communities is treated and is safe and of recommended quality of drinking water for supply to communities. The communities boil the water, sieve it, cover it, use clean water collection vessels, allow it to sediment, and use the supernatant and sometimes use safeguard to treat their drinking water before use. For safety, I believe that for national water and sewerage cooperation, safety is maintained through treatment to address specific issues such as microbiology nuclei and all that, as well as disinfection, so the water is disinfected, and then there is the aspect of filtration to remove these other suspended materials, so there is this deliberate effort to treat the water so that it meets the standards that we require. So there is an effort in terms of personnel and resources, PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 9 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation and the entire site has the idea in mind that this water must be treated to this acceptable stan- dard, so the safety is guaranteed (KII IV). We boil water, especially when it comes to drinking, and then we filter it to limit the level of contamination since after boiling, there is some dirt that remains on the bottom (FGD II). Communities have a vast number of alternative sources of drinking water that range from, open wells, protected springs, boreholes, gravity flow, and tap water and rain harvest tanks. They use an alternative source depending on what they want to use the water for and the avail- ability, accessibility, safety and quality of water. We can choose to utilize rain water since we have been spoiled by taps, and as a result, one can build a house without a gutter. Collecting rain water would also help, but the problem is that once collected, one person dips a cup to fetch water and another person brings a jug, mak- ing it unsafe, but it would be one of the best ways because that water is free, and I feel like if I could get a crest tank and put it on my house to harvest clean water, could it be a solution and once I get it, I get period to wash it and by the time water enters into that tank I make sure there is a sieve to prevent large things from entering the tank, and once the water is finished, you cleanse the tank and boil water from that tank for drinking (FGD II). Community members have resorted to putting to use the overgrowing garbage and plastic volume to use. Garbage and plastics are being collected and used to make brickets. Plastics are collected and recycled into other accessories like beads and mats. Organic and inorganic plastics are separated for possible recycling since we cannot do away with this because as a country, we still need jobs, so how can we have these jobs without dam- aging the environment and River Rwizi (FGD II). For example, we make bricks from garbage, so if a person is aware that garbage is important and will benefit from it, he or she will take responsibility, and the responsible companies will come and pick it up. However, people should also be aware of which rubbish has value so that one knows the exact amount he is likely to receive (FGD VI). Community member’s perceived solutions to ensure safe and quality of water Participants believe that engaging stakeholders in the catchment area on water source protec- tion guidelines and the need to alert communities/stakeholders in case of contamination, and enforcing laws through political leaders can help ensure safe and quality water to the communities. When the catchment is not proper, all of these will come down, so when we go to individuals, we must be aware that when certain things are not done correctly, one suffers, and when someone is not aware that chemicals for agriculture once sprayed, such chemicals will come back to me, so those people are not aware. So that is the information we are talking about, the safety of water and how this safety is important to all of us, not just you and me, but all of us, and even the people outside there, because otherwise, we would be treating the symptoms rather than the core cause of the problem (KII I). Participants believe that community sensitisation on the need for safe and quality drinking water will help greatly in changing the mind set of communities. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 10 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation Sensitization, that is it, the community may not be aware of pollutants rods and, major con- taminants of water, so they need to come up with an approach of sensitization in our commu- nities about the dangers of drinking contaminated water by informing the communities that if you use contaminated water, it affects them like getting water borne diseases, so that they can come to understand that they must protect the water sources (KII I). When you go to the villages in these town councils, you will see what I mean, you will see everywhere is garbage and so on, so sometimes we apply law and sometimes you can find a leader in the village does not have a latrine, does not have anything to use for sanitation, you find somebody’s compound is full of funny things and is a leader, so those are the things, but we will keep on community sensitization. Even if we lack resources, we will continue to edu- cate the community, and those who wish to spread the word will go out and improve their sur- roundings (FGD II). There is need to educate communities to create awareness and ensure that there are buffer zones in water catchment areas. Awareness has been raised through educating communities and limiting development around waterways. NEMA regulates all projects where it is not possible; NEMA has not authorized any developments beside water resources, and when they are permitted, limitations are imposed. When we look at the River Rwizi, we tried to engage a number of stakeholders, including encroachers along the river basins, so that they can vacate and the buffer is well pro- tected, and when the buffer is well protected, it means that the water is fine. We have also engaged industrialists in managing the effluents coming from their industries, so that the water released from the manufacturing processes is treated before it is discharged into the river, and even before it is discharged into the river (KII IV). Participants believe that following guidelines set to ensure safe and quality drinking water is key to in maintaining safe and quality drinking water. We must adhere to the drinking water guidelines. What is the distance from the latrine to the water source, sometimes people come and start cultivating near the water sources, so some meters are required from the water source, and even the water source itself has some meters’ standard like 50 by 50 so that if there is run off, it should not filtrate easily into the water source, so after putting the measures at the source, we know that that source is ok (KII II). Participants believe that most factors that affect the safety and quality of water in Mbarara city are due to the behaviour and mind set of community members. National water and sewer- age cooperation tries its level best to supply safe and quality drinking water at the recom- mended standard for home use not at the bottled water standard. Most communities access this resource through illegal connections that makes the cost of supply and maintenance expensive for the service provider thereby making it expensive for the consumer. Communities are aware that the water available for use needs to be boiled before drinking it but for personal reasons like lack of firewood, ignorance and time, they resort to drinking it half boiled or unboiled. City authorities have put in place provisions for waste disposal but communities continue to dispose waste as they wish. We have a lot of pollution from industrial developments, as well as problems with improper waste management, all of which end in our waters. Leaving that aside, we have a number of illegal construction and illegal activities that are taking place outside of the 100-meter zone PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 11 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation that is the protection zone along the river, thus ending up in the river, so the quality of water is not up to date due to poor waste disposal, population growth, and direct influent discharge from industries that is not even treated, and all of that ends up in our water sources (KII IV). Individuals illegally connect to the water supply, and we have many such incidents in Mbar- ara, largely from private plumbers. Connection is also important since someone will connect you where the settlement is and where they do not encourage consumers to be connected, but you will find individuals connecting illegally (KI III). We would be boiling the water, but most of the time we do not have enough money to buy charcoal because it is expensive, and sometimes you can have food but you cannot cook because you do not have charcoal, so there is no charcoal to boil water, so most people drink it without boiling it, which has caused typhoid infections. People have been talking about some- one who just grabs a cup, pours directly from a jerrican, and drinks (FGD I). To ensure safe and quality drinking water, there is need for collaboration between commu- nities and water service providers. The community needs to be engaged and encouraged to participate in activities aimed at ensuring stable and sustainable supply and use of safe and quality drinking water. There is need to set up community water committees, catchment man- agement committees and school sanitation committees through which information pertaining the use and maintenance of safe and quality water and practices to ensure proper use and maintenance of safe and quality water are shared between communities and water service providers. we normally have the water user committees to check whoever gets water but they also have a challenge themselves. There is need to make committee on sanitation to ensure things like toi- lets, hand washing facility, abcd are introduced in the community so that they can reduce the risks and if someone comes from the toilet, there should be a jerrican and soap on the toilet to wash hands so those are the measures we are putting up but I told you that is a behavioural change strategy with its many challenges (K II IV). We have not gone to the household level, but we have managed to get to catchment organiza- tions, which are made up of many stakeholders, including local governments, so from local governments, we establish a committee of that catchment organization called the catchment management committee. The organization is comprised of structures that comprise the execu- tive arm, which meets to address issues. There are many entities in that catchment manage- ment committee, such as district local government, which brings on board district water officials, chief administrative officers, and LC 5 (Local council) chairpersons (KII I). Participants believe treating drinking water from drinking water sources in Mbarara city at supply system level with chemicals and at home with safeguard will greatly help in improving the quality and safety. This could be by providing chemicals to help in home treatment of drinking water and general treatment of water before it is supplied for use on the taps. There is need to mechanically remove the water weeds/plants, clear bushes around water sources and regularly cleaning the open wells. Water, in my opinion, should be collected in tanks and then purified at various treatment sta- tions before being released. But my heart continues telling me that maybe National Water and Sewerage Corporation obtains water from a source and store it in tanks, but they don’t treat it before distributing it, or the tanks aren’t washed on a regular basis, or the treatment PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 12 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation they use is insufficient. You are aware that in Uganda, less treatment can be used than is rec- ommended, which cannot be sufficient for effective water treatment (FGD II). For safety, I believe that national water and sewerage cooperation maintains safety through water treatment to address specific issues such as microbiological nuclei. Water is disinfected, and there is also the issue of filtration to remove these other suspended things, so there is a concerted effort to clean the water so that it meets the acceptable standards we require. There is constant monitoring to verify that these criteria are met, including the availability of per- sons and resources to ensure that this water is treated to appropriate levels and that safety is ensured (KII I). Participants suggest that water service providers should ensure that the water they supply is safe and of quality. They should put provisions in place to ensure that the quality is maintained throughout the supply chain by routine monitoring and surveillance and ensuring that any pit- falls are addressed in a timely manner. I believe that National Water and Sewerage Company should take the time to walk around and observe what is going on, not only to appear to collect their money but also to learn about the kind of water they supply. There is a need for communities to set aside time to meet with individuals and discuss what to do, like we are doing now, and we also know if the problem is here or there. They do not have that time they only come when they want their money but I think giving time to people is also crucial. They should visit different locations since the water may be polluted in some locations but clean in others (FGD II). Participants believe that so many factors contribute to ensuring safe and quality drinking water supply. It is these same factors if not properly addressed that will lead to deterioration of water quality. By engaging stakeholders, it is a great step towards provision and sustaining clean, safe and quality water. Stakeholders should provide community with feasible solutions, keep the process in check and hence the safety and quality of drinking water is achieved and maintained. We are executing a project in Rubanda, Kabale, Ntungamo, and Rukiga where they are attempting to engage with people to preserve water in their farmland in order to maintain it for a longer period of time. But when it is running, it runs a way with soil and they see that some diseases are becoming prevalent and they start asking themselves that they never used to get these diseases so where are they coming from not knowing that it is due to mishandling some aspects of the environment such as hormonal birth control measures and when we shared those things they understood (KII I). We have national wetland, river bank, lake beaches management in place for water source protection, and as part of our mandate, we attempt to engage communities and stakeholders in water source conservation and protection, and the struggle is still ongoing (KII III). Discussion This study explored community perceptions and practices about drinking water quality and safety from various water sources in Mbarara, Uganda. We wanted to know what community members thought about the quality of water drawn from drinking water sources, what is responsible for the quality, what they do to ensure the drinking water is safe and of good PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 13 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation quality, and what possible solutions there are to ensure the water is safe and of good quality. The findings show that populations in Mbarara, south western Uganda, regard the quality of drinking water drawn for use as poor, dirty, tastes salty, and is generally unsafe for human use, as well as being limited in supply to communities. Community member’s perceptions of the quality and safety of drinking water Based on the color, taste, and presence of physical pollutants, community members perceive the safety and quality of drinking water to be poor, dirty, and salty. This perspective was ech- oed by members of the community and key informants. They believe that the safety and qual- ity of drinking water is poor and that it does not meet established standards for human consumption, yet they continue to consume it since it is the only water available to them. Simi- larly, a study by Apecu and co-authors on quality of water sources in South-western Uganda using the compartment bag test (CBT) found out that most of the water sources in the study areas were not fit for human consumption without prior treatment [25]. This is odd given that this is a city neighbourhood where social services should be of higher quality. This however is not unique to Mbarara city alone since the World Health Organization estimates that 2 billion people lack safely managed services, including 1.2 billion with basic services, 282 million with limited services, 367 million using unimproved sources, and 122 million drinking surface water, when the United Nations Sustainable Development Goal6 is to ensure universal access to water and sanitation by 2030 [26]. In addition, without point-of-use treatment systems, at least four billion people worldwide do not have access to clean drinking water or are under the impression that it is unsafe to drink [27]. This is owing to increased water demand, reduced water supplies, and increased water pollution as a result of tremendous population and eco- nomic expansion. In many underdeveloped nations’ urban areas, badly polluted little water sources are widely used [28]. Contamination concerns are considerable in urban areas due to increased population. Yet, because most populations in urban areas cannot afford the expense of a treated water system and lack access to infrastructure, their sense of quality relies on mod- est water systems or various sources to supply their drinking water demands [29]. In addition, perceptions of worsening water quality have been observed all across the world, in both rich and developing nations, and ’Thousands have survived without love, not one without decent water quality [30]. It should be emphasized that the types, magnitudes, and extents of water quality concerns vary from country to country, and even from region to region within a coun- try. This might be the result of uneven growth, accessibility, and water demands. Issues may be resolved via trust, political will, and social will, albeit the methods vary depending on the region of the country. It should be noted that; trust enables water delivery businesses to achieve both social and commercial benefits [31]. Community member’s perceived factors responsible for safety and quality of water According to the findings of this study, the poor drinking water quality in Mbarara city is mostly attributable to improper waste management, poor water treatment, poor system main- tenance, political interference, population increase, and water hyacinth. Some variables do not remain constant throughout time. These factors are not static, but instead vary over time. Some of these factors are human made while others are beyond the communities’ control. Issues such as flooding and water cost fluctuate between wet and dry seasons; changes in the water supply; changes in the community’s/family’s ability to maintain quality, household income, and level of awareness within a given community [32]. Similarly, to our study, the PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 14 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation decline in water quality is caused by increased demand for water, reduced water supplies, and increased water pollution as a result of dramatic population and economic growth [33]. This is due to the discharge of essential pollutants from anthropogenic activities such as industrial applications (solid/liquid wastes, chemical compounds, mining activities, spills, and leaks), urban development (municipal wastes, land use practices, and others), and agricultural prac- tices (pesticides and fertilizers) that affect the safety and quality of water in urban communities [34]. There are other key pollutants emitted by natural processes that contribute to climate change, natural catastrophes, geological causes, soil matrix, and hyporheic exchange in the aquatic environment, all of which might have a detrimental impact (e.g. Endocrine disrup- tions, DNA damage, cancerogenicity). These elements, together with rising temperatures, accelerated remobilisation processes, and hormone pollution, have a greater impact and may disrupt natural environmental equilibrium. It should be noted that, as indicated in this study, greater population expansion frequently coincides with the demand for more food and food production, forcing communities to encroach on water catchment areas for agriculture [35]. Because of the requirement for improved yields on a short plot of land, fertilizers and crop insecticides are used indiscriminately. These changes in land-use/land-cover (LULC) pattern degrade water quality. This is due to the interdependence of population and economic growth, as well as water consumption, resources, and pollution, all of which contribute to water short- age [36]. Moreover, population growth leads to deforestation to support agricultural develop- ment and urban expansion in Mbarara city, necessitating the need for water quality protection to meet urgent human requirements while also ensuring the long-term quality of water resources. There is a lot of garbage produced in the midst of economic issues, making it hard to properly dispose of or pay for proper disposal through structured public services, resulting in waste buildup. Occasionally garbage is dumped in available water sources or catchment areas. This not only affects the quality of drinking water, but it also raises the cost of treated water since more sophisticated procedures are used to assure that the water supplied to com- munities is treated and of the required standard. A study to investigate the impact of drinking water quality and sanitation on child health: Evidence from rural Ethiopia demonstrated that uncontaminated stored drinking water and safe child stool disposal are related with 18 and 20 percentage point decreases in child diarrhoea rates, respectively [37]. Community member’s practices for safe and quality drinking water To ensure that drinking water from sources in Mbarara city is safe and of quality for use, as well as available and accessible in quantity, service providers use a holistic approach, water rationing, changing chemicals as often as possible depending on the quality of water available for treatment, and the water treatment process is quality controlled internally at National Water and Sewerage Corporation facility treatment centers and externally at Uganda National Bureau of Standards. National environment Management Authority issues permits for any developments that would result in waste to be dumped in River Rwizi or any developments close to the water catchment areas. This can be traced to the fact that, National water and sew- erage cooperation, through their service accelerated program has created awareness for the need and maintenance of safe and quality water through radio talk shows, school health sanita- tion program and in churches. The communities, on the other hand, ensure that bushes are cleared around water sources, that adults and children in company of adults have access to these sources, that overhead tanks are installed and maintained, and that drinking water is boiled. This is crucial in increasing the availability, accessibility, and appropriate quantity of quality and safe drinking water since they are a primary measure for preventing different water-borne infections, poisoning, disease outbreaks, and human deaths in urban settings PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 15 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation [38]. A healthy population is critical for health and long-term socioeconomic growth. Clean drinking water is a crucial component of Primary health care and plays an important role in poverty alleviation, hence boosting economic growth [24]. Due to the exponential growth in water demand and the decrease in usable freshwater due to various climate, environmental, and anthropogenic events, rain water harvesting has become a useful practice because it is inexpensive and low risk if the roof catchment, collection system, and storage are well main- tained [39]. Similar to the findings of this study, there is a need to better understand social fac- tors such as governance and increased understanding of diverse physical and social influences that lead to a more comprehensive understanding, knowledge, and need for clean, safe, and quality water, as well as water security, which is defined as a reliable and adequate supply of safe and quality water to support humans and ecosystems at all times [40]. Furthermore, there is a need to raise awareness about the need of clean, safe, and high-quality drinking water, as well as the necessity for other government stakeholders to work together to enhance water quality for improved health [41]. As a result, there should be a continuous extensive water quality monitoring program of drinking water sources across urban areas and their adjacent settings to guarantee population health and environmental balance [42]. However, this requires policymakers and managers to use Artificial Neural Networks (ANNs) and risk analysis techniques to predict water quality because such predictions indicate the level of risk (low, moderate, or high) to the inhabitants, allowing for the implementation of preven- tive measures to avoid illness or disease outbreaks. This can be achieved through engaging in socio-hydrological research and data analysis to help improve the current understanding and management of the quantity and quality society dynamics for drinking water quality and safety [40]. Community member’s perceived solutions for safe and quality of water The participants in this study feel that stakeholder involvement, community awareness, estab- lishing catchment plan rules and regulations, water treatment and maintenance, surveillance, and monitoring might all assist to improve and maintain the quality and safety of drinking water in Mbarara. This is due to an increasing number of people turning to alternative sources of drinking water, such as rainwater harvesting, to reduce their environmental footprint, because rainwater harvesting (RWH), while not economically feasible, provides protection against damage caused by increasing precipitation frequency and intensity [43]. Similarly, Anjana and colleagues in India advocated training people on drinking water treatment meth- ods, sanitation, and hand washing habits since participants believed their drinking water was pure and didn’t need any further treatment [44]. Furthermore, an Ochilova and colleagues study recommended the need for rational use and protection of water resources, as well as ensuring and guaranteeing citizens’ right to a favorable natural environment, as well as helping to protect land, subsoil, forests, flora and fauna, atmospheric air, natural resources, and improving healthy family life [45]. There is a need to connect rural and urban areas. The two communities are mutually reliant. Water streams come from rural communities to feed water to urban cities; food production is mostly done in rural communities but is consumed in both rural and urban areas. The rural people should be given policy attention to the ecosystem ser- vices that rural areas provide, and the rural area’s ecology should be conserved for long-term service delivery, reducing the need to farm in water catchment areas that exist in already over- crowded urban areas [46]. Most importantly, there is a need to invest in implementing sustain- able technologies for future water supply and sanitation because the amount of time and money spent by developed, developing, and underdeveloped countries on water investments, operation, and maintenance has changed dramatically in recent decades [47]. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 16 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation Strengths and limitations of the study The findings of this study represent the perspectives and opinions of community members and stakeholders in Mbarara City’s water provision and maintenance. The study’s main strength is the unanimity in their thoughts and beliefs. Our capacity to interact with communi- ties and stakeholders in water service supply to investigate their perceptions and practices about the safety and quality of drinking water is our strength. Key informants in this study were water service providers; this may have worked against us since they were afraid to completely voice their ideas and opinions for fear of acting against the expectations of their employers. Nonetheless, we ensured all of our participants of anonymity and confidentiality during the informed consent procedure. We acknowledge that this study presents views and opinions of communities and stakeholders in the water service provision and maintenance in Mbarara city. Conclusion Residents in Mbarara perceive the quality of drinking water drawn for use as not good, dirty and salty, and generally unfit for human consumption and limited in supply to communities. Increased population expansion and accompanying human activities, political intervention, flooding, and deficiencies in water treatment, supply, and management are all contributing to poor quality of drinking water in Mbarara city. The service providers use water rationing, offer permits for developments in the city and most importantly in water catchment areas, the water is treated and the water supply system is quality controlled both internally and externally, water sources are protected from contamination by clearing bushes and fencing, and alterna- tive sources are used to supply drinking water in the event of suspected contamination. Perspective and recommendation We recommend a comprehensive approach to the provision, use, and management of drink- ing water sources. Policymakers and stakeholders should collaborate to increase knowledge, sensitization, and practices aimed at providing, using, and maintaining safe and high-quality drinking water from drinking water sources in Mbarara, south-western Uganda. Acknowledgments We thank all those who participated in this research project. We acknowledge the study partic- ipants and the Alex Tumusiime (AT) and Patience Nabaasa the study Research Assistants and Owokuhaisa Judith(OJ) who read through the transcripts and coded We acknowledge the reviewers who are going to review and provide constructive comments that will help perfect this manuscript. Author Contributions Conceptualization: Abaasa N. Catherine, Savino Ayesiga, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. Data curation: Abaasa N. Catherine, Savino Ayesiga, Imelda K. Tamwesigire. Formal analysis: Abaasa N. Catherine, Savino Ayesiga. Funding acquisition: Abaasa N. Catherine. Investigation: Abaasa N. Catherine, Savino Ayesiga, Imelda K. Tamwesigire. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 17 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation Methodology: Abaasa N. Catherine, Savino Ayesiga, Godfrey Zari Rukundo, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. Project administration: Abaasa N. Catherine, Savino Ayesiga, Julius B. Lejju, Frederick Byar- ugaba, Imelda K. Tamwesigire. Resources: Abaasa N. Catherine. Software: Abaasa N. Catherine. Supervision: Godfrey Zari Rukundo, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. Validation: Abaasa N. Catherine, Savino Ayesiga, Godfrey Zari Rukundo, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. Visualization: Abaasa N. Catherine, Savino Ayesiga, Godfrey Zari Rukundo, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. Writing – original draft: Abaasa N. Catherine. Writing – review & editing: Abaasa N. Catherine, Savino Ayesiga, Godfrey Zari Rukundo, Julius B. Lejju, Frederick Byarugaba, Imelda K. Tamwesigire. References 1. Lin J, Ganesh AJIjoehr. Water quality indicators: bacteria, coliphages, enteric viruses. 2013; 23(6):484– 506. 2. Morgan MJ, Halstrom S, Wylie JT, Walsh T, Kaksonen AH, Sutton D, et al. Characterization of a drink- ing water distribution pipeline terminally colonized by Naegleria fowleri. Environmental science & tech- nology. 2016; 50(6):2890–8. https://doi.org/10.1021/acs.est.5b05657 PMID: 26853055 3. ESCAP U. Sustainable Development Goal 6: ensure availability and sustainable management of water and sanitation for all. 2016. 4. Dinka MOJWcoauw. Safe drinking water: concepts, benefits, principles and standards. 2018;163. 5. Alhamlan FS, Al-Qahtani AA, Al-Ahdal MNA. Recommended advanced techniques for waterborne pathogen detection in developing countries. The Journal of Infection in Developing Countries. 2015; 9 (02):128–35. https://doi.org/10.3855/jidc.6101 PMID: 25699486 6. Johnson DM, Hokanson DR, Zhang Q, Czupinski KD, Tang J. Feasibility of water purification technol- ogy in rural areas of developing countries. J Environ Manage. 2008; 88(3):416–27. https://doi.org/10. 1016/j.jenvman.2007.03.002 PMID: 17459569. 7. Daly SW, Lowe J, Hornsby GM, Harris ARJJoW, Health. Multiple water source use in low-and middle- income countries: a systematic review. 2021; 19(3):370–92. 8. Kumpel E, Peletz R, Bonham M, Khush R. Assessing drinking water quality and water safety manage- ment in sub-Saharan Africa using regulated monitoring data. Environmental science & technology. 2016; 50(20):10869–76. https://doi.org/10.1021/acs.est.6b02707 PMID: 27559754 9. Unicef. Progress on drinking water, sanitation and hygiene. 2017. 10. Mooney SO’Dwyer J, Lavallee S, Hynds P. Private groundwater contamination and extreme weather events: The role of demographics, experience and cognitive factors on risk perceptions of Irish private well users. Science of the Total Environment. 2021; 784:147118. 11. Lucier KJ, Schuster-Wallace CJ, Skead D, Skead K, Dickson-Anderson SEJBPH. “Is there anything good about a water advisory?”: an exploration of the consequences of drinking water advisories in an indigenous community. 2020; 20(1):1–12. 12. Ramya N, Reddy MM, Kamath PB. Community Perception vs. Biochemical Confirmation: A Mixed- Methods Study on Water Quality From South India. Cureus. 2021; 13(10). 13. Liu H, Sun S, Fang C, van den Berg P, Dane G, Li J, et al. Public perceptions of physical and virtual water in China. Science of the Total Environment. 2022; 812:151460. https://doi.org/10.1016/j. scitotenv.2021.151460 PMID: 34762958 14. Organization WH. Guidelines for drinking-water quality: first addendum to the fourth edition. 2017. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 18 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation 15. Benameur T, Benameur N, Saidi N, Tartag S, Sayad H, Agouni AJEM, et al. Predicting factors of public awareness and perception about the quality, safety of drinking water, and pollution incidents. 2022; 194 (1):22. 16. Delpla I, Legay C, Proulx F, Rodriguez MJJSotTE. Perception of tap water quality: Assessment of the factors modifying the links between satisfaction and water consumption behavior. 2020; 722:137786. https://doi.org/10.1016/j.scitotenv.2020.137786 PMID: 32208246 17. Bitew BD, Gete YK, Biks GA, Adafrie TT. Barriers and Enabling Factors Associated with the Implemen- tation of Household Solar Water Disinfection: A Qualitative Study in Northwest Ethiopia. The American Journal of Tropical Medicine and Hygiene. 2020; 102(2):458. https://doi.org/10.4269/ajtmh.18-0412 PMID: 31837131 18. Ondieki J, Akunga D, Warutere P, Kenyanya OJH. Socio-demographic and water handling practices affecting quality of household drinking water in Kisii Town, Kisii County, Kenya. 2022; 8(5):e09419. https://doi.org/10.1016/j.heliyon.2022.e09419 PMID: 35600434 19. Okumah M, Yeboah AS, Bonyah SKJLUP. What matters most? Stakeholders’ perceptions of river water quality. 2020; 99:104824. 20. Peres MR, Ebdon J, Purnell S, Taylor HJIJoH, Health E. Potential microbial transmission pathways in rural communities using multiple alternative water sources in semi-arid Brazil. 2020; 224:113431. 21. Uganda Bureau of Statistics U. Population of Mbarara. online: https://all-populations.com/en/ug/ population-of-mbarara.html;2022; 2022. 22. MDL G. Water and Sanitation. Https://ww.mbarara.go.ug/services/water-and-sanitation. online2022. 23. Gachango F, Andersen L, Pedersen S, editors. Danish farmers’ perception of water quality, nutrient reduction measures and their implementation strategy. International Conference on Sustainable Water Resources Management; 2015. 24. Kaur K, Singh GJND. Access to Safe Drinking Water and Economic Development: A Comparative Anal- ysis of Developed and Developing Countries.76. 25. Apecu RO, Ampaire L, Mulogo EM, Bagenda FN, Traore A, Potgieter N, et al. Quality of water sources in Southwestern Uganda using the compartment bag test (CBT): a cross-sectional descriptive study. 2019; 9(4):683–93. 26. Organization WH. Progress on household drinking water, sanitation and hygiene 2000–2020: five years into the SDGs. 2021. 27. Biswas AK, Tortajada CJIJoWRD. Water quality management: a globally neglected issue. Taylor & Francis; 2019. p. 913–6. 28. Zhao Y, Li MJIJoER, Health P. Effect of water-saving society policy on water consumption in the cities of china: a propensity score matching analysis. 2020; 17(21):8171. 29. Howard G, Bartram J, Luyima P. Small water supplies in urban areas of developing countries. Providing 30. 31. Safe Drinking Water in Small Systems: Routledge; 2019. p. 83–93. Inwald JF, Bruine de Bruin Wn, Yaggi M, A´ rvai JJES, Technology. Public Concern about Water Safety, Weather, and Climate: Insights from the World Risk Poll. 2023; 57(5):2075–83. Jabłoński A, Jabłoński MJIJoER, Health P. Business models in water supply companies—Key implica- tions of trust. 2020; 17(8):2770. 32. Price H, Adams E, Quilliam RSJSotte. The difference a day can make: The temporal dynamics of drink- ing water access and quality in urban slums. 2019; 671:818–26. 33. Juma DW, Wang H, Li FJES, Research P. Impacts of population growth and economic development on water quality of a lake: case study of Lake Victoria Kenya water. 2014; 21:5737–46. 34. Akhtar N, Syakir Ishak MI, Bhawani SA, Umar KJW. Various natural and anthropogenic factors respon- sible for water quality degradation: A review. 2021; 13(19):2660. 35. Njagi DM, Routh J, Odhiambo M, Luo C, Basapuram LG, Olago D, et al. A century of human-induced environmental changes and the combined roles of nutrients and land use in Lake Victoria catchment on eutrophication. 2022; 835:155425. 36. Naeem M, Farid HU, Madni MA, Ahsen R, Khan ZM, Dilshad A, et al. Remotely sensed image interpre- tation for assessment of land use land cover changes and settlement impact on allocated irrigation water in Multan, Pakistan. 2022; 194(2):98. https://doi.org/10.1007/s10661-021-09732-5 PMID: 35031930 37. Usman MA, Gerber N, von Braun JJTJoDS. The impact of drinking water quality and sanitation on child health: Evidence from rural Ethiopia. 2019; 55(10):2193–211. 38. Musoke D, Ndejjo R, Halage AA, Kasasa S, Ssempebwa JC, Carpenter DOJJoe, et al. Drinking water supply, sanitation, and hygiene promotion interventions in two slum communities in Central Uganda. 2018;2018. https://doi.org/10.1155/2018/3710120 PMID: 29623096 PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 19 / 20 PLOS WATERQuality and safety of drinking water; A qualitative investigation 39. Senevirathna S, Ramzan S, Morgan JJPS, Protection E. A sustainable and fully automated process to treat stored rainwater to meet drinking water quality guidelines. 2019; 130:190–6. 40. Gunda T, Hess D, Hornberger GM, Worland SJWS. Water security in practice: The quantity-quality- society nexus. 2019; 6:100022. 41. Ravindra K, Mor S, Pinnaka VLJES, Research P. Water uses, treatment, and sanitation practices in rural areas of Chandigarh and its relation with waterborne diseases. 2019; 26:19512–22. 42. Adeola Fashae O, Abiola Ayorinde H, Oludapo Olusola A, Oluseyi Obateru RJAWS. Landuse and sur- face water quality in an emerging urban city. 2019; 9:1–12. 43. Hofman-Caris R, Bertelkamp C, de Waal L, van den Brand T, Hofman J, van der Aa R, et al. Rainwater harvesting for drinking water production: a sustainable and cost-effective solution in the Netherlands? 2019; 11(3):511. 44. Kuberan A, Singh AK, Kasav JB, Prasad S, Surapaneni KM, Upadhyay V, et al. Water and sanitation hygiene knowledge, attitude, and practices among household members living in rural setting of India. 2015; 6(Suppl 1):S69. https://doi.org/10.4103/0976-9668.166090 PMID: 26604623 45. Ochilova NR, Muratova GS, Karshieva DRJCAJoM, Science N. The Importance of Water Quality and Quantity in Strengthening the Health and Living Conditions of the Population. 2021; 2(5):399–402. 46. Gebre T, Gebremedhin BJGe, conservation. The mutual benefits of promoting rural-urban interdepen- dence through linked ecosystem services. 2019; 20:e00707. 47. Palansooriya KN, Yang Y, Tsang YF, Sarkar B, Hou D, Cao X, et al. Occurrence of contaminants in drinking water sources and the potential of biochar for water quality improvement: A review. 2020; 50 (6):549–611. PLOS Water \| https://doi.org/10.1371/journal.pwat.0000075 May 30, 2023 20 / 20 PLOS WATER
10.2196_42978	JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al Original Paper Long-Term Impact of a Smartphone App on Prescriber Adherence to Antibiotic Guidelines for Adult Patients With Community-Acquired Pneumonia: Interrupted Time-Series Study Chang Ho Yoon1,2; Imogen Nolan2; Gayl Humphrey3; Eamon J Duffy2; Mark G Thomas4; Stephen R Ritchie4 1Big Data Institute, Oxford, United Kingdom 2Infectious Diseases Department, Auckland City Hospital, Auckland, New Zealand 3National Institute for Health Innovation, University of Auckland, Auckland, New Zealand 4School of Medical Sciences, University of Auckland, Auckland, New Zealand Corresponding Author: Chang Ho Yoon Big Data Institute Old Campus Road Oxford, OX3 7LF United Kingdom Phone: 44 7925818791 Email: [email protected] Abstract Background: Mobile health platforms like smartphone apps that provide clinical guidelines are ubiquitous, yet their long-term impact on guideline adherence remains unclear. In 2016, an antibiotic guidelines app, called SCRIPT, was introduced in Auckland City Hospital, New Zealand, to provide local antibiotic guidelines to clinicians on their smartphones. Objective: We aimed to assess whether the provision of antibiotic guidelines in a smartphone app resulted in sustained changes in antibiotic guideline adherence by prescribers. Methods: We analyzed antibiotic guideline adherence rates during the first 24 hours of hospital admission in adults diagnosed with community-acquired pneumonia using an interrupted time-series study with 3 distinct periods post app implementation (ie, 3, 12, and 24 months). Results: Adherence increased from 23% (46/200) at baseline to 31% (73/237) at 3 months and 34% (69/200) at 12 months, reducing to 31% (62/200) at 24 months post app implementation (P=.07 vs baseline). However, increased adherence was sustained in patients with pulmonary consolidation on x-ray (9/63, 14% at baseline; 23/77, 30% after 3 months; 32/92, 35% after 12 month; and 32/102, 31% after 24 months; P=.04 vs baseline). Conclusions: An antibiotic guidelines app increased overall adherence, but this was not sustained. In patients with pulmonary consolidation, the increased adherence was sustained. (J Med Internet Res 2023;25:e42978) doi: 10.2196/42978 KEYWORDS app; antimicrobial stewardship; antibiotic adherence; community; pneumonia; smartphone; mobile health; mHealth; antibiotic; behavior; adults; diagnosis; pulmonary; patient Introduction Antibiotic stewardship programs in hospitals and community clinics strive to improve rates of appropriate antibiotic prescribing through a wide variety of methods (from clinical decision support tools to educational sessions) both to optimize the treatment of patients with bacterial infections and to reduce inappropriate antibiotic prescribing [1]. Greater adherence to antibiotic guidelines (ie, prescription of antibiotics consistent with guidelines) is associated with better treatment outcomes and reduced antibiotic resistance [1,2], yet rates of adherence remain low [3-7]. Despite the ubiquity and promise of mobile health (mHealth) platforms like smartphone apps to overcome some of the causes of low adherence, such as limited access to guidelines, the long-term impact of mHealth apps on guideline adherence remains unclear [8]. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al A small number of studies have suggested that apps displaying antibiotic guidelines improve antibiotic prescribing behavior in the short term [8], while the only study to have measured adherence beyond 12 months post app implementation suggested that any improvements are not necessarily sustained [9]. Therefore, the long-term influence of such apps requires further investigation, with implications for their cost-benefit analysis and long-term utility in antibiotic stewardship programs. In 2016, we developed an antibiotic guidelines app, “SCRIPT,” at Auckland City Hospital (ACH) in New Zealand, which displayed antibiotic guidelines for community-acquired pneumonia (CAP) and urinary tract infection to prescribing clinicians on their smartphones and observed improvement in adherence for patients with CAP but not for patients with urinary tract infection [6]. However, this study was limited by only 3 months of follow-up and a short lead-in time for app adoption. Since then, SCRIPT has increased in popularity among local prescribers and become a standard part of the prescribing repertoire, evidenced by 5600 unique users accounting for 600,000 app sessions in 2020 and over 700,000 in 2021 [10]. The app is freely available (as “SCRIPT ADHB” via Google Play and App Store) and sports a simple, user-friendly interface (Figure 1). Given SCRIPT’s widespread adoption in ACH, we aimed to assess its long-term impact on prescriber adherence to antibiotic guidelines in patients with CAP at 3, 12, and 24 months after the SCRIPT guidelines were made available. Figure 1. Succession of screenshots (left to right) from the SCRIPT smartphone app, which displays the user interface in accessing antibiotic guidelines for low-risk community-acquired pneumonia as defined by a CURB-65 score of 0-1 (middle screenshot). Methods Aims and Study Setting We performed an interrupted time-series study to test the hypothesis that the provision of the SCRIPT app would increase prescriber adherence to antibiotic guidelines for hospitalized adult patients with CAP. We further hypothesized that adherence to antibiotic guidelines would be higher in cases with chest x-ray evidence of CAP than in cases without chest x-ray evidence of CAP (because the diagnosis of CAP is questionable in cases without chest x-ray evidence [11,12]). The initial impact of SCRIPT’s implementation on adherence to local ACH antibiotic guidelines has been described previously [6]. In the first 2 weeks of the intervention period, educational sessions, posters, and intranet advertisements were employed to socialize the app and facilitate its uptake by ACH clinicians. Thereafter, the app was promoted periodically in newsletters and posters and at each orientation for new rotations of junior doctors. ACH has a multifaceted approach to antimicrobial stewardship (including formulary management, regular audit and feedback, expert consultation services, and surveillance of antibiotic use), which continued unaltered throughout the study period. The hospital antibiotic guidelines remained on the hospital intranet. No other interventions impacting CAP management were introduced during the study period. Study Cohort We retrospectively collected data during 4 periods, as follows: “baseline” pre-app implementation (January 1 to May 31, 2016); “immediate” post-app implementation (June 1 to August 31, 2016); 12-month post-app implementation (June 1 to October 31, 2017); and 24-month post-app implementation (June 1 to October 31, 2018). Adult patients (aged ≥18 years) admitted to ACH for ≥4 hours with a discharge diagnosis of CAP (International Classification of Diseases-10 codes: J10-18 and J22) were included. Patients were excluded if they were not diagnosed with CAP during the first 24 hours of admission; had incorrectly coded diagnoses (eg, “empyema”); or were transferred from another secondary or tertiary care facility where antibiotics had been administered. All CAP cases during each period were identified. We used Microsoft Excel’s random number generator to randomly select ≥200 cases per period (200 at “baseline”; 237 in the “immediate” post-app period; 200 at 12 months; and 200 at 24 months). We calculated that inclusion of these case numbers would achieve 90% power to detect an absolute 15% increase in guideline adherence (=.05) [6]. All patients had a chest x-ray at admission to detect radiological features of consolidation, defined as 1 or more opacities in the lung fields consistent with the diagnosis of pneumonia. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al Data Collection and Definitions Electronic health record data were collected using REDCap (version 6.5.15; Vanderbilt University) to record demographic (eg, age, sex, and ethnicity) and clinical data (eg, admission date; diagnostic impression at admission; vital signs at admission—documentation of confusion the patient, respiratory rate, systolic and diastolic blood pressures; urea; and presence of consolidation on chest x-ray at admission, as reported by a radiologist) as well as antibiotics prescribed (eg, drug name, route, and duration) during the first 24 hours post admission. in Adherence was defined as prescription of antibiotic(s), including dose(s) and route(s) of administration, according to local guidelines, during the first 24 hours post admission. The ACH antibiotic guidelines for CAP vary by the CURB-65 pneumonia severity score, where a point is given for each of the prognostic features (C: confusion, U: increased serum urea concentration, R: respiratory rate ≥30 breaths/min, B: systolic blood pressure <90 mmHg or diastolic blood pressure ≤60 mmHg, and 65: age ≥65 years). Cases with a total CURB-65 score of 0-1 were considered to be at low risk (<10%) of mortality; those with a score of 2 were at intermediate risk (10%-20%) of mortality; and a score of 3-5 indicated high risk (20%-60%) of mortality [13]. In cases whose serum urea concentration had not been measured, CRB-65 scores were calculated. CRB-65 is a validated alternative to CURB-65, shown to be predictive of mortality in patients hospitalized with pneumonia [14]. A CRB-65 score of 0 would equate to a CURB-65 score of 0 at best and 1 at worst; thus, we elected to use CURB-65 score ranges (0-1, 1-2, 2-3, and 3-5). If the attending clinicians had not documented the CURB-65 score in the clinical records, we calculated the patient’s CURB-65 score using relevant data available to the clinicians when selecting antibiotic management. Antibiotic guideline adherence was assessed according to the actual or highest possible CURB-65 score. Other antibiotic(s), prescribed in addition to guideline-adherent antibiotic(s), were considered unnecessary additional antibiotics. Undertreatment was defined as prescription of an inappropriately narrow-spectrum regimen (eg, prescription of amoxicillin alone for severe CAP). These definitions were applied by 2 physicians (CHY and SRR) and an infectious diseases specialist pharmacist (EJD) based on the assumption that the patient had CAP, regardless of the presence of pulmonary consolidation on chest x-ray (a defining characteristic of CAP, the absence of which does not preclude the diagnosis of CAP) [15]. Analysis Statistical analyses were performed using R (version 4.0.3; The R Core Team). Rates of adherence, use of unnecessary additional antibiotics, and undertreatment were compared between study periods and between cases with or without pulmonary consolidation on the admission chest x-ray (based on the reporting radiologist’s assessment), using Pearson chi-square test or Fisher exact test (significance level: α=.05). One case, in the immediate follow-up group, did not have a chest x-ray and was excluded from analyses that compared patients with or without pulmonary consolidation. Ethics Approval All analyses were performed in accordance with the study protocol for which ethics approval was granted (New Zealand Health and Disabilities Ethics Committee reference number: 16/STH/6). Results Demographic And Clinical Features The sex, median ages, and ethnicities of the patients in the 4 cohorts were broadly similar (Table 1). The proportions of patients with consolidation on chest x-ray (an initial diagnostic impression of pneumonia) and prescriber-documented CURB-65 scores were higher in the 12-month and 24-month cohorts compared to the baseline cohort. In all 4 cohorts, most patients with consolidation on chest x-ray (43/63, 68% at baseline; 54/77, 70% in the immediate post-app period; 65/92, 71% at 12 months; and 69/102, 68% at 24 months) had an initial diagnostic impression of “pneumonia.” By contrast, in all 4 cohorts, a minority of patients without consolidation on chest x-ray (25/137, 18% at baseline; 14/159, 9% in the immediate post-app period; 31/108, 29% at 12 months; 24/98, 24% at 24 months) had an initial diagnostic impression of “pneumonia” (P<.001). https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al Table 1. Demographic and clinical features and overall adherence to antibiotic guidelines for patients with community-acquired pneumonia admitted to Auckland City Hospital in the baseline, immediate, 12-month, and 24-month cohorts. Cohort Baseline (n=200) Immediate (n=237) 12-month (n=200) 24-month (n=200) Age (years), median (IQR) 62 (46-77) 64 (44-79) 70 (53-82) 67 (51-80) Sex, n (%) Female Male Ethnicity, n (%) Asian or other Māori New Zealand European Pacific Chest x-ray consolidation, n (%) Initial diagnostic impression n (%) Pneumonia Lower respiratory tract infections (un- specified) Viral illness Bronchitis or other 96 (48) 104 (52) 47 (24) 15 (7.5) 91 (46) 47 (24) 63 (32) 68 (34) 103 (52) 15 (7.5) 14 (7) CURB-65 score estimated from clinical data, n (%) 0-1 1-2 2-3 3-5 87 (44) 68 (34) 40 (20) 5 (2.5) 139 (59) 98 (41) 32 (14) 29 (12) 121 (51) 55 (23) 77 (33) 68 (29) 97 (41) 61 (26) 11 (4.6) 102 (43) 95 (40) 34 (14) 6 (2.5) 94 (47) 106 (53) 41 (20) 14 (7) 101 (50) 44 (22) 92 (46) 96 (48) 59 (30) 27 (14) 18 (9) 62 (31) 84 (42) 39 (20) 15 (7.5) 112 (56) 88 (44) 38 (19) 21 (10) 95 (48) 46 (23) 102 (51) 93 (46) 59 (30) 28 (14) 20 (10) 68 (34) 70 (35) 51 (26) 11 (5.5) Length of stay (days), median (IQR) 2.0 (1.0-4.0) 2.0 (1.0-4.0) 2.0 (1.0-5.0) 2.0 (1.0-4.2) Adherence to antibiotic guidelines, n (%) 46 (23) 73 (31) 69 (34) 62 (31) Overall Antibiotic Guideline Adherence Compared with the baseline cohort (46/200, 23%), there was a nonsignificant increase in prescriber adherence to the antibiotic guideline in the immediate cohort (73/237, 31%) but a significant increase in adherence in the 12-month cohort (69/200, 34%; P=.01), which was not sustained in the 24-month cohort (62/200, 31%; Table 1). Antibiotic Guideline Adherence in Patients With Pulmonary Consolidation For patients with consolidation on chest x-ray, antibiotic guideline adherence increased from 14% (9/63) in the baseline cohort to 30% (23/77) in the immediate cohort—a change that was sustained in the 12-month cohort (32/92, 35%) and in the 24-month cohort (32/102, 31%; P=.04; Table 2). There were no significant differences between cohorts in the prescription of unnecessary additional antibiotics or in undertreatment. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al Table 2. Adherence to antibiotic guidelines, use of additional unnecessary antibiotics, undertreatment, and diagnostic features for cases with or without consolidation on admission chest x-ray (a definitive diagnosis of pneumonia requires radiographic evidence of consolidation, but the absence of consolidation does not necessarily preclude the diagnosis) [15]. Characteristics Consolidation No consolidation Baseline (n=63), n (%) Immediate (n=77), n (%) 12 months (n=92), n (%) 24 months (n=102), n (%) P valuea Baseline (n=137), n (%) Immediate (n=159), n (%) 12 months (n=108), n (%) 24 months (n=98), n (%) P valuea Adherence Adherent 9 (14) 23 (30) 32 (35) 32 (31) Nonadherent 54 (86) 54 (70) 60 (65) 70 (69) Unnecessary additional antibiotics No Yes 38 (60) 50 (65) 59 (64) 62 (61) 25 (40) 27 (35) 33 (36) 40 (39) Undertreatment No Yes 47 (75) 64 (83) 78 (85) 82 (80) 16 (25) 13 (17) 14 (15) 20 (20) Initial diagnostic impression .04 .91 .43 .18 37 (27) 50 (31) 37 (34) 30 (31) 100 (73) 109 (69 71 (66) 68 (69) 103 (75) 107 (67) 96 (89) 70 (71) 34 (25) 52 (33) 12 (11) 28 (29) 99 (72) 123 (77) 77 (71) 76 (78) 38 (28) 36 (23) 31 (29) 22 (22) .67 .001 .55 <.001 Pneumonia 43 (68) 54 (70) 65 (71) 69 (68) 25 (18) 14 (8.8) 31 (29) 24 (24) LRTIb (un- specified) 16 (25) 18 (23) 14 (15) 17 (17) 87 (64) 78 (49) 45 (42) 42 (43) Viral illness 0 (0) Bronchitis or other 4 (6.3) 3 (3.9) 2 (2.6) 7 (7.6) 6 (6.5) 5 (4.9) 11 (11) 15 (11) 58 (36) 20 (19) 23 (23) 10 (7.3) 9 (5.7) 12 (11) 9 (9.2) CURB-65c score documented by prescriber 17 (27) 35 (45) 33 (36) 28 (27) .046 15 (11) 15 (9.4) 17 (16) 16 (16) .26 CURB-65 score calculated from clinical data .80 .002 0-1 1-2 2-3 3-5 22 (35) 31 (40) 31 (34) 35 (34) 65 (47) 71 (45) 31 (29) 33 (34) 24 (38) 27 (35) 36 (39) 32 (31) 44 (32) 67 (42) 48 (44) 38 (39) 15 (24) 14 (18) 17 (18) 26 (25) 25 (18) 20 (13) 22 (20) 25 (26) 2 (3.2) 5 (6.5) 8 (8.7) 9 (8.8) 3 (2.2) 1 (0.6) 7 (6.5) 2 (2) aChi-square test and Fisher exact test. bLRTI: lower respiratory tract infection. cPneumonia severity score (C: confusion, U: increased serum urea concentration, R: respiratory rate ≥30 breaths/min, B: systolic blood pressure <90 mmHg or diastolic blood pressure ≤60 mmHg, and 65: age ≥65 years). Discussion In patients with CAP and pulmonary consolidation on chest x-ray, there was a sustained improvement in guideline adherence. However, in patients with CAP without consolidation, where the most common diagnostic impression was “viral illness” or “lower respiratory tract infections (unspecified),” guideline adherence was not sustained. The sustained improvement in adherence to the guidelines for treatment of CAP in patients with consolidation on chest x-ray indicates that clinicians were adapting their use of the guideline to increase their use of it in those patients for whom they thought the guideline was most appropriate. This evolution of prescriber use of the guideline over time is an encouraging feature, particularly given the absence of other initiatives to improve prescribing for CAP, suggesting that prescribers were intellectually engaging with the guideline. An appropriate response by those responsible for maintaining and updating the guideline might be to include the presence or absence of consolidation on the chest x-ray as a decision point in the treatment algorithm. The only other published study of the long-term impact of an antibiotic guidelines app on prescriber adherence was performed in 3 hospitals in west London, where baseline rates of adherence were high (75%-90%) [9]. The introduction of a smartphone app resulted in a significant increase in the rate of adherence for surgical patients, sustained at 24 months. However, in medical patients, a nonsignificant increase in the rate of adherence was followed by a gradual decline toward preintervention levels. In our study of medical patients with https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al CAP, preintervention rates of adherence were low (46/200, 23%) but improved significantly to 34% (69/200) at 12 months post app implementation, before then declining to 31% (62/200). Our findings are broadly consistent with those of Charani et al [9], who found an initial increase followed by a subsequent decline in guideline adherence in medical patients. It should be noted that adherence to guidelines in our study required that the antibiotic, dose, and mode of administration be as stated in our guidelines; however, the definition used in the Charani et al [9] study required that only the antibiotic were that stated in their guidelines and did not require the dose and mode of administration to be the same as those in the guidelines. Although very high uptake and use of the app at ACH (>1000 new downloads each year, over half of which are by junior doctors) enabled this real-world evidence study, there were no data directly matching the use of the SCRIPT app by the clinicians whose antibiotic prescriptions were analyzed in this study, that is, we were not able to measure the direct influence of using the app on individual cases of antibiotic prescription but rather the average net effect of making such an app available. Other limitations included the unmeasured impact of team-based decisions (vs individual decisions) for antibiotic prescriptions and of junior doctors changing clinical jobs every few months at ACH, moving to or from other hospitals, which would periodically and variably diminish the proportions of doctors using the app at ACH. We were not able to assess the app’s impact relative to other antibiotic stewardship methods nor to other variables that may influence guideline adherence, such as the prescriber’s level of seniority, where they had previously worked, their specialty, and patient-related factors like comorbidity and illness acuity. A range of technological advances, including antibiotic guidelines apps and computerized decision support systems appear to offer opportunities to dramatically improve adherence to prescriber guidelines. However, as with our study, it is rare that such advances provide a silver bullet for the widespread, recalcitrant problem of low adherence to antimicrobial prescribing guidelines. Instead, it is common for such advances to provide modest improvements, commonly of a 10%-20% absolute improvement in guideline adherence, when a 30%-50% absolute increase would have been required to achieve adherence rates above 90% [16-19]. Although mHealth solutions have been perceived to be convenient and effective in improving guideline adherence, their high cost would be more justified should their impact be more long-term; this is especially pertinent in multimodal antibiotic stewardship programs, where there would be further opportunity costs. Causes of failure to achieve large changes in antibiotic guideline adherence include within-team dynamics that may contribute to lack of support for changes in prescriber behavior. Junior clinicians, who write almost all prescriptions, may be more influenced by the entrenched opinions of their senior colleagues than by the advice contained in a guideline [20,21]. Other causes of low adherence may pertain to app-related factors like usability, acceptability, and app fatigue, although SCRIPT was designed using state-of-the-art co-design approaches through interactions between designers and end-user stakeholders [22,23]. Moreover, rates of SCRIPT use at Auckland Hospital have steadily increased rather than declining, suggesting that the app has high usability with no evidence of app fatigue. SCRIPT can only provide guidelines, not actively reinforce them. e-Prescribing may be able to address this gap and could be the subject of future studies in antibiotic guideline adherence. Overall, our results suggest that a highly used antibiotic guidelines app can help to increase overall rates of prescriber adherence, especially in those patients with the strongest evidence that they fall into the diagnostic group the treatment advice is intended for and in those patients with more severe diseases. Sustaining increased rates of adherence likely requires refinement of the app algorithms in response to evidence that prescribers are selective in their adherence to guidelines and may respond to clinical features that are not included in the app algorithms. As with all innovations, a continuous process of development, testing, analysis, and modification is necessary to achieve the best results. Acknowledgments The authors wish to acknowledge and thank the information technology and systems teams at Auckland City Hospital (ACH), Emma Mills for her contribution to data collection, and Rachel Chen for statistical consultation. We wish to thank the Design for Health and Wellbeing Lab (ACH) for their exceptional creative skills in creating the look of SCRIPT; the Development and Design Team at the National Institute for Health Innovation (University of Auckland, New Zealand), who helped to develop the app; and the medical professionals who used our app and provided invaluable feedback. The research was supported by a Health Research Council Research Partnerships for New Zealand Health Delivery grant (15/665) and an ACH A+ Research grant (6969). No funding sources had any role in study design, data collection, or preparation of the manuscript. Data Availability The data analyzed are not publicly available as they contain personal data but may be made available subject to an application and research proposal meeting the ethical and governance requirements of accessing the data. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al Authors' Contributions All authors contributed to the study conception and design. All authors contributed to material preparation, data collection, and data analysis. The first draft of the manuscript was written by CHY, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Conflicts of Interest None declared. References 1. 2. 3. Davey PM, Marwick CA, Scott CL, Charani E, McNeil K, Brown E, et al. Interventions to improve antibiotic prescribing practices for hospital inpatients. Cochrane Database Syst Rev 2017 Feb 09;2(2):CD003543 [FREE Full text] [doi: 10.1002/14651858.CD003543.pub4] [Medline: 28178770] Curtis CE, Al Bahar F, Marriott JF. The effectiveness of computerised decision support on antibiotic use in hospitals: a systematic review. PLoS One 2017 Aug 24;12(8):e0183062 [FREE Full text] [doi: 10.1371/journal.pone.0183062] [Medline: 28837665] Schuts EC, Hulscher MEJL, Mouton JW, Verduin CM, Stuart JWTC, Overdiek HWPM, et al. Current evidence on hospital antimicrobial stewardship objectives: a systematic review and meta-analysis. The Lancet Infectious Diseases 2016 Jul;16(7):847-856. [doi: 10.1016/s1473-3099(16)00065-7] 4. Mol PG, Rutten WJ, Gans RO, Degener JE, Haaijer-Ruskamp FM. Adherence barriers to antimicrobial treatment guidelines in teaching hospital, the Netherlands. Emerg Infect Dis 2004 Mar;10(3):522-525 [FREE Full text] [doi: 10.3201/eid1003.030292] [Medline: 15109428] Dellit TH, Owens RC, McGowan JE, Gerding DN, Weinstein RA, Burke JP, Infectious Diseases Society of America, Society for Healthcare Epidemiology of America. Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America guidelines for developing an institutional program to enhance antimicrobial stewardship. Clin Infect Dis 2007 Jan 15;44(2):159-177. [doi: 10.1086/510393] [Medline: 17173212] Yoon CH, Ritchie SR, Duffy EJ, Thomas MG, McBride S, Read K, et al. Impact of a smartphone app on prescriber adherence to antibiotic guidelines in adult patients with community acquired pneumonia or urinary tract infections. PLoS One 2019 Jan 29;14(1):e0211157 [FREE Full text] [doi: 10.1371/journal.pone.0211157] [Medline: 30695078] Aikman KL, Hobbs MR, Ticehurst R, Karmakar GC, Wilsher ML, Thomas MG. Adherence to guidelines for treating community-acquired pneumonia at a New Zealand Hospital. Journal of Pharmacy Practice and Research 2015 Apr 13;43(4):272-275. [doi: 10.1002/j.2055-2335.2013.tb00273.x] Helou RI, Foudraine DE, Catho G, Peyravi Latif A, Verkaik NJ, Verbon A. Use of stewardship smartphone applications by physicians and prescribing of antimicrobials in hospitals: A systematic review. PLoS One 2020 Sep 29;15(9):e0239751 [FREE Full text] [doi: 10.1371/journal.pone.0239751] [Medline: 32991591] Charani E, Gharbi M, Moore L, Castro-Sanchéz E, Lawson W, Gilchrist M, et al. Effect of adding a mobile health intervention to a multimodal antimicrobial stewardship programme across three teaching hospitals: an interrupted time series study. J Antimicrob Chemother 2017 Jun 01;72(6):1825-1831 [FREE Full text] [doi: 10.1093/jac/dkx040] [Medline: 28333297] Secondary SCRIPT BigQuery dashboard. SCRIPT BigQuery Dashboard. URL: https://datastudio.google.com/reporting/ 1VBgnaE7hECYAhpadopwsUEfCyw2lGYk8/page/gnwl?s=qNALSiYx5FE [accessed 2023-04-20] 5. 6. 7. 8. 9. 10. 11. Drug Therapeutics Bulletin. An introduction to patient decision aids. BMJ 2013 Jul 23;347:f4147. [doi: 10.1136/bmj.f4147] [Medline: 23881944] 12. Wootton D, Feldman C. The diagnosis of pneumonia requires a chest radiograph (x-ray)-yes, no or sometimes? Pneumonia (Nathan) 2014 Jun 19;5(Suppl 1):1-7 [FREE Full text] [doi: 10.15172/pneu.2014.5/464] [Medline: 31641570] 13. Lim W, van der Eerden MM, Laing R, Boersma WG, Karalus N, Town GI, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax 2003 May;58(5):377-382 [FREE Full text] [doi: 10.1136/thorax.58.5.377] [Medline: 12728155] 14. McNally M, Curtain J, O'Brien KK, Dimitrov BD, Fahey T. Validity of British Thoracic Society guidance (the CRB-65 rule) for predicting the severity of pneumonia in general practice: systematic review and meta-analysis. Br J Gen Pract 2010 Oct 01;60(579):e423-e433. [doi: 10.3399/bjgp10x532422] 15. Durrington HJ, Summers C. Recent changes in the management of community acquired pneumonia in adults. BMJ 2008 Jun 21;336(7658):1429-1433 [FREE Full text] [doi: 10.1136/bmj.a285] [Medline: 18566081] 16. Laka M, Milazzo A, Merlin T. Can evidence-based decision support tools transform antibiotic management? A systematic review and meta-analyses. J Antimicrob Chemother 2020 May 01;75(5):1099-1111. [doi: 10.1093/jac/dkz543] [Medline: 31960021] Paul M, Andreassen S, Tacconelli E, Nielsen AD, Almanasreh N, Frank U, TREAT Study Group. Improving empirical antibiotic treatment using TREAT, a computerized decision support system: cluster randomized trial. J Antimicrob Chemother 2006 Dec;58(6):1238-1245. [doi: 10.1093/jac/dkl372] [Medline: 16998208] 17. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Yoon et al 18. Demonchy E, Dufour J, Gaudart J, Cervetti E, Michelet P, Poussard N, et al. Impact of a computerized decision support system on compliance with guidelines on antibiotics prescribed for urinary tract infections in emergency departments: a multicentre prospective before-and-after controlled interventional study. J Antimicrob Chemother 2014 Oct;69(10):2857-2863. [doi: 10.1093/jac/dku191] [Medline: 24898019] 19. Nachtigall I, Tafelski S, Deja M, Halle E, Grebe MC, Tamarkin A, et al. Long-term effect of computer-assisted decision support for antibiotic treatment in critically ill patients: a prospective 'before/after' cohort study. BMJ Open 2014 Dec 22;4(12):e005370 [FREE Full text] [doi: 10.1136/bmjopen-2014-005370] [Medline: 25534209] 20. Broom A, Broom J, Kirby E. Cultures of resistance? A Bourdieusian analysis of doctors' antibiotic prescribing. Soc Sci Med 2014 Jun;110:81-88. [doi: 10.1016/j.socscimed.2014.03.030] [Medline: 24727665] 21. Broom J, Broom A, Anstey C, Kenny K, Young S, Grieve D, et al. Barriers-enablers-ownership approach: a mixed methods analysis of a social intervention to improve surgical antibiotic prescribing in hospitals. BMJ Open 2021 May 10;11(5):e046685 [FREE Full text] [doi: 10.1136/bmjopen-2020-046685] [Medline: 33972342] 22. Ball E, Rivas C. Health apps require co-development to be acceptable and effective. Front Psychol 2021 Jul 16;12:714453 23. [FREE Full text] [doi: 10.3389/fpsyg.2021.714453] [Medline: 34335428] Slattery P, Saeri AK, Bragge P. Research co-design in health: a rapid overview of reviews. Health Res Policy Syst 2020 Feb 11;18(1):17 [FREE Full text] [doi: 10.1186/s12961-020-0528-9] [Medline: 32046728] Abbreviations ACH: Auckland City Hospital CAP: community-acquired pneumonia mHealth: mobile health Edited by A Mavragani; submitted 26.09.22; peer-reviewed by H Islam, C Xie; comments to author 25.01.23; revised version received 12.04.23; accepted 14.04.23; published 02.05.23 Please cite as: Yoon CH, Nolan I, Humphrey G, Duffy EJ, Thomas MG, Ritchie SR Long-Term Impact of a Smartphone App on Prescriber Adherence to Antibiotic Guidelines for Adult Patients With Community-Acquired Pneumonia: Interrupted Time-Series Study J Med Internet Res 2023;25:e42978 URL: https://www.jmir.org/2023/1/e42978 doi: 10.2196/42978 PMID: 37129941 ©Chang Ho Yoon, Imogen Nolan, Gayl Humphrey, Eamon J Duffy, Mark G Thomas, Stephen R Ritchie. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 02.05.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e42978 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e42978 \| p. 8 (page number not for citation purposes)
10.1371_journal.ppat.1011251	RESEARCH ARTICLE MoErv14 mediates the intracellular transport of cell membrane receptors to govern the appressorial formation and pathogenicity of Magnaporthe oryzae Bin Qian1,2, Xiaotong Su1,2, Ziyuan Ye1,2, Xinyu Liu1,2, Muxing Liu1,2, Haifeng Zhang1,2, Ping Wang3, Zhengguang ZhangID 1,2* 1 Department of Plant Pathology, College of Plant Protection, Nanjing Agricultural University, and Key Laboratory of Integrated Management of Crop Diseases and Pests, Ministry of Education, Nanjing, China, 2 The Key Laboratory of Plant Immunity, Nanjing Agricultural University, Nanjing, China, 3 Department of Microbiology, Immunology, and Parasitology, Louisiana State University Health Sciences Center, New Orleans, Louisiana, United States of America * [email protected] Abstract Magnaporthe oryzae causes rice blasts posing serious threats to food security worldwide. During infection, M. oryzae utilizes several transmembrane receptor proteins that sense cell surface cues to induce highly specialized infectious structures called appressoria. However, little is known about the mechanisms of intracellular receptor tracking and their function. Here, we described that disrupting the coat protein complex II (COPII) cargo protein MoErv14 severely affects appressorium formation and pathogenicity as the ΔMoerv14 mutant is defective not only in cAMP production but also in the phosphorylation of the mito- gen-activated protein kinase (MAPK) MoPmk1. Studies also showed that either externally supplementing cAMP or maintaining MoPmk1 phosphorylation suppresses the observed defects in the ΔMoerv14 strain. Importantly, MoErv14 is found to regulate the transport of MoPth11, a membrane receptor functioning upstream of G-protein/cAMP signaling, and MoWish and MoSho1 function upstream of the Pmk1-MAPK pathway. In summary, our studies elucidate the mechanism by which the COPII protein MoErv14 plays an important function in regulating the transport of receptors involved in the appressorium formation and virulence of the blast fungus. Author summary During the pathogenic interaction with rice, Magnaporthe oryzae senses cell surface cues and forms a highly specialized infectious structure called an appressorium that initiates the infection. Studies have shown several transmembrane proteins essential for the per- ception of rice host cues. However, little is known about how these transmembrane pro- teins are trafficked intracellularly to impact appressorium formation and pathogenicity. In this study, we found MoErv14 is important in the growth and pathogenicity of the blast a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Qian B, Su X, Ye Z, Liu X, Liu M, Zhang H, et al. (2023) MoErv14 mediates the intracellular transport of cell membrane receptors to govern the appressorial formation and pathogenicity of Magnaporthe oryzae. PLoS Pathog 19(4): e1011251. https://doi.org/10.1371/journal. ppat.1011251 Editor: Jie Zhang, Institute of Microbiology, Chinese Academy of Sciences, CHINA Received: September 12, 2022 Accepted: February 28, 2023 Published: April 3, 2023 Copyright: © 2023 Qian et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Funding: This work was supported by the Natural Science Foundation of China-German Research Foundation Mobility Programme (31861133017 to ZZG), and the China National Funds for Innovative Research Groups (Grant No.31721004 to ZZG), NSFC (31772110 to ZZG). QB received support from Natural Science Foundation of China Youth PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 1 / 20 PLOS PATHOGENSProgramme (NSFC 32202240), and grant number BK20200543 fromYouth Program for Natural Science Foundation of Jiangsu Province. WP received support from grant number AI156254 and AI168867 of the National Institutes of Health (USA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. MoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae fungus, and MoErv14 regulates the transport of several membrane sensors, including MoPth11, MoWish, and MoSho1, that function upstream of either G-protein/cAMP sig- naling or the Pmk1-MAPK pathway. Importantly, we revealed that MoErv14 mediates the transport of these receptor proteins to promote the appressorium formation and pathoge- nicity of M. oryzae. Introduction Rice blast is one of the most devastating fungal diseases on rice, and it seriously threats world food supply security [1]. When infecting the host, the blast fungus Magnaporthe oryzae recog- nizes external physical and chemical signals, such as surface hardness and hydrophobicity, to produce specific infectious structures called appressoria. The robust turgor generated in appressoria leads to the penetration of the host epidermis and the production of invasive hyphae. Previous studies indicated that several cell surface transmembrane receptor proteins, including MoPth11, play a role in sensing the host surface for appressorium production. MoPth11 is an atypical G protein-coupled receptor (GPCR) that senses surface hydrophobicity leading to endocytosis, which regulates the cAMP-signal pathway and appressorium formation [2,3]. In addition, the receptor proteins MoMsb2 and MoSho1 recognize the host surface kera- tinous monomer to activate the MoMst11-MoMst7-MoPmk1 MAP kinase signaling pathway that also controls appressorium formation. A recent study suggested that a GPCR protein, MoWish, is involved in appressorium formation by also recognizing host hydrophobic signals [4]. Although the above studies showed that surface transmembrane proteins are important for signal perception and appressorium induction, the molecular mechanism by which these transmembrane proteins are transported intracellularly and exert their function remains unknown. In eukaryotic cells, protein transport and exchange of macromolecular substances between different organelles depend on vesicle trafficking. Vesicles are membrane-coated cargos con- taining various macromolecules that shuttle between different subcellular structures or organ- elles. According to their different forms of membrane coating, vesicles are divided into clathrin vesicles, COPI (cytoplasmic envelope complex I or coat protein complex I) vesicles, and COPII (cytoplasmic envelope complex II, coat protein complex II) vesicles. Clathrin vesi- cles mainly function in the transport of cargo from the Golgi apparatus to lysosomes and cell membranes and the endocytosis of extracellular or membrane substances into the cell. COPI vesicles mediate transport from the Golgi back to the endoplasmic reticulum (ER), and COPII vesicles mediate the material transport from the ER to the Golgi [5]. The transport of substances from the ER to the Golgi apparatus mediated by COPII is an essential component of the secretion pathway in eukaryotes that is essential for cells to perform their normal functions and maintain the dynamic balance of various organelles [6]. For exam- ple, in the budding yeast Saccharomyces cerevisiae, the Sec23/Sec24 complex forms the inner membrane, while the Sec13/Sec31 complex forms the outer membrane when coating on COPII vesicles [7]. ERV (ER-derived vesicles) proteins, divided according to their molecular weights [8,9], are a group of conserved cargo receptor proteins that is mainly responsible for recruiting specific cargo proteins and directing their ER to Golgi transport [9]. Different ERV proteins recognize specific cargo proteins [10], but the transport of the cargo containing trans- membrane receptor proteins remains unclear. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 2 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae We previously reported that MoErv29 specifically mediates the secretion of protein effec- tors during infection [11]. Here, we characterized the function of MoErv14, another member of the COPII vesicle receptor family in M. oryzae, and we revealed that MoErv14 mediates the transport of various transmembrane receptor proteins that promote the appressorium forma- tion and pathogenicity of M. oryzae. Results Identification and characterization by host infection of ERV family genes in M. oryzae We previously reported that the COPII cargo receptor MoErv29 promotes the apoplastic effec- tor secretion and the virulence of M. oryzae [11]. In order to systematically elucidate functions of the ERV COPII cargo receptor family proteins, we screened the available genomes of M. oryzae by BLAST using the S. cerevisiae ERV proteins’ sequence as a reference and identified five MoERV genes in M. oryzae: MoERV14 (MGG_08132), MoERV25 (MGG_08210), MoERV26 (MGG_08706), MoERV41 (MGG_00949), and MoERV46 (MGG_01245) (Fig 1A and 1B). To examine the roles of the above-identified ERV proteins in M. oryzae, we generated respective mutant strains that were verified by Southern blot analysis (S1 Fig) and carried out pathogenicity studies. Conidial suspensions of the mutants were sprayed onto 2-week-old rice seedlings (Oryza sativa cv. CO-39), along with the wild-type strain Guy11. Seven days after inoculation, ΔMoerv14 mutants failed to produce any leaf lesions, while ΔMoerv25, ΔMoerv26, ΔMoerv41, and ΔMoerv46 all produced numerous spindle-shaped lesions, comparable to Guy11. Similar results were also observed on detached barley leaves. However, in the rice- sheath penetration assay, lesions produced by the ΔMoerv14 mutant showed 95% Type 1 and 5% Type 2 infectious hyphal growth, compared to 9% Type 1, 20% Type 2, 31% Type 3, and 40% Type 4 in Guy11. The severe defect in the pathogenicity of ΔMoerv14 was restored in the complemented strain (Fig 1C–1F). We concluded that MoErv14 plays a critical role in host penetration and invasive hyphae growth. MoErv14 is involved in the vegetative growth, conidium formation, and appressorium development of M. oryzae We selected MoErv14 for further studies based on its importance in pathogenicity. Firstly, to determine whether Erv14 is conserved among species, we used cDNA of MoErv14 to comple- ment a yeast Δerv14 mutant. As expected, MoErv14 could partially suppress the growth defect of the yeast Δerv14 mutant upon dithiothreitol (DTT)-induced stress (S2 Fig). Despite the par- tial suppression, it suggested that MoErv14 could be functionally homologous to ScErv14. We then observed the vegetative growth of ΔMoerv14 mutants that showed significant growth reduction on CM, MM, SDC, and OM media in comparison to Guy11 and the complemented strain (Fig 2A–2B). The colony formed by the ΔMoerv14 mutant was much thinner on day 7 (Fig 2C). We also found that ΔMoerv14 mutants produced significantly fewer conidia on corn- meal agar (SDC) (Fig 2D–2E). These results indicated that MoErv14 is important for the hyphae growth and asexual development of M. oryzae. To further assess hyphae growth and asexual development as the cause of decreased viru- lence of the ΔMoerv14 mutant, we estimated conidia germination and appressorium formation at various time points [12–14]. The conidia germination of the ΔMoerv14 mutant was not sig- nificantly different from Guy11. However, its appressorium formation rate was significantly lower than Guy11, as ~20% of the conidia developed appressoria in 24 hours (Fig 2F–2G). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 3 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 1. Prediction, phylogenetic analysis, and virulence of ERV (ER-derived vesicles) proteins in M. oryzae. (A) Phylogenetic trees of ERV proteins from several organisms were constructed using the CLUSTAL_W, and MEGA 5.1 programs by the neighbor-joining method with 1000 bootstrap replicates. (B) Schematic representation of ERV proteins in M. oryzae. (C) Two- week-old rice seedlings (Oryza sativa cv. CO-39) were sprayed with tested with a conidial suspension of different mutants (5×104 conidia/ml). The diseased leaves were harvested and photographed 7 days after inoculation. Three independent experiments were performed. (D) Quantification of lesion types (per 1.5 cm2) on susceptible rice leaves. 0, no lesion; 1, dark-brown pinpoint lesions; 2, 1.5 mm brown spots; 3, 2–3 mm lesions with brown margins; 4, eyespot lesions longer than 3 mm; 5, coalesced lesions infecting 50% or more of the leaf maximum size. Asterisks represent significant differences (p<0.01). (E) Separated barley leaves were dropped with serial dilution (1×105, 1×104, 1×103 conidia /ml) of conidial suspensions, and diseased leaves were photographed 5 days after inoculation. (F) Conidial suspensions (1×105 spores/ml) of three different strains were injected into separate rice sheaths, and the infection severity was observed at 24 h post-inoculation (hpi). Percentages of different types of infectious hyphae in rice cells were counted at 24 hpi. Sample size (n) = 50. Error bars represent the standard deviations. Type 1, no infection structures; Type 2, only with a appressorium; Type 3, only with a single invasive hypha (IH); Type 4, with 1–3 branches but restricted in one cell. https://doi.org/10.1371/journal.ppat.1011251.g001 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 4 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 2. MoErv14 is involved in the vegetative growth, conidium formation, and appressorium development. (A) The tested strains Guy11, ΔMoerv14 mutants, and complemented strain were inoculated on CM, MM, OM, and SDC media, 28˚C for 7 days, and then photographed. (B) Statistical analysis of the colony diameter of wild-type Guy11, ΔMoerv14 mutants and complemented strain on different media. Error bars represent the standard deviations; Asterisks denote statistical significances (p<0.01). (C) The ΔMoerv14 mutant appears as a flat colony with thin aerial hyphal growth when compared with the wild type Guy11. (D) Conidia of different strains were observed under a light microscope after illumination for 24 h and then photographed. (E) The number of conidia was calculated and analyzed from the tested strains following incubation on SDC medium for 7 days. Error bars represent the standard deviations. Asterisks represent a significant difference (p<0.01). (F) A comparative time-lapse observation about the conidia germination and appressorium formation of the ΔMoerv14 mutant and the wild type Guy11 at 4, 8, 12, and 24 h time points. (G) Statistical analysis of the appressorium formation observed at 4, 8, 12 and 24 h time point. Error bars represent the standard deviations. Asterisks represent significant differences. (p<0.01). https://doi.org/10.1371/journal.ppat.1011251.g002 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 5 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Meanwhile, the appressoria of the ΔMoerv14 mutant showed a higher collapse ratio than the control, indicating a decreased turgor pressure (S3 Fig). In addition, MoErv14 appeared not to be involved in a septin ring formation (S4 Fig). This results suggested that MoErv14 is impor- tant for appressorium formation and function. MoErv14 is a vesicle traffic-associated protein We then generated a MoErv14-GFP fusion protein and found MoErv14-GFP is located at vesi- cle-like spots that could be stained by the lipophilic dye FM4-64 (Fig 3A). With reference to RFP-tagged MoLhs1 and MoSft2 proteins that are markers of ER and Golgi, respectively Fig 3. MoErv14 shuttles between ER to Golgi in M. oryzae. (A) MoErv14-GFP were expressed in ΔMoerv14, then stained with FM4-64 for 3min and observed by Axio Observer A1 Zeiss inverted microscope. The left shown is the tip of the hypha and the right shown is the middle of the hyphae. Asterisks represent the labeled and merged vesicles. Bar = 10 nm. (B) The localization pattern of MoErv14 in the hypha. Shown are confocal fluorescent images (Zeiss LSM710, 63×oil). Bar = 10 nm. (C) Time-lapse images of cells expressing the MoErv14-GFP reporter at different time intervals with or without the treatment of ER-Golgi protein trafficking inhibitor Brefeldin A (BFA), 50 μg/ml, DMSO treatment was used as a control, Bar = 10 nm. White frames indicated dynamic MoErv14 and red frames indicated detained MoErv14. https://doi.org/10.1371/journal.ppat.1011251.g003 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 6 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae [15–17], we observed about 25% of MoErv14-GFP co-localized or was surrounded by MoLhs1, and there was nearly 75% MoErv14-GFP co-localized with MoSft2 (Fig 3B). Similar results were also observed during conidia and appressorium stages in that about ~80% MoErv14 localized in Golgi and ~20% MoErv14 in ER, respectively (S5A and S5B Fig). Moreover, by using gradient centrifugation to fractionate subcellular organelles, we con- firmed that MoErv14 was mainly accumulated in Golgi and ER (S5C Fig). We further co- expressed MoErv14-S and MoSec24-2-GFP in Guy11 and extracted the vesicular proteins. Proteins were then analyzed by Western blotting using S tag and GFP-specific antibodies. The results showed that MoErv14 and MoSec24-2 were detected in total vesicular proteins, which indicated that MoErv14 is a vesicular protein (S5D Fig). In yeast, Erv14 was reported to func- tion as a COPII component shuttling between ER and Golgi. We then employed Brefeldin A (BFA, an inhibitor of ER to Golgi protein transport) and live-cell time-lapse imaging analysis revealed that the dynamic localization pattern of MoErv14 can be specifically inhibited by BFA (Fig 3C) [18]. Taken together, these results suggested that MoErv14 is a COPII component that shuttles between ER and Golgi. We also examined whether the ΔMoerv14 mutant is defective in endocytosis. By staining cells with FM4-64, we observed its internalization, but there was no apparent difference between the ΔMoerv14 mutant and Guy11 2 and 5 min after staining (S6 Fig). MoErv14 mediates cAMP and MoPmk1-MAPK signaling pathways Previous studies demonstrate that both cAMP and MoPmk1-MAPK signaling pathways are implicated in transducing hydrophobic surface signals to regulate appressorium formation in M. oryzae [1]. When MoPth11 senses surface signals, cAMP signaling is activated to promote appressorium formation. MoMsb2 and MoSho1, on the other hand, control the formation of appressoria by sensing host surface signals to activate the Mst11-Mst7-Pmk1 signaling path- way through protein phosphorylation. We measured cellular levels of cAMP and MoPmk1 phosphorylation, respectively, and the results showed that both cAMP levels and MoPmk1 phosphorylation were decreased significantly in the ΔMoerv14 mutant (S7A–S7B Fig). We further analyzed the effect of externally supplementing cAMP on the ΔMoerv14 mutant. At 10 and 20 mM levels, the ΔMoerv14 mutant can germinate and form appressoria, but not at the levels of the wild-type strain (Fig 4A–4B). Exogenous cAMP also suppressed the virulence defect of ΔMoerv14 mutants on rice and barley (Fig 4C). We also transformed a MoMST7S212D, T216E (MoMST7DE) construct into the ΔMoerv14 mutant to constitutively activate the MoPmk1-MAPK1 pathway (Fig 5A). The results showed that MoMst7DE effectively suppressed the defect of the ΔMoerv14 mutant in appres- sorium formation (Fig 5B–5C). Seven days after inoculation, the virulence of ΔMoerv14/ MoMST7DE was partially restored when compared with the ΔMoerv14 mutant. ΔMoerv14/ MoMST7DE also exhibited a higher penetration rate and enhanced infectious hyphal growth (Fig 5D–5F). Based on these results, we concluded that MoErv14 is involved in regulating cAMP and MoPmk1-MAPK signaling pathways required for appressorium formation and pathogenicity. MoErv14 interacts with and mediates the transport of MoPth11, MoWish, and MoSho1 Previous studies suggested that transmembrane receptors, including MoPth11, MoWish, and MoSho1, are important for cell surface perception and appressorium formation in M. oryzae [3,4,19]. We demonstrated that MoErv14 regulates cAMP and MoPmk1-MAPK signaling pathways during appressorium formation and pathogenicity. To test whether MoErv14 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 7 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 4. MoErv14 is important for the appressorium formation and virulence by regulating cAMP level in M. oryzae. (A) Exogenous cAMP suppressed the appressorium formation defect of the ΔMoerv14 mutant. (B) Statistical analysis of the appressorium formation rate. Error bars represent ±standard deviations (SD), and different numbers denote statistical significances (p< 0.01). (C) Two-week- old rice seedlings (Oryza sativa cv. CO-39) were sprayed with tested with a conidial suspension of different mutants (5×104 conidia/ml). The diseased leaves were harvested and photographed 7 days after inoculation. Three independent experiments were performed. Separated barley leaves were dropped with serial dilution (5×104 conidia /ml) of conidial suspensions, and diseased leaves were photographed 5 days after inoculation. https://doi.org/10.1371/journal.ppat.1011251.g004 interacts with these receptors, yeast two-hybrid (Y2H) and co-immunoprecipitation (co-IP) assays were conducted that showed MoErv14 interacting with all three proteins (Fig 6A–6F). Also, bimolecular fluorescence complementation (BiFC) was used to confirm the interaction between MoErv14 and these proteins following BFA treatment (Fig 6G–6I). However, no interaction was detected among these transmembrane receptor proteins (S8A–S8C Fig). As MoErv14 functions in COPII vesicle trafficking, we speculated that MoErv14 also inter- acts with these proteins to control their transport. Hence, we fused GFP with these proteins and expressed them in Guy11 and the ΔMoerv14 mutant, respectively. The results showed that MoPth11, MoWish, and MoSho1 were all located at the cell membrane surface of Guy11. However, in ΔMoerv14 mutants, they were all retained in ER (Fig 7A–7C), which was similar with the BFA treatment of Guy11 (S9 Fig). Using gradient centrifugation to fractionate subcel- lular organelles, we found that MoPth11-GFP, MoWish-GFP, and MoSho1-GFP were mainly distributed in ER fractions of the ΔMoerv14 mutant, along with MoLhs1 (Fig 7D–7F). To distinguish whether these proteins were transported to the cell surface, we observed their locations with or without Latrunculin B (LatB) which inhibits endocytosis. In Guy11 and during the appressorium formation stage, these proteins were transported into the cytoplasm by endocytosis from the cell membrane following signal perception to mediate appressorial formation. In the presence of LatB, these proteins were retained in the cell membrane surface. In contrast, these proteins failed in transport to the cell surface and were trapped on the ER in PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 8 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 5. Activating the Pmk1-MAPK signaling pathway suppresses the defect of appressorium formation in the ΔMoerv14 mutant. (A) Pmk1 phosphorylation level analysis with proteins extracted from the mycelia of Guy11, ΔMoerv14 mutant, and ΔMoerv14/Mst7DE. The phosphorylation levels of Pmk1 (42-kDa) were detected using a phosphor-MAPK antibody (upper panel). The endogenous Pmk1 was detected using a MAPK antibody (lower panel). (B) Appressorium formation assay on the hydrophobic surfaces at 4 h, 8 h, 12 h and 24 h, respectively. (C) Appressorium formation rates were calculated and statistically analyzed. Asterisks represent significant differences (p<0.01). (D) Pathogenicity assay on detached barley leaves. (E) Pathogenicity assay on rice (CO-39) and the quantification of the lesion numbers per 5 cm length of rice leaf. Error bars represent SD and different numbers denote statistical significances (p< 0.01). (F) Penetration assays in rice sheath. IH growth on rice cells was observed at 24 hpi and 4 types of IH were quantified and statistically analyzed. Error bars represent SD. Asterisks represent expanded IH. https://doi.org/10.1371/journal.ppat.1011251.g005 the mutants, with or without LatB treatment (Fig 8A–8C). We further compared the ΔMoerv14 mutant and Guy11 in the transport of MoPth11, MoWish, and MoSho1 by fluores- cence recovery after photobleaching (FRAP). We intended to bleach the fluorescence from the regions where MoPth11-GFP, MoWish-GFP, and MoSho1-GFP were accumulated in germ tubes, and the recovery of fluorescence can reflect the rate of endocytosis. In the FRAP assay, we bleached 90% of fluorescence of a region using 488 nm light. For MoPth11-GFP, fluores- cence was recovered at post-photobleach 35 s in Guy11, but not in the ΔMoerv14 mutant. In addition, the recovery level of MoWish-GFP and MoSho1-GFP in the ΔMoerv14 mutant was PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 9 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 6. MoErv14 interacts with MoPth11, MoWish, and MoSho1. (A-C) Yeast two-hybrid assay for interactions of MoErv14 with MoPth11, MoWish, and MoSho1. (D-F) Co-immunoprecipitation (Co-IP) assays to detect interactions of MoErv14 with MoPth11, MoWish, and MoSho1. Total proteins were extracted from MoErv14-S tag/MoPth11-GFP, MoErv14-S tag/MoWish-GFP, MoErv14-S tag/MoWish-GFP co-expressing strains, and the MoErv14-S tag/GFP- empty strain. Proteins were then incubated with anti-GFP agarose before being eluted and detected with anti-S tag and anti-GFP antibodies, respectively. (G-I) BiFC assay for the MoPth11 & MoErv14, MoWish & MoErv14 and MoSho1 & MoErv14 interaction in vivo. Hyphae were treated with BFA and examined by DIC and fluorescence microscopy. The strains expressing the MoPth11-YFPN & YFPC, MoErv14-YFPC & empty YFPN, MoWish-YFPN & YFPC, MoSho1-YFPN & YFPC, and empty YFPN and empty YFPC constructs were used as controls. Dye ER-Tracker Red was used to image ER in the interacting strains and merged. Bars = 10 μm. https://doi.org/10.1371/journal.ppat.1011251.g006 also significantly impaired when compared with Guy11 at the same post-photobleach (Fig 8D–8F). Collectively, these results suggested that MoErv14 is involved in the transport of MoPth11, MoWish, and MoSho1 to the Golgi, and MoErv14 affects the transportation of these proteins to the cell membrane. Discussion Understanding the mechanisms by which pathogen transmembrane receptors transduce plant surface signals promotes the development of strategies in controlling the rice blast. Previous studies revealed that MoEnd3 and MoCrn1 undergo endocytosis following the perception of external signals in M. oryzae [20,21]. This current study demonstrated that MoErv14 mediated the signal transduction of receptors MoPth11 and MoSho1 by governing their intracellular transport. MoPth11 and MoSho1 regulate cAMP and Pmk1-MAPK signaling pathways, respectively, to promote appressorium formation and pathogenicity of the blast fungus. Based on these results, we concluded that COPII coat complex proteins, such as MoErv14, played critical roles in recognizing and transmitting host surface signals (Fig 9). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 10 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 7. MoErv14 is involved in the transport of MoPth11, MoWish, and MoSho1. (A-C) The subcellular localization of MoPth11-GFP, MoWish-GFP, and MoSho1-GFP in Guy11 and the ΔMoerv14 mutant during hyphal stage. MoPth11-GFP, MoWish-GFP, and MoSho1-GFP were localized in the cell membrane of the wild-type and in ER of the ΔMoerv14 mutant. Bars, 10 μm. (D-F) Organelles from M. oryzae protoplasts were partially separated by centrifugation. The ER distribution and gradient fractions were analyzed by Western blotting using RFP antibodies against the ER marker MoLhs1 fused with RFP. Distribution of MoPth11 was detected by the anti-GFP antibody. https://doi.org/10.1371/journal.ppat.1011251.g007 Previous studies also showed that COPII coat complex proteins-mediated ER to Golgi transport is an important component of eukaryotic secretion pathways [22]. In S. cerevisiae, individual ERV proteins recognize the different groups of cargo proteins and are important in ER-Golgi transport [8,23–25]. In M. oryzae, this process is closely linked to its physiology and pathogenicity. Following our previous studies of MoErv29 promoting apoplastic effector secre- tion contributing to the virulence of M. oryzae [11], we here showed that another ERV protein, MoErv14, was involved in appressorium formation and pathogenicity by governing intracellu- lar transport of several receptor proteins. In addition, even though we found a severe defect of appressorium formation in the ΔMoerv14 mutant, no defect was found in conidium germina- tion. Similar results were also observed in the ΔMowish and ΔMosho1 mutants [4,19]. These observations indicated that conidium germination may be regulated by different signal pathways. The increased knowledge of each specific cargo receptor makes it feasible to hypothesize the presence of specificities between cargo coat proteins and cargo contents. Indeed, the exam- ination of the cargo spectrum relying on ScErv14 suggested that this large and non-homoge- nous group of proteins is recognized based on the length of their transmembrane domains (TMDs), rather than their sequences [26]. We have noticed that MoPth11, MoSho1, and PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 11 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 8. MoErv14 regulates trafficking of MoPth11-GFP, MoWish-GFP, and MoSho1-GFP during conidial germination. (A) MoPth11-GFP localization patterns in germ tubes of Guy11 and ΔMoerv14 mutant at 3 h and MoPth11-GFP localization patterns with LatB treatment in germ tubes of Guy11 at 3 h. Bars = 10 μm. (B) MoWish- GFP localization patterns in germ tubes of Guy11 and ΔMoerv14 mutant at 3 h. MoWish-GFP localization patterns with LatB treatment in germ tubes of Guy11 at 3 h. Bars = 10 μm. (C) MoSho1-GFP localization patterns in germ tubes of Guy11 and ΔMoerv14 mutant at 3 h. MoSho1-GFP localization patterns with LatB treatment in germ tubes of Guy11 at 3 h. Bars = 10 μm. (D) Representative images of FRAP analysis for diffusion at MoPth11-GFP localized regions in germ tubes of Guy11 and ΔMoerv14. The fluorescence of MoPth11-GFP significantly recovered at 35 s post- photobleaching in Guy11 but not in ΔMoerv14. The FRAP curves of MoPth11-GFP localized regions in Guy11 and ΔMoerv14. 20 regions from different cells were subjected to FRAP analysis for each strain. Intervals: 5 s. Bars = 5 μm. (E) Representative images of FRAP analysis for diffusion at MoWish-GFP localized regions in germ tubes of Guy11 and ΔMoerv14. The fluorescence of MoWish-GFP significantly recovered at 35 s post-photobleaching in Guy11 but not in ΔMoerv14. The FRAP curves of MoWish-GFP localized regions in Guy11 and ΔMoerv14. 20 regions from different cells were subjected to FRAP analysis for each strain. Intervals: 5 s. Bars = 5 μm. (F) Representative images of FRAP analysis for diffusion at MoSho1-GFP localized regions in germ tubes of Guy11 and ΔMoerv14. The fluorescence of MoSho1-GFP significantly recovered at 35 s post-photobleaching in Guy11 but not in ΔMoerv14. The FRAP curves of MoSho1-GFP localized regions in Guy11 and ΔMoerv14. 20 regions from different cells were subjected to FRAP analysis for each strain. Intervals: 5 s. Bars = 5 μm. https://doi.org/10.1371/journal.ppat.1011251.g008 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 12 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Fig 9. A proposed working model of MoErv14 function. During germ tube development, GPCR Pth11, Wish, and membrane sensor MoSho1 are recognized and transported by MoErv14. After seining the plant surface signals the membrane sensors are regulated by MoEnd3-mediated endocytosis. Following transport to endosomal systems by endocytic vesicles, MoPth11, MoWish, and MoSho1 can trigger a downstream Pmk1-MAPK and CWI MAPK cascade. The MAPK cascade facilitates successful appressorium formation, penetration, and pathogenicity. https://doi.org/10.1371/journal.ppat.1011251.g009 MoWish all contain a long TMD; however, MoErv14 is not known to regulate MoMsb2 that also contains a long TMD. These findings indicated that the mechanisms of MoErv14 in recog- nizing these transmembrane proteins may be more complex. Attenuation of appressorium formation would provide a means for managing the rice blast. Despite this extraordinary potential, however, relatively few fungicides targeting appressoria have been developed. Previously, identifying chemicals inhibiting very long-chain fatty acids (VLCFA) biosynthesis, which prevents septin-mediated appressorium formation, offers an effective strategy for controlling fungal diseases [27]. Our study could provide another avenue for controlling the disease by interfering with MoErv14-mediated ER-Golgi trafficking. Methods and materials Strains and cultural conditions The M. oryzae strain Guy11 was used as the wild-type strain for transformation in this research. All strains were cultured on a complete medium (CM) at 28˚C [28,29]. For vegetative growth, small blocks were cut from the edge of 7-day-old cultures and placed onto fresh media, followed by incubation in the dark at 28˚C. The radial growth was then measured after PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 13 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae incubation for seven days. Other media, including oatmeal agar medium and minimal medium, were prepared as described previously [27]. Liquid CM was used to prepare the vege- tative mycelia for DNA and RNA extraction. For conidiation, mycelial blocks were inoculated on straw decoction and corn agar medium (SDC) (100 g of rice straw decoction was boiled in 1 L of ddH2O for 20 min and filtered, and the filtrate was mixed with 40 g of cornmeal and 15 g of agar and adjusted to 1 L with ddH2O) at 28˚C for 7 d in the dark and followed by 3 days of continuous illumination under fluorescent light [30]. Co-immunoprecipitation (co-IP) assay The empty GFP construct was obtained in our previous work [28]. The MoERV14 DNA frag- ment fused with the S tag was inserted into the pXY203 construct (MoERV14-S) containing the hygromycin resistance, and the MoPTH11, MoWISH, and MoSHO1 DNA fragment fused with GFP fluorescent protein was inserted into the pYF11 construct, respectively. Then MoPth11-GFP/MoErv14-S, MoWish-GFP/MoErv14-S, MoSho1-GFP/MoErv14-S, and MoErv14-S/GFP fusion constructs were co-transformed into wild-type strain Guy11, and transformants resistant to hygromycin and bleomycin were isolated. Total proteins were extracted from the transformants using protein lysis buffer (50 mM Tris-HCl, pH 7.4, with 150 mM NaCl, 1 mM EDTA, and 1% Triton X-100 [Sigma-Aldrich, T8787]) and incubated with anti-GFP agarose for 4 h, followed by washing the affinity gel with Tris-buffered saline (TBS) (50 mM Tris-HCl, 150 mM NaCl, pH 7.4) four times. The proteins bound to the affinity gel were eluted by 0.1 M glycine HCl (pH 2.5) and were detected by anti-S and anti-GFP antibod- ies (Abmart, Shanghai, China) [11]. Yeast two-hybrid assay Full-length cDNA of MoPth11, MoWish, and MoSho1 (see primers in S1 Table) was cloned into the prey pGADT7 and the bait pGBKT7-MoErv14 constructs, respectively, and the plas- mid pairs were co-transformed into the yeast strain AH109 according to the description (Clontech, USA). The transformants from SD-Trp-Leu plates were then isolated and spotted on SD-Trp-Leu-His and SD-Trp-Leu-His-Ade media for further growth testing. MoERV gene family deletion and complementation Five ΔMoerv mutants were generated by using the one-step gene replacement strategy. Two fragments with 1.0 kb of sequences flanking the targeted gene were PCR amplified with primer pairs (S1 Table). The resulting PCR products were ligated to the hygromycin resistance cas- sette (HPH) released from pCX62. The 3.4-kb fragment, which included the flanking sequences and the HPH cassette, was transformed into Guy11 protoplasts [31]. Putative mutants were screened by PCR and further confirmed by Southern blot analysis. The comple- ment fragment, which contains the entire MoERV14 gene coding region and its native pro- moter region, was amplified by PCR with primers (S1 Table) and inserted into the pYF11 construct to complement the respective mutant strain [32]. Conidial germination and appressorium formation Conidial germination and appressorium formation were measured on a hydrophobic surface as described before [33–35]. Conidial suspensions of 25 μL (5×104 spores/mL) were dropped onto a hydrophobic surface and placed in a humidified box at 28˚C. The appressorium forma- tion rate was counted at 24 h post inoculation (hpi) under a microscope, and more than 200 appressoria were counted for each strain. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 14 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae Host penetration and pathogenicity assays Conidia were collected from mutants grown on SDC agar for 10 days and re-suspended by 0.2% (w/v) gelatin solution to a concentration of 5×104 spores/ml. For spraying assay, two- week-old rice seedlings (O. sativa cv. CO39) were sprayed with 4 ml of the conidial suspension of each treatment and kept in a growth chamber at 25˚C with 90% humidity in the dark for the first 24 h, followed by a 12-h-light and 12-h-dark cycle [36,37]. Lesion formation was daily checked and photographed after seven days of inoculation. The 7-day-old barley leaves were drop-inoculated with three droplets (20 μl) of conidial suspension, and photographs were taken 5 days after infection. Each experiment was repeated more than three times, and the experimental condition was kept consistent (e.g., temperature, humidity, illumination, and the age of the plants). For the ‘relative fungal growth’ assay, total DNA was extracted from 1.5 g disease leaves and tested by qRT-PCR (HiScript II Reverse Transcriptase, Vazyme Biotech Co., Nanjing, China) with 28S/Rubq1 primers (S1 Table) [38,39]. For rice sheath penetration and invasive hyphae expansion, conidial suspension (1×105 spores/ml) was inoculated into the sheaths. After incubation for 36 h at 28˚C, the sheath cuticle cells were observed under a Zeiss Axio Observer A1 inverted microscope [40]. Organelle isolation Protoplasts were prepared and maintained in a buffer containing 100 mM Tris-HCl, 0.1 mM MgCl2, 10 mM DTT, and 1.1 M sorbitol plus proteinases inhibitor mix (8215, Sigma-Aldrich), as previously described [32]. Organelles were isolated according to the established protocols with minor modifications as follows. Briefly, protoplasts were lysed in a buffer containing 10 mM Tris-HCl, 0.5 mM MgCl2, 8% Ficoll, and a proteinase inhibitor mixture, and the lysates were placed on the top of a centrifuge tube containing a bottom layer of the lysis buffer (10 ml) and a top layer of the same buffer containing 4% Ficoll (10 ml). Intact organelles, which are viewed as an opaque layer at the top of the centrifuge tube following centrifugation (50 000 g, 45 min), were collected. The supernatants were loaded into a cushion of 18% iodixanol (Sigma-Aldrich) for density gradient self-generation. After precipitation at 100 000 g for 2 h in a swinging bucket rotor, the crude membranes were collected at the interface, adjusted to 16% in iodixanol, and spun again at 350 000 g for 3 h. Fractions of 0.5 ml were harvested and washed with 0.8 ml 160 mM Na2CO3 for 30 min at 4˚C. The membrane fraction was precipi- tated (100 000 g), washed with water, and then precipitated again. Membranes were solubilized in a buffer (0.1 ml of 25 mM triethylammonium bicarbonate/8 M urea/2% Triton X-100/0.1% SDS), and concentrations were determined. FRAP assay Germinated conidia with 3 h of incubation were treated with cycloheximide and benomyl as described [19]. FRAP was performed using a fluorescence microscope Zeiss LSM710. Regions containing MoPth11-GFP, MoWish-GFP, and MoSho1-GFP in germ tubes were selected for photo-bleaching. Photobleaching was carried out using an Argon-multiline laser at a wave- length of 488 nm with 90% laser power and 150 iterations in ROI. Images were acquired with 2% laser power at a wavelength of 488 nm every 5 sec. For quantitative analyses, the fluores- cence intensity was measured using the ZEISS ZEN blue software, and fluorescence recovery curves were fitted using the following formula: F(t) = Fmin + (Fmax—Fmin)(1-exp-kt), where F(t) is the intensity of fluorescence at time t, Fmin is the intensity of fluorescence immediately post-bleaching, Fmax is the intensity of fluorescence following complete recovery, and k is the rate constant of the exponential recovery. Mobile Fraction was calculated as the following for- mula: Mf = (Fend—F0)/(Fpre—F0), where Fend is the stable fluorescent intensity of the PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 15 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae punctae after sufficient recovery, F0 is the fluorescent intensity immediately after bleaching, and Fpre is the fluorescent intensity before bleaching. The germinated conidia (3 h) with cycloheximide inhibit protein biosynthesis, which may prevent Golgi resident MoPth11-GFP, MoWish-GFP, and MoSho1-GFP from entering endosomes. The germinated conidia also treated with benomyl for 10 min to inhibit endosome trafficking. The extraction and purification of vesicles About 5 g mycelia were ground in liquid nitrogen with a mortar and pestle, and the mycelia powder was suspended in 15 ml of 0.1 M sodium acetate (containing 0.07% β-mercaptoetha- nol), then stirred at 4˚C for 2 h. The mixture was centrifuged at 6500 g for 30 min and the supernatant was transferred to a new centrifuge tube. The precipitation was resuspended in 15 ml 0.1 M sodium acetate containing 0.07% β-mercaptoethanol. After being well-suspended, the mixture was centrifuged at 6500 rpm for 30min, and the supernatant was mixed with the previous one. In order to remove the debris, the mixed supernatant was centrifuged at 7500 g for 30 min at 4˚C. The supernatant was again centrifuged at 50000 g for 90 min. The superna- tant was removed and the precipitate dissolved in 1 ml of TMD buffer (50 mM Tris pH 7.5, 10 mM MgCl2, 5 mM DTT) for 30 min. The mixture was again precipitated at 8000 g for 3 min at 4˚C, and glycerol was added to a final concentration of 15% to the supernatant. The suspen- sion was stored at -80˚C. The lysis and extraction of vesicle proteins For protein extraction, 400 μl vesicle samples were dissolved with 200 μl of lysis buffer (5 M urea, 2 M thiourea, 2% CHAPS, 2% SB 3–10, 40 mM Tris, 0.07% β-mercaptoethanol, 1 mM PMSF) and mixed uniformly. Proteins were precipitated by adding two volumes of pre-cold acetone containing 10% TCA and 0.07% β-mercaptoethanol and kept at -20˚C for 30 min before precipitating at 13000 g, 4˚C for 30 min. The pellet was washed with pre-cold acetone 3 times and air dried. The protein sample was re-dissolved in 200 μl of lysis buffer and centri- fuged at 13000 g to increase impurity. Further purification was carried out using the 2D Clean-Up kit (GE Healthcare). Supporting information S1 Fig. Targeted genes knockout strategy and confirmation by Southern blot analysis. (A-E) The strategy of knocking out target genes in M. oryzae genome. Thin lines below the arrows indicate the probe sequence of each gene. Southern blot analysis was used to confirm the MoERV14 deletion and the copy of the HPH gene. (TIF) S2 Fig. MoErv14 is homolog of ScErv14. MoERV14 could partially suppress the growth defect of the yeast ΔScerv14 mutant under 30 μg/ml DTT stress. The yeast ΔScERV14 mutant was complemented with MoERV14 cDNA. The yeast wild type-strain BY4741 and the ΔScerv14 mutant transformed with the empty pYES2 vector were used as controls. Serial dilutions of cell suspensions of each strain were spotted on SD and SD+DTT plates for 5 days and photo- graphed. (TIF) S3 Fig. MoErv14 is involved in appressorium turgor generation. Statistical analysis of the collapsed appressoria on hydrophobic surfaces after 24 h incubation. Error bars represent ±SD, and asterisks represent significant differences (p < 0.01). (TIF) PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 16 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae S4 Fig. MoErv14 is dispensable for septin-actin assembly. Septin network in appressoria (24 h) of Guy11 and ΔMoerv14 mutant.Bars, 5 μm. (TIF) S5 Fig. MoErv14 is a vesicle protein and located at ER and Golgi. (A and B) The localization pattern of MoErv14 in conidia and appressorium phase. Left, the observation of MoErv14-GFP with ER marker MoLsh1. Right, the observation of MoErv14-GFP with Golgi marker MoSft2. Bars, 10 μm (C) Organelles from M. oryzae protoplasts were partially sepa- rated by centrifugation. The ER and Golgi distribution were analyzed by Western blotting, using RFP antibodies against the ER marker MoLhs1 and Golgi marker MoSft2 fused with RFP. Distribution of MoErv14 was detected by anti-s antibody. (D) The extraction of the vesic- ular proteins from MoErv14-GFP/MoSec24-2-RFP co-transformed strains and used the west- ern blot to detect MoErv14 and MoSec24-2-RFP. The MoSec24-2-RFP and anti-tubulin were used as references. (TIF) S6 Fig. MoErv14 is dispensable for endocytosis. Hyphae stained by FM4-64 were examined by using fluorescence microscopy at different time points to observe the FM4-64 uptake. Bars, 10 μm. (TIF) S7 Fig. MoErv14 is important for intracellular cAMP generation and MoPmk1 phosphory- lation. (A) Loss of MoERV14 leads to decreased accumulation of cAMP. Bar chart showing quantification of intracellular cAMP in the mycelia of the indicated strains cultured for 2 days in complete medium. Two biological repetitions with three replicates were assayed. The error bars represent SD of three replicates. The asterisks denote statistical significances (p<0.01). (B) The total protein of Guy11 and ΔMoerv14 mutant strains were isolated from mycelia for detecting the MoPmk1 phosphorylation level using the anti-phospho-p44/42 MAP kinase antibody (Cell Signaling Technology) and the anti-p44/42 MAP kinase antibody (Cell Signal- ing Technology) was used as control. Three independent experiments were replicated that showed similar results. The asterisks denote statistical significance (p<0.01). (TIF) S8 Fig. There is no interaction detected among MoPth11, MoWish, and MoSho1. (A-C) Yeast two-hybrid assay for interactions among MoPth11, MoWish, and MoSho1. (TIF) S9 Fig. MoPth11, MoWish and MoSho1 are detained in ER with BFA treatment. Three cell membrane proteins fused with GFP, then the transformats were treated with BFA and co- localized with MoLhs1-RFP. Images were observed by Axio Observer A1 Zeiss inverted micro- scope. Bar = 10 μm. (TIF) S1 Table. Primers used in this study. This work was supported by the Natural Science Foun- dation of China-German Research Foundation Mobility Programme (31861133017 to ZZG), and the China National Funds for Innovative Research Groups (Grant No.31721004 to ZZG), NSFC (31772110 to ZZG). QB received support from Natural Science Foundation of China Youth Programme (NSFC 32202240), and grant number BK20200543 fromYouth Program for Natural Science Foundation of Jiangsu Province. WP received support from grant number AI156254 and AI168867 of the National Institutes of Health (USA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 17 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae manuscript. (DOCX) Author Contributions Conceptualization: Bin Qian, Zhengguang Zhang. Data curation: Bin Qian, Xiaotong Su, Xinyu Liu, Zhengguang Zhang. Formal analysis: Bin Qian, Haifeng Zhang, Zhengguang Zhang. Funding acquisition: Bin Qian, Zhengguang Zhang. Investigation: Bin Qian, Ziyuan Ye, Xinyu Liu, Haifeng Zhang, Ping Wang, Zhengguang Zhang. Project administration: Bin Qian, Muxing Liu, Zhengguang Zhang. Resources: Xiaotong Su, Ziyuan Ye, Haifeng Zhang, Zhengguang Zhang. Software: Bin Qian, Xinyu Liu, Muxing Liu. Supervision: Xinyu Liu. Validation: Xiaotong Su, Muxing Liu, Zhengguang Zhang. Visualization: Xiaotong Su, Zhengguang Zhang. Writing – original draft: Bin Qian, Ping Wang, Zhengguang Zhang. Writing – review & editing: Ping Wang, Zhengguang Zhang. References 1. Zhang H, Zheng X, Zhang Z. The Magnaporthe grisea species complex and plant pathogenesis. Mol Plant Pathol. 2016; 17: 796–804. 2. Kulkarni RD, Thon MR, Pan HQ, Dean RA. Novel G-protein-coupled receptor-like proteins in the plant pathogenic fungus Magnaporthe grisea. Genome Biol. 2005; 6(3): R24. 3. DeZwaan TM, Carroll AM, Valent B, Sweigard JA. Magnaporthe grisea Pth11p is a novel plasma mem- brane protein that mediates appressorium differentiation in response to inductive substrate cues. Plant Cell. 1999; 11(10): 2013–2130. 4. Sabnam N, Barman SR. WISH, a novel CFEM GPCR is indispensable for surface sensing, asexual and pathogenic differentiation in rice blast fungus. Fungal Genet Biol. 2017; 105: 37–51. https://doi.org/10. 1016/j.fgb.2017.05.006 PMID: 28576657 5. Futerman AH, Hirschberg K, Meivarlevy I, Rapaport E, Schwarz A, Sofer A, et al. Vesicle transport dur- ing Cell-Growth and in the maintenance of cell polarity. Biochem Soc Trans. 1995; 23(3): 530–534. https://doi.org/10.1042/bst0230530 PMID: 8566408 6. Kunz PJ, Barthel L, Meyer V, King R. Vesicle transport and growth dynamics in Aspergillus niger: Micro- scale modeling of secretory vesicle flow and centerline extraction from confocal fluorescent data. Bio- technol Bioeng. 2020; 117(9): 2875–2886. 7. Lord C, Ferro-Novick S, Miller EA. The highly conserved COPII coat complex sorts cargo from the endo- plasmic reticulum and targets it to the golgi. Cold Spring Harb Perspect Biol. 2013; 5(2): a013367. https://doi.org/10.1101/cshperspect.a013367 PMID: 23378591 8. Belden WJ, Barlowe C. Erv25p, a component of COPII-coated vesicles, forms a complex with Emp24p that is required for efficient endoplasmic reticulum to Golgi transport. J Biol Chem. 1996; 271(43): 26939–26946. https://doi.org/10.1074/jbc.271.43.26939 PMID: 8900179 9. Otte S, Belden WJ, Heidtman M, Liu J, Jensen ON and Barlowe C. Erv41p and Erv46p: New compo- nents of COPII vesicles involved in transport between the ER and Golgi complex. J Cell Biol. 2001; 152 (3): 503–518. https://doi.org/10.1083/jcb.152.3.503 PMID: 11157978 10. Powers J, Barlowe C. Erv14p directs a transmembrane secretory protein into COPII-coated transport vesicles. Mol Biol Cell. 2002; 13(3): 880–891. https://doi.org/10.1091/mbc.01-10-0499 PMID: 11907269 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 18 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae 11. Qian B, Su X, Ye Z, Liu X, Liu M, Shen D, et al. MoErv29 promotes apoplastic effector secretion contrib- uting to virulence of the rice blast fungus Magnaporthe oryzae. New Phytol. 2022; 233(3):1289–1302. 12. 13. Zhang H, Tang W, Liu K, Huang Q, Zhang X, Yan X, et al. Eight RGS and RGS-like proteins orchestrate growth, differentiation, and pathogenicity of Magnaporthe oryzae. PLoS Pathog. 2011; 7(12): e1002450. Liu X, Yang J, Qian B, Cai Y, Zou X, Zhang H, et al. MoYvh1 subverts rice defense through functions of ribosomal protein MoMrt4 in Magnaporthe oryzae. PLoS Pathog. 2018; 14(4): e1007016. 14. Yin Z, Feng W, Chen C, Xu J, Li Y, Yang L, et al. Shedding light on autophagy coordinating with cell wall integrity signaling to govern pathogenicity of Magnaporthe oryzae. Autophagy. 2020; 16(5): 900–916. 15. Zhang S, Liu X, Li L, Yu R, He J, Zhang H, et al. The ArfGAP protein MoGlo3 regulates the development and pathogenicity of Magnaporthe oryzae. Environ Microbiol. 2017; 19(10): 3982–3996. 16. Yi M, Chi M, Khang C, Park S, Kang S, Valent B, et al. The ER Chaperone LHS1 is involved in asexual development and rice infection by the blast fungus Magnaporthe oryzae. Plant Cell. 2009; 21(2): 681– 695. 17. Zhang S, Yang L, Li L, Zhong K, Wang W, Zhang H, et al. System-Wide characterization of MoArf GTPase family proteins and adaptor protein MoGga1 involved in the development and pathogenicity of Magnaporthe oryzae. Mbio. 2019; 10(5):e02398–19. 18. Chardin P, McCormick F. Brefeldin A: The advantage of being uncompetitive. Cell. 1999; 97(2): 153– 155. https://doi.org/10.1016/s0092-8674(00)80724-2 PMID: 10219235 19. 20. 21. Liu W, Zhou X, Li G, Li L, Kong L, Wang C, et al. Multiple plant surface signals are sensed by different mechanisms in the rice blast fungus for appressorium formation. PLoS Pathog. 2011; 7(1): e1001261. https://doi.org/10.1371/journal.ppat.1001261 PMID: 21283781 Li X, Gao C, Li L, Liu M, Yin Z, Zhang H, et al. MoEnd3 regulates appressoriumformation and virulence through mediating endocytosis in rice blast fungus Magnaporthe oryzae. PLoS Pathog. 2017; 13(6): e1006449. Li X, Zhong K, Yin Z, Hu J, Wang W, Li L, et al. The seven transmembrane domain protein MoRgs7 functions in surface perception and undergoes coronin MoCrn1-dependent endocytosis in complex with G subunit MoMagA to promote cAMP signaling and appressorium formation in Magnaporthe oryzae. PLoS Pathog. 2019; 15(2): e1007382. 22. Han P, Watanabe S, Shimada H, Sakamoto A. Dynamics of the leaf endoplasmic reticulum modulate beta-glucosidase-mediated stress-activated ABA production from its glucosyl ester. J Exp Bot. 2020; 71(6): 2058–2071. 23. Belden WJ, Barlowe C. Role of Erv29p in collecting soluble secretory proteins into ER-derived transport vesicles. Science. 2001; 294(5546): 1528–1531. https://doi.org/10.1126/science.1065224 PMID: 11711675 24. Bue CA, Bentivoglio CM, Barlowe C. Erv26p directs pro-alkaline phosphatase into endoplasmic reticu- lum-derived coat protein complex II transport vesicles. Mol Biol Cell. 2006; 17(11): 4780–4789. https:// doi.org/10.1091/mbc.e06-05-0455 PMID: 16957051 25. Otte S, Barlowe C. The Erv41p-Erv46p complex: multiple export signals are required in trans for COPII- dependent transport from the ER. Embo Journal. 2002; 21(22): 6095–6104. https://doi.org/10.1093/ emboj/cdf598 PMID: 12426381 26. Herzig Y, Sharpe H, Elbaz Y, Munro S, Schuldiner M. A systematic approach to pair secretory cargo receptors with their cargo suggests a mechanism for cargo selection by Erv14. PLoS Biol. 2012; 10(5): e1001329. https://doi.org/10.1371/journal.pbio.1001329 PMID: 22629230 27. He M, Su J, Xu Y, Chen J, Chern M, Lei M, et al. Discovery of broad-spectrum fungicides that block sep- tin-dependent infection processes of pathogenic fungi. Nature Microbiology. 2020; 5(12): 1565–1575. https://doi.org/10.1038/s41564-020-00790-y PMID: 32958858 28. Qian B, Liu X, Ye Z, Zhou Q, Liu P, Yin Y, et al. Phosphatase-associated protein MoTip41 interacts with the phosphatase MoPpe1 to mediate crosstalk between TOR and cell wall integrity signalling during infection by the rice blast fungus Magnaporthe oryzae. Environ Microbiol. 2021; 23(2): 791–809. 29. Qian B, Liu X, Jia J, Cai Y, Chen C, Zhang H, et al. MoPpe1 partners with MoSap1 to mediate TOR and cell wall integrity signalling in growth and pathogenicity of the rice blast fungus Magnaporthe oryzae. Environ Microbiol. 2018; 20(11): 3964–3979. 30. Feng W, Yin Z, Wu H, Liu P, Liu X, Liu M, et al. Balancing of the mitotic exit network and cell wall integ- rity signaling governs the development and pathogenicity in Magnaporthe oryzae. PLoS Pathog. 2021; 17(1): e1009080. 31. Yin Z, Chen C, Yang J, Feng W, Liu X, Zuo R, et al. Histone acetyltransferase MoHat1 acetylates autop- hagy-related proteins MoAtg3 and MoAtg9 to orchestrate functional appressorium formation and PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 19 / 20 PLOS PATHOGENSMoErv14 mediates transmembrane sensor transportation to control appressorium formation in M. oryzae pathogenicity in Magnaporthe oryzae. Autophagy. 2019; 15(7): 1234–1257. https://doi.org/10.1080/ 15548627.2019.1580104 PMID: 30776962 32. Yu R, Shen X, Liu M, Liu X, Yin Z, Li X, et al. The rice blast fungus MoRgs1 functioning in cAMP signal- ing and pathogenicity is regulated by casein kinase MoCk2 phosphorylation and modulated by mem- brane protein MoEmc2. PLoS Pathog. 2021; 17(6): e1009657. https://doi.org/10.1371/journal.ppat. 1009657 PMID: 34133468 33. Guo M, Chen Y, Du Y, Dong Y, Guo W, Zhai S, et al. The bZIP transcription factor MoAP1 mediates the oxidative stress response and is critical for pathogenicity of the rice blast fungus Magnaporthe oryzae. PLoS Pathog. 2011; 7(2): e1001302. 34. Guo Z, Liu X, Wang N, Mo P, Shen J, Liu M, et al. Membrane component ergosterol builds a platform for promoting effector secretion and virulence in Magnaporthe oryzae. New Phytol. 2023; 237(3): 930– 943. 35. 36. 37. Li Y, Liu X, Liu M, Wang Y, Zou Y, You Y, et al. Magnaporthe oryzae auxiliary activity protein MoAa91 functions as chitin-binding protein to induce appressorium formation on artificial inductive surfaces and suppress plant immunity. mBio. 2020 11(2):e03304–19. https://doi.org/10.1128/mBio.03304-19 PMID: 32209696 Liu M, Hu J, Zhang A, Dai Y, Chen W, He Y, et al. Auxilin-like protein MoSwa2 promotes effector secre- tion and virulence as a clathrin uncoating factor in the rice blast fungus Magnaporthe oryzae. New Phy- tol. 2021; 230(2): 720–736. Liu X, Zhou Q, Guo Z, Liu P, Shen L, Chai N, et al. A self-balancing circuit centered on MoOsm1 kinase governs adaptive responses to host-derived ROS in Magnaporthe oryzae. Elife. 2020; 9: e61605. https://doi.org/10.7554/eLife.61605 PMID: 33275098 38. Qi Z, Liu M, Dong Y, Zhu Q, Li L, Li B, et al. The syntaxin protein (MoSyn8) mediates intracellular traf- ficking to regulate conidiogenesis and pathogenicity of rice blast fungus. New Phytol. 2016; 209(4): 1655–1667. https://doi.org/10.1111/nph.13710 PMID: 26522477 39. 40. Zhong K, Li X, Le X, Kong X, Zhang H, Zheng X, et al. MoDnm1 dynamin mediating peroxisomal and mitochondrial fission in complex with MoFis1 and MoMdv1 is important for development of functional appressorium in Magnaporthe oryzae. PLoS Pathog. 2016; 12(8): e1005823. Liu M, Zhang S, Hu J, Sun W, Padilla J, He Y, et al. Phosphorylation-guarded light-harvesting complex II contributes to broad-spectrum blast resistance in rice. Proc Natl Acad Sci U S A. 2019; 116(35): 17572–17577. https://doi.org/10.1073/pnas.1905123116 PMID: 31405986 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1011251 April 3, 2023 20 / 20 PLOS PATHOGENS
10.2196_44330	JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Original Paper Text Analysis of Trends in Health Equity and Disparities From the Internal Revenue Service Tax Documentation Submitted by US Nonprofit Hospitals Between 2010 and 2019: Exploratory Study Emily Hadley, MS; Laura Haak Marcial, PhD; Wes Quattrone, MA; Georgiy Bobashev, PhD RTI International, Durham, NC, United States Corresponding Author: Emily Hadley, MS RTI International 3040 East Cornwallis Road Durham, NC, 27514 United States Phone: 1 919 541 6000 Email: [email protected] Abstract Background: Many US hospitals are classified as nonprofits and receive tax-exempt status partially in exchange for providing benefits to the community. Proof of compliance is collected with the Schedule H form submitted as part of the annual Internal Revenue Service Form 990 (F990H), including a free-response text section that is known for being ambiguous and difficult to audit. This research is among the first to use natural language processing approaches to evaluate this text section with a focus on health equity and disparities. Objective: This study aims to determine the extent to which the free-response text in F990H reveals how nonprofit hospitals address health equity and disparities, including alignment with public priorities. Methods: We used free-response text submitted by hospital reporting entities in Part V and VI of the Internal Revenue Service Form 990 Schedule H between 2010 and 2019. We identified 29 main themes connected to health equity and disparities, and 152 related key phrases. We tallied occurrences of these phrases through term frequency analysis, calculated the Moran I statistic to assess geographic variation in 2018, analyzed Google Trends use for the same terms during the same period, and used semantic search with Sentence-BERT in Python to understand contextual use. Results: We found increased use from 2010 to 2019 across all the 29 phrase themes related to health equity and disparities. More than 90% of hospital reporting entities used terms in 2018 and 2019 related to affordability (2018: 2117/2131, 99.34%; 2019: 1620/1627, 99.57%), government organizations (2018: 2053/2131, 96.33%; 2019: 1577/1627, 96.93%), mental health (2018: 1937/2131, 90.9%; 2019: 1517/1627, 93.24%), and data collection (2018: 1947/2131, 91.37%; 2019: 1502/1627, 92.32%). The themes with the largest relative increase were LGBTQ (lesbian, gay, bisexual, transgender, and queer; 1676%; 2010: 12/2328, 0.51%; 2019: 149/1627, 9.16%) and social determinants of health (958%; 2010: 68/2328, 2.92%; 2019: 503/1627, 30.92%). Terms related to homelessness varied geographically from 2010 to 2018, and terms related to equity, health IT, immigration, LGBTQ, oral health, rural, social determinants of health, and substance use showed statistically significant (P<.05) geographic variation in 2018. The largest percentage point increase was for terms related to substance use (2010: 403/2328, 17.31%; 2019: 1149/1627, 70.62%). However, use in themes such as LGBTQ, disability, oral health, and race and ethnicity ranked lower than public interest in these topics, and some increased mentions of themes were to explicitly say that no action was taken. Conclusions: Hospital reporting entities demonstrate an increasing awareness of health equity and disparities in community benefit tax documentation, but these do not necessarily correspond with general population interests or additional action. We propose further investigation of alignment with community health needs assessments and make suggestions for improvements to F990H reporting requirements. (J Med Internet Res 2023;25:e44330) doi: 10.2196/44330 KEYWORDS text mining; natural language processing; health care disparities; hospital administration https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Introduction Background Nonprofit hospitals in the United States are exempt from federal taxes. In exchange for this exemption, these hospitals have an obligation to provide community benefit [1]. The proof of compliance is collected with Schedule H, a form submitted as part of the annual Form 990 (F990) Internal Revenue Service (IRS) tax documentation for nonprofit hospitals. A substantial section of the F990 Schedule H (F990H) is composed of free-response (unstructured) text fields, where reporting entities can voluntarily provide details on community benefit spending. This may include discussion of community needs and the measures a hospital has or has not taken to address these needs. Community needs can and do include topics related to health equity and disparities. Health equity is commonly understood as an opportunity for all individuals to be healthy, regardless of membership in a group that has historically been economically or socially disadvantaged [2]. Health disparities are defined as a particular type of health difference that is worse among socially disadvantaged individuals, namely members of disadvantaged race or ethnicity groups, or economically disadvantaged people within any racial or ethnic group [2]. Addressing social determinants of health (SDOH), defined as the economic and social conditions that impact the health of people and communities, is considered a primary approach for reducing health disparities and achieving health equity [3]. In recent years, legislators and other stakeholders have paid increasing attention to whether hospitals are providing adequate community benefits to justify their tax-exempt status [4]. The IRS is required to review each tax-exempt hospital’s community benefit activities at least once every 3 years, although historically, this requirement has been ambiguous and difficult to track [5]. In 2020, the Government Accountability Office completed a review of the IRS’s implementation of requirements for tax-exempt hospitals and made a series of recommendations [5]. One recommendation was that the IRS establish a well-documented process for identifying hospitals at risk of noncompliance with the community benefit standard; the IRS added instructions in April and July 2021 for employees to document case files with relevant facts and circumstances considered during their review to determine whether the hospital organization satisfied the community benefit standard [5]. One unfulfilled recommendation is updating F990H to ensure that the community benefit a hospital is providing is clear and can be easily identified by Congress and the public [5]. The IRS recognizes that 3 of the factors currently addressed through open-ended narrative responses are not part of the quantitative, machine-readable files and that a revised F990H could more clearly, consistently, and comprehensively provide community benefit information to the public [5]. A related open recommendation is that Congress should specify which hospital services and activities are sufficient for community benefit [5]. These recommendations provide an opportunity for an explicit alignment with approaches to address health equity and disparities. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX Most existing studies that use data from F990H have focused on financial data. Empirical studies suggest that nonprofit hospital community benefit spending focuses on charity care and patient care services with little effort to improve community health [6-8]. Limited literature has explored community benefit spending with a focus on health equity or disparities, generally finding that increased IRS reporting clarification or explicit goals to address health disparities could more directly address community needs [9-11]. The text in F990H is unstructured data that varies in detail and length and has been historically challenging to analyze or review in large quantities. Only 1 study by Chen et al [12] reviewed this text in depth using a manual review of a small sample of 47 hospitals from 2015 to 2017. Recent advances in natural language processing techniques have made text analysis with much larger and longer text data sets more accessible [13]. Objective We present a novel text analysis of F990H tax documentation to understand if and how US nonprofit hospitals address health equity and disparities through community benefit programming. Our research is the first known work to analyze the F990H free-response text on a national scale across a 9-year period using text analytics approaches. We contribute to a larger body of work regarding hospital community benefits programming, including the limited existing discussion on how hospitals use community benefits programming to address health equity and disparities. By comparing our results with public search trends and identifying gaps in term use and action, we provide findings that stakeholders can use to advocate for community benefit approaches and improvements to F990H to better address health equity and disparities. Methods Data Source The data for this analysis come from free-response text in Part V and VI of Schedule H from F990s submitted by US nonprofit hospitals for tax years between 2010 and 2019 [14]. Detailed descriptions of the specific IRS requirements for Parts V and VI are included in Multimedia Appendix 1. Data from 2020 onward were not available at the time of analysis because of a data lag that has been exacerbated by the COVID-19 pandemic. F990H is submitted annually (although sometimes delayed by extensions) by a hospital facility or, in many cases, by a hospital organization with a shared employee identification number for multiple hospital facilities. All free-response answers for Part V and VI were combined for this analysis. Data were collected and maintained through the Community Benefit Insight project [15]. Analysis was completed in Python (Python Software Foundation) using pandas, numpy, nltk, PyTorch, and SentenceTransformers. Visualizations were created in Tableau (Tableau) and R (R Foundation for Statistical Computing) using ggplot2. The free-response text sections include answers to several questions regarding community health needs assessments (CHNAs), financial eligibility assistance programs, and descriptions of whether and how identified community needs are addressed by a hospital facility. With a few exceptions, these J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al questions are often ambiguous, and hospitals voluntarily choose the level of detail they provide [16]. The F990H for each hospital is reviewed by the IRS at least once every 3 years but rarely audited; an audit is more complicated and thorough than a review. Even if audited, there is no clear definition of the activities and services that are sufficient to prove community benefit [5]. Text responses are generally full sentences and paragraphs. Colloquial terminology and misspellings are infrequent. Figure 1 shows the number of hospital entities that report each year. As portions of the free-response text are required in the IRS Schedule H Instructions (Multimedia Appendix 1), every nonprofit hospital reporting entity is expected to have a free-response text entry; an average of 99.9% of hospital reporting entities that submitted the IRS Form 990 have completed the free-response text in Schedule H. Of the 2131 reporting hospital entities in 2018, a total of 90.5% (n=1930) had continuously submitted free-response text data since 2010. However, the overall number of hospital reporting entities decreased from 2010 to 2018, likely reflecting national trends in hospital consolidation and closure [17,18]. The number of reporting hospitals was notably low in 2019 (n=1627), likely because of the reporting extensions permitted in 2020 during the COVID-19 pandemic. Though we anticipate that 2019 is likely missing data from some hospitals, we decided to retain the 2019 data in the analysis as we prioritized the timeliness of the findings. The median number of words in text responses increased from 1629 in 2010 to 3439 in 2019, whereas the average number of words increased from 2840 in 2010 to 10,123 in 2019. The average was skewed by hospital reporting entities in California, Arizona, and Utah, which submitted average responses of over 30,000 words. Most hospitals (2406/2554, 94.21%) do not submit duplicate text across years. Figure 1. Count of hospital reporting entities from 2010 to 2019. The number of unique hospital reporting entities with free-response text data in Schedule H each year from 2010 to 2019. Outcomes and Variable Construction Term frequency analysis is a type of lexical analysis that searches for an exact word or phrase using a bag-of-words model [19]. In this analysis, we used the term frequency to flag whether a word or phrase was used one or more times by a particular hospital organization in each tax year. Term frequency analysis was proposed as an option in the first step of computational grounded theory in sociology, which combines expert human knowledge with the processing power and pattern recognition of computers for content analysis [20]. This study leveraged an opportunity from stakeholders that did not align with the more exploratory nature of the computational grounded theory approach. However, we used the key principles of computational grounded theory by deriving a list of terms, regularly seeking expert feedback, and validating our results. We also evaluated a semisupervised topic modeling approach that could have supported finding additional terms and topics related to key anchor words; however, we to use clearly defined expectations found that the suggested topics were too broad or included terms that were unrelated to the topic, as defined by stakeholders [21,22]. Improvements in semisupervised topic modeling in this context could be an area of future research. Term frequency analysis requires a list of words or phrases to begin with. A limitation of term frequency analysis is that it will only consider the exact words and phrases searched for, so a thorough and nontrivial multistep process combining text analysis approaches and stakeholder input was used to create the term set. We used a 3-step process to build this list. The first step in key term selection was providing stakeholders the opportunity to suggest specific words, phrases, and themes related to health equity and disparities. Stakeholders included 11 subject matter experts and programming staff with experience in health and community benefits from the Robert Wood Johnson Foundation and the Robert Wood Johnson Foundation grantees Community Catalyst and Healthy Food in Health Care. These stakeholders provided suggestions and agreed on words, https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al phrases, and themes in meetings and by email from fall 2021 to fall 2022. The terms included both single words (rural) and phrases (data collection). Terms with singular and specific meanings were selected. For example, specific drugs such as opioids and fentanyl were used for the substance use theme, as opposed to drug, which is broader and not always related to substance use. In some cases, both singular and plural versions of a term were included (ie, equity and equities). The second step involved n-gram tokenization of the text, followed by a search for the 1000 most common single words, bigrams, and trigrams. We reviewed the most common words and phrases. Common terms closely related to the themes suggested by the stakeholders were added to the full list of terms. For example, the terms listening tour and focus group were both added to the data collection theme through this process. The third step involved a review of the SDOH literature associated with the Healthy People 2030 initiative led by the US Office of Disease Prevention and Health Promotion for any other words or phrases that should be included [23]. For example, this review led to the addition of lead-based paint and air pollution to the environment theme. The final step was to provide the list of terms to a variety of stakeholders, including those from the first step, for feedback. The resulting term set spanned 29 themes and included 152 words and phrases. These words and themes are shown in Figure 2. A clean version of the text was created such that it was all lower case with no symbols or punctuation. Term frequency analysis was performed using this text. We ranked the use of these terms in 2019. We also calculated the percentage and percentage point change from 2010 to 2019. Percent change measurements are useful for understanding the relative increase in term use, especially for less-frequent themes. Percentage point change measurements are useful for understanding the raw change in term use. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Figure 2. Percent of hospital reporting entities with one or more uses of term in theme by tax year, ranked by use in 2019. This heat map highlights the number of hospital reporting entities with one or more uses of any term from a health equity or disparity theme by tax year. Darker tones indicate a higher use. The figure is sorted by ranking of use in the 2019 tax year. The figure also includes the relative percentage and percentage point changes from 2010 to 2019. BIPOC: Black, Indigenous, (and) People of Color; DACA: Deferred Action for Childhood Arrivals; LGBTQ: lesbian, gay, bisexual, transgender, and queer. Geographic Variation in F990H Use Community benefit programming is intended to align with community needs, which may vary by geographic region. To assess the alignment of F990H theme use by geography, we first aggregated the percentage of hospital reporting entities with one or more uses of a term by state for each year from 2010 to 2019 and mapped the findings. We visually reviewed the maps for trends in changes over time. We then calculated the Moran I statistic for 2018 to evaluate the presence of spatial autocorrelation to determine if there is a pattern of similarity between observations that are geographically close to one https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al another [24]. The data from 2018 were used as it is the most recent year with the most complete data. Neighboring states were identified through centroids up to 1000 km apart. We performed a 1-sided statistical significance test with an alternative hypothesis that the observed spatial autocorrelation in the data was significantly greater than what would be expected by chance under a null hypothesis of no spatial autocorrelation. Statistical significance was assessed at α values of .001, .01, and .05. We reviewed the top 20 sentences returned for each query and reported summary findings from the themes with the 3 largest percent or percentage point increase. We selected up to 3 sentences for each theme that best reflected the dichotomy between taking action and not taking action. Not all themes had examples of both action and inaction, and we intentionally did not seek to quantify the results from the semantic search, as it is an imperfect method that can return ambiguous or unrelated sentences. Comparison With Google Trends One criticism of hospital community benefits programming (and related spending) is that it is out of touch with community-identified needs [8]. Google Trends has emerged as a source for investigating how social trends change over time [25]. We used Google Trends as a proxy for the general population’s interest in topics related to health equity and disparities. We obtained the relative frequency of term use in Google searches related to health (as categorized by Google) between January 1, 2010, and December 31, 2019, in the United States, for the words and phrases in the term list. We note a limitation in that it is unclear how Google determines the category to which a search query belongs, and we are unable to determine the scope of the health category; however, for the purposes of this work, filtering to the health category is still preferable to using all search categories. Further details on Google Trends are provided in Multimedia Appendix 2. Considering only the searches in 2019, we ranked public interest in each of the 29 themes. We compared this Google Trends ranking with the ranking of use in F990H in 2019. We assigned similar relative usage to themes within 5 rankings of each other (eg, Nutrition has rank 11 in 2019 F990H use and rank 10 in 2019 in Google Trends searches; Nutrition is considered a theme with “similar relative usage” as rank 11 is only 1 rank difference from rank 10). A difference of 6 to 18 rankings was considered large, whereas a difference of more than 18 rankings was considered very large. These thresholds were selected to ensure bands of similar width in the figure were used for comparison of the Google Trends and F990H results. Semantic Search A major criticism of term frequency analysis is that, although it is useful for determining whether a keyword or phrase is used, it is challenging to determine the context or meaning of the term. Therefore, a semantic search was used to augment term frequency analysis. This methodology is used by major search engines to search for meaning by evaluating both the searcher’s intent and the contextual meaning of the terms. We used the question-answer retrieval implementation of Sentence-BERT, a model pretrained on the Natural Questions data set which uses real questions from Google Search with annotated data from Wikipedia as the answers [26]. This approach is best for an asymmetrical search task in which a short query (such as a question or keyword) is used to find a sentence or paragraph. For this project, we built a semantic search model and used two search queries for each theme: (1) “took action on <theme>” and (2) “did not take action on <theme>.” https://www.jmir.org/2023/1/e44330 XSL•FO RenderX Ethical Considerations This research was completed using publicly available secondary data from hospital reporting entities and did not require institutional review board review because it did not involve human participants. All research was conducted with an ethic individuals, and of respect for cultures, communities, independent knowledge. Feedback was obtained from stakeholders likely to be impacted by the findings of this study. Results Overview Figure 2 shows the results of the term frequency analysis sorted by prevalence in 2019. Figures S1-S29 in Multimedia Appendix 3 provide the detailed disaggregation for each theme. Figure 2 illustrates that nearly every hospital organization uses a term related to affordability (2018: 2117/2131, 99.34%; 2019: 1620/1627, 99.57%), and more than 90% of hospital reporting entities used a term in 2018 and 2019 related to government organizations (2018: 2053/2131, 96.33%; 2019: 1577/1627, 96.93%), mental health (2018: 1937/2131, 90.9%; 2019: 1517/1627, 93.24%), and data collection (2018: 1947/2131, 91.37%; 2019: 1502/1627, 92.32%). The least used themes, with a prevalence of less than 10% in 2018 and 2019, were related to immigration (2018: 191/2131, 8.96%; 2019: 149/1627, 9.16%), LGTBQ (2018: 161/2131, 7.56%; 2019: 149/1627, 9.16%), and the environment (2018: 104/2131, 4.88%; 2019: 100/1627, 6.14%). Figure 2 provides additional details on the percentage increase and raw percentage increase in the use of a term at least once from 2010 to 2019. All 29 themes showed an increase in use in Schedule H of F990, as indicated by both relative and percentage point change. Although LGBTQ (lesbian, gay, bisexual, transgender, and queer)-related terms were used by only a small percentage of hospital reporting entities (Figure 2), this theme saw the largest relative increase from 2010 to 2019 (1676%; 2010: 12/2328, 0.51%; 2019: 149/1627, 9.16%). Other themes with large relative increases were SDOH (958%; 2010: 68/2328, 2.92%; 2019: 503/1627, 30.92%) and environment (522%; 2010: 23/2328, 0.99%; 2019: 100/1627, 6.15%). The themes with the smallest relative increase include affordability (2.06%; 2010: 2270/2328, 97.51%; 2019: 1620/1627, 99.57%) and insurance (13.39%; 2010: 1763/2328, 75.73%; 2019: 1450/1627, 89.12%). Terms related to substance use saw the largest raw percentage point increase: less than a fifth of hospital reporting entities used any substance use language in 2010 (403/2328, 17.31%), and more than two-thirds of hospital reporting entities used a substance use term in 2019 (1149/1627, 70.62%). Other themes J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al with notable increases included keywords related to oral health (48.26 percentage points, 2010: 429/2328, 18.43%; 2019: 1085/1627, 66.69%) and nutrition (40.31 percentage points, 2010: 853/2328, 36.64%; 2019: 1252/1627, 76.95%). Themes with the smallest percentage point increase included affordability (2.06 percentage points, 2010: 2270/2328, 97.51%; 2019: 1620/1627, 99.57%) and environment (5.16 percentage points, 2010: 23/2328, 0.99%; 2019: 100/1627, 6.15%). Geographic Variation in Themes From 2010 to 2019 Theme use can vary across states and time. In Figure 3, we highlight the theme with the most visually clear example of change in geographic variability, homelessness. In 2010, the percentage of hospital reporting entities using one or more key terms related to homelessness was small and similar across states. From 2012 to 2015, the proportion of hospital reporting entities using one or more key terms related to homelessness increased in states on the West Coast. In 2018 and 2019, the majority of hospital reporting entities on the West Coast (2018: 99/168, 58.9%; 2019: 115/161, 71.4%) used one or more terms related to homelessness. The geographic maps for all the themes are available in Multimedia Appendix 4. Although homelessness was the most visually striking theme for showing change across time, Table 1 highlights the results of the Moran test for spatial autocorrelation among themes in 2018. A total of 8 themes showed statistically significant (equity: P=.001; health IT: P=.02; immigration: P=.002; LGBTQ: P=.007; oral health: P=.04; rural: P<.001; SDOH: P<.001; substance use: P=.003) positive spatial autocorrelation, suggesting some degree of neighboring state clustering in the discussion of themes in F990H. The maps in Multimedia Appendix 4 help illustrate where clusters may occur. The equity theme was used in clusters of states on the East and West coasts, with less use in the Midwest, excluding a small cluster around Indiana. The health IT theme had higher use in a cluster of neighboring states, including Nebraska, South Dakota, Colorado, Montana, Utah, and Iowa as well as another cluster, including Indiana, Michigan, Ohio, Kentucky, and Illinois. The use of the immigration theme was clustered in the Pacific Northwest (Washington, Idaho, and Oregon) and a smaller cluster in the Northeast (Massachusetts, New York, New Jersey, and Connecticut). The LGBTQ theme was almost exclusively used in 3 clusters: West Coast (Washington, Oregon, and California), Midwest (Minnesota, Wisconsin, Illinois, Kentucky, and Oklahoma), and New England (Maine, Vermont, Massachuetts, and Connecticut), with some mention in New York and New Jersey. The oral health theme had a number of clusters, including Mountain West (Idaho, Montana, Wyoming, and Colorado), mid-Atlantic (Pennsylvania, New Jersey, Maryland, and Virginia), and New England (New Hampshire, Vermont, Massachusetts, Connecticut, and Rhode Island). The use of rural was higher in the Midwest (Wisconsin, Minnesota, Iowa, Missouri, Arkansas, Oklahoma, Kansas, North Dakota, and South Dakota) and Mountain West (Montana, Idaho, Utah, Colordado, and Oregon). The use of terms in the SDOH theme was higher in the West Coast (Washington, Oregon, and California) and North Atlantic (Maine, Massachusetts, New York, Rhode Island, New Jersey, and Maryland). Finally, the substance use theme had clusters in the Southwest (Utah, Arizona, New Mexico, and Colorado), Midwest (Wisoconsin, Ilinois, Missouri, Iowa, Ohio, Kentucky, and Tennessee) and on the East Coast (excluding North Carolina and Georgia). Figure 3. Geographic variation from 2010 to 2019 of the percentage of hospitals in each state using one or more key terms related to homelessness. A darker color indicates a higher use. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Table 1. Results from Moran test for spatial autocorrelation among states by theme in 2018. Theme Activities Affordability Child Chronic illness Criminal justice Data collection Disability Environment Equity Exercise Firearms Government organizations Health IT Homelessness Immigration Insurance Language LGBTQa Mental health Nutrition Oral health Poverty Race and ethnicity Rural SDOHb Senior Sexual health Substance use Tobacco Moran I statistic Variance P value 0.006 0.035 –0.048 –0.006 0.047 –0.036 0.012 –0.019 0.149 –0.036 –0.03 –0.021 0.082 0.055 0.118 –0.008 –0.049 0.099 0.011 –0.062 0.066 –0.03 0.038 0.179 0.208 0.059 0.063 0.12 0.037 0.002 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.002 0.003 0.003 0.002 0.003 0.003 0.002 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 0.003 .26 .12 .70 .38 .09 .61 .26 .48 .001 .61 .56 .50 .02 .07 .002 .40 .70 .007 .26 .78 .04 .56 .12 <.001 <.001 .05 .05 .003 .12 aLGBTQ: lesbian, gay, bisexual, transgender, and queer. bSDOH: social determinants of health. Google Trends Figure 4 shows the relative rank of use of a theme in hospital F990H reporting versus Google Trends in 2019. A rank of 1 was the most used, whereas a rank of 29 was the least used. Items in the middle band (eg, government organizations, chronic illness, and immigration) reflect similar relative use. Items in the bands near the top-left corner of the figure (eg, activity and insurance) reflect themes where the relative rank in Schedule H reporting is higher than that in Google Trends. Items in the bands toward the bottom-right corner of the figure (eg, LGBTQ, oral health, and disability) reflect themes where the relative rank in Google Trends is higher than in Schedule H. A total of 2 themes (government organizations and mental health) ranked in the top 5 and 1 theme (environment) ranked in the bottom 5 for relative use in both Google Trends and Schedule H in 2019. Figure 5 highlights the percentage change in relative use from 2010 to 2019 by hospitals for Google Trends versus F990H. Although the LGBTQ theme has greater use in Google Trends in Figure 4, this theme saw a much larger relative increase in use in the analysis timeframe in F990H. The SDOH and environment themes also increased considerably in use in F990H and saw some increase in use in Google Trends. The right panel in Figure 4 highlights that many themes had small and sometimes negative changes in Google Trends, even though the theme saw a substantial increase in use in F990H. Oral health and substance use were 2 themes with more than a 75% https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al increase in Google Trends searches (oral health weighted average Google Trends increase: 98.9%, 2010 less frequent Google Trends terms average use: 16.7, 2019 less frequent Google Trends terms average use: 32.78, 2010 more frequent Google Trends terms average use: 14.6, 2019 more frequent Google Trends terms average use: 29.6; substance use weighted average Google Trends increase: 77.7%, 2010 less frequent Google Trends terms average use: 1.9, 2019 less frequent Google Trends terms average use: 4.8, 2010 more frequent Google Trends terms average use: 9.4, 2019 more frequent Google Trends terms average use: 10.9) and more than a 250% increase in use in F990H text from 2010 to 2019 (oral health 261.9%, 2010: 429/2328, 18.43%; 2019: 1085/1627, 66.69%; substance use: 308% 2010: 403/2328, 17.31%; 2019: 1149/1627, 70.62%). Figure 4. Relative rank of themes in 2019 in Google Trends versus F990 Schedule H. This figure considers the relative rank of a health equity or disparity theme in Google Trends versus F990H tax documentation in 2019. The figure is segmented into bands that indicate when Google Trends’ relative use in 2019 is larger than F990H’s relative use, relative use is similar, or relative use in F990H is larger than in Google Trends. LGBTQ: lesbian, gay, bisexual, transgender, and queer; SDOH: social determinants of health. Figure 5. Percentage change in relative use of theme from 2010 to 2019 in Google Trends versus in F990 Schedule H. This plot compares the percentage change in the relative use of a health equity or disparity theme in Google Trends with the percentage change in relative use in F990 Schedule H forms from 2010 to 2019. The left panel shows all 29 themes. The right panel shows the 26 themes. LGBTQ: lesbian, gay, bisexual, transgender, and queer; SDOH: social determinants of health. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Semantic Search Key findings from the semantic search are shown in Table 2 for the most common phrases in themes with large percentage point increases in F990 Schedule H use from 2010 to 2019. These results highlight the presence of both action and nonaction statements in the free-response text submitted by hospitals. Table 2. Results from semantic search, separated into results that imply action and results that imply no action. Theme LBGTQa LBGTQ SDOHb SDOH SDOH Statements implying action “A LGBT+ business group was formed and is working on addressing diversity and inclusion in health care.” Statements implying no action None documented in search results. “We earned top marks for our commitment to equitable inclusive care for LGBT patients and their families.” None documented in search results. “This program tracked social determinant of health data.” None documented in search results. “The model addresses social determinants of health.” None documented in search results. “Social determinants of health were deemed an underlying current of all priorities.” None documented in search results. Environment “Helped educate participants about the dangers of excessive heat and injuries caused by extensive heat.” “The hospital will not address ‘climate and health’ be- cause the topic is not the hospital’s area of expertise.” Environment “For many years, we have been conducting lead risk assessments.” Environment “The program addressed enforcement of pest control and regular air filter changes with the housing authority in appropriate housing units.” “We have concluded that addressing these environmental issues falls outside of our mission.” “Improvement of air quality is not a part of the hospital’s mission.” Substance use “These programs target substance abuse prevention treatment and justice efforts.” “Will not take action on mental health and substance abuse.” Substance use “Promote mental health and prevent substance abuse.” “The only need not being address is substance abuse.” Substance use “Provide in-kind leadership and support to the implementation of the substance abuse action plan developed by the—behavioral health collaborative.” “The organization has chosen not to address alcoholism and alcohol abuse in adults.” Oral health Oral health Oral health Nutrition Nutrition “Reduce the number of oral health related emergency department visits.” “Also not being addressed at this time will be oral health.” “A consensus was reached that ongoing oral health efforts should be sustained.” “Oral health: the dental provider of the institution left the organization in FY18.” “Providers promote the importance of oral health to patients and where they can seek services in the area.” “Limited resources excluded this as an area chosen for action: oral health” “In 2014 the group introduced and implemented interventions to promote nutrition and reduce childhood obesity.” “The nutrition services provided are no longer covered by insurance or Medicare.” “This intervention aimed to increase activity levels and nutrition education among students at title 1 schools.” “Access to healthy food: this priority will not be ad- dressed.” aLGBTQ: lesbian, gay, bisexual, transgender, and queer. bSDOH: social determinants of health. Discussion Principal Findings Given the increased attention paid to health equity and disparities [27], it is gratifying to see greater use by hospital reporting entities of language related to these terms across all 29 health equity and disparity themes in this analysis. This increased use may suggest an improved hospital awareness of these issues or community needs related to these issues. It is worth noting that this increased use of words and phrases related to these specific themes is, with 2 exceptions, not required by the IRS and is therefore mostly voluntary on the part of hospitals. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX The 2 exceptions to voluntary reporting are the affordability and data collection themes. Both topics require explicit descriptions in community benefit documentation. Data collection is of particular interest, because the 2010 Affordable Care Act required the implementation of CHNAs, including data collection and focus groups, by 2013. Data collection was one of the themes with the largest percentage point increase, with a particularly large jump (approximately 22 percentage points) between 2012 and 2013. This reflects the impact that data collection legislation can have on both the implementation and documentation of meeting requirements. Some of the relative changes in theme use in the F990 documentation were parallel to national events. Substance use was a theme with a substantial percentage and percentage J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al increase from 2010 to 2019 in F990H. This corresponds to the second wave of the US opioid epidemic, which began in 2010 with a rapid increase in heroin overdoses, and the third wave, which began in 2013 with significant overdoses resulting from synthetic opioids [28]. Heroin overdoses continued to increase in 2016, whereas synthetic opioid overdoses continued to increase in 2019. Nutrition, a theme that includes the term obesity, also increased from 2010 to 2019 in F990H. This use likely reflects an increase in obesity in the United States, and obesity-related conditions are among the leading causes of preventable premature death [29]. SDOH gained increased attention in 2010 when the World Health Organization published the “Conceptual Framework for Action on the Social Determinants of Health” [30]. The substantial increase in use during a similar timeframe suggests growing awareness among hospitals regarding the SDOH framework. Variation in use can also be attributed to geography. Figure 3 shows that the clearest depiction of geographic differences as the percentage of reporting entities in states on the West Coast, using key terms related to homelessness, grew substantially between 2010 and 2019. This may reflect on the fact that WA, OR, and CA were among the top 10 states with the highest rates of homelessness in 2019 [31]. However, of these 3 states, only CA was also among the states with a high percentage increase in homelessness from 2010 to 2019. Other states with a large increase in documented homelessness from 2010 to 2019 included NY, SD, KS, and MA, but among these, only MA saw a corresponding increase in homelessness terms in F990H [31]. Chen et al [12] also documented lower mentions of investment in housing in F990H, with only 12 of 47 hospital organizations in 5 cities with high rates of homelessness reporting housing-related spending between 2015 and 2017. These findings suggest a lack of concordance in community needs and hospital community benefits spending in regions with high current or markedly increasing rates of homelessness. Chen et al [12] suggested that hospitals should be provided with evidence-based strategies from early adopters of homelessness strategies to see how housing may fit within their purview and that F990H instructions should be updated so that hospitals are getting adequate credit for housing investment. The Moran results in Table 1 highlight other themes with discrepancies in F990H theme use between neighboring states and regions with a documented need. For example, oral health has clusters of F990H use in the Mountain West, mid-Atlantic, and New England, but all the states in the clusters with higher F990H oral health use are also among states with medium to high proportions of adults reporting in 2018 that they had visited a dentist or dental clinic within the past year [32]. As regular dental visits are important to both oral health and overall wellness, it is notable that no state among those with the lowest proportion of previous year adult dental visits was among the states with the highest mentions of oral health in F990H [33]. The F990H use of the immigration theme was clustered in the Pacific Northwest and mid-Atlantic and distinctly lower in the 3 states where nearly half (45%) of US immigrants live: California, Texas, and Florida [34]. This insight aligns with the documented finding that adult immigrants, regardless of immigration and citizenship status, are underserved in the US https://www.jmir.org/2023/1/e44330 XSL•FO RenderX health care system [35]. Not all F990H use was misaligned—the use of rural was high in states from more rural regions (Midwest and Mountain West) and showed some overlap with the health IT theme, potentially reflecting the pre–COVID-19 pandemic emphasis on the use of health IT approaches such as telehealth in rural communities [36,37]. The substance use theme was highest in the Southwest, Midwest, and East Coast, all regions where most states maintained similar opioid overdose rates between 2017 and 2018 [38]. However, the remaining themes geographic with clustering—equity,LGBTQ, and SDOH—are all themes with national applicability that transcend the few clusters with higher F990H use. The clustered use in the West Coast and North Atlantic, with some sporadic clustered use in the Midwest, suggests that these terms may be politically charged and more commonly used in politically liberal states [39]. statistically significant Themes including LGBTQ, disability, oral health, and race and ethnicity are more highly ranked in Google Trends than in F990H, suggesting a degree of misalignment in general public interests and hospital activities as described in F990H. Individuals may be more interested in how these personal topics impact them, whereas hospital reporting entities discuss insurance and activity themes much more in F990H than in the broader population. This is likely expected with the current version of F990H as it includes specific questions related to how hospitals address uninsured patients but has ambiguous questions requesting information on how a hospital is addressing “significant needs” as identified in a CHNA or why a significant need is not being addressed. Given that a major motivation of community benefits reporting is to ensure that nonprofit hospitals are addressing community needs, the misalignment in ranking of some themes in Google Trends health searches and F990H free-response text suggests that a revamp of F990H with more explicit and granular community needs questions may generate greater accountability for addressing health equity and disparities. A notable limitation of the Google Trends comparison is that the trends are from national searches, whereas hospital reporting entities generally aim to align with local community needs. Although national searches may not always be applicable to a local community, attention to larger trends may still help hospital entities that are missing opportunities to address important rising topics that are still relevant to their community. For example, the increased attention in the past decade to LGBTQ rights and disparities by race and ethnicity suggests that both LGBTQ and race and ethnicity have national applicability, and the Google Trends results suggest that they are currently more prioritized by the public than by hospitals. More hospitals could seek to discuss how they are addressing these themes in the F990H. The increased use of terms related to health equity and disparities is promising, but the semantic search results in Table 2 make it clear that increased use of terms may not necessarily correspond to increased community benefit programming. Of the 6 themes with the largest relative or raw increase, 4 themes had results with hospitals explicitly stating their inaction. A common reason for lack of action often appears to be a lack of resources or services or misalignment with the hospital’s J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al mission. Young et al [7] found that most benefit-related expenditures were related to patient care, rather than community health improvement. Sapirstein et al [9] found no evidence of dramatic shifts in community benefit spending from 2014 to 2019. Hospitals may be mentioning community-related themes more but may not actually take additional action or allocate funding to new community benefits themes. Governing authorities such as the IRS could better scrutinize these hospital statements, such as the inconsistency in Table 2, where 1 hospital says that improvement of air quality does not align with a hospital’s mission, whereas a different hospital says that it provides air filters to a housing authority. Similar to the recommendation of Chen et al [12] for improved reporting of efforts to address homelessness in F990H, the IRS should clarify that providing air filters to meet a community need is mission-aligned and credit hospitals reporting efforts such as this on F990H [12]. Although it is exciting to see that hospitals use more language related to health equity and disparities, it cannot be presumed that there is a corresponding increase in activities related to these needs; mentions in text alone cannot prove action. Policy makers should consider additional language in F990H that requires a clear description of health equity and disparities, including explicit recognition of work on SDOH as a community benefit [10,11]. The Government Accountability Office has called for updating Form 990, including Schedule H, to more clearly, consistently, and comprehensively describe community benefit activities, as well as for Congress to specify which services and activities are sufficient to meet community benefit standards [5]. These updates could be used to improve auditing of community benefits, contribute to efforts to score hospitals on community benefits programming, or highlight innovative hospitals providing exemplary community benefits. Greater transparency, documentation of activities, and community benefits–specific IRS audit processes for F990H could lead to increased accountability and action by hospitals to address community health equity and disparities. In their study, Chen et al [12] also recommended that the IRS seek to ensure greater alignment of F990H activities with CHNAs [12]. The study by Carlton and Singh [40] found that joint CHNAs with hospital-local health department collaboration encouraged greater hospital investment in community health improvement activities. Further research can use text analytics to explore the programs that hospitals describe in F990H and assess their alignment with implementation plans in CHNAs. Conclusions We created a health equity and disparities term list and showed increased use of terms across all 29 themes by hospital reporting entities in free-response text submitted annually from 2010 to 2019 in F990H. We found variations across years and geographies. We suggest that hospitals demonstrate an increased awareness of health equity and disparities yet also show potential misalignment with public interests, as demonstrated through Google Trends and varying changes in action or programming with semantic search. Further research can continue to explore the degree to which hospitals have satisfactorily addressed community needs, as described in the free-response text. Policy changes to the F990H could improve transparency and accountability related to hospital community benefit efforts to address health equity and disparities. Acknowledgments The authors would like to thank Mark Rukavina and colleagues at the Community Catalyst and the staff at the Robert Wood Johnson Foundation for their inputs. Funding for this work was provided by the Robert Wood Johnson Foundation grant 77387. Conflicts of Interest None declared. Multimedia Appendix 1 Details the requirements of Form 990 Schedule H Part V Section C and Schedule H Part VI. [DOCX File , 18 KB-Multimedia Appendix 1] Multimedia Appendix 2 Additional details on the Google Trends methodology. [DOCX File , 12 KB-Multimedia Appendix 2] Multimedia Appendix 3 Details of the percentage of hospital reporting entities with one or more uses of a word or phrase in each theme in a given tax year. The figures are sorted according to the percentage of hospitals using one or more terms in 2019. [DOCX File , 2070 KB-Multimedia Appendix 3] Multimedia Appendix 4 The figures provide additional details of the geographic variation for each theme by illustrating the percentage of hospital reporting entities by state with one or more uses of a particular theme. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al [DOCX File , 6044 KB-Multimedia Appendix 4] References 1. 2. 3. 4. 5. 6. 7. 8. 9. Ortiz A. Tracking community benefit spending. The Medical Care Blog. 2021. URL: https://www.themedicalcareblog.com/ community-benefit-spending/ [accessed 2022-08-25] Braveman P. What are health disparities and health equity? We need to be clear. Public Health Rep 2014 Jan;129 Suppl 2(Suppl 2):5-8 [FREE Full text] [doi: 10.1177/00333549141291S203] [Medline: 24385658] Frequently asked questions, NCHHSTP social determinants of health. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/nchhstp/socialdeterminants/faq.html [accessed 2022-08-29] Atkeson A. How states can hold hospitals accountable for their community benefit expenditures. National Academy for State Health Policy. 2021 Mar 15. URL: https://www.nashp.org/ states-can-hold-hospitals-accountable-for-their-community-benefit-expenditures/ [accessed 2022-08-29] Tax administration: opportunities exist to improve oversight of hospitals' tax-exempt status. U.S. Government Accountability Office. 2020 Sep 17. URL: https://www.gao.gov/assets/gao-20-679.pdf [accessed 2022-02-01] Schneider H, Yilmaz H. Hospital community benefits and the effect of Schedule H: a difference-in-difference approach. Health 2013 Oct;5(10):1681-1688 [FREE Full text] [doi: 10.4236/health.2013.510226] Young GJ, Chou CH, Alexander J, Lee SY, Raver E. Provision of community benefits by tax-exempt U.S. hospitals. N Engl J Med 2013 Apr 18;368(16):1519-1527 [FREE Full text] [doi: 10.1056/nejmsa1210239] Rubin DB, Singh SR, Young GJ. Tax-exempt hospitals and community benefit: new directions in policy and practice. Annu Rev Public Health 2015 Mar 18;36:545-557. [doi: 10.1146/annurev-publhealth-031914-122357] [Medline: 25785895] Sapirstein A, Rao A, Steimle LN. Alignment of community benefit spending and initiatives to improve community health: is there evidence of progress? medRxiv 2022 Aug [FREE Full text] [doi: 10.1101/2022.08.17.22278878] 10. Rozier MD. Nonprofit hospital community benefit in the U.S.: a scoping review from 2010 to 2019. Front Public Health 2020 Mar 11;8:72 [FREE Full text] [doi: 10.3389/fpubh.2020.00072] [Medline: 32219089] 11. Rozier M, Goold S, Singh S. How should nonprofit hospitals' community benefit be more responsive to health disparities? AMA J Ethics 2019 Mar 01;21(3):E273-E280 [FREE Full text] [doi: 10.1001/amajethics.2019.273] [Medline: 30893042] 12. Chen KL, Chen K, Holaday LW, Lopez 3rd L. Assessing concordance across nonprofit hospitals' public reporting on housing as a community health need in the era of the affordable care act. J Public Health Manag Pract 2022 Mar;28(2):E615-E618 [FREE Full text] [doi: 10.1097/PHH.0000000000001357] [Medline: 33938486] 13. Hirschberg J, Manning CD. Advances in natural language processing. Science 2015 Jul 17;349(6245):261-266. [doi: 10.1126/science.aaa8685] [Medline: 26185244] 14. About schedule H (Form 990), hospitals. Internal Revenue Service. URL: https://www.irs.gov/forms-pubs/ about-schedule-h-form-990 [accessed 2022-08-29] 15. Ortiz A, Quattrone W, Underwood M, Zmuda M, Goode LA, Saur C, et al. The development and management of community benefit insight: a web-based resource that aggregates US-based nonprofit hospital community benefit spending data. RTI Press. 2022 Jul. URL: https://www.rti.org/rti-press-publication/development-and-management-community-benefit-insight [accessed 2022-08-29] 16. Martha HS, Gayle DN, Carl HM. Hospital community benefits after the ACA: the state law landscape. The Hilltop Institute. 2012. URL: https://mdsoar.org/bitstream/handle/11603/13115/ Hospital-Community-Benefits-after-the-ACA-The-State-Law-Landscape-pdf.pdf?sequence=1&isAllowed=y [accessed 2022-08-29] 17. Williams Jr D, Reiter KL, Pink GH, Holmes GM, Song PH. Rural hospital mergers increased between 2005 and 2016-what did those hospitals look like? Inquiry 2020 Jan;57:46958020935666 [FREE Full text] [doi: 10.1177/0046958020935666] [Medline: 32684072] Saghafian S, Song LD, Raja AS. Towards a more efficient healthcare system: opportunities and challenges caused by hospital closures amid the COVID-19 pandemic. Health Care Manag Sci 2022 Jun;25(2):187-190 [FREE Full text] [doi: 10.1007/s10729-022-09591-7] [Medline: 35292872] 18. 19. Azam N, Yao JT. Comparison of term frequency and document frequency based feature selection metrics in text categorization. Expert Syst Appl 2012 Apr;39(5):4760-4768 [FREE Full text] [doi: 10.1016/j.eswa.2011.09.160] 20. Nelson LK. Computational grounded theory: a methodological framework. Sociol Methods Res 2020 Feb;49(1):3-42 [FREE 21. 22. Full text] [doi: 10.1177/0049124117729703] Fang Z, He Y, Procter R. A query-driven topic model. In: Proceedings of the Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. 2021 Presented at: ACL-IJCNLP '21; August 1-6, 2021; Virtual Event p. 1764-1777 URL: https://aclanthology.org/2021. findings-acl.154.pdf [doi: 10.18653/v1/2021.findings-acl.154] van Kessel P. Overcoming the limitations of topic models with a semi-supervised approach. Pew Research Center. 2019 Apr 11. URL: https://medium.com/pew-research-center-decoded/ overcoming-the-limitations-of-topic-models-with-a-semi-supervised-approach-b947374e0455 [accessed 2023-02-20] https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al 23. Social determinants of health literature summaries: healthy people 2030. U.S. Department of Health and Human Services and Office of Disease Prevention and Health Promotion. URL: https://health.gov/healthypeople/priority-areas/ social-determinants-health/literature-summaries [accessed 2023-02-14] 24. Moran PA. Notes on continuous stochastic phenomena. Biometrika 1950 Jun;37(1-2):17-23 [FREE Full text] [doi: 10.1093/biomet/37.1-2.17] 25. Nghiem LT, Papworth SK, Lim FK, Carrasco LR. Analysis of the capacity of google trends to measure interest in conservation topics and the role of online news. PLoS One 2016 Mar 30;11(3):e0152802 [FREE Full text] [doi: 10.1371/journal.pone.0152802] [Medline: 27028399] Semantic search: sentence-transformers documentation. GitHub. URL: https://www.sbert.net/examples/applications/ semantic-search/README.html [accessed 2022-08-29] 26. 27. Cash-Gibson L, Rojas-Gualdrón DF, Pericàs JM, Benach J. Inequalities in global health inequalities research: a 50-year bibliometric analysis (1966-2015). PLoS One 2018;13(1):e0191901 [FREE Full text] [doi: 10.1371/journal.pone.0191901] [Medline: 29385197] 28. Understanding the epidemic: drug overdose. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/ drugoverdose/epidemic/index.html [accessed 2022-08-29] 29. Obesity is a common, serious, and costly disease. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/ obesity/data/adult.html [accessed 2022-08-29] 30. Osmick MJ, Wilson M. Social determinants of health-relevant history, a call to action, an organization's transformational 31. story, and what can employers do? Am J Health Promot 2020 Mar;34(2):219-224. [doi: 10.1177/0890117119896122d] [Medline: 31931600] 2022 AHAR: Part 1 - PIT estimates of homelessness in the U.S. United States Department of Housing and Urban Development. 2022 Dec. URL: https://www.huduser.gov/portal/datasets/ahar/ 2022-ahar-part-1-pit-estimates-of-homelessness-in-the-us.html [accessed 2023-02-06] 32. Oral health data: explore by topic. Centers for Disease Control and Prevention. URL: https://nccd.cdc.gov/oralhealthdata/ rdPage.aspx?rdReport=DOH_DATA.ExploreByTopic&islTopic=ADT [accessed 2023-03-09] 33. Oral health. US Department of Health and Human Services. URL: https://health.gov/our-work/national-health-initiatives/ healthy-people/healthy-people-2020/healthy-people-2020-law-and-health-policy/oral-health [accessed 2023-03-09] 34. Budiman A. Key findings about U.S. immigrants. Pew Research Center. 2020 Aug 20. URL: https://www.pewresearch.org/ fact-tank/2020/08/20/key-findings-about-u-s-immigrants/ [accessed 2023-03-09] 35. Bustamante AV, Chen J, Félix Beltrán L, Ortega AN. Health policy challenges posed by shifting demographics and health trends among immigrants to the United States. Health Aff (Millwood) 2021 Jul;40(7):1028-1037 [FREE Full text] [doi: 10.1377/hlthaff.2021.00037] [Medline: 34228519] 36. Gajarawala SN, Pelkowski JN. Telehealth benefits and barriers. J Nurse Pract 2021 Mar;17(2):218-221 [FREE Full text] [doi: 10.1016/j.nurpra.2020.09.013] [Medline: 33106751] 37. Butzner M, Cuffee Y. Telehealth interventions and outcomes across rural communities in the United States: narrative review. J Med Internet Res 2021 Aug 26;23(8):e29575 [FREE Full text] [doi: 10.2196/29575] [Medline: 34435965] 38. NCHS data brief, no. 356. Centers for Disease Control and Prevention. 2020 Jan. URL: https://www.cdc.gov/nchs/products/ databriefs/db356.htm [accessed 2023-03-09] 39. Most liberal states 2023. World Population Review. URL: https://worldpopulationreview.com/state-rankings/ most-liberal-states [accessed 2023-03-09] 40. Carlton EL, Singh SR. Joint community health needs assessments as a path for coordinating community-wide health improvement efforts between hospitals and local health departments. Am J Public Health 2018 May;108(5):676-682. [doi: 10.2105/AJPH.2018.304339] [Medline: 29565662] Abbreviations CHNA: community health needs assessment IRS: Internal Revenue Service LGBTQ: lesbian, gay, bisexual, transgender, and queer SDOH: social determinants of health https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 14 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Hadley et al Edited by A Mavragani; submitted 15.11.22; peer-reviewed by K Chen, D Valdes, M Torii; comments to author 20.12.22; revised version received 13.03.23; accepted 14.04.23; published 24.05.23 Please cite as: Hadley E, Marcial LH, Quattrone W, Bobashev G Text Analysis of Trends in Health Equity and Disparities From the Internal Revenue Service Tax Documentation Submitted by US Nonprofit Hospitals Between 2010 and 2019: Exploratory Study J Med Internet Res 2023;25:e44330 URL: https://www.jmir.org/2023/1/e44330 doi: 10.2196/44330 PMID: 37223985 ©Emily Hadley, Laura Haak Marcial, Wes Quattrone, Georgiy Bobashev. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 24.05.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e44330 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44330 \| p. 15 (page number not for citation purposes)
10.1590_2317-1782_20232021221en	Original Article Artigo Original Luciéle Dias Oliveira1  Anaelena Bragança de Moraes2  Sabrina Felin Nunes1  Inaê Costa1  Ana Paula Ramos de Souza3,4  Relationship between enunciative signs of language acquisition and language assessment through the Bayley III scale at 24 months Relação entre sinais enunciativos de aquisição da linguagem e a avaliação de linguagem pela escala Bayley III aos 24 meses Keywords ABSTRACT Language Development Risk Factors Childhood Assessment Purpose: To analyze the correlation between the results obtained on the SEAL and the Bayley III Scale and compare babies with and without delay in language acquisition at 24 months concerning the performance obtained by them and their mothers on the SEAL from 3 to 24 months. Methods: The SEAL collection consists of 15-minute footages of 45 babies aged from 3 to 24 months old in interaction with their mothers, who were assessed by two trained speech therapists for the use of the SEAL. At 24 months, the 45 babies were assessed using the Bayley III Scale and the item language was selected to classify them with and without delay. These results were statistically analyzed through a Pearson’s correlation test and a Fisher’s exact test. Results: In average, eighteen signs of typical development as we obtained, while a mean of 12 delay signs were found. By comparing the presence and absence of signs between the groups with and without delay in language acquisition, eight signs from the baby and one from the mother differed statistically in the sample. The analysis using the SEAL for cases of delay showed that the maternal factor was as important as the infant factor to understand the babies’ language functioning. Conclusion: There was a significant correlation between the SEAL performance from 3 to 24 months and the language outcome at 24 months assessed by the Bayley III Scale in this sample. Descritores RESUMO Linguagem Desenvolvimento Infantil Fatores de Risco Infância Avaliação Objetivo: Analisar a correlação entre resultados obtidos no Sinais Enunciativos de Aquisição da Linguagem (SEAL) e na Escala Bayley III e comparar bebês com e sem atraso na aquisição da linguagem aos 24 meses no desempenho obtido por ele e sua mãe no SEAL dos 3 aos 24 meses. Método: A coleta do SEAL constou de filmagens de 45 bebês, realizadas nas faixas etárias de 3 a 24 meses em interação com suas mães, com duração de 15 minutos, que foram avaliados por duas fonoaudiólogas treinadas no uso do SEAL. Aos 24 meses, os 45 bebês foram avaliados pela Escala Bayley III e selecionado o item linguagem para classificá-los com e sem atraso. Sobre tais resultados realizaram-se as análises estatísticas com o teste de correlação de Pearson e o teste exato de Fisher. Resultados: Obtiveram-se as médias de sinais no desenvolvimento típico que foi 18 sinais e, em casos de atraso, a média foi de 12 sinais. Na comparação da relação de presença e ausência dos sinais entre os grupos com e sem atraso na aquisição da linguagem, oito sinais do bebê e um da mãe diferiram estatisticamente na amostra. O fator materno apresentou-se tão importante quanto o infantil na compreensão do funcionamento de linguagem dos bebês na análise realizada com o SEAL nos casos de atraso. Conclusão: Houve correlação significativa entre o desempenho no SEAL entre 3 e 24 meses e o desfecho de linguagem aos 24 meses avaliado pela Escala Bayley III nesta amostra. Correspondence address: Ana Paula Ramos de Souza Departamento de Saúde da Comunicação Humana, Instituto de Psicologia, Universidade Federal do Rio Grande do Sul – UFRGS Rua Ramiro Barcelos, 2600, Santa Cecília, Porto Alegre (RS), Brasil, CEP: 90035-003. E-mail: [email protected] Received: August 20, 2021 Accepted: March 02, 2022 Study conducted at Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 1Programa de Pós-graduação em Distúrbios da Comunicação Humana, Departamento de Fonoaudiologia, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 2 Programa de Pós-graduação em Distúrbios da Comunicação Humana, Departamento de Estatística, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 3 Departamento de Saúde da Comunicação Humana, Instituto de Psicologia, Universidade Federal do Rio Grande do Sul – UFRGS - Porto Alegre (RS), Brasil. 4 Programa de Pós-graduação em Distúrbios da Comunicação Humana, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. Financial support: research productivity grant from the National Council for Scientific and Technological Development (CNPq), Level 2, process nº 306510/2013-8 for the last author when the research was carried out. Conflict of interests: nothing to declare. This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 1/7 ISSN 2317-1782 (Online version)INTRODUCTION Infant language acquisition should be assessed in the ealry years of life since there is specialized childcare monitoring that analyzes this developmental aspect. The Bayley III Scale(1) is one of the instruments that allows for this assessment in the first two years of life, considered the gold-standard examination for infant development assessment and widely used by the scientific community(2-10) for differentiating receptive communication (49 items) from expressive communication (48 items) in the infant grammar domain. According to Madaschi et al.(11) in a study on cross-cultural adaptation and psychometric properties, the Brazilian version of the Bayley III scale showed a highly convergent validity, as well as good internal consistency and homogeneity of items for children aged 12 to 42 months, thus corroborating its effectiveness for research purposes. Although this scale has diagnostic value for the grammatical domain, it involves some application time (one to two sessions) and depends on the collaboration of the child, as well as on specialized training by the examiner and acquisition of high- cost materials in the context of the common reality of Brazilian professionals. Furthermore, it does not investigate the adult’s participation in the language acquisition process. The Enunciative Signs of Language Acquisition [ESLA – in Portuguese Sinais Enunciativos de Aquisição da Linguagem (SEAL)](12-14) tool was preliminarily validated to provide an instrument to address the adult-child dialogue so that it could be easily applied in the process of language acquisition follow-up, based on the contributions of clinical studies from the enunciative perspective(15-18) and the enunciative study of language acquisition(19). The ESLA signs consider the semiotic level (grammatical domain of the language) and the language semantization process. This process is related to the subject’s appropriation of their linguistic knowledge (semiotic level) in the dialogue support, which allows for identifying the emergence and support of a place of enunciation for the baby(16-19). From this perspective, the child’s potentialities (biological, cognitive, and subjective) and the enunciative support offered by the adult are important(17,18). The ESLA signs capture whether the language acquisition process is proceeding as expected or it presents some impediment through an indicative paradigm, i.e., if the signs are present, the process is possibly satisfactory; if they are absent, the child and their family members should be monitored through shorter sessions to verify the language progress and establish a potential demand for timely intervention. ESLA studies indicate a child factor and a maternal factor in the language functioning between babies and their mothers, which contributes to the understanding of obstacles to language acquisition. ESLA is not aimed at diagnosis but at monitoring language acquisition. In this context, herein we consider the results from previous studies on the grammatical domain and the language semantization process. The goal is to analyze the child’s conditions of occupying the enunciation place, as well as the adult’s conditions of sustaining this place(12-19), in addition to the scientific evidence generated by the Bayley Scale III(1-11). Thereby, this research analyzes the correlation between the results obtained by the ESLA and the Bayley Scale III, and compares babies with and without delayed language acquisition at 24 months of age concerning their performance and their mothers’ in the ESLA from 3 to 24 months. METHODS This is a quantitative, longitudinal, and prospective study that was approved by the Research Ethics Committee of an educational institution in a medium-sized city in the Rio Grande do Sul state – CAAE number: 28586914.0.0000.5346. This study is in line with the regulatory norms and guidelines for research with human beings established in Resolution 466/12 of the National Health Council. It provides for data confidentiality, thus ensuring both the secrecy and privacy of the subjects’ identities by the signing of the Confidentiality Agreement. In addition, the families who agreed to participate in the research and signed the Free and Informed Consent Form (FICF) were instructed on the objectives and procedures to the research. Those responsible for the babies answered the interview on sociodemographic, obstetric, and psychosocial data adapted from the original version by Schwengber and Piccinini(20). Initially, the sample for the ESLA assessment included 101 babies, out of which only 45 children remained, 19 born at term and 26 pre-terms, followed by the ESLA from 3 to 24 months, who were assessed at 24 months based on the Bayley Scale III. Our research analyzes the language item, all other developmental aspects were analyzed based on the Bayley Scale III by other professionals participating in a larger research. For babies born prematurely, the corrected age was considered in the assessment. The babies and their families were invited to participate in the research during the follow-up sessions of premature babies at a university hospital, and during the Guthrie test at a Primary Health Care Unit nearby. The following inclusion criteria were applied: babies without biological limitations, such as neurological lesions or syndromes, or sensory deficits (visual, auditory, etc.). These aspects were assessed by pediatricians and the research team, including speech therapists, psychologists, physiotherapists, and occupational therapists. In case of doubt, they were removed from the study and referred to a neuro pediatrician or geneticist. The language analysis based on ESLA involves filming the interaction of the mother, or whoever performed this function for the baby, which occurred in different ways throughout the research stages. The filming was carried out from two angles, frontal and lateral, encompassing a time of 15 minutes, depending on the baby’s age and other aspects to be analyzed. The babies were positioned in a baby-comfort or seated on an EVA mat to interact with their mothers in a lighted and comfortable environment in terms of temperature. They should be in a good state of wakefulness, well fed, and sanitized. The filming was performed using two JVC Everio GZ-MG 630 camcorders placed in two positions: two meters Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 2/7 away from the mother and the baby, with the mother positioned with her back to the camcorder and the baby in front, and another placed one meter away, with the mother and the baby interacting face to face at a side angle. The postures were standardized so that the baby would be observed interacting with their mother. The babies were born at term or were late preterm infants, in the case of the latter, the corrected age was considered. The babies filmed for the ESLA analysis, according to the four six-monthly instruments created(12-14), were dividied into the following age groups: Phase 1 – 3 months and 1 day to 4 months and 29 days – The baby sitting in the baby carrier (9 minutes). The mother was instructed to sing (3 minutes) (ambiance), talk to the baby (3 minutes), and offer an object - e.g., a rubber dog without noise (3 minutes). Phase 2 – 8 months and 1 day to 9 months and 29 days – Mother and baby seated on the EVA mat were filmed in the interaction, and the mother was asked to sing to the baby for 3 minutes, talk for another 6 minutes, and play with an object (the rubber dog) offered by the examiner (6 minutes). If the baby did not have trunk control, they could use a comfort baby. Phase 3 – 17 months and one day to 18 months and 29 days and Phase 4 – 23 months and one day to 24 months and 29 days – At these phases, the baby was observed in free activity with the mother playing with a box of thematic toys (animals, a baby with a bottle, small pans, etc.), in addition to plays and linguistic interactions between the mother and the baby. The mother was instructed to remain on the EVA mat with the baby during the filming. Over the first 10 minutes, the interaction of the mother with the baby was filmed and in the last 5 minutes, the examiner participated in the interaction to observe some signs that covered the dialogue with different interlocutors. The videos were watched by two qualified speech therapists who assigned the signs of the ESLA instruments to the babies, thus allowing for verifying an agreement of 95 and 100% between both of them. Our analysis considered, the values assigned by the main researcher, the main author of this article. Table 1 summarizes the enunciative signs analyzed. All subjects were also assessed at phase 4 (aged 24 months) using the Bayley Scale III(1,11) by a qualified professional. Herein, the language subscale (receptive communication and expressive communication) was considered. Initially, the starting point was found in the test of each baby based on their age. The assessment started as soon as the baby consecutively scored the first three questions (basepoint) and ended upon five consecutive errors. The statistical analysis used an Excel database to organize the language data generated from the presence and absence of ESLA signs in each age group and total, as well as the Bayley III scores obtained for 24 months. All statistical analyses of the results were performed on the STATISTICA 9.1 software. Herein we consider the significance level of p ≤ 0.05. Pearson’s correlation and Fisher’s exact test were also used. RESULTS We analyzed forty-five infants and verified the correlation between the total number of ESLA signs obtained through the four six-monthly instruments and the language scores generated using the Bayley Scale III at 24 months. The results scored a Pearson correlation of 0.718 and a p value of 0.001, thus indicating statistical significance (p<0.05). It allows us to infer that the higher the ESLA score the higher the Bayley III score. The comparative analysis between children with and without language delay, based on the results from the language assessment by the Bayley Scale III at 24 months, showed the average number of enunciative signs of language acquisition in each group, as shown in Table 2. On average, the babies with no delay in language acquisition presented 18 signs, whereas the babies with delay presented 12 signs, i.e., which reflected on the difference between both groups when comparing the absence and presence of each sign, as shown in Table 3. The signs 9, 10, 16, 17, 18, 21, 22, and 23, related to the infant’s enunciative aspects, and sign 24, related to the maternal position in the last age group, differed between the group with language acquisition delay and the group without delay, in terms of the sample statistical comparison . These signs were more markedly present in the babies with typical development than in the babies with delay. Table 4 indicates the data descriptive analysis from the 20 children who presented language alteration according to the Bayley Scale III, at 24 months, indicating both the present and absent signs. The children with alterations according to the ESLA reached borderline, low, or very low scores on the Bayley Scale III. The children who developed some delay in the assessment by the Bayley Scale III at 24 months showed altered child’s signs (child factor) and signs related to the maternal activity of dialogue support (maternal factor). Furthermore, from this group of 20 children with alteration by the Bayley Scale III, four were not at risk in the ESLA, according to the average number of signs (Table 2), nor did they present any alterations in the maternal factor (signs in bold). The four children without risk in the ESLA but with delay by the Bayley Scale III were assessed between borderline and low average. None of them received an extremely low rating. Another relevant aspect in Table 4 is associated with the varying scores from the Bayley Scale III (borderline, very low, extremely low), highlighting most children in the extremely low category, a greater indication of severe delay, with the lowest values in the ESLA (2 to 7), except for subject 3 (S3) with 12 signs. However, a low value shows an agreement at least regarding the aspects concerning the child factor between both tests. The children within the other categories (borderline or very low) scored values between 9 and 17 signs in the ESLA. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 3/7 Table 1. Enunciative Signs of Language Acquisition (ESLA) Signs from 2 to 6 months and 29 days 1. The child reacts to motherese, through vocalizations, body movements, or looks. 2. The child fills its place in the interlocution with verbal sounds such as vowels and/or consonants. 3. The child fills her place in the interlocution with non-verbal sounds in tune with the enunciative context (smile, cry, cry, cough, grumble). 4. The child fills his place in the interlocution silently only with body movements and looks attuned to the enunciative context. 5. Child initiates conversation or proto-conversation. Analyzed speaker baby baby baby baby baby 6. The child and the mother (or her substitute) exchange glances during the interaction. baby-mother 7. The mother (or her substitute) assigns meaning to the baby’s verbal and non-verbal manifestations and sustains this proto-conversation or conversation when the baby initiates it. 8. The mother (or her substitute) uses motherese by talking to the child in a way that is attuned to what is happening in the context and waiting for the baby’s responses. Signs from 7 to 12 months and 29 days 9. The child fills its place in the interlocution (enunciation) with verbal sounds (syllables with varied vowels and consonants - at least two points and two consonant articulatory modes). 10. The child outlines the production of proto words by mirroring the mother’s (or substitute’s) speech. 11. Child outlines the production of proto words spontaneously. 12. When the mother (or substitute) is called upon to enunciate by the child, she produces her enunciation and waits for the child’s answer. Signs from 13 to 17 months and 29 days 13. The child names spontaneously and intelligibly to the adult interlocutor, objects that are absent in the context. 14. The child names in a spontaneous way, but not intelligible to the interlocutor adult, objects that are absent in the context, seeking in the prosody a way to be understood. 15. The child names in a spontaneous and intelligible way to the interlocutor adult, objects, people, and actions, which are present in the enunciative context. 16. The child makes gestures to try to make himself/herself understood when the adult interlocutor does not understand him/her. 17. The child repeats what the adult interlocutor says as a way of zorganizing or reorganizing his or her utterance, for example by improving the syntactic or phonological form, the choice of the lexical item, or even by accentuating some item prosodically. 18. The child talks to different adult interlocutors (father, mother, examiner). 19. The adult interlocutor attributes a possible meaning to the child’s verbal productions, that is, in a tuned way. Signs 18 to 24 months and 29 days 20. The child requests objects and/or asks for clarifications from the interlocutor adult, marking his position as speaker. 21. The child uses distinct phonemic forms to convey different meanings in his/her utterance (at least two articulatory points - labial and alveolar - and two distinct consonantal sound classes - at least nasal and plosive). 22. The child uses different forms (words) to convey different meanings in his/her enunciation. 23. The child combines words, in direct or inverted form, to convey different meanings. 24. When the child presents verbal productions distinct from adult speech, the adult interlocutor reacts by making a neutral repair request (what) or by correctly repeating the child’s speech, or offering a lexical item compatible with the infant’s communicative intention. Source: Crestani et al.(12,13), Fattore et al.(14). Significant signs in the factor analysis are in bold mother mother baby baby baby mother baby baby baby baby baby baby mother baby baby baby baby mother Table 2. Comparison of ESLA total score versus Bayley scale III TESTS TOTAL ESLA WITHOUT RISK (LANGUAGE) Bayley III at 24 months Maximum 22.00 Mean (±SD) 18.85 (± 2.92) Minimum 15.00 N 25 WITH RISK (LANGUAGE) Bayley III at 24 months N 20 Mean (±SD) 12.39 (± 5.13) Minimum 2.00 Maximum 22.00 p_value 0.001* Significant by Mann-Whitney U test Caption: ESLA = Enunciative Signs of Language Acquisition; SD = Standard Deviation; N = Number of Subjects Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 4/7 Table 3. Comparative analysis by a sign of the babies with and without language delay according to the Bayley III scale Babies with delay n=20 Babies without delay n=25 SIGNS ESLA Dyads with missing signs Dyads with present signs Dyads with missing signs Dyads with present signs Comparison 1 (B) 2 (B) 3 (B) 4 (B) 5 (B) 6 (B) 7 (A) 8 (A) 9 (B) 10 (B) 11 (B) 12 (A) 13 (B) 14 (B) 15 (B) 16 (B) 17 (B) 18 (B) 19 (A) 20 (B) 21 (B) 22 (B) 23 (B) 24 (A) n 2 6 2 2 10 3 2 5 11 13 11 2 20 20 15 9 16 14 11 8 12 17 15 10 % 10 30 10 10 50 15 10 25 55 65 55 10 100 100 75 45 80 70 55 40 60 85 75 50 n 18 14 18 18 10 17 18 15 9 7 9 17 0 0 5 11 4 6 9 12 8 3 5 10 % 90 70 90 90 50 85 90 75 45 35 45 85 0 0 25 55 20 30 45 60 40 15 25 50 N 2 3 2 2 11 2 1 3 6 7 9 3 22 25 4 0 4 3 7 4 1 4 4 1 % 8 12 8 8 44 8 4 12 24 28 36 12 88 100 6 0 6 12 28 6 4 6 6 4 n 23 22 23 23 14 23 24 22 19 18 16 22 3 0 21 25 21 22 18 21 24 21 21 24 % 92 88 92 92 56 92 96 88 76 72 64 88 12 0 94 100 94 88 72 94 96 84 84 96 p-value 0.606 0.131 0.606 0.606 0.460 0.392 0.415 0.229 0.034* 0.014* 0.165 0.632 0.625 1.000 0.0001* 0.0002* 0.0000* 0.0001* 0.063 0.071 0.0000* 0.0000* 0.0001* 0.0005* Significant Fisher’s exact test Caption: Bold = Important Signs in Factor Analysis (Crestani et al.(12,13), Fattore et al.(14)); B = Infant Shows Sign; A = Adult Shows Sign; n = Number of Infants; ESLA = Enunciative Signs of Language Acquisition Table 4. Descriptive analysis of Enunciative Signs of Language Acquisition (ESLA) in the cases of language delay in Bayley III at 24 months Subject Present Signs ESLA Absent Signs ESLA Total ESLA Present Bayley III at 24m S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 S19 S20 1,2,3,4,5,6,7,8,12,16,18,19,20, 21,24 9,10,11, 13, 14, 15,17, 22,23 1,3,4,5,6,7,8,16,18, 1,2,3,4,5,6,7,8,9,10,11,12, 1,3,4,6,7,8,12, 16,19 2,9,10,11,12, 13,14,15,17 19, 20,21,22,23,24 13,14,15,16,17,18,19,2021,22,23, 24 2,5,9,10,11,13,14,15,17,18, 20,21,22,23,24 1,2,3,4,5,6,7,8,9,10,11,12, 15,16,18,19,20,21,24 13,14,17,22,23 1,2,3,4,5,6,7,8,12, 20,21,24 9,10,11,13,14,15,16,17, 18,19, 22,23 6,12 1,2,3,4,5,7,8,9,10,11,13, 14,15,16,17,18,19,20,21,22 ,23,24 1,3,4,6,7,8,12,20,21,22,23,24 2,5,9,10,11,13,14,15,16, 17,18,19 1,2,3,4,5,6,7,8,9,10,11,12,16,19,20,21,22 13,14,15,17,18,23,24 1,2,3,4,7,11,12, 5,6,8,9,10,13,14,15,16,17,18,19,20,21,22,23,24 1,2,3,4,5,6,10,11,12,20,21,22,23 7,8,9,13,14,15,16,17,18,19,24 1,2,3,4,5,6,7,8,9,10,11,12,15,16,17,18,19,20,21,22,2 3,24 13,14 1,2,3,4,7,9,11 5,6,8,10,12,13,14,15,16,17,18,19,20,21,22,23,24 2,7,8,15,16,17,18,19,20,21,22,24 1,3,4,5,6,9,1011,12,13,14,23 1,3,4,6,7,12,16,20,21,24 1,2,3,4,5,6,7,8,9,10,11,12,16, 2,5,8,9,1011,13,14,15,17,18,19,22,23 13,14,15,17,18,19,20,21,22,23,24 1,2,3,4,6,7,8,9,12,15,16,17,18,19,20,21,22,23,24 1,2,3,4,5,6,7,8,9,10,11,12,15,16,17,18,19,20,21,22,2 3,24 5,10,11,13,14 13,14 1,3,4,6,7,8,12,13,20,21 2,5,9,10,11,14,15,16,17,18,19, 22,23,24 1,2,3,4,6,7,8,9,12,20,21,22,23,24 5,10,11,13,14,15,16,17,18,19 Caption: B = Borderline; EL = Extremely Low; LM = Low Medium; ESLA = Enunciative Signs of Language Acquisition 15 9 12 9 19 12 2 12 17 7 13 22 7 12 10 13 19 22 10 14 79 (B) 71 (B) 59 (EL) 77 (B) 83 (LM) 79 (B) 65 (EL) 47 (EL) 85 (LM) 59 (EL) 74 (B) 77 (B) 47 (EL) 79 (B) 79 (B) 77 (B) 79 (B) 83 (LM) 83 (LM) 89 (LM) Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 5/7 DISCUSSION The positive correlation between cases of language delay assessed by the Bayley III Scale and risk cases in the ESLA demonstrates the effectiveness of this examination as a screening test for the sample studied. Table 4 shows that 16 of the 20 cases assessed as delay by the Bayley III Scale had significant alterations in the ESLA. In this research, the ESLA was assigned by filming during childcare follow-up sessions, although it may be analyzed by a qualified professional by observing the mother-baby interaction in an outpatient clinic, which would be less expensive in terms of time and cost for insertion in the Universal Health System [in Portuguese Sistema Único de Saúde (SUS)]. In this condition, the suspected cases according to the ESLA would be referred to a diagnostic test through the Bayley Scale III or other diagnostic scales for language and development(21,22). The number of mother-baby dyads participating in the ESLA investigation in all age groups was much higher (101) than in the group that attended the two assessment meetings for the Bayley Scale III (45). This sample loss suggests that the adherence to the more time-consuming test (the Bayley Scale III) by users demands a change of culture and an improvement in the access to the health service, which is not expected to change on a short-term perspective. In contrast, the comparison between the group with and without delay allowed us to establish an average of 18 enunciative signs out of 24 assessed as the absence of risk in the ESLA. These data suggest the need to continue investigating the ESLA in terms of establishing criteria per age group and in the total test, which was not possible from the small sample obtained in this research. Table 3 shows signs, such as the 13 and 14, that were absent in both groups, although sign 14 was indicated as relevant in the factor analysis(14). Among the signs that statistically differed when comparing infants with and without delay, signs 9, 10, 21, 22, and 23 showed the ability of infants without language delay to occupy their place in enunciation with increasingly complex vocalization and speech (phonological and lexical diversity, and initial use of syntax). In turn, babies without these signs may show a potential delay in language acquisition. Sign 16 is related to the use of gestures as a form of communication, which is predicted by language acquisition studies that claim some continuity and synchrony between verbal skills and baby gestures(23). Sign 17 is related to the baby’s ability to anchor themselves in the mother’s speech to improve what they say, a strategy identified by Silva(19) in the language acquisition process. It is also related to an adult’s willingness to help the child speaking, assessed in sign 24. Therefore, it is important to observe the strength of both signs simultaneously in the sample studied. In addition to those already mentioned, sign 18, related to the amplitude of interlocutors, was fundamental to assess not only the disjunction in terms of the enunciative acquisition relation but also of the mother-baby separation process. Such a scenario was observed by Flores and Souza(24), who found that babies in psychological suffering and facing difficulties in the separation process and operation of the paternal function showed difficulty in speaking with distinct interlocutors. The lack of distinctive sign when comparing the two groups in the first age group (zero to six months) indicates the need for further investigations and improving the instrument. Likewise, the factor analysis showed three signs at this phase that were related to a child factor and a maternal factor in a larger sample of subjects at the same phase(13). These results suggest the need for continuous studies on validation criteria. Table 4 shows that subjects 5, 12, 17, and 19, presented 18 or more enunciative signs. It is worth mentioning that these four subjects presented no alterations in the mother factor, nor did subject 5, with delay according to both ESLA and Bayley Scale III, whereas all others did. They also did not obtain the “extremely low” classification in the Bayley Scale III(1). These data allow us to observe that, in most cases, the maternal factor contributed to the emergence and understanding of language functioning in cases of delayed language acquisition. In other words, the way the adult carries out enunciative support must be considered in the assessment and intervention for delayed language acquisition(17,18). Several studies in the enunciative field(12-19) have evidenced that both language acquisition and clinical practice with young children should invest in the analysis of the mother-baby dialogue to propose a hypothesis of language functioning(16) that allows proposing intervention lines. This language operating hypothesis foresees the relation I (child) – YOU (adult) in the understanding of the suffering arising from language delay or disorder. Based on this theoretical perspective and our results, the ESLA is a promising tool for assessing such a factor, was as it was revealed in 16 out of 20 cases assessed as language delay by the Bayley III Scale. It is important highlighting that children with extremely low values on the Bayley III Scale were the same ones who received lower ESLA values. Considering the numerical limitation of our sample, the results suggest the need to continue investigating the language of infants and young children using ESLA since it is an effective way to monitor language acquisition in childcare and propose interventions in time to prevent the crystallization of language symptoms(17,18). Such a scenario requires to establish criteria for the test in larger samples. Ours is a clinical study of a smaller proportion, which included babies who attended the assessment using the Bayley Scale III at the end of the research, at two years of age, a number much smaller than should be desired. ESLA assessments have no diagnostic purposes since the baby is undergoing the process of linguistic constitution, rather they seek to offer timely interventions to favor the convergence between family members and the baby or small child. In this context, facilitating the maternal factor, an important element in the factorial studies, is a way of strengthening the convergence and linguistic synchrony between the mother (or her substitute) and the baby. This shows that the field of speech therapy could benefit from studies centered on dialogue as the analysis focus in research on language acquisition since children’s abilities to occupy their place of enunciation are as relevant in the acquisition process as the adult’s support of an enunciative place. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 6/7 CONCLUSION Our findings allow us to suggest a significant correlation between the performance in the ESLA between 3 and 24 months and the language outcome at 24 months assessed by the Bayley Scale III. The comparison between babies with and without delay in language acquisition allowed us to establish averages of signs in the ESLA. Additionally, some signs from the baby and one from the mother showed statistical differences when comparing the two groups of the sample studied as to their presence and absence, especially from the second age group studied. These data allow us to conclude that the ESLA has some potential as a screening test and should be investigated in larger samples since it involves a short application time requiring only to observe the mother-baby interaction during the first and second years of life in a context of spontaneous play with materials that are accessible to examiners and families. REFERENCES 1. Bayley N. Bayley scales of infant and toddler development. 3rd ed. San Antonio: The Psychological Corporation; 2006. 2. Godamunne P, Liyanage C, Wimaladharmasooriya N, Pathmeswaran A, Wickremasinghe AR, Patterson C, et al. Comparison of performance of Sri Lankan and US children on cognitive and motor scales of the Bayley scales of infant development. BMC Res Notes. 2014;7(1):300. http://dx.doi. org/10.1186/1756-0500-7-300. PMid:24886547. 3. Ballot DE, Ramdin T, Rakotsoane D, Agaba F, Davies VA, Chirwa T, et al. Use of the Bayley scales of infant and toddler development, third edition, to assess developmental outcome in infants and young children in an urban setting in South Africa. Int Sch Res Notices. 2017;2017:1631760. http:// dx.doi.org/10.1155/2017/1631760. PMid:28835912. 4. Springer PE, Slogrove AL, Laughton B, Bettinger JA, Saunders HH, Molteno CD, et al. Neurodevelopmental outcome of HIV – exposed but uninfected infants in the mother an infants health study, Cape Town, South Africa. Trop Med Int Health. 2018;23(1):69-78. http://dx.doi.org/10.1111/ tmi.13006. PMid:29131457. 5. Shapiro KA, Kim H, Mandelli ML, Rogers EE, Gano D, Ferriero DM, et al. Early changes in brain structure correlate with language outcomes in children with neonatal encephalopathy. Neuroimage Clin. 2017;15:572-80. http:// dx.doi.org/10.1016/j.nicl.2017.06.015. PMid:28924555. 6. Goh SKY, Tham EKH, Magiati I, Sim L, Sanmugam S, Qiu A, et al. Analysis of item-level bias in the Bayley-III language subscales: the validity and utility of standardized language assessment in a multilingual setting. J Speech Lang Hear Res. 2017;60(9):2663-71. http://dx.doi. org/10.1044/2017_JSLHR-L-16-0196. PMid:28813555. 7. Fairbairn N, Galea C, Loughran-Fowlds A, Hodge A, Badawi N, Walker K. Prediction of three year outcomes using the Bayley-III for surgical, cardiac and healthy Australian infants at one year of age. Early Hum Dev. 2018;117:57- 61. http://dx.doi.org/10.1016/j.earlhumdev.2017.12.012. PMid:29288912. 8. Khurana S, Shivakumar M, Reddy GVSK, Jayashree P, Bhat YR, Lewis LES, et al. Long-term neurodevelopment outcome of caffeine versus aminophylline therapy for apnea of prematurity. J Neonatal Perinatal Med. 2017;10(4):355-62. http://dx.doi.org/10.3233/NPM-16147. PMid:29286928. 9. Wang H, Zhou H, Zhang Y, Wang Y, Sun J. Association of maternal depression with dietary intake, growth, and development of preterm infants: a cohort study in Beijing, China. Front Med. 2018;12(5):533-41. http:// dx.doi.org/10.1007/s11684-017-0591-y. PMid:29181690. 10. Nunes SF, Moraes AB, Busanello-Stella AR, Roth-Hoogstraten AM, Souza APR. Risco psíquico e desenvolvimento infantil: importância da detecção precoce na puericultura. Saúde. 2020;46(2):e47856. http://dx.doi. org/10.5902/2236583447856. 11. Madaschi V, Mecca TP, Macedo EC, Paula CS. Bayley-III scales of infant and toddler development: transcultural adaptation and psychometric properties. Paidéia. 2016;26(64):189-97. 12. Crestani AH, Moraes AB, Souza APR. Validação de conteúdo: clareza, pertinência, fidedignidade e consistência interna de sinais enunciativos de aquisição da linguagem. CoDAS. 2017;29(4):e20160180. http://dx.doi. org/10.1590/2317-1782/201720160180. PMid:28813071. 13. Crestani AH, Moraes AB, Souza AM, Souza APR. Construct validation of enunciative signs of language acquisition for the first year of life. CoDAS. 2020;32(3):e20180279. http://dx.doi.org/10.1590/2317-1782/20202018279. PMid:32578837. 14. Fattore IM, Moraes AB, Crestani AH, Souza AM, Souza APR. Validação de conteúdo e construto de sinais enunciativos de aquisição da linguagem no segundo ano de vida. CoDAS. 2022;34(2):e20200252. http://dx.doi. org/10.1590/2317-1782/20202020252. PMid:34932657. 15. Cardoso JL, Sachetti JK. Enunciação e distúrbios de linguagem: sobre a questão da nomeação. RevLet. 2010;2(2):105-19. 16. Bender S, Surreaux LM. Os efeitos da fala da criança: a escuta do sintoma na clínica de linguagem. Cad IL. 2011;42:129-45. 17. Oliveira LD, Ramos-Souza AP. O distúrbio de linguagem em dois sujeitos com risco ao desenvolvimento em uma perspectiva enunciativa do funcionamento de linguagem. Rev CEFAC. 2014;16(5):1700-12. http:// dx.doi.org/10.1590/1982-0216201410713. 18. Nazario CG, Rechia IC, Fattore IM, Nunes SF, Souza APR. Comparação entre avaliações de linguagem na infância e sua relação com risco psíquico. Distúrbios Comun. 2019;31(1):104-18. http://dx.doi.org/10.23925/2176- 2724.2019v31i1p104-118. 19. Silva CLC. Os movimentos enunciativos da criança na linguagem. Rev ABRALIN. 2011;10(4):77-94. http://dx.doi.org/10.5380/rabl.v10i4.32424. 20. Schwengber DDS, Piccinini CA. Depressão materna e interação mãe-bebê no final do primeiro ano de vida. Psic Teor Pesq. 2004;20(3):233-40. http:// dx.doi.org/10.1590/S0102-37722004000300004. 21. Rocha SR, Dornelas LF, Magalhães LC. Instrumentos utilizados para avaliação do desenvolvimento de recém-nascidos pré-termo no Brasil: revisão da literatura. Cad Ter Ocup UFSCar. 2013;21(1):109-17. http:// dx.doi.org/10.4322/cto.2013.015. 22. Miranda JR, Silva MA, Mendonça EJ. Fo, Bandeira DR. Evidências de validade de critério do Inventário Dimensional de Avaliação do Desenvolvimento Infantil para rastreio do transtorno do espectro autismo. Neuropsicol Latinoam. 2020;12(3):19-29. 23. Cavalcante MCB. Contribuições dos estudos gestuais para as pesquisas em aquisição da linguagem. Ling Ensino. 2018;21:5-35. 24. Flores MR, Souza APR. Diálogo de pais e bebês em situação de risco ao desenvolvimento. Rev CEFAC. 2014;16(3):840-52. http://dx.doi. org/10.1590/1982-0216201411412. Author contributions LDO was responsible for study design, data collection, analysis, and article writing; ABM was responsible for statistical analysis; SFN was responsible for data collection of Bayley-III Scale; IC was responsible for statistical analysis; APRS was responsible for study design, and article review. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221en 7/7 Artigo Original Original Article Luciéle Dias Oliveira1  Anaelena Bragança de Moraes2  Sabrina Felin Nunes1  Inaê Costa1  Ana Paula Ramos de Souza3,4  Relação entre sinais enunciativos de aquisição da linguagem e a avaliação de linguagem pela escala Bayley III aos 24 meses Relationship between enunciative signs of language acquisition and language assessment through the Bayley III scale at 24 months Descritores RESUMO Linguagem Desenvolvimento Infantil Fatores de Risco Infância Avaliação Objetivo: Analisar a correlação entre resultados obtidos no Sinais Enunciativos de Aquisição da Linguagem (SEAL) e na Escala Bayley III e comparar bebês com e sem atraso na aquisição da linguagem aos 24 meses no desempenho obtido por ele e sua mãe no SEAL dos 3 aos 24 meses. Método: A coleta do SEAL constou de filmagens de 45 bebês, realizadas nas faixas etárias de 3 a 24 meses em interação com suas mães, com duração de 15 minutos, que foram avaliados por duas fonoaudiólogas treinadas no uso do SEAL. Aos 24 meses, os 45 bebês foram avaliados pela Escala Bayley III e selecionado o item linguagem para classificá-los com e sem atraso. Sobre tais resultados realizaram-se as análises estatísticas com o teste de correlação de Pearson e o teste exato de Fisher. Resultados: Obtiveram-se as médias de sinais no desenvolvimento típico que foi 18 sinais e, em casos de atraso, a média foi de 12 sinais. Na comparação da relação de presença e ausência dos sinais entre os grupos com e sem atraso na aquisição da linguagem, oito sinais do bebê e um da mãe diferiram estatisticamente na amostra. O fator materno apresentou-se tão importante quanto o infantil na compreensão do funcionamento de linguagem dos bebês na análise realizada com o SEAL nos casos de atraso. Conclusão: Houve correlação significativa entre o desempenho no SEAL entre 3 e 24 meses e o desfecho de linguagem aos 24 meses avaliado pela Escala Bayley III nesta amostra. Keywords ABSTRACT Language Development Risk Factors Childhood Assessment Purpose: To analyze the correlation between the results obtained on the SEAL and the Bayley III Scale and compare babies with and without delay in language acquisition at 24 months concerning the performance obtained by them and their mothers on the SEAL from 3 to 24 months. Methods: The SEAL collection consists of 15-minute footages of 45 babies aged from 3 to 24 months old in interaction with their mothers, who were assessed by two trained speech therapists for the use of the SEAL. At 24 months, the 45 babies were assessed using the Bayley III Scale and the item language was selected to classify them with and without delay. These results were statistically analyzed through a Pearson’s correlation test and a Fisher’s exact test. Results: In average, eighteen signs of typical development as we obtained, while a mean of 12 delay signs were found. By comparing the presence and absence of signs between the groups with and without delay in language acquisition, eight signs from the baby and one from the mother differed statistically in the sample. The analysis using the SEAL for cases of delay showed that the maternal factor was as important as the infant factor to understand the babies’ language functioning. Conclusion: There was a significant correlation between the SEAL performance from 3 to 24 months and the language outcome at 24 months assessed by the Bayley III Scale in this sample. Endereço para correspondência: Ana Paula Ramos de Souza Departamento de Saúde da Comunicação Humana, Instituto de Psicologia, Universidade Federal do Rio Grande do Sul – UFRGS Rua Ramiro Barcelos, 2600, Santa Cecília, Porto Alegre (RS), Brasil, CEP: 90035-003. E-mail: [email protected] Recebido em: Agosto 20, 2021 Aceito em: Março 02, 2022 Trabalho realizado na Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 1 Programa de Pós-graduação em Distúrbios da Comunicação Humana, Departamento de Fonoaudiologia, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 2 Programa de Pós-graduação em Distúrbios da Comunicação Humana, Departamento de Estatística, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. 3 Departamento de Saúde da Comunicação Humana, Instituto de Psicologia, Universidade Federal do Rio Grande do Sul – UFRGS - Porto Alegre (RS), Brasil. 4 Programa de Pós-graduação em Distúrbios da Comunicação Humana, Universidade Federal de Santa Maria – UFSM - Santa Maria (RS), Brasil. Fonte de financiamento: bolsa de produtividade em Pesquisa do Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Nível2, processo nº 306510/2013-8 para o último autor quando realizada a pesquisa. Conflito de interesses: nada a declarar. Este é um artigo publicado em acesso aberto (Open Access) sob a licença Creative Commons Attribution, que permite uso, distribuição e reprodução em qualquer meio, sem restrições desde que o trabalho original seja corretamente citado. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 1/7 ISSN 2317-1782 (Online version)INTRODUÇÃO A aquisição da linguagem infantil deve ser avaliada nos primeiros anos de vida do bebê, desde que haja um acompanhamento especializado na puericultura que contemple uma análise deste aspecto do desenvolvimento. Um dos instrumentos que permite uma avaliação nos dois primeiros anos de vida é a Escala Bayley III(1), considerada padrão-ouro para avaliação do desenvolvimento infantil, amplamente utilizada pela comunidade científica(2-10) por diferenciar a comunicação receptiva (49 itens) da expressiva (48 itens) no domínio gramatical infantil. Na pesquisa de Madaschi et al. (11) sobre a adaptação transcultural e propriedades psicométricas, a versão brasileira da escala Bayley III apresentou alta validade convergente e boa consistência interna e homogeneidade de itens para crianças de 12 a 42 meses, corroborando sua efetividade para fins de pesquisas. Embora possua valor diagnóstico para o domínio gramatical, a referida escala demanda tempo de aplicação (de uma a duas sessões) e depende da colaboração da criança, bem como formação especializada do examinador e aquisição de materiais de alto custo, ao considerar a realidade habitual dos profissionais no Brasil. Além disso, não investiga a participação do adulto no processo de aquisição da linguagem. Com o propósito de oferecer um instrumento que abordasse o diálogo adulto-criança e de fácil aplicação no processo de acompanhamento da aquisição da linguagem, foram validados preliminarmente os Sinais Enunciativos de Aquisição da Linguagem - SEAL(12-14), que consideraram as contribuições de estudos clínicos na perspectiva enunciativa(15-18) e estudo enunciativo de aquisição da linguagem(19). Os sinais do SEAL consideram o nível semiótico (domínio gramatical da língua) e o processo de semantização da língua. Esse processo relaciona-se à apropriação que o sujeito faz de seu conhecimento linguístico (nível semiótico) na sustentação do diálogo, que permite que se identifique a emergência e sustentação de um lugar de enunciação para o bebê(16-19). Nessa perspectiva, são importantes as potencialidades da criança (biológicas, cognitivas e subjetivas) e a sustentação enunciativa que o adulto oferece(17,18). Os sinais do SEAL captam se o processo de aquisição da linguagem está transcorrendo dentro do esperado ou se apresenta algum impedimento por meio de um paradigma indiciário, ou seja, se os sinais estão presentes possivelmente o processo está a contento, se ausentes, a criança e seus familiares devem ser acompanhados em consultas de menor intervalo de tempo para verificar o progresso em linguagem e estabelecer uma eventual demanda por intervenção oportuna. Os estudos do SEAL indicam um fator infantil e um fator materno no funcionamento de linguagem entre o bebê e sua mãe, que contribuem na compreensão de obstáculos à aquisição da linguagem(12-14), demonstrando que é possível realizar intervenções oportunas para revertê-los a partir de uma escuta sensível da família. O SEAL não tem objetivo diagnóstico, mas de acompanhamento da aquisição da linguagem. Assim, considerando os resultados obtidos de estudos anteriores sobre o domínio gramatical e processo de semantização da língua para analisar as condições da criança na ocupação do lugar de enunciação e o do adulto na sustentação desse lugar(12-19), e as evidências científicas obtidas com a Escala Bayley III(1-11), esta pesquisa analisa a correlação entre os resultados obtidos pelo SEAL e a Escala Bayley III, e compara bebês com e sem atraso na aquisição da linguagem aos 24 meses, em relação ao desempenho obtido por ele e sua mãe no SEAL dos 3 aos 24 meses. MÉTODO Trata-se de um estudo quantitativo, longitudinal e prospectivo, aprovado pelo Comitê de Ética em Pesquisa de uma instituição de ensino em cidade de porte médio do Rio Grande do Sul, sob número de CAAE: 28586914.0.0000.5346. O estudo respeita as normas e diretrizes regulamentadoras para pesquisa com seres humanos que estão na Resolução 466/12 do Conselho Nacional de Saúde, prevendo a confidencialidade dos dados, garantindo sigilo e privacidade da identidade dos sujeitos, por meio da assinatura do Termo de Confidencialidade e do esclarecimento dos objetivos e procedimentos às famílias que assinaram o TCL (Termo de Consentimento Livre e Esclarecido) após terem aceitado participar da pesquisa. Os responsáveis pelos bebês responderam à entrevista sobre dados sociodemográficos, obstétricos e psicossociais adaptados na versão original de Schwengber e Piccinini(20). A amostra inicial na avaliação do SEAL foi de 101 bebês, mas destes apenas 45 crianças, 19 nascidas a termo e 26 pré- termo, acompanhadas pelo SEAL de 3 a 24 meses, foram avaliadas aos 24 meses pela Escala Bayley III. Nesta pesquisa será analisado o item linguagem. Os demais aspectos do desenvolvimento foram analisados a partir da Escala Bayley III por outros profissionais participantes da pesquisa maior. Para os bebês nascidos prematuros considerou-se a idade corrigida na avaliação. Os bebês e seus familiares foram convidados a participar da pesquisa no seguimento de prematuros de um hospital universitário e no teste do Pezinho em uma Unidade Básica de Saúde próxima ao local. Como critérios de inclusão buscou-se bebês sem limitações biológicas como lesões neurológicas ou síndromes, bem como déficits sensoriais (visual, auditivo, etc). Esses aspectos foram avaliados pelos pediatras, pela equipe de pesquisa que constou de fonoaudiólogos, psicólogos, fisioterapeutas e terapeutas ocupacionais. Em caso dúvida foram retirados da pesquisa e encaminhados para o neuropediatra ou geneticista. Para análise de linguagem pelo SEAL foi realizada uma filmagem da interação da mãe ou quem desempenhava essa função para o bebê, que ocorreu de diferentes modos nas etapas do estudo. A filmagem foi realizada por dois ângulos: frontal e lateral, com tempo de 15 minutos, a depender da idade do bebê e demais aspectos a serem analisados. Os bebês foram posicionados em um bebê conforto ou sentados em tapete de EVA para interagir com suas mães, em ambiente iluminado e confortável em termos de temperatura. Deveriam estar em bom estado de vigília, bem alimentados e higienizados. A filmagem foi realizada com duas filmadoras JVC Everio GZ-MG 630 colocadas em duas posições: há dois metros da mãe e do bebê, pegando a mãe posicionada de costas para filmadora e o bebê de frente, Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 2/7 e outra colocada há um metro, pegando a interação face a face da mãe com o bebê em ângulo lateral. nível de significância p ≤ 0,05. Foram utilizados os testes de correlação de Pearson e exato de Fisher. Houve uma padronização de posturas em que a criança foi observada na interação com sua mãe. Os bebês eram nascidos a termo e prematuros tardios. No caso deste grupo considerou-se a idade corrigida. As faixas etárias em que os bebês foram filmados para análise do SEAL de acordo com os quatro instrumentos semestrais criados(12-14) foram: Fase 1 – 3 meses e 1 dia a 4 meses e 29 dias – O bebê sentado no bebê conforto (9 minutos). A mãe foi instruída a cantar (3 minutos) (ambientação), conversar com o bebê (3 minutos) e oferecer um objeto -ex- um cachorro de borracha sem barulho (3 minutos). Fase 2- 8 meses e 1 dia a 9 meses e 29 dias – A mãe e bebê sentados no tapete EVA, foram filmados na interação, com a solicitação de a mãe cantar para o bebê por 3 minutos, conversar por mais 6 minutos e a brincar com um objeto (o cachorro de borracha) oferecido pelo examinador (6 minutos). Se o bebê não tivesse ainda domínio de tronco poderia utilizar bebê conforto. Fase 3 - 17 meses e um dia a 18 meses e 29 dias e Fase 4 - 23 meses e um dia a 24 meses e 29 dias – Nestas fases o bebê foi observado em atividade livre com a mãe com uma caixa de brinquedos temáticos (animais, um bebê com mamadeira, panelinhas, etc.) e observou-se o brincar e a interação linguística entre mãe e bebê. A mãe foi orientada a permanecer sobre o tapete com o bebê durante a filmagem. Nos primeiros 10 minutos foi filmada a interação da mãe com o bebê e nos últimos 5 minutos o examinador participou da interação para observar alguns sinais que abrangiam o diálogo com distintos interlocutores. Os vídeos foram assistidos por duas fonoaudiólogas habilitadas que atribuíram os sinais dos instrumentos do SEAL aos bebês, o que permitiu verificar uma concordância entre ambas 95 e 100%. Para a análise aqui apresentada foram considerados os valores atribuídos pela pesquisadora principal, primeira autora deste artigo. Na Tabela 1 estão resumidos os sinais enunciativos analisados. Todos os sujeitos também foram avaliados na fase 4 (faixa etária de 24 meses) por meio da Escala Bayley III(1,11) por uma profissional habilitada para esta avaliação, sendo considerada nesta pesquisa a subescala de linguagem (comunicação receptiva e comunicação expressiva). Inicialmente foi encontrado o ponto de partida no teste de cada bebê, com base na sua idade. A avaliação teve início assim que o bebê pontuou consecutivamente as três primeiras questões (ponto base), e terminou com cinco erros seguidos. Para a análise estatística utilizou-se um banco de dados em Excel em que foram organizados os dados de linguagem obtidos para presença e ausência de sinais do SEAL em cada faixa etária e total, bem como também os escores do Bayley III obtidos para 24 meses. Os resultados foram analisados estatisticamente no software STATISTICA 9.1. Para este estudo, considerou-se o RESULTADOS Foram analisados 45 bebês, foi verificada a correlação entre o número total de sinais do SEAL obtido por meio dos quatro instrumentos semestrais e os resultados de linguagem obtidos por meio da Escala Bayley III aos 24 meses. Os resultados pontuaram uma correlação Pearson de 0,718 e um p_valor de 0,001, indicando significância estatística (p<0,05). Isso permite concluir que quanto maior a pontuação do SEAL, maior a pontuação no Bayley III. Na análise comparativa entre crianças com atraso de linguagem e sem atraso de linguagem por meio dos resultados obtidos na avaliação de linguagem na Escala Bayley III aos 24 meses, foi possível identificar o número médio de sinais enunciativos de aquisição da linguagem de cada grupo, conforme indicado na Tabela 2. Observou-se que, em média, os bebês sem atraso na aquisição da linguagem apresentaram 18 sinais presentes, e que os bebês com atraso, 12 sinais, fato que se refletiu na diferença entre ambos grupos quando comparados a ausência e presença de cada sinal como se observa na Tabela 3. Os sinais 9, 10, 16, 17, 18, 21, 22 e 23, referentes a aspectos enunciativos do bebê, e o sinal 24, relativo à posição materna na última faixa etária, diferenciaram o grupo com atraso na aquisição da linguagem do grupo sem atraso na comparação estatística da amostra. Os bebês com desenvolvimento típico tiveram maior presença desses sinais do que os bebês com atraso. A análise descritiva dos dados das 20 crianças que apresentaram alteração de linguagem pela Escala Bayley III, aos 24 meses, está descrita na Tabela 4, com a indicação dos sinais presentes e ausentes. Observou-se que as crianças com alterações pelo SEAL tiveram pontuação limítrofe, baixa ou muito baixa pela Escala Bayley III. As crianças que desenvolveram o atraso na avaliação pela Escala Bayley III aos 24 meses, apresentaram alterações em sinais da criança (fator infantil) e nos sinais relativos à atividade materna de sustentação do diálogo (fator materno). Ainda, deste grupo de 20 crianças com alteração pela Escala Bayley III, quatro não obtiveram risco no SEAL, de acordo com o número médio de sinais (Tabela 2), bem como não apresentaram alterações do fator materno (sinais em negrito). As quatro crianças sem risco pelo SEAL, mas com atraso pela Escala Bayley III, apresentaram avaliações entre limítrofe e média baixa. Nenhuma delas recebeu a atribuição de extremamente baixa. Outro aspecto relevante na Tabela 4 é a variação na pontuação obtida na Escala Bayley III (limítrofe, muito baixa, extremamente baixa) que ressalta a maior parte das crianças na categoria extremamente baixa, mais indicativa de grave atraso, com os menores valores no SEAL (2 a 7) na exceção do sujeito 3 (S3) com 12 sinais presentes, ainda um valor baixo, o que evidencia a concordância ao menos nos aspectos referentes ao fator infantil entre ambos testes. As crianças com as outras categorias (limítrofe e muito baixa) pontuaram valores entre 9 e 17 sinais no SEAL. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 3/7 Tabela 1. Sinais Enunciativos de Aquisição da Linguagem (SEAL) Sinais de 2 a 6 meses e 29 dias 1. A criança reage ao manhês, por meio de vocalizações, movimentos corporais ou olhar. 2. A criança preenche seu lugar na interlocução com sons verbais como vogais e/ou consoantes. 3. A criança preenche seu lugar na interlocução com sons não verbais de modo sintonizado ao contexto enunciativo (sorriso, grito, choro, tosse, resmungo). 4. A criança preenche seu lugar na interlocução silenciosamente apenas com movimentos corporais e olhares sintonizados ao contexto enunciativo. 5. A criança inicia a conversação ou protoconversação. Locutor analisado bebê bebê bebê bebê bebê 6. A criança e a mãe (ou sua substituta) trocam olhares durante a interação. bebê-mãe 7. A mãe (ou sua substituta) atribui sentido às manifestações verbais e não verbais do bebê, e sustenta essa protoconversação ou conversação, quando o bebê a inicia. 8. A mãe (ou sua substituta) utiliza o manhês falando com a criança de modo sintonizado ao que está acontecendo no contexto e aguardando as respostas do bebê. Sinais de 7 a 12 meses e 29 dias 9. A criança preenche seu lugar na interlocução (enunciado) com sons verbais (sílabas com vogais e consoantes variadas - ao menos dois pontos e dois modos articulatórios de consoantes). 10. A criança esboça a produção de protopalavras por espelhamento à fala da mãe (ou substituto). 11. A criança esboça a produção de protopalavras espontaneamente. 12. Quando a mãe (ou substituta) é convocada a enunciar pela criança, a mesma produz seu enunciado e aguarda a resposta da criança. Sinais de 13 a 17 meses e 29 dias 13. A criança nomeia de modo espontâneo e inteligível ao adulto interlocutor, objetos que estão ausentes no contexto. 14. A criança nomeia de modo espontâneo, mas não inteligível ao adulto interlocutor, objetos que estão ausentes no contexto, buscando na prosódia uma forma de ser compreendida. 15. A criança nomeia de modo espontâneo e inteligível ao adulto interlocutor, objetos, pessoas, ações, que estão presentes no contexto enunciativo. 16. A criança faz gestos para tentar fazer-se entender quando o adulto interlocutor não a compreende. 17. A criança repete o dizer do adulto interlocutor como forma de organizar ou reorganizar sua enunciação, por exemplo, aprimorando a forma sintática, ou fonológica, ou a escolha do item lexical ou mesmo acentuando algum item prosodicamente. 18. A criança conversa com diferentes interlocutores adultos (pai, mãe, examinador). 19. O adulto interlocutor atribui um sentido possível às produções verbais da criança, ou seja, de modo sintonizado. Sinais 18 a 24 meses e 29 dias 20. A criança solicita objetos e/ou pede esclarecimentos ao adulto interlocutor, marcando sua posição como locutor. 21. A criança utiliza formas fonêmicas distintas para veicular sentidos diferentes em sua enunciação (ao menos dois pontos articulatórios – labial e alveolar- e duas classes sonoras consonantais distintas – ao menos nasais e plosivas). 22. A criança utiliza distintas formas (palavras) para veicular sentidos diferentes em sua enunciação. 23. A criança combina palavras, na forma direta ou inversa, para veicular sentidos diferentes. 24. Quando a criança apresenta produções verbais distintas da fala adulta, o adulto interlocutor reage fazendo um pedido de reparo neutro (o que) ou repetindo corretamente a fala infantil ou oferecendo item lexical compatível com a intenção comunicativa do bebê. Fonte: Crestani et al.(12,13), Fattore et al.(14). Em negrito sinais significativos na análise fatorial Tabela 2. Comparação pontuação total SEAL versus escala Bayley III SEM RISCO LINGUAGEM Bayley III 24 meses N Média (±DP) Mínimo Máximo COM RISCO LINGUAGEM Bayley III 24 meses N Média (±DP) Mínimo Máximo mãe mãe bebê bebê bebê mãe bebê bebê bebê bebê bebê bebê mãe bebê bebê bebê bebê mãe p_valor 25 18,85 (± 2,92) 15,00 22,00 20 12,39 (± 5,13) 2,00 22,00 0,001 Significativo pelo teste U de Mann-Whitney Legenda: SEAL = Sinais Enunciativos de Aquisição da Linguagem; DP = Desvio Padrão; N = Número de Sujeitos Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 4/7 TESTES SEAL TOTAL Tabela 3. Análise comparativa por sinal dos bebês com e sem atraso em linguagem na escala Bayley III SINAIS SEAL 1 (B) 2 (B) 3 (B) 4 (B) 5 (B) 6 (B) 7 (A) 8 (A) 9 (B) 10 (B) 11 (B) 12 (A) 13 (B) 14 (B) 15 (B) 16 (B) 17 (B) 18 (B) 19 (A) 20 (B) 21 (B) 22 (B) 23 (B) 24 (A) Bebês com atraso n=20 Bebês sem atraso n=25 Díades com sinal ausente Díades com sinal presente Díades com sinal ausente Díades com sinal presente Comparação n 2 6 2 2 10 3 2 5 11 13 11 2 20 20 15 9 16 14 11 8 12 17 15 10 % 10 30 10 10 50 15 10 25 55 65 55 10 100 100 75 45 80 70 55 40 60 85 75 50 n 18 14 18 18 10 17 18 15 9 7 9 17 0 0 5 11 4 6 9 12 8 3 5 10 % 90 70 90 90 50 85 90 75 45 35 45 85 0 0 25 55 20 30 45 60 40 15 25 50 N 2 3 2 2 11 2 1 3 6 7 9 3 22 25 4 0 4 3 7 4 1 4 4 1 % 8 12 8 8 44 8 4 12 24 28 36 12 88 100 6 0 6 12 28 6 4 6 6 4 n 23 22 23 23 14 23 24 22 19 18 16 22 3 0 21 25 21 22 18 21 24 21 21 24 % 92 88 92 92 56 92 96 88 76 72 64 88 12 0 94 100 94 88 72 94 96 84 84 96 p-valor 0,606 0,131 0,606 0,606 0,460 0,392 0,415 0,229 0,034* 0,014* 0,165 0,632 0,625 1,000 0,0001* 0,0002* 0,0000* 0,0001* 0,063 0,071 0,0000* 0,0000* 0,0001* 0,0005* *Significativo teste exato de Fisher Legenda: Negrito = sinais importantes na análise fatorial (Crestani et al.(12,13), Fattore et al.(14)); B = Sinal do Bebê; A = Sinal do Adulto; n = Número de Bebês; SEAL = Sinais Enunciativos de Aquisição da Linguagem Tabela 4. Análise descritiva dos Sinais Enunciativos de Aquisição da Linguagem nos casos de atraso em linguagem no Bayley III aos 24meses Sujeito Sinais Presentes SEAL Sinais Ausentes SEAL Total SEAL Presentes Bayley III 24m S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 S19 S20 1,2,3,4,5,6,7,8,12,16,18,19,20, 21,24 9,10,11, 13, 14, 15,17, 22,23 1,3,4,5,6,7,8,16,18, 1,2,3,4,5,6,7,8,9,10,11,12, 1,3,4,6,7,8,12, 16,19 2,9,10,11,12, 13,14,15,17 19, 20,21,22,23,24 13,14,15,16,17,18,19,2021,22,23, 24 2,5,9,10,11,13,14,15,17,18,20,21,22,23,24 1,2,3,4,5,6,7,8,9,10,11,12,15,16,18,19,20,21,24 13,14,17,22,23 1,2,3,4,5,6,7,8,12, 20,21,24 9,10,11,13,14,15,16,17, 18,19, 22,23 6,12 1,2,3,4,5,7,8,9,10,11,13, 14,15,16,17,18,19,20,21,22, 23,24 1,3,4,6,7,8,12,20,21,22,23,24 2,5,9,10,11,13,14,15,16, 17,18,19 1,2,3,4,5,6,7,8,9,10,11,12,16,19,20,21,22 13,14,15,17,18,23,24 1,2,3,4,7,11,12, 5,6,8,9,10,13,14,15,16,17,18,19,20,21,22,23,24 1,2,3,4,5,6,10,11,12,20,21,22,23 7,8,9,13,14,15,16,17,18,19,24 1,2,3,4,5,6,7,8,9,10,11,12,15,16,17,18,19,20,21,22,2 3,24 13,14 1,2,3,4,7,9,11 5,6,8,10,12,13,14,15,16,17,18,19,20,21,22,23,24 2,7,8,15,16,17,18,19,20,21,22,24 1,3,4,5,6,9,1011,12,13,14,23 1,3,4,6,7,12,16,20,21,24 1,2,3,4,5,6,7,8,9,10,11,12,16, 2,5,8,9,1011,13,14,15,17,18,19,22,23 13,14,15,17,18,19,20,21,22,23,24 1,2,3,4,6,7,8,9,12,15,16,17,18,19,20,21,22,23,24 1,2,3,4,5,6,7,8,9,10,11,12,15,16,17,18,19,20,21,22,2 3,24 5,10,11,13,14 13,14 1,3,4,6,7,8,12,13,20,21 2,5,9,10,11,14,15,16,17,18,19, 22,23,24 1,2,3,4,6,7,8,9,12,20,21,22,23,24 5,10,11,13,14,15,16,17,18,19 Legenda: L = Limítrofe; EB = Extremamente Baixa; MB = Média Baixa; SEAL = Sinais Enunciativos de Aquisição da Linguagem 15 9 12 9 19 12 2 12 17 7 13 22 7 12 10 13 19 22 10 14 79 (L) 71 (L) 59 (EB) 77 (L) 83 (MB) 79 (L) 65 (EB) 47 (EB) 85 (MB) 59 (EB) 74 (L) 77 (L) 47 (EB) 79 (L) 79 (L) 77 (L) 79 (L) 83 (MB) 83 (MB) 89 (MB) Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 5/7 DISCUSSÃO A correlação positiva entre casos com atraso de linguagem avaliados pela Escala Bayley III e casos de risco pelo SEAL permite afirmar que este teste foi efetivo como teste de triagem na amostra estudada. Na Tabela 4, observou-se que 16 dos 20 casos avaliados com atraso pela Escala Bayley III, estiveram com alteração importante no SEAL. Nesta pesquisa, o SEAL foi atribuído por filmagens, em acompanhamentos de puericultura, embora possa ser analisado pelo profissional habilitado em sua formação, durante uma observação da interação mãe-bebê em ambulatório, o que seria menos dispendioso em termos de tempo e custo para inserção no SUS. Nesta condição, os casos suspeitos pelo SEAL seriam enviados para um teste diagnóstico com a Escala Bayley III ou outras escalas diagnósticas em linguagem e desenvolvimento(21,22). Cabe destacar que o número de díades mãe-bebê participantes na investigação do SEAL em todas as faixas etárias foi muito superior (101) ao grupo que compareceu aos dois encontros de avaliação da Escala Bayley III (45). Essa perda amostral sugere que a adesão ao realizar o teste mais demorado (a Escala Bayley III) por parte dos usuários demanda uma mudança de cultura e melhoria no acesso ao serviço de saúde, algo que não se vislumbra mudar em uma perspectiva de curto prazo. Já a comparação realizada entre o grupo com atraso e sem atraso permitiu estabelecer uma média de 18 sinais enunciativos, entre os 24 avaliados, como ausência de risco. Esse dado sugere a necessidade de dar continuidade na investigação do SEAL em termos de estabelecimento de critérios por faixa etária e no total do teste, o que não foi possível com a pequena amostra obtida nesta pesquisa. Na Tabela 3, observa-se que há sinais como o 13 e 14 que estiveram ausentes em ambos grupos, embora o sinal 14 tenha sido indicado como relevante na análise fatorial(14). Entre os sinais que diferiram estatisticamente na comparação entre bebês com atraso e sem atraso, os Sinais 9, 10, 21, 22 e 23 evidenciam habilidades de o bebê sem atraso na linguagem de ocupar seu lugar de enunciação com vocalização e fala crescentemente mais complexas (diversidade fonológica e lexical, e uso inicial de sintaxe). Já bebês sem esses sinais pode evidenciar o possível atraso na aquisição da linguagem. O sinal 16 se relaciona, por outro lado, ao uso da gestualidade como forma de comunicação, o que é previsto por estudos de aquisição da linguagem que afirmam haver uma continuidade e sincronia entre habilidades verbais e gestualidade do bebê(23). O sinal 17 relaciona-se à capacidade de o bebê se ancorar na fala da mãe para aprimorar o seu dizer, uma estratégia identificada por Silva(19) no processo de aquisição da linguagem. Ela também está relacionada a uma disposição do adulto em socorrer a criança em seu dizer, avaliada no sinal 24. Portanto, é importante observar a força de ambos sinais simultaneamente na amostra estudada. Além deles, o sinal 18 que se relaciona à amplitude de interlocutores, é fundamental na avaliação não só da disjunção em termos de relação enunciativa de aquisição, mas também do processo de separação mãe-bebê, como visto no estudo de Flores e Souza(24), em que bebês em sofrimento psíquico e dificuldades no processo de separação e operação da função paterna, evidenciaram dificuldade de falar com distintos interlocutores. O fato de nenhum sinal ter sido distintivo na comparação entre ambos grupos na primeira faixa etária (zero a seis meses) indica a necessidade de continuar investigando e aprimorando o instrumento, visto que a análise fatorial demonstrou três sinais nessa fase relativos a um fator infantil e um materno em uma amostra maior de sujeitos nesta fase(13). Esses resultados indicam a necessidade de continuar os estudos de validação de critério. Nos dados observados na Tabela 4, cabe destacar que os sujeitos 5, 12, 17 e 19, apresentaram 18 ou mais sinais enunciativos presentes. Cumpre ressaltar que esses quatro sujeitos não apresentaram alterações no fator materno, assim como o sujeito 5 com atraso pelo SEAL e pela Escala Bayley III, enquanto todos os demais sim. Eles também não obtiveram a classificação “extremamente baixa” na Escala Bayley III(1). Esse dado permite observar que, na maior parte dos casos, houve a contribuição do fator materno para a emergência e compreensão do funcionamento de linguagem nos casos de atraso na aquisição da linguagem, ou seja, a forma como o adulto realiza a sustentação enunciativa é um aspecto que deve ser considerado na avaliação e intervenção junto aos atrasos de aquisição da linguagem(17,18). Diversos estudos do campo enunciativo(12-19) têm evidenciado que tanto a aquisição da linguagem, quanto a clínica com crianças pequenas, devem investir na análise do diálogo mãe-bebê para propor uma hipótese de funcionamento de linguagem(16) que permita propor linhas de intervenção. Essa hipótese de funcionamento de linguagem prevê a relação EU (criança) – TU (adulto) no entendimento do sofrimento advindo do atraso ou distúrbio de linguagem. A partir dessa perspectiva teórica e dos resultados observados nesta pesquisa, observou-se que o SEAL é promissor por avaliar tal fator cuja importância foi identificada na presença do mesmo em 16 dos 20 casos avaliados como atraso de linguagem pela Escala Bayley III. Destaca-se a importância de que crianças com valores extremamente baixos pela Escala Bayley III também foram as com menores valores no SEAL. Considerando a limitação numérica da amostra deste estudo, os resultados encontrados sugerem a necessidade de continuar investigando a linguagem de bebês e crianças pequenas com o SEAL, pois realça um modo efetivo de acompanhar a aquisição da linguagem na puericultura e propor intervenções a tempo de impedir a cristalização de sintomas de linguagem(17,18). Para tanto, será necessário estabelecer critérios para o teste em amostras maiores, visto que este foi um estudo clínico, de menor proporção, com os bebês que compareceram à avaliação realizada com a Escala Bayley III no desfecho da pesquisa aos dois anos, número muito inferior ao desejado. As avaliações com o SEAL não possuem fins diagnósticos, uma vez que o bebê está em processo de constituição linguística, mas buscam oferecer intervenções oportunas que possam favorecer um encontro entre os familiares e o bebê ou criança pequena, pois facilitar o fator materno, que se apresentou importante nos estudos fatoriais, é uma forma de fortalecer o encontro e sincronia linguística entre a mãe (ou sua substituta) e o bebê(13- 15). Isso evidencia que a Fonoaudiologia poderia se beneficiar Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 6/7 de estudos que tomem o diálogo como foco da análise nas pesquisas de aquisição da linguagem, pois as habilidades infantis para ocupar seu lugar de enunciação são tão relevantes no processo de aquisição, quanto a sustentação que o adulto faz de um lugar de enunciação. CONCLUSÃO É possível concluir que houve correlação significativa entre o desempenho no SEAL entre 3 e 24 meses e o desfecho de linguagem aos 24 meses avaliado pela Escala Bayley III. A comparação entre bebês com e sem atraso na aquisição da linguagem permitiu estabelecer médias de sinais no SEAL e em alguns sinais do bebê e um da mãe apresentaram diferenças estatísticas na comparação dos dois grupos da amostra estudada quanto a sua presença e ausência, sobretudo a partir da segunda faixa etária estudada. Tais dados permitem afirmar que o SEAL tem potencial como teste de triagem e deve ser investigado em amostras maiores, pois é rápido e exige apenas a observação da interação mãe-bebê no primeiro e segundo anos de vida, em meio a uma brincadeira naturalista com materiais acessíveis aos examinadores e às famílias. REFERÊNCIAS 1. Bayley N. Bayley scales of infant and toddler development. 3. ed. San Antonio: The Psychological Corporation; 2006. 2. Godamunne P, Liyanage C, Wimaladharmasooriya N, Pathmeswaran A, Wickremasinghe AR, Patterson C, et al. Comparison of performance of Sri Lankan and US children on cognitive and motor scales of the Bayley scales of infant development. BMC Res Notes. 2014;7(1):300. http://dx.doi. org/10.1186/1756-0500-7-300. PMid:24886547. 3. Ballot DE, Ramdin T, Rakotsoane D, Agaba F, Davies VA, Chirwa T, et al. Use of the Bayley scales of infant and toddler development, third edition, to assess developmental outcome in infants and young children in an urban setting in South Africa. Int Sch Res Notices. 2017;2017:1631760. http:// dx.doi.org/10.1155/2017/1631760. PMid:28835912. 4. Springer PE, Slogrove AL, Laughton B, Bettinger JA, Saunders HH, Molteno CD, et al. Neurodevelopmental outcome of HIV – exposed but uninfected infants in the mother an infants health study, Cape Town, South Africa. Trop Med Int Health. 2018;23(1):69-78. http://dx.doi.org/10.1111/ tmi.13006. PMid:29131457. 5. Shapiro KA, Kim H, Mandelli ML, Rogers EE, Gano D, Ferriero DM, et al. Early changes in brain structure correlate with language outcomes in children with neonatal encephalopathy. Neuroimage Clin. 2017;15:572-80. http:// dx.doi.org/10.1016/j.nicl.2017.06.015. PMid:28924555. 6. Goh SKY, Tham EKH, Magiati I, Sim L, Sanmugam S, Qiu A, et al. Analysis of item-level bias in the Bayley-III language subscales: the validity and utility of standardized language assessment in a multilingual setting. J Speech Lang Hear Res. 2017;60(9):2663-71. http://dx.doi. org/10.1044/2017_JSLHR-L-16-0196. PMid:28813555. 7. Fairbairn N, Galea C, Loughran-Fowlds A, Hodge A, Badawi N, Walker K. Prediction of three year outcomes using the Bayley-III for surgical, cardiac and healthy Australian infants at one year of age. Early Hum Dev. 2018;117:57-61. http://dx.doi.org/10.1016/j.earlhumdev.2017.12.012. PMid:29288912. 8. Khurana S, Shivakumar M, Reddy GVSK, Jayashree P, Bhat YR, Lewis LES, et al. Long-term neurodevelopment outcome of caffeine versus aminophylline therapy for apnea of prematurity. J Neonatal Perinatal Med. 2017;10(4):355-62. http://dx.doi.org/10.3233/NPM-16147. PMid:29286928. 9. Wang H, Zhou H, Zhang Y, Wang Y, Sun J. Association of maternal depression with dietary intake, growth, and development of preterm infants: a cohort study in Beijing, China. Front Med. 2018;12(5):533-41. http:// dx.doi.org/10.1007/s11684-017-0591-y. PMid:29181690. 10. Nunes SF, Moraes AB, Busanello-Stella AR, Roth-Hoogstraten AM, Souza APR. Risco psíquico e desenvolvimento infantil: importância da detecção precoce na puericultura. Saúde. 2020;46(2):e47856. http://dx.doi. org/10.5902/2236583447856. 11. Madaschi V, Mecca TP, Macedo EC, Paula CS. Bayley-III scales of infant and toddler development: transcultural adaptation and psychometric properties. Paidéia. 2016;26(64):189-97. 12. Crestani AH, Moraes AB, Souza APR. Validação de conteúdo: clareza, pertinência, fidedignidade e consistência interna de sinais enunciativos de aquisição da linguagem. CoDAS. 2017;29(4):e20160180. http://dx.doi. org/10.1590/2317-1782/201720160180. PMid:28813071. 13. Crestani AH, Moraes AB, Souza AM, Souza APR. Construct validation of enunciative signs of language acquisition for the first year of life. CoDAS. 2020;32(3):e20180279. http://dx.doi.org/10.1590/2317-1782/20202018279. PMid:32578837. 14. Fattore IM, Moraes AB, Crestani AH, Souza AM, Souza APR. Validação de conteúdo e construto de sinais enunciativos de aquisição da linguagem no segundo ano de vida. CoDAS. 2022;34(2):e20200252. http://dx.doi. org/10.1590/2317-1782/20202020252. PMid:34932657. 15. Cardoso JL, Sachetti JK. Enunciação e distúrbios de linguagem: sobre a questão da nomeação. RevLet. 2010;2(2):105-19. 16. Bender S, Surreaux LM. Os efeitos da fala da criança: a escuta do sintoma na clínica de linguagem. Cad IL. 2011;42:129-45. 17. Oliveira LD, Ramos-Souza AP. O distúrbio de linguagem em dois sujeitos com risco ao desenvolvimento em uma perspectiva enunciativa do funcionamento de linguagem. Rev CEFAC. 2014;16(5):1700-12. http:// dx.doi.org/10.1590/1982-0216201410713. 18. Nazario CG, Rechia IC, Fattore IM, Nunes SF, Souza APR. Comparação entre avaliações de linguagem na infância e sua relação com risco psíquico. Distúrbios Comun. 2019;31(1):104-18. http://dx.doi.org/10.23925/2176- 2724.2019v31i1p104-118. 19. Silva CLC. Os movimentos enunciativos da criança na linguagem. Rev ABRALIN. 2011;10(4):77-94. http://dx.doi.org/10.5380/rabl.v10i4.32424. 20. Schwengber DDS, Piccinini CA. Depressão materna e interação mãe-bebê no final do primeiro ano de vida. Psic Teor Pesq. 2004;20(3):233-40. http:// dx.doi.org/10.1590/S0102-37722004000300004. 21. Rocha SR, Dornelas LF, Magalhães LC. Instrumentos utilizados para avaliação do desenvolvimento de recém-nascidos pré-termo no Brasil: revisão da literatura. Cad Ter Ocup UFSCar. 2013;21(1):109-17. http:// dx.doi.org/10.4322/cto.2013.015. 22. Miranda JR, Silva MA, Mendonça EJ. Fo, Bandeira DR. Evidências de validade de critério do Inventário Dimensional de Avaliação do Desenvolvimento Infantil para rastreio do transtorno do espectro autismo. Neuropsicol Latinoam. 2020;12(3):19-29. 23. Cavalcante MCB. Contribuições dos estudos gestuais para as pesquisas em aquisição da linguagem. Ling Ensino. 2018;21:5-35. 24. Flores MR, Souza APR. Diálogo de pais e bebês em situação de risco ao desenvolvimento. Rev CEFAC. 2014;16(3):840-52. http://dx.doi. org/10.1590/1982-0216201411412. Contribuição dos autores LDO foi responsável pela concepção do estudo, coleta, análise e redação do artigo; ABM realizou a análise estatística; SFN realizaou a coleta Bayley III; IC foi responsável pela análise estatística; APRS foi responsável pela concepção do estudo e revisão do artigo. Oliveira et al. CoDAS 2023;35(3):e20210221 DOI: 10.1590/2317-1782/20232021221pt 7/7
10.2196_43669	JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Original Paper Effectiveness of an Immersive Telemedicine Platform for Delivering Diabetes Medical Group Visits for African American, Black and Hispanic, or Latina Women With Uncontrolled Diabetes: The Women in Control 2.0 Noninferiority Randomized Clinical Trial Suzanne E Mitchell1,2,3, MSc, MD; Alexa Bragg3, BS; Barbara A De La Cruz2, MPH; Michael R Winter4, MPH; Matthew J Reichert5, PhD; Lance D Laird3, ThD; Ioana A Moldovan2, BA; Kimberly N Parker2, MS; Jessica Martin-Howard3, MA, MPH; Paula Gardiner1, MPH, MD 1Department of Family Medicine and Community Health, University of Massachusetts Chan Medical School, Worcester, MA, United States 2Department of Family Medicine, Boston Medical Center, Boston, MA, United States 3Department of Family Medicine, Boston University School of Medicine, Boston, MA, United States 4Biostatistics and Epidemiology Data Analytics Center, Boston University School of Public Health, Boston, MA, United States 5Weatherhead Center for International Affairs, Harvard University, Cambridge, MA, United States Corresponding Author: Suzanne E Mitchell, MSc, MD Department of Family Medicine and Community Health University of Massachusetts Chan Medical School 55 Lake Avenue North Worcester, MA, 01655 United States Phone: 1 9789856033 Email: [email protected] Abstract Background: Medically underserved people with type 2 diabetes mellitus face limited access to group-based diabetes care, placing them at risk for poor disease control and complications. Immersive technology and telemedicine solutions could bridge this gap. Objective: The purpose of this study was to compare the effectiveness of diabetes medical group visits (DMGVs) delivered in an immersive telemedicine platform versus an in-person (IP) setting and establish the noninferiority of the technology-enabled approach for changes in hemoglobin A1c (HbA1c) and physical activity (measured in metabolic equivalent of task [MET]) at 6 months. Methods: This study is a noninferiority randomized controlled trial conducted from February 2017 to December 2019 at an urban safety net health system and community health center. We enrolled adult women (aged ≥18 years) who self-reported African American or Black race or Hispanic or Latina ethnicity and had type 2 diabetes mellitus and HbA1c ≥8%. Participants attended 8 weekly DMGVs, which included diabetes self-management education, peer support, and clinician counseling using a culturally adapted curriculum in English or Spanish. In-person participants convened in clinical settings, while virtual world (VW) participants met remotely via an avatar-driven, 3D VW linked to video teleconferencing. Follow-up occurred 6 months post enrollment. Primary outcomes were mean changes in HbA1c and physical activity at 6 months, with noninferiority margins of 0.7% and 12 MET-hours, respectively. Secondary outcomes included changes in diabetes distress and depressive symptoms. Results: Of 309 female participants (mean age 55, SD 10.6 years; n=195, 63% African American or Black; n=105, 34% Hispanic or Latina; n=151 IP; and n=158 in VW), 207 (67%) met per-protocol criteria. In the intention-to-treat analysis, we confirmed noninferiority for primary outcomes. We found similar improvements in mean HbA1c by group at 6 months (IP: –0.8%, SD 1.9%; VW: –0.5%, SD 1.8%; mean difference 0.3, 97.5% CI –∞ to 0.3; P<.001). However, there were no detectable improvements in physical activity (IP: –6.5, SD 43.6; VW: –9.6, SD 44.8 MET-hours; mean difference –3.1, 97.5% CI –6.9 to ∞; P=.02). The proportion of participants with significant diabetes distress and depressive symptoms at 6 months decreased in both groups. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Conclusions: In this noninferiority randomized controlled trial, immersive telemedicine was a noninferior platform for delivering diabetes care, eliciting comparable glycemic control improvement, and enhancing patient engagement, compared to IP DMGVs. Trial Registration: ClinicalTrials.gov NCT02726425; https://clinicaltrials.gov/ct2/show/NCT02726425 (J Med Internet Res 2023;25:e43669) doi: 10.2196/43669 KEYWORDS type 2 diabetes mellitus; virtual world; women; digital health; diabetes self-management education; self-management; health equity; Hispanic or Latina; Black or African American; group visit; shared medical appointment Introduction Methods Minority and low-income women with type 2 diabetes mellitus (T2DM) face widening disparities in diabetes care and clinical outcomes, highlighting the pressing need to improve diabetes care for underserved communities [1-5]. Diabetes medical group visits (DMGVs) are shared appointments where groups of patients receive diabetes self-management education (DSME), peer support, and a clinical visit within a 2-hour appointment. Compared to usual care for adults living with diabetes, the in-person (IP) DMGV model has been associated with improved diabetes outcomes and lower costs [6-9]. Moreover, receiving care as a group can reduce disparities by fostering more equitable patient-provider relationships, creating relationships of care between patients, and improving health literacy and self-management skills [10]. Yet, many disadvantaged communities report poor access to DMGVs as health systems find them difficult to implement [11,12]. Patient engagement in group-based diabetes care is also often low due to social stigma, lack of transportation, and time constraints [11,13]. Telehealth solutions have gained unprecedented traction with the onset of the COVID-19 pandemic. Early evidence has shown that virtual worlds (VWs) and virtual reality platforms are feasible and potentially more effective alternatives to IP programming [14,15]. A VW is a 3D, computer-based simulated environment where users engage in immersive, experiential learning with animated educational content [16]. Users create avatars, digital manifestations of self, to engage in peer group programming virtually [17]. This environment is intrinsically designed for users to enact behavioral change among peers and restructure old habits [18,19]. To our knowledge, the possibilities of avatar-based VW DSME have not been rigorously tested. We developed an immersive telemedicine platform, linking an interactive VW learning environment with videoconferencing software, to overcome the common barriers to diabetes group-based care while maintaining clinical effectiveness at scale. We implemented Women in Control 2.0 (WIC2) in 2015 to study the comparative effectiveness of delivering DMGVs in a VW versus the traditional IP classroom for women from Black, African American, Hispanic, or Latina backgrounds with uncontrolled T2DM (trial protocol in Multimedia Appendix 1) [20]. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX Trial Design From February 2017 to October 2019, we recruited 17 cohorts of African American or Black or Hispanic or Latina women with uncontrolled T2DM. A total of 309 participants were enrolled and randomly assigned to the VW or IP DMGV conditions. Participants attended 8 weekly DMGVs and were followed for 6 months. Participants Eligible participants were adult women (≥18 years) who self-identified as African American, Black, Hispanic, or Latina with uncontrolled T2DM, defined by a hemoglobin A1c (HbA1c) value ≥8%. Participants were English-speaking or Spanish-speaking, had telephone access, permanent or stable housing, a clinician-supervised diabetes treatment plan, and could provide informed consent. Exclusion criteria included scheduling conflicts with DMGV programming, enrollment in another program, a history of diabetic ketoacidosis, oxygen-dependent chronic obstructive pulmonary disease, stroke within the last 6 months, and an acute coronary event or chronic heart condition within the last year. Pregnancy, recent glucocorticoid therapy, dialysis, active substance abuse, active cancer treatment, and any medical contraindications to study dietary recommendations were also exclusionary. Recruitment Overview We identified participants from Boston Medical Center and a local community health center using a weekly electronic medical record query. We contacted eligible participants with an introductory letter and follow-up call [21]. Additional recruitment strategies included participant or provider referrals and posted flyers. We screened participants by phone and reconfirmed eligibility at an IP enrollment appointment. Participants provided written informed consent and were eligible for up to US $300 in compensation or a new laptop. Randomization and Masking After stratification by language, we used one-to-one block randomization (alternating blocks of 6 and 8) to assign participants to the VW or IP DMGV conditions. A biostatistician generated the randomization sequence, and randomization occurred after informed consent. We randomized participants prior to obtaining baseline data. Investigators were blinded to the randomization process, but assignments were revealed to participants and investigators post consent. J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Intervention Assigned to cohorts of 6-12 participants based on study arm, participants convened in clinical or virtual settings for 8 weekly DMGVs. Each session lasted approximately 120 minutes and started with the completion of an intake form to document acute or chronic symptoms, health system usage, and self-management activities, followed by the measurement of vital signs and the delivery of DSME. Sessions included a one-on-one clinical consult. Study clinicians were 4 board-certified physicians and 2 nurse practitioners. Nonclinical group facilitators received training on core DSME topics and facilitation skills from lead faculty (SEM, PG). All participants received a paper curriculum booklet. Prior to the first virtual DMGV, staff provided laptops and wireless internet to VW participants and conducted IP computer training. All participants then met weekly for 8 weeks, according to a session schedule. During DMGVs, all participants received the same WIC2 curriculum, which was adapted from Power to Prevent [22] and consisted of 8 modules highlighting topics such as diabetes self-monitoring, preventative care, healthy eating, exercise, and stress management. Three bilingual staff, who were native Spanish speakers and included a certified interpreter, used the forward-backward method of translation, in tandem, to produce a culturally equivalent, Spanish-language Figure 1. Illustration of avatars in the virtual world. curriculum [23,24]. The curriculum content was reviewed by 2 patient advisory groups. To develop avatar-driven learning experiences and incorporate chat and telehealth capabilities, instructional design was adapted for a VW environment using game design theory [25]. VW participants customized avatars to represent themselves and engage in DMGVs, including practicing positive health behaviors such as dance and social support (Figure 1). During each session, a clinician met individually with participants (in a separate physical space or via secure telehealth platform or telephone depending on study group) to review blood glucose readings and hyper or hypo glycemic data, conduct diabetes medication reconciliation, and address concerns. Recommendations for medication adjustments were based on an algorithm [26] and shared with primary care providers via progress notes in the electronic health record. To ensure fidelity of DMGV protocols and standard operating procedures, we used checklists, audits of session recordings, and participant observation field notes. Following the 8-week DMGV sessions, participants entered a 16-week maintenance period. They were encouraged to self-monitor (tracking blood glucose, blood pressure, diet, and exercise) using a paper booklet or mobile app. No formal DMGVs occurred. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Outcomes The primary outcomes were mean changes in (1) HbA1c and (2) physical activity (metabolic equivalent of task [MET]-hours) by study arm from baseline to 6 months. Secondary outcomes included mean changes in HbA1c, physical activity, and medication changes at 9 weeks after enrollment and changes in the 17-item Diabetes Distress (DD) scale, depressive symptoms (Patient Health Questionnaire-8 [PHQ-8]), physical function (Patient-Reported Outcomes Measurement Information System [PROMIS-29] measure), 13-item Patient Activation Measure (PAM), weight, and step count at 6 months [27-31]. data included collection Data Collection and Management Baseline sociodemographic characteristics, HbA1c values (blood draw), physical activity (per accelerometers), and survey measures. Baseline HbA1c was obtained within 30 days of the first DMGV. Physical activity was measured within a 14-day window, with participants wearing an accelerometer on the wrist for 7 consecutive days. Follow-up data collection occurred 9 weeks and 6 months post enrollment (within a 28-day window). Study data were stored using secure Research Electronic Data Capture (REDCap) software hosted by Boston University [32,33]. Unique study identification numbers were used to label all participant forms. Sample Size Calculations We used the average overtime change in HbA1c and physical activity by study arm as coprimary outcomes, measured from baseline to 6-month follow-up. Data obtained from the WIC 1.0 pilot study was used to estimate the sample size necessary to establish the noninferiority margin of VW DMGVs compared to IP DMGVs at reducing HbA1c and increasing total physical activity levels [34]. For HbA1c, we assumed a noninferiority margin of 0.7 based on a clinically meaningful decrease [35,36], a pooled SD of 2, an α of .05, and a power of 80% based on the pilot study results [34]. For physical activity, we assumed a noninferiority margin of 12, a pooled SD of 35, an α of .05, and a power of 80%. We required 106 participants per arm. We did not expect the dropout rate for WIC2 to exceed 7%; thus, we aimed to enroll and randomize 228 and retain 212 participants. Statistical Analysis Sociodemographic characteristics were compared by arm using chi-square and Fisher exact tests as appropriate for categorical variables and 2 sample t tests or Wilcoxon rank sum tests for continuous variables. Within-group changes from baseline to follow-up on mean HbA1c and mean physical activity between the VW and IP study arms were assessed with paired t tests; between-group changes were assessed with multiple linear regression models, both at an α level of .025 after applying a Bonferroni correction for multiple testing. Between-group differences in the likelihood of achieving a 0.4% reduction or more in HbA1c at follow-up were examined by logistic regression. One-sided P values and 97.5% CIs were calculated to assess the noninferiority hypothesis, using margins of 0.7% for HbA1c and 12 MET-hours for physical activity. Other statistical intervals were 2-sided. Per-protocol (PP) analyses were limited to participants who tests and confidence https://www.jmir.org/2023/1/e43669 XSL•FO RenderX completed the protocol as intended, by attending ≥6 out of 8 DMGVs. PP and intention-to-treat (ITT) analyses were conducted on a full data set that used multiple imputation via predictive mean matching to impute missing baseline, 9-week, and 6-month primary and secondary outcomes. Analyses were replicated on unimputed data to check the sensitivity of results to imputation. Accelerometry data was used to calculate participants’ mean change in physical activity behavior from baseline to 6 months. For each participant, we randomly selected the 2 weekdays with the longest wear-time. We considered missing wear time data in a 24-hour day as sedentary activity. For each weekday, we estimated total MET-hours by a weighted sum of the number of hours in light (1.5 MET), moderate (4 MET), vigorous (6 MET), and very vigorous (8 MET) activity as measured by the accelerometer using the Freedson et al cut points [37,38]. We then averaged the estimated MET-hours for the 2 weekdays with the longest wear-time to obtain weekday average MET-hours per participant. Sensitivity analyses were performed to evaluate the influence of language preference on our primary outcome results. Participant characteristics with, versus without, baseline HbA1c were assessed to detect potential bias from missing data. Characteristics of participants who adhered to the session protocol (attended ≥6 vs <6 sessions) were also assessed, and primary outcome PP analyses were replicated controlling for characteristics found to be correlated with protocol adherence. All analyses were performed using SAS/STAT software (SAS version 9.4; SAS Institute) or the R programming language (version 3.4.3; R Core Team). Ethics Approval This study was conducted according to the CONSORT (Consolidated Standards of Reporting Trials) guidelines [39] and approved by the Boston University/Boston Medical Center Institutional Review Board (H-34220). Results Study Population Of 1960 potentially eligible patients, 1349 were screened, and 309 participants were randomized; 29 patients had a change in eligibility status before the first DMGV (Figure 2). The PP sample included 207 (108 VW and 99 IP; 67%) participants who met a priori criteria by attending ≥6 DMGVs. At baseline, participants’ mean age was 55 (SD 10.6) years, their mean weight was 195 (SD 41.8) lb, and their mean physical activity was 104.1 (SD 34.3) MET-hours. All participants were female, with 63% (195/309) of African American or Black race and 34% (105/309) Hispanic or Latina ethnicity (Table 1). A majority (219/309, 71%) were insured by Medicaid, Medicare, or both, and 59.6% (184/309) had home internet. The mean DD score was 2.3 (SD 1.0), which is moderately high, and the PHQ-8 score for depressive symptoms was 5.5 (SD 5.0), which is mild. More IP participants owned a smartphone. Mean HbA1c values differed by VW and IP groups (mean 9.7%, SD 1.7% vs mean 10.2%, SD 1.8%), respectively. Participant characteristics with complete versus missing baseline HbA1c data and changes J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al in eligibility status were compared (Tables S1 and S2 in Multimedia Appendix 2 [27-31]). Because no participant characteristic was identified as accountable for the imbalance in mean baseline HbA1c, we attributed the difference in baseline HbA1c to random imbalance and controlled for baseline HbA1c in our outcome analyses. Figure 2. Flowchart of Women in Control 2.0 (WIC2) participants. DMGV: diabetes medical group visit; ITT: intention-to-treat. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Table 1. Baseline characteristics of Women in Control 2.0 (WIC2) study participants. Characteristic Age (years) Mean (SD) Participants, n Race, n (%) African American or Black White Missing Otherb Hispanic, Latina, or Spanish ethnicity, n (%) Spanish-speaking, n (%) Insurance, n (%) Commercial Medicare or Medicaid No insurance Education, n (%) High school graduate or less Any college, vocational, or trade school Any postgraduate Missing Employment status, n (%) Full-time Part-time Not employed Other Missing Financial insecurity, n (%) Overall (n=309)a In-person (n=151) Virtual world (n=158) 55.4 (10.6) 55.20 (9.6) 55.54 (11.5) 300 146 154 195 (63.1) 26 (8.4) 10 (3.2) 78 (25.2) 105 (34.0) 73 (23.6) 69 (22.3) 219 (70.9) 4 (1.3) 152 (49.2) 132 (42.7) 14 (4.5) 11 (3.6) 75 (24.3) 44 (14.2) 156 (50.5) 12 (3.9) 22 (7.1) 96 (63.6) 13 (8.6) 5 (3.3) 37 (24.5) 49 (32.5) 35 (23.2) 31 (20.5) 109 (72.2) 2 (1.3) 74 (49.0) 65 (43.1) 6 (4.0) 6 (4.0) 36 (23.8) 24 (15.9) 75 (49.7) 6 (4.0) 10 (6.6) 99 (62.7) 13 (8.2) 5 (3.2) 41 (25.9) 56 (35.4) 38 (24.1) 38 (24.1) 110 (69.6) 2 (1.3) 78 (49.4) 67 (42.4) 8 (5.1) 5 (3.2) 39 (24.7) 20 (12.7) 81 (51.3) 6 (3.8) 12 (7.6) Does not have enough money to make ends meet 161 (52.1) 77 (51.0) 84 (53.2) Annual income, n (%) ≤US $29,999 ≥US $30,000 Refused to answer or don’t know Missing Marital status, n (%) Married Single with partner Single for any reason Missing Health care head of household, n (%) Has internet access, n (%) Has smartphone, n (%) Low health literacy, n (%) https://www.jmir.org/2023/1/e43669 XSL•FO RenderX 166 (53.7) 59 (19.1) 75 (24.3) 9 (2.9) 74 (24.0) 34 (11.0) 192 (62.1) 9 (2.9) 118 (38.2) 184 (59.6) 237 (76.7) 87 (28.2) 87 (57.6) 24 (15.9) 35 (23.2) 5 (3.3) 34 (22.5) 17 (11.3) 95 (62.9) 5 (3.3) 57 (37.8) 90 (59.6) 125 (82.8) 44 (29.1) 79 (50.0) 35 (22.2) 40 (25.3) 4 (2.5) 40 (25.3) 17 (11.8) 97 (61.4) 4 (2.5) 61 (38.6) 94 (59.5) 112 (70.9) 43 (27.2) J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Characteristic BMI (kg/m2) Mean (SD) Participants, n HbA1c c, % Mean (SD) ≥9.0, n (%) Participants, n PHQ-8 scored Mean (SD) Participants, n Diabetes distress scoree Mean (SD) Participants, n Overall (n=309)a In-person (n=151) Virtual world (n=158) 34.12 (6.9) 33.95 (6.7) 34.29 (7.2) 285 138 147 9.93 (1.7) 181 (58.6) 284 5.49 (5.0) 299 2.69 (1.4) 298 10.16 (1.7) 95 (62.9) 138 5.10 (4.8) 145 2.65 (1.5) 145 9.70 (1.7) 86 (54.4) 146 5.86 (5.2) 154 2.72 (1.4) 153 aPercentages are based on the n value of each column (column %). bOther category: 66 participants reported “some other race”; 8 reported “multiple races”; 1 reported Native Hawaiian, Pacific Islander; and 3 reported American Indian or Alaska Native. cHbA1c: hemoglobin A1c. dAssessed using the Patient Health Questionnaire-8 (PHQ-8), which ranges from 1 to 8 [29]. eAssessed using the Diabetes Distress Scale- 17 [27,28]. Fidelity The 17 study cohorts were conducted with 98% fidelity to the 8-week curriculum. The median number of sessions attended by participants was 6 in the IP arm and 7 in the VW arm. Among participants who attended WIC2 DMGVs, 98.2% (1618/1648 total events) completed the clinician consult and intake forms. In the VW condition, participants completed the clinical consult via telehealth (540/823, 65.6%), telephone (54/823, 6.6%), or either modality (230/823, 27.9%). Coprimary Outcomes—ITT Changes in HbA1c and physical activity are reported in Tables 2 and 3. In the ITT sample, we found within-group HbA1c improvements of 0.8% among IP participants, from 10.2% at baseline to 9.4% at 6 months, and 0.5% among VW participants, from 9.7% at baseline to 9.2% at 6 months. Improvements in HbA1c values were not statistically significant between groups and were noninferior (the mean difference across study arms was 0.3 (97.5% CI –∞ to 0.3); P<.001). The upper limit did not cross the predetermined noninferiority margin of 0.7%. It would require a noninferiority margin of less than 0.3% to fail to reject the null hypothesis of inferiority. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Table 2. Changes in HbA1c among Women in Control 2.0 (WIC2) participants. Analysis type and study arm HbA1c a, % Baseline to 6 months HbA1c noninferiority (margin of 0.7) Baseline 6 months Change Mean difference (1-sided 97.5% CI) P value 0.2 (–∞ to 0.3) <.001 PPb (n=207) IPc, mean (SD) IP, mmol/mol VWe, mean (SD) 10.1 (1.8) 9.4 (2.1) 87 79 9.6 (1.7) 9.1 (1.8) VW, mmol/mol 81 76 ITTf (n=309) IP, mean (SD) IP, mmol/mol VW, mean (SD) 10.2 (1.8) 88 9.7 (1.7) VW, mmol/mol 83 9.4 (2.2) 79 9.2 (2.1) 77 aHbA1c: hemoglobin A1c. bPP: per-protocol. cIP: in-person. dN/A: not applicable. eVW: virtual world. fITT: intention-to-treat. –0.7 (1.8) N/Ad –0.5 (1.6) N/A –0.8 (1.9) N/A –0.5 (1.8) N/A 0.3 (–∞ to 0.3) <.001 Table 3. Changes in physical activity among Women in Control 2.0 (WIC2) participants. Analysis type and study arm Work week METa-hours Baseline to 6 months Work week MET-hours noninferiority (margin of 12) Baseline 6 months Change Mean difference (one-sided 97.5% CI) P value PPb (n=207) IPc, mean (SD) VWd, mean (SD) ITTe (n=309) 105.3 (35.3) 101.4 (34.9) –5.2 (37.7) 106.6 (35.4) 100.6 (38.4) –8.1 (37.7) –3.0 (–8.9 to ∞) –3.1 (–6.9 to ∞) .008 .02 IP, mean (SD) 105.5 (40.1) 106.3 (45.4) VW, mean (SD) 106.4 (37.1) 105.1 (46.9) –6.5 (43.6) –9.6 (44.8) aMET: metabolic equivalent of task. bPP: per-protocol. cIP: in-person. dVW: virtual world. eITT: intention-to-treat. For physical activity, IP and VW participants had mean within-group decreases of 6.5 MET-hours and 9.6 MET-hours, respectively. Between-group differences were not detected from baseline to post intervention. Still, the noninferiority of the VW approach was confirmed (the mean difference across arms was –3.1 MET-hours, 97.5% CI –6.9 to ∞; P=.02). It would require a noninferiority margin of less than 9 MET-hours to fail to reject the null hypothesis of inferiority. PP Results Among the 207 participants who attended at least 6 DMGVs, within-group mean HbA1c values improved by 0.7% among IP participants, from 10.1% at baseline to 9.4% at 6 months, and by 0.5% among VW participants, from 9.6% at baseline to 9.1% at 6 months. Noninferiority was confirmed with a mean difference of 0.2 (97.5% CI –∞ to 0.3; P<.001) in HbA1c across study arms. Improvements of ≥0.4% were achieved by 56% (56/99) of IP and 52% (56/108) of VW participants from baseline to 6 months, while nearly one-third (75/207, 36.02%) achieved a 1% improvement. IP and VW participants’ physical activity decreased, on average, by 5.2 MET-hours and 8.1 MET-hours, respectively. Noninferiority was confirmed in the PP sample (the mean difference across arms was –3.0, 97.5% CI –8.9 to ∞; P=.008). Similar to the ITT analyses, https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al between-group changes in HbA1c and accelerometer-measured physical activity were not statistically significant. Secondary Outcomes We analyzed mean changes in HbA1c and physical activity data at 9 weeks from baseline, which were similar to the 6-month results (Table S3 in Multimedia Appendix 2). We compared the mean change in DD, depression symptom burden, physical functioning, patient activation, weight, and step count by study arm, and adjusted for the baseline values. We observed substantial within-group improvements in both study arms for total DD and some DD subscales (emotional burden, regimen, and interpersonal). We observed substantive but nonsignificant improvements in both study arms for depression symptom burden, physical functioning, and patient activation, and mixed results for weight (Table 4). Notably, the total participants reporting moderate DD (scores of ≥2) decreased from 53% (156/294) to 33% (77/237), and the proportion with clinically meaningful depressive symptoms (scores of ≥5) decreased from 48% (142/295) to 40% (95/237) from baseline to 6 months (Table S4 in Multimedia Appendix 2). Physical activity assessment using step count revealed high baseline step counts, small decreases at 6 months, and overall null findings (Table 4). Results of sensitivity and unimputed analyses are in Table S5-S10 in Multimedia Appendix 2. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Table 4. Group changes in secondary outcomes among Women in Control 2.0 (WIC2) study participants. Baseline, mean (SD) Changes from baseline to 9 weeks Changes from baseline to 6 months LSa, mean (SE) Between-group P value LS, mean (SE) Between-group P value Total DDb PPc IPd VWe ITTf IP VW DD: emotional burden PP ITT IP VW IP VW DD: physician PP ITT IP VW IP VW DD: regimen PP ITT IP VW IP VW DD: interpersonal PP ITT PHQ-8g PP IP VW IP VW IP VW 2.3 (1.0) 2.3 (1.0) 2.2 (1.0) 2.3 (1.1) 2.7 (1.5) 2.7 (1.5) 2.6 (1.5) 2.7 (1.4) 1.5 (0.9) 1.5 (1.0) 1.5 (0.9) 1.6 (1.1) 2.7 (1.3) 2.7 (1.4) 2.6 (1.3) 2.7 (1.3) 2.0 (1.3) 2.0 (1.3) 1.9 (1.3) 2.0 (1.3) 4.9 (4.8) 5.5 (5.1) .14 .35 .01 .26 .27 .53 .08 .34 .77 .98 .66 –0.5 (0.1) –0.3 (0.1) –0.4 (0.1) –0.2 (10.1) –0.7 (0.2) –0.4 (0.2) –0.6 (0.2) –0.3 (0.2) –0.2 (0.1) 0.0 (0.1) –0.1 (0.1) 0.0 (0.1) –0.7 (0.2) –0.4 (0.2) –0.5 (0.2) –0.3 (0.2) –0.3 (0.2) –0.3 (0.2) –0.2 (0.1) –0.2 (0.2) –0.7 (0.7) –1.1 (0.7) .33 .69 .28 .68 .31 .98 .76 .56 .62 .98 .12 –0.3 (0.1) –0.2 (0.1) –0.2 (0.1) –0.2 (0.1) –0.3 (0.1) –0.2 (0.1) –0.2 (0.1) –0.2 (0.1) –0.1 (0.1) –0.1 (0.1) 0.0 (0.1) –0.1 (0.1) –0.3 (0.1) –0.3 (0.1) –0.2 (0.1) –0.2 (0.1) –0.2 (0.1) –0.2 (0.1) –0.1 (0.1) –0.1 (0.1) –0.5 (0.3) –0.3 (0.4) https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Baseline, mean (SD) Changes from baseline to 9 weeks Changes from baseline to 6 months LSa, mean (SE) Between-group P value LS, mean (SE) Between-group P value ITT IP VW Physical functioningh PP ITT IP VW IP VW Patient activationi PP ITT IP VW IP VW Weight (lb) PP ITT IP VW IP VW Step count PP ITT IP VW IP VW 5.1 (4.8) 5.9 (5.3) –0.4 (0.6) –0.7 (0.6) 45.8 (9.8) 46.5 (9.0) 45.8 (9.2) 46.1 (8.9) 67.5 (20.4) 68.8 (17.4) 66.5 (21.1) 66.0 (20.3) 193.2 (38.7) 199.7 (45.6) 194.0 (40.5) 196.2 (44.7) 2.3 (1.3) 1.0 (1.2) 2.6 (1.2) 1.1 (1.2) 7.7 (2.8) 0.1 (2.5) 5.1 (2.5) 0.4 (2.3) –0.6 (5.6) 0.8 (6.1) –0.2 (5.0) 3.1 (5.1) 9779.4 (3187.1) 397.0 (483.2) 10,008.1 (2977.4) –476.7 (417.1) 9631.4 (3304.2) 297.2 (424.4) 9798.9 (3060.2) –317.9 (378.7) .86 .42 .53 .02 .12 .02 .30 .01 .03 .99 .30 .36 .63 .36 .91 .83 .24 .35 –0.2 (0.3) –0.3 (0.3) 1.0 (0.7) 0.3 (0.6) 0.9 (0.6) 0.7 (0.5) 1.6 (1.4) 0.7 (1.2) 1.1 (1.2) 1.1 (1.2) –0.6 (2.8) –0.6 (3.1) –0.6 (2.4) 1.13 (2.9) –219.8 (217.6) –338.2 (202.4) –150.9 (198.6) –168.3 (175.0) aLS: least squares. bAssessed using the Diabetes Distress [DD] Scale-17, which ranges from 1 to 6 [27,28]. cPP: per-protocol. dIP: in-person. eVW: virtual world. fITT: intention-to-treat. gAssessed using the Patient Health Questionnaire-8 (PHQ-8), which ranges from 1 to 8 [29]. hAssessed using the physical function subscale on the Patient-Reported Outcomes Measurement Information System (PROMIS-29) measure [30]. iAssessed using the Patient Activation Measure (PAM)-13 [31]. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al to detect changes Lifestyle Behaviors Self-management behaviors were assessed through a weekly self-report in diet, exercise, and diabetes-related medication. Nearly a third of all participants (89/309, 28.8%) reported ≥1 dietary change, with a greater proportion in the VW group compared to the IP group (55/158, 34.8% vs 34/151, 22.5%). Of all participants, 65.7% (203/309) engaged in at least 20 minutes of exercise weekly during the intervention (VW: 106/158, 67.1% vs IP: 97/151, 64.2%). Only 17.2% (53/309) reported any type of change (increase, decrease, or switch) in their diabetes medication regimen (VW: 29/158, 18.4% vs IP: 24/151, 15.9%). Adverse Events One study-related severe adverse event in the VW group occurred due to emotional distress. Discussion Principal Findings To our knowledge, this is the first fully powered clinical trial to demonstrate the effectiveness of delivering DMGVs using an immersive 3D telemedicine platform versus IP care. Both approaches were similarly effective in reducing mean HbA1c over 6 months. Our PP sample is indicative of a high patient retention rate. No significant changes in physical activity were detected. Altogether, this research demonstrates that 3D immersive telemedicine DMGVs are an effective alternative to IP group diabetes care for high-risk patients in a safety net health system. Our preliminary study compared IP versus immersive DSME delivery among 89 low-income African American women with uncontrolled T2DM [34]. Results showed substantial improvements in mean HbA1c. Other pilot studies demonstrated positive impacts on patient outcomes but faced methodological limitations, including lack of a comparison group, small sample size, and inadequate power, and none used a DMGV format [40,41]. In contrast, the WIC2 study was randomized with an active DMGV control condition and fully powered to rigorously test the primary outcomes. Nearly half of our participants at baseline had measurable depressive symptoms and diabetes distress. Prior research has revealed a strong correlation between depression, diabetes distress, and uncontrolled diabetes [42,43]. Interestingly, we found that the proportion of WIC2 participants with depressive symptoms (PHQ-8≥5) and diabetes distress (DD≥2) decreased from baseline to 6-month follow-up, indicating the WIC2 intervention improves glucose control and mental health. This finding is important as interventions that address both physical and mental health can reduce patients’ treatment burden. Given our pilot study showed increased physical activity among study participants, the null finding in physical activity in WIC2 was unexpected [34]. We experienced challenges with accelerometry wear due to participants’ discomfort with the that devices. A accelerometry-measured activity for middle-aged women with chronic disease has limitations [44], such that physical activity can be underestimated or inconsistent across the life span [37,45,46]. More research is needed to establish activity assessment guidelines for older adults. literature revealed review Limitations We acknowledge several study limitations. We had a small imbalance in HbA1c at baseline. After careful assessment of participant characteristics, it was determined that this imbalance occurred at random and was unrelated to the fidelity of the study protocol. It is not possible to rule out unobserved confounding of protocol adherence, such as participants’ access to transportation, digital literacy, work or childcare conflicts, or financial constraints impacting access to medication. Finally, this study was conducted with women in an urban safety net health system, which may limit its generalizability. The ongoing challenges with access to digital resources and digital literacy for underserved communities may also limit the immediate generalizability of our study findings to similar populations. Conclusions Immersive technologies can reduce disparities by improving effectiveness and access to evidence-based diabetes care. We showed that when given the tools, adults from digitally underserved communities robustly adopt health technology tools with improved health outcomes. More effort is warranted to design technology tailored to the needs, capabilities, and life perspectives of diverse communities to avoid leaving behind those most in need of better health care. Acknowledgments We extend our gratitude to the clinicians and research study staff who supported Women in Control 2.0 (WIC2). Clinicians A Mansa Semenya, MD, MPH; Elena Hill, MD; Adi Rattner, MD; Aissatou Gueye, NP; and Cleopatra Ferrao, NP delivered the WIC2 curriculum and performed clinical consults with participants. Katherine Melo, Jenna Bhaloo, Jennifer Albuquerque, Shakiyla Woods, Maria Pompeya Gomez, and Eddie V Gomez supported participant recruitment, data collection, and intervention activities as support staff members. All study personnel received compensation for their work. We additionally thank all the women who participated in WIC2 and the community advocacy groups who helped us achieve our mission. Research reported in this publication was supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under Award Number R01DK106531. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al manuscript; and decision to submit the manuscript for publication. SEM had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Data Availability The WIC2 team agrees to share deidentified individual participant data that underlie the results reported in this study, the study protocol, and the statistical analysis plan. Data will be available for 6 months following publication for 5 years. Data will only be shared with academic researchers who provide a methodologically sound proposal to achieve aims related to primary or secondary outcomes and upon completion of a data use agreement. Requests to [email protected]. should be directed Authors' Contributions SEM and PG conceived and designed the study. SEM, PG, AB, IAM, KNP, and JM-H carried out the study intervention and collected participant data. MRW, MJR, AB, IAM, and BAD performed data validation and analyses. SEM, MRW, MJR, AB, and BAD interpreted results. All authors contributed to the preparation of the manuscript, critically reviewed the results, and approved the manuscript for submission. SEM obtained funding. JM-H, KNP, AB, and BAD provided administrative, technical, and material support. Conflicts of Interest SEM is a consultant on health communication and relationship-centered care and has provided workshops and lectures on this topic funded by pharmaceutical and other industry sponsors. No product endorsement is permitted during these programs. SEM also holds equity in See Yourself Health LLC, a digital health service provider. Multimedia Appendix 1 Women in Control 2.0 (WIC2) trial protocol. [PDF File (Adobe PDF File), 286 KB-Multimedia Appendix 1] Multimedia Appendix 2 Additional statistical analyses. [DOCX File , 92 KB-Multimedia Appendix 2] Multimedia Appendix 3 CONSORT eHEALTH checklist (V 1.6.1). [PDF File (Adobe PDF File), 14076 KB-Multimedia Appendix 3] References 1. 2. Peyrot M, Egede LE, Campos C, Cannon AJ, Funnell MM, Hsu WC, et al. Ethnic differences in psychological outcomes among people with diabetes: USA results from the second diabetes attitudes, wishes, and needs (DAWN2) study. Curr Med Res Opin 2014;30(11):2241-2254. [doi: 10.1185/03007995.2014.947023] [Medline: 25079662] Chow EA, Foster H, Gonzalez V, McIver L. The disparate impact of diabetes on racial/ethnic minority populations. Clin Diabetes 2012;30(3):130-133 [FREE Full text] [doi: 10.2337/diaclin.30.3.130] 3. Marquez I, Calman N, Crump C. A framework for addressing diabetes-related disparities in US Latino populations. J 4. Community Health 2019;44(2):412-422. [doi: 10.1007/s10900-018-0574-1] [Medline: 30264184] Peek ME, Cargill A, Huang ES. Diabetes health disparities: a systematic review of health care interventions. Med Care Res Rev 2007;64(5 Suppl):101S-156S [FREE Full text] [doi: 10.1177/1077558707305409] [Medline: 17881626] 5. Walker RJ, Strom Williams J, Egede LE. Influence of race, ethnicity and social determinants of health on diabetes outcomes. 6. 7. 8. Am J Med Sci 2016;351(4):366-373 [FREE Full text] [doi: 10.1016/j.amjms.2016.01.008] [Medline: 27079342] Burke RE, Ferrara SA, Fuller AM, Kelderhouse JM, Ferrara LR. The effectiveness of group medical visits on diabetes mellitus type 2 (dm2) specific outcomes in adults: a systematic review. JBI Libr of Syst Rev 2011;9(23):833-885. [doi: 10.11124/jbisrir-2011-143] Quiñones AR, Richardson J, Freeman M, Fu R, O'Neil ME, Motu'apuaka M, et al. Educational group visits for the management of chronic health conditions: a systematic review. Patient Educ Couns 2014;95(1):3-29. [doi: 10.1016/j.pec.2013.12.021] [Medline: 24468199] Edelman D, Gierisch JM, McDuffie JR, Oddone E, Williams JW. Shared medical appointments for patients with diabetes mellitus: a systematic review. J Gen Intern Med 2015;30(1):99-106 [FREE Full text] [doi: 10.1007/s11606-014-2978-7] [Medline: 25107290] https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al 9. Cunningham SD, Sutherland RA, Yee CW, Thomas JL, Monin JK, Ickovics JR, et al. Group medical care: a systematic review of health service performance. Int J Environ Res Public Health 2021;18(23):12726 [FREE Full text] [doi: 10.3390/ijerph182312726] [Medline: 34886452] 10. Thompson-Lastad A. Group medical visits as participatory care in community health centers. Qual Health Res 2018;28(7):1065-1076 [FREE Full text] [doi: 10.1177/1049732318759528] [Medline: 29781398] 11. Adjei Boakye E, Varble A, Rojek R, Peavler O, Trainer AK, Osazuwa-Peters N, et al. Sociodemographic factors associated with engagement in diabetes self-management education among people with diabetes in the United States. Public Health Rep 2018;133(6):685-691 [FREE Full text] [doi: 10.1177/0033354918794935] [Medline: 30223759] 12. Rutledge SA, Masalovich S, Blacher RJ, Saunders MM. Diabetes self-management education programs in nonmetropolitan counties - United States, 2016. MMWR Surveill Summ 2017;66(10):1-6 [FREE Full text] [doi: 10.15585/mmwr.ss6610a1] [Medline: 28448482] Shaw K, Killeen M, Sullivan E, Bowman P. Disparities in diabetes self-management education for uninsured and underinsured adults. Diabetes Educ 2011;37(6):813-819. [doi: 10.1177/0145721711424618] [Medline: 22021026] 13. 14. Rush KL, Hatt L, Janke R, Burton L, Ferrier M, Tetrault M. The efficacy of telehealth delivered educational approaches for patients with chronic diseases: a systematic review. Patient Educ Couns 2018;101(8):1310-1321. [doi: 10.1016/j.pec.2018.02.006] [Medline: 29486994] 15. Mitchell SE, Mako M, Sadikova E, Barnes L, Stone A, Rosal MC, et al. The comparative experiences of women in control: diabetes self-management education in a virtual world. J Diabetes Sci Technol 2014;8(6):1185-1192 [FREE Full text] [doi: 10.1177/1932296814549829] [Medline: 25212580] 16. Boulos MN, Hetherington L, Wheeler S. Second life: an overview of the potential of 3-D virtual worlds in medical and health education. Health Info Libr J 2007;24(4):233-245 [FREE Full text] [doi: 10.1111/j.1471-1842.2007.00733.x] [Medline: 18005298] Peterson M. Learning interaction in an avatar-based virtual environment: a preliminary study. PacCALL J 2005;1(1):29-40 [FREE Full text] Fox J, Bailenson JN. Virtual self-modeling: the effects of vicarious reinforcement and identification on exercise behaviors. Media Psychol 2009;12(1):1-25. [doi: 10.1080/15213260802669474] 17. 18. 19. Webb TL, Joseph J, Yardley L, Michie S. Using the internet to promote health behavior change: a systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. J Med Internet Res 2010;12(1):e4 [FREE Full text] [doi: 10.2196/jmir.1376] [Medline: 20164043] 20. Mitchell S, Gardiner PM, Weigel G, Rosal M. Women in control: pioneering diabetes self-management medical group visits in the virtual world. J Clin Trials 2016;6(3):272 [FREE Full text] [doi: 10.4172/2167-0870.1000272] [Medline: 35495550] 22. 21. Mitchell S, Bragg A, Moldovan I, Woods S, Melo K, Martin-Howard J, et al. Stigma as a barrier to participant recruitment of minority populations in diabetes research: development of a community-centered recruitment approach. JMIR Diabetes 2021;6(2):e26965 [FREE Full text] [doi: 10.2196/26965] [Medline: 33938811] Power to prevent: A family lifestyle approach to diabetes prevention. National Diabetes Education Program. URL: http:/ /www.adph.org/diabetes/assets/powertoprevent2007.pdf [accessed 2023-03-28] Sousa VD, Rojjanasrirat W. Translation, adaptation and validation of instruments or scales for use in cross-cultural health care research: a clear and user-friendly guideline. J Eval Clin Pract 2011;17(2):268-274. [doi: 10.1111/j.1365-2753.2010.01434.x] [Medline: 20874835] 23. 24. Maneesriwongul W, Dixon JK. Instrument translation process: a methods review. J Adv Nurs 2004;48(2):175-186. [doi: 10.1111/j.1365-2648.2004.03185.x] 25. Koster R. A Theory of Fun for Game Design. Sebastopol: O'Reilly Media Inc; 2013. 26. Garber AJ, Handelsman Y, Grunberger G, Einhorn D, Abrahamson MJ, Barzilay JI, et al. Consensus statement by the 27. American Association of Clinical Endocrinologists and American College of Endocrinology on the comprehensive type 2 diabetes management algorithm - 2020 executive summary. Endocr Pract 2020;26(1):107-139. [doi: 10.4158/CS-2019-0472] [Medline: 32022600] Fisher L, Hessler DM, Polonsky WH, Mullan J. When is diabetes distress clinically meaningful?: establishing cut points for the diabetes distress scale. Diabetes Care 2012;35(2):259-264 [FREE Full text] [doi: 10.2337/dc11-1572] [Medline: 22228744] Polonsky WH, Fisher L, Earles J, Dudl RJ, Lees J, Mullan J, et al. Assessing psychosocial distress in diabetes: development of the diabetes distress scale. Diabetes Care 2005;28(3):626-631. [doi: 10.2337/diacare.28.3.626] [Medline: 15735199] 29. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression 28. in the general population. J Affect Disord 2009;114(1-3):163-173. [doi: 10.1016/j.jad.2008.06.026] [Medline: 18752852] 30. Craig BM, Reeve BB, Brown PM, Cella D, Hays RD, Lipscomb J, et al. US valuation of health outcomes measured using the PROMIS-29. Value Health 2014;17(8):846-853 [FREE Full text] [doi: 10.1016/j.jval.2014.09.005] [Medline: 25498780] 31. Hibbard JH, Mahoney ER, Stockard J, Tusler M. Development and testing of a short form of the patient activation measure. Health Serv Res 2005 Dec;40(6 Pt 1):1918-1930 [FREE Full text] [doi: 10.1111/j.1475-6773.2005.00438.x] [Medline: 16336556] https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 14 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al 32. Harris PA, Taylor R, Minor BL, Elliott V, Fernandez M, O'Neal L, REDCap Consortium. The REDCap consortium: building an international community of software platform partners. J Biomed Inform 2019;95:103208 [FREE Full text] [doi: 10.1016/j.jbi.2019.103208] [Medline: 31078660] 33. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42(2):377-381 [FREE Full text] [doi: 10.1016/j.jbi.2008.08.010] [Medline: 18929686] 34. Rosal MC, Heyden R, Mejilla R, Capelson R, Chalmers KA, Rizzo DePaoli M, et al. A virtual world versus face-to-face intervention format to promote diabetes self-management among African American women: a pilot randomized clinical trial. JMIR Res Protoc 2014;3(4):e54 [FREE Full text] [doi: 10.2196/resprot.3412] [Medline: 25344620] 35. Taveira TH, Friedmann PD, Cohen LB, Dooley AG, Khatana SA, Pirraglia PA, et al. Pharmacist-led group medical appointment model in type 2 diabetes. Diabetes Educ 2010;36(1):109-117. [doi: 10.1177/0145721709352383] [Medline: 19966072] 36. Khatana SA, Taveira TH, Choudhary G, Eaton CB, Wu WC. Change in hemoglobin A(1c) and C-reactive protein levels in patients with diabetes mellitus. J Cardiometab Syndr 2009;4(2):76-80. [doi: 10.1111/j.1559-4572.2008.00042.x] [Medline: 19614793] Freedson PS, Melanson E, Sirard J. Calibration of the computer science and applications, Inc. accelerometer. Med Sci Sports Exerc 1998;30(5):777-781. [doi: 10.1097/00005768-199805000-00021] [Medline: 9588623] 37. 39. 38. Matthews CE, Freedson PS, Hebert JR, Stanek EJ, Merriam PA, Ockene IS. Comparing physical activity assessment methods in the seasonal variation of blood cholesterol study. Med Sci Sports Exerc 2000;32(5):976-984. [doi: 10.1097/00005768-200005000-00015] [Medline: 10795789] Schulz KF, Altman DG, Moher D, CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMC Med 2010;8:18 [FREE Full text] [doi: 10.1186/1741-7015-8-18] [Medline: 20334633] Johnson C, Feinglos M, Pereira K, Hassell N, Blascovich J, Nicollerat J, et al. Feasibility and preliminary effects of a virtual environment for adults with type 2 diabetes: pilot study. JMIR Res Protoc 2014;3(2):e23 [FREE Full text] [doi: 10.2196/resprot.3045] [Medline: 24713420] 40. 41. Ruggiero L, Moadsiri A, Quinn LT, Riley BB, Danielson KK, Monahan C, et al. Diabetes island: preliminary impact of a virtual world self-care educational intervention for African Americans with type 2 diabetes. JMIR Serious Games 2014;2(2):e10 [FREE Full text] [doi: 10.2196/games.3260] [Medline: 25584346] Fisher L, Mullan JT, Arean P, Glasgow RE, Hessler D, Masharani U. Diabetes distress but not clinical depression or depressive symptoms is associated with glycemic control in both cross-sectional and longitudinal analyses. Diabetes Care 2010;33(1):23-28 [FREE Full text] [doi: 10.2337/dc09-1238] [Medline: 19837786] 42. 43. Katon WJ, Von Korff M, Lin EH, Simon G, Ludman E, Russo J, et al. The pathways study: a randomized trial of collaborative care in patients with diabetes and depression. Arch Gen Psychiatry 2004;61(10):1042-1049. [doi: 10.1001/archpsyc.61.10.1042] [Medline: 15466678] 44. Moldovan IA, Bragg A, Nidhiry AS, De La Cruz BA, Mitchell SE. The physical activity assessment of adults with type 2 diabetes using accelerometer-based cut points: scoping review. Interact J Med Res 2022;11(2):e34433 [FREE Full text] [doi: 10.2196/34433] [Medline: 36066937] 45. Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc 2008;40(1):181-188. [doi: 10.1249/mss.0b013e31815a51b3] [Medline: 18091006] 46. Miller NE, Strath SJ, Swartz AM, Cashin SE. Estimating absolute and relative physical activity intensity across age via accelerometry in adults. J Aging Phys Act 2010;18(2):158-170 [FREE Full text] [doi: 10.1123/japa.18.2.158] [Medline: 20440028] Abbreviations CONSORT: Consolidated Standards of Reporting Trials DD: diabetes distress DMGV: diabetes medical group visit DSME: diabetes self-management education HbA1c: hemoglobin A1c IP: in-person ITT: intention-to-treat MET: metabolic equivalent of task PHQ-8: Patient Health Questionnaire-8 PP: per-protocol T2DM: type 2 diabetes mellitus VW: virtual world WIC2: Women in Control 2.0 https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 15 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Mitchell et al Edited by A Mavragani; submitted 19.10.22; peer-reviewed by M Lall, S Pesälä; comments to author 10.01.23; revised version received 12.01.23; accepted 10.03.23; published 10.05.23 Please cite as: Mitchell SE, Bragg A, De La Cruz BA, Winter MR, Reichert MJ, Laird LD, Moldovan IA, Parker KN, Martin-Howard J, Gardiner P Effectiveness of an Immersive Telemedicine Platform for Delivering Diabetes Medical Group Visits for African American, Black and Hispanic, or Latina Women With Uncontrolled Diabetes: The Women in Control 2.0 Noninferiority Randomized Clinical Trial J Med Internet Res 2023;25:e43669 URL: https://www.jmir.org/2023/1/e43669 doi: 10.2196/43669 PMID: 37163341 ©Suzanne E Mitchell, Alexa Bragg, Barbara A De La Cruz, Michael R Winter, Matthew J Reichert, Lance D Laird, Ioana A Moldovan, Kimberly N Parker, Jessica Martin-Howard, Paula Gardiner. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 10.05.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e43669 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e43669 \| p. 16 (page number not for citation purposes)
10.3390_ijms22073558	Article Changes in Serum MicroRNAs after Anti-IL-5 Biological Treatment of Severe Asthma Manuel J. Rial 1,2, José A. Cañas 2,3 Beatriz Sastre 2,3 , Joaquín Sastre 1,3 and Victoria del Pozo 2,3,* , José M. Rodrigo-Muñoz 2,3 , Marcela Valverde-Monge 1 , 1 Allergy Unit, Hospital Universitario Fundación Jiménez Díaz, 28040 Madrid, Spain; [email protected] (M.J.R.); [email protected] (M.V.-M.); [email protected] (J.S.) 2 Department of Immunology, IIS-Fundación Jiménez Díaz, 28040 Madrid, Spain; [email protected] (J.A.C.); [email protected] (J.M.R.-M.); [email protected] (B.S.) 3 CIBER de Enfermedades Respiratorias (CIBERES), Instituto de Salud Carlos III, 28029 Madrid, Spain * Correspondence: [email protected]; Tel.: +34-9155-048-91 Abstract: There is currently enough evidence to think that miRNAs play a role in several key points in asthma, including diagnosis, severity of the disease, and response to treatment. Cells release different types of lipid double-membrane vesicles into the extracellular microenvironment, including exosomes, which function as very important elements in intercellular communication. They are capable of distributing genetic material, mRNA, mitochondrial DNA, and microRNAs (miRNAs). Serum miRNA screening was performed in order to analyze possible changes in serum miRNAs in 10 patients treated with reslizumab and 6 patients with mepolizumab after 8 weeks of treatment. The expression of miR-338-3p was altered after treatment (p < 0.05), although no signiﬁcant differences between reslizumab and mepolizumab were found. Bioinformatic analysis showed that miR-338-3p regulates important pathways in asthma, such as the MAPK and TGF-β signaling pathways and the biosynthesis/degradation of glucans (p < 0.05). However, it did not correlate with an improvement in lung function. MiRNA-338-3p could be used as a biomarker of early response to reslizumab and mepolizumab in severe eosinophilic asthmatic patients. In fact, this miRNA could be involved in airway remodeling, targeting genes related to MAPK and TGF-β signaling pathways. Citation: Rial, M.J.; Cañas, J.A.; Rodrigo-Muñoz, J.M.; Valverde-Monge, M.; Sastre, B.; Sastre, J.; del Pozo, V. Changes in Serum MicroRNAs after Anti-IL-5 Biological Treatment of Severe Asthma. Int. J. Mol. Sci. 2021, 22, 3558. Keywords: severe asthma; biomarkers; microRNAs; anti-IL5 biologics; mepolizumab; reslizumab https://doi.org/10.3390/ijms22073558 Academic Editor: Nicola Scichilone Received: 26 February 2021 Accepted: 27 March 2021 Published: 30 March 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1. Introduction Airway inﬂammation in asthma is distinguished by an imbalance between T1/T2 immune responses. T-helper 2 (Th2) cells produce interleukin (IL)-4, IL-5, IL-6, and IL-13, which are responsible for the allergic immune response [1]. A consequence of the upregula- tion of T2 cytokines is airway eosinophilic inﬂammation and the subsequent manifestations of asthma. Eosinophils participate in the initiation phase towards Th2 polarization, in the suppression of the Th1/Th17 pathways in the pulmonary lymph nodes, in the recruit- ment of Th2 cells in the lung, and in the mechanisms of resolution of inﬂammation that restore pulmonary homeostasis [2]. All these characteristics of eosinophils imply that they play an important role in asthma, not only as a biomarker of severity and/or control of asthma symptoms but also as a regulator of multiple functions. In fact, there is cur- rently an important therapeutic development underway with biologics, whose target is the eosinophil directly (anti-IL5Rα: benralizumab) [3] or indirectly (anti-IL-5: mepolizumab and reslizumab) [4,5]. Additionally, eosinophils can release exosomes containing speciﬁc and nonspeciﬁc proteins as well as microRNAs (miRNAs); those have recently gained im- portance as regulatory elements that can be transferred to recipient cells. To date, asthmatic patients have been identiﬁed to have differential expression of more than 100 miRNAs when compared with healthy subjects [6]. Int. J. Mol. Sci. 2021, 22, 3558. https://doi.org/10.3390/ijms22073558 https://www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2021, 22, 3558 2 of 9 MiRNAs are short RNA sequences (around 19–22 nucleotides in length) with crucial functions in post-transcriptional regulation. They are involved in the development and continuation of pathogenic mechanisms of several diseases, including asthma [7]. One of the unique characteristics of miRNAs is that they are resistant to degradation by nucleases, making them very promising and stable biomarkers. It is possible that asthmatic inﬂam- mation may be maintained as a consequence of altered miRNA expression proﬁles, which regulate some of the complex inﬂammatory processes that occur in asthma [8]. All these characteristics make miRNAs an emerging ﬁeld in allergic asthma, speciﬁcally as possi- ble biomarkers. For clinical application, the most important criteria for a diagnostic and prognostic biomarker should be high sensitivity and speciﬁcity. However, there are several limitations that hinder the use of miRNAs in clinical practice, such as the lack of studies with a large sample size, the criteria used in clinical applications (like age and gender), and the methods used (such as equipment, techniques or qualiﬁed staff) [9]. Regarding current clinical practice, there is a lack of standardized protocols on the use of miRNAs; however, there is promising evidence to believe that they constitute a useful tool for future use [10,11]. For example, our research group has demonstrated that serum miRNAs can differentiate between different asthma severities grades [12]. Likewise, a study made by our lab group has recently described that miRNAs can be used as a useful tool to predict early response to benralizumab in severe eosinophilic asthmatics [13]. The usefulness of miRNAs as markers of the therapeutic response to inhaled treatment has been previously explored [14]; however, the change in these miRNAs, secondary to biological treatment, has been poorly addressed. The purpose of this study is to analyze possible changes in serum miRNAs in patients treated with reslizumab and mepolizumab after 8 weeks of treatment. Both treatments are humanized immunoglobulin G (IgG) monoclonal antibodies with a high afﬁnity for IL-5, neutralizing this cytokine by binding to epitopes on the IL-5-Rα binding domain. 2. Results 2.1. Administration of Anti-IL-5 Drugs Improves Asthma Symptoms To develop this study, a total of 16 medicated severe eosinophilic patients with anti-IL- 5 drugs were included; 6 were treated with mepolizumab and 10 with reslizumab. Clinical and demographic characteristics before the ﬁrst administration of biologicals are shown in Table 1. The studied population consisted of adult patients (58 ± 13 years), and mostly, they were women (68.75%). Regarding inﬂammatory characteristics, more than half of the recruited patients were atopic, and a high mean of immunoglobulin E (IgE) was observed (Table 1). Additionally, this asthmatic population showed peripheral blood eosinophil counts higher than normal reference levels as well as raised fractional exhaled nitric oxide (FeNO) levels (Table 1). After 8 weeks of biological treatment, 87.5% of the patients improved their lung function in a signiﬁcant way (1.81 ± 0.93 vs. 2.14 ± 0.92 L; p < 0.001). Additionally, with respect to peripheral blood eosinophils, all asthmatic patients presented a decrease in peripheral eosinophil counts, reaching normal levels. This reduction was very signiﬁcant (649.86 ± 798.19 vs. 58.81 ± 40.49 cells/mm3; p < 0.0001), corroborating that biological drugs were able to decrease the peripheral blood eosinophils (Figure 1). In view of these data, mepolizumab and reslizumab are able to improve clinical parameters from severe asthmatics patients in a time-point of 8 weeks. 2.2. MiRNA Deregulation after Anti-IL5 Treatment A miRNA screening was performed with serum samples from nine treated asthmatic patients using the miRNA PCR array, evaluating 179 miRNAs. Results from the miRNA array showed that miR-195-5p and miR-27b-3p were downregulated (p < 0.05), while miR-1260a (p < 0.05), miR-193a-5p (p < 0.01), and miR-338-3p (p < 0.05) were upregulated at 8 weeks (Figure 2). Int. J. Mol. Sci. 2021, 22, 3558 3 of 9 Afterward, these results were validated by RT-qPCR of serum samples from 16 severe asthmatics. Nine of them were the same patients that we used in the miRNA PCR array, and the rest were different patients treated with mepolizumab or reslizumab. We conﬁrmed that only miR-338-3p was signiﬁcantly upregulated (p < 0.05) in these patients after 8 weeks of treatment (Figure 3). Table 1. Initial demographic and clinical characteristics of the 16 patients. Demographic and Clinical Characteristics Age 1 Female (%) Age at onset <30 years (%) >30 years (%) Body Mass Index 1 Smoking status Never (%) Passive (%) Former smoker (%) Smoker (%) Atopy (%) Total IgE 1 (kU/L) Eosinophils (cells/mm3)1 FeNO 1 (ppb) FEV1 Pre-BD (%) 1 FEV1 Post-BD (%) 1 FVC Pre-BD (%) 1 FVC Post-BD (%) 1 FEV1/FVC Pre-BD 1 FEV1/FVC Post-BD 1 ACT 1 58 ± 13 11 (68.75) 35.7 64.3 26.90 ± 5.29 62.5 6.25 25 6.25 53.8 603.7 ± 663.3 493 ± 321 56.08 ± 38.1 74.69 ± 29.21 80.25 ± 31.98 86.87 ± 20.24 87.62 ± 38.85 69.13 ± 11.64 70.2 ± 9.73 13.77 ± 6.2 Demographic Inﬂammatory characteristics Functional parameters Questionaries Results are expressed as mean ± SD. Figure 1. Anti-IL-5 biologics improved lung function and decreased peripheral eosinophils counts. Lung function (a) and peripheral eosinophil levels (b) were measured at baseline and an 8-week follow-up visit. After 8 weeks of mepolizumab or reslizumab drug administration, patients recovered their FEV1 and lessened their peripheral eosinophils counts. * p < 0.001, p < 0.0001. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 10 FEV1 Post-BD (%) 1 80.25 ± 31.98 FVC Pre-BD (%) 1 86.87 ± 20.24 FVC Post-BD (%) 1 87.62 ± 38.85 FEV1/FVC Pre-BD 1 69.13 ± 11.64 FEV1/FVC Post-BD 1 70.2 ± 9.73 Questionaries ACT 1 13.77 ± 6.2 1 Results are expressed as mean ± SD. The studied population consisted of adult patients (58 ± 13 years), and mostly, they were women (68.75%). Regarding inflammatory characteristics, more than half of the re-cruited patients were atopic, and a high mean of immunoglobulin E (IgE) was observed (Table 1). Additionally, this asthmatic population showed peripheral blood eosinophil counts higher than normal reference levels as well as raised fractional exhaled nitric oxide (FeNO) levels (Table 1). After 8 weeks of biological treatment, 87.5% of the patients improved their lung func-tion in a significant way (1.81 ± 0.93 vs. 2.14 ± 0.92 L; p < 0.001). Additionally, with respect to peripheral blood eosinophils, all asthmatic patients presented a decrease in peripheral eosinophil counts, reaching normal levels. This reduction was very significant (649.86 ± 798.19 vs. 58.81 ± 40.49 cells/mm3; p < 0.0001), corroborating that biological drugs were able to decrease the peripheral blood eosinophils (Figure 1). Figure 1. Anti-IL-5 biologics improved lung function and decreased peripheral eosinophils counts. Lung function (a) and peripheral eosinophil levels (b) were measured at baseline and an 8-week follow-up visit. After 8 weeks of mepolizumab or reslizumab drug administration, patients recov-ered their FEV1 and lessened their peripheral eosinophils counts. * p < 0.001, **** p < 0.0001. In view of these data, mepolizumab and reslizumab are able to improve clinical pa-rameters from severe asthmatics patients in a time-point of 8 weeks. 2.2. MiRNA Deregulation after Anti-IL5 Treatment A miRNA screening was performed with serum samples from nine treated asthmatic patients using the miRNA PCR array, evaluating 179 miRNAs. Results from the miRNA array showed that miR-195-5p and miR-27b-3p were downregulated (p < 0.05), while miR-1260a (p < 0.05), miR-193a-5p (p < 0.01), and miR-338-3p (p < 0.05) were upregulated at 8 weeks (Figure 2). Int. J. Mol. Sci. 2021, 22, 3558 4 of 9 Figure 2. Serum miRNA deregulation in severe asthmatic patients treated with anti-IL-5 drugs. Among the nine patients analyzed, ﬁve were severe asthmatics treated with mepolizumab and four with reslizumab. Eosinophilic asthmatic patients showed an altered expression of miR-195-5p, miR-27b-3p, miR-1260a, miR-423-3p, miR-193a-5p, and miR-338-3p at eight weeks after anti-IL-5 administration. Relative miRNA expression is expressed as 2−∆∆Ct. * p < 0.05, ** p < 0.01. Figure 3. Individual variation of miR-195-5p, miR-27b-3p, miR-1260a, miR-423-3p, miR-193a-5p, and miR-338-3p after 8 weeks of treatment. Relative expressions of these miRNAs were validated by RT-qPCR in serum samples from 16 asthmatic patients treated with mepolizumab or reslizumab. Among these miRNAs, miR-338-3p was the only one that modiﬁed its expression in a signiﬁcant way. All experiments were performed in triplicate. Relative miRNA expression is expressed as 2−∆Ct. Blue dots and lines represent patients treated with mepolizumab. Black dots and lines represent patients treated with reslizumab. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 4 of 10 Figure 2. Serum miRNA deregulation in severe asthmatic patients treated with anti-IL-5 drugs. Among the nine patients analyzed, five were severe asthmatics treated with mepolizumab and four with reslizumab. Eosinophilic asthmatic patients showed an altered expression of miR-195-5p, miR-27b-3p, miR-1260a, miR-423-3p, miR-193a-5p, and miR-338-3p at eight weeks after anti-IL-5 admin-istration. Relative miRNA expression is expressed as 2−ΔΔCt. * p < 0.05, ** p < 0.01. Afterward, these results were validated by RT-qPCR of serum samples from 16 se-vere asthmatics. Nine of them were the same patients that we used in the miRNA PCR array, and the rest were different patients treated with mepolizumab or reslizumab. We confirmed that only miR-338-3p was significantly upregulated (p < 0.05) in these patients after 8 weeks of treatment (Figure 3). Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 5 of 10 Figure 3. Individual variation of miR-195-5p, miR-27b-3p, miR-1260a, miR-423-3p, miR-193a-5p, and miR-338-3p after 8 weeks of treatment. Relative expressions of these miRNAs were validated by RT-qPCR in serum samples from 16 asth-matic patients treated with mepolizumab or reslizumab. Among these miRNAs, miR-338-3p was the only one that modi-fied its expression in a significant way. All experiments were performed in triplicate. Relative miRNA expression is ex-pressed as 2−ΔCt. Blue dots and lines represent patients treated with mepolizumab. Black dots and lines represent patients treated with reslizumab. In view of these results, we tried to study the effect of mepolizumab and reslizumab separately. However, no statistical differences were found when miRNA expression was compared between the mepolizumab groups and the reslizumab group (data not shown). 2.3. MiR-338-3p Regulates Important Pathways in Asthma but It Does Not Correlate with Clinical Parameters Afterward, we performed an in-silico analysis with the bioinformatic tool DIANA-mirPath to obtain the putative target genes of miR-338-3p and the altered pathways. In this analysis, an alteration in some target genes and pathways related to several functions and processes in asthma were observed, such as mitogen-activated protein kinases (MAPK) and transforming growth factor beta (TGF-β) signaling pathways and glycan bi-osynthesis/degradation (Table 2). Int. J. Mol. Sci. 2021, 22, 3558 5 of 9 In view of these results, we tried to study the effect of mepolizumab and reslizumab separately. However, no statistical differences were found when miRNA expression was compared between the mepolizumab groups and the reslizumab group (data not shown). 2.3. MiR-338-3p Regulates Important Pathways in Asthma but It Does Not Correlate with Clinical Parameters Afterward, we performed an in-silico analysis with the bioinformatic tool DIANA- mirPath to obtain the putative target genes of miR-338-3p and the altered pathways. In this analysis, an alteration in some target genes and pathways related to several func- tions and processes in asthma were observed, such as mitogen-activated protein kinases (MAPK) and transforming growth factor beta (TGF-β) signaling pathways and glycan biosynthesis/degradation (Table 2). Table 2. KEGG pathways signiﬁcantly altered by miR-338-3p. KEGG Pathway Prion diseases Fatty acid biosynthesis Fatty acid metabolism Other types of O-glycan biosynthesis p-Value 2.69−31 2.42−29 5.22−7 1.49−6 MAPK signaling pathway 0.015 Other glycan degradation TGF-beta signaling pathway Glutathione metabolism Cell cycle Mucin type O-Glycan biosynthesis 0.026 0.029 0.031 0.032 0.036 Target Genes PRNP FASN FASN OGT, POMT2, EOGT, POFUT1 FOS, CACNG8, DUSP2, ELK4, CDC25B, TAOK2, MAP4K3, MAP2K3, RASA1, ZAK, RAPGEF2, NFKB2, MAPKAP3, HSPA8, CACNA1H, MAP3K2, DUSP5, RPSKA4, NFATC3, DUSP1 NEU3 SKP1, DCN, SMAD4, SMAD5, SP1, BAMBI SRM, CDC1, GSTP1, GGT6, RRM1 YWHAH, CCNB1, CDC25B, MCM4, BUB3, SKP1, CCND1, SMAD4, CDC14B, PRKDC, MDM2, MCM3, CDC25A GALNT7, GALNT16 On the other hand, we also studied whether any of the miRNAs were correlated with circulating eosinophils at 8 weeks. Although a mild negative correlation was observed between ∆Ct values of miR-338-3p and eosinophil counts, no signiﬁcant differences were reached (r = −0.2426, p > 0.05). We also evaluated whether the levels of miR-338-3p correlated with lung function in forced expiratory volume in 1 s (FEV1) terms, but no signiﬁcant differences were found (r = 0.138; p > 0.05). 3. Discussion In this study, we described that miRNA-338-3p could be used as a biomarker of early anti-IL5 biologic response (reslizumab and mepolizumab) in severe eosinophilic asthmatic patients. The usefulness of miRNAs as markers of the therapeutic response to inhaled treatment in asthma has been previously explored [14]; however, the variation in miRNAs due to biological treatment has only been analyzed after benralizumab treatment [13]; it has never been addressed with anti-IL5 drugs such as mepolizumab and reslizumab. We observed that miRNA-338-3p is upregulated in serum from severe asthmatic patients after 8 weeks of anti-IL-5 treatment. This data concurs with results obtained Int. J. Mol. Sci. 2021, 22, 3558 6 of 9 previously [13]. In this report, our group showed that three miRNAs were altered after 8 weeks of benralizumab administration, and miR-338-3p was one of them. MiR-338 has been described to regulate differentiation, apoptosis, and probably tissue degeneration [15]. Other authors have speculated that inflammation and cell proliferation at the base of the remodeling processes can be promoted by the activation of miR-338 [16]. However, the role of miRNA-338 in lung functions has not been clearly identified, but the findings to date appear to implicate it in the pathogenesis of obstructive lung diseases [15,16]. In view of these results, we could speculate that changes in the expression of miR- 338-3p could translate into signiﬁcant changes in lung function. In most of the patients recruited, a great improvement in FEV1 was observed 8 weeks after the introduction of the anti-IL-5 drug (FEV1 baseline 1.81 ± 0.93 L; FEV1 8 weeks 2.14 ± 0.92 L); however, this improvement in lung function seems to behave independently of the variation of miR-338 at 8 weeks in this population, as correlation results have shown. One possible explanation is that this fact may be due to the small sample size used, which is one of the main limitations of the study. On the other hand, it could be related to the fact that FEV1 is not the best parameter that translates the miR-338-3p clinical expression, nor is it the parameter that changes the most with treatment, compared to other parameters such as quality of life or reduction of oral corticosteroids. So, it would be very interesting to be able to ﬁnd a biomarker that could predict a good clinical response to a biological drug in a short period of time since this would save time and money for healthcare providers in the ﬁeld of severe asthma. However, currently, there is a lack of standardized protocols about the use of miRNAs in clinical practice [10,11]; hence, further studies are needed. Despite the important ﬁndings of this study, there are some limitations. First, this study was performed with a small sample size, although the results and differences of miR-338-3p expression between the groups were found to be signiﬁcant. Second, we have included a unique population treated with anti-IL-5 biologics (reslizumab or mepolizumab); it would be interesting to study the effects of miRNA expression of reslizumab and mepolizumab separately. Hence, more studies are necessary to conﬁrm this ﬁnding on a larger scale. To our knowledge, this study supports the idea that there are differences in the expression of certain miRNAs after the introduction of an anti-IL-5 biological treatment. The expression of miR-338-3p changed after treatment; therefore, this could be used as an early response biomarker to anti-IL-5 biological drugs. However, more studies are needed to relate these changes to any clinical improvement observed after 8 weeks of treatment with an anti-IL-5 drug in patients with severe asthma. 4. Materials and Methods 4.1. Patient Selection Sixteen severe eosinophilic asthmatic patients treated with anti-IL-5 drugs were re- cruited from Hospital Universitario Fundación Jiménez Díaz in Madrid. Six of them were treated with mepolizumab and the rest with reslizumab. Asthma diagnosis and treatment were made according to the Global Initiative for Asthma (GINA) guidelines [17]. The patient inclusion criterion for the administration of mepolizumab and reslizumab was the presence of more than 300 and 400 eosinophils/µL in peripheral blood (in at least one registry in the previous year), respectively. The electronic medical record included pulmonary function, blood count, ACT, medication and visits to the emergency room. However, not all data were available in the electronic registry for all patients, so some parameters could not be included for analysis. Patients received all necessary information, and they signed a form of written informed consent to participate. The study was conducted following the principles of the Declaration of Helsinki and approved by Fundación Jiménez Díaz Ethics Committee. Int. J. Mol. Sci. 2021, 22, 3558 7 of 9 4.2. Blood Processing Peripheral blood samples were collected in anticoagulant-free tubes (Becton Dick- inson, Franklin Lakes, NJ, USA). Serum samples were obtained by blood clotting and centrifugation at 3000 rpm for 10 min at 4 ◦C. Then, they were stored at −80 ◦C until use. 4.3. MiRNAs Isolation MiRNAs were obtained from 200 µL of serum using a miRNeasy serum/plasma advanced kit (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. Three synthetic miRNA spike-ins (SP2, SP4, and SP5) were added to evaluate optimal RNA extraction (miRCURY LNA RNA Spike-in kit, Qiagen, Hilden, Germany). 4.4. cDNA Retrotranscription Serum miRNAs were retrotranscribed to cDNA using the miRCURY LNA RT Kit (Qiagen, Hilden, Germany), following the manufacturer’s protocol. Brieﬂy, 2 µL of total RNA was mixed with reverse transcription enzyme and with another synthetic miRNA Spike-in (Sp6) and cel-miR-39-3p, which accounts for control of a correct retrotranscription to cDNA, yielding a total volume of 10 µL. The reaction was performed for 60 min at 42 ◦C, then 5 min at 95 ◦C, and immediately at 4 ◦C in a Veriti Thermal Cycler (Applied Biosystems, Warrington, UK). 4.5. Serum miRNA PCR Panel In order to perform an initial serum miRNA screening of the asthmatic patients, Serum/Plasma miRNA PCR Panels (Qiagen, Hilden, Germany) to evaluate 179 miRNAs were used. Additionally, the PCR panels contained several controls for RNA isolation (UniSp2, UniSp4, and UniSp5); they also monitored cDNA synthesis (UniSp6 and cel- miR-39-3p) and checked that reaction was successful (UniSp3). Additionally, these panels included miR-451 and miR-23a, which were used as hemolysis markers. Moreover, miR- 191-5p, miR-let-7a-5p, and cel-miR-39-3p were utilized as endogenous controls for data normalization. MiRNA expression was calculated by using the 2−∆∆Ct method [18], where ∆Ct = CtmiRNA − X (CtmiR-191-5p + Ctlet-7a-5p + Ctcel-miR-39-3p) and ∆∆Ct = ∆Ct8wk − ∆CtBaseline. Nine serum samples from severe asthmatics were analyzed; ﬁve of them were treated with mepolizumab and four with reslizumab. Immediately after cDNA synthesis, it was diluted 1:30 in RNase-free water, and the reaction was performed in a Light Cycler® 96 thermocycler (Roche, Basel, Switzerland). The incubation program was carried out for 45 cycles of 95 ◦C during 10 s and 60 ◦C for 1 min. DNA melting was performed by heating at 95 ◦C for 5 s, then 65 ◦C for 1 min, and ﬁnally at 97 ◦C for 1 s. Samples were cooled for 10 s at 40 ◦C. Cycle threshold (Ct) data were taken in the last step of the melting cycle in a continuous way. 4.6. MiRNA Validation MiRNAs were validated in 16 patients, including the 9 patients used in the initial screening. Validation was performed by RT-qPCR using a miRCURY LNA SYBR Green PCR Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. Brieﬂy, cDNA from serum was diluted 1:30 in RNase-free water and then mixed with SYBR Green and the suitable miRNA probes (miR-195-5p, miR-27b-3p, miR-1260a, miR-423-3p, miR193a-3p, and miR-338-3p) (Qiagen, Hilden, Germany). Additionally, miR-191-5p, let-7a-5p, and cel-miR-39 were used as endogenous controls (Qiagen, Hilden, Germany). MiR-451a and miR-23a-3p were utilized for hemolysis control. All samples were run in triplicate, and the reaction was performed in a Light Cycler® 96 thermocycler (Roche, Basel, Switzerland); cycle threshold (Ct) values were analyzed with LightCycler® 96 SW 1.1 (Roche, Basel, Switzerland) software. MiRNA expression was calculated by using the 2−∆Ct method [18], where ∆Ct = CtmiRNA − (CtmiR-191-5p + Ctlet-7a-5p + Ctcel-miR-39-3p). Int. J. Mol. Sci. 2021, 22, 3558 8 of 9 4.7. Analysis of Pathway Enrichment In order to ﬁnd the speciﬁc pathways and biological functions that miR-338-3p is involved in, an in-silico study of enrichment analysis with this miRNA was performed by using DIANA-mirPath v3 (DIANA LAB, University of Thessaly, Thessaly, Greece) and the DIANA-TarBase database v7.0 (DIANA LAB, University of Thessaly, Thessaly, Greece) [19]. We report KEGG pathways with a p-value and a false discovery rate (FDR) of less than 0.05. 4.8. Statistical Analysis A comparison between groups was performed using the paired two-tailed t-test for parametric data, with the Wilcoxon matched-pair test for non-Gaussian data. Unpaired groups were compared by a two-tailed Student t-test for Gaussian parameters and the Mann–Whitney U-test for non-Gaussian samples. Normality was analyzed using the Shapiro–Wilk test. For the analysis of categorical variables, Fisher’s exact test was used. A p-value < 0.05 was considered signiﬁcant. Statistical calculations and graphs were performed using GraphPad Prism 8.4 (GraphPad Software Inc., San Diego, CA, USA). 5. Conclusions In this study, we report evidence supporting the potential use of miR-338-3p as a biomarker of an early anti-IL5 biologic response in severe eosinophilic asthmatic patients. In fact, this miRNA could be involved in airway remodeling, targeting genes related to MAPK and TGF-β signaling pathways. Author Contributions: M.J.R., J.A.C. and V.d.P. were involved in the design of the study; V.d.P. in funding acquisition; M.J.R., M.V.-M. and J.S. in the recruitment of patients, biological sampling, and data collection; M.J.R., J.A.C., J.M.R.-M., B.S. and V.d.P. in data curation and formal analyses; J.A.C., J.M.R.-M. and B.S., in sample processing; V.d.P. in the project supervision; M.J.R., J.A.C. and V.d.P. participated in the preparation and writing of the original draft. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Fondo de Investigación Sanitaria (FIS) and FEDER (Fondo Europeo de Desarrollo Regional; PI18/00044, and FI16/00036), CIBER de Enfermedades Respiratorias (CIBERES), Merck Health Foundation, and Ministerio de Ciencia, Innovación y Universidades (RTC- 2017-6501-1). Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Instituto de Investigación Sanitaria Fundación Jiménez Díaz. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to it includes personal data of patients. Conﬂicts of Interest: M.J.R. has received pay for lectures from Astra-Zéneca, Chiesi, Merck, GSK, Allergy Therapeutics, Novartis, ALK, TEVA, and Shire. J.A.C. has received pay for lectures from Astra Zeneca. J.S. reports having served as a consultant to Thermoﬁsher, MEDA, Novartis, Sanoﬁ, Leti, Faes Farma, Mundipharma, and GSK and having been paid lecture fees by Novartis, GSK, Stallergenes, Leti, and Faes Farma as well as having received grant support for research from Thermoﬁsher, Sanoﬁ, and ALK. V.d.P. reports having served as a consultant to Astra Zeneca and GSK and having been paid lecture fees by Astra Zeneca and GSK. The rest of the authors declare no conﬂict of interest. References 1. Murphy, D.M.; O’Byrne, P.M. Recent advances in the pathophysiology of asthma. Chest 2010, 137, 1417–1426. [CrossRef] [PubMed] 2. Melo, R.C.N.; Liu, L.; Xenakis, J.J.; Spencer, L.A. Eosinophil-derived cytokines in health and disease: Unraveling novel mecha- 3. nisms of selective secretion. Allergy 2013, 68, 274–284. [CrossRef] [PubMed] Saco, T.V.; Pepper, A.N.; Lockey, R.F. Benralizumab for the treatment of asthma. Expert Rev. Clin. Immunol. 2017, 13, 405–413. [CrossRef] [PubMed] Int. J. Mol. Sci. 2021, 22, 3558 9 of 9 4. 5. Castro, M.; Zangrilli, J.; Wechsler, M.E.; Bateman, E.D.; Brusselle, G.G.; Bardin, P.; Murphy, K.; Maspero, J.F.; O’Brien, C.; Korn, S. Reslizumab for inadequately controlled asthma with elevated blood eosinophil counts: Results from two multicentre, parallel, double-blind, randomised, placebo-controlled, phase 3 trials. Lancet Respir. Med. 2015, 3, 355–366. [CrossRef] Ortega, H.G.; Liu, M.C.; Pavord, I.D.; Brusselle, G.G.; FitzGerald, J.M.; Chetta, A.; Humbert, M.; Katz, L.E.; Keene, O.N.; Yancey, S.W.; et al. Mepolizumab Treatment in Patients with Severe Eosinophilic Asthma. N. Engl. J. Med. 2014, 371, 1198–1207. [CrossRef] [PubMed] 6. Mousavi, S.R.; Ahmadi, A.; Jamalkandi, S.A.; Salimian, J. Involvement of microRNAs in physiological and pathological processes 7. 8. in asthma. J. Cell. Physiol. 2019, 234, 21547–21559. [CrossRef] [PubMed] Sastre, B.; Cañas, J.A.; Rodrigo-Muñoz, J.M.; del Pozo, V. Novel modulators of asthma and allergy: Exosomes and microRNAs. Front. Immunol. 2017, 8, 826. [CrossRef] [PubMed] Oglesby, I.K.; McElvaney, N.G.; Greene, C.M. MicroRNAs in inﬂammatory lung disease—Master regulators or target practice? Respir. Res. 2010, 11, 148. [CrossRef] [PubMed] 9. Wang, H.; Peng, R.; Wang, J.; Qin, Z.; Xue, L. Circulating microRNAs as potential cancer biomarkers: The advantage and disadvantage. Clin. Epigenetics 2018, 10, 59. [CrossRef] [PubMed] 10. Bonneau, E.; Neveu, B.; Kostantin, E.; Tsongalis, G.J.; De Guire, V. How close are miRNAs from clinical practice? A perspective on the diagnostic and therapeutic market. Electron. J. Int. Fed. Clin. Chem. Lab. Med. 2019, 30, 114–127. 11. Condrat, C.E.; Thompson, D.C.; Barbu, M.G.; Bugnar, O.L.; Boboc, A.; Cretoiu, D.; Suciu, N.; Cretoiu, S.M.; Voinea, S.C. miRNAs as Biomarkers in Disease: Latest Findings Regarding Their Role in Diagnosis and Prognosis. Cells 2020, 9, 276. [CrossRef] [PubMed] 12. Rodrigo-Muñoz, J.M.; Cañas, J.A.; Sastre, B.; Rego, N.; Greif, G.; Rial, M.; Mínguez, P.; Mahíllo-Fernández, I.; Fernández-Nieto, M.; Mora, I.; et al. Asthma diagnosis using integrated analysis of eosinophil microRNAs. Allergy 2019, 74, 507–517. [CrossRef] 13. Cañas, J.A.; Valverde-Monge, M.; Rodrigo-Muñoz, J.M.; Sastre, B.; Gil-Martínez, M.; García-Latorre, R.; Rial, M.J.; Gómez- Cardeñosa, A.; Fernández-Nieto, M.; Pinillos-Robles, E.J.; et al. Serum micrornas as tool to predict early response to benralizumab in severe eosinophilic asthma. J. Pers. Med. 2021, 11, 76. [CrossRef] 14. Lambert, K.A.; Roff, A.N.; Panganiban, R.P.; Douglas, S.; Ishmael, F.T. MicroRNA-146a is induced by inﬂammatory stimuli in airway epithelial cells and augments the anti-inﬂammatory effects of glucocorticoids. PLoS ONE 2018, 13, e0205434. [CrossRef] [PubMed] 15. Kos, A.; Olde Loohuis, N.F.M.; Wieczorek, M.L.; Glennon, J.C.; Martens, G.J.M.; Kolk, S.M.; Aschraﬁ, A. A potential regulatory role for intronic microrna-338-3p for its host gene encoding apoptosis-associated Tyrosine Kinase. PLoS ONE 2012, 7, e31022. [CrossRef] [PubMed] 16. Lacedonia, D.; Palladino, G.P.; Foschino-Barbaro, M.P.; Scioscia, G.; Elisiana, G. Carpagnano Expression proﬁling of miRNA-145 and miRNA-338 in serum and sputum of patients with COPD, asthma, and asthma–COPD overlap syndrome phenotype. Int. J. COPD 2017, 12, 1811–1817. [CrossRef] [PubMed] 17. Global INitiative for Asthma. Global Strategy for Asthma Management and Prevention. Available online: www.ginasthma.org (accessed on 10 December 2020). 18. Livak, K.J.; Schmittgen, T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2−∆∆CT method. Methods 2001, 25, 402–408. [CrossRef] [PubMed] 19. Vlachos, I.S.; Zagganas, K.; Paraskevopoulou, M.D.; Georgakilas, G.; Karagkouni, D.; Vergoulis, T.; Dalamagas, T.; Hatzigeorgiou, A.G. DIANA-miRPath v3.0: Deciphering microRNA function with experimental support. Nucleic Acids Res. 2015, 43, 460–466. [CrossRef] [PubMed]
10.3390_ijerph16091638	Article Sickle Cell Anaemia Prevalence among Newborns in the Brazilian Amazon-Savanna Transition Region Rayane Cristina Souza 1, Pedro Agnel Dias Miranda Neto 2, Jessﬂan Rafael Nascimento Santos 3, Sílvio Gomes Monteiro 4 Rodrigo Assuncao Holanda 5 and Julliana Ribeiro Alves Santos 6,* , Maria Cláudia Gonçalves 4 , Fabrício Brito Silva 4 , 1 2 Biomédica and Mestranda em Meio Ambiente da Universidade CEUMA, São Luís 65075-120, MA, Brazil; [email protected] Faculdade Pitágoras, São Luís 65075-120, MA, Brazil; [email protected] Laboratório de Geotecnologias, Universidade CEUMA, São Luís 65075-120, MA, Brazil; jessﬂ[email protected] 3 4 Mestrado em Meio Ambiente da Universidade CEUMA, São Luís 65075-120, MA, Brazil; [email protected] (S.G.M.); mcgﬁ[email protected] (M.C.G.); [email protected] (F.B.S.) 5 Mestrado em Biologia Microbiana da Universidade CEUMA, São Luís 65075-120, MA, Brazil; [email protected] 6 Mestrado em Meio Ambiente e Mestrado em Biologia Microbiana da Universidade CEUMA, São Luís 65075-120, MA, Brazil * Correspondence: [email protected]; Fax: +98-3214-4127 Received: 16 April 2019; Accepted: 8 May 2019; Published: 10 May 2019 Abstract: Sickle cell anaemia is one of the most common hemoglobinopathies worldwide and an important public health problem in Brazil. This study evaluated the prevalence of sickle cell anaemia and its traits in newborns from the Amazon-Savanna Transition Region in the state of Maranhão, Brazil. A cross-sectional study was carried out, based on data from neonatal screening tests performed in 2013–2015 in Maranhão. The Hardy-Weinberg theorem was applied to analyse the frequency of expected homozygotes based on HbSS phenotype. A spatial-temporal distribution analysis was performed to delimit the regions with the greatest number of newborn cases with sickle cell anaemia. Of 283,003 newborns, 162 were found to have sickle cell anaemia, while 10,794 had a sickle cell trait, with a prevalence of 0.05% and 3.8%, respectively. The prevalence of expected homozygotes was higher in the North Region and in the state capital of Maranhão. This study may contribute to existing social and public health actions or the creation of new strategies for sickle cell disease in endemic areas in Brazil to improve the quality of life. Keywords: sickle cell anaemia; spatio-temporal distribution; epidemiology 1. Introduction Due to its incidence and biopsychosocial complexity, sickle cell anaemia is considered one of the main public health problems worldwide [1]. It originated in Africa and expanded to Saudi Arabia and India. The mutations causing sickle cell anaemia and other hemoglobinopathies were previously non-existent in the Americas; however, the occurrence in Brazil is due to the trade of slaves from the African continent. In Brazil, the distribution of sickle cell anaemia is heterogeneous, with a higher incidence in states such as Maranhão with socio-demographic proﬁles featuring higher poverty indexes and greater numbers of individuals of African descent [2]. The disease is an autosomal recessive genetic disorder characterised by a change in the beta-globin gene, resulting in the production of abnormal haemoglobin. The haemoglobin change is caused by the substitution of glutamic acid by valine at position 6 of the polypeptide chain. Glutamic acid is a Int. J. Environ. Res. Public Health 2019, 16, 1638; doi:10.3390/ijerph16091638 www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2019, 16, 1638 2 of 8 polar amino acid, unlike valine; this change in electrical charge gives the molecule the physicochemical diﬀerence responsible for the eructation of red blood cells [3]. Sickle cell anaemia is characterised by haemolysis, chronic and acute inﬂammation, vaso-occlusive complications, multiple organ damage, and reduced patient survival. The pathophysiology of haemoglobin S (HbS) is complex and inﬂuenced by hypoxia, acidosis, and cell dehydration, which result in the polymerization of HbS, leading to erythrocyte deformity. Perturbations in blood ﬂow may increase oxidative stress, causing vaso-occlusive episodes and clinical manifestations such as acute chest syndrome, stroke, priapism, and leg ulcer, among others [4]. Sickle cell trait is the heterozygous form of the HbS gene. It is a condition in which the individual, although not presenting anaemia in routine examination, presents about 40% of haemoglobin S content within erythrocytes. Patients with sickle cell trait are generally asymptomatic, and their quality of life is similar to that of the unaﬀected population [5]. The screening for anaemia and sickle cell trait in newborns is performed through the neonatal screening that was integrated into the Uniﬁed Health System (SUS) in 1992 in Brazil; while mandatory testing was required for all live births, only some states participated. In 2001, GM/MS Ordinance Nº. 822 of Ministry of Health was published, giving rise to the National Neonatal Screening Program, which aims to cover all live births in the Brazilian territory. Since then, all states in the country have become part of the program, which allowed the timely diagnosis of asymptomatic diseases in the neonatal period for early intervention with prophylactic measures and speciﬁc treatments to reduce or even prevent physical and mental sequelae [6–8]. The treatment of sickle cell anaemia is non-speciﬁc; thus, patient survival and quality of life depend on general prevention and treatment measures. The methodologies employed in the screening for sickle cell anaemia serve to identify other types of hemoglobinopathies as well as sickle cell carriers [9,10]. The method used to identify hemoglobinopathies is high-performance liquid chromatography (HPLC). Considering the importance of this disease in the State of Maranhão, this study aimed to evaluate the neonatal screening program coverage and to characterise the traits and sickle cell prevalence in newborns from the Brazilian Amazon-Savanna Transition Region between 2013 and 2015. 2. Materials and Methods This cross-sectional study was carried out in 283,003 newborns in Maranhão who underwent neonatal screening tests between 2013 and 2015. This is a prevalence study, and the data were analysed on a single occasion. Maranhão has 217 municipalities divided into 19 regional health centres. According to data from the Brazilian Institute of Geography and Statistics (IBGE, 2010), the population of Maranhão is 6,954,036. The study was based on documentary data sources, in which the results of all neonatal screening tests with FS and FAS standards (homozygosity and heterozygosity for HbS, respectively) were analyzed in the State of Maranhão during the study period. The data analyzed in this article refer to the database of the Neonatal Screening Reference Service of Maranhão, located in the Association of Parents and Friends of the Exceptional (APAE) in São Luís, Maranhão. The Neonatal Screening Reference Service also identiﬁes other variants of hemoglobin and thalassemias, which were not included because they did not ﬁt the objective of this study. The laboratory tests were performed using high-performance liquid chromatography (HPLC). The initial stage was based on the survey of the number of live births in Maranhão in 2013–2015 from the Information System of Live Births (SINASC) database of the Ministry of Health. The data analyses were based on descriptive statistics, using the calculation of the prevalence coeﬃcient. The prevalence coeﬃcient was calculated by dividing the number of cases by the total number of neonates screened by the National Neonatal Screening Program. The following formula was used to calculate the coverage: (number of children tested/number of children born alive) × 100. Chi-square tests were used to analyse categorical variables, and the Hardy-Weinberg theorem was applied to analyse the frequency of expected and observed homozygotes for the sickle hemoglobin Int. J. Environ. Res. Public Health 2019, 16, 1638 3 of 8 (HbSS). We used an exploratory analysis of spatial data to analyse the spatiotemporal dynamics of sickle cell anaemia cases in Maranhão in 2013, 2014, and 2015 using ArcGIS 10.2.2 (esri, Redlands, CA, USA). This research was submitted and approved by the Ethics Committee in Research—CEP of the University CEUMA, under opinion number 2,319,707, in compliance with the requirements of Resolution 466/12 of the National Health Council. 3. Results Between 1 January 2013, and 31 December 2015, 349,176 children were born in the state of Maranhão, 283,003 of which participated in the Neonatal Screening Program. Among the newborns submitted to the test, 162 presented sickle cell anaemia (HbSS) and, in relation to the sickle cell trait, 10,794 had HbAS. The overall test coverage was 81%, ranging from 82.6% in 2013 to 80.3% in 2015 (Table 1). Table 1. Neonatal screening coverage and the total number of sickle cell anaemia and sickle-trace cases in newborns in the State of Maranhão, 2013–2015. 9 2013 2014 2015 Total LB * (N ◦ ) ◦ Screening (N ) Coverage (%) ◦ N HbSS * ◦ N HbAS * 115,332 117,181 116,663 349,176 95,329 93,956 93,718 283,003 82.6% 80.2% 80.3% 81.0% 49 67 46 162 3811 3683 3300 10,794 * LB: Live births; HbSS: sickle cell anaemia; HbAS: sickle-trace. The distributions of sickle cell trait (10,794) and sickle cell anaemia (162) identiﬁed in the neonatal screening per year of research (2013–2015) were: 3811 (35.3%) and 49 (30.3%), 3,683 (34.1%) and 67 (41.3%), and 3300 (30.6%) and 46 (28.4%), respectively. During the study period, there was variability in the numbers of cases of sickle cell anaemia, with the lowest percentage occurring in 2015. The incidence of sickle cell anaemia among newborns in the State of Maranhão, as determined from samples from children submitted to neonatal screening program from 2013 to 2015, was 57:100,000, while the incidence of the sickle trait was 3814:100,000 (Table 2). Regarding the distribution of the HbSS phenotype among newborns, girls represented 48.8% of the cases, while boys represented 51.2% of the cases diagnosed with sickle cell anaemia (Table 3). Table 2. Prevalence of haemoglobin S in newborns in the state of Maranhão, 2013–2015. Phenotype Prevalence (Live Newborns) HbAS HbSS 3814: 100,000 (1: 26) 57: 100,000 (1: 1,754) ◦ N 10,794 162 % 3.8 0.05 HbSS: sickle cell anaemia; HbAS: sickle-trace. Table 3. Distribution of sickle cell anaemia by sex of newborns in the state of Maranhão, 2013–2015. HbSS Female Male ◦ N 79 83 % 48.8 51.2 HbSS: sickle cell anaemia. Assessment of the geographical distribution of sickle cell anaemia in Maranhão was evaluated according to the 19 regional health centres in the state and revealed a greater distribution in the capital and North regions. Thirty-two cases were diagnosed in the São Luís region, 15 cases were diagnosed in the Itapecuru-Mirim region, and 15 cases were diagnosed in the Pinheiro region. Int. J. Environ. Res. Public Health 2019, 16, 1638 4 of 8 We employed an exploratory analysis to evaluate the spatiotemporal dynamics of sickle cell anaemia cases in the State of Maranhão in 2013–2015. In 2013, the state capital (São Luís) had 10 cases, followed by the municipality of Itapecuru-Mirim with four cases (Figure 1a). In 2014, the municipality of São Luís presented 12 cases, while the municipalities of Codó, Presidente Dutra, São José de Ribamar, Timon, and Turiaçu presented three cases each (Figure 1b). Finally, in 2015, the municipalities of Caxias and São Luís presented three new cases each (Figure 1c). Figure 1. Spatiotemporal distribution of sickle cell anaemia cases in newborns in Maranhão in 2013 (a), 2014 (b), and 2015 (c). The observed distributions of sickle cell anaemia expected-genotypes were compared to the expected distributions according to the Hardy-Weinberg theorem. The statistically signiﬁcant result of χ2 adherence tests (χ2 = 26.57 p < 0.0001) indicated that the distribution of these expected-genotypes did not occur according to Hardy-Weinberg equilibrium. The frequency of the sickle-cell allele (βS) in this sample was greater than 1% (1.96%), indicating that the genetic polymorphism seems to be being Int. J. Environ. Res. Public Health 2019, 16, 1638 5 of 8 maintained in the population. The high χ2 value (χ2 = 26.57 p < 0.0001) was due to the deviations determined by the class of expected (βS/βS) homozygotes (χ2 = 25.54), with 53 more cases observed than the 109 cases expected according to Hardy-Weinberg equilibrium. In the two other expected-genotype classes, the observed (272,047 β/β and 10,794 β/βS) and expected (271,994.2 β/β and 10,899.6 β/βS) cases did not present signiﬁcant deviations (Table 4). Table 4. Distributions of sickle cell anaemia expected-genotypes in newborns identiﬁed in the databases and the numbers and percentage of cases according Chi-square tests. Expected-Genotype βS/βS β/βS β/β Total (N) newborns 162 10,794 272,047 283,003 % 0,06 3.81 96.13 100 χ2 (p) 26.57 (<0.0001) Alleles β = 0.9804 βS = 0.0196 4. Discussion Maranhão is in Phase IV of the National Neonatal Screening Program, which is responsible for screening for phenylketonuria, congenital hypothyroidism, hemoglobinopathies, and cystic ﬁbrosis. The São Luís’s APAE is responsible for all State Neonatal Screening coverage and oﬀers the complementary service of the Maranhão Hemocenter (HEMOMAR). All 217 municipalities are contracted with PNTN and have 474 collection points, with an average of 2.18 posts per municipality [2]. The results of the present study indicated that the coverage rate was highest in 2013, in which 82.6% of newborns were screened in Maranhão. Another survey carried out in Maranhão by Lopes et al. [11], which screened 99,498 children, showed a PNTN coverage rate of 81.57% in 2008. Another PNTN coverage study in the state of Amapá observed a rate of 31.2%. The deﬁciency in coverage of Amapá occurred because only 16 municipalities in the state are contracted to the Institute of Hematology and Hemotherapy responsible for the execution of screening throughout the state [12]. In contrast, the coverage rate was 100% of live births in the state of Paraná [13]. During the study period in Paraná, 548,810 newborns underwent the neonatal screening program, and there were 482,094 live births. The total coverage of the state of Paraná likely occurred due to the registration of all hospitals and maternities and the possible inclusion of some border regions of neighbouring countries. The PNTN coverage rates vary in diﬀerent regions of the country. Full screening coverage of the population is hampered by socioeconomic and cultural problems, lack of knowledge about the importance of the test, and obstacles related to the movement of the test units. In addition, the coverage of live births may be underestimated since examinations carried out in private networks are not counted. Increased coverage rates should be prioritised, either by the inclusion of the tests performed in the private network or with public actions to meet this need. The early detection of PNTN-treated disorders provides a better prognosis for the disease carrier [14,15]. The appearance of the HbS gene is likely linked to African slavery in Maranhão since the second half of the 17th century [16]. The high prevalence of the disease in the state can be explained by the regular entry of blacks into the states of Grão-Pará and Maranhão, assuring the monopoly of the slave trade, as well as the sale of these slaves to the residents who utilised their labour in the countryside and in the city [17]. Homozygous and heterozygous βS gene are distributed heterogeneously nationwide and are frequently observed in populations with higher proportions of black ancestors. Despite the predominance of these genes in black and mulatto populations, other population studies have demonstrated the increasing presence of βS in Caucasian populations. As a consequence, the North and Northeast Regions, which had the greatest inﬂuence of the black race, show higher frequencies of hemoglobinopathies [18]. Int. J. Environ. Res. Public Health 2019, 16, 1638 6 of 8 The prevalence of sickle cell anaemia in this study indicated that one in every 1754 live births had the βS/βS homozygous form, supposedly. This high frequency is possibly explained by the large proportion of individuals of African descent in the state of Maranhão [19]. However, one of the limitations of the present study was the impossibility of analysing the percentages of participants with anaemia and sickle cell trait by colour or race due to the lack of this information in the databases. Although sickle cell anaemia is a non-sex-linked genetic disease, there was a higher prevalence of expected βS/βS in male newborns. The present study corroborated data reported in studies carried out at Campana Laboratory and the Assis Chateaubriand Maternity-School in Fortaleza, which analysed blood samples from 1303 and 389 newborns, respectively, and also observed a higher prevalence in male newborns [20,21]. In Maranhão, the frequency of sickle cell trait, as determined by the results from samples submitted to the neonatal screening program, was 3.81% (1:26) in 2013–2015. In Salvador, a survey of 590 newborns reported a 9.8% (1:10) frequency of sickle cell trait, the highest percentage of sickle cell trait in Brazil [15]. A study carried out in Mato Grosso do Sul from 2000 to 2005 of 190,809 newborns submitted to the PNTN reported an average annual HbAS frequency of 1.6%. In 2000, the frequency was 2.4%, the highest in relation to other years, probably due to the low coverage of the PNTN that year [22]. The Neonatal Screening Program in Rio de Janeiro assessed 99,260 samples from newborns born between August 2000 and November 2001. The prevalence of heterozygotes was 4.7% (1:27) [23]. In a study carried out in Santa Catarina, Ellen et al. reported a sickle cell trait frequency of 0.8% among 730,412 children [24]. Lopes et al. analysed neonatal hemoglobinopathy screening data in all municipalities of the State of Maranhão, reporting that 4.9% (1:25) of 99,498 newborns had heterozygous HbS [11]. The results of comparative analysis of the ﬁndings of the present study with those of this previous research suggested that the diﬀerence in results was related to the size of the studied populations. Lopes et al. analysed samples only from 2008, whereas the present study analysed samples from three consecutive years. Another factor that may have impacted the reduction in the number of cases is prevention actions, such as genetic counselling, directed at parents and individuals identiﬁed as having sickle cell trait [25]. Our ﬁndings in Maranhão are similar to those in other Northeast states including Bahia [15,26], Rio Grande do Norte [27], and Rio de Janeiro [28], and Minas Gerais [29] in the Southeast, emphasizing the correlation with the history of Brazilian colonisation and reﬂecting the inﬂux of immigrants of African descent which occurred mainly in these states. The increased distribution of cases of sickle cell anaemia among newborns observed in the Northeast Region and state capital can be explained by the ethnic composition of these populations and the union of people of the same ethnic groups, which may contribute to a signiﬁcant increase in the number of homozygous individuals in certain regions of the state. The frequency of expected homozygous individuals indicates that this locus is changing over time in the state of Maranhão. 5. Conclusions Increased biological and epidemiological knowledge of sickle cell anaemia provides new In addition, the mapping of sickle cell anaemia cases and knowledge therapeutic approaches. about its occurrence in Maranhão may contribute to existing social and public health actions and the creation of new epidemiological surveillance strategies to improve the quality of life of the aﬀected population. Author Contributions: R.C.S. contributed to the conceptualization, formal analysis, investigation, methodology, validation, display, writing of the original draft preparation, and writing-review & editing. P.A.D.M.N. contributed to the formal analysis, validation, display, and writing-review & editing. J.R.N.S. contributed to the formal analysis, methodology, validation, display, and writing-review & editing. S.G.M. contributed to the data curation, formal analysis, investigation, validation, display, and writing-review & editing. M.C.G. contributed to the formal analysis, investigation, validation, display, writing of the original draft preparation, and writing-review & editing. F.B.S. contributed to the contributed to the formal analysis, investigation, methodology, validation, display, and writing-review & editing. R.A.H. contributed to the data curation, formal analysis, validation, display, Int. J. Environ. Res. Public Health 2019, 16, 1638 7 of 8 and writing-review & editing. J.R.A.S. contributed to the data curation, formal analysis, funding acquisition, investigation, project administration, supervision, validation, display, writing of the original draft preparation, and writing-review & editing. Funding: This research received no external funding. J.R.A.S. is a research fellow of the FAPEMA (Grant BEPP-02494/18). Acknowledgments: We would like to thank the Universidade CEUMA (UNICEUMA- São Luis-MA, Brazil), and Fundação de Amparo a Pesquisa e Desenvolvimento do Estado do Maranhão—FAPEMA. Conﬂicts of Interest: The authors declare no competing interests. References 1. Castro, I.P.S.; Viana, M.B. Cognitive proﬁle of children with sickle cell anemia compared to healthy controls. J. Pediatr. 2018. [CrossRef] [PubMed] 2. Ministério da Saúde. Doença Falciforme: Condutas Básicas para Tratamento; Ministério da Saúde, Secretaria de 3. 4. Atenção à Saúde, Departamento de Atenção Especializada: Brasília, DF, Brazil, 2012. Da Silva, P.H.; Alves, H.B.; Henneberg, R.; Merlin, J.C.; Stinghen, S.T.; Comar, S.C. Hematologia Laboratorial: Teoria e Procedimentos, 1st ed.; Artmed Editora S.A.: Porto Alegre, Brazil, 2016. Aleluia, M.M.; da Guarda, C.C.; Santiago, R.P.; Fonseca, T.C.; Neves, F.I.; de Souza, R.Q.; Farias, L.A.; Pimenta, F.A.; Fiuza, L.M.; Pitanga, T.N. Association of classical markers and establishment of the dyslipidemic sub-phenotype of sickle cell anemia. Lipids Health Dis. 2017, 16, 74. [PubMed] 5. Murão, M.; Ferraz, M.H.C. Sickle cell trait: Heterozygous for the hemoglobin S. Rev. Bras. Hematol. Hemoter. 6. 2007, 29, 223–225. Souza, I.M.; Araújo, E.M. Doença Falciforme e triagem neonatal: Um debate necessário. RSC UEFS 2015, 5, 51–58. [CrossRef] 7. Menezes, F.L.; Gracioli, M.D.S.A.; de Freitas, H.M.B.; Diaz, C.M.G.; da Rocha, B.D.; Gomes, I.E.M.; Bordignon, J.S. Conhecimento das mães acerca do teste do pezinho. Revista Espaço para a Saúde-REpS 2016, 17, 220–228. [CrossRef] 8. Ministério da Saúde. Doença Falciforme: Conhecer para Cuidar; Ministério da Saúde, Secretaria de Atenção à 9. 10. Saúde, Departamento de Atenção Hospitalar e de Urgência: Brasília, Brazil, 2015. Braga, J.A.P. General measures in the treatment of sickle cell disease. Rev. Bras. Hematol. Hemoter. 2007, 29, 233–238. Ferraz, M.H.C.; Murão, M. Laboratorial diagnosis of sickle cell disease in the neonate and after the sixth month of life. Rev. Bras. Hematol. Hemoter. 2007, 29, 218–222. 11. Lopes, T.D.C.; Sarmento, L.D.M.; Fróz, R.C.; Marinho, H.T.; Noronha, E.P.; Oliveira, R.A.G. Assessment of National Neonatal Screening Program for Hemoglobinopathies. Rev. Inst. Adolfo Lutz. 2011, 70, 417–421. 12. Lacerda, G.S.L.; Costa, F.S.; de Souza Dantas, D.; Costa, É.R.G.; Resque, R.L.; do Nascimento, A.A.; Gomes, M.R.F. Triagem neonatal: O panorama atual no estado do Amapá. Vigilância Sanitária em Debate: Sociedade, Ciência & Tecnologia 2017, 5, 89–96. 13. Watanabe, A.M.; Pianovski, M.A.D.; Zanis Neto, J.; Lichtvan, L.C.; Chautard-Freire-Maia, E.A.; Domingos, M.T.; Wittig, E.O. Prevalence of hemoglobin S in the State of Paraná, Brazil, based on neonatal screening. Cadernos de Saúde Pública 2008, 24, 993–1000. [CrossRef] [PubMed] 14. Ramalho, R.J.R.; Valido, D.P.; Aguiar-Oliveira, M.H. Avaliação do programa de triagem para o hipotireoidismo congênito no estado de Sergipe. Arq. Bras. Endocrinol. Metabol. 2000, 44, 157–161. [CrossRef] 15. Adorno, E.V.; Couto, F.D.; Moura Neto, J.P.D.; Menezes, J.F.; Rêgo, M.; Reis, M.G.D.; Gonçalves, M.S. Hemoglobinopathies in newborns from Salvador, Bahia, Northeast Brazil. Cadernos de Saúde Pública 2005, 21, 292–298. [CrossRef] 16. Meirelles, M.M. Os Negros do Maranhão; UFMA: São Luís, Brazil, 1983. 17. Dias, M.N. Fomento e Mercantilismo: A Companhia Geral do Grão-Pará e Maranhão (1755-1778); Universidade 18. 19. Federal do Pará: Belém, Brazil, 1970. Siqueira, F.A.M.; Fett-ConteII, A.C.; Borin, L.N.B.; Bonini-Domingos, C.R. Diagnosis of hemoglobinopathies in newborn babies in Hospital de Base, São José do Rio Preto, SP, Brazil. Rev. Bras. Hematol. Hemoter. 2002, 24, 302–305. [CrossRef] IBGE. Instituto Brasileiro de Geograﬁa e Estatística. Censo Demográﬁco do Maranhão 2010. Available online: http://ibge.gov.br (accessed on 22 May 2017). Int. J. Environ. Res. Public Health 2019, 16, 1638 8 of 8 20. Campana, P.G.C. Prevalência de hemoglobinopatias no Laboratório Campana. News Lab. 2001, 49, 100–110. 21. Pinheiro, L.S.; Gonçalves, R.P.; Tomé, C.A.; Alcântara, A.E.; Marques, A.R.; Silva, M.M.D. The prevalence of hemoglobin S in newborns from Fortaleza, Brazil: The importance of neonatal research. Rev. Bras. Ginecol. Obstet. 2006, 28, 122–125. 22. Holsbach, D.R.; Ivo, M.L.; Honer, M.R.; Rigo, L.; Botelho, C.A.D.O. Ocorrência de hemoglobina S no estado de Mato Grosso do Sul, Brasil. J. Bras. Patol. Med. Lab. 2008, 44, 277–282. [CrossRef] 23. Lobo, L.C.; Bueno, M.L.; Moura, P.; Ogeda, L.L.; Castilho, S.; Farias, C.S.M. Triagem neonatal para hemoglobinopatias no Rio de Janeiro, Brazil. Revista Panamericana de Salud Pública 2003, 13, 154–159. [CrossRef] 24. Eller, R.; da Silva, D.B. Evaluation of a neonatal screening program for sickle-cell disease. J. Pediatr. 2015, 92, 409–413. [CrossRef] 25. Giordano, P.C.; Harteveld, C.L.; Bakker, E. Genetic Epidemiology and Preventive Healthcare in Multiethnic Societies: The Hemoglobinopathies. Int. J. Environ. Res. Public Health 2014, 11, 6136–6146. [CrossRef] [PubMed] Silva, W.S.; Oliveira, R.F.; Ribeiro, S.B.; Silva, I.B.; Araújo, E.M.; Baptista, A.F. Screening for Structural Hemoglobin Variants in Bahia, Brazil. Int. J. Environ. Res. Public Health 2016, 13, 225. [CrossRef] [PubMed] 27. Bezerra, T.M.; Andrade, S.R. Investigação sobre a prevalência de hemoglobinas anormais entre doadores de 26. 28. sangue. Rev. Bras. Anal. Clin. 1991, 23, 117–118. Silva-Filho, I.; Gonçalves, M.S.; Adôrno, E.V.; Campos, D.P.; Fleury, M.K. Screening of abnormal haemoglobin and the evaluation of oxidative degeneration of haemoglobin among workers with the sickle cell traits (HbAS), exposed to occupational hazards. Rev. Bras. Hematol. Hemoter. 2005, 27, 183–187. 29. Melo, S.M.A.; Arantes, S.C.F.; Botelho-Filho, A.; Rocha, A.F.S. Prevalência de hemoglobinopatias em doadores de sangue do hemocentro regional de Uberlândia-MG. Rev. Bras. Hematol. Hemoter. 2000, 22. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.2196_44990	JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al Original Paper Examining Twitter-Derived Negative Racial Sentiment as Indicators of Cultural Racism: Observational Associations With Preterm Birth and Low Birth Weight Among a Multiracial Sample of Mothers, 2011-2021 Thu T Nguyen1, MSPH, ScD; Junaid S Merchant1, BA, MS; Shaniece Criss2, MPH, MPA, ScD; Katrina Makres1, BS; Krishik N Gowda1, BS, MSc; Heran Mane1, BS; Xiaohe Yue1, MS; Yulin Hswen3, ScD; M Maria Glymour4, MS, ScD; Quynh C Nguyen1, MSPH, PhD; Amani M Allen5, MPH, PhD 1Department of Epidemiology & Biostatistics, University of Maryland School of Public Health, College Park, MD, United States 2Department of Health Sciences, Furman University, Greenville, SC, United States 3Department of Epidemiology and Biostatistics, Bakar Computational Health Sciences Institute, University of California, San Francisco, San Francisco, CA, United States 4Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA, United States 5Divisions of Community Health Sciences and Epidemiology, University of California, Berkeley, Berkeley, CA, United States *these authors contributed equally Corresponding Author: Thu T Nguyen, MSPH, ScD Department of Epidemiology & Biostatistics University of Maryland School of Public Health 4254 Stadium Dr. College Park, MD, 20742 United States Phone: 1 301 405 1484 Email: [email protected] Abstract Background: Large racial and ethnic disparities in adverse birth outcomes persist. Increasing evidence points to the potential role of racism in creating and perpetuating these disparities. Valid measures of area-level racial attitudes and bias remain elusive, but capture an important and underexplored form of racism that may help explain these disparities. Cultural values and attitudes expressed through social media reflect and shape public norms and subsequent behaviors. Few studies have quantified attitudes toward different racial groups using social media with the aim of examining associations with birth outcomes. Objective: We used Twitter data to measure state-level racial sentiments and investigate associations with preterm birth (PTB) and low birth weight (LBW) in a multiracial or ethnic sample of mothers in the United States. Methods: A random 1% sample of publicly available tweets from January 1, 2011, to December 31, 2021, was collected using Twitter’s Academic Application Programming Interface (N=56,400,097). Analyses were on English-language tweets from the United States that used one or more race-related keywords. We assessed the sentiment of each tweet using support vector machine, a supervised machine learning model. We used 5-fold cross-validation to assess model performance and achieved high accuracy for negative sentiment classification (91%) and a high F1 score (84%). For each year, the state-level racial sentiment was merged with birth data during that year (~3 million births per year). We estimated incidence ratios for LBW and PTB using log binomial regression models, among all mothers, Black mothers, racially minoritized mothers (Asian, Black, or Latina mothers), and White mothers. Models were controlled for individual-level maternal characteristics and state-level demographics. Results: Mothers living in states in the highest tertile of negative racial sentiment for tweets referencing racial and ethnic minoritized groups had an 8% higher (95% CI 3%-13%) incidence of LBW and 5% higher (95% CI 0%-11%) incidence of PTB compared to mothers living in the lowest tertile. Negative racial sentiment referencing racially minoritized groups was associated with adverse birth outcomes in the total population, among minoritized mothers, and White mothers. Black mothers living in states in the highest tertile of negative Black sentiment had 6% (95% CI 1%-11%) and 7% (95% CI 2%-13%) higher incidence https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al of LBW and PTB, respectively, compared to mothers living in the lowest tertile. Negative Latinx sentiment was associated with a 6% (95% CI 1%-11%) and 3% (95% CI 0%-6%) higher incidence of LBW and PTB among Latina mothers, respectively. Conclusions: Twitter-derived negative state-level racial sentiment toward racially minoritized groups was associated with a higher risk of adverse birth outcomes among the total population and racially minoritized groups. Policies and supports establishing an inclusive environment accepting of all races and cultures may decrease the overall risk of adverse birth outcomes and reduce racial birth outcome disparities. (J Med Internet Res 2023;25:e44990) doi: 10.2196/44990 KEYWORDS birth outcomes; health disparities; machine learning, racial sentiment; social media Introduction Low birth weight (LBW) and preterm birth (PTB) are widely used indicators of reproductive health [1,2] and are associated with an increased risk of infant mortality [3], developmental delays [4], and cardiometabolic disorders in adulthood [5,6]. Large racial disparities in these birth outcomes persist in the United States [7,8]. Black mothers in particular have substantially higher rates of LBW and PTB, infant mortality, and maternal morbidity compared to White mothers. For example, pregnancy-related mortality is over 3 times higher among Black compared to White women, and LBW rates in 2020 were 14.2% for Black mothers, but 6.8% for White mothers [7]. These disparities cannot be fully accounted for by sociodemographic and individual-level factors [7-9], and there is increasing evidence for the role of racism in creating and perpetuating race-based disparities in birth outcomes [10,11]. interpersonal), Racism is a well-established social determinant of health that operates at and across multiple levels. This includes the internalized, personally mediated (also referred to as individual or institutional, structural, and cultural dimensions of racism [12]. Internalized racism refers to the acceptance of negative beliefs about their own race by individuals of racially stigmatized groups [13]. Personally mediated racism refers to racial prejudice (attitudes) and stereotypes (beliefs and assumptions) according to race and discrimination (differential treatment based on race) enacted between individuals [14]. Institutional racism refers to laws, policies, and practices of particular institutions in providing advantages and disadvantages, differentially, according to race [15]. Structural racism involves the coordination and interaction of multiple institutions and systems, such as those involved in housing, prison, banking, and education, to provide differential access and resources according to racial group identity [16]. Research has found experiences of racism to be associated with a wide variety of health outcomes [17,18]. However, most of this research has examined how racism operates at the individual level, whereas work investigating how racism operates at other levels has been relatively limited. Exposure to racism is also a psychosocial stressor that, when experienced chronically, has demonstrated negative health effects that contribute to birth outcome disparities [11,19]. Maternal stress can alter neuroendocrine function, impact immune and inflammatory responses, and affect the vascular system. Racism stress, in particular, is associated with increases the systemic corticotropin-releasing hormone, which in https://www.jmir.org/2023/1/e44990 XSL•FO RenderX stimulates the release of prostaglandins from the placenta and facilitates oxytocin’s role in initiating contractions earlier in pregnancy [20]. Stress can prompt vasoconstriction, increasing blood pressure and reducing blood flow to the fetus [21]. Stress also impacts the immune system, hindering its ability to fight infections that linger in the body longer, and is associated with an increased likelihood of preterm labor [21]. Racism-related stress can also lead to maladaptive coping behaviors, such as smoking and alcohol use. Additionally, women who are born with LBW are more likely to give birth to children with LBWs [19], perpetuating the impacts of racism intergenerationally. Racism may also influence birth outcomes by limiting access to resources and opportunities, such as education, employment, health care, and housing [16,17]. For example, historical practices like redlining and discriminatory home loan lending led to racial residential segregation and disinvestment in communities of color, which continue to have measurable effects on health disparities today, including racial birth outcome disparities [22]. An emerging body of research has revealed the foundational role of cultural racism in perpetuating race-based disparities. Cultural racism is the infusion of the ideology of racial hierarchy into the values, language, imagery, symbols, and unstated assumptions of the larger society [17]. It is displayed through the media, stereotyping, and norms within society and its institutions [23]. Importantly, cultural racism produces a context that supports the other dimensions of racism to maintain and reinforce health inequities [17]. Cultural racism, or the accepted norms, values, and ideologies of a racialized society, becomes realized in policies and practices within and across institutions [24]. In this way, cultural and structural racism are mutually reinforcing. One way of tapping into cultural racism is to assess community members’ attitudes toward other racial groups [25,26]. A promising line of work has used such approaches to assess the impact of area-level measures of racial prejudice (also referred to as sentiment or animus) on health. A 2022 systematic review of the literature on area-level prejudice and health revealed that all studies found an association between higher area-level prejudice and worse health outcomes among racially minoritized groups, and 4 studies even showed the negative impact of prejudice among White samples [27]. However, research on birth outcomes is limited as only 4 of the studies in the aforementioned review investigated birth outcomes [27]. Moreover, measuring racial sentiment using traditional survey approaches can be costly, time-consuming, and subject to self-report biases [28-30]. Alternatively, social media data J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al provide unique opportunities for assessing population-level racial sentiment, which can be used as an indicator of cultural racism. Twitter represents a rapid mode of communication, with millions of tweets sent daily by users across the globe, 80%-90% of whom have publicly accessible Twitter profile [31]. Cultural values and attitudes expressed on the web through social media both reflect and shape public norms, beliefs, and subsequent behaviors [32,33]. Few studies have attempted to quantify attitudes toward different racial groups using Twitter data with the aim of examining how it relates to birth outcomes. Using state-level data from 2015 to 2017, our previous study demonstrated that state-level racial sentiment was associated with implicit and explicit racial bias [10], and exhibited associations with birth outcomes of different groups of minority women [34]. This study aims to extend the findings of previous work by (1) examining a wider time frame (2011-2021) to assess whether the associations between Twitter-derived racial sentiment and birth outcomes are persistent across the years, and (2) through a closer examination of the distribution of attitudes toward different racial groups at the state- and county-level to see how they related to birth outcomes within and across racial or ethnic groups. Methods A random sample comprising 1% of publicly available tweets from January 1, 2011 to December 31, 2021 was collected using Twitter’s Academic Application Programming Interface. We restricted our analyses to tweets that were in English, from the United States, and used one or more race-related keywords (Multimedia Appendix 1, Table S1). These race-related keywords were constructed based racial and ethnic terms from the US Census, previous studies examining race-related web-based conversations [35], and a web-based database of racial slurs [36]. We included tweets with a unique tweet id. We dropped duplicate tweets according to their “tweet_id.” Retweets and quoted tweets are included in this data. Our analytic sample included 56,400,097 tweets from 3,699,646 unique users. Sentiment Analysis We assessed the sentiment of each tweet using a support vector machine (SVM), a supervised machine learning model. A full description of our model has been previously described by Nguyen and colleagues [34]. Our training data was comprised of manually labeled tweets obtained from Sentiment140 (n=498) [37], Kaggle (n=7086) [38], Sanders (n=5113) [39], as well as 6481 tweets labeled by our research group. Sentiment140, Sanders, and Kaggle datasets are all publicly available training datasets specifically labeled for sentiment analysis. We used 5-fold cross-validation to assess model performance and achieved a high level of accuracy for the negative sentiment classification (91%) and a high F1 score (84%). Accuracy is measured as the number of posts with the correct prediction divided by the total number of tweets in the testing data set. The F1 score is a measure that balances precision (positive predicted value and recall [sensitivity]); a high F1 score suggests a model is robust in predicting posts that are labeled as 1. The https://www.jmir.org/2023/1/e44990 XSL•FO RenderX trained SVM model was used to analyze our Twitter data set for negative sentiment classification. Please see the code for the data collection and sentiment analysis model in Multimedia Appendix 1. To assess historical trends in sentiment, average negative sentiment scores for each racial and ethnic group along with the sentiment scores referencing racially minoritized groups were plotted as a function of time. The line chart was plotted using the “Matplotlib” library in Python [40]. We visualized the average negative sentiment scores with geographic plots using the Tableau software. The sentiment scores were obtained from the SVM model and grouped using the state Federal Information Processing Standards (FIPS) codes. Individual-Level Birth Outcomes Data Individual birth outcomes data through 2011-2021 were obtained from restricted US natality files that included geographic identifiers. The latest year the natality files are currently available is 2021. We chose 2011 to examine trends in the last decade. In addition, Twitter introduced Twitter “Places” for geotagging tweets in June 2010 [41]. The natality files include all births in the United States during this time period. The data come from birth certificates filed in each state. The analysis was restricted to singleton births with no congenital abnormalities. These exclusion criteria helped ensure that associations with our birth outcomes were not due to congenital abnormalities [42] or twins, triplets, and other higher order multiple births, which are known to increase the risk for LBW and PTB [43]. The primary outcomes were LBW (defined as birth weight <2499 g) and PTB (defined as gestational age <37 weeks). Covariates We adjusted for potential confounders at the individual- and area-level when assessing the association between racial sentiment and birth outcomes. Individual-level covariates in our models included birth year, maternal age (linear spline with knots at 19, 25, 29, 33, and 38 years), race (non-Hispanic White, non-Hispanic Black, non-Hispanic Asian), Hispanic ethnicity, marital status (married or unmarried), education (less than high school, high school or General Education Development, some college, bachelor’s degree, master’s degree, or doctorate), and first birth (yes or no), and prenatal care initiation during the first trimester (yes or no). We also adjusted for state-level characteristics, including proportions of non-Hispanic Black and Latinx individuals, population density (per square mile), and economic disadvantage (standardized factor score) [44,45] summarizing the following variables (%): unemployed; some college education, high school diploma, children in poverty, single parent household, and median household income) to account for state-level compositional differences in demographic and economic characteristics. Use of the factor score has been previously established [46]. State-level covariates were derived from 2011 to 2021 from the American Community Survey [47]. Statistical Analyses For each year, the state-level racial sentiment was merged with birth data during that year. The racial sentiment was coded in tertiles for analysis assessing associations with birth outcomes. J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al evaluated characteristics. We We estimated incidence ratios for LBW and PTB (separately) using log binomial regression models with state-level racial sentiment as our independent variable of interest, and controlled for individual-level maternal characteristics and state-level demographic statistical significance at P<.05. Our primary models examined the association between state-level sentiment toward all racially minoritized groups on birth outcomes among the full sample, racially minoritized mothers, and White mothers. Follow-up analyses examined associations between state-level sentiment toward individual racial groups (Asian, Black, Latinx, and White) and birth outcomes for those groups specifically. Sensitivity analyses were conducted to examine the association between county-level racial sentiment and birth outcomes. Although all tweets collected had place information that permitted the state, only 60% (n=33,840,058) of tweets collected had place information to identify the county. Thus, we consider the county-level analyses to be supplementary. Sensitivity analyses also excluded tweets from users who tweeted more than 1000 times per year, which represented less than 1% (n=1066) of all tweets. Stata MP 16 [48] was used for statistical analysis. identification of the Ethical Considerations This study was determined exempt by the University of Maryland College Park Institutional Review Board (1797788-1). Results Among the terms assessed, the top 20 terms by year were present in 86% (n=48,374,805) of all tweets concerning a racial or ethnic minority group (Multimedia Appendix 1, Table S2). The proportion of tweets referencing Black people, Latinx, and White people that were negative increased over time, peaking in 2019. For tweets referencing White people and Latinx, we saw a plateauing or declining trend in 2020-2021. The proportion of tweets referencing Asians that were negative climbed until 2020 and declined slightly in 2021 (Figure 1). Black people experienced the highest negative sentiment of all groups. After a brief period of decline in 2020, the proportion of tweets referencing Black people that were negative began to rise again. Higher proportions of negative tweets referencing Black Americans were found in the Southern and Northeastern US regions. Geographic distribution of tweets referencing racially minoritized groups, Latinx, and Asians are presented in Multimedia Appendix 1 (Figures S1-3). Approximately 55% of births were to White mothers; 22% of the samples were Latinx. Black and Asian mothers made up 15% (n=5,152,595) and 6% (n=2,116,785) of the sample, respectively. The majority of women were US born (78%, n=27,630,726), and 31% (n=11,088,533) completed college or beyond. The state-level negative racial sentiment was consistently associated with adverse birth outcomes controlling for individual maternal characteristics and state-level sociodemographic characteristics (Tables 1 and 2). For the total population, mothers living in states in the highest (third) tertile of negative racial sentiment for tweets referencing racial and ethnic minoritized groups have an 8% higher (95% CI 3%-13%) incidence of LBW and a 5% higher (95% CI 0%-11%) incidence of PTB compared to mothers living in the lowest (first) tertile of negative racial sentiment. Moreover, associations were found for mothers living in states in the second vs lowest tertile of negative racial sentiment. Negative racial sentiment referencing racially minoritized groups was associated with birth outcomes in the total population, among minoritized mothers and White mothers. Associations are slightly higher for racially minoritized mothers in all but one of the models where racially minoritized and White mothers show similar associations with racial sentiment (Figure 2). When examining race- and ethnic-specific sentiment and birth outcomes, the strongest and most consistent associations were found for negative sentiment in tweets referencing Black people and birth outcomes among Black mothers. Black mothers living in states in the third tertile of negative Black sentiment had a 6% higher incidence of LBW and a 7% higher incidence of PTB compared to mothers living in the lowest tertile states. Negative sentiment in tweets referencing Latinx individuals was associated with a 6% and 3% higher incidence of LBW and PTB, respectively, comparing Latinx mothers living in the highest compared to the lowest tertile states. Negative sentiments referencing White individuals was not associated with birth outcomes among White mothers (Table 3). Multimedia Appendix 1 tables present these associations by year (Tables S3-6). Associations between state-level negative racial sentiment for tweets referencing minoritized groups tended to be more strongly associated with birth outcomes among minoritized mothers compared to the total population of birthing people or of White mothers. There were small fluctuations in estimates across the years. The overall negative state-level racial sentiment was associated with adverse birth outcomes for most years. Negative sentiment referencing Black Americans was consistently associated with an elevated risk of LBW and PTB among Black mothers. Negative sentiment in tweets referencing Latinx and Asian individuals was associated with adverse birth outcomes among Latinx and Asian mothers for selected years. Sensitivity analyses excluded tweets from users who tweeted more than 1000 per year. The estimates changed slightly, but all main conclusions remain the same (Multimedia Appendix 1, Tables S7 and S8). Sensitivity analyses examining the associations between county-level racial sentiment and birth outcomes (Multimedia Appendix 1, Tables S9-14) found attenuated associations compared to state-level analyses. The restricted natality files only included state- or county-geographic identifiers, so more granular analyses were not possible with that data. Compared to state-level analyses, small increases in risks of PTB and LBW among minoritized women were observed when comparing mothers living in the highest versus lowest tertile counties (see combined county-level in Multimedia Appendix 1, Tables S6-7). https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al Figure 1. Temporal trends in proportion of tweets that were negative by racial and ethnic group. Data source: Twitter data from 2011 to 2021. Table 1. Characteristics of mothers giving birth from 2011 to 2021 (N=35,267,177). Characteristic Age (years), mean (SD) Married, n (%) White, n (%) Black, n (%) Asian, n (%) Latinx, n (%) US born, n (%) Education, n (%) Less than high school High school Some college College Master’s or doctorate First birth, n (%) Prenatal care during first trimester, n (%) Birth outcomes, n (%) Low birth weight Preterm birth aData source: US natality files from 2011 to 2021. Valuea 28.50 (5.85) 21,068,841 (59.74) 19,236,693 (54.55) 5,152,595 (14.61) 2,116,785 (6.00) 7,932,703 (22.49) 27,630,726 (78.35) 5,018,290 (14.23) 8,999,298 (25.61) 10,161,059 (28.81) 7,014,807 (19.89) 4,073,726 (11.55) 11,318,052 (32.09) 26,832,305 (76.09) 2,215,586 (6.28) 2,770,462 (7.86) https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al Table 2. Associations using incidence rate ratios between state-level racial sentiment toward minoritized and birth outcomes for full sample, minoritized mothers, and White mothers, 2011-2021.a Data Low birth weight Total, incidence rate ratio (95% CI) Minoritized groups, incidence rate ratio (95% CI) White mothers, incidence rate ratio (95% CI) Second tertile vs first (lowest) 1.07 (1.02-1.13) 1.08 (1.02-1.15) Third tertile vs first (lowest) 1.08 (1.03-1.13) 1.09 (1.03-1.15) Sample, n Preterm 35,267,177 17,190,739 Second tertile vs first (lowest) 1.06 (1.01-1.11) 1.07 (1.02-1.12) Third tertile vs first (lowest) 1.05 (1.00-1.11) 1.06 (1.01-1.11) Sample, n 35,288,014 17,200,907 aData source: US natality files and Twitter data from 2011 to 2021. 1.07 (1.01-1.13) 1.08 (1.03-1.14) 19,236,695 1.05 (1.0-1.11) 1.06 (0.99-1.13) 19,247,910 Figure 2. Geographic distribution of averaged negative sentiment of tweets referencing Black people. Data source: Twitter data from 2011 to 2021. https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al Table 3. Race and ethnic associations using incidence rate ratios between state-level negative sentiment and birth outcomes, 2011-2021.a Data Negative Black sentiment, birth outcomes among Black mothers, incidence rate ratio (95% CI) Negative Asian sentiment, birth outcomes among Asian mothers, incidence rate ratio (95% CI) Negative Latinx sentiment, birth outcomes among Black mothers, incidence rate ratio (95% CI) Negative White sentiment, birth outcomes among White mothers, incidence rate ratio (95% CI) Low birth weight Second tertile vs first (lowest) Third tertile vs first (low- est) 1.08 (1.03-1.13) 0.99 (0.96-1.03) 1.00 (0.97-1.02) 1.00 (0.97-1.04) 1.06 (1.01-1.11) 1.01 (0.96-1.05) 1.06 (1.01-1.11) 1.00 (0.97-1.04) Sample, n 5,152,595 2,116,785 7,932,703 19,236,695 Preterm Second tertile vs first (lowest) third tertile vs first (low- est) 1.08 (1.03-1.14) 1.01 (0.97-1.05) 0.99 (0.98-1.0) 1.01 (0.98-1.04) 1.07 (1.02-1.13) 1.03 (0.99-1.07) 1.03 (1.00-1.06) 0.99 (0.96-1.03) Sample, n 5,157,600 2,117,808 7,935,836 19,247,910 aData source: US natality files and Twitter data from 2011 to 2021. Discussion This study used a combination of social media data and comprehensive US birth records to assess how spatial patterns in attitudes toward different racial groups on Twitter related to birth outcomes. We calculated negative sentiment toward different racial groups at the state level and found that negative attitudes toward minoritized groups were associated with increased risk for low birth-weight babies and PTB for the total population, and this remained generally consistent across each of the years we analyzed. Furthermore, we found that negative sentiment toward minoritized groups was comparably associated with birth outcomes for each racial group, including White women when stratifying outcomes by race. In contrast, negative sentiment toward White people was not associated with adverse birth outcomes for White mothers. In fully stratified models wherein exposures and outcomes were stratified by race, negative Black sentiment was associated with adverse birth outcomes for Black mothers. Results were less consistent when examining associations between Latinx and Asian sentiment and birth outcomes for Latinx and Asian mothers, respectively, and null for White sentiment on White mothers’ birth outcomes. Together, our findings suggest that area-level sentiment toward minoritized groups is associated with negative birth outcomes for the population as a whole. Furthermore, the greatest and most consistent disparities were observed for Black mothers. This study extends our understanding of racial health disparities. Racism is an organized system of racial hierarchy that structures risks, opportunities, and resources within a society [17]. Expressions of negativity toward minoritized groups demonstrating an ideology of racial inferiority in the values, language, imagery, symbols, and unstated assumptions of the larger society are particularly salient, working to uphold other forms of racism (eg, structural, institutional, personally mediated, and internalized). Our work suggests that negative sentiment of racial or ethnic minoritized groups explains birth outcomes among White mothers also. This aligns with research https://www.jmir.org/2023/1/e44990 XSL•FO RenderX showing that policies attempting to limit the rights of a minority group have negative ramifications for the majority group also. For instance, McGhee [49] outlines how the closing of public pools across America in the 1950s—policy decisions made to deny Black people access—resulted in the elimination of this public amenity for all people, including White people. Structurally racist policies disproportionately impact people of color but also have negative consequences for all, including health outcomes for White people [49-51]. For example, racial animus toward racially minoritized groups has led to opposition to social policies and programs, such as the expansion of the Affordable Care Act [52] a policy that would also benefit a large proportion of White people. Previous research on structural racism shows that racial segregation is associated with adverse birth outcomes among racial minorities and White people [53-55], exemplifying Camara P Jones’ (former President of the American Public Health Association) [56] statement that “racism saps the strength of the whole society.” Our findings add to the growing body of work that examines racism and health disparities for multiple racial or ethnic groups. the research has examined Although pregnancy-related health disparities for Black people, emerging research is beginning to tease apart commonalities and differences experienced by different racial groups. the majority of Our study also advances research aiming to understand how the social and cultural environment can shape health. Social media represent a fertile space for stigmatizing language, stereotype representations, slurs, and hostile speech. As more social interactions take place on internet, it is imperative to track and monitor this space and its potential impact on health and well-being, particularly for minoritized and stigmatized racial groups. Over the last decade, we have seen a trend of increasing negative racial sentiment referencing racial and ethnic minorities. We have previously found that racialized events have led to shifts in expressed racial sentiment. For example, the killing of George Floyd was followed by a temporary decline J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al in negative Black sentiment in 2020 [57], and negative Asian sentiment spiked during March 2020 with the emergence of the COVID-19 pandemic [58]. These events, including the resurgence of the Black Lives Matter movement in 2020, changed the sentiment and volume of discussions related to race and racism on the web [57]. These events were commonly characterized by a rapid change in racial sentiment, lasting a few weeks, followed by a return to near baseline levels [57-59]. In 2020, we saw associations between negative sentiment and adverse birth outcomes remaining comparable to other years. The current findings also extend our understanding of the geography of racial health disparities in the United States. Our analysis included area-level racial sentiment measured at the state level. States vary in social norms, laws, and policies. However, cultural racism can also vary at lower geographic levels. There may be a bidirectional influence of tweets and state-level policies such that policies may impact tweets and tweets may in turn influence the support of new policies. We conducted analyses at the county level but considered these sensitivity analyses because approximately 40% of the Twitter data had missing county location information. Furthermore, Twitter data for less populated counties were relatively sparse, which may lead to biased estimates of racial sentiment. The geographic identifiers in the US natality files only include state and county identifiers. A valuable future direction would be to examine associations at a more granular resolution for studies looking at cultural racism and health in other data. A major strength of this study comes from the use of social media–derived racial sentiment. Traditional approaches to measuring racial attitudes, such as survey measures or experimental approaches, are often time-consuming and expensive to conduct. These approaches provide a time-limited snapshot of sentiment, unlike the temporally rich data that can be obtained from social media posts. This temporally and geographically rich data thus makes it possible to look at changes in health disparities as a function of local health policies as well as race-related events, such as Black Lives Matter protests [57,58]. This study is not without limitations. For this paper, we used our trained SVM model for sentiment analysis, as we have used this model in previous publications [57,58], including our paper examining racial sentiment and birth outcomes for 2015-2017 [34]. Changing the sentiment analysis model may change the findings. We have found a high level of accuracy and a high F1 score with this model. To be most comparable to our other papers, we used this sentiment model here. The keyword list is not exhaustive, but we attempted to be as comprehensive as possible by using words used by the US Census, a racial slurs database, and previous studies, while also balancing search constraints that limited the number of characters that can be used for searches to 1024 characters. Data collected is what people were willing to express on Twitter, where negative content tends to spread faster [60] and, therefore, may be overestimated. Furthermore, we analyzed tweets referencing different racial groups for sentiment or emotional tone, which is different from previous work that has more directly focused on hate speech and racial slurs, for example [61-63]. However, we believe this approach to be sensitive to subtle expressions https://www.jmir.org/2023/1/e44990 XSL•FO RenderX exclusive limitations include our of racial attitudes that are more representative of the area-level culture and have previously found that such measures are associated with explicit and implicit forms of racial bias [64]. Other focus on English-language tweets. Future research can examine non-English language tweets referencing minoritized populations to evaluate how the racial sentiment of English and non-English tweets may differ. Approximately 1 in 5 US adults use Twitter [31]. Racial or ethnic minoritized populations are slightly overrepresented on Twitter compared to the US general population. For example, 17% of adult Twitter users are Latinx compared to 15% of the US adult population [65]. Twitter users skew younger than the general US population, with 42% of Twitter users being between the ages of 18 and 29 years [66], whereas those aged 19-34 years account for only 20.8% of the US population [67]. However, Twitter users have become more similar to the US population in terms of education over time. In 2016, Twitter users tended to be more educated than the US population [68]. In 2021, 33% of adult Twitter users have a college or graduate degree [66] compared to 38% of US adults 25 years and older with a college or graduate degree [69]. Visitors tweeting from a location may express different sentiments compared to residents of places. Twitter users differ in their frequency of tweeting with the majority of Twitter users being less frequent users. However, in our data on race-related tweets, tweets from users tweeting more than 1000 times per year represent less than 1% (n=1066) of the data. Despite these limitations, the Twitter data provide a broad signal for the social context and the online environment people may be exposed to. The focus is not on individual users at one particular time point or place but rather on the patterns of associations with birth outcomes in the United States by using a continuous random sample of US tweets from 2011 to 2021. The state-level aggregation of racial sentiment also makes it challenging to tease apart the contribution of state politics (structural) and sentiment (cultural) to birth outcomes. For instance, Medicaid is known to mitigate racial health disparities [70], but states vary in the degree to which they expanded Medicaid benefits to their constituents [71]. Sentiment toward minority race groups is an important, yet hard-to-measure factor that may contribute to the instantiation of policies intended to impact these groups [17]. Given that cultural and structural racism are mutually reinforcing factors, areas that are more welcoming to racial or ethnic minorities may have greater safety net programs for example. Future work can build on the current findings to further examine the mutually reinforcing nature of state-level sentiment and policies contributing to racial health disparities. Policies that support racial literacy and cultural competency training as well as policies and supports that promote racial equity more broadly (housing, education, criminal justice, media, and city planning) and establish an inclusive social environment accepting of all races and cultures may improve population health, decreasing the overall risk of adverse birth outcomes and reducing racial-birth outcome disparities. Additionally, our results suggest that policies and practices that ostracize, stigmatize, or otherwise isolate certain racial or ethnic groups and increase hostility directed toward them may exacerbate J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al adverse birth outcomes, which would perpetuate the impact of racism or discrimination on future generations. Conclusions This study represents advancement in our ability to assess racial sentiment at a more fine-grained level as a means to provide more detailed explanations for racial disparities in birth outcomes. Social media provide a unique opportunity to examine time, geographic location, and for racial sentiment across different racial groups. By examining the relationship between sentiment and birth outcomes across racial groups, we are better able to assess the specificity and generalizability of cultural racism in the outcomes for different groups. Future work can build on our findings, and use this work to inform policy aimed at reducing racial disparities in health. Acknowledgments Research reported in this publication was supported by the National Institute on Minority Health and Health Disparities (R00MD012615 (TTN), R01MD015716 (TTN), R01MD016037 (QCN)), the National Library of Medicine (R01LM012849 (QCN)). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Data Availability Twitter data were collected using Twitter’s Application Programming Interface for Academic Research. More information on applying for access can be found at [72]. The paper also uses restricted US natality files with geographic identifiers. The files were obtained after submitting a research proposal to and obtaining approval for data access from the National Center for Health Statistics. More information on applying can be found at [73]. We have added example code to run the analyses in Multimedia Appendix 1. Conflicts of Interest None declared. Multimedia Appendix 1 Supplementary figures, tables, and code. [DOCX File , 3769 KB-Multimedia Appendix 1] References 1. 2. 3. 4. 5. 6. 7. 8. 9. Centers for Disease Control and Prevention (CDC). Infant mortality and low birth weight among black and white infants–United States, 1980-2000. MMWR Morb Mortal Wkly Rep 2002;51(27):589-592 [FREE Full text] [Medline: 12139201] US Department of Health and Human Services. Healthy People 2010: Understanding and Improving Health. Washington, DC: US Department of Health and Human Services; 2000. Kim D, Saada A. The social determinants of infant mortality and birth outcomes in Western developed nations: a cross-country systematic review. Int J Environ Res Public Health 2013;10(6):2296-2335 [FREE Full text] [doi: 10.3390/ijerph10062296] [Medline: 23739649] Hille ET, den Ouden AL, Bauer L, van den Oudenrijn C, Brand R, Verloove-Vanhorick SP. School performance at nine years of age in very premature and very low birth weight infants: perinatal risk factors and predictors at five years of age. Collaborative project on preterm and small for gestational age (POPS) infants in The Netherlands. J Pediatr 1994;125(3):426-434. [doi: 10.1016/s0022-3476(05)83290-1] [Medline: 8071753] Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008;371(9608):261-269. [doi: 10.1016/S0140-6736(08)60136-1] [Medline: 18207020] Couzin J. Quirks of fetal environment felt decades later. Science 2002;296(5576):2167-2169. [doi: 10.1126/science.296.5576.2167] [Medline: 12077397] Artiga S, Pham O, Orgera K, Ranji U. Racial disparities in maternal and infant health: an overview - issue brief. KFF. 2020. URL: https://www.kff.org/report-section/racial-disparities-in-maternal-and-infant-health-an-overview-issue-brief/ [accessed 2022-09-16] Leonard SA, Main EK, Scott KA, Profit J, Carmichael SL. Racial and ethnic disparities in severe maternal morbidity prevalence and trends. Ann Epidemiol 2019;33:30-36 [FREE Full text] [doi: 10.1016/j.annepidem.2019.02.007] [Medline: 30928320] Almeida J, Bécares L, Erbetta K, Bettegowda VR, Ahluwalia IB. Racial/ethnic inequities in low birth weight and preterm birth: the role of multiple forms of stress. Matern Child Health J 2018;22(8):1154-1163. [doi: 10.1007/s10995-018-2500-7] [Medline: 29442278] https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al 10. Alhusen JL, Bower KM, Epstein E, Sharps P. Racial discrimination and adverse birth outcomes: an integrative review. J Midwifery Womens Health 2016;61(6):707-720 [FREE Full text] [doi: 10.1111/jmwh.12490] [Medline: 27737504] 11. Dominguez TP, Dunkel-Schetter C, Glynn LM, Hobel C, Sandman CA. Racial differences in birth outcomes: the role of general, pregnancy, and racism stress. Health Psychol 2008;27(2):194-203 [FREE Full text] [doi: 10.1037/0278-6133.27.2.194] [Medline: 18377138] Jones CP. Levels of racism: a theoretic framework and a gardener's tale. Am J Public Health 2000;90(8):1212-1215. [doi: 10.2105/ajph.90.8.1212] [Medline: 10936998] 12. 13. David EJR, Schroeder TM, Fernandez J. Internalized racism: a systematic review of the psychological literature on racism's most insidious consequence. J Soc Issues 2019;75(4):1057-1086. [doi: 10.1111/josi.12350] 14. Williams O, Ovbiagele B. Stroking out while black-the complex role of racism. JAMA Neurol 2020;77(11):1343-1344. [doi: 10.1001/jamaneurol.2020.3510] [Medline: 32821899] 15. Elias A, Paradies Y. The costs of institutional racism and its ethical implications for healthcare. J Bioeth Inq 2021 Mar;18(1):45-58 [FREE Full text] [doi: 10.1007/s11673-020-10073-0] [Medline: 33387263] 16. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and interventions. Lancet 2017;389(10077):1453-1463. [doi: 10.1016/S0140-6736(17)30569-X] [Medline: 28402827] 17. Williams DR, Lawrence JA, Davis BA. Racism and health: evidence and needed research. Annu Rev Public Health 2019;40:105-125 [FREE Full text] [doi: 10.1146/annurev-publhealth-040218-043750] [Medline: 30601726] Paradies Y. A systematic review of empirical research on self-reported racism and health. Int J Epidemiol 2006;35(4):888-901. [doi: 10.1093/ije/dyl056] [Medline: 16585055] 18. 19. Giscombé CL, Lobel M. Explaining disproportionately high rates of adverse birth outcomes among African Americans: the impact of stress, racism, and related factors in pregnancy. Psychol Bull 2005;131(5):662-683 [FREE Full text] [doi: 10.1037/0033-2909.131.5.662] [Medline: 16187853] 20. Wadhwa PD, Culhane JF, Rauh V, Barve SS. Stress and preterm birth: neuroendocrine, immune/inflammatory, and vascular mechanisms. Matern Child Health J 2001;5(2):119-125. [doi: 10.1023/a:1011353216619] [Medline: 11573837] 21. Nepomnaschy PA, Sheiner E, Mastorakos G, Arck PC. Stress, immune function, and women's reproduction. Ann N Y Acad 22. Sci 2007;1113:350-364. [doi: 10.1196/annals.1391.028] [Medline: 17978283] Swope CB, Hernández D, Cushing LJ. The relationship of historical redlining with present-day neighborhood environmental and health outcomes: a scoping review and conceptual model. J Urban Health 2022;99(6):959-983 [FREE Full text] [doi: 10.1007/s11524-022-00665-z] [Medline: 35915192] 23. Rodat S. Cultural racism: a conceptual framework. Revista de tiin e Politice/ Revue des Sciences Politiques 2017(54):129-140 [FREE Full text] 24. Hicken MT, Miles L, Haile S, Esposito M. Linking history to contemporary state-sanctioned slow violence through cultural and structural racism. Ann Am Acad Pol Soc Sci 2021;694(1):48-58 [FREE Full text] [doi: 10.1177/00027162211005690] [Medline: 34446942] 25. Axt JR. The best way to measure explicit racial attitudes is to ask about them. Soc Psychol Personal Sci 2017;9(8):896-906. [doi: 10.1177/1948550617728995] 26. McConnell AR, Leibold JM. Relations among the implicit association test, discriminatory behavior, and explicit measures of racial attitudes. J Exp Soc Psychol 2001;37(5):435-442. [doi: 10.1006/jesp.2000.1470] 27. Michaels EK, Board C, Mujahid MS, Riddell CA, Chae DH, Johnson RC, et al. Area-level racial prejudice and health: a systematic review. Health Psychol 2022;41(3):211-224. [doi: 10.1037/hea0001141] [Medline: 35254858] 28. Amodio DM, Harmon-Jones E, Devine PG. Individual differences in the activation and control of affective race bias as assessed by startle eyeblink response and self-report. J Pers Soc Psychol 2003;84(4):738-753. [doi: 10.1037/0022-3514.84.4.738] [Medline: 12703646] 29. Quillian L. New approaches to understanding racial prejudice and discrimination. Annu Rev Sociol 2006;32(1):299-328. [doi: 10.1146/annurev.soc.32.061604.123132] 30. Rosenman R, Tennekoon V, Hill LG. Measuring bias in self-reported data. Int J Behav Healthc Res 2011;2(4):320-332 [FREE Full text] [doi: 10.1504/IJBHR.2011.043414] [Medline: 25383095] 31. Odabas M. 10 facts about Americans and Twitter. Pew Research Center. 2022. URL: https://www.pewresearch.org/fact-tank/ 2022/05/05/10-facts-about-americans-and-twitter/ [accessed 2022-11-26] 32. Albarracin D, Shavitt S. Attitudes and attitude change. Annu Rev Psychol 2017;69(1):1-29. [doi: 10.1146/annurev-psych-122216-011911] 33. Radwan M. Effect of social media usage on the cultural identity of rural people: a case study of Bamha village, Egypt. Humanit Soc Sci Commun 2022;9(1):258. [doi: 10.1057/s41599-022-01268-4] 34. Nguyen TT, Adams N, Huang D, Glymour MM, Allen AM, Nguyen QC. The association between state-level racial attitudes assessed from Twitter data and adverse birth outcomes: observational study. JMIR Public Health Surveill 2020;6(3):e17103 [FREE Full text] [doi: 10.2196/17103] [Medline: 32298232] 35. Bartlett J, Reffin J, Rumball N, Williamson S. Anti-social media. DEMOS. 2014. URL: http://www.demos.co.uk/files/ DEMOS_Anti-social_Media.pdf [accessed 2023-02-03] https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al 36. The Racial Slur Database. URL: http://www.rsdb.org/ [accessed 2023-02-03] 37. Go A, Bhayani R, Huang L. Twitter sentiment classification using distant supervision. Sentiment140. 2009. URL: http:/ /help.sentiment140.com/home [accessed 2022-11-28] 38. UMICH SI650 - sentiment classification. Kaggle. 2018. URL: https://kaggle.com/competitions/si650winter11 [accessed 39. 2022-11-26] Sanders N. Twitter sentiment corpus. Sanders Analytics. 2011. URL: https://github.com/zfz/twitter_corpus [accessed 2022-11-26] 40. Hunter JD. Matplotlib: a 2D graphics environment. Comput Sci Eng 2007;9(3):90-95. [doi: 10.1109/mcse.2007.55] 41. Tweet metadata timeline. Twitter Developer Platform. URL: https://developer.twitter.com/en/docs/twitter-api/enterprise/ data-dictionary/tweet-timeline [accessed 2023-03-13] 42. Dolan SM, Gross SJ, Merkatz IR, Faber V, Sullivan LM, Malone FD, et al. The contribution of birth defects to preterm birth and low birth weight. Obstet Gynecol 2007;110(2 Pt 1):318-324. [doi: 10.1097/01.AOG.0000275264.78506.63] [Medline: 17666606] 43. Luke B, Brown MB. The changing risk of infant mortality by gestation, plurality, and race: 1989-1991 versus 1999-2001. Pediatrics 2006;118(6):2488-2497 [FREE Full text] [doi: 10.1542/peds.2006-1824] [Medline: 17142535] 44. DeVellis RF, Thorpe CT. Scale Development : Theory and Applications. Thousand Oaks, CA: SAGE Publications; 2021. 45. Elmståhl S, Gullberg B. Bias in diet assessment methods--consequences of collinearity and measurement errors on power and observed relative risks. Int J Epidemiol 1997;26(5):1071-1079. [doi: 10.1093/ije/26.5.1071] [Medline: 9363530] 46. Nguyen QC, Li D, Meng HW, Kath S, Nsoesie E, Li F, et al. Building a national neighborhood dataset from geotagged Twitter data for indicators of happiness, diet, and physical activity. JMIR Public Health Surveill 2016;2(2):e158 [FREE Full text] [doi: 10.2196/publichealth.5869] [Medline: 27751984] 47. American community survey (ACS). US Census Bureau. URL: https://www.census.gov/programs-surveys/acs [accessed 2022-11-26] StataCorp LLC. Stata. URL: https://www.stata.com/company/ [accessed 2022-11-26] 48. 49. McGhee H. The Sum of Us : What Racism Costs Everyone and How We Can Prosper Together. New York: One World; 2022. 50. Akala A. Cost Of racism: U.S. economy lost $16 trillion because of discrimination, bank says. NPR. 2020. URL: https:/ /tinyurl.com/2s3th4ye [accessed 2022-11-03] 51. Boyle P. Racial resentment hurts White people too, physician tells colleagues. AAMC. 2022. URL: https://www.aamc.org/ news-insights/racial-resentment-hurts-white-people-too-physician-tells-colleagues [accessed 2022-11-26] 52. Tesler M. The spillover of racialization into health care: how President Obama polarized public opinion by racial attitudes and race. Am J Pol Sci 2012;56(3):690-704. [doi: 10.1111/j.1540-5907.2011.00577.x] 53. Debbink MP, Bader MDM. Racial residential segregation and low birth weight in Michigan's metropolitan areas. Am J Public Health 2011;101(9):1714-1720. [doi: 10.2105/AJPH.2011.300152] [Medline: 21778487] 54. Williams AD, Wallace M, Nobles C, Mendola P. Racial residential segregation and racial disparities in stillbirth in the 55. 56. United States. Health Place 2018;51:208-216 [FREE Full text] [doi: 10.1016/j.healthplace.2018.04.005] [Medline: 29715639] no authors listed. The relationship of fetal and infant mortality to residential segregation: an inquiry into social epidemiology. 1949. Am J Public Health 2015;105(2):278-281. [doi: 10.2105/AJPH.2014.1052278] [Medline: 25574696] Jones CP. Toward the science and practice of anti-racism: launching a national campaign against racism. Ethn Dis 2018;28(suppl 1):231-234 [FREE Full text] [doi: 10.18865/ed.28.S1.231] [Medline: 30116091] 57. Nguyen TT, Criss S, Michaels EK, Cross RI, Michaels JS, Dwivedi P, et al. Progress and push-back: how the killings of Ahmaud Arbery, Breonna Taylor, and George Floyd impacted public discourse on race and racism on Twitter. SSM Popul Health 2021;15:100922 [FREE Full text] [doi: 10.1016/j.ssmph.2021.100922] [Medline: 34584933] 58. Nguyen TT, Criss S, Dwivedi P, Huang D, Keralis J, Hsu E, et al. Exploring U.S. shifts in anti-asian sentiment with the emergence of COVID-19. Int J Environ Res Public Health 2020;17(19):7032 [FREE Full text] [doi: 10.3390/ijerph17197032] [Medline: 32993005] 59. Criss S, Nguyen TT, Michaels EK, Gee GC, Kiang MV, Nguyen QC, et al. Solidarity and strife after the Atlanta spa shootings: a mixed methods study characterizing Twitter discussions by qualitative analysis and machine learning. Front Public Health 2023;11:952069 [FREE Full text] [doi: 10.3389/fpubh.2023.952069] [Medline: 36825140] 60. Bellovary AK, Young NA, Goldenberg A. Left- and right-leaning news organizations use negative emotional content and elicit user engagement similarly. Affect Sci 2021;2(4):391-396 [FREE Full text] [doi: 10.1007/s42761-021-00046-w] [Medline: 34423311] 61. Chae DH, Clouston S, Martz CD, Hatzenbuehler ML, Cooper HL, Turpin R, et al. Area racism and birth outcomes among Blacks in the United States. Soc Sci Med 2018;199:49-55 [FREE Full text] [doi: 10.1016/j.socscimed.2017.04.019] [Medline: 28454665] 62. Chae DH, Clouston S, Hatzenbuehler ML, Kramer MR, Cooper HLF, Wilson SM, et al. Association between an internet-based measure of area racism and Black mortality. PLoS One 2015;10(4):e0122963 [FREE Full text] [doi: 10.1371/journal.pone.0122963] [Medline: 25909964] https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Nguyen et al 63. Arcila-Calderón C, Amores JJ, Sánchez-Holgado P, Vrysis L, Vryzas N, Oller Alonso M. How to detect online hate towards migrants and refugees? Developing and evaluating a classifier of racist and xenophobic hate speech using shallow and deep learning. Sustainability 2022;14(20):13094. [doi: 10.3390/su142013094] 64. Nguyen TT, Huang D, Michaels EK, Glymour MM, Allen AM, Nguyen QC. Evaluating associations between area-level Twitter-expressed negative racial sentiment, hate crimes, and residents' racial prejudice in the United States. SSM Popul Health 2021;13:100750 [FREE Full text] [doi: 10.1016/j.ssmph.2021.100750] [Medline: 33665332] 65. Wojcik S, Hughes A. Sizing up Twitter users. Pew Research Center. URL: https://www.pewresearch.org/internet/2019/04/ 24/sizing-up-twitter-users/ [accessed 2023-03-13] 66. Auxier B, Anderson M. Social Media Use in 2021. Pew Research Center. 2021. URL: https://www.pewresearch.org/internet/ 67. 2021/04/07/social-media-use-in-2021/ [accessed 2022-11-28] Population distribution by age. KFF. 2022. URL: https://www.kff.org/other/state-indicator/distribution-by-age/ [accessed 2022-11-28] 68. Greenwood S, Perrin A, Duggan M. Social media update 2016. Pew Research Center. 2016. URL: https://www. pewresearch.org/internet/2016/11/11/social-media-update-2016/ [accessed 2022-11-28] 69. Annual social and economic supplements. US Census Bureau. 2021. URL: https://www.census.gov/data/datasets/time-series/ demo/cps/cps-asec.html [accessed 2022-11-28] 70. Guth M, Artiga S. Medicaid and racial health equity. KFF. 2022. URL: https://www.kff.org/medicaid/issue-brief/ medicaid-and-racial-health-equity/ [accessed 2022-11-26] 71. Cross-Call J. Medicaid expansion has helped narrow racial disparities in health coverage and access to care. Center on Budget and Policy Priorities. 2020. URL: https://www.cbpp.org/research/health/ medicaid-expansion-has-helped-narrow-racial-disparities-in-health-coverage-and [accessed 2022-11-26] Preparing for the application. Twitter Developer Platform. URL: https://developer.twitter.com/en/products/twitter-api/ academic-research/application-info [accessed 2023-04-20] 72. 73. Restricted-use vital statistics data. Centers for Disease Control and Prevention. URL: https://www.cdc.gov/nchs/nvss/ nvss-restricted-data.htm [accessed 2023-04-20] Abbreviations FIPS: Federal Information Processing Standards LBW: low birth weight PTB: preterm birth SVM: support vector machine Edited by A Mavragani; submitted 12.12.22; peer-reviewed by R Gore, K Marchi; comments to author 08.03.23; revised version received 22.03.23; accepted 28.03.23; published 28.04.23 Please cite as: Nguyen TT, Merchant JS, Criss S, Makres K, Gowda KN, Mane H, Yue X, Hswen Y, Glymour MM, Nguyen QC, Allen AM Examining Twitter-Derived Negative Racial Sentiment as Indicators of Cultural Racism: Observational Associations With Preterm Birth and Low Birth Weight Among a Multiracial Sample of Mothers, 2011-2021 J Med Internet Res 2023;25:e44990 URL: https://www.jmir.org/2023/1/e44990 doi: 10.2196/44990 PMID: 37115602 ©Thu T Nguyen, Junaid S Merchant, Shaniece Criss, Katrina Makres, Krishik N Gowda, Heran Mane, Xiaohe Yue, Yulin Hswen, M Maria Glymour, Quynh C Nguyen, Amani M Allen. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 28.04.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e44990 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e44990 \| p. 12 (page number not for citation purposes)
10.3390_ijerph16091557	Article Exploring Food Access and Sociodemographic Correlates of Food Consumption and Food Insecurity in Zanzibari Households Maria Adam Nyangasa 1,† Antje Hebestreit 1,* , Christoph Buck 1,† , Soerge Kelm 2, Mohammed Sheikh 3 and 1 Leibniz Institute for Prevention Research and Epidemiology—BIPS, 28359 Bremen, Germany; [email protected] (M.A.N.); [email protected] (C.B.) 2 Center for Biomolecular Interactions Bremen, University Bremen, 28334 Bremen, Germany; 3 [email protected] Environmental Analytical Chemistry and Eco-toxicology, Zanzibar State University, P.O. Box 146 Unguja, Zanzibar; [email protected] * Correspondence: [email protected]; Tel.: +49-421-218-56849; Fax: +49-421-218-56821 † These authors contributed equally to the work. Received: 25 February 2019; Accepted: 1 May 2019; Published: 4 May 2019 Abstract: Rapid growth of the Zanzibari population and urbanization are expected to impact food insecurity and malnutrition in Zanzibar. This study explored the relationship between food access (FA) and sociodemographic correlates with food consumption score and food insecurity experience scale. Based on cross-sectional data of 196 randomly selected households, we ﬁrst investigated the association between sociodemographic correlates and Food Consumption Score (FCS) and Food Insecurity Experience Scale using multilevel Poisson regression. Secondly, the role of FA in these associations was investigated by interaction with the respective correlates. About 65% of households had poor food consumption, and 32% were severely food-insecure. Poor FA was more prevalent in households with poor food consumption (71%). Polygamous households and larger households had a higher chance for severe food insecurity. In the interaction with FA, only larger households with poor FA showed a higher chance for severe food insecurity. In households having no vehicle, good FA increased the chance of having acceptable FCS compared to poor FA. By contrast, urban households with good FA had a twofold chance of acceptable FCS compared to rural household with poor FA. Poor FA, poor food consumption and food insecurity are challenging; hence, facilitating households’ FA may improve the population’s nutrition situation. Keywords: demographic correlates; food access; household; food insecurity experience scale; Zanzibar; sub-Saharan Africa 1. Introduction The world population is expected to increase by 2.5 billion between 2007 and 2050, with most of the growth foreseen to occur in urban areas of developing countries [1]. This rapid growth and urbanization are expected to increase poverty and negatively impact the food security environment of urban dwellers, leading to food insecurity and malnutrition [2]. According to the World Bank report from 2015 [3], 29% of Tanzanians could not meet their basic consumption needs, and about 10% of the population could not aﬀord to buy basic food stuﬀ. In Tanzania and other developing countries, the leading factor in household food insecurity in urban areas is the dependency on food purchase [2,4,5]. Therefore, a slight increase in food prices has a major impact on vulnerable households, pushing them into hunger and poverty [6]. About 80% of the household food requirement in peri-urban areas of Int. J. Environ. Res. Public Health 2019, 16, 1557; doi:10.3390/ijerph16091557 www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2019, 16, 1557 2 of 15 Unguja Island, Zanzibar is purchased [7,8], while about 60% of food consumed in rural areas is obtained through home gardening and farming [9]. Home gardening and farming contribute substantially to households’ own food production [10] and may thus enhance household food availability and food security in rural areas. Own food production increases purchasing power, due to savings on food bills [4,11] and income from selling of the produce. It also provides a diversity of nutritious food that helps to improve the health status of the household [11] as well as serves as a means of food provision during food shortage. Studies conducted in Nigeria [12] and Ethiopia [13] have shown that male heads of household (HH) play a substantial role in determining household food security, while others reported that female HH are more likely to spend most of their income on food, thus guaranteeing food security for their households [14,15]. Household size has also been shown to inﬂuence food security and acceptable food consumption, with smaller household size being associated with household food security [16] and acceptable food consumption, and large household size with poor food consumption and perceived food insecurity [17]. Food insecurity has also been found to be associated with low monthly income [18]. Income earned from any source improves the food situation of a household [19]; thus, households with more employed adult members are likely to have a better food situation compared to households with more unemployed adult members [20]. Several studies conducted in low-income countries, especially in Africa [4,5,16,18], have investigated the determinants of either household food security or food consumption behavior in households. However, this is the ﬁrst study in Zanzibar that recruited randomly selected households and enrolled all members of a household for further insights on correlates of food insecurity and food consumption. The present study aimed to explore the relationship between food access (FA) and sociodemographic household factors with the Food Consumption Score (FCS) and Food Insecurity Experience Scale (FIES). Findings from this study can provide baseline information on the interaction between food access and household factors, and the collected data can be used for further research on health interventions to improve food consumption and food security in Zanzibar. 2. Materials and Methods 2.1. Study Area, Population, and Sampling A population-based cross-sectional study was conducted in 2013 in Unguja Island, Zanzibar, whereby entire households were enrolled as sampling units. For the purposes of this study, a household is deﬁned according to Beaman and Dillon [21], with emphasis on eating from the same pot, i.e., having the same food provider. This is particularly important as our study population consisted not only of monogamous but also of polygamous families, who do not necessarily live together in the same house or compound. Household aspects were reported by the head of household in a questionnaire-administered personal interview. In total, 244 randomly selected households were contacted from 80 Shehias (wards), and 239 (97.9%) participated in the survey. Due to missing information on socioeconomic status, demographic correlates, and responses from FCS and FIES instruments, 43 households were excluded from the analysis, resulting in a ﬁnal sample of 196 (82%) households. Further details on sampling procedures, data collection, and quality management are provided elsewhere [22]. Prior to the data collection process, all participants gave written, informed consent. All procedures applied in this study were approved by the Ethics Committees of the University of Bremen (in September 26, 2013) as well as the Zanzibar Ministry of Health and the Zanzibar Medical Research and Ethics Committee (ZAMREC/0001/AUGUST/013) in accordance with the ethical standards according to the 1964 Declaration of Helsinki and its later amendments. Int. J. Environ. Res. Public Health 2019, 16, 1557 3 of 15 2.2. Questionnaires A structured household questionnaire was used to collect general household information. The information included data on socioeconomic and demographic indicators of the household and of the head of household, such as area of residence, number of animals owned, number of vehicles belonging to the household, household size, marital status, education level of the HH, occupation of the HH, etc. In cases of polygamy, household information was also collected from the households of the other wife or wives. Information on indicators of household food consumption was collected using a standardized questionnaire (Food Consumption Score, FCS), adapted from the United Nations World Food Programme (UNWFP) [23]. Household food insecurity was measured at the household level using a standardized questionnaire for the Food Insecurity Experience Scale (FIES), which was adapted from the Food and Agriculture Organization (FAO) [24]. All questionnaires were developed in English, translated into Swahili and back-translated to check for translation errors. 2.2.1. Food Consumption Score (FCS) FCS is a composite score constructed from (1) household dietary diversity based on nine food groups (staples, pulses, fruits, vegetables, meat and ﬁsh, dairies, sugar, oil and fat, condiments) consumed during the 7 days preceding the survey, (2) food frequency, counted as the number of days each food group was consumed during the 7 days preceding the survey, and (3) relative nutritional importance of diﬀerent food groups, applying a weighting system [23], thus reﬂecting the quality and quantity of food consumed in the household. Higher weights were given to energy-dense foods with proteins of high quality and a range of bioavailable micronutrients, while lower weights were given to oil and sugar, which are energy-dense but contain—if any—proteins of low quality and low levels of micronutrients [25]. Cut-oﬀ points established by the UNWFP [23] were used to classify FCS. They were computed by summing up the weighted frequencies of the diﬀerent food groups consumed in the household. FCS ≤ 28 was categorized as “poor”, FCS > 28 and < 42 as “borderline”, and FCS ≥ 42 as “acceptable” [23]. For the regression analysis in this study, poor consumption and borderline were merged to a new category, “poor”, resulting in two categories of food consumption score, i.e., “poor” and “acceptable”. 2.2.2. Food Insecurity Experience Scale (FIES) Food insecurity was assessed using the Food Insecurity Experience Scale (FIES), a standardized set of questions developed by the FAO [24] that has been applied in a large number of countries following a standardized procedure. The scale, which is an experience-based metric of severity of food insecurity that relies on people’s direct/actual responses, includes components of uncertainty and worry about food, inadequate food quality, and insuﬃcient food quantity. It consists of a set of eight items that assess food-related behaviors associated with diﬃculties in accessing food due to limited resources. The instrument measures the degree of food insecurity/hunger experienced by individuals during the 12 months preceding the survey. Household scores of food insecurity on the eight items were scaled based on a Rasch model as an application of the item–response theory (IRT) [26]. In IRT, the response to each item is modelled as a function of item and household parameters to measure the position of households on a latent trait, independently of the item diﬃculty. We conducted the Rasch model using the eRm package 0.16–2 [27] in R 3.5.1 to derive household scores for the following regression analyses. Households who responded ‘no’ to any of the eight items received the lowest value. Item characteristic curves indicated item reliability, and the person separation index was found to be high (0.84). Household scores ranged from −3.57 to 3.76 and clustered either to the lowest negative scores (<−2), around 0 (−2 to 2) or to the highest scores. Based on these cut-oﬀ points, we categorized household scores into mild, moderate, and severe (hunger), which are the three categories used by the FAO [24] to deﬁne levels of food insecurity/hunger. For the Poisson regression analysis, food security, mild and moderate Int. J. Environ. Res. Public Health 2019, 16, 1557 4 of 15 food insecurity were dichotomized into “mild to moderate” (0) and “severe food insecurity” (1). A household is considered as food-secure if members have always had enough food and no hunger worries [24]. 2.3. Correlates Correlates of FCS and FIES were assessed for the HH (gender, education level, number of jobs, marital status) and at household level (household size, number of types of animals kept in the household, vehicles owned by any of the household members, area of residence, and food access). The highest education level of the HH was assessed using the International Standard Classiﬁcation of Education (ISCED) [28] and was categorized for the analysis as low education level (primary school and below) and high education level (secondary school and above). Number of jobs of HH for assessing main household income was deﬁned as “no job” and “≥1 job”. Marital status of HH was calculated in three categories: Married monogamous, married polygamous, and other (single, widowed, cohabitating or divorced). To facilitate interpretation, two categories for marital status were derived: Married (monogamous or polygamous) and not married (single, widow, cohabitation, divorced). Cohabitation was categorized as “not married” since it is characterized by a diﬀerent socioeconomic status compared to those households with married HH. Using the mean household size of 6 members in this study population as a cut-oﬀ, household size was classiﬁed as large (six or more members) and small (less than six members). To assess household wealth, the number of types of animals kept in the household from a list of six items including ducks, goats, sheep, cows, ﬁsh, and chicken, and the number of assets from a list of eight items including electricity, radio, mobile phone, iron, kerosene lamps, television, refrigerator, and non-mobile phone was summed up. The median number of animals and assets was ﬁve, and this ﬁgure was used to categorize wealth into wealthy (≥5) and poor (<5). The number of vehicles per household was assessed as type of transportation owned by any member of the household from a list of 5 categories (bicycle, car/truck, boat, motorcycle, none). Two categories were built: “none” and “at least one type of vehicle”. Area of residence was assessed as rural or urban. An important component of food consumption and food insecurity included in the analysis was FA. In this study, FA was deﬁned as the ability of a household to acquire adequate amount of food through mixed strategies. Indicators for FA were assessed, and derived variables were combined to a composite score (see Table 1 below). The derived variables were: (a) Food source; main source of food consumed during the last seven days (purchased, borrowed, own production, traded food/barter, received as gift, food aid, other), (b) food purchased; types of food (cereals, starchy vegetables/ tubers, vegetables, fruits, legumes, meat, egg, milk, ﬁsh, oils and fats, any kind of beverage, other), frequently bought from shop/market during the last seven days, (c) own food; types of foods (from those listed above) of own household production, and (d) market distance; distance to the nearest market or shop (<30 min, 30 to 60 min, 1 h to 2 h, >2 h). To classify households as having poor or good FA, a composite score for FA was computed as the mean of the values of the derived variables multiplied by 4 (number of all derived variables). The FA-score ranged from 4–8, and FA was then categorized as “poor food access” (≤6) and “good food access” (>6). The cut-oﬀ was set according to the distribution of FA-status in the study population, as half of the population included in the study had FA score of 6 and below. Int. J. Environ. Res. Public Health 2019, 16, 1557 5 of 15 Table 1. Overview of measured indicators for FA and derivation of variables for the composite score. Indicator of Food Access Derived Variable Categories Main source of food consumed One main source to be selected from 6 categories: purchased, borrowed, traded food/barter, received as gift, food aid, own production Types of food groups frequently bought from shop/market Food group (e.g., cereals) out of 11 food groups was 0: not bought, 1: bought Types of food groups of own household production Food groups out of 11 food groups 0: not produced (purchased, borrowed, traded food/barter, received as gift, food aid), 1: own production Distance to the nearest market/shop far (>30 min walking distance); near (<30 min walking distance) 2.4. Statistical Analysis Food source = one main source per household 1: borrowed, received as gift, food aid, other 2: own production, traded food/barter, purchased Food purchased = sum of all food groups purchased 1: ≤4 food groups, 2: >4 food groups Own food = sum of all food groups with own production 1: ≤2 food groups, 2: >2 food groups Market distance 1: far, 2: near Study characteristics such as socioeconomic and demographic variables were calculated for categories of food consumption and food insecurity. First, the associations between the exposure variables (food access, socioeconomic and demographic correlates) with either food consumption (FCS; Model 1) or food insecurity (FIES; Model 2) as outcome variables were explored. To explore these associations, linear, Poisson, and logistic regression models were considered and evaluated with regard to model ﬁt and Pearson residuals as well as quantile-quantile (Q–Q) plots. Eventually, we conducted multilevel logistic regression models to calculate odds ratios (OR) and 95% conﬁdence limits (CI) and to account for clustering within Shehia level using a random intercept. Secondly, statistical interactions between each correlate and food access (correlate*FA) were investigated. Hence, in Models 3a–h and 4a–h, the predictive power of each socioeconomic and demographic factor with FA on FCS and FIES was tested separately. Each model was again adjusted for the remaining correlates, including a random intercept for the Shehia level. All statistical analyses were performed using SAS 9.3 (SAS Institute, Cary, NC, USA). Due to the exploratory design of our study, we only considered conﬁdence limits as a precision measure of the point estimates but did not apply a level of signiﬁcance. Moreover, we did not adjust for multiple testing. Noteworthy associations are presented considering higher (OR > 1.5) or lower (OR < 0.66) chances for the modelled response category. 3. Results 3.1. Household Characteristics The sample data were based on the responses of the HH. The majority of the households were headed by men (63%, 123/196), and about 55% of the HH (107/196) were in a monogamous marriage. More than half of the HH had one or more sources of income that he/she contributed to the household. Most of the households were in rural areas, and the overall mean household size of the participating households was 6 persons. More than 60% of the households had a good socioeconomic status, with one or more than one animal kept and at least one vehicle owned by a member belonging to the Int. J. Environ. Res. Public Health 2019, 16, 1557 6 of 15 household. Overall, about 65% of the households had poor food consumption, and about 32% were severely food-insecure (Table 2). Acceptable food consumption was more prevalent in households with higher-educated HH (40%), in monogamous households (38%), and in larger households (six or more members) (38%). Severe food insecurity was more prevalent in polygamous households (40%), in households with low-educated HH (40%) and in larger households (six or more members) (40%). Looking at each question of the FIES, the majority of the households (73.5%) indicated having eaten few kinds of food in the last 12 months due to lack of money, and 26% went without eating for a whole day due to lack of money (Table 3). About 54% (106/196) of the study population had poor FA, of which 71% had poor food consumption and 35% experienced severe food insecurity. Table 2. Proportion of food consumption and food insecurity experience scale according to demographic and socioeconomic factors. All Household Demographics Gender Male Female Marital status of HH Not married a Married monogamous Married polygamous Education level Low High Number of jobs No job One or more jobs Area Rural Urban Household size b Small Large Socioeconomic Factors Wealth (household assets and animals) c Poor Wealthy Number of vehicles None At least one vehicle Food access d Poor Good Food Consumption Score Food Insecurity Experience Scale Poor Acceptable Mild to Moderate Severe Total N 128 72 56 27 66 35 69 59 61 67 105 23 68 60 61 67 31 97 75 53 % 65.3 58.5 76.7 79.4 61.7 63.6 71.1 59.6 70.1 61.5 68.6 53.5 68.7 61.9 73.5 59.3 72.1 63.4 70.8 58.9 N 68 51 17 7 41 20 28 40 26 42 48 20 31 37 22 46 12 56 31 37 % 34.7 41.5 23.3 20.6 38.3 36.4 28.9 40.4 29.9 38.5 31.4 46.5 31.3 38.1 26.5 40.7 27.9 36.6 29.2 41.1 N 134 80 54 28 73 33 59 75 57 75 102 32 75 59 51 83 28 106 69 65 % 68.4 65.0 74.0 82.4 68.2 60.0 60.8 75.8 65.5 68.8 66.7 74.4 75.8 60.8 61.4 73.5 65.1 69.3 65.1 72.2 N 62 43 19 6 34 22 38 24 30 24 51 11 24 38 32 30 15 47 37 25 % 31.6 35.0 26.0 17.6 31.8 40.0 39.2 24.2 34.5 22.2 33.3 25.6 24.2 39.2 38.6 26.5 34.9 30.7 34.9 27.8 N 196 123 73 34 107 55 97 99 87 109 153 43 99 97 83 113 43 153 106 90 a Not married includes single, divorced, widow, and cohabitation; b cut-oﬀ was derived from the mean number of household members in this study, small (≤6) and large (>6); c wealth (poor <5 and wealthy ≥5, calculated as the median number of animals and assets in the household); d cut-oﬀ (poor ≤6, good >6). Table 3. Questions of the Food Insecurity Experience Scale and aﬃrmatively answered questions by the study population in Zanzibar (N = 196). During the last 12 months, was there a time when . . . No 1 2 3 4 5 6 7 8 Food Insecurity Experience Scale Questions You were worried you would run out of food because of a lack of money? You were unable to eat healthy and nutritious food because of a lack of money? You ate only a few kinds of foods because of a lack of money? You had to skip a meal because there was not enough money to get food? You ate less than you thought you should because of a lack of money? Your household ran out of food because of a lack of money? You were hungry but did not eat because there was not enough money for food? You went without eating for a whole day because of a lack of money? N 112 134 144 100 117 103 76 51 % 57.1 68.4 73.5 51.0 59.7 52.6 38.8 26.0 Int. J. Environ. Res. Public Health 2019, 16, 1557 7 of 15 3.2. Correlates of Food Consumption and Food Insecurity Households with a higher-educated HH had a lower chance of reporting severe food insecurity (OR 0.53; 95% CI 0.28–1.08) compared to those with lower-educated HH (Table 4). Those HH married in monogamy had a higher chance of reporting acceptable food consumption (OR 1.71; 95% CI 0.55–5.38) but at the same time a higher chance of severe food insecurity (OR 1.83; 95%CI 0.55; 6.08) compared to those HH not married (single, widowed, cohabitating or divorced) (Table 4). Polygamous households had a higher chance of severe food insecurity (OR 3.95; 95% CI 1.17–13.4) and also reported higher chance for acceptable food consumption (OR 1.78; 95% CI 0.54–5.83) compared to those not married. Larger households had a higher chance of severe food insecurity (OR 2.44; 95% CI 1.16–5.13) than smaller households, while wealthy households had a lower chance of severe food insecurity (OR 0.52; 95% CI 0.25–1.11) than poor households. Table 4. Associations of socioeconomic and demographic correlates of 196 households with food consumption (Model 1) and food insecurity (Model 2) in terms of odds ratios (OR) and 95% conﬁdence intervals (CI) as well as model ﬁt (generalized chi-square/degrees of freedom), respectively. Model ﬁt Between Shehia variance (SE) Gender (ref: female) Marital status of HH (ref: not married) monogamous polygamous Education (ref: low) Number of jobs (ref: no job) Area of residence (ref: rural) Household size (ref: small) Wealth (ref: poor ≥ 5) Number of vehicles (ref: none) Food access (ref: poor ≤ 6) Model 1: Food Consumption (Ref: Poor) χ2/DF OR 1.76 1.71 1.78 1.36 1.22 2.08 1.02 1.35 1.04 1.56 0.87 0.48 (0.38) (95% CI) (0.75–4.11) (0.55–5.38) (0.54–5.83) (0.69–2.70) (0.59–2.50) (0.85–5.10) (0.51–2.05) (0.64–2.83) (0.42–2.57) (0.79–3.10) Model 2: Food Insecurity (Ref: Mild to Moderate) χ2/DF 0.86 0.37 (0.38) (95% CI) (0.68–4.01) (0.055–6.08) (1.17–13.4) (0.26–1.08) (0.28–1.22) (0.24–1.70) (1.16–5.13) (0.25–1.11) (0.32–1.89) (0.33–1.43) RR 1.65 1.83 3.95 0.53 0.58 0.64 2.44 0.52 0.78 0.69 3.3. Role of Food Access on the Correlates, Food Consumption and Food Insecurity There were few relevant changes in the chances observed in the interaction of food access with gender, marital status of HH, number of jobs, and wealth on both food consumption and food insecurity. However, urban households with good FA showed a higher chance of acceptable food consumption compared to rural households with poor FA (Table 5, Model 3e). Households with no vehicle had a higher chance of acceptable food consumption if they had good FA compared to those with poor FA (OR 6.21; 95% CI 1.20–32.3). However, having at least one vehicle tentatively increased the chance of having acceptable food consumption for both good and poor FA (OR 2.17; 95% CI 0.68–6.87; OR 1.83; 95% CI 0.59–5.71, respectively) (Model 3h). In comparison to households with low-educated HH and poor FA, we observed a lower chance of severe food insecurity in households with higher-educated HH either with good FA (OR 0.42; 95% CI 0.15–1.17) or poor FA (OR 0.40; 95% CI 0.16; 1.02) and in households with lower-educated HH and good FA (OR 0.49; 95% CI 0.18–1.30) (Model 4c). Considering poor FA, larger households (six or more members) had a higher chance of severe food insecurity (OR 3.42; 95% CI 1.29–9.10) compared to smaller households (Model 4f). Int. J. Environ. Res. Public Health 2019, 16, 1557 8 of 15 Table 5. Results of the multilevel logistic regressions in terms of odds ratios (OR) and 95% conﬁdence limits as well as model ﬁt (generalized chi-square/degrees of freedom) to investigate the interaction of food access with socioeconomic and demographic correlates on food consumption (Model 3a–h) and food insecurity (Model 4a–h), each adjusted for the remaining correlates. Model Covariate Food Access Model 3: Food Consumption (Ref: Poor) Model 4: Food Insecurity (Ref: Mild to Moderate) Ref: N (%) OR (95% CI) Between Shehia Variance (SE) χ2/DF Ref: N (%) OR (95%CI) Between Shehia Variance (SE) χ2/DF a b c d e f g Gender Male Male Female Female Marital status a Married Married Not married Not married Education High High Low Low Number of jobs One or more One or more No Job No Job Area of residence Urban Urban Rural Rural Household size Large Large Small Small Wealth Wealthy Wealthy Poor Poor Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access Good access Poor access 34 (55.7) 38 (61.3) 19 (65.5) 37 (84.1) 43 (55.8) 58 (68.2) 10 (76.9) 17 (81.0) 28 (58.3) 31 (60.8) 25 (59.5) 44 (80.0) 36 (60.0) 31 (63.3) 17 (56.7) 44 (77.2) 6 (35.3) 17 (65.4) 47 (64.4) 58 (72.5) 23 (52.3) 37 (69.8) 30 (65.2) 38 (71.1) 28 (53.8) 39 (63.9) 25 (65.8) 36 (80.0) 3.03 2.26 2.24 1.00 1.96 1.09 0.62 1.00 2.21 2.09 2.53 1.00 2.05 1.95 2.81 1.00 5.48 1.21 1.16 1.00 1.56 0.70 1.08 1.00 2.21 1.59 1.90 1.00 (0.96–9.62) (0.73–7.00) (0.66–7.66) (0.49–7.88) (0.28–4.25) (0.09–4.20) (0.82–5.98) (0.80–5.48) (0.91–7.06) (0.80–5.24) (0.73–5.17) (0.96–8.25) (1.42–21.2) (0.39–3.71) (0.53–2.51) (0.59–4.12) (0.26–1.86) (0.41–2.84) (0.78–6.28) (0.57–4.48) (0.63–5.71) 0.48 (0.38) 0.88 0.50 (0.39) 0.85 0.49 (0.39) 0.87 0.55 (0.40) 0.86 0.46 (0.38) 0.88 0.46 (0.38) 0.88 0.48 (0.38) 0.87 41 (67.2) 39 (62.9) 24 (82.8) 30 (68.2) 53 (68.8) 53 (62.4) 12 (92.3) 16 (76.2) 37 (77.1) 38 (74.5) 28 (66.7) 31 (56.4) 42 (70.0) 35 (71.4) 23 (76.7) 34 (59.6) 14 (82.4) 18 (69.2) 51 (69.9) 51 (63.8) 30 (68.2) 29 (54.7) 35 (76.1) 40 (75.5) 40 (76.9) 43 (70.5) 25 (65.8) 26 (57.8) 0.85 0.99 0.44 1.00 1.67 2.28 0.44 1.00 0.42 0.40 0.49 1.00 0.44 0.36 0.32 1.00 0.44 0.62 0.70 1.00 1.56 3.42 1.15 1.00 0.38 0.42 0.51 1.00 (0.29–2.53) (0.36–2.78) (0.12–1.61) (0.43–6.39) (0.63–8.20) (0.04–5.01) (0.15–1.17) (0.16–1.02) (0.18–1.30) (0.17–1.13) (0.14–0.95) (0.10–1.01) (0.09–2.08) (0.19–1.96) (0.32–1.53) (0.54–4.48) (1.29–9.10) (0.41–3.22) (0.13–1.09) (0.16–1.14) (0.18–1.47) 0.41 (0.39) 0.87 0.39 (0.39) 0.87 0.38 (0.38) 0.88 0.50 (0.42) 0.83 0.39 (0.38) 0.87 0.38 (0.38) 0.87 0.43 (0.40) 0.86 Int. J. Environ. Res. Public Health 2019, 16, 1557 9 of 15 Model Covariate Food Access Model 3: Food Consumption (Ref: Poor) Model 4: Food Insecurity (Ref: Mild to Moderate) Ref: N (%) OR (95% CI) Between Shehia Variance (SE) χ2/DF Ref: N (%) OR (95%CI) Between Shehia Variance (SE) χ2/DF Table 5. Cont. h Number of Vehicles At least one At least one None None a Two categories for marital status of HH were used (1 = not married (single, widow, divorce, cohabitation) 2 = married (monogamous or polygamous)). Good access Poor access Good access Poor access 48 (60.8) 49 (66.2) 5 (45.5) 26 (81.3) 56 (70.9) 50 (67.6 9 (81.8) 19 (59.4) (0.68–6.87) (0.59–5.71) (1.20–32.3) (0.18–1.43) (0.24–1.69) (0.05–2.06) 2.17 1.83 6.21 1.00 0.51 0.63 0.31 1.00 0.44 (0.38) 0.89 0.38 (0.38) 0.86 Int. J. Environ. Res. Public Health 2019, 16, 1557 10 of 15 4. Discussion This exploratory study aimed at adding to the on-going debate on how good FA may help to improve the nutrition situation of the Zanzibar population. We observed that poor FA, poor food consumption, and food insecurity are a problem in many Zanzibari households. Poor FA was more prevalent in households with poor food consumption and severe food insecurity. In particular, polygamous households and larger households had a higher chance of severe food insecurity. Good FA increased the chance of acceptable food consumption for urban households and households with no vehicle, whereas poor FA increased the chance for severe food insecurity for larger households. 4.1. Proportions of FCS and FIES in the Study Population The proportion of households with acceptable food consumption was about 35% in the present study. This is lower than reported in Ethiopia (73%) [29] and in the Nyarugusa refugee camp in Tanzania (86%) with a sample of 343 households in a WFP/United Nations High Commissioner for Refugees (UNHCR) study [30]. Food consumption was assessed using comparable instruments in the present study and in the UN World Food Programme Study [23]; thus, the higher proportions in the Nyarugusa camp could have been due to the fact that 83% of the households’ main source of food was from food aid, unlike in Unguja Island, where households relied mostly on purchase and on their own production. The overall proportion of mild to moderate food insecurity in our study population was about 68%, slightly higher than the 49% observed in Burkina Faso among 330 surveyed households [31], and doubles that in Ethiopia (34%) [32]. This survey was conducted during the short-rainy season (October–December 2013), which is a time for sowing and growing of food products and when most household stocks are depleted [33]. This may have contributed to the relatively low proportions of households with acceptable food consumption. Still, the proportion of households with experienced mild to moderate food insecurity was higher compared to other household surveys, which were also conducted during the lean seasons [29,31,32,34]. More than three quarters of the households in our study resided in the rural area and hence depended mostly on own food production, and food products grown in the rainy season are harvested between January and February. A survey conducted in Burkina Faso conﬁrmed the seasonal eﬀects on food security of 1056 households with a lower food security during the lean season compared to the post-harvest season [34]. Hence, data collection should also be conducted during the post-harvest season in order to gain insight into FCS and FIES when food availability improves on Unguja Island. 4.2. Role of Food Access on the Correlates, Food Consumption and Food Insecurity In the present study, higher education level of the HH seemed to inﬂuence the level of food consumption and food insecurity of the households with poor FA, agreeing with ﬁndings from Rwanda [35] and Uganda [36], reﬂecting that “some level of education is important to household food security”. It is assumed that a literate HH has a greater capacity of adapting to improved technologies and coping strategies, thus increasing production/food supply in the household [16]. The lower chance for severe food insecurity in households with higher-educated HH and good FA, higher-educated HH and poor FA, and lower-educated HH and good FA indicated that education level of the HH as well as FA both played a comparable role in impacting the food insecurity status of the household. The fact that in this study, larger households had higher chances for severe food insecurity than smaller households may be explained by the fact that smaller household sizes are generally better manageable in terms of food demand and supply. The latter can be improved through own production, which increases food consumption and decreases food insecurity as also reported in other studies [18,37]. In contrast to our ﬁndings, studies from Niger and Nigeria observed a decreasing likelihood of a household being food-insecure with an increasing household size [38]. However, to estimate the eﬀect of household size on food insecurity is challenging, as this would require verifying numerous other factors such as the number of all active members in the household (contributors of Int. J. Environ. Res. Public Health 2019, 16, 1557 11 of 15 income or food), income sources (salary), and expenditure on food. While the number of jobs of the HH showed no eﬀect on food insecurity, wealthier households, on the other hand, revealed a lower chance of having severe food insecurity compared to poor households. Interestingly, our results showed that larger households with poor FA had a higher chance for severe food insecurity compared to smaller households, while those with good FA only had a smaller chance for severe food insecurity compared to small households with poor FA, this further conﬁrms the role FA played in impacting food insecurity in the study population. Studies investigating the role of FA on food consumption and food insecurity in Sub-Saharan Africa are scarce, and more research is advisable. In our study, households with polygamous HH had a higher chance for of severe food insecurity than households with not married (e.g., widowed or divorced) HH. While this is in line with previous ﬁndings from Tanzania mainland [39], other studies reported contrasting ﬁndings [40,41]. The latter postulated that the large number of individuals in polygamous households means that a great number of individuals can provide ﬁnancial and labor support among each other, thus reducing the chance of having food insecurity. However, these studies compared polygamous households against monogamous households. When we compared monogamous against unmarried households, we observed a relatively lower chance of experiencing severe food insecurity. This is supporting our ﬁndings that larger households were more likely to be food-insecure compared to smaller households. The ﬁnding that urban households with good FA experienced a higher chance of acceptable food consumption than rural households with poor FA may be explained by the good infrastructure in urban areas, which enables good accessibility for foods and may potentially enhance the food diversity of households. The good infrastructure includes factors such as the quality of roads or market density, which facilitate food distribution and transport into the communities. Studies in Malawi and Kenya [42,43] also reported that food insecurity for rural households was aﬀected by long distance to the market and poor market access. Furthermore, in Malawi, the distance to the market was reported to aﬀect food consumption for both rural and urban households [42]. The fact that in our study, even households without a vehicle had a higher chance of acceptable food consumption when they had good FA compared to those with poor FA indicates the importance of other aspects of the FA—in addition to infrastructure—such as borrowing foods, receiving foods as a gift, and own food production or animal rearing. The interplay of multiple factors facilitating food accessibility should be considered in future intervention studies. 4.3. Strengths and Limitations This exploratory study provides valuable data on the interplay of food access, socioeconomic and demographic correlates with food consumption and food insecurity of Zanzibari households. Even though our study was conducted during the short-rainy season, which made it diﬃcult to have access to some of the villages, an important strength was the overall high proportion of households that participated in the study. Further, the highly standardized study protocol using partly validated and pretested methods and instruments is a clear strength of the study. When comparing the study characteristics such as HH demography (gender, marital status, education level, occupation), household demography (rural/urban, household size), and the socioeconomic status of the household of the full survey sample and the sample presented in this study sample, no substantial diﬀerences were observed (results not shown). Thus, a selection bias can be ruled out. The study, however, has limitations. Firstly, the FIES was explicitly developed for cross-cultural comparability and assesses correlates of food insecurity across diﬀerent areas that have the same climatic or agricultural calendar [44]. We, however, conducted our study only in Unguja Island and hence have no basis for a comparison with other Zanzibari islands such as Pemba Islands, or with the population of Mainland Tanzania. As household information was based on self-reports, social desirability could have inﬂuenced the responses given. The survey was conducted during October–December, which is the time of the year when most household stocks are depleted (lean season) [33]; this may have aﬀected our results on the food consumption and food insecurity situation Int. J. Environ. Res. Public Health 2019, 16, 1557 12 of 15 of the households and must be acknowledged as a limitation. Further, the collection of cross-sectional data means that eﬀects of seasonal variations could not be investigated. To overcome this limitation, it would be advisable to collect longitudinal data. One major limitation we like to address is that the overall study was planned, powered, and conducted to estimate the prevalence of malnutrition in the Zanzibari population [22]. However, with this study, we intended to explore important household survey data on a broader level even though for this particular approach, the sample size was underpowered. Nevertheless, the data are a useful source for exploring the role of FA in the association between sociodemographic household factors with food consumption and food insecurity. Findings from this study will add knowledge and inform the development of intervention strategies and policies aiming at improving food consumption and food security in Zanzibar. Still, more research with a larger sample size is advisable. 5. Conclusions Based on our ﬁndings, poor access to food may be seen as a modiﬁable factor for food consumption and perceived food insecurity in Zanzibari households, in particular for the association with educational level and household size. To improve food and nutrition security in Zanzibar, implementation of policies and programs that address education activities and diﬀerent forms of practical coping strategies, such as eﬃcient food storage techniques and home gardening, in their agendas are needed, particularly in rural areas. In parallel, strategies should consider improvement of infrastructure to facilitate distribution of produce within the rural–urban areas, as well as education campaigns on food quality and utilization, emphasizing on the importance of food group and balanced diets. Ethics Approval and Consent to Participate: Ethical approval was obtained from the Ethics Committees of the University of Bremen in Germany with a reference number 06-3 and of the Zanzibar Ministry of Health and the Zanzibar Medical Research and Ethics Committee in Zanzibar, Tanzania with a reference number ZAMREC/0001/AUGUST/013. Written informed consents were taken from all participants and parents/guardians gave a written informed consent for their children. The consent forms were approved by the Institutional Ethics Committee. Author Contributions: This manuscript represents original work that has not been published previously and is currently not considered by another journal. The authors’ responsibilities were as follows: A.H. and M.A.N. had the idea of the analysis; A.H., M.A.N., S.K., and M.S. were responsible for data collection. M.A.N. and C.B. conducted statistical analyses; M.A.N. and C.B. did the analysis and data interpretation; M.N. wrote the manuscript and had primary responsibility for ﬁnal content and submitting the manuscript for publication; M.A.N., S.K., M.S., C.B., and A.H. were responsible for critical revisions and ﬁnal approval of the manuscript. Funding: This research was funded by Leibniz-Gemeinschaft: SAW-2012-ZMT-4. Acknowledgments: This work was done as part of the Leibniz Graduate School SUTAS (Sustainable Use of Tropical Aquatic Systems; http://www.zmt-bremen.de/SUTAS.html). This study would not have been possible without the voluntary collaboration of the Zanzibari families who participated in the extensive examinations. We are grateful for the support from regional and local community leaders and municipalities. The authors gratefully acknowledge the assistance from all the ﬁeldworkers. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations FA FANTA FAO FCS FIES GLIMMIX HH ISCED SNNPR SSA Food Access Food and Nutrition Technical Assistance Food and Agriculture Organization Food Consumption Score Food Insecurity Experience Scale Generalized Linear Mixed Models Head of Household International Standard Classiﬁcation of Education South Nation’s, Nationalities and Peoples Region Sub-Saharan Africa Int. J. Environ. Res. Public Health 2019, 16, 1557 13 of 15 SUTAS TDHS UNHCR WFP WHO References Sustainable Use of Tropical Aquatic Systems Tanzanian Demographic and Health Survey United Nations High Commissioner for Refugees World Food Programme World Health Organization 1. 2. United Nations (UN). United Nations Population Prospects. The 2007 Revision Population Database; Department of Economic and Social Aﬀairs Population Division: New York, NY, USA, 2008; p. 244. United Nations World Food Programme (UNWFP). Comprehensive Food Security and Vulnerability Analysis (CFSVA) and Nutrition Assessment 2010. Kenya High Density Urban Areas; United Nations World Food Programme Headquarters: Rome, Italy, 2012; p. 91. 3. World Bank. Main Report: Tanzania Mainland Poverty Assessment; World Bank: Washington, DC, USA, 2015; 4. 5. 6. 7. 8. 9. p. 180. Kimani-Murage, E.W.; Schoﬁeld, L.; Wekesah, F.; Mohamed, S.; Mberu, B.; Ettarh, R.; Egondi, T.; Kyobutungi, C.; Ezeh, A. Vulnerability to food insecurity in Urban slums: Experiences from Nairobi, Kenya. J. Urban Health 2014, 91, 1098–1113. [CrossRef] Leduka, R.; Crush, J.; Frayne, B.; Mccordic, C.; Matobo, T.; Makoa, T.E.; Mphale, M.; Phaila, M.; Letsie, M. The State of Poverty and Food Insecurity in Maseru, Lesotho. No.21 ed.; African Food Security Urban Network (AFSUN): Cape Town South Africa, 2015; pp. 1–79. Gustafson, D.J. Rising food costs & global food security: Key issues & relevance for India. Indian J. Med. Res. 2013, 138, 398–410. The Revolutionary Government of Zanzibar (RGoZ). Zanzibar food security & nutrition situational analysis; Ministry of Agriculture, Livestock and Environment and Ministry of Health Social Welfare: Unguja, Zanzibar, 2006. The United Republic of Tanzania. Comprehensive Food Security and Nutrition Assessment Report. Available online: www.ipcinfo.org/ﬁleadmin/user_upload/ipcinfo/docs/IPC_Tanzania_AFI_Situation_2018Feb.pdf (accessed on 1 February 2017). Revolutionary Government of Zanzibar (RGoZ). The Zanzibar Strategy for Growth and Reduction of Poverty; The Revolutionary Government of Zanzibar: Unguja, Zanzibar, March 2007. 10. Musotsi, A.A.; Sigot, A.J.; Onyango, M.O.A. The role of home gardening in household food security in Butere division of Western Kenya. Afr. J. Food Agric. Nutr. Dev. 2008, 8, 4. [CrossRef] 11. Koyenikan, M.J. Perception of home garden potentials among women in Edo South ecological zone, Nigeria. Gend. Behav. 2007, 5, 1042–1052. [CrossRef] 12. Obayelu, A. Comparative analysis of households’ socioeconomic and demographic characteristics and food security status in urban and rural areas of Kwara and Kogi states of North-central Nigeria. Afr. J. Food Agric. Nutr. Dev. 2012, 12, 6027–6054. 13. Kebede, M. The gender perspective of household food security in Meskan district of the Gurage zone, Southern Ethiopia. Afr. Res. Rev. 2009, 3. [CrossRef] 14. Masinde, G.V. Food security coping strategies in female and male headed households in Kenyan slums: 15. The case of Kawangware, Nairobi. Int. J. Soc. Sci. Entrep. 2014, 1, 36–54. Food and Agriculture Organization of the United Nations (FAO). Women Play a Decisive Role in Household Food Security, Dietary Diversity and Children’s Health. Available online: http://www.fao.org/gender/gender- home/gender-programme/gender-food/en/ (accessed on 1 July 2016). 16. Olayemi, A.O. Eﬀects of family size on household food security in Osun State, Nigeria. Asian J. Agric. Rural Dev. 2012, 2, 136. 17. Olson, C.M.; Rauschenbach, B.S.; Frongillo, E.A., Jr.; Kendall, A. Factors Contributing to Housefold Food Insecurity in a Rural Upstate New York County. Available online: https://www.irp.wisc.edu/publications/ dps/pdfs/dp110796.pdf (accessed on 1 September 1996). 18. Aidoo, R.; Mensah, J.O.; Tuﬀour, T. Determinants of household food security in the Sekyere-Afram plains district of Ghana. Eur. Sci. J. 2013. [CrossRef] Int. J. Environ. Res. Public Health 2019, 16, 1557 14 of 15 19. Bogale, A.; Shimelis, A. Household level determinants of food insecurity in rural areas of Dire Dawa, Eastern Ethiopia. Afr. J. Food Agric. Nutr. Dev. 2009, 9, 1914–1926. 20. Ali Naser, I.; Jalil, R.; Muda, W.; Manan, W.; Nik, W.; Suriati, W.; Mohd Shariﬀ, Z.; Abdullah, M.R. Association between household food insecurity and nutritional outcomes among children in Northeastern of peninsular Malaysia. Nutr. Res. Pract. 2014, 8, 304–311. [CrossRef] [PubMed] 21. Beaman, L.; Dillon, A. Do Household Deﬁnitions Matter Available online: http://www.fao.org/ﬁleadmin/templates/ess/documents/meetings_and_workshops/ICAS5/PDF/ ICASV_1.2_109_Paper_Beaman.pdf (accessed on 14 July 2016). in Survey Design. 22. Nyangasa, M.A.; Kelm, S.; Sheikh, M.A.; Hebestreit, A. Design, response rates, and population characteristics of a cross-sectional study in Zanzibar, Tanzania. JMIR Res. Protoc. 2016, 5, e235. [CrossRef] [PubMed] 23. United Nations World Food Programme (WFP). Technical Guidance Sheet—Food Consumption Analysis: Calculation and Use of the Food Consumption Score in Food Security Analysis. Available online: https://www.wfp.org/content/technical-guidance-sheet-food-consumption-analysis-calculation-and- use-food-consumption-score-food-s (accessed on 1 February 2018). Food and Agriculture Organization of the United Nations (FAO). The Food Insecurity Experience Scale—Development of a Global Standard for Monitoring Hunger Worldwide. Available online: http: //www.fao.org/ﬁleadmin/templates/ess/voh/FIES_Technical_Paper_v1.1.pdf (accessed on 1 October 2013). 25. Leroy, J.L.; Ruel, M.; Frongillo, E.A.; Harris, J.; Ballard, T.J. Measuring the food access dimension of food security: A critical review and mapping of indicators. Food Nutr. Bull. 2015, 36, 167–195. [CrossRef] 26. Caﬁero, C.; Viviani, S.; Nord, M. Food security measurement in a global context: The food insecurity 24. experience scale. Measurement 2018, 116, 146–152. [CrossRef] 27. Maire, P.; Hatzinger, R. Extended rasch modeling: The erm package for the application of IRT models in R. J. Stat. Softw. 2007, 20, 1–20. 28. United Nations Educational Scientiﬁc and Cultural Organization (UNESCO). International Standard Classiﬁcation of Education; UNESCO Institute for Statistics: Montreal, QC, Canada, 2011. 29. Ethiopia—Comprehensive Food Security and Vulnerability Analysis (CFSVA). Available online: https: //www.wfp.org/content/ethiopia-comprehensive-food-security-and-vulnerability-analysis-2014 (accessed on 25 February 2019). 30. Tanzania—Community and Household Surveillance in North Western Tanzania: Programme Outcome Monitoring in Nyarugusu Refugee Camp. Available online: https://www.wfp.org/content/tanzania- community-and-household-surveillance-north-western-june-2011 (accessed on 25 February 2019). 31. Melgar-Quinonez, H.R.; Zubieta, A.C.; MkNelly, B.; Nteziyaremye, A.; Gerardo, M.F.; Dunford, C. Household food insecurity and food expenditure in Bolivia, Burkina Faso, and the Philippines. J. Nutr. 2006, 136, 1431S–1437S. [CrossRef] 32. Tadesse Tantu, A.; Demissie Gamebo, T.; Kuma Sheno, B.; Yohannis Kabalo, M. Household food insecurity and associated factors among households in Wolaita Sodo Town, 2015. Agric. Food Secur. 2017, 6, 19. [CrossRef] Food and Agriculture Organization of the United Nations (FAO). Food security snapshot. United Republic of Tanzania. Available online: http://www.fao.org/giews/countrybrief/country.jsp?code=TZA (accessed on 11 April 2019). 33. 34. Becquey, E.; Delpeuch, F.; Konate, A.M.; Delsol, H.; Lange, M.; Zoungrana, M.; Martin-Prevel, Y. Seasonality of the dietary dimension of household food security in Urban Burkina Faso. Br. J. Nutr. 2012, 107, 1860–1870. [CrossRef] [PubMed] 35. Habyarimana, J.B. Determinants of household food insecurity in developing countries evidences from a probit model for the case of rural households in Rwanda. Sustain. Agric. Res. 2015, 4, 78. [CrossRef] 36. Bahiigwa, G. Household food security in Uganda: An empirical analysis; Economic Policy Research Center: Kampala, Uganda, 1999. 37. Deaton, A.; Paxson, C. Economies of scale, household size, and the demand for food. J. Polit. Econ. 1998, 106, 897–930. [CrossRef] 38. Zakari, S.; Ying, L.; Song, B. Factors inﬂuencing household food security in West Africa: The case of Southern Niger. Sustainability 2014, 6, 1191–1202. [CrossRef] Int. J. Environ. Res. Public Health 2019, 16, 1557 15 of 15 39. Lawson, D.W.; James, S.; Ngadaya, E.; Ngowi, B.; Mﬁnanga, S.G.; Borgerhoﬀ Mulder, M. No evidence that polygynous marriage is a harmful cultural practice in Northern Tanzania. Proc. Natl. Acad. Sci. USA 2015, 112, 13827–13832. [CrossRef] [PubMed] 40. Owoo, N.S. Food insecurity and family structure in Nigeria. SSM Popul. Health 2018, 4, 117–125. [CrossRef] [PubMed] 41. Meludu, N.T.; Iﬁe, P.A.; Akinbile, L.A.; Adekoya, E.A. The role of women in sustainable food security in Nigeria: A case of Udu local government area of Delta state. J. Sustain. Agric. 1999, 15, 87–97. [CrossRef] 42. Tembo, D.; Simtowe, F. The Eﬀects of Market Accessibility on Household Food Security: Evidence from Malawi. In Proceedings of the Conference on International Research on Food Security, Natural Resource Management and Rural Development, Hamburg, Germany, 6–8 October 2009; German Institute for Agriculture in the Tropics and Subtropics: Witzenhausen, Germany. 43. Ndegwa, P.B. Assessment of Factors Inﬂuencing Food Security in Wenje Division, Tana River County-Kenya. 44. Food Sci. Qual. Manag. 2015, 44. Food and Agriculture Organization of the United Nations (FAO). Methods for Estimating Comparable Rates of Food Insecurity Experienced by Adults throughout the World; Food and Agriculture Organization (FAO): Rome, Italy, 2016. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_ijms241210379	Article Comparative Evaluation of Four Commercially Available Immunoassays for Therapeutic Drug Monitoring of Inﬂiximab and Adalimumab Florian Rissel 1, Yoann Cazaubon 2 Thierry Vincent 1 and Alexandre Jentzer 1,* , Syrine Saffar 1, Romain Altwegg 3, Mélanie Artasone 1, Claire Lozano 1, 1 Department of Immunology, Saint Eloi, Montpellier University Hospital, Montpellier University, 2 34295 Montpellier, France; [email protected] (F.R.) Institute Desbrest of Epidemiology and Public Health, Institut National de la Santé et de la Recherche Médicale (INSERM), Department of Pharmacology and Toxicology, Montpellier University Hospital, Montpellier University, 34090 Montpellier, France 3 Department of Hepato-Gastroenterology, Saint Eloi, Montpellier University Hospital, Montpellier University, 34295 Montpellier, France * Correspondence: [email protected] Abstract: Therapeutic drug monitoring (TDM) of anti-TNF-α is an important tool in clinical practice for inﬂammatory diseases. In this study, we have evaluated the performance of several assays for drug and antidrug antibodies (ADA) measurement in the serum. 50 sera from patients treated with inﬂiximab (IFX) and 49 sera from patients treated with adalimumab (ADAL) were monitored with four immunoassays. We have compared Promonitor, i-Track10®, and ez-track1 assays to our gold standard Lisa Tracker® ELISA using Cohen’s kappa, Passing-Bablok, and Bland–Altman analysis. The qualitative analysis evaluated by Cohen’s kappa values found for IFX measurements an “almost perfect” concordance for Promonitor, “moderate” for i-Track10® and “substantial” for ez-Track1. For ADAL, kappa values were “moderate” for all tested methods. For anti-IFX, kappa values were “almost perfect” for Promonitor, “fair” for i-Track10®, and “substantial” for ez-Track1. For anti-ADAL, kappa values were “almost perfect” for all three assays. For quantitative analysis of drug measurements, Pearson’s r values were all above 0.9 and Lin’s concordance coefﬁcients of all immunoassays were around 0.80. Performances of the four evaluated immunoassays were acceptable for TDM based on our laboratory experience. Nevertheless, concordance between the four methods for IFX measurement was not perfect and we recommend the use of the same assay for the follow-up of a given patient. The performances of the four immunoassays evaluated were similar and are acceptable for TDM based on our laboratory experience. Keywords: adalimumab; inﬂiximab; antidrug antibodies; therapeutic drug monitoring; immunoas- says comparison 1. Introduction Tumor necrosis factor alpha (TNF-α) is a cytokine with pleiotropic effects initially recognized as a necrosis factor. TNF-α binds to the receptors TNFR1 and TNFR2 which initiate signal transduction pathways leading to inﬂammation and cell death [1]. Physio- logically, TNF-α is crucial for a normal immune response. However, the inappropriate or excessive production of TNF-α leads to inﬂammatory or autoimmune diseases [1]. There- fore, TNF-α has become the target of therapeutic monoclonal antibodies with the aim of blocking inﬂammatory dysregulation [2]. Thus, anti-TNF-α monoclonal antibodies are broadly used and are efﬁcient in particular for the treatment of rheumatoid arthritis (RA), inﬂammatory bowel disease (IBD), psoriasis, psoriatic arthritis, ankylosing spondylitis (AS), and juvenile idiopathic arthritis (JIA) [2]. In this article, we will focus more particularly Citation: Rissel, F.; Cazaubon, Y.; Saffar, S.; Altwegg, R.; Artasone, M.; Lozano, C.; Vincent, T.; Jentzer, A. Comparative Evaluation of Four Commercially Available Immunoassays for Therapeutic Drug Monitoring of Inﬂiximab and Adalimumab. Int. J. Mol. Sci. 2023, 24, 10379. https://doi.org/10.3390/ ijms241210379 Academic Editor: Jesús Cosín-Roger Received: 26 May 2023 Revised: 13 June 2023 Accepted: 16 June 2023 Published: 20 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Mol. Sci. 2023, 24, 10379. https://doi.org/10.3390/ijms241210379 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10379 2 of 10 on two anti-TNF-α biotherapies approved by the FDA (Food and Drug Administration) and the EMA (European Medicines Agency) in these pathologies, adalimumab (ADAL) and inﬂiximab (IFX), which are a fully human and a chimeric IgG1 monoclonal antibody respectively [2]. Despite good efﬁcacy, primary nonresponse, loss of response, and adverse events can be related to the development of antidrug antibodies (ADA) with reduced clinical response and an increased incidence of infusion reactions and injection-site reac- tions [3]. A large meta-analysis found frequent immunizations against these biotherapies with 25.3% of anti-IFX antibodies (95% CI 19.5–32.3) and 14.1% of anti-ADAL antibodies (95% CI 8.6–22.3) [4] that should be taken into account in current medical practice. In a prospective multicenter cohort study with 1650 anti-TNF-α naïve patients with active luminal Crohn’s disease (CD), the authors have concluded that anti-TNF-α treatment failure is common and can be predicted by low drug concentrations mediated in part by immunogenicity [5]. This multivariable analysis showed that the only factor independently associated with primary nonresponse at week 14 and nonremission at week 54 was low IFX or ADAL serum concentration. In detail, primary nonresponse occurred in 23.8% (95% CI 21.4–26.2) at week 14, nonremission at week 54 occurred in 63.1% (95% CI 60.3–65.8), and adverse events curtailed treatment in 7.8% (95% CI 6.6–9.2) [5]. The immunogenicity, which is precocious [5], might be mitigated by early dose optimization, minimizing the loss of response. Several studies have demonstrated a correlation between exposition and clinical outcomes [6,7]. One important aspect of optimizing the therapeutic response to these drugs is monitoring levels in the patient’s bloodstream, particularly the serum-trough level. Monitoring trough provides valuable insights into the drug’s pharmacokinetics, helping clinicians assess if the drug is being maintained in the therapeutic area. To better understand this time-varying concentration–effect relationship where the concentrations required at induction should be higher than during the maintenance phase [8], the TMDD (target-mediated drug disposition) is a speciﬁc PK/PD modeling approach that considers the interaction of a drug with its target molecule in the body. In the case of adalimumab and inﬂiximab, which are monoclonal antibodies targeting TNF-α, TMDD modeling has been employed to better understand their pharmacokinetic and pharmacodynamic prop- erties. TMDD models typically incorporate the following key components: free drug compartment inﬂuencing pharmacokinetic behavior, target compartment (TNF-α) leads to the pharmacological effects, internalization processes of the drug–target complex, and production/elimination of the TNF-α which can further impact the overall drug interaction. By utilizing these components, TMDD models could help in predicting drug concentrations, target occupancy, and pharmacological effects over time. They provide insights into the optimal dosing regimens, dosing intervals, and target saturation levels required to achieve the desired therapeutic outcomes [9]. This type of model could be useful to improve the use of therapeutic drug monitoring (TDM). Especially for IBD, two therapeutic TDM strategies exist: reactive TDM that should be used for all biologics for both primary nonresponse and secondary loss of response with therapeutic optimization guided by drug and ADA measurements and proactive TDM after induction and at least once during maintenance in order to achieve adequate serum concentrations of biological drugs [10]. Proactive TDM may also be used in de-escalating anti-TNF-α therapy in patients with clinical remission. In economic terms, TDM related to anti-TNF-α therapy seems to result in cost savings in both IBD and RA patients with no negative impact on efﬁcacy compared to routine IFX dose escalation without TDM [11,12]. Concerning proactive TDM, its implementation in clinical practice is controversial due to several randomized controlled trials that failed to demonstrate its beneﬁt compared with the conventional approach [13]. However, recently, Syversen et al. demonstrated the superiority of proactive TDM during the ﬁrst 3 years of maintenance dose with inﬂiximab for various inﬂammatory conditions including IBD compared to a conventional approach based on clinical and biological monitoring [14]. Finally, TDM based on clinical data regarding disease activity and especially estima- tion of pharmacokinetic parameters of the anti-TNF-α might be useful to guide clinical decision-making and to optimize the dose at the individual level [3]. In the aim of per- Int. J. Mol. Sci. 2023, 24, 10379 3 of 10 forming TDM, several methods have been developed to measure drugs and ADA lev- els, including enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), liquid-chromatography-based homogeneous mobility shift assay (HMSA), and chemilu- minescence immunoassay (CLIA). Although there is no gold-standard technique, ELISA represents the most commonly used assay for TDM of anti-TNF-α biotherapies and ADA measurements in clinical practice because it has the advantage of being relatively simple and inexpensive. However, ELISA requires several days to report the results for clinicians without random-access ﬂexibility, making it difﬁcult to obtain results in a timely fashion. Consequently, drug administration or therapeutic optimization based on drug concentra- tion or the presence of ADA will be delayed. On the other hand, RIA and HMSA are more expensive and reserved for specialized laboratories. Concerning drug measurement, the correlation between different methods has been shown to be relatively good [15]. With regard to ADA detection, some assays such as ﬁrst-generation ELISA are “drug sensitive”, meaning that they can only detect ADA in the absence of a detectable drug in the serum. Although all ADA assays are drug sensitive to some degree, as the majority of them rely on the capture of the ADA by drug, next-generation ELISA, as well as RIA and HMSA, are described as “drug tolerant” so that they can measure ADA even in the presence of a drug in the serum. Therefore, the former assays will detect only free ADA while the latter will measure total antidrug antibodies. Our study aims to evaluate the performance of several commercially available im- munoassays for drugs and ADA measurements with an ELISA (Promonitor assays, GRI- FOLS™), a chemiluminescence immunoassay (CLIA) (i-Track10®, THERADIAG™) and a point-of-care method (POC) based on a time-resolved ﬂuorescence (TRF) (ez-track1, THERADIAG™) in comparison with an ELISA (Lisa Tracker® (LT DS2), THERADIAG™), our gold standard used in the department of Immunology in the Montpellier Univer- sity Hospital. 2. Results 2.1. Immunoassays The immunoassays characteristics used in this study are summarized in Table 1. The measurement ranges for drugs and ADA are substantially equivalent for the four assays carried out in this study. Nevertheless, the use of arbitrary units (AU) for ADA measured by the ELISA Promonitor assays (GRIFOLS™) and ez-Track1 (THERADIAG™), and the capacity for i-Track10® (THERADIAG™) to measure ADA up to 2000 (µg/mL) without any dilution should be noted. The manufacturer’s instructions mentioned no interference with hemolysis, bilirubin, triglyceride, and rheumatoid factors for THERADIAGTM assays and no interference for rheumatoid factor for the GRIFOLSTM assay. Table 1. Characteristics of assays used for method comparison. Company Method Promonitor GRIFOLSTM LisaTracker® THERADIAGTM i-Track10® THERADIAGTM ez-track1 THERADIAGTM ELISA (Automated) ELISA (Automated) CLIA TRF IFX 0.3–14.4 (µg/mL) 0.3–20 (µg/mL) 0.3–24 (µg/mL) 0.2–50 (µg/mL) Measurement Range Anti-IFX 2–144 (AU/mL) 10–200 (ng/mL) 10–2000 (ng/mL) 4–250 (AU/mL) ADAL 0.25–12 (µg/mL) 0.3–20 (µg/mL) 0.5–24 (µg/mL) 0.2–50 (µg/mL) Anti-ADAL 6–400 (AU/mL) 10–160 (ng/mL) 10–2000 (ng/mL) 3–200 (AU/mL) ELISA: enzyme-linked immunoassay CLIA: chemiluminescence immunoassay; TFF: time-resolved ﬂuorescence; IFX: inﬂiximab; ADAL: adalimumab; AU: arbitrary unit. 2.2. Imprecision The mean concentrations for each parameter (10 values/parameter) (IFX, ADAL, and ADA) and the imprecision values of LT DS2, our gold standard used in routine, are represented in Table 2. The coefﬁcient of variation (CV) of the intrarun and inter-run Int. J. Mol. Sci. 2023, 24, 10379 4 of 10 imprecisions were acceptable for all the assays (CV < 20%) according to the Food and Drug Administration [16] (FDA) and the European Medicines Agency (EMA) [17] guidelines. Indeed, the intrarun imprecision CV is from 2.6% to 12.2% and the inter-run imprecision CV is from 3.3% to 12.6%. Table 2. Imprecision of LT DS2 with the use of low and high IMMUNO-TROL® IFX, ADAL, and ADA. IFX Anti-IFX ADAL Anti-ADAL Intrarun Inter-Run Intrarun Inter-Run Intrarun Inter-Run Intrarun Inter-Run (µg/mL) (CV%) (ng/mL) (CV%) (µg/mL) (CV%) (ng/mL) (CV%) Low High 4 (4.3) 9.6 (6.4) 3.3 (12.6) ND 41 (4.2) 129 (2.6) 39.9 (10.1) ND 3.3 (8.6) 11.5 (12.2) 3.5 (11.6) ND 29 (10.2) 111 (4.8) 37 (11) ND IFX: inﬂiximab; ADAL: adalimumab; ADA: antidrug antibodies; CV: coefﬁcient of variation; ND: not determined. 2.3. Qualitative Analysis between LT DS2 and Promonitor DS2/i-Track10®/ez-Track1 for IFX, ADAL, Anti-IFX, and Anti-ADAL Measurement The results of each immunoassay for drug measurement were stratified into three cate- gories according to the standard concentration range [6]: subtherapeutic (<5 µg/mL for IFX and <8 µg/mL for ADAL), maintenance therapeutic (5–10 µg/mL for IFX and 8–12 µg/mL for ADAL), and supratherapeutic (>10 µg/mL for IFX and >12 µg/mL for ADAL) (Table 3). For IFX, kappa values were “almost perfect” for Promonitor (kappa value = 0.904), “moder- ate” (kappa value = 0.565) for i-Track10, and “substantial” (kappa value = 0.752) for ez-Track1. For ADAL, kappa values were “moderate” for all tested methods: kappa value = 0.455 for Promonitor, kappa value = 0.517 for i-Track10, and kappa value = 0.401 for ez-Track1. The results of each immunoassay for antidrug antibodies measurements were stratified into two categories: <10 ng/mL and ≥10 ng/mL, assuming that UA/mL is similar to ng/mL (Table 2) with supplier recommendations defining positives antibodies as >10 UA/mL for ez-Track1 and anti-ADAL Promonitor and >5 UA/mL for anti-IFX Promonitor. For anti-IFX, kappa values were “almost perfect” for Promonitor (kappa value = 0.877), “fair” (kappa value = 0.345) for i-Track10, and “substantial” (kappa value = 0.788) for ez-Track1. For anti- ADAL, kappa values were “almost perfect” for the 3 assays: kappa value = 1 for Promonitor, kappa value = 0.936 for i-Track10, and kappa value = 1 for ez-Track1. Table 3. Data agreement between LT DS2 (gold standard) and Promonitor DS2/i-Track10®/ez-Track1 values for IFX, ADAL stratiﬁcations according to the therapeutic window in inﬂammatory bowel diseases during maintenance therapy and for anti-IFX, anti-ADAL stratiﬁcation according to the limit of quantiﬁcation (LOQ) of LT DS2 (10 ng/mL). IFX Promonitor DS2 LT DS2 aIFX LT DS2 (ng/mL) Promonitor DS2 (UA/mL) i-Track10 <5 µg/mL 5–10 µg/mL >10 µg/mL Total i-Track10 (ng/mL) ≥10 <10 Total ez-Track1 POC <5 µg/mL 5–10 µg/mL >10 µg/mL Total Kappa value 19 17 21 1 4 0 0 0 0 20 21 21 0.904 0.565 0.752 1 1 7 14 8 11 1 10 0 16 19 18 0 0 0 0 0 1 12 12 11 12 12 12 20 18 28 15 12 12 13 22 11 48 52 51 ez-Track1 POC (UA/mL) ≥10 <10 Total Kappa value 0 13 3 42 31 41 42 44 44 4 20 10 43 31 41 47 50 51 4 7 7 1 0 0 5 7 7 0.877 0.345 0.788 Int. J. Mol. Sci. 2023, 24, 10379 5 of 10 Table 3. Cont. ADAL Promonitor DS2 LT DS2 aADAL LT DS2 (ng/mL) Promonitor DS2 (UA/mL) i-Track10 <8 µg/mL 8–12 µg/mL >12 µg/mL Total i-Track10 (ng/mL) ≥10 <10 Total ez-Track1 POC <8 µg/mL 8–12 µg/mL >12 µg/mL Total Kappa value 26 24 21 7 11 11 0 0 0 33 35 32 0.455 0.517 0.401 0 0 1 9 7 4 2 4 6 11 11 11 0 0 0 5 0 0 2 7 7 7 7 7 26 24 22 21 18 15 4 11 13 51 53 50 ez-Track1 POC (UA/mL) ≥10 <10 Total Kappa value 0 1 0 42 43 41 42 44 41 8 10 8 42 43 41 50 53 49 8 9 8 0 0 0 8 9 8 1.000 0.936 1.000 2.4. Quantitative Analysis between LT DS2 and Promonitor DS2/i-TRACK10®/ez-Track1 for IFX, ADAL, Anti-IFX and Anti-ADAL Measurement A comparative analysis of clinical applicability between LT DS2 and Promonitor DS2/i- TRACK10®/ez-Track1 is shown in Figure 1 which reveals that only Promonitor was consistent with no constant or proportional deviation for IFX (Promonitor DS2 = 0.18 (95% CI, −0.014, 1.57) + 0.90 (95% CI, 0.72, 1.00) × LT DS2 (Figure 1A)) and for ADAL (Promonitor DS2 = 0.68 (95% CI, −0.97, 1.61) + 1.08 (95% CI, 0.93, 1.32) × LT DS2 (Figure 1D)). For i-Track10, there was a proportional deviation for IFX (i-Track10 = 0.019 (95% CI, −2.10, 0.84) + 1.31 (95% CI, 1.13, 1.60) × LT DS2 (Figure 1B)) and for ADAL (i-Track10 = 0.55 (95% CI, −0.22, 1.22) + 1.21 (95% CI, 1.11, 1.33) × LTDS2 (Figure 1E). For ez-Track1, there was a constant deviation for IFX (ez-Track1 = −0.66 (95% CI, −1.64, −0.13) + 0.92 (95% CI, 0.77, 1.05) x LTDS2 (Figure 1C)) and a proportional deviation for ADAL (ez-Track1 = 0.16 (95% CI, −1.10, 0.71) + 1.34 (95% CI, 1.18, 1.52) × LTDS2 (Figure 1F)). Concerning Lin’s concordance coefﬁcient, all coefﬁcients were above 0.80 except for the comparison concerning ADAL for ez-Track POC. A Lin’s concordance coefﬁcient greater than 0.8 was considered as excellent. Bland–Altman plots revealed mean differences between our “Gold Standard” (LT DS2) and the three other immunoassays tested (Figure 2). For Promonitor, IFX concentrations were quite similar but 0.45% higher (95% CI: −17.9%, 66.1%) compared with LT DS2, with three points outside the limit of agreement (95% CI: −2.7, 3.6; bias 0.45). ADAL concentrations were 18.8% lower on average (95% CI: −63.3, 22.2%), with two points outside the limit of agreement (95% CI: −3.2, 0.932; bias −1.1). For i-TRACK10®, IFX concentrations were on average 28.9% lower (95% CI: −71.4%, 13.7%) compared with LT DS2, with two points outside the limit of agreement (95% CI: −6.1, 1.8; bias −2.13). ADAL concentrations were 26.7% lower on average (95% CI: −49.1, 11.2%), with three points outside the limit of agreement (95% CI: −4.2, 0.04; bias −2.1). For ez-Track1, IFX concentrations were 24.1% higher on average (95% CI: −17.9%, 66.1%) compared with LT DS2, with two points outside the limit of agreement (95% CI: −6.1, 1.8; bias −2.13). ADAL concentrations were 28.7% lower on average (95% CI: −72.4, 21.9%), with two points outside the limit of agreement (95% CI: −5.6, 1.2; bias −2.2). For antidrug antibodies, Passing–Bablok, Lin’s concordance correlation coefﬁcient, and Bland–Altman are presented in Supplementary Figures S1 and S2. Interpretation is not possible due to the number of values available in the calibration range. Int. J. Mol. Sci. 2023, 24, 10379 6 of 10 Figure 1. Passing–Bablok regression ﬁt between LT DS2 and Promonitor DS2 (A,D)/i-TRACK10® (B,E)/ez-Track1 (C,F) immunoassays for IFX and ADAL. Pearson’s r-values are shown for all linear correlations. The solid blue lines indicate the Passing–Bablok regression. Red dashed lines are identity lines (y = x). Grey shade areas are the 95% conﬁdence bounds calculated with the bootstrap (quantile) method. Figure 2. Bland–Altman analysis verifies the difference in IFX/ADAL measurements between LT DS2 and Promonitor DS2 (A,D)/i-TRACK10® (B,E)/ez-Track1 (C,F). The difference between the two measurements is plotted on the y-axis and the average of the two measurements on the x-axis. Dashed blue lines represent the bias and dashed red lines the 95% limit of agreement (LOA) for each comparison. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 6 of 11 x LTDS2 (Figure 1C)) and a proportional deviation for ADAL (ez-Track1 = 0.16 (95% CI, −1.10, 0.71) + 1.34 (95% CI, 1.18, 1.52) × LTDS2 (Figure 1F)). Concerning Lin’s concordance coeﬃcient, all coeﬃcients were above 0.80 except for the comparison concerning ADAL for ez-Track POC. A Lin’s concordance coeﬃcient greater than 0.8 was considered as excellent. Bland–Altman plots revealed mean diﬀerences between our “Gold Standard” (LT DS2) and the three other immunoassays tested (Figure 2). For Promonitor, IFX concentra-tions were quite similar but 0.45% higher (95% CI: −17.9%, 66.1%) compared with LT DS2, with three points outside the limit of agreement (95% CI: −2.7, 3.6; bias 0.45). ADAL con-centrations were 18.8% lower on average (95% CI: −63.3, 22.2%), with two points outside the limit of agreement (95% CI: −3.2, 0.932; bias −1.1). For i-TRACK10®, IFX concentrations were on average 28.9% lower (95% CI: −71.4%, 13.7%) compared with LT DS2, with two points outside the limit of agreement (95% CI: −6.1, 1.8; bias −2.13). ADAL concentrations were 26.7% lower on average (95% CI: −49.1, 11.2%), with three points outside the limit of agreement (95% CI: −4.2, 0.04; bias −2.1). For ez-Track1, IFX concentrations were 24.1% higher on average (95% CI: −17.9%, 66.1%) compared with LT DS2, with two points out-side the limit of agreement (95% CI: −6.1, 1.8; bias −2.13). ADAL concentrations were 28.7% lower on average (95% CI: −72.4, 21.9%), with two points outside the limit of agreement (95% CI: −5.6, 1.2; bias −2.2) For antidrug antibodies, Passing–Bablok, Lin’s concordance correlation coeﬃcient, and Bland–Altman are presented in Supplementary Figures S1 and S2. Interpretation is not possible due to the number of values available in the calibration range. Figure 1. Passing–Bablok regression ﬁt between LT DS2 and Promonitor DS2 (A,D)/i-TRACK10® (B,E)/ez-Track1 (C,F) immunoassays for IFX and ADAL. Pearson’s r-values are shown for all linear correlations. The solid blue lines indicate the Passing–Bablok regression. Red dashed lines are iden-tity lines (y = x). Grey shade areas are the 95% conﬁdence bounds calculated with the bootstrap (quantile) method. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 7 of 11 Figure 2. Bland–Altman analysis veriﬁes the diﬀerence in IFX/ADAL measurements between LT DS2 and Promonitor DS2 (A,D)/i-TRACK10® (B,E)/ez-Track1 (C,F). The diﬀerence between the two measurements is plotted on the y-axis and the average of the two measurements on the x-axis. Dashed blue lines represent the bias and dashed red lines the 95% limit of agreement (LOA) for each comparison. 3. Discussion Our study evaluated the performance of ELISA (Promonitor assays, GRIFOLS™), CLIA (i-Track10®, THERADIAG™), and ez-track1 (THERADIAG™) in comparison with the ELISA LT DS2 (THERADIAG™) we used routinely in the Department of Immunology of the Montpellier University Hospital. This method has been internally validated and is periodically evaluated and validated by external quality controls. Currently, LT DS2 is the most used commercial kit in France. Thus, LT DS2 was used as the reference method for our study. For IFX measurements, the qualitative analysis evaluated by Cohen’s kappa values found in order of best to worst correlation: Promonitor, ez-Track1, and i-Track10 (0.904, 0.752, and 0.565, respectively). For ADAL, the three methods seemed equivalent (Cohen’s kappa values 0.455, 0.517, and 0.401, respectively) and the concordance with our gold standard was not satisfactory. In our study, 50 samples were analyzed. Nevertheless, ad-ditional data with measurements distributed over the entire dosage range could bring more robustness to our study. For the detection of ADA, anti-ADAL antibodies were well identiﬁed by the three assays but i-Track10 failed to correctly identify anti-IFX in sera, possibly due to a low number of immunizations. Promonitor has already been identiﬁed as an acceptable assay in drug [16] and ADA [17,18] TDM. Ez-Track1 had quite acceptable results even if this assay was not strictly used as a POC because sera were used in the laboratory instead of whole-blood samples in care units. A further validation on whole blood in care units should therefore be preferable even if the supplier’s recommendations indicate that both serum and whole blood can be used with this device. In our study, Int. J. Mol. Sci. 2023, 24, 10379 7 of 10 3. Discussion Our study evaluated the performance of ELISA (Promonitor assays, GRIFOLS™), CLIA (i-Track10®, THERADIAG™), and ez-track1 (THERADIAG™) in comparison with the ELISA LT DS2 (THERADIAG™) we used routinely in the Department of Immunology of the Montpellier University Hospital. This method has been internally validated and is periodically evaluated and validated by external quality controls. Currently, LT DS2 is the most used commercial kit in France. Thus, LT DS2 was used as the reference method for our study. For IFX measurements, the qualitative analysis evaluated by Cohen’s kappa values found in order of best to worst correlation: Promonitor, ez-Track1, and i-Track10 (0.904, 0.752, and 0.565, respectively). For ADAL, the three methods seemed equivalent (Cohen’s kappa values 0.455, 0.517, and 0.401, respectively) and the concordance with our gold standard was not satisfactory. In our study, 50 samples were analyzed. Nevertheless, additional data with measurements distributed over the entire dosage range could bring more robustness to our study. For the detection of ADA, anti-ADAL antibodies were well identiﬁed by the three assays but i-Track10 failed to correctly identify anti-IFX in sera, possibly due to a low number of immunizations. Promonitor has already been identiﬁed as an acceptable assay in drug [16] and ADA [17,18] TDM. Ez-Track1 had quite acceptable results even if this assay was not strictly used as a POC because sera were used in the laboratory instead of whole-blood samples in care units. A further validation on whole blood in care units should therefore be preferable even if the supplier’s recommendations indicate that both serum and whole blood can be used with this device. In our study, qualitative analysis of i-Track10® measurements provided a less strong correlation with LT DS2 for IFX, anti-IFX, and anti-ADA than previously shown [19]. This could be explained by a different choice of stratiﬁcation range. Indeed, we referred to the publication of Adam S Cheifetz et al. for the measurements of the drugs (IFX: 5–10 µg/mL; ADAL: 8–12 µg/mL) [6] rather than older guidelines (IFX: 3–7 µg/mL; ADAL: 5–8 µg/mL) [20,21]. For ADA detection, we focused on the threshold of positivity of our gold standard (10 ng/mL) which seemed the most sensitive and appropriate value to discriminate whether ADA were absent or present. Berger AE et al. used a threshold of 100 ng/mL in order to discriminate low afﬁnity or transient ADAs and high ADA values which would be associated with a decrease of circulating TNF-α and a loss of treatment efﬁcacy but with the risk to underestimate the presence of some ADAs. Nevertheless, the quantitative analysis indicated a good concordance between the three evaluated assays and our gold standard. Indeed, all serum drug quantiﬁcations (except ADAL measured by ez-Track1) had a concerning Lin’s concordance coefﬁcient above 0.80 which is considered by Altman as excellent. For the i-Track10, the Bland– Altman analysis revealed a tendency to overestimate the drug concentrations that was not found elsewhere [19]. For the ez-Track, we could observe the same tendency. The assay with the closest results to our gold standard remains Promonitor, likely because LT DS2 (THERADIAG™) and Promonitor (GRIFOLS™) are both ELISA. For the ADA, the quantitative analysis has been reported in the Supplementary Data due to different measurement ranges and units between the assays. The interpretation was not possible because of the small number of positive values related to the few numbers of immunized patients in our cohort. More data need to be collected to compare the different ADA assays. In our study, we used different methods to measure drugs and ADA in a large number of sera by ELISA, at the patient’s bedside in POC or with an automated method with random access for chemiluminescence. ELISA is a robust test to accurately quantify drugs and ADA with good sensitivity and speciﬁcity but requires working in series without ﬂexibility making it difﬁcult to obtain timely results for an efﬁcient and rapid therapeutic drug optimization in case of treatment failure. POC assays give results to clinicians one by one every 15 min and can be adapted to urgent requests but in small numbers. The mainte- nance of devices and the management of controls and calibrators in clinical departments Int. J. Mol. Sci. 2023, 24, 10379 8 of 10 are often problematic in the implementation of delocalized biology. However, low drug concentration and/or the presence of ADA could be characterized quickly at the time of the consultation and provide valuable information on the mechanism of therapeutic failure useful for a rapid treatment optimization. The automated CLIA is a random-access assay with an acceptable time to result for clinicians (about 35 min to obtain the ﬁrst result) and the technical handling time is less than that of ELISAs. To conclude, the four immunoassays evaluated seem acceptable for TDM in clinical practice based on our laboratory experience. Each assay has advantages that are technique dependent: ELISA makes it possible to measure a large number of sera in series, CLIA is a random-access instrument that decreases the time to results for clinicians, and ez- Track1 could rather be used as a POC method in healthcare centers without specialized immunology laboratories. Nevertheless, it is strongly recommended to systematically use the same assay method for the follow-up of a given patient. 4. Materials and Methods 4.1. Patients and Samples Samples were obtained from patients treated with IFX or ADAL in Montpellier Hos- pital or Nimes Hospital between September and November 2022. Blood samples were collected before anti-TNF-α administration. The sera were stored at 4 ◦C before immunoas- says which were performed within 3 weeks. A total of 50 sera from patients treated by IFX were monitored with the 4 immunoassays tested for IFX and anti-IFX and a total of 49 sera from patients treated by ADAL were monitored with the 4 immunoassays tested for ADAL and anti-ADAL. Low and High IMMUNO-TROL® IFX, ADAL, and anti-drug antibodies (THERADIAGTM, Croissy Beaubourg, France) were measured as patient samples and used for imprecision assays. 4.2. Immunoassay Method Four assays were compared in this study: Promonitor ELISA tests (GRIFOLS™, Barcelona, Spain) and Lisa-Tracker® ELISA (THERADIAG™) (LT DS2) using an auto- mated ELISA DS2® analyzer (Dynex Technologies, Chantilly, VA, USA); the chemilumines- cence immunoassay (CLIA) i-Track10 from THERADIAG™; and the POC method e-track1 from THERADIAG™, based on a time-resolved ﬂuorescence (TRF). The assays were per- formed in the Department of Immunology, Saint-Eloi Montpellier University Hospital. Lisa-Tracker® ELISA (THERADIAG™) was used as the gold standard. Internal quality controls were performed at each experiment as current practice in medical biology. The assay procedures were performed according to the manufacturer’s instructions and the speciﬁc protocol for each instrument. 4.3. Data Analysis For qualitative analysis, results obtained for LT DS2, Promonitor, i-Track10, and ez- Track1 were evaluated using Cohen’s kappa for agreement of each pair of assays (LT DS2 versus Tested assays). Cohen’s Kappa coefﬁcient was used to measure the level of agreement between two raters or judges who each classiﬁed items into predeﬁned categories. It reﬂects the concordance that is higher the closer its value is to 1. By following the classiﬁcation of Landis and Koch [22], the agreement interpretation of Kappa results is as follows: <0 poor, 0–0.2 slight, 0.21–0.40 fair, 0.41–0.60 moderate, 0.61–0.80 substantial, and 0.81–1,00 almost perfect. For quantitative analysis, Passing–Bablok regression, Lin’s concordance coefﬁcient, and Bland–Altmann were adopted to analyze the agreement between LT DS2 (gold stan- dard) and tested methods. The Passing–Bablok method is a nonparametric, robust method used for comparing and estimating the relationship between two continuous variables (analytical methods). It is particularly useful when analyzing data that may not satisfy the assumptions of traditional linear regression, such as when there are outliers, heteroscedasticity, or non- Int. J. Mol. Sci. 2023, 24, 10379 9 of 10 normal distribution of data. It does not assume a speciﬁc functional form for the relationship between variables. Instead, it estimates the line of best ﬁt by comparing the ranks of observations between the two variables. It provides estimates of the slope, intercept, and their conﬁdence intervals. Passing-Bablok regression is especially valuable when analyzing real-world data that may have characteristics that derogate linear regression assumptions. Lin’s concordance correlation coefﬁcient (ρc) is a measure of agreement or reliability between two sets of continuous measurements. The utility of Lin’s concordance correlation coefﬁcient lies in its ability to capture both the precision (Pearson’s correlation coefﬁcient) and accuracy (Cβ) of the agreement between the two sets of measurements. Cβ is a bias correction factor, a measure of how far a line of best ﬁt is from the identity line: y = x. It takes into account both the systematic differences (bias) and the dispersion (variability) between the measurements. Its interpretation is by no means set in stone. Altman considers the interpretation of ρc > 0.80 as excellent whereas McBride suggests the following ranges: <0.90 poor, 0.90–0.95: moderate, 0.95–0.99: substantial, and >0.99 almost perfect. Bland–Altman analysis is a statistical technique used to assess the agreement between two quantitative measurements. The Bland–Altman plot displays the difference between the two measurements (y-axis) against their mean (x-axis), representing the mean difference between the two measurements and the limits of agreement, which are calculated as the mean difference plus or minus two standard deviations of the differences. The agreement was sufﬁcient if the concentration difference was within ±1.96 SD of the mean concentration difference for ≥67% of the sample pairs [23]. Statistical analysis and graphs were performed using Rstudio (version 4.2.0, ggplot, mcr, DescTools, and blandr packages). 4.4. Ethics The Institutional Review Board (IRB) of Montpellier University Hospital has approved the study (IRB Accreditation number: 198711). The approval number assigned by the IRB was IRB-MTP_2022_07_202201163. Supplementary Materials: The supporting information can be downloaded at: https://www.mdpi. com/article/10.3390/ijms241210379/s1. Author Contributions: Conceptualization, Y.C., T.V. and A.J.; Data curation, F.R., Y.C., C.L., T.V. and A.J.; Formal analysis, Y.C. and A.J.; Funding acquisition, T.V. and A.J.; Investigation, F.R., S.S., R.A. and M.A.; Methodology, Y.C. and A.J.; Project administration, A.J.; Resources, T.V. and A.J.; Software, Y.C. and A.J.; Supervision, Y.C., T.V. and A.J.; Validation, Y.C., T.V. and A.J.; Visualization, A.J.; Writing—original draft, F.R., Y.C., R.A., C.L., T.V. and A.J.; Writing—review & editing, Y.C. and A.J. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: Institutional Review Board (IRB) of Montpellier University Hospital has approved the study (IRB Accreditation number: 198711). Informed Consent Statement: The approval Number assigned by the IRB was IRB-MTP_2022_07_ 202201163. Data Availability Statement: Data are available upon request. Acknowledgments: Assay kits were kindly provided by THERADIAG™ and GRIFOLSTM. Conﬂicts of Interest: The authors have no conﬂict of interest to declare. References 1. 2. 3. Jang, D.; Lee, A.-H.; Shin, H.-Y.; Song, H.-R.; Park, J.-H.; Kang, T.-B.; Lee, S.-R.; Yang, S.-H. The Role of Tumor Necrosis Factor Alpha (TNF-α) in Autoimmune Disease and Current TNF-α Inhibitors in Therapeutics. Int. J. Mol. Sci. 2021, 22, 2719. [CrossRef] Kalliolias, G.D.; Ivashkiv, L.B. TNF Biology, Pathogenic Mechanisms and Emerging Therapeutic Strategies. Nat. Rev. Rheumatol. 2016, 12, 49–62. [CrossRef] [PubMed] Atiqi, S.; Hooijberg, F.; Loeff, F.C.; Rispens, T.; Wolbink, G.J. Immunogenicity of TNF-Inhibitors. Front. Immunol. 2020, 11, 312. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10379 10 of 10 4. 5. 6. 7. 8. 9. Thomas, S.S.; Borazan, N.; Barroso, N.; Duan, L.; Taroumian, S.; Kretzmann, B.; Bardales, R.; Elashoff, D.; Vangala, S.; Furst, D.E. Comparative Immunogenicity of TNF Inhibitors: Impact on Clinical Efﬁcacy and Tolerability in the Management of Autoimmune Diseases. A Systematic Review and Meta-Analysis. BioDrugs 2015, 29, 241–258. [CrossRef] [PubMed] Kennedy, N.A.; Heap, G.A.; Green, H.D.; Hamilton, B.; Bewshea, C.; Walker, G.J.; Thomas, A.; Nice, R.; Perry, M.H.; Bouri, S.; et al. Predictors of Anti-TNF Treatment Failure in Anti-TNF-Naive Patients with Active Luminal Crohn’s Disease: A Prospective, Multicentre, Cohort Study. Lancet Gastroenterol. Hepatol. 2019, 4, 341–353. [CrossRef] [PubMed] Yanai, H.; Lichtenstein, L.; Assa, A.; Mazor, Y.; Weiss, B.; Levine, A.; Ron, Y.; Kopylov, U.; Bujanover, Y.; Rosenbach, Y.; et al. Levels of Drug and Antidrug Antibodies Are Associated with Outcome of Interventions after Loss of Response to Inﬂiximab or Adalimumab. Clin. Gastroenterol. Hepatol. 2015, 13, 522–530.e2. [CrossRef] Kelly, O.B.; Donnell, S.O.; Stempak, J.M.; Steinhart, A.H.; Silverberg, M.S. Therapeutic Drug Monitoring to Guide Inﬂiximab Dose Adjustment Is Associated with Better Endoscopic Outcomes than Clinical Decision Making Alone in Active Inﬂammatory Bowel Disease. Inﬂamm. Bowel Dis. 2017, 23, 1202–1209. [CrossRef] [PubMed] Amiot, A.; Hulin, A.; Belhassan, M.; Andre, C.; Gagniere, C.; Le Baleur, Y.; Farcet, J.-P.; Delchier, J.-C.; Hüe, S. Therapeutic Drug Monitoring Is Predictive of Loss of Response after De-Escalation of Inﬂiximab Therapy in Patients with Inﬂammatory Bowel Disease in Clinical Remission. Clin. Res. Hepatol. Gastroenterol. 2016, 40, 90–98. [CrossRef] Ternant, D.; Le Tilly, O.; Picon, L.; Moussata, D.; Passot, C.; Bejan-Angoulvant, T.; Desvignes, C.; Mulleman, D.; Goupille, P.; Paintaud, G. Inﬂiximab Efﬁcacy May Be Linked to Full TNF-α Blockade in Peripheral Compartment—A Double Central- Peripheral Target-Mediated Drug Disposition (TMDD) Model. Pharmaceutics 2021, 13, 1821. [CrossRef] 11. 10. Cheifetz, A.S.; Abreu, M.T.; Aﬁf, W.; Cross, R.K.; Dubinsky, M.C.; Loftus, E.V.; Osterman, M.T.; Sarouﬁm, A.; Siegel, C.A.; Yarur, A.J.; et al. A Comprehensive Literature Review and Expert Consensus Statement on Therapeutic Drug Monitoring of Biologics in Inﬂammatory Bowel Disease. Am. J. Gastroenterol. 2021, 116, 2014–2025. [CrossRef] Steenholdt, C.; Brynskov, J.; Thomsen, O.Ø.; Munck, L.K.; Fallingborg, J.; Christensen, L.A.; Pedersen, G.; Kjeldsen, J.; Jacobsen, B.A.; Oxholm, A.S.; et al. Individualised Therapy Is More Cost-Effective than Dose Intensiﬁcation in Patients with Crohn’s Disease Who Lose Response to Anti-TNF Treatment: A Randomised, Controlled Trial. Gut 2014, 63, 919–927. [CrossRef] [PubMed] 12. Martelli, L.; Olivera, P.; Roblin, X.; Attar, A.; Peyrin-Biroulet, L. Cost-Effectiveness of Drug Monitoring of Anti-TNF Therapy in Inﬂammatory Bowel Disease and Rheumatoid Arthritis: A Systematic Review. J. Gastroenterol. 2017, 52, 19–25. [CrossRef] Feuerstein, J.D.; Nguyen, G.C.; Kupfer, S.S.; Falck-Ytter, Y.; Singh, S. American Gastroenterological Association Institute Clinical Guidelines Committee American Gastroenterological Association Institute Guideline on Therapeutic Drug Monitoring in Inﬂammatory Bowel Disease. Gastroenterology 2017, 153, 827–834. [CrossRef] Syversen, S.W.; Jørgensen, K.K.; Goll, G.L.; Brun, M.K.; Sandanger, Ø.; Bjørlykke, K.H.; Sexton, J.; Olsen, I.C.; Gehin, J.E.; Warren, D.J.; et al. Effect of Therapeutic Drug Monitoring vs Standard Therapy During Maintenance Inﬂiximab Therapy on Disease Control in Patients With Immune-Mediated Inﬂammatory Diseases: A Randomized Clinical Trial. JAMA 2021, 326, 2375–2384. [CrossRef] 13. 14. 15. Vande Casteele, N. Assays for Measurement of TNF Antagonists in Practice. Frontline Gastroenterol. 2017, 8, 236–242. [CrossRef] [PubMed] 16. Lee, M.W.M.; Connor, S.; Ng, W.; Toong, C.M.-L. Comparison of Inﬂiximab Drug Measurement across Three Commercially 17. Available ELISA Kits. Pathology 2016, 48, 608–612. [CrossRef] Sam, M.J.; Connor, S.J.; Ng, W.W.-S.; Toong, C.M.-L. Comparative Evaluation of 4 Commercially Available ELISA Kits for Measuring Adalimumab and Anti-Adalimumab Antibodies. Ther. Drug Monit. 2020, 42, 821–828. [CrossRef] 18. Laserna-Mendieta, E.J.; Salvador-Martín, S.; Marín-Jiménez, I.; Menchén, L.A.; López-Cauce, B.; López-Fernández, L.A.; Lucendo, A.J. Comparison of a New Rapid Method for Determination of Serum Anti-Adalimumab and Anti-Inﬂiximab Antibodies with Two Established ELISA Kits. J. Pharm. Biomed. Anal. 2021, 198, 114003. [CrossRef] [PubMed] 19. Berger, A.E.; Gleizes, A.; Waeckel, L.; Roblin, X.; Krzysiek, R.; Hacein-Bey-Abina, S.; Soriano, A.; Paul, S. Validation Study of a New Random-Access Chemiluminescence Immunoassay Analyzer i-TRACK10® to Monitor Inﬂiximab and Adalimumab Serum Trough Levels and Anti-Drug Antibodies. Int. J. Mol. Sci. 2022, 23, 9561. [CrossRef] [PubMed] 20. Lamb, C.A.; Kennedy, N.A.; Raine, T.; Hendy, P.A.; Smith, P.J.; Limdi, J.K.; Hayee, B.; Lomer, M.C.E.; Parkes, G.C.; Selinger, C.; et al. British Society of Gastroenterology Consensus Guidelines on the Management of Inﬂammatory Bowel Disease in Adults. Gut 2019, 68, s1–s106. [CrossRef] 21. Maaser, C.; Sturm, A.; Vavricka, S.R.; Kucharzik, T.; Fiorino, G.; Annese, V.; Calabrese, E.; Baumgart, D.C.; Bettenworth, D.; Borralho Nunes, P.; et al. ECCO-ESGAR Guideline for Diagnostic Assessment in IBD Part 1: Initial Diagnosis, Monitoring of Known IBD, Detection of Complications. J. Crohns Colitis 2019, 13, 144–164. [CrossRef] [PubMed] 22. Landis, J.R.; Koch, G.G. The Measurement of Observer Agreement for Categorical Data. Biometrics 1977, 33, 159–174. [CrossRef] [PubMed] 23. Nuland, M.; Rosing, H.; Schellens, J.H.M.; Beijnen, J.H. Bioanalytical LC–MS/MS Validation of Therapeutic Drug Monitoring Assays in Oncology. Biomed. Chromatogr. 2020, 34, e4623. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_genes14061253	Article Non-Pharmaceutical Interventions against COVID-19 Causing a Lower Trend in Age of LHON Onset Yuxi Zheng 1,†, Xiaoyun Jia 1,†, Shiqiang Li 1, Xueshan Xiao 1, Qingjiong Zhang 1,* and Panfeng Wang 1,2,* 1 State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangdong Provincial Key Laboratory of Ophthalmology and Visual Science, 54 Xianlie Road, Guangzhou 510060, China; [email protected] (Y.Z.); [email protected] (S.L.); [email protected] (X.X.) 2 Gene Diagnostic Laboratory, Genetic Eye Clinic, Zhongshan Ophthalmic Center, Sun Yat-sen University, 54 Xianlie Road, Guangzhou 510060, China * Correspondence: [email protected] or [email protected] (Q.Z.); [email protected] (P.W.) † These authors contributed equally to this work. Abstract: Leber hereditary optic neuropathy (LHON) is a monogenic but multifactorial disease vulnerable to environmental triggers. Little is known about how LHON onset changed during the COVID-19 pandemic and how non-pharmaceutical interventions (NPHIs) against COVID-19 impact LHON onset. One hundred and forty-seven LHON patients with the m.11778G>A mutation complaining of vision loss were involved between January 2017 and July 2022. The onset time points, age of onset, and possible risk factors were evaluated. Analyses were conducted among 96 LHON patients in the Pre-COVID-19 group and 51 in the COVID-19 group. The median (IQR) age of onset decreased signiﬁcantly from 16.65 (13.739, 23.02) in pre-COVID-19 to 14.17 (8.87, 20.29) during COVID-19. Compared with the Pre-COVID-19 group, the COVID-19 group exhibited bimodal distribution with an additional peak at six; the ﬁrst quarter of 2020 also witnessed a relatively denser onset, with no subsequent second spike. NPHIs against COVID-19 signiﬁcantly changed patients’ lifestyles, including higher secondhand smoke exposure (p < 0.001), adherence to masks (p < 0.001), reduction in time spent outdoors for leisure (p = 0.001), and prolonged screen time (p = 0.007). Multivariate logistic regression revealed that secondhand smoke exposure and mask-wearing were independent risk factors of younger LHON onset. Lower age of onset of LHON appeared after the breakout of the COVID-19 pandemic, and novel risk factors were detected, including secondhand exposure and long mask-wearing. Carriers of LHON mtDNA mutations, especially teenagers or children, should be advised to avoid secondhand smoke exposure and there are possible adverse outcomes of longer mask-wearing. Keywords: COVID-19; Leber hereditary optic neuropathy; m.11778G>A; age of onset; risk factors 1. Introduction Leber hereditary optic neuropathy (LHON, OMIM #535000) is the most common form of mitochondrial disease in young adults, with the peak age of onset preferentially in the second and third decades of life, and this coincides with the fact that more than 90% of affected Chinese carriers develop the disease younger than 35 [1,2]. Although the pathogenic mtDNA mutations are homoplastic or nearly homoplastic in most LHON pedigrees, the penetrance can vary among different families carrying the same mutation and even within the same family among other branches, which cannot be due to mtDNA mutations alone [3,4]. The discordant onset in two pairs of monozygotic twins and bet- ter prognosis in childhood-onset LHON further supported the fact that environmental modiﬁers interact with the primary pathogenic mtDNA variants and thus regulate the incomplete penetrance of LHON [5,6]. Citation: Zheng, Y.; Jia, X.; Li, S.; Xiao, X.; Zhang, Q.; Wang, P. Non-Pharmaceutical Interventions against COVID-19 Causing a Lower Trend in Age of LHON Onset. Genes 2023, 14, 1253. https://doi.org/ 10.3390/genes14061253 Received: 27 April 2023 Revised: 6 June 2023 Accepted: 10 June 2023 Published: 12 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Genes 2023, 14, 1253. https://doi.org/10.3390/genes14061253 https://www.mdpi.com/journal/genes genesG C A TT A C GG C A TGenes 2023, 14, 1253 2 of 11 Since the breakout of the COVID-19 pandemic, unprecedented life-altering effects have been brought on adults and children. The implementation of strict non-pharmaceutical interventions (NPHIs) to contain COVID-19, including mask-wearing, school and work- place closure, and home quarantine, has been reported to be associated with the onset or progression of complex diseases, such as a boom in myopia [7], exacerbation in diabetes [8], and increase in preterm stillbirth [9]. A similar situation was also observed in older sporadic patients whom LHON attacked because of increased alcohol consumption or smoking during the COVID-19 pandemic [10]. So, we carried out this retrospective study to sys- temically study the effect of lifestyle changes caused by COVID-19 on the onset of LHON and to ﬁnd possible novel risk factors. On the other hand, it cannot be neglected that the risk factors of certain diseases are distinct across human populations with different genetic backgrounds [11]. Therefore, most of the research that investigates the effects of risk factors related to NPHIs on complex diseases might be limited since the genetic backgrounds of most of them were hard to consider [12,13]. LHON, a monogenic but multifactorial disease, acts as an ideal disease model to investigate the effect of COVID-19 containment measures on the disease when the participants were under controllable genetic differences [14]. Herein, we conducted a cohort study of LHON patients with the m.11778G>A mu- tation to assess the changes in LHON onset during the COVID-19 pandemic and try to uncover the potential risk factors for penetrance. 2. Materials and Methods 2.1. Study Design and Population Patients complaining of sequentially bilateral, painless, subacute visual failure onset from January 2017 to July 2022 and who were genetically diagnosed with the mtDNA m.11778G>A mutation using Sanger sequencing were included. Before peripheral blood was collected for the genetic test, the probands or their legal guardians signed written informed consent. This study conformed to the tenets of the Declaration of Helsinki, and ethical approval was obtained from the institutional review board of Zhongshan Ophthalmic Centre, Guangzhou, China. The symptom onset time was divided into two phases: pre-COVID-19 (January 2017 to December 2019), corresponding to the period prior to the COVID-19 breakout, and COVID-19 (January 2020 to July 2022), corresponding to the period in which social restric- tions were still in place. 2.2. Data Collection Detailed disease history of all participants, including the age, gender, time of vision loss, previous treatment, family history, and residence, was obtained by senior ophthal- mologists during the ﬁrst visit. Potential risk factors were selected based on literature reviews [10,15]. A quantitative questionnaire survey via phone or in an outpatient clinic was used to collect reliable patient information between July 2022 and September 2022. Socioeconomic characteristics prior to symptom onset included education level, occupation, and annual family income (CNY). Clinical-related and lifestyle indicators before the vision loss included alcohol consumption and smoke exposure, mask-wearing habits, COVID-19 vaccination, daily screen time, and outdoor activity time. Secondhand smoke is deﬁned as the parents or any other family members living with the patients smoking at home [16]. Participants are allocated to the secondhand smoke exposure if there is at least one smoker at their home; those who are nonsmokers and have no smoker at home are classiﬁed as non-smoke exposed. Outdoor time comprises time spent outdoors for sports and tome spent outdoors for leisure (walking to and from school, post-dinner walking, playing in the park) [17]. Daily digital screen time consists of time spent on TV, computer, smartphone, and tablet. Based on the assumption that a non-response phone call—deﬁned as no answer after ten calls, a dead number, or a redirection of calls, is entirely random, non-answered phone call participants were excluded for deeper analysis. Genes 2023, 14, 1253 3 of 11 2.3. Statistical Analysis and Visualization Statistical analysis and visualization were performed in open source R (version 4.1.2) and RStudio. Categorical variables are displayed as numbers and the different prevalence between groups was analyzed using a Chi-square test or Fisher’s exact test. Continuous variables that passed the normality test are presented as mean (SD) and were assessed using Student’s t-test; otherwise, the Mann–Whitney test was used. Statistical signiﬁcance was deﬁned as p < 0.05. Sociodemographic characteristics and COVID-19-related pressures that were assessed as clinically relevant or those with a signiﬁcant level of p < 0.05 in univariate analysis were included for the multivariate logistic regression model. Given the rarity of LHON and the number of outcome events, the ﬁnal model was constructed by carefully limiting the number of variables included. 3. Results 3.1. Comparison of LHON Onset between the Pre-COVID-19 Group and the COVID-19 Group There were 147 patients with the m.11778G>A mutation complaining of vision loss between January 2017 and July 2022. The median age of onset of LHON signiﬁcantly decreased to 14.17 (8.97, 20.29) in the COVID-19 group compared to 16.65 (13.73, 23.02) in the Pre-COVID-19 group (p = 0.016), in accordance with the median age of onset of 16 in our prior research (Figure 1A) [2]. Unexpectedly, in addition to the forward shift of the age range from 16.65 to 14.17, a novel and pronounced peak at six occurred, different from pre- vious observations [18]. To further reveal the underlying reason for this phenomenon, we illustrated the frequency of symptom onset at the different times between the two groups using kernel density plots associated with scatter plots in Figure 1B. No regular trend of LHON onset throughout the different time points was observed in the Pre-COVID-19 group (blue curve). The yellow density curve of LHON onset in the COVID-19 group had developed an elevation since January 2020, the beginning of the COVID-19 lockdowns imposed, which peaked three months later and then fell rapidly; no subsequent second COVID-19-associated spike was seen. The median age of onset declined signiﬁcantly to 11.30 (7.79, 18.31) from January to March 2020 (early pandemic), with a Hodges–Lehmann median estimation difference of 5.356 (95% CI 0.3781–9.553) (p = 0.040) when compared with that in the Pre-COVID-19 group (January 2017 to December 2019). Box-violin plots in- dicated that seasonal variations had no statistically signiﬁcant effect on LHON age of onset (Figure S1). The distribution difference among three age of onset groups in two periods shown in Table S1 achieved statistical signiﬁcance (p = 0.001). Notably, the male-to-female ratio of the probands in the whole cohort was 15.38:1 (139:8), with 23.00:1 (92:4) for the Pre-COVID-19 group and 11.75:1 (47:4) for the COVID-19 group (Table S1). When the affected family members were included in the analysis, the male-to-female ratio was 6.76:1 (115:17) for the Pre-COVID-19 group and 3.47:1 (59:17) for the COVID-19 group. 3.2. Socio-Demographic Characteristics of Study Patients To further analyze how the NPHIs altered the age of onset, we retroactively inter- viewed patients about their lifestyles. Because up to 97% of the patients in this study manifested vision failure before they were 35 years old, consistent with the dominant age range in our prior research and a previous study on Chinese probands, we then focused on the patients with an age of onset of younger than 35 years [2,19]. Patients were divided into four groups according to the onset time being before or after COVID-19 and using a cut-off point of 16 years old, which is the end of the nine-year compulsory education according to China’s Compulsory Education law [2]: teenager-onset in pre-COVID-19 (TO-pre, onset ≤ 16 years), teenager-onset in COVID-19 (TO-post, onset ≤ 16 years), adult- onset in pre-COVID-19 (AO-pre, 16 < onset ≤ 35 years), and adult-onset in COVID-19 (AO-pre, 16 < onset ≤ 35 years). Socio-demographic characteristics and COVID-19-related characteristics were then compared among the four groups. Genes 2023, 14, 1253 4 of 11 Figure 1. Distribution of LHON onset at different time points and ages. (A) Quantiﬁcation of age of onset in the Pre-COVID-19 group (blue) and the COVID-19 group (yellow). The color bar on the horizontal axis represents the three-age of onset groups. Blue and yellow dots indicate the median age of onset of pre-COVID-19 and COVID-19, respectively. The Kernel density curve was generated and arranged using the R package ggridges and ggplot2 with adjust equal to 0.5. (B) A combined scatter plot and histogram of LHON onset in the pre-COVID-19 era (blue) and the COVID-19 era (yellow). The horizontal axis represents the time (months) and the vertical axis is the age of onset. Each dot represents the timepoint and age of onset of individuals when experiencing sudden vision loss. Each year was divided into 12 segments on the x-axis. Out of the 137 participants with an age of onset younger than 35 years, 114 ﬁnished the phone or ﬁeld questionnaire, a response rate of 83.21%, with a median age of 15.16 (11.70, 19.17) at onset. The 114 interviewed participants and the complete 137 participants were not different in any of the general socio-demographics (Table S2), indicating that loss during Genes 2023, 14, x FOR PEER REVIEW 4 of 12 Figure 1. Distribution of LHON onset at different time points and ages. (A) Quantification of age of onset in the Pre-COVID-19 group (blue) and the COVID-19 group (yellow). The color bar on the horizontal axis represents the three-age of onset groups. Blue and yellow dots indicate the median age of onset of pre-COVID-19 and COVID-19, respectively. The Kernel density curve was generated Genes 2023, 14, 1253 5 of 11 follow-up was less likely to affect the overall conclusions. The general characteristics and COVID-19-related lifestyle of the interviewed patients in two time periods are demonstrated in Table 1. Among the 114 participants, 73 were from pre-COVID-19, and 41 were from COVID-19, the median ages of onset of which were 16.53 (13.77, 21.37) and 13.22 (8.61, 15.67) (p < 0.001), respectively. The signiﬁcant difference in age of onset also led to the differences in education (p < 0.001) and occupation (p = 0.013). Males were roughly 93.86% overall, with a male-to-female ratio of 15.28:1 based on probands and 5.12:1 based on affected family members. Most participants were students (72.80%) and reported an annual family income of less than 80,000 CNY (81.58%). The median education years of the non-students were 9.00 (9.00, 11.00), mostly with secondary education. There was no statistical difference in gender, geographical location, residence, the season of onset, family yearly income, and alcohol consumption between pre-COVID-19 and COVID-19 (Table 1). Table 1. Sociodemographic characteristics of patients with LHON. Characteristics Sociodemographic characteristics Overall (n = 114) Patients, No. (%) (n = 114) Pre-COVID-19 (n = 73) COVID-19 (n = 41) p Value Age of onset, median (IQR) 15.16 (11.70, 19.17) 16.53 (13.77, 21.37) 13.22 (8.61, 15.67) <0.001 Age of onset group, % age ≤ 16 16 < age ≤ 35 Gender Male Female Annual family income, % ≤80,000 RMB >80,000 RMB Geographical location, % North China South China Residence, % Rural areas Urban areas Season of onset, % Cold season Warm season Education years, median (IQR) Occupation, % Student Non-student COVID-19 related pressures Time spent outdoors Outdoors for sports, median (IQR), h/day Outdoors for leisure, mean (SD), h/day Time spent on screen-based devices, h/day Smoke exposed Firsthand smoke 66 (57.89) 48 (42.11) 107 (93.86) 7 (6.14) 93 (81.58) 21 (18.42) 3 (2.63) 111 (97.37) 71 (62.28) 43 (37.72) 50 (43.86) 64 (56.14) 32 (43.84) 41 (56.16) 68(94.44) 4(5.56) 60 (82.19) 13 (17.81) 2 (2.74) 71 (97.26) 46 (63.01) 27 (37.99) 30 (41.10) 43 (58.90) 34 (82.93) 7 (17.07) 38 (92.68) 3 (7.32) 33 (80.49) 8 (19.51) 1 (2.44) 40 (97.561) 25 (60.98) 16 (39.02) 20 (48.78) 21 (51.22) <0.001 0.989 0.822 >0.999 0.989 0.44 7.50 (6.00, 9.00) 9.00 (7.00, 10.0) 6.00 (2.00, 8.00) <0.001 83 (72.81) 31 (27.19) 47 (64.38) 26 (35.62) 36 (87.80) 5 (12.20) 1.50 (0.75, 2.50) 1.00 (0.00, 2.00) 0.75 (0.50, 1.00) 4.00 (2.00, 5.00) 1.75 (0.75, 3.00) 1.00 (0.00, 2.00) 0.75 (0.50, 1.00) 3.00 (2.00, 5.00) 1.00 (0.50, 2.25) 0.50 (0.00, 1.50) 0.50 (0.50, 0.75) 5.00 (3.00, 6.00) 0.013 0.074 0.583 0.001 0.007 17 (14.9) 15 (20.5) 2 (4.88) <0.001 Genes 2023, 14, 1253 6 of 11 Table 1. Cont. Characteristics Secondhand smoke Non-exposed Alcohol consumption YES NO Vaccination YES NO Mask-wearing habits YES NO Overall (n = 114) 49 (43.0) 48 (42.1) 13 (11.4) 101 (88.6) - - - - Patients, No. (%) (n = 114) Pre-COVID-19 (n = 73) 21 (28.8) 37 (50.7) 10 (13.7) 63 (86.3) - - 0 (0.00) 73 (100.00) COVID-19 (n = 41) 28 (68.3) 11 (26.8) 3 (7.32) 38 (92.7) 19 (46.34) 22 (53.66) 41 (100.00) 0 (0.00) p Value 0.372 0.839 <0.001 3.3. COVID-19-Related Characteristics As a monogenic multifactorial disease, LHON is susceptible to environmental factors. Since the outbreak of COVID-19, people’s lifestyle has been modiﬁed by interventions to oppose the pandemic, such as mask usage, home quarantine, and COVID-19 vaccination. Therefore, patients or guardians were asked about wearing a mask when going out, smoke exposure, daily screen time, and outdoor activity time, including outdoors for sports and outdoor for leisure, to investigate the intervention of NPHIs in LHON onset. Signiﬁcant increases in secondhand smoke exposure (p < 0.001), adherence to masks (p < 0.001), prolonged screen time (p = 0.007), and decreased time spent outdoors for leisure (p = 0.001) were observed when comparing the COVID-19-related changes between the two periods (Table 1). Therefore, further analysis focused on these four potential risk factors. Time spent on screen was dichotomized at 4 h/d, the median in the total questioned population. There was no statistical difference in TO-pre or TO-post when compared to the adult-onset group of the same time point, respectively (p = 0.488) (p = 0.673) (Figure 2A). It is also noteworthy that the duration of time spent outdoors for leisure in the COVID-19 group decreased to 0.50 (0.50, 0.75) h/day (p = 0.001), probably due to the policy of reduction in nonessential activity. Prolonged screen time and limited outdoor leisure time collectively indicate a longer time at home and might increase the risk of secondhand exposure, especially for those whose family members are smoking. A higher prevalence of secondhand smoke exposure was observed in TO-pre or TO-post when, respectively, compared with the AO-pre and AO-post (p < 0.001) (p = 0.025), indicating teenagers had more secondhand smoke exposure (Figure 2B). We distinguished those who reported wearing a mask when going out from those not reporting this preventive strategy. All the participants in the COVID-19 group reported wearing a mask, as mask usage is a cost-effective measure to alleviate the risk of COVID-19 transmissions. After adjusting for other potentially confounding factors, the impact of mask usage ﬁnds a signiﬁcant effect on teenager-onset LHON (3.73; 95%CI, 1.18–13.47) (p = 0.031), and secondhand smoke was associated with nearly a ﬁve-fold increased risk of teenager-onset LHON with high heterogeneity (4.82; 95%CI, 1.69–14.23) (p = 0.004) (Figure 2C). As tobacco is banned from being sold to minors in China, only one of the juveniles in this study smoked; that is why active smoking was calculated as negatively associated with teenager-onset LHON. For those who developed the disease after the COVID-19 breakout, we additionally inquired about COVID-19 vaccination and the date of doses given. Among the 19 patients who reported they were vaccinated, the coverage rate was 47.06% (16/34) in the TO-post Genes 2023, 14, 1253 7 of 11 group, while it was 42.86% (3/7) in the AO-post. There was no statistical difference in COVID-19 vaccination between the TO-post and AO-post (p = 0.839). Figure 2. Uncovering potential risk factors for teenager-onset LHON. (A) Stack bar plot showing the proportion of digital screen time between the period before and after the COVID-19 pandemic categorized by teenager-onset group and adult-onset group. (B) Stack bar plot showing the proportion of smoke-exposed between the period before and after the COVID-19 pandemic categorized by teenager-onset group and adult-onset group. (C) Forest plot for putative triggers. Odds and 95% (CI) were computed and visualized using R package forestplot (version 3.1.1). OR, odds ratio. 4. Discussion Our study reports an increase in teenager-onset LHON after the breakout of COVID-19, with a proportion of 68.63% (35/51). The bimodal distribution of the COVID-19 group is observed with two peaks of nearly equal altitude, roughly at age six and thirteen. Even though childhood-onset LHON is correlated with a better prognosis when a higher spontaneous visual recovery rate was indicated, complete restoration of vision is not yet observed, and many of these teenagers sustain permanent vision impairment [6]. Therefore, Genes 2023, 14, x FOR PEER REVIEW 7 of 12 COVID-19 group decreased to 0.50 (0.50, 0.75) h/day (p = 0.001), probably due to the policy of reduction in nonessential activity. Figure 2. Uncovering potential risk factors for teenager-onset LHON. (A) Stack bar plot showing the proportion of digital screen time between the period before and after the COVID-19 pandemic categorized by teenager-onset group and adult-onset group. (B) Stack bar plot showing the propor-tion of smoke-exposed between the period before and after the COVID-19 pandemic categorized by teenager-onset group and adult-onset group. (C) Forest plot for putative triggers. Odds and 95% (CI) were computed and visualized using R package forestplot (version 3.1.1). OR, odds ratio. Prolonged screen time and limited outdoor leisure time collectively indicate a longer time at home and might increase the risk of secondhand exposure, especially for those whose family members are smoking. A higher prevalence of secondhand smoke exposure was observed in TO-pre or TO-post when, respectively, compared with the AO-pre and AO-post (p < 0.001) (p = 0.025), indicating teenagers had more secondhand smoke expo-sure (Figure 2B). Genes 2023, 14, 1253 8 of 11 it is of critical importance to uncover the risk factors behind this phenomenon. In the present study, the detailed retrospective investigation of the lifestyle habits allowed us ﬁrst to propose secondhand smoke exposure and mask-wearing as potential triggers of teenager-onset LHON. Several extrinsic factors have already been proposed for LHON onset, including gen- der, haplogroup, cigarette smoke, alcohol consumption, head injury, excessive blood loss, acute illness, and anti-retroviral and anti-tuberculosis drugs [20,21]. Even though cigarette smoke has been long recognized as a signiﬁcant environmental trigger of LHON by reduc- ing the copy number of mtDNA, most studies explored the inﬂuence of proactive instead of passive smoke [15]. Secondhand smoke is a signiﬁcantly higher risk factor for teenager- onset LHON after adjustment for other potential triggers. Very scarce association between secondhand smoke exposure and ocular adverse effect has been reported, including refrac- tive error, AMD, choroidal thinning, and cataracts; advocating hazards of environmental smoke should also be taken into consideration in ophthalmological perspectives in future studies [22,23]. Research conducted in Hong Kong indicated secondhand smoke expo- sure was associated with thinner peripapillary retinal ﬁber layers, which was a sensitive biomarker of axonal damage and the damage level of retinal ganglion degeneration [16]. Moreover, the concentration of secondhand smoke in the air is intense, especially in a conﬁned space during the home quarantine, and teenagers have more narrow airways and inhale faster, making teenagers themselves more susceptible to environmental hazards [24]. According to a large cross-sectional study among 1745 individuals in mainland China, the mask-wearing rates were 20 of 1745 individuals (1.1%) before the COVID-19 pandemic, while during the COVID-19, the mask-wearing rate increased to 1090 of 1097 (99.4%) when the Chinese government launched a mandatory mask wearing policy in January 2020 [25]. It is reported that modest changes in healthy individuals’ gas exchange, pulmonary function, and psychological effects showed up after wearing a mask during exercising [26]. The subtle cerebral hemodynamics and oxygenation changes brought by daily mask usage raise a concern about exacerbation in the dysfunction of mitochondria in LHON carriers, as LHON is sensitive to mitochondrial biogenesis alteration and oxidative stress [27]. So, it is rational to treat mask-wearing as a candidate risk factor for LHON during COVID-19. Further analysis is needed to ensure that mask usage associated with teenager-onset LHON does not persist beyond the state of the COVID-19 pandemic. The ﬁrst spike in the time–density curve of LHON onset was in accordance with the mass quarantine in early 2020. Our ﬁgures show that the change in the age of LHON onset is not subject to seasonal variation, we suggest that the teenager-onset LHON boom is likely to be an effect of the COVID-19 home quarantine. Individuals who have been exposed to secondhand smoke may have already encountered this exposure from cohabitants prior to the COVID-19 pandemic; however, due to the prolonged time spent at home during lockdown measures, the intensity of secondhand smoke exposure might have increased, leading to the accumulation of smoke to a critical level. Additionally, changes in lifestyle practices, such as wearing a mask, might cause damage before potential protective pathways have had time to reach their full potential, leading to the onset of illness. As surgical masks are primarily designed to prevent the release of droplets, including saliva or respiratory droplets, from the wearer to others, instead of ﬁltering airborne particles, such as the ﬁne particulate matter found in secondhand smoke, it is possible that the combined effects of secondhand smoke exposure during home conﬁnement and the potential impact of prolonged mask usage on oxygen intake may play a role in triggering LHON onset [28]. The following slight ﬂuctuation reﬂects the neuroplasticity as restrictions were gradually eased, consistent with this hypothesis and a previous study on cortical plasticity linked to retinal ganglion cell loss [29,30]. Most of the teenagers in our study might be genetically primed to adult-onset or even lifelong asymptomatic LHON but became symptomatic early in life due to passive smoking and mask usage. However, it should be noted that there were still 22.73% (15/66) of the teenager- onset LHON in our study neither exposed to secondhand smoke nor mask-wearing. The Genes 2023, 14, 1253 9 of 11 complex scenario between different environmental triggers and mtDNA haplogroup or even modiﬁers in nuclear genes requires further exploration. Although increased digital screen time and reduced outdoor time were revealed to be insigniﬁcantly associated with teenager-onset LHON, whether these short-term changes will persist or deteriorate remains to be determined as strict containment measures are still imposed to mitigate the spread of COVID-19. It has been reported that the burden of ophthalmic risk factors was higher among those of lower socioeconomic status, despite the accessibility of health care ser- vices [31,32]. There are about 81.6% (93/114) of patients in our cohort with a lower family income, and 62.3% (71/114) living in rural areas. The overall prevalence of COVID-19 vaccination rate in the COVID-19 group is 46.34% (19/41), with 42.86% (3/7) in AO-post and 47.06% (16/34) in TO-post, the rate of which is lower than the ﬁrst dose vaccine coverage rate of 98.02% observed in the general population in China [33]. Chief factors leading to vaccination hesitancy might be fears caused by some unscientiﬁc opinions such as vaccinations causing disease or deterioration of vision, especially for the mutation carriers and those with a family history of LHON. Our research indicates that COVID-19 vaccination was not a risk factor for earlier LHON onset, and the reasons lie in two factors: (1) The declined age of onset of LHON mainly happened from January to March 2020 when the COVID-19 vaccines were unavailable. (2) There was no statistical difference in COVID-19 vaccination between TO-post and AO-post. Given the efﬁcacy of COVID-19 vaccines to lessen the hospitalization with COVID-19 pneumonia and the mortality rate [34], whether the LHON unaffected carriers or patients should refuse the vaccine or not remains conservative before a much larger sample size shows evidence of an increased risk for LHON onset after vaccination. According to the results of our questionnaire, none of the 147 individuals were infected with COVID-19 at the time of the survey. This is comparable to the cumulative number of conﬁrmed COVID-19 cases reported in the general population in China, which stood at 250,449 by the end of September 2022, as documented by the China Center for Disease Control and Prevention (http://www.nhc.gov.cn/xcs/xxgzbd/gzbd_index.shtml, accessed on 9 June 2023), with an estimated COVID-19 infection rate of 0.17%, based on the seventh national census (https://www.gov.cn/guoqing/2021-05/13/content_5606149.htm, assessed on 9 June 2023). The COVID-19 infection was not treated as an independent risk factor in this study because none of the enrolled individuals were infected with the COVID-19 virus before September 2022. The inﬂuence of the COVID-19 infection on LHON onset will be based on data from December 2022 when China’s massive COVID-19 outbreak peaked in our future study. Still, our study has some limitations. First, our study is limited by its retrospective nature and its associated recall bias. However, the rarity of LHOH makes it unrealistic to evaluate a more signiﬁcant number of LHON patients prospectively. Second, the rarity of LHON imposes restrictions on the size of our cohort. However, the number of patients in our different groups satisﬁed the demand for statistical analysis. Third, it is likely that some of the LHON patients in the COVID-19 group might have been inﬂuenced by factors that happened before COVID-19. Almost all of the recent studies on extra abnormality onset or progression of several complex diseases during COVID-19 set lockdowns as the cut-off point when groups were concerned [17,25,35]. Some of them also pointed out the limitation that it is difﬁcult to estimate the standpoint since the exact duration of exposure is unclear [35,36]. In summary, to minimize inﬂuences caused by different genetic backgrounds, we used the most common point mutation m.11778G>A of LHON as the study subject to reveal that the primary age of LHON onset shifted to an earlier time during the COVID-19 pandemic. These ﬁndings were associated with increased secondhand smoke exposure and mask-wearing habits, which should have a practical role in clinical implications and genetic counseling. For LHON carriers, secondhand smoke should also be strongly avoided, in addition to avoiding proactive smoking and efforts to relieve the potential adverse effects of mask-wearing should be considered. Genes 2023, 14, 1253 10 of 11 Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/genes14061253/s1, Figure S1: Box-violin plots for age of onset of LHON in four seasons before and during COVID-19; Table S1: Socio-demographic characteristics of all patients with LHON onset from January 2017 to July 2022; Table S2: Comparison of the demographic characteristics of patients with LHON onset younger than 35 years between the complete dataset and data from interviewed participants. Author Contributions: Conceptualization, Y.Z., X.J., Q.Z. and P.W.; methodology, P.W. and Y.Z.; soft- ware, P.W., S.L., X.J. and Y.Z.; validation, P.W. and Y.Z.; formal analysis, P.W. and Y.Z.; investigation, P.W., X.J., X.X. and Y.Z.; resources, X.J., X.X., Q.Z. and P.W.; writing—original draft preparation, X.J. and Y.Z.; writing—review and editing, Q.Z. and P.W.; visualization, P.W. and Y.Z.; supervision, Q.Z. and P.W.; project administration, Q.Z. and P.W.; funding acquisition, Q.Z. and P.W. All authors have read and agreed to the published version of the manuscript. Funding: This study was supported by the Science and Technology Planning Projects of Guangzhou (grant 202102010271) and the Fundamental Research funds of the State Key Laboratory of Ophthal- mology at Zhongshan Ophthalmic Center. Institutional Review Board Statement: This study conformed to the tenets of the Declaration of Helsinki, and ethical approval (2011KYNL012) was obtained from the institutional review board of Zhongshan Ophthalmic Centre, Guangzhou, China. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented are available on request from the corresponding author. The data are not publicly available due to ethical privacy. Acknowledgments: The authors thank all patients and their family members for participation in this study. Conﬂicts of Interest: All authors in this study declared no conﬂict of interest. References 1. 2. Yu-Wai-Man, P.; Grifﬁths, P.G.; Chinnery, P.F. Mitochondrial Optic Neuropathies—Disease Mechanisms and Therapeutic Strategies. Prog. Retin. Eye Res. 2011, 30, 81–114. [CrossRef] [PubMed] Jia, X.; Li, S.; Xiao, X.; Guo, X.; Zhang, Q. Molecular Epidemiology of Mtdna Mutations in 903 Chinese Families Suspected with Leber Hereditary Optic Neuropathy. J. Hum. Genet. 2006, 51, 851–856. [CrossRef] [PubMed] 3. Nikoskelainen, E.K.; Savontaus, M.L.; Wanne, O.P.; Katila, M.J.; Nummelin, K.U. Leber’s Hereditary Optic Neuroretinopathy, a Maternally Inherited Disease. A Genealogic Study in Four Pedigrees. Arch. Ophthalmol. 1987, 105, 665–671. [CrossRef] 4. Newman, N.J.; Lott, M.T.; Wallace, D.C. The Clinical Characteristics of Pedigrees of Leber’s Hereditary Optic Neuropathy with 5. 6. the 11778 Mutation. Am. J. Ophthalmol. 1991, 111, 750–762. [CrossRef] Johns, D.R.; Smith, K.H.; Miller, N.R.; Sulewski, M.E.; Bias, W.B. Identical Twins Who Are Discordant for Leber’s Hereditary Optic Neuropathy. Arch. Ophthalmol. 1993, 111, 1491–1494. [CrossRef] [PubMed] Siedlecki, J.; Koenig, S.; Catarino, C.; Schaumberger, M.M.; Schworm, B.; Priglinger, S.G.; Rudolph, G.; von Livonius, B.; Klopstock, T.; Priglinger, C.S. Childhood Versus Early-Teenage Onset Leber’s Hereditary Optic Neuropathy: Visual Prognosis and Capacity for Recovery. Br. J. Ophthalmol. 2022. Online ahead of print. [CrossRef] 7. Ma, M.; Xiong, S.; Zhao, S.; Zheng, Z.; Sun, T.; Li, C. COVID-19 Home Quarantine Accelerated the Progression of Myopia in 8. Children Aged 7 to 12 Years in China. Invest. Ophthalmol. Vis. Sci. 2021, 62, 37. [CrossRef] Valabhji, J.; Barron, E.; Gorton, T.; Bakhai, C.; Kar, P.; Young, B.; Khunti, K.; Holman, N.; Sattar, N.; Wareham, N.J. Associations between Reductions in Routine Care Delivery and Non-COVID-19-Related Mortality in People with Diabetes in England During the COVID-19 Pandemic: A Population-Based Parallel Cohort Study. Lancet Diabetes Endocrinol. 2022, 10, 561–570. [CrossRef] 9. Hui, L.; Marzan, M.B.; Potenza, S.; Rolnik, D.L.; Pritchard, N.; Said, J.M.; Palmer, K.R.; Whitehead, C.L.; Sheehan, P.M.; Ford, J.; et al. Increase in Preterm Stillbirths in Association with Reduction in Iatrogenic Preterm Births During COVID-19 Lockdown in Australia: A Multicenter Cohort Study. Am. J. Obs. Gynecol. 2022, 227, 491.e1–491.e17. [CrossRef] 10. Zaslavsky, K.; Margolin, E.A. Leber’s Hereditary Optic Neuropathy in Older Individuals Because of Increased Alcohol Consump- tion During the COVID-19 Pandemic. J. Neuroophthalmol. 2021, 41, 316–320. [CrossRef] 11. Roberts, N.J.; Vogelstein, J.T.; Parmigiani, G.; Kinzler, K.W.; Vogelstein, B.; Velculescu, V.E. The Predictive Capacity of Personal Genome Sequencing. Sci. Transl. Med. 2012, 4, 133ra58. [CrossRef] 12. Lai, S.; Ruktanonchai, N.W.; Zhou, L.; Prosper, O.; Luo, W.; Floyd, J.R.; Wesolowski, A.; Santillana, M.; Zhang, C.; Du, X.; et al. Effect of Non-Pharmaceutical Interventions to Contain COVID-19 in China. Nature 2020, 585, 410–413. [CrossRef] 13. Corona, E.; Chen, R.; Sikora, M.; Morgan, A.A.; Patel, C.J.; Ramesh, A.; Bustamante, C.D.; Butte, A.J. Analysis of the Genetic Basis of Disease in the Context of Worldwide Human Relationships and Migration. PLoS Genet. 2013, 9, e1003447. [CrossRef] Genes 2023, 14, 1253 11 of 11 14. Carelli, V.; Ross-Cisneros, F.N.; Sadun, A.A. Mitochondrial Dysfunction as a Cause of Optic Neuropathies. Prog. Retin. Eye Res. 2004, 23, 53–89. [CrossRef] 15. Kirkman, M.A.; Yu-Wai-Man, P.; Korsten, A.; Leonhardt, M.; Dimitriadis, K.; De Coo, I.F.; Klopstock, T.; Chinnery, P.F. Gene- Environment Interactions in Leber Hereditary Optic Neuropathy. Brain 2009, 132, 2317–2326. [CrossRef] [PubMed] 16. Li, J.; Yuan, N.; Chu, W.K.; Cheung, C.Y.; Tang, S.; Li, F.F.; Chen, L.J.; Kam, K.W.; Young, A.L.; Ip, P.; et al. Exposure to Secondhand Smoke in Children Is Associated with a Thinner Retinal Nerve Fiber Layer: The Hong Kong Children Eye Study. Am. J. Ophthalmol. 2021, 223, 91–99. [CrossRef] [PubMed] 17. Zhang, X.; Cheung, S.S.L.; Chan, H.N.; Zhang, Y.; Wang, Y.M.; Yip, B.H.; Kam, K.W.; Yu, M.; Cheng, C.Y.; Young, A.L.; et al. Myopia Incidence and Lifestyle Changes among School Children During the COVID-19 Pandemic: A Population-Based Prospective Study. Br. J. Ophthalmol. 2021, 106, 1772–1778. [CrossRef] [PubMed] 18. Yu-Wai-Man, P.; Grifﬁths, P.G.; Brown, D.T.; Howell, N.; Turnbull, D.M.; Chinnery, P.F. The Epidemiology of Leber Hereditary Optic Neuropathy in the North East of England. Am. J. Hum. Genet. 2003, 72, 333–339. [CrossRef] 19. Guo, H.; Li, S.; Dai, L.; Huang, X.; Yu, T.; Yin, Z.; Bai, Y. Genetic Analysis in a Cohort of Patients with Hereditary Optic Neuropathies in Southwest of China. Mitochondrion 2019, 46, 327–333. [CrossRef] 20. Dimitriadis, K.; Leonhardt, M.; Yu-Wai-Man, P.; Kirkman, M.A.; Korsten, A.; De Coo, I.F.; Chinnery, P.F.; Klopstock, T. Leber’s Hereditary Optic Neuropathy with Late Disease Onset: Clinical and Molecular Characteristics of 20 Patients. Orphanet. J. Rare. Dis. 2014, 9, 158. [CrossRef] 21. Tsao, K.; Aitken, P.A.; Johns, D.R. Smoking as an Aetiological Factor in a Pedigree with Leber’s Hereditary Optic Neuropathy. Br. J. Ophthalmol. 1999, 83, 577–581. [CrossRef] 22. Yuan, N.; Li, J.; Tang, S.; Li, F.F.; Lee, C.O.; Ng, M.P.H.; Cheung, C.Y.; Tham, C.C.; Pang, C.P.; Chen, L.J.; et al. Association of Secondhand Smoking Exposure with Choroidal Thinning in Children Aged 6 to 8 Years: The Hong Kong Children Eye Study. JAMA Ophthalmol. 2019, 137, 1406–1414. [CrossRef] 23. Lois, N.; Abdelkader, E.; Reglitz, K.; Garden, C.; Ayres, J.G. Environmental Tobacco Smoke Exposure and Eye Disease. Br. J. Ophthalmol. 2008, 92, 1304–1310. [CrossRef] 24. Etzel, R.A. Indoor and Outdoor Air Pollution: Tobacco Smoke, Moulds and Diseases in Infants and Children. Int. J. Hyg. Env. Health 2007, 210, 611–616. [CrossRef] 25. Chen, Y.J.; Qin, G.; Chen, J.; Xu, J.L.; Feng, D.Y.; Wu, X.Y.; Li, X. Comparison of Face-Touching Behaviors before and During the Coronavirus Disease 2019 Pandemic. JAMA Netw. Open. 2020, 3, e2016924. [CrossRef] 26. Zheng, C.; Poon, E.T.; Wan, K.; Dai, Z.; Wong, S.H. Effects of Wearing a Mask During Exercise on Physiological and Psychological Outcomes in Healthy Individuals: A Systematic Review and Meta-Analysis. Sport. Med. 2022, 53, 125–150. [CrossRef] 27. Wu, H.; Chen, Q. Hypoxia Activation of Mitophagy and Its Role in Disease Pathogenesis. Antioxid. Redox Signal. 2015, 22, 1032–1046. [CrossRef] 28. Douglas, J.D.M.; McLean, N.; Horsley, C.; Higgins, G.; Douglas, C.M.; Robertson, E. COVID-19: Smoke Testing of Surgical Mask and Respirators. Occup. Med. 2020, 70, 556–563. [CrossRef] 29. Mateus, C.; d’Almeida, O.C.; Reis, A.; Silva, E.; Castelo-Branco, M. Genetically Induced Impairment of Retinal Ganglion Cells at the Axonal Level Is Linked to Extrastriate Cortical Plasticity. Brain Struct. Funct. 2016, 221, 1767–1780. [CrossRef] 30. Vasalauskaite, A.; Morgan, J.E.; Sengpiel, F. Plasticity in Adult Mouse Visual Cortex Following Optic Nerve Injury. Cereb. Cortex. 2019, 29, 1767–1777. [CrossRef] 31. Wong, E.S.; Choy, R.W.; Zhang, Y.; Chu, W.K.; Chen, L.J.; Pang, C.P.; Yam, J.C. Global Retinoblastoma Survival and Globe Preservation: A Systematic Review and Meta-Analysis of Associations with Socioeconomic and Health-Care Factors. Lancet Glob. Health 2022, 10, e380–e389. [CrossRef] [PubMed] 32. Etty, M.C.; Michaelsen, S.; Yelle, B.; Beaulieu, K.; Jacques, P.; Ettaleb, S.; Samaha, D.; Tousignant, B.; Druetz, T. The Sociodemo- graphic Characteristics and Social Determinants of Visual Impairment in a Homeless Population in the Montreal Area. Can. J. Public Health 2022, 114, 113–124. [CrossRef] 33. Wu, J.; Guo, X.; Zhou, X.; Wang, M.; Gu, J.; Miao, Y.; Tarimo, C.S.; He, Y.; Xing, Y.; Ye, B. The Pattern from the First Three Rounds of Vaccination: Declining Vaccination Rates. Front. Public Health 2023, 11, 1124548. [CrossRef] [PubMed] 34. Kelly, J.D.; Leonard, S.; Hoggatt, K.J.; Boscardin, W.J.; Lum, E.N.; Moss-Vazquez, T.A.; Andino, R.; Wong, J.K.; Byers, A.; Bravata, D.M.; et al. Incidence of Severe COVID-19 Illness Following Vaccination and Booster with Bnt162b2, Mrna-1273, and Ad26.Cov2.S Vaccines. JAMA 2022, 328, 1427–1437. [CrossRef] [PubMed] 35. Molina, R.L.; Tsai, T.C.; Dai, D.; Soto, M.; Rosenthal, N.; Orav, E.J.; Figueroa, J.F. Comparison of Pregnancy and Birth Outcomes before Vs During the COVID-19 Pandemic. JAMA Netw. Open. 2022, 5, e2226531. [CrossRef] 36. Pellegrini, M.; Bernabei, F.; Scorcia, V.; Giannaccare, G. May Home Conﬁnement During the COVID-19 Outbreak Worsen the Global Burden of Myopia? Graefes Arch. Clin. Exp. Ophthalmol. 2020, 258, 2069–2070. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.31557_apjcp.2020.21.10.2979	RESEARCH ARTICLE Editorial Process: Submission:04/13/2020 Acceptance:10/10/2020 Knowledge, Attitudes, and Practices Regarding Cervical Cancer Screening among HIV-infected Women at Srinagarind Hospital: A Cross-Sectional Study Athiwat Songsiriphan1, Lingling Salang1, Woraluk Somboonporn1, Nuntasiri Eamudomkarn1, Wilasinee Nhokaew1, Chusri Kuchaisit2, Pornnipa Harnlakorn3 Abstract Introduction: In recent years, the lives of HIV-infected patients in Thailand have improved significantly due to continuous advances in treatment. However, the rate of cancer related to HIV infection (especially cervical cancer) is likely to increase. Although the World Health Organization (WHO) recommends Papanicolaou testing in all HIV-infected women, few of these patients receive this kind of screening in Thailand. Therefore, we conducted this study to evaluate the knowledge, attitudes, and practices of these patients with regard to cervical cancer screening. Materials and Methods: This cross-sectional study was conducted in HIV-infected women aged 18-65 years from April to November 2019 via a self-administered cervical cancer screening questionnaire, which consisted of four parts: demographic data, knowledge, attitudes, and practices. Results: Three hundred HIV-infected women were recruited. Most of the participants had good attitudes toward screening and practiced adequate screening (75.3% and 71.3%, respectively). However, only 62 participants (20.7%) demonstrated adequate knowledge. The crucial factors that were associated with adequate screening practice were age 40-49 years-old (AOR =3.26, 95%CI=1.02-10.37), CD4 cell count (AOR = 3.41, 95%CI = 1.29-8.99), having been advised about cervical cancer screening (AOR= 6.23, 95%CI 1.84-21.07), and attitude toward screening (AOR= 5.7, 95%CI = 2.23-14.55). The major reasons for not undergoing screening were embarrassment (41.86%), lack of symptoms (41.86%), fear of the results (36.04%), and fear of pain (36.04%). Conclusion: The reasons for inadequate testing were disregard and misconceptions about the procedure. To prevent invasive cervical lesions in HIV-infected women, health care providers should inform these patients about the importance of regular cervical cancer screening. Keywords: Knowledge- attitudes- practices - cervical cancer screening- HIV-infected women Asian Pac J Cancer Prev, 21 (10), 2979-2986 Introduction Human immunodeficiency virus (HIV) is a transmittable disease that affects the immunity of its host. Currently, there are many anti-retroviral drugs available to decrease viral load including highly active antiretroviral therapy (HARRT). However, while this regimen is effective in decreasing viral load, it does not reduce the risk of other long-term complications such as malignancy-related HIV infection (Kietpeerakool, 2005). Women who have coinfection with HIV and human papillomavirus (HPV) have a greater chance of cervical transformation. Previous studies have found that HIV-infected women had a 6.7 times greater risk of abnormal cytological test results and four times greater risk of cervical cancer than those without HIV infection (Vafaei et al., 2015; Abraham et al., 2013). In Thailand, cervical cancer was the HIV-related malignancy with the highest incidence (Kiertiburanakul et al., 2007). A previous study found the prevalence rates of HPV infection, abnormal cytological test results, and cervical cancer in these patients to be 38.6%, 20.4%, and 1.9%, respectively (Sirivongrangson et al., 2007), suggesting that HIV-infected women had an increased risk of cervical cancer. Despite this, another study found that only 14.7% of HIV-infected women undergo Pap smear screening (Chen et al., 2013). According to GLOBOCAN (2018), cervical cancer was the third most common cancer among women worldwide. In 2018, there were 569,000 new cases of 1Department of Obstetrics and Gynecology, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. 2Retired Government Official, Nursing Division, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. 3Senior Professional level AIDs Unit, Nursing Division, Srinagarind Hospital, Faculty of Medicine, Khon Kaen University, Khon Kaen, Thailand. For Correspondence: [email protected] 2979 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.10.2979Knowledge, Attitudes, and Practices regarding Cervical Cancer Screeningcervical cancer, or 13.1 patients per 100,000, with 311,400 deaths (7.5% of all female cancer deaths; Bray et al., 2018). Previous studies conducted in Thailand found that cervical cancer was the second most common malignancy in women, with 8,622 new cases and an incidence of 16.2 patients per 100,000 in 2018. In total, 5,015 patients die of cervical cancer annually (an average of 13 patients per day; Chaowawanit et al., 2016.; Bray et al., 2018) Human papillomavirus is a sexually transmitted disease, some serotypes of which (especially HPV 16 and HPV 18) are major causes of cervical cancer. Although the immune response in most HPV-infected patients leads to remission, some patients experience persistent infection and develop cervical cancer within 10-20 years (Saslow et al., 2012; Walboomers et al., 1999). Because it has a long preinvasive period, invasive cervical cancer is considered a preventable disease. If a preinvasive lesion is detected and treated early, the risks of morbidity and mortality decrease significantly (Rukrungtam et al., 2017; Anderson et al, 2019). The aims of this study were to evaluate the knowledge, attitudes, and practices of HIV-infected women in Thailand with regard to cervical cancer screening and to evaluate the factors that affect whether or not these patients undergo screening. Materials and Methods This cross-sectional study was conducted in HIV-infected women at Srinagarind Hospital in Khon Kaen, Thailand. Prior to the study, a questionnaire was developed to evaluate knowledge, attitudes, and practices with regard to cervical cancer screening, which was then tested in 20 HIV-infected women who sought medical care at Srinagarind Hospital’s Infectious Disease Clinic as a pilot study. The final questionnaire was validated by three gynecologic oncologists (not involved in the study). The study, conducted from April to November 2019, was approved by the Human Research Ethics Committee of Khon Kaen University as per the Helsinki Declaration and Good Clinical Practice Guidelines (HE621021). The primary endpoint of this study was to evaluate knowledge, attitudes, and practices of HIV-infected women in Thailand with regard to cervical cancer screening. A previous study by Chaowawanit et al., (2016) found that 26.4% of women had good attitudes about screening and adequate knowledge. Another study by Budkaew et al., (2014) found that 32.3% of women underwent cervical screening. Assuming a non-response rate of 5%, we estimated that a sample size of 300 participants would be required. Women who had been diagnosed with HIV at least five years prior, were aged 18 to 65, underwent follow-up examination at Srinagarind Hospital’s Infectious Disease Clinic, and could read and write in Thai were included in this study. We excluded women with history of pre-invasive or invasive cervical lesions or other gynecologic cancers, hysterectomy for any reason, or HPV vaccination, or who were pregnant at the time of the study, or who could not provide the necessary information. An information sheet about the project was given 2980 to all participants who met the inclusion criteria. A verbal explanation was also provided if anything was unclear. All women gave verbal consent and completed a Thai-language questionnaire, either by themselves with assistance if necessary for comprehension. Any questions about the questionnaire were answered by our research assistants before proceeding. Questionnaire design The questionnaire was a revised version of that used in a study by Chaowawanit et al. (2016). It was validated by three experts (not involved in this study), and its reliability was tested in a pilot study with 20 patients. Its reliability for knowledge outcomes as calculated using the Kuder-Richarson coefficient (KR20) was 0.77 and for attitude outcomes according to Cronbach’s alpha was 0.84 (similar to the 0.78 found in a previous study by Charoenmak et al. (2013)). A KR-20 coefficient and Cronbach’s alpha equal to or greater than 0.7 and 0.75, respectively, were considered acceptable (Anderson et al., 2002; Feldt, 1965). Data collection The questionnaire was divided into four parts: 1) demographic data, 2) knowledge regarding cervical cancer screening, 3) attitudes about cervical cancer screening, and 4) screening practice. It required 30 minutes to complete and was self-administered in a private room. Research assistants collected the finished questionnaires and checked them for completion. There were 13 questions in the knowledge section, each worth 1 point. We used modified Bloom’s cut-off points (Narayana et al., 2017) to classify knowledge level, with scores of 0-9 indicating “inadequate knowledge” and 10-13 indicating “adequate knowledge.” The attitude section consisted of 10 questions, which were adapted from the Health Belief Model, each worth up to 4 points. A score of 10-30 was classified as a “poor attitude” and 31-40 as a “good attitude.” Screening practice was classified into three categories according to Thailand’s 2017 national guidelines on HIV/AIDS treatment and prevention: 1) adequate 2) inadequate, and 3) never. Adequate practice was defined as annual screening if CD4 levels are <500 cell/mm3 and cytological testing every two years and co-testing every 3 years if CD4 levels were ≥500 cell/mm3. Inadequate practice was defined as having undergone screening but less frequently than was considered adequate. Statistical analysis We used STATA/SE version 10.1 to analyze participants’ baseline characteristics, knowledge, attitudes, and practice in terms of mean, median, percentage, standard deviation (SD), and 95% confidence interval (95% CI). Univariate analysis using a chi-square test, Fisher’s exact, and t-test was conducted to evaluate the associations between socio-demographic characteristics and knowledge, attitudes, and practice. We performed binary logistic regression analysis to predict knowledge, attitudes, and practice using binary and multinomial logistic regression. A p-value < 0.05 and odds ratio with Athiwat Songsiriphan et alAsian Pacific Journal of Cancer Prevention, Vol 21a 95% CI were used to determine any associations. Table 1. Sociodemographic Data Results Three-hundred HIV-infected women were recruited for this study. The baseline characteristics were presented in Table 1. The mean age of the participants was 45 ± 9.5 years, with most (68%) under 50 years of age. Most were married, and the average age at first intercourse was 22.2 years. The most common occupations were farmer, company employee, and public servant (35%, 29.6%, and 22.6% of participants, respectively). In terms of education, 46.3% held a bachelor’s degree or higher. Most were non-smokers (97%), Buddhist (98%), multiparous (44.9%), and had menstrual period (65.7%). Most (82.7%) used contraception such as oral contraceptive pills (OCPs) (47.3%), condoms (44.3%), and tubal resection (TR; 22.3%). The average duration since HIV diagnosis was 12.8±7.4 years. The mean CD4 level was 533.8±285.5 cell/mm3, and 148 (49.3%) participants had CD4 levels of less than 500 cell/mm3. The route of HIV infection in the majority of cases (90.3%) was sexual intercourse. Two hundred seventy-six of the participants (92%) had been advised about cervical cancer screening. Overall Evaluation of Knowledge, Attitude, and Practice The average knowledge and attitude scores were 6.39 ±3.18 points and 33.98 ±5.39, respectively. Although only 62 participants (20.7%) demonstrated adequate knowledge of cervical cancer screening, 226 (75.3%) had good attitude scores. Cervical cancer screening practice was adequate in 214 participants (71.3%), inadequate in 55 participants (18%), and 31 (10.3%) had never undergone screening (Table 2). Factors associated with Attitude and Knowledge High level of education, more pregnancies, contraception use, and high income were significantly correlated with adequate knowledge. However, only education level and number of pregnancies remained after multivariable regression analysis. Women with a bachelor’s degree or higher were 2.24 times more likely to have adequate knowledge than those without (adjusted odds ratio [AOR]= 2.41; 95%CI = 1.4-4.4). A higher number of parities was also significantly associated with adequate knowledge (parity = 1; AOR = 4.1, 95%CI = 1.5 - 11.23 and parity ≥ 2; AOR = 3.67, 95%CI =1.37- 9.83). However, high income and contraceptive use were less likely to be associated (AOR= 1.67, 95%CI = 0.74- 3.76; AOR = 2.24, 95%CI = 0.82-6.981, respectively; Table 3). Number of parities and history of sexual intercourse were significantly associated with attitude toward cervical cancer screening according to multivariable regression analysis. Women who had been pregnant at least once were about 2-3 times more likely to have a good attitude than those who had not (parity = 1: AOR = 3.39, 95%CI = 1.55-7.42; parity ≥ 2: AOR = 2.11, 95%CI = 1.07-4.19), and those with a history of sexual intercourse were 3.36 times more likely than those who had never had sexual Characteristics Mean age, year (2SD) Marital status Single, n (%) Married, n (%) Divorced, n (%) Occupation Public servant, n (%) Employee, n (%) Unemployed/Housewife, n (%) Farmer, n (%) Education level Primary school, n (%) High school, n (%) Bachelor’s degree, n (%) Master’s degree or higher, n (%) Age <50 years n (%) ≥50 years n (%) Smoking status Yes n (%) No n (%) Religion Buddhist, n (%) Christian, n (%) Others, n (%) Income (Baht)/month <10,000, n (%) 10,000-20,000, n (%) 20,001-30,000, n (%) ≥30,001, n (%) Parity 0, n (%) 1, n (%) 2, n (%) 3, n (%) ≥4, n (%) Menstruation Yes, n (%) No, n (%) Sexual intercourse Yes, n (%) Average age at first intercourse Median number of partners Contraception history Yes, n (%) Condom, n (%) Oral contraceptive pills, n (%) Tubal ligation, n (%) Data 45 (± 9.5) 76 (26.3) 162 (54) 59 (19.7) 78 (22.6) 89 (29.6) 27 (9) 106 (35) 49 (16.3) 110 (36.6) 120 (40) 19 (6.3) 204 (68) 96 (32) 9 (3) 291 (97) 294 (98) 4 (1.3) 2 (0.67) 87 (29) 96 (32) 39 (13) 78 (26) 72 (24) 93 (31) 97 (32.3) 31 (10.3) 7 (2.3) 197 (65.7) 103 (34.3) 286 (95.3) 22.2 (years) 2 partners 248 (82.7) 133 (44.3) 142 (47.3) 67 (22.3) 2981 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.10.2979Knowledge, Attitudes, and Practices regarding Cervical Cancer ScreeningTable 1. Continued Characteristics HIV infection history Data Duration of diagnosis, Mean ± SD 12.8±7.4 CD4 level, Mean ± SD CD<500 cell/mm3, n (%) CD≥500 cell/mm3, n (%) Route of infection Sexual intercourse, n (%) Blood, n (%) Others, n (%) Advised about cervical cancer screening Yes, n (%) No, n (%) 533.8±285.5 148 (49.3) 152 (50.7) 271 (90.3) 20 (6.7) 9 (3) 276 (92) 24 (8) Table 2. Knowledge, Attitudes, and Practice Regarding Cervical Cancer Screening in HIV-Infected Women characteristics 95% confident interval N (%) Knowledge Knowledge (MD±SD) 6.39 ±3.18 Poor [0-9], n (%) 238 (79.3) 6.02-6.75 74.3-83.77 Good [10-13], n (%) 62 (20.7) 16.23-25.70 Attitude Attitude (MD±SD) 33.98 ±5.39 Poor [10-30], n (%) 74 (24.7) Good [31-40], n (%) 226 (75.3) 33.37-34.60 21.49-29.95 70.05-80.11 Practice Adequate, n (%±SD) 214 (71.3±2.6) 66.19-76.48 Inadequate, n (%±SD) 55 (18±2.2) 13.93-22.73 Never, n (%±SD) 31 (10.3±1.8) 6.87-13.80 intercourse (AOR = 3.36, 95%CI=1.07-10.67). Factors associated with Cervical Cancer Screening Practice The associated factors related to cervical cancer screening practice were age, parity, mean age at first intercourse, CD4 level, whether or not the patient had been advised regarding cervical cancer screening, and attitude toward screening. Participants aged 40-49 years were 3.26 times more likely to practice adequate screening compared to women < 40 years old (AOR = 3.26, 95%CI = 1.02-10.37), while women ≥ 50 years of age were 2.35 times more likely than those under 40 (AOR = 2.35, 95%CI = 0.67-8.32). However, there were no significant differences in the proportions of those practicing inadequate screening versus never undergoing screening by age group. Although there was no statistically significant association between parity and cervical cancer screening practice, those who had been pregnant once or were multiparous more likely to practice adequate screening than those who were nulliparous (AOR = 3.33, 95%CI = 0.96-11.49 and AOR =2.02, 95%CI = 0.71-5.78, respectively). They were also more likely to practice Table 3. Association with Knowledge and Attitude Regarding Cervical Cancer Screening Variable Outcome Crude Odds ratio (95%CI) Adjusted Odds ratio (95%CI) Knowledge of cervical cancer screening Education Poor n (%) Good N (%) < Bachelor’s degree 31 (16.40) 158 (83.60) 1 1 ≥ Bachelor’s degree 31 (27.93) 80 (72.7) 1.96 (1.12 - 3.48) 12.24 (1.4 - 4.4) Number of parities 0 1 ≥ 2 Income 6 (8.3) 24 (25.8) 32 (23.7) 66 (91.7) 69 (74.2) 1 3.83 (1.47 - 9.95) 103 (76.3) 3.41 (1.36 - 8.62) 1 4.1 (1.5 - 11.23) 3.67 (1.37 – 9.83) < 10,000 baht/month 11 (12.6) 76 (76.4) 1 1 ≥ 10,000 baht/month 51 (23.9%) 162 (76.2) 2.16 (1.07 - 4.41) 11.67 (0.74 - 3.76) Contraception use No Yes Attitude of cervical cancer screening Number of parities 0 1 ≥ 2 History of sexual intercourse No Yes 2982 5 (9.6) 57 (23) 47 (90.4) 191 (77) 1 1 2.81 (1.07 -7.39) 2.24 (0.82 – 6.18) 46 (61.1) 79 (85) 28 (38.9) 14 (15) 1 3.59 (1.71 – 7.53) 103 (76.3) 32 (23.7) 2.04 (1.1 – 3.8) 1 3.39 (1.55 – 7.42) 2.11 (1.07 – 4.19) 7 (50) 7 (50) 1 1 219 (76.6) 67 (23.4) 3.27 (1.11 – 9.65) 3.36 (1.07 – 10.67) Athiwat Songsiriphan et alAsian Pacific Journal of Cancer Prevention, Vol 21Table 4. Factors Associated Cervical Cancer Screening Practice Variable Practice on cervical cancer screening Adequaten (%) Inadequate n (%) Never n (%) Crude Odds ratio1 (95%CI) Adjusted Odds ratio1 (95%CI) Crude Odds ratio2 (95%CI) Adjusted Odds ratio2 (95%CI) Age < 40 40-49 ≥ 50 Age of first intercourse < 20 ≥20 CD4 level 41 (61.2) 12 (17.9) 14 (20.9) 1 1 1 1 105 (76.6) 24 (17.5) 8 (5.8) 4.48 (1.75–11.48) 3.26 (1.02-10.37) 3.5 (1.15-10.63) 3.01 (0.89-10.21) 86 (70.8) 19 (19.8) 9 (9.4) 2.58 (1.03-6.49) 2.35 (0.67-8.32) 2.46 (0.81-7.44) 2.48 (0.64-4.96) 89 (57.4) 23 (18) 21 (14.2) 1 1 1 1 129 (84.9) 32 (18.7) 10 (6.6) 1.49 (1.43-7.1) 1.10 (1.01-1.19) 1.48 (0.61-3.59) 0.89 (0.31-.252) < 500 cell/mm3 85 (57.4) 42 (28.4) 21 (14.2) 1 1 1 1 ≥ 500 cell/mm3 129 (84.9) 13 (8.6) 10 (6.6) 3.19 (1.43-7.1) 3.41 (1.29-8.99) 0.65 (0.24-1.72) 0.60 (0.21-1.79) Number of parities 0 1 ≥ 2 40 (55.6) 17 (23.6) 15 (20.8) 1 1 1 1 74 (79.6) 14 (15.1) 5 (5.4) 5.55 (1.87-16.39) 3.33 (0.96-11.49) 2.47 (0.71-8.49) 2.27 (0.61-8.5) 100 (74.1) 24 (17.8) 11 (8.2) 3.41 (1.44-8.06) 2.02 (0.71-5.78) 1.93 (0.71-5.21) 1.56 (0.51-4.79) Advised about cervical cancer screening No Yes 9 (37.5) 6 (25) 9 (37.5) 1 1 1 1 205 (74.3) 49 (17.8) 22 (8) 9.32 (3.35-25.92) 6.23 (1.84-21.07) 3.34 (1.6-10.54) 2.62 (0.75-9.19) Attitude regarding cervical cancer screening Poor Good 32 (43.2) 25.3 (33.8) 17 (23) 1 1 1 1 182 (80.5) 30 (13.3) 14 (6.2) 6.91 (3.1-15.38) 5.7 (2.23-14.55) 1.46 (0.6-3.53) 1.35 (0.51-3.59) Crude Odds ratio1, Compared adequate with never screening; Crude Odds ratio2, Compared Inadequate with never screening; Adjusted Odds ratio1, Compared adequate with never screening; Adjusted Odds ratio2, Compared Inadequate with never screening inadequate screening than to not undergo screening. There was a significant difference in terms of mean age at first intercourse between women who practiced adequate screening and those who had never been screened. Those who first engaged in intercourse at ≥ 20 years old were 1.10 times more likely to practice adequate versus inadequate screening than those who did so when they were < 20 (AOR= 1.10, 95%CI =1.01-1.19). However, there was no statistically significant correlation between age at first intercourse and inadequate versus no screening (AOR=0.89, 95%CI =0.31-2.52). Participants with CD4 levels greater than 500 cell/ mm3 were 3.41 times more likely to practice adequate versus no screening than women with CD4 levels < 500 cell/mm3 (AOR = 3.41, 95%CI=1.29-8.99), but there were slightly significant difference when comparing between inadequate practice and never practice (AOR = 0.61, 95%CI=0.21-1.79) . Not having been advised about cervical cancer screening was a major factor associated with non-adherence in terms of both adequate screening versus never having been screened (AOR = 6.32, 95%CI = 1.84-21.07) and inadequate screening versus never having been screened (AOR = 2.62, 95%CI = 0.75-9.19). Women with good attitudes toward screening were 5.7 times more likely to practice adequate versus no screening compared with those who had poor attitudes (AOR = 5.7, 95% CI = 2.23 – 14.55). Conversely, those with good attitudes were less likely to practice inadequate versus no screening than those with poor attitudes (AOR = 1.35, 95%CI=0.51-3.59). By contrast, women with a history of pregnancy were less likely than nulliparous women to practice adequate versus no screening (AOR = 3.33, 95%CI = 0.96-11.49). This was also true for multiparous versus nulliparous women (AOR=2.20, 95%CI = 0.71-5.78) in comparing in both adequate versus no screening and inadequate versus no screening, as shown in Table 4. Reasons for inadequate cervical cancer screening The 86 participants who practiced inadequate screening gave the following reasons: embarrassment, lack of symptoms, fear of pain, fear of the results, feeling Table 5. The Attitude or Reasons why Women have Inadequate Cervical Cancer Screening Attitude or reason for ignoring screening Number (%) Self-perception (answer more than 1) Unnecessary No risk Lack of symptom Fear of pain Embarrassment Fear of the results 22(25.58) 25(29.07) 36 (41.86) 31 (36.04) 36 (41.86) 31 (36.05) Health care-provider (answer more than 1) Bad impression with health services 11 (12.79) Procedure (answer more than 1) Expensive cost of screening 24 (27.91) 2983 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.10.2979Knowledge, Attitudes, and Practices regarding Cervical Cancer Screeningthey were not at risk, cost, feeling that screening was unnecessary, and having a bad impression of health services (41.86%, 41.86%, 36.04%, 36.04%, 29.07%, 27.91%, 25.58%, and 12.79%, respectively; Table 5). Discussion Only 20.7% of the 300 HIV-infected women in our study had sufficient knowledge about cervical cancer screening. However, 75.3% of participants had good attitude scores and 71.3% practiced adequate cervical screening. These results are similar to those of a study by Olivia et al., (2016). in which only 21.6% of HIV-infected women demonstrated adequate knowledge of cervical cancer screening. By contrast, a study in Kenya found that more than 90% of participants knew about cervical cancer screening (Rosser et al., 2015). Although most of our participants demonstrated inadequate knowledge, they were aware of the factors that increase the risk of cervical cancer (HIV infection, sexual intercourse, smoking, and parity). However, most (65%) misunderstood that cervical cancer patients have prodrome symptoms before the disease reaches an advanced stage. In addition, only 35% knew that the HPV vaccine was recommended to prevent cervical cancer, which is higher than in Nigeria (3.1%) but lower than in Belgium (50%) (Donders et al., 2008; Rabiu et al., 2011). The factors that were associated with adequate knowledge holding a bachelor’s degree or higher, income > 10,000 baht per month, multiparity, and contraception use, a finding that is consistent with those of a previous study conducted in Laos (Sichanh et al., 2014). Seventy-five of participants had high attitude scores, most of whom (83%) indicated that they were satisfied with their routine cervical cancer screening, and 70% of whom strongly desired screening. Four of the major reasons for inadequate screening were embarrassment, lack of symptoms, fear of pain, and fear of the test results. Approximately 80% of HIV-infected women were aware that they were at greater risk of developing cervical cancer, which is higher than in a study conducted Songkla province in southern Thailand (66%) (Charoenmak et al., 2013). These results were similar to those of a study conducted in Laos, which found that awareness of cervical cancer was four times higher in HIV-infected women than in those without HIV (Sichanh et al., 2014). In a previous study by Rukrungtan et al., (2017) 71% of participants practiced adequate cervical screening according to Thailand’s national guidelines regarding HIV/AIDS treatment and prevention, 18% practiced inadequate screening (low frequency), and 10% had never undergone screening. However, the previously mentioned study in Songkhla found that 83% of participants practiced adequate screening. The reason for this disparity may be that frequency of screening was not included in the previous study’s analysis (Charoenmak et al., 2013). Other studies conducted in Canada, Italy, and the United States found the percentage of HIV-infected women who received adequate screening to be 58%, 61%, and 77%, respectively (Maso et al., 2010; Oster et al. 2009; Pamela et al., 2010). However, studies in Ethiopia and 2984 Laos found the rate of adequate screening to be less than 25% (Erku et al., 2017; Nega et al., 2018; Sichanh et al., 2014). These differences may be due to variation in sociocultural, educational, or economic conditions. In addition, non-HIV-infected women in Thailand appear to have lower rates of adequate screening than those with HIV. Studies in northeast Thailand and Bangkok found that only 41% and 42%, respectively, of non-HIV-infected women underwent adequate screening (Chaowawanit et al., 2016; Mongsawaeng et al., 2016). We found that the women who had children were 1.8 times more likely to be screened for cervical cancer. Similar results were found in Ethiopia, in which women who had children were three times more likely to be screened. This could be due to women with children visiting healthcare facilities more often than those without (Budkaew and Chumworathayi, 2014). In addition, HIV-infected women with CD4 ≥500 cell/mm3 were three times more likely to undergo adequate screening than those with CD4 <500 cell/mm3. This may be due to higher CD4 count leading to greater adherence to HIV treatment and cervical cancer screening (Suwannanobon et al., 2018). Participants who had been advised about cervical cancer screening were nine times more likely to undergo adequate screening compared with those who had not. Women with high attitude scores were nearly seven times more likely to undergo adequate screening. This suggests that all health care providers, especially gynecologists and those in infectious disease clinics, should inform HIV-infected women about their greater risk, reasons to undergo screening, and screening methods. They should also work to clear up any misperceptions regarding cervical cancer screening, such as those involving embarrassment, fear of pain, and fear of test results, and encourage these patients to undergo screening regularly. Strengths and limitations To our knowledge, this is the first study in Thailand to evaluate the knowledge, attitudes, and practices regarding cervical cancer screening among HIV-infected women. In addition, this was a large trial study that revealed the major reasons that HIV-infected women did not undergo cervical cancer screening. One limitation of this study was that the data may not represent all HIV-infected women in Thailand. Moreover, data were gathered via a self-administered questionnaire and not verified through medical records, which may have resulted in recall bias with regard to screening practice. Implications for practice and further research Our results suggest that many HIV-infected women undergo inadequate screening, mainly due to misperceptions regarding cervical cancer. This data may be useful in the development of government policy aimed at educating HIV-infected women with regard to screening in order to reduce the incidence of invasive cervical cancer. In conclusion, we found that 71% of participants underwent adequate cervical cancer screening and that the factors that associated with screening were parity, CD4 levels, whether or not patients had been advised Athiwat Songsiriphan et alAsian Pacific Journal of Cancer Prevention, Vol 21about screening, and attitude toward screening. In order to promote cervical cancer screening in HIV-infected women, all healthcare providers should inform their patients about screening, correct any misconceptions, and ensure that screenings are being performed. Routine cervical cancer screening in HIV clinics should be included in a national prevention program in order to ensure that precancerous cervical lesions are detected early and appropriate treatment can be performed. Author contributions All of the authors participated in the writing of this paper and have read the finished manuscript. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments The authors would like to thank Dylan Southard for assistance with the English-language presentation of the manuscript. We would like to thank prof. James A. for editing this MS via the KKU Publication Clinic (Thailand). In addition, we would like to thank the Khon Kaen University Faculty of Medicine Research and Development Fund (Thailand). Potential conflicts of interest The authors have no conflicts of interest. References Anderson DL, Fosher KM, Norman GJ (2002). Development and evaluation of the conceptual inventory of natural selection. J Res Sci Tech, 39, 952-78. Anderson DM, Lee J, John C (2019). Cervical and vaginal cancer. In Berek JS (eds). Berek & Novak’s gynecology 16th edition. Philadelphia; Wolters Kluwer, China, pp 1038-66. Bray F, Ferlay J, Soerjomataram I, et al (2018). Global Cancer Statistics 2018: GLOBOCAN Estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin, 68, 394-424. Budkaew J, Chumworathayi B (2014). Factors associated with decision to attend cervical cancer screening women aged 30-60 years in Chatapadung contracting medical unit, Thailand. Asian Pac J Cancer Prev, 15, 4903-7. Chaowawanit W, Tabgjitgamol S, Kantathavorn N, et al (2016). Knowledge, attitude and behavior of Bangkok metropolitian women regarding cervical cancer screening. Asian Pac J Cancer Prev, 17, 945-52. Charoenmak B, Siripaitoon P, Hortiwakul T (2013). The uptake rate and patient perception of cervical cancer screening in HIV-infected women attending infectious disease, medicine clinic, Songklanagarind hospital. Songkla Med J, 31, 1-9. Chen Y-C, Liu H-Y, Li C-Y, et al (2013). Low Papanicolaou smear screening rate of women with HIV infection: a nationwide population-based study in Taiwan. J Womens Health, 22, 1016–22. Donders GG, Gabrovska M, Bellen G, et al (2008). Knowledge of cervix cancer, human papilloma virus (HPV) and HPV vaccination at the moment of introduction of the vaccine in women in Belgium. Arch Gynecol Obstet, 277, 291-8. Erku DA, Netere AK, Mersha AG, et al (2017). Comprehensive knowledge and uptake of cervical cancer screening is low among women living with HIV/AIDS in n Northwest Ethiopia. Gynecol Oncol Res Pract, 4, 20-7. Feldt L (1965). The approximate sampling distribution of Kuder-Richarderson reliability coefficient twenty. Phychometrika, 30, 357-70. Kietpeerakool C (2005). Cervical intraepithelial neoplasia and cancer in HIV – Epidermic era. Srinagarind Med J, 20, 48-54. Kiertiburanakul S, Likhitpongwit S, Ratanasiri S, Sungkanuparph S (2007). Malignancies in HIV-infected Thai patients. HIV Med, 8, 322-3. Maso LD, Franceschi S, Lise M, et al (2010). Self-reported history of pap-smear in HIV-positive women in northern Italy: a cross-sectional study. BMC Cancer, 10, 310-7. Mongsawaeng C, Kokorn N, Kujapun J, et al (2016). Knowledge, attitude, and practice regarding cervical cancer among rural community women in Northeast Thailand. Asian Pac J Cancer Prev, 17, 85-88. Narayana G, Sychitra MJ, Sunanda G, et al (2017). Knowledge, attitude, and practice regarding cervical cancer among women attending obstetrics and Gynacology Department: A cross-sectional, hospital-based survey in South Indian. Indian J Cancer, 54, 481-7. Nega AD, Woldetsadik MA, Gelagay AA (2018). Low uptake of cervical cancer screening among HIV positive women in Gondar university referral hospital. Northwest Ethiopia. BMC Womens Health, 18, 87-94. Olivia M, Ramalivhana NJ, Kekana M, Augustine N, Maxwell M (2016). Knowledge, attitudes and practices of HIV-infected women on cervical cancer screening at a Church-affiliated hospital. IOSR -JDMS, 15, 119-26. Oster AM, Sullivan PS, Blair JM (2009). Prevalence of cervical cancer screening of HIV-infected women in the United States. J Acquir Immune Defic Syndr, 51, 430-6. Pamela L, Claire K, Claire T, Kevin P, Jonathan BA (2010). Cervical cancer screening among HIV-positive women retrospective cohort study from a tertiary care HIV clinic. Can Fam Physician, 56, 425–31. Rabiu KA, Akinbami AA, Adewunmi AA , Akinola OI, Wright KO (2011). The need to incorporate routine cervical cancer counselling and screening in the management of HIV positive women in Nigeria. Asian Pac J Cancer Prev, 12, 1211–4. Rosser JI, Njoroge B, Huchko MJ (2015). Cervical cancer screening knowledge and behavior among women attending an urban HIV clinic in western Kenya. J Cancer Educ, 30, 567-72. Rukrungtam K, Poothanakit T, Puthajareon O, et al (2017). Thailand national guidelines on HIV/AIDS treatment and prevention. Sex Transm Dis, 34, 104-7. Saslow D, Solomom D, Lawson HW, et al (2012). American cancer society, American society for colposcopy and cervical pathology, and American society for clinical pathology screening guidelines for the prevention and early detection of cervical cancer. CA Cancer J Clin, 62, 147-72. Sichanh C, Quet F, Chanthavilay P, et al (2014). Knowledge, awareness and attitudes about cervical cancer among women attending or not an HIV treatment center in Lao PDR. BMC Cancer, 14, 161-73. Sirivongrangson P, Bollen JL, Chaovavanich A, et al (2007). Screening HIV-infected women for cervical cancer in Thailand. Sex Transm Dis, 34, 104-7. Suwannanobon N, Anansawad S, Jaiaia R (2018). Success factors and barriers to develop the caring system for people 2985 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.10.2979Knowledge, Attitudes, and Practices regarding Cervical Cancer Screeningliving with HIV and AIDS. J Royal Thai Army Nurses, 17, 70-8. Vafaei H, Asadi N, Foroughinia L, et al (2015). Comparison of abnormal cervical cytology from HIV positive women, female sex workers and general population. Int J Community Based Nurs Midwifery, 3, 76-83. Walboomers JM, Jacobs MV, Manos MM, et al (1999). Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol, 189, 12-19. This work is licensed under a Creative Commons Attribution- Non Commercial 4.0 International License. 2986 Athiwat Songsiriphan et alAsian Pacific Journal of Cancer Prevention, Vol 21
10.2196_20920	JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Original Paper Evaluating Behavioral and Linguistic Changes During Drug Treatment for Depression Using Tweets in Spanish: Pairwise Comparison Study Angela Leis, PsyM; Francesco Ronzano, PhD; Miguel Angel Mayer, MD, MPH, PhD; Laura I Furlong, PhD; Ferran Sanz, Prof Dr Research Programme on Biomedical Informatics, Hospital del Mar Medical Research Institute, Department of Experimental and Health Sciences, Pompeu Fabra University, Barcelona, Spain *these authors contributed equally Corresponding Author: Ferran Sanz, Prof Dr Research Programme on Biomedical Informatics Hospital del Mar Medical Research Institute Department of Experimental and Health Sciences, Pompeu Fabra University Barcelona Biomedical Research Park Carrer Dr Aiguader 88 Barcelona, 08003 Spain Phone: 34 933160540 Fax: 34 933160550 Email: [email protected] Abstract Background: Depressive disorders are the most common mental illnesses, and they constitute the leading cause of disability worldwide. Selective serotonin reuptake inhibitors (SSRIs) are the most commonly prescribed drugs for the treatment of depressive disorders. Some people share information about their experiences with antidepressants on social media platforms such as Twitter. Analysis of the messages posted by Twitter users under SSRI treatment can yield useful information on how these antidepressants affect users’ behavior. Objective: This study aims to compare the behavioral and linguistic characteristics of the tweets posted while users were likely to be under SSRI treatment, in comparison to the tweets posted by the same users when they were less likely to be taking this medication. Methods: In the first step, the timelines of Twitter users mentioning SSRI antidepressants in their tweets were selected using a list of 128 generic and brand names of SSRIs. In the second step, two datasets of tweets were created, the in-treatment dataset (made up of the tweets posted throughout the 30 days after mentioning an SSRI) and the unknown-treatment dataset (made up of tweets posted more than 90 days before or more than 90 days after any tweet mentioning an SSRI). For each user, the changes in behavioral and linguistic features between the tweets classified in these two datasets were analyzed. 186 users and their timelines with 668,842 tweets were finally included in the study. Results: The number of tweets generated per day by the users when they were in treatment was higher than it was when they were in the unknown-treatment period (P=.001). When the users were in treatment, the mean percentage of tweets posted during the daytime (from 8 AM to midnight) increased in comparison to the unknown-treatment period (P=.002). The number of characters and words per tweet was higher when the users were in treatment (P=.03 and P=.02, respectively). Regarding linguistic features, the percentage of pronouns that were first-person singular was higher when users were in treatment (P=.008). Conclusions: Behavioral and linguistic changes have been detected when users with depression are taking antidepressant medication. These features can provide interesting insights for monitoring the evolution of this disease, as well as offering additional information related to treatment adherence. This information may be especially useful in patients who are receiving long-term treatments such as people suffering from depression. (J Med Internet Res 2020;22(12):e20920) doi: 10.2196/20920 http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al KEYWORDS depression; antidepressant drugs; serotonin uptake inhibitors; mental health; social media; infodemiology; data mining Introduction Background Depression is one of the most common mental disorders [1]. According to the World Health Organization, depression affects more than 322 million people of all ages globally, being a leading cause of disability worldwide [2]. The proportion of people with depression went up by around 18% between 2005 and 2015 [3]. This mental disorder constitutes a challenge for society and health care systems due to devastating personal and social consequences and the associated economic costs [4-13]. In spite of the high prevalence of depression and the efforts of health care services to improve its management, this health condition remains underdiagnosed and undertreated [14]. In the case of moderate and severe forms of depression, pharmacological treatment can improve the quality of life of these patients [4]. There are several types of antidepressant drugs, and among them, selective serotonin reuptake inhibitors (SSRIs) are currently the most prescribed antidepressants around the world. For instance, according to the Spanish Agency for Medicines and Health Products [15], SSRIs constitute more than 70% of all antidepressants prescribed in Spain. They have fewer side effects than other antidepressants [16], show a good risk-benefit ratio [17,18], are safer and better tolerated [19], and exhibit a reduced risk of toxicity in overdose in comparison to tricyclic antidepressants [20]. They are commonly used as first-line treatment for depression [21-23] and are usually prescribed as maintenance therapy to prevent relapse [4,23-26]. SSRIs include the following drugs: fluvoxamine, fluoxetine, paroxetine, sertraline, citalopram, and escitalopram [17]. Furthermore, although social media platforms have typically not been created with health-related purposes in mind [27,28], millions of people publicly share personal health information on social media platforms every day [29,30]. For this reason, these platforms represent an important source of health information that is faster and more broadly available than other sources of health information, being unsolicited, spontaneous, and up to date. Infodemiology approaches have been developed and applied to better understand the dynamics of these platforms when used as a health information source [31-33]. In this context, social media users share health-related information, such as experiences with prescribed drugs [34], cancer patients’ sentiments [35], opinions on vaccines [36], or online conversations on epidemic outbreaks [37]. The massive data from social media can be monitored and analyzed by using natural language processing and machine learning technologies, providing new possibilities to better understand users’ behavior [30], including automatic identification of early signs of mental disorders [38-40]. In particular, it is typical for people suffering from depression to talk about their illness and the drugs they are taking [41-43]. amount of data that can collected in real time [28,30,33,45-48]. Twitter users post short messages about facts, feelings, and opinions, including about health conditions [49]. in Mining of drug-related information from Twitter has been applied the pharmacovigilance field [27,50]. Some pharmacovigilance studies carried out on Twitter studied specific cohorts by identifying users’ mentions of drug intake [37,51-53]. Other studies focused on adverse drug reactions, analyzing users’ tweets regarding adverse events and side effects associated with drug use, which were identified by means of generic or brand names [29,47,54,55]. In our previous study [49], we observed that Twitter users who are potentially suffering from depression show particular behavioral and linguistic features in their tweets. These features were related to an increase in their activity during the night, a different style of writing with increased use of the first-person singular pronoun, fewer characters in their tweets, an increase in the frequency of words related to sadness and disgust emotions, and more frequent presence of negation words and negative polarity. This information can be used as a complementary tool to detect signals of depression and for monitoring and supporting patients using Twitter. Objectives In this paper, we aim to enrich our previous study [49] by focusing on analysis of the changes in behavioral and linguistic features of Twitter users in Spanish language, which may be associated with the antidepressant medication these users are taking. It is worth mentioning that users from Spanish-speaking countries are among the most active on Twitter in the world [56]. The study is focused on Twitter users who mention treatment with SSRIs, which are the most frequently prescribed antidepressants [15]. In particular, this study compares the characteristics of the tweets posted while users were probably taking SSRIs versus the tweets posted by the same users when they have a lower probability of taking this antidepressant medication. This analysis can contribute to better understanding how these drugs affect users’ mood. Although we found two additional studies describing changes in Twitter users’ language in some mental disorders [57,58], to the best of our knowledge, there are no other studies that analyze Twitter posts in Spanish language to detect behavioral and linguistic changes when the users are taking antidepressant medication. Methods Study Design This study was designed with the aim of analyzing the behavioral patterns and linguistic features of users who mention SSRIs in their Twitter timeline. The study was developed in several steps and focused on tweets written in Spanish. The flow diagram of the study is depicted in Figure 1. Twitter is a very popular microblogging platform with more than 330 million active users worldwide [44]. Tweets, freely available in almost 90% of users’ accounts, provide a huge As shown in Figure 1, two nonoverlapping datasets of tweets from users mentioning treatment with SSRIs were obtained: (1) The in-treatment tweets dataset was made up of the tweets http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al posted throughout the 30 days after the publication date of any tweet mentioning SSRI intake. We assumed that these tweets were posted while the users had a high probability of being in treatment with an SSRI. (2) The unknown-treatment tweets dataset was made up of the tweets that were posted more than 90 days before or more than 90 days after the publication date of any tweet mentioning SSRI intake. We assumed that these tweets were posted while users had a lower probability of being in treatment with an SSRI than in the previous dataset. These datasets were designed in a way that made it possible to carry out intrasubject comparisons, since the in-treatment tweets and unknown-treatment tweets datasets were obtained from the same Twitter users. The strategy for the selection of the tweets included in the two datasets is depicted in Figure 2. Figure 1. Flow diagram of the study process. SSRI: selective serotonin reuptake inhibitor. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Figure 2. The in-treatment and unknown-treatment dataset selection strategy. SSRI: selective serotonin reuptake inhibitor. Data Collection and User Selection The selection of the tweets and their users was based on the filtered real-time streaming support provided by the Twitter application programming interface [59]. In the first step, we selected tweets in Spanish that mention any of the SSRI generic and brand names used around the world. To obtain the generic and brand names, we performed searches on the following databases and resources: DrugBank [60], the Anatomical Therapeutic Chemical Classification System and the Defined Daily Dose of the World Health Organization [61], Wikipedia [62], and the Database for Pharmacoepidemiological Research in Primary Care [63]. The list of 135 generic and brand names obtained is shown in Table 1. Table 1. Selective serotonin reuptake inhibitors (SSRIs) used in the study. Generic name Brand names Fluvoxamina (fluvoxamine) Dumirox, Faverin, Floxyfral, Fluvoxin, Luvox, Uvox Fluoxetina (fluoxetine) Paroxetina (paroxetine) Prozac, Reneuron, Adofen, Luramon, Sarafem Seroxat, Motivan, Frosinor, Praxil, Daparox, Xetin, Sertralina (sertraline) Citalopram (citalopram) Apo-oxpar, Appoxar, Aropax, Aroxat, Aroxat CR, Bectam, Benepax, Casbol, Cebrilin, Deroxat, Hemtrixil, Ixicrol, Loxamine, Meplar, Olane, Optipar, Oxetine, Pamax, ParadiseCR, Paradox, Paraxyle, Parexis, Paroxat, Paroxet, Paxan, Paxera, Paxil, Paxil CR, Pexot, Plasare, Pondera, Posivyl, Psicoasten, Rexetin, Seretran, Sereupin, Tiarix, Tamcere, Traviata, Xerenex, Xetroran Aremis, Besitran, Zoloft, Altisben, Aserin, Altruline, Ariale, Asertral, Atenix, Eleval, Emergen, Dominium, Inosert, Irradial, Sedora, Serolux, Sertex Seropram, Celexa, Akarin, C Pram S, Celapram, Celica, Ciazil, Cilate, Cilift, Cimal, Cipralex, Cipram, Cipramil, Cipraned, Cinapen, Ciprapine, Ciprotan, Citabax, Citaxin, Citalec, Citalex, Citalo, Citalopram, Citol, Citox, Citrol, Citta, Dalsan, Denyl, Elopram, Estar, Humorup, Humorap, Oropram, Opra, Pram, Pramcit, Procimax, Recital, Sepram, Szetalo, Talam, Temperax, Vodelax, Zentius, Zetalo, Cipratal, Zylotex Escitalopram (escitalopram) Cipralex, Diprex, Esertia, Essential, Heipram, Lexapro The following 7 brand names of medicines have been excluded due to their semantic ambiguity: Essential, Motivan, Estar, Traviata, Pondera, Recital, and Emergen. These commercial names are, at the same time, very common words used with different meanings in Spanish, as we verified after reviewing a random sample of 200 tweets with mentions of these words. The number of tweets excluded because of their semantic ambiguity was 21,104. In the manual check of a random sample of 200 tweets, the mentions of SSRIs when using these words were 0% (0/200) in some cases, such as for Motivan and Estar, and 0.5% (1/200) for Recital. The final list of words included 128 generic and brand names of SSRIs. Using the aforementioned 128 SSRI names, we collected 3651 tweets in Spanish posted during November 2019 with occurrences of the words listed in Table 1. These tweets were posted by 3138 different Twitter users and mentioned 33 different words from the list. The frequencies of these 32 words are shown in Table 2. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Table 2. Frequencies of SSRI names mentioned in Spanish tweets during November 2019. SSRI mentions Prozac Fluoxetina Sertralina Escitalopram Citta Citalo Paroxetina Pram Fluvoxamina Citalopram Seroxat Eleval Lexapro Opra Casbol Ariale Zoloft Altruline Paxil Akarin Heipram Aremis Cimal Tiarix Seretran Dominium Citox Atenix Aserin Talam Dalsan Celexa Frequency 998 756 542 248 210 109 69 49 40 33 22 21 20 18 14 11 9 9 7 7 4 4 3 2 2 2 2 2 2 1 1 1 In a second step, we crawled the public Twitter timelines of the 3138 users (until the 3200 most recent tweets for each user were retrieved). Given that retweets are not useful for analyzing the linguistic behavior of a particular user, the third step consisted of excluding the retweets and checking if the remaining tweets from each timeline included the mention of at least one SSRI. 1800 users were excluded by this filter, leaving a total of 1338 Twitter users. We obtained 3,791,609 tweets after compiling the timelines from these 1338 users. From these timelines, 4872 tweets mentioning at least one of the SSRIs from the list were automatically detected. These 4872 tweets were independently reviewed by two experts, a psychologist and a family physician, both with clinical experience. These experts manually selected the tweets that suggested that the user who posted the tweet was taking an SSRI on the date of posting. Examples of these tweets are shown in Textbox 1. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Textbox 1. Examples of tweets that positively or negatively suggest whether the user is taking an SSRI. Positive examples: • • “Eso de tener sueños raros debido a la fluoxetina se está saliendo de control.” (“Having odd dreams due to fluoxetine is getting out of control.”) “Yo tomo sertralina, como me lo receta el doctor y aún así a veces siento que el mundo donde estoy no es para mi. Ese susto esa angustia esas ganas de correr es algo que sólo el que lo padece lo entiende” (“I take sertraline as my doctor prescribes it to me and, even so, sometimes I feel that the world I’m living in is not for me. This fear this anxiety this desire to run out is something that only one who suffers from it can understand”) Negative examples: • • “Ella debería tomar prozac, como Tic Tac” (“She should take prozac, like Tic Tac” [a candy brand]) “La Paroxetina es un medicamento que pertenece a la familia de los antidepresivos inhibidores de la recaptación de la serotonina ¡Conoce más sobre él!” (“Paroxetine is a drug that belongs to the antidepressant family of serotonin reuptake inhibitors. Find out more about it!”) The agreement between reviewers was 93.1% (4537/4872) with a Cohen kappa score of 0.68, indicating that there was substantial agreement between raters. The reviewers discussed and reached a consensus on the classification of the 335 tweets they classified differently. Finally, we obtained a total number of 518 tweets with one or more SSRI mentions, suggesting that the users who posted these tweets were taking an SSRI at the moment of posting. These tweets corresponded to 279 different users. Therefore, these users had two characteristics: first, the tweets on their timeline included at least one mention of SSRIs, and second, the text of tweets mentioning SSRIs suggested that the user was taking the antidepressant. In addition, we analyzed the tweets posted by each user that belonged to the two datasets (in-treatment and unknown-treatment; see Figure 1) by trying different minimum numbers of tweets per dataset (10, 30, 60, and 100 tweets) in order to include a user in the study. 10 tweets contained little information in terms of number of words or posting characteristics. In the cases of 60 and 100 tweets, the number of users included dropped dramatically. For this reason, we applied a requirement of a minimum of 30 tweets in both in-treatment and unknown-treatment datasets to keep the balance between the number of tweets and the number of users to be included in the study. After applying this requirement, 187 users were finally included in the study. The complete timelines of these users were compiled, totaling 668,842 tweets, which were reduced to 482,338 once retweets were removed. Out of these, 168,369 more tweets were excluded because they were posted on dates located outside the periods that qualified a tweet for being included in the in-treatment or the unknown-treatment datasets. Finally, 57,525 tweets were included in the in-treatment dataset and 256,444 in the unknown-treatment dataset. Data Analysis The two datasets of tweets, in-treatment and unknown-treatment, were compared in order to determine the existence of behavioral and linguistic differences between the tweets generated by the users in each period. The features that were analyzed are listed in Table 3. Table 3. Features of the tweets analyzed. Features Analyses performed Distribution over time Tweets per hour, tweets during daytime vs night, tweets per day, tweets during weekdays vs weekend Length Number of characters, number of words Part-of-speech (POS) Number of words by grammatical categories (part-of-speech tags) Emotion analysis Frequencies of emotion types Negations Polarity Frequencies of negation words Polarity of tweets on the basis of Spanish Sentiment Lexicon Paired data statistical significance tests (paired t tests) were carried out whenever possible. The Benjamini-Hochberg false discovery rate was applied for multiple testing correction analysis [64]. The P values provided incorporate it. The textual content of each tweet was analyzed using the same methodology and tools used in our previous study [49]. The textual content of each tweet was analyzed by means of the following steps: tokenization performed based on a customized Twitter tokenizer included in the Natural Language Toolkit [65]; part-of-speech (POS) tagging performed by means of the FreeLing Natural Language Processing tool in order to analyze the usage patterns of grammatical categories, such as verbs, nouns, pronouns, adverbs, and adjectives, in the text of tweets [66]; identification of negations performed by building upon a customized list of Spanish negation expressions, such as nada (nothing), nadie (nobody), no (no), nunca (never), and similar; identification of positive and negative words inside the text of each tweet using the Spanish Sentiment Lexicon [67]; and identification of words and expressions associated with emotions such as happiness, anger, fear, disgust, surprise, and sadness [68] by using the Spanish Emotion Lexicon [69]. The statistical analyses were carried out using Python 3.7, the Tweepy, SciPy, and Natural Language Toolkit libraries, and R version 3.6.2 (R Development Core Team), including the R “psych” package 1.9.12.31. All the aforementioned software tools are publicly available. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Ethical Approval The protocol used in this study was reviewed and approved by the Ethics Committee of Parc Salut Mar (approval number 2017/7234/1). in-treatment periods; the percentage went down to 74.40% (SD 5.31) in unknown-treatment periods, with a mean percentage difference of 1.56% (SD 8.9) that implies statistically significant differences between the two periods (t186=2.39; P=.02). Results Distribution Over Time Several types of distribution-over-time analysis were performed in order to study the potential influence of being in in-treatment periods in comparison to unknown-treatment ones. The tweet hours were adjusted by the users’ time zone. The mean duration of the time period analyzed of all the users was 28.2 months (SD 24.7); the mean of the total number of tweets analyzed was 307.6 (SD 336.0) for in-treatment periods and 1371.4 (SD 748.2) in the case of unknown-treatment periods. The mean number of tweets per day generated by users during in-treatment periods was 11.44 (SD 10.05); this number dropped to 9.07 (SD 7.21) in the unknown-treatment dataset with a mean difference of 2.37 (SD 9.72) between periods, which shows statistically significant differences between the two datasets (t186=3.33; P<.001). The mean percentage of tweets posted during daytime (between 8 AM and midnight) was 64.30% (SD 14.83) when the users were in-treatment periods; this percentage fell to 61.78% (SD 13.69) during the unknown-treatment periods, with a mean percentage difference of 2.52% (SD 11.81), which implies statistically significant differences (t186=3.07; P=.004). in periods (SD 6.70) The mean number of tweets generated during the weekdays (from Monday to Friday) was 12.28 (SD 11.05) during in-treatment the and 9.33 unknown-treatment periods, with a mean difference of 2.95 (SD 10.23) and statistically significant differences between the datasets (t186=3.93; P<.001). For the mean number of tweets generated during the weekends (Saturday and Sunday), it was 9.35 (SD 9.31) in the in-treatment period and 8.41 (SD 9.82) in the unknown-treatment period, with a mean difference of 0.94 (SD 10.92) that implies statistically significant differences between the datasets (t186=1.18; P=.23). The mean percentage of tweets posted on weekdays was 75.95% (SD 9.17) during Length The average number of characters per tweet was 88.03 (SD 30.74) and 85.19 (SD 28.82) in the in-treatment and unknown-treatment datasets, respectively, with a mean difference of 2.84 (SD 17.70) and statistically significant differences between the periods (t186=2.19; P=.03). As for the number of words per tweet, the mean was 15.68 (SD 5.75) in the the unknown-treatment dataset, with a mean difference of 0.59 (SD 3.54) and statistically significant differences (t186=2.28; P=.02). in-treatment dataset and 15.09 (SD 5.20) in Links and Mentions to Other Users The mean percentages of tweets that include at least one link were 23.10% (SD 16.16) and 23.27% (SD 15.29) in the in-treatment and unknown-treatment datasets, respectively, with a mean difference of −0.17 (SD 10.94), which is not statistically significant (t186=−0.23; P=.82).The mean percentages of tweets that include at least one mention of another Twitter user were 45.79% (SD 24.77) and 43.52% (SD 24.71) in the in-treatment and unknown-treatment datasets, respectively, with a mean difference of 2.27% (SD 12.13), which is statistically significant (t186=2.56; P=.01). Part-of-Speech As for the analysis of the number of words by grammatical category (ie, part-of-speech) in each tweet, we also compared the in-treatment and unknown-treatment datasets. The mean percentage of words per grammatical category over the total number of words in each dataset is shown in Table 4. We considered the most relevant lexical POS such as verbs, nouns, pronouns, adverbs, and adjectives, excluding conjunctions, interjections, punctuations, determiners, adpositions, numbers, and dates. Regarding the different types of pronouns, the mean percentages of personal pronouns in each dataset are shown and compared in Table 5. Table 4. Percentages of part-of-speech words compared between in-treatment and unknown-treatment datasets. POSa Verbs Nouns Pronouns Adverbs Adjectives in-treatment (%), mean unknown-treatment (%), mean Difference (%), mean (SD) Paired t test P value 18.50 19.50 9.19 6.42 6.05 18.20 19.94 8.93 6.36 6.21 0.3 (1.28) −0.44 (2.57) 0.26 (1.33) 0.06 (0.84) −0.16 (0.95) 3.15 −2.35 2.61 0.97 −2.34 .002 .02 .01 .34 .02 aPOS: part-of-speech. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Table 5. Mean percentages of personal pronouns compared between in-treatment and unknown-treatment datasets. Personal pronouns in-treatment (%), mean unknown-treatment (%), mean Difference (%), mean (SD) Paired t test P value 1st person singular 2nd person singular 3rd person singular 1st person plural 2nd person plural 3rd person plural 49.50 14.77 22.13 3.44 1.00 5.60 47.80 16.07 22.86 3.43 1.00 5.39 Emotion Analysis The mean percentages of the different emotions, obtained using the Spanish Sentiment Lexicon on the tweets posted in the two periods, are shown in Table 6. 1.7 (8.68) −1.3 (6.17) −0.73 (5.79) 0.01 (3.43) 0 (1.22) 0.21 (3.68) 2.67 −2.88 −1.72 0.04 −0.01 0.77 .008 .004 .08 .96 .98 .44 Table 6. Mean percentages of different emotions compared between in-treatment and unknown-treatment datasets. Emotion in-treatment (%), mean unknown-treatment (%), mean Difference (%), mean (SD) Paired t test P value Happiness Sadness Fear Anger Disgust Surprise 26.93 10.01 3.20 5.52 3.11 5.59 25.94 9.76 3.02 5.20 3.06 5.06 Negation Analysis The mean percentages of tweets, among all users, that included one or more negation words were 27.66% (SD 10.54) and 26.59% (SD 9.87) for the in-treatment and unknown-treatment datasets, respectively, with a mean difference of 1.07% (SD 6.99), which is statistically significant (t186=2.10; P=.04). Polarity Analysis As for the polarity of tweets, the percentage of tweets, among all users, with one or more positive words inside the text was 15.13% (SD 6.56) in the in-treatment dataset and 14.50% (SD 5.43) in the unknown-treatment dataset, with a mean percentage difference of 0.63% (SD 5.22; t186=1.66; P=.09). The percentage of tweets with one or more negative words was 7.97% (SD 4.40) in the in-treatment dataset and 7.54% (SD 3.52) in the unknown-treatment dataset, with a mean percentage difference of 0.43% (SD 3.58) (t186=1.64; P=.10). No statistically significant differences were detected in this analysis. Discussion Principal Findings Social media platforms in general, and Twitter in particular, may provide useful information on how patients respond when they receive a pharmacological treatment, as has been shown in several studies in which social media has been used as a complementary source of pharmacovigilance and monitoring [34,70]. In this study, we analyzed the tweets of users who mentioned they were taking antidepressant drugs, in particular SSRIs, with the aim of detecting behavioral changes when they http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX 0.99 (5.82) 0.25 (4.20) 0.18 (1.94) 0.32 (2.71) 0.05 (1.97) 0.53 (2.42) 2.32 0.81 1.23 1.62 0.38 2.98 .02 .41 .21 .11 .69 .003 are more likely to be in treatment in comparison to periods in which they are less likely to be in treatment (“in-treatment” vs “unknown-treatment” periods). The results of this study show that, in general, Twitter users significantly increased their activity of posting tweets during the in-treatment periods. This increase was more pronounced during weekdays than during weekends. We also observed a significantly greater proportion of tweets posted during the daytime during the in-treatment periods. These results are consistent with the results of our previous paper [49], in which we observed that the control group without signs of depression showed more tweet posting activity than the group of users with signs of depression, especially during the daytime and the weekdays. These results are also consistent with another paper that described the behavior in social media of people with self-reported depression [41], as well as with a study on the diurnal mood variation of patients suffering from major depressive disorder [71]. In summary, we can state that when considering tweet posting activity, the behavior of individuals suffering from depression becomes more similar to that of the general population when they are in treatment with SSRIs. Likewise, the average number of characters and words per tweet were significantly higher when the Twitter users were in treatment with SSRIs, a finding that again points toward an increase in the activity of these treated users. In addition, the increase in the number of mentions per tweet can reflect a greater interest in interacting with other people. All these changes may be due to some improvement in their anhedonic symptoms because of the medication. J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al the and in-treatment changes between Regarding the linguistic analysis, we observed quantitatively slight the unknown-treatment periods, although in some cases they are statistically significant. These slight findings are not easily interpretable. In general, given that the style of writing of people suffering from depression is characterized by self-focus attention, which is associated with negative emotional states and psychological distancing in order to connect with others [72], we can conclude that when the studied subjects were in treatment, they improved some traits related to their posting activity as previously mentioned, but at the same time, their language maintained the features of people suffering from depression without a clear influence of the medication. Emotion is another important aspect that characterizes people suffering from depression, and it was consequently analyzed. When the users were in treatment, they showed small but statistically significant increases in the happiness and surprise emotions, but not in sadness or other emotions (ie, anger, fear and disgust). As for the number of negations, the users slightly increased their use of these types of words during the in-treatment period. However, the polarity analysis did not show differences between the periods. The increased activity observed on Twitter when the users were likely to be in treatment with SSRIs can be linked to improved emotional status in their happiness and surprise emotions. These changes are consistent with our previous observations on mood states of Twitter users without depression compared to those with depression [49]. However, the traits that are related to language, as indicated by the POS analysis and the use of negations, maintained a similar profile to that of subjects with depression, independently of the pharmacological treatment detected. These results denote that users with depression who are taking SSRIs show some mood improvements while receiving antidepressant treatment, but at the same time maintain an altered language pattern, which may be indicative of incomplete recovery. On the basis of our statistically significant results, we may state that Twitter timelines can be used as a complementary tool to monitor subjects in order to detect adherence to treatment, which is an important problem in this kind of patient. Adherence to treatment is essential for disease remission [73-76]. According to some studies, it is common for patients suffering from depression to not maintain the duration of antidepressant treatment that is clinically recommended [4,18,77]. In summary, the follow-up of behavioral and language changes in users’ Twitter timelines can be useful for monitoring the evolution of depressive symptoms and the effect of treatments. Limitations and Future Directions This type of study in general, and this one in particular, presents some limitations. For instance, we considered tweets written in Spanish and from public Twitter users’ timelines, and these users may be not representative of the general population or people suffering from depression [33,49,78,79]. Some studies have shown that Twitter users are often urban people with high levels of education, and they are generally younger than the general population [33,49,78,80,81]. We should also take into account that SSRIs are used in different types of depressive disorders and in other mental conditions. Moreover, we have no information about whether these drugs were taken in the context of a prescribed medical treatment or as a result of an inappropriate self-medication decision. Another limitation may be the fact that Twitter users who share their personal drug intake may use words or expressions not included in the list of drug names employed in this study for streaming tweets, even though we tried to be exhaustive in the list of names used. Twitter texts are informal and limited by the number of characters, and they commonly include abbreviations, errors, or slang language [33,45]. All these issues can make it difficult to automatically extract drug mentions and link them to a formal lexicon [28,30,50,53,55]. Unlike clinical records that could be linked to domain resources, the lack of lay vocabularies related to health concepts and terminologies hinders the processing of social media texts [55]. In addition, the results obtained may depend on the particular drugs selected for the study [33], as well as on the periods of time set up for classifying the tweets into the in-treatment and unknown-treatment datasets. On the basis of the strategy applied for defining the groups of tweets to be compared (tweets generated just after mentions to SSRI intake vs tweets generated in periods far from any mention to the SSRI intake), there is some chance of misclassification; it is likely that not all the tweets in the first group were generated by users under actual SSRI treatment, and it is probable that some tweets of the second group have been generated by users under SSRI treatment. Furthermore, we must take into account that data from social media posts contain irrelevant information. Although the proportion of useful information for the specific research purpose can be quite limited, it constitutes a useful starting point [28,30,51,53]. In this scenario, the human curation of tweets is a necessary step in this kind of analysis [34]. Even so, due to the different nuances that a tweet can involve, it is not easy to detect real drug intakes or firsthand experiences [24,46,52]. Conclusions Social media can be used to monitor the health status of people and, in particular, to detect symptoms or features related to diseases or health conditions by means of analysis of the users’ behavior and language on social media platforms. Moreover, the detection of changes in symptoms or other features when patients are taking medications can provide interesting insights for monitoring pharmacological treatments, as well as for following up on the evolution of the disease, detecting side effects, or providing information related to treatment adherence. Changes in some features, such as a decrease in activity on Twitter or of the length of tweets, an increase of self-focus through the use of the first-person singular pronoun, and changes in the happiness and surprise emotions could be used as the to detect complementary psychological status of users suffering from depression, as well as to perceive lack of adherence to treatment. This information may be especially useful in patients suffering from chronic diseases who are receiving long-term treatments, as is the case for mental disorders in general and depression in particular. However, it is not possible to determine the specific reasons why individuals change their behavior and language on social the worsening of tools http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al media platforms in the framework of a disease and its treatment without performing a clinical assessment. Overall, this study shows the relevance of monitoring behavioral and linguistic changes in the tweets of persons taking antidepressants. These changes are likely to be influenced by the diverse stages of the disease and the therapeutic effects of the treatment that these Twitter users are receiving, opening a new line of research to better understand and follow up on depression through social media. Acknowledgments We received support from the Agency for Management of University and Research Grants in Catalonia (Spain) for the incorporation of new research personnel (FI2016) and from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement number 802750 (FAIRplus) with the support of the European Union’s Horizon 2020 research and innovation programme and European Federation of Pharmaceutical Industries and Associations Companies. The Research Programme on Biomedical Informatics is a member of the Spanish National Bioinformatics Institute, funded by Instituto de Salud Carlos III and the European Regional Development Fund (PRB2-ISCIII), and it is supported by grant PT17/0009/0014. The Department of Experimental and Health Sciences, Universitat Pompeu Fabra, is a “Unidad de Excelencia María de Maeztu”, funded by the Ministry of Economy, Spain [MDM-2014-0370]. Funding for the open access charge is from the Agència de Gestió d’Ajuts Universitaris i de Recerca Generalitat de Catalunya (2017 SGR 00519). The Database for Pharmacoepidemiological Research in Primary Care, from the Spanish Agency for Medicines and Health Products of the Ministry of Health and Consumer Affairs and Social Welfare of the Government of Spain, was used to obtain useful information about the prescription frequency of antidepressants. Conflicts of Interest None declared. References 1. World Health Organization. Depression: Key Facts. 2019. URL: https://www.who.int/news-room/fact-sheets/detail/depression [accessed 2020-01-09] 2. World Health Organization. Depression and Other Common Mental Disorders: Global Health Estimates. 2017. URL: http:/ 3. 4. /apps.who.int/iris/bitstream/10665/254610/1/WHO-MSD-MER-2017.2-eng.pdf?ua=1 [accessed 2020-01-09] Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med 2013 Nov;10(11):e1001547 [FREE Full text] [doi: 10.1371/journal.pmed.1001547] [Medline: 24223526] Royal College of Psychiatrists. Position statement on antidepressants and depression. PS04/19. London, United Kingdom: Royal College of Psychiatrists; May 2019. 6. 5. Marcus M, Yasamy MT, van Ommeren M, Chisholm D, Saxena S, WHO Department of Mental Health and Substance Abuse. Depression: a global public health concern.: World Health Organization; 2012. URL: http://www.who.int/ mental_health/management/depression/who_paper_depression_wfmh_2012.pdf [accessed 2020-11-12] Global BODS2C. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: a systematic analysis for the global burden of disease study 2013. Lancet Aug 22 2015;386(9995):743-800. [Medline: 26063472] Trautmann S, Rehm J, Wittchen H. The economic costs of mental disorders: Do our societies react appropriately to the burden of mental disorders? EMBO Rep 2016 Sep;17(9):1245-1249 [FREE Full text] [doi: 10.15252/embr.201642951] [Medline: 27491723] Patel V, Chisholm D, Parikh R, Charlson F, Degenhardt L, Dua T. Global priorities for addressing the burden of mental, neurological, substance use disorders. In: Patel V, Chisholm D, Dua T, Laxminarayan R, Medina-Mora ME, Vos T. eds. Disease Control Priorities: Mental, Neurological, and Substance Use Disorders, 3rd ed. (Vol 4). Washington, DC: The World Bank; 2016:1-27. 8. 7. 9. Whiteford HA, Degenhardt L, Rehm J, Baxter AJ, Ferrari AJ, Erskine HE, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 2013 Nov 9;382(9904):1575-1586. [doi: 10.1016/S0140-6736(13)61611-6] [Medline: 23993280] 10. Wongkoblap A, Vadillo MA, Curcin V. Researching mental health disorders in the era of social media: Systematic review. J Med Internet Res 2017 Dec 29;19(6):e228 [FREE Full text] [doi: 10.2196/jmir.7215] [Medline: 28663166] 11. Working Group of the Clinical Practice Guideline on the Management of Depression in adults. Ministry of Health, Social Services and Equality. Galician Agency for Health Technology Assessment (avalia-t). Clinical practice guideline on the management of depression in adults. 2014. URL: https://portal.guiasalud.es/wp-content/uploads/2018/12/ GPC_534_Depresion_Adulto_Avaliat_compl_en.pdf [accessed 2020-11-12] 12. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006 Nov;3(11):e442 [FREE Full text] [doi: 10.1371/journal.pmed.0030442] [Medline: 17132052] http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al 13. Vigo D, Thornicroft G, Atun R. Estimating the true global burden of mental illness. Lancet Psychiatry 2016 Feb;3(2):171-178. [doi: 10.1016/S2215-0366(15)00505-2] [Medline: 26851330] 14. Kraus C, Kadriu B, Lanzenberger R, Zarate CA, Kasper S. Prognosis and improved outcomes in major depression: a review. Transl Psychiatry 2019 Apr 03;9(1):127 [FREE Full text] [doi: 10.1038/s41398-019-0460-3] [Medline: 30944309] 15. Ministerio de Sanidad, Servicios Sociales e Igualdad. Agencia Española de Medicamentos y Productos Sanitarios (AEMPS). Utilización de medicamentos antidepresivos en España durante el periodo 2000-2013. Informe de utilización de medicamentos U/AD/V1/14012015. Madrid: Ministerio de Sanidad, Servicios Sociales e Igualdad. Agencia Española de Medicamentos y Productos Sanitarios (AEMPS); 2015. 16. Luo Y, Kataoka Y, Ostinelli EG, Cipriani A, Furukawa TA. National prescription patterns of antidepressants in the treatment of adults with major depression in the US between 1996 and 2015: A population representative survey based analysis. Front Psychiatry 2020;11:35 [FREE Full text] [doi: 10.3389/fpsyt.2020.00035] [Medline: 32116850] Fasipe OJ. The emergence of new antidepressants for clinical use: Agomelatine paradox versus other novel agents. IBRO Rep 2019 Jun;9(6):95-110 [FREE Full text] [doi: 10.1016/j.ibror.2019.01.001] [Medline: 31211282] 17. 18. Depression in adults: recognition and management. Clinical guideline CG90. National Institute for Health and Care Excellence (NICE); 2009 Oct 28. URL: https://www.nice.org.uk/guidance/cg90 [accessed 2020-11-12] 19. Gelenberg A, Freeman M, Markowitz J, Rosenbaum J, Thase M, Trivedi M, et al. Practice guideline for the treatment of patients with major depressive disorder. 3rd ed. Washington, DC: American Psychiatric Association; 2010. 20. Lane R, Baldwin D, Preskorn S. The SSRIs: advantages, disadvantages and differences. J Psychopharmacol 1995 Jan;9(2 21. Suppl):163-178. [doi: 10.1177/0269881195009002011] [Medline: 22297235] Pundiak TM, Case BG, Peselow ED, Mulcare L. Discontinuation of maintenance selective serotonin reuptake inhibitor monotherapy after 5 years of stable response: a naturalistic study. J Clin Psychiatry 2008 Nov;69(11):1811-1817. [doi: 10.4088/jcp.v69n1117] [Medline: 19026252] 22. Emslie GJ, Mayes TL, Ruberu M. Continuation and maintenance therapy of early-onset major depressive disorder. Paediatr Drugs 2005;7(4):203-217. [doi: 10.2165/00148581-200507040-00001] [Medline: 16117558] 23. Garnock-Jones KP, McCormack PL. Escitalopram: a review of its use in the management of major depressive disorder in adults. CNS Drugs 2010 Sep;24(9):769-796. [doi: 10.2165/11204760-000000000-00000] [Medline: 20806989] 24. Clevenger SS, Malhotra D, Dang J, Vanle B, IsHak WW. The role of selective serotonin reuptake inhibitors in preventing relapse of major depressive disorder. Ther Adv Psychopharmacol 2018 Jan;8(1):49-58 [FREE Full text] [doi: 10.1177/2045125317737264] [Medline: 29344343] Sim K, Lau WK, Sim J, Sum MY, Baldessarini RJ. Prevention of relapse and recurrence in adults with major depressive disorder: Systematic review and meta-analyses of controlled trials. Int J Neuropsychopharmacol 2015 Jul 07;19(2):pyv076 [FREE Full text] [doi: 10.1093/ijnp/pyv076] [Medline: 26152228] Peselow ED, Tobia G, Karamians R, Pizano D, IsHak WW. Prophylactic efficacy of fluoxetine, escitalopram, sertraline, paroxetine, and concomitant psychotherapy in major depressive disorder: outcome after long-term follow-up. Psychiatry Res 2015 Feb 28;225(4):680-686. [doi: 10.1016/j.psychres.2014.11.022] [Medline: 25496869] Salathé M. Digital pharmacovigilance and disease surveillance: Combining traditional and Big-Data systems for better public health. J Infect Dis 2016 Dec 01;214(suppl_4):S399-S403 [FREE Full text] [doi: 10.1093/infdis/jiw281] [Medline: 28830106] Sarker A, Ginn R, Nikfarjam A, O'Connor K, Smith K, Jayaraman S, et al. Utilizing social media data for pharmacovigilance: A review. J Biomed Inform 2015 Apr;54:202-212 [FREE Full text] [doi: 10.1016/j.jbi.2015.02.004] [Medline: 25720841] 25. 26. 27. 28. 29. Adrover C, Bodnar T, Huang Z, Telenti A, Salathé M. Identifying adverse effects of HIV drug treatment and associated sentiments using Twitter. JMIR Public Health Surveill 2015;1(2):e7 [FREE Full text] [doi: 10.2196/publichealth.4488] [Medline: 27227141] 30. Nikfarjam A, Sarker A, O'Connor K, Ginn R, Gonzalez G. Pharmacovigilance from social media: mining adverse drug reaction mentions using sequence labeling with word embedding cluster features. J Am Med Inform Assoc 2015 May;22(3):671-681 [FREE Full text] [doi: 10.1093/jamia/ocu041] [Medline: 25755127] Salathé M. Digital epidemiology: what is it, and where is it going? Life Sci Soc Policy 2018 Jan 04;14(1):1 [FREE Full text] [doi: 10.1186/s40504-017-0065-7] [Medline: 29302758] 31. 32. Eysenbach G. Infodemiology and infoveillance: framework for an emerging set of public health informatics methods to analyze search, communication and publication behavior on the Internet. J Med Internet Res 2009 Mar 27;11(1):e11 [FREE Full text] [doi: 10.2196/jmir.1157] [Medline: 19329408] 33. Lardon J, Bellet F, Aboukhamis R, Asfari H, Souvignet J, Jaulent M, et al. Evaluating Twitter as a complementary data source for pharmacovigilance. Expert Opin Drug Saf 2018 Aug;17(8):763-774. [doi: 10.1080/14740338.2018.1499724] [Medline: 29991282] Freifeld CC, Brownstein JS, Menone CM, Bao W, Filice R, Kass-Hout T, et al. Digital drug safety surveillance: monitoring pharmaceutical products in twitter. Drug Saf 2014 May;37(5):343-350 [FREE Full text] [doi: 10.1007/s40264-014-0155-x] [Medline: 24777653] 34. 35. Crannell WC, Clark E, Jones C, James TA, Moore J. A pattern-matched Twitter analysis of US cancer-patient sentiments. J Surg Res 2016 Dec;206(2):536-542. [doi: 10.1016/j.jss.2016.06.050] [Medline: 27523257] http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al 36. Surian D, Nguyen DQ, Kennedy G, Johnson M, Coiera E, Dunn AG. Characterizing Twitter discussions about HPV vaccines using topic modeling and community detection. J Med Internet Res 2016;18(8):e232 [FREE Full text] [doi: 10.2196/jmir.6045] [Medline: 27573910] 37. Chen E, Lerman K, Ferrara E. Tracking Social Media Discourse About the COVID-19 Pandemic: Development of a Public Coronavirus Twitter Data Set. JMIR Public Health Surveill 2020 May 29;6(2):e19273 [FREE Full text] [doi: 10.2196/19273] [Medline: 32427106] 38. Reece AG, Reagan AJ, Lix KLM, Dodds PS, Danforth CM, Langer EJ. Forecasting the onset and course of mental illness with Twitter data. Sci Rep 2017 Oct 11;7(1):13006 [FREE Full text] [doi: 10.1038/s41598-017-12961-9] [Medline: 29021528] Park M, Cha C, Cha M. Depressive moods of users portrayed in Twitter. In: Proceedings of the ACM SIGKDD Workshop on Health Informatics. 2012 Presented at: HI-KDD'12; August 12-16; Beijing, China p. 1-8. 39. 40. Conway M, O'Connor D. Social Media, Big Data, and Mental Health: Current Advances and Ethical Implications. Curr Opin Psychol 2016 Jun;9:77-82 [FREE Full text] [doi: 10.1016/j.copsyc.2016.01.004] [Medline: 27042689] 41. De Choudhury C, Gamon M, Counts S, Horvitz E. Predicting depression via social media. In: Proceedings of the Seventh International Conference on Weblogs and Social Media. 2013 Presented at: AAA'13; July 8-11; Cambridge, MA p. 128-138. 42. Cavazos-Rehg P, Krauss M, Sowles S, Connolly S, Rosas C, Bharadwaj M, et al. A content analysis of depression-related tweets. Comput Human Behav 2016 Jan 01;54:351-357 [FREE Full text] [doi: 10.1016/j.chb.2015.08.023] [Medline: 26392678] 43. Nguyen T, O’Dea B, Larsen M, Phung D, Venkatesh S, Christensen H. Using linguistic and topic analysis to classify sub-groups of online depression communities. Multimed Tools Appl 2015 Dec 21;76(8):10653-10676. [doi: 10.1007/s11042-015-3128-x] Statista. Number of Monthly Active Twitter Users Worldwide From 1st Quarter 2010 to 2nd Quarter 2019 (in Millions). 2019. URL: https://www.statista.com/statistics/282087/number-of-monthly-active-twitter-users/ [accessed 2020-02-03] 45. Audeh B, Calvier F, Bellet F, Beyens M, Pariente A, Lillo-Le Louet A, et al. Pharmacology and social media: Potentials 44. and biases of web forums for drug mention analysis-case study of France. Health Informatics J 2019 Sep 30;26(2):1253-1272. [doi: 10.1177/1460458219865128] [Medline: 31566468] 46. Alvaro N, Conway M, Doan S, Lofi C, Overington J, Collier N. Crowdsourcing Twitter annotations to identify first-hand experiences of prescription drug use. J Biomed Inform 2015 Dec;58:280-287 [FREE Full text] [doi: 10.1016/j.jbi.2015.11.004] [Medline: 26556646] Pierce CE, Bouri K, Pamer C, Proestel S, Rodriguez HW, Van LH, et al. Evaluation of Facebook and Twitter monitoring to detect safety signals for medical products: An analysis of recent FDA safety alerts. Drug Saf 2017 Apr;40(4):317-331 [FREE Full text] [doi: 10.1007/s40264-016-0491-0] [Medline: 28044249] 47. 48. Bian J, Topaloglu U, Yu F. Towards Large-scale Twitter mining for drug-related adverse events. SHB12 2012 Oct 29;2012:25-32 [FREE Full text] [doi: 10.1145/2389707.2389713] [Medline: 28967001] 49. Leis A, Ronzano F, Mayer MA, Furlong LI, Sanz F. Detecting signs of depression in tweets in Spanish: Behavioral and linguistic analysis. J Med Internet Res 2019 Jun 27;21(6):e14199 [FREE Full text] [doi: 10.2196/14199] [Medline: 31250832] 50. O'Connor K, Pimpalkhute P, Nikfarjam A, Ginn R, Smith KL, Gonzalez G. Pharmacovigilance on twitter? Mining tweets for adverse drug reactions. AMIA Annu Symp Proc 2014;2014:924-393 [FREE Full text] [Medline: 25954400] 51. Mahata D, Friedrichs J, Shah RR, Jiang J. Detecting personal intake of medicine from Twitter. IEEE Intell Syst 2018 Jul;33(4):87-95. [doi: 10.1109/mis.2018.043741326] 52. Klein A, Sarker A, Rouhizadeh M, O'Connor K, Gonzalez G. Detecting personal medication intake in Twitter: an annotated corpusbaseline classification system. 2017 Presented at: the 16th Biomedical Natural Language Processing (BioNLP); Aug 4; Vancouver, BC, Canada p. 1-11 URL: https://www.aclweb.org/anthology/W17-2316.pdf 53. Kiritchenko S, Mohammad S, Morin J, de BB. NRC-Canada at SMM4H shared task: classifying tweets mentioning adverse drug reactions and medication intake. 2017 Presented at: In Proceedings of 2nd Social Media Mining for Health Applications Workshop co-located with the American Medical Informatics Association Annual Symposium (AMIA); Nov 4, 2017; Washington DC p. 1-11 URL: http://ceur-ws.org/Vol-1996/paper1.pdf 55. 54. Nikfarjam A, Ransohoff JD, Callahan A, Jones E, Loew B, Kwong BY, et al. Early detection of adverse drug reactions in social health networks: A natural language processing pipeline for dignal detection. JMIR Public Health Surveill 2019 Jun 03;5(2):e11264 [FREE Full text] [doi: 10.2196/11264] [Medline: 31162134] Segura-Bedmar I, Martínez P, Revert R, Moreno-Schneider J. Exploring Spanish health social media for detecting drug effects. BMC Med Inform Decis Mak 2015;15 Suppl 2:S6 [FREE Full text] [doi: 10.1186/1472-6947-15-S2-S6] [Medline: 26100267] Statista. Leading Countries Based on Number of Twitter Users as of April 2020 (in Millions). 2020. URL: https://www. statista.com/statistics/242606/number-of-active-twitter-users-in-selected-countries/ [accessed 2020-07-05] 56. 57. Liu J, Weitzman E, Chunara R. Assessing behavioral stages from social media data. In: Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing. 2017 Presented at: CSCW 2017; February 25-March 1, 2017; Portland, Oregon p. 1320-1333. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al 58. Saha K, Sugar B, Torous J, Abrahao B, Kiciman E, De Choudhury M. A social media study on the effects of psychiatric medication use. In: Proceedings of the thirteenth International Conference on Web and Social Media. 2019 Presented at: ICWSM 2019; 2019; Munich, Germany p. 440-451. 59. Twitter Developer. URL: https://developer.twitter.com/en.html [accessed 2019-12-16] 60. Wishart DS, Knox C, Guo AC, Cheng D, Shrivastava S, Tzur D, et al. DrugBank: a knowledgebase for drugs, drug actions and drug targets. Nucleic Acids Res 2008;36(Database issue):D901-D906 [FREE Full text] [doi: 10.1093/nar/gkm958] [Medline: 18048412] 61. WHO Collaborating Centre for Drug Statistics Methodology, Norwegian Institute of Public Health. The Anatomical Therapeutics Chemical Classification System (ATC). Last updated December 16. 2019 Jan 16. URL: https://www.whocc.no/ [accessed 2019-05-06] Selective serotonin reuptake inhibitor.: Wikipedia; 2019 May 07. URL: https://en.wikipedia.org/wiki/ Selective_serotonin_reuptake_inhibitor [accessed 2019-05-07] Spanish Agency for Medicines and Health Products (AEMPS), Ministry of Health and Consumer Affairs and Social Welfare. Multiregional primary care database of Spanish population (BIFAP). 2019. URL: http://www.bifap.org/ [accessed 2019-05-07] 62. 63. 64. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple hypothesis testing. J R Stat Soc B 1995;57(1):289-300. [doi: 10.1111/j.2517-6161.1995.tb02031.x] 65. Natural Language Tool Kit. URL: https://www.nltk.org/api/nltk.tokenize.html [accessed 2019-12-10] 66. Padró L, Stanilovsky E. FreeLing 3.0: Towards wider multilinguality. In: Proceedings of the Eighth International Conference on Language Resources and Evaluation. 2012 Presented at: LREC'12; May 21-27; Istanbul, Turkey p. 2473-2479 URL: http://www.lrec-conf.org/proceedings/lrec2012/pdf/430_Paper.pdf Perez-Rosas V, Banea C, Mihalcea R. Learning sentiment lexicons in Spanish. In: Proceedings of the Eighth International Conference on Language Resources and Evaluation. 2012 Presented at: LREC'12; May 21-27; Istanbul, Turkey p. 3077-3081 URL: http://www.lrec-conf.org/proceedings/lrec2012/pdf/1081_Paper.pdf 67. 68. Ekman P, Friesen WV, O'Sullivan M, Chan A, et al. Universals and cultural differences in the judgments of facial expressions of emotion. J Pers Soc Psychol 1987 Oct;53(4):712-717. [doi: 10.1037//0022-3514.53.4.712] [Medline: 3681648] Sidorov G, Miranda-Jiménez S, Viveros-Jiménez F, Gelbukh A, Castro-Sánchez N, Castillo F. Empirical study of opinion mining in Spanish tweets. In: Proceedings of the 11th Mexican International Conference on Artificial Intelligence. 2012 Presented at: MICAI'12; October 27-November 4, 2012; San Luis Potosí, Mexico p. 1-4. [doi: 10.1007/978-3-642-37798-3_4] 69. 70. Carbonell P, Mayer MA, Bravo A. Exploring brand-name drug mentions on Twitter for pharmacovigilance. Stud Health Technol Inform 2015;210:55-59. [Medline: 25991101] 71. Morris DW, Rush AJ, Jain S, Fava M, Wisniewski SR, Balasubramani GK, et al. Diurnal mood variation in outpatients with major depressive disorder: Implications for DSM-V from an analysis of the sequenced treatment alternatives to relieve depression study data. J Clin Psychiatry 2007 Sep;68(9):1339-1347. [Medline: 17915971] Pennebaker JW, Mehl MR, Niederhoffer KG. Psychological aspects of natural language use: our words, our selves. Annu Rev Psychol 2003;54:547-577. [doi: 10.1146/annurev.psych.54.101601.145041] [Medline: 12185209] 72. 73. Anghelescu IG, Kohnen R, Szegedi A, Klement S, Kieser M. Comparison of Hypericum extract WS 5570 and paroxetine in ongoing treatment after recovery from an episode of moderate to severe depression: results from a randomized multicenter study. Pharmacopsychiatry 2006 Nov;39(6):213-219. [doi: 10.1055/s-2006-951388] [Medline: 17124643] Sansone RA, Sansone LA. Antidepressant adherence: are patients taking their medications? Innov Clin Neurosci 2012 May;9(5-6):41-46 [FREE Full text] [Medline: 22808448] 74. 75. Mitchell AJ. Depressed patients and treatment adherence. Lancet 2006 Jun 24;367(9528):2041-2043. [doi: 10.1016/S0140-6736(06)68902-2] [Medline: 16798371] 76. De las Cuevas C, Peñate W, Sanz EJ. Risk factors for non-adherence to antidepressant treatment in patients with mood disorders. Eur J Clin Pharmacol 2014 Jan;70(1):89-98. [doi: 10.1007/s00228-013-1582-9] [Medline: 24013851] 77. De Choudhury M, De S. Mental health discourse on Reddit: Self-disclosure, social support, and anonymity. In: Proceedings of the 8th International Conference on Weblogs and Social Media. 2014 Presented at: AAAI'14; June 1-4; Ann Arbor, Michigan p. 71-80 URL: https://www.aaai.org/ocs/index.php/ICWSM/ICWSM14/paper/view/8075/8107 Sadah SA, Shahbazi M, Wiley MT, Hristidis V. A study of the demographics of web-based health-related social media users. J Med Internet Res 2015 Aug 06;17(8):e194 [FREE Full text] [doi: 10.2196/jmir.4308] [Medline: 26250986] 78. 79. Eichstaedt JC, Schwartz HA, Kern ML, Park G, Labarthe DR, Merchant RM, et al. Psychological language on Twitter predicts county-level heart disease mortality. Psychol Sci 2015 Feb;26(2):159-169 [FREE Full text] [doi: 10.1177/0956797614557867] [Medline: 25605707] 80. Yom-Tov E, Johansson-Cox I, Lampos V, Hayward AC. Estimating the secondary attack rate and serial interval of influenza-like illnesses using social media. Influenza Other Respir Viruses 2015 Jul;9(4):191-199 [FREE Full text] [doi: 10.1111/irv.12321] [Medline: 25962320] Paul MJ, Dredze M. Discovering health topics in social media using topic models. PLoS One 2014;9(8):e103408 [FREE Full text] [doi: 10.1371/journal.pone.0103408] [Medline: 25084530] 81. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Leis et al Abbreviations POS: part-of-speech SSRIs: selective serotonin reuptake inhibitors Edited by G Eysenbach, R Kukafka; submitted 01.06.20; peer-reviewed by F Lopez Segui, E Yom-Tov; comments to author 22.06.20; revised version received 01.09.20; accepted 12.11.20; published 18.12.20 Please cite as: Leis A, Ronzano F, Mayer MA, Furlong LI, Sanz F Evaluating Behavioral and Linguistic Changes During Drug Treatment for Depression Using Tweets in Spanish: Pairwise Comparison Study J Med Internet Res 2020;22(12):e20920 URL: http://www.jmir.org/2020/12/e20920/ doi: 10.2196/20920 PMID: 33337338 ©Angela Leis, Francesco Ronzano, Miguel Angel Mayer, Laura I Furlong, Ferran Sanz. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 18.12.2020. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included. http://www.jmir.org/2020/12/e20920/ XSL•FO RenderX J Med Internet Res 2020 \| vol. 22 \| iss. 12 \| e20920 \| p. 14 (page number not for citation purposes)
10.3390_ijerph17072588	Article Cooperation as the Secret Ingredient in the Recipe to Foster Internal Technological Eco-Innovation in the Agri-Food Industry Adrián Rabadán 1,* , Ángela Triguero 2 and Ángela Gonzalez-Moreno 2 1 Higher Technical School of Agricultural and Forestry Engineering, University of Castilla-La Mancha, 2 02071 Albacete, Spain Faculty of Economics and Business Administration, University of Castilla-La Mancha, 02071 Albacete, Spain; [email protected] (Á.T.); [email protected] (Á.G.-M.) * Correspondence: [email protected] Received: 2 March 2020; Accepted: 7 April 2020; Published: 10 April 2020 Abstract: Although eco-innovation in the agri-food sector is receiving increasing amounts of attention, there is a lack of information about the speciﬁc conditions that encourage ﬁrms to develop eco-innovation strategies internally. Our empirical method relies on the data of Spanish ﬁrms operating in the agri-food sector, and uses the Qualitative Comparative Analysis (QCA). Speciﬁcally, we identify the recipes of antecedent conditions that eﬀectively foster the internal development of technological eco-innovation, and then we analyze whether diﬀerences exist in the internal development of product and process eco-innovations. The results show that diﬀerent combinations of conditions can yield internally developed eco-innovation, but all of them indicate that cooperation with stakeholders is the key to fostering technological eco-innovation in this industry. This conclusion encourages the creation of policies and incentives to promote cooperation in order to improve the sustainability of the sector. Keywords: eco-innovation; eco-product innovation; eco-process innovation; QCA; agri-food 1. Introduction Firms’ environmental responsibility is in today’s conversations among practitioners and on policy makers’ agendas. Managers and entrepreneurs are everyday becoming more aware of the consequences of their ﬁrms’ activities on the environment. This environmental consciousness is also fostered from external pressures from stakeholders, such as regulators or consumers [1]. Additionally, some ﬁrms recognize that environmentally friendly behavior can lead them to obtain and maintain competitive advantages [2]. Therefore, some ﬁrms try to become greener by introducing changes in their products and manufacturing processes, enabling more eﬃcient and responsible use of natural resources and energy. This behavior leads to the development and adoption of the so-called technological eco-innovations that take the form of new environmentally friendly products or processes. The academic literature on the drivers of eco-innovation has been increasing lately [2–4], and three main drivers have been identiﬁed: the market pull [5], the regulatory push/pull [6], and the technology push [7]. However, several gaps have been found. First of all, the literature tends to avoid small- and medium-sized enterprises (SMEs) with very few exceptions [1,8,9]. Research has mainly focused on high-tech industries and on large corporations [10]. Usually, SMEs ﬁnd it diﬃcult to convert environmentally friendly practices into competitive advantages [11] and, hence, are reluctant to include environmental concerns in their management practices [12]. Secondly, the agri-food sector, more than any other industry, is characterized by a signiﬁcant dependence on natural resources [13]. Int. J. Environ. Res. Public Health 2020, 17, 2588; doi:10.3390/ijerph17072588 www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2020, 17, 2588 2 of 19 However, although some agri-food companies are currently introducing eco-innovation strategies into their business models [14], there is a lack of research on this topic in traditional sectors, such as the agri-food industry that are typically characterized as low-tech with notable exceptions [1,15–17]. A distinctive feature in the ﬁrms operating in the agri-food industry is the fact that although technological innovation has been found to be critical in these companies, especially for co-operatives [18], the sector has low research and development (R&D) intensity, while producing a signiﬁcant number of innovations [19]. Additionally, most innovations in this industry are more incremental than radical—that is, these innovations are improvements to new food products, or variations of existing ones [20]. Acosta, et al. [21] argued that ﬁrms in this particular industry are aware of external sources of knowledge, including business relationships, a well-developed inter-industry network, and equipment and material purchases. As a result, food ﬁrms take advantage of external knowledge and might have no need to generate this knowledge through internal R&D expenditures. Hence, most of the innovations and eco-innovations that are adopted by ﬁrms in the agri-food industry are acquired from external sources [1], and a few are internally developed. Therefore, there is a lack in the literature of papers that focus on SMEs and on low-tech industries, such as the agri-food industry. Eco-innovation, aimed at providing new business models, products, and manufacturing processes which incorporate new environmentally friendly formats and materials, such as bottling and packaging technology, is considered one of the research priorities for agri-food research, at least in Spain [22]. However, recent literature reviews on eco-innovation have found no single paper that explains internal development of eco-innovation in SMEs in the agri-food industry [2,23,24]. The present paper seeks to ﬁll this gap in the literature, and this is its main contribution. Additionally, it is important to diﬀerentiate between eco-product and eco-process innovations in this industry. In this respect, the former increases the demand for new food products, while the latter reduces the use of energy, materials, and/or natural resources in the manufacturing processes, thereby increasing the ﬁrm’s productivity and competitiveness [16]. Thus, material recycling, energy recovery, waste management, solid waste collection, and water pollution abatement are considered as examples of process eco-innovations, while organic production and the development of more sustainable food production systems would be treated as eco-product innovations. Hence, the purpose of this paper is to determine which factors lead to the internal development of eco-innovation. By means of Qualitative Comparative Analysis (QCA), the present paper will explore the diﬀerent factors that determine the eﬀects of ﬁrms’ resources, capabilities, and cooperative activities on the internal development of eco-innovations in the agri-food industry in Spain. This particular method is suitable for research with small data samples, yet it allows for the generalization of the results, conclusions, and implications. Consequently, in this paper we attempt to answer the following research questions: (1) What drives the internal development of eco-innovations by agri-food SMEs considered the limited access to knowledge and resource in a mature and low-tech industry? In other words, is it possible to develop eco-innovations internally in a mature and traditional industry made up of small companies without ﬁnancial resources or internal R&D? (2) How do ﬁnancial resources and proﬁt levels inﬂuence internal adoption of eco-innovation? What is the role of technological and organizational capabilities of the ﬁrm? What is the inﬂuence of knowledge cooperation to foster internal development of eco-innovations? (3) What are the “optimal” combinations of resources and capabilities to boost the adoption of internal development of eco-innovations by these SMEs? (4) How will the factors interact with each other and how will the interactions aﬀect the overall performance of eco-product and process innovations by agri-food SMEs? (5) Does the adoption of eco-product and process innovations require similar or diﬀerent combinations? Is there any interdependence between both types of eco-innovation? What is the key factor to enhance each type of eco-innovation in the food industry? Int. J. Environ. Res. Public Health 2020, 17, 2588 3 of 19 This paper contributes to the literature by focusing on internal development of eco-innovation in a low-tech industry, by diﬀerentiating between eco-product and eco-process innovations, and by applying QCA, which permits research with small data samples. The paper is structured as follows. In Section 2, we present the literature review and the theoretical background, which are followed by the methodology in Section 3. Then, we show the results of our empirical analysis (Section 4) and we ﬁnish in Section 5 with the conclusions, implications, and limitations of our study. 2. Theoretical Framework Eco-innovation is deﬁned in the Oslo Manual as “the production, assimilation, or exploitation of a product, production process, service, or management or business method that is novel to the organization (developing or adopting it) and which results, throughout its life cycle, in a reduction of environmental risk, pollution, and other negative impacts of resource use (including energy use) compared to relevant alternatives” [25] (p. 8). Attending to this deﬁnition, we can distinguish between technological and non-technological eco-innovation. The former refers to eco-product and eco-production processes, while the latter refers to management, marketing, or business methods that reduce the impact of the ﬁrm’s activities on the environment. The literature on eco-innovation has increased exponentially [24]. While some of the recent papers deal with the relationship between eco-innovation and other emerging topics, such as circular economy [26,27], the majority of the highly cited papers on the topic (see Table 1) are focused on the identiﬁcation of drivers and factors that foster the development and adoption of eco-innovation. Table 1. Highly cited papers on eco-innovation from the Web of Science, 2017–2019. Authors Mavi et al. (2019) [28] Zhang et al. (2019) [29] Kusi-Sarpong et al. (2019) [30] Stucki (2019) [31] Saidani et al. (2019) [26] Kiefer et al. (2019) [32] Díaz-Lopez et al. (2019) [33] Comments Empirical analysis on the joint eﬀect of eco-eﬃciency and eco-innovation on economic growth Empirical analysis on the eﬀect of eco-innovation on performance Empirical analysis on sustainable supply chains in manufacturing companies Empirical analysis on the eﬀect of eco-innovation (green energy) on performance Literature review on eco-innovation and circular economy indicators Resources and capabilities as drivers of diﬀerent eco-innovations Empirical analysis on the implementation of resource eﬃciency measures Kirchherr et al. (2018) [34] Empirical analysis on circular economy barriers Li et al. (2018) [35] Empirical analysis on the role of eco-innovation as a mediator on corporate carbon disclosure Prieto-Sandoval et al. (2018) [27] Literature review on eco-innovation as a precursor of circular economy Ben Arﬁ et al. (2018) [36] Yuan and Xiang (2018) [37] De Jesus and Mendonca (2018) [38] Cai and Li (2018) [39] Tang et al. (2018) [40] Watson et al. (2018) [41] Empirical analysis on the role of external knowledge as driver of eco-innovation Empirical analysis on the role of regulation as a driver of eco-innovation Literature review on the drivers of eco-innovation and circular economy Empirical analysis on the drivers of eco-innovation and its impact on performance Analysis of the moderating role of management on the eﬀect of eco-innovation on performance Literature review on engaging stakeholders in environmental innovation Int. J. Environ. Res. Public Health 2020, 17, 2588 4 of 19 Choi (2018) [42] Feng and Chen (2018) [43] Huang and Li (2017) [44] Gupta and Barua (2017) [45] Costantini et al. (2017) [46] Beltran-Esteve and Picazo-Tadeo (2017) [47] Jansson et al. (2017) [48] Zhang et al. (2017) [49] Notarnicola et al. (2017) [50] Table 1. Cont. Analysis of the interactions between technology-push and demand-pull factors and the role of industry life cycles and domestic market status in the electric vehicle sector Analysis of the role of environmental regulation in the impact of green innovation on industrial green development This paper identiﬁes the factors inﬂuencing green innovation and the relationship between green innovation and performance The paper presents a framework for supplier selection by large companies considering green innovation Empirical analysis of the role played by selected characteristics of the policy mix in inducing innovation in energy eﬃciency technologies The paper assesses environmental performance in the European Union using Luenberger productivity indicators, directional distance functions and Data Envelopment Analysis techniques Analysis of the relationship between market orientation and entrepreneurial orientation in relation to sustainability practices in SMEs Estimation of the eﬀect of environmental innovation on carbon emissions in China Analysis of the environmental impact of food consumption using a lifecycle assessment approach Highly Cites Papers are those papers cited on top 1% of their ﬁeld in their year of publication. Source: Own elaboration. The literature on the drivers of eco-innovation has been increasing [2–4] and three main drivers have been identiﬁed: the market pull [4,5,51], the regulatory push/pull [1,6,52], and the technology push [1,3,7,53–55] (see Figure 1). Figure 1. Drivers of eco-innovation. Adapted from Triguero et al. [16]. Recent research has shown a clear openness of consumers toward product innovation in the agri-food industry [56]. An increase in the consumer demand for greener products and services and an increased willingness to pay extra for environmentally friendly products and/or services has been identiﬁed as a market pull towards eco-innovation [16,51]. On the other hand, the literature has argued that the use of ﬁscal incentives and subsidies fosters the introduction of eco-innovation, thus making its beneﬁts higher than the costs of paying ﬁnes to governments for non-compliance [52]. Additionally, regulation has enabled the agri-food industry to address a prominent issue involving the processing of waste materials, as well as sustainable production systems [13]. Int. J. Environ. Res. Public Health 2020, 17, x 4 of 19 Beltran-Esteve and Picazo-Tadeo (2017) [47] The paper assesses environmental performance in the European Union using Luenberger productivity indicators, directional distance functions and Data Envelopment Analysis techniques Jansson et al. (2017) [48] Analysis of the relationship between market orientation and entrepreneurial orientation in relation to sustainability practices in SMEs Zhang et al. (2017) [49] Estimation of the effect of environmental innovation on carbon emissions in China Notarnicola et al. (2017) [50] Analysis of the environmental impact of food consumption using a lifecycle assessment approach Highly Cites Papers are those papers cited on top 1% of their field in their year of publication. Source: Own elaboration. The literature on the drivers of eco-innovation has been increasing [2–4] and three main drivers have been identified: the market pull [4,5,51], the regulatory push/pull [1,6,52], and the technology push [1,3,7,53–55] (see Figure 1). Figure 1. Drivers of eco-innovation. Adapted from Triguero et al. [16]. Recent research has shown a clear openness of consumers toward product innovation in the agri-food industry [56]. An increase in the consumer demand for greener products and services and an increased willingness to pay extra for environmentally friendly products and/or services has been identified as a market pull towards eco-innovation [16,51]. On the other hand, the literature has argued that the use of fiscal incentives and subsidies fosters the introduction of eco-innovation, thus making its benefits higher than the costs of paying fines to governments for non-compliance [52]. Additionally, regulation has enabled the agri-food industry to address a prominent issue involving the processing of waste materials, as well as sustainable production systems [13]. Finally, the technology push is also considered another key driver of eco-innovation in this industry. A firm’s resources and capabilities enable them to develop the necessary knowledge base to promote eco-innovations [53]. The role of the technology push could also come from the creation of technological networks through which firms collaborate with stakeholders, such as clients, suppliers, and universities [7,54]. This is especially relevant for SMEs. There has been a recent call for studying SMEs’ openness and their knowledge networks [24], as they constitute a key element fostering eco-innovation that deserves further analysis [16]. In this line, some of the latest research has been focusing on the relationship between the cooperation strategies of firms and the adoption and development of eco-innovations [36,57,58]. Regarding the agri-food industry, to the best of our knowledge, only a few studies have investigated the specific drivers of the eco-innovations in this particular sector. Recently, Triguero, Fernández and Sáez-Martinez [16] proposed the framework in Figure 1 to study the influence of open innovation strategies on the adoption of radical and incremental eco-innovations in this industry. This approach is somehow similar to the framework proposed by Marotta and Nazzaro [59] on the issue of the determinants of value creation processes on farms, and further developed by Marotta and Nazzaro [13] on their recent Int. J. Environ. Res. Public Health 2020, 17, 2588 5 of 19 Finally, the technology push is also considered another key driver of eco-innovation in this industry. A ﬁrm’s resources and capabilities enable them to develop the necessary knowledge base to promote eco-innovations [53]. The role of the technology push could also come from the creation of technological networks through which ﬁrms collaborate with stakeholders, such as clients, suppliers, and universities [7,54]. This is especially relevant for SMEs. There has been a recent call for studying SMEs’ openness and their knowledge networks [24], as they constitute a key element fostering eco-innovation that deserves further analysis [16]. In this line, some of the latest research has been focusing on the relationship between the cooperation strategies of ﬁrms and the adoption and development of eco-innovations [36,57,58]. Regarding the agri-food industry, to the best of our knowledge, only a few studies have investigated the speciﬁc drivers of the eco-innovations in this particular sector. Recently, Triguero, Fernández and Sáez-Martinez [16] proposed the framework in Figure 1 to study the inﬂuence of open innovation strategies on the adoption of radical and incremental eco-innovations in this industry. This approach is somehow similar to the framework proposed by Marotta and Nazzaro [59] on the issue of the determinants of value creation processes on farms, and further developed by Marotta and Nazzaro [13] on their recent analysis of value creation in wineries. Triguero et al. [16] concluded that customer pressure fosters eco-innovation and high standards and requirements related to food safety. Regarding ﬁrms’ R&D resources, their ﬁndings are not conclusive. Similarly, Cuerva, Triguero-Cano and Córcoles [1] corroborated part of the proposed framework by comparing eco-friendly and non-eco-friendly innovations in the Spanish food and beverage industry. They showed that these three driving factors exercised diﬀerent inﬂuences on eco-product and eco-process innovations compared to non-environmental agri-food ﬁrms. Additionally, Bossle, De Barcellos and Vieira [15], in their analysis of the Brazilian food industry, proposed a relatively diﬀerent framework distinguishing between internal (e.g., resources) and external factors (e.g., collaboration with partners). Finally, although not focused on eco-innovation, Cainelli, Mazzanti, and Zoboli [60] stresses the inﬂuence of cooperation in the French food industry for the development of innovations. Based on the proposed framework in Figure 1 and considering that the market and regulatory factors are equal and constant for all ﬁrms in this industry, we will focus on the technology push factors. Therefore, we can argue that the ﬁrm’s resources, capabilities, and cooperation with stakeholders will make a diﬀerence in fostering eco-innovation in this particular sector. Hence, we will analyze the diverse combinations of resources, capabilities, and cooperation activities that result in the development of eco-product and eco-process innovations in the agri-food industry. Despite recent eﬀorts on the role of internal factors on eco-innovation, such as environmental management [52], skilled personnel [57], equipment renewal [5], technological capabilities [4], or cooperation [61], research on their inﬂuence is still very limited [62]. According to the resource-based view (RBV) of the ﬁrm [63,64] and from a Dynamic Capabilities perspective [65], certain ﬁrm resources and capabilities (valuable, rare, and imperfectly imitable) may be required to successfully develop and adopt eco-innovations. Therefore, the RBV provides an appropriate theoretical basis for analyzing eco-innovation, although the literature shows that there is an overlap between eco-innovation and general innovation processes [66]. A ﬁrm’s eco-innovation capacity will be connected to the pool of knowledge, resources, and capabilities that is available within the company [67]. However, most research on the topic addresses ﬁrms’ resources and capabilities that are not speciﬁc to eco-innovation and often not internally diﬀerentiated [68]. In this paper, we will explain why some ﬁrms internally develop eco-innovation through the analysis of the combinations of resources and capabilities that increase their eco-innovation performance. In this sense, Horbach [3] contends that internal R&D, high investment intensity, and improvements in a company´s innovative capacity are important drivers of eco-innovation, since the “availability of greater technical knowledge within a company moderates its vulnerability in the face of the demands of new environmental regulations” [69] (p. 307). Eco-innovative activity depends directly on R&D activity, which is inﬂuenced by past activities (dependence on the technological trajectory) and activities of Int. J. Environ. Res. Public Health 2020, 17, 2588 6 of 19 other companies in the same industry/sector. The empirical literature is not conclusive. While some empirical works show that R&D is essential for all types of eco-innovation [70], other research focused on the food industry ﬁnds inconclusive results [16]. Although R&D investment is considered to be a source for eco-innovation [71] that provides ﬁrms with a competitive advantage in it [72], the inﬂuence of technological capabilities on eco-innovation processes and its causal relationship has not been thoroughly elucidated to date [2]. Furthermore, apart from technological capabilities, eco-innovation activities will require the ﬁrm to have access to ﬁnancial resources [53], just as any other type of innovation does. The lack of ﬁnancial resources is one of the barriers to eco-innovation that is identiﬁed in the literature [72]. Having access to one’s own ﬁnancial resources or to private or public funding will allow the ﬁrm to conduct the necessary investments to internally develop environmental innovations as the availability of ﬁnancial resources themselves or ﬁnancial slack inﬂuences eco-innovation [32]. In this sense, own-ﬁnancing will allow ﬁrms to approach their eco-innovation activities with greater independence [73]. Regarding proﬁtability, Przychodzen and Przychodzen [74] studied the relationship between the ﬁnancial performance and eco-innovation activities of a sample of Polish and Hungarian ﬁrms. According to them, eco-innovative companies have lower proﬁles of exposure to ﬁnancial risk. “The information asymmetries could imply that the cost of ﬁnancial resources increases and spreads due to a worsening in proﬁtability from the higher risk level of the investments in eco-innovation” [73] (p. 260). Hence, we can expect that higher ﬁnancial performance will increase eco-innovation behavior through indebtedness and reduced ﬁnancial risk. Organizational capability is also a valuable resource to be considered as a driver of eco-innovations [5], specially, for internal development. In this sense, Environmental Management Systems and other eco-organizational innovations and their implementation create organizational capacities and lead to the development of technological eco-innovations [1]. Additionally, several studies have identiﬁed the positive eﬀects of incorporating external knowledge, and, compared with other innovations, “eco-innovation activities seem to require more external sources of knowledge and information” [3] (p.523). Cooperation is of high importance for eco-innovation because of its characteristics, such as double externality, including positive spillovers. Moreover, the transition towards more sustainable production and consumption patterns necessarily involves several private and public actors in a system [75]. Eco-innovations require more cooperation than other innovations, given their systemic and complex character, and that eco-innovators have to leverage on the competences of external partners to a higher extent than other innovators [7]. Companies cooperate in order to reduce and share the risk, costs, and uncertainty that are associated with R&D activities [76,77]. External knowledge from customers, suppliers, and other agents are keys to environmental innovation [61,75]. Despite the limited number of studies on the inﬂuence of open innovation modes on eco-innovation in food ﬁrms, some interesting research shows that the use of a variety of external knowledge sources has a positive inﬂuence on eco-innovation in the manufacturing sector [70,72]. According to Acosta, Coronado, Ferrándiz, León and Moreno [21], food ﬁrms take advantage of external knowledge and might have no need to generate this knowledge through internal R&D expenditures. A distinctive characteristic of the food industry in Europe is that ﬁrms have low R&D intensity while producing a signiﬁcant number of innovations [19]. The agri-food industry is dominated by SMEs, which lack knowledge on how to commercialize their own technology [78], and most innovations are mere improvements to new food products or variations of existing ones [20]. However, these ﬁrms are continuously exposed to external sources of knowledge, professional relationships, and a well-developed inter-industry network [21]. This paper analyses the inﬂuence of cooperative activities and the combination of resources and capabilities on the internal development of eco-innovations by ﬁrms in a traditional industry—the agri-food industry in Spain (see Figure 2). Int. J. Environ. Res. Public Health 2020, 17, 2588 7 of 19 Figure 2. The inﬂuence of cooperative activities and the combination of resources and capabilities on the internal development of eco-innovations by ﬁrms. In our paper, we will also include two additional variables: the size and group of ﬁrms. Both reﬂect the ﬁrm’s availability to ﬁnancial, human, and even organizational resources and capabilities due to its bigger size and/or from being part of a bigger corporation [79]. In this regard, size has also been analyzed as a source for eco-innovation [17], since larger companies are supposed to have higher levels of external ﬁnancing for eco-innovation [79]. Additionally, the availability of ﬁnancial resources is related to R&D, since ﬁrms will invest if they can access suﬃcient ﬁnancing at a reasonable cost, and this availability depends, among other things, on the characteristics of the ﬁrm, such as its size [80]. An additional goal of this study is the distinction between process and product eco-innovations. Although it is true that if a company decides to put an eco-innovative product on the market, it will necessarily have to implement eco-innovative production processes, and vice versa (i.e., if the company adopts eco-innovative production processes, the ﬁnal product of these eco-innovative processes will obviously be eco-innovative), there is a gap between theory and ﬁrm innovation behavior. Regarding eco-innovation performance, ﬁrms can adopt signiﬁcant changes in their production processes by adopting cleaner technologies. In these cases, the ﬁnal product may be more eco-friendly due to the reduction of environmental harm, but it does not mean that these ﬁrms are introducing eco-product innovations. Food ﬁrms introduce cleaner processes to reduce energy use or waste so as to increase their production eﬃciency through cost reduction, and they also implement End-of-Pipe technologies to comply with environmental legislation [4,11]. Both are process eco-innovations, but there are innovations related to the improvements to existing products or the development of new eco-products that achieve other purposes [81,82]. Although previous empirical research states that “ﬁrms adopt both types of eco-innovations to improve their competitive advantage, because one type of innovation often requires the other” [58] (p.16), the adoption of eco-product and eco-process innovation relies on diﬀerent resources, capabilities, and knowledge bases. Thus, the study of the speciﬁc conditions that encourage ﬁrms to develop each type of eco-innovation is considered separately. In addition, the conditions and core competencies that foster internal development of eco-product and eco-process innovations by Spanish SME food ﬁrms are heterogeneous, due to the complexity being higher for eco-innovation than for traditional innovation [57,70]. In fact, the introduction of sustainable processes (green manufacturing) and eco-products is a major innovative trend in the food industry, but each company shows innovative Int. J. Environ. Res. Public Health 2020, 17, x 6 of 19 information asymmetries could imply that the cost of financial resources increases and spreads due to a worsening in profitability from the higher risk level of the investments in eco-innovation” [73] (p. 260). Hence, we can expect that higher financial performance will increase eco-innovation behavior through indebtedness and reduced financial risk. Organizational capability is also a valuable resource to be considered as a driver of eco-innovations [5], specially, for internal development. In this sense, Environmental Management Systems and other eco-organizational innovations and their implementation create organizational capacities and lead to the development of technological eco-innovations [1]. Additionally, several studies have identified the positive effects of incorporating external knowledge, and, compared with other innovations, “eco-innovation activities seem to require more external sources of knowledge and information” [3] (p.523). Cooperation is of high importance for eco-innovation because of its characteristics, such as double externality, including positive spillovers. Moreover, the transition towards more sustainable production and consumption patterns necessarily involves several private and public actors in a system [75]. Eco-innovations require more cooperation than other innovations, given their systemic and complex character, and that eco-innovators have to leverage on the competences of external partners to a higher extent than other innovators [7]. Companies cooperate in order to reduce and share the risk, costs, and uncertainty that are associated with R&D activities [76,77]. External knowledge from customers, suppliers, and other agents are keys to environmental innovation [61,75]. Despite the limited number of studies on the influence of open innovation modes on eco-innovation in food firms, some interesting research shows that the use of a variety of external knowledge sources has a positive influence on eco-innovation in the manufacturing sector [70,72]. According to Acosta, Coronado, Ferrándiz, León and Moreno [21], food firms take advantage of external knowledge and might have no need to generate this knowledge through internal R&D expenditures. A distinctive characteristic of the food industry in Europe is that firms have low R&D intensity while producing a significant number of innovations [19]. The agri-food industry is dominated by SMEs, which lack knowledge on how to commercialize their own technology [78], and most innovations are mere improvements to new food products or variations of existing ones [20]. However, these firms are continuously exposed to external sources of knowledge, professional relationships, and a well-developed inter-industry network [21]. This paper analyses the influence of cooperative activities and the combination of resources and capabilities on the internal development of eco-innovations by firms in a traditional industry—the agri-food industry in Spain (see Figure 2). Int. J. Environ. Res. Public Health 2020, 17, 2588 8 of 19 performances. They neither have the resources and capabilities, nor are able to combine the resources and skills to meet the challenges involved in each type of eco-innovation in the same way. 3. Materials and Methods 3.1. Database The food industry is one of the most important branches of the national economy in Spain and the European Union, with high relevance for employment and economic output. Spanish food ﬁrms are mainly process innovation-oriented [16]. In this industry, new technologies are developed by upstream industries, and innovation occurs through equipment and capital good investments [83]. The original sample contained the data of 277 agri-food companies operating in Spain in 2016. Taking into account the fact that the food industry is a low-tech and mature sector, the adoption of eco-processes is more habitual than the introduction of eco-products. This evidence is shown by our data. From the initial sample, 79 companies had developed technological eco-innovations (products or processes) in the last 3 years. Speciﬁcally, 21 companies had developed product eco-innovations, and 66 had developed process eco-innovations. Within the companies that developed product eco-innovations, 16 relied on internal innovations, and ﬁve acquired that innovation. For the process eco-innovations, up to 32 companies developed the processes internally, while 34 acquired the eco-innovations. 3.2. Methodology This study used qualitative comparative analysis (QCA). Compared to traditional methods, QCA oﬀers a series of advantages that made it appropriate for this study. Contrary to traditional multiple regression analysis, QCA relies on asymmetrical relationships overcoming the limitations that appear due to the linearity and complementary associations between variables [84]. The goal of traditional methods has been to analyze the eﬀect of a single variable on a particular outcome. In this regard, QCA allows one to discover the combination of the antecedent conditions (independent variables) that lead to a given outcome (in this study, the internal development of technological eco-innovations). QCA entails equiﬁnality, since diﬀerent associations between variables can result in the same outcome [85]. Each one of the possible associations or combinations of variables is known as a recipe. QCA considers both the presence and the absence of antecedent conditions [86]. Moreover, it is an appropriate method for the analysis of the data of this study since it oﬀers valid responses when using small-to-intermediate research designs [87]. In this study, two speciﬁc QCA methods have been employed: crisp-set qualitative comparative analysis (csQCA) and fuzzy-set qualitative comparative analysis (fsQCA). csQCA is used for binary variables (i.e., the company develops/does not develop internal product eco-innovation). csQCA calibration uses categorical conditions based on a dichotomy, assigning full membership (value of 1) and full non-membership (value of 0) to each condition. On the other hand, fsQCA is appropriate for variables with continuous values (i.e., the number of employees of a company). fsQCA categorizes the variables into meaningful groups of cases [88]. The cut-oﬀ values range from full membership (0.95) to full non-membership (0.05) with the 0.5 case representing the maximum ambiguity. Fuzzy logic calibration combines qualitative and quantitative methods and requires theoretical and substantive knowledge of the context [87,89]. After calibrating the variables, the analysis of the necessity is done. The goal is to identify if all, or nearly all, instances of the outcome have the same condition for some of the considered variables. A condition is necessary if its consistency is particularly high (>0.95) and its coverage is not too low (>0.5). The creation of the truth table is the next step. The truth table sorts the cases according to the combinations of the causal conditions they exhibit (2k rows). It considers all logically possible combinations of conditions, even those without empirical instances, and assesses the consistency of the cases in each row with respect to the outcome. Each empirical case (i.e., company) corresponds to a conﬁguration (a row of the truth table) depending on the antecedent conditions that it meets [87,90]. Int. J. Environ. Res. Public Health 2020, 17, 2588 9 of 19 The next step is the reduction of the cases (rows) using algorithms. A version of the Quine-McCluskey algorithm is the most commonly used algorithm to perform the logical reduction of the statements [91]. By using Boolean algebra, QCA identiﬁes the minimal set of causal conditions that are suﬃcient to produce the outcome. The goodness-of-ﬁt of the row reduction depends on two criteria: coverage and consistence. Similarly, to the traditional R2 value, the coverage refers to the number of cases for which a conﬁguration is valid. The consistency refers to the percentage of causal conﬁgurations with similar compositions that result in the same outcome value [84,92]. The starting point of the study is the consideration of the diﬀerent factors stimulating product and process eco-innovations in companies [93]. For that reason, diﬀerent models are proposed to evaluate those diﬀerences. First, a general model including all companies of the sample that developed internal/acquired technological eco-innovations was performed. After that, speciﬁc models, including the companies that developed internal/acquired product eco-innovations and internal/acquired process eco-innovations, were developed. Table 2 shows the description of the variables that were considered and the transformation values of the outcome and the antecedent conditions into fuzzy and crisp set terms. Table 2. Variable deﬁnitions and calibration values. Condition Description Internal technological Eco-Innovation Internal Product Eco-Innovation Internal Process Eco-Innovation Eco-Organizational Capabilities Group Cooperation R&D Size Capital The company develops an internal product or process eco-innovation The company develops an internal product eco-innovation The company develops an internal process eco-innovation The company develops a non-technological eco-innovation (marketing, organizational) The company is part of a company group Number of internal or external partners the company cooperates with in the development of eco-innovations R&D expenditures as a percentage of sales Number of employees Company capital (thousands of euros) Proﬁtability Company proﬁt margin (%) 4. Empirical Results and Discussion Membership Threshold Values Full Non-Membership (0.05) Crossover Point (0.5) Full Membership (0.95) 0 0 0 0 0 0 0 4.9 3 −5.77 1 1 1 1 1 2 15 361.4 21219.5 15.28 0.95 1.95 76.0 1125 3.47 The individual eﬀect of each antecedent condition on the development of internal technological eco-innovation is shown in Table 3. The same data are shown in Table 4 for the internal development of product eco-innovations, and in Table 5 for the internal development of process eco-innovations. The antecedent conditions alone are insuﬃcient for the outcome. In the case of the absence of the Group variable for product eco-innovation, a high value appears. This is the result of the small number of companies in the sample belonging to a company group. Moreover, as the value is lower than 0.95, the condition is not considered suﬃcient. Int. J. Environ. Res. Public Health 2020, 17, 2588 10 of 19 Table 3. Companies that developed internal technological eco-innovations (any type). Analysis of the necessary conditions. Conditions Tested * Consistency Coverage Group ~Group Eco-Organizational Capabilities ~ Eco-Organizational Capabilities Cooperation ~Cooperation R&D ~ R&D Size ~Size Capital ~Capital Proﬁtability ~Proﬁtability 0.090909 0.909091 0.454545 0.545455 0.752500 0.247500 0.457955 0.542046 0.417045 0.582955 0.359964 0.641136 0.491591 0.508409 1.000000 0.533333 0.606061 0.521739 0.839503 0.275278 0.647286 0.498224 0.523538 0.583618 0.494210 0.599575 0.584121 0.533000 * The symbol (~) represents the negation of the characteristic. Table 4. Companies developing internal product eco-innovations. Analysis of the necessary conditions. Conditions Tested * Consistency Coverage Group ~Group Eco-Organizational Capabilities ~ Eco-Organizational Capabilities Cooperation ~Cooperation R&D ~ R&D Size ~Size Capital ~Capital Proﬁtability ~Proﬁtability Process eco-innovation ~ Process eco-innovation 0.058824 0.921176 0.588235 0.411765 0.732941 0.267059 0.419412 0.580588 0.357647 0.642353 0.454118 0.545882 0.442941 0.442941 0.529412 0.470588 1.000000 0.761905 0.769231 0.666667 0.980330 0.488698 0.839811 0.730570 0.784516 0.766316 0.845564 0.721057 0.726133 0.726133 0.900000 0.666667 * The symbol (~) represents the negation of the characteristic. Table 5. Companies developing internal process innovations. Analysis of the necessary conditions. Conditions Tested * Consistency Coverage Group ~Group Eco-Organizational Capabilities ~ Eco-Organizational Capabilities Cooperation ~Cooperation R&D ~ R&D Size ~Size Capital ~Capital Proﬁtability ~Proﬁtability Product eco-innovation ~ Product eco-innovation 0.093750 0.906250 0.406250 0.593750 0.755313 0.244688 0.500938 0.499062 0.409375 0.590625 0.292500 0.707500 0.470000 0.530000 0.156250 0.843750 1.000000 0.460317 0.500000 0.475000 0.747603 0.232551 0.603085 0.405124 0.438714 0.522966 0.356165 0.569990 0.514716 0.461120 0.555556 0.473684 * The symbol (~) represents the negation of the characteristic. Table 6 shows the results of the model predicting the development of technological eco-innovation in agri-food companies. By using the notation introduced by Ragin and Fiss [94], black circles indicate the presence of the condition (•), white circles indicate the absence of the condition ( ), and the absence (cid:35) Int. J. Environ. Res. Public Health 2020, 17, 2588 11 of 19 of a circle indicates that the condition is not binding in that conﬁguration. Up to four diﬀerent recipes (conﬁgurations) result in the internal development of technological eco-innovations in companies. All of the paths (conﬁgurations 1 to 4, Table 6) require cooperation with internal and external partners, while the inﬂuence of R&D is not binding for most of the conﬁgurations. Previously, some studies have suggested that R&D is less important to eco-innovation when compared with collaboration strategies in the industrial sector [61]. The limited eﬀect of R&D spending on technological eco-innovation was also reported by Cuerva, Triguero-Cano and Córcoles [1] who suggested that even though R&D promotes mainstream innovation, the case was not the same for eco-innovation. Table 6. Model predicting the development of internal technological eco-innovations (product or process) in agri-food companies. Conﬁguratio no. Group Eco-Organ. Capab Cooperation R&D Size Capital Proﬁtability 1 2 3 4 (cid:35) • (cid:35) • (cid:35) • • • • • (cid:35) (cid:35) • (cid:35) (cid:35) • (cid:35) (cid:35) • • • Coverage Raw Unique 0.2780 0.2564 0.0266 0.0316 0.1259 0.1043 0.0172 0.0223 Consistency 0.9154 0.8558 1.0000 1.0000 Solution coverage: 0.4311 Solution consistency: 0.9068 Eco-Organizational Capabilities (Eco-Organ. Capab.). Frequency threshold = 1, and consistency threshold = 0.8186. (cid:35) Up to 27.8% of companies developing internal technological eco-innovations are small companies that neither belong to a company group nor adopt eco-organizational capabilities. They rely only on cooperation (conﬁguration 1, Table 6). Eﬀectively, cooperation enables the acquisition of complex and new knowledge required for eco-innovation. Moreover, cooperative food ﬁrms with high proﬁt ratios also have a high probability of adopting internal eco-innovations (up to 25.6% of companies in conﬁguration 2, Table 5). Cooperation improves ﬁrm eﬃciency and proﬁts [76], and it is also considered an essential part of the open innovation concept [16]. Cooperation with partners has recently been identiﬁed as a driver for the development of eco-innovations in the manufacturing sector in general [95,96] and for the introduction of radical eco-innovations, speciﬁcally in the food and beverage sector [16]. Other conditions, such as ﬁrm size or technological capabilities, are only important in conﬁgurations that include companies that belong to a company group and have high capital ratio and/or proﬁtability (conﬁgurations 3 and 4, Table 6), but the percentage of food ﬁrms is lower (around 2.7 and 3.2 percent, respectively). Although some studies show that ﬁrm size has a positive inﬂuence on eco-innovation, the empirical evidence is not all conclusive. Firm size has been identiﬁed as crucial, but also as an indeterminate factor in explaining eco-innovations in the manufacturing industry [3,93]. This result does not mean that size or R&D do not have an inﬂuence on the adoption of internal eco-innovation by food companies, but it shows that cooperation has a more essential role than ﬁnancial and technological capabilities related to size. The model with the conﬁgurations resulting in the development of internal product eco-innovation is shown in Table 7. The coverage value of the model is high (0.56), meaning that the model is valid for a large number of agri-food companies. Up to seven diﬀerent conﬁgurations lead to the internal development of the less frequent type of technological eco-innovation in the food industry. Conﬁgurations 1 and 4 have high raw coverages, and thus deserve further attention since they explain the conditions of more companies (14.6 and 14.8%, respectively). Conﬁguration 1 includes small companies that do not belong to a company group and have low proﬁtability. However, they succeed in developing internal product eco-innovations through their eco-organizational capabilities and knowledge cooperation. These ﬁrm capabilities with suﬃcient capital availability foster the adoption of product eco-innovations. As pointed out by Dora et al. [97], eco-organizational capabilities related to quality assurance methods, such as food safety, Hazard Analysis and Critical Control Points (HACCP), British Retail Consortium (BRC), International Organization for Standardization (ISO), or microbiological issues, are directly linked to product innovations in the food sector. Cooperation is also crucial for the adoption of product eco-innovations by SMEs due to the low technological Int. J. Environ. Res. Public Health 2020, 17, 2588 12 of 19 opportunities by SMEs compared to large companies [14]. Higher levels of ﬁnancial performance and R&D capabilities have been also previously proposed as drivers of the internal development of eco-product innovations [73,74,98]. Hence, the introduction of novel eco-products in the food industry, such as functional foods, plant-based meats, or foodstuﬀs using genetic engineering are normally carried out by large food companies with corporate proﬁtability and R&D departments. In conﬁguration 4, high proﬁtability and R&D expenditure replace a high capital ratio in conﬁguration 1, explaining how these factors predict the adoption of eco-product innovation by about 14.7% of the sample. The adoption of eco-process innovations interacts with other factors to explain the development of eco-product innovations: large ﬁrms with high proﬁtability that cooperate and adopt eco-organizational innovations (conﬁguration 2); ﬁrms with high proﬁtability that cooperate and adopt eco-organizational innovations (conﬁguration 3); and ﬁrms with high proﬁtability and capital ratios that cooperate (conﬁguration 5). The two former conﬁgurations show how the complementarity of eco-organizational innovations and eco-process innovations explain the adoption of eco-product innovations, while the latter shows the role of capital in adopting eco-process innovations that enhances eco-product innovations. The ﬁrst result is in line with the existence of complementarities across the diﬀerent types of eco-innovation activities shown in the related literature [2,4], as well as the positive inﬂuence of proactive environmental management and incremental organizational eco-innovations in the adoption of eco-innovations by Spanish food ﬁrms [99]. The second one indicates the capital requirements needed by the food industry to do eco-process innovations, often related to the acquisition of new machinery and equipment [16]. Table 7. Model predicting the development of internal product eco-innovations in agri-food companies. Conﬁguration no. Group Eco-Organ Capab Cooperation R&D Size Capital Proﬁtability Process Eco-Innovation Coverage Raw Unique Consistency 1 2 3 4 5 6 7 (cid:35) (cid:35) (cid:35) (cid:35) (cid:35) • (cid:35) • • • • (cid:35) (cid:35) (cid:35) • • • • • • • • (cid:35) (cid:35) • (cid:35) • • (cid:35) (cid:35) (cid:35) • (cid:35) • (cid:35) • (cid:35) • (cid:35) • • • (cid:35) • • • (cid:35) 0.1464 0.0894 0.0865 0.1477 0.1018 0.1165 0.0312 0.0841 0.0506 0.0047 0.0580 0.1018 0.1165 0.0311 0.9614 0.9682 0.9671 0.9436 0.9454 1.0000 1.0000 Solution coverage: 0.5635 Solution consistency: 0.9746 Frequency threshold = 1, and consistency threshold = 0.8630. (cid:35) To summarize, all conﬁgurations need cooperation to develop internal product eco-innovations. The importance of external sources of knowledge and information in the development of eco-innovation activities has been previously illustrated by Horbach [3]. However, other conditions are also important, since they appear in most of the conﬁgurations for internal product eco-innovations. This is the case for eco-organizational capabilities, high corporate proﬁtability, and the development of process eco-innovations. Table 8 shows the model predicting the development of internal process eco-innovations in agri-food companies. The solution consistency of the models ranges from 0.84 to 1.00, which are higher than the minimum value (0.8) recommended by Ragin [88]. Similar to the results for the companies developing internal product eco-innovations, the conﬁguration that includes most of the companies (28.5% of the sample) relies only on cooperation with partners (conﬁguration 1). Previously, cooperation had been identiﬁed as the main driver of continuous process innovation in a review covering the fertilizer and agricultural sector and other related studies [100]. Unlike other types of eco-innovation, the absence of eco-organizational innovations and product eco-innovation in food companies is not a constraint to developing internal process eco-innovation in the resource and capability combinations that include most of the companies (conﬁgurations 1 and 2, Table 8). The development of process eco-innovations is considered less demanding of complementary eco-innovations related to non-technological and technological capabilities than product eco-innovations. Int. J. Environ. Res. Public Health 2020, 17, 2588 13 of 19 Table 8. Model predicting the development of internal process eco-innovations in agri-food companies. Conﬁguration no. Group Eco-Organ Capab. Cooperation R&D Size Capital Proﬁtability Product Eco-Innovation Coverage Raw Unique Consistency 1 2 3 4 5 (cid:35) • (cid:35) • (cid:35) (cid:35) • (cid:35) • • • • • • • • • (cid:35) • (cid:35) (cid:35) (cid:35) (cid:35) (cid:35) (cid:35) • (cid:35) • • (cid:35) • (cid:35) (cid:35) • (cid:35) 0.2853 0.2097 0.0434 0.0200 0.0200 0.1322 0.0566 0.0306 0.0200 0.0072 0.8986 0.8483 1.0000 0.8530 1.0000 Solution coverage: 0.4126 Solution consistency: 0.9097 Frequency threshold = 1, and consistency threshold = 0.8138. (cid:35) Since innovation in the agri-food sector is mostly incremental [101] and relies mainly on external knowledge through cooperation [21], the eﬀect of company R&D spending on internal technological eco-innovations was expected to be limited, compared to the results obtained for Spanish manufacturing ﬁrms in general [102]. Previously, some studies have shown the limitations of R&D for promoting eco-innovation in the sector [1]. Hence, the same results are obtained for technological eco-innovation in general (Table 6). However, the joint use of R&D and cooperation for food ﬁrms (conﬁguration 2) is also identiﬁed by about 21.0% of process eco-innovators. This result shows that about one-ﬁfth of companies adopting process eco-innovations also invest in R&D, but in a complementary mode to cooperate with external actors. Similarly, cooperation interacts with the rest of the ﬁnancial, technological, and eco-organizational capabilities to predict the development of internal process eco-innovations (conﬁgurations 3, 4, and 5). To sum up, most food ﬁrms adopting internal product and process eco-innovation depend on cooperation. However, our empirical models show diﬀerent conﬁgurations to predict the development of each type of eco-innovation. Results are diﬀerent when the sample is divided into internal product and process eco-innovations. Once the sample is divided, high R&D spending can be considered a beneﬁcial condition to developing both product eco-innovations (conﬁgurations 4 and 6, Table 7) and process eco-innovations (conﬁgurations 2 and 4, Table 8) under speciﬁc circumstances. Financial capabilities related to size, proﬁt, and capital also interact in a diﬀerent way to predict each eco-innovative path. However, the most insightful result is regarding the complementarities between eco-organizational capabilities in each type of eco-innovation. Speciﬁcally, the adoption of eco-process innovations (technological capabilities) and eco-organizational capabilities are relevant for product eco-innovation (Table 7). Although this type of eco-innovation also depends on market pull factors [4], these ﬁndings make a lot of sense. On one hand, the implementation of process eco-innovations in the upstream stages must lead to the development of downstream product eco-innovations. On the other hand, eco-organizational capabilities enhance the internal development of product eco-innovations, probably because environmental management strategies in the food industry (such as food safety and quality systems or labelling) also contributes to the success of product eco-innovations. In particular, some certiﬁcations and labels (i.e., EMS) provide valuable information to the consumer aﬀecting their conﬁdence about new products in a traditional sector, such as the food and beverage industry [103]. However, the opposite does not apply for process eco-innovations. Product eco-innovations are not a crucial factor to the development of internal process eco-innovation by agri-food companies, where cooperation and R&D are more relevant (Table 8). 5. Conclusions Although the number of studies about eco-innovation has signiﬁcantly increased in the last decade [2], some aspects about companies’ eco-innovation drivers and conditions remain unclear. In general, studies have focused on the whole industrial sector [61,93,96], with few paying attention to speciﬁc sectors, such as the agri-food industry [16]. Regarding speciﬁc eco-innovations, the diﬀerences between product and process eco-innovations are persistent in the current studies. In this scenario, this study went beyond that diﬀerentiation and analyzed the conditions that promote internal technological Int. J. Environ. Res. Public Health 2020, 17, 2588 14 of 19 eco-innovations, distinguishing between products and processes within a company. To achieve this, a new method (QCA) with proven guarantees in the business and management area has been used [92]. The study showed that the proposed conditions are useful for explaining the development of internal technological eco-innovations in the agri-food industry. All models and conﬁgurations concur with the idea that cooperation with external partners is the key to success in the development of internal technological eco-innovations in general, and internal product and process eco-innovations, more speciﬁcally. It must be considered that the agri-food sector is mainly composed of SMEs, and empirical evidence has found that these small companies can obtain the best results in developing technological eco-innovations through the recipe of cooperation. This conclusion encourages the design of policies and incentives to promote cooperation between companies operating in the agri-food industry so as to enhance innovative patterns and make them more environmentally friendly, which also allows for the increase of competitive advantages and the eﬃcient use of capabilities and limited resources by small ﬁrms operating in a traditional low-tech sector. Although similarities in the conditions of developing product and process eco-innovations have appeared, some diﬀerences also exist. Companies that develop eco-organizational capabilities and process eco-innovations tend also to develop product eco-innovations. However, these conditions do not apply to the development of process eco-innovations. This may be the result of the stage of production in which each type of eco-innovation is made. Upstream eco-innovation (process) encourages the development of subsequent eco-innovations (product), but the opposite does not apply. This analysis of the eﬀects of the individual types of eco-innovation provides important information regarding the design and planning of eco-innovation strategies by SMEs. According to our ﬁndings, ﬁrms should cooperate with external partners to gather the exploitation of inbound information ﬂow that is valuable to eco-innovation. In addition, ﬁrms enabling access to necessary knowledge through these cooperative relationships can improve their eco-processes and eco-products. At that point, the adoption of process eco-innovations can develop skills and capabilities that can be used to improve and introduce eco-products. According to our results, food companies that are more committed to eco-process and eco-organizational changes are also more likely to introduce product eco-innovations, taking advantage of the complementarity and synergies derived from open innovation schemes in an industry based on natural resources, such as the agri-food industry. In this regard, our ﬁndings are in line with the previous literature showing that the integration of potential solutions for the production stage (eco-process), the consumption stage (eco-products), and the production-supply-disposal chain (eco-organizational) can help in the transition towards a circular system and more sustainable innovative practices in the food industry [104]. Limitations in this study also appeared, due to the low number of companies that were analyzed. Due to the outcome’s speciﬁcity, the number of companies that was considered in the analysis of each outcome within the original sample was inevitably small. However, this limitation is considered to have been solved by using QCA. Although the coverage and the consistency of the models is adequate, the question regarding whether the variables that were considered in this study were the best proxies for capturing the development of the technological eco-innovations in the sector remains unanswered. Author Contributions: Conceptualization, Á.T. and Á.G.-M.; Formal analysis, A.R. and Á.G.-M.; Methodology, A.R.; Project administration, Á.T. and Á.G.-M.; Resources, Á.T. and Á.G.-M.; Software, A.R.; Supervision, Á.T. and Á.G.-M.; Writing—original draft, A.R., Á.T. and Á.G.-M. All authors have read and agreed to the published version of the manuscript. Funding: Spanish Ministry of Science, Innovation and Universities Grant: RTI2018-101867-B-100. European Regional Development Fund: 2018/11744. Acknowledgments: This study was partially supported by the Spanish Ministry of Science, Innovation and Universities (RTI2018-101867-B-100), the University of Castilla-La Mancha and the European Regional Development Fund (2018/11744). Conﬂicts of Interest: Authors declare no conﬂict of interest. Int. J. Environ. Res. Public Health 2020, 17, 2588 15 of 19 References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Cuerva, M.C.; Triguero-Cano, Á.; Córcoles, D. Drivers of green and non-green innovation: Empirical evidence in Low-Tech SMEs. J. Clean. Prod. 2014, 68, 104–113. [CrossRef] Díaz-García, C.; González-Moreno, Á.; Sáez-Martínez, F.J. Eco-innovation: Insights from a literature review. Innov. Manag. Policy Pract. 2015, 17, 6–23. [CrossRef] Horbach, J. Determinants of environmental innovation-New evidence from German panel data sources. Res. Policy 2008, 37, 163–173. [CrossRef] Triguero, A.; Moreno-Mondéjar, L.; Davia, M.A. Drivers of diﬀerent types of eco-innovation in European SMEs. Ecol. Econ. 2013, 92, 25–33. [CrossRef] Kesidou, E.; Demirel, P. On the drivers of eco-innovations: Empirical evidence from the UK. Res. Policy 2012, 41, 862–870. [CrossRef] Ashford, N.A.; Hall, R.P. The Importance of Regulation-Induced Innovation for Sustainable Development. Sustainability 2011, 3, 270–292. [CrossRef] De Marchi, V. Environmental innovation and R&D cooperation: Empirical evidence from Spanish manufacturing ﬁrms. Res. Policy 2012, 41, 614–623. [CrossRef] Popa, S.; Soto-Acosta, P.; Martinez-Conesa, I. Antecedents, moderators, and outcomes of innovation climate and open innovation: An empirical study in SMEs. Technol. Forecast. Soc. Chang. 2017, 118, 134–142. [CrossRef] Klewitz, J.; Hansen, E.G. Sustainability-oriented innovation of SMEs: A systematic review. J. Clean. Prod. 2014, 65, 57–75. [CrossRef] Schiederig, T.; Tietze, F.; Herstatt, C. Green Innovation in technology and innovation management—An exploratory literature review. R. D. Manag. 2012, 42, 180–192. [CrossRef] 11. Del Río González, P. Analysing the factors inﬂuencing clean technology adoption: A study of the Spanish pulp and paper industry. Bus. Strategy Environ. 2005, 14, 20–37. [CrossRef] 12. Revell, A.; Rutherfoord, R. UK environmental policy and the small ﬁrm: Broadening the focus. Bus. Strategy Environ. 2003, 12, 26–35. [CrossRef] 13. Nazzaro, C.; Stanco, M.; Marotta, G. The life Cycle of Corporate Social Responsibility in Agri-Food: Value Creation Models. Sustainability 2020, 12, 1287. [CrossRef] 14. Amara, D.B.; Chen, H.; Hafeez, M. Role of entrepreneurial opportunity identiﬁcation factors in the eco-innovation of agribusiness. Bus. Strategy Environ. 2020. [CrossRef] 15. Bossle, M.B.; De Barcellos, M.D.; Vieira, L.M. Why food companies go green? The determinant factors to adopt eco-innovations. Br. Food J. 2016, 118, 1317–1333. [CrossRef] 16. Triguero, A.; Fernández, S.; Sáez-Martinez, F.J. Inbound open innovative strategies and eco-innovation in the Spanish food and beverage industry. Sustain. Prod. Consum. 2018, 15, 49–64. [CrossRef] 17. García-Granero, E.M.; Piedra-Muñoz, L.; Galdeano-Gómez, E. Multidimensional Assessment of Eco-Innovation Implementation: Evidence from Spanish Agri-Food Sector. Int. J. Environ. Res. Public Health 2020, 17, 1432. [CrossRef] 18. Luo, J.; Guo, H.; Jia, F. Technological innovation in agricultural co-operatives in China: Implications for agro-food innovation policies. Food Policy 2017, 73, 19–33. [CrossRef] 19. Capitanio, F.; Coppola, A.; Pascucci, S. Indications for drivers of innovation in the food sector. Br. Food J. 2009, 111, 820–838. [CrossRef] 20. Baregheh, A.; Rowley, J.; Sambrook, S.; Davies, D. Innovation in food sector SMEs. J. Small Bus. Enterp. Dev. 2012, 19, 300–321. [CrossRef] 21. Acosta, M.; Coronado, D.; Ferrándiz, E.; León, M.D.; Moreno, P.J. The geography of university scientiﬁc production in Europe: An exploration in the ﬁeld of Food Science and Technology. Scientometrics 2017, 112, 215–240. [CrossRef] 22. García, M.; Alonso, A.; Tello, M.L.; De la Poza, M.; Villalobos, N.; Lansac, R.; Melgarejo, P. Idenﬁtying agri-food research priorities for Spain: 2017 results. Span. J. Agric. Res. 2018, 16. [CrossRef] 23. Mazzanti, M. Eco-innovation and sustainability: Dynamic trends, geography and policies. J. Environ. Plan. Manag. 2018, 61, 1851–1860. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 2588 16 of 19 24. Sáez-Martínez, F.J.; Triguero, A.; González-Moreno, A. A review of Open-innovation and Eco-innovation strategies in SMEs. In Research on Open-Innovation Strategies and Eco-Innovation in Agro-Food Industries; Triguero, A., González-Moreno, Á., Eds.; Chartridge Books Oxford: Witney, UK, 2019; pp. 9–23. 25. OECD. The Measurement of Scientiﬁc and Technological Activities: Guidelines for Collecting and Interpreting 26. Innovation Data; OCDE: Paris, France, 2005. Saidani, M.; Yannou, B.; Leroy, Y.; Cluzel, F.; Kendall, A. A taxonomy of circular economy indicators. J. Clean. Prod. 2019, 207, 542–559. [CrossRef] 27. Prieto-Sandoval, V.; Jaca, C.; Ormazabal, M. Towards a consensus on the circular economy. J. Clean. Prod. 2018, 179, 605–615. [CrossRef] 28. Kiani Mavi, R.; Saen, R.F.; Goh, M. Joint analysis of eco-eﬃciency and eco-innovation with common weights in two-stage network DEA: A big data approach. Technol. Forecast. Soc. Chang. 2019, 144, 553–562. [CrossRef] 29. Zhang, D.; Rong, Z.; Ji, Q. Green innovation and ﬁrm performance: Evidence from listed companies in China. Resour. Conserv. Recycl. 2019, 144, 48–55. [CrossRef] 30. Kusi-Sarpong, S.; Gupta, H.; Sarkis, J. A supply chain sustainability innovation framework and evaluation 31. methodology. Int. J. Prod. Res. 2019, 57, 1990–2008. [CrossRef] Stucki, T. Which ﬁrms beneﬁt from investments in green energy technologies?—The eﬀect of energy costs. Res. Policy 2019, 48, 546–555. [CrossRef] 32. Kiefer, C.P.; Carrillo-Hermosilla, J.; Del Río, P. Building a taxonomy of eco-innovation types in ﬁrms. A quantitative perspective. Resour. Conserv. Recycl. 2019, 145, 339–348. [CrossRef] 33. Diaz Lopez, F.J.; Bastein, T.; Tukker, A. Business model innovation for resource-eﬃciency, circularity and cleaner production: What 143 cases tell us. Ecol. Econ. 2019, 155, 20–35. [CrossRef] 34. Kirchherr, J.; Piscicelli, L.; Bour, R.; Kostense-Smit, E.; Muller, J.; Huibrechtse-Truijens, A.; Hekkert, M. Barriers to the circular economy: Evidence from the European Union (EU). Ecol. Econ. 2018, 150, 264–272. [CrossRef] 35. Li, D.; Huang, M.; Ren, S.; Chen, X.; Ning, L. Environmental legitimacy, green innovation, and corporate carbon disclosure: Evidence from CDP China 100. J. Bus. Ethics 2018, 150, 1089–1104. [CrossRef] 36. Ben Arﬁ, W.; Hikkerova, L.; Sahut, J.M. External knowledge sources, green innovation and performance. Technol. Forecast. Soc. Chang. 2018, 129, 210–220. [CrossRef] 37. Yuan, B.; Xiang, Q. Environmental regulation, industrial innovation and green development of Chinese manufacturing: Based on an extended CDM model. J. Clean. Prod. 2018, 176, 895–908. [CrossRef] 38. De Jesus, A.; Mendonça, S. Lost in transition? Drivers and barriers in the eco-innovation road to the circular economy. Ecol. Econ. 2018, 145, 75–89. [CrossRef] 39. Cai, W.; Li, G. The drivers of eco-innovation and its impact on performance: Evidence from China. J. Clean. Prod. 2018, 176, 110–118. [CrossRef] 40. Tang, M.; Walsh, G.; Lerner, D.; Fitza, M.A.; Li, Q. Green innovation, managerial concern and ﬁrm performance: An empirical study. Bus. Strategy Environ. 2018, 27, 39–51. [CrossRef] 41. Watson, R.; Wilson, H.N.; Smart, P.; Macdonald, E.K. Harnessing diﬀerence: A capability-based framework for stakeholder engagement in environmental innovation. J. Prod. Innov. Manag. 2018, 35, 254–279. [CrossRef] 42. Choi, H. Technology-push and demand-pull factors in emerging sectors: Evidence from the electric vehicle 43. market. Ind. Innov. 2018, 25, 655–674. [CrossRef] Feng, Z.; Chen, W. Environmental regulation, green innovation, and industrial green development: An empirical analysis based on the spatial Durbin model. Sustainability 2018, 10, 223. [CrossRef] 44. Huang, J.W.; Li, Y.H. Green innovation and performance: The view of organizational capability and social reciprocity. J. Bus. Ethics 2017, 145, 309–324. [CrossRef] 45. Gupta, H.; Barua, M.K. Supplier selection among SMEs on the basis of their green innovation ability using BWM and fuzzy TOPSIS. J. Clean. Prod. 2017, 152, 242–258. [CrossRef] 46. Costantini, V.; Crespi, F.; Palma, A. Characterizing the policy mix and its impact on eco-innovation: A patent analysis of energy-eﬃcient technologies. Res. Policy 2017, 46, 799–819. [CrossRef] 47. Beltrán-Esteve, M.; Picazo-Tadeo, A.J. Assessing environmental performance in the European Union: 48. Eco-innovation versus catching-up. Energy Policy 2017, 104, 240–252. [CrossRef] Jansson, J.; Nilsson, J.; Modig, F.; Hed Vall, G. Commitment to sustainability in small and medium-sized enterprises: The inﬂuence of strategic orientations and management values. Bus. Strategy Environ. 2017, 26, 69–83. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 2588 17 of 19 49. Zhang, Y.J.; Peng, Y.L.; Ma, C.Q.; Shen, B. Can environmental innovation facilitate carbon emissions reduction? Evidence from China. Energy Policy 2017, 100, 18–28. [CrossRef] 50. Notarnicola, B.; Tassielli, G.; Renzulli, P.A.; Castellani, V.; Sala, S. Environmental impacts of food consumption in Europe. J. Clean. Prod. 2017, 140, 753–765. [CrossRef] 51. McDonagh, P.; Prothero, A. Sustainability marketing research: Past, present and future. J. Mark. Manag. 52. 53. 54. 2014, 30, 1186–1219. [CrossRef] Frondel, M.; Horbach, J.; Rennings, K. What triggers environmental management and innovation? Empirical evidence for Germany. Ecol. Econ. 2008, 66, 153–160. [CrossRef] Segarra-Oña, M.V.; Peiró-Signes, A.; Mondéjar-Jiménez, J. Identifying variables aﬀecting the proactive environmental orientation of ﬁrms: An empirical study. Pol. J. Environ. Stud. 2013, 22, 873–880. Sáez-Martínez, F.J.; González-Moreno, Á.; Hogan, T. The role of university in eco-entrepreneurship: Evidence from the eurobarometer survey on attitudes of european entrepreneurs towards eco-innovation. Environ. Eng. Manag. J. 2014, 13, 2541–2549. 55. Triguero, A.; Córcoles, D.; Cuerva, M.C. Persistence of innovation and ﬁrm’s growth: Evidence from a panel of sme and large spanish manufacturing ﬁrms. Small Bus. Econ. 2014, 43, 787–804. [CrossRef] 56. Nazzaro, C.; Lerro, M.; Stanco, M.; Marotta, G. Do consumers like food product innovation? An analysis of willingness to pay for innovative food attributes. Br. Food J. 2019, 121, 1413–1427. [CrossRef] 57. Ghisetti, C.; Marzucchi, A.; Montresor, S. The open eco-innovation mode. An empirical investigation of eleven European countries. Res. Policy 2015, 44, 1080–1093. [CrossRef] 58. Mothe, C.; Nguyen-Thi, U.T.; Triguero, Á. Innovative products and services with environmental beneﬁts: Design of search strategies for external knowledge and absorptive capacity. J. Environ. Plan. Manag. 2018, 61, 1934–1954. [CrossRef] 59. Marotta, G.; Nazzaro, C. Il portafoglio di valori nell’impresa agricola multifunzionale: Nuovi approcci teorico-metodologici. Riv. Econ. Agrar. 2012, 2, 7–36. 60. Tanguy, C. Cooperation in the food industry: Contributions and limitations of the open innovation model. J. Innov. Econ. Manag. 2016, 1, 61–86. [CrossRef] 61. Cainelli, G.; Mazzanti, M.; Zoboli, R. Environmental innovations, complementarity and local/global cooperation: Evidence from North-East Italian industry. Int. J. Technol. Policy Manag. 2011, 11, 328–368. [CrossRef] 62. Kiefer, C.P.; González, P.D.R.; Carrillo-hermosilla, J. Drivers and barriers of eco-innovation types for sustainable transitions: A quantitative perspective. Bus. Strategy Environ. 2018, 28, 155–172. [CrossRef] 63. Wernerfelt, B. A resource-based view of the ﬁrm. Strateg. Manag. J. 1984, 5, 171–180. [CrossRef] 64. Barney, J. Firm resources and sustained competitive advantage. J. Manag. 1991, 17, 99–120. [CrossRef] 65. Teece, D.J.; Pisano, G.; Shuen, A. Dynamic capabilities and strategic management. Strateg. Manag. J. 1997, 18, 509–533. [CrossRef] 66. Aragon-Correa, J.A.; Leyva-de la Hiz, D.I. The inﬂuence of technology diﬀerences on corporate environmental patents: A resource-based versus an institutional view of green innovations. Bus. Strategy Environ. 2016, 25, 421–434. [CrossRef] 67. Doran, J.; Ryan, G. Firms’ skills as drivers of radical and incremental innovation. Econ. Lett. 2014, 125, 107–109. [CrossRef] 68. Lee, K.H.; Min, B. Green R&D for eco-innovation and its impact on carbon emissions and ﬁrm performance. J. Clean. Prod. 2015, 108, 534–542. [CrossRef] 69. Cañón-De-Francia, J.; Garcés-Ayerbe, C.; Ramírez-Alesón, M. Are more innovative ﬁrms less vulnerable to new environmental regulation? Environ. Resour. Econ. 2007, 36, 295–311. [CrossRef] 70. Marzucchi, A.; Montresor, S. Forms of knowledge and eco-innovation modes: Evidence from Spanish manufacturing ﬁrms. Ecol. Econ. 2017, 131, 208–221. [CrossRef] 71. Triguero, Á.; Cuerva, M.C.; Álvarez-Aledo, C. Environmental Innovation and Employment: Drivers and Synergies. Sustainability 2017, 9, 2057. [CrossRef] 72. Ghisetti, C.; Mancinelli, S.; Mazzanti, M.; Zoli, M. Financial barriers and environmental innovations: Evidence 73. from EU manufacturing ﬁrms. Clim. Policy 2017, 17, S131–S147. [CrossRef] Scarpellini, S.; Marín-Vinuesa, L.M.; Portillo-Tarragona, P.; Moneva, J.M. Deﬁning and measuring diﬀerent dimensions of ﬁnancial resources for business eco-innovation and the inﬂuence of the ﬁrms’ capabilities. J. Clean. Prod. 2018, 204, 258–269. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 2588 18 of 19 74. Przychodzen, J.; Przychodzen, W. Relationships between eco-innovation and financial performance—Evidence from publicly traded companies in Poland and Hungary. J. Clean. Prod. 2015, 90, 253–263. [CrossRef] 75. Carrillo-Hermosilla, J.; Del Río, P.; Könnölä, T. Diversity of eco-innovations: Reﬂections from selected case studies. J. Clean. Prod. 2010, 18, 1073–1083. [CrossRef] 76. Van Kranenburg, H.; Hagedoorn, J.; Pennings, J. Measurement of international and product diversiﬁcation in the publishing industry. J. Media Econ. 2004, 17, 87–104. [CrossRef] 77. Tether, B.S. Who co-operates for innovation, and why An empirical analysis. Res. Policy 2002, 31, 947–967. 78. [CrossRef] Fryer, P.J.; Versteeg, C. Processing technology innovation in the food industry. Innov. Manag. Policy Pract. 2008, 10, 74–90. [CrossRef] 79. Leitner, A.; Wehrmeyer, W.; France, C. The impact of regulation and policy on radical eco-innovation: The need for a new understanding. Manag. Res. Re View 2010, 33, 1022–1041. [CrossRef] 80. Ociepa-Kubicka, A.; Pachura, P. Eco-innovations in the functioning of companies. Environ. Res. 2017, 156, 81. 284–290. [CrossRef] Sezen, B.; Çankaya, S.Y. Eﬀects of green manufacturing and eco-innovation on sustainability performance. Procedia Soc. Behav. Sci. 2013, 99, 154–163. [CrossRef] 82. Cheng, C.C.; Shiu, E.C. Validation of a proposed instrument for measuring eco-innovation: An implementation perspective. Technovation 2012, 32, 329–344. [CrossRef] 83. Capitanio, F.; Coppola, A.; Pascucci, S. Product and process innovation in the Italian food industry. Agribusiness 2010, 26, 503–518. [CrossRef] 84. Woodside, A.G. Moving beyond multiple regression analysis to algorithms: Calling for adoption of a paradigm shift from symmetric to asymmetric thinking in data analysis and crafting theory. J. Bus. Res. 2013, 66, 463–472. [CrossRef] Samara, G.; Berbegal-Mirabent, J. Independent directors and family ﬁrm performance: Does one size ﬁt all? Int. Entrep. Manag. J. 2018, 14, 149–172. [CrossRef] 85. 86. Wu, P.L.; Yeh, S.S.; Huan, T.C.; Woodside, A.G. Applying complexity theory to deepen service dominant logic: Conﬁgural analysis of customer experience-and-outcome assessments of professional services for personal transformations. J. Bus. Res. 2014, 67, 1647–1670. [CrossRef] Fiss, P.C. A set-theoretic approach to organizational conﬁgurations. Acad. Manag. Rev. 2007, 32, 1190–1198. [CrossRef] 87. 88. Ragin, C.C. Redesigning Social Inquiry Fuzzy Sets and Beyond; University of Chicago Press: Chicago, IL, USA, 2008. 89. Vis, B. The comparative advantages of fsQCA and regression analysis for moderately large-N analyses. Sociol. Methods Res. 2012, 40, 168–198. [CrossRef] 90. Berbegal-Mirabent, J.; Ribeiro-Soriano, D.E.; Sánchez García, J.L. Can a magic recipe foster university spin-oﬀ creation? J. Bus. Res. 2015, 68, 2272–2278. [CrossRef] 91. Quine. The problem of simplifying truth functions. Am. Math. Mon. 1952, 59, 521–531. [CrossRef] 92. Roig-Tierno, N.; Gonzalez-Cruz, T.F.; Llopis-Martinez, J. An overview of qualitative comparative analysis: 93. A bibliometric analysis. J. Innov. Knowl. 2017, 2, 15–23. [CrossRef] Sanni, M. Drivers of eco-innovation in the manufacturing sector of Nigeria. Technol. Forecast. Soc. Chang. 2018, 131, 303–314. [CrossRef] 94. Ragin, C.C.; Fiss, P.C. Net eﬀects analysis versus conﬁgurational analysis: An empirical demostration. In Redesigning Social Inquiry: Fuzzy Sets and Beyond; University of Chicago Press: Chicago, IL, USA, 2008; pp. 190–2012. 95. Tumelero, C.; Sbragia, R.; Evans, S. Cooperation in R & D and eco-innovations: The role in companies’ socioeconomic performance. J. Clean. Prod. 2019, 207, 1138–1149. [CrossRef] 96. Da Silva Rabêlo, O.; De Azevedo Melo, A.S.S. Drivers of multidimensional eco-innovation: Empirical evidence from the Brazilian industry. Environ. Technol. (U. K.) 2019, 40. [CrossRef] [PubMed] 97. Dora, M.; Kumar, M.; Van Goubergen, D.; Molnar, A.; Gellynck, X. Food quality management system: Reviewing assessment strategies and a feasibility study for European food small and medium-sized enterprises. Food Control 2013, 31, 607–616. [CrossRef] 98. Rabadán, A.; González-Moreno, Á.; Sáez-Martínez, F.J. Improving Firms’ Performance and Sustainability: The Case of Eco-Innovation in the Agri-Food Industry. Sustainability 2019, 11, 5590. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 2588 19 of 19 99. Peiró-Signes, Á.; Miret-Pastor, L.; De-Miguel-Molina, B. Analysing the determinants of better performance through eco management tools at the food industry: An empirical study. In EcoProduction and Logistics; Springer: Berlin/Heidelberg, Germany, 2013. 100. Hasler, K.; Olfs, H.W.; Omta, O.; Bröring, S. Drivers for the Adoption of Diﬀerent Eco-Innovation Types in the Fertilizer Sector: A review. Sustainability 2017, 9, 2216. [CrossRef] 101. Costa, A.I.A.; Jongen, W.M.F. New insights into consumer-led food product development. Trends Food Sci. Technol. 2006, 17, 457–465. [CrossRef] 102. Anzola-Román, P.; Bayona-Sáez, C.; García-Marco, T. Organizational innovation, internal R&D and externally sourced innovation practices: Eﬀects on technological innovation outcomes. J. Bus. Res. 2018, 91, 233–247. [CrossRef] 103. Avermaete, T.; Viaene, J.; Morgan, E.J.; Crawford, N. Determinants of innovation in small food ﬁrms. Eur. J. Innov. Manag. 2003, 6, 8–17. [CrossRef] 104. Jurgilevich, A.; Birge, T.; Kentala-Lehtonen, J.; Korhonen-Kurki, K.; Pietikäinen, J.; Saikku, L.; Schösler, H. Transition Towards Circular Economy in the Food System. Sustainability 2016, 8, 69. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_ijms20184562	Article Sesquiterpenes Are Agonists of the Pregnane X Receptor but Do Not Induce the Expression of Phase I Drug-Metabolizing Enzymes in the Human Liver Michaela Šadibolová 1, Tomáš Zárybnický 1 Petra Matoušková 1 , Lenka Skálová 1 and Iva Boušová 1,* , Tomáš Smutný 2, Petr Pávek 2, Zdenˇek Šubrt 3,4, 1 Department of Biochemical Sciences, Faculty of Pharmacy in Hradec Králové, Charles University, 500 05 Hradec Králové, Czech Republic; [email protected] (M.S.); [email protected] (T.Z.); [email protected] (P.M.); [email protected] (L.S.) 2 Department of Pharmacology and Toxicology, Faculty of Pharmacy in Hradec Králové, Charles University, 500 05 Hradec Králové, Czech Republic; [email protected] (T.S.); [email protected] (P.P.) 3 Department of General Surgery, Third Faculty of Medicine and University Hospital Královské Vinohrady, Charles University, 100 34 Prague, Czech Republic; [email protected] 4 Department of Surgery, University Hospital Hradec Králové, 500 05 Hradec Králové, Czech Republic * Correspondence: [email protected]; Tel.: +420-495-067-406 Received: 13 August 2019; Accepted: 12 September 2019; Published: 14 September 2019 Abstract: Sesquiterpenes, the main components of plant essential oils, are bioactive compounds with numerous health-beneﬁcial activities. Sesquiterpenes can interact with concomitantly administered drugs due to the modulation of drug-metabolizing enzymes (DMEs). The aim of this study was to evaluate the modulatory eﬀects of six sesquiterpenes (farnesol, cis-nerolidol, trans-nerolidol, α-humulene, β-caryophyllene, and caryophyllene oxide) on the expression of four phase I DMEs (cytochrome P450 3A4 and 2C, carbonyl reductase 1, and aldo-keto reductase 1C) at both the mRNA and protein levels. For this purpose, human precision-cut liver slices (PCLS) prepared from 10 patients and transfected HepG2 cells were used. Western blotting, quantitative real-time PCR and reporter gene assays were employed in the analyses. In the reporter gene assays, all sesquiterpenes signiﬁcantly induced cytochrome P450 3A4 expression via pregnane X receptor interaction. However in PCLS, their eﬀects on the expression of all the tested DMEs at the mRNA and protein levels were mild or none. High inter-individual variabilities in the basal levels as well as in modulatory eﬃcacy of the tested sesquiterpenes were observed, indicating a high probability of marked diﬀerences in the eﬀects of these compounds among the general population. Nevertheless, it seems unlikely that the studied sesquiterpenes would remarkably inﬂuence the bioavailability and eﬃcacy of concomitantly administered drugs. Keywords: sesquiterpene; mRNA expression; protein expression; precision-cut liver slices; gene reporter assay; cytochrome P450 3A4; pregnane X receptor 1. Introduction In view of promoting healthier living as well as addressing various health issues, the tendency of the population to seek out and use herbal remedies and nutraceuticals has signiﬁcantly increased over the past years [1]. Moreover, the intake of natural remedies and herbal supplements concomitantly with prescription drugs has rapidly increased in terms of frequency. Although many of these herbal remedies have demonstrated a number of beneﬁcial and health-promoting activities, information concerning possible herb–drug interactions is limited. Concurrent intake may indeed lead to undesirable herb–drug interactions at both the pharmacokinetic and pharmacodynamic level, which might result in adverse Int. J. Mol. Sci. 2019, 20, 4562; doi:10.3390/ijms20184562 www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2019, 20, 4562 2 of 15 drug reactions with severe consequences [2,3]. Elderly people are particularly at risk, as more than other populations they suﬀer from various comorbidities and age-related pathophysiological changes as well as participate in polypharmaceutical treatment regimens that contribute to a reduction in drug clearance [4,5]. Recently, sesquiterpenes have drawn the attention of the research community owing to their considerable anti-inﬂammatory, antitumorigenic, antioxidant, and antiparasitic activities. Carrying these promising characteristics as the major components of plant essential oils, they are often present in various herbal medicines and nutraceuticals, especially widely used in the cosmetics and pharmaceutical industries [6]. Furthermore, as sesquiterpenes are inherent components of many spices, traditional delicacies, and beverages, they are often present in the human diet [7]. Their lipophilic character allows them to be easily absorbed following oral (or even topical) administration, with a single oral dose leading to measurable plasma concentrations [8–10]. We can infer from current evidence that sesquiterpenes are able to modulate the activity of some drug-metabolizing enzymes (DME) and could therefore interfere with the biotransformation of concurrently administered drugs despite the fact that information addressing this problem is scarce [6]. In a recent study, the cyclic sesquiterpenes α-humulene (HUM), β-caryophyllene (CAR), and caryophyllene oxide (CAO) were found to inhibit the activity of cytochrome P450 3A (CYP3A) in human and rat hepatic microsomes (with CAO being the strongest enzyme inhibitor), but did not aﬀect the activities of other CYPs, carbonyl-reducing enzymes, or even conjugation enzymes [7]. In a similar manner, the linear sesquiterpenes farnesol (FAR), cis-nerolidol (cNER), and trans-nerolidol (tNER) inhibited the activities of some CYP isoforms, namely, CYP1A, CYP2B, and CYP3A, also in both human and rat hepatic microsomal fractions, but aﬀected neither carbonyl-reducing enzymes nor conjugation enzymes [11]. Although sesquiterpenes inhibited CYPs activities in vitro, in vivo administration of CAO and tNER to mice increased CYPs expression and activity in the liver and small intestine [12]. To address these seemingly contradictory ﬁndings, the present study was designed to develop more detailed information on the possible modulatory activity of six commonly used sesquiterpenes (HUM, CAR, CAO, FAR, cNER, and tNER; structures shown in Figure 1) on the expression of the main CYPs and carbonyl reducing enzymes. Human hepatic HepG2 cell line transfected with the human pregnane X receptor (PXR) and the aryl hydrocarbon receptor (AhR) along with human precision-cut liver slices (PCLS) were used for this purpose. Human PCLS represent a miniature in vitro tissue model which comprises all cell types of the tissue in their natural environment. PCLS have been successfully used in drug clearance, metabolism, and toxicity studies due to their relatively stable expression of DMEs and drug transporters [13,14]. Furthermore, their employment in enzyme induction studies has been substantiated [13,15]. Unlike cell culture models, PCLS preserve complex cell–cell and cell–matrix interactions, and are therefore more relevant in terms of reﬂecting the multicellular characteristics of the liver in vivo [14]. In our study, PCLS from 10 patients were incubated with sesquiterpenes, following which the changes in enzyme expression of CYP3A4, CYP2C, aldo-keto reductase 1C (AKR1C), and carbonyl reductase 1 (CBR1) were determined at the mRNA and protein levels. In addition, the sesquiterpenes were tested for their ability to interact with the receptors PXR and AhR which regulate the expression of two important CYP families, CYP3A and CYP1, respectively. This study contributes to the risk and safety assessment of the concomitant administration of the studied sesquiterpenes with prescription drugs. Int. J. Mol. Sci. 2019, 20, 4562 3 of 15 Figure 1. Chemical structures of the studied sesquiterpenes. 2. Results and Discussion The liver samples employed in this study were selected to be as healthy as possible, even with the varying clinical conditions of the patients. Human PCLS were prepared from liver samples obtained from ten patients undergoing partial hepatectomy for a malignant disease. Except for patient L6, plasma levels of total bilirubin (5.0–20.2 µkat/L), alanine aminotransferase (0.29–0.55 µkat/L), aspartate aminotransferase (0.32–0.47 µkat/L), and alkaline phosphatase (0.61–1.80 µkat/L), all of which provide information about liver functions, were within the physiological range in all patients. In patient L6, total bilirubin level was twice increased (48.9 µmol/L) along with a 3.6-times elevated level of conjugated bilirubin (12.3 µmol/L), which suggests the presence of intrahepatic biliary obstruction caused by ongoing malignancy [16]. The level of γ-glutamyltransferase (0.35–1.97 µkat/L) exceeded the physiological range 1.1–2.9-times in six out of the ten patients. The elevation in γ-glutamyltransferase levels is often seen in patients with biliary tract diseases including malignancies [17] as well as in patients with colorectal carcinoma with liver metastases [18]. Scoring of all liver samples for steatosis (score of 0–1) and ﬁbrosis (score of zero) was performed by a pathologist. Based on the obtained scores, no or mild signs of liver disease were found in the patients’ biopsies. 2.1. The Eﬀect of Sesquiterpenes on AhR and PXR Activation Initially, all the sesquiterpenes were tested for the ability to activate the PXR and AhR nuclear receptors, which are known to be involved in the xenobiotic-induced increase of cytochrome P450 (CYP) 3A4 and CYP1A enzyme expression, respectively [19–21]. A luciferase reporter gene assay in HepG2 cells was employed for this purpose. The cells were treated with the sesquiterpenes in two concentrations (10 and 30 µM). Rifampicin (RIF, 10 µM) and methylcholanthrene (MC, 10 µM), well-known PXR and AhR ligands, respectively, were used as positive controls. Neither of the tested compounds showed the potential to interact with the AhR nuclear receptor (Figure 2A), whereas marked interaction of MC was observed. On the other hand, all the sesquiterpenes were able to activate the PXR signaling pathway. The treatment with a higher concentration (30 µM) resulted in more pronounced PXR activation by all the studied compounds. However, tNER, cNER, HUM, and CAR managed to activate PXR even at lower concentrations (10 µM). The PXR activators tNER and cNER appeared to be the most potent at both the lower (2.4- and 2.1-fold, respectively) and higher (5.5- and 4.3-fold, respectively) concentrations (Figure 2B). Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 3 of 15 Figure 1. Chemical structures of the studied sesquiterpenes. 2. Results and Discussion The liver samples employed in this study were selected to be as healthy as possible, even with the varying clinical conditions of the patients. Human PCLS were prepared from liver samples obtained from ten patients undergoing partial hepatectomy for a malignant disease. Except for patient L6, plasma levels of total bilirubin (5.0–20.2 µkat/L), alanine aminotransferase (0.29–0.55 µkat/L), aspartate aminotransferase (0.32–0.47 µkat/L), and alkaline phosphatase (0.61–1.80 µkat/L), all of which provide information about liver functions, were within the physiological range in all patients. In patient L6, total bilirubin level was twice increased (48.9 µmol/L) along with a 3.6-times elevated level of conjugated bilirubin (12.3 µmol/L), which suggests the presence of intrahepatic biliary obstruction caused by ongoing malignancy [16]. The level of γ-glutamyltransferase (0.35–1.97 µkat/L) exceeded the physiological range 1.1–2.9-times in six out of the ten patients. The elevation in γ-glutamyltransferase levels is often seen in patients with biliary tract diseases including malignancies [17] as well as in patients with colorectal carcinoma with liver metastases [18]. Scoring of all liver samples for steatosis (score of 0–1) and fibrosis (score of zero) was performed by a pathologist. Based on the obtained scores, no or mild signs of liver disease were found in the patients’ biopsies. 2.1. The Effect of Sesquiterpenes on AhR and PXR Activation Initially, all the sesquiterpenes were tested for the ability to activate the PXR and AhR nuclear receptors, which are known to be involved in the xenobiotic-induced increase of cytochrome P450 (CYP) 3A4 and CYP1A enzyme expression, respectively [19–21]. A luciferase reporter gene assay in HepG2 cells was employed for this purpose. The cells were treated with the sesquiterpenes in two concentrations (10 and 30 µM). Rifampicin (RIF, 10 µM) and methylcholanthrene (MC, 10 µM), well-known PXR and AhR ligands, respectively, were used as positive controls. Neither of the tested compounds showed the potential to interact with the AhR nuclear receptor (Figure 2A), whereas marked interaction of MC was observed. On the other hand, all the sesquiterpenes were able to activate the PXR signaling pathway. The treatment with a higher concentration (30 µM) resulted in more pronounced PXR activation by all the studied compounds. However, tNER, cNER, HUM, and CAR managed to activate PXR even at lower concentrations (10 µM). The PXR activators tNER and cNER appeared to be the most potent at both the lower (2.4- and 2.1-fold, respectively) and higher (5.5- and 4.3-fold, respectively) concentrations (Figure 2B). Int. J. Mol. Sci. 2019, 20, 4562 4 of 15 Figure 2. The eﬀect of sesquiterpenes on the AhR and PXR receptors. HepG2 cells were transiently transfected with either p1A1-luc (A) or p3A4-luc in combination with expression vectors pSG5-PXR and pSG5-RXRα (B). The next day, the cells were treated with the tested compounds for 24 h. AhR and PXR as well as the well-known ligands methylcholanthrene (MC, 10 µM) and rifampicin (RIF, 10 µM) were used as positive controls. The samples were subsequently assayed by a Dual-Luciferase Reporter Assay System (Promega). The results are presented as the relative change to DMSO-treated controls deﬁned as 100% (n = 3). * p < 0.05. In our experiments, all the studied sesquiterpenes caused mild to intermediate activation of human PXR, reaching 2.6- to 5.5-fold at the 30 µM concentration. In another study, the sesquiterpenes zederone and germacrone caused the signiﬁcant and dose-dependent activation of mouse PXR, while their eﬀect on the activation of human PXR was weaker and comparable to our obtained results. In accordance with our ﬁndings, zederone and germacrone did not activate human AhR at 1–30 µM concentrations [22]. It has been reported that the antimalarial drug artemisinin and its derivatives were also able to moderately activate human PXR [23,24]. 2.2. Basal mRNA and Protein Expression of CYP3A4, CYP2C, CBR1, and AKR1C3 in PCLS Based on the results of the gene reporter assay, the eﬀect of sesquiterpenes on the gene and protein expression of selected phase I DMEs have been studied in human PCLS. As none of the selected sesquiterpenes signiﬁcantly activated the AhR-responsive luciferase construct, their eﬀect on the expression of CYP1A1/2 has not been tested. Four major phase I DMEs, namely CYP3A4, CYP2C, CBR1, and AKR1C3, were selected. Both CYP3A4 and CYP2C, the most abundant CYPS in the human liver and the main DME involved in oxidative biotransformation of drugs, are downstream targets of PXR/CAR nuclear receptors [25]. The transcription regulation of CBR1 and AKR1C, the main DME for drugs bearing the carbonyl group, proceeds mainly by the nuclear factor erythroid 2-related factor 2 (Nrf2) system via the antioxidant-response element (ARE), which is present in their gene promotor [26,27]. As several sesquiterpenes and sesquiterpene lactones have been reported to activate the Nrf2-ARE-dependent detoxiﬁcation pathway [28,29], CBR1 and AKR1C expression was tested in the present study. In the control PCLS, basal expressions of four selected DMEs at the mRNA and protein level were measured. Concerning mRNA expression, CYP2C was the DME with the highest variability, while CBR1 was the most stably expressed gene (Figure 3A). The mRNA levels of CYP2C and CBR1 among samples with the lowest and the highest expression diﬀered 92.2-times and 2.9-times, respectively. With regards to protein expression, the situation was reversed and CBR1 exerted the highest variability among the studied enzymes, while CYP2C was stably expressed in all liver samples (Figure 3B). Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 4 of 15 Figure 2. The effect of sesquiterpenes on the AhR and PXR receptors. HepG2 cells were transiently transfected with either p1A1-luc (A) or p3A4-luc in combination with expression vectors pSG5-PXR and pSG5-RXRα (B). The next day, the cells were treated with the tested compounds for 24 h. AhR and PXR as well as the well-known ligands methylcholanthrene (MC, 10 µM) and rifampicin (RIF, 10 µM) were used as positive controls. The samples were subsequently assayed by a Dual-Luciferase Reporter Assay System (Promega). The results are presented as the relative change to DMSO-treated controls defined as 100% (n = 3). * p < 0.05. In our experiments, all the studied sesquiterpenes caused mild to intermediate activation of human PXR, reaching 2.6- to 5.5-fold at the 30 µM concentration. In another study, the sesquiterpenes zederone and germacrone caused the significant and dose-dependent activation of mouse PXR, while their effect on the activation of human PXR was weaker and comparable to our obtained results. In accordance with our findings, zederone and germacrone did not activate human AhR at 1–30 µM concentrations [22]. It has been reported that the antimalarial drug artemisinin and its derivatives were also able to moderately activate human PXR [23,24]. 2.2. Basal mRNA and Protein Expression of CYP3A4, CYP2C, CBR1, and AKR1C3 in PCLS Based on the results of the gene reporter assay, the effect of sesquiterpenes on the gene and protein expression of selected phase I DMEs have been studied in human PCLS. As none of the selected sesquiterpenes significantly activated the AhR-responsive luciferase construct, their effect on the expression of CYP1A1/2 has not been tested. Four major phase I DMEs, namely CYP3A4, CYP2C, CBR1, and AKR1C3, were selected. Both CYP3A4 and CYP2C, the most abundant CYPS in the human liver and the main DME involved in oxidative biotransformation of drugs, are downstream targets of PXR/CAR nuclear receptors [25]. The transcription regulation of CBR1 and AKR1C, the main DME for drugs bearing the carbonyl group, proceeds mainly by the nuclear factor erythroid 2-related factor 2 (Nrf2) system via the antioxidant-response element (ARE), which is present in their gene promotor [26,27]. As several sesquiterpenes and sesquiterpene lactones have been reported to activate the Nrf2-ARE-dependent detoxification pathway [28,29], CBR1 and AKR1C expression was tested in the present study. In the control PCLS, basal expressions of four selected DMEs at the mRNA and protein level were measured. Concerning mRNA expression, CYP2C was the DME with the highest variability, while CBR1 was the most stably expressed gene (Figure 3A). The mRNA levels of CYP2C and CBR1 among samples with the lowest and the highest expression differed 92.2-times and 2.9-times, respectively. With regards to protein expression, the situation was reversed and CBR1 exerted the highest variability among the studied enzymes, while CYP2C was stably expressed in all liver samples (Figure 3B). Int. J. Mol. Sci. 2019, 20, 4562 5 of 15 Figure 3. Inter-individual variability in the basal expression of selected mRNAs (A) and proteins (B) in PCLS from ten patients. The horizontal line represents the median, and whiskers represent the maximum and minimum values. In our results, marked inter-individual diﬀerences in the basal expression of all the selected DMEs among the individual liver samples were observed. However, a good correlation between mRNA levels of CYP3A4 and AKR1C (r = 0.688, p = 0.0278), the protein levels of CYP3A4 and AKR1C3 (r = 0.699, p = 0.0244), and the protein levels of CBR1 and AKR1C3 (r = 0.691, p = 0.0248) were all found in human PCLS (untreated controls). A meta-analysis of 50 studies dealing with the abundance of human hepatic cytochrome P450 enzymes in Caucasian adult livers showed a strong positive correlation between the expression levels of CYP3A4 and CYP2C8/9 [30]. As was reported previously, the PCLS represent individuals exhibiting large variations in basal mRNA levels as well as in responsiveness to potential inducers [15,31,32]. 2.3. The Eﬀect of Sesquiterpenes on the mRNA Expression of the Studied Enzymes As sesquiterpenes are important components of popular nutraceuticals and dietary supplements, their ability to modulate the activity and/or expression of DMEs and drug transporters becomes an important question. Recently, the inhibitory eﬀect of linear (cNER, tNER, and FAR) and cyclic (HUM, CAR, and CAO) sesquiterpenes on the activity of the CYP3A subfamily in human and rat hepatic subcellular fractions was observed, while the activities of carbonyl-reducing and conjugating enzymes were not signiﬁcantly inﬂuenced [7,11]. In human liver microsomes, other sesquiterpenes, zederone and germacrone, moderately inhibited CYP2B6 and CYP3A4 activities, with IC50 values below 10 µM [22]. The sesquiterpene lactone alantolactone acted as non-competitive inhibitor of CYP3A4 in human liver microsomes, with an IC50 equal to 3.6 µM [33]. On the other hand, a marked increase in CYP2B and CYP3A activity as well as in mRNA levels was observed after 24 h in the liver and small intestine of mice orally treated with tNER and CAO (50 mg/kg) [12]. In the present study, the eﬀect of FAR, tNER, cNER, HUM, CAR, and CAO on the mRNA expression of CYP3A4, CYP2C, CBR1, and AKR1C was studied in human PCLS prepared from 10 liver samples. As the number of PCLS prepared from one tissue sample was insuﬃcient for testing all six sesquiterpenes, ﬁve samples were used for the cyclic sesquiterpenes and ﬁve samples for the linear ones. RIF was used in all PCLS as a positive control. The PCLS were incubated in the presence of individual sesquiterpenes, DMSO (control), and RIF (positive control) for 24 h. PCR primers for CYP2C and AKR1C were designed to amplify all four human CYP2C isoforms (namely 2C8, 2C9, 2C18, and 2C19) and all four AKR1C isoforms (namely AKR1C1–4), respectively. The linear sesquiterpenes FAR, cNER, and tNER showed some eﬀects on the mRNA expression of DMEs. Results are presented in Figure 4. In patient L7, FAR and tNER caused a signiﬁcant decrease in the mRNA level of all four studied enzymes. This inhibitory eﬀect was most pronounced in the case of CYP3A4, in which FAR and tNER reduced the level of mRNA by 76.3% and 60.8%, respectively. In this patient, basal expression of CYP3A4, CYP2C, and AKR1C ranked among the highest expression levels. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 5 of 15 Figure 3. Inter-individual variability in the basal expression of selected mRNAs (A) and proteins (B) in PCLS from ten patients. The horizontal line represents the median, and whiskers represent the maximum and minimum values. In our results, marked inter-individual differences in the basal expression of all the selected DMEs among the individual liver samples were observed. However, a good correlation between mRNA levels of CYP3A4 and AKR1C (r = 0.688, p = 0.0278), the protein levels of CYP3A4 and AKR1C3 (r = 0.699, p = 0.0244), and the protein levels of CBR1 and AKR1C3 (r = 0.691, p = 0.0248) were all found in human PCLS (untreated controls). A meta-analysis of 50 studies dealing with the abundance of human hepatic cytochrome P450 enzymes in Caucasian adult livers showed a strong positive correlation between the expression levels of CYP3A4 and CYP2C8/9 [30]. As was reported previously, the PCLS represent individuals exhibiting large variations in basal mRNA levels as well as in responsiveness to potential inducers [15,31,32]. 2.3. The Effect of Sesquiterpenes on the mRNA Expression of the Studied Enzymes As sesquiterpenes are important components of popular nutraceuticals and dietary supplements, their ability to modulate the activity and/or expression of DMEs and drug transporters becomes an important question. Recently, the inhibitory effect of linear (cNER, tNER, and FAR) and cyclic (HUM, CAR, and CAO) sesquiterpenes on the activity of the CYP3A subfamily in human and rat hepatic subcellular fractions was observed, while the activities of carbonyl-reducing and conjugating enzymes were not significantly influenced [7,11]. In human liver microsomes, other sesquiterpenes, zederone and germacrone, moderately inhibited CYP2B6 and CYP3A4 activities, with IC50 values below 10 µM [22]. The sesquiterpene lactone alantolactone acted as non-competitive inhibitor of CYP3A4 in human liver microsomes, with an IC50 equal to 3.6 µM [33]. On the other hand, a marked increase in CYP2B and CYP3A activity as well as in mRNA levels was observed after 24 h in the liver and small intestine of mice orally treated with tNER and CAO (50 mg/kg) [12]. In the present study, the effect of FAR, tNER, cNER, HUM, CAR, and CAO on the mRNA expression of CYP3A4, CYP2C, CBR1, and AKR1C was studied in human PCLS prepared from 10 liver samples. As the number of PCLS prepared from one tissue sample was insufficient for testing all six sesquiterpenes, five samples were used for the cyclic sesquiterpenes and five samples for the linear ones. RIF was used in all PCLS as a positive control. The PCLS were incubated in the presence of individual sesquiterpenes, DMSO (control), and RIF (positive control) for 24 h. PCR primers for CYP2C and AKR1C were designed to amplify all four human CYP2C isoforms (namely 2C8, 2C9, 2C18, and 2C19) and all four AKR1C isoforms (namely AKR1C1–4), respectively. The linear sesquiterpenes FAR, cNER, and tNER showed some effects on the mRNA expression of DMEs. Results are presented in Figure 4. In patient L7, FAR and tNER caused a significant decrease in the mRNA level of all four studied enzymes. This inhibitory effect was most pronounced in the case of CYP3A4, in which FAR and tNER reduced the level of mRNA by 76.3% and 60.8%, respectively. In this patient, basal expression of CYP3A4, CYP2C, and AKR1C ranked among the Int. J. Mol. Sci. 2019, 20, 4562 6 of 15 In patient L9, tNER induced AKR1C expression 1.4-times. Taken together, FAR signiﬁcantly inﬂuenced the expression of CYP3A4 and CBR1 in one patient and the expression of CYP2C and AKR1C in two patients; cNER reduced only CBR1 expression in one patient; while tNER inhibited the mRNA level of CYP3A4, CYP2C, CBR1, and AKR1C in one patient and induced AKR1C expression in another. In contrast to the linear sesquiterpenes, the cyclic sesquiterpenes HUM, CAR, and CAO showed no signiﬁcant eﬀect on the mRNA expression of all the studied DMEs (Figure S1). The applicability of the chosen model system was proved by 10 µM RIF (positive control, a prototypical ligand of the human PXR), which caused signiﬁcant induction on CYP3A4 expression in PCLS from all patients. Figure 4. Inter-individual diﬀerences in the eﬀect of linear sesquiterpenes (10 µM) and RIF (10 µM) on the normalized mRNA expression of CYP3A4 (A), CYP2C (B), CBR1 (C) and AKR1C (D) in human PCLS from ﬁve patients after 24 h (n = 3). The normalized expression level was calculated using −∆∆Ct method with the geometric mean of GAPDH and SDHA as a reference gene. Results are the 2 presented as the mean ± SD (n = 3). Statistical analyses were performed using one-way ANOVA with Dunnett’s test: p < 0.05 (). In human PCLS, mRNA expression of the studied DMEs was only mildly inﬂuenced by the tested sesquiterpenes. Their eﬀect is noticeably lower than might be expected based on the results of the gene reporter assay, in which 10 µM cNER induced the mRNA level of CYP3A4 2.4-times. This discrepancy can be explained by the diﬀerent nature of model systems, i.e., HepG2 cells and PCLS. In human PCLS, the AhR-, PXR- and CAR-mediated induction of major CYP mRNAs can be detected, although the extent of the induction is often lower than in the primary hepatocytes [15,31,32]. Moreover, inter-individual variability in responses is often seen when using PCLS as a model system [34]. However, PCLS represent a miniature model of liver tissue with preserved cell–cell and cell–matrix interactions and all cell types are present, therefore, PCLS are more relevant to a physiological state than cells in a cell culture. On the other hand, HepG2 cells possess very low basal CYPs enzymatic activity. Therefore, Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 6 of 15 highest expression levels. In patient L9, tNER induced AKR1C expression 1.4-times. Taken together, FAR significantly influenced the expression of CYP3A4 and CBR1 in one patient and the expression of CYP2C and AKR1C in two patients; cNER reduced only CBR1 expression in one patient; while tNER inhibited the mRNA level of CYP3A4, CYP2C, CBR1, and AKR1C in one patient and induced AKR1C expression in another. In contrast to the linear sesquiterpenes, the cyclic sesquiterpenes HUM, CAR, and CAO showed no significant effect on the mRNA expression of all the studied DMEs (Figure S1). The applicability of the chosen model system was proved by 10 µM RIF (positive control, a prototypical ligand of the human PXR), which caused significant induction on CYP3A4 expression in PCLS from all patients. Figure 4. Inter-individual differences in the effect of linear sesquiterpenes (10 µM) and RIF (10 µM) on the normalized mRNA expression of CYP3A4 (A), CYP2C (B), CBR1 (C) and AKR1C (D) in human PCLS from five patients after 24 h (n = 3). The normalized expression level was calculated using the 2−ΔΔCt method with the geometric mean of GAPDH and SDHA as a reference gene. Results are presented as the mean ± SD (n = 3). Statistical analyses were performed using one-way ANOVA with Dunnett’s test: p < 0.05 (). In human PCLS, mRNA expression of the studied DMEs was only mildly influenced by the tested sesquiterpenes. Their effect is noticeably lower than might be expected based on the results of the gene reporter assay, in which 10 µM cNER induced the mRNA level of CYP3A4 2.4-times. This discrepancy can be explained by the different nature of model systems, i.e., HepG2 cells and PCLS. In human PCLS, the AhR-, PXR- and CAR-mediated induction of major CYP mRNAs can be detected, although the extent of the induction is often lower than in the primary hepatocytes [15,31,32]. Moreover, inter-individual variability in responses is often seen when using PCLS as a model system [34]. However, PCLS represent a miniature model of liver tissue with preserved cell–cell and cell–matrix interactions and all cell types are present, therefore, PCLS are more relevant to a physiological state than cells in a cell culture. On the other hand, HepG2 cells possess very low basal CYPs Int. J. Mol. Sci. 2019, 20, 4562 7 of 15 this discrepancy may be explained by the fast degradation of sesquiterpenes in liver slices, but not in the HepG2 cells. Various factors can contribute to inter-individual variability in the response to administered drugs/compounds. One of them is a level of constitutive expression of individual DMEs and nuclear receptors, which is aﬀected by sex, genetic polymorphism, food, environmental factors, medication, In addition, the induction eﬀect could be inﬂuenced by the level of pathological conditions etc. sesquiterpene, which depends on the rate of its metabolism. For example, three metabolites of farnesol (i.e., hydroxyfarnesol, farnesyl glucuronide and hydroxyfarnesyl glucuronide) have been identiﬁed in human liver microsomes [35]. If such metabolites of sesquiterpenes are less active in DME induction than parent compounds, higher eﬀect of sesquiterpenes in some PCLS could be attributed to lower activity of CYP and/or UGT in those individuals. In some PCLS, inhibitory eﬀect of FAR and tNER was observed. One of the mechanisms, which could explain sesquiterpenes-mediated inhibition of DMEs’ gene expression, could be based on their possible involvement in the epigenetic regulation of those genes. As was reported earlier, sesquiterpene lactone parthenolide inﬂuenced the level of DNA methylation by decreasing expression and activity of human DNA methyltransferase 1 in several human cell lines and also aﬀected chromatin remodeling by downregulation of histone deacetylase 1 level via a proteasome-dependent degradation [36,37]. In addition, expression of DMEs can be regulated by miRNAs either directly or indirectly by targeting DME regulators (e.g., nuclear receptors) [38]. Regulation of miRNA expression by several sesquiterpenes have been described [39,40], however, no report describing sesquiterpene-miRNA-PXR interaction was found. 2.4. The Eﬀect of Sesquiterpenes on the Protein Expression of Studied Enzymes Subsequently, the inﬂuence of the studied sesquiterpenes on the protein expression of DMEs was studied in human PCLS. Homogenates from individual PCLS, which were treated in the same way as was the case in the mRNA expression study, were prepared and pooled. Calnexin, protein present in the endoplasmic reticulum was used as a loading control. The primary antibodies against CYP3A4, i.e., CBR1, AKR1C3, and CYP2C8 + 2C9 + 2C19 + 2C12, were used to detect protein expression of the corresponding DMEs. The studied sesquiterpenes possessed mild or no inhibitory inﬂuence on the protein expression of DMEs in individual patients. The highest inhibition was observed in the case of sample L11, in which tNER reduced the protein expression of CBR1 by 53.5%. On the other hand, this sesquiterpene increased the protein expression of CYP2C and AKR1C3 1.36-times and 1.45-times, respectively, in sample L7 (Figure S2). The eﬀects of individual sesquiterpenes diﬀered among individual liver samples and DMEs, e.g., CAR elevated CBR1 expression in L38, while it decreased CBR1 level in sample L6. Results are presented in Figure 5. Figure 5. Cont. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 8 of 15 Figure 5. Inter-individual differences in the effect of sesquiterpenes (10 µM) and RIF (10 µM) on the normalized protein expression of CYP3A4 (A,B) and CBR1 (C,D) in human PCLS from ten patients after 24 h (n = 3). The protein expression was calculated using calnexin as a loading control. Results are presented as the mean ± SD (n = 4), with controls set to 100%. Statistical analyses were performed using one-way ANOVA with Dunnett’s test: p < 0.05 (). The observed effect of the studied sesquiterpenes on the protein expression of the four DMEs was only weak, but certain inter-individual variability was found. However, the changes in the protein expression of DMEs are probably not biologically relevant. Knowledge regarding the sesquiterpenes‘ effects on the DMEs protein expression is scarce, as the effects have been studied only sporadically. For example, sesquiterpene lactone deoxyelephantopin (10 µM) showed no effect on the CYP3A4 protein expression, while the enzymatic activity of CYP3A4 was reduced by 45% in HepG2 cells [41]. 3. Materials and Methods 3.1. Chemicals and Reagents Sesquiterpenes α-humulene, β-caryophyllene (CAR), caryophyllene oxide (CAO), farnesol (FAR), cis-nerolidol (cNER) and trans-nerolidol (tNER), rifampicin (RIF), methylcholanthrene (MC) and fetal bovine serum (FBS) were purchased from Sigma Aldrich (Prague, Czech Republic). All other chemicals were of analytical grade or higher. Stock solutions of sesquiterpenes (10 mM) were prepared in dimethyl sulfoxide (DMSO) and stored at 4 °C in the dark. 3.2. Cell Culture Human hepatoblastoma-derived (HepG2) cells were purchased from the European Collection of Authenticated Cell Cultures (ECACC, Salisbury, UK) and maintained in antibiotic-free Dulbecco’s Int. J. Mol. Sci. 2019, 20, 4562 8 of 15 Figure 5. Inter-individual diﬀerences in the eﬀect of sesquiterpenes (10 µM) and RIF (10 µM) on the normalized protein expression of CYP3A4 (A,B) and CBR1 (C,D) in human PCLS from ten patients after 24 h (n = 3). The protein expression was calculated using calnexin as a loading control. Results are presented as the mean ± SD (n = 4), with controls set to 100%. Statistical analyses were performed using one-way ANOVA with Dunnett’s test: p < 0.05 (). The observed eﬀect of the studied sesquiterpenes on the protein expression of the four DMEs was only weak, but certain inter-individual variability was found. However, the changes in the protein expression of DMEs are probably not biologically relevant. Knowledge regarding the sesquiterpenes‘ eﬀects on the DMEs protein expression is scarce, as the eﬀects have been studied only sporadically. For example, sesquiterpene lactone deoxyelephantopin (10 µM) showed no eﬀect on the CYP3A4 protein expression, while the enzymatic activity of CYP3A4 was reduced by 45% in HepG2 cells [41]. 3. Materials and Methods 3.1. Chemicals and Reagents Sesquiterpenes α-humulene, β-caryophyllene (CAR), caryophyllene oxide (CAO), farnesol (FAR), cis-nerolidol (cNER) and trans-nerolidol (tNER), rifampicin (RIF), methylcholanthrene (MC) and fetal bovine serum (FBS) were purchased from Sigma Aldrich (Prague, Czech Republic). All other chemicals were of analytical grade or higher. Stock solutions of sesquiterpenes (10 mM) were prepared in dimethyl sulfoxide (DMSO) and stored at 4 C in the dark. ◦ 3.2. Cell Culture Human hepatoblastoma-derived (HepG2) cells were purchased from the European Collection of Authenticated Cell Cultures (ECACC, Salisbury, UK) and maintained in antibiotic-free Dulbecco’s modiﬁed Eagle’s medium (Thermo Fisher Scientiﬁc, Waltham, MA, USA) supplemented with 10% FBS at 37 C in a humidiﬁed incubator under 5% CO2. ◦ 3.3. Plasmids The expression plasmid encoding human PXR receptor (pSG5-PXR) was a generous gift from Dr. S. Kliewer (University of Texas, Dallas, TX, USA) and the pSG5-RXRα construct was kindly provided by Dr. C. Carlberg (University of Kuopio, Kuopio, Finland), while pRL-TK was obtained from Promega (Madison, WI, USA). The p3A4-luc reporter vector carries a distal XREM (–7836/–7208) (cid:48) and a basal promoter sequence (prPXRE, –362/+53) of the CYP3A4 gene 5 -ﬂanking region inserted to pGL3-Basic reporter vector [42]. The p1A1-luc plasmid bears the promoter region (–1566 to +73) of human CYP1A1 gene [43]. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 8 of 15 Figure 5. Inter-individual differences in the effect of sesquiterpenes (10 µM) and RIF (10 µM) on the normalized protein expression of CYP3A4 (A,B) and CBR1 (C,D) in human PCLS from ten patients after 24 h (n = 3). The protein expression was calculated using calnexin as a loading control. Results are presented as the mean ± SD (n = 4), with controls set to 100%. Statistical analyses were performed using one-way ANOVA with Dunnett’s test: p < 0.05 (*). The observed effect of the studied sesquiterpenes on the protein expression of the four DMEs was only weak, but certain inter-individual variability was found. However, the changes in the protein expression of DMEs are probably not biologically relevant. Knowledge regarding the sesquiterpenes‘ effects on the DMEs protein expression is scarce, as the effects have been studied only sporadically. For example, sesquiterpene lactone deoxyelephantopin (10 µM) showed no effect on the CYP3A4 protein expression, while the enzymatic activity of CYP3A4 was reduced by 45% in HepG2 cells [41]. 3. Materials and Methods 3.1. Chemicals and Reagents Sesquiterpenes α-humulene, β-caryophyllene (CAR), caryophyllene oxide (CAO), farnesol (FAR), cis-nerolidol (cNER) and trans-nerolidol (tNER), rifampicin (RIF), methylcholanthrene (MC) and fetal bovine serum (FBS) were purchased from Sigma Aldrich (Prague, Czech Republic). All other chemicals were of analytical grade or higher. Stock solutions of sesquiterpenes (10 mM) were prepared in dimethyl sulfoxide (DMSO) and stored at 4 °C in the dark. 3.2. Cell Culture Human hepatoblastoma-derived (HepG2) cells were purchased from the European Collection of Authenticated Cell Cultures (ECACC, Salisbury, UK) and maintained in antibiotic-free Dulbecco’s Int. J. Mol. Sci. 2019, 20, 4562 9 of 15 3.4. Luciferase Reporter Gene Assays The HepG2 cells were seeded into 48-well plates (30 000 cells/well) overnight and transfected either with gene reporter vector p1A1-luc (150 ng/well) or p3A4-luc (150 ng/well) in combination with expression vectors pSG5-PXR (100 ng/well) and pSG5-RXRα (50 ng/well), and co-transfected with pRL-TK (30 ng/well) for transfection normalization. The transfection was performed by Lipofectamine 3000 Reagent (Thermo Fisher Scientiﬁc, Waltham, MA, USA) following the manufacturer’s recommendations. After 24 h, the HepG2 cells were treated with the tested compounds at the indicated concentrations for an additional 24 h. The compounds were diluted in Opti-MEM I Reduced Serum Medium (Thermo Fisher Scientiﬁc, Waltham, MA, USA) supplemented with 5% FBS. Final concentration of the vehicle (DMSO) in media did not exceed 0.1% in all the experiments. After treatment, the cells were lysed and measured for both ﬁreﬂy and Renilla luciferase activities using a Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA). 3.5. Ethics Committee Statement All the experimental procedures were approved by the Ethics Committee of the University Hospital Hradec Králové, Czech Republic (Permission No. 201703 S14P, 2 March 2017). An informed consent for tissue procurement for research purposes was obtained from all subjects. 3.6. Human Liver Tissue Human liver tissue was provided by the University Hospital Hradec Králové as healthy surplus tissue from 10 patients (5 males and 5 females, 45–81 years old) undergoing partial hepatectomy due to the presence of a tumor. Table 1 summarizes a brief medical history of the liver tissue donors. The resected liver tissue was placed directly into a chilled vessel with Euro–Collins solution and transported to the laboratory for immediate handling. The liver tissue was regarded as healthy based on the results of biochemical tests and a histopathological examination. Routine biochemical tests (i.e., plasma levels of bilirubin, alanine aminotransferase, aspartate aminotransferase, γ-glutamyltransferase and alkaline phosphatase) were performed before the surgery. Histopathological examination of the liver tissue for signs of ﬁbrosis and/or steatosis was performed by a pathologist. Table 1. Brief medical history of liver tissue donors. Sex (Age) Reason of Surgery Comorbidities Long-Term Pharmacotherapy Liver Sample L5 L6 L7 L9 L11 L14 L16 L19 male (63) Colorectal carcinoma male (69) Colorectal carcinoma male (69) Colorectal carcinoma male (81) female (57) female (45) female (59) Colorectal carcinoma Colorectal carcinoma Benign focal nodular hyperplasia Colorectal carcinoma female (65) Colorectal carcinoma L36 female (78) L38 male (59) Cholangiocellular carcinoma Cholangiocellular carcinoma HTN, arterial hypertension; DM, diabetes mellitus; HLD, hyperlipidemia. HTN, hyperuricemia, type 2 DM HTN HTN, s/p CVA HTN, dyslipidemia none none HLD, ovarian cancer HTN, HLD, impaired glucose tolerance HTN, HLD, coronary artery disease, atrial ﬁbrillation Ramipril, atorvastatin, metformin, allopurinol Hydrochlorothiazide Acetylsalicylic acid, nitrendipine Betaxolol none none none Amlodipine Bisoprolol, furosemide, ramipril, simvastatin, enoxaparin, zolpidem none none s/p CVA, status post cerebrovascular accident; Int. J. Mol. Sci. 2019, 20, 4562 10 of 15 3.7. Preparation of Precision-Cut Liver Slices and Experimental Treatment The liver slices were prepared as described previously [44]. Brieﬂy, small cylindrical cores were cut out of the liver tissue and sliced using the Krumdieck tissue slicer MD4000 (Alabama Research and Development, Munford, AL, USA) ﬁlled with an ice-cold Krebs–Henseleit buﬀer saturated with carbogen and containing 25 mM d-glucose, 25 mM NaHCO3, and 10 mM HEPES (Carl Roth, Karlsruhe, Germany). The liver slices (8 mM in diameter, 150–170 µm in thickness) were preincubated individually in 1 mL of Williams’ Medium E (with L-glutamine, Invitrogen, Paisley, UK) supplemented with 25 mM d-glucose and 50 µg/mL gentamycin in 12-well plates under continuous supply of 85% O2 and 5% CO2 with continuous shaking (90 times/min) at 37 C for 60 min. Afterwards, the liver slices were transferred to new 12-well plates and incubated individually in 1.3 mL of fresh Williams‘ Medium E supplemented by either the tested compounds or DMSO (control) for 24 h. The ﬁnal DMSO concentration did not exceed 0.2%. Due to the limited number of liver slices that could be prepared from one tissue sample, three linear (FAR, tNER, and cNER) and three cyclic (HUM, CAR, CAO) sesquiterpenes were studied separately. All the experiments were performed in triplicates using the liver tissue from ﬁve diﬀerent patients. ◦ 3.8. RNA Isolation, cDNA Synthesis and Quantitative Real-Time PCR (RT-qPCR) ◦ ◦ ◦ ◦ ◦ The liver slices were collected after 24 h of incubation. All treatments were performed in triplicates and every slice was placed separately into 500 µL of TriReagent and stored at −80 C until use. Total RNA from every slice was isolated using TriReagent according to the manufacturer’s instructions (Biotech, Praha, Czech Republic). The homogenization of each sample was performed using a single steel bead in a 2 mL Eppendorf tube using a microhomogenizer. The puriﬁed RNA was dissolved in 40 µL of diethyl pyrocarbonate (DEPC)-treated water (0.01% DEPC in HPLC water, autoclaved) and stored at −80 C. The measurement of the absorbance at 260 and 280 nm using the NanoDrop ND-1000 UV–vis Spectrophotometer (Thermo Fisher Scientiﬁc, Pardubice, Czech Republic) was used to determine RNA yields and purity. Subsequently, RNA (4 µg) was treated with 2 U of DNase I (New England Biolabs, Ipswich, MA, USA) in a ﬁnal volume of 30 µL for 20 min at 37 C, 1.5 µL of 0.1 M EDTA was added and the DNAse was inactivated by heat (10 min at 75 C). The solution was diluted to a concentration of 0.2 µg/µL by adding 8.5 µL of DEPC water. The DNAse I treated RNA was stored at −80 C until further analyses. The ﬁrst strand cDNA was synthesized from 1 µg of total RNA and 1 µL of 50 µM random hexamers (Generi Biotech, Hradec Kralove, Czech Republic) using ProtoScript II reverse transcriptase (New England Biolabs, Ipswich, MA, USA). After initial heat denaturation of total C for 5 min), 4 µL 5× ProtoScript II RT Reaction Buﬀer, 2 µL 10× DTT, 2 µL dNTP Mix 5 mM, RNA (65 3.5 µL H2O and 0.5 µL ProtoScript II 200 U/µL were added and mixed by pipetting. The reactions (20 µL) were incubated for 10 min at 25 C. The obtained C, for 50 min at 42 cDNAs were diluted 1:6 by DEPC water. The qPCR analyses were carried out using QuantStudio 6 Flex (Applied Biosystems, Foster City, CA, USA) with SYBR green I (Xceed qPCR SG Mix, Institute of Applied Biotechnologies, Prague, Czech Republic) detection according to the manufacturer’s protocol. The samples contained both forward and reverse primers (both 250 nM) and 5 µL of diluted cDNA. Primer sequences are listed in Table 2. The PCR reactions started with a denaturation step (10 min, 95 C) and annealing and extension (40 s, 60 C). Fluorescence data were recorded at the end of each ampliﬁcation step. Relative expression levels of the target genes were calculated as fold changes in triplicates for each −∆∆Ct method. [45]. The normalized expression level was expressed using a geometric group using the 2 mean of reference genes (glyceraldehyde 3-phosphate dehydrogenase, GAPDH; subunit A of succinate dehydrogenase complex, SDHA). C) followed by 40 cycles of ampliﬁcation which consisted of denaturation (10 s, 95 C and for 5 min at 80 ◦ ◦ ◦ ◦ ◦ ◦ ◦ Int. J. Mol. Sci. 2019, 20, 4562 11 of 15 Table 2. List of primers used for RT-qPCR analysis of the selected genes. Gene CYP3A4 CYP2C CBR1 AKR1C Forward Primer Reverse Primer CCCCTGAAATTAAGCTTAGGAGG TTTGGGATGGGGAAGAGGAG TTGGTACCCGAGATGTGTGC ATGAGGAGCAGGTTGGACTG CTGGTGTTCTCAGGCACAGA GGAGCACAGCCCAGGAT CTTGGGGTTTTATTAGAGGGAG GCTTTGAAGTGTAGAATATGTCTTCT 3.9. Western Blotting ◦ The liver slices were collected after 24 h of incubation. All treatments were performed in triplicates, with every slice was placed separately into 500 µL of lysis buﬀer and stored at −80 C until use. Equal volumes of once homogenized and centrifuged triplicate samples were pooled together. Protein concentration was measured using the BCA protein assay (Sigma Aldrich, Prague, Czech Republic) according to the manufacturer‘s instructions. The proteins (25 µg) were loaded onto a 10% sodium dodecyl sulfate (SDS; w/v)–polyacrylamide gel (with 4% stacking gel) and separated by SDS-PAGE electrophoresis. The proteins were transferred onto a nitrocellulose membrane using the Trans-Blot Turbo Transfer System (Bio-Rad, Hercules, CA, USA). The membrane blocking was performed in a 5% non-fat dry milk/TRIS-buﬀered saline-Tween-20 (TBS-T) solution at room temperature for C. Following primary 2 h. Incubation with primary antibodies was accomplished overnight at 4 antibodies were employed: Anti-Calnexin (ab75801, 1:2000), anti-AKR1C3 (ab27491, 0.1 µg/mL), anti-CBR1 (ab4148, 1:5000), anti-CYP2C (ab22596, 1:1000) (Abcam, Cambridge, UK), and anti-CYP3A4 (NB600-1396, 1:10,000) (Novus Biologicals, Cambridge, UK). Calnexin, a housekeeping protein, was used as a loading control. Afterwards, the membrane was washed with 0.3% TBS-T solution for 6 × 5 min, incubated with respective secondary antibodies conjugated with horseradish peroxidase (bovine anti-rabbit (sc2370) and bovine anti-goat (sc2350), 1:10,000) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) at room temperature for 1 h, and rinsed with TBS-T solution for 6 × 5 min. Visualization of protein bands was carried out using the chemiluminescence kit (GE Healthcare, Buckinghamshire, UK) and Carestream BioMax light ﬁlm (Sigma Aldrich, Prague, Czech Republic). The relative protein signal intensities were determined densitometrically using ImageJ software (National Institutes of Health, Bethesda, MD, USA). ◦ 3.10. Statistical Analysis All calculations were performed in Microsoft Excel and GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). The results of the luciferase reporter gene assays are presented as the relative change in Renilla-normalized ﬁreﬂy luciferase activities compared to the vehicle-treated control activities set as 100%. The presented results are based on at least three independent experiments (n = 3). A p-value of < 0.05 was considered to be statistically signiﬁcant. In the RT-qPCR analysis, three liver samples (a triplicate) were measured individually, whereas in the Western blot analysis, a pooled sample was prepared from a triplicate and the experiment was repeated four times. The results are expressed as the mean ± SD. One-way ANOVA followed by Dunnett‘s post hoc test was used for the statistical evaluation of diﬀerences between the treated samples and control. Diﬀerences of p < 0.05 were considered as statistically signiﬁcant. 4. Conclusions Despite the fact that the studied sesquiterpenes acted as agonists of the PXR receptor, they possessed only a weak or no inﬂuence on the expression of CYP3A4, CYP2C, CBR1, and AKR1C at both the mRNA and protein levels. Moreover, high inter-individual variability both in the basal levels of the tested enzymes and in the modulatory eﬀect of the sesquiterpenes in individual PCLS were observed. It seems improbable that the studied sesquiterpenes could signiﬁcantly inﬂuence the bioavailability Int. J. Mol. Sci. 2019, 20, 4562 12 of 15 and eﬃcacy of concomitantly administered drugs. Therefore, serious herb–drug interactions with the studied sesquiterpenes are not expected. Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/18/ 4562/s1. Author Contributions: Conceptualization, L.S. and I.B.; Data curation, M.Š., T.Z. and T.S.; Formal analysis, M.Š., T.S. and P.M.; Funding acquisition, P.P., L.S. and I.B.; Investigation, M.Š., T.Z., T.S. and Z.Š.; Methodology, M.Š., T.Z., T.S. and P.M.; Project administration, I.B.; Resources, P.P., Z.Š., L.S. and I.B.; Supervision, I.B.; Validation, M.Š., T.Z., P.P. and P.M.; Visualization, M.Š. and T.S.; Writing—original draft, M.Š., P.P., P.M., L.S. and I.B.; Writing—review and editing, P.P., P.M., L.S. and I.B. All authors read and approved the ﬁnal manuscript. Funding: This research was funded by the Czech Science Foundation (grant number 18-09946S) and by Charles University (Research Project SVV 260 416). Lenka Skálová, Tomáš Smutný and Petr Pávek were partly supported by the project EFSA-CDN [CZ.02.1.01/0.0/0.0/16_019/0000841], co-funded by ERDF. Acknowledgments: We thank Daniel Paul Sampey for English revision. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations aldo-keto reductase 1C caryophyllene oxide β-caryophyllene carbonyl reductase 1 cis-nerolidol cytochrome P450 diethyl pyrocarbonate diabetes mellitus drug-metabolizing enzyme dimethyl sulfoxide farnesol fetal bovine serum glyceraldehyde 3-phosphate dehydrogenase hyperlipidemia α-humulene hypertension methylcholanthrene rifampicin reverse transcription-quantitative polymerase chain reaction succinate dehydrogenase complex, subunit A status post cerebrovascular accident sodium dodecyl sulfate–polyacrylamide gel electrophoresis TRIS-buﬀered saline-Tween-20 trans-nerolidol AKR1C CAO CAR CBR1 cNER CYP DEPC DM DME DMSO FAR FBS GADPH HLD HUM HTN MC RIF RT-qPCR SDHA s/p CVA SDS-PAGE TBST-T tNER References 1. 2. 3. 4. Ekor, M. The growing use of herbal medicines: Issues relating to adverse reactions and challenges in monitoring safety. Front. Pharm. 2014, 4, 177. [CrossRef] [PubMed] Kennedy, D.A.; Seely, D. Clinically based evidence of drug-herb interactions: A systematic review. Expert. Opin. Drug Saf. 2010, 9, 79–124. [CrossRef] [PubMed] Awortwe, C.; Bruckmueller, H.; Cascorbi, I. Interaction of herbal products with prescribed medications: A systematic review and meta-analysis. Pharm. Res. 2019, 141, 397–408. [CrossRef] [PubMed] Agbabiaka, T.B.; Wider, B.; Watson, L.K.; Goodman, C. Concurrent Use of Prescription Drugs and Herbal Medicinal Products in Older Adults: A Systematic Review. Drugs Aging 2017, 34, 891–905. [CrossRef] [PubMed] Int. J. Mol. Sci. 2019, 20, 4562 13 of 15 5. 6. Tonner, P.H.; Kampen, J.; Scholz, J. Pathophysiological changes in the elderly. Best Pr. Res. Clin. Anaesthesiol. 2003, 17, 163–177. [CrossRef] Bartikova, H.; Hanusova, V.; Skalova, L.; Ambroz, M.; Bousova, I. Antioxidant, pro-oxidant and other biological activities of sesquiterpenes. Curr. Top. Med. Chem. 2014, 14, 2478–2494. [CrossRef] 8. 9. 7. Nguyen, L.T.; Mysliveckova, Z.; Szotakova, B.; Spicakova, A.; Lnenickova, K.; Ambroz, M.; Kubicek, V.; Krasulova, K.; Anzenbacher, P.; Skalova, L. The inhibitory eﬀects of β-caryophyllene, β-caryophyllene oxide and α-humulene on the activities of the main drug-metabolizing enzymes in rat and human liver in vitro. Chem. Biol. Interact. 2017, 278, 123–128. [CrossRef] [PubMed] Chaves, J.S.; Leal, P.C.; Pianowisky, L.; Calixto, J.B. Pharmacokinetics and tissue distribution of the sesquiterpene α-humulene in mice. Planta. Med. 2008, 74, 1678–1683. [CrossRef] Liu, H.; Yang, G.; Tang, Y.; Cao, D.; Qi, T.; Qi, Y.; Fan, G. Physicochemical characterization and pharmacokinetics evaluation of β-caryophyllene/β-cyclodextrin inclusion complex. Int. J. Pharm. 2013, 450, 304–310. [CrossRef] Saito, A.Y.; Sussmann, R.A.; Kimura, E.A.; Cassera, M.B.; Katzin, A.M. Quantiﬁcation of nerolidol in mouse plasma using gas chromatography-mass spectrometry. J. Pharm. Biomed. Anal. 2015, 111, 100–103. [CrossRef] Spicakova, A.; Szotakova, B.; Dimunova, D.; Mysliveckova, Z.; Kubicek, V.; Ambroz, M.; Lnenickova, K.; Krasulova, K.; Anzenbacher, P.; Skalova, L. Nerolidol and Farnesol Inhibit Some Cytochrome P450 Activities but Did Not Aﬀect Other Xenobiotic-Metabolizing Enzymes in Rat and Human Hepatic Subcellular Fractions. Molecules 2017, 22, 590. [CrossRef] [PubMed] 11. 10. 12. Lnenickova, K.; Svobodova, H.; Skalova, L.; Ambroz, M.; Novak, F.; Matouskova, P. The impact of sesquiterpenes β-caryophyllene oxide and trans-nerolidol on xenobiotic-metabolizing enzymes in mice in vivo. Xenobiotica 2018, 48, 1089–1097. [CrossRef] [PubMed] 13. De Graaf, I.A.; Olinga, P.; de Jager, M.H.; Merema, M.T.; de Kanter, R.; van de Kerkhof, E.G.; Groothuis, G.M. Preparation and incubation of precision-cut liver and intestinal slices for application in drug metabolism and toxicity studies. Nat. Protoc. 2010, 5, 1540–1551. [CrossRef] [PubMed] 14. Olinga, P.; Schuppan, D. Precision-cut liver slices: A tool to model the liver ex vivo. J. Hepatol. 2013, 58, 1252–1253. [CrossRef] 15. Edwards, R.J.; Price, R.J.; Watts, P.S.; Renwick, A.B.; Tredger, J.M.; Boobis, A.R.; Lake, B.G. Induction of cytochrome P450 enzymes in cultured precision-cut human liver slices. Drug Metab. Dispos. 2003, 31, 282–288. [CrossRef] [PubMed] Faugeras, L.; Dili, A.; Druez, A.; Krug, B.; Decoster, C.; D’Hondt, L. Treatment options for metastatic colorectal cancer in patients with liver dysfunction due to malignancy. Crit. Rev. Oncol. Hematol. 2017, 115, 59–66. [CrossRef] 16. 17. Ciombor, K.K.; Goﬀ, L.W. Current therapy and future directions in biliary tract malignancies. Curr. Treat. Opt. Oncol. 2013, 14, 337–349. [CrossRef] [PubMed] 18. He, W.Z.; Guo, G.F.; Yin, C.X.; Jiang, C.; Wang, F.; Qiu, H.J.; Chen, X.X.; Rong, R.M.; Zhang, B.; Xia, L.P. Gamma-glutamyl transpeptidase level is a novel adverse prognostic indicator in human metastatic colorectal cancer. Colorectal Dis. 2013, 15, e443–e452. [CrossRef] 19. Lehmann, J.M.; McKee, D.D.; Watson, M.A.; Willson, T.M.; Moore, J.T.; Kliewer, S.A. The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J. Clin. Invest. 1998, 102, 1016–1023. [CrossRef] 20. Whitlock, J.P., Jr. Induction of cytochrome P4501A1. Annu. Rev. Pharm. Toxicol. 1999, 39, 103–125. [CrossRef] 21. Nebert, D.W.; Dalton, T.P.; Okey, A.B.; Gonzalez, F.J. Role of aryl hydrocarbon receptor-mediated induction of the CYP1 enzymes in environmental toxicity and cancer. J. Biol. Chem. 2004, 279, 23847–23850. [CrossRef] [PubMed] 22. Pimkaew, P.; Kublbeck, J.; Petsalo, A.; Jukka, J.; Suksamrarn, A.; Juvonen, R.; Auriola, S.; Piyachaturawat, P.; Honkakoski, P. Interactions of sesquiterpenes zederone and germacrone with the human cytochrome P450 system. Toxicol. In Vitro 2013, 27, 2005–2012. [CrossRef] [PubMed] 23. Burk, O.; Arnold, K.A.; Nussler, A.K.; Schaeﬀeler, E.; Eﬁmova, E.; Avery, B.A.; Avery, M.A.; Fromm, M.F.; Eichelbaum, M. Antimalarial artemisinin drugs induce cytochrome P450 and MDR1 expression by activation of xenosensors pregnane X receptor and constitutive androstane receptor. Mol. Pharm. 2005, 67, 1954–1965. [CrossRef] [PubMed] Int. J. Mol. Sci. 2019, 20, 4562 14 of 15 24. Burk, O.; Piedade, R.; Ghebreghiorghis, L.; Fait, J.T.; Nussler, A.K.; Gil, J.P.; Windshugel, B.; Schwab, M. Diﬀerential eﬀects of clinically used derivatives and metabolites of artemisinin in the activation of constitutive androstane receptor isoforms. Br. J. Pharm. 2012, 167, 666–681. [CrossRef] [PubMed] 25. Zanger, U.M.; Schwab, M. Cytochrome P450 enzymes in drug metabolism: Regulation of gene expression, enzyme activities, and impact of genetic variation. Pharmacol. Therapeut. 2013, 138, 103–141. [CrossRef] [PubMed] 26. Bousova, I.; Skalova, L.; Soucek, P.; Matouskova, P. The modulation of carbonyl reductase 1 by polyphenols. Drug Metab. Rev. 2015, 47, 520–533. [CrossRef] [PubMed] 27. Penning, T.M.; Wangtrakuldee, P.; Auchus, R.J. Structural and Functional Biology of Aldo-Keto Reductase 28. Steroid-Transforming Enzymes. Endocr. Rev. 2019, 40, 447–475. [CrossRef] [PubMed] Fischedick, J.T.; Standiford, M.; Johnson, D.A.; De Vos, R.C.; Todorovic, S.; Banjanac, T.; Verpoorte, R.; Johnson, J.A. Activation of antioxidant response element in mouse primary cortical cultures with sesquiterpene lactones isolated from Tanacetum parthenium. Planta. Med. 2012, 78, 1725–1730. [CrossRef] [PubMed] 29. Nakamura, Y.; Yoshida, C.; Murakami, A.; Ohigashi, H.; Osawa, T.; Uchida, K. Zerumbone, a tropical ginger sesquiterpene, activates phase II drug metabolizing enzymes. Febs. Lett. 2004, 572, 245–250. [CrossRef] 30. Achour, B.; Barber, J.; Rostami-Hodjegan, A. Expression of hepatic drug-metabolizing cytochrome p450 enzymes and their intercorrelations: A meta-analysis. Drug Metab. Dispos. 2014, 42, 1349–1356. [CrossRef] [PubMed] 31. Martin, H.; Sarsat, J.P.; de Waziers, I.; Housset, C.; Balladur, P.; Beaune, P.; Albaladejo, V.; Lerche-Langrand, C. Induction of cytochrome P450 2B6 and 3A4 expression by phenobarbital and cyclophosphamide in cultured human liver slices. Pharm. Res. 2003, 20, 557–568. [CrossRef] [PubMed] 32. Persson, K.P.; Ekehed, S.; Otter, C.; Lutz, E.S.; McPheat, J.; Masimirembwa, C.M.; Andersson, T.B. Evaluation of human liver slices and reporter gene assays as systems for predicting the cytochrome p450 induction potential of drugs in vivo in humans. Pharm. Res. 2006, 23, 56–69. [CrossRef] [PubMed] 33. Qin, C.Z.; Lv, Q.L.; Wu, N.Y.; Cheng, L.; Chu, Y.C.; Chu, T.Y.; Hu, L.; Cheng, Y.; Zhang, X.; Zhou, H.H. Mechanism-based inhibition of Alantolactone on human cytochrome P450 3A4 in vitro and activity of hepatic cytochrome P450 in mice. J. Ethnopharmacol. 2015, 168, 146–149. [CrossRef] [PubMed] 34. Pelkonen, O.; Turpeinen, M.; Hakkola, J.; Honkakoski, P.; Hukkanen, J.; Raunio, H. Inhibition and induction of human cytochrome P450 enzymes: Current status. Arch. Toxicol. 2008, 82, 667–715. [CrossRef] [PubMed] Staines, A.G.; Sindelar, P.; Coughtrie, M.W.; Burchell, B. Farnesol is glucuronidated in human liver, kidney and intestine in vitro, and is a novel substrate for UGT2B7 and UGT1A1. Biochem. J. 2004, 384, 637–645. [CrossRef] [PubMed] 35. 36. Liu, Z.; Liu, S.; Xie, Z.; Pavlovicz, R.E.; Wu, J.; Chen, P.; Aimiuwu, J.; Pang, J.; Bhasin, D.; Neviani, P.; et al. Modulation of DNA methylation by a sesquiterpene lactone parthenolide. J. Pharm. Exp. 2009, 329, 505–514. [CrossRef] [PubMed] 37. Gopal, Y.N.; Arora, T.S.; Van Dyke, M.W. Parthenolide speciﬁcally depletes histone deacetylase 1 protein and induces cell death through ataxia telangiectasia mutated. Chem. Biol 2007, 14, 813–823. [CrossRef] [PubMed] 38. Pan, Y.Z.; Gao, W.; Yu, A.M. MicroRNAs regulate CYP3A4 expression via direct and indirect targeting. Drug Metab. Dispos. 2009, 37, 2112–2117. [CrossRef] [PubMed] 39. Wen, S.W.; Zhang, Y.F.; Li, Y.; Xu, Y.Z.; Li, Z.H.; Lu, H.; Zhu, Y.G.; Liu, Z.X.; Tian, Z.Q. Isoalantolactone Inhibits Esophageal Squamous Cell Carcinoma Growth Through Downregulation of MicroRNA-21 and Derepression of PDCD4. Dig. Dis. Sci. 2018, 63, 2285–2293. [CrossRef] 40. Zuo, W.; Wang, Z.Z.; Xue, J. Artesunate induces apoptosis of bladder cancer cells by miR-16 regulation of COX-2 expression. Int. J. Mol. Sci. 2014, 15, 14298–14312. [CrossRef] 41. Koe, X.F.; Lim, E.L.; Seah, T.C.; Amanah, A.; Wahab, H.A.; Adenan, M.I.; Sulaiman, S.F.; Tan, M.L. Evaluation of in vitro cytochrome P450 induction and inhibition activity of deoxyelephantopin, a sesquiterpene lactone from Elephantopus scaber L. Food Chem. Toxicol. 2013, 60, 98–108. [CrossRef] [PubMed] Svecova, L.; Vrzal, R.; Burysek, L.; Anzenbacherova, E.; Cerveny, L.; Grim, J.; Trejtnar, F.; Kunes, J.; Pour, M.; Staud, F.; et al. Azole antimycotics diﬀerentially aﬀect rifampicin-induced pregnane X receptor-mediated CYP3A4 gene expression. Drug Metab. Dispos. 2008, 36, 339–348. [CrossRef] [PubMed] 42. Int. J. Mol. Sci. 2019, 20, 4562 15 of 15 43. Dvorak, Z.; Vrzal, R.; Henklova, P.; Jancova, P.; Anzenbacherova, E.; Maurel, P.; Svecova, L.; Pavek, P.; Ehrmann, J.; Havlik, R.; et al. JNK inhibitor SP600125 is a partial agonist of human aryl hydrocarbon receptor and induces CYP1A1 and CYP1A2 genes in primary human hepatocytes. Biochem. Pharm. 2008, 75, 580–588. [CrossRef] [PubMed] 44. Zarybnicky, T.; Matouskova, P.; Lancosova, B.; Subrt, Z.; Skalova, L.; Bousova, I. Inter-Individual Variability in Acute Toxicity of R-Pulegone and R-Menthofuran in Human Liver Slices and Their Inﬂuence on miRNA Expression Changes in Comparison to Acetaminophen. Int. J. Mol. Sci. 2018, 19, 1805. [CrossRef] [PubMed] 45. Livak, K.J.; Schmittgen, T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001, 25, 402–408. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_genes14061211	Article Genome-Wide Assessment of Runs of Homozygosity by Whole-Genome Sequencing in Diverse Horse Breeds Worldwide Chujie Chen 1,†, Bo Zhu 2,†, Xiangwei Tang 1, Bin Chen 1, Mei Liu 1 and Jingjing Gu 1,* , Ning Gao 1 , Sheng Li 3,* 1 Hunan Provincial Key Laboratory for Genetic Improvement of Domestic Animal, College of Animal Science and Technology, Hunan Agricultural University, Changsha 410128, China; [email protected] (C.C.); [email protected] (X.T.); [email protected] (B.C.); [email protected] (M.L.); [email protected] (N.G.) 2 Novogene Bioinformatics Institute, Beijing 100015, China; [email protected] 3 Maxun Biotechnology Institute, Changsha 410024, China * Correspondence: [email protected] (S.L.); [email protected] (J.G.) † These authors contributed equally to this work. Abstract: In the genomes of diploid organisms, runs of homozygosity (ROH), consecutive segments of homozygosity, are extended. ROH can be applied to evaluate the inbreeding situation of individuals without pedigree data and to detect selective signatures via ROH islands. We sequenced and analyzed data derived from the whole-genome sequencing of 97 horses, investigated the distribution of genome- wide ROH patterns, and calculated ROH-based inbreeding coefﬁcients for 16 representative horse varieties from around the world. Our ﬁndings indicated that both ancient and recent inbreeding occurrences had varying degrees of impact on various horse breeds. However, recent inbreeding events were uncommon, particularly among indigenous horse breeds. Consequently, the ROH- based genomic inbreeding coefﬁcient could aid in monitoring the level of inbreeding. Using the Thoroughbred population as a case study, we discovered 24 ROH islands containing 72 candidate genes associated with artiﬁcial selection traits. We found that the candidate genes in Thoroughbreds were involved in neurotransmission (CHRNA6, PRKN, and GRM1), muscle development (ADAMTS15 and QKI), positive regulation of heart rate and heart contraction (HEY2 and TRDN), regulation of insulin secretion (CACNA1S, KCNMB2, and KCNMB3), and spermatogenesis (JAM3, PACRG, and SPATA6L). Our ﬁndings provide insight into horse breed characteristics and future breeding strategies. Keywords: ROH; whole-genome sequencing; inbreeding; horse; Thoroughbred 1. Introduction Domestication of horses began approximately 5500 years ago in the Eurasian steppe [1–3]. Since then, selective breeding and acclimatization have shaped the horse genome, resulting in more than 500 horse breeds worldwide [4]. Horses are employed in transportation, warfare, agriculture, and entertainment and can be categorized according to their usage (racing, sport, endurance, local, and gait), appearance (body size, coat color, and confor- mation), and temperament (hot, warm, and cold). Horse genomics has progressed rapidly since the establishment of the horse reference genome [5,6] and advancements in genomics technology. The genetic mechanisms of many horse traits have been investigated using single nucleotide polymorphism (SNP) chips and resequencing of the whole genome [7]. In contrast to SNP chips, whole-genome sequencing can repeatedly cover the entire genome, resulting in greater resolution and accuracy. Inbreeding is inevitable in the horse population, and breeds subjected to intense artiﬁ- cial selection and/or those with a small population size are more likely to experience the negative effects of inbreeding (such as inbreeding depression). Calculating the inbreeding Citation: Chen, C.; Zhu, B.; Tang, X.; Chen, B.; Liu, M.; Gao, N.; Li, S.; Gu, J. Genome-Wide Assessment of Runs of Homozygosity by Whole-Genome Sequencing in Diverse Horse Breeds Worldwide. Genes 2023, 14, 1211. https://doi.org/10.3390/ genes14061211 Academic Editor: Chunjiang Zhao Received: 25 April 2023 Revised: 29 May 2023 Accepted: 30 May 2023 Published: 1 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Genes 2023, 14, 1211. https://doi.org/10.3390/genes14061211 https://www.mdpi.com/journal/genes genesG C A TT A C GG C A TGenes 2023, 14, 1211 2 of 12 coefﬁcient from pedigree-based data [8] is the conventional method for measuring the in- breeding level. However, pedigree mistakes in farm animals [9] and horse populations [10] are prevalent. Runs of homozygosity (ROH) are continuous stretches of homozygosity regions spread across diploid genomes resulting from the transmission of identical haplo- types from common ancestors [11]. ROHs were ﬁrst identiﬁed in the human genome [12] and have been used to deﬁne the degree of inbreeding [13]. The ROH-based genomic inbreeding coefﬁcient (FROH) is described by measuring the proportion of the ratio of the sum of each individual’s ROH lengths to the total genome length [14]. Due to the fact that inbreeding is one of the primary causes of ROH [15], ROH is able to be applied to evaluate the inbreeding situation of individuals without pedigree data. In general, long ROHs indicate recent genome-wide inbreeding events, whereas short ROHs indicate ancient inbreeding. Additionally, population bottlenecks, genetic drift, and selection may contribute to the emergence of ROHs [16]. ROH are not distributed indistinguishably across the genome and accumulate in particular regions of the genome in various populations. The regions of the genome with the highest ROH occurrence in a population are known as “ROH islands” [17]. Genomic regions with selective signatures frequently overlap with ROH islands [18]. ROH islands can therefore be used to identify potentially selected genomic regions and identify the genetic basis of commercially valuable traits in farm animal populations [19]. In recent years, ROH detections on horses have become increasingly prevalent. However, most ROH studies on horses have focused on SNP chip data, and only a few have utilized whole-genome sequencing for ROH analysis [20]. We sequenced and utilized whole-genome sequencing data from 97 horses to identify and analyze ROH patterns in 16 globally representative horse breeds. Using the Thorough- bred population as a case study, we further investigated ROH islands containing potential candidate genes for performance traits. Our ﬁndings provide insight into horse breed characteristics and future breeding strategies. 2. Materials and Methods 2.1. Ethics Statement The Hunan Agricultural University’s Biomedical Research Ethics Committee approved this study (No. 202046). No horses were injured during or after the sample collection, and they remained healthy. 2.2. Sampling and Whole-Genome Sequencing In our horse panel, 37 horses were whole-genome sequenced at high coverage (~30×). Using a standard phenol-chloroform method, DNAs were obtained from freshly collected blood samples. Following instructions provided by the manufacturer, sequencing libraries were constructed and sequenced using an Illumina HiSeq 4000 sequencer to generate 150 bp paired-end reads. We also retrieved whole-genome sequencing data for more diverse horse breeds from NCBI (BioProject accession numbers: PRJEB10098, PRJEB10854, PRJNA168142, PRJNA205517, PRJNA230019, PRJNA233529, PRJNA288817, and PRJNA291776). We ana- lyzed a diverse horse panel (breed n = 16; total sample n = 97) with distinct appearances, breed-deﬁning traits, and geographic origins. The horse breeds included Arabian, An- dalusian, Akhal-Teke, Criollo, Debao, Friesian, Hanoverian, Jeju, Mongolian, Franches- Montagnes, Przewalskii, American Quarter Horse, Shetland pony, Standardbred, Thor- oughbred, and Yakutian. 2.3. Quality Controls and SNP Genotyping All raw sequencing reads were preprocessed for quality control and ﬁltered using FastQC. After quality control, the BWA program [21] was employed to map clean reads to the equine reference genome (EquCab3). Population-scale SNP calling was performed using the Bayesian approach in the SAMtools package [22]. The EquCab3 genome was used to conduct SNP annotation using ANNOVAR [23]. According to their genomic location, SNPs were classiﬁed into the following classes: exonic, intronic, splicing sites, upstream, Genes 2023, 14, 1211 3 of 12 downstream, and intergenic. Exonic SNPs were further classiﬁed as synonymous, non- synonymous, stop-gain, and stop-loss SNPs. 2.4. Runs of Homozygosity Detection ROH were calculated utilizing Plink v1.9 [24]. We scanned the entire genome of each horse using a sliding window strategy to identify the ROH regions. The criteria used to identify ROH were as follows: (1) the size of the sliding window was set to 500 kb; (2) the lowest SNP density was one per 50 kb; (3) 1 Mb was the maximum distance between SNPs; (4) based on the ROH length, 1 heterozygote was allowed in a sliding window; (5) a maximum of 4 missing genotypes were allowed. The deﬁned ROHs were categorized according to their length: <1 Mb, 1–5 Mb, 5–10 Mb, and >10 Mb. 2.5. Inbreeding Coefﬁcients As reported by McQuillan et al. [14], genome-wide inbreeding coefﬁcients were com- puted. In each individual, to calculate the inbreeding coefﬁcients for each of the ﬁve ROH categories, the total length of each ROH category was divided by the total length of the autosomes (2280.94 Mb) in the sequenced horse genome. The inbreeding coefﬁcients were recorded as FROH < 1 Mb (<1 Mb), FROH 1–5 Mb (1 to 5 Mb), FROH 5–10 Mb (5 to 10 Mb), FROH > 10 Mb (>10 Mb), and FROH all (including ROHs of all lengths). 2.6. Detection of ROH Islands in Thoroughbreds and Candidate Genes To determine the ROH islands in the Thoroughbred population (n = 22), we calculated the frequency of each SNP across all ROH regions in the entire Thoroughbred population. Potential ROH islands were identiﬁed as the top 1% of SNPs based on their occurrence frequency in the empirical distribution [17]. Using information from the Ensembl Genome Browser (www.ensembl.org, accessed on 20 February 2023), genes contained in the ROH islands were annotated. Functional analysis of the candidate genes was performed using Gene Ontology (GO) Biological Process enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses in DAVID 2021 [25], with an adjusted p-value greater than 0.05 indicating signiﬁcance. 3. Results 3.1. Whole-Genome Sequencing Using the whole-genome sequencing method, we sequenced and obtained a total of 56,768.2 million clean reads for 97 horse individuals, and the mean entire genome coverage for each horse was 25.6× (Table S1). We obtained 22,539,736 informative SNPs that were evenly dispersed across the equine genome (10 SNPs per kb on average) following a stringent quality control ﬁltering process. Using the Ensembl horse gene annotation set (Release 106), these population SNPs were annotated. A total of 8,461,302 (37.5%) SNPs were mapped within the gene regions, including 7,835,178 SNPs in introns, 208,369 SNPs in exons, and 417,052 SNPs in untranslated regions. 3.2. ROHs in the 16 Horse Breeds In this study, ROHs were identiﬁed in 16 diverse horse breeds that represented dif- ferent phenotypes and levels subject to selection (Figure 1). To understand the ROH characteristics of the studied horse population, we ﬁrst examined the average total length of the population ROH and the average number of total ROH for each horse breed. We found that the three highest average numbers of total ROH per horse breed were discov- ered in three sport horse breeds: Friesian (637), Arabian (621), and Thoroughbred (568). The three lowest average numbers of total ROH per horse were observed in Przewalskii primitive horses (180) and two local horse breeds, Debao (167) and Yakutian (102). Genes 2023, 14, 1211 4 of 12 Figure 1. Box plots of ROHs detected in 16 different horse breeds. The horse breeds were classiﬁed according to their main usages. The horse breeds included Arabian (AB), Andalusian (AL), Akhal- Teke (AT), Criollo (CR), Debao (DB), Friesian (FS), Hanoverian (HAN), Jeju (JEJU), Mongolian (MG), Franches-Montagnes (MON), Przewalskii (PRZ), American Quarter Horse (QT), Shetland pony (ST), Standardbred (STD), Thoroughbred (TB) and Yakutian (YAK). Hollow dots represent the outliers. Furthermore, the average total length of ROH maintained the same pattern as the average number of total ROH for each breed. Friesian horses had the largest average total length of ROH (635.69 Mb), followed by Arabian (602.63 Mb) and Thoroughbred (614.86 Mb). The lowest average total length of ROH was still found in the primitive and local horse breeds (Przewalskii: 159.15 Mb, Debao: 118.61 Mb, and Yakutian: 65.69 Mb). Of the ROH segments in the four length categories, most are short ROH segments (<1 Mb), followed by ROH segments of 1–5 Mb, accounting for 69.55% and 29.83% of the total number of ROHs, respectively. ROH segments (5–10 Mb) were present in 12 horse breeds, with Thoroughbred having the most abundant (117). ROHs greater than 10 Mb were also the highest in Thoroughbred (10), followed by Standardbreds and Franches- Montagnes, each with only one long ROH. No long ROH fragments (>10 Mb) were found in the other horse breeds. Table 1 provides a summary of the ROH segment statistics for the 16 horse breeds. Table 1. Summary statistics of the runs of homozygosity (ROH) based on length classes. Horse Population No. of Samples Total No. a Mean No. b Friesian Thoroughbred Arabian Shetland pony Andalusian Akhal-Teke Standardbred Hanoverian American Quarter Horse 5 22 5 3 4 5 4 4 7 3183 12,490 3104 1435 2102 2112 1549 1282 2173 637 568 621 478 526 422 387 321 310 Total Length (Mb) c 3178.47 13,526.86 3013.17 1545.37 2047.94 1997.03 1550.21 1207.67 1997.14 Total Mean Length (Mb) d Max. Length (Mb) e 635.69 614.86 602.63 515.12 511.99 399.41 387.55 301.92 285.31 7.82 13.89 7.34 8.18 8.47 8.37 11.38 7.32 8.99 Classiﬁcation of ROH by Length <1 Mb 1–5 Mb 5–10 Mb >10 Mb 2116 8233 2119 918 1428 1508 1094 918 1587 1060 4130 970 504 667 599 437 360 575 7 117 15 13 7 5 17 4 11 0 10 0 0 0 0 1 0 0 Genes 2023, 14, x FOR PEER REVIEW 4 of 12 3.2. ROHs in the 16 Horse Breeds In this study, ROHs were identified in 16 diverse horse breeds that represented dif-ferent phenotypes and levels subject to selection (Figure 1). To understand the ROH char-acteristics of the studied horse population, we first examined the average total length of the population ROH and the average number of total ROH for each horse breed. We found that the three highest average numbers of total ROH per horse breed were discovered in three sport horse breeds: Friesian (637), Arabian (621), and Thoroughbred (568). The three lowest average numbers of total ROH per horse were observed in Przewalskii primitive horses (180) and two local horse breeds, Debao (167) and Yakutian (102). Figure 1. Box plots of ROHs detected in 16 different horse breeds. The horse breeds were classified according to their main usages. The horse breeds included Arabian (AB), Andalusian (AL), Akhal-Teke (AT), Criollo (CR), Debao (DB), Friesian (FS), Hanoverian (HAN), Jeju (JEJU), Mongolian (MG), Franches-Montagnes (MON), Przewalskii (PRZ), American Quarter Horse (QT), Shetland pony (ST), Standardbred (STD), Thoroughbred (TB) and Yakutian (YAK). Hollow dots represent the out-liers. Furthermore, the average total length of ROH maintained the same pattern as the average number of total ROH for each breed. Friesian horses had the largest average total length of ROH (635.69 Mb), followed by Arabian (602.63 Mb) and Thoroughbred (614.86 Mb). The lowest average total length of ROH was still found in the primitive and local horse breeds (Przewalskii: 159.15 Mb, Debao: 118.61 Mb, and Yakutian: 65.69 Mb). Of the ROH segments in the four length categories, most are short ROH segments (<1 Mb), followed by ROH segments of 1–5 Mb, accounting for 69.55% and 29.83% of the total number of ROHs, respectively. ROH segments (5–10 Mb) were present in 12 horse breeds, with Thoroughbred having the most abundant (117). ROHs greater than 10 Mb were also the highest in Thoroughbred (10), followed by Standardbreds and Franches-Montagnes, each with only one long ROH. No long ROH fragments (>10 Mb) were found in the other horse breeds. Table 1 provides a summary of the ROH segment statistics for the 16 horse breeds. Genes 2023, 14, 1211 5 of 12 Table 1. Cont. Horse Population Franches- Montagnes Criollo Jeju Przewalskii Mongolian Debao Yakutian No. of Samples Total No. a Mean No. b 6 2 2 10 5 5 7 1476 491 501 1802 993 836 711 246 246 251 180 199 167 102 Total Length (Mb) c 1649.65 403.68 369.51 1591.60 785.24 593.08 459.81 Total Mean Length (Mb) d Max. Length (Mb) e 274.94 201.84 184.76 159.16 157.05 118.62 65.69 10.56 4.12 4.13 8.74 4.70 5.06 2.91 Classiﬁcation of ROH by Length <1 Mb 1–5 Mb 5–10 Mb >10 Mb 939 390 417 1378 809 712 639 520 101 84 423 184 123 72 16 0 0 1 0 1 0 1 0 0 0 0 0 0 a Total No.: The overall amount of ROH found in a horse population. b Mean No.: the average number of total ROH per horse breed. c Total Length: sum of all ROH lengths obtained within a horse population. d Total Mean Length: the average of the total length of ROH in each population. e Max. Length: maximum length of ROH segment detected in a horse population. 3.3. Assessment of Inbreeding Coefﬁcients According to the different ROH length categories, the inbreeding coefﬁcient was calculated for each horse, and then the average inbreeding coefﬁcient within the horse breed was calculated. Friesian had the highest value of FROH all (2.79 × 10−1), followed by Arabian (2.64 × 10−1) and Thoroughbred (2.58 × 10−1). Primitive and indigenous breeds, such as Przewalskii (6.98 × 10−2), Mongolian (6.89 × 10−2), Debao (5.20 × 10−2) and Yakutian (2.88 × 10−2), had relatively low inbreeding coefﬁcient values. In the <1 Mb and 1–5 Mb ROH range divisions, Friesian had the highest FROH (<1 Mb) (1.16 × 10−1) and FROH (1–5 Mb) (1.59 × 10−1), whereas Yakutian had the lowest FROH (<1 Mb) (2.26 × 10−2) and FROH (1–5 Mb) (6.20 × 10−3) among all the horse breeds. In the 5–10 Mb and >10 Mb long ROH range divisions, Thoroughbreds had the highest inbreeding coefﬁcients FROH (5–10 Mb) (1.42 × 10−2) and FROH (>10 Mb) (2.27 × 10−3) compared to the rest of the horse breeds. The mean genomic inbreeding coefﬁcients (FROH) for ROH of different length categories in horse populations are shown in Table 2. Table 2. Mean genomic inbreeding coefﬁcients (FROH) for ROH of different length categories in horse populations. Horse Population Horse Population No. of Samples FS AB TB AL ST AT STD HAN QT JEJU CR MON MG DB PRZ YAK Friesian Arabian Thoroughbred Andalusian Shetland pony Akhal-Teke Standardbred Hanoverian American Quarter Horse Jeju Criollo Franches- Montagnes Mongolian Debao Przewalskii Yakutian 5 5 23 4 3 5 4 4 7 2 2 5 5 5 10 7 ROH Length Category (Mb) FROH (<1 Mb) 1.16 × 10−1 1.15 × 10−1 9.73 × 10−2 9.69 × 10−2 8.26 × 10−2 8.14 × 10−2 7.33 × 10−2 6.00 × 10−2 5.94 × 10−2 5.41 × 10−2 5.19 × 10−2 4.96 × 10−2 4.26 × 10−2 3.54 × 10−2 3.50 × 10−2 2.26 × 10−2 FROH (1–5 Mb) 1.59 × 10−1 1.41 × 10−1 1.44 × 10−1 1.23 × 10−1 1.31 × 10−1 9.06 × 10−2 8.39 × 10−2 6.95 × 10−2 6.15 × 10−2 2.69 × 10−2 3.66 × 10−2 8.53 × 10−2 2.63 × 10−2 1.62 × 10−2 3.44 × 10−2 6.20 × 10−3 FROH (5–10 Mb) 3.81 × 10−3 7.50 × 10−3 1.42 × 10−2 4.87 × 10−3 1.19 × 10−2 3.06 × 10−3 1.15 × 10−2 2.79 × 10−3 4.22 × 10−3 0 0 8.80 × 10−3 0 4.44 × 10−4 3.83 × 10−4 0 FROH (>10 Mb) 0 0 2.27 × 10−3 0 0 0 1.25 × 10−3 0 0 0 0 9.26 × 10−4 0 0 0 0 FROH all SD 2.79 × 10−1 2.64 × 10−1 2.58 × 10−1 2.24 × 10−1 2.26 × 10−1 1.75 × 10−1 1.70 × 10−1 1.32 × 10−1 1.25 × 10−1 8.10 × 10−2 8.85 × 10−2 1.45 × 10−1 6.89 × 10−2 5.20 × 10−2 6.98 × 10−2 2.88 × 10−2 3.11 × 10−4 3.12 × 10−4 4.25 × 10−4 3.13 × 10−4 3.93 × 10−4 3.17 × 10−4 3.93 × 10−4 3.25 × 10−4 3.26 × 10−4 ND ND 4.43 × 10−4 2.21 × 10−4 1.90 × 10−4 3.10 × 10−4 1.33 × 10−4 Note: FROH was calculated using this formula: FROH = LROH/LAUTO. The total length of ROH on autosomes is denoted by LROH. LAUTO is the total autosomal length (2280.94 Mb). ND: not detected. SD: standard deviation (SD is only calculated for population sample sizes greater than 3). Genes 2023, 14, 1211 6 of 12 3.4. The ROH Islands and Candidate Genes in Thoroughbreds Since Thoroughbreds have been selectively bred for racing performance for more than 300 years, we further analyzed the ROH genome-wide distribution patterns using the Thoroughbred population as a case study. In total, 10,631 ROHs were identiﬁed in 22 Thoroughbred horses (Table S2). We found that ROH segments were not evenly distributed across chromosomes. Figure 2 displays the number of ROH and percentage of genomic ROH coverage in the Thoroughbred population on each chromosome. With a high coverage ratio of 28.2%, chromosome 1 of Equus caballus (ECA1) contains the most ROH segments (997). In contrast, ECA29 had the fewest ROH segments (102), and its coverage ratio is the second lowest (11.97%). ECA17 had the highest percentage of coverage (31.63%), while ECA12 had the lowest (11.15%). Figure 2. Distribution of ROH in Thoroughbred population. The bars represent the sum of number of ROH, and the line represents the percentage of genomic ROH coverage on horse chromosomes 1 to 31. Next, we examined the ROH islands in the Thoroughbred population to identify genomic regions that might have been subjected to selection pressure. We calculated the frequency of SNPs occurring in ROHs and selected the top 1% as an indicator of the ROH islands. The frequency of SNP occurrence within the ROH regions was plotted against the locations of the SNPs along the chromosome for each individual using the Manhattan plot. A total of 24 ROH islands containing 72 candidate genes were identiﬁed on ECA7, 10, 16, 19, 23, 25, 27, 29, 30, and 31 (Figure 3). The longest ROH island was identiﬁed on ECA16 with 3325 contiguous SNPs, whereas the shortest was observed on ECA31. ECA30 had the largest number of ROH islands (six ROH islands, including ﬁve candidate genes). Most identiﬁed ROH islands in Thoroughbreds contained candidate genes. However, six ROH islands on ECA25, 29, 30, and 31 did not contain any annotated protein-coding genes. Enrichment analyses for GO and KEGG on all identiﬁed candidate genes were conducted. Nine signiﬁcant GO biological process terms and three signiﬁcant KEGG pathways are listed in Supplementary Table S3. The most signiﬁcantly enriched GO terms were neurological signaling, neuronal development, positive regulation of heart rate and contraction, and metabolic processes. KEGG pathways were signiﬁcantly enriched in cholinergic synapses, retrograde endocannabinoid signaling, and insulin secretion. We found that the candidate genes were involved in neurotransmission (CHRNA6, PRKN, and GRM1), muscle development (ADAMTS15 and QKI), positive regulation of heart rate and contraction (HEY2 and TRDN), regulation of insulin secretion (CACNA1S, KCNMB2, and KCNMB3), and spermatogenesis (JAM3, PACRG, and SPATA6L). Genes 2023, 14, x FOR PEER REVIEW 6 of 12 QT American Quar-ter Horse 7 5.94 × 10−2 6.15 × 10−2 4.22× 10−3 0 1.25 × 10−1 3.26 × 10−4 JEJU Jeju 2 5.41 × 10−2 2.69 × 10−2 0 0 8.10 × 10−2 ND CR Criollo 2 5.19 × 10−2 3.66 × 10−2 0 0 8.85 × 10−2 ND MON Franches-Monta-gnes 5 4.96 × 10−2 8.53 × 10−2 8.80× 10−3 9.26 × 10−4 1.45 × 10−1 4.43 × 10−4 MG Mongolian 5 4.26 × 10−2 2.63 × 10−2 0 0 6.89 × 10−2 2.21 × 10−4 DB Debao 5 3.54 × 10−2 1.62 × 10−2 4.44 × 10−4 0 5.20 × 10−2 1.90 × 10−4 PRZ Przewalskii 10 3.50 × 10−2 3.44 × 10−2 3.83 × 10−4 0 6.98 × 10−2 3.10 × 10−4 YAK Yakutian 7 2.26 × 10−2 6.20× 10−3 0 0 2.88 × 10−2 1.33 × 10−4 Note: FROH was calculated using this formula: FROH = LROH/LAUTO. The total length of ROH on auto-somes is denoted by LROH. LAUTO is the total autosomal length (2280.94 Mb). ND: not detected. SD: standard deviation (SD is only calculated for population sample sizes greater than 3). 3.4. The ROH Islands and Candidate Genes in Thoroughbreds Since Thoroughbreds have been selectively bred for racing performance for more than 300 years, we further analyzed the ROH genome-wide distribution patterns using the Thoroughbred population as a case study. In total, 10,631 ROHs were identified in 22 Thoroughbred horses (Table S2). We found that ROH segments were not evenly distrib-uted across chromosomes. Figure 2 displays the number of ROH and percentage of ge-nomic ROH coverage in the Thoroughbred population on each chromosome. With a high coverage ratio of 28.2%, chromosome 1 of Equus caballus (ECA1) contains the most ROH segments (997). In contrast, ECA29 had the fewest ROH segments (102), and its coverage ratio is the second lowest (11.97%). ECA17 had the highest percentage of coverage (31.63%), while ECA12 had the lowest (11.15%). Figure 2. Distribution of ROH in Thoroughbred population. The bars represent the sum of number of ROH, and the line represents the percentage of genomic ROH coverage on horse chromosomes 1 to 31. Next, we examined the ROH islands in the Thoroughbred population to identify ge-nomic regions that might have been subjected to selection pressure. We calculated the frequency of SNPs occurring in ROHs and selected the top 1% as an indicator of the ROH islands. The frequency of SNP occurrence within the ROH regions was plotted against the locations of the SNPs along the chromosome for each individual using the Manhattan plot. A total of 24 ROH islands containing 72 candidate genes were identified on ECA7, 10, 16, 19, 23, 25, 27, 29, 30, and 31 (Figure 3). The longest ROH island was identified on ECA16 with 3325 contiguous SNPs, whereas the shortest was observed on ECA31. ECA30 had the largest number of ROH islands (six ROH islands, including five candidate genes). Genes 2023, 14, 1211 7 of 12 Figure 3. Manhattan plot of the occurrences (%) of each SNP within ROH regions in Thoroughbred population. Each colorful dot stands for an SNP. The horizontal red dotted line represents the cutoff level (top 1%). 4. Discussion 4.1. Distribution and Patterns of ROH in 16 Horse Populations In the diploid genome, ROHs are the contiguous regions in which all SNPs at any position are homozygous in an individual [13]. In our study, we examined the length patterns of ROH in 16 diverse horse populations. In general, short ROHs (1 Mb) were the most prevalent, followed by medium (1–5 Mb) and medium-long ROHs (5–10 Mb), with only a dozen ultra-long ROHs (>10 Mb) detected. The ROH lengths may approximate the period during which inbreeding occurs. For instance, short ROHs indicate a history of ancestral inbreeding, whereas long ROHs usually result from recent inbreeding events. We found that the average length of the short ROHs was much longer in horse breeds (such as Friesian, Thoroughbred, and Arabian) that had been subjected to strong artiﬁcial selection than in native horse breeds (such as Mongolian, Debao, and Yakutian). In conjunction with the number and average length of ROHs based on the length cate- gories, the results suggested that ancient and recent inbreeding events may have varying degrees of inﬂuence on various horse breeds. However, very recent instances of inbreeding were uncommon, particularly among indigenous horse breeds. It is worth noting that inbreeding events are not the only factor affecting ROH length. Owing to dynamic ran- domness and recombination during gamete formation, the generation and evolution of ROHs are random events to a certain extent [26]. In addition, reduced population size and bottlenecks may alter the properties of short ROH (<4 Mb) [27]. 4.2. ROH-Based Genomic Inbreeding Coefﬁcients Traditionally, the inbreeding coefﬁcient has been calculated primarily using data ob- tained from pedigrees. However, the horse pedigree records often contain errors that may have occurred long ago and could not be tracked. On the other hand, some native horse breeds did not even have pedigree records. Recently, calculating inbreeding coefﬁcients using the genome-wide SNP data of livestock is now achievable thanks to the advent of high-density SNP genotyping technology (such as SNP chips and whole-genome se- quencing) [19]. SNP data are more advantageous than pedigree data for evaluating the impact of inbreeding [28]. Moreover, SNP-based calculations of the inbreeding coefﬁcients demonstrated authentic relationships between individuals [29]. In our study, we used the whole-genome sequencing method to estimate unbiased genome-wide inbreeding coefﬁcients. We found that horse breeds that required breed Genes 2023, 14, x FOR PEER REVIEW 7 of 12 Most identified ROH islands in Thoroughbreds contained candidate genes. How-ever, six ROH islands on ECA25, 29, 30, and 31 did not contain any annotated protein-coding genes. Enrichment analyses for GO and KEGG on all identified candidate genes were conducted. Nine significant GO biological process terms and three significant KEGG pathways are listed in Supplementary Table S3. The most significantly enriched GO terms were neurological signaling, neuronal development, positive regulation of heart rate and contraction, and metabolic processes. KEGG pathways were significantly enriched in cho-linergic synapses, retrograde endocannabinoid signaling, and insulin secretion. We found that the candidate genes were involved in neurotransmission (CHRNA6, PRKN, and GRM1), muscle development (ADAMTS15 and QKI), positive regulation of heart rate and contraction (HEY2 and TRDN), regulation of insulin secretion (CACNA1S, KCNMB2, and KCNMB3), and spermatogenesis (JAM3, PACRG, and SPATA6L). Figure 3. Manhattan plot of the occurrences (%) of each SNP within ROH regions in Thoroughbred population. Each colorful dot stands for an SNP. The horizontal red dotted line represents the cutoff level (top 1%). 4. Discussion 4.1. Distribution and Patterns of ROH in 16 Horse Populations In the diploid genome, ROHs are the contiguous regions in which all SNPs at any position are homozygous in an individual [13]. In our study, we examined the length pat-terns of ROH in 16 diverse horse populations. In general, short ROHs (1 Mb) were the most prevalent, followed by medium (1–5 Mb) and medium-long ROHs (5–10 Mb), with only a dozen ultra-long ROHs (>10 Mb) detected. The ROH lengths may approximate the period during which inbreeding occurs. For instance, short ROHs indicate a history of ancestral inbreeding, whereas long ROHs usually result from recent inbreeding events. We found that the average length of the short ROHs was much longer in horse breeds (such as Friesian, Thoroughbred, and Arabian) that had been subjected to strong artificial selection than in native horse breeds (such as Mongolian, Debao, and Yakutian). In conjunction with the number and average length of ROHs based on the length categories, the results suggested that ancient and recent inbreeding events may have var-ying degrees of influence on various horse breeds. However, very recent instances of in-Genes 2023, 14, 1211 8 of 12 registrations and had studbooks had high overall inbreeding coefﬁcients (high FROH all). For example, due to the limited number of Thoroughbred founders, their effective popula- tion size is modest. In contrast, indigenous horse breeds showed relatively low degrees of inbreeding (low FROH all). We further calculated the FROH using different lengths of ROH as follows: FROH < 1 Mb, FROH 1–5 Mb, FROH 5–10 Mb, and FROH > 10 Mb, which reﬂect, respectively, ancestral inbreeding events that happened 50 generations, 10–50 generations, 5–10 gener- ations, and 5 generations ago [30]. All 16 horse breeds have historical inbreeding events dating back to 50 generations. Only three horse breeds (Thoroughbred, Standardbred, and Franches-Montagnes) had FROH > 10 Mb, indicating that inbreeding events occurred within ﬁve generations. Overall, the ROH-based genomic inbreeding coefﬁcient can be useful for estimating the inbreeding levels of individual horses lacking pedigree information. It could also provide useful indicators for monitoring increases in inbreeding, preserving horse breeds, and minimizing the adverse impacts of inbreeding on horse populations. 4.3. Candidate Genes in ROH Islands in Thoroughbreds Are Associated with Artiﬁcial Selection Traits ROH can be employed to deﬁne genomic regions subject to selection pressure and to characterize the occurrence of selective sweeps. Using the Thoroughbred population as a case study, we evaluated the candidate genes within the ROH islands. In contrast to other domesticated animals, horses are valued for their temperament. Important for the breeding, selection, and training of horses, temperament is deﬁned as an innate neurological charac- teristic. Due to the fact that the Thoroughbred horse has traditionally been characterized as a “hot blood” breed and their temperament has been described as extremely prone to nervousness [31], several candidate genes discovered by our analysis have been reported to play crucial roles in neurotransmission. For example, CHRNA6 encodes an α subunit of the neuronal nicotinic acetylcholine receptor that regulates dopaminergic neurotransmission. In humans, mutations in this gene most likely result in neuropsychiatric disorders (autism, depression, bipolar disorder, and schizophrenia), neurodegenerative diseases (Parkinson’s and Alzheimer’s disease), and lung cancer [32,33]. PRKN encodes Parkin, a component of the E3 ubiquitin ligase complex, and mutations in this gene have been implicated in Parkinson’s disease [34] and Autism spectrum disorder [35]. In addition, Prkn-deﬁcient mice exhibit autistic-like behavior and defective synaptogenesis [36]. The metabotropic glutamate receptor, which is encoded by the GRM1 gene, is involved in learning, synaptic activity, and neuroprotection. It is also associated with inherited cerebellar ataxia [37]. Thoroughbreds are considered to have great athletic ability because their maximum oxygen uptake (VO2max) is nearly double that of elite human athletes [38,39]. Equine scientists and breeders believe that Thoroughbreds must strengthen their cardiorespiratory capacity and muscle adaptation to obtain such high athletic ability. Consequently, it is possible that the cardiovascular and muscular systems of Thoroughbreds have been subjected to intense artiﬁcial selection. Several candidate genes associated with cardiac development have been identiﬁed. For example, HEY2 encodes a member of the basic Helix-Loop-Helix (bHLH) subfamily. It has been suggested that HEY2 controls heart growth by limiting cardiomyocyte proliferation [40] and is considered a crucial regulator of human cardiac development [41]. Triadin, one of the major cardiac sarcoplasmic reticulum proteins encoded by TRDN, stimulates muscle contraction via calcium-induced calcium release [42]. Humans and mice exhibited aberrant heart rates due to the loss of function of TRDN [43]. In addition, we identiﬁed candidate genes associated with muscle development, such as myoblast fusion (ADAMTS15) [44] and vascular smooth cell differentiation (QKI) [45]. Insulin is secreted by pancreatic β-cells to increase glucose consumption by promoting glucose uptake, glycogen synthesis, and adipogenesis in muscle and adipose tissue [46]. Insulin is essential for maintaining glucose homeostasis in the body. Studies have demon- strated that insulin secretion is a complex process in which sodium, potassium, and calcium channels in the membrane of pancreatic β-cells play crucial roles [47,48]. Thoroughbred horses are insulin-sensitive [49], and insulin stimulates muscle and protein synthesis [50]. Genes 2023, 14, 1211 9 of 12 Several candidate genes were signiﬁcantly associated with insulin secretion regulation in our study. For instance, KCNMB2 and KCNMB3 are two potassium calcium-activated channel genes inherited in the linkage region, and CACNA1S encodes the voltage-gated calcium channel subunit α CaV1.1, which may be jointly involved in regulating insulin secretion in Thoroughbreds. Since the vast majority of the sequenced Thoroughbreds we used were males, we also identiﬁed candidate genes involved in spermatogenesis (JAM3, PACRG, and SPATA6L). The adhesion of germ and Sertoli cells regulates the dynamic process of spermatogenesis. Junctional adhesion molecule-C (JAM-C, encoded by JAM3) is expressed by germ cells and localizes to the junctions between germ and Sertoli cells. JAM-C participates in the formation of acrosomes and germ cell polarity [51]. The development of the ﬂagellum is a crucial step in spermiogenesis because it enables sperm to reach the egg for fertilization. A MEIG1/PACRG complex in the manchette transports cargo to the centrioles, which are used to construct sperm tails [52]. Although SPATA6L (encoding spermatogenesis-associated 6-like protein) is predicted to be located in sperm connecting pieces and to be involved in spermatogenesis, its molecular function remains unknown. An important paralog of SPATA6L is SPATA6, which is necessary for the correct assembly of the sperm connecting component and head-tail junction [53]. In the artiﬁcial selection of Thoroughbreds for breeding, athletic performance and superior pedigree lines take precedence over reproduc- tive ﬁtness. Therefore, almost no selection pressure was exerted on fertility traits [54,55]. Typically, the conception rate of Thoroughbreds is lower than that of other livestock breeds, at about 60% per conception cycle [56]. All registered foals in the Thoroughbred horse industry must be born naturally, and artiﬁcial reproduction techniques are prohibited. In addition, breeding seasons in the Northern and Southern Hemispheres are strictly regulated by the industry. We hypothesized that the relaxation of reproductive traits could result in the accumulation of deleterious mutations that could diminish the reproductive ability of Thoroughbred stallions. These candidate genes associated with spermatogenesis may serve as targets for the future selection of Thoroughbreds in an effort to improve stallion fertility. 5. Conclusions The present study examined the distribution of ROH and estimated inbreeding coefﬁ- cients based on ROH in 16 diverse horse breeds using whole-genome sequencing data from 97 horses. Our data suggest that ancient and recent inbreeding may affect horse breeds differently, but recent inbreeding is uncommon, particularly among indigenous horse breeds. The ROH-based genomic inbreeding coefﬁcient is useful for estimating horse in- breeding levels in horses without pedigree data and for monitoring inbreeding increments in the horse population. Moreover, we identiﬁed 24 ROH islands containing 72 candidate genes associated with artiﬁcial selection traits in Thoroughbreds. These candidate genes are associated with neurotransmission, muscle development, positive regulation of heart rate and contraction, regulation of insulin secretion, and spermatogenesis. These ﬁndings provide insight into the characteristics of horse breeds and future breeding strategies. Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/genes14061211/s1, Table S1: Mapping results of clean reads against horse reference genome; Table S2: The statistics of ROH on each chromosome in the Thoroughbred population; Table S3: The top functional categories enriched for candidate genes located in ROH islands in Thoroughbreds. Author Contributions: Conceptualization, S.L. and J.G.; methodology, S.L. and J.G.; software, B.Z., S.L. and J.G.; validation, C.C., B.Z., S.L. and J.G.; formal analysis, C.C., B.Z., S.L. and J.G.; investigation, C.C. and B.Z.; resources, X.T., B.C., M.L. and N.G.; data curation, C.C. and B.Z.; writing—original draft preparation, C.C. and B.Z.; writing—review and editing, C.C., B.Z., S.L. and J.G.; visualization, J.G.; supervision, J.G.; project administration, J.G.; funding acquisition, J.G. All authors have read and agreed to the published version of the manuscript. Genes 2023, 14, 1211 10 of 12 Funding: This research was funded by a grant from the National Natural Science Foundation of China (No. 31501000). Institutional Review Board Statement: This work has been approved by the Biomedical Research Ethics Committee of Hunan Agricultural University (No. 202046). Informed Consent Statement: Not applicable. Data Availability Statement: The whole genome data used in this manuscript are available in the GenBank database under BioProject accession PRJNA416233, PRJEB10098, PRJEB10854, PRJNA168142, PRJNA205517, PRJNA230019, PRJNA233529, PRJNA288817 and PRJNA291776. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. 2. 3. 4. Gaunitz, C.; Fages, A.; Hanghoj, K.; Albrechtsen, A.; Khan, N.; Schubert, M.; Seguin-Orlando, A.; Owens, I.J.; Felkel, S.; Bignon- Lau, O.; et al. Ancient genomes revisit the ancestry of domestic and Przewalski’s horses. Science 2018, 360, 111–114. [CrossRef] [PubMed] Librado, P.; Fages, A.; Gaunitz, C.; Leonardi, M.; Wagner, S.; Khan, N.; Hanghoj, K.; Alquraishi, S.A.; Alfarhan, A.H.; Al-Rasheid, K.A.; et al. The Evolutionary Origin and Genetic Makeup of Domestic Horses. Genetics 2016, 204, 423–434. [CrossRef] [PubMed] Outram, A.K.; Stear, N.A.; Bendrey, R.; Olsen, S.; Kasparov, A.; Zaibert, V.; Thorpe, N.; Evershed, R.P. The earliest horse harnessing and milking. Science 2009, 323, 1332–1335. [CrossRef] [PubMed] Petersen, J.L.; Mickelson, J.R.; Cothran, E.G.; Andersson, L.S.; Axelsson, J.; Bailey, E.; Bannasch, D.; Binns, M.M.; Borges, A.S.; Brama, P.; et al. Genetic diversity in the modern horse illustrated from genome-wide SNP data. PLoS ONE 2013, 8, e54997. [CrossRef] 6. 5. Wade, C.M.; Giulotto, E.; Sigurdsson, S.; Zoli, M.; Gnerre, S.; Imsland, F.; Lear, T.L.; Adelson, D.L.; Bailey, E.; Bellone, R.R.; et al. Genome sequence, comparative analysis, and population genetics of the domestic horse. Science 2009, 326, 865–867. [CrossRef] Kalbﬂeisch, T.S.; Rice, E.S.; DePriest, M.S., Jr.; Walenz, B.P.; Hestand, M.S.; Vermeesch, J.R.; BL, O.C.; Fiddes, I.T.; Vershinina, A.O.; Saremi, N.F.; et al. Improved reference genome for the domestic horse increases assembly contiguity and composition. Commun. Biol. 2018, 1, 197. [CrossRef] Petersen, J.L.; Coleman, S.J. Next-Generation Sequencing in Equine Genomics. Vet. Clin. N. Am. Equine Pract. 2020, 36, 195–209. [CrossRef] 7. 8. Wright, S. Coefﬁcients of inbreeding and relationship. Am. Nat. 1922, 56, 330–338. [CrossRef] 9. Oliehoek, P.A.; Bijma, P. Effects of pedigree errors on the efﬁciency of conservation decisions. Genet. Sel. Evol. 2009, 41, 9. [CrossRef] 10. Hill, E.W.; Bradley, D.G.; Al-Barody, M.; Ertugrul, O.; Splan, R.K.; Zakharov, I.; Cunningham, E.P. History and integrity of thoroughbred dam lines revealed in equine mtDNA variation. Anim. Genet. 2002, 33, 287–294. [CrossRef] 11. Ceballos, F.C.; Joshi, P.K.; Clark, D.W.; Ramsay, M.; Wilson, J.F. Runs of homozygosity: Windows into population history and trait architecture. Nat. Rev. Genet. 2018, 19, 220–234. [CrossRef] [PubMed] 12. Broman, K.W.; Weber, J.L. Long homozygous chromosomal segments in reference families from the centre d’Etude du polymor- phisme humain. Am. J. Hum. Genet. 1999, 65, 1493–1500. [CrossRef] [PubMed] 13. Gibson, J.; Morton, N.E.; Collins, A. Extended tracts of homozygosity in outbred human populations. Hum. Mol. Genet. 2006, 15, 789–795. [CrossRef] [PubMed] 14. McQuillan, R.; Leutenegger, A.L.; Abdel-Rahman, R.; Franklin, C.S.; Pericic, M.; Barac-Lauc, L.; Smolej-Narancic, N.; Janicijevic, B.; Polasek, O.; Tenesa, A.; et al. Runs of homozygosity in European populations. Am. J. Hum. Genet. 2008, 83, 359–372. [CrossRef] 15. Curik, I.; Ferenˇcakovi´c, M.; Sölkner, J. Inbreeding and runs of homozygosity: A possible solution to an old problem. Livest. Sci. 2014, 166, 26–34. [CrossRef] Falconer, D.S. Introduction to Quantitative Genetics; Chennai, India: Pearson Education India, 1996. 16. 17. Pemberton, T.J.; Absher, D.; Feldman, M.W.; Myers, R.M.; Rosenberg, N.A.; Li, J.Z. Genomic patterns of homozygosity in worldwide human populations. Am. J. Hum. Genet. 2012, 91, 275–292. [CrossRef] 18. Bosse, M.; Megens, H.J.; Madsen, O.; Paudel, Y.; Frantz, L.A.; Schook, L.B.; Crooijmans, R.P.; Groenen, M.A. Regions of homozygosity in the porcine genome: Consequence of demography and the recombination landscape. PLoS Genet. 2012, 8, e1003100. [CrossRef] 19. Peripolli, E.; Munari, D.P.; Silva, M.; Lima, A.L.F.; Irgang, R.; Baldi, F. Runs of homozygosity: Current knowledge and applications in livestock. Anim. Genet. 2017, 48, 255–271. [CrossRef] 20. Gorssen, W.; Meyermans, R.; Janssens, S.; Buys, N. A publicly available repository of ROH islands reveals signatures of selection in different livestock and pet species. Genet. Sel. Evol. 2021, 53, 2. [CrossRef] 21. Li, H.; Durbin, R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 2010, 26, 589–595. [CrossRef] 22. Li, H.; Handsaker, B.; Wysoker, A.; Fennell, T.; Ruan, J.; Homer, N.; Marth, G.; Abecasis, G.; Durbin, R.; Genome Project Data Processing, S. The Sequence Alignment/Map format and SAMtools. Bioinformatics 2009, 25, 2078–2079. [CrossRef] [PubMed] Genes 2023, 14, 1211 11 of 12 23. Wang, K.; Li, M.; Hakonarson, H. ANNOVAR: Functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res. 2010, 38, e164. [CrossRef] [PubMed] 24. Purcell, S.; Neale, B.; Todd-Brown, K.; Thomas, L.; Ferreira, M.A.; Bender, D.; Maller, J.; Sklar, P.; de Bakker, P.I.; Daly, M.J.; et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 2007, 81, 559–575. [CrossRef] Sherman, B.T.; Hao, M.; Qiu, J.; Jiao, X.; Baseler, M.W.; Lane, H.C.; Imamichi, T.; Chang, W. DAVID: A web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022, 50, W216–W221. [CrossRef] [PubMed] 25. 26. Liu, G.E.; Hou, Y.; Zhu, B.; Cardone, M.F.; Jiang, L.; Cellamare, A.; Mitra, A.; Alexander, L.J.; Coutinho, L.L.; Dell’Aquila, M.E.; et al. Analysis of copy number variations among diverse cattle breeds. Genome Res. 2010, 20, 693–703. [CrossRef] 27. Kirin, M.; McQuillan, R.; Franklin, C.S.; Campbell, H.; McKeigue, P.M.; Wilson, J.F. Genomic runs of homozygosity record population history and consanguinity. PLoS ONE 2010, 5, e13996. [CrossRef] 28. Keller, M.C.; Visscher, P.M.; Goddard, M.E. Quantiﬁcation of inbreeding due to distant ancestors and its detection using dense single nucleotide polymorphism data. Genetics 2011, 189, 237–249. [CrossRef] 29. Visscher, P.M.; Medland, S.E.; Ferreira, M.A.; Morley, K.I.; Zhu, G.; Cornes, B.K.; Montgomery, G.W.; Martin, N.G. Assumption- free estimation of heritability from genome-wide identity-by-descent sharing between full siblings. PLoS Genet. 2006, 2, e41. [CrossRef] 30. Zanella, R.; Peixoto, J.O.; Cardoso, F.F.; Cardoso, L.L.; Biegelmeyer, P.; Cantao, M.E.; Otaviano, A.; Freitas, M.S.; Caetano, A.R.; Ledur, M.C. Genetic diversity analysis of two commercial breeds of pigs using genomic and pedigree data. Genet. Sel. Evol. 2016, 48, 24. [CrossRef] Sackman, J.E.; Houpt, K.A. Equine Personality: Association With Breed, Use, and Husbandry Factors. J. Equine Vet. Sci. 2019, 72, 47–55. [CrossRef] 31. 32. Tabares-Seisdedos, R.; Rubenstein, J.L. Chromosome 8p as a potential hub for developmental neuropsychiatric disorders: Implications for schizophrenia, autism and cancer. Mol. Psychiatry 2009, 14, 563–589. [CrossRef] [PubMed] 33. Wen, L.; Yang, Z.; Cui, W.; Li, M.D. Crucial roles of the CHRNB3-CHRNA6 gene cluster on chromosome 8 in nicotine dependence: Update and subjects for future research. Transl. Psychiatry 2016, 6, e843. [CrossRef] [PubMed] 34. Blauwendraat, C.; Nalls, M.A.; Singleton, A.B. The genetic architecture of Parkinson’s disease. Lancet Neurol. 2020, 19, 170–178. [CrossRef] 35. Barone, R.; Cirnigliaro, L.; Saccuzzo, L.; Valdese, S.; Pettinato, F.; Prato, A.; Bernardini, L.; Fichera, M.; Rizzo, R. PARK2 microdeletion in a multiplex family with autism spectrum disorder. Int. J. Dev. Neurosci. 2023, 83, 121–131. [CrossRef] [PubMed] 36. Huo, Y.; Lu, W.; Tian, Y.; Hou, Q.; Man, H.Y. Prkn knockout mice show autistic-like behaviors and aberrant synapse formation. iScience 2022, 25, 104573. [CrossRef] [PubMed] 37. Rossi, P.I.; Vaccari, C.M.; Terracciano, A.; Doria-Lamba, L.; Facchinetti, S.; Priolo, M.; Ayuso, C.; De Jorge, L.; Gimelli, S.; Santorelli, F.M.; et al. The metabotropic glutamate receptor 1, GRM1: Evaluation as a candidate gene for inherited forms of cerebellar ataxia. J. Neurol. 2010, 257, 598–602. [CrossRef] 38. Ohmura, H.; Matsui, A.; Hada, T.; Jones, J.H. Physiological responses of young thoroughbred horses to intermittent high-intensity treadmill training. Acta Vet. Scand. 2013, 55, 59. [CrossRef] 39. Ohmura, H.; Mukai, K.; Takahashi, Y.; Takahashi, T.; Jones, J.H. Hypoxic training increases maximal oxygen consumption in Thoroughbred horses well-trained in normoxia. J. Equine Sci. 2017, 28, 41–45. [CrossRef] 40. Davis, R.L.; Turner, D.L. Vertebrate hairy and Enhancer of split related proteins: Transcriptional repressors regulating cellular differentiation and embryonic patterning. Oncogene 2001, 20, 8342–8357. [CrossRef] 41. Gerrard, D.T.; Berry, A.A.; Jennings, R.E.; Piper Hanley, K.; Bobola, N.; Hanley, N.A. An integrative transcriptomic atlas of organogenesis in human embryos. eLife 2016, 5, e15657. [CrossRef] 42. Chopra, N.; Yang, T.; Asghari, P.; Moore, E.D.; Huke, S.; Akin, B.; Cattolica, R.A.; Perez, C.F.; Hlaing, T.; Knollmann-Ritschel, B.E.; et al. Ablation of triadin causes loss of cardiac Ca2+ release units, impaired excitation-contraction coupling, and cardiac arrhythmias. Proc. Natl. Acad. Sci. USA 2009, 106, 7636–7641. [CrossRef] [PubMed] 43. Chopra, N.; Knollmann, B.C. Triadin regulates cardiac muscle couplon structure and microdomain Ca(2+) signalling: A path 44. towards ventricular arrhythmias. Cardiovasc. Res. 2013, 98, 187–191. [CrossRef] [PubMed] Stupka, N.; Kintakas, C.; White, J.D.; Fraser, F.W.; Hanciu, M.; Aramaki-Hattori, N.; Martin, S.; Coles, C.; Collier, F.; Ward, A.C.; et al. Versican processing by a disintegrin-like and metalloproteinase domain with thrombospondin-1 repeats proteinases-5 and -15 facilitates myoblast fusion. J. Biol. Chem. 2013, 288, 1907–1917. [CrossRef] [PubMed] 45. van der Veer, E.P.; de Bruin, R.G.; Kraaijeveld, A.O.; de Vries, M.R.; Bot, I.; Pera, T.; Segers, F.M.; Trompet, S.; van Gils, J.M.; Roeten, M.K.; et al. Quaking, an RNA-binding protein, is a critical regulator of vascular smooth muscle cell phenotype. Circ. Res. 2013, 113, 1065–1075. [CrossRef] [PubMed] 46. Ding, Y.; Li, Y.; Chen, Q.; Niu, B. Advances in Antidiabetic Drugs Targeting Insulin Secretion. Curr. Pharm. Des. 2018, 24, 3990–3997. [CrossRef] 47. Thompson, B.; Satin, L.S. Beta-Cell Ion Channels and Their Role in Regulating Insulin Secretion. Compr. Physiol. 2021, 11, 1–21. [CrossRef] Genes 2023, 14, 1211 12 of 12 48. Hiriart, M.; Velasco, M.; Larque, C.; Diaz-Garcia, C.M. Metabolic syndrome and ionic channels in pancreatic beta cells. Vitam. Horm. 2014, 95, 87–114. [CrossRef] 49. Breuhaus, B.A. Glucose and Insulin Responses to an Intravenous Glucose Load in Thoroughbred and Paso Fino Horses. J. Equine Vet. Sci. 2019, 81, 102793. [CrossRef] 50. Urschel, K.L.; Escobar, J.; McCutcheon, L.J.; Geor, R.J. Insulin infusion stimulates whole-body protein synthesis and activates the upstream and downstream effectors of mechanistic target of rapamycin signaling in the gluteus medius muscle of mature horses. Domest. Anim. Endocrinol. 2014, 47, 92–100. [CrossRef] 51. Cartier-Michaud, A.; Bailly, A.L.; Betzi, S.; Shi, X.; Lissitzky, J.C.; Zarubica, A.; Serge, A.; Roche, P.; Lugari, A.; Hamon, V.; et al. Genetic, structural, and chemical insights into the dual function of GRASP55 in germ cell Golgi remodeling and JAM-C polarized localization during spermatogenesis. PLoS Genet. 2017, 13, e1006803. [CrossRef] 52. Li, W.; Tang, W.; Teves, M.E.; Zhang, Z.; Zhang, L.; Li, H.; Archer, K.J.; Peterson, D.L.; Williams, D.C., Jr.; Strauss, J.F., 3rd; et al. A MEIG1/PACRG complex in the manchette is essential for building the sperm ﬂagella. Development 2015, 142, 921–930. [CrossRef] [PubMed] 53. Yuan, S.; Stratton, C.J.; Bao, J.; Zheng, H.; Bhetwal, B.P.; Yanagimachi, R.; Yan, W. Spata6 is required for normal assembly of the sperm connecting piece and tight head-tail conjunction. Proc. Natl. Acad. Sci. USA 2015, 112, E430–E439. [CrossRef] [PubMed] 54. Novak, S.; Smith, T.A.; Paradis, F.; Burwash, L.; Dyck, M.K.; Foxcroft, G.R.; Dixon, W.T. Biomarkers of in vivo fertility in sperm and seminal plasma of fertile stallions. Theriogenology 2010, 74, 956–967. [CrossRef] 55. Gibb, Z.; Lambourne, S.R.; Aitken, R.J. The paradoxical relationship between stallion fertility and oxidative stress. Biol. Reprod. 2014, 91, 77. [CrossRef] 56. Nath, L.C.; Anderson, G.A.; McKinnon, A.O. Reproductive efﬁciency of Thoroughbred and Standardbred horses in north-east Victoria. Aust. Vet. J. 2010, 88, 169–175. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijms21144955	Article Black Soldier Fly Larvae Adapt to Diﬀerent Food Substrates through Morphological and Functional Responses of the Midgut Marco Bonelli 1,† Ling Tian 4 , Daniele Bruno 2,† , Gianluca Tettamanti 2,5,* , Matteo Brilli 1,3 , Silvia Caccia 6,* , Novella Gianfranceschi 1, and Morena Casartelli 1,5,* 1 Department of Biosciences, University of Milano, 20133 Milano, Italy; [email protected] (M.B.); [email protected] (M.B.); [email protected] (N.G.) 2 Department of Biotechnology and Life Sciences, University of Insubria, 21100 Varese, Italy; [email protected] Pediatric Clinical Research Center “Romeo ed Enrica Invernizzi”, University of Milano, 20133 Milano, Italy 3 4 Guangdong Provincial Key Laboratory of Agro-animal Genomics and Molecular Breeding/Guangdong Provincial Sericulture and Mulberry Engineering Research Center, College of Animal Science, South China Agricultural University, Guangzhou 510642, China; [email protected] BAT Center—Interuniversity Center for Studies on Bioinspired Agro-Environmental Technology, University of Napoli Federico II, 80138 Napoli, Italy 5 6 Department of Agricultural Sciences, University of Napoli Federico II, 80055 Portici (NA), Italy * Correspondence: [email protected] (G.T.); [email protected] (S.C.); [email protected] (M.C.) † These authors contributed equally. Received: 21 June 2020; Accepted: 9 July 2020; Published: 13 July 2020 Abstract: Modulation of nutrient digestion and absorption is one of the post-ingestion mechanisms that guarantees the best exploitation of food resources, even when they are nutritionally poor or unbalanced, and plays a pivotal role in generalist feeders, which experience an extreme variability in diet composition. Among insects, the larvae of black soldier ﬂy (BSF), Hermetia illucens, can grow on a wide range of feeding substrates with diﬀerent nutrient content, suggesting that they can set in motion post-ingestion processes to match their nutritional requirements. In the present study we address this issue by investigating how the BSF larval midgut adapts to diets with diﬀerent nutrient content. Two rearing substrates were compared: a nutritionally balanced diet for dipteran larvae and a nutritionally poor diet that mimics fruit and vegetable waste. Our data show that larval growth performance is only moderately aﬀected by the nutritionally poor diet, while diﬀerences in the activity of digestive enzymes, midgut cell morphology, and accumulation of long-term storage molecules can be observed, indicating that diet-dependent adaptation processes in the midgut ensure the exploitation of poor substrates. Midgut transcriptome analysis of larvae reared on the two substrates showed that genes with important functions in digestion and absorption are diﬀerentially expressed, conﬁrming the adaptability of this organ. Keywords: Hermetia illucens; insect midgut; post-ingestion regulation; diet composition; midgut transcriptome; waste management 1. Introduction In animals, the regulation of food intake and post-ingestion mechanisms are two important and strategic aspects to ensure optimal performance and healthy conditions [1]. With regard to nutritional demands, studies performed on insects and vertebrates have established that diet balancing and Int. J. Mol. Sci. 2020, 21, 4955; doi:10.3390/ijms21144955 www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2020, 21, 4955 2 of 27 protein:carbohydrate ratio (P:C ratio) are two sides of the same coin [2–4]. Indeed, food intake is primarily regulated by the amount of these macronutrients, and thus the P:C ratio inﬂuences not only growth and development, but also body composition, reproduction, aging, gut microbiota, metabolic homeostasis, and immune functions [2–5]. The optimal dietary P:C ratio varies among species and it can also change within species, depending on the developmental stage and even sex as a result of diﬀerent physiological needs [3,6,7]. The optimization of food intake is undoubtedly a prerequisite to match nutrient demand. Nevertheless, ﬁtness maximization is ultimately and decisively guaranteed by the ability of animals to implement eﬀective and targeted post-ingestion adjustments [1]. The presence and robustness of these mechanisms is of particular relevance in generalist feeders. Their responses, which include the modulation of nutrient digestion and absorption, the redirection of metabolism, and regulation of excretory system activity, have evolved not only to maximize food exploitation in order to meet nutritional requirements in the case of poor (i.e., with diluted or extremely diluted nutrients) or unbalanced diets, but also to compensate for the extreme variability of intake composition [1,7]. In this scenario, the gut plays a key homeostatic role, which strongly impacts on overall animal performance and ﬁtness [1]. The black soldier ﬂy (BSF), Hermetia illucens (Linnaeus, 1758) (Diptera: Stratiomyidae), represents a very attractive model for studying these aspects. Currently, BSF has a leading role in studies concerning the development of waste management strategies [8]. The interest in this saprophagous insect stems from multiple reasons. BSF larvae have a high feed conversion ratio [8] and exhibit an astonishing capacity to grow on a wide variety of organic materials and residues of disparate origin that are transformed into valuable biomass [6,9–13]. The high nutritional value of the larvae, and in particular the extent and the qualitative proﬁle of their protein content, makes them exploitable as animal feed constituent, while their grease represents suitable raw material for biodiesel production [8,12,14]. In addition, BSF larvae have been used as a model for in vitro digestion studies [15] and described as a source of bioactive compounds with high biotechnological and medical potential such as chitin (and its derivative, chitosan) and antimicrobial peptides [16–18]. Finally, proteins obtained from BSF larvae have been recently used for the production of bioplastics [19]. While the striking bioconversion potential of BSF is well proven, knowledge on how rearing substrate composition and moisture, which in turn may condition intestinal microbiota, and rearing conditions (e.g., temperature and insect density) can aﬀect the performance and body composition of this insect still needs to be improved [8,12,20,21]. In this respect, the literature on BSF clearly shows that carbohydrate and protein content in the rearing substrate signiﬁcantly inﬂuences larval developmental time, larval and pupal weight, and their nutritional value, and thus the overall performance of the biotransformation process [8,12,20,21]. The present study aims to investigate whether and how BSF larvae are able to exploit nutritionally poor diets thanks to post-ingestion midgut responses. We explored this issue by performing a comparative analysis of the midgut morphology and physiology of larvae reared on two diﬀerent diets, i.e., a nutritionally balanced diet for dipteran larvae and a diet that mimics fruit and vegetable waste composition. Our choice was motivated by the challenging nature of fruit and vegetable waste as raw material for BSF-mediated bioconversion processes [22,23]. Although nutritionally poor, it has been successfully used to rear H. illucens larvae [10,11,24] and strategies to reduce or reuse this valuable biomass are already underway [22,23,25]. In addition, fruit and vegetable waste is produced in large amounts by large-scale retail trade and wholesale markets, and it is allowed by the European Union as a rearing substrate for insects destined for ﬁsh feed (EU Regulation 2017/893) [26]. This study, which takes advantage of the knowledge obtained in our previous studies on the morphology and physiology of BSF larval midgut [24,27], evaluates the ﬂexibility of this organ in response to changes in diet composition. Along with basic parameters of larval performance on diﬀerent diets, we provide here a detailed morphofunctional characterization of the midgut responses. In addition, a complementary transcriptomic analysis of this organ does not simply support the Int. J. Mol. Sci. 2020, 21, 4955 3 of 27 structural and biochemical analyses performed herein, but also represents a well-stocked platform on which future work concerning the BSF larval midgut can be developed. 2. Results 2.1. Nutrient Content of the Rearing Substrates Since this study aims to analyze the response of the BSF larval midgut to diﬀerent nutrient availability, we preliminarily determined the nutrient composition of the two diets used to rear the larvae. Diﬀerences in the nutrient content of rearing substrates, especially carbohydrate and protein concentration and their ratio, can trigger insect behavioral and physiological adaptations [2–4]. Table 1 reports the composition of Standard Diet (SD) [28], a balanced diet conventionally used to rear BSF larvae, and Vegetable Mix Diet (VMD), a diet mimicking fruit and vegetable waste, expressed on “as fed” and “dry matter” bases. The former refers to the diet as fed to the larvae (taking into consideration the amount of moisture), the latter is computed on a moisture-free basis. Even though the P:C ratio of the two diets (calculated from the crude protein content and the available carbohydrates, i.e., starch, glucose, and fructose) was comparable, being 0.6 for SD and 0.4 for VMD, the content of these nutrients was quite diﬀerent. In particular, considering the diet composition expressed on “as fed” basis, crude protein and carbohydrate supply was lower in VMD, and this diet was nutritionally poorer than SD due to the dilution of these macronutrients (Table 1). Also lipids, which have fundamental structural and functional roles, were 12-times more diluted in VMD (Table 1). Therefore, the two diets can be eﬀectively used to compare the growth performance of BSF larvae and investigate if post-ingestion responses of the midgut are activated to compensate low nutrient supply of the feeding substrate. Table 1. Chemical composition and moisture content of the two experimental diets. Values are expressed as g per 100 g of diet. Component Crude protein Crude lipid Crude ﬁber a Nitrogen-free extract b Ash Hemicellulose c Cellulose c Lignin c Starch Glucose and fructose Moisture Standard Diet Vegetable Mix Diet As Fed Dry Matter As Fed Dry Matter 6.4 1.2 4.9 30.2 2.3 9.8 4.4 1.7 8.5 1.5 55 14.1 2.7 10.8 67.3 5.1 21.3 9.7 3.7 18.8 3.3 - 1.2 0.1 0.5 9.6 0.6 0.4 0.6 0.2 1.4 1.5 88 10.3 0.7 4.4 80.0 4.6 3.6 4.6 1.3 11.6 12.8 - a Includes most of cellulose and insoluble lignin. b Includes sugars, organic acids, pectins, soluble lignin, hemicellulose and a small percentage of cellulose. c Values calculated from neutral and acid detergent ﬁber analyses. Since insects, and BSF larvae in particular [18], are rich in minerals and their micronutrient proﬁle depends on the dietary substrate [18], the content of speciﬁc mineral micronutrients in the two diets was evaluated, too (Table 2). A comparison between SD and VMD evidenced that the former was richer in minerals, especially iron (the concentration was 10- and 40-times higher when considering the diet composition expressed on “dry matter” or “as fed”, respectively) (Table 2). Int. J. Mol. Sci. 2020, 21, 4955 4 of 27 Table 2. Heavy metal content of the two experimental diets. Values are expressed as mg per Kg of diet. Standard Diet Vegetable Mix Diet Component Iron Copper Nickel Zinc Moisture As Fed 119.1 4.2 0.8 22.4 55 Dry Matter As Fed Dry Matter 261.8 9.3 1.7 49.3 - 3.1 0.8 0.1 2.1 88 26.9 6.8 0.7 18.1 - 2.2. Larval Growth Rate The growth performance of BSF larvae reared on SD and VMD was evaluated. Figure 1 reports a typical experiment and shows that the duration of the larval stage, the end of which coincides with the attainment of the maximum weight before pupation (see “Measurement of larval growth rate” in Materials and Methods), was shorter when insects were grown on SD than on VMD (17 vs. 24 days, respectively). The maximum weight reached by the larvae was higher when larvae were reared on SD than on VMD (226.2 ± 3.7 mg vs. 188.5 ± 7.9 mg, respectively, mean ± s.e.m. of at least 20 samples, unpaired t-test, p-value < 0.001). These results are in agreement with our previous data [24] and show that larval growth performance, although aﬀected, was not dramatically altered by the diﬀerent nutritional content of the diets. Moreover, both larval groups were able to pupate and reached the adult stage. Figure 1. Growth rate of H. illucens larvae. The weight of the larvae reared on diﬀerent substrates was recorded until 25% of insects reached pupal stage. The day in which the larvae reached the maximum weight was considered the end of the larval stage (arrows). Then insects entered the prepupal stage and stopped feeding. 2.3. pH of Diets and Midgut Lumen The pH of freshly made SD and VMD was compared. The pH of SD was almost neutral, with a value of 6.8 ± 0.2 (mean ± s.e.m. of 5 samples), while the pH of VMD was acidic, with a value of 4.5 ± 0.2 (mean ± s.e.m. of 5 samples). Due to the diﬀerent pH values of the diets, the possible eﬀect of the feeding substrates on the pH of the midgut lumen was evaluated. This parameter has a critical role in midgut functionality and presents variations along the midgut of brachycerous Diptera, conferring peculiar functional features to each region of this organ [27,29]. pH was therefore measured Int. J. Mol. Sci. 2020, 21, 4955 4 of 27 Table 2. Heavy metal content of the two experimental diets. Values are expressed as mg per Kg of diet. Standard Diet Vegetable Mix Diet Component As Fed Dry Matter As Fed Dry Matter Iron 119.1 261.8 3.1 26.9 Copper 4.2 9.3 0.8 6.8 Nickel 0.8 1.7 0.1 0.7 Zinc 22.4 49.3 2.1 18.1 Moisture 55 - 88 - 2.2. Larval Growth Rate The growth performance of BSF larvae reared on SD and VMD was evaluated. Figure 1 reports a typical experiment and shows that the duration of the larval stage, the end of which coincides with the attainment of the maximum weight before pupation (see “Measurement of larval growth rate” in Materials and Methods), was shorter when insects were grown on SD than on VMD (17 vs. 24 days, respectively). The maximum weight reached by the larvae was higher when larvae were reared on SD than on VMD (226.2 ± 3.7 mg vs. 188.5 ± 7.9 mg, respectively, mean ± s.e.m. of at least 20 samples, unpaired t-test, p-value < 0.001). These results are in agreement with our previous data [24] and show that larval growth performance, although affected, was not dramatically altered by the different nutritional content of the diets. Moreover, both larval groups were able to pupate and reached the adult stage. Figure 1. Growth rate of H. illucens larvae. The weight of the larvae reared on different substrates was recorded until 25% of insects reached pupal stage. The day in which the larvae reached the maximum weight was considered the end of the larval stage (arrows). Then insects entered the prepupal stage and stopped feeding. 2.3. pH of Diets and Midgut Lumen The pH of freshly made SD and VMD was compared. The pH of SD was almost neutral, with a value of 6.8 ± 0.2 (mean ± s.e.m. of 5 samples), while the pH of VMD was acidic, with a value of 4.5 ± 0.2 (mean ± s.e.m. of 5 samples). Due to the different pH values of the diets, the possible effect of the feeding substrates on the pH of the midgut lumen was evaluated. This parameter has a critical role in midgut functionality and presents variations along the midgut of brachycerous Diptera, conferring peculiar functional features to each region of this organ [27,29]. pH was therefore measured in the anterior, middle, and posterior midgut of BSF larvae reared on SD and VMD. As reported in Table 3, Int. J. Mol. Sci. 2020, 21, 4955 5 of 27 in the anterior, middle, and posterior midgut of BSF larvae reared on SD and VMD. As reported in Table 3, no signiﬁcant diﬀerences were found among the pH values recorded. These results are in agreement with pH values reported previously [27], being acidic in the lumen of the anterior midgut, strongly acidic in the middle, and alkaline in the posterior. Therefore, unlike previous reports on BSF larvae [30], the pH of the feeding substrate did not inﬂuence the midgut luminal pH, at least in our experimental conditions. Table 3. pH values in the lumen of H. illucens midgut regions. Mean ± s.e.m., number of replicates in parenthesis. No statistically signiﬁcant diﬀerences were recorded among diet groups for each midgut region (unpaired t-test). Anterior midgut Middle midgut Posterior midgut SD 5.8 ± 0.1 (6) 2.4 ± 0.2 (6) 8.3 ± 0.4 (6) VMD 6.0 ± 0.1 (6) 1.8 ± 0.2 (7) 8.8 ± 0.1 (7) 2.4. Enzymatic Assays To assess if H. illucens larvae could modulate digestive activity in response to nutrient content of the feeding substrate, the activity of midgut enzymes involved in protein, carbohydrate, and lipid digestion was measured in larvae reared on both diets. As reported in Figure 2A, a signiﬁcantly higher proteolytic activity was observed in all midgut regions of larvae reared on VMD, which has a lower protein concentration compared to SD (Table 1). The highest proteolytic activity was detected in the posterior region for both diets (Figure 2A), therefore, attention was focused on this tract for further analyses on protein digestion. Previous data on larvae reared on SD indicated that serine proteases, endopeptidases with an alkaline optimum pH, are the main enzymes involved in the initial phase of protein digestion in the posterior region of BSF larval midgut [27]. To evaluate if serine proteases were responsible for the increase of total proteolytic activity in this midgut tract in larvae reared on VMD, their activity was measured at acidic pH (pH = 5), which is very far from the pH optimum of these enzymes; in this condition a signiﬁcant decrease of total proteolytic activity was observed (5-fold reduction, compared to pH = 8.5; total proteolytic activity at pH = 8.5: 169.7 ± 25.7 U, at pH = 5.0: 39.2 ± 7.2 U, mean ± s.e.m. of 4 experiments, paired t-test: p-value < 0.01). Given this evidence, the activity of trypsin and chymotrypsin, the two major serine proteases in insects [27,31,32], was measured in the posterior midgut. While no signiﬁcant diﬀerence in trypsin-like activity among larvae reared on the two diets was observed (Figure 2B), chymotrypsin-like activity in larvae reared on VMD was double of that measured in larvae grown on SD (Figure 2C). In addition to enzymes involved in the initial phase of protein digestion, enzymes responsible for the ﬁnal phase of digestion were also considered. Among them, aminopeptidase N (APN), which is anchored to the midgut brush border membranes [31], was selected as representative of exopeptidase activity. APN activity in the posterior midgut was signiﬁcantly higher in larvae reared on VMD, reaching values 6-fold higher than those measured in larvae grown on SD (Figure 2D). To evaluate carbohydrate digestion, the activity of α-amylase was measured [27,31,32]. As reported in Figure 2E, no signiﬁcant diﬀerence in α-amylase activity was observed in the anterior region and no activity was detected in both samples in the middle midgut, while larvae reared on VMD showed a signiﬁcantly lower activity in the posterior region, with a 60-fold reduction compared to that measured in larvae reared on SD. Finally, lipase activity was measured in anterior and posterior midgut. Middle midgut was not considered since lipase activity was not detectable in this region [27]. Signiﬁcant activity was measured in both anterior and posterior midgut of larvae reared on SD (Figure 2F), in accordance with our previous data [27]. On the contrary, no lipase activity was recorded in both districts from larvae reared on VMD. Int. J. Mol. Sci. 2020, 21, 4955 6 of 27 Figure 2. Enzymatic activities in midgut juice (A–C,E,F) or midgut homogenate (D) from larvae reared on SD (white bars) and VMD (grey bars). Total proteolytic activity in midgut juice extracted from anterior, middle, and posterior midgut (A). Trypsin- (B) and chymotrypsin- (C) like activity in midgut juice extracted from posterior midgut. Aminopeptidase N activity in the homogenate of posterior midgut (D). α-amylase activity in midgut juice extracted from anterior, middle, and posterior midgut (E). Lipase activity in midgut juice extracted from anterior and posterior midgut; n.d. non-detectable activity (F). The values are reported as mean ± s.e.m. of at least 3 experiments. Asterisks indicate statistically signiﬁcant diﬀerences between diet groups (unpaired t-test: * p-value < 0.05, p-value < 0.01, * p-value < 0.001). 2.5. Morphological Analysis of the Larval Midgut To evaluate if the feeding substrate could aﬀect the morphology of midgut cells, a thorough microscopy analysis of the midgut districts from larvae grown on SD and VMD was done. The anterior midgut did not show any relevant modiﬁcations (Figure 3A,B): columnar cells of larvae reared on both substrates showed a big central nucleus, basal infoldings, and a well-developed brush border. Moreover, a large amount of dark vesicles, probably containing digestive enzymes, was clearly visible under the microvilli. In the middle midgut of H. illucens larvae, the ﬁrst tract formed by an epithelium containing copper cells is followed by a district in which a thinner epithelium presents large and ﬂat cells [27,29]. Both cell types showed their typical morphological features regardless of the feeding substrate: copper cells exhibited the typical cup shape with a big central nucleus and developed microvilli (Figure 3C,D), while large ﬂat cells presented an elongated nucleus and very short microvilli (Figure 3E,F). The most consistent change in the morphology of the epithelium was observed in the posterior midgut (Figure 3G,H). In fact, the brush border of columnar cells showed a diﬀerent length depending on the diet. In particular, columnar cells of larvae reared on VMD were characterized by microvilli that were longer than in larvae reared on SD, suggesting an increase in the absorbing surface. Int. J. Mol. Sci. 2020, 21, 4955 6 of 27 Figure 2. Enzymatic activities in midgut juice (A–C,E,F) or midgut homogenate (D) from larvae reared on SD (white bars) and VMD (grey bars). Total proteolytic activity in midgut juice extracted from anterior, middle, and posterior midgut (A). Trypsin- (B) and chymotrypsin- (C) like activity in midgut juice extracted from posterior midgut. Aminopeptidase N activity in the homogenate of posterior midgut (D). α-amylase activity in midgut juice extracted from anterior, middle, and posterior midgut (E). Lipase activity in midgut juice extracted from anterior and posterior midgut; n.d. non-detectable activity (F). The values are reported as mean ± s.e.m. of at least 3 experiments. Asterisks indicate statistically significant differences between diet groups (unpaired t-test: * p-value < 0.05, p-value < 0.01, * p-value < 0.001). 2.5. Morphological Analysis of the Larval Midgut To evaluate if the feeding substrate could affect the morphology of midgut cells, a thorough microscopy analysis of the midgut districts from larvae grown on SD and VMD was done. The anterior midgut did not show any relevant modifications (Figure 3A,B): columnar cells of larvae reared on both substrates showed a big central nucleus, basal infoldings, and a well-developed brush border. Moreover, a large amount of dark vesicles, probably containing digestive enzymes, was clearly visible under the microvilli. In the middle midgut of H. illucens larvae, the first tract formed by an epithelium containing copper cells is followed by a district in which a thinner epithelium presents large and flat cells [27,29]. Both cell types showed their typical morphological features regardless of the feeding substrate: copper cells exhibited the typical cup shape with a big central nucleus and developed microvilli (Figure 3C,D), while large flat cells presented an elongated nucleus and very short microvilli (Figure 3E,F). The most consistent change in the morphology of the epithelium was observed in the posterior midgut (Figure 3G,H). In fact, the brush border of columnar cells showed a different length depending on the diet. In particular, columnar cells of larvae reared on VMD were characterized by microvilli that were longer than in larvae reared on SD, suggesting an increase in the absorbing surface. Int. J. Mol. Sci. 2020, 21, 4955 7 of 27 Figure 3. Morphological comparison of midgut from larvae reared on SD and VMD. (A,B): cross-sections of the anterior midgut. (C–F): copper (C,D) and large ﬂat cells (E,F) in the middle midgut of H. illucens larvae. (G,H): cross-sections of the posterior midgut. Columnar cells of larvae grown on VMD (H) show microvilli (arrowheads) that are longer than those of columnar cells of larvae reared on SD (G). Arrows: dark vesicles under the brush border. c: copper cells; e: epithelium; l: lumen; mc: muscle cells; pm: peritrophic matrix. Bars: 10 µm (A,B), 20 µm (C–H). 2.6. Histochemical Characterization of the Larval Midgut To get insights into the storage eﬃciency of the epithelium in relation to the diet, a histochemical approach was used. In particular, the storage of glycogen, which is important for energy production, was evaluated. Moreover, due to the importance of iron homeostasis in insects [33,34], the storage of iron was examined. It should be noted that iron metabolism initiates with its uptake from the diet by midgut cells, which also store iron by binding it to proteins [33,35]. Int. J. Mol. Sci. 2020, 21, 4955 7 of 27 Figure 3. Morphological comparison of midgut from larvae reared on SD and VMD. (A,B): cross-sections of the anterior midgut. (C–F): copper (C,D) and large flat cells (E,F) in the middle midgut of H. illucens larvae. (G,H): cross-sections of the posterior midgut. Columnar cells of larvae grown on VMD (H) show microvilli (arrowheads) that are longer than those of columnar cells of larvae reared on SD (G). Arrows: dark vesicles under the brush border. c: copper cells; e: epithelium; l: lumen; mc: muscle cells; pm: peritrophic matrix. Bars: 10 μm (A,B), 20 μm (C–H). 2.6. Histochemical Characterization of the Larval Midgut To get insights into the storage efficiency of the epithelium in relation to the diet, a histochemical approach was used. In particular, the storage of glycogen, which is important for energy production, was evaluated. Moreover, due to the importance of iron homeostasis in insects [33,34], the storage of iron was examined. It should be noted that iron metabolism initiates with its uptake from the diet by midgut cells, which also store iron by binding it to proteins [33,35]. Int. J. Mol. Sci. 2020, 21, 4955 8 of 27 The comparison of glycogen deposits in the midgut of larvae reared on the two diets evidenced a major diﬀerence in the anterior midgut, where glycogen accumulation was more consistent in larvae reared on SD than on VMD (Figure 4A,B). Moreover, the deposits in this region were sparsely distributed throughout the cytoplasm rather than localized in a speciﬁc area (Figure 4A,B). In the middle and posterior midgut, glycogen deposits showed almost the same localization and abundance in larvae grown on both diets (Figure 4C–F), although the distribution diﬀered in the two midgut regions. In detail, glycogen deposits were sparse into the cytoplasm in the middle midgut (Figure 4C,D), while in the posterior midgut they were localized in the apical region of columnar cells (Figure 4E,F). Figure 4. Comparison of glycogen accumulation in the three midgut regions of larvae reared on SD and VMD—Periodic Acid-Schiﬀ (PAS) staining. (A,B): anterior midgut of larvae reared on SD (A) shows a higher accumulation of glycogen (arrowheads) than VMD (B). (C–F): middle (C,D) and posterior (E,F) midgut of larvae reared on the two diets do not show signiﬁcant diﬀerences in glycogen accumulation. e: epithelium; l: lumen. Bars: 50 µm (A,B,E,F), 20 µm (C,D). Since iron plays important physiological functions in insects [33,34] and our analyses showed that its concentration signiﬁcantly diﬀers in the two diets (Table 2), we examined the iron content in midgut cells. This mineral, which is particularly abundant in SD, showed a broader distribution in the midgut of larvae grown on this diet compared to those reared on VMD (Figure 5). The presence of this element in the ﬁrst part of the posterior midgut in larvae reared on the two diets (Figure 5C,D) could be attributed to iron cells, a peculiar cell type that has been identiﬁed in the midgut epithelium of other Diptera [35]. Apart from this characteristic localization, in the anterior (Figure 5A,B) and in the second part of the posterior midgut (Figure 5E,F), the presence of iron in larvae reared on SD was higher than Int. J. Mol. Sci. 2020, 21, 4955 8 of 27 The comparison of glycogen deposits in the midgut of larvae reared on the two diets evidenced a major difference in the anterior midgut, where glycogen accumulation was more consistent in larvae reared on SD than on VMD (Figure 4A,B). Moreover, the deposits in this region were sparsely distributed throughout the cytoplasm rather than localized in a specific area (Figure 4A,B). In the middle and posterior midgut, glycogen deposits showed almost the same localization and abundance in larvae grown on both diets (Figure 4C–F), although the distribution differed in the two midgut regions. In detail, glycogen deposits were sparse into the cytoplasm in the middle midgut (Figure 4C,D), while in the posterior midgut they were localized in the apical region of columnar cells (Figure 4E,F). Figure 4. Comparison of glycogen accumulation in the three midgut regions of larvae reared on SD and VMD—Periodic Acid-Schiff (PAS) staining. (A,B): anterior midgut of larvae reared on SD (A) shows a higher accumulation of glycogen (arrowheads) than VMD (B). (C–F): middle (C,D) and posterior (E,F) midgut of larvae reared on the two diets do not show significant differences in glycogen accumulation. e: epithelium; l: lumen. Bars: 50 μm (A,B,E,F), 20 μm (C,D). Since iron plays important physiological functions in insects [33,34] and our analyses showed that its concentration significantly differs in the two diets (Table 2), we examined the iron content in midgut cells. This mineral, which is particularly abundant in SD, showed a broader distribution in the midgut of larvae grown on this diet compared to those reared on VMD (Figure 5). The presence of this element in the first part of the posterior midgut in larvae reared on the two diets (Figure 5C,D) could be attributed to iron cells, a peculiar cell type that has been identified in the midgut epithelium of other Diptera [35]. Apart from this characteristic localization, in the anterior (Figure 5A,B) and in the second part of the posterior midgut (Figure 5E,F), the presence of iron in larvae reared on SD was Int. J. Mol. Sci. 2020, 21, 4955 9 of 27 on VMD (Figure 5A,B,E,F). Regardless of the feeding substrate, the middle midgut did not show any staining (data not shown). Figure 5. Comparison of iron accumulation between the three midgut regions of larvae reared on SD and VMD—Perls’ method. (A,B): a higher iron accumulation in the anterior midgut of larvae reared on SD (A) compared to VMD (B) can be observed. (C,D): iron region in the ﬁrst part of the posterior midgut of larvae reared on the two diets. (E,F): higher accumulation of iron in the second part of the posterior midgut of larvae grown on SD than VMD is visible. Arrowheads: transition zone between middle and posterior midgut. 2.7. De Novo Transcriptome Analysis RNA-Seq was performed on midgut samples isolated from H. illucens larvae reared on SD and VMD. Gene expression analysis was performed in triplicate for each diet (6 libraries in total), obtaining over 516 million reads (with an average of about 67 million reads per replicate). The reads were quality trimmed and the high-quality reads were normalized and used as input to perform de novo transcriptome assembly as reported in Materials and Methods (“De novo transcriptome and functional annotation”). Reads normalization reduced the redundancy of the dataset, allowing a less computationally intensive analysis and increasing the quality of the ﬁnal assembly. This procedure resulted in a raw assembly of 32,643 transcripts with size ranging from 201 nt to 26,717 nt (average 1309 nt) and an N50 of 2392 nt. The average GC content was 41%. This preliminary assembly was ﬁltered by using three diﬀerent strategies (see “De novo transcriptome and functional annotation” in Materials and Methods for details). First, transcripts with very low expression level were removed, as they could be the outcome of erroneous reads and, moreover, they were extremely diﬃcult to identify as diﬀerentially expressed, as the dispersion of the gene expression estimate is inversely correlated with the expression level. Second, very similar transcripts were merged with CD-HIT-EST to minimize the redundancy [36]. Third, all the transcripts with a best match to sequences from organisms outside Int. J. Mol. Sci. 2020, 21, 4955 9 of 27 higher than on VMD (Figure 5A,B,E,F). Regardless of the feeding substrate, the middle midgut did not show any staining (data not shown). Figure 5. Comparison of iron accumulation between the three midgut regions of larvae reared on SD and VMD—Perls’ method. (A,B): a higher iron accumulation in the anterior midgut of larvae reared on SD (A) compared to VMD (B) can be observed. (C,D): iron region in the first part of the posterior midgut of larvae reared on the two diets. (E,F): higher accumulation of iron in the second part of the posterior midgut of larvae grown on SD than VMD is visible. Arrowheads: transition zone between middle and posterior midgut. 2.7. De Novo Transcriptome Analysis RNA-Seq was performed on midgut samples isolated from H. illucens larvae reared on SD and VMD. Gene expression analysis was performed in triplicate for each diet (6 libraries in total), obtaining over 516 million reads (with an average of about 67 million reads per replicate). The reads were quality trimmed and the high-quality reads were normalized and used as input to perform de novo transcriptome assembly as reported in Materials and Methods (“De novo transcriptome and functional annotation”). Reads normalization reduced the redundancy of the dataset, allowing a less computationally intensive analysis and increasing the quality of the final assembly. This procedure resulted in a raw assembly of 32,643 transcripts with size ranging from 201 nt to 26,717 nt (average 1309 nt) and an N50 of 2392 nt. The average GC content was 41%. This preliminary assembly was filtered by using three different strategies (see “De novo transcriptome and functional annotation” in Materials and Methods for details). First, transcripts with very low expression level were removed, as they could be the outcome of erroneous reads and, moreover, they were extremely difficult to identify as differentially expressed, as the dispersion of the gene expression estimate is inversely correlated with the expression level. Second, very similar transcripts were merged with CD-HIT-EST to minimize the redundancy [36]. Third, all the transcripts with a best match to sequences from organisms outside the arthropods were removed, as midgut samples contain associated contaminating microorganisms. The features of the resulting transcriptome are reported in Table 4. Int. J. Mol. Sci. 2020, 21, 4955 10 of 27 the arthropods were removed, as midgut samples contain associated contaminating microorganisms. The features of the resulting transcriptome are reported in Table 4. Table 4. Overview of the de novo transcriptome assembly of H. illucens midgut. Sequencing and Assembly Parameters Total number of transcripts a Minimum length (nt) Maximum length (nt) Average length (nt) N50 of transcripts (nt) b Reads remapped (%) Value 27,102 201 26,717 1170 2137 86.81 a Number of transcripts resulting after the assembly by Trinity and the subsequent collapsing step by CD-HIT-EST, and after ﬁltering out non-Arthropoda sequences. b N50 value represents the threshold delimiting 50% of the transcripts in the entire assembly which are equal to or larger than the reported value. One important question mark in de novo transcriptome assembly is what fraction of the entire coding potential of the species of interest has been recovered, as it aﬀects the understanding of processes under analysis. To assess the true coding potential, ﬁrstly the reads were mapped back to the assembly showing that 86.8% of them were included in the transcripts (86% map uniquely), and secondly, the transcriptome was compared against a dataset of genes that are universal in related genomes. For this purpose, BUSCO v3 [37], a tool that exploits a set of conserved single copy genes in a meaningful set of genomes to provide a measure of the completeness and redundancy of a genome/transcriptome assembly, was run. BUSCO pipeline was run with two diﬀerent datasets: the ﬁrst one contained a set of 303 proteins, which are conserved across Eukaryotes, while the second was more speciﬁc for our organism and comprised 1066 proteins that are conserved in all arthropod genomes. The output of the analysis showed that the present transcriptome was almost complete (Supplementary Table S1): it contained about 97% of the conserved single copy eukaryotic genes, 89% of which were estimated to be full-length and in single copy; consequently, only a small fraction of duplicated or fragmented genes was present. Similarly, our transcriptome contained about 94% of the Arthropoda conserved genes, 84% of which were complete and in single copy. The number of missing genes was about 3.5%. Taken together, these results show that the assembly can be considered a high-quality representative of the H. illucens midgut transcriptome. Once the quality and the degree of completeness of the de novo transcriptome of H. illucens midgut was veriﬁed, functional annotation of the transcripts was performed (Supplementary Table S2 in Supplementary ﬁle 1). To this aim, the Automatic assignment of Human Readable Descriptions pipeline (AHRD, [38]) was applied. The method is based on a similarity search between the transcripts and a reference set of proteins for which the annotation is known. Arthropoda proteins were used to assign a functional description to the transcripts, obtaining a gene ontology (GO) annotation for a total of 13,360 transcripts (Supplementary Table S3 in Supplementary ﬁle 1). The analysis of the annotation unveiled the abundance of transcripts related to “proteolysis” and “transport” in the Biological Processes (BP) category (Figure 6), in line with the prominent role of the midgut in digestion and transport of nutrients. Moreover, a high proportion of transcripts in BP category were involved in “phosphorylation” and “oxidation–reduction” processes (Figure 6). Within the Molecular Functions (MF) category, a bulk of transcripts pertained to “hydrolase activity” (Figure 6), conﬁrming the importance of such function in H. illucens midgut. Other MF categories with a high number of transcripts were “metal ion binding”, “nucleic acid binding”, “transferase activity”, and “ATP binding”, suggestive of a metabolic active organ (Figure 6). Int. J. Mol. Sci. 2020, 21, 4955 11 of 27 Figure 6. Graphical representation of the functional annotation of the transcriptome assembled in this work. Pie charts are realized using the CRAN platform and show the percentage of the 10 most represented gene ontology (GO) terms for Biological Processes and Molecular Functions. The categories are not terminal nodes in the GO hierarchy. The full list of categories is reported in Supplementary Table S3. Two H. illucens genome assemblies are available at the NCBI genome assembly database [39]. The ﬁrst, ASM101489v1 (GCA_001014895), has over 319K scaﬀolds and a N50 of only 1212 bp and was obtained in the framework of a work on transitions of sex chromosomes among Diptera [40]; the second assembly, indicated as “representative genome” for this species, is ASM983516v1 (GCA_009835165) and consists of 2806 scaﬀolds, and a N50 of over 1.69 Mbp [41]. By comparing our transcriptome assembly with the latter, a blast hit for 98.6% of our transcripts was found, 88.3% of which (i.e., 24,367 transcripts) sharing an identity level along the alignment of at least 90% together with a query coverage of 90% or more. The unavailability of gene or transcript sequences for the ASM983516v1 assembly did not allow further comparisons, but the above data conﬁrmed the degree of completeness and general quality of our transcriptome. 2.8. Diﬀerential Gene Expression Analysis GO annotation suggests the functional meaning of gene expression diﬀerences in midgut of H. illucens larvae reared on SD and VMD. The identiﬁcation of diﬀerentially expressed genes provides a high-resolution view of the changes occurring in this organ as a consequence of the diﬀerent rearing substrate, although it does not help understand large-scale functional modiﬁcations induced by the diﬀerentially expressed genes. For this reason, the examination of the transcriptomic changes in the midgut due to diets with diﬀerent nutritional content was ﬁrstly performed by discussing the enrichment analysis of BP that exploits the GO annotation of the diﬀerentially expressed genes. This kind of approach provides a less detailed description of the transcriptomic changes (Figure 7; Figure 8; Supplementary Tables S3–S7 in Supplementary ﬁle 1), but it is highly informative because it shows the functional consequences associated with the multitude of transcripts that changed their expression level. Figure 7 shows that upregulated genes in VMD were signiﬁcantly associated to GO categories related to proteolysis, translation, transport, and several metabolic processes (for instance “Glycogen metabolic process”). This evidence can be tentatively explained by the lower energy supply and protein content of the VMD that drives the midgut to optimize protein catabolism, increase nutrient uptake, and maintain high protein synthesis in support of the massive secretory activity (“Signal peptide processing” was also enriched). Conversely, the enrichment analysis of the genes signiﬁcantly decreasing their expression level (Figure 8) highlights a reorganization of cytoskeleton and several metabolic processes, together with a high representation of categories related to iron. As VMD has a lower nutrient content, a reduced activity of some metabolic and catabolic processes is needed but suﬃcient to process the ingested nutrients. Int. J. Mol. Sci. 2020, 21, 4955 11 of 27 Figure 6. Graphical representation of the functional annotation of the transcriptome assembled in this work. Pie charts are realized using the CRAN platform and show the percentage of the 10 most represented gene ontology (GO) terms for Biological Processes and Molecular Functions. The categories are not terminal nodes in the GO hierarchy. The full list of categories is reported in Supplementary Table S3. Two H. illucens genome assemblies are available at the NCBI genome assembly database [39]. The first, ASM101489v1 (GCA_001014895), has over 319K scaffolds and a N50 of only 1212 bp and was obtained in the framework of a work on transitions of sex chromosomes among Diptera [40]; the second assembly, indicated as “representative genome” for this species, is ASM983516v1 (GCA_009835165) and consists of 2806 scaffolds, and a N50 of over 1.69 Mbp [41]. By comparing our transcriptome assembly with the latter, a blast hit for 98.6% of our transcripts was found, 88.3% of which (i.e., 24,367 transcripts) sharing an identity level along the alignment of at least 90% together with a query coverage of 90% or more. The unavailability of gene or transcript sequences for the ASM983516v1 assembly did not allow further comparisons, but the above data confirmed the degree of completeness and general quality of our transcriptome. 2.8. Differential Gene Expression Analysis GO annotation suggests the functional meaning of gene expression differences in midgut of H. illucens larvae reared on SD and VMD. The identification of differentially expressed genes provides a high-resolution view of the changes occurring in this organ as a consequence of the different rearing substrate, although it does not help understand large-scale functional modifications induced by the differentially expressed genes. For this reason, the examination of the transcriptomic changes in the midgut due to diets with different nutritional content was firstly performed by discussing the enrichment analysis of BP that exploits the GO annotation of the differentially expressed genes. This kind of approach provides a less detailed description of the transcriptomic changes (Figure 7; Figure 8; Supplementary Tables S3–S7 in Supplementary file 1), but it is highly informative because it shows the functional consequences associated with the multitude of transcripts that changed their expression level. Figure 7 shows that upregulated genes in VMD were significantly associated to GO categories related to proteolysis, translation, transport, and several metabolic processes (for instance “Glycogen metabolic process”). This evidence can be tentatively explained by the lower energy supply and protein content of the VMD that drives the midgut to optimize protein catabolism, increase nutrient uptake, and maintain high protein synthesis in support of the massive secretory activity (“Signal peptide processing” was also enriched). Conversely, the enrichment analysis of the genes significantly decreasing their expression level (Figure 8) highlights a reorganization of cytoskeleton and several metabolic processes, together with a high representation of categories related to iron. As VMD has a lower nutrient content, a reduced activity of some metabolic and catabolic processes is needed but sufficient to process the ingested nutrients. Int. J. Mol. Sci. 2020, 21, 4955 12 of 27 Figure 7. Upregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are upregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way, categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO on the basis of the size −3 of each category in a background database (SwissProt [42]). Only categories with FDR ≤ 1.0 × 10 were selected, see full list in Supplementary Tables S3–S7. Figure 8. Downregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are downregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO on the basis of the size of each category in a background database (SwissProt [42]). Only categories with FDR ≤ 1.0 × 10 −3 were selected, see full list in Supplementary Tables S3–S7. Int. J. Mol. Sci. 2020, 21, 4955 12 of 27 Figure 7. Upregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are upregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way, categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO on the basis of the size of each category in a background database (SwissProt [42]). Only categories with FDR ≤ 1.0 × 10−3 were selected, see full list in Supplementary Tables S3–S7. Figure 8. Downregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are downregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO Int. J. Mol. Sci. 2020, 21, 4955 12 of 27 Figure 7. Upregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are upregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way, categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO on the basis of the size of each category in a background database (SwissProt [42]). Only categories with FDR ≤ 1.0 × 10−3 were selected, see full list in Supplementary Tables S3–S7. Figure 8. Downregulated genes associated to the GO category “Biological Processes”. Starting from the enrichment analysis of genes that are downregulated in midguts of larvae reared on VMD compared to SD, REVIGO was used to group similar biological processes on the basis of the SimRel semantic similarity metric; in this way categories with similar descriptions are close in the plot. As shown in the scale on the right, the color of the bubble identifying each biological process is a function of log10 (p value) for the false discovery rate for the enrichment of each process. Bubble size indicates the frequency of each GO term (larger size indicates larger categories) and was calculated by REVIGO Int. J. Mol. Sci. 2020, 21, 4955 13 of 27 The analysis of diﬀerentially expressed genes in the midgut of H. illucens larvae reared on SD and VMD identiﬁed 843 upregulated and 1067 downregulated transcripts, for a total of 1910. To unravel the functional adaptations of H. illucens midgut in response to the diﬀerent rearing substrates, diﬀerential expression of genes coding for digestive enzymes and transport proteins was analyzed in detail. Proteolytic enzymes were among transcripts increasing their expression in larvae reared on VMD (Figure 9). In particular, transcriptomic analysis showed that trypsin and trypsin-like transcripts were strongly enriched (29/97 transcripts), with fold change ranging from 2 to over 168, whereas only 4 out of 97 transcripts were downregulated. Similarly, 29/66 chymotrypsin and chymotrypsin-like transcripts were signiﬁcantly upregulated for VDM (in a range of fold change from 2 to 570), whereas only 10/66 transcripts were downregulated. Although among serine proteases only 5 out of 29 transcripts were diﬀerentially expressed, all of them were strongly upregulated. It is noteworthy that exopeptidases were strongly upregulated in larvae reared on VMD (Figure 10). Conversely, the expression of transcripts coding for enzymes involved in carbohydrate and lipid digestion was downregulated in the midgut of larvae reared on VMD (Figure 9). The transcriptomic data showed that 18/40 transcripts coding for α-amylases were signiﬁcantly downregulated with transcripts that were up to 20 times less abundant than in larvae reared on SD. Finally, lipases were strongly downregulated (Figure 9), too, with 19/35 transcripts diﬀerentially expressed, mostly downregulated (18/19). Figure 9. Log fold change of diﬀerentially expressed transcripts assigned to molecular functions corresponding to hydrolytic activity related to digestion in midguts of larvae reared on VMD compared to SD. For each function the number of genes (G.), transcripts (T.), and transcripts diﬀerentially expressed (D.) identiﬁed by the transcriptome analysis are shown together with their log fold change in our transcriptomic data. Int. J. Mol. Sci. 2020, 21, 4955 13 of 27 on the basis of the size of each category in a background database (SwissProt [42]). Only categories with FDR ≤ 1.0 × 10−3 were selected, see full list in Supplementary Tables S3–S7. The analysis of differentially expressed genes in the midgut of H. illucens larvae reared on SD and VMD identified 843 upregulated and 1067 downregulated transcripts, for a total of 1910. To unravel the functional adaptations of H. illucens midgut in response to the different rearing substrates, differential expression of genes coding for digestive enzymes and transport proteins was analyzed in detail. Proteolytic enzymes were among transcripts increasing their expression in larvae reared on VMD (Figure 9). In particular, transcriptomic analysis showed that trypsin and trypsin-like transcripts were strongly enriched (29/97 transcripts), with fold change ranging from 2 to over 168, whereas only 4 out of 97 transcripts were downregulated. Similarly, 29/66 chymotrypsin and chymotrypsin-like transcripts were significantly upregulated for VDM (in a range of fold change from 2 to 570), whereas only 10/66 transcripts were downregulated. Although among serine proteases only 5 out of 29 transcripts were differentially expressed, all of them were strongly upregulated. It is noteworthy that exopeptidases were strongly upregulated in larvae reared on VMD (Figure 10). Conversely, the expression of transcripts coding for enzymes involved in carbohydrate and lipid digestion was downregulated in the midgut of larvae reared on VMD (Figure 9). The transcriptomic data showed that 18/40 transcripts coding for α-amylases were significantly downregulated with transcripts that were up to 20 times less abundant than in larvae reared on SD. Finally, lipases were strongly downregulated (Figure 9), too, with 19/35 transcripts differentially expressed, mostly downregulated (18/19). Figure 9. Log fold change of differentially expressed transcripts assigned to molecular functions corresponding to hydrolytic activity related to digestion in midguts of larvae reared on VMD Int. J. Mol. Sci. 2020, 21, 4955 14 of 27 Figure 10. Log fold change of diﬀerentially expressed transcripts assigned to molecular functions corresponding to exopeptidases, membrane transport, lipid binding and transport, and iron binding in midguts of larvae reared on VMD compared to SD. For each function the number of genes (G.), transcripts (T.) and transcripts diﬀerentially expressed (D.) identiﬁed by the transcriptome analysis are shown together with their log fold change in our transcriptomic data. A focus on transcripts involved in transport and binding of nutrients (Figure 10) revealed that only 14 transcripts out of 452 annotated as transporters were diﬀerentially expressed, and the expression changes were in both directions. A similar pattern was observed for lipocalins and fatty acid binding proteins, for which, however, 11/16 transcripts were diﬀerentially expressed. Finally, most of ferritin transcripts showed coordinated downregulation (4/6) in larvae reared on VMD. 3. Discussion Insects are widely distributed in terrestrial ecosystems. They are characterized by an incredible variety of feeding habits thanks to an unmatched diversity of morphofunctional specializations and adaptability to cope with changes in diet composition [29]. The digestive apparatus, which is responsible for food ingestion and processing, represents a major player in this adaptation [29]. In this scenario, the study of the mechanisms that allow strict dietary specializations of insects is attractive, especially in the case of feeding substrates that are indigestible from a human perspective [43,44]. On the other hand, the mechanisms underlying the high eﬃciency of the digestive tract of highly polyphagous insects are worthy of investigation. These insects need to adapt to variations in nutrient composition of the ingested food, and H. illucens larvae, which are able to grow on a variety of organic waste materials and residues, are a relevant model to address this issue [6,9–13,25]. Moreover, the use of BSF in the feed sector has been promoted by recent changes in the EU legislation that partially Int. J. Mol. Sci. 2020, 21, 4955 14 of 27 compared to SD. For each function the number of genes (G.), transcripts (T.), and transcripts differentially expressed (D.) identified by the transcriptome analysis are shown together with their log fold change in our transcriptomic data. A focus on transcripts involved in transport and binding of nutrients (Figure 10) revealed that only 14 transcripts out of 452 annotated as transporters were differentially expressed, and the expression changes were in both directions. A similar pattern was observed for lipocalins and fatty acid binding proteins, for which, however, 11/16 transcripts were differentially expressed. Finally, most of ferritin transcripts showed coordinated downregulation (4/6) in larvae reared on VMD. Figure 10. Log fold change of differentially expressed transcripts assigned to molecular functions corresponding to exopeptidases, membrane transport, lipid binding and transport, and iron binding in midguts of larvae reared on VMD compared to SD. For each function the number of genes (G.), transcripts (T.) and transcripts differentially expressed (D.) identified by the transcriptome analysis are shown together with their log fold change in our transcriptomic data. 3. Discussion Insects are widely distributed in terrestrial ecosystems. They are characterized by an incredible variety of feeding habits thanks to an unmatched diversity of morphofunctional specializations and adaptability to cope with changes in diet composition [29]. The digestive apparatus, which is responsible for food ingestion and processing, represents a major player in this adaptation [29]. In this scenario, the study of the mechanisms that allow strict dietary specializations of insects is attractive, especially in the case of feeding substrates that are indigestible from a human perspective [43,44]. On the other hand, the mechanisms underlying the high efficiency of the digestive tract of highly polyphagous insects are worthy of investigation. These insects need to adapt to variations in nutrient composition of the ingested food, and H. illucens larvae, which are able to grow on a variety Int. J. Mol. Sci. 2020, 21, 4955 15 of 27 lifted the feed ban rules regarding the use of processed animal proteins from insects (EU Regulation 2017/893) [26]. Thus, knowledge on the mechanisms that allow BSF larvae to exploit substrates with diﬀerent nutritional quality is also relevant from an applied perspective. In the present work we evaluated larval growth performance, as well as investigated the activity of digestive enzymes, morphological features of cells, changes of long-term storage molecules, and diﬀerential gene expression in the midgut of BSF larvae reared on two diﬀerent diets, i.e., Standard Diet (SD) for dipteran larvae [28] and Vegetable Mix Diet (VMD). Whereas the former is a nutritionally balanced feeding substrate, the latter is nutritionally poor due to the lower concentration of protein, lipid, starch, and minerals compared to SD. These diets are thus particularly suitable to understand if and how the midgut sets in motion post-ingestion responses to compensate variations in nutrient composition of the diet because, although highly diﬀerent from a nutritional point of view, they allow comparable larval growth performance, as demonstrated in the present study and in another recent work [24]. We initially focused attention on the digestion of proteins, in which the concentration in VMD is 5-fold less than in SD and could thus represent a limitation for larval performance. It has been demonstrated that BSF larvae strongly rely on serine proteases for protein digestion, which is mainly accomplished in the posterior midgut where most of tryptic and chymotryptic activities have been measured [27]. Enzymatic assays performed on larvae reared on the two diets showed an increase of total proteolytic activity in all midgut tracts of the larvae reared on VMD. These functional data were supported by an increase of transcripts coding for proteolytic enzymes (i.e., trypsin, trypsin-like, chymotrypsin, chymotrypsin-like, and other serine proteases). However, conversely to chymotrypsin and chymotrypsin-like enzymes, we did not observe any signiﬁcant increase of proteolytic activity mediated by trypsin and trypsin-like enzymes. This could be due to the diﬀerent extent of the regulation of genes coding for the two serine proteases. In fact, while chymotrypsin and chymotrypsin-like transcripts related to 27 of the 38 annotated genes were diﬀerentially expressed (mostly upregulated), a relatively limited number of genes were subjected to regulation for trypsin and trypsin-like enzymes (i.e., 24 out of 79). Thus, since for trypsin and trypsin-like proteins the enzymatic activity from constitutively expressed genes represents the bulk of the total activity, the impact of upregulated transcripts might be negligible. The overall increase of total proteolytic activity associated with enzymes involved in the initial phase of protein digestion was accompanied and corroborated by an increase of exopeptidases transcription (aminopeptidases and carboxypeptidases), which are involved in the ﬁnal protein digestion [31]. Both functional (aminopeptidase activity) and transcriptomic data showed an increased ability of the midgut to obtain free amino acids from the diet. These are in turn internalized into midgut cells by apical transporters, the transcripts of which were also upregulated in larvae reared on VMD. This scenario suggests that, due to the low protein concentration in VMD, the larvae of H. illucens optimize the diﬀerent steps of proteolysis and maximize the internalization of amino acids to meet their nitrogen requirements. Many studies in insects demonstrated that the expression and activity of serine proteases increase with the amount of proteins in the dietary substrate when these are not limiting nutrients [45,46], while complete lack of protein, as during starvation [47–50] or between blood meals in hematophagous insects [45,51,52], leads their activity to decrease or even drop. In our study, proteins, which are scarcely represented in VMD, might be a limiting nutrient, and therefore H. illucens larvae set in motion compensatory mechanisms to make the best use of this rearing substrate. Our data also demonstrate a signiﬁcant regulation of α-amylases in relation to diet composition. These enzymes catalyze the hydrolysis of α-1,4 glycosidic bonds in polysaccharides (essentially starch and glycogen) to produce oligosaccharides that are then hydrolyzed into glucose units by α-glucosidases. The regulation of midgut α-amylase activity by diet composition has been observed in several insects, although responses are quite variable even within the same order [53]. Indeed, in polyphagous lepidopteran larvae, midgut α-amylase activity responds to carbohydrate composition of the diet, but the nature of the correlation is inconsistent and discrepant results are reported [54–57]. Int. J. Mol. Sci. 2020, 21, 4955 16 of 27 For example, Sarate et al. [57] demonstrated that amylase activity in the midgut of Helicoverpa armigera (Lepidoptera: Noctuidae) is inversely proportional to carbohydrate content in the diet, in contrast to Kotkar et al. [55] that showed no direct correlation in this insect. On the other hand, gut amylase activity in Spodoptera frugiperda (Lepidoptera: Noctuidae) larvae increases with food consumption and carbohydrate amount [54,56]. In H. illucens larvae, α-amylase activity is mainly localized in the lumen of anterior and posterior midgut [27], whereas it is negligible in the middle midgut. The present study indicates that diet composition is able to strongly regulate α-amylase activity associated to the posterior midgut, while the activity is unaﬀected in the anterior midgut. In particular, the low starch concentration in VMD compared to SD was associated to the nearly complete absence of α-amylase activity in the posterior midgut. This trend was supported by the analysis of diﬀerential gene expression that showed an overall decrease of α-amylase transcripts in the whole midgut. The apparent discrepancy between the overall expression decline and the presence of unaltered α-amylase activity in anterior midguts may be due to the limited impact of transcripts associated to the relatively short anterior midgut compared to the posterior tract [27]. Alternatively, since about a half of the transcripts annotated as α-amylase were not diﬀerentially expressed, α-amylase genes could be constitutively expressed in the anterior midgut, while their expression could be regulated in the posterior midgut in response to carbohydrate content of the diets. Considering that in our transcriptome the number of genes and transcripts assigned to α-amylase are 29 and 40, respectively, and that the downregulated transcripts (18) in larvae reared on VDM originate from 14 genes, it is possible to conclude that the regulation of amylolytic activity involves a signiﬁcant number of genes. Although in larvae reared on VMD starch hydrolysis occurs only in the anterior midgut, the activity of α-amylase in this tract and the amount of free sugars present in the diet appeared to be suﬃcient to meet larval requirements. Nevertheless, the larvae responded to low starch concentration in VMD with a slight increase of α-glucosidases, enzymes involved into the ﬁnal phases of starch digestion (i.e., transcript DN11894_c0_g1_i2), and, a strong increase (7–fold) in the expression of sugar transporters to maximize the internalization of free sugars (i.e., transcript DN12897_c0_g1_i1). In accordance with these data, α-amylase activity in larvae of Drosophila spp. (Diptera: Drosophilidae) is increased by dietary starch and a similar trend was observed in other insect species, such as Periplaneta americana (Blattodea: Blattidae) and Gryllus bimaculatus (Orthoptera: Gryllidae) [58–63]. Interestingly, in the phytophagous insect Locusta migratoria (Orthoptera: Acrididae) α-amylase activity is regulated by the P:C ratio of the diet rather than by carbohydrate content [1]. In particular, when carbohydrate content of the diet is high, α-amylase activity is reduced only in the presence of low protein content. The occurrence in H. illucens larvae of glucose repression (i.e., the inhibition of α-amylase activity by the ﬁnal products of starch digestion) observed in several Drosophila spp. [53,59] could not be determined in our experimental conditions (i.e., with ﬁxed amounts of starch), but such regulation is worthy of further investigations. Along with glycogen, lipids represent essential energy reservoirs and insects meet lipid requirements through de novo lipogenesis, which mainly occurs in the fat body, and dietary lipid digestion in the midgut lumen [64,65]. Lipases produce free fatty acids in the midgut lumen, that are then internalized into the midgut cells and converted into intracellular storage molecules or released as diacylglycerols into the hemolymph where they are shuttled by lipoproteins to other tissues [64]. VMD contains a 12-fold lower concentration of crude lipids than SD. H. illucens larvae grown on VDM did not increase the digestion of these nutrients in the lumen, but rather reduced the expression of lipases, which was accompanied by the drop of lipolytic activity. Accordingly, over half of the transcripts related to fatty acid binding proteins, that likely mediate the intracellular movement of absorbed fatty acids [64], were downregulated, too. The ability to regulate the expression and the activity of digestive lipases in response to the quantity and quality of dietary lipids has been detected in many insect species, and lipid deprivation, such as during starvation, is not always accompanied by lipolytic decrease [57,64–68]. In the case of H. illucens larvae reared on VMD, a basal, faintest lipase activity apparently guarantees the hydrolysis of available lipids. Indeed, although lower than in SD, the concentration of crude lipids is apparently suﬃcient to support larval growth. Int. J. Mol. Sci. 2020, 21, 4955 17 of 27 Our results demonstrate that diﬀerences in the nutrient composition of the feeding substrate also induced modiﬁcations of midgut cells at a morphological level. Previous studies have shown that the diet can induce ultrastructural changes in insect midgut cells as ﬂuctuations in the number and structure of lysosomes [69], proliferation of smooth and rough endoplasmic reticulum [70], and increase of the basal labyrinth surface [71]. Here, we observed a relevant change in the morphology of posterior midgut cells in larvae reared on VMD, which showed a signiﬁcant increase in the length of microvilli. This modiﬁcation, restricted to the region that is mainly involved in nutrient absorption [27], may respond to the need of a higher absorbing surface and likely represents an adaptation to the low nutritional content of VMD. Histochemical analysis revealed diﬀerences in glycogen accumulation in H. illucens larval midgut. Glycogen reserves are crucial to sustain metabolic homeostasis throughout the life cycle and, in holometabolous insects, stored energy is useful during metamorphosis [72,73]. In H. illucens, this process occurs in about 12 days and glycogen deposits are mobilized and progressively reduced during this period [74]. Our data demonstrate that larvae reared on VMD show lower accumulation of this long-term storage molecule than larvae reared on SD, especially in the anterior midgut, suggesting that glycogen accumulation is reduced in favor of survival and growth when larvae experience a nutritionally poor diet. The higher presence of glycogen reserves in larvae reared on SD could be related to the higher content of nutrients in this diet that allows glycogen accumulation in the midgut. We observed that glycogen accumulation occurs diﬀerentially in the three midgut regions regardless of the feeding substrate. In particular, the anterior and the posterior midgut are mainly involved in glycogen storage, indicating that the cells in the midgut epithelium accomplish diﬀerent metabolic functions [75]. Evidence suggesting the regionalization of glycogen and lipid metabolism in the midgut has already been obtained in adult Drosophila melanogaster [76,77]. Transcriptomic analysis was not particularly helpful in the interpretation of histochemical data. Glycogen metabolism is mainly controlled by two enzymes, i.e., glycogen synthase and glycogen phosphorylase. From a survey of diﬀerentially expressed genes, it emerged that two transcripts were annotated as glycogen synthase (i.e., transcripts DN12700_c0_g1_i1 and DN12700_c0_g1_i2): the ﬁrst was not diﬀerentially expressed, whereas the second was upregulated in larvae reared on VDM. Moreover, the unique transcript annotated as phosphoglucomutase (i.e., transcript DN12272_c0_g1_i) was not diﬀerentially expressed. Phosphoglucomutase controls the availability of glucose-1-phosphate, the precursor for glycogen synthesis, and is strongly downregulated in D. melanogaster mutants with reduced glycogen storage capacity [78]. On the other hand, although no transcripts annotated as glycogen phosphorylase were found, a number of transcripts that may correlate to a decrease in glycogen synthesis were downregulated in larvae reared on VDM. Indeed, 3 out of 4 serine/threonine kinase transcripts (i.e., enzymes that positively regulate glycogen synthase activity) were downregulated up to 4-fold, although the other one is upregulated. This not completely clear picture deriving from transcriptomic analysis may be due to the speciﬁc function of each midgut region, which, in turn, determines the diﬀerent glycogen accumulation in the three districts. Finally, we evaluated the accumulation of microelements in midgut cells of larvae grown on the two diets, focusing our attention on iron. Several studies demonstrated that this element is fundamental for the development of Diptera [79,80] due to its role as a cofactor for diﬀerent enzymes involved in crucial physiological functions [81–84]. Ferritin is the major protein responsible for iron storage in insects [85] and is constitutively expressed in iron cells [86]. In D. melanogaster larvae these peculiar cells are located between the middle and posterior midgut, in a tract called “iron region” [87]. However, the expression of ferritin is inducible in other midgut cells in the fruit ﬂy [86,88]. In fact, diets rich in iron induce the expression of ferritin encoding genes also in the anterior and posterior midgut cells, allowing iron accumulation in these regions [86,88]. In this study, we obtained comparable results. Cells able to accumulate iron are distributed in H. illucens larval midgut similarly to D. melanogaster and iron accumulation in these cells occurs regardless of the diet. By contrast, iron was accumulated also in the anterior and posterior midgut region only in larvae reared on SD, which is characterized by Int. J. Mol. Sci. 2020, 21, 4955 18 of 27 a higher iron content compared to VMD. In agreement with these data, ferritin encoding transcripts (4 out of 6) are downregulated in larvae reared on VMD. 4. Materials and Methods 4.1. Insect Rearing H. illucens larvae used in this work were derived from a colony established in 2015 at the University of Insubria (Varese, Italy). Two speciﬁc substrates, with diﬀerent nutrient composition, were used to rear the larvae: (i) Standard Diet (SD) for dipteran larvae [28], composed of 50% wheat bran, 30% corn meal, and 20% alfalfa meal, mixed in the ratio 1:1 dry matter:water; (ii) Vegetable Mix Diet (VMD), composed by fruits and vegetables (apple, banana, pear, broccoli, zucchini, potato, and carrot) mixed in equal quantity and appropriately minced (i.e., cut into small pieces of about 5 mm). VMD mimics fruit and vegetable waste, and ingredients available throughout the year were chosen in order to standardize the experimental conditions. The rearing methods of the larvae on both diets were previously described [27,89]. Larvae were maintained at 27.0 ± 0.5 C, 70 ± 5% relative humidity, in the dark. For all the experiments, last instar larvae were used. ◦ 4.2. Determination of Nutrient Content of the Diets The analyses were conducted at the Department of Agronomy, Food, Natural Resources, Animals and Environment, University of Padua (Agripolis, Legnaro, Italy). Three samples of fresh VMD were lyophilized with a freeze-dryer (Alpha 2-4 LD plus, Martin Christ GmbH, Osterode, Germany) under 12–15 mbar at –80 C, and then analyzed to determine nutrient content. Samples of SD powder were analyzed as they were. Diet samples were analyzed for crude protein, crude lipid, crude ﬁber, nitrogen-free extract, and ash following the protocols of AOAC International [90,91]. Hemicellulose, cellulose, and lignin content was calculated from the concentration of neutral detergent ﬁber, acid detergent ﬁber, and acid detergent lignin determined as reported elsewhere [92] and following the protocols of AOAC International [90,91]. Starch was determined by enzymatic digestion followed by glucose quantiﬁcation with HPLC. Free glucose and fructose were also quantiﬁed by HPLC. ◦ 4.3. Measurement of Larval Growth Rate Batches of 300 larvae were grown in plastic containers. Starting from the fourth day after hatching, 20 individuals were randomly sampled every two days, washed in lukewarm tap water to remove rearing substrate debris from the body, wiped dry, and weighed. The weight was recorded until 25% of insects reached pupal stage. The day in which the larvae reached the maximum weight was considered the end of the larval stage, subsequently the insect entered the prepupal stage and stopped feeding (deﬁnition and description of the developmental stages of H. illucens are reported elsewhere [74]). 4.4. pH of Diet and Midgut Lumen To evaluate pH of SD and VMD, 5 samples (1 g each) of fresh diets were placed in a plastic tube with 200 µL of distilled water. After mixing and short spinning, the liquid fraction was withdrawn and pH was measured by pH indicator strips with a resolution of 0.5 pH unit (Hydrion Brillant pH Dip Stiks, Sigma-Aldrich, Milano, Italy). pH of the midgut juice (obtained as described below, section “Isolation of midgut samples”) from anterior, middle, and posterior region of last instar larvae reared on the two diets was measured using the same pH indicator strips. The experiment was repeated on six independent samples obtained from larvae reared on SD and VMD. Int. J. Mol. Sci. 2020, 21, 4955 19 of 27 4.5. Isolation of Midgut Samples After anaesthetization on ice with CO2, larvae were dissected and the midgut was isolated in Phosphate Buﬀer Saline (PBS) (137 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.76 mM KH2PO4, pH = 7.4) at 4 C. ◦ For measurements of midgut lumen pH, enzymatic assays, morphological analyses, and Periodic Acid-Schiﬀ (PAS) staining, anterior, middle, and posterior midgut were collected, as previously reported [27]. In particular, for pH measurement and enzymatic assays (except for aminopeptidase N activity assay), the peritrophic matrix from diﬀerent midgut regions, with the enclosed intestinal content, was isolated, centrifuged at 15,000× g for 10 min at 4 C to remove the insoluble material, and supernatant (midgut juice) was collected. The midgut juice from 15 larvae was used as a fresh C for the enzymatic assays. The epithelium sample for the luminal pH measurements or stored at –80 of the posterior midgut, devoid of the peritrophic matrix and collected from 15 larvae, was placed into cryovials and stored in liquid nitrogen for aminopeptidase N activity assay. For morphological analyses and PAS staining, the three midgut regions were processed as described below (“Optical microscopy analysis of the midgut epithelium” and “Histochemical analysis of the larval midgut”). For Perls’ staining and transcriptome analysis, the whole midgut epithelium, with the enclosed midgut content, was isolated and processed as described below (“Histochemical analysis of the larval midgut” and “RNA isolation and Illumina sequencing”). ◦ ◦ 4.6. Enzymatic Assays Total proteolytic activity in midgut juice samples from anterior, middle, and posterior region, was assayed with azocasein (Sigma-Aldrich, Milano, Italy), measuring its degradation by release of azo chromophore [93], as previously reported [27]. For each midgut tract, the enzymatic assay was performed at pH as close as possible to that of the lumen (i.e., pH = 6.0 for the anterior midgut, pH = 5.0 for the middle midgut and pH = 8.5 for posterior). One unit (U) of total proteolytic activity with azocasein was deﬁned as the amount of enzyme that causes an increase in absorbance by 0.1 unit per min per mg of proteins. Chymotrypsin- and trypsin-like proteolytic activity in midgut juice samples were assayed with N-succinyl-Ala-Ala-Pro-Phe p-nitroanilide (SAAPPpNA, Sigma-Aldrich, Milano, Italy) and Nα-Benzoyl-d,l-arginine p-nitroanilide hydrochloride (BApNA, Sigma-Aldrich, Milano, Italy), respectively, measuring their degradation by release of p-nitroaniline (pNA) [27]. These assays were performed at pH = 8.5 on midgut juice obtained from the posterior region. One unit (U) of chymotrypsin- and trypsin-like proteolytic activity was deﬁned as the amount of enzyme that causes an increase in absorbance by 0.1 unit per min per mg of proteins. The activity of APN was assayed using l-leucine p-nitroanilide as substrate, measuring its degradation by release of pNA, as previously reported [27]. Assays were performed on the posterior midgut epithelium that, after thawing, was homogenized in 50 mM Tris-HCl, pH = 7.5 (1 mL/mg tissue). One unit (U) of APN activity was deﬁned as the amount of enzyme that releases 1 µmol of pNA per min per mg of proteins. α-amylase activity in midgut juice samples obtained from anterior, middle, and posterior midgut, was assayed with starch as substrate, measuring its hydrolysis by the amount of maltose released, as previously reported [27]. The assay was performed at pH = 6.9. One unit (U) of α-amylase activity was deﬁned as the amount of enzyme necessary to produce 1 mg of maltose per min per mg of proteins. Lipase activity in midgut juice samples obtained from anterior and posterior midgut was assayed using a Lipase Activity Colorimetric Assay Kit (BioVision, Milpitas, CA, USA) according to the manufacturer’s instructions. Int. J. Mol. Sci. 2020, 21, 4955 20 of 27 4.7. Statistical Analyses for pH Values, Larval Growth Parameters, and Enzymatic Activities Statistical analyses were performed with R statistical software (ver. 3.6.1) [94]. Paired and unpaired t-tests were done. Statistical diﬀerences between groups were considered signiﬁcant at p-value ≤ 0.05. The statistical analysis performed for each experiment and the p-values are reported in the captions to ﬁgures. 4.8. Optical Microscopy Analysis of the Midgut Epithelium The three regions of the larval midgut, with the enclosed intestinal content, were processed for morphological analysis as previously described [27]. Brieﬂy, after ﬁxation in glutaraldehyde 4% (v/v) in 0.1 M Na-cacodylate buﬀer, pH = 7.4, specimens were dehydrated in an increasing ethanol series and then embedded in epoxy resin (Epon/Araldite 812 mixture). Sections of 0.6-µm-thickness were obtained with a Leica Reichert Ultracut S (Leica, Wetzlar, Germany), stained with crystal violet and basic fuchsin, and then observed under Eclipse Ni-U microscope (Nikon, Tokyo, Japan) equipped with TrueChrome II S digital camera (Tucsen photonics, Fuzhou, China). 4.9. Histochemical Analysis of the Larval Midgut For glycogen detection, after dissection of the larva, the three regions of the midgut, with the enclosed intestinal content, were immediately ﬁxed in 4% (w/v) paraformaldehyde in PBS for 2 h at room temperature and then overnight at 4 C. After dehydration in increasing ethanol series, specimens were embedded in paraﬃn [72] and 7-µm-thick sections were obtained using a Jung Multicut 2045 microtome (Leica, Wetzlar, Germany). After deparaﬃnization, sections were stained with PAS kit (Bio-Optica, Milano, Italy), according to the manufacturer’s instructions, to detect the glycogen deposits in the midgut tissues, and then analyzed under Eclipse Ni-U microscope (Nikon, Tokyo, Japan) equipped with digital camera (Tucsen photonics, Fuzhou, China). ◦ For the detection of ferric iron, whole mount staining of the entire midgut was performed. After isolation, the tissue was ﬁxed in 4% (w/v) paraformaldehyde in PBS for 20 min, and then stained with Perls’ staining kit (Bio-Optica, Milano, Italy) according to the manufacturer’s instructions. Each region of the midgut was analyzed under NSZ-606 Zoom Stereo Microscope (Xiamen Phio Scientiﬁc Instruments, Xiamen, China) equipped with TrueChrome II S digital camera (Tucsen photonics, Fuzhou, China). 4.10. RNA Isolation and Illumina Sequencing Larvae were reared on SD and VMD (3 replicates for each diet), anesthetized on ice, and washed in 70% ethanol (v/v in water) before dissection under sterile conditions. Midguts were isolated in autoclaved PBS in a sterile Petri dish (5.5 × 1.3 cm). For both rearing substrates, pools of 10 midguts for each of the 3 experimental replicates (a total of 6 samples) were collected in a cryovial containing TRIzol reagent (Thermo Fisher Scientiﬁc, Waltham, MA, USA) and kept at –80 C until extraction of total RNA, which was performed according to the manufacturer’s instructions. Total RNA preparations were then treated with TURBO DNase I (Thermo Fisher Scientiﬁc, Waltham, MA, USA), according to the manufacturer’s instructions. Next Generation Sequencing (NGS) experiments, including samples quality control, were performed by Genomix4life S.R.L. (Salerno, Italy). RNA concentration in each sample was assayed with NanoDrop 1000 spectrophotometer (Thermo Fisher Scientiﬁc, Waltham, MA, USA) and its quality assessed with Agilent TapeStation 4200 (Agilent Technologies, Santa Clara, CA, USA). Indexed libraries were prepared from 1 µg of puriﬁed RNA from each sample with TruSeq Stranded mRNA Sample Prep Kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. Libraries were quantiﬁed using Agilent TapeStation 4200 (Agilent Technologies, Santa Clara, CA, USA) and pooled such that each index-tagged sample was present in equimolar amounts, with a 2 nM ﬁnal concentration of the pooled samples. The pooled samples were ◦ Int. J. Mol. Sci. 2020, 21, 4955 21 of 27 subjected to cluster generation and sequencing using an Illumina Nextseq 500 (Illumina, San Diego, CA, USA) in a 2 × 100 paired-end format at 1.8 pmol ﬁnal concentration. 4.11. De novo Transcriptome and Functional Annotation The raw sequence ﬁles (fastq ﬁles), generated as reported above for each of the six samples, underwent quality control analysis using FastQC [95]. Quality check on the raw sequencing data allowed to remove low quality sequences while preserving the high-quality part of NGS reads. Filtering was performed with BBDuk [96] by specifying a minimum length of 35 nucleotides (nt), and a sequencing quality of at least 35. Supplementary Table S8 reports sequencing output for the six samples before and after quality ﬁltering. The high-quality reads were normalized to reduce redundancy and then assembled using Trinity v2.1.1 [97]. The raw transcriptome was then ﬁltered as follows: (i) reads were mapped back on the transcriptome and transcripts with no expression, as determined using Kallisto [98], were removed; (ii) redundant sequences were merged using CD-HIT-EST [36]; and (iii) all the transcripts having a match to non-Arthropoda sequences were ﬁltered out. The quality of the assembly was estimated by (i) mapping the reads back to the assembly and (ii) using the BUSCO v3 pipeline [37] to get an estimate of the degree of completeness of the transcriptome assembly. In order to obtain the expression quantiﬁcation of the assembled transcripts in the 6 samples, Kallisto and the trimmed reads were used; abundance for all transcripts was expressed as TPM (number of transcripts per million); all the analyses in this study were based on this transcript abundance unit, after normalization using the trimmed mean of M-values formula [99]. The AHRD pipeline [38] was used to infer the putative function of the assembled transcripts. This method is based on a similarity search between the transcripts and a set of SwissProt databases [42] (Supplementary Table S2 in Supplementary ﬁle 1). For the analysis, the Arthropoda proteins were used; 48% of the input proteins were classiﬁed as “Unknown Protein”. The functional enrichment of the diﬀerent samples was calculated using GO annotation of transcripts [100] and the hypergeometric cumulative distribution. The enrichment p-values were adjusted for multiple testing using the Benjamini–Hochberg correction (Supplementary Table S2 in Supplementary ﬁle 1). 4.12. Diﬀerential Gene Expression Analysis The identiﬁcation of the diﬀerentially expressed genes was performed with the package NOISeq [101], the threshold for signiﬁcance used was FDR (false discovery rate) ≤ 0.05. A total of 4096 diﬀerential transcripts were found, of which 1671 were upregulated and 2425 downregulated (Supplementary Tables S3–S7 in Supplementary ﬁle 1). The subset of transcripts with at least a \|log2(fold change) \| ≥ 1 was considered (i.e., at least a 2-fold change, up or down, was considered of interest when statistically signiﬁcant). 4.13. Availability of Data The raw sequences are available at [102] with the study accession number ERP122672. 5. Conclusions The present work deﬁnitely demonstrates that the midgut of H. illucens larvae is able to adapt to diets with diﬀerent nutrient content and gives an important contribution to the ability of these insects to grow on a variety of feeding substrates. As a consequence of nutritionally poor diet, modiﬁcations of the digestive enzymatic machinery can be observed: (i) an increase in proteolytic activity and in the abundance of transcripts associated to serine-proteases, and (ii) a decrease in α-amylase and lipase activity and in abundance of related transcripts. Moreover, an increase in the length of microvilli of midgut cell ensures a higher absorbing surface and likely represents an adaptation to diet with a low nutritional content. Finally, a reduction of glycogen accumulation occurs. The overall picture suggests Int. J. Mol. Sci. 2020, 21, 4955 22 of 27 that midgut functions are regulated to modulate nutrient digestion and absorption with the aim of optimizing larval growth. The study of the biochemical and molecular mechanisms underlying midgut plasticity, at the midgut level (e.g., role of peptides secreted by midgut endocrine cells) and/or involving other tissues (e.g., fat body, and nervous system) will be pivotal to complete the scenario here unveiled. Moreover, since an alteration of midgut microbiota was observed in larvae reared on the two substrates used in this study [24], the contribution of microorganisms in midgut digestive processes and thus in larvae adaptation to diﬀerent substrates also represents a key issue to be investigated. Supplementary Materials: Supplementary Materials can be found at http://www.mdpi.com/1422-0067/21/14/ 4955/s1. Supplementary ﬁle 1 includes: Table S2. Functional annotation of the transcriptome by using SwissProt. Table S3. Functional annotation of the transcriptome using GO. Table S4. Expression levels of all transcripts in transcripts per million (TPM). Table S5. Diﬀerential expression analysis. Table S6. GO Enrichment analysis for downregulated transcripts. Table S7. GO Enrichment analysis for upregulated transcripts. Author Contributions: Conceptualization, S.C., M.C., G.T.; data curation, M.B. (Marco Bonelli), M.B. (Matteo Brilli), D.B., S.C., L.T.; formal analysis, M.B. (Marco Bonelli), M.B. (Matteo Brilli), D.B., L.T.; funding acquisition, M.C., G.T.; investigation, M.B. (Marco Bonelli), M.B. (Matteo Brilli), D.B., N.G.; methodology, S.C., M.C., G.T.; project administration, G.T.; supervision, M.C., G.T.; writing—original draft preparation, S.C., M.C., G.T. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by Fondazione Cariplo (grant n. 2014-0550) to M.C. and G.T. Acknowledgments: The authors acknowledge Ilaria di Lelio (University of Napoli Federico II) for her help in the preparation of samples for transcriptomic analysis. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations AHRD APN BApNA BP BSF P:C ratio FDR GO MF NGS PAS PBS pNA SAAPPpNA SD TPM VMD U References automatic assignment of Human Readable Descriptions pipeline aminopeptidase N Nα-Benzoyl-d,l-arginine p-nitroanilide hydrochloride biological processes black soldier ﬂy protein:carbohydrate ratio false discovery rate gene onthology molecular functions next generation sequencing periodic acid-Schiﬀ phosphate buﬀer saline p-nitroaniline N-succinyl-Ala–Ala-Pro-Phe p-nitroanilide standard diet number of transcripts per million vegetable mix diet unit of enzymatic activity 1. 2. 3. Clissold, F.J.; Tedder, B.J.; Conigrave, A.D.; Simpson, S.J. The gastrointestinal tract as a nutrient-balancing organ. Proc. R. Soc. B Biol. Sci. 2010, 277, 1751–1759. [CrossRef] [PubMed] Lee, K.P.; Simpson, S.J.; Clissold, F.J.; Brooks, R.C.; Ballard, J.W.O.; Taylor, P.W.; Soran, N.; Raubenheimer, D. Lifespan and reproduction in Drosophila: New insights from nutritional geometry. Proc. Natl. Acad. Sci. USA 2008, 105, 2498–2503. [CrossRef] Simpson, S.J.; Le Couteur, D.G.; Raubenheimer, D. Putting the Balance Back in Diet. Cell 2015, 161, 18–23. [CrossRef] [PubMed] Int. J. Mol. Sci. 2020, 21, 4955 23 of 27 8. 7. 4. 5. 6. Solon-Biet, S.M.; McMahon, A.C.; Ballard, J.W.O.; Ruohonen, K.; Wu, L.E.; Cogger, V.C.; Warren, A.; Huang, X.; Pichaud, N.; Melvin, R.G.; et al. The ratio of macronutrients, not caloric intake, dictates cardiometabolic health, aging, and longevity in ad libitum-fed mice. Cell Metab. 2014, 19, 418–430. [CrossRef] [PubMed] Simpson, S.J.; Raubenheimer, D. Macronutrient balance and lifespan. Aging 2009, 1, 875–880. [CrossRef] Barragan-Fonseca, K.B.; Dicke, M.; van Loon, J.J.A. Nutritional value of the black soldier ﬂy (Hermetia illucens L.) and its suitability as animal feed—A review. J. Insects Food Feed. 2017, 3, 105–120. [CrossRef] Raubenheimer, D.; Simpson, S.J. Nutritional ecology and foraging theory. Curr. Opin. Insect Sci. 2018, 27, 38–45. [CrossRef] Gold, M.; Tomberlin, J.K.; Diener, S.; Zurbrügg, C.; Mathys, A. Decomposition of biowaste macronutrients, microbes, and chemicals in black soldier ﬂy larval treatment: A review. Waste Manag. 2018, 82, 302–318. [CrossRef] Nguyen, T.T.X.; Tomberlin, J.K.; VanLaerhoven, S. Ability of black soldier ﬂy (Diptera: Stratiomyidae) larvae to recycle food waste. Environ. Entomol. 2015, 44, 406–410. [CrossRef] Jucker, C.; Erba, D.; Leonardi, M.G.; Lupi, D.; Savoldelli, S. Assessment of vegetable and fruit substrates as potential rearing media for Hermetia illucens (Diptera: Stratiomyidae) larvae. Environ. Entomol. 2017, 46, 1415–1423. [CrossRef] Spranghers, T.; Ottoboni, M.; Klootwijk, C.; Ovyn, A.; Deboosere, S.; De Meulenaer, B.; Michiels, J.; Eeckhout, M.; De Clercq, P.; De Smet, S. Nutritional composition of black soldier ﬂy (Hermetia illucens) prepupae reared on diﬀerent organic waste substrates. J. Sci. Food Agric. 2016, 97, 2594–2600. [CrossRef] 12. Wang, Y.S.; Shelomi, M. Review of black soldier ﬂy (Hermetia illucens) as animal feed and human food. Foods 11. 10. 9. 2017, 6, 91. [CrossRef] [PubMed] 13. Bava, L.; Jucker, C.; Gislon, G.; Lupi, D.; Savoldelli, S.; Zucali, M.; Colombini, S. Rearing of Hermetia illucens on diﬀerent organic by-products: Inﬂuence on growth, waste reduction, and environmental impact. Animals 2019, 9, 289. [CrossRef] 14. Wang, H.; Rehman, K.U.; Liu, X.; Yang, Q.; Zheng, L.; Li, W.; Cai, M.; Li, Q.; Zhang, J.; Yu, Z. Insect bioreﬁnery: A green approach for conversion of crop residues into biodiesel and protein. Biotechnol. Biofuels 2017, 10, 304. [CrossRef] [PubMed] 15. Gold, M.; Egger, J.; Scheidegger, A.; Zurbrügg, C.; Bruno, D.; Bonelli, M.; Tettamanti, G.; Casartelli, M.; Schmitt, E.; Kerkaert, B.; et al. Estimating black soldier ﬂy larvae biowaste conversion performance by simulation of midgut digestion. Waste Manag. 2020, 112, 40–51. [CrossRef] [PubMed] 16. Rumpold, B.A.; Klocke, M.; Schlüter, O. Insect biodiversity: Underutilized bioresource for sustainable applications in life sciences. Reg. Environ. Chang. 2016, 17, 1445–1454. [CrossRef] 17. Vogel, H.; Müller, A.; Heckel, D.G.; Gutzeit, H.; Vilcinskas, A. Nutritional immunology: Diversiﬁcation and diet-dependent expression of antimicrobial peptides in the black soldier ﬂy Hermetia illucens. Dev. Comp. Immunol. 2018, 78, 141–148. [CrossRef] 18. Liu, C.; Wang, C.; Yao, H. Comprehensive resource utilization of waste using the black soldier ﬂy (Hermetia illucens (L.)) (Diptera: Stratiomyidae). Animals 2019, 9, 349. [CrossRef] [PubMed] 19. Barbi, S.; Messori, M.; Manfredini, T.; Pini, M.; Montorsi, M. Rational design and characterization of bioplastics from Hermetia illucens prepupae proteins. Biopolymers 2018, 110, e23250. [CrossRef] 20. Cammack, J.A.; Tomberlin, J.K. The impact of diet protein and carbohydrate on select life-history traits of the black soldier ﬂy Hermetia illucens (L.) (Diptera: Stratiomyidae). Insects 2017, 8, 56. [CrossRef] [PubMed] 21. Pleissner, D.; Rumpold, B.A. Utilization of organic residues using heterotrophic microalgae and insects. 22. Waste Manag. 2018, 72, 227–239. [CrossRef] [PubMed] San Martin, D.; Ramos, S.; Zufía, J. Valorization of food waste to produce new raw materials for animal feed. Food Chem. 2016, 198, 68–74. [CrossRef] 23. Plazzotta, S.; Manzocco, L.; Nicoli, M.C. Fruit and vegetable waste management and the challenge of fresh-cut salad. Trends Food Sci. Technol. 2017, 63, 51–59. [CrossRef] 24. Bruno, D.; Bonelli, M.; De Filippis, F.; Di Lelio, I.; Tettamanti, G.; Casartelli, M.; Ercolini, D.; Caccia, S. The intestinal microbiota of Hermetia illucens larvae is aﬀected by diet and shows a diverse composition in the diﬀerent midgut regions. Appl. Environ. Microbiol. 2019, 85, 01864–18. [CrossRef] [PubMed] 25. Cappellozza, S.; Leonardi, M.G.; Savoldelli, S.; Carminati, D.; Rizzolo, A.; Cortellino, G.; Terova, G.; Moretto, E.; Badaile, A.; Concheri, G.; et al. A ﬁrst attempt to produce proteins from insects by means of a circular economy. Animals 2019, 9, 278. [CrossRef] [PubMed] Int. J. Mol. Sci. 2020, 21, 4955 24 of 27 26. Publications Oﬃce of the EU; European Commission. Commission Regulation (EU) 2017/893 of 24 May 2017 Amending Annexes I and IV to Regulation (EC) no 999/2001 of the European Parliament and of the Council and Annexes X, XIV and XV to Commission Regulation (EU) no 142/2011 as Regards the Provisions on Processed Animal Protein. Oﬀ. J. Eur. Union 2017, L138, 92–116. Available online: https://eur-lex.europa. eu/legal-content/EN/TXT/PDF/?uri=CELEX:32017R0893&from=EN (accessed on 7 July 2020). 27. Bonelli, M.; Bruno, D.; Caccia, S.; Sgambetterra, G.; Cappellozza, S.; Jucker, C.; Tettamanti, G.; Casartelli, M. Structural and functional characterization of Hermetia illucens larval midgut. Front. Physiol. 2019, 10, 204. [CrossRef] 28. Hogsette, J.A. New diets for production of house ﬂies and stable ﬂies (Diptera: Muscidae) in the laboratory. J. Econ. Entomol. 1992, 85, 2291–2294. [CrossRef] 29. Caccia, S.; Casartelli, M.; Tettamanti, G. The amazing complexity of insect midgut cells: Types, peculiarities, and functions. Cell Tissue Res. 2019, 377, 505–525. [CrossRef] [PubMed] 30. Callegari, M.; Marasco, R.; Jucker, C.; Mapelli, F.; Fusi, M.; Borin, S.; Daﬀonchio, D.; Savoldelli, S.; Crotti, E. Developmental stage and diet drive the bacterial community diversity in the food-waste reducing insect Hermetia illucens. In Proceedings of the 4th International Conference on Microbial Diversity “Drivers of Microbial diversity 2017”, Bari, Italy, 24–26 October 2017; pp. 239–242. 31. Terra, W.R.; Ferreira, C. Insect digestive enzymes: Properties, compartmentalization and function. Comp. Biochem. Physiol. Part B Comp. Biochem. 1994, 109, 1–62. [CrossRef] 32. Terra, W.R.; Ferreira, C.; Jordão, B.P.; Dillon, R.J. Digestive enzymes. In Biology of the Insect Midgut; Springer Science and Business Media LLC: Berlin, Germany, 1996; pp. 153–194. 33. Nichol, H.; Law, J.H.; Winzerling, J.J. Iron metabolism in insects. Annu. Rev. Entomol. 2002, 47, 535–559. [CrossRef] 34. Tang, X.; Zhou, B. Iron homeostasis in insects: Insights from Drosophila studies. IUBMB Life 2013, 65, 863–872. [CrossRef] 35. Mehta, A.; Deshpande, A.; Bettedi, L.; Missirlis, F. Ferritin accumulation under iron scarcity in Drosophila 36. iron cells. Biochimie 2009, 91, 1331–1334. [CrossRef] [PubMed] Fu, L.; Niu, B.; Zhu, Z.; Wu, S.; Li, W. CD-HIT: Accelerated for clustering the next-generation sequencing data. Bioinformatics 2012, 28, 3150–3152. [CrossRef] [PubMed] 37. Waterhouse, R.M.; Seppey, M.; Simão, F.A.; Manni, M.; Ioannidis, P.; Klioutchnikov, G.; Kriventseva, E.V.; Zdobnov, E.M. BUSCO applications from quality assessments to gene prediction and phylogenomics. Mol. Biol. Evol. 2017, 35, 543–548. [CrossRef] [PubMed] 38. GitHub. Available online: https://github.com/groupschoof/AHRD (accessed on 7 July 2020). 39. NCBI Assembly. Available online: https://www.ncbi.nlm.nih.gov/assembly/ (accessed on 7 July 2020). 40. Vicoso, B.; Bachtrog, D. Numerous transitions of sex chromosomes in Diptera. PLoS Biol. 2015, 13, e1002078. [CrossRef] [PubMed] 41. Zhan, S.; Fang, G.; Cai, M.; Kou, Z.; Xu, J.; Cao, Y.; Bai, L.; Zhang, Y.; Jiang, Y.; Luo, X.; et al. Genomic landscape and genetic manipulation of the black soldier ﬂy Hermetia illucens, a natural waste recycler. Cell Res. 2019, 30, 50–60. [CrossRef] [PubMed] 42. The UniProt Consortium. UniProt Consortium UniProt: A worldwide hub of protein knowledge. Nucleic Acids Res. 2018, 47, D506–D515. [CrossRef] 43. Ni, J.; Tokuda, G. Lignocellulose-degrading enzymes from termites and their symbiotic microbiota. Biotechnol. 44. Adv. 2013, 31, 838–850. [CrossRef] Saadeddin, A. The complexities of hydrolytic enzymes from the termite digestive system. Crit. Rev. Biotechnol. 2012, 34, 115–122. [CrossRef] 45. Lehane, M.; Blakemore, D.; Williams, S.; Moﬀatt, M.R. Regulation of digestive enzyme levels in insects. Comp. Biochem. Physiol. Part B Biochem. Mol. Biol. 1995, 110, 285–289. [CrossRef] 46. Lazarevi´c, J.; Jankovi´c-Tomani´c, M. Dietary and phylogenetic correlates of digestive trypsin activity in insect pests. Entomol. Exp. Appl. 2015, 157, 123–151. [CrossRef] 47. Broehan, G.; Kemper, M.; Driemeier, D.; Vogelpohl, I.; Merzendorfer, H. Cloning and expression analysis of midgut chymotrypsin-like proteinases in the tobacco hornworm. J. Insect Physiol. 2008, 54, 1243–1252. [CrossRef] Int. J. Mol. Sci. 2020, 21, 4955 25 of 27 48. Zhang, C.; Zhou, D.; Zheng, S.; Liu, L.; Tao, S.; Yang, L.; Hu, S.; Feng, Q. A chymotrypsin-like serine protease cDNA involved in food protein digestion in the common cutworm, Spodoptera litura: Cloning, characterization, developmental and induced expression patterns, and localization. J. Insect Physiol. 2010, 56, 788–799. [CrossRef] [PubMed] 49. Zhan, Q.; Zheng, S.; Feng, Q.; Liu, L. A midgut-speciﬁc chymotrypsin cDNA (Slctlp1) from Spodoptera litura: Cloning, characterization, localization and expression analysis. Arch. Insect Biochem. Physiol. 2010, 76, 130–143. [CrossRef] [PubMed] Spit, J.; Zels, S.; Dillen, S.; Holtof, M.; Wynant, N.; Broeck, J.V. Eﬀects of diﬀerent dietary conditions on the expression of trypsin- and chymotrypsin-like protease genes in the digestive system of the migratory locust, Locusta migratoria. Insect Biochem. Mol. Biol. 2014, 48, 100–109. [CrossRef] [PubMed] 50. 51. Borovsky, V. Biosynthesis and control of mosquito gut proteases. IUBMB Life 2003, 55, 435–441. [CrossRef] 52. [PubMed] Santiago, P.B.; de Araújo, C.N.; Motta, F.N.; Praça, Y.R.; Charneau, S.; Bastos, I.M.D.; Santana, J.M. Proteases of haematophagous arthropod vectors are involved in blood-feeding, yolk formation and immunity—A review. Parasit. Vectors 2017, 10, 79. [CrossRef] 53. Da Lage, J.L. The Amylases of Insects. Int. J. Insect Sci. 2018, 10, 1–14. [CrossRef] 54. Alfonso, J.; Ortego, F.; Sanchez-Monge, R.; García-Casado, G.; Pujol, M.; Castañera, P.; Salcedo, G. Wheat and barley inhibitors active towards α-Amylase and trypsin-like activities from Spodoptera frugiperda. J. Chem. Ecol. 1997, 23, 1729–1741. [CrossRef] 55. Kotkar, H.M.; Sarate, P.J.; Tamhane, V.A.; Gupta, V.; Giri, A.P. Responses of midgut amylases of Helicoverpa armigera to feeding on various host plants. J. Insect Physiol. 2009, 55, 663–670. [CrossRef] 56. Lwalaba, D.; Hoﬀmann, K.H.; Woodring, J. Control of the release of digestive enzymes in the larvae of the 57. fall armyworm, Spodoptera frugiperda. Arch. Insect Biochem. Physiol. 2009, 73, 14–29. [CrossRef] [PubMed] Sarate, P.; Tamhane, V.; Kotkar, H.; Ratnakaran, N.; Susan, N.; Gupta, V.; Giri, A.P. Developmental and digestive ﬂexibilities in the midgut of a polyphagous pest, the cotton bollworm, Helicoverpa armigera. J. Insect Sci. 2012, 12, 1–16. [CrossRef] [PubMed] 58. Hickey, D.A.; Benkel, B. Regulation of amylase activity in Drosophila melanogaster: Eﬀects of dietary 59. carbohydrate. Biochem. Genet. 1982, 20, 1117–1129. [CrossRef] Inomata, N.; Kanda, K.; Cariou, M.; Tachida, H.; Yamazaki, T. Evolution of the response patterns to dietary carbohydrates and the developmental diﬀerentiation of gene expression of α-amylase in Drosophila. J. Mol. Evol. 1995, 41, 1076–1084. [CrossRef] 60. Klarenberg, A.J.; Vermeulen, J.W.C.; Jacobs, P.J.M.; Scharloo, W. Genetic and dietary regulation of tissue-speciﬁc expression patterns of α-amylase in larvae of Drosophila melanogaster. Comp. Biochem. Physiol. Part B Comp. Biochem. 1988, 89, 143–146. [CrossRef] 61. Chng, W.A.; Sleiman, M.S.B.; Schüpfer, F.; Lemaitre, B. Transforming Growth Factor β/Activin signaling functions as a sugar-sensing feedback loop to regulate digestive enzyme expression. Cell Rep. 2014, 9, 336–348. [CrossRef] Sakai, T.; Satake, H.; Takeda, M. Nutrient-induced α-amylase and protease activity is regulated by crustacean cardioactive peptide (CCAP) in the cockroach midgut. Peptides 2006, 27, 2157–2164. [CrossRef] 62. 63. Weidlich, S.; Müller, S.; Hoﬀmann, K.H.; Woodring, J. Regulation of amylase, cellulase and chitinase secretion in the digestive tract of the two-spotted ﬁeld cricket, Gryllus bimaculatus. Arch. Insect Biochem. Physiol. 2013, 83, 69–85. [CrossRef] 64. Canavoso, L.E.; Jouni, Z.E.; Karnas, K.J.; Pennington, J.E.; Wells, M.A. Fat metabolism in insects. Annu. Rev. Nutr. 2001, 21, 23–46. [CrossRef] 65. Heier, C.; Kühnlein, R.P. Triacylglycerol metabolism in Drosophila melanogaster. Genetics 2018, 210, 1163–1184. [CrossRef] 66. Loidl, A.; Crailsheim, K. Free fatty acids digested from pollen and triolein in the honeybee (Apis mellifera carnica Pollmann) midgut. J. Comp. Physiol. B 2001, 171, 313–319. [CrossRef] [PubMed] 67. Christeller, J.T.; Amara, S.; Carrière, F. Galactolipase, phospholipase and triacylglycerol lipase activities in the midgut of six species of lepidopteran larvae feeding on diﬀerent lipid diets. J. Insect Physiol. 2011, 57, 1232–1239. [CrossRef] 68. Weidlich, S.; Hoﬀmann, K.H.; Woodring, J. Secretion of lipases in the digestive tract of the cricket Gryllus bimaculatus. Arch. Insect Biochem. Physiol. 2015, 90, 209–217. [CrossRef] [PubMed] Int. J. Mol. Sci. 2020, 21, 4955 26 of 27 69. Sutherland, P.W.; Burgess, E.P.J.; Philip, B.A.; McManus, M.T.; Watson, L.; Christeller, J.T. Ultrastructural changes to the midgut of the black ﬁeld cricket (Teleogryllus commodus) following ingestion of potato protease inhibitor II. J. Insect Physiol. 2002, 48, 327–336. [CrossRef] 70. Houk, E.I.; Hardy, J.L. Midgut cellular responses to bloodmeal digestion in the mosquito, Culex tarsalis Coquillett (Diptera: Culicidae). Int. J. Insect Morphol. Embryol. 1982, 11, 109–119. [CrossRef] 71. Rudin, W.; Hecker, H. Functional morphology of the midgut of Aedes aegypti L. (Insecta, Diptera) during 72. blood digestion. Cell Tissue Res. 1979, 200, 193–203. [CrossRef] Franzetti, E.; Romanelli, D.; Caccia, S.; Cappellozza, S.; Congiu, T.; Rajagopalan, M.; Grimaldi, A.; de Eguileor, M.; Casartelli, M.; Tettamanti, G. The midgut of the silkmoth Bombyx mori is able to recycle molecules derived from degeneration of the larval midgut epithelium. Cell Tissue Res. 2015, 361, 509–528. [CrossRef] 73. Yamada, T.; Habara, O.; Yoshii, Y.; Matsushita, R.; Kubo, H.; Nojima, Y.; Nishimura, T. The role of glycogen in development and adult ﬁtness in Drosophila. Development 2019, 146, dev176149. [CrossRef] 74. Bruno, D.; Bonelli, M.; Cadamuro, A.G.; Reguzzoni, M.; Grimaldi, A.; Casartelli, M.; Tettamanti, G. The digestive system of the adult Hermetia illucens (Diptera: Stratiomyidae): Morphological features and functional properties. Cell Tissue Res. 2019, 378, 221–238. [CrossRef] 75. Turunen, S.; Crailsheim, K. Lipid and sugar absorption. In Biology of the Insect Midgut; Springer Science and Business Media LLC: Berlin, Germany, 1996; pp. 293–320. 76. Buchon, N.; Osman, D.; David, F.P.; Fang, H.Y.; Boquete, J.P.; Deplancke, B.; Lemaitre, B. Morphological and molecular characterization of adult midgut compartmentalization in Drosophila. Cell Rep. 2013, 3, 1725–1738. [CrossRef] 77. Marianes, A.; Spradling, A.C. Physiological and stem cell compartmentalization within the Drosophila midgut. eLife 2013, 2, 00886. [CrossRef] 78. Ruaud, A.F.; Lam, G.; Thummel, C.S. The Drosophila NR4A nuclear receptor DHR38 regulates carbohydrate metabolism and glycogen storage. Mol. Endocrinol. 2011, 25, 83–91. [CrossRef] [PubMed] 79. Law, J.H. Insects, oxygen, and iron. Biochem. Biophys. Res. Commun. 2002, 292, 1191–1195. [CrossRef] [PubMed] 80. Missirlis, F.; Kosmidis, S.; Brody, T.; Mavrakis, M.; Holmberg, S.; Odenwald, W.F.; Skoulakis, E.M.C.; Rouault, T.A. Homeostatic mechanisms for iron storage revealed by genetic manipulations and live imaging of Drosophila ferritin. Genetics 2007, 177, 89–100. [CrossRef] [PubMed] 81. Clark, D.V. Molecular and genetic analyses of Drosophila prat, which encodes the ﬁrst enzyme of de novo purine biosynthesis. Genetics 1994, 136, 547–557. [PubMed] 82. Chávez, V.M.; Marqués, G.; Delbecque, J.P.; Kobayashi, K.; Hollingsworth, M.; Burr, J.E.; Natzle, J.; O’Connor, M.B. The Drosophila disembodied gene controls late embryonic morphogenesis and codes for a cytochrome P450 enzyme that regulates embryonic ecdysone levels. Development 2000, 127, 4115–4126. 83. Warren, J.T.; Petryk, A.; Marqués, G.; Jarcho, M.; Parvy, J.P.; Dauphin-Villemant, C.; O’Connor, M.B.; Gilbert, L.I. Molecular and biochemical characterization of two P450 enzymes in the ecdysteroidogenic pathway of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 2002, 99, 11043–11048. [CrossRef] 84. Navarro, J.A.; Ohmann, E.; Sanchez, D.; Botella, J.A.; Liebisch, G.; Moltó, M.D.; Ganfornina, M.D.; Schmitz, G.; Schneuwly, S. Altered lipid metabolism in a Drosophila model of Friedreich’s ataxia. Hum. Mol. Genet. 2010, 19, 2828–2840. [CrossRef] 85. Pham, D.Q.D.; Winzerling, J.J. Insect ferritins: Typical or atypical? Biochim. Biophys. Acta Gen. Subj. 2010, 1800, 824–833. [CrossRef] 86. Mandilaras, K.; Pathmanathan, T.; Missirlis, F. Iron absorption in Drosophila melanogaster. Nutrients 2013, 5, 87. 1622–1647. [CrossRef] Filshie, B.K.; Poulson, D.F.; Waterhouse, D.F. Ultrastructure of the copper-accumulating region of the Drosophila larval midgut. Tissue Cell 1971, 3, 77–102. [CrossRef] 88. Poulson, D.F.; Bowen, V.T. Organization and function of the inorganic constituents of nuclei. Exp. Cell Res. 1952, 2, 161–180. 89. Pimentel, A.C.; Montali, A.; Bruno, D.; Tettamanti, G. Metabolic adjustment of the larval fat body in Hermetia illucens to dietary conditions. J. Asia-Paciﬁc Entomol. 2017, 20, 1307–1313. [CrossRef] 90. Horwitz, W. AOAC International Oﬃcial Methods of Analysis of AOAC International, 17th ed.; The Association of Oﬃcial Analytical Chemists: Gaithersburg, MD, USA, 2000. Int. J. Mol. Sci. 2020, 21, 4955 27 of 27 91. Latimer, G.W. AOAC International Oﬃcial Methods of Analysis of AOAC International, 20th ed.; The Association of Oﬃcial Analytical Chemists: Gaithersburg, MD, USA, 2016. 92. Van Soest, P.V.; Robertson, J.B.; Lewis, B.A. Methods for dietary ﬁber, neutral detergent ﬁber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 1991, 74, 3583–3597. [CrossRef] 93. Charney, J.; Tomarelli, R.M. A colorimetric method for the determination of the proteolytic activity of duodenal juice. J. Biol. Chem. 1947, 171, 501–505. 94. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: 95. Vienna, Austria, 2019. Available online: https://www.R-project.org/ (accessed on 7 July 2020). FastQC. Available online: 7 July 2020). http://www.bioinformatics.babraham.ac.uk/projects/fastqc/ (accessed on 96. Bushnell, B. BBMap. 2015. Available online: https://sourceforge.net/projects/bbmap/ (accessed on 7 July 2020). 97. Grabherr, M.G.; Haas, B.J.; Yassour, M.; Levin, J.Z.; Thompson, D.A.; Amit, I.; Adiconis, X.; Fan, L.; Raychowdhury, R.; Zeng, Q.; et al. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat. Biotechnol. 2011, 29, 644–652. [CrossRef] 98. Bray, N.L.; Pimentel, H.; Melsted, P.; Pachter, L. Near-optimal probabilistic RNA-seq quantiﬁcation. Nat. Biotechnol. 2016, 34, 525–527. [CrossRef] [PubMed] 99. Robinson, M.D.; Oshlack, A. A scaling normalization method for diﬀerential expression analysis of RNA-seq data. Genome Biol. 2010, 11, R25. [CrossRef] 100. The Gene Ontology Consortium; Carbon, S.; Douglass, E.; Dunn, N.; Good, B.; Harris, N.L.E.; Lewis, S.; Mungall, C.J.; Basu, S.; Chisholm, R.L.; et al. The Gene Ontology Resource: 20 years and still GOing strong. Nucleic Acids Res 2018, 47, D330–D338. [CrossRef] 101. Tarazona, S.; Furió-Tarí, P.; Turrà, D.; Di Pietro, A.; Nueda, M.J.; Ferrer, A.; Conesa, A. Data quality aware analysis of diﬀerential expression in RNA-seq with NOISeq R/Bioc package. Nucleic Acids Res. 2015, 43, e140. [CrossRef] [PubMed] 102. ENA European Nucleotide Archive. Available online: https://www.ebi.ac.uk/ena (accessed on 7 July 2020). © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.31557_apjcp.2020.21.1.157	RESEARCH ARTICLE Editorial Process: Submission:09/05/2019 Acceptance:11/09/2019 Expression of Ki-67 and Βeta-Catenin in Pseudoepitheliomatous Hyperplasia and Squamous Cell Carcinoma in Oral Mucosal Biopsies : An Immunohistochemical Study Bismah Ahmad1, Mohammad Asif1, Anam Ali1, Shahid Jamal2, Muhammad Zaib Khan3, Mohammad Tahir Khadim1 Abstract Objective: To examine the immunohistochemical expression of Ki-67 and beta-catenin in pseudoepitheliomatous hyperplasia and squamous cell carcinoma (SCC) in oral mucosal biopsies. Methods: In this comparative cross sectional study, 70 cases of each PEH and OSCC were taken from the patients of both genders and in all age groups. Study was conducted at Armed Forces Institute of Pathology (AFIP), Rawalpindi from Dec 2017 to March 2019. Statistical analysis was done with the help of SPSS Version 24.0. We used Chi-Squared test with p value of < 0.05 which was considered as statistically significant. Results: In the current study, 80 (57.1%) male and 60 (42.8%) female patients with the mean age of 51.69 ± 16.121 (mean ± SD) years were included. It was found that 6-25% Ki-67 labeling index was observed in all (70) PEH cases, which involved only basal layer of the epithelium. Whereas, Ki-67 labeling index was highly expressed in tumor of high grade malignancy than tumor of low grade malignancy. On the other hand, expression of membranous beta-catenin was higher in PEH and cytoplasmic beta-catenin expression was higher in OSCC. Conclusion : It is concluded that Ki-67 and beta-catenin showed significant expression in PEH and OSCC in oral mucosal biopsies especially those with intense inflammation or unoriented tissue, helping the clinicians to arrive at a final diagnosis before planning any surgical intervention. Keywords: Squamous cell carcinoma- Pseudoepitheliomatous hyperplasia- Beta-catenin- Ki-67 Asian Pac J Cancer Prev, 21 (1), 157-161 Introduction Oral cancer is an important problem of global public health, constituting approximately 5% of all cancers (Vig et al., 2015). Out of this 5%, almost 90% constitutes oral squamous cell carcinoma. In Pakistan, it accounts for 10% of all malignancies (Wahid et al., 2005). According to Pakistan Medical Research Council (PMRC), OSCC accounts for the majority of the cancers among males than in females because of heavier consumption of smoked tobacco and alcohol in males with tongue and lips are the most frequent occupied sites (Jamal et al., 2006). The major risk factor for OSCC is use of tobacco which may be in the form of cigarettes, cigars, pipes and chewing betel nuts. Other risk factors include heavy consumption of alcohol, nutritional deficiencies, sepsis, genetic mutation involving p53 gene, human papillomavirus (HPV) infection, Epstein-Barr virus (EBV) infection and poor dental hygiene (Markopolous, 2012). These risk factors leads to changes in epithelium of oral mucosa causing transformation of normal keratinocytes into abnormal keratinocytes, squamous cell atypical and invasive features (Feller et al., 2013). O n e o f t h e b e n i g n m i m i c s o f O S C C i s pseudoepitheliomatous hyperplasia (PEH) which is defined as a histopathological response to numerous stimuli, such as shock, disease, irritation and neoplasia (Zayour and Lazova, 2011). Histologically, it is seen as tongue like epithelial propagations penetrating into connective tissue (Zarovnaya and Black, 2005). When the mucosal biopsies are of small size, ulcerated, unoriented and inflamed or when the profound rete pegs are cut obliquely, it become extremely difficult to differentiate between PEH and OSCC. Hence, it is important to differentiate between PEH and OSCC as the whole treatment plan of the surgeons and oncologists relies on correct diagnosis (Nayak et al., 2015). Ki-67 is a monoclonal antibody for proliferation. It is strictly associated with cell proliferation and aggressiveness of malignant tumours (Dadfarnia et 1Department of Histopathology, Armed forces Institute of Pathology (AFIP), 2Department of Histopathology, Watim Medical and Dental College, 3Department of Endodontics , Margalla Institute of Health Sciences, Rawalpindi, Pakistan. For Correspondence: [email protected] 157 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.1.157 Ki-67 and β-Catenin in PEH and OSCCal., 2012). During cell cycle, the presence of Ki-67 protein is seen during prophase and metaphase (G1, S, G2 and M phase) but is absent during resting cells (G0) (Scholzen et al., 2000). Whereas; beta-catenin is a subunit of cadherin protein complex. beta-catenin is a 92 kDa protein normally found in cytoplasm of the cell in sub-membranous location which participate in regulation and organization of cell–cell adhesion and transcriptional regulator (Liu at al., 2010). Mutation or overexpression in the beta-catenin gene results in nuclear accumulation of the protein. Reduced expression of membranous beta-catenin and amplified expression of cytoplasmic / nuclear beta- catenin has been associated with increase grading and invasiveness of tumor (Rosado et al., 2013). So far, the expression of Ki-67 and beta-catenin protein has not been completely explained in human PEH and OSCC samples by immunohistochemistry (IHC). The aim of this study was to examine the expression of Ki-67 and beta-catenin by IHC in paraffin-embedded oral mucosal tissues exhibiting PEH and OSCC. Marterials and Methods An institutional ethical committee approval (IRB, AFIP) (letter no. MP-ORP16-10/READ-IRB/17/394) was taken before the start of the study. A total of 140 oral mucosal biopsy samples, 70 cases each of PEH and OSCC, along with normal mucosal biopsy as control group were collected from Armed Force Institute of Pathology (AFIP), Rawalpindi, Pakistan from December 2017 to March 2019. Along with clinical histories of each case, the data of age, gender and site were noted. Immunohistochemistry was performed on 140 mucosal biopsies along with adjacent controls. Scanty and poorly fixed specimens were excluded from the study. Antibodies used for IHC were: monoclonal rabbit Ki-67 (clone no. EP5, catalogue no. BSB 5709, ready to use) from Bio SB (Santa Barbara, CA, USA); and monoclonal mouse beta-catenin (clone no. 17C2, catalogue no. PA0083, ready to use) from Leica Bio system (Newcastle, UK). Results were interpreted on light microscope (Binocular Olympus (Tokyo, Japan) Model CX-21) using high power field objective (10x, 40x) and further counter checked by consultant pathologist. Evaluation of Ki-67 and beta-catenin was performed as follow: The intensity of brown colored nuclear Ki-67 staining which is confined to spinous layer or both basal and parabasal layer or only basal layer of epithelium is graded as (Humayun et al., 2011) : Mild……. +light brown color Moderate……. ++ dark brown color Severe……… +++ very dark brown color The pattern of staining was assessed in percentage by calculating the positive cells per 100 basal cells, parabasal or spinous cell layer of epithelium. The percentage of positive cells or labeling index (LI) was as follows (Humayun et al., 2011): Negative = 0-5% basal layer staining = 6-25% 158 basal and parabasal layer staining = 26-60% basal, parabasal and spinous layer staining = 61-99%. On the other hand, beta-catenin positive cell is defined as brown staining of nucleus, cytoplasm or cell membrane of epithelium and is expressed in the form of immunoreactivity as follows (Zaid et al., 2015): The proportion score (P) was interpreted as: 0……………. Negative 1……………. <10% positive cells 2……………. 10-50% positive cells 3……………. 50-80% positive cells 4……………. >80% positive cells The intensity score (I) was defined as: 0……………. Negative / No staining 1……………. Weak staining 2……………. Moderate staining 3……………. Strong staining Immunoreactive combined score (T) = proportion score (P) x intensity score (I). Immunoreactivity scores (T) of beta-catenin were categorized into three groups based on the final score: Score 0……………. Negative Score 1-4……………. Weak Staining Score > 4……………. Strong Staining Results In a total of 140 cases, 70 (50%) cases showed positive stained cells ranging from 6-25% Ki -67 labeling index. Whereas, 24 (17.1%) and 46 (32.9%) cases showed positive stained cells ranging from 26-50% and 51-99% Ki -67 labeling index respectively (Table 1). Among 70 cases of PEH, all 70 (100%) cases showed positive staining with Ki-67 antibody in the basal layer of the epithelium. The labelling index of Ki-67 for PEH is 6-25%. Out of 70 cases of OSCC, 18 (45%) and 22 (55%) cases of well-differentiated OSCC (WDOSSC) showed Ki-67 labelling index of 26-50% and 51-99 % respectively. No staining pattern of WDOSCC was observed between Ki-67 labelling index of 6-25%. Similarly, 6 (22%) and 21 (77.8%) cases of moderately differentiated OSCC (MDOSCC) showed Ki-67 labelling index of 26-50% in cases and 51-99 % respectively. While, no staining pattern of MDOSCC was observed between Ki-67 labelling indexes of 6-25%. Interestingly, poorly differentiated OSCC (PDOSCC) showed only 51- 99 % Ki-67 labelling index in 3(100%) cases. Whereas, no staining pattern was observed between index of 6-25% and 26-50% in PDOSCC (Figure 1). Among 140 cases, score 0 was not found in any single case. Whereas, 52(37.1%) cases showed weak staining and 88(62.9%) cases showed strong staining (table 2). Out of 140 cases (Table 3), 78 cases of the oral mucosal biopsies exhibited membranous expression of β- catenin in which 7.7% of membranous β- catenin was expressed in WDOSCC, 2.6% in MDOSCC, 0% in PDOSCC. In PEH, 87.7% of membranous β- catenin was expressed. Twenty-five (17.9%) cases of the epithelial cells exhibited cytoplasmic expression of β- catenin in which 56% of cytoplasmic β- catenin expression was seen in Bismah Ahmad et alAsian Pacific Journal of Cancer Prevention, Vol 21Figure 1. Expression of Ki-67 in Different Oral Mucosal Regions; (A) Ki -67 in PEH showing nuclear staining limited to basal layer (red arrow); (B) Ki-67 in WDOSCC showing 26-50% of dispersed positive nuclei; (C) Ki-67 in PDOSCC showing numerous intense (51-99%) scattered positive nuclei (immunohistochemistry staining, original magnification 400x). Table 1. Expression of Ki-67 in OSCC and PEH (n = 140). Ki-67 expression ( 6 - 25%) n =70 ( 26 - 50%) n =24 ( 51 - 99%) n =46 Total n = 140(%) Histological Diagnosis PEH n (%) WDOSCC n (%) MDOSCC n (%) PDOSCC n (%) 70 (100%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 18 (45%) 6 (22.2%) 0 (0%) 0 (0%) 22 (55%) 21 (77.8%) 3 (100%) 70 (50%) 40 (28.6%) 27 (19.3%) 3 (2.1%) WDOSCC, 44 % in MDOSCC, 0% in PDOSCC and PEH. Only 1 case of the PDOSCC exhibited nuclear staining of β- catenin. But since, in this case, a little bit of cytoplasmic staining was also found, so, this case was included in cytoplasmic/nuclear expression. Eighteen cases of the oral mucosal biopsies exhibited membranous / cytoplasmic expression of β- catenin in which 64.3% of membranous / cytoplasmic β- catenin was expressed in WDOSCC, and 25.7% in MDOSCC. No membranous β- catenin expression found in PDOSCC and PEH. Nine cases of the oral mucosal biopsies exhibited cytoplasmic / nuclear expression of β- catenin in which 12.5% of cytoplasmic / nuclear β- catenin was Figure 2. Expression of Beta-Catenin in Different Oral Mucosal regions; (A), Beta - catenin in PEH showing intense membranous staining (red arrow) throughout the epithelium; (B), Beta - catenin in WDOSSC showing membranous (black arrow) and cytoplasmic (red arrow) staining; (C)Beta-catenin in PDOSCC showing dispersed altered cytoplasmic (black arrow) and nuclear staining (red arrow) throughout the cancerous islands (immunohistochemistry staining, original magnification 400x). Table 2. Patterns of Beta-Catenin Expression in PEH and OSCC (n = 140). Pattern of Beta-catenin Membranous Cytoplasmic Membranous / Cytoplasmic Cytoplasmic / Nuclear Histological Diagnosis PEH n (%) 70 (89.7%) 0 (0%) 0 (0%) 0 (0%) WDOSCC n (%) 6 (7.7%) 14 (56%) 18 (64.3%) 1 (12.5%) MDOSCC n (%) 2 (2.6%) 11 (44%) 10 (35.7%) 5 (50%) PDOSCC n (%) 0 (0%) 0 (0%) 0 (0%) 3 (37.5%) Total (140) 78 25 28 9 159 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.1.157 Ki-67 and β-Catenin in PEH and OSCCexpressed in WDOSCC, and 50% in MDOSCC. 37.5% in PDOSCC. In PEH, no cytoplasmic / nuclear β- catenin was expressed (Fig. 2). So, the statistically significant relation was observed between pattern of β-catenin expression and the histological diagnosis i.e. OSCC and PEH of p = .000 (p<0.05) . Discussion Oral squamous cell carcinoma is the most frequent malignant neoplasm of oral region (Vig, 2015). There are a lot of structural features which help in differentiating oral squamous cell carcinoma from pseudoepitheliomatous hyperplasia on H / E sections. But, there has always been argument whether the mucosal biopsies are diagnosed as OSCC or PEH, when they are of small size, ulcerated, unoriented and inflamed or when the profound rete pegs are cut obliquely, as this will lead to unnecessary removal of tissue or additional therapy such as chemo radiotherapy (Zarovnaya and Black, 2005). Current study investigated the immunohistochemical expression of Ki-67 proliferative marker and beta-catenin on pseudoepitheliomatous hyperplasia and PEH. In our study, PEH showed positive staining with Ki-67 antibody in the basal layer of the epithelium with a labelling index of 6-25% while in most of the cases of OSCC; the entire epithelium showed continuous and prominent Ki-67 labelling. The results are similar to the study made by Humayun and Prasad, (2011). which showed that Ki-67 labelling index in WDOSCC is low (28 ± 20.82) while in MDOSCC and PDOSCC a significantly higher number of Ki-67 positive cases with a mean of 47 ± 6.3 ( mean ± SD ) and 60.3 ± 2.8 ( mean ± SD ) respectively were seen. Similarly, in another study by Ashraf and his co-workers (Ashraf et al., 2010) studied 54 patients showing Ki- 67 staining increases as the tumor leads towards its severe stage. Moreover, in a study of Finland by Luukkaa showed that the Ki-67 and p53 as tumor markers and revealed that in PDOSCC, Ki-67 expression was more diffuse and intense as the cells were less differentiated than WDOSCC and MDOSCC (Luukkaa et al., 2006). Similar finding have been seen in our study in which all PDOSCC cases showed 51-99% Ki-67 expression range implying that high grade tumor showed great proliferative activity which has a poor prognosis. On the other hand, beta- catenin is a protein normally found in cytoplasm of the cell in sub-membranous location which participate in regulation and organization of cell–cell adhesion and transcriptional regulator (Purcell et al., 2011). In a research carried out in China, Jiang and his fellows (Jiang et al., 2004) showed that increased membranous appearance of beta-catenin antibody was seen in pseudoepitheliomatous hyperplasia exhibiting good prediction of recovery- of the patient. Likewise, in the study by purcell (Purcell et al., 2011) illustrated that when there was increased expression of cytoplasmic/ nuclear beta-catenin in squamous cell carcinoma of pharynx, the prognosis is poor. This was parallel to the present study in which all PDOSCC cases showed cytoplasmic/nuclear staining. 160 In the present study, number of positive cases of beta-catenin and its intensity decreased as the cancer differentiation become poor. In our study, nine number of cases of the oral mucosal biopsies showed cytoplasmic/nuclear beta-catenin expression with all PDOSCC cases having low immunoreactivity score of 1-4. While, 78 number of cases showed membranous beta-catenin expression with all PEH cases having high immunoreactivity score of > 4. Our results are concordant to those of previous studies performed by Yu et al., (2005), who compared expression of beta-catenin in normal tissue to cancerous tissue and concluded that in regular ordinary tissue beta-catenin expression was high while in cancerous tissue its expression was low showing poor prognosis. Our current results are parallel to those of Japanese pathologist, (Yun et al., 2010), who studied high Ki-67 expression and low beta-catenin expression which is significantly linked with poorly and moderately differentiated carcinoma (p<0.05). A point noteworthy is that the sample size of the current study is huge making the results more reliable and statistically significant. On the basis of our study, the present results can be interpreted that in OSCC, the entire thickness of epithelium showed nuclear staining with Ki-67, while in PEH there was increased nuclear staining only in the basal layer of the epithelium. On the other hand, number of positive cases and immunoreactive score of beta-catenin decreased as the grade of oral squamous cell carcinoma increases. Hence, we concluded that opposite expression of Ki-67 and beta-catenin play a vital role in forecasting histological grades of differentiation in oral mucosal biopsies and prognosis of the lesion. Therefore, helping clinicians to arrive at a final diagnosis before planning any surgical management. But because of small tissue estimate, the utility of this panel is constrained. One of the limitations of this study is that the size of mucosal biopsies is so tiny or small, if we want more sections of the tissue for further study; it is difficult to choose if the sufficient tissue is cleared out in the block. For this purpose, we had also learnt from our present study to slice the tissue fragment for immunohistochemical analysis at the same time when routine sections were sliced so that there is reduce shaving and we have more left over tissue. One of the greatest limitations of our present study is that the sample size of PDOSCC was scarce owing to the patient turnout in our set up. No follow ups were possible owing to the short duration of present study. Acknowledgements I am also very thankful to Dr. Prof. Mian Hammad Nazir, Research Officer, University of Wales, Cardiff, Wales, UK for his valuable time and professional guidance in assembling my results. I am sincerely indebted to the staff of Histopathology Department, AFIP for helping me out during my training and my research work and providing me a friendly research environment. Bismah Ahmad et alAsian Pacific Journal of Cancer Prevention, Vol 21Statement Conflict of Interest The authors have no affiliation and have No conflict of interest (personal, commercial, political, academic or financial interest). Neither NO financial interest which may include employment, research funding, stock or share ownership, payment for lecture or travel, consultancies and company support for staff exists. Funding partially funded by university which is National University of Medical Sciences (NUMS) and partially self- funded. Approved By Any Scientific Body An institutional ethical committee approval (IRB, AFIP) (letter no. MP-ORP16-10/READ-IRB/17/394) was taken before the start of the study. Later on, defence of thesis was taken and approved by external and internal supervisor appointed by the university after correction of minor mistakes. Thus, this article is the part of approved thesis with the same title. How The Ethical Issue Was Handled By collecting as much information as possible, discussing the problems with supervisor very often , human subject protection and avoid copying others work. References Ashraf MJ, Maghbul M, Azarpira N, Khademi B (2010). Expression of Ki67 and P53 in primary squamous cell carcinoma of the larynx. Indian J Pathol Microbiol, 53, 661-5. Dadfarnia T, Mohammed BS, Eltorky MA (2012). Significance of Ki-67 and p53 immunoexpression in the differential diagnosis of oral necrotizing sialometaplasia and squamous cell carcinoma. Ann Diagn Pathol, 16, 171–6. Feller LL, Khammisa RR, Kramer BB, Lemmer JJ (2013). Oral squamous cell carcinoma in relation to field precancerisation: pathobiology. Cancer Cell Int, 13, 31. Humayun S, Prasad VR (2011). Expression of p53 protein and ki-67 antigen in oral premalignant lesions and oral squamous cell carcinomas: An immunohistochemical study. Natl J Maxillofac Surg, 2, 38-46. Jamal S, Mamoon N, Mushtaq S, Luqman M (2006). Oral cancer: a clinicopathological analysis of 723 cases. Pak Armed Forces Med J(PAFMJ), 56, 295-9. Jiang DY, Fu XB, Sheng ZY, et al (2004). Study of mechanism on loss of some components from basement membrane in epithelial-interstitial junction in cutaneous pseudoepitheliomatous hyperplasia lesion. Pediatr Crit Care Med, 16, 36-41. Liu LK, Jiang XY, Zhou XX, et al (2010). Upregulation of vimentin and aberrant expression of e-cadherin/β-catenin complex in oral squamous cell carcinomas: correlation with the clinicopathological features and patient outcome. Mod Pathol, 23, 213-24. Luukkaa H, Klemi P, Levivo I, Vahlberg T, Grenman R (2006). Prognostic significance of Ki-67 and p53 as tumor markers in salivary gland malignancies in Finland: An evaluation of 212 cases. Acta Oncol (Madr), 45, 669-75. Markopoulos KA (2012). Current aspects on oral squamous cell carcinoma. Open Dent J, 6, 126-30. Nayak VN, Uma K, Girish H, et al (2015). Pseudoepitheliomatous Hyperplasia in Oral Lesions: A Review. J Int Oral Heal, 7, 148-52. Purcell R, Childs M, Maibach R, et al (2011). HGF/c-Met related activation of β-catenin in hepatoblastoma. J Exp Clin Cancer Res, 30, 96. Rosado p, Lequeric-Fernandez P, Fernandez S, et al (2013). E-cadherin and β-catenin expression in well-differentiated and moderately-differentiated oral squamous cell carcinoma: Relations with clinical variables. Br J Oral Maxillofac Surg, 51, 149-56. Scholzen T, Gerdes J (2000). The ki-67 protein: from the known and the unknown. J Cell Physiol, 182, 311-22. Vig N, Mackenzie IC, Biddle A (2015). Phenotypic plasticity and epithelial to mesenchymal transition in the behaviour and therapeutic response of oral squamous cell carcinoma. J Oral Pathol Med, 44, 649-55. Wahid A, Ahmad S, Sajid M (2005). Pattern of carcinoma of oral cavity presenting at Dental Department of Ayub Medical College. J Ayub Med Coll Abbottabad, 17, 65-6. Yu Z, Wein berger PM, Provost E, et al (2005). Beta-Catenin functions mainly as an adhesion molecule in patients with squamous cell cancer of the head and neck. Clin Cancer Res, 11, 2471-77. Yun X, Wang L, Cao L, Okada N, Miki Y (2010) Immunohistochemical study of β-catenin and functionally related molecular markers in tongue squamous cell carcinoma and its correlation with cellular proliferation. Oncol Lett, 1, 437-43. Zaid KW (2014). Immunohistochemical assessment of E-cadherin and beta-catenin in the histological differentiations of oral squamous cell carcinoma. Asian Pac J Cancer Prev, 15, 8847-53. Z a r o v n a y a E , B l a c k C ( 2 0 0 5 ) . D i s t i n g u i s h i n g pseudoepitheliomatous hyperplasia from squamous cell carcinoma in mucosal biopsy specimens from the head and neck. Arch Pathol Lab Med, 129, 1032-36. Zayour M, Lazova R (2011). Pseudoepitheliomatous hyperplasia: A review. Am J Dermatopathol, 33, 112-16. This work is licensed under a Creative Commons Attribution- Non Commercial 4.0 International License. 161 Asian Pacific Journal of Cancer Prevention, Vol 21DOI:10.31557/APJCP.2020.21.1.157 Ki-67 and β-Catenin in PEH and OSCC
10.3390_ijms24129783	Article Casein Kinase 2 Alpha Inhibition Protects against Sepsis-Induced Acute Kidney Injury Jeung-Hyun Koo 1,†, Hwang Chan Yu 1,†, Seonhwa Nam 2, Dong-Chan Kim 2 and Jun Ho Lee 2,3,* 1 Department of Biochemistry and Molecular Biology, Jeonbuk National University Medical School, Jeonju 54896, Republic of Korea; [email protected] (J.-H.K.); [email protected] (H.C.Y.) 2 Department of Anesthesiology and Pain Medicine, Jeonbuk National University Medical School and Hospital, Jeonju 54896, Republic of Korea; [email protected] (S.N.); [email protected] (D.-C.K.) 3 Research Institute of Clinical Medicine of Jeonbuk National University-Biomedical Research Institute of Jeonbuk National University Hospital, Jeonju 54896, Republic of Korea * Correspondence: [email protected] † These authors contributed equally to this work. Abstract: Sepsis-induced acute kidney injury (AKI) is a common complication in critically ill patients, often resulting in high rates of morbidity and mortality. Previous studies have demonstrated the effectiveness of casein kinase 2 alpha (CK2α) inhibition in ameliorating ischemia–reperfusion-induced AKI. In this study, our aim was to investigate the potential of the selective CK2α inhibitor, 4,5,6,7- tetrabromobenzotriazole (TBBt), in the context of sepsis-induced AKI. To assess this, we initially conﬁrmed an upregulation of CK2α expression following a cecum ligation and puncture (CLP) procedure in mice. Subsequently, TBBt was administered to a group of mice prior to CLP, and their outcomes were compared to those of sham mice. The results revealed that, following CLP, the mice exhibited typical sepsis-associated patterns of AKI, characterized by reduced renal function (evidenced by elevated blood urea nitrogen and creatinine levels), renal damage, and inﬂammation (indicated by increased tubular injury score, pro-inﬂammatory cytokine levels, and apoptosis index). However, mice treated with TBBt demonstrated fewer of these changes, and their renal function and architecture remained comparable to that of the sham mice. The anti-inﬂammatory and anti-apoptotic properties of TBBt are believed to be associated with the inactivation of the mitogen-activated protein kinase (MAPK) and nuclear factor κB (NF-κB) signaling pathways. In conclusion, these ﬁndings suggest that inhibiting CK2α could be a promising therapeutic strategy for treating sepsis- induced AKI. Keywords: sepsis; acute kidney injury; CK2α; TBBt; NF-κB 1. Introduction Sepsis, a severe systemic inﬂammatory disease with high mortality, is caused by dysregulated host response to bacterial infection [1]. Multiple organ failure is a major complication that leads to high mortality in patients with sepsis [2]. When acute kidney injury (AKI) occurs during or after sepsis, it increases the mortality rate by approximately 70% [3]. Of note, levels of circulating inﬂammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin (IL)-6 are positively associated with the increased risk of mortality in AKI patients [4,5]. It is widely accepted that lipopolysaccharide (LPS) released from Gram-negative bacteria binds to Toll-like receptor 4 (TLR4) and transduces signals to activate several signaling pathways, including mitogen-activated protein kinases (MAPKs) and nuclear factor κB (NF-κB). These pathways ultimately increase transcription of pro-inﬂammatory cytokines and other inﬂammatory mediators [6]. Despite the growing understanding of the pathophysiological processes of sepsis-induced AKI, however, the treatment of septic AKI has still been unsatisfactory. Citation: Koo, J.-H.; Yu, H.C.; Nam, S.; Kim, D.-C.; Lee, J.H. Casein Kinase 2 Alpha Inhibition Protects against Sepsis-Induced Acute Kidney Injury. Int. J. Mol. Sci. 2023, 24, 9783. https://doi.org/10.3390/ijms24129783 Academic Editor: Manoocher Soleimani Received: 23 April 2023 Revised: 27 May 2023 Accepted: 1 June 2023 Published: 6 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Mol. Sci. 2023, 24, 9783. https://doi.org/10.3390/ijms24129783 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 9783 2 of 11 Casein kinase 2 (CK2), also known as protein kinase 2, is a protein serine/threonine kinase; it is a tetrameric enzyme composed of two catalytic subunits (α and/or α(cid:48)) and two regulatory beta subunits [7]. A large number of studies show that CK2 has been found to phosphorylate various transcription factors regulating the inﬂammatory response, including NF-κB [8]. Speciﬁcally, CK2 phosphorylates inhibitory κB (IκB)α and degrades it through the proteasomal pathway [9]. In addition to IκBα, CK2 also directly phosphorylates p65, thereby amplifying its transcriptional activity [10]. Overall, inhibition of CK2 activity implicates attenuating inﬂammation and cytokine signaling via the suppression of NF-κB. Over the past two decades, numerous inhibitors of CK2, such as emodin, apigenin, and 4,5,6,7-tetrabromobenzotriazole (TBBt) have been discovered and developed. Among these inhibitors, TBBt is the most effective cell-permeant inhibitor of CK2. TBBt selectivity is obtained via a hydrophobic pocket adjacent to the ATP/GTP binding site, which is smaller in CK2 than in other protein kinases [11]. Ka et al. previously showed that TBBt is useful to ameliorate ischemia–reperfusion-induced AKI [12]. Mechanistically, the renoprotective effects of TBBt were associated with the suppression of MAPKs and NF-κB pathways. Because MAPK and NF-κB activation are critical events for septic AKI, it can be hypothesized that CK2α inhibition would be beneﬁcial against septic AKI. To address this question, a cecum ligation and puncture (CLP)-induced AKI model was generated and investigated for the potential renoprotective effects of TBBt against septic AKI. 2. Results 2.1. CK2α Expression Is Increased in Kidney Tissues of Septic AKI Mice Septic AKI was induced in C57BL/6J mice, and kidney and blood samples were obtained at various time points (Figure 1A). To determine whether CK2α is involved in septic AKI pathogenesis, protein levels were measured for CK2α in kidney tissues. Time- course analyses showed that the CK2α protein level reached peak levels at 3 h, continued to increase up to 6 h, and returned to the normal level at 24 h after CLP (Figure 1B). Figure 1. Alterations of CK2α expression in kidney tissues of septic AKI mice. (A) C57BL/6J mice were intraperitoneally injected with vehicle or TBBt and subjected to CLP. Kidney tissues and blood samples were collected at indicated times after CLP for each experiment. (B) Kidney tissues prepared from mice with CLP at the indicated time points were used to analyze CK2α expression. Protein intensity was measured. Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. time 0. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; WB, Western blotting; ELISA, enzyme-linked immunosorbent assay; BUN, blood urea nitrogen; SCr, serum creatinine; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling. Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 2 of 12 Casein kinase 2 (CK2), also known as protein kinase 2, is a protein serine/threonine kinase; it is a tetrameric enzyme composed of two catalytic subunits (α and/or α′) and two regulatory beta subunits [7]. A large number of studies show that CK2 has been found to phosphorylate various transcription factors regulating the inﬂammatory response, includ-ing NF-κB [8]. Speciﬁcally, CK2 phosphorylates inhibitory κB (IκB)α and degrades it through the proteasomal pathway [9]. In addition to IκBα, CK2 also directly phosphory-lates p65, thereby amplifying its transcriptional activity [10]. Overall, inhibition of CK2 activity implicates attenuating inﬂammation and cytokine signaling via the suppression of NF-κB. Over the past two decades, numerous inhibitors of CK2, such as emodin, apigenin, and 4,5,6,7-tetrabromobenzotriazole (TBBt) have been discovered and developed. Among these inhibitors, TBBt is the most eﬀective cell-permeant inhibitor of CK2. TBBt selectivity is obtained via a hydrophobic pocket adjacent to the ATP/GTP binding site, which is smaller in CK2 than in other protein kinases [11]. Ka et al. previously showed that TBBt is useful to ameliorate ischemia–reperfusion-induced AKI [12]. Mechanistically, the reno-protective eﬀects of TBBt were associated with the suppression of MAPKs and NF-κB pathways. Because MAPK and NF-κB activation are critical events for septic AKI, it can be hypothesized that CK2α inhibition would be beneﬁcial against septic AKI. To address this question, a cecum ligation and puncture (CLP)-induced AKI model was generated and investigated for the potential renoprotective eﬀects of TBBt against septic AKI. 2. Results 2.1. CK2α Expression Is Increased in Kidney Tissues of Septic AKI Mice Septic AKI was induced in C57BL/6J mice, and kidney and blood samples were ob-tained at various time points (Figure 1A). To determine whether CK2α is involved in sep-tic AKI pathogenesis, protein levels were measured for CK2α in kidney tissues. Time-course analyses showed that the CK2α protein level reached peak levels at 3 h, continued to increase up to 6 h, and returned to the normal level at 24 h after CLP (Figure 1B). Figure 1. Alterations of CK2α expression in kidney tissues of septic AKI mice. (A) C57BL/6J mice were intraperitoneally injected with vehicle or TBBt and subjected to CLP. Kidney tissues and blood samples were collected at indicated times after CLP for each experiment. (B) Kidney tissues pre-Int. J. Mol. Sci. 2023, 24, 9783 3 of 11 2.2. CK2α Inhibition Increases Survival Outcome of Spetic AKI Mice The survival of CLP mice with or without TBBt treatment was investigated next. Mortality was 80% within 48 h after CLP without TBBt treatment. However, in septic mice treated with 1 mg/kg of TBBt, the survival rate was 38%, and the survival rate was 55% when 2 mg/kg of TBBt was administered at 7 days after CLP (Figure 2). Figure 2. Improved survival outcome with CK2α inhibitor. Kaplan–Meier survival curves of CLP- operated mice treated with or without TBBt. Mice were randomly divided into three groups: sham, CLP, and CLP with TBBt (either 1 or 2 mg/kg). For the CLP plus TBBt group, TBBt was intraperitoneal administered 3 h before CLP. Other groups received the same volume of DMSO as a control. Survival was monitored for 7 d. Values are mean ± SD (n = 10 mice per group). p < 0.01 vs. sham and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; DMSO, demethyl sulfoxide. 2.3. CK2α Inhibition Protects Renal Function and Alleviates Kidney Injury The renoprotective effects of TBBt were identiﬁed by measuring renal injury biomark- ers, including serum creatinine and BUN. After 24 h of CLP, the levels of creatinine and BUN were elevated in CLP mice, while those levels were signiﬁcantly reduced in TBBt-treated CLP mice (Figure 3A). The extent of renal injury was assessed via histological observation with H&E and PAS staining. In the CLP group, there was severe architectural disruption of the kidney, including tubular dilatation, brush border loss, and necrosis. However, less tubular injury and necrosis were observed in TBBt-treated mice (Figure 3B,C). 2.4. CK2α Inhibition Suppresses Apoptosis in Mice with Septic AKI Although necrosis is the major cause of cell death in septic AKI, apoptotic cell death also contributes to histopathological changes in the kidney [13]. Therefore, we evaluated the extent of apoptosis via TUNEL staining. The number of TUNEL-positive apoptotic cells was considerably increased in CLP mice compared to sham mice (Figure 4A,B). Consistently, increased protein levels of the proapoptotic proteins cleaved caspase-3 and Bax, and decreased protein levels of the antiapoptotic protein Bcl-2 were observed in CLP mice (Figure 4C,D). Treatment with TBBt attenuated all of these changes. 2.5. CK2α Inhibition Decreases Inﬂammatory Responses in Mice with Septic AKI Cytokines play an essential role in the initiation and progression of systemic inﬂam- mation in septic mice [4,5]. Therefore, we investigated the effect of TBBt as an inhibitor of inﬂammation during sepsis. The mice 12 h after CLP-induced sepsis had marked elevations in TNF-α, IL-6, and INF-γ levels in serum (Figure 5A). However, following sepsis induction, pretreatment with TBBt showed noticeably lower levels of these cytokines compared to CLP mice without TBBt. The cytokine levels were also analyzed in peritoneal ﬂuid, and Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 3 of 12 pared from mice with CLP at the indicated time points were used to analyze CK2α expression. Pro-tein intensity was measured. Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. time 0. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; WB, Western blotting; ELISA, enzyme-linked immunosorbent assay; BUN, blood urea nitrogen; SCr, serum creatinine; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling. 2.2. CK2α Inhibition Increases Survival Outcome of Spetic AKI Mice The survival of CLP mice with or without TBBt treatment was investigated next. Mor-tality was 80% within 48 h after CLP without TBBt treatment. However, in septic mice treated with 1 mg/kg of TBBt, the survival rate was 38%, and the survival rate was 55% when 2 mg/kg of TBBt was administered at 7 days after CLP (Figure 2). Figure 2. Improved survival outcome with CK2α inhibitor. Kaplan–Meier survival curves of CLP-operated mice treated with or without TBBt. Mice were randomly divided into three groups: sham, CLP, and CLP with TBBt (either 1 or 2 mg/kg). For the CLP plus TBBt group, TBBt was intraperito-neal administered 3 h before CLP. Other groups received the same volume of DMSO as a control. Survival was monitored for 7 d. Values are mean ± SD (n = 10 mice per group). p < 0.01 vs. sham and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; DMSO, demethyl sulfoxide. 2.3. CK2α Inhibition Protects Renal Function and Alleviates Kidney Injury The renoprotective eﬀects of TBBt were identiﬁed by measuring renal injury bi-omarkers, including serum creatinine and BUN. After 24 h of CLP, the levels of creatinine and BUN were elevated in CLP mice, while those levels were signiﬁcantly reduced in TBBt-treated CLP mice (Figure 3A). The extent of renal injury was assessed via histological observation with H&E and PAS staining. In the CLP group, there was severe architectural disruption of the kidney, including tubular dilatation, brush border loss, and necrosis. However, less tubular injury and necrosis were observed in TBBt-treated mice (Figure 3B,C). Int. J. Mol. Sci. 2023, 24, 9783 4 of 11 as with the results in the serum, the inhibitory effect of TBBt on cytokine levels was also conﬁrmed in the peritoneal ﬂuid. Figure 3. Reduced acute renal injury with TBBt. Mice were injected with 2 mg/kg of TBBt 3 h before CLP. (A) After 24 h CLP, blood samples were obtained for measurement of creatinine and BUN. (B) Kidney tissue were stained with H&E and PAS. Bars = 250 µm. (C) Histopathologic scoring and quantiﬁcation of the necrotic area were performed in a blind fashion. The tubular injury score was given based upon epithelial simpliﬁcation, tubular dilatation, vacuolization, and red blood cell (RBC) and hyaline casts (score 0: <1%, 1: ≥1~10%. 2: ≥10~25%, 3: ≥25~50%, 4: >50%). Bars show the mean ± SEM (n = 5). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; H&E, hematoxylin and eosin; PAS, periodic acid–Schiff. Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 4 of 12 Figure 3. Reduced acute renal injury with TBBt. Mice were injected with 2 mg/kg of TBBt 3 h before CLP. (A) After 24 h CLP, blood samples were obtained for measurement of creatinine and BUN. (B) Kidney tissue were stained with H&E and PAS. Bars = 250 µm. (C) Histopathologic scoring and quantiﬁcation of the necrotic area were performed in a blind fashion. The tubular injury score was given based upon epithelial simpliﬁcation, tubular dilatation, vacuolization, and red blood cell (RBC) and hyaline casts (score 0: <1%, 1: ≥1~10%. 2: ≥10~25%, 3: ≥25~50%, 4: >50%). Bars show the mean ± SEM (n = 5). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; H&E, hematoxylin and eosin; PAS, periodic acid–Schiﬀ. 2.4. CK2α Inhibition Suppresses Apoptosis in Mice with Septic AKI Although necrosis is the major cause of cell death in septic AKI, apoptotic cell death also contributes to histopathological changes in the kidney [13]. Therefore, we evaluated the extent of apoptosis via TUNEL staining. The number of TUNEL-positive apoptotic cells was considerably increased in CLP mice compared to sham mice (Figure 4A,B). Con-sistently, increased protein levels of the proapoptotic proteins cleaved caspase-3 and Bax, and decreased protein levels of the antiapoptotic protein Bcl-2 were observed in CLP mice (Figure 4C,D). Treatment with TBBt attenuated all of these changes. Int. J. Mol. Sci. 2023, 24, 9783 5 of 11 Figure 4. Suppression of CLP-induced apoptosis with TBBt. Mice were injected with 2 mg/kg of TBBt, and kidney tissues were collected 24 h after CLP. (A) Tissues were stained with TUNEL (×400). Bars = 250 µm. (B) Apoptotic cells were counted and expressed by a percentage which represented all glomerular and tubular cells. (C,D) The expression levels of Bcl-2, cleaved caspase-3, and Bax were examined via Western blotting at 24 h after CLP. Protein intensity was measured. Values are expressed as the mean ± SD (n = 5 mice per group). * p < 0.05 and p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling. Figure 5. Decreased inﬂammatory responses with TBBt. Mice were injected with 2 mg/kg of TBBt 3 h before CLP. (A) Protein levels of inﬂammatory cytokines in blood and peritoneal ﬂuid were measured via ELISA 12 h after CLP. (B) Immunohistochemistry to identify inﬁltrating macrophages (F4/80+) at 24 h after CLP. Bars = 50 µm. F4/80+-positive macrophages (indicated with arrows) were counted in at least 5 photographs at ×400 magniﬁcation per animal and expressed as the number of positive macrophages per high-power ﬁeld (HPF). Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; ND, not detected. Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 5 of 12 Figure 4. Suppression of CLP-induced apoptosis with TBBt. Mice were injected with 2 mg/kg of TBBt, and kidney tissues were collected 24 h after CLP. (A) Tissues were stained with TUNEL (×400). Bars = 250 µm. (B) Apoptotic cells were counted and expressed by a percentage which represented all glomerular and tubular cells. (C,D) The expression levels of Bcl-2, cleaved caspase-3, and Bax were examined via Western blotting at 24 h after CLP. Protein intensity was measured. Values are expressed as the mean ± SD (n = 5 mice per group). * p < 0.05 and p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; TUNEL, terminal deoxynucleotidyl transferase dUTP nick-end labeling. 2.5. CK2α Inhibition Decreases Inﬂammatory Responses in Mice with Septic AKI Cytokines play an essential role in the initiation and progression of systemic inﬂam-mation in septic mice [4,5]. Therefore, we investigated the eﬀect of TBBt as an inhibitor of inﬂammation during sepsis. The mice 12 h after CLP-induced sepsis had marked eleva-tions in TNF-α, IL-6, and INF-γ levels in serum (Figure 5A). However, following sepsis induction, pretreatment with TBBt showed noticeably lower levels of these cytokines com-pared to CLP mice without TBBt. The cytokine levels were also analyzed in peritoneal ﬂuid, and as with the results in the serum, the inhibitory eﬀect of TBBt on cytokine levels was also conﬁrmed in the peritoneal ﬂuid. Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 6 of 12 Figure 5. Decreased inﬂammatory responses with TBBt. Mice were injected with 2 mg/kg of TBBt 3 h before CLP. (A) Protein levels of inﬂammatory cytokines in blood and peritoneal ﬂuid were meas-ured via ELISA 12 h after CLP. (B) Immunohistochemistry to identify inﬁltrating macrophages (F4/80+) at 24 h after CLP. Bars = 50 µm. F4/80+-positive macrophages (indicated with arrows) were counted in at least 5 photographs at ×400 magniﬁcation per animal and expressed as the number of positive macrophages per high-power ﬁeld (HPF). Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture; ND, not detected. Next, we determined F4/80, a speciﬁc surface marker of macrophages, in the kidney tissues of CLP mice. The results of the IHC staining showed that the number of F4/80+ macrophages was dramatically increased in CLP mice, whereas in TBBt-pretreated mice it was signiﬁcantly decreased (Figure 5B). 2.6. CK2α Inhibition Suppresses MAPK-NF-κB Signaling Pathway in Mice with Septic AKI To explore the mechanisms responsible for mediating the anti-inﬂammatory eﬀect of TBBt, we measured NF-κB activation, which is a major signaling pathway of inﬂamma-tion, in the kidney tissues of mice. Upon CLP, we observed the translocation of NF-κB subunits p50 and p65 into the nucleus (Figure 6A,B). In contrast, in the presence of TBBt, levels of p50 and p65 in the nucleus were reduced. Additionally, we observed an increase in the phosphorylation of IκB kinase (IKK) in CLP mice, whereas the ratio of p-IKK/IKK was decreased in TBBt-pretreated mice. Int. J. Mol. Sci. 2023, 24, 9783 6 of 11 Next, we determined F4/80, a speciﬁc surface marker of macrophages, in the kidney tissues of CLP mice. The results of the IHC staining showed that the number of F4/80+ macrophages was dramatically increased in CLP mice, whereas in TBBt-pretreated mice it was signiﬁcantly decreased (Figure 5B). 2.6. CK2α Inhibition Suppresses MAPK-NF-κB Signaling Pathway in Mice with Septic AKI To explore the mechanisms responsible for mediating the anti-inﬂammatory effect of TBBt, we measured NF-κB activation, which is a major signaling pathway of inﬂammation, in the kidney tissues of mice. Upon CLP, we observed the translocation of NF-κB subunits p50 and p65 into the nucleus (Figure 6A,B). In contrast, in the presence of TBBt, levels of p50 and p65 in the nucleus were reduced. Additionally, we observed an increase in the phosphorylation of IκB kinase (IKK) in CLP mice, whereas the ratio of p-IKK/IKK was decreased in TBBt-pretreated mice. Figure 6. Suppression of MAPKs and NF-κB pathways in kidney tissues of septic mice with TBBt. (A,B) Mice were injected with 2 mg/kg of TBBt 3 h before CLP. Cytosolic (CE) and nuclear extract (NE) were prepared from kidney tissues 3 h after CLP. Nuclear translocation of p50 and p65 subunits and phosphorylation of cytoplasmic IKK were analyzed via Western blotting (n = 5 mice per group). (C,D) Total lysates prepared from kidney tissues 1 h after CLP were used to determine the total and phospho-forms of p38 MAPK and ERK. Protein intensity was measured. Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture. To investigate the mechanism by which TBBt suppresses the NF-κB signaling pathway, we compared the mitogen-activated protein kinase (MAPK) signal transduction path- ways. We analyzed the lysates prepared from kidney tissues via Western blotting. Our results showed that CLP upregulated the phosphorylated forms of extracellular-signal- regulated kinase (ERK) and p38 MAPK, which were markedly suppressed by TBBt treat- ment (Figure 6C,D). These results suggest that TBBt suppresses inﬂammatory responses through the downregulation of the MAPKs–IKK–NF–κB axis in septic AKI mice. Int. J. Mol. Sci. 2023, 23, x FOR PEER REVIEW 7 of 12 Figure 6. Suppression of MAPKs and NF-κB pathways in kidney tissues of septic mice with TBBt. (A,B) Mice were injected with 2 mg/kg of TBBt 3 h before CLP. Cytosolic (CE) and nuclear extract (NE) were prepared from kidney tissues 3 h after CLP. Nuclear translocation of p50 and p65 subu-nits and phosphorylation of cytoplasmic IKK were analyzed via Western blotting (n = 5 mice per group). (C,D) Total lysates prepared from kidney tissues 1 h after CLP were used to determine the total and phospho-forms of p38 MAPK and ERK. Protein intensity was measured. Values are mean ± SD (n = 5 mice per group). p < 0.01 vs. sham; # p < 0.05; and ## p < 0.01 vs. CLP + vehicle. TBBt, 4,5,6,7-tetrabromobenzotriazole; CLP, cecum ligation and puncture. To investigate the mechanism by which TBBt suppresses the NF-κB signaling path-way, we compared the mitogen-activated protein kinase (MAPK) signal transduction pathways. We analyzed the lysates prepared from kidney tissues via Western blotting. Our results showed that CLP upregulated the phosphorylated forms of extracellular-sig-nal-regulated kinase (ERK) and p38 MAPK, which were markedly suppressed by TBBt treatment (Figure 6C,D). These results suggest that TBBt suppresses inﬂammatory re-sponses through the downregulation of the MAPKs–IKK–NF–κB axis in septic AKI mice. 3. Discussion The aim of this study was to evaluate the potential eﬀects of CK2α inhibition on septic AKI in mice. The results of this study clearly demonstrated that treatment with CK2α in-hibitor TBBt alleviated sepsis-induced histopathological damage and improved renal function. Furthermore, TBBt not only increased the survival time but also the survival rate of septic mice. These beneﬁcial eﬀects were accompanied by decreases in inﬂammation and apoptosis. The underlying molecular mechanism appeared to involve the suppres-sion of MAPKs and NF-κB activation. Inﬂammation plays a critical role in the pathogenesis of septic AKI [14]. A recent study demonstrated that myeloid speciﬁc CK2α knockout mice are resistant to systemic bacterial infection [15], highlighting the importance of CK2α in inﬂammatory responses in an infectious bacterial setting. In agreement with this study, CK2α inhibition attenuated CLP-induced systemic inﬂammation, as evidenced by the decreased levels of cytokines Int. J. Mol. Sci. 2023, 24, 9783 7 of 11 3. Discussion The aim of this study was to evaluate the potential effects of CK2α inhibition on septic AKI in mice. The results of this study clearly demonstrated that treatment with CK2α inhibitor TBBt alleviated sepsis-induced histopathological damage and improved renal function. Furthermore, TBBt not only increased the survival time but also the survival rate of septic mice. These beneﬁcial effects were accompanied by decreases in inﬂammation and apoptosis. The underlying molecular mechanism appeared to involve the suppression of MAPKs and NF-κB activation. Inﬂammation plays a critical role in the pathogenesis of septic AKI [14]. A recent study demonstrated that myeloid speciﬁc CK2α knockout mice are resistant to systemic bacterial infection [15], highlighting the importance of CK2α in inﬂammatory responses in an infectious bacterial setting. In agreement with this study, CK2α inhibition attenuated CLP- induced systemic inﬂammation, as evidenced by the decreased levels of cytokines (TNF-α, IFN-γ, and IL-6) in blood and peritoneum and also the reduced numbers of macrophages in kidney tissues. Macrophages are initiators of inﬂammation via the release of pro- inﬂammatory cytokines and mediators in response to endotoxin. Excessive macrophage inﬁltration directly affects renal parenchyma and promotes tubular cell apoptosis, which ultimately induces the occurrence of AKI [14]. The role of pro-inﬂammatory cytokines in sepsis-induced tissue damage has been well established. Pro-inﬂammatory cytokines such as TNF-α, IFN-γ, and IL-6 act as endogenous pyrogens, promoting the synthesis of other pro-inﬂammatory cytokines and inﬂammatory mediators by macrophages and mesenchymal cells. They are also known to stimulate the production of acute-phase proteins [16]. IL-6 levels are higher in patients who died from severe sepsis [17,18] and among pro-inﬂammatory cytokines, and plasma IL-6 levels have the best correlation with the mortality rate of sepsis patients [5], indicating that IL-6 is the key cytokine in the pathophysiology of severe sepsis. Plasma levels of TNF-α and IFN-γ are also markedly increased in patients with sepsis and in animal models [17,19]. Since the increase in macrophage inﬁltration and consequent release of pro-inﬂammatory cytokines appear to be an essential part of the pathogenesis in the inﬂammation process in septic AKI, this study provides a pharmacological basis for TBBt in the management of inﬂammation in septic AKI. What is the underlying molecular mechanism by which TBBt exhibits an anti-inflammatory role? NF-κB is believed to be a master regulator of inﬂammation. Once activated, NF-κB initiates tissue injury by releasing a variety of pro-inﬂammatory cytokines and mediators. These released cytokines also activate NF-κB, which leads to a vicious cycle [17]. In this regard, several NF-κB inhibitors derived from either natural products or synthetic design have been proven to be highly effective for the treatment of septic AKI in mice [20–23]. In this study, increased phosphorylation of IKK in the cytosol and increased levels of NF-κB subunits in the nucleus were observed in the CLP mice, indicating that IKK-mediated phosphorylation of NF-κB subunits in the cytosol and concomitant nuclear translocation of NF-κB subunits. LPS-triggered MAPK activation is a critical part of signal transduction in modulating the activation of NF-κB [24]. This study demonstrates that TBBt treatment inhibits the phosphorylation of p38 MAPK and ERK and decreases NF-κB activity in the kidney tissues of CLP mice. In consistency with the suppression of pro-inﬂammatory cytokines (TNF-α, IFN-γ, and IL-6), it can be suggested that pretreatment with TBBt effectively relieves AKI, and its protective effects may be associated with the regulation of the MAPK-NF-κB-cytokines signaling pathway. It is well known that the activation of NF-κB has been suggested to be positively correlated to apoptosis [25]. Additionally, the expression of caspase-3 is regulated by NF- κB in the kidney tissues of septic patients [26]. Consistent with these reports, pretreatment with TBBt decreased the protein levels of cleaved caspase-3 and the number of TUNEL- positive apoptotic cells in kidney tissues of CLP mice. This suggests that TBBt inhibits the activation of NF-κB, thereby exerting anti-apoptotic effects in septic AKI mice. Int. J. Mol. Sci. 2023, 24, 9783 8 of 11 Apart from inﬂammation and apoptotic tissue injury, hemodynamic factors also play a critical role in the development and progression of AKI. Multiple mechanisms may cause microcirculatory dysfunction in septic AKI, such as endothelial injury, shedding of the glycocalyx, autonomic nervous system activation, and microthrombi formation [14]. In this regard, Ka et al. previously reported that TBBt treatment protected mice against bilateral renal ischemia–reperfusion injury. TBBt-treated mice presented preservation of renal function, histologically less tubular damage, and reduced inﬁltration of inﬂammatory cells [12]. However, an initial clinical trial with goal-directed therapy has shown little improvement in the mortality rates of septic AKI [27]. One explanation for these unexpected outcomes is that tissue damage might occur before microvascular alteration. Nevertheless, further study is required to ﬁgure out if TBBt could affect hemodynamic alterations in mice with septic AKI. In the current study, a treatment regime of 2 mg/kg TBBt signiﬁcantly attenuated the renal impairment. In a previous study, the same dose of TBBt could effectively protect against renal ischemia–reperfusion injury [12]. Because no noticeable adverse effects were not found in two studies, a 2 mg/kg dose of TBBt seems to be the most renoprotective dose in AKI animal models. This dose is less than other disease models. Intraperitoneal injection of 12.5 mg/kg of TBBt suppresses NF-κB activation in a murine model of asthma [28]. In a mouse model of bleomycin-induced dermal ﬁbrosis, TBBt (2.5 mg/kg/day for three weeks) suppresses JAK2/STAT3 signaling [29]. In summary, a selective CK2 inhibitor, TBBt, alleviates sepsis-induced AKI by sup- pressing inﬂammation and apoptosis. Therefore, these results provide CK2 as a promising therapeutic target for treatment of sepsis-associated AKI. 4. Materials and Methods 4.1. Experimental Animals and Materials Pathogen-free male 8-week-old C57BL/6J mice (body weight 20 ± 2 g) were purchased from Orient Bio (Sungnam, Republic of Korea). The mice were housed in a laminar airﬂow cabinet with a 12-h light/dark cycle and maintained on standard laboratory chow ad libitum. TBBt was purchased from Tocris (#2275, Bristol, UK) and dissolved in dimethyl sulfoxide (DMSO) to a concentration of 50 mg/mL. All other reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise noted. 4.2. Animal Model of Cecal Ligation & Puncture To investigate the effect of TBBt during sepsis-induced AKI, mice were randomly divided into three groups: (i) sham-operated mice as the control group, (ii) the CLP group injected with normal saline, and (iii) the TBBt group with either 1 or 2 mg/kg in 200 µL of 5% DMSO in CLP-operated mice. TBBt was administered intraperitoneally once 3 h before CLP. CLP was performed as previously described elsewhere, with minor modiﬁcations [30]. Brieﬂy, mice were anesthetized with a ketamine–xylazine mixture through an i.p. injection. To expose the cecum, the mouse abdomen was incised about 1 cm, and the cecum was ligated with a 4-0 silk suture (5 mm) from the base of the ileocecal valve. Next, the cecum was doubly perforated using a 24-gauge needle. A small amount of the bowel contents was extracted from both holes, and the cecum was inserted back into the abdomen. The skin of the abdomen was then closed, followed by a subcutaneous injection of resuscitative normal saline. Twenty-four hours after the CLP surgical procedure, mice were sacriﬁced via an overdose of sodium pentobarbital. Blood was collected from the hearts for biochemical analysis. Kidneys were collected, cut in half, and one half was immediately frozen in liquid nitrogen for protein analysis, while the other half was ﬁxed in 4% formalin at room temperature for parafﬁn-embedded histological analysis. Another group of CLP mice with or without TBBt pre-treatment was observed to determine their survival rate for 7 days. All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals, published by the US National Institutes of Health (NIH Publication No. 85-23, revised 2011). The current study protocol was approved by the Int. J. Mol. Sci. 2023, 24, 9783 9 of 11 Institutional Animal Care and Use Committee of Jeonbuk National University (Approval No. JBNU 2017-0088). 4.3. Western Blot Analysis Kidney tissues were homogenized with proteinase and phosphatase inhibitors in a pro- tein extraction solution (Intron Biotechnology, Burlington, MA, USA). The homogenates, which contained 20 µg of protein, were separated via 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to polyvinylidene ﬂuoride (PVDF) membranes. The blot was probed with 1:2500 diluted primary antibodies against CK2α (#2656), p-ERK (#4377), ERK (#4695), p-p38 MAPK (#4511), p38 MAPK (#8690), p-IKK (#2697), IKK (#2682), p50 (#3035), p65 (#8262), Bcl2 (#3498), cleaved caspase-3 (#9664), Bax (#5023, Cell Signaling, Beverly, MA, USA), lamin B1 (SC-374015, Santa Cruz Biotechnology, Dallas, TX, USA), and GAPDH (#A351, Bioworld Technology, St. Louis Park, MN, USA). Signals were de- tected with a Las-4000 imager (GE Healthcare Life Science, Pittsburgh, PA, USA). All the densitometry ﬁgures were obtained through Image J software (Version 1.53t). Fractionation of the kidney tissue was performed using speciﬁc reagents for nuclear and cytosolic extraction (#78833, Thermo Scientiﬁc, Rockford, IL, USA). After centrifugation at 15,000× g for 10 min, the supernatants from the lysates were used for cytosolic extraction, while the pellets were used for nuclear extraction. All experiments were performed on ice. 4.4. Biochemical Analysis Blood urea nitrogen (BUN) and serum creatinine were measured using speciﬁc assay kits (#K002-H1, Arbor Assays, Ann Arbor, MI, USA). TNF-α (#BMS607-3), IL-6 (#KHC- 0061), and IFN-γ (#KMC4021, Invitrogen, Carlsbad, CA, USA) were measured using speciﬁc ELISA kits following the manufacturer(cid:48)s instructions. 4.5. Histopathologic Assessment The kidneys were collected and ﬁxed with 4% formalin at room temperature for 24 h and were then parafﬁn-embedded. Tissues were cut into 4-µm thick sections, which were stained with hematoxylin and eosin (H&E) at room temperature and visualized under a microscope (Leica DM 2500; Leica Microsystems GmbH, Wetzlar, Germany). Histopathological damage was deﬁned as swelling of the tubular epithelia, loss of brush border, degeneration of vacuole, necrosis of tubules, cast formation, and desquamation. The degree of tubular injury was estimated at a ×200 magniﬁcation using 5 randomly selected ﬁelds for each kidney according to the following standard: 0, normal; 1, damage involving <25% of tubules; 2, damage involving 25–50% of tubules; 3, damage involving 50–75% of tubules; and 4, damage involving 75–100% of tubules. For Periodic acid–Schiff (PAS) staining, sections were hydrated in ethanol and stained with a PAS reagent (#ab150680, Abcam, Cambridge, UK). Tubular necrosis was quantitated as the percentage of tubules in the outer medulla in which epithelial necrosis or necrotic debris was observed in PAS- stained sections. 4.6. Immunohistochemistry Immunohistochemical staining was performed using the DAKO Envision system (Carpinteria, CA, USA), which utilizes dextran polymers conjugated with horseradish per- oxidase to prevent contamination with endogenous biotin. After deparafﬁnization, tissue sections were subjected to a microwave antigen-retrieval procedure in a 10-mM sodium citrate buffer. Subsequently, endogenous peroxidases were blocked, and the sections were incubated with Protein Block Serum-Free (DAKO) to prevent non-speciﬁc staining. The sections were then immunostained with an antibody against F4/80 (#ab6640, Abcam) in 1:100 dilutions overnight. After washing with TRIS buffer saline (TBS, Sigma, St. Louis, MO, USA) 3 times for 10 min each, stained tissues were covered with a coverslip of an ap- propriate size. The stained sections were scanned under a microscope (Leica DM 2500) and Int. J. Mol. Sci. 2023, 24, 9783 10 of 11 quantiﬁed as the number of F4/80 positive cells per ﬁeld using iSolution DT 36 software (Version 36.0) (Carl Zeiss, Oberkochen, Germany). 4.7. TUNEL Assay TUNEL staining was used to detect apoptotic cells (#G3250, Promega, Madison, WI, USA). In brief, after treatment with a nucleotide mix and rTdT (terminal deoxynucleotide transferase), tissue sections were incubated at 37 ◦C for 1 h. Apoptotic cells, counterstained with hematoxylin, were covered with an appropriately sized coverslip. Apoptotic cells were counted under a microscope (×200) and expressed as the apoptosis index (num- ber of apoptotic bodies/100 cells). Each group was assessed in triplicate, and the data were averaged. 4.8. Statistical Analysis Data were expressed as the mean ± SD. Statistical comparisons were performed using one-way analysis of variance, followed by Bonferroni’s post hoc analysis. p values less than 0.05 were considered signiﬁcant. Author Contributions: J.-H.K.: Investigation, formal analysis, writing—original draft. H.C.Y.: In- vestigation, formal analysis, writing—original draft. S.N.: Investigation. D.-C.K.: Formal analysis, methodology. J.H.L.: Conceptualization, methodology, supervision, writing—original draft, review, and editing. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2020R1I1A1A01071984); Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare (HR22C1832); and the Biomedical Research Institute of Jeonbuk National University Hospital. Institutional Review Board Statement: All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals, published by the US National Institutes of Health (NIH Publication No. 85-23, revised 2011). The current study protocol was approved by the Institutional Animal Care and Use Committee of Jeonbuk National University (Approval No. JBNU 2017-0088). Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors have no conﬂict of interest to declare. References 1. 2. 3. 4. 5. 6. 7. 8. Singer, M.; Deutschman, C.S.; Seymour, C.W.; Shankar-Hari, M.; Annane, D.; Bauer, M.; Bellomo, R.; Bernard, G.R.; Chiche, J.D.; Coopersmith, C.M.; et al. The Third International Consensus Deﬁnitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016, 315, 801–810. [CrossRef] Arina, P.; Singer, M. Pathophysiology of sepsis. Curr. Opin. Anaesthesiol. 2021, 34, 77–84. [CrossRef] [PubMed] Uchino, S.; Kellum, J.A.; Bellomo, R.; Doig, G.S.; Morimatsu, H.; Morgera, S.; Schetz, M.; Tan, I.; Bouman, C.; Macedo, E.; et al. Acute renal failure in critically ill patients: A multinational, multicenter study. JAMA 2005, 294, 813–818. [CrossRef] [PubMed] Gonçalves, M.C.; Horewicz, V.V.; Lückemeyer, D.D.; Prudente, A.S.; Assreuy, J. Experimental sepsis severity score associated to mortality and bacterial spreading is related to bacterial load and inﬂammatory proﬁle of different tissues. Inﬂammation 2017, 40, 1553–1565. [CrossRef] [PubMed] Simmons, E.M.; Himmelfarb, J.; Sezer, M.T.; Chertow, G.M.; Mehta, R.L.; Paganini, E.P.; Soroko, S.; Freedman, S.; Becker, K.; Spratt, D.; et al. Plasma cytokine levels predict mortality in patients with acute renal failure. Kidney Int. 2004, 65, 1357–1365. [CrossRef] Chen, F.; Zou, L.; Williams, B.; Chao, W. Targeting Toll-like receptors in sepsis: From bench to clinical trials. Antioxid. Redox Signal. 2021, 35, 1324–1339. [CrossRef] [PubMed] Graham, K.C.; Litchﬁeld, D.W. The regulatory β subunit of protein kinase CK2 mediates formation of tetrameric CK2 complexes. J. Biol. Chem. 2000, 275, 5003–5010. [CrossRef] Singh, N.N.; Ramji, D.P. Protein kinase CK2, an important regulator of the inﬂammatory response? J. Mol. Med. 2008, 86, 887–897. [CrossRef] Int. J. Mol. Sci. 2023, 24, 9783 11 of 11 9. Mathes, E.; O’Dea, E.L.; Hoffmann, A.; Ghosh, G. NF-κB dictates the degradation pathway of IκBα. EMBO J. 2008, 27, 1357–1367. 10. 11. [CrossRef] Ji, H.; Lu, Z. The role of protein kinase CK2 in glioblastoma development. Clin. Cancer Res. 2013, 19, 6335–6337. [CrossRef] Sajnaga, E.; Kubinski, K.; Szyszka, R. Catalytic activity of mutants of yeast protein kinase CK2α. Acta Biochim. Pol. 2008, 55, 767–776. [CrossRef] [PubMed] 12. Ka, S.O.; Hwang, H.P.; Jang, J.H.; Hyuk Bang, I.; Bae, U.J.; Yu, H.C.; Cho, B.H.; Park, B.H. The protein kinase 2 inhibitor tetrabromobenzotriazole protects against renal ischemia reperfusion injury. Sci. Rep. 2015, 5, 14816. [CrossRef] [PubMed] 13. Bantel, H.; Schulze-Osthoff, K. Cell death in sepsis: A matter of how, when, and where. Crit. Care 2009, 13, 173. [CrossRef] 14. Peerapornratana, S.; Manrique-Caballero, C.L.; Gomez, H.; Kellum, J.A. Acute kidney injury from sepsis: Current concepts, epidemiology, pathophysiology, prevention and treatment. Kidney Int. 2019, 96, 1083–1099. [CrossRef] [PubMed] 15. Larson, S.R.; Bortell, N.; Illies, A.; Crisler, W.J.; Matsuda, J.L.; Lenz, L.L. Myeloid cell CK2 regulates inﬂammation and resistance to bacterial infection. Front. Immunol. 2020, 11, 590266. [CrossRef] [PubMed] 16. Chaudhry, H.; Zhou, J.; Zhong, Y.; Ali, M.M.; McGuire, F.; Nagarkatti, P.S.; Nagarkatti, M. Role of cytokines as a double-edged sword in sepsis. Vivo 2013, 27, 669–684. 17. Mera, S.; Tatulescu, D.; Cismaru, C.; Bondor, C.; Slavcovici, A.; Zanc, V.; Carstina, D.; Oltean, M. Multiplex cytokine proﬁling in patients with sepsis. Apmis 2011, 119, 155–163. [CrossRef] 18. Gouel-Cheron, A.; Allaouchiche, B.; Guignant, C.; Davin, F.; Floccard, B.; Monneret, G.; AzuRea, G. Early interleukin-6 and slope of monocyte human leukocyte antigen-DR: A powerful association to predict the development of sepsis after major trauma. PLoS ONE 2012, 7, e33095. [CrossRef] 19. Romero, C.R.; Herzig, D.S.; Etogo, A.; Nunez, J.; Mahmoudizad, R.; Fang, G.; Murphey, E.D.; Toliver-Kinsky, T.; Sherwood, E.R. The role of interferon-γ in the pathogenesis of acute intra-abdominal sepsis. J. Leukoc. Biol. 2010, 88, 725–735. [CrossRef] Ibrahim, Y.F.; Moussa, R.A.; Bayoumi, A.M.A.; Ahmed, A.F. Tocilizumab attenuates acute lung and kidney injuries and improves survival in a rat model of sepsis via down-regulation of NF-κB/JNK: A possible role of P-glycoprotein. Inﬂammopharmacology 2020, 28, 215–230. [CrossRef] 20. 21. Chen, L.; Yang, S.; Zumbrun, E.E.; Guan, H.; Nagarkatti, P.S.; Nagarkatti, M. Resveratrol attenuates lipopolysaccharide-induced acute kidney injury by suppressing inﬂammation driven by macrophages. Mol. Nutr. Food Res. 2015, 59, 853–864. [CrossRef] [PubMed] 22. Zhong, W.; Qian, K.; Xiong, J.; Ma, K.; Wang, A.; Zou, Y. Curcumin alleviates lipopolysaccharide induced sepsis and liver failure by suppression of oxidative stress-related inﬂammation via PI3K/AKT and NF-κB related signaling. Biomed. Pharmacother. 2016, 83, 302–313. [CrossRef] [PubMed] 23. Xu, X.; Liao, L.; Hu, B.; Jiang, H.; Tan, M. Roﬂumilast, a phosphodiesterases-4 (PDE4) inhibitor, alleviates sepsis induced acute kidney injury. Med. Sci. Monit. 2020, 26, e921319. [CrossRef] [PubMed] 24. Vanden Berghe, W.; Plaisance, S.; Boone, E.; De Bosscher, K.; Schmitz, M.L.; Fiers, W.; Haegeman, G. p38 and extracellular signal- regulated kinase mitogen-activated protein kinase pathways are required for nuclear factor-κB p65 transactivation mediated by tumor necrosis factor. J. Biol. Chem. 1998, 273, 3285–3290. [CrossRef] 25. Meteoglu, I.; Erdogdu, I.H.; Meydan, N.; Erkus, M.; Barutca, S. NF-κB expression correlates with apoptosis and angiogenesis in clear cell renal cell carcinoma tissues. J. Exp. Clin. Cancer Res. 2008, 27, 53. [CrossRef] 26. Liu, Z.; Tang, C.; He, L.; Yang, D.; Cai, J.; Zhu, J.; Shu, S.; Liu, Y.; Yin, L.; Chen, G.; et al. The negative feedback loop of NF-κB/miR-376b/NFKBIZ in septic acute kidney injury. JCI Insight 2020, 5, e142272. [CrossRef] 27. Dudley, C. Maximizing renal preservation in acute renal failure. BJU Int. 2004, 94, 1202–1206. [CrossRef] 28. Kim, J.M.; Kim, H.K.; Im, Y.N.; Bae, Y.S.; Im, S.Y.; Lee, H.K. FcγR/ROS/CK2α is the key inducer of NF-κB activation in a murine model of asthma. Int. Arch. Allergy Immunol. 2018, 175, 16–25. [CrossRef] 29. Zhang, Y.; Dees, C.; Beyer, C.; Lin, N.Y.; Distler, A.; Zerr, P.; Palumbo, K.; Susok, L.; Kreuter, A.; Distler, O.; et al. Inhibition of casein kinase II reduces TGFβ induced ﬁbroblast activation and ameliorates experimental ﬁbrosis. Ann. Rheum. Dis. 2015, 74, 936–943. [CrossRef] 30. Rittirsch, D.; Huber-Lang, M.S.; Flierl, M.A.; Ward, P.A. Immunodesign of experimental sepsis by cecal ligation and puncture. Nat. Protoc. 2009, 4, 31–36. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijerph16152767	Article The Role of Parenting Styles on Behavior Problem Proﬁles of Adolescents Bárbara Lorence 1, Victoria Hidalgo 1 , Javier Pérez-Padilla 2,* and Susana Menéndez 3 2 1 Faculty of Psychology, University of Seville, Calle Camilo José Cela, s/n, 41018 Sevilla, Spain Faculty of Humanities and Education Sciences, University of Jaen, Campus Las Lagunillas, s/n, 23071 Jaén, Spain Faculty of Education, Psychology and Sports Sciences, University of Huelva, Campus El Carmen, Avenida de las Fuerzas Armadas, s/n, 21007 Huelva, Spain * Correspondence: [email protected]; Tel.: +34-953-212-614 3 Received: 11 June 2019; Accepted: 1 August 2019; Published: 2 August 2019 Abstract: Parental behavior is one of the most inﬂuential factors on the development of adolescent externalizing and internalizing behavior problems. These behavioral problems are closely related and often co-occur. The objectives of this work were: (i) to identify adolescents proﬁles according to their behavior problems; (ii) to explore individual, family, and social characteristics associated with these proﬁles; and (iii) to analyze the potential role of parenting styles in belonging to adolescents’ proﬁles. A total of 449 Spanish adolescents (223 from families declared at-risk and enrolled in Child Welfare Services and 226 from families from the general population) participated in this study. The analyses revealed three proﬁles of adolescents based on external and internal behavior problems (adjusted, external maladjustment, and internal maladjustment). Parenting styles explained the adolescents’ belonging to diﬀerent proﬁles, in which the indulgent style was the most favorable in general terms. The distinctive role of parenting styles on two types of maladjustment proﬁles was conﬁrmed. The relationship between parenting styles and adolescent adjustment is a key component that should be included in interventions according to adolescents’ behavior problem proﬁles. Furthermore, the results shed light on the need that family interventions are complemented with individualized interventions with adolescents that accumulate stressful life events. Keywords: parenting styles; adolescent; behavior problems; stressful life events 1. Introduction There is broad consensus that the goal of parenting is to establish positive relationships with children and adolescents within the family to ensure their development and well-being. The contemporary concept of positive parenting [1] implies that parent–child relationships should be based on aﬀection, support, communication, stimulation, and structuring in routines, in the establishment of limits, norms, and consequences, as well as in the involvement in the daily life of children and adolescents [2]. However, positive parenting is a diﬃcult task, especially during adolescence, when there is a tendency toward an increase in family conﬂict, which is due in part to developmental changes and new challenges faced by boys and girls [3,4]. In view of this reality, current family support policies in most countries provide parental education programs for the parents of adolescents [1]. In these interventions, parenting styles are often a central component in promoting positive parenting. This paper explores the relationship between parenting styles and adolescent adjustment proﬁles, with the overall aim of identifying key components of parenting styles for preventing internalizing and externalizing problems. Int. J. Environ. Res. Public Health 2019, 16, 2767; doi:10.3390/ijerph16152767 www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2019, 16, 2767 2 of 17 1.1. Psychosocial Adjustment in Adolescence: Behavior Problems and Self-Concept Normative changes that occur during adolescence have been related to an increase in self-perceived maladjustment during this developmental stage. Psychological adjustment in adolescence is explained by the interaction of the multiple changes that take place at these ages [5,6]. In terms of self-development, there tends to be a drop in boys’ and girls’ assessment of themselves during early adolescence, with this gradually recovering as they reach adulthood [7]. Self-concept is an indicator of adjustment associated with behavior problems [8]. These types of problems aﬀect many adolescents [9], although few persist with behavioral problems into adulthood [10,11]. There is an increase in externalizing problems, such as delinquency, substance abuse, aggressiveness, or breaking of rules [12,13]; and internal problems related to social isolation, withdrawal, anxiety, and depression [14,15]. It has been found that the prevalence of certain disorders follows diﬀerent patterns in males and females. Most of the studies that make sex comparisons highlight a higher rate of internalizing problems in girls [16,17] and a higher rate of externalizing problems in boys [6,17–19], although not all recent studies found such a clear-cut diﬀerence along sex lines [20,21]. The co-occurrence, which is referred to the simultaneous presence of both internalizing and externalizing problems, has also come under close scrutiny in adolescent studies [9,22]. The previous research goals tend to focus exclusively on internalizing or externalizing problems, without taking into account the development of co-occurring symptoms for some children. The studies of the etiology or treatment of internalizing and externalizing symptoms in adolescents should account for co-occurrence (e.g., identifying existing adolescent adjustment proﬁles). In addition to normative changes, several studies report a strong association between the accumulation of stressful life events in adolescents and both internalizing and externalizing problems, regardless of sex [23–26]. The accumulative risk hypothesis understands that experiencing stressful circumstances has a negative impact on people’s development, so that the greater the number of stressful events, the greater the presence of clinical problems [27]. Studies with adolescents living in at-risk families enrolled in Child Welfare Services (CWS) found complicated risk trajectories in these youngsters, ﬁnding that they were particularly susceptible to developing this type of behavioral problem [28–30]. That high level of stress may be more harmful to the development of boys and girls in these surroundings than to the adults themselves due, amongst other reasons, to adolescents’ lack of psychosocial maturity to cope with these adversities [28–30]. Self-concept studies have shown that those adolescents who have experienced a greater number of signiﬁcant negative experiences, particularly in areas which are important to them, present more diﬃculties and ﬂuctuations in self-concept [7]. 1.2. Parenting and Psychosocial Adjustment in Adolescence Parental behavior is one of the most inﬂuential factors in terms of the development of externalizing and internalizing behaviors of adolescents. A great amount of research has focused on which parenting styles are best at promoting the growth and development of adolescents [31,32]. The typological perspective of parenting is the most widely used, and includes a constellation of parental dimensions such as aﬀection, punishment, dialogue, or indiﬀerence [31,32]. This perspective allows a multidimensional approach on parenting [33]. the relations between parenting styles and child adjustment have been studied from the classical two-dimensional orthogonal model of parental socialization. Based on these two dimensions (acceptance/involvement and strictness/imposition), Musitu and García [34] identiﬁed four parental socialization styles: authoritative (characterized by the use of high strictness/imposition and high acceptance/involvement); neglectful (low strictness/imposition and low acceptance/involvement); indulgent (low strictness/imposition and high acceptance/involvement); and authoritarian (high strictness/imposition and low acceptance/involvement). This typology emerges from a contextual [31] and situational [35] approach to parenting. Traditionally, The relation between parenting styles and adolescent adjustment has a great deal of empirical evidence. In general, ﬁndings from previous studies have shown better adolescent outcomes with authoritative parents and poorer with neglectful parents; while the authoritarian and indulgent Int. J. Environ. Res. Public Health 2019, 16, 2767 3 of 17 styles have occupied an intermediate position [36–39]. In general, the available empirical data show adolescents from authoritative families perform better academically [40] are more optimistic [41], make better use of adaptive strategies [36], have less drug use [38], higher self-concept [42], more behavioral regulation [43] and are more resilient [44] than adolescents from families with other parenting styles. Additionally, this relationship remains invariant with regard to the sex of the participants in previous studies. [36,45,46]. Although in general terms the positive eﬀects of the authoritative style and the negative eﬀects of the neglectful style have been widely documented, some studies have shown variations in the eﬀects of the diﬀerent styles depending on the type of behavioral problem analyzed. In relation to externalizing problems, the review by Marcone, Aﬀuso, and Borrone [43] showed that the most positive style for promoting external adolescent adjustment is the authoritative, and the most harmful is the authoritarian style. This same review did not ﬁnd such clear-cut results in relation to the indulgent (sometimes associated with positive eﬀects) and neglectful styles (sometimes associated with negative eﬀect styles). These authors concluded that parental control has been linked to beneﬁcial eﬀects if it is exerted as supervision, behavioral control, or guide; in contrast, psychological, coercive, or restrictive control is associated with externalizing behaviors. There is less research into internalizing problems. The systematic review by Rose, Roman, Mwaba and Ismail [47] of the 2012–2015 period only found ﬁve studies focusing speciﬁcally on parenting styles and their relation to internalization (articles with children over the age of 12 years and which had a medical focus). These authors found that the lowest levels of internalizing problems were found in adolescents with authoritative parents, and the uninvolved parenting style (neglectful) was associated to the highest level of internalizing problems. In turn, they also found that hostile and punitive parenting was identiﬁed as a risk factor for internalization problems in several studies. Despite these results, the systematic review performed by Yap, Pilkington, Ryan, and Jorm [48] showed that the empirical evidence of the relation between authoritarian, authoritative, and indulgent styles and internalizing problems is weak and remains equivocal. The authors concluded that more empirical evidence is needed about the role of parenting styles in belonging to internalizing problems, because other signiﬁcant factors have been identiﬁed for these types of behavior problems (temperamental response styles and emotion regulation strategies) [49]. The association between parenting and behavior problems seems to vary when co-occurrence of internalizing and externalizing problems is controlled. Likewise, Lorence et al. [50] found that the association between parenting and externalizing problems remains invariant once this co-occurrence is controlled, but not for internalizing problems. These authors found that the eﬀect of parenting styles on internal adjustment disappeared while the inﬂuence of other variables, as the emotional impact of stressful life events accumulation, is maintained. Finally, the role of a parenting style varies according to the socio-cultural context [51–55]. Some studies in European or Latin American countries support the indulgent style as the optimum parenting style [56–61]. Parenting practices based on acceptance, support, and reasoned communication promote adolescent adjustment regardless of the cultural context, although according to García and Gracia [59], this is particularly beneﬁcial in collectivist cultures, where cooperation and egalitarian relations are emphasized [58,62]. However, ﬁndings about the eﬀects of parental control are more controversial when comparing diﬀerent cultures [55,59,63,64] and speciﬁc social environments [65–67]. For all of the above, the results on parenting should be understood in the socio-cultural context (country) in which these are studied. 1.3. Objectives This research is proposed to add knowledge to the relationships between mothers’ parenting styles and adolescent adjustment in Spain. Speciﬁcally, the objectives of this study were the following: • • To identify proﬁles of adolescents according to their behavioral problems. To explore the individual, family, and social characteristics of these proﬁles. Int. J. Environ. Res. Public Health 2019, 16, 2767 4 of 17 • To analyze the potential role of parenting styles in belonging to adolescents’ proﬁles. 2. Methods 2.1. Participants A total of 449 adolescents (211 females and 238 males; M = 13.65 years, SD = 1.89) from southern Spain participated in this research. The sample was split into two groups: 223 adolescents living in families declared at-risk and enrolled in Child Welfare Services (CWS), and 226 adolescents living in families not receiving intervention (community adolescents). Both groups were similar in age (t(447) = −0.250, p = 0.803) and sex (χ2(1, N = 449) = 3.95, p = 0.208). In regard to the families of the participants, the average number of children under 18 in these homes was 2.11 (SD = 0.95), and 29.9% were single-parent families (100% single-mother families). The fathers’ mean age was 44.94 years (SD = 6.60), the majority had completed primary education (31.3%) or had ﬁnished secondary education (29.9%), and 89% of the fathers were working. Regarding the mothers, the mean age was 42.15 years (SD = 6.86), and 21.7% had not completed primary education, 37.6% had completed primary education, 25.5% had ﬁnished secondary education, and 15.2% had a university degree. As for employment, 62.4% of the mothers were working at the time. 2.2. Measures Socio-demographic proﬁle: We developed a questionnaire to collect socio-demographic information about the adolescent (age, sex, and educational level), family (structure of family) and parents (age, sex, educational level, and employment situation). A short version of the Inventory of Stressful Life Events (ISLE, [68]) was administered to adolescents. Fifteen items from the original 29 negative or potentially problematic events were included in this study (situations not directly controllable by the adolescent have been chosen), covering stressful experiences at individual (“severe illness”, “sexual abuse”), family (“family member’s addiction”, “move”, “parental divorce”, “parental conﬂicts”, “new couple”, “economic hardship”, “death of family members”, “illness of family members”, “a family member moves/leaves home), and peer-related levels (“isolation”, “couple breakup”, “couple betrayal”, “ﬁght with friends”). The occurrence of these kind of events in the previous ﬁve years was rated using a dichotomous scale (0 = no, 1 = yes). The calculation of the accumulated score of SLEs refers to the total number of stressful life events experienced by the subject. Parental Socialization Scale for Adolescents (ESPA29, [34]). This instrument assesses parental socialization strategies in 29 situations of everyday family life: 13 compliance situations (e.g., “a teacher calls your mother and tells her that you are behaving well in class”) and 16 non-compliance situations (e.g., “you arrive home late”). In each of the compliance situations, the adolescent had to rate the parenting practices of Aﬀection (“my mother shows aﬀection”) and Indiﬀerence (“my mother seems indiﬀerent”), and in each of the non-compliance situations, the adolescent had to rate the parenting reactions on Dialogue (“my mother talks to me”), Detachment (“my mother doesn’t tell me anything”), Verbal Scolding (“my mother scolds me”), Physical Punishment (“my mother spanks me”), and Revoking Privileges (“my mother takes something away from me”). ESPA29 consists of 116 items ranging from 1 (never) to 4 (always), providing us with two parenting dimensions: (1) the Acceptance/Involvement dimension, the result of averaging the responses on Aﬀection, Dialogue, Indiﬀerence, and Detachment (the last two subscales are reverse coded); and (2) the Strictness/Imposition dimension, by averaging the responses on Verbal Scolding, Physical Punishment, and Revoking Privileges. According to the ESPA29 original authors [34], after dichotomizing the total sample on acceptance/involvement and strictness/imposition, and examining the two parenting dimensions simultaneously, four parenting styles emerged: Authoritative (above the 50th percentile on both dimensions), Authoritarian (above the 50th percentile on strictness/imposition but below the 50th percentile on acceptance/involvement), Indulgent (above the 50th percentile on acceptance/involvement Int. J. Environ. Res. Public Health 2019, 16, 2767 5 of 17 but below the 50th percentile on strictness/imposition) and Neglectful (below the 50th percentile in both dimensions). This study is focused exclusively on mothers’ parenting styles (main caregiver according to adolescents). Youth Self-Report (YSR, [18]) assesses behavior problems in adolescence. YSR consists of 112 items with response options ranging from 0 (not true) to 2 (very true), which measure eight behavior problems: Withdrawn (e.g., “I would rather be alone than with others”), Somatic complaints (e.g., “I feel dizzy or lightheaded”), Anxiety and depression (e.g., “I cry a lot”), Social problems (e.g., “I’m too dependent on adults”), Thought problems (e.g., “I see things that other people think aren’t there”), Attention problems (e.g., “I feel confused or in a fog”), Aggressive behavior (e.g., “I destroy things belonging to others”), and Rule-breaking behaviors (e.g., “I steal from places other than home”). The ﬁrst three subscales are related to Internalizing problems, whereas the last two correspond to Externalizing problems. Cronbach’s alphas in this study were α = 0.79 for internalizing problems and α = 0.86 for externalizing problems. Multidimensional Self-Concept Scale (AF5, [69]). This scale considers that self-concept has diﬀerent, but related, components. This instrument consists of 30 items with a ﬁve-point response scale (ranging from 1 “complete disagreement” to 5 “complete agreement”), which is designed to measure ﬁve self-concept dimensions: Academic, Social, Emotional, Family, and Physical. Cronbach’s alpha in this study for each dimension was: Academic, 0.82, Social, 0.70, Emotional, 0.77, Family, 0.80, and Physical, 0.78. 2.3. Procedure Adolescents from at-risk families were selected by CWS professionals according to the following inclusion criteria: (1) adolescents between 11–17 years old, without a diagnosed mental health condition; and (2) living in a family being assessed by CWS professionals at a medium level of psychosocial risk. Community adolescents were selected based on a random sample stratiﬁed by conglomerates, bearing in mind (in addition to age and sex) the type of school. Fourteen schools were chosen randomly in southern Spain, and the selection of participants was also random between students per academic course of each school. Adolescents from both samples were living in the same neighborhoods. All the subjects participated in the study voluntarily, after signing an informed consent form in accordance with the Declaration of Helsinki. The aims of the research project were explained, and all the participants were assured that their anonymity would be protected. Ethics approval was obtained from the ethics committee of the Andalusian Health Services (code 22/0509). No monetary incentives were oﬀered. 2.4. Data Analyses Statistical analyses were conducted by using IBM SPSS 20 software for Windows [70]. For the univariate analyses, the descriptive statistics were presented through the means, standard deviations, and the minimum and maximum values (Min–Max) of the quantitative variables, thereby describing their central tendency, variability, and range. For the bivariate analyses, t-tests for two independent samples were computed, using Student’s t-statistic. For the multivariate analyses, a series of cluster analyses were computed in order to describe the joint variability of the sample with respect to behavior problems. Cluster analysis is a multivariate classiﬁcation technique that identiﬁes groups of individuals deﬁned by similarities on multiple dimensions, so that members of the resulting groups are as similar as possible to others within their group (high within-group homogeneity) and as diﬀerent as possible to those in the other groups (low between-group homogeneity) [33]. Prior to clustering, all the selected measures were z-standardized to equate the variables. Although this statistical technique is very robust if assumptions for parametric statistics are violated, following Pérez [71], both the presence of inﬂuential extreme cases and the existence of linearity problems were examined before computing the analysis. Initial groupings were obtained using hierarchical cluster analysis with squared Euclidean distance, Int. J. Environ. Res. Public Health 2019, 16, 2767 6 of 17 and the nearest-neighbor method was used as the measure of linkage. The best solution (number of clusters) was determined by examining both the agglomeration schedule and the dendogram. Then, the centroids of these initial clusters were submitted to an iterative clustering procedure (K-means cluster analysis) to reﬁne ﬁnal cluster membership. Once the cluster dimensions had been determined through conﬁrmatory K-means clustering, the quality of the solution was checked by analyzing atypical subjects, calculating the t-distribution of the sample, comparing standardized residuals for all the variables selected, and checking for outliers. To analyze the diﬀerences between clusters, post hoc ANOVAs were computed with the selected psychosocial variables. Finally, multinomial logistic models were performed, taking into consideration deviance distribution to calculate goodness-of-ﬁt, as well as the rate of correct classiﬁcation of the observed and predicted subjects of the resulting models. Nagelkerke’s pseudo-R2 statistic was used to assess the resulting models’ degree of explanation. After creating the models and satisfactory conﬁrming its viability, the meaning and direction of the coeﬃcients using the Wald statistic and odds ratios (OR) were examined. 3. Results 3.1. Characteristics of Adolescents: Parenting Styles, Stressful Life Events, Behavior Problems, and Self-Concept The results of the descriptive analyses of the individual, family, and social aspects evaluated in the study are presented in Table 1. It shows that parenting styles were homogeneously distributed, with the Neglectful style being the most frequently reported and Indulgent being the least. On the other hand, adjustment problems obtained a similar average with moderate variability in their scores. In addition, the analyses focusing on the dimension of self-concept reflected that the dimension referring to the family presented the highest average score, but also the lowest academic score. To conclude the descriptive analyses, adolescents had experienced an accumulation of four SLE over the previous five years. Table 1. Descriptions of parenting styles, behavior problems, self-concept, and stressful life events. Parenting styles Authoritative Indulgent Authoritarian Neglectful Stressful life events (SLE) Accumulation SLE Behavior problems Internalizing Externalizing Self-Concept Family Social Emotional Academic Physical Percentage 28.73% 17.15% 22.49% 30.73% M 4.19 0.41 0.41 4.37 4.09 3.53 3.31 3.34 SD Minimum Maximum 2.52 0.22 0.26 0.68 0.65 0.77 0.96 0.79 0 0 0 1.33 1.67 1 1 1.17 14 1.26 1.44 5 5 5 5 5 Int. J. Environ. Res. Public Health 2019, 16, 2767 7 of 17 3.2. Two-Step Cluster Analysis: Adjustment Proﬁles of Adolescents To accomplish the ﬁrst aim of this study—to identify adolescent proﬁles according to externalizing and internalizing problems—two-step cluster analysis was computed. The box and whiskers plots were examined for each variable, and the calculation of the Mahalanobis distance did not reveal the existence of either univariate or multivariate extreme cases, respectively. The bivariate correlation indices between the internal and external adjustment problems (r = 0.39) did not exceed the value r = 0.80, indicating that there were no collinearity problems. After computing a hierarchical agglomerative cluster analysis, and repeating the procedure with diﬀerent clustering methods (between-groups linkage, furthest neighbor, and Ward’s method), the visual examination of each dendogram revealed the existence of three initial clusters. To conﬁrm this solution, a K-means cluster analysis was performed requesting the deﬁnition of three groups. Iteration history showed that convergence (and thus the absence of center changes in each cluster) was achieved in the 10th iteration. Table 2 contains the results of the means in each cluster of behavior problems and an analysis of variance between the conglomerates. Furthermore, Figure 1 shows the centers of the clusters resulting from the analysis. Table 2. Descriptive and ANOVA between the conglomerates according to internalizing and externalizing problems. AG (49.67%) M (SD) MEPG (22.05%) M (SD) MIPG (26.95%) M (SD) ANOVA F (df) Bonferroni Internalizing problems 0.26 (0.11) 0.45 (0.19) 0.66 (0.17) Externalizing problems 0.24 (0.13) 0.77 (0.19) 0.41 (0.17) 285.95 * (2,440) 406.09 * (2,440) 1–2 * 1–3 * 2–3 * 1–2 * 1–3 * 2–3 * * p < 0.001; AG: Adjusted Group, MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problems Group. Figure 1. Conglomerate centers oﬀered in typiﬁed scores. As Table 2 shows, the three groups were formed diﬀerentially depending on both the internal and external maladjustment dimensions. The ﬁrst cluster (AG; Adjusted Group) diﬀered signiﬁcantly from the other two, indicating lower scores of internalizing and externalizing problems. The second cluster (MEPG; Maladjustment with Externalizing Problems Group) was characterized by medium Int. J. Environ. Res. Public Health 2018, 15, x FOR PEER REVIEW 8 of 17 Figure 1. Conglomerate centers offered in typified scores. To complete the analysis, we computed bivariate tests including the three resulting clusters and some individual, family, and social variables (see Table 3). First, we observed that adolescents were not significantly distributed according to sex (χ2 = 3.41, p = 0.182). As Table 3 shows, adolescents from the AG tended to be community and experience authoritative and indulgent parenting styles, high levels of self-concept and a low accumulation of SLE, and did not perceive authoritarian and neglectful styles. In turn, the MEPG consisted mainly of at-risk adolescents supported by CWS with low self-concept (barring the social one) and a greater accumulation of SLEs. Regarding parenting styles, the adolescents were characterized as experiencing a neglectful parenting style if they did not perceive authoritative and indulgent styles. Lastly, the MIPG cluster was not defined by adolescent background and parenting styles; however, their social, emotional, and physical self-concept tended to be low compared to the other groups, and was characterized by an accumulation of SLEs similar to the MEPG and superior to the AG. Table 3. Descriptive and bivariate analyses including the three resulting clusters and different individual, family, and social variables. AG (49.67%) MEPG (22.05%) MIPG (26.95%) Crosstabs %(ras) %(ras) %(ras) χ2 (df) V Cramer Background Community 30.70% (4.2) 7.67% (−3.8) 12.64% (−1.2) 20.97 * (2, 441) 0.22 *** Child Welfare Services 19.64% (−4.2) 14.67% (3.8) 14.67% (1.2) Parenting styles Authoritative (yes/no) 17.08%/33.71% (2.1/−2.1) 3.87%/17.99% (−2.8/2.8) 8.20%/19.13% (0.2/−0.2) 8.29 * (2, 437) 0.14 * Indulgent (yes/no) 11.85%/38.95% (3.4/−3.4) 1.59%/20.27% (−2.9/2.9) 3.87%/23.46 (−1.1/1.1) 13.18 (2, 437) 0.17 Authoritarian (yes/no) 9.11%/41.68% (−2.6/2.6) 6.38%/15.49% (1.6/−1.6) 7.52%/19.82% (1.4/−1.4) 6.60 * (2, 437) 0.12 * Neglectful (yes/no) 12.76%/38.04% (−2.5/2.5) 10.02%/11.85% (3.7/−3.7) 7.74%/19.58% (−0.6/0.6) 13.96 (2, 437) 0.18 ANOVA M (SD) M (SD) M (SD) F(df) Bonferroni Age 13.40 (1.89) 14.25 (1.84) 13.66 (1.92) 5.41 (2, 440) 1–2 * 1–3 2–3 -1.5-1-0.500.511.5AGMEPGMIPGInternalizing behavioursExternalizing behavioursInt. J. Environ. Res. Public Health 2019, 16, 2767 8 of 17 scores in internalizing problems and higher scores in externalizing problems. The third cluster (MIPG; Maladjustment with Internalizing Problems Group) revealed high scores of internalizing problems, and medium scores of externalizing problems (see Figure 1). To complete the analysis, we computed bivariate tests including the three resulting clusters and some individual, family, and social variables (see Table 3). First, we observed that adolescents were not signiﬁcantly distributed according to sex (χ2 = 3.41, p = 0.182). As Table 3 shows, adolescents from the AG tended to be community and experience authoritative and indulgent parenting styles, high levels of self-concept and a low accumulation of SLE, and did not perceive authoritarian and neglectful styles. In turn, the MEPG consisted mainly of at-risk adolescents supported by CWS with low self-concept (barring the social one) and a greater accumulation of SLEs. Regarding parenting styles, the adolescents were characterized as experiencing a neglectful parenting style if they did not perceive authoritative and indulgent styles. Lastly, the MIPG cluster was not deﬁned by adolescent background and parenting styles; however, their social, emotional, and physical self-concept tended to be low compared to the other groups, and was characterized by an accumulation of SLEs similar to the MEPG and superior to the AG. 3.3. Parenting Styles and Group Membership Multinomial logistic regression analyses were performed to analyze the potential role of parenting styles in adolescent’s belonging to each cluster. The resulting models estimates the factors associated with the probability that an adolescent from the maladjusted groups will become part of the AG, that is, the one characterized by better indicators in terms of internalizing and externalizing problems. The ﬁrst important indicator that allows us to ascertain the correspondence of the models with the data is the goodness-of-ﬁt, which is measured through the deviance (Authoritative Model, χ2 (46) = 43.86, p = 0.562; Indulgent Model, χ2 (44) = 39.17, p = 0.678; Authoritarian Model, χ2 (48) = 42.74, p = 0.688; and Neglectful Model, χ2 (46) = 38.87, p = 0.763), concluding that the models were adequate. Final model likelihood ratio tests (−2LL) were statistically diﬀerent from the initial ones for all models (Authoritative Model, χ2 (4) = 43.54, p < 0.001; Indulgent Model, χ2 (4) = 46.41, p < 0.001; Authoritarian Model, χ2 (4) = 38.10, p < 0.001; and Neglectful Model, χ2 (4) = 50.64, p < 0.001). Thus, we proceeded to interpret the models by examining the OR scores, which allowed us to establish whether each selected variable constituted, on its own, an element that increased the probability of belonging to the reference group. The factors associated with adjustment problems, taking the AG as reference, are presented in Table 4. Speciﬁcally, the Authoritative Model explained 10.86% of the variance of the scores, and correctly predicted 53.09% of the subjects belonging to its reference group. As Table 4 shows, the analysis of the OR values of the MEPG showed that the presence of the authoritative style increased the probability that an adolescent would become part of this AG by 77%. This relationship was not observed in the MPIG. Likewise, for each increase in an SLE, the probability that an adolescent belong to the MEPG or MIPG increased by 32% and 22%, respectively. The Indulgent Model explained 11.54% of the variance and correctly predicted 52.17% of the subjects belonging to its reference group. As Table 5 shows, the analysis of OR values showed that the presence of the indulgent style increased the probability that an adolescent from the MEPG would become part of the AG by 72%, and that an adolescent from the MIPG would become part of the AG by 43%. In turn, for each increase in an SLE, the probability that an adolescent belonged to the MEPG or MIPG, increased by 31% and 21%, respectively. Int. J. Environ. Res. Public Health 2019, 16, 2767 9 of 17 Table 3. Descriptive and bivariate analyses including the three resulting clusters and diﬀerent individual, family, and social variables. Background Community Child Welfare Services Parenting styles Authoritative (yes/no) Indulgent (yes/no) Authoritarian (yes/no) Neglectful (yes/no) AG (49.67%) %(ras) MEPG (22.05%) %(ras) MIPG (26.95%) %(ras) Crosstabs χ2 (df) V Cramer 30.70% (4.2) 19.64% (−4.2) 7.67% (−3.8) 12.64% (−1.2) 14.67% (3.8) 14.67% (1.2) 20.97 *** (2, 441) 17.08%/33.71% (2.1/−2.1) 11.85%/38.95% (3.4/−3.4) 9.11%/41.68% (−2.6/2.6) 12.76%/38.04% (−2.5/2.5) 3.87%/17.99% (−2.8/2.8) 1.59%/20.27% (−2.9/2.9) 6.38%/15.49% (1.6/−1.6) 10.02%/11.85% (3.7/−3.7) 8.20%/19.13% (0.2/−0.2) 3.87%/23.46 (−1.1/1.1) 7.52%/19.82% (1.4/−1.4) 7.74%/19.58% (−0.6/0.6) 8.29 * (2, 437) 13.18 ** (2, 437) 6.60 * (2, 437) 13.96 (2, 437) 0.22 * 0.14 * 0.17 ** 0.12 * 0.18 M (SD) M (SD) M (SD) F(df) Bonferroni ANOVA Age 13.40 (1.89) 14.25 (1.84) 13.66 (1.92) 5.41 (2, 440) Stressful life events Accumulation SLE 3.52 (2.23) 5.25 (2.50) 4.62 (2.68) Self-Concept Family 4.58 (0.46) 3.99 (0.84) 4.34 (0.70) Social 4.16 (0.55) 4.19 (0.73) 3.86 (0.72) Emotional 3.79 (0.62) 3.47 (0.76) 3.07 (0.80) Academic 3.58 (0.84) 2.61 (0.99) 3.38 (0.88) Physical 3.45 (0.74) 3.31 (0.87) 3.17 (0.77) 19.75 * (2, 438) 30.57 * (2, 438) 10.66 * (2, 439) 40.51 * (2, 434) 39.48 * (2, 431) 5.12 ** (2, 438) 1–2 * 1–3 2–3 1–2 * 1–3 * 2–3 1–2 * 1–3 2–3 * 1–2 1–3 * 2–3 * 1–2 1–3 * 2–3 * 1–2 * 1–3 2–3 * 1–2 *** 1–3 * 2–3 ras = adjusted standardized residuals; p < 0.05, p < 0.01, **p < 0.001; AG: Adjusted Group, MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problem Group. The Authoritarian Model explained 9.56% of the variance and correctly predicted 51.49% of the subjects belonging to its reference group. As Table 6 shows, the analysis of the OR values showed that the presence of the authoritarian style increased the probability that an adolescent from the AG would become part of the MEPG or MIPG by 58% and 53%, respectively, although the p-values did not report statistical signiﬁcance. In addition, for each increase in SLE, the MEPG and MIPG increased by 31% and 21%, respectively. Int. J. Environ. Res. Public Health 2019, 16, 2767 10 of 17 Table 4. Multinomial logistic regression model parameters using the AG as a reference and authoritative style as covariate. Authoritative Model B χ2 Wald p OR OR LB 95% OR UB 95% MEPG Intercept SLE Authoritative style MIPG Intercept SLE Authoritative style −1.81 0.28 −0.88 −1.36 0.20 −0.19 44.62 28.82 8.42 32.04 17.29 0.59 <0.001 <0.001 0.004 <0.001 <0.001 0.442 1.32 0.23 1.22 0.83 1.19 0.23 1.11 0.51 1.46 0.75 1.34 1.34 MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problem Group. Table 5. Multinomial logistic regression model parameters using the AG as a reference and indulgent style as covariate. Indulgent Model B χ2 Wald p OR OR LB 95% OR UB 95% MEPG Intercept SLE Indulgent style MIPG Intercept SLE Indulgent style −1.80 0.27 −1.28 −1.29 0.19 −0.56 47.81 28.39 9.86 31.53 17.24 3.63 <0.001 <0.001 0.002 <0.001 <0.001 0.057 1.31 0.28 1.21 0.57 1.18 0.12 1.11 0.32 1.44 0.62 1.39 1.01 MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problem Group. Table 6. Multinomial logistic regression model parameters using the AG as a reference and authoritarian style as covariate. Authoritarian Model B χ2 Wald p OR OR LB 95% OR UB 95% MEPG Intercept SLE Authoritarian style MIPG Intercept SLE Authoritarian style −2.09 0.27 0.46 −1.48 0.19 0.42 64.40 28.63 2.71 42.47 16.64 2.68 <0.001 <0.001 0.100 <0.001 <0.001 0.101 1.31 1.58 1.21 1.53 1.18 0.92 1.10 0.92 1.44 2.73 1.32 2.54 MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problem Group. Lastly, the Neglectful Model explained 12.53% of the variance and correctly predicted 53.55% of the subjects belonging to its reference group (AG). As Table 7 shows, the analysis of OR values showed that the presence of the neglectful style increased the probability that an adolescent from AG would become part of the group with externalizing problems (MEPG) by 190%, while no statistical signiﬁcance was observed in the group with internalizing problems (MIPG). Likewise, for each increase Int. J. Environ. Res. Public Health 2019, 16, 2767 11 of 17 in an SLE, the probability that an adolescent belong to the MEPG and MIPG, increased by 34% and 22% respectively. Table 7. Multinomial logistic regression model parameters using the AG as a reference and neglectful style as covariate. Neglectful Model B χ2 Wald p OR OR LB 95% OR UB 95% MEPG Intercept SLE Neglectful style MIPG Intercept SLE Neglectful style −2.46 0.29 1.06 −1.49 0.20 0.26 76.83 34.90 18.40 41.98 19.85 1.18 <0.001 <0.001 <0.001 <0.001 <0.001 0.276 1.34 2.90 1.22 0.81 1.22 1.78 1.12 0.81 1.48 4.72 1.33 2.08 MEPG: Maladjustment with Externalizing Problems Group, MIPG: Maladjustment with Internalizing Problem Group. 4. Discussion Within the framework of a broader study analyzing adolescence from a global and positive perspective, the central objective of this work was to analyze the diﬀerential eﬀects of parenting styles on adolescent maladjustment in Spain. Speciﬁcally, diﬀerent proﬁles of adolescents have been identiﬁed considering internal and external adjustment problems together, exploring the individual, family, and social attributes that characterize these proﬁles, and studying the potential role of parenting styles and the accumulation of stressful events in belonging to these proﬁles. While the normative changes that take place in adolescence are conducive to the emergence of adjustment problems [12–15], not all adolescents experience them in the same way. In this sense, the analyses related to the ﬁrst objective of the study showed the existence of three diﬀerent adolescent proﬁles depending on the prevalence of the type of adjustment problems. Speciﬁcally, while the ﬁrst group (AG) of adolescents was characterized by low levels of both internal and external maladjustment, the analyses carried out showed two other groups with medium–high levels of adjustment problems. The second group (MEPG) was characterized by moderate levels of internal maladjustment and high external adjustment problems, while the third group (MIPG) had moderate levels of external maladjustment and high levels of internal maladjustment. These results underline the importance of studying together both internal and external adjustment problems in adolescence, as they often co-occur in adolescence [9,22]. The second objective of this study was to characterize each cluster by variables on the individual (sex, age, self-concept, stressful life events), family (parenting styles), and social level (community or CWS users). On the individual level, we did not found diﬀerences in sex between clusters. Although studies analyzing how internal and external adjustment problems are related to sex have tended to identify the former with girls [16,17] and the latter with boys [6,17–19], the relevance of this socio-demographic dimension seems to be disappear when taking into account the co-occurrence of adjustment problems. Self-concept is another of the individual dimensions that has traditionally been considered as an indicator of the presence of behavioral problems [8]. In the present study, adolescents from diﬀerent clusters scored diﬀerently in self-concept, with AG scoring better than the maladjusted (MEPG and MPIG) in most of the dimensions comprising self-concept. In addition, these results reveal the presence of a complex proﬁle in adolescents with behavioral problems. On the one hand, the members of the MEPG had a negative perception of their family and academic self-concept. These dimensions refer to the two most relevant developmental contexts for children (family and school) Int. J. Environ. Res. Public Health 2019, 16, 2767 12 of 17 where, amongst other issues, they have to comply with rules of behavior. Thus, the adolescents from the MEPG, which were characterized by problems in complying with established norms [12,13], deﬁned themselves more negatively due to their experience in these contexts. In turn, the members of the group characterized especially by internalizing problems (MIPG) stood out due to their low social and emotional self-concept. Problems of loneliness, anxiety, and depression present in boys and girls with these diﬃculties [14,15] may interfere in their self-description and their self-assessment in these important developmental areas. We observed diﬀerences in the distribution of parenting styles on the three adjustment proﬁles. The group of adjusted adolescents (AG) was characterized by indulgent and authoritative parenting styles, while those in the MEPG perceived a more neglectful style. These results follow the line of previous studies that stress the importance of the family environment, and more speciﬁcally, the type of parent–child interaction and how this aﬀects adolescent adjustment [36,37]. In addition, while the eﬀects of authoritative and neglectful styles on adolescent adjustment have been widely recognized in previous research [36–39], this paper also highlights the positive role of the indulgent style in deﬁning the group of adjusted adolescents (AG). In contrast to the studies in which the indulgent style is placed in an intermediate position, the results of this work demonstrate that this parental style is related to the absence of adjustment problems. A possible explanation for this is the socio-cultural environment in which this study was performed. Compared to other more individualistic socio-cultural environments, Spain is considered a country with a collectivist-type culture, and previous studies have shown that this style appears to be particularly favorable for adolescent adjustment [56–59,61]. In the social sphere, the adjusted group (AG) consisted mainly of adolescents from the community sample, while the MEPG was made up of users of CWS, and the MIPG comprises adolescents from both subsamples to equal extent. Thus, those adolescents who break rules and have less self-control and greater aggressiveness tend to be detected and attended by CWS to a greater extent than those who showed internal adjustment problems (loneliness, depression, anxiety...). However, adolescents from both MPEG and MIPG were characterized by a high accumulation of stressful life events around ﬁve negative events that were especially relevant in the lives of these boys and girls compared to the adjusted group (AG). The results regarding the accumulation of stressful life events follow the line found in other studies carried out both in community samples [23–26] and with CWS users [28,29]. As for the last objective of the study, and due to parental behavior being an important predictor of adolescent maladjustment, we explored whether parenting styles could explain the belonging to the groups. We also took into consideration the accumulation of stressful life events as an explanatory dimension. As a result, this study conﬁrms the accumulative hypothesis of risk and its impact on adolescent development [27], since in all of the computed analyses, a greater accumulation of stressful events predicted a change from the AG to the MIPG or MEPG. In turn, less experience of such stressful circumstances increased the likelihood of adolescents with maladjustment proﬁles (MIPG or MEPG) belonging to the AG. Regarding parenting styles, MEPG adolescents beneﬁt to the same degree from authoritative or indulgent parenting styles, because a greater presence of these styles meant they were more likely to pass from this proﬁle to a more adjusted one. In contrast, a greater degree of neglect increased the likelihood of moving from the adjusted group (AG) to the MEPG. These results follow the line of previous studies, which point, in general terms, to the positive eﬀects of the authoritative style and the negative eﬀects of the neglectful style (e.g., [43]), plus that the indulgent style has a similar inﬂuence to the authoritative style for this type of adjustment proﬁle. On the other hand, belonging to the MIPG was not signiﬁcantly related to any of the parenting styles. This result is similar to the conclusions of the review by Yap et al. [48], where they pointed out that empirical evidence regarding the relationship between parenting styles and internalizing problems is fragile and equivocal. The indulgent style is the only one that seems to be able to explain the internal adjustment problems. It is not surprising that a style centered on aﬀection and communication has some impact on the development of internal adolescent adjustment [56–59,61]. However, as mentioned above, there may Int. J. Environ. Res. Public Health 2019, 16, 2767 13 of 17 be other personal factors with greater predictive power concerning the adolescent [49], as in the case of the accumulation of stressful events [50]. This study has a number of limitations, amongst which is that the information has been obtained from a single informant. In addition, information has only been collected about the mother’s parenting style, ignoring the role of fathers in adolescent adjustment problems. Having data from diﬀerent members of the family system and about the parenting styles of both parents would have provided a clearer picture of the problems surrounding these adolescents, as well as how their families functioned. Additionally, it would have been interesting to examine the role of others personal measures of the adolescents (emotional regulation, coping strategies, or locus of control) as well as others related to their social support network (size, composition, or satisfaction). Finally, this study is cross-sectional, and a longitudinal approach would have allowed us to analyze the short-term and long-term eﬀects of parenting styles on adolescent adjustment proﬁles. Despite the aforementioned limitations, the main contribution of this study is its potential to improve interventions for families with adolescents. To prevent negative consequences in adulthood, it is essential to develop family interventions that provide adolescents that have problematic proﬁles with the resources they need to cope with stressful situations [27,28,30]. The results reported and discussed here suggest that it’s important to take into account the risk trajectories of adolescents, as well as their adjustment proﬁles, to tailor the objectives of family intervention programs. The power of the accumulation of stressful events to explain both internal and external maladjustment proﬁles makes it essential for interventions to have the empowerment of adolescents as a central objective. This study also highlights the importance of positive parenting practices based on aﬀection, support, communication, and dialogue for adolescent adjustment, in line with previous studies [2]. However, results obtained show that the eﬀects of diﬀerent parenting styles are not the same for adolescent with diﬀerent adjustment proﬁles. Thus, interventions should be tailored to the speciﬁc and diﬀering needs of adolescents with internalizing problems and those with a more externalizing maladjustment proﬁle. Speciﬁcally, the promotion of authoritative and indulgent parenting styles signiﬁcantly favors the improvement of the adjustment of adolescents with an external maladjustment proﬁle, but not for those with a maladjustment proﬁle in which internalizing problems predominate. 5. Conclusions The results reported here point out the diﬀering inﬂuence of parenting styles on adjustment during adolescence depending on the speciﬁc problems (internalizing versus externalizing) experienced by adolescents, and seems to be a central topic to the planning of positive parenting interventions. These results suggest the need to diversify the interventions to ﬁt the speciﬁc needs of adolescents with diﬀerent risk proﬁles and trajectories (accumulation of stressful life events). Girls and boys with complicated risk trajectories may require individual interventions, with specialists working speciﬁcally and directly with them. In addition, this work notes the need to combine this type of intervention with another aimed at promoting positive parenting styles (authoritative and indulgent) in families with adolescents who present a problematic, predominantly externalizing adjustment proﬁle. Author Contributions: Project administration, supervision and methodology: B.L., V.H. and S.M. Data curation: B.L., and J.P.-P. Resources: V.H. Conceptualization, formal analysis and writing: B.L., V.H., J.P.-P. and S.M. All authors have read and approved the ﬁnal manuscript. Funding: This work was also supported by the Spanish Government (MINECO, Ministry of Economics and Competitiveness). Project reference: EDU2013-41441-P. Acknowledgments: The authors would like to thank the families and professionals for their collaboration. Conﬂicts of Interest: The authors declare no conﬂict of interest. Int. J. Environ. Res. Public Health 2019, 16, 2767 14 of 17 References 1. 2. 3. 4. 5. 6. 7. 8. 9. Council of Europe. Recommendation Rec (2006) 19 of the Committee of Ministers to Members States on Policy to Support Positive Parenting. Available online: http://www.coe.int/es (accessed on 21 April 2016). Daly, M.; Bray, R.; Bruckauf, Z.; Byrne, J.; Margaria, A.; Pécnik, N.; Samms-Vaughan, M. Family and Parenting Support: Policy and Provision in a Global Context; Innocenti Insight, UNICEF Oﬃce of Research: Florence, Italy, 2015. Collins, W.A.; Laursen, B. Parent-adolescent relationships and inﬂuences. In Handbook of Adolescent Psychology; Lerner, R.M., Steinberg, L., Eds.; Willey: New York, NY, USA, 2004; pp. 331–361, ISBN 978-0-471-69044-3. Larson, R.W.; Richards, M.H.; Moneta, G.; Holmbeck, G.; Duckett, E. Changes in adolescents’ daily interactions with their families from ages 10 to 18: Disengagement and transformation. Dev. Psychol. 1996, 32, 744–754. [CrossRef] Achenbach, T.M.; Bird, H.R.; Canino, G.; Phares, V.; Gould, M. Epidemiological comparisons of Puerto Rican and US Mainland children: Parent, teacher, and self-reports. J. Am. Acad. Child. Adolesc. Psychiatry 1990, 29, 84–93. [CrossRef] [PubMed] Lemos, S.; Fidalgo, A.M.; Calvo, P.; Menéndez, P. Salud mental de los adolescentes asturianos Mental health in Asturian adolescents. Psicothema 1992, 4, 21–48. Rodríguez, C.; Caño, A. Autoestima en la adolescencia: Análisis y estrategias de intervención [Self-concept in adolescence: Analysis and intervention strategies]. Int. J. Psychol. Psychol. Therapy 2012, 12, 389–403. Garaigordobil, M.; Pérez, J.I.; Mozaz, M. Self-concept, self-concept and psychopathological symptoms. Psicothema 2008, 20, 114–123. [CrossRef] [PubMed] Reitz, E.; Dekovic, M.; Meijer, M. The structure and stability of externalizing and internalizing problem behavior during early adolescence. J. Youth Adolesc. 2005, 34, 577–588. [CrossRef] 10. Moﬃtt, T.E. Adolescence-limited and life-course persistent antisocial behavior: A developmental taxonomy. Psychol 1993, 100, 674–701. [CrossRef] 11. Hoeve, M.; Blokland, A.; Dubas, J.S.; Loeber, R.; Gerris, J.R.M.; Van der Laan, P. Trajectories of delinquency and parenting styles. J. Abnorm. Child. Psychol. 2008, 36, 223–235. [CrossRef] 12. Abad, J.; Forns, M.; Gómez, J. Emotional and behavioral problems as measured by the YSR: Gender and age diﬀerences in Spanish adolescents. Eur. J. Psychol. Assess. 2002, 18, 149–157. [CrossRef] 13. Rutter, M.; Giller, H.; Hagel, A. La Conducta Antisocial de los Jóvenes; Cambridge University Press: Madrid, Spain, 2000; ISBN 9788483231265. 14. Hamza, C.; Willoughby, T. Perceived parental monitoring, adolescent disclosure, and adolescent depressive symptomology: A longitudinal examination. J. Youth Adolesc. 2011, 40, 902–915. [CrossRef] 15. Bouma, E.M.; Ormel, J.; Verhulst, F.C.; Oldehinkel, A.J. Stressful life events and depressive problems in early adolescent boys and girls: The inﬂuence of parental depression, temperament and family environment. J. Aﬀect. Dis. 2008, 105, 185–193. [CrossRef] [PubMed] 16. Domènech, E.; Subirà, S.; Cuxart, F. Trastornos del ánimo en la adolescencia temprana. La labilidad afectiva [Mood disorders in early adolescence. Emotional lability]. In Psicopatología en Niños y Adolescentes. Desarrollos Actuales Child and Adolescent Psychopathology. Current Developments; Buendía, J., Ed.; Pirámide: Madrid, Spain, 1996; pp. 265–277. ISBN 9788436810264. 17. Zubeidat, I.; Fernández-Parra, A.; Ortega, J.; Vallejo, M.A.; Sierra, J.C. Características psicosociales y psicopatológicas en una muestra de adolescentes españoles a partir del Youth Self-Report/11-18 Psychosocial and psychopathological characteristics of Spanish adolescents sample through Youth Self-Report/11-18. An. Psicol. 2009, 25, 60–69. 18. Achenbach, T.M. Manual for the Child Behavior Checklist/4–18 and 1991 Proﬁle; University of Vermont: Burlington, VT, USA, 1991; ISBN 978-0938565086. 19. Dekovic, M. Risk and protective factors in the development of problem behavior during adolescence. J. Youth Adolesc. 1999, 28, 667–685. [CrossRef] 20. Kim, K.J.; Conger, R.D.; Elder, G.H.; Lorenz, F.O. Reciprocal inﬂuences between stressful life events and adolescent internalizing and externalizing problems. Child. Dev. 2003, 74, 127–143. [CrossRef] [PubMed] 21. Yakin, J.A.; McMahon, S.D. Risk and resiliency: A test of a theoretical model for urban, African-American youth. J. Prev. Interv. Commun. 2003, 26, 5–19. [CrossRef] Int. J. Environ. Res. Public Health 2019, 16, 2767 15 of 17 22. Zahn-Waxler, C.; Klimes-Dougan, B.; Slattery, M.J. Internalizing problems of childhood and adolescence: Prospects, pitfalls and progress in understanding the development of anxiety and depression. Dev. Psychopathol. 2000, 22, 443–466. [CrossRef] 23. Costello, E.J.; Keeler, G.P.; Angold, A. Poverty, race/ethnicity and psychiatric disorder: A study of rural 24. children. Am. J. Public Health 2001, 91, 1494–1498. [CrossRef] Friedman, R.J.; Chase-Lansdale, P.L. Chronic adversities. In Child and Adolescent Psychiatry; Rutter, M., Taylor, E., Eds.; Blackwell Science: Oxford, UK, 2002; pp. 261–276. ISBN 0865428808. 25. Wadsworth, M.E.; Raviv, T.; Compas, B.E.; Connor-Smith, J.K. Parent and adolescent responses to J. Child. Fam. Stud. 2005, poverty-related stress: Tests of mediated and moderated coping models. 14, 283–298. [CrossRef] 26. Williams, S.; Anderson, J.; McGee, R.; Silva, P.A. Risk factors for behavioral and emotional disorder in preadolescent children. J. Am. Acad. Child. Adolesc. Psychiatry 1990, 29, 413–419. [CrossRef] [PubMed] Sameroﬀ, A.J.; Fiese, B.H. transactional regulation: The developmental ecology of early intervention. In Handbook of Early Childhood Intervention; Shonkoﬀ, J.P., Meisels, S.J., Eds.; Cambridge University Press: New York, NY, USA, 2000; pp. 135–159. ISBN 0521585732. 27. 28. Lorence, B.; Hidalgo, M.V.; Menéndez, S.; Pérez, J. Los problemas de internalización y externalización en adolescentes de familias en situación de riesgo psicosocial: Un estudio comparativo Internalizing and externalizing problems in adolescents from at-risk families: A comparative study. In Avances en Psicología Clínica Advances in Clinical Psychology; Quevedo, R., Quevedo, V., Eds.; AEPC: Granada, Spain, 2012; pp. 503–507, ISBN 978-84-695-3599-8. 29. Lorence, B.; Jiménez, L.; Sánchez, J. Un análisis de los sucesos vitales estresantes experimentados por adolescentes que crecen en familias usuarias de los Servicios Sociales Comunitarios An analysis of stressful life events experienced by adolescents growing up in at-risk families. Portularia Revista Trabajo Social 2009, 9, 115–126. 30. Aneshensel, C.S.; Gore, S. Development, stress, and role restructuring: Social transitions of adolescence. In The Social Context of Coping; Eckenrode, J., Ed.; Plenum Press: New York, NY, USA, 1991; pp. 55–77. ISBN 030643783X. 31. Darling, N.; Steinberg, L. Parenting style as context: An integrative model. Psychol. Bull. 1993, 113, 487–496. [CrossRef] 32. O’Connor, T.G. Annotation: The ‘eﬀects’ of parenting reconsidered. Findings, challenges and applications. J. Child. Psychol. Psychiatry 2002, 43, 555–572. [CrossRef] [PubMed] 33. Henry, D.B.; Tolan, P.H.; Gorman-Smith, D. Cluster analysis in family psychology research. J. Fam. Psychol. 2005, 19, 121–132. [CrossRef] [PubMed] 34. Musitu, G.; García, F. ESPA29: Escala de Socialización Parental en la Adolescencia ESPA29: Parental Socialization 35. Scale in Adolescence; TEA: Madrid, Spain, 2001; ISBN 84-7174-783-9. Smetana, J.G. Parenting styles and conceptions of parental authority during adolescence. Child. Dev. 1995, 66, 299–316. [CrossRef] [PubMed] 36. Aunola, K.; Stattin, H.; Nurmi, J.E. Parenting styles and adolescents’ achievement strategies. J. Adolesc. 2000, 23, 205–222. [CrossRef] [PubMed] 37. Lamborn, S.D.; Mounts, N.S.; Steinberg, N.L.; Dornbusch, S.M. Pattern of competence and adjustment among adolescents from authoritative, authoritarian, indulgent and neglectful families. Child. Dev. 1991, 62, 1049–1065. [CrossRef] [PubMed] 38. Montgomery, C.; Fisk, J.E.; Craig, L. The eﬀects of perceived parenting style on the propensity for illicit drug use: The importance of parental warmth and control. Drug Alcohol Rev. 2008, 27, 640–649. [CrossRef] 39. Ruiz-Hernández, J.A.; Moral-Zafra, E.; Llor-Esteban, B.; Jiménez-Barbero, J.A. Inﬂuence of parental styles and other psychosocial variables on the development of externalizing behaviors in adolescents: A sytematic review. Eur. J. Psychol. Appl. Leg. Context 2019, 11, 9–21. [CrossRef] 40. Cohen, D.A.; Rice, J. Parenting styles, adolescent substance use and academic achievement. J. Drug Educ. 1997, 27, 199–211. [CrossRef] 41. Cenk, D.S.; Demir, A. The relationship between parenting style, gender and academic achievement with optimism among Turkish adolescents. Curr. Psychol. J. Divers. Perspect. Divers. Psychol. Issues 2016, 35, 720–728. [CrossRef] Int. J. Environ. Res. Public Health 2019, 16, 2767 16 of 17 42. Barber, B.K.; Chadwick, B.A.; Oerter, R. Parental Behaviors and Adolescent Self-concept in the United States and Germany. J. Marriage Fam. 1992, 54, 128–141. [CrossRef] 43. Marcone, R.; Aﬀuso, G.; Borrone, A. Parenting styles and children’s internalizing-externalizing behavior: The mediating role of behavioral regulation. Curr. Psychol. 2017, 1–12. [CrossRef] 44. Kritzas, N.; Grobler, A.A. The relationship between perceived parenting styles and resilience during adolescence. J. Child. Adolesc. Ment. Health 2005, 17, 1–12. [CrossRef] [PubMed] 45. Barton, A.L.; Kirtley, M.S. Gender diﬀerences in the relationships among parenting styles and college student mental health. J. Am. Coll. Health 2012, 60, 21–26. [CrossRef] [PubMed] 46. Gracia, E.; Fuentes, M.C.; Garcia, F.; Lila, M. Perceived neighborhood violence, parenting styles, J. Commun. Psychol. 2012, 40, 1004–1021. and developmental outcomes among Spanish adolescents. [CrossRef] 47. Rose, J.; Roman, N.; Mwaba, K.; Ismail, K. The relationship between parenting and internalizing behaviours of children: A systematic review. Early Child. Dev. Care 2018, 188, 1468–1486. [CrossRef] 48. Yap, M.B.; Pilkington, P.D.; Ryan, S.M.; Jorm, A.F. Parental factors associated with depression and anxiety in young people: A systematic review and meta-analysis. J. Aﬀect. Disord. 2014, 156, 8–23. [CrossRef] [PubMed] 49. Betts, J.; Gullone, E.; Allen, S. An examination of emotion regulation, temperament, and parenting style as potential predictors of adolescent depression risk status: A correlational study. Br. J. Dev. Psychol. 2009, 27, 473–485. [CrossRef] [PubMed] 50. Lorence, B.; Hidalgo, M.V.; Dekovic, M. Adolescent adjustment in at-risk families: The role of parental socialization and psychosocial stress. Salud Ment. 2013, 36, 49–57. [CrossRef] 51. Bornstein, M.H. Form and function: Implications for studies of culture and human development. Cult. Psychol. 1995, 1, 123–137. [CrossRef] 52. Chao, R. Extending research on the consequences of parenting styles for Chinese Americans and European Americans. Child. Dev. 2001, 72, 1832–1843. [CrossRef] [PubMed] 53. Deater-Deckard, K.; Lansford, J.E.; Malone, P.S.; Alampay, L.P.; Sorbring, E.; Bacchini, D.; Bombi, A.S.; Bornstein, M.H.; Chang, L.; Di Giunta, L.; et al. The association between parental warmth and control in thirteen cultural groups. J. Fam. Psychol. 2011, 25, 790–794. [CrossRef] [PubMed] 54. Dwairy, M.; Achoui, M. Adolescents-family connectedness: A ﬁrst cross-cultural research on parenting and psychological adjustment of children. J. Child. Fam. Stud. 2010, 19, 8–15. [CrossRef] 55. Lansford, J.E.; Deater-Deckard, K.; Dodge, K.A.; Bates, J.E.; Pettit, G.S. Ethnic diﬀerences in the link between discipline and later adolescent externalizing behaviors. J. Child. Psychol. Psychiatry 2004, 45, 801–812. [CrossRef] [PubMed] 56. Garaigordobil, M.; Aliri, J. Parental socialization styles, parents’ educational level and sexist attitudes in adolescence. Span. J. Psychol. 2012, 15, 592–603. [CrossRef] [PubMed] 57. DiMaggio, R.; Zappulla, C. Mothering, fathering, and Italian adolescents’ problem behaviors and life satisfaction: Dimensional and typological approach. J. Child. Fam. Stud. 2013, 23, 567–580. [CrossRef] 58. García, F.; Gracia, E. Is always authoritative the optimum parenting style? Evidence from Spanish families. Adolesc 2009, 44, 101–131. [CrossRef] 59. García, F.; Gracia, E. The indulgent parenting style and developmental outcomes in South European and Latin American Countries. In Parenting Across Cultures: Childrearing, Motherhood and Fatherhood in Non-Western Cultures; Selin, H., Ed.; Springer: Dordrecht, The Netherlands, 2014; Volume 7, pp. 419–434. ISBN 9789400775022. 60. Lorence, B.; Hidalgo, M.V.; Menéndez, S. Parenting style and adolescent adjustment in contexts at psychosocial risk: Evidence from Spanish families. In Parenting: Cultural Inﬂuences and Impact on Childhood Health and Well-Being; García, F., Ed.; NOVA publishers: New York, NY, USA, 2015; pp. 77–92. ISBN 978-1-63482-493-4. 61. Garcia, O.C.; Serra, E. Raising children with poor school performance: Parenting styles and short- and long-term consequences for adolescent and adult development. Int. J. Environ. Res. Public Health 2019, 16, 1089. [CrossRef] [PubMed] 62. Musitu, G.; García, J.F. Consequences of families socialization in the Spanish culture. Psychol. Spain 2005, 9, 34–40. 63. Aucoin, K.J.; Frick, P.J.; Bodin, S.D. Corporal punishment and child adjustment. J. Appl. Dev. Psychol. 2006, 27, 527–541. [CrossRef] Int. J. Environ. Res. Public Health 2019, 16, 2767 17 of 17 64. Nunes, C.; Bodden, D.; Lemos, I.; Lorence, B.; Jiménez, L. Parenting practices and quality of life in Dutch and Portuguese adolescents: A cross-cultural study. Rev. Psicodidácica 2014, 19, 327–346. [CrossRef] 65. Leventhal, T.; Brooks-Gunn, J. The neighborhoods they live in: The eﬀects of neighborhood residence on 66. child and adult outcomes. Psychol. Bull. 2000, 126, 309–336. [CrossRef] [PubMed] Schonberg, M.A.; Shaw, D.S. Do the pre-dictors of child conduct problems vary by high- and low-levels of socioeconomic and neighborhood risk? Clin. Child. Fam. Psychol. 2007, 10, 101–136. [CrossRef] [PubMed] 67. Gracia, E.; Fuentes, M.C.; García, F. Neighborhood risk, parental socialization styles and adolescent conduct problems. Psychosoc. Interv. 2010, 19, 265–278. [CrossRef] 68. Oliva, A.; Jiménez, J.; Parra, A.; Sánchez-Queija, I. Acontecimientos vitales estresantes, resiliencia y ajuste adolescente [Stressful life events, resilience and adolescent adjustment]. Revista de Psicopatología y Psicología Clínica 2008, 13, 53–62. [CrossRef] 69. García, F.; Musitu, G. AF5: Autoconcepto Forma 5 [AF5: Self-concept form 5, 3rd ed.; TEA: Madrid, Spain, 2009; ISBN 8415262981. IBM Corp. Released. IBM SPSS Statistics for Windows 20.0 Version; IBM Corp.: Armonk, NY, USA, 2011. 70. 71. Pérez, C. Técnicas de Análisis Multivariante de Datos. Aplicaciones con SPSS Multivariate Data Analysis Techniques. Applications with SPSS; Pearson-Prentice Hall: Madrid, Spain, 2004; ISBN 9788420541044. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_ijms241210191	Article A Comprehensive Look at the -13910 C>T LCT Gene Polymorphism as a Molecular Marker for Vitamin D and Calcium Levels in Young Adults in Central and Eastern Europe: A Preliminary Study Magdalena Kowalówka 1,* , Grzegorz Kosewski 1, Daniel Lipi ´nski 2 and Juliusz Przysławski 1 1 Department of Bromatology, Pozna ´n University of Medical Sciences, Rokietnicka 3 Street, 60-806 Pozna ´n, Poland; [email protected] (G.K.); [email protected] (J.P.) 2 Department of Biochemistry and Biotechnology, Pozna ´n University of Life Sciences, Dojazd 11 Street, 60-647 Pozna ´n, Poland; [email protected] * Correspondence: [email protected] Abstract: Intolerance to dairy products resulting from the abnormal digestion of milk sugar (lactose) is a common cause of human gastrointestinal disorders. The aim of this study was to show that the -13910 C>T LCT gene polymorphism, together with genotypes of selected VDR gene polymorphisms and diet and nutritional status parameters, can impact the prevalence of vitamin D and calcium deﬁciency in young adults. This study was conducted on a group of 63 people, which comprised 21 individuals with primary adult lactase deﬁciency, and a control group of 42 individuals with no hypolactasia. The LCT and VDR gene genotypes were assessed using PCR restriction fragment length polymorphism (PCR-RFLP) analysis. A validated HPLC method was used to determine serum concentrations of 25(OH)D2 and 25(OH)D3. Atomic absorption spectrometry was used to determine calcium levels. Their diets (self-reported 7-day estimated food record), estimated calcium intakes based on the ADOS-Ca questionnaire and basic anthropometric parameters were assessed. The CC genotype associated with hypolactasia was found in 33.3% of the subjects. The presence of the CC variant of the LCT gene polymorphism in the study group of young Polish adults was found to be associated with signiﬁcantly lower milk (134.7 ± 66.7 g/d vs. 342.5 ± 176 g/d; p = 0.012) and dairy product consumption (78.50 ± 36.2 g/d vs. 216.3 ± 102 g/d; p = 0.008) compared with lactase persistence. At the same time, people with adult-type primary intolerance were found to have statistically signiﬁcant lower serum levels of vitamin D and calcium (p < 0.05). There was a higher chance of vitamin D and calcium deﬁciency and a lower intake in the group exhibiting lactase non-persistence (OR > 1). The AA variant of the VDR gene’s BsmI polymorphism present in people with hypolactasia may further contribute to an increased risk of vitamin D deﬁciency. Exclusion of lactose from the diet, combined with impaired vitamin D metabolism, may also lead to inhibited calcium absorption by the body. Further research should be carried out on a larger group of subjects to clarify the relationship between lactase activity and vitamin D and calcium levels in young adults. Keywords: LCT polymorphism; lactase non-persistence; lactase persistence; nutrition; vitamin D; calcium; VDR polymorphism; milk drinking Citation: Kowalówka, M.; Kosewski, G.; Lipi ´nski, D.; Przysławski, J. A Comprehensive Look at the -13910 C>T LCT Gene Polymorphism as a Molecular Marker for Vitamin D and Calcium Levels in Young Adults in Central and Eastern Europe: A Preliminary Study. Int. J. Mol. Sci. 2023, 24, 10191. https://doi.org/ 10.3390/ijms241210191 Academic Editor: Alfred King-Yin Lam Received: 23 May 2023 Revised: 6 June 2023 Accepted: 10 June 2023 Published: 15 June 2023 Copyright: © 2023 by the authors. 1. Introduction Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Lactose digestion disorders, as well as milk and dairy product intolerance, are among the most common human gastrointestinal dysfunctions. Lactase non-persistence is a worldwide phenomenon, but it affects people with varying degrees of severity. According to estimates, between 31 and 37% of adults in Poland are affected [1–3]. Lactose is the main and unique carbohydrate in milk. It is also present in all milk products. Lactose may also be an additive in a variety of foods, such as pastries and Int. J. Mol. Sci. 2023, 24, 10191. https://doi.org/10.3390/ijms241210191 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10191 2 of 24 confectionery, ice creams, breads, frozen meals, cured meats, powdered soups or margarine. Some spices, sweetening agents, aromas and chewing gums contain small amounts of lactose. Furthermore, milk sugar is one of the most popular and versatile ﬁlling agents and excipients used for tableting and encapsulation within the pharmaceutical industry [4–6]. It is used as a carrier for the actual active substance in the drug manufacturing process. This milk sugar is the main source of energy for newborn mammals. It also reduces appetite. Lactose is involved in the absorption and retention of important minerals such as calcium, magnesium and zinc. It is also the only dietary source of galactose, essential for synthesizing macromolecules such as oligosaccharides, glycoproteins and glycolipids, which are used as components of nerve cell membranes [7]. Lactose exhibits probiotic properties and thus has a beneﬁcial effect on the functioning of the intestinal epithelium and the intestines themselves [8,9]. The endogenous enzyme lactase-phlorizin hydrolase known as lactase (LCT) is required to digest lactose into monosaccharides. There are three main types of lactase deﬁciency: • • Alactasia, caused by a lack of lactase in the body (rare); Secondary lactase deﬁciency, either transient or chronic, depending on the type and duration of the agent damaging the small intestinal mucosa; Primary lactase deficiency, which manifests itself in adolescence or early adulthood [10–12]. • In primary lactase1 deﬁciency, the activity of the enzyme responsible for breaking down milk sugar decreases with age. This is a natural condition called adult type hypolac- tasia (ATH) or lactase non-persistence (LPN). It is a genetically determined type of reduced lactase [13–16]. There are two distinct phenotypes: a persistently high enzymatic activity of lactase throughout a lifetime and one where its expression decreases with age due to a decrease in precursor protein synthesis in epithelial cells. Lactase deﬁciency in adults may be caused by a recessively inherited polymorphism of the LCT gene. The lactase persistence phenotype is inherited as an autosomal dominant trait [17–19]. Within the European population, the persistence or non-persistence of lactase expres- sion is primarily closely related to a single nucleotide polymorphism (SNP), LCT-13910 C>T (rs4988235), which is located in the promoter region of the gene encoding LCT. This polymorphic variant occurs as a CC, CT or TT genotype [13,20,21]. The CC genotype is a good predictor of reduced intestinal lactase expression, whereas the TT genotype is a predictor of persistently high enzymatic expression. The CT genotype is characterized by intermediate levels of lactase expression, which are usually sufﬁcient to digest lactose [22–24]. Reduced or absent lactase expression, and the consequent presence of milk sugar in the intestines, causes an increase in osmotic pressure in the gut, leading to the occurrence of diarrhea (stools exhibit a characteristic sour smell), abdominal pain and sometimes nausea and vomiting [25,26] (Figure 1). Genetic factors, as well as ethnicity, impact the prevalence of primary lactose intoler- ance. Looking at Europe, ATH affects 5% of white adults in Great Britain, 6% in Denmark, 15% in Germany, 17% in Finland and northern France and up to 40% in Mediterranean countries [27]. On the other hand, the ﬁgures for South America, Africa and Asia are much higher, with more than 50% of the population suffering from this type of condition; in some Asian countries it is even close to 100% [28,29]. Dietary history and ﬁnding an association between the diet and the dyspeptic symp- toms play an important role in diagnosing ATH. Excluding milk and its products from the diet can lead to calcium and vitamin D deﬁciency. Calcium homeostasis is affected by vitamin D status, calcitriol production and nuclear vitamin D receptor (VDR) expression. Vitamin D increases the absorption of calcium and phosphate from the gastrointestinal tract. It supports bone ossiﬁcation and mineralization and optimizes the mineral density of bones. It is essential for the proper functioning of the nervous, muscular, immune and endocrine systems [30,31]. Int. J. Mol. Sci. 2023, 24, 10191 3 of 24 Figure 1. Diagram of lactose digestion during lactase persistence (A) and lactase non-persistence (B) modiﬁed from [12]. The VDR receptor, a member of the nuclear steroid hormone receptor family, is respon- sible for the pleiotropic effects of vitamin D. The function of the VDR gene is determined by its polymorphism. Polymorphic variants in genes encoding proteins involved in vitamin D metabolism and transport can affect the levels of that vitamin in the body [32,33]. There are four main VDR gene polymorphisms. Some of its SNPs may contribute to reduced vitamin D concentrations; analysis of the serum levels of this vitamin and its metabolites may be relevant for many diseases [34]. In humans, vitamin D occurs in two main inactive forms: vitamin D2 and vitamin D3, of which vitamin D3 accounts for more than 90% of the total vitamin D level [35]. Cutaneous synthesis and diet are the primary sources of vitamin D for humans. As a result of hydroxylation in the liver, inactive vitamin D is converted to the metabolite 25(OH)D, which is a vitamin D status biomarker [36]. According to guidelines, 25(OH)D < 20 ng/mL is deﬁned as vitamin D deﬁciency and 20–30 ng/mL 25(OH)D is suboptimal, while the optimal vitamin D concentration is 25(OH)D ≥ 30 ng/mL [37,38]. This preliminary study was carried out to verify the hypothesis that the -13910 C>T LCT gene polymorphism in association with selected variants of the VDR gene, which encodes the vitamin D receptor, as well as diet and nutritional status, may inﬂuence the occurrence of vitamin D and calcium deﬁciency in the body and the pathogenesis of diseases associated with these deﬁciencies in young adults. 2. Results A ﬂow chart of this study is shown in Figure 2. Out of the seventy-one people included in the study, eight did not meet the designated criteria. All the remaining subjects (n = 63) were genotyped for the LCT polymorphism (-13910 C>T). The gene encoding lactase is located on the long arm of chromosome 2 in position 21.3 (2q21.3). The ability to produce lactase is determined by the presence of the -13910 C>T polymorphism. The variant is located in a non-coding region in the genome in intron 13 of the minichromosome maintenance type 6 gene (MCM6) [39] (Figure 3A). A change of the single nucleotide C to T in position -13910 signiﬁcantly affects LCT transcription. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 3 of 25 Figure 1. Diagram of lactose digestion during lactase persistence (A) and lactase non-persistence (B) modiﬁed from [12]. Genetic factors, as well as ethnicity, impact the prevalence of primary lactose intoler-ance. Looking at Europe, ATH aﬀects 5% of white adults in Great Britain, 6% in Denmark, 15% in Germany, 17% in Finland and northern France and up to 40% in Mediterranean countries [27]. On the other hand, the ﬁgures for South America, Africa and Asia are much higher, with more than 50% of the population suﬀering from this type of condition; in some Asian countries it is even close to 100% [28,29]. Dietary history and ﬁnding an association between the diet and the dyspeptic symp-toms play an important role in diagnosing ATH. Excluding milk and its products from the diet can lead to calcium and vitamin D deﬁciency. Calcium homeostasis is aﬀected by vitamin D status, calcitriol production and nuclear vitamin D receptor (VDR) expression. Vitamin D increases the absorption of cal-cium and phosphate from the gastrointestinal tract. It supports bone ossiﬁcation and min-eralization and optimizes the mineral density of bones. It is essential for the proper func-tioning of the nervous, muscular, immune and endocrine systems [30,31]. The VDR receptor, a member of the nuclear steroid hormone receptor family, is re-sponsible for the pleiotropic eﬀects of vitamin D. The function of the VDR gene is deter-mined by its polymorphism. Polymorphic variants in genes encoding proteins involved in vitamin D metabolism and transport can aﬀect the levels of that vitamin in the body [32,33]. There are four main VDR gene polymorphisms. Some of its SNPs may contribute to reduced vitamin D concentrations; analysis of the serum levels of this vitamin and its metabolites may be relevant for many diseases [34]. In humans, vitamin D occurs in two main inactive forms: vitamin D2 and vitamin D3, of which vitamin D3 accounts for more than 90% of the total vitamin D level [35]. Cutane-ous synthesis and diet are the primary sources of vitamin D for humans. As a result of hydroxylation in the liver, inactive vitamin D is converted to the metabolite 25(OH)D, which is a vitamin D status biomarker [36]. According to guidelines, 25(OH)D < 20 ng/mL is deﬁned as vitamin D deﬁciency and 20–30 ng/mL 25(OH)D is suboptimal, while the optimal vitamin D concentration is 25(OH)D ≥ 30 ng/mL [37,38]. This preliminary study was carried out to verify the hypothesis that the -13910 C>T LCT gene polymorphism in association with selected variants of the VDR gene, which encodes the vitamin D receptor, as well as diet and nutritional status, may inﬂuence the occurrence of vitamin D and calcium deﬁciency in the body and the pathogenesis of dis-eases associated with these deﬁciencies in young adults. Int. J. Mol. Sci. 2023, 24, 10191 4 of 24 Figure 2. Flowchart of the study. Ampliﬁcation of the promoter fragment of the LCT gene encoding lactase resulted in a 201 bp product for the polymorphism in question. The result of hydrolysis of the PCR product was revealed in an agarose gel with the HinfI restriction enzyme. Larger DNA fragments corresponded to C alleles, while smaller ones corresponded to T alleles (or, more precisely, to the larger product of hydrolysis (177 bp), as the smaller one (24 bp) was not visible in agarose gel. Genotypes were identiﬁed by the presence or absence of an appropriate restriction site. Recessive homozygote (CC) was characterized by the presence of a single strand (201 bp) (Figure 3B). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of 25 2. Results A ﬂow chart of this study is shown in Figure 2. Out of the seventy-one people in-cluded in the study, eight did not meet the designated criteria. All the remaining subjects (n = 63) were genotyped for the LCT polymorphism (-13910 C>T). Figure 2. Flowchart of the study. The gene encoding lactase is located on the long arm of chromosome 2 in position 21.3 (2q21.3). The ability to produce lactase is determined by the presence of the -13910 C>T polymorphism. The variant is located in a non-coding region in the genome in intron 13 of the minichromosome maintenance type 6 gene (MCM6) [39] (Figure 3A). A change of the single nucleotide C to T in position -13910 signiﬁcantly aﬀects LCT transcription. Int. J. Mol. Sci. 2023, 24, 10191 5 of 24 Figure 3. (A) Polymorphism in LCT gene (-13910 C>T), red of box—locus LCT gene, green—locus MCM 6 gene; (B) PCR-RFLP-based genotyping. Path M: DNA marker GPB1000 bp; path 1: negative control; paths 5,7: homozygote CC; paths 2,3,4,6,9,10,11,12,13: heterozygote CT; path 8: homozygote TT. Table 1 shows the obtained allele and genotype frequencies for the study group. Those with the CC genotype, indicating hypolactasia, accounted for 33.3% (n = 21); 50.8% had CT heterozygotes (n = 32); 15.9% (n = 10) had TT homozygotes. The CT and TT variants were associated with a lactase persistence phenotype (normolactasia). Table 1. Distribution of the LCT (-13910 C>T) genotype and allelic frequencies in the surveyed population. SNP rs4988235 Allele/Genotype Allele Frequency/ Genotypes N(%) C/T 74/52 (58.7/41.3) TT CT CC 10 (15.9) 32 (50.8) 21 (33.3) Data presented as absolute value (%), p = 0.635, χ2—0.076, df2, a chi-square test (χ2) was used to test the Hardy–Weinberg equilibrium. Values < 3.84 were consistent with the H-W equilibrium law. For the CT genotype, lactase deﬁciency is rarely identiﬁed, as this is due to compen- sation for the presence of the lactase persistence (T) allele, which has a dominant effect. Hence, it is believed that most heterozygous individuals produce sufﬁcient lactase. The C and T allele frequencies in this study group were 58.7% and 41.3%, respectively. There was no statistically signiﬁcant difference between the expected genotype and allele values and the actual values for the subjects (p > 0.05); genotype frequencies were Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 5 of 25 (A) (B) Figure 3. (A) Polymorphism in LCT gene (-13910 C>T), red of box—locus LCT gene, green—locus MCM 6 gene; (B) PCR-RFLP-based genotyping. Path M: DNA marker GPB1000 bp; path 1: negative control; paths 5,7: homozygote CC; paths 2,3,4,6,9,10,11,12,13: heterozygote CT; path 8: homozygote TT. Ampliﬁcation of the promoter fragment of the LCT gene encoding lactase resulted in a 201 bp product for the polymorphism in question. The result of hydrolysis of the PCR product was revealed in an agarose gel with the HinfI restriction enzyme. Larger DNA fragments corresponded to C alleles, while smaller ones corresponded to T alleles (or, more precisely, to the larger product of hydrolysis (177 bp), as the smaller one (24 bp) was not visible in agarose gel. Genotypes were identiﬁed by the presence or absence of an appropriate restriction site. Recessive homozygote (CC) was characterized by the presence of a single strand (201 bp) (Figure 3B). Table 1 shows the obtained allele and genotype frequencies for the study group. Those with the CC genotype, indicating hypolactasia, accounted for 33.3% (n = 21); 50.8% had CT heterozygotes (n = 32); 15.9% (n = 10) had TT homozygotes. The CT and TT vari-ants were associated with a lactase persistence phenotype (normolactasia). Table 1. Distribution of the LCT (-13910 C>T) genotype and allelic frequencies in the surveyed pop-ulation. SNP Allele/Genotype Allele Frequency/ Genotypes N(%) rs4988235 C/T 74/52 (58.7/41.3) TT 10 (15.9) Int. J. Mol. Sci. 2023, 24, 10191 6 of 24 within the Hardy–Weinberg equilibrium (χ2 < 3.46, α = 0.05). The frequencies of the investigated alleles and genotypes of the LCT gene were consistent with data published in the NCBI SNP database [40]. Those with the CC genotype were allocated to the LNP study group (arm I) and those with the CT and TT variants (n = 42) were assigned to the LP control group (arm II) (Figure 2). The mean values and medians for the selected anthropometric parameters are shown in Table 2. Women comprised 63% of the study group and men made up the remaining 37%. No relationship between genotypes and gender was found. The groups were homogeneous in terms of age. Study participants were between 21 and 31 years old and the mean median age of young adults in the LNP group was 23.0 ± 1.50 years and 24.0 ± 1.00 years for those in the control group. Body weights were between 47.4 kg and 93.3 kg. There were no signiﬁcant differences in age, height or body weight between the groups. Table 2. The level of selected anthropometric parameters according to the LCT-13910 C>T genotype (CC vs. CT + TT). Parameters Lactase Non-Persistence LNP (N = 21) Lactase Persistence LP (N = 42) p-Value Male (N %) Female (N %) # Age (years) # Height (cm) Weight (kg) BMI (kg/m2) # Body fat content (%) # Bone mass (kg) # Body water content (%) 10 (16) 11 (17) 23.0 ± 1.50 174 ± 6.50 68.5 ± 12.4 22.8 ± 2.68 21.4 ± 3.37 2.40 ± 0.25 54.5 ± 3.60 13 (21) 29 (46) 24.0 ± 1.00 172 ± 6.50 67.7 ± 15.9 22.8 ± 4.63 20.8 ± 6.70 2.60 ± 0.70 55.2 ± 5.20 - - NS NS NS NS NS NS NS Mean values ± SD, Student’s t-test, # Me ± IQR (interquartile ratio), Mann–Whitney U-test, p < 0.05, NS—no statistical differences. Indicators of nutritional status were assessed based on the performed anthropometric analysis. Mean BMI was approximately 23 kg/m2 for both study groups (BMI in the LNP group was between 17.6 kg/m2 and 28.4 kg/m2; in the LP group it was between 18.8 kg/m2 and 31.0 kg/m2). According to the norms for this index, the majority of the subjects had a normal body weight. Despite lower bone mass in those with impaired ATH, there was no statistically signiﬁcant difference compared with the control group (p > 0.05). In addition, there were no signiﬁcant differences for body fat (%) and water (%) (Table 2). Estimated total energy and protein intakes were similar: 2033 ± 542 kcal/d for lactase non-persistent subjects and 2094 ± 499 kcal/d for healthy subjects (Table 3). Daily protein intakes were also similar at 85.8 ± 27.4 g/d vs. 84.1 ± 21.9 g/d; 17% of energy was obtained from protein. A lower intake of fats was shown for the LNP group (68.8 ± 16.8 g/d) compared with the control group (80.3 ± 29.5 g/d); however, no signiﬁcant differences were observed (p > 0.05). A total of 32.7 ± 4.42 % kcal of energy was obtained from fat in the ATH group and 34.3 ± 5.32 % kcal from fat in the LP group. Carbohydrate intakes were at similar statistically non-signiﬁcant levels in both participant groups subject to analysis (253 ± 60.3 vs. 262 ± 66.8 g/d). The total vitamin D intake for those with hypolactasia was 2.32 ± 0.36 µg and was sig- niﬁcantly lower compared with those with CT and TT variants (4.57 ± 0.50 µg) (p = 0.045). Calcium intake was also found to be lower in the group with the CC genotype (785 ± 97.5 mg/d). In healthy subjects, the calcium intake was at 881 ± 128 mg/d (Table 3); this difference was statistically signiﬁcant (p = 0.048). Int. J. Mol. Sci. 2023, 24, 10191 7 of 24 Table 3. Intake levels of selected dietary components according to the LCT-13910 C>T genotype (CC vs. CT + TT). Parameters Male (N %) Female (N %) # Total energy (kcal/d) Protein (g/d) Protein (% kcal) # Fat (g/d) Fat (% kcal) Carbohydrate (g/d) Carbohydrate (% kcal) # Vitamin D intake (µg/d) # Calcium intake (mg/d) Phosphorus intake (mg/d) LNP (N = 21) 10 (16) 11 (17) 2033 ± 542 85.8 ± 27.4 17.0 ± 2.58 68.8 ± 16.8 32.7 ± 4.42 253 ± 60.3 48.9 ± 4.18 2.32 ±0.36 a 785 ± 97.5 a 1357 ± 421 LP (N = 42) 13 (21) 29 (46) 2094 ± 499 84.1 ± 21.9 16.5 ± 1.69 80.3 ± 29.5 34.3 ± 5.32 262 ± 66.8 50.4 ± 5.36 4.57 ± 0.50 b 881 ± 128 b 1566 ± 502 p-Value - - NS NS NS NS NS NS NS 0.045 0.048 NS Mean values ± SD, Student’s t-test, # Me ± IQR, Mann–Whitney U-test, a,b signiﬁcant with p < 0.05, NS—no statistical differences. Unfortunately, the intake of calcium and vitamin D for the majority of the subjects studied was below the estimated average requirement (EAR) in Poland. No statistically signiﬁcant differences were found with regard to phosphorus intake in the groups subject to analysis. An assessment of the intake of milk and dairy products, as well as calcium from dairy products, was also carried out using the ADOS-Ca questionnaire. There was a signiﬁcantly (p = 0.008) lower intake of milk (78.50 ± 36.2 g/d) and dairy products (134.7 ± 66.7 g/d, p = 0.012) for people with reduced lactase activity (LNP) compared with the control group (342.5 ± 176; 216.3 ± 102 g/d, respectively) (Table 4). Table 4. Estimated average intake of dairy products according to LCT polymorphism. Parameters LNP (N = 21) LP (N = 42) p-Value Milk (g/d) Total dairy products (g/d) 134.7 ± 66.7 a 0.012 78.50 ± 36.2 a 0.008 Me ± IQR, Mann–Whitney U-test, statistically signiﬁcant differences (p < 0.05) are marked with an a,b. 342.5 ± 176 b 216.3 ± 102 b In the study population, the prevalence of inadequate calcium intake and the as- sociated high risk of calcium deﬁciency among those with identiﬁed hypolactasia was statistically signiﬁcantly (p < 0.05) higher (91%) than for those in the LP group (62%). In addition, 31% of the control group had a low calcium intake and 7% of the subjects showed a normal intake. A total of 9% of those with the CC genotype had an average risk of calcium deﬁciency and no one in the LNP group showed a normal intake (Figure 4). Analysis of serum biochemical parameters in an unadjusted model showed signiﬁcant differences in 25(OH)D3 metabolite (17.6 ± 2.19 ng/mL, p = 0.025) and total 25(OH)D levels (18.1 ± 2.91 ng/mL, p = 0.022) in hypolactasia vs. TT variant subjects (21.5 ± 3.99 ng/mL and 22.6 ± 4.46 ng/mL, respectively). Furthermore, total calcium levels were also signiﬁcantly lower (2.27 ± 0.29 mg/dL, p = 0.049) for those with the CC genotype compared with those with the TT genotype (2.62 ± 0.26 mg/dL) (Table 5). Analysis of the mutual relations between calcium intake from diet and serum concen- trations showed a weak positive correlation in the hypolactasia group (r = 0.28) and in the control group (r = 0.27) (Figure 5A,B). A low positive correlation between vitamin D intake and serum vitamin D concentration (total 25(OH)D) was identiﬁed for those with ATH (r = 0.21) (Figure 6A). Int. J. Mol. Sci. 2023, 24, 10191 8 of 24 Figure 4. Estimated calcium intakes based on the ADOS-Ca questionnaire in the study group. Statistically signiﬁcant differences (p < 0.05) are marked with an . Table 5. Vitamin 25(OH)D2, 25(OH)D3 and 25(OH) (total), as well as calcium serum, levels in the study subjects according to the LCT-13910C>T genotype. Parameters 25(OH)D2 (ng/mL) 25(OH)D3 (ng/mL) Total (D2+D3) (ng/mL) CC (N = 21) CT (N = 32) TT (N = 10) p-Value 1.42 ± 0.61 (1.25–2.40) 17.6 ± 2.19 a (8.58–23.9) 18.1 ± 2.91 a (9.58–24.8) Median ± IQR (Min–Max) 1.99 ± 0.36 (1.20–2.82) 19.5 ± 2.79 ab (9.26–32.6) 20.8 ± 2.97 ab (9.28–33.0) 1.47 ± 0.22 (1.31–2.75) 21.5 ± 3.99 b (16.8–43.7) 22.6 ± 4.46 b (16.9–44.1) Mean values ± SD (Min–Max) NS 0.025 0.022 0.049 Calcium (mg/dL) 2.27 ± 0.29 a (1.76–2.75) 2.56 ± 0.35 ab (1.57–3.17) 2.62 ± 0.26 b (2.28–3.02) No normal distribution: Kruskal–Wallis ANOVA, * normal distribution: post hoc Tukey’s RIR test. Values marked a,b are statistically signiﬁcantly different, p < 0.05. Figure 5. Scatter plot of variables: relationship between calcium intake and its concentration in blood serum in a group of people with hypolactasia (A) and control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 8 of 25 In the study population, the prevalence of inadequate calcium intake and the associated high risk of calcium deﬁciency among those with identiﬁed hypolactasia was statistically signiﬁcantly (p < 0.05) higher (91%) than for those in the LP group (62%). In addition, 31% of the control group had a low calcium intake and 7% of the subjects showed a normal intake. A total of 9% of those with the CC genotype had an average risk of calcium deﬁciency and no one in the LNP group showed a normal intake (Figure 4). Figure 4. Estimated calcium intakes based on the ADOS-Ca questionnaire in the study group. Statistically signiﬁcant diﬀerences (p < 0.05) are marked with an . Analysis of serum biochemical parameters in an unadjusted model showed signiﬁcant diﬀerences in 25(OH)D3 metabolite (17.6 ± 2.19 ng/mL, p = 0.025) and total 25(OH)D levels (18.1 ± 2.91 ng/mL, p = 0.022) in hypolactasia vs. TT variant subjects (21.5 ± 3.99 ng/mL and 22.6 ± 4.46 ng/mL, respectively). Furthermore, total calcium levels were also signiﬁcantly lower (2.27 ± 0.29 mg/dL, p = 0.049) for those with the CC genotype compared with those with the TT genotype (2.62 ± 0.26 mg/dL) (Table 5). Table 5. Vitamin 25(OH)D2, 25(OH)D3 and 25(OH) (total), as well as calcium serum, levels in the study subjects according to the LCT-13910C>T genotype. Parameters CC (N = 21) CT (N = 32) TT (N = 10) p-Value Median ± IQR (Min–Max) 25(OH)D2 (ng/mL) 1.42 ± 0.61 1.99 ± 0.36 1.47 ± 0.22 NS (1.25–2.40) (1.20–2.82) (1.31–2.75) 25(OH)D3 (ng/mL) 17.6 ± 2.19 a 19.5 ± 2.79 ab 21.5 ± 3.99 b 0.025 (8.58–23.9) (9.26–32.6) (16.8–43.7) Total (D2+D3) (ng/mL) 18.1 ± 2.91 a 20.8 ± 2.97 ab 22.6 ± 4.46 b 0.022 (9.58–24.8) (9.28–33.0) (16.9–44.1) Mean values ± SD (Min–Max) Calcium (mg/dL) 2.27 ± 0.29 a 2.56 ± 0.35 ab 2.62 ± 0.26 b 0.049 (1.76–2.75) (1.57–3.17) (2.28–3.02) No normal distribution: Kruskal–Wallis ANOVA, * normal distribution: post hoc Tukey’s RIR test. Values marked a,b are statistically signiﬁcantly diﬀerent, p < 0.05 Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 9 of 25 Analysis of the mutual relations between calcium intake from diet and serum concentrations showed a weak positive correlation in the hypolactasia group (r = 0.28) and in the control group (r = 0.27) (Figure 5A,B). A low positive correlation between vitamin D intake and serum vitamin D concentration (total 25(OH)D) was identiﬁed for those with ATH (r = 0.21) (Figure 6A). (A) (B) Figure 5. Scaer plot of variables: relationship between calcium intake and its concentration in blood serum in a group of people with hypolactasia (A) and control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. (A) (B) Figure 6. Scaer plot of the studied variables: dietary vitamin D intake and serum vitamin D concentration (total) in subjects with hypolactasia (A) and in the control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. A multivariate logistic regression model showed a signiﬁcant relationship between impaired lactose tolerance due to the CC variant of the lactase gene polymorphism and serum levels of calcium and vitamin D (total) metabolites and their intake (Table 6). A higher chance of vitamin D and calcium deﬁciency and lower intake in the group exhibiting lactase non-persistence compared with the control group was demonstrated (OR > 1). Table 6. Relationship between lactase non-persistence (CC genotype) and serum total vitamin D and calcium concentrations and their intake levels. Variables Raw Structure Model p-Value Adjusted OR (95% Cl) Total vitamin D (D2 + D3) in serum 0.035 1.21 (0.92; 1.40) Int. J. Mol. Sci. 2023, 24, 10191 9 of 24 Figure 6. Scatter plot of the studied variables: dietary vitamin D intake and serum vitamin D concentration (total) in subjects with hypolactasia (A) and in the control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. A multivariate logistic regression model showed a signiﬁcant relationship between impaired lactose tolerance due to the CC variant of the lactase gene polymorphism and serum levels of calcium and vitamin D (total) metabolites and their intake (Table 6). A higher chance of vitamin D and calcium deﬁciency and lower intake in the group exhibiting lactase non-persistence compared with the control group was demonstrated (OR > 1). Table 6. Relationship between lactase non-persistence (CC genotype) and serum total vitamin D and calcium concentrations and their intake levels. Variables Total vitamin D (D2 + D3) in serum Calcium in serum Vitamin D intake Calcium intake OR—odds ratio; Cl—conﬁdence interval. Raw Structure Model p-Value Adjusted OR (95% Cl) 0.035 0.022 0.040 0.032 1.21 (0.92; 1.40) 1.15 (0.84; 1.23) 1.11 (1.05; 1.37) 1.20 (0.98; 1.28) Table 7 shows the allele and genotype frequencies of the VDR gene polymorphism SNPs, BsmI (1024 + 283 G > A, rs1544410) and VDR FokI (c.2T > C, rs2228570), in the study group. The BsmI polymorphism was shown to have a G allele frequency of 50.8% and an A allele frequency of 49.2%. In contrast, for the FokI polymorphism, the frequencies of T and C alleles were 38.9.2% and 61.1%, respectively. The frequencies of the investigated alleles and genotypes of the VDR gene were consistent with data published in the NCBI SNP database [40]. Table 7. Allele and genotype frequencies of VDR BsmI (1024 + 283 G > A) and VDR FokI (c.2T > C) in the study group. SNP Allele/Genotype Allele Frequency/ Genotypes N (%) BsmI (rs1544410) G/A 64/62 (50.8/49.2) GG GA AA 15 (23.8) 34 (54.0) 14 (22.2) Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 9 of 25 Analysis of the mutual relations between calcium intake from diet and serum concentrations showed a weak positive correlation in the hypolactasia group (r = 0.28) and in the control group (r = 0.27) (Figure 5A,B). A low positive correlation between vitamin D intake and serum vitamin D concentration (total 25(OH)D) was identiﬁed for those with ATH (r = 0.21) (Figure 6A). (A) (B) Figure 5. Scaer plot of variables: relationship between calcium intake and its concentration in blood serum in a group of people with hypolactasia (A) and control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. (A) (B) Figure 6. Scaer plot of the studied variables: dietary vitamin D intake and serum vitamin D concentration (total) in subjects with hypolactasia (A) and in the control group (B). The red line is a regression line with a 95% conﬁdence interval, bounded by dashed lines. The blue circles are the values of the analyzed variables. A multivariate logistic regression model showed a signiﬁcant relationship between impaired lactose tolerance due to the CC variant of the lactase gene polymorphism and serum levels of calcium and vitamin D (total) metabolites and their intake (Table 6). A higher chance of vitamin D and calcium deﬁciency and lower intake in the group exhibiting lactase non-persistence compared with the control group was demonstrated (OR > 1). Table 6. Relationship between lactase non-persistence (CC genotype) and serum total vitamin D and calcium concentrations and their intake levels. Variables Raw Structure Model p-Value Adjusted OR (95% Cl) Total vitamin D (D2 + D3) in serum 0.035 1.21 (0.92; 1.40) Int. J. Mol. Sci. 2023, 24, 10191 10 of 24 Table 7. Cont. SNP Allele/Genotype Allele Frequency/ Genotypes N (%) FokI (rs2228570) T/C 49/77 (38.9/61.1) TT TC CC 12 (19.0) 25 (39.7) 26 (41.3) Data presented as absolute value (%), a chi-square test (χ2) was used to test the Hardy–Weinberg equilibrium. Values < 3.84 were consistent with the H-W equilibrium law. Figures 7 and 8 show the obtained genotyping products after restriction enzyme digestion (RFLP) and agarose gel electrophoresis. Figure 7. PCR-RFLP-based genotyping in VDR genes (BsmI). Path M: DNA marker GPB1000 bp; path 3: homozygote GG; paths 2,7,8,9: heterozygote GA; paths 4,5,6,10: homozygote AA; path 1: negative control (-), where with proper cleanliness work, no product will be created, (most often without matrix). Figure 8. PCR-RFLP-based genotyping in VDR genes (FokI). Path M: DNA marker GPB1000 bp; paths 2–6,9,10: homozygote TT; paths 8,11,13,14: heterozygote TC; paths 7,12: homozygote CC; path 1: negative control (-). Rs1544410 (BsmI) genotyping showed no statistically signiﬁcant differences in the distribution of VDR gene polymorphism genotypes between the LNP group and the healthy subjects (Table 8). In the study population, the GA genotype of the BsmI polymorphism was dominant in both the LNP and LP groups (47.6% vs. 57.2%). Table 8. Distribution of VDR genotypes and allelic frequencies in dependence on lactase persistence. VDR Genotypes BsmI GG GA AA LNP (N = 21) LP (N = 42) N (%) 6 (28.6) 10 (47.6) 5 (23.8) 9 (21.4) 24 (57.2) 9 (21.4) FokI TT TC CC 8 (38.1) 9 (42.9) 4 (19.0) The signiﬁcance of difference was analyzed at p < 0.05; NS—no statistical differences. 8 (19.0) 16 (38.1) 18 (42.9) p-Value NS NS NS NS NS 0.042 Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 10 of 25 Calcium in serum 0.022 1.15 (0.84; 1.23) Vitamin D intake 0.040 1.11 (1.05; 1.37) Calcium intake 0.032 1.20 (0.98; 1.28) OR—odds ratio; Cl—conﬁdence interval. Table 7 shows the allele and genotype frequencies of the VDR gene polymorphism SNPs, BsmI (1024 + 283 G > A, rs1544410) and VDR FokI (c.2T > C, rs2228570), in the study group. The BsmI polymorphism was shown to have a G allele frequency of 50.8% and an A allele frequency of 49.2%. In contrast, for the FokI polymorphism, the frequencies of T and C alleles were 38.9.2% and 61.1%, respectively. The frequencies of the investigated alleles and genotypes of the VDR gene were consistent with data published in the NCBI SNP database [40]. Table 7. Allele and genotype frequencies of VDR BsmI (1024 + 283 G > A) and VDR FokI (c.2T > C) in the study group. SNP Allele/Genotype Allele Frequency/ Genotypes N (%) BsmI (rs1544410) G/A 64/62 (50.8/49.2) GG 15 (23.8) GA 34 (54.0) AA 14 (22.2) FokI (rs2228570) T/C 49/77 (38.9/61.1) TT 12 (19.0) TC 25 (39.7) CC 26 (41.3) Data presented as absolute value (%), a chi-square test (χ2) was used to test the Hardy–Weinberg equilibrium. Values < 3.84 were consistent with the H-W equilibrium law. Figures 7 and 8 show the obtained genotyping products after restriction enzyme digestion (RFLP) and agarose gel electrophoresis. Figure 7. PCR-RFLP-based genotyping in VDR genes (BsmI). Path M: DNA marker GPB1000 bp; path 3: homozygote GG; paths 2,7,8,9: heterozygote GA; paths 4,5,6,10: homozygote AA; path 1: negative control (-), where with proper cleanliness work, no product will be created, (most often without matrix). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 10 of 25 Calcium in serum 0.022 1.15 (0.84; 1.23) Vitamin D intake 0.040 1.11 (1.05; 1.37) Calcium intake 0.032 1.20 (0.98; 1.28) OR—odds ratio; Cl—conﬁdence interval. Table 7 shows the allele and genotype frequencies of the VDR gene polymorphism SNPs, BsmI (1024 + 283 G > A, rs1544410) and VDR FokI (c.2T > C, rs2228570), in the study group. The BsmI polymorphism was shown to have a G allele frequency of 50.8% and an A allele frequency of 49.2%. In contrast, for the FokI polymorphism, the frequencies of T and C alleles were 38.9.2% and 61.1%, respectively. The frequencies of the investigated alleles and genotypes of the VDR gene were consistent with data published in the NCBI SNP database [40]. Table 7. Allele and genotype frequencies of VDR BsmI (1024 + 283 G > A) and VDR FokI (c.2T > C) in the study group. SNP Allele/Genotype Allele Frequency/ Genotypes N (%) BsmI (rs1544410) G/A 64/62 (50.8/49.2) GG 15 (23.8) GA 34 (54.0) AA 14 (22.2) FokI (rs2228570) T/C 49/77 (38.9/61.1) TT 12 (19.0) TC 25 (39.7) CC 26 (41.3) Data presented as absolute value (%), a chi-square test (χ2) was used to test the Hardy–Weinberg equilibrium. Values < 3.84 were consistent with the H-W equilibrium law. Figures 7 and 8 show the obtained genotyping products after restriction enzyme digestion (RFLP) and agarose gel electrophoresis. Figure 7. PCR-RFLP-based genotyping in VDR genes (BsmI). Path M: DNA marker GPB1000 bp; path 3: homozygote GG; paths 2,7,8,9: heterozygote GA; paths 4,5,6,10: homozygote AA; path 1: negative control (-), where with proper cleanliness work, no product will be created, (most often without matrix). Int. J. Mol. Sci. 2023, 24, 10191 11 of 24 For the FokI polymorphism (rs2228570), it was shown that individuals with hypolacta- sia were mainly characterized by the TC (42.9%) and TT (38.1%) genotypes. The CC variant was dominant in LP subjects (42.9%), while the impaired lactose tolerance genotype was least common (19.0%). This difference was statistically signiﬁcant (p = 0.042) (Table 8). In the study groups, the frequencies of BsmI and FokI polymorphism genotypes were within the Hardy–Weinberg equilibrium (χ2 < 3.46, α = 0.05) (Table 9). Table 9. χ2 value of the tested VDR gene polymorphisms in the LNP group and the LP control group. Polymorphism LNP LP BsmI FokI 0.05 0.23 A chi-square test (χ2) was used to test the Hardy–Weinberg equilibrium. Values < 3.84 were consistent with the H-W equilibrium law. 0.85 1.41 Analysis of vitamin D (total 25(OH)D) levels and serum calcium levels in subjects with hypolactasia showed that carriers of the AA genotype of the VDR gene BsmI poly- morphism had statistically signiﬁcantly lower levels (10.4 ± 1.25 ng/mL, p = 0.017 and 2.28 ± 0.12 mg/dL, p = 0.049, respectively) compared with those with the GG genotype. A similar relation was found for lactose-tolerant individuals. Individuals with the GG genotype had signiﬁcantly higher (p = 0.048) serum vitamin D levels (22.6 ± 3.53 ng/mL) than those with the AA variant (17.0 ± 2.79 ng/mL) (Table 10). Table 10. Vitamin D (total) and serum calcium levels according to VDR genotype (BsmI and FokI) and the LCT-13910 C>T genotype. SNP BsmI GG GA AA p-value LNP LP Total Vitamin D (ng/mL) (Min–Max) Calcium (mg/dL) (Min–Max) Total Vitamin D (ng/mL) (Min–Max) Calcium (mg/dL) (Min–Max) 21.6 ± 0.67 a (19.3–24.9) 18.1 ± 1.99 ab (10.6–23.4) 10.4 ± 1.25 b (9.58–16.8) 0.017 2.46 ± 0.26 a (2.05–2.75) 2.38 ± 0.37 ab (1.87–2.75) 2.28 ± 0.12 b (2.11–2.43) 0.049 22.6 ± 3.53 a (9.56–43.7) 19.3 ± 2.76 ab (9.26–32.6) 17.0 ± 2.79 b (12.2–23.7) 0.048 2.50 ± 0.26 (2.21–3.07) 2.71 ± 0.28 (2.0–3.17) 2.33 ± 0.32 (1.60–2.70) NS FokI CC TC 2.04 ± 0.06 a (1.96–2.11) 2.52 ± 0.21 b (2.13–2.75) 2.37 ± 0.31 b (1.87–2.75) 0.05 Me ± IQR and mean values ± SD. No normal distribution: Kruskal–Wallis ANOVA; normal distribution: post hoc Tukey’s RIR test. a,b,c signiﬁcant with p < 0.05, NS—no statistical differences. 13.2 ± 2.16 a (10.4–22.4) 19.7 ± 1.30 b (12.3–24.9) 17.2 ± 2.29 b (9.58–20.8) 0.05 2.17 ± 0.26 a (1.57–2.38) 2.55 ± 0.24 b (2.21–3.17) 2.70 ± 0.21 c (2.37–3.07) 0.036 20.5 ± 5.19 (9.56–28.7) 21.8 ± 2.36 (10.0–32.6) 20.0 ± 2.43 (9.26–43.7) NS p-value TT Analysis of the FokI polymorphism among individuals with LNP showed the lowest serum vitamin D (13.2 ± 2.16 ng/mL) and calcium (2.04 ± 0.06 mg/dL) concentrations among those with the TT genotype compared with the TC and CC variants. The obtained differences were at the threshold of statistical signiﬁcance (p = 0.05). For lactose-tolerant individuals, there were no signiﬁcant differences between vitamin D and calcium levels for different FokI genotypes (Table 10). Lactase-deﬁcient participants had signiﬁcantly lower (p < 0.05) total 25(OH)D concen- trations compared with the milk sugar-tolerant group. A total of 71% of subjects had severe Int. J. Mol. Sci. 2023, 24, 10191 12 of 24 vitamin D deﬁciency and 29% had suboptimal concentrations 20 < (25(OH)D < 30 ng/mL. For the control group, 40% of the subjects had a vitamin D deﬁcit and 10% had optimal concentrations (Figure 9). Figure 9. The percentage of people with vitamin D2 + D3 deﬁciency in patients depending on lactase persistence and non-persistence. There were no signiﬁcant differences within the scope of body weight, BMI and bone mass and basal dietary components between the study groups (LNP vs. LP) in relation to vitamin D status. Serum vitamin D levels appeared to be partially dependent on vitamin D intake. The group with hypolactasia and severe serum vitamin D deﬁciency (<20 ng/mL) had the lowest intake of this vitamin (2.23 ± 0.72 µg/d). The obtained differences were at the threshold of statistical signiﬁcance (p = 0.05). Participants with severe vitamin D deﬁciency were also observed in the lactose-tolerant group; however, their vitamin D intake was higher (3.92 ± 1.22 µg/d). In addition, individuals with optimal serum vitamin D levels and the highest intake (>30 ng/mL, 5.13 ± 0.89 µg/d) were identiﬁed. The resulting differences in vitamin D intake according to vitamin D supply status in the LP group were statistically signiﬁcant (p = 0.048) (Table 11). Table 11. Anthropometric parameters, intake levels of selected nutrients and serum calcium levels and VDR polymorphisms (BsmI and FokI) according to LCT-13910 C>T genotype and serum vitamin D levels (deﬁciency, subclinical and optimal levels). Parameters Weight (kg) BMI (kg/m2) # Bone mass (kg) # Total energy (kcal/d) Protein (g/d) Fat (g/d) Carbohydrate ((g/d) # Vitamin D intake (µg/d) # Calcium intake (mg/d) LNP LP Total Vitamin D in Serum <20 ng/mL >20 <30 ng/mL p-Value <20 ng/mL >20 <30 ng/mL >30 ng/mL p-Value * 68.4 ± 10.8 23.0 ± 2.46 2.30 ± 0.40 2022 ± 277 85.4 ± 15.3 70.0 ± 17.3 258 ± 48.4 68.9 ± 16.9 22.4 ± 3.38 2.50 ± 0.25 1638 ± 380 64.6 ± 16.1 64.1 ± 18.0 241 ± 58.0 2.23 ± 0.72 a 3.07 ± 0.51 b 812 ± 133 830 ± 65.9 NS NS NS NS NS NS NS 0.050 NS 63.5 ± 13.7 21.5 ± 3.27 2.35 ± 0.15 1884 ± 453 85.1 ± 22.9 81.7 ± 14.9 260 ± 67.0 71.2 ± 17.7 23.8 ± 5.61 2.50 ± 0.45 2239 ± 466 84.9 ± 22.3 79.2 ± 14.1 266 ± 70.3 69.0 ± 12.6 22.9 ± 2.53 2.80 ± 0.55 1933 ± 380 72.4 ± 13.5 75.9 ± 14.1 245 ± 57.1 3.92 ± 1.22 a 4.37 ± 0.84 ab 5.13 ± 0.89 b 899 ± 144 915 ± 175 1018 ± 173 NS NS NS NS NS NS NS 0.048 NS Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 12 of 25 Table 10. Vitamin D (total) and serum calcium levels according to VDR genotype (BsmI and FokI) and the LCT-13910 C>T genotype. SNP LNP LP Total Vitamin D (ng/mL) (Min–Max) Calcium (mg/dL) (Min–Max) Total Vitamin D (ng/mL) (Min–Max) Calcium (mg/dL) (Min–Max) BsmI GG 21.6 ± 0.67 a 2.46 ± 0.26 a 22.6 ± 3.53 a 2.50 ± 0.26 (19.3–24.9) (2.05–2.75) (9.56–43.7) (2.21–3.07) GA 18.1 ± 1.99 ab 2.38 ± 0.37 ab 19.3 ± 2.76 ab 2.71 ± 0.28 (10.6–23.4) (1.87–2.75) (9.26–32.6) (2.0–3.17) AA 10.4 ± 1.25 b 2.28 ± 0.12 b 17.0 ± 2.79 b 2.33 ± 0.32 (9.58–16.8) (2.11–2.43) (12.2–23.7) (1.60–2.70) p-value 0.017 0.049 0.048 NS FokI CC 13.2 ± 2.16 a 2.04 ± 0.06 a 20.5 ± 5.19 2.17 ± 0.26 a (10.4–22.4) (1.96–2.11) (9.56–28.7) (1.57–2.38) TC 19.7 ± 1.30 b 2.52 ± 0.21 b 21.8 ± 2.36 2.55 ± 0.24 b (12.3–24.9) (2.13–2.75) (10.0–32.6) (2.21–3.17) TT 17.2 ± 2.29 b 2.37 ± 0.31 b 20.0 ± 2.43 2.70 ± 0.21 c (9.58–20.8) (1.87–2.75) (9.26–43.7) (2.37–3.07) p-value 0.05 0.05 NS 0.036 Me ± IQR and mean values ± SD. No normal distribution: Kruskal–Wallis ANOVA; normal distribution: post hoc Tukey’s RIR test. a,b,c signiﬁcant with p < 0.05, NS—no statistical diﬀerences. Lactase-deﬁcient participants had signiﬁcantly lower (p < 0.05) total 25(OH)D concentrations compared with the milk sugar-tolerant group. A total of 71% of subjects had severe vitamin D deﬁciency and 29% had suboptimal concentrations 20 < (25(OH)D < 30 ng/mL. For the control group, 40% of the subjects had a vitamin D deﬁcit and 10% had optimal concentrations (Figure 9). Figure 9. The percentage of people with vitamin D2 + D3 deﬁciency in patients depending on lactase persistence and non-persistence. There were no signiﬁcant diﬀerences within the scope of body weight, BMI and bone mass and basal dietary components between the study groups (LNP vs. LP) in relation to vitamin D status. Serum vitamin D levels appeared to be partially dependent on vitamin D intake. The group with hypolactasia and severe serum vitamin D deﬁciency (<20 ng/mL) had the lowest intake of this vitamin (2.23 ± 0.72 µg/d). The obtained diﬀerences were at Int. J. Mol. Sci. 2023, 24, 10191 13 of 24 Parameters Phosphorus intake (mg/d) Calcium (mg/dL) BsmI (N%) GG GA AA FokI (N%) CC TC TT Table 11. Cont. LNP Total Vitamin D in Serum LP <20 ng/mL >20 <30 ng/mL p-Value <20 ng/mL >20 <30 ng/mL >30 ng/mL p-Value * 1398 ± 369 2.28 ± 0.29 1256 ± 457 2.32 ± 0.14 1 (4.76) 9 (42.9) 5 (23.8) 3 (14.3) 5 (23.8) 7 (33.3) 5 (23.8) 1 (4.76) 0 1 (4.76) 4 (19.0) 1 (4.76) NS NS NS NS - NS NS NS 1528 ± 449 2.39 ± 0.38 1619 ± 570 2.57 ± 0.30 1419 ± 361 2.58 ± 0.17 2 (4.76) 12 (28.6) 4 (9.52) 4 (9.52) 5 (11.9) 1 (2.38) 6 (14.3) 11 (26.2) 2 (4.76) 2 (4.76) 6 (14.3) 15 (35.7) 2 (4.76) 1 (2.38) 0 2 (4.76) 5 (11.9) 2 (4.76) NS NS NS NS NS NS NS NS # Me ± IQR and mean values ± SD. No normal distribution: Mann–Whitney U test; normal distribution: Student’s t-test. * No normal distribution: Kruskal–Wallis ANOVA; normal distribution: post hoc Tukey’s RIR test. a,b signiﬁcant with p < 0.05. NS—no statistical differences. Individuals with hypolactasia and severe serum vitamin D deﬁciency predominantly had the GA (42.9%) and the AA genotypes (23.8%) of the VDR gene BsmI polymorphism. For the FokI polymorphism, it was shown that more people with ATH and deﬁcient vitamin D levels had the TT genotype (33.3%). LP individuals and those with higher serum vitamin D concentrations were mainly of the CC and TC genotype (Table 11). 3. Discussion Research on lactase phenotypes has been ongoing for many years. Despite many studies, the molecular mechanisms that determine the development of lactase deﬁciency with age are still not understood completely. Today, modern molecular biology diagnostic methods make it possible to identify the genetic basis of this condition. This study analyzed polymorphic variants of the LCT gene associated with different lactase activity and their possible relationship with serum vitamin D and calcium levels, diet and nutritional status, as well as the vitamin D receptor promoter genotype (BsmI and FokI), in a group of young Polish adults. A total of 33% of the study population had the CC genotype and reported gastroin- testinal symptoms of varying severity after consuming dairy products. The frequencies of genotypes obtained in the study were within the Hardy–Weinberg equilibrium and were consistent with previous studies performed by other facilities [1]. Overall, it is estimated that around two-thirds of people worldwide have a problem digesting lactose in adulthood [18]. Suppressive transcriptional epigenetic changes (e.g., DNA methylation) that accu- mulate over time and lead to decreased lactase levels in adulthood impact the severity of the symptoms associated with hypolactasia [41]. The presence of the T allele means that age-dependent DNA methylation is not signiﬁcant and does not reduce lactase levels in adulthood [42,43]. Hypolactasia depends not only on lactase expression but also on the amount of lactose in the diet, intestinal ﬂora, gastrointestinal motility, small intestinal bacterial overgrowth and the sensitivity of the gastrointestinal tract to the production of gas and other lactose fermentation products [44]. Individuals with adult-type hypolactasia have a signiﬁcantly lower clearance of short-chain fatty acids, i.e., propionate, acetate and butyrate (SCFA). In patients with ATH, lactose intake of more than 12 g is estimated to contribute to adverse symptoms due to milk sugar consumption, whereas low lactose intake is most often not a problem [45–48]. Int. J. Mol. Sci. 2023, 24, 10191 14 of 24 The ability to ferment undigested lactose depends mainly on the composition of the colonic and intestinal microbiota. A positive correlation between the occurrence of hypo- lactasia and probiotics of a speciﬁc strain and concentration has been demonstrated [49]. Alleviation or exacerbation of lactase deﬁciency symptoms has been observed depending on the abundance of certain bacterial strains in the colon. Colonic bacteria effective in lactose fermentation help to reduce osmotic shock which causes diarrhea but can result in an increased production of gasses. Interestingly, heterozygous carriers of LCT-13910 C>T and LCT-22018 G>A showed intermediate enzymatic activity causing symptoms of lactose intolerance in stressful situations or during intestinal infections [12]. The research ﬁndings we presented applied to: 3.1. Potential Differences in Anthropometric Characteristics Nutritional status parameters for individuals with lactase non-persistence were not signiﬁcantly different from the control group. Bone mass was lower in those with the CC genotype, but the difference was not signiﬁcant. In contrast, Mnich et al. [50], in their study, demonstrated a signiﬁcant correlation between lower bone density and the CC genotype. It was also noted that individuals with lactase deﬁciency were more likely to suffer from osteoporosis, however these values were not statistically signiﬁcant compared with individuals with CT and TT genotypes [50]. Popadowska et al. [51] did not show an association between the -13910 C>T LCT gene polymorphism and obesity. However, a subsequent study by Popadowska and Kempinska-Podhorodecka [52] revealed that the CC genotype was linked with a reduced intake of milk and dairy products, as well as higher lean mass and larger forearm circumference, which may have implications for dietary management of ATH. According to Alharbi and El-Sohema [43], individuals with the CC genotype were shorter than those with the TT genotype; presumably, as conﬁrmed by numerous studies, this determined lower milk consumption [53–56]. 3.2. Vitamin D and Calcium Intake vs. Their Serum Concentrations Multivariate analysis showed that adult-type hypolactasia was associated with a higher risk of vitamin D and calcium deﬁciency, as well as lower calcium intake, compared with lactase-persistent (LP) individuals. Vitamin D intake in the Polish population has been very low for years and is far from meeting dietary recommendations, which have been set at an adequate intake (AI) level of 15 µg/day [57]. For adults, the intake is within the range of 1.4–5.1 µg per day. Hence, according to recent recommendations, depending on body weight and dietary vitamin D supply, year-round vitamin D supplementation may be recommended in Poland [58]. The mean intake of vitamin D in other European countries is also low (less than 5 µg/day (200 IU/day) in most countries). It is highest in Scandinavian countries and also for the Innuit population due to the consumption of oily ﬁsh and cod liver oil and fortiﬁed dairy products [59]. The relationship between the LCT-13910 C>T polymorphism genotypes and the con- sumption of milk and dairy products reﬂected with high probability the vitamin D and calcium serum levels in the subjects. The mean concentrations of the vitamin in question in the LNP and LP groups were found to be below the recommended ﬁgures, but those in the LNP group had statistically signiﬁcantly lower hypovitaminosis D (p < 0.05) compared with those with the TT variant. A similarly signiﬁcant relationship was shown for calcium concentrations. The mean calcium level for individuals with diagnosed hypolactasia was at the lower limit of the recommended range, as opposed to those with the TT genotype. Vitamin D plays an essential role in maintaining phosphorus and calcium homeostasis and in stimulating bone mineralization. Calcium deﬁciency in lactose-intolerant individuals is often associated with lower bone mineral density [60,61]. In a Canadian population study, Alharbi and El-Sohemy [43] also demonstrated lower serum vitamin D levels due to reduced dairy product intake. Int. J. Mol. Sci. 2023, 24, 10191 15 of 24 Milk and dairy products are the richest source of easily assimilated calcium in the diet, mainly due to the presence of lactose, milk phosphopeptides and a favorable 1.4:1 calcium: phosphorus ratio. They are the main source of calcium in the diets of Poles and other Europeans, providing between 60% and 80% of the total amount [57]. Calcium homeostasis in the human body depends on vitamin D status, the production of 1,25(OH)2D3 and the expression of the nuclear receptor VDR, which is known to regulate the expression of genes important for calcium balance and bone metabolism, as well as through the presence of other absorption promoters. Calcium in dairy products reduces fat absorption and may therefore prevent the development of cardiovascular diseases [62]. At the same time, it has been suggested that calcium deﬁciency may be a factor which increases the risk of obesity through excessive fat accumulation and contributes to the development of type 2 diabetes [62]. The beneﬁts associated with dairy product consumption also include improved hypertension control, weight gain and reduced risk of colorectal cancer. The research carried out within the scope of this study showed a statistically signiﬁ- cantly lower calcium intake in the LNP group subjects (p < 0.05). The demonstrated lower total intake of milk and of dairy products for people with reduced lactase activity compared with the control group may be the cause of this unfavorable phenomenon. The application of the ADOS-Ca questionnaire enabled a more accurate characteriza- tion of this intake in the study population by distinguishing those with inadequate, low and correct recommended intake levels of this mineral. Through this, it was found that study participants who avoided eating milk and other dairy products in particular due to symptoms associated with the presence of hypolactasia were at a greater risk of calcium deﬁciency. This was in line with a number of other studies. Di Stefano et al. demonstrated that lactose-intolerant subjects consumed statistically signiﬁcantly less calcium in their diet compared with lactose-tolerant subjects [63]. Furthermore, young adults with lactase deﬁciency exhibited elevated parathormone (PHT) levels [64]. Koek et al. [65] demonstrated a clear relationship between dietary calcium intake and serum ionized calcium levels. However, Yahya et al. [66] showed that, although young Malaysian adults had a high prevalence of ATH, there was no direct effect on bone health, unlike calcium intake, which was low. Enattah et al. [24] showed that hypolactasia and abnormal lactose digestion did not alter calcium absorption and bone turnover rates nor did it interfere with reaching peak bone mass. 3.3. Assessment of LCT and VDR Gene Polymorphisms Interactions Our results allowed us to assess gene interactions between lactase gene polymor- phisms and genetic VDR gene changes, as well as the effect of polymorphism on vitamin D status. The 1.25(OH)2D3 encoding VDR gene activates a rapid receptor binding to regula- tory regions of target genes and causes changes in transcription [67]. Expression of vitamin D receptors is associated with the occurrence of polymorphisms in the VDR encoding gene. The most common polymorphisms include the morphs referred to as FokI, BsmI, TaqI and ApaI. These polymorphisms may affect vitamin 25(OH)D serum levels [68]. The vitamin D receptor gene mediates the action of the hormone system in calcium homeostasis and the VDR genotype also impacts the gut’s ability to absorb calcium. In relation to the pleiotropic effects of vitamin D, correlations have been demonstrated to exist between VDR polymorphisms and various diseases such as insulin resistance, type 2 diabetes, abdominal obesity and responses to calcium and vitamin D supplementation [69]. Vitamin D receptor polymorphisms may be the reason for why not all individuals beneﬁt from vitamin D supplementation [70]. Of the many polymorphisms of the VDR gene described, two affect VDR molecular signaling. These are BsmI, located in the gene intron sequence and FokI in the encoding section. Changes in VDR expression associated with polymorphisms impact vitamin D- dependent functions. Individuals with the GG genotype (BsmI) exhibit increased plasma concentrations of 1α,25-dihydroxyvitamin D3 [71]. Int. J. Mol. Sci. 2023, 24, 10191 16 of 24 Analysis of serum vitamin D and calcium levels according to the study group (LNP and LP) and VDR genotype showed that subjects with adult type hypolactasia had signiﬁcantly lower levels when it came to carriers of the AA BsmI genotype compared with carriers of the GG variant in both study groups, who exhibited the highest 25(OH)D and calcium levels. A similar relation was shown by Abouzid et al. [34], where levels of 25(OH)D3 and 3-epi-25(OH)D3 were signiﬁcantly higher for the GG genotype as compared to the GA variant. For individuals with lactase deﬁciency (LNP) and the TT variant of the FokI polymor- phism, we demonstrated the lowest statistically signiﬁcant vitamin D and calcium levels. A similar relation was not found for lactose tolerant individuals. Divanoglou [72] suggested that epigenetic VDR modiﬁcations regulate the conversion of vitamin D to its metabolites via CYP450, thereby affecting vitamin D concentrations. San- tos et al. [73] demonstrated the association of wild-type BsmI, ApaI and TaqI variants of the VDR gene with low 25(OH)D levels. However, these results were not consistent. In contrast, Cobayashi et al. [74] observed that it was the presence of the mutant A allele of the BsmI polymorphism that constituted an increased risk of vitamin D deﬁciency. Valtuena et al. [75] showed no association between the BsmI polymorphism and vitamin D concentrations in adolescents under 18 years of age. Furthermore, according to Jakubowska-Pietkiewicz et al. [76], the BsmI and FokI polymorphisms of the VDR gene did not directly impact the calcium–phosphate metabolism in young individuals. Discrepancies between different study populations can be partly explained by ethnic, geographical and genetic differences. In order to alleviate the symptoms resulting from impaired lactase activity, sufferers often choose to eliminate milk and dairy products from their diets. This action exposes them to a risk of both vitamin D and calcium deﬁciencies; however, only congenital lactase deﬁciency requires complete elimination of milk sugar from the diet. Hypolactasia does not require complete elimination of lactose from the diet, only a reduction of its intake to approximately 10–12 g/day [12,77]. Some probiotics can improve the digestion of lactose and thus alleviate dyspepsia symptoms. Therefore, fermented dairy products can still be consumed by people with ATH as they exhibit high β-galactosidase activity, partial lactose hydrolysis and slower intestinal transit times. This is also true for maturing hard cheeses and brie-type cheeses. It is only recommended to limit the consumption of milk in its pure form to 50–100 mL, depending on individual sensitivity to milk sugar [6]. Lactose tolerance may be increased by taking probiotics, which alter the colonic microﬂora. Furthermore, in people with LNP, a continuous intake of small amounts of lactose leads to the microbiome adapting, resulting in altered metabolomes [78,79]. Today, more and more reduced-lactose products are becoming available. An enzymatic lactose hydrolysis process is usually applied to such products. The market for this type of dairy product is the fastest growing dairy industry sector. However, the lactose hydrolysis process increases the cost of producing milk and dairy products, which translates into a higher price on the shelves. Limitations The main limitation of the present study was the relatively small sample size, which probably resulted in other relationships between genotyping and environmental factors not being revealed. However, this was a preliminary study. Financial issues and the limited time available to conduct the survey also made it difﬁcult to put together a large study group. Furthermore, the genotyping results obtained did not exclude the possibility that the risk of vitamin D and calcium deﬁciency could also be affected by other SNPs in the VDR gene. Therefore, further assessments of vitamin D status and calcium levels are required. This was also true for genotyping other major VDR polymorphisms, such as TaqI, ApaI and Cdx-2. The study population was not divided into male and female groups, since genotyping was not shown to be related to gender and the sample size was small (although the calculated minimum sample size was reached). This perhaps meant that signiﬁcant Int. J. Mol. Sci. 2023, 24, 10191 17 of 24 differences in levels and intakes of vitamin D and calcium, as well as the other parameters studied, were not demonstrated. In addition, the samples for the present study were taken in autumn and winter, so the seasonal variation in vitamin D levels in the body, associated with cutaneous synthesis, was not taken into account. The strengths of the study included the fairly high homogeneity of baseline results between the analyzed groups of people with lactase non-persistence and lactase persistence, as well as the use of an ethnically homogeneous population and the use of multiple predictors to determine the intake, as well as serum vitamin D and calcium status. 4. Materials and Methods 4.1. Design of the Study The study was carried out on a group of 71 subjects recruited via paper ﬂyers left in doctors’ surgeries in the city of Pozna ´n. Subjects who, in their dietary histories, declared dyspeptic symptoms following the consumption of milk and dairy products (such as frequent ﬂatulence, cramps, chronic or recurrent diarrhea, symptoms of gastro-esophageal reﬂux), as well as asymptomatic volunteers, were invited to participate in the study. The study was conducted in accordance with the procedure (Figure 2). The type I error probability of α = 0.05 was used to calculate the minimum sample size. The sample size was calculated using Statistica StatSoft 13.3 data analysis software based on a hypothesis test for a difference between the two population means of total serum vitamin D levels (23.3 ng/mL vs. 18.04 ng/mL) with standard deviation (Sigma = 7.09). The calculated minimum sample size was 21 people, with a test power of 0.7167. The participants were made aware of the purpose of the study, the manner in which it was going to be conducted, the fact that participation was voluntary and of their option to opt out at any stage without providing a reason. This project and study were in compliance with the Declaration of Helsinki guidelines. This study was carried out pursuant to Approval No. 1109/18 and 1068/19 granted by the Pozna ´n University of Medical Sciences. Written consent, including for blood sampling and genetic testing, was obtained from each participant. The following exclusion criteria were applied: less than 18 years old, systemic diseases (diabetes, kidney and liver diseases, chronic inﬂammatory conditions), coeliac disease, Crohn’s disease, inﬂammatory bowel disease, malignancies, vitamin D or calcium metabolism disorders, obesity, pregnancy, antibiotic therapy during the study period and use of corticosteroids. Individuals using vitamin D or calcium supplements were also excluded. Sixty-three individuals between 21 and 31 years of age were included in the study. Eight patients did not meet the inclusion criteria. The study was conducted between November 2019 and March 2020 during the autumn and winter period to minimize the effects of UV-B-induced vitamin D biosynthesis. An amount of 10 mL of blood was collected from each participant 12 h after their last meal. The blood samples were taken from an elbow vein venipuncture into an anticoagulant (EDTA) tube at the Central Gynaecological Obstetric Laboratory at the Medical University of Pozna ´n Clinical Hospital. Qualiﬁed professional laboratory staff collected the blood samples. All safety precautions were strictly observed whilst collecting blood. The blood obtained for DNA extraction and the serum for biochemical analyses were stored at −80 ◦C. Subjects’ dietary history and genotyping were used to diagnose hypolactasia. 4.2. Genotypes The incidence of the following gene polymorphism genotypes and allele was deter- mined: LCT-13910 C>T, VDR BsmI (1024 + 283 G > A) and VDR FokI (c.2T > C), using the polymerase chain reaction–restriction fragment length polymorphism method (PCR-RFLP). The 200 µL blood samples were collected for the needs of the study. Genomic DNA was extracted from peripheral blood leukocytes using a blood mini kit from A&A Biotechnology (Poland). The manufacturer’s guidelines were followed in the process. A 7415 nanospec- trophotometer (Jenway®, Chicago, IL, USA) was used to determine the obtained DNA Int. J. Mol. Sci. 2023, 24, 10191 18 of 24 concentrations. Values of the A260/A280 coefﬁcients within the range of 1.8–2.0. were calculated on the basis of absorbance measurements at wavelengths of 260 and 280 nm. Single nucleotide polymorphisms of the LCT and VDR gene were selected using Variation Viewer: https://www.ncbi.nlm.nih.gov/variation/ accessed on 9 June 2023. DNA was ampliﬁed in thermal cycles using the PCR master mix plus reagent (A&A Biotechnology, Gda ´nsk, Poland). The mix contained optimal concentrations of Taq DNA polymerase, PCR buffer, MgCl2, nucleotides and stabilizers to capture polymerization reaction inhibitors, dye and loading buffers, as well as appropriate primers (Laboratory of Sequencing and Oligonucleotide Synthesis (PAS, Warsaw, Poland). Starters were selected on the basis of [80]. Their sequences were additionally checked using the Primer3Plus soft- ware [81] (https://www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi, accessed on 9 June 2023). • • • • The following starters were used: For LCT: Forward—5(cid:48)-GCTGGCAATACAGATAAGATAATGGA-3(cid:48) Reverse—5(cid:48)-CTGCTTTGGTTGAAGCGAAGAT-3(cid:48) For VDR BsmI: F—5(cid:48)-GGCAACCAGACTACAAGTACC-3(cid:48) R—5(cid:48)-TCTTCTCACCTCTAACCAGCG-3(cid:48) For VDR FokI: F—5(cid:48)-AGCTGGCCCTGGCACTGACTCTGCTCT-3(cid:48) R—5(cid:48)-ATGGAAACACCTTGCTTCTTCTCCCTC–3(cid:48) PCR reaction conditions to determine the lactose intolerance polymorphism were as follows: initial denaturation: 180 s at 94 ◦C followed by 35 cycles; denaturation: 45 s at 94 ◦C; primer binding: 45 s at 58 ◦C; elongation: 120 s at 72 ◦C; ﬁnal synthesis: 300 s at 72 ◦C. • • PCR conditions for VDR polymorphisms were as follows: initial denaturation: 240 s at 94 ◦C; denaturation: 40 s at 94 ◦C; primer binding: 40 s at 60 ◦C for BsmI and at 55 ◦C for FokI; elongation: 100 s at 72 ◦C; 31 cycles for BsmI and 35 cycles for FokI; ﬁnal synthesis: 180 s at 72 ◦C. Independent PCR reactions were carried out for inconclusive genotypes. PCR products were digested using the Hinf I, Mva1269I and BseGI restriction enzymes (EURx, Gda ´nsk, Poland). The procedure was carried out according to the manufacturer’s instructions. It generated fragments of different lengths depending on the presence of a polymorphic restriction site at one or both ends. The resulting digestion products were identiﬁed by 3% agarose gel electrophoresis. The results were visualized under UV light and photographed. Genotypes were identiﬁed by the presence or absence of an appropriate restriction site. The obtained electrophoresis results were read independently by two people. The following genotypes were found for the LCT gene polymorphism: CC (201 bp, no restriction site), CT (201 + 177 + 24 bp) and TT (177 + 24 bp). For the BsmI polymorphism: AA (837 bp, no restriction site), GA (837 + 648 +189 bp) and GG (648 + 189 bp); for FokI: genotype CC (265 bp, no restriction site), CT (265 + 196 + 69 bp) and TT (196 + 69 bp). 4.3. Anthropometric Parameters A body composition analysis using a Tanita MC-780 body composition analyzer (Poland) was carried out for each participant using the bioelectrical impedance method. Body weight, BMI, body fat (%) and water (%), as well as bone mass, were measured. The tests were performed methodologically at the same time of day and at a minimum of 3 h after a meal and physical activity. Measurements were taken using a step-on analyzer. Body height was measured using a measuring rod (Seca, Hamburg, Germany), with an accuracy of up to 0.5 cm. Int. J. Mol. Sci. 2023, 24, 10191 19 of 24 4.4. Assessment of Nutrition Nutrient intake was assessed on the basis of a dietary history interview conducted in accordance with the guidelines of the Food and Nutrition Institute in Warsaw. As part of the assessment, the respondents, supervised by a dietician, completed a 7-day diet questionnaire. Detailed instructions were given to participants during their ﬁrst visit. Digital databases developed on the basis of “Food composition table” were used to analyze the results of the qualitative and quantitative questionnaire on the composition of total daily food intake. The energy values of diets, intake and energy percentage from protein and fat and carbohydrates in daily food portions, as well as estimated intakes of vitamin D, calcium and phosphorus, were determined using the Diet 6.0 certiﬁed nutrition software package (Institute of Food and Nutrition, Poland). 4.5. Calcium Intake from Dairy Products The assessment of the frequency (times/person/day) and amount (g/person/day) of dairy products habitually consumed in the past 6 months and the amount of calcium consumed with dairy products (mg/person/day) and as part of a daily food portion (mg/person/day) was performed using the validated ADOS-Ca questionnaire (a diagnostic test to assess calcium intake) [82]. The core part of the ADOS-Ca questionnaire comprised (closed) questions on the habitual intake frequency and amount for 11 dairy products. These included milk, but- termilk/keﬁr, hard cheese, fresh cheese, processed cheese, natural yoghurt, fruit yoghurt, cream, ice-cream (during and out of season), homogenized cheese and “Fromage” -type cheese. Multiple choice answers to questions on intake amounts were chosen individ- ually for each product and expressed in common household measures. The ADOS-Ca questionnaire also included questions on current and past nutrition habits, lifestyle and other osteoporosis risk factors. The list of dairy products was developed based on an analysis of the composition of food consumed by Poles. Three classes of calcium intake were established, with 66.7% and 90% of the recommended dietary intake (RDI) amount adopted as safe thresholds: Ca < 66.7% of the recommended amount, 66.7% ≤ Ca < 90% of the recommended amount and Ca ≥ 90% of the recommended amount. 4.6. Biochemical Research 4.6.1. Determination of Vitamin D Metabolites A validated HPLC method (1220 Inﬁnity II, Agilent Technologies, USA) was used to determine vitamin D metabolites (25(OH)D2 and 25(OH)D3). Separation was carried out in reversed-phase mode with UV detection. Absorbance was measured at λ = 265 nm. Analytes were separated in a LiChroCART®250-4 Superspher®60R-P-select B analytical column (MERCK, Darmstadt, Germany), 250 mm × 4 mm, load: 3.2 cm3. Methanol and water (80:20, v/v) were used in the mobile phase at a ﬂow rate of 1 cm3/min. The 25(OH)D2 and 25(OH)D3 standards (Santa Cruz Biotechnology, Dallas, TX, USA) were used to generate a standard curve. For the determination of vitamin D metabolites, 50 µL of blood serum and 50 µL of internal standard (retinol at 10 µg/mL) were added to 500 µL of 50 g/L human albumin solution. Salting out was performed by adding 350 µL of methanol and propanol mixture (8:2, v/v), mixing for 30 s and then adding 2000 µL of n-hexane. The samples were mixed and centrifuged for 10 min at 3000× g. Then, the top hexane layer was collected. An amount of 2000 µL of n-hexane was added again to the remaining solution. Then, it was centrifuged and the top layer of the solution was collected and combined with the previous one. The organic layer was evaporated at 40 ◦C under a stream of nitrogen. The resulting residue was dissolved in 120 µL of phase (methanol/water (80:20, v/v), then it was mixed and centrifuged. An amount of 100 µL was collected and injected into the HPLC system. A linear method over a concentration range of 1–100 ng/mL was used. Method accuracy, expressed by relative standard deviation, was 18.5%. An approximate Int. J. Mol. Sci. 2023, 24, 10191 20 of 24 recovery rate of 72% was obtained for 25(OH)D2 and 25(OH)D3 from human serum [83]. An example chromatogram of the separation of vitamin D metabolites is presented in the Supplementary Materials (Figure S1). 4.6.2. Determination of Calcium Validated atomic absorption spectrometry (iCE 3000 Series, AAS, Thermo Scientiﬁc, Cambridge, UK) was used to determine serum calcium concentration. Sample dissolution was carried out using a microwave accelerated reaction system (MARS 6, CEM Corporation, Matthews, NC, USA). For this purpose, 1 mL of blood serum was collected in a microwave vessel and 7 mL of 69% ultra-pure nitric acid (ROMIL, Cambridge, UK) was added. The vessels were gently mixed and left open for 15 min to facilitate initial sample digestion. The microwave heating program consisted of two stages: increase from ambient temperature to 180 ◦C in 20 min and then this temperature was maintained for a further 20 min. After cooling, the resulting solutions were transferred to 50 mL volumetric ﬂasks and diluted to 20 mL with ultra-pure water. All analyses were repeated three times. The correct serum calcium concentration range was between 2.25 and 2.75 mmol/L [84]. 4.7. Statistical Analysis Statistica 13.3 software (StatSoft, TIBCO Software Inc., Palo Alto, CA, USA) was used to perform statistical analyses. Descriptive statistics were used to describe the basic nutritional parameters, as well as serum vitamin D and calcium concentrations of the subjects. The Shapiro–Wilk test was used to test the data distribution for normality. The statistical signiﬁcance of the differences between the study groups of individuals for data with a normal distribution was analyzed using the Student’s t-test; the data are presented as means with standard deviations. In the absence of a normal distribution, the Mann– Whitney U test was used and data were presented as medians with interquartile ranges (IQR). Minimum and maximum values were calculated. When comparing the signiﬁcance of differences between multiple groups, a Kruskal– Wallis ANOVA was used (in the absence of a normal distribution) and a Tukey’s RIR post hoc analysis test was used for normal distributions. For all tests used, p < 0.05 was considered statistically signiﬁcant. A χ2 test was used to test the Hardy–Weinberg equilibrium, as well as allele and genotype frequencies. Pearson’s and Spearman’s correlation coefﬁcient tests were used to determine the correlation between the parameters studied. A multivariate logistic regression model was used to determine the relationship between impaired lactose tolerance and serum vitamin D, as well as calcium status and intake. The odds ratio (OR) and a 95% conﬁdence interval (CI) were calculated. 5. Conclusions To sum up, the -13910 CC genotype associated with hypolactasia signiﬁcantly affects the consumption of milk and dairy products. Eliminating dairy products from the diet is associated with a reduced intake and vitamin D and calcium deﬁciency in the Polish population. The mutated A allele of VDR gene BsmI polymorphism present in people with hypolactasia may contribute to an increased risk of vitamin D deﬁciency. Exclusion of milk sugar from the diet, combined with impaired vitamin D metabolism, may also lead to inhibited calcium absorption by the human body. The body’s calcium and vitamin D status should therefore be monitored. Further studies on a larger number of participants should be carried out to clarify the role of molecular markers of vitamin D and calcium supply status in individuals suffering from hypolactasia. Supplementary Materials: The supporting information can be downloaded at: https://www.mdpi. com/article/10.3390/ijms241210191/s1. Author Contributions: Conceptualization and visualization, M.K.; methodology, M.K. and D.L.; analysis of data and their interpretation, M.K.; validation, M.K.; performance of experiments and Int. J. Mol. Sci. 2023, 24, 10191 21 of 24 preparation of results, M.K. and G.K.; writing—original draft preparation, M.K.; writing—review and editing, M.K., G.K., D.L. and J.P.; supervision, J.P. and D.L.; graphic and photographic documentation, M.K. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of Pozna ´n University of Medical Sciences (protocol code 1109/18 and 1068/19). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy protection. Conﬂicts of Interest: The authors declare no conﬂict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript or in the decision to publish the results. References 1. Płoszaj, T.; J˛edrychowska-Da ´nska, K.; Witas, H. Frequency of Lactase Persistence Genotype in a Healthy Polish Population. Cent. Eur. J. Biol. 2011, 6, 176–179. [CrossRef] 2. Marasz, A. Frequency and clinical overview of hypolactasia among children, adolescents and students of Szczecin schools. 3. Pomeranian J. Life Sci. 2015, 61, 207–213. [CrossRef] Tomczonek-Moru´s, J.; Wojtasik, A.; Zeman, K.; Smolarz, B.; B ˛ak-Romaniszyn, L. 13910C>T and 22018G>A LCT Gene Polymor- phisms in Diagnosing Hypolactasia in Children. United Eur. Gastroenterol. J. 2019, 7, 210–216. [CrossRef] [PubMed] 4. Mill, D.; Dawson, J.; Johnson, J.L. Managing Acute Pain in Patients Who Report Lactose Intolerance: The Safety of an Old 5. 6. 7. 8. Excipient Re-Examined. Ther. Adv. Drug Saf. 2018, 9, 227–235. [CrossRef] [PubMed] Fassio, F.; Facioni, M.S.; Guagnini, F. Lactose Maldigestion, Malabsorption, and Intolerance: A Comprehensive Review with a Focus on Current Management and Future Perspectives. Nutrients 2018, 10, 1599. [CrossRef] [PubMed] Facioni, M.S.; Raspini, B.; Pivari, F.; Dogliotti, E.; Cena, H. Nutritional Management of Lactose Intolerance: The Importance of Diet and Food Labelling. J. Transl. Med. 2020, 18, 260. [CrossRef] Toca, M.D.C.; Fernández, A.; Orsi, M.; Tabacco, O.; Vinderola, G. Lactose Intolerance: Myths and Facts. An Update. Arch. Argent. Pediatr. 2022, 120, 59–66. [CrossRef] Szilagyi, A. Adaptation to Lactose in Lactase Non Persistent People: Effects on Intolerance and the Relationship between Dairy Food Consumption and Evalution of Diseases. Nutrients 2015, 7, 6751–6779. [CrossRef] Szilagyi, A. Review Article: Lactose--a Potential Prebiotic. Aliment. Pharmacol. Ther. 2002, 16, 1591–1602. [CrossRef] 9. 10. M ˛adry, E.; Ewa, F.; Walkowiak, J. Lactose Intolerance—Current State of Knowledge. Acta Sci. Pol. Technol. Aliment. 2010, 9, 343–350. 11. Amiri, M.; Diekmann, L.; von Köckritz-Blickwede, M.; Naim, H.Y. The Diverse Forms of Lactose Intolerance and the Putative Linkage to Several Cancers. Nutrients 2015, 7, 7209–7230. [CrossRef] [PubMed] 12. Porzi, M.; Burton-Pimentel, K.J.; Walther, B.; Vergères, G. Development of Personalized Nutrition: Applications in Lactose Intolerance Diagnosis and Management. Nutrients 2021, 13, 1503. [CrossRef] [PubMed] 13. Mattar, R.; do Socorro Monteiro, M.; Villares, C.A.; dos Santos, A.F.; Carrilho, F.J. Single Nucleotide Polymorphism C/T-13910, Located Upstream of the Lactase Gene, Associated with Adult-Type Hypolactasia: Validation for Clinical Practice. Clin. Biochem. 2008, 41, 628–630. [CrossRef] [PubMed] 14. Campbell, A.K.; Waud, J.P.; Matthews, S.B. The Molecular Basis of Lactose Intolerance. Sci. Prog. 2005, 88, 157–202. [CrossRef] [PubMed] 15. Di Rienzo, T.; D’Angelo, G.; D’Aversa, F.; Campanale, M.C.; Cesario, V.; Montalto, M.; Gasbarrini, A.; Ojetti, V. Lactose Intolerance: From Diagnosis to Correct Management. Eur. Rev. Med. Pharmacol. Sci. 2013, 17 (Suppl. 2), 18–25. 16. Lukito, W.; Malik, S.G.; Surono, I.S.; Wahlqvist, M.L. From “lactose Intolerance” to “Lactose Nutrition”. Asia Pac. J. Clin. Nutr. 2015, 24 (Suppl. 1), S1–S8. [CrossRef] 17. Ridefelt, P.; Håkansson, L.D. Lactose Intolerance: Lactose Tolerance Test versus Genotyping. Scand. J. Gastroenterol. 2005, 40, 822–826. [CrossRef] 18. Deng, Y.; Misselwitz, B.; Dai, N.; Fox, M. Lactose Intolerance in Adults: Biological Mechanism and Dietary Management. Nutrients 2015, 7, 8020–8035. [CrossRef] 19. Ugidos-Rodríguez, S.; Matallana-González, M.C.; Sánchez-Mata, M.C. Lactose Malabsorption and Intolerance: A Review. Food Funct. 2018, 9, 4056–4068. [CrossRef] 20. M ˛adry, E.; Lisowska, A.; Kwiecie ´n, J.; Marciniak, R.; Korzon-Burakowska, A.; Drzymała-Czy ˙z, S.; Mojs, E.; Walkowiak, J. Adult-Type Hypolactasia and Lactose Malabsorption in Poland. Acta Biochim. Pol. 2010, 57, 585–588. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10191 22 of 24 21. Catanzaro, R.; Sciuto, M.; Marotta, F. Lactose Intolerance: An Update on Its Pathogenesis, Diagnosis, and Treatment. Nutr. Res. 2021, 89, 23–34. [CrossRef] [PubMed] 22. Enattah, N.S.; Sahi, T.; Savilahti, E.; Terwilliger, J.D.; Peltonen, L.; Järvelä, I. Identiﬁcation of a Variant Associated with Adult-Type Hypolactasia. Nat. Genet. 2002, 30, 233–237. [CrossRef] [PubMed] 23. De Luca, P.; Iaconis, D.; Biffali, E.; Enza, C.; de Magistris, L.; Riegler, G.; Pappalardo, D.; Amato, M.R.; Iardino, P.; Montanino, C.; et al. Development of a Novel SNP Assay to Detect Lactase Persistence Associated Genetic Variants. Mol. Biol. Rep. 2021, 48, 7087–7093. [CrossRef] [PubMed] 24. Enattah, N.S.; Jensen, T.G.K.; Nielsen, M.; Lewinski, R.; Kuokkanen, M.; Rasinpera, H.; El-Shanti, H.; Seo, J.K.; Alifrangis, M.; Khalil, I.F.; et al. Independent Introduction of Two Lactase-Persistence Alleles into Human Populations Reﬂects Different History of Adaptation to Milk Culture. Am. J. Hum. Genet. 2008, 82, 57–72. [CrossRef] [PubMed] Ségurel, L.; Bon, C. On the Evolution of Lactase Persistence in Humans. Annu. Rev. Genomics Hum. Genet. 2017, 18, 297–319. [CrossRef] 25. 26. Hammer, H.F.; Fox, M.R.; Keller, J.; Salvatore, S.; Basilisco, G.; Hammer, J.; Lopetuso, L.; Benninga, M.; Borrelli, O.; Dumitrascu, D.; et al. European Guideline on Indications, Performance, and Clinical Impact of Hydrogen and Methane Breath Tests in Adult and Pediatric Patients: European Association for Gastroenterology, Endoscopy and Nutrition, European Society of Neurogastroenterology and Motility, and European Society for Paediatric Gastroenterology Hepatology and Nutrition Consensus. United Eur. Gastroenterol. J. 2022, 10, 15–40. [CrossRef] Storhaug, C.L.; Fosse, S.K.; Fadnes, L.T. Country, Regional, and Global Estimates for Lactose Malabsorption in Adults: A Systematic Review and Meta-Analysis. Lancet Gastroenterol. Hepatol. 2017, 2, 738–746. [CrossRef] 27. 28. Misselwitz, B.; Butter, M.; Verbeke, K.; Fox, M.R. Update on Lactose Malabsorption and Intolerance: Pathogenesis, Diagnosis and 29. Clinical Management. Gut 2019, 68, 2080–2091. [CrossRef] Itan, Y.; Jones, B.L.; Ingram, C.J.E.; Swallow, D.M.; Thomas, M.G. A Worldwide Correlation of Lactase Persistence Phenotype and Genotypes. BMC Evol. Biol. 2010, 10, 36. [CrossRef] 30. Alonso, N.; Zelzer, S.; Eibinger, G.; Herrmann, M. Vitamin D Metabolites: Analytical Challenges and Clinical Relevance. Calcif. Tissue Int. 2023, 112, 158–177. [CrossRef] 31. Nikolac Gabaj, N.; Unic, A.; Miler, M.; Pavicic, T.; Culej, J.; Bolanca, I.; Herman Mahecic, D.; Milevoj Kopcinovic, L.; Vrtaric, A. In Sickness and in Health: Pivotal Role of Vitamin D. Biochem. Med. 2020, 30, 020501. [CrossRef] [PubMed] 32. Wang, J.; Shu, B.; Li, C.-G.; Xie, X.-W.; Liang, D.; Chen, B.-L.; Lin, X.-C.; Wei, X.; Wang, L.; Leng, X.-Y.; et al. Polymorphisms of Genes Related to Vitamin D Metabolism and Transportation and Its Relationship with the Risk of Osteoporosis: Protocol for a Multicentre Prospective Cohort Study in China. BMJ Open 2019, 9, e028084. [CrossRef] [PubMed] 33. Ammar, M.; Heni, S.; Tira, M.S.; Khalij, Y.; Hamdouni, H.; Amor, D.; Ksibi, S.; Omezzine, A.; Bouslama, A. Variability in Response to Vitamin D Supplementation According to Vitamin D Metabolism Related Gene Polymorphisms in Healthy Adults. Eur. J. Clin. Nutr. 2023, 77, 189–194. [CrossRef] 35. 34. Abouzid, M.; Kruszyna, M.; Burchardt, P.; Kruszyna, Ł.; Główka, F.K.; Kara´zniewicz-Łada, M. Vitamin D Receptor Gene Polymorphism and Vitamin D Status in Population of Patients with Cardiovascular Disease—A Preliminary Study. Nutrients 2021, 13, 3117. [CrossRef] [PubMed] Sun, H.; Long, S.R.; Li, X.; Ge, H.; Liu, X.; Wang, T.; Yu, F.; Wang, Y.; Xue, Y.; Li, W. Serum Vitamin D Deﬁciency and Vitamin D Receptor Gene Polymorphism Are Associated with Increased Risk of Cardiovascular Disease in a Chinese Rural Population. Nutr. Res. 2019, 61, 13–21. [CrossRef] [PubMed] Seamans, K.M.; Cashman, K.D. Existing and Potentially Novel Functional Markers of Vitamin D Status: A Systematic Review. Am. J. Clin. Nutr. 2009, 89, 1997S–2008S. [CrossRef] 36. 37. Ross, A.C.; Manson, J.E.; Abrams, S.A.; Aloia, J.F.; Brannon, P.M.; Clinton, S.K.; Durazo-Arvizu, R.A.; Gallagher, J.C.; Gallo, R.L.; Jones, G.; et al. The 2011 Report on Dietary Reference Intakes for Calcium and Vitamin D from the Institute of Medicine: What Clinicians Need to Know. J. Clin. Endocrinol. Metab. 2011, 96, 53–58. [CrossRef] 38. Lips, P.; Cashman, K.D.; Lamberg-Allardt, C.; Bischoff-Ferrari, H.A.; Obermayer-Pietsch, B.; Bianchi, M.L.; Stepan, J.; El-Hajj Fuleihan, G.; Bouillon, R. Current Vitamin D Status in European and Middle East Countries and Strategies to Prevent Vitamin D Deﬁciency: A Position Statement of the European Calciﬁed Tissue Society. Eur. J. Endocrinol. 2019, 180, P23–P54. [CrossRef] Järvelä, I.; Torniainen, S.; Kolho, K.-L. Molecular Genetics of Human Lactase Deﬁciencies. Ann. Med. 2009, 41, 568–575. [CrossRef] 39. 40. Home—SNP—NCBI. Available online: https://www.ncbi.nlm.nih.gov/snp/ (accessed on 11 May 2023). 41. Nowak, J.K.; Dybska, E.; Dworacka, M.; Tsikhan, N.; Kononets, V.; Bermagambetova, S.; Walkowiak, J. Ileal Lactase Expression Associates with Lactase Persistence Genotypes. Nutrients 2021, 13, 1340. [CrossRef] 42. Labrie, V.; Buske, O.J.; Oh, E.; Jeremian, R.; Ptak, C.; Gasi ¯unas, G.; Maleckas, A.; Petereit, R.; Žvirbliene, A.; Adamonis, K.; et al. Lactase Non-Persistence Is Directed by DNA Variation-Dependent Epigenetic Aging. Nat. Struct. Mol. Biol. 2016, 23, 566–573. [CrossRef] 43. Alharbi, O.; El-Sohemy, A. Lactose Intolerance (LCT-13910C>T) Genotype Is Associated with Plasma 25-Hydroxyvitamin D Concentrations in Caucasians: A Mendelian Randomization Study. J. Nutr. 2017, 147, 1063–1069. [CrossRef] [PubMed] 44. Di Costanzo, M.; Berni Canani, R. Lactose Intolerance: Common Misunderstandings. Ann. Nutr. Metab. 2018, 73 (Suppl. 4), 30–37. [CrossRef] [PubMed] Int. J. Mol. Sci. 2023, 24, 10191 23 of 24 45. Corgneau, M.; Scher, J.; Ritie-Pertusa, L.; Le, D.T.L.; Petit, J.; Nikolova, Y.; Banon, S.; Gaiani, C. Recent Advances on Lactose Intolerance: Tolerance Thresholds and Currently Available Answers. Crit. Rev. Food Sci. Nutr. 2017, 57, 3344–3356. [CrossRef] [PubMed] 46. Lomer, M.C.E.; Parkes, G.C.; Sanderson, J.D. Review Article: Lactose Intolerance in Clinical Practice--Myths and Realities. Aliment. Pharmacol. Ther. 2008, 27, 93–103. [CrossRef] [PubMed] 47. Yuan, S.; Sun, J.; Lu, Y.; Xu, F.; Li, D.; Jiang, F.; Wan, Z.; Li, X.; Qin, L.-Q.; Larsson, S.C. Health Effects of Milk Consumption: Phenome-Wide Mendelian Randomization Study. BMC Med. 2022, 20, 455. [CrossRef] 48. Chin, E.L.; Huang, L.; Bouzid, Y.Y.; Kirschke, C.P.; Durbin-Johnson, B.; Baldiviez, L.M.; Bonnel, E.L.; Keim, N.L.; Korf, I.; Stephensen, C.B.; et al. Association of Lactase Persistence Genotypes (Rs4988235) and Ethnicity with Dairy Intake in a Healthy U.S. Population. Nutrients 2019, 11, 1860. [CrossRef] 49. Vitellio, P.; Celano, G.; Bonfrate, L.; Gobbetti, M.; Portincasa, P.; De Angelis, M. Effects of Biﬁdobacterium Longum and Lactobacillus Rhamnosus on Gut Microbiota in Patients with Lactose Intolerance and Persisting Functional Gastrointestinal Symptoms: A Randomised, Double-Blind, Cross-Over Study. Nutrients 2019, 11, 886. [CrossRef] 50. Mnich, B.; Spinek, A.E.; Chyle ´nski, M.; Sommerfeld, A.; Dabert, M.; Juras, A.; Szostek, K. Analysis of LCT-13910 Genotypes and Bone Mineral Density in Ancient Skeletal Materials. PLoS ONE 2018, 13, e0194966. [CrossRef] 51. Popadowska, A.; Kempi ´nska-Podhorodecka, A. LCT-13910 C>T gene polymorphism and obesity in women. Med. ´Sr. Environ. Med. 2012, 2, 48–54. 52. Popadowska, A.; Kempinska-Podhorodecka, A. Relation of the C/T-13910 LCT Polymorphism with Body Composition Measures and Their Modulation by Dairy Products in a Caucasian Men. Am. J. Men’s Health 2021, 15, 15579883211007272. [CrossRef] [PubMed] Setty-Shah, N.; Maranda, L.; Candela, N.; Fong, J.; Dahod, I.; Rogol, A.D.; Nwosu, B.U. Lactose Intolerance: Lack of Evidence for Short Stature or Vitamin D Deﬁciency in Prepubertal Children. PLoS ONE 2013, 8, e78653. [CrossRef] [PubMed] 53. 54. Almon, R.; Sjöström, M.; Nilsson, T.K. Lactase Non-Persistence as a Determinant of Milk Avoidance and Calcium Intake in Children and Adolescents. J. Nutr. Sci. 2013, 2, e26. [CrossRef] [PubMed] 55. de Beer, H. Dairy Products and Physical Stature: A Systematic Review and Meta-Analysis of Controlled Trials. Econ. Hum. Biol. 2012, 10, 299–309. [CrossRef] 56. Angelin, T.C.; Bardosono, S.; Shinta, D.; Fahmida, U. Growth, Dietary Intake, and Vitamin D Receptor (VDR) Promoter Genotype 57. in Indonesian School-Age Children. Nutrients 2021, 13, 2904. [CrossRef] Jarosz, M.; Rychlik, E.; Sto´s, K.; Charzewska, J. Normy ˙Zywienia Dla Populacji Polski i Ich Zastosowanie; Nutrition Standards for the Polish Population; Narodowy Instytut Zdrowia Publicznego-Pa ´nstwowy Zakład Higieny: Warsaw, Poland, 2020. 58. Płudowski, P.; Kos-Kudła, B.; Walczak, M.; Fal, A.; Zozuli ´nska-Ziółkiewicz, D.; Sieroszewski, P.; Peregud-Pogorzelski, J.; Lauterbach, R.; Targowski, T.; Lewi ´nski, A.; et al. Guidelines for Preventing and Treating Vitamin D Deﬁciency: A 2023 Update in Poland. Nutrients 2023, 15, 695. [CrossRef] 59. Petrenya, N.; Lamberg-Allardt, C.; Melhus, M.; Broderstad, A.R.; Brustad, M. Vitamin D Status in a Multi-Ethnic Population of Northern Norway: The SAMINOR 2 Clinical Survey. Public Health Nutr. 2020, 23, 1186–1200. [CrossRef] 60. Hodges, J.K.; Cao, S.; Cladis, D.P.; Weaver, C.M. Lactose Intolerance and Bone Health: The Challenge of Ensuring Adequate Calcium Intake. Nutrients 2019, 11, 718. [CrossRef] 61. Obermayer-Pietsch, B.M.; Bonelli, C.M.; Walter, D.E.; Kuhn, R.J.; Fahrleitner-Pammer, A.; Berghold, A.; Goessler, W.; Stepan, V.; Dobnig, H.; Leb, G.; et al. Genetic Predisposition for Adult Lactose Intolerance and Relation to Diet, Bone Density, and Bone Fractures. J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 2004, 19, 42–47. [CrossRef] 63. 62. Rana, S.; Morya, R.K.; Malik, A.; Bhadada, S.K.; Sachdeva, N.; Sharma, G. A Relationship between Vitamin D, Parathyroid Hormone, Calcium Levels and Lactose Intolerance in Type 2 Diabetic Patients and Healthy Subjects. Clin. Chim. Acta Int. J. Clin. Chem. 2016, 462, 174–177. [CrossRef] Stefano, M.D.; Veneto, G.; Malservisi, S.; Cecchetti, L.; Minguzzi, L.; Strocchi, A.; Corazza, G.R. Lactose Malabsorption and Intolerance and Peak Bone Mass. Gastroenterology 2002, 122, 1793–1799. [CrossRef] [PubMed] Jasielska, M.; Grzybowska-Chlebowczyk, U. Hypocalcemia and Vitamin D Deﬁciency in Children with Inﬂammatory Bowel Diseases and Lactose Intolerance. Nutrients 2021, 13, 2583. [CrossRef] [PubMed] 64. 65. Koek, W.N.H.; van Meurs, J.B.; van der Eerden, B.C.J.; Rivadeneira, F.; Zillikens, M.C.; Hofman, A.; Obermayer-Pietsch, B.; Lips, P.; Pols, H.A.; Uitterlinden, A.G.; et al. The T-13910C Polymorphism in the Lactase Phlorizin Hydrolase Gene Is Associated with Differences in Serum Calcium Levels and Calcium Intake. J. Bone Miner. Res. Off. J. Am. Soc. Bone Miner. Res. 2010, 25, 1980–1987. [CrossRef] [PubMed] 66. Yahya, N.F.S.; Daud, N.M.; Makbul, I.A.A.; Aziz, Q.A.S.A. Association of Calcium Intake, Lactose Intolerance and Physical Activity with Bone Health Assessed via Quantitative Ultrasound among Young Adults of a Malaysian University. Arch. Osteoporos. 2021, 16, 14. [CrossRef] 67. Pike, J.W.; Meyer, M.B. The Vitamin D Receptor: New Paradigms for the Regulation of Gene Expression by 1,25-Dihydroxyvitamin D3. Endocrinol. Metab. Clin. N. Am. 2010, 39, 255–269. [CrossRef] 68. Barry, E.L.; Rees, J.R.; Peacock, J.L.; Mott, L.A.; Amos, C.I.; Bostick, R.M.; Figueiredo, J.C.; Ahnen, D.J.; Bresalier, R.S.; Burke, C.A.; et al. Genetic Variants in CYP2R1, CYP24A1, and VDR Modify the Efﬁcacy of Vitamin D3 Supplementation for Increasing Serum 25-Hydroxyvitamin D Levels in a Randomized Controlled Trial. J. Clin. Endocrinol. Metab. 2014, 99, E2133–E2137. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10191 24 of 24 69. Wysocza ´nska-Klaczy ´nska, A.; ´Sl˛ezak, A.; Hetman, M.; Barg, E. The impact of VDR gene polymorphisms on obesity, metabolic changes, bone mass disorders and neoplastic processes. Pediatr. Endocrinol. Diabetes Metab. 2018, 24, 96–105. [CrossRef] 70. Usategui-Martín, R.; De Luis-Román, D.-A.; Fernández-Gómez, J.M.; Ruiz-Mambrilla, M.; Pérez-Castrillón, J.-L. Vitamin D Receptor (VDR) Gene Polymorphisms Modify the Response to Vitamin D Supplementation: A Systematic Review and Meta- Analysis. Nutrients 2022, 14, 360. [CrossRef] Serrano, J.C.E.; De Lorenzo, D.; Cassanye, A.; Martín-Gari, M.; Espinel, A.; Delgado, M.A.; Pamplona, R.; Portero-Otin, M. Vitamin D Receptor BsmI Polymorphism Modulates Soy Intake and 25-Hydroxyvitamin D Supplementation Beneﬁts in Cardiovascular Disease Risk Factors Proﬁle. Genes Nutr. 2013, 8, 561–569. [CrossRef] 71. 72. Divanoglou, N.; Komninou, D.; Stea, E.A.; Argiriou, A.; Papatzikas, G.; Tsakalof, A.; Pazaitou-Panayiotou, K.; Georgakis, M.K.; Petridou, E. Association of Vitamin D Receptor Gene Polymorphisms with Serum Vitamin D Levels in a Greek Rural Population (Velestino Study). Lifestyle Genom. 2021, 14, 81–90. [CrossRef] Santos, B.R.; Mascarenhas, L.P.G.; Satler, F.; Boguszewski, M.C.S.; Spritzer, P.M. Vitamin D Deﬁciency in Girls from South Brazil: A Cross-Sectional Study on Prevalence and Association with Vitamin D Receptor Gene Variants. BMC Pediatr. 2012, 12, 62. [CrossRef] [PubMed] 73. 74. Cobayashi, F.; Lourenço, B.H.; Cardoso, M.A. 25-Hydroxyvitamin D3 Levels, BsmI Polymorphism and Insulin Resistance in Brazilian Amazonian Children. Int. J. Mol. Sci. 2015, 16, 12531–12546. [CrossRef] [PubMed] 75. Valtueña, J.; González-Gross, M.; Huybrechts, I.; Breidenassel, C.; Ferrari, M.; Mouratidou, T.; Gottrand, F.; Dallongeville, J.; Azzini, E.; Sioen, I.; et al. Factors Associated with Vitamin D Deﬁciency in European Adolescents: The HELENA Study. J. Nutr. Sci. Vitaminol. 2013, 59, 161–171. [CrossRef] Jakubowska-Pietkiewicz, E.; Fendler, W.; Porczy ´nski, M.; Chlebna-Sokół, D. Vitamin D receptor gene (BsmI and FokI) polymor- phism and calcium-phoshporus indices in children from Lodz region. Pediatr. Endocrinol. 2016, 15, 31–40. [CrossRef] Szilagyi, A.; Ishayek, N. Lactose Intolerance, Dairy Avoidance, and Treatment Options. Nutrients 2018, 10, 1994. [CrossRef] [PubMed] 77. 76. 79. 78. da Silva, A.T.; de Lima, J.J.; Reis, P.; Passos, M.; Baumgartner, C.G.; Sereno, A.B.; Krüger, C.C.H.; Cândido, L.M.B. Application of Lactose-Free Whey Protein to Greek Yogurts: Potential Health Beneﬁts and Impact on Rheological Aspects and Sensory Attributes. Foods 2022, 11, 3861. [CrossRef] [PubMed] Silanikove, N.; Leitner, G.; Merin, U. The Interrelationships between Lactose Intolerance and the Modern Dairy Industry: Global Perspectives in Evolutional and Historical Backgrounds. Nutrients 2015, 7, 7312–7331. [CrossRef] [PubMed] Seremak-Mrozikiewicz, A.; Drews, K.; Mrozikiewicz, P.M.; Bartkowiak-Wieczorek, J.; Marcinkowska, M.; Wawrzyniak, A.; Slomski, R.; Kalak, R.; Czerny, B.; Horst-Sikorska, W. Correlation of Vitamin D Receptor Gene (VDR) Polymorphism with Osteoporotic Changes in Polish Postmenopausal Women. Neuro Endocrinol. Lett. 2009, 30, 540–546. 80. 81. Untergasser, A.; Cutcutache, I.; Koressaar, T.; Ye, J.; Faircloth, B.C.; Remm, M.; Rozen, S.G. Primer3--New Capabilities and 82. Interfaces. Nucleic Acids Res. 2012, 40, e115. [CrossRef] Szymelfejnik, E.J.; W ˛adołowska, L.; Cichon, R.; Przysławski, J.; Bolesławska, I. Dairy Products Frequency Questionnaire (ADOS- Ca) Calibration for Calcium Intake Evaluation. Pol. J. Food Nutr. Sci. 2006, 15, 229–236. 83. Turpeinen, U.; Hohenthal, U.; Stenman, U.-H. Determination of 25-Hydroxyvitamin D in Serum by HPLC and Immunoassay. Clin. Chem. 2003, 49, 1521–1524. [CrossRef] [PubMed] 84. Dembi ´nska-Kie´c, A.; Solnica, B.; Naskalski, J. Laboratory Diagnostics with Elements of Clinical Biochemistry; Edra Urban & Partner: Wrocław, Poland, 2017; ISBN 978-83-65625-50-2. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijerph20126087	Article Health Workers’ Burnout and COVID-19 Pandemic: 1-Year after—Results from a Repeated Cross-Sectional Survey Eleonora Gambaro 1,, Carla Gramaglia 1,2, Debora Marangon 1, Manuela Probo 3, Marco Rudoni 3 and Patrizia Zeppegno 1,2, 1 Department of Translational Medicine, Università del Piemonte Orientale, 13100 Vercelli, Italy; [email protected] (C.G.) Psychiatry Unit, Maggiore della Carità Hospital, 28100 Novara, Italy 2 3 Department of Mental Health, ASL NOVARA, 28100 Novara, Italy * Correspondence: [email protected] (E.G.); [email protected] (P.Z.) Abstract: (1) Background: This study evaluates, one year later, the levels of burnout, anxious–depressive, and post-traumatic symptoms and the general health status in the Health Workers (HWs) involved in the SARS-COVID-19 pandemic in the Novara area. (2) Methods: The survey was sent via a link in an email to doctors, nurses, and other operators during the period between June and August 2021. The survey collected socio-demographic data and contained some self-administered questionnaires. (3) Results: A total of 688 HWs completed the survey, 53% were aged 30–49 years, 68% were female, 76% were cohabiting, 55% had children, 86% reported family habit changes, and 20% had non-COVID related health problems. Only a few of the respondents had a follow-up by a specialist (12%), of which there were even less in recent times (6%). It was observed that the respondents had undergone burnout; a poor state of general mental health (62%); depressive symptoms (70%); post-traumatic symptoms (29%); and less frequently, anxious symptoms (16%). The data of this study are in line with other studies in the literature. (4) Conclusions: The data indicate that psychological-based suffering was no longer markedly concentrated in some speciﬁc bands of HWs. In conclusion, it would be essential to enhance HW support strategies. Keywords: COVID-19; health-worker; burnout; cross-sectional; follow-up 1. Introduction Despite our knowledge concerning terms such as “outbreak”, “epidemic”, and “pan- demic”, in recent years, we have had unexpected and painful ﬁrsthand experience of their meaning, not only as health professionals, but also as ordinary citizens. To date, however, the COVID-19 pandemic is far from over, with conﬁrmed cases in the world equal to 759,408,703, including 6,866,434 deaths, as reported to the WHO on 7 March 2023—even despite the administration of a total of 13,232,780,775 vaccines [1]. The COVID-19 pandemic has had an impact on mental health problems [2], which were already an issue for healthcare workers (HWs) before the COVID-19 pandemic [3], and this has been acknowledged by the World Health Organization (WHO) as well [4]. Actually, Health Workers’ mental and psychological health problems might include burnout as well as depressive, anxiety, stress, and post-traumatic stress symptoms, which can inﬂuence their working function [5,6]. Several studies have focused on burnout syndrome during the COVID-19 pandemic, sometimes showing mixed results, with high levels of burnout among HWs (doctors and nurses) [7,8] speciﬁcally engaged in COVID-19 wards (the so- called “frontline” HWs, such are intensive care health workers, emergency room staff, and emergency service rescuers [9], residents) [10] as well as in “non-frontline” HWs [11,12]. Even in Italy, among the first European nations to be hit hard by the pandemic, sev- eral studies have been assessed, and they will be briefly summarized as follows. An in- crease in the three dimensions of the Maslach burnout scale (emotional exhaustion—EE, Citation: Gambaro, E.; Gramaglia, C.; Marangon, D.; Probo, M.; Rudoni, M.; Zeppegno, P. Health Workers’ Burnout and COVID-19 Pandemic: 1-Year after—Results from a Repeated Cross-Sectional Survey. Int. J. Environ. Res. Public Health 2023, 20, 6087. https://doi.org/10.3390/ ijerph20126087 Academic Editors: Albert Nienhaus and Paul B. Tchounwou Received: 11 April 2023 Revised: 7 May 2023 Accepted: 29 May 2023 Published: 8 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Environ. Res. Public Health 2023, 20, 6087. https://doi.org/10.3390/ijerph20126087 https://www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public HealthInt. J. Environ. Res. Public Health 2023, 20, 6087 2 of 22 depersonalization—D, and personal accomplishment—PA) has been found [13] in a popu- lation of 330 HWs in a medical facility in Northern Italy. With more detail, moderate and severe levels were found in 35.7 and 31.9% for EE; 14% and 12.1% for D; and 40.1% and 34.3% experienced a worsening in PA, respectively. A multicenter Italian study found higher rates of burnout in nurses (high EE in 77.4%, increased risk of D in 68.7%, and 77.9% exhibited an increased risk of minimized individual PA), especially in those working in COVID wards or in intensive care units [14]. These results from single studies were supported by some systematic reviews [15,16]; furthermore, [17] highlighted a high psychological impact and high emotional fatigue in Italian HWs. Suggestions have been raised [18] that paying attention to mental health issues, reduc- ing the workload of HWs by regulating their shifts, reducing work-related stressors, and creating a healthy work environment can prevent or reduce burnout. In addition to burnout, during the current pandemic, many studies have focused on HW’s experiences of depressive symptoms, stress, and anxiety. A systematic review and meta-analysis [19] identiﬁed 13 studies assessing the prevalence of mental post-traumatic stress symptoms in HWs, with a 23.1% prevalence for anxiety and 22.8% for depressive symptoms. Higher rates of both anxiety and depression were found in women, especially in nursing roles. The variety of results found in the very rich literature available about the topic is difﬁcult to interpret considering that most studies rely on self-report measures and that these are not consistently used across studies. Despite these limits, studies are consistent in underscoring the presence of high levels of depressive, post-traumatic stress, and anxiety symptoms, especially in frontline HWs [20–24]. Overall, post-traumatic stress and the above-described symptoms were related to feelings of vulnerability, loss of control and concerns about one’s health, COVID-19 infection-related physical symptoms, the possibility of spreading the virus to family members and acquaintances, changes in work, isolation. With more detail, post-traumatic stress symptoms were related, among others, to the unpredictability of daily workloads, the management of the expectations of patients and their families in unforeseen critical cases/situations, decision load, high daily mortality rates, constant changes, and updates to hospital procedures [24]. Post-traumatic stress disorder (PTSD) is a common psychiatric condition, occurring after direct or indirect exposure to a traumatic event [22,25], such as the COVID-19 pan- demic. HWs have faced unprecedented demands, both professionally and personally, in managing an unclear disease with a high mortality rate. They found themselves forced to make difﬁcult ethical decisions and to work in conditions where they feared for both themselves and for their loved ones [23]. Several studies have been performed worldwide to identify possible targets to decrease the mental and psychological health impact of the COVID-19 pandemic [26,27]. Our research group has focused particularly on the issue of burnout and mental health among HWs. In previous works involving HWs working in different settings during the very ﬁrst wave of the COVID-19 pandemic in the Maggiore della Carità University Hospital, in Novara, Italy [20,28], a strong correlation was found between depressive symptoms, mental post-traumatic stress, health perception, and anxiety symptoms. Furthermore, in HWs, there was found an effect of job burnout on stress and anxiety, post-traumatic stress, and depressive symptoms levels, with evidence of an inner increase in mental suffering (depressive symptoms and job burnout) in the event of changes in job tasks and responsibilities. Brieﬂy, to summarize the above-described evidence: the COVID-19 pandemic, both in the emergency and in the current phase, had a strong impact on HW’s psychological and physical stress, which at the beginning were even worsened by the discomfort of isolation linked to containment measures. Therefore, this research intends to assess, one year after the previous survey, how burnout rates have evolved, as well as the extent of the anxious–depressive and post- Int. J. Environ. Res. Public Health 2023, 20, 6087 3 of 22 traumatic symptoms in the study population. It is very important to reevaluate the suf- fering of HWs after some time, given that the previous study highlighted high levels of psychological burden during the 2019 coronavirus pandemic [20,28]. Therefore, the objective of this study was to repeat the measure of the levels of burnout (primary outcome), depressive, anxiety and post-traumatic stress symptoms, and general health status (secondary outcomes) in the HWs who were employed in the COVID-19 pandemic, both at the hospital and territorial level in Novara, through the administration of a test protocol that included Maslach Burnout [29], the Beck Anxiety Inventory [30], a Beck Depression Inventory-II (BDI-II) [31], the Impact of Event Scale—Revised (IES) [32], and the 12 Item General Health Questionnaire (GHQ-12), respectively. 2. Materials and Methods This cross-sectional study involved the population of HWs (medical doctors/ physicians—including medical executives and residents in training—nurses and “oth- ers” such as psychologists, social workers, radiology, and laboratory technicians, educators) employed at the Maggiore della Carità University Hospital or the Mental Health Center of the Local Health Unit in Novara, the general practitioners and the related continuity of care doctors to the Local Health Unit, medical and nursing staff of the Health Emergency Service Territorial 118 (SEST118) of the Maggiore della Carità University Hospital, active during the period of the COVID-19 health emergency. The reference population was composed of 897 HWs, initially contacted. The catchment area of the Maggiore della Carità University Hospital includes about 725,000 inhabitants and it is the hub hospital of the area, to which the 5 hospitals spoken of refer. During the emergency phase, the hospital was re-arranged to obtain 77 beds available for COVID patients: 47 ordinary beds, 14 sub-intensive beds, and 16 intensive beds. The medical staff was rearranged as well, to address the needs of the emergency, and this re-arrangement went on even later to adapt to the ongoing changes of the pandemic course. The current research project follows the rules for clinical research [29] approved by the Intercompany Ethics Committee of Novara (Protocol 82/20). This research follows an amendment that was requested in March 2021 to continue the research project of which the results were published in two previous publications [20,28], to which certain non- substantial changes have been made: willingness to participate in a follow-up assessment, which was decided based on the persistence of the pandemic emergency. The second administration took place from 1 June to 31 August 2021. Between-subjects analyses were carried out; two separate cross-sectional samples were collected, with only partly overlapping participants. The primary and secondary outcomes were assessed with an online survey predis- posed ad hoc, featuring two parts: the ﬁrst aimed at collecting information about partici- pants and the second was a group of standardized and validated psychometric measures (see below for more detail). The survey was implemented with the REDCap platform and e-mailed at the end of the third wave of the COVID pandemic emergency crisis period (in June 2021), on behalf of the human resources ofﬁces in charge of the healthcare institutions, who have access to the mailing lists including the institutional e-mail contacts of all HW employees. This strategy was adopted in order to offer everyone the opportunity to take part in the survey, while granting anonymity with the use of the REDCap link to ﬁll in the survey. Data gathering closed at the end of August 2021. The ﬁrst part of the survey gathered a wide range of information: socio-demographic and clinical information (such as gender, age, ethnicity, marital status and children, medical history of psychiatric type), information concerning employment (as a job category and job change during the emergency, information on the use of Personal protective equipment (PPE), any variation in the number of working hours and if the HW has vaccinated against SARS-CoV-2), concerning the COVID-19 infection and pathology (such as, for example, the ﬁnding of positivity, the onset of related symptoms or non-COVID19 related health problems, the presence of affected relatives) and the change of habits during the pandemic) Int. J. Environ. Res. Public Health 2023, 20, 6087 4 of 22 and a test part (Maslach Burnout [29], the Beck Anxiety Inventory [30], a Beck Depression Inventory-II (BDI-II) [31], the Impact of Event Scale—Revised (IES) [32], and the 12-Item General Health Questionnaire [33]), which can be ﬁlled in online. The link for the online anonymous survey was sent; it was possible to ﬁll in the protocol in one or more sessions; its completion required a total duration of about 30 min. Participant HWs had to give consent to participate in the study and were granted anonymity. In the ﬁrst screen, the participants were offered a thorough explanation of the research protocol, and after that, they were asked to agree/disagree on consent for participation. No participant exclusion criteria were applied, except for failure. In our sample, four main subgroups of HWs could be identiﬁed: medical doc- tors/physicians, residents in training (i.e., graduated medical doctors attending specializa- tion schools), nurses, and “others” (including psychologists, social workers, radiology, and laboratory technicians, educators). The methods used in this study have been described in detail elsewhere [20,28]. 2.1. Statistical Analysis of Data Data are presented in aggregate form, and it is not possible to trace information or make comparisons at the individual level. The data were synthesized by the median for the continu- ous variables and with absolute percentages and frequencies for the categorical data. Group comparisons were made using Wilcoxon’s test (Mann–Whitney and Kruskal–Wallis tests) for continuous variables and Pearson chi-square tests or Fisher’s exact test for categorical variables. The estimates were adjusted in a logistic model that considered the covariates as possible confounding factors that can be related, for example, to participants’ gender or age. The analysis was performed using the R 3.6.2 software (R Foundation for statistical computing, Vienna, Austria) for Windows. In all performed analyses, a signiﬁcance criterion equal to or less than 0.05 was used to determine statistical signiﬁcance. A p-value < 0.05 and p-value < 0.001 were the cutoffs for statistical signiﬁcance and for strong statistical signiﬁcance. 2.2. Data Processing and Ethical Evaluation Survey The information obtained during the research was processed in compliance with the provisions of the code regarding the protection of personal data (D.L. 196/2003). The study did not present either any risk for participants or any ethical issue. 3. Results 3.1. Description of the First Assessment A total of 897 HWs joined the survey. However, only 653 (73%) of them responded to all questions (i.e., 244 surveys were incomplete), and it was on these that the data analysis was focused. The descriptive data of the sample and the test results are shown in Table 1 and Figure 1, respectively. 3.2. Description of the Second Assessment—1 Year after A total of 688 HWs joined the survey. The descriptive data of the sample and the test results are shown in Table 2 and Figure 2, respectively. Int. J. Environ. Res. Public Health 2023, 20, 6087 5 of 22 Table 1. Socio-demographic and COVID-19 data (n = 653) during the ﬁrst phase. N = 653 Age (mean age = 44.28 years) Gender Marital status Sons Job category Change in job? Changing family habits? Have you tested positive for COVID-19? Did you have any symptoms related to COVID-19? Have you had any health problems that are unrelated to COVID-19? Has anyone dear to you tested positive for the virus? Categories >50 years 30–49 years 18–29 years Female Married/cohabiting Lives alone Yes Doctor Nurse Other Resident doctor Yes Yes No No No No N 227 (35%) 334 (51%) 92 (14%) 443 (68%) 413 (63%) 240 (37%) 358 (55%) 286 (44%) 137 (21%) 131 (20%) 99 (15%) 331 (51%) 564 (86%) 564 (86%) 528 (81%) 556 (85%) 454 (70%) Figure 1. Levels of burnout, anxiety, depression, post-traumatic stress, and mental health in 653 HWs (data expressed as observed frequencies) during the ﬁrst phase. Abbreviations: MBI =Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonalization; PA = Personal Accomplish- ment; SD = Standard Deviation; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES Impact of Event Scale; GHQ = General Health Quesionnaire. Int. J. Environ. Res. Public Health 2023, 20, x 5 of 23 EE±SD D±SD PA±SD BAI±SD BDI±SD IES±SD GHQ±SD Median 20.0 ± 11 11.4 ± 5.5 31.7 ± 5.6 Median 11.3 ± 10.6 28 ± 17 6.1 ± 8.3 13.6 ± 9.5 Figure 1. Levels of burnout, anxiety, depression, post-traumatic stress, and mental health in 653 HWs (data expressed as observed frequencies) during the ﬁrst phase. Abbreviations: MBI =Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonalization; PA = Personal Accomplishment; SD = Standard Deviation; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES Impact of Event Scale; GHQ = General Health Quesionnaire. Table 1. Socio-demographic and COVID-19 data (n = 653) during the ﬁrst phase. N = 653 Categories N Age (mean age = 44.28 years) >50 years 227 (35%) 30–49 years 334 (51%) 18–29 years 92 (14%) Gender Female 443 (68%) Marital status Married/cohabiting 413 (63%) Lives alone 240 (37%) Sons Yes 358 (55%) Job category Doctor 286 (44%) Nurse 137 (21%) Other 131 (20%) Resident doctor 99 (15%) Change in job? Yes 331 (51%) Changing family habits? Yes 564 (86%) Have you tested positive for COVID-19? No 564 (86%) Did you have any symptoms related to COVID-19? No 528 (81%) Have you had any health problems that are unrelated to COVID-19? No 556 (85%) Has anyone dear to you tested positive for the virus? No 454 (70%) 3.2. Description of the Second Assessment—1 Year after A total of 688 HWs joined the survey. The descriptive data of the sample and the test results are shown in Table 2 and Figure 2, respectively. Int. J. Environ. Res. Public Health 2023, 20, 6087 6 of 22 Table 2. Socio-demographic data, COVID-19, and psychiatric history (n = 688) during the second phase. N = 688 Age (mean age = 43.99 years) Gender Marital status Sons Have you tested positive for COVID-19? Have you had symptoms related to COVID-19? Have you had health problems unrelated to COVID-19? Has anyone dear to you tested positive for COVID-19? Changing family habits? Change in job during emergency? Did you participate in the previous edition of the survey? Job category Do you carry out urgent or emergent activities? Area Have you ever been followed by a specialist psychiatrist? Have you taken psychopharmacological therapy in the past? Are you currently followed by a mental health specialist? Do you currently take psychopharmacological therapy? Did you feel protected in the workplace during the pandemic? Do you think you have had the correct number of PPE available during your work? Do you think you have had a correct number of works shifts in the context of COVID-19-related work? How has the COVID-19 emergency inﬂuenced the number of working hours? Have you had the COVID-19 vaccine? Categories >50 years 30–49 years 18–29 years Female Married/cohabiting Lives alone Yes No No No Yes Yes No Yes Doctor Nurse Resident in training Freelance doctor Other Yes Clinic Surgical Services Emergency/Urgency Yes Yes Yes Yes Yes No Not always Yes No Not always Yes No Not always Stable Decreased Increased Yes N 165 (35%) 245 (53%) 55 (12%) 316 (68%) 354 (76%) 111 (24%) 257 (55%) 343 (74%) 338 (73%) 368 (80%) 253 (54%) 399 (86%) 249 (54%) 287 (62%) 143 (31%) 130 (28%) 63 (13%) 38 (8%) 91 (20%) 187 (40%) 196 (42%) 66 (14%) 102 (22%) 101 (22%) 56 (12%) 72 (16%) 29 (6%) 36 (7%) 108 (23%) 98 (21%) 259 (56%) 174 (37%) 77 (17%) 214 (46%) 258 (55%) 97 (21%) 110 (24%) 169 (59%) 22 (5%) 274 (36%) 451 (97%) The sample was mainly composed of HWs aged between 30 and 49 years (53%; mean = 43.99 years), females (68%), married/cohabiting (76%), and with children (55%). A total of 44% were doctors (13% in residents in training). In 46% of cases, HWs had to change their duties and in 86%, also their family habits due to the pandemic emergency. The majority never tested positive to COVID-19 (74%) and never had COVID-19-related symptoms (73%) or other health problems (80%). The family members of the HWs involved in 54% of the cases had tested positive for the virus. Int. J. Environ. Res. Public Health 2023, 20, 6087 7 of 22 Figure 2. Levels of burnout, anxiety, depression, stress, and mental health in the 688 HWs (data expressed as observed frequencies) during the second phase. Abbreviations: MBI =Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonalization; PA = Personal Accomplishment; SD = Standard Deviation; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES = Impact of Event Scale; GHQ = General Health Quesionnaire. Forty percent of the sample had worked in the emergency/urgency ﬁeld, with a prevalence in the clinical area (42%), followed by the service area (22%), and ﬁnally, by the surgical area (14%). As for the psychiatric history, 12% had a lifetime history of treatment on behalf of a psychiatrist while 6% were currently referred to a psychiatrist. Sixteen percent had a previous history of psychiatric medication, while 7% were currently taking it. Most HWs did not feel protected in the workplace during the pandemic (77%), and they also believed that they did not receive adequate PPE (63%). Most believed that they have had a proper number of shifts in the context of their COVID-related activity (55%), and 59% reported that their working hours had remained stable (neither increased nor decreased) despite the pandemic. Almost all the HWs received a COVID vaccine (97%) (Table 1). Concerning the mental health outcomes analyzed, moderate/high scores at the test assessment were considered clinically relevant. Signs of burnout were found in most HWs; in particular, a low level in PA (93%, medium/low, indicative of high burnout; median 30.9—high level of burnout) was found, in more than half (56%) medium/high levels of EE (median 22—moderate level of burnout) were observed; in most participants (86%), there was medium/high D (median 11.9—high level of burnout). Anxiety symptoms as suggested by BAI scores were found in about 15% of cases I (median BAI = 11.8—minimum levels) and depressive symptoms as assessed by the BDI were recorded in 19% of HWs (median BDI = 27.0—moderate levels). In 40% of cases, there were post-traumatic stress symptoms (median IES = 6.2—low level) and in 83% of cases, a poor state of mental health (median GHQ = 13.4—low discomfort) was observed (Figure 2). 3.2.1. Burnout With more detail, concerning burnout, we found higher mean EE levels in the female than in the male population (p < 0.001). The univariate analysis showed that for HWs, being younger was a protective factor against high EE scores (OR age < 33 vs. >54 = 0.68, 95% CI = 0.515–0.899, p = 0.01), while having experienced changes in extra-work habits was correlated with higher EE scores (OR no change habits vs. change habits = 0.366, IC 95% = 0.226–0.592, p < 0.001). Other risk factors for higher EE levels included not having children (OR no children vs. children= 14,793, 95% CI = 10,826–20,213, p = 0.01) but also having non-COVID related health problems (OR non-COVID health problems vs. No = 18,969, 95% CI = 12,945–27,797 (p < 0.001). p = 0). Int. J. Environ. Res. Public Health 2023, 20, x 6 of 23 EE±SD D±SD PA±SD BAI±SD BDI±SD IES±SD GHQ±SD Median 22 ± 12 11.9 ± 6.2 30.9 ± 6.4 Median 11.8 ± 11.9 27 ± 17 6.2 ± 8.9 13.5 ± 9.9 Figure 2. Levels of burnout, anxiety, depression, stress, and mental health in the 688 HWs (data expressed as observed frequencies) during the second phase. Abbreviations: MBI =Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonalization; PA = Personal Accomplishment; SD = Standard Deviation; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES Impact of Event Scale; GHQ = General Health Quesionnaire. Table 2. Socio-demographic data, COVID-19, and psychiatric history (n = 688) during the second phase. N = 688 Categories N Age (mean age = 43.99 years) >50 years 165 (35%) 30–49 years 245 (53%) 18–29 years 55 (12%) Gender Female 316 (68%) Marital status Married/cohabiting 354 (76%) Lives alone 111 (24%) Sons Yes 257 (55%) Have you tested positive for COVID-19? No 343 (74%) Have you had symptoms related to COVID-19? No 338 (73%) Have you had health problems unrelated to COVID-19? No 368 (80%) Has anyone dear to you tested positive for COVID-19? Yes 253 (54%) Changing family habits? Yes 399 (86%) Change in job during emergency? No 249 (54%) Did you participate in the previous edition of the survey? Yes 287 (62%) Job category Doctor 143 (31%) Nurse 130 (28%) Resident in training 63 (13%) Freelance doctor 38 (8%) Other 91 (20%) Do you carry out urgent or emergent activities? Yes 187 (40%) Area Clinic 196 (42%) Surgical 66 (14%) Services 102 (22%) Emergency/Urgency 101 (22%) Int. J. Environ. Res. Public Health 2023, 20, 6087 8 of 22 The highest EE scores (≥30), were observed in doctors working in the clinical area (51%; p = 0.001) and by those who reported an increase in their working hours (70%; p ≤ 0.001). Being younger and not having children emerged, respectively, as a protective and a risk factor also as far as D scores are concerned (respectively: OR < 33 vs. >54 = 0.520, 95% CI = 0.392–0.688, p ≤ 0.001; OR no children vs. children = 17,061, 95% CI = 12,428–23,421, p ≤ 0.001). This was also the case for those who modiﬁed their family habits (OR no change habits vs. change habits = 0.505, IC95% = 0.330–0.773, p ≤ 0.001), but also in those whose habits changed due to the fear of infecting their loved ones (OR no modiﬁcation for fear vs. changes for fear = 0.447, IC95% = 0.248–0.804, p = 0.01). As for EE, the highest D scores (≥12), were found in HWs in the clinical area (50%, p = 0.005) and in those who had increased their work hours (68%, p ≤ 0.001). Less statistically significant correlations emerged for the PA scale. However, when aggregating family status variables (Single/Married/Divorced/Cohabiting/Widower/In a relationship) to only the categories of “lives alone” and “cohabitant”, higher levels of PA were found in HWs living alone (OR lives vs. lives alone = 15,586, IC95% = 10.311–2.356, p = 0.04). The indicative values of burnout in the two scales of the Maslach test were observed in those who changed their job due to the emergency when compared to those who continued to perform their work (EE p ≤ 0.001; D p ≤ 0.001) (Tables 3 and 4). Table 3. Focus Burnout measured with Maslach Burnout Inventory (MBI) scale: categorical data analysis based on biographical data and test results (n, % column) during the second phase. Only signiﬁcant differences (p < 0.05) are shown in the table. N = 688 Cut-off Subscale, N Participants Burnout Level OVER CUT OFF—EE OVER CUT OFF—D OVER CUT OFF—PA ≤17 N = 393 Low 18–29 N = 139 Middle ≥30 N = 156 High p ≤05 N = 238 Low 06–11 N = 196 Middle ≥12 N = 254 High p ≤34 N = 538 High 35–39 N = 114 Middle ≥40 N = 36 Low p Unmarried 15% (39) 22% (30) 22% (34) 17% (18) 16% (31) 21% (54) Married 55% (144) 46% (64) 47% (73) 56% (61) 57% (111) 43% (109) Marital status (I) Divorced 7% (18) 6% (8) 1% (2) 0.049 6% (7) 7% (13) 3% (8) 0.026 Cohabiting 13% (34) 18% (25) 22% (35) 11% (12) 13% (26) 22% (56) Widower 0% (1) 1% (1) 1% (1) 0% (0) 1% (1) 1% (2) In relation 10% (26) 8% (11) 7% (11) 9% (10) 7% (13) 10% (25) Living situation (II) Lives alone Cohabit Have children Non-COVID 19 related health issues Change habits due to the pandemic Change habits due to the fear of infecting loved ones Change in job Participation in the ﬁrst survey Yes No Yes No Yes No Yes No Yes No Yes No General practition- ers 22% (88) 33% (37) 25% (9) 78% (320) 67% (76) 75% (27) 0.048 59% (155) 50% (70) 47% (74) 0.044 58% (63) 63% (122) 45% (114) 41% (107) 50% (69) 53% (82) 42% (45) 37% (73) 55% (140) 18% (46) 17% (24) 33% (51) 82% (215) 83% (115) 67% (105) 78% (205) 89% (124) 92% (144) <0.001 <0.001 79% (85) 82% (159) 90% (229) 22% (57) 11% (15) 8% (12) 21% (23) 18% (36) 10% (25) 33% (87) 53% (73) 58% (90) <0.001 87% (74) 89% (141) 95% (218) 13% (11) 11% (18) 5% (11) 24% (26) 42% (81) 56% (143) 67% (175) 47% (66) 42% (66) 76% (82) 58% (114) 44% (111) 54% (140) 70% (96) 67% (103) 46% (117) 30% (42) 33% (51) 0.004 <0.001 0.005 0.02 <0.001 25% (66) 35% (48) 30% (47) 22% (24) 27% (52) 33% (85) 28% (116) 32% (36) 25% (9) Nurse 32% (83) 22% (30) 22% (35) 31% (33) 35% (69) 18% (46) 28% (114) 24% (27) 19% (7) Type HW Residents in Training Freelance doctors 15% (38) 19% (26) 10% (15) 0.027 12% (13) 11% (22) 17% (44) 0.004 15% (60) 14% (16) 8% (3) 0.046 9% (24) 6% (9) 10% (15) 8% (9) 8% (16) 9% (23) 7% (28) 10% (11) 25% (9) Other 19% (51) 19% (26) 28% (44) 27% (29) 18% (36) 22% (56) 22% (90) 20% (23) 22% (8) Int. J. Environ. Res. Public Health 2023, 20, 6087 9 of 22 Table 3. Cont. N = 688 Cut-off Subscale, N Participants Burnout Level OVER CUT OFF—EE OVER CUT OFF—D OVER CUT OFF—PA ≤17 N = 393 Low 18–29 N = 139 Middle ≥30 N = 156 High p ≤05 N = 238 Low 06–11 N = 196 Middle ≥12 N = 254 High p ≤34 N = 538 High 35–39 N = 114 Middle ≥40 N = 36 Low p Clinic 37% (95) 47% (65) 51% (78) 34% (35) 40% (76) 50% (127) 40% (160) 58% (64) 39% (14) Surgical 11% (28) 20% (27) 12% (19) 9% (9) 15% (29) 14% (36) 14% (57) 9% (10) 19% (7) Medical area Services 27% (69) 17% (23) 22% (34) 0.001 31% (32) 27% (51) 17% (43) 0.005 25% (102) 15% (17) 19% (7) 0.033 Never been followed by psychiatrist Past psychopharma- cologic treatments Noncurrent psy- chopharmacological treatment Feeling protected at work Enough PPE available Sufﬁcient work shifts Emergency/ Urgency Yes No Yes No Yes No Yes No 26% (66) 16% (22) 14% (22) 27% (28) 19% (36) 18% (46) 20% (82) 18% (20) 22% (8) 11% (28) 7% (10) 17% (27) 89% (232) 93% (129) 83% (128) 12% (30) 12% (17) 23% (35) 88% (228) 88% (122) 77% (119) 4% (10) 9% (12) 12% (19) 96% (250) 91% (125) 88% (136) 0.02 0.005 0.005 36% (94) 17% (23) 8% (12) 37% (40) 27% (53) 14% (36) 11% (28) 20% (27) 38% (60) <0.001 7% (8) 12% (24) 33% (83) <0.001 Not always 53% (139) 64% (87) 54% (84) 55% (59) 60% (117) 53% (134) Yes No 44% (114) 37% (50) 24% (37) 43% (46) 40% (78) 31% (77) 11% (28) 12% (17) 31% (48) <0.001 9% (10) 11% (22) 24% (61) <0.001 Not always 45% (118) 51% (69) 46% (71) 47% (50) 48% (94) 45% (114) Yes No 44% (114) 37% (50) 24% (37) 75% (80) 64% (124) 41% (102) 11% (28) 12% (17) 31% (48) <0.001 2% (2) 12% (24) 34% (86) <0.001 Not always 45% (118) 51% (69) 46% (71) 23% (24) 23% (45) 25% (63) Stable 47% (122) 32% (44) 27% (42) 51% (55) 42% (82) 28% (71) Working hours Decreased 5% (12) 6% (8) 3% (4) <0.001 3% (3) 5% (10) 4% (11) <0.001 Increased 49% (127) 62% (86) 70% (109) 46% (49) 53% (102) 68% (171) Vaccine for COVID-19 Yes No 98% (398) 98% (110) 89% (32) 2% (10) 2% (2) 11% (4) 0.009 Abbreviations: HW = Health Worker; MBI = Maslach Burnout Inventory EE = Emotional Exhaustion; D = Depersonal- ization; PA = Personal Accomplishment; PPE = Personal Protective Equipment p = p Value. Table 4. Focus Burnout measured with Maslach Burnout Inventory (MBI): categorical data analysis based on biographical data and text results (n, % column) during the second phase. Only signiﬁcant differences (p < 0.05) are shown in the table. N = 688 Cut-off Subscale, N Participants Burnout Level MBI—EE MBI—D MBI—PA ≤17 N = 393 Low 18–29 N = 139 Middle ≥30 N = 156 High p ≤05 N = 238 Low 06–11 N = 196 Middle ≥12 N = 254 High p ≤34 N = 538 High 35–39 N = 114 Middle ≥40 N = 36 Low p Minimum (0–21) 95% (215) 86% (115) 64% (96) 95% (69) 89% (170) 77% (187) BAI Medium (22–35) 4% (9) 1% (2) 11% (15) 22% (33) <0.001 2% (3) 13% (20) 5% (4) 0% (0) 8% (16) 15% (37) <0.001 3% (5) 8% (20) High (>36) Minimal (0–13) Low (14–19) Moderate (20–28) High (29–63) BDI IES Subclinical (0–8) 85% (336) 59% (82) 47% (73) 93% (222) 68% (133) 54% (136) Mild (9–25) 13% (50) 37% (52) 40% (63) Moderate (26–43) Severe (>44) 2% (6) 0% (1) 3% (4) 1% (1) 10% (16) 3% (4) <0.001 6% (15) 27% (52) 39% (98) 0% (1) 0% (0) 4% (8) 2% (3) 7% (17) 1% (3) <0.001 No problem (0–14) 55% (217) 17% (24) 13% (20) 76% (182) 20% (39) 16% (40) 42% (227) 21% (24) 28% (10) GHQ Some problems (15–19) 36% (141) 40% (55) 19% (29) <0.001 22% (52) 46% (90) 33% (83) <0.001 29% (158) 43% (49) 50% (18) <0.001 Several problems (20–36) 9% (35) 43% (60) 69% (107) 2% (4) 34% (67) 52% (131) 28% (153) 36% (41) 22% (8) Abbreviations: HW = Health Worker; MBI = Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonal- ization; PA = Personal Accomplishment; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES = Impact of Event Scale; GHQ = General Health Questionnaire; p = p Value. Int. J. Environ. Res. Public Health 2023, 20, 6087 10 of 22 3.2.2. Anxiety, Depression and Post-Traumatic Stress Symptoms Women, compared to men, scored higher on most of the analyzed scales: they more fre- quently showed anxious, depressive, and post-traumatic stress symptoms (p < 0.001) (Table 5). The univariate analysis showed that the male sex was protective for anxiety levels as mea- sured by the BAI (BAI: OR males vs. females = 0.475, 95% CI = 0.266–0.849, p = 0.01) (Table 6). Having non-COVID-related health problems was a risk factor for high anxiety scores (BAI: OR non-COVID health problems vs. No = 28,677, 95% CI = 17,261–47,644, p ≤ 0.001) while it seemed to play a protective role against high depressive scores (BDI: OR non- COVID-19 health problems vs. No = 0.428, CI95% = 0.236–0.775, p = 0.01). A change in family habits was a risk factor for post-traumatic stress symptoms as measured by the IES (IES: OR no change in habits vs. change in habits = 0.411, 95% CI = 0.234–0.720, p ≤ 0.001). In addition, by also looking at the categorical data, it can be seen how 91% of those who scored high values (≥20) in the GHQ had changed their habits (p = 0.003) and that they were also those who had obtained the highest values (≥43) in the IES (p = 0.016). HWs living alone (67%) had higher IES scores (≥43) than those living with other people (p = 0.04 and p = 0.039). HWs with the highest BAI scores (≥36) were in most cases (84%) working in the context of emergency/urgency (p = 0.032). HWs with high BDI values (≥20) were mostly working in the clinical (43%) and emergency urgency (22%) areas (p = 0.025). 3.2.3. Mental Health in HWs Women, compared to men, reported overall worse mental health (Table 5). As with what was found for burnout, GHQ scores were higher (≥20) in 67% of those who increased working hours during the pandemic (Table 6). Having non-COVID-19-related health problems was a risk factor for worse overall mental health (GHQ: OR non-COVID-19 health problems vs. No = 17,678, 95% CI = 11,946–26,159, p ≤ 0.001). A change in family habits was a risk factor for lower global health (GHQ: OR no change in habits vs. change in habits = 0.476, 95% CI = 0.310–0.732, p ≤ 0.001. 3.2.4. Categorical Data Analysis based on Biographical data (n, % Column) during the Second Phase In the group of HWs with high EE scores, 35% showed moderate-to-severe anxiety symptoms (mean EE = 13%; Low EE = 5%), 53% had post-traumatic stress symptoms (mean EE = 41%; Low EE = 15%), and 88% had moderate-to-severe mental health problems as suggested by GHQ scores (mean EE = 83%; Low EE = 45%). Among the HWs who had elevated D levels, 23% had moderate-to-severe anxiety symptoms (mean D = 11%; D low = 5%), 47% had post-traumatic stress symptoms that were all the way up to severe (D medium = 33%; D low = 6%), and 85% had a mental state that was characterized by moderate-to-severe problems (D mean = 80%; D mild = 24%). Among the HWs who had low levels of PA, this factor appeared to be indicative of high burnout. Speciﬁcally, 72% had general mental health with moderate-to-severe problems (mean PA = 79%; High PA = 57%), while the other associations were not signiﬁcant (Table 7). Int. J. Environ. Res. Public Health 2023, 20, 6087 11 of 22 Table 5. Focus on Anxiety, Depression, Stress and Mental Health: categorical data analysis based on biographical data (n, % column) during the second phase. Only signiﬁcant differences (p < 0.05) are shown in the table. N = 688 BAI BDI IES GHQ Cut-off Subscale N Participants Minimum ≤21 N = 426 Medium 22–35 N = 57 High ≥36 N = 25 p Minimal 0–13 N = 186 Medium 14–19 N = 20 High ≥ 20 N = 482 p Subclinical 0–8 N = 491 Mild 9–25 N = 165 Moderate 26–43 N = 26 Severe ≥43 N = 6 p No Problem 0–14 N = 261 Some Problems 15–19 N = 225 Several Problems 20–36 N = 202 p 34% (144) 66% (281) 19% (11) 81% (46) 20% (5) 80% (20) 0.037 54% (231) 46% (194) 17% (74) 83% (351) 63% (36) 37% (21) 37% (21) 63% (36) 32% (8) 68% (17) 40% (10) 60% (15) 0.033 <0.001 37% (20) 63% (34) 45% (9) 55% (11) 19% (92) 81% (390) <0.001 Gender Living situation (II) Marital status (I) Have children Non-COVID 19 related health issues Changed habits due to the pandemic Male Female Lives alone Cohabits Unmarried Married Divorced Cohabiting Widower In relation Yes No Yes No Yes No 24% (86) 76% (274) 19% (68) 51% (183) 5% (17) 15%(53) 0% (1) 21% (35) 79% (130) 15% (25) 51% (84) 5% (9) 22 (36) 1% (1) 11%(38) 6% (10) 35% (9) 65% (17) 23% (6) 46% (12) 8% (2) 19% (5) 4% (1) 0% (0) 67% (4) 33% (2) 67% (4) 33% (2) 0% (0) 0% (0) 0% (0) 0% (0) 0.04 0.039 81% (293) 19% (67) 91% (150) 9% (15) 92% (24) 8% (2) 100% (6) 0% (0) 0.016 43% (56) 57% (74) 21% (27) 79% (102) 78% (101) 22% (29) 61% (137) 52% (106) 39% (88) 14% (32) 48% (96) 31% (62) 86% (193) 69% (140) 84% (188) 91% (184) 16% (37) 9% (18) 0.005 <0.001 0.003 Int. J. Environ. Res. Public Health 2023, 20, 6087 12 of 22 Table 5. Cont. N = 688 BAI BDI IES GHQ Cut-off Subscale N Participants Minimum ≤21 N = 426 Medium 22–35 N = 57 High ≥36 N = 25 p Minimal 0–13 N = 186 Medium 14–19 N = 20 High ≥ 20 N = 482 p Subclinical 0–8 N = 491 Mild 9–25 N = 165 Moderate 26–43 N = 26 Severe ≥43 N = 6 p No Problem 0–14 N = 261 Some Problems 15–19 N = 225 Several Problems 20–36 N = 202 Change in job HW Job Categories Emergency professions Medical area Yes No Doctor Nurse Resident in training Freelance doctor Other No Yes Clinic Surgical Services Emergency/ xcvbUr- gency 32% (137) 27% (114) 15% (64) 9% (37) 17% (73) 41% (175) 59% (248) 19% (11) 30% (17) 11% (6) 9% (5) 32% (18) 35% (20) 65% (37) 16% (4) 28% (7) 8% (2) 4% (1) 44% (11) 16% (4) 84% (21) 0.013 0.032 18% (10) 20% (11) 15% (8) 9% (5) 38% (21) 20% (11) 80% (43) 44% (23) 10% (5) 38% (20) 8% (4) 25% (5) 15% (3) 5% (1) 10% (2) 45% (9) 20% (4) 80% (16) 58% (11) 5% (1) 26% (5) 11% (2) 30% (146) 28% (134) 15% (70) 9% (41) 19% (91) 40% (194) 60% (286) 43% (204) 14% (68) 21% (101) 22% (104) 0.011 0.004 0.025 Past psychophar- macological treatment Yes No 12% (49) 88% (374) 32% (18) 68% (39) 25% (6) 75% (18) <0.001 11% (40) 89% (316) 20% (33) 80% (130) 27% (7) 73% (19) 33% (2) 67% (4) 0.007 38% (49) 62% (81) 19% (25) 25% (32) 18% (24) 9% (12) 28% (37) 41% (52) 13% (16) 29% (37) 17% (21) 40% (90) 60% (135) 29% (65) 30% (67) 14% (32) 7% (16) 20% (45) 39% (87) 12% (28) 25% (57) 23% (52) 55% (111) 45% (91) 35% (71) 24% (49) 11% (23) 10% (20) 19% (39) 50% (99) 15% (30) 16% (32) 19% (37) p 0.001 0.045 0.049 Int. J. Environ. Res. Public Health 2023, 20, 6087 13 of 22 Table 5. Cont. N = 688 BAI BDI IES GHQ Cut-off Subscale N Participants Minimum ≤21 N = 426 Medium 22–35 N = 57 High ≥36 N = 25 p Minimal 0–13 N = 186 Medium 14–19 N = 20 High ≥ 20 N = 482 p Subclinical 0–8 N = 491 Mild 9–25 N = 165 Moderate 26–43 N = 26 Severe ≥43 N = 6 p No Problem 0–14 N = 261 Some Problems 15–19 N = 225 Several Problems 20–36 N = 202 p 11% (6) 89% (49) 13% (7) 87% (48) 20% (4) 80% (16) 25% (5) 75% (15) 5% (24) 95% (455) 6% (29) 94% (448) 0.007 0.002 Followed by psychiatrist in the present Current psychopharmaco- logical treatment Feeling protected at work Sufﬁcient availability of PPE Sufﬁcient work shifts Yes No Yes No Yes No Not always Yes No Not always Yes No 5% (23) 95% (397) 25% (105) 17% (74) 58% (245) 39% (164) 13% (57) 48% (202) 58% (246) 16% (9) 84% (48) 12% (7) 32% (18) 55% (31) 27% (15) 32% (18) 41% (23) 44% (24) 19% (82) 29% (16) Not always 23% (96) 27% (15) Stable Working hours Decreased Increased 0.01 <0.001 <0.001 0.033 12% (3) 88% (22) 8% (2) 44% (11) 48% (12) 24% (6) 36% (9) 40% (10) 32% (8) 36% (9) 32% (8) 28% (102) 16% (57) 56% (199) 41% (147) 13% (47) 14% (23) 30% (49) 56% (92) 27% (45) 23% (37) 46% (162) 50% (82) 61% (215) 17% (61) 22% (79) 47% (76) 27% (44) 26% (43) 12% (3) 31% (8) 58% (15) 23% (6) 31% (8) 46% (12) 42% (11) 27% (7) 31% (8) 17% (1) 17% (1) 67% (4) 50% (3) 17% (1) 33% (2) 67% (4) 0% (0) 33% (2) <0.001 0.008 0.036 32% (41) 15% (19) 53% (69) 40% (51) 14% (18) 46% (59) 60% (75) 15% (19) 25% (32) 46% (59) 2% (3) 52% (67) 28% (62) 17% (37) 13% (26) 29% (59) 56% (125) 58% (116) 43% (97) 13% (30) 43% (96) 69% (154) 13% (30) 18% (40) 42% (93) 5% (11) 26% (53) 22% (45) 51% (103) 38% (77) 32% (63) 30% (60) 28% (56) 5% (10) 54% (120) 67% (135) <0.001 0.003 <0.001 0.006 Abbreviations: Cat. HW = Health Worker Category; BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES = Impact of Event Scale; GHQ = General Health Questionnaire; p = p Value. Int. J. Environ. Res. Public Health 2023, 20, 6087 14 of 22 Table 6. Univariable model, categorical data collected during the second phase. OR CI (95%) x-Square Df p GHQ BDI IES BAI OVER CUTOFF-D Non-COVID 19 related health issues Changed habits due to the pandemic Non-COVID 19 related health issues Non-COVID 19 related health issues Changed habits due to the pandemic Age Have children Changed habits due to the pandemic Yes vs. No 17.678 11.946–26.159 8.12 No vs. Yes 0.476 0.310–0.732 11.45 Yes vs. No 0.428 0.236–0.775 Gender Male vs. Female 0.475 0.266–0.849 Yes vs. No 28.677 17.261–47.644 16.64 No vs. Yes 0.411 0.234–0.720 <33 vs. >54 No vs. Yes 0.520 17.061 0.392–0.688 12.428–23.421 No vs. Yes 0.505 0.330–0.773 Changed habits due to the fear of infecting loved ones No vs. Yes 0.447 0.248–0.804 OVER CUT OFF-PA Civil status OVER CUT OFF-EE Age Have children Non-COVID 19 related health issues Changed habits due to the pandemic Cohabiting vs. Lives alone <33 vs. >54 No vs. Yes 15.586 10.311–2.356 0.680 14.793 0.515–0.899 10.826–20.213 Yes vs. No 18.969 12.945–27.797 10.78 No vs. Yes 0.366 0.226–0. 592 16. 73 7.84 6.3 9.65 20.82 10.92 9.89 7.23 4.43 7.36 6.04 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 <0.001 <0.001 0.01 0.01 <0.001 <0.001 <0.001 <0.001 <0.001 0.01 0.04 0.01 0.01 <0.001 <0. 001 Abbreviation: MBI = Maslach Burnout Inventory; EE = Emotional Exhaustion; D = Depersonalization; PA = Personal Accomplishment; GHQ = General Health Questionnaire; BDI = Beck Depression Inventory; BAI = Beck Anxiety Inventory; IES = Impact of Event Scale; HW = Health Workers; OR = Odds Ratio; CI = Conﬁdence Interval; p = p Value. Anxiety symptoms, post-traumatic stress symptoms, and overall mental health can also inﬂuence each other, independently of burnout. Speciﬁcally, those with a high score on the BAI scale (≥36) had, in 68% of cases, symptoms of post-traumatic stress (IES- ≥ 9) (medium BAI 22–35 = 62%; low BAI ≤ 21 = 34%). In 92%, there were moderately severe mental health problems (GHQ ≥ 15) (medium BAI 22–35 = 88%; low BAI ≤ 21 = 83%). In the group of HWs who had moderate–severe BDI scores (≥20), 14% had moderate–high anxiety symptoms (BAI ≥ 22) (medium BDI 14–19 =50%; low BDI 0–13 = 50%), 39% had post-traumatic stress symptoms (IES ≥ 9) (medium BDI 14–19 = 50%; low BDI 0–13 = 0%), and 85% reported moderately severe mental health problems (GHQ ≥ 15) (medium BDI 14–19 = 95%; low BDI 0–13 = 0%). HWs scoring high on the IES scale, assessing post-traumatic stress symptoms (with score ≥43), had a moderate–high BAI scores in 67% of cases (≥22) (moderate IES 26–43 = 53%; mild IES 9–25 = 20%; subclinical IES 0–8 =10%) and moderately severe mental health problems in 100% of HWs (GHQ ≥ 15) (moderate IES 26–43 =100%; mild IES 9–25 = 92%; subclinical IES 0–8 = 49%). Last, in the group of HWs with high GHQ values (≥20), 29% had moderate–severe anxiety symptoms (BAI ≥ 22) (moderate GHQ 15–19 = 6%; low GHQ 0–14 = 11%), 100% reported moderate–severe depressive symptoms (BDI ≥ 20) (moderate GHQ 15–19 = 100%; low GHQ 0–14 = 28%), and 56% showed high scores for post-traumatic stress symptoms (IES ≥ 9) (GHQ moderate 15–19 = 31%; GHQ low 0–14 = 0%). Those with higher levels of depression also tended to have more anxiety and post-traumatic stress symptoms, and vice versa (Table 7). BAI BDI IES GHQ Minimum (0–21) Medium (22–35) High (>36) Minimal (0–13) Low (14–19) Moderate (20–28) High (29–63) Subclinical (0–8) Mild (9–25) Moderate (26–43) Severe (>44) No problem (0–14) Some problems (15–19) Several problems (20–36) Int. J. Environ. Res. Public Health 2023, 20, 6087 15 of 22 Table 7. Focus on Anxiety, Depression, Stress, and Mental Health: categorical data analysis based on text scores (n, % column) during the second phase. Only signiﬁcant differences (p < 0.05) are shown in the table: p < 0.001 was observed for all comparisons. N = 688 BAI BDI IES Cut-off Subscale, N Participants Minimum ≤21 N = 426 Medium 22–35 N = 57 High ≥36 N = 25 Minimal 0–13 N = 186 Medium 14–19 N = 20 High ≥20 N = 482 Subclinical 0–8 N = 491 Mild 9–25 N = 165 Moderate 26–43 N = 26 Severe ≥43 N = 6 50% (3) 50% (10) 33% (2) 25% (5) 86% (413) 10% (50) 90% (281) 7% (22) 17% (1) 25% (5) 4% (19) 3% (8) 79% (131) 13% (22) 7% (12) 46% (12) 38% (10) 15% (4) 33% (2) 50% (3) 17% (1) 66% (281) 39% (22) 31% (131) 39% (22) 3% (12) 18% (10) 32% (8) 48% (12) 16% (4) 100% (186) 50% (10) 0% (0) 30% (6) 61% (295) 33% (159) 0% (0) 15% (3) 5% (23) No Problem 0–14 N = 261 GHQ Some Problems 15–19 N = 225 Several Problems 20–36 N = 202 89% (72) 93% (210) 71% (144) 9% (7) 4% (10) 20% (40) 2% (2) 2% (5) 9% (18) 71% (186) 0% (0) 0% (0) 0% (0) 0% (0) 0% (0) 0% (1) 2% (4) 7% (15) 28% (74) 98% (221) 93% (187) 95% (248) 68% (154) 44% (89) 5% (13) 28% (64) 44% (88) 0% (0) 2% (5) 10% (21) 0% (2) 5% (3) 4% (1) 0% (0) 5% (1) 1% (5) 0% (0) 1% (2) 2% (4) 17% (72) 12% (7) 8% (2) 100% (186) 5% (1) 49% (210) 18% (10) 34% (144) 70% (40) 20% (5) 72% (18) 0% (0) 20% (4) 0% (0) 75% (15) 15% (74) 46% (221) 39% (187) 51% (248) 31% (154) 18% (89) 8% (13) 39% (64) 53% (88) 0% (0) 0% (0) 19% (5) 81% (21) 33% (2) 67% (4) Abbreviations: BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; IES = Impact of Event Scale; GHQ = General Health Questionnaire; p = p Value. 4. Discussion 4.1. Discussion of the Results The results of the second administration of the survey showed a substantial overlap with those of the ﬁrst administration. Indeed, the study results describe high levels of burnout and the presence of many general mental health problems in HWs, as represented by the MBI and GHQ scales in the face of an apparently mild level of stress, anxiety, and depressive symptoms, data that were still present in the HWs sample during the ﬁrst administration. Though the current study, as with similar ones, does not allow one to assess the actual change of HW mental health conditions from the pre-pandemic period, as baseline measurements are not available, it can still be hypothesized, based on the data from the literature, that the pandemic had an impact on burnout levels. A systematic review performed in 2015 [34] reported the presence of burnout, which was measured with the MBI, in 30% of HWs working in emergency–urgency settings. An Italian report of 2008 [35] analyzing the level of burnout among general practitioners found medium/high EE and D in 32% and 53% of participants, respectively, and low/medium PA in 32% of cases. As these scores are lower than those observed in the current study, it can be suggested that the pandemic may have generated adverse psychological outcomes in HWs. A further limitation of the widely varied existing literature about the topic is the difﬁculty of comparing results from different studies, as far as both the method used for symptom assessment and the selected populations are concerned. Int. J. Environ. Res. Public Health 2023, 20, 6087 16 of 22 4.1.1. Burnout Burnout has been an important topic of research over the years, especially with re- spect to HWs [36–38]. During the current pandemic, this condition has been addressed by several studies in the literature, with different populations and approaches; most of them, as with this study, used the MBI scale [39–42] to assess burnout, while others used different scales [15,19,43–54]. Regarding the populations targeted by available studies, in most cases, these have included ward physicians—especially in intensive or emergency care settings—nurses, and general practitioners. Additionally, a small number of studies have focused on residents in training. For example, high burnout rates have been found particularly in frontline HWs [55], in those working in intensive and sub-intensive care units [46,56–58]. This may lead to evidence that the department in which a HW works may be associated with a higher risk of burnout. Focusing on the studies that used MBI [39–42], it is easier to make comparisons to the current data. Moreover, among the HWs under study, against levels of EE that are comparable to or lower than those known in the literature, there were higher levels of burnout, which were expressed as D increase and PA reduction. It is widely acknowledged [58] that higher levels of burnout can be associated with both individual work-related and non-work factors. In the current study, it was observed that individual factors related to work or extra work inﬂuenced the following (as was the case in [59]): in the ﬁrst phase of the study [20], higher levels of burnout were observed in the female population, in participants under the age of 30, in those who changed their extra-work habits, in those who did not have children, and, above all, in those who have had to change their job or those who were postgraduates. Contrary to most of the data in the literature [42–48,60,61], and to the data from our ﬁrst analysis [20,28], at the follow-up, 1 year after assessment following the ﬁrst pandemic wave, we noted that female sex and younger age were no longer determining factors for high levels of burnout when compared to other variables, which could have a greater weight since the emergency phase has continued over these two long years. According to some of the studies analyzed [45,60], high levels of burnout correlated with an increase in the number of working hours, while in other studies [62] higher levels of burnout were even found in HW males with more than 15 years of work experience. The study of [58] also did not detect correlations with low values of the MBI-PA scale, which is similar to the results that were collected by our research group in this phase of the study. Interestingly, some evidence from the literature [49,51,54] has shown that HWs, speciﬁ- cally those who normally operate in the emergency–urgency ﬁeld and in critical conditions, were less vulnerable to the development of burnout in the current pandemic. This result would seem to contrast with the idea that COVID-19 exposes front-line staff to high risk and to requests for increasing work commitment with consequentially greater emotional impact; however, it was highlighted how the ability to “have the situation under control” could protect the HWs from the development of greater stress in the workplace [29]. This is in addition to the fact that these HWs would perceive greater personal fulﬁlment by being able to apply their knowledge, making themselves effectively indispensable in terms of facing the pandemic emergency with subsequent recognition by the community [63,64]. On the other hand, the HWs who remained in their wards or who, in any case, were forced to change their duties, were instead predisposed to greater stress: those who remained conﬁned to the clinical activity that was carried out previously had less chance of treating their patients given the reallocation of resources aimed at emergency support. Meanwhile, as is evidenced by the data presented here, those who changed their job found themselves carrying out non-habitual tasks, thus committing errors more frequently that often put a greater strain on the HWs, as such, reducing personal satisfaction and increasing stress. 4.1.2. Anxiety, Depression, and Stress In addition to burnout, we assessed also symptoms of anxiety, depression, post- traumatic stress, and overall mental health in the general population. The existing literature Int. J. Environ. Res. Public Health 2023, 20, 6087 17 of 22 showed a high variability as far as assessment tools are concerned, thus making it difﬁcult to generalize and compare results. Rates of reported depressive symptoms range from 30.2% to 57.6%; anxiety has been described in up to 46.6% of HWs [43]. Both anxiety and depression have been associated with female sex, a university hospital setting, and ethical issues [49]. A study carried out in Italy [65], in addition to the high levels of burnout (40.7% EE, 30.2% D, and 35.4% PA), found increased anxiety (through the questionnaire The State-Trait Anxiety Inventory—Form Y) and post-traumatic stress symptoms (IES) in women, and in nurses; however, burnout did not show differences between doctors or nurses. Interestingly the Portuguese study of [46] on burnout found normal levels of anxiety (66.9%), depression (70.6%), and stress (63.4%) in their HWs. In the American study of [18], depressive symptoms were found in 27.2% of HWs, anxiety in 18.6%, and post-traumatic stress in 24.7% of cases. Additionally, in the Spanish study of [50], severe anxiety disorder was found in 20.7% of cases, severe depression in 5.3%, and in 83%, the HWs obtained moderate/high scores according to the IES (here, 36%). The current study Is in line with other ones available in the literature [2]), showing with low anxiety symptoms (15% moderate–high) and moderate post-traumatic symptoms (40% moderate–high), but where high depressive symptoms (70% moderate–high) were highlighted in the total population. Additionally, in this case, the possible work and extra work factors were investigated to understand the possible predisposing factors for the adverse psychological outcomes, such as anxiety, depression, or post-traumatic symptoms. Populations that were identiﬁed as more fragile and at risk of burnout were also more likely to show more severe anxiety, depression, post-traumatic stress, and worse mental health symptoms: female HWs who changed their habits and who also had non-COVID- related health problems. Furthermore, in the literature, anxious–depressive symptoms have been found in similar populations at risk. Female sex [18,50,52,66], younger age [50,52,66], and being unmarried [18] were factors associated with anxiety and depressive symptoms, and female sex was associated with post-traumatic stress symptoms as measured with the IES [67], as well. Changing job was associated with higher levels of depression in women [52]. In general, therefore, we can conclude that, despite showing differences from our previous analysis, the current results are still consistent with ﬁndings in the literature. The reasons why the populations that are most at risk are in emergency environments could be many. First, regarding women, anxious–depressive symptoms are normally more frequent than in men. The change in job duties exposes HWs to factors they are not used to coping with, increasing feelings of devaluation and incapacity, which is probably at the base of the depressive and post-traumatic stress symptoms [68,69]. As far as the extra-work environment is concerned, worse mental health was found in those who had to change their daily home habits, likely due to the fear of infecting their loved ones or to the need of managing the family situation in the absence of school support or caregivers; this was also the case in those who have had a positive family member. HWs fearing to be the cause of infection of family members tend to isolate themselves, reducing contact with family and friends. These factors increase the sense of loneliness, anxiety, depression, and post-traumatic stress [70–72]. Surprisingly, positivity to the infection or the presence of COVID-19 symptoms did not signiﬁcantly inﬂuence burnout or psychological symptoms in any of the analysis performed; furthermore, in the face of a worse mental state, having positive family members did not affect the level of burnout. The literature is still lacking data about mental health in HWs who were infected with COVID-19. Surely, in this study sample, this population was the minority when compared to the total. Therefore, the data may not be signiﬁcant for this reason (only 26% for those with COVID-19 and 27% for those with symptoms). However, we must consider both the factors that could worsen mental health or not in these HWs. Although the distance from the healthcare environment can trigger feelings of guilt toward colleagues, the fear of having infected family members or collaborators, and the Int. J. Environ. Res. Public Health 2023, 20, 6087 18 of 22 consequences of the disease, were further worsened by isolation. This was probably due to distance from the difﬁcult and stressful health situation itself having maybe balanced these concerns. As for the role that positive family members may have had on mental health, it can be thought that these outcomes inﬂuence more of the concerns and the depressive sphere than the attitude and work attitudes that were demonstrated by the remaining results. 4.1.3. Mental Health in HWs It emerges, from the analysis of comparison and association between the burnout and other psychological outcomes, that there exists an interaction and possible inﬂuence of each factor on the others. How they affect each other is certainly known in the literature, but few studies have studied these ongoing inﬂuences of the current pandemic. It can therefore be said that high levels of burnout are associated with greater anxious and post-traumatic stress symptoms, as well as with an overall impairment of mental health. This relationship can be understood in two ways: burnouts appear to determine psychological suffering by promoting the onset of such symptoms, and psychological fragility could make HWs more vulnerable to the development of burnout. 4.2. Limitations and Strengths: Possible Future Developments The present study was carried out on a large, varied population, giving a broad picture of the local reality of one of the hub hospitals in Piedmont, Italy—a region that was greatly affected by COVID-19. Some limitations should be underscored. First, there is no possibility of comparison for these parameters before the event, and thus, no possibility for giving an idea of the real increase due to the pandemic situation, as well as subsequent change that occurred during the pandemic. Second, data were collected from a single center, thus limiting the possibility of generalizing results. The cross-sectional design of the study did not allow for one to derive the causal relationships that exist between the variables under study. Moreover, an assessment based exclusively on self-administered questionnaires entails possible biases and does not allow one to make clinical diagnoses of any disorder. Finally, although the current study was carried out one year after the ﬁrst, a post- traumatic stress assessment tool that was validated speciﬁcally for COVID-19, i.e., the COVID-19 Peritraumatic Distress Index (CPDI), has not yet been adopted in this survey. However, it should be acknowledged that most of the limitations described above are shared by similar studies in the literature, as they are strictly linked to the type of study performed. On the other hand, some strengths should be underscored, as well. The current study employed validated tools for mental health assessment, including burnout, anxiety, depression, post-traumatic stress symptoms, and mental health in general. In addition, the sample consists of both frontline and non-frontline health personnel who were recruited both in the hospital (the hub of the Piedmont Region, an area that was greatly affected by COVID-19) and in extra-hospital contexts. This allowed an in-depth understanding of the impact of the pandemic on health workers at different levels. Information was collected on different socio-demographic variables, such as those related to work habits and the pandemic. 5. Conclusions It is undeniable that increased levels of burnout and adverse psychological outcomes have been observed during this long pandemic. While in the ﬁrst phase of the study, some gender and age-related differences were found as far as the psychological and mental health impact of the pandemic is concerned, in the second and current phase of the study, these were less marked, thus suggesting more widespread distress and suffering. The GHQ scores, indicative for general mental health problems, seem to support this hypothesis as well as those of the other protocol tests Int. J. Environ. Res. Public Health 2023, 20, 6087 19 of 22 (BAI, BDI, IES) highlighting anxiety, depression, and post-traumatic stress symptoms in the HWs population. Surely, these problems cannot and must not be underestimated: the institutions must not forget that HWs’ psychological well-being should be prioritized in order to avoid the reduced work performance that would come with a greater expenditure of short- and long-term resources. Therefore, the development of HWs support techniques should be strengthened, with particular attention being directed to the most fragile and at-risk populations. One of the most immediate strategies could be a greater access to psychological support services (such as the telephone counseling service offered to the employees of the Maggiore della Carità University Hospital) that not only give a chance to listen and discuss, but also teach self-care strategies in order to better manage difﬁcult situations in the workplace and beyond. Author Contributions: Conceptualization, C.G. and P.Z.; methodology, E.G.; software, E.G.; valida- tion, E.G., C.G. and P.Z.; formal analysis, C.G.; investigation, D.M. resources, M.P., M.R.; data curation, D.M.; writing—original draft preparation, E.G.; writing—review and editing, P.Z.; visualization, M.P.; supervision, C.G.; project administration, C.G. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Intercompany Ethics Committee of Novara (Protocol 82/20) for studies involving humans. This research follows an amendment that was requested in March 2021 to continue the research project of which the results were published in two previous publications. Informed Consent Statement: Informed consent was obtained from all participants involved in the study. Written informed consent has been obtained from the patient(s) to publish this paper. Data Availability Statement: No new data were created or analyzed in this study. Data sharing is not applicable to this article. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. WHO. Coronavirus (COVID-19) Dashboard. Available online: https://covid19.who.int (accessed on 19 November 2021). 2. Hao, Q.; Wang, D.; Xie, M.; Tang, Y.; Dou, Y.; Zhu, L.; Wu, Y.; Dai, M.; Wu, H.; Wang, Q. Prevalence and Risk Factors of Mental Health Problems Among Healthcare Workers During the COVID-19 Pandemic: A Systematic Review and Meta-Analysis. Front. Psychiatry 2021, 12, 567381. [CrossRef] [PubMed] 3. Weinberg, A.; Creed, F. Stress and psychiatric disorder in healthcare professionals and hospital staff. Lancet 2000, 355, 533–537. 4. 5. 6. 7. 8. 9. [CrossRef] [PubMed] Sonali, R. World Failing in ‘Our Duty of Care’ to Protect Mental Health and Well-Being of Health and Care Workers, Finds Report on Impact of COVID-19. Available online: https://www.who.int/news/item/05-10-2022-world-failing-i----our-duty-of-car----to- protect-mental-health-and-wellbeing-of-health-and-care-worker----finds-report-on-impact-of-covid-19 (accessed on 28 May 2023). Cochrane Effective Practice and Organisation of Care Group; Pollock, A.; Campbell, P.; Cheyne, J.; Cowie, J.; Davis, B.; McCallum, J.; McGill, K.; Elders, A.; Hagen, S.; et al. Interventions to support the resilience and mental health of frontline health and social care professionals during and after a disease outbreak, epidemic or pandemic: A mixed methods systematic review. Cochrane Database Syst. Rev. 2020, 11, CD013779. [PubMed] De Hert, S. Burnout in Healthcare Workers: Prevalence, Impact and Preventative Strategies. Local Reg. Anesth. 2020, 13, 171–183. [CrossRef] Kaﬂe, B.; Bagale, Y.; Kaﬂe, S.; Parajuli, A.; Pandey, S. Depression, Anxiety and Stress among Healthcare Workers during COVID-19 Pandemic in a Tertiary Care Centre of Nepal: A Descriptive Cross-sectional Study. JNMA J. Nepal Med. Assoc. 2021, 59, 239–242. [CrossRef] Sung, C.W.; Chen, C.H.; Fan, C.Y.; Chang, J.H.; Hung, C.C.; Fu, C.M.; Wong, L.P.; Huang, E.P.; Lee, T.S. Mental health crisis in healthcare providers in the COVID-19 pandemic: A cross-sectional facility-based survey. BMJ Open 2021, 11, e052184. [CrossRef] Sahin, T.; Aslaner, H.; Eker, O.O.; Gokcek, M.B.; Dogan, M. Effect of COVID-19 pandemic on anxiety and burnout levels in emergency healthcare workers: A questionnaire study. Res. Square 2020, 12, 987. [CrossRef] 10. Kannampallil, T.G.; Goss, C.W.; Evanoff, B.A.; Strickland, J.R.; McAlister, R.P.; Duncan, J. Exposure to COVID-19 patients increases physician trainee stress and burnout. PLoS ONE 2020, 15, e0237301. [CrossRef] Int. J. Environ. Res. Public Health 2023, 20, 6087 20 of 22 11. Ulbrichtova, R.; Svihrova, V.; Tatarkova, M.; Svihra, J., Jr.; Novak, M.; Hudeckova, H. Prevalence of Burnout Syndrome in COVID-19 and Non-COVID-19 Units in University Hospital: A Cross-Sectional Study. Int. J. Environ. Res. Public Health 2022, 19, 12664. [CrossRef] 12. Gramaglia, C.M.; Marangon, D.; Azzolina, D.; Guerriero, C.; Lorenzini, L.; Probo, M.; Rudoni, M.; Gambaro, E.; Zeppegno, P. The Mental Health Impact of 2019-nCOVID on Healthcare Workers From North-Eastern Piedmont, Italy. Focus on Burnout. Front. Public Health 2021, 9, 667379. [CrossRef] 13. Giusti, E.M.; Pedroli, E.; D’Aniello, G.E.; Stramba Badiale, C.; Pietrabissa, G.; Manna, C.; Stramba Badiale, M.; Riva, G.; Castelnuovo, G.; Molinari, E. The Psychological Impact of the COVID-19 Outbreak on Health Professionals: A Cross-Sectional Study. Front. Psychol. 2020, 11, 1684. [CrossRef] 14. Damico, V.; Murano, L.; Demoro, G.; Russello, G.; Cataldi, G.; D’Alessandro, A. Burnout syndrome among Italian nursing staff 15. 16. during the COVID 19 emergency. Multicenter fact-ﬁnding survey. Prof. Inferm. 2020, 73, 250–257. Sanghera, J.; Pattani, N.; Hashmi, Y.; Varley, K.F.; Cheruvu, M.S.; Bradley, A.; Burke, J.R. The impact of SARS-CoV-2 on the mental health of healthcare workers in a hospital setting—A Systematic Review. J. Occup. Health 2020, 62, E12175. [CrossRef] Jalili, M.; Niroomand, M.; Hadavand, F.; Zeinali, K.; Fotouhi, A. Burnout among healthcare professionals during COVID-19 pandemic: A cross-sectional study. Int. Arch. Occup. Environ. Health 2021, 94, 1345–1352. [CrossRef] 17. Barello, S.; Palamenghi, L.; Grafﬁgna, G. Burnout and somatic symptoms among frontline healthcare professionals at the peak of the Italian COVID-19 pandemic. Psychiatry Res. 2020, 290, 113129. [CrossRef] 18. de Pablo, G.S.; Vaquerizo-Serrano, J.; Catalan, A.; Arango, C.; Moreno, C.; Ferre, F.; Shin, J.I.; Sullivan, S.; Brondino, N.; Solmi, M.; et al. Impact of coronavirus syndromes on physical and mental health of health care workers: Systematic review and meta-analysis. J. Affect. Disord. 2020, 275, 48–57. [CrossRef] 19. Pappa, S.; Ntella, V.; Giannakas, T.; Giannakoulis, V.; Papoutsi, E.; Katsaounou, P. Prevalence of depression, anxiety, and insomnia among healthcare workers during the COVID-19 pandemic: A systematic review and meta-analysis. Brain Behav. Immun. 2020, 88, 901–907. [CrossRef] 20. Buchner, D.M.; Cress, M.E.; Esselman, P.C.; Margherita, A.J.; de Lateur, B.J.; Campbell, A.J.; Wagner, E.H. Factors Associated With Changes in Gait Speed in Older Adults. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 1996, 51, M297–M302. [CrossRef] 21. Huang, J.Z.; Han, M.F.; Luo, T.D.; Ren, A.K.; Zhou, X.P. Mental health survey of medical staff in a tertiary infectious disease hospital for COVID-19. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi Zhonghua Laodong Weisheng Zhiyebing Zazhi Chin. J. Ind. Hyg. Occup. Dis. 2020, 38, 192–195. Shalev, A.; Liberzon, I.; Marmar, C. Post-Traumatic Stress Disorder. N. Engl. J. Med. 2017, 376, 2459–2469. [CrossRef] 22. 23. Benfante, A.; Di Tella, M.; Romeo, A.; Castelli, L. Traumatic Stress in Healthcare Workers During COVID-19 Pandemic: A Review of the Immediate Impact. Front. Psychol. 2020, 11, 569935. [CrossRef] [PubMed] 24. Carmassi, C.; Bertelloni, C.A.; Avella, M.T.; Cremone, I.; Massimetti, E.; Corsi, M.; Dell’Osso, L. PTSD and Burnout are Related to Lifetime Mood Spectrum in Emergency Healthcare Operator. Clin. Pract. Epidemiol. Ment. Health CP EMH 2020, 16, 165–173. [CrossRef] [PubMed] 25. Couette, M.; Mouchabac, S.; Bourla, A.; Nuss, P.; Ferreri, F. Social cognition in post-traumatic stress disorder: A systematic review. Br. J. Clin. Psychology 2020, 59, 117–138. [CrossRef] [PubMed] 26. Chen, Q.; Liang, M.; Li, Y.; Guo, J.; Fei, D.; Wang, L.; He, L.I.; Sheng, C.; Cai, Y.; Li, X.; et al. Mental health care for medical staff in China during the COVID-19 outbreak. Lancet Psychiatry 2020, 7, E15–E16. [CrossRef] [PubMed] 27. Maslach, C.; Jackson, S.E. The measurement of experienced burnout. J. Organ. Behav. 1981, 2, 99–113. [CrossRef] 28. Gambaro, E.; Gramaglia, C.; Marangon, D.; Azzolina, D.; Probo, M.; Rudoni, M.; Zeppegno, P. The Mediating Role of Gender, Age, COVID-19 Symptoms and Changing of Mansion on the Mental Health of Healthcare Workers Operating in Italy during the First Wave of the COVID-19 Pandemic. Int. J. Environ. Res. Public Health 2021, 18, 13083. [CrossRef] 29. Maslach, C.; Jackson, S.; Leiter, M. Maslach Burnout Inventory Manual, 3rd ed.; Consulting Psychologists Press: Palo Alto, CA, USA, 1996. 30. Beck, A.T.; Epstein, N.; Brown, G.; Steer, R.A. An inventory for measuring clinical anxiety: Psychometric properties. J. Consult. Clin. Psychol. 1988, 56, 893–897. [CrossRef] 31. Beck, A.T.; Steer, R.A.; Carbin, M.G. Psychometric properties of the Beck Depression Inventory: Twenty-ﬁve years of evaluation. Clin. Psychol. Rev. 1988, 8, 77–100. [CrossRef] 32. Horowitz, M.; Wilner, N.; Alvarez, W. Impact of event scale: A measure of participantive stress. Psychosom. Med. 1979, 41, 209–218. [CrossRef] 33. Goldberg, D. Manual of the General Health Questionnaire; NFER: Windsor, ON, Canada, 1978; 32p. 34. van Mol, M.M.C.; Kompanje, E.J.O.; Benoit, D.D.; Bakker, J.; Nijkamp, M.D. The Prevalence of Compassion Fatigue and Burnout among Healthcare Professionals in Intensive Care Units: A Systematic Review. PLoS ONE 2015, 10, e0136955. [CrossRef] 35. Padula, M.S.; Ilari, G.; Baraldi, S.; Guaraldi, G.P.; Ferretti, E.; Musiani, V.; Svampa, E.; Venuta, M. Il burnout nella Medicina Generale: Personalità del medico e personalità del paziente. Riv. Della Soc. Ital. Med. Gen. 2008, 4, 42–47. 36. Maslach, C.; Jackson, S. Maslach Burnout Inventory; Consulting Psychologists Press, Inc.: Palo Alto, CA, USA, 1981. 37. Tang, R.; Feng, O.; Chong, J.J.; Wang, A. Evaluating the impact of coronavirus disease on burnout among healthcare workers using maslach burnout inventory tool: A systematic review. Proc. Singap. Healthc. 2022, 31, 20101058221117390. [CrossRef] Int. J. Environ. Res. Public Health 2023, 20, 6087 21 of 22 38. Morse, G.; Salyers, M.P.; Rollins, A.L.; Monroe-DeVita, M.; Pfahler, C. Burnout in Mental Health Services: A Review of the Problem and Its Remediation. Adm. Policy Ment. Health Ment. Health Serv. Res. 2012, 39, 341–352. [CrossRef] 39. Di Monte, C.; Monaco, S.; Mariani, R.; Di Trani, M. From Resilience to Burnout: Psychological Features of Italian General Practitioners During COVID-19 Emergency. Front. Psychol. 2020, 11, 2476. [CrossRef] 40. Zhang, Y.; Wang, C.; Pan, W.; Zheng, J.; Gao, J.; Huang, X.; Cai, S.; Zhai, Y.; Latour, J.M.; Zhu, C. Stress, Burnout, and Coping Strategies of Frontline Nurses During the COVID-19 Epidemic in Wuhan and Shanghai, China. Front. Psychiatry 2020, 11, 1154. [CrossRef] 41. Orrù, G.; Marzetti, F.; Conversano, C.; Vagheggini, G.; Miccoli, M.; Ciacchini, R.; Panait, E.; Gemignani, A. Secondary Traumatic Stress and Burnout in Healthcare Workers during COVID-19 Outbreak. Int. J. Environ. Res. Public Health 2021, 18, 337. [CrossRef] 42. Lasalvia, A.; Amaddeo, F.; Porru, S.; Carta, A.; Tardivo, S.; Bovo, C.; Ruggeri, M.; Bonetto, C. Levels of burn-out among healthcare workers during the COVID-19 pandemic and their associated factors: A cross-sectional study in a tertiary hospital of a highly burdened area of north-east Italy. BMJ Open 2021, 11, E045127. [CrossRef] 43. Huo, L.; Zhou, Y.; Li, S.; Ning, Y.; Zeng, L.; Liu, Z.; Qian, W.; Yang, J.; Zhou, X.; Liu, T.; et al. Burnout and Its Relationship with Depressive Symptoms in Medical Staff During the COVID-19 Epidemic in China. Front. Psychol. 2021, 12, 616369. [CrossRef] 44. Liu, X.; Chen, J.; Wang, D.; Li, X.; Wang, E.; Jin, Y.; Ma, Y.; Yu, C.; Luo, C.; Zhang, L.; et al. COVID-19 Outbreak Can Change the Job Burnout in Health Care Professionals. Front. Psychiatry 2020, 11, 1362. [CrossRef] 45. Alsulimani, L.K.; Farhat, A.M.; Borah, R.A.; AlKhalifah, J.A.; Alyaseen, S.M.; Alghamdi, S.M.; Bajnaid, M.J. Health care worker burnout during the COVID-19 pandemic. Saudi Med. J. 2021, 42, 306–314. [CrossRef] 46. Duarte, I.; Teixeira, A.; Castro, L.; Marina, S.; Ribeiro, C.; Jácome, C.; Martins, V.; Ribeiro-Vaz, I.; Pinheiro, H.C.; Silva, A.R.; et al. Burnout among Portuguese healthcare workers during the COVID-19 pandemic. BMC Public Health 2020, 20, 1885. [CrossRef] [PubMed] 47. Khasne, R.W.; Dhakulkar, B.S.; Mahajan, H.C.; Kulkarni, A.P. Burnout among Healthcare Workers during COVID-19 Pandemic in India: Results of a Questionnaire-based Survey. Indian J. Crit. Care Med. Peer Rev. Off. Publ. Indian Soc. Crit. Care Med. 2020, 24, 664–671. 48. Meynaar, I.A.; Ottens, T.; Zegers, M.; van Mol, M.M.C.; van der Horst, I.C.C. Burnout, resilience and work engagement among Dutch intensivists in the aftermath of the COVID-19 crisis: A nationwide survey. J. Crit. Care 2021, 62, 1–5. [CrossRef] [PubMed] 49. Azoulay, E.; De Waele, J.; Ferrer, R.; Staudinger, T.; Borkowska, M.; Povoa, P.; Iliopoulou, K.; Artigas, A.; Schaller, S.J.; Hari, M.S.; et al. Symptoms of burnout in intensive care unit specialists facing the COVID-19 outbreak. Ann. Intensive Care 2020, 10, 110. [CrossRef] [PubMed] 50. Luceño-Moreno, L.; Talavera-Velasco, B.; García-Albuerne, Y.; Martín-García, J. Symptoms of posttraumatic stress, anxiety, depression, levels of resilience and burnout in spanish health personnel during the COVID-19 pandemic. Int. J. Environ. Res. Public Health 2020, 17, 5514. [CrossRef] 51. Dimitriu, M.C.T.; Pantea-Stoian, A.; Smaranda, A.C.; Nica, A.A.; Carap, A.C.; Constantin, V.D.; Socea, B. Burnout syndrome in Romanian medical residents in time of the COVID-19 pandemic. Medical Hypotheses 2020, 144. [CrossRef] 52. Hu, D.; Kong, Y.; Li, W.; Han, Q.; Zhang, X.; Zhu, L.X.; Zhu, J. Frontline nurses’ burnout, anxiety, depression, and fear statuses and their associated factors during the COVID-19 outbreak in Wuhan, China: A large-scale cross-sectional study. eClinicalMedicine 2020, 24, 100424. [CrossRef] 53. Arora, M.; Asha, S.; Chinnappa, J.; Diwan, A.D. Review article: Burnout in emergency medicine physicians. Emerg. Med. Australas. 2013, 25, 491–495. [CrossRef] 54. Wu, Y.; Wang, J.; Luo, C.; Hu, S.; Lin, X.; Anderson, A.E.; Qian, Y. A Comparison of Burnout Frequency among Oncology Physicians and Nurses Working on the Frontline and Usual Wards During the COVID-19 Epidemic in Wuhan, China. J. Pain Symptom Manag. 2020, 60, e60–e65. [CrossRef] 55. Torrente, M.; Sousa, P.A.; Sánchez-Ramos, A.; Pimentao, J.; Royuela, A.; Franco, F.; Collazo-Lorduy, A.; Menasalvas, E.; Provencio, M. To burn-out or not to burn-out: A cross-sectional study in healthcare professionals in Spain during COVID-19 pandemic. BMJ Open 2021, 11, E044945. [CrossRef] 56. Chen, R.; Sun, C.; Chen, J.J.; Jen, H.J.; Kang, X.L.; Kao, C.C.; Chou, K.R. A Large-Scale Survey on Trauma, Burnout, and Posttraumatic Growth among Nurses during the COVID-19 Pandemic. Int. J. Ment. Health Nurs. 2021, 30, 102–116. [CrossRef] 57. Abdelhaﬁz, A.S.; Ali, A.; Ziady, H.H.; Maaly, A.M.; Alorabi, M.; Sultan, E.A. Prevalence, Associated Factors, and Consequences of Burnout Among Egyptian Physicians During COVID-19 Pandemic. Front. Public Health 2020, 8, 590190. [CrossRef] 58. Elhadi, M.; Msherghi, A.; Elgzairi, M.; Alhashimi, A.; Bouhuwaish, A.; Biala, M.; Abuelmeda, S.; Khel, S.; Khaled, A.; Alsoufi, A.; et al. Burnout Syndrome Among Hospital Healthcare Workers During the COVID-19 Pandemic and Civil War: A Cross-Sectional Study. Front. Psychiatry 2020, 11, 579563. [CrossRef] 59. Dillon, E.C.; Stults, C.D.; Deng, S.; Martinez, M.; Szwerinski, N.; Koenig, P.T.; Gregg, L.; Cobb, J.K.; Mahler, E.; Frosch, D.L.; et al. Women, Younger Clinician”, and Caregiver” Experiences of Burnout and Well-being During COVID-19 in a US Healthcare System. J. Gen. Intern. Med. 2002, 37, 145–153. [CrossRef] 60. Leo, C.G.; Sabina, S.; Tumolo, M.R.; Bodini, A.; Ponzini, G.; Sabato, E.L.; Mincarone, P. Burnout among Healthcare Workers in the COVID 19 Era: A Review of the Existing Literature. Front. Public Health 2021, 9, 750529. [CrossRef] 61. Çelmeçe, N.; Menekay, M. The Effect of Stress, Anxiety and Burnout Levels of Healthcare Professionals Caring for COVID-19 Patients on Their Quality of Life. Front. Psychol. 2020, 11, 597624. [CrossRef] Int. J. Environ. Res. Public Health 2023, 20, 6087 22 of 22 62. Pniak, B.; Leszczak, J.; Adamczyk, M.; Rusek, W.; Matłosz, P.; Guzik, A. Occupational burnout among active physiotherapists working in clinical hospitals during the COVID-19 pandemic in south-eastern Poland. Work 2021, 68, 285–295. [CrossRef] 63. Grigorescu, S.; Cazan, A.M.; Rogozea, L.; Grigorescu, D.O. Predictive Factors of the Burnout Syndrome Occurrence in the Healthcare Workers During the COVID-19 Pandemic. Front. Med. 2022, 9, 842457. [CrossRef] 64. Asai, M.; Morita, T.; Akechi, T.; Sugawara, Y.; Fujimori, M.; Akizuki, N.; Nakano, T.; Uchitomi, Y. Burnout and psychiatric morbility among physicians engaged in end-to-life care for cancer patients: A cross-sectional nationwide survey in Japan. Psychooncology 2007, 16, 421–428. [CrossRef] 65. Naldi, A.; Vallelonga, F.; Di Liberto, A.; Cavallo, R.; Agnesone, M.; Gonella, M.; Sauta, M.D.; Lochner, P.; Tondo, G.; Bragazzi, N.L.; et al. COVID-19 pandemic-related anxiety, distress and burnout: Prevalence and associated factors in healthcare workers of North-West Italy. BJPsych Open 2021, 7, e27. [CrossRef] 66. Cyr, S.; Marcil, M.J.; Houchi, C.; Marin, M.F.; Rosa, C.; Tardif, J.C.; Guay, S.; Guertin, M.C.; Genest, C.; Forest, J.; et al. Evolution of burnout and psychological distress in healthcare workers during the COVID-19 pandemic: A 1-year observational study. BMC Psychiatry 2022, 22, 809. [CrossRef] [PubMed] 67. Biber, J.; Ranes, B.; Lawrence, S.; Malpani, V.; Trinh, T.T.; Cyders, A.; English, S.; Staub, C.L.; McCausland, K.L.; Kosinski, M.; et al. Mental health impact on healthcare workers due to the COVID-19 pandemic: A U.S. cross-sectional survey study. J. Patient-Rep. Outcomes 2022, 6, 63. [CrossRef] [PubMed] 68. Pala, A.N.; Chuang, J.C.; Chien, A.; Krauth, D.M.; Leitner, S.A.; Okoye, N.M.; Costello, S.C.; Rodriguez, R.M.; Sheira, L.A.; Solomon, G.; et al. Depression, anxiety, and burnout among hospital workers during the COVID-19 pandemic: A cross-sectional study. PLoS ONE 2022, 17, e0276861. [CrossRef] [PubMed] 69. Moore, R.; Zielinski, M.J.; Thompson, R.G., Jr.; Willis, D.E.; Purvis, R.S.; McElﬁsh, P.A. “This Pandemic Is Making Me More Anxious about My Welfare and the Welfare of Others:” COVID-19 Stressors and Mental Health. Int. J. Environ. Res. Public Health 2021, 18, 5680. [CrossRef] 70. Adams, J.G.; Walls, R.M. Supporting the Health Care Workforce During the COVID-19 Global Epidemic. JAMA 2020, 323, 1439–1440. [CrossRef] 71. Hwang, S.; Kwon, K.T.; Lee, S.H.; Kim, S.W.; Chang, H.H.; Kim, Y.; Bae, S.; Cheong, H.S.; Park, S.Y.; Kim, B.; et al. Correlates of burnout among healthcare workers during the COVID-19 pandemic in South Korea. Sci. Rep. 2023, 13, 3360. [CrossRef] 72. Kalin, N.H. The Critical Relationship Between Anxiety and Depression. Am. J. Psychiatry 2020, 177, 365–367. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.2196_41005	JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Original Paper An Unguided, Computerized Cognitive Behavioral Therapy Intervention (TreadWill) in a Lower Middle-Income Country: Pragmatic Randomized Controlled Trial Arka Ghosh1, PhD; Rithwik J Cherian1,2, MSc; Surbhit Wagle1,3, MTech; Parth Sharma4, MTech; Karthikeyan R Kannan1, BTECH; Alok Bajpai5, MBBS, MD, DPM; Nitin Gupta1,6, PhD 1Department of Biological Sciences and Bioengineering, Indian Institute of Technology Kanpur, Kanpur, India 2Department of Cognitive Science, Indian Institute of Technology Kanpur, Kanpur, India 3Institute of Physiological Chemistry, University Medical Center Mainz, Mainz, Germany 4Department of Computer Science and Engineering, Indian Institute of Technology Kanpur, Kanpur, India 5Counseling Service, Indian Institute of Technology Kanpur, Kanpur, India 6Mehta Family Center for Engineering in Medicine, Indian Institute of Technology Kanpur, Kanpur, India Corresponding Author: Nitin Gupta, PhD Department of Biological Sciences and Bioengineering Indian Institute of Technology Kanpur IIT Campus Kanpur, 208016 India Phone: 91 5122594384 Email: [email protected] Abstract Background: Globally, most individuals who are susceptible to depression do not receive adequate or timely treatment. Unguided computerized cognitive behavioral therapy (cCBT) has the potential to bridge this treatment gap. However, the real-world effectiveness of unguided cCBT interventions, particularly in low- and middle-income countries (LMICs), remains inconclusive. Objective: In this study, we aimed to report the design and development of a new unguided cCBT–based multicomponent intervention, TreadWill, and its pragmatic evaluation. TreadWill was designed to be fully automated, engaging, easy to use, and accessible to LMICs. Methods: To evaluate the effectiveness of TreadWill and the engagement level, we performed a double-blind, fully remote, and randomized controlled trial with 598 participants in India and analyzed the data using a completer’s analysis. Results: The users who completed at least half of the modules in TreadWill showed significant reduction in depression-related (P=.04) and anxiety-related (P=.02) symptoms compared with the waitlist control. Compared with a plain-text version with the same therapeutic content, the full-featured version of TreadWill showed significantly higher engagement (P=.01). Conclusions: Our study provides a new resource and evidence for the use of unguided cCBT as a scalable intervention in LMICs. Trial Registration: ClinicalTrials.gov NCT03445598; https://clinicaltrials.gov/ct2/show/NCT03445598 (J Med Internet Res 2023;25:e41005) doi: 10.2196/41005 KEYWORDS computerized cognitive behavioral therapy; cCBT; depression; digital intervention; mobile phone https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 1 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Introduction Background Globally, >264 million individuals experience depressive disorders [1]. Despite the availability of evidence-based pharmacological and psychological treatment approaches, 76% to 85% of the individuals experiencing mental health disorders do not receive any treatment in low- and middle-income countries (LMICs) [2]. Barriers to accessing treatment for mental health disorders include the lack of access to treatment options, high cost, the fear of social stigma, and an inclination to self-manage the problem [3-5]. In India, there is a treatment gap of 85.2% for major depressive disorders [6]. One approach to bridging this treatment gap is to deliver computerized psychotherapy. The first therapeutic chatbot, ELIZA, was developed in 1966 [7]; it was a rudimentary program based on text rephrasing rather than evidence-based methods. The first computer-assisted cognitive behavioral therapy (CBT)–based program for depression was delivered in 1982 [8]. However, in the past 2 decades, the advent of stable internet connection and the pervasiveness of smartphones and computers have made it feasible to deploy technological interventions at scale. Computerized CBT (cCBT) has gained traction as a viable treatment modality, with >200 trials conducted to date [9]. cCBT for depressive disorders, both guided and unguided, has been evaluated in several clinical trials worldwide. In both guided and unguided cCBT interventions, the intervention is provided by a software; in guided interventions, a guide or a coach is additionally involved who provides encouragement, technical assistance, and explanations of the intervention, whereas in a strictly unguided intervention, the user should not have any interaction with a human guide. Recent studies and meta-analyses have indicated that for depressive symptoms, guided cCBT interventions are more beneficial than unguided cCBT interventions [10-14]. Carlbring et al [15] showed equivalent effects between guided cCBT interventions and face-to-face CBT. Including guided cCBT intervention with treatment as usual does not add any extra benefits [16]. Moreover, although guided cCBT intervention can be a feasible option in high-income countries [17], it is not feasible in LMICs because of the acute shortage of mental health professionals who can act as qualified guides [18]. Unguided cCBT interventions have the potential to bridge the treatment gap in LMICs. The evidence for unguided cCBT interventions is mixed, with some meta-analyses showing that they are effective with a small or medium effect size [14,19,20] and some showing that they are not effective [21-23]. The effectiveness of the unguided interventions is reduced by the high dropout rates. Note that the unguided studies included in the meta-analyses often involved initial contact with humans for diagnostic interviews [24-29], weekly telephone contact support [30,31], or treatment as usual [16,32]. Even minimal human contact can increase adherence to the interventions compared with a study without any such contact [33,34]. Indeed, Fleming et al [35] found that adherence rates observed in trial settings failed to translate into the real world. Recent https://www.jmir.org/2023/1/e41005 XSL•FO RenderX meta-analyses have reported a positive correlation between treatment adherence and treatment effects [14,19]. In addition, a recent meta-analysis found that existing guided or unguided cCBT interventions had low acceptability among patients, which was even less than that of waitlist [10]. interventions have been conducted Studies on cCBT predominantly in high-income countries [36]; however, systematic reviews on depression and mental health disorders in LMICs have been done by Martínez et al [37] and Fu et al [38], respectively. A recent meta-analysis reported that 92% of the studies on diagnosed depression had been conducted in Western Europe, North America, and Australia [39]. The interventions have been developed, evaluated, and made available for free only in these high-income regions. There is a need for unguided interventions that are more effective, have higher adherence, and are available free of cost for wide accessibility in LMICs. Objective In this study, we developed and evaluated such a cCBT-based multicomponent intervention, TreadWill. We included several features in TreadWill that could increase adherence to and improve the effectiveness of a completely unguided intervention. We also developed an active control version of TreadWill that presented the same therapeutic content without these features. We designed a fully remote 3-armed randomized controlled trial (RCT)—an experimental version of TreadWill, a plain-text version of TreadWill with the same therapeutic CBT content (active control), and a waitlist control. We hypothesized that the participants in the experimental group would show significantly greater improvement in depressive and anxiety symptom severity. We also hypothesized that the participants in the experimental group would show significantly more engagement in terms of modules completed and absolute time used compared with the active control group participants. Methods Study Design We designed a fully remote RCT to test the effectiveness of the experimental version of TreadWill compared with an active control version and a waitlist control version. We planned to recruit 600 participants with a 1:1:1 distribution across the 3 groups. We implemented simple randomization using an automated randomization function (developed in Python; version 3.4.3; Python Software Foundation). This trial was registered at ClinicalTrials.gov before commencement (NCT03445598). Participant Recruitment and Screening We recruited the participants using both offline and web-based publicity. We displayed flyers in residential hostels, research buildings, and lecture halls at the Indian Institute of Technology, Kanpur. A press release helped with coverage in newspapers and social media. The publicity material included a website link to join this study. The link opened a web page that provided information regarding the study and accepted the email ID of the interested participants. Over the next 3 steps, the web page collected the J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 2 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al it (the demographic data, baseline Patient Health Questionnaire-9 (PHQ-9) score [40], and informed consent from the potential participants. The entire participant recruitment process was automated (including self-reports and self-administered questionnaires) to eliminate human contact and maintain scalability. To be eligible to participate in the study, an individual must be an Indian resident aged between 16 and 35 years. They must be fluent in English and have had access to an internet-enabled computer or tablet device. They must have had scored between 5 and 19 (both inclusive) in the PHQ-9 with a score of 0 on the ninth question. We decided to include participants with mild symptoms of depression (a score of 5-9 in the PHQ-9) and exclude those with severe symptoms (a score >19) because our program was targeted not at the clinical population but at a wider population with susceptibility for depression. We excluded individuals who were unemployed, had a diagnosis of bipolar disorder or psychosis, or reported that they just wanted to check out the site and did not plan to complete trial commencement to exclude casual visitors to the website). Because of the pragmatic nature of our study, we included participants regardless of whether they were receiving treatment for depression. Once a potential participant met the inclusion and exclusion criteria and provided informed consent (for individuals aged between 16 and 17 years, informed consent was also required from a parent or guardian), the individual was scheduled to be recruited in the study. After 18 hours, the individual was randomized to 1 of the 3 groups and received a unique link via email. The delay of 18 hours was included to prevent individuals from signing up using disposable temporary email IDs. They were counted as participants in the study only after clicking on the unique link and were led to a sign-up page (for participants assigned to the experimental or active control groups). The participants assigned to the waitlist control group were led to a page to collect their baseline Generalized Anxiety Disorder-7 (GAD-7) scores [41]; GAD-7 scores of the experimental and active control groups were taken just before the start of the first module in the intervention. The participants did not receive any monetary compensation. last condition was added after Ethics Approval The Institutional Ethics Committee of the Indian Institute of Technology Kanpur provided ethical clearance to conduct this study (IITK/IEC/2017-18 II/1). Safety Check At any stage in the intervention, if we detected severe depressive symptoms or suicidal ideation, we blocked access to TreadWill. Severe depressive symptoms were determined as a total score of >19 on the PHQ-9. Suicidal ideation was detected by a score of >0 on the ninth question of the PHQ-9 and a total score of >4 on the Suicidal Intent Questionnaire [42]. In such cases, email and SMS text messaging alerts were sent to the participants (and their buddy, if they had one in the program), requesting them to seek professional help. For participants aged between 16 and 17 years, an email notification was sent to the parent or guardian as well. The participants had not been informed of this exclusion criterion; therefore, they did not https://www.jmir.org/2023/1/e41005 XSL•FO RenderX intentionally suppress their scores for the sake of continuing the intervention. Automated Notifications Participant contact was minimal and automated. Participants who initiated the recruitment process but did not complete it were sent automated email reminders encouraging them to complete the process. Participants also received periodic automated email and SMS text messaging reminders nudging them to use TreadWill (Table S1 in Multimedia Appendix 1 provides details). The program asked the participants about their preferred time to log in; using this information, email and SMS text messaging alerts were sent 10 minutes in advance to remind the participants. The research team did not initiate any direct contact with the participants. Technical support via email was provided in case the participants sent an email requesting for it. Active Control Version The active control version presented the same CBT content as the experimental version in the same 6 modules, but used plain text instead of slides, videos, and conversations. Each module had Introduction, Learn, and Discuss sections, but the Practice section was excluded. The content was not tailored according to the participant. The active control version included the CBT forms, but excluded games, such as SupportGroup, PeerGroup, and the option to involve a buddy. The participants received only essential email notifications (Table S1 in Multimedia Appendix 1 presents the details of notifications). The active control version was introduced to test whether the additional interactive elements introduced in the experimental version increased user engagement. Development of the Intervention We used the Django framework (Django Software Foundation) for developing the TreadWill website. We used Google Slides (Google LLC) to embed the slides and YouTube (Google LLC) to embed the videos on the website. We used images with a Creative Commons license for use in slides and videos. We used images from the internet for the Identify the friendly face game [43]. The content and the website underwent multiple rounds of checking by the development team and other volunteers to fix errors before launching the trial. Assessments We used the PHQ-9 [40] and GAD-7 [41] questionnaires to measure depressive and anxiety symptom severity, respectively. For the experimental and active control group participants, the PHQ-9 and GAD-7 were administered before the beginning of each module, after completing all the modules, and at the 90-day follow-up time point. The first PHQ-9 (administered before randomization, as it was an inclusion-exclusion criterion) and the first GAD-7 (administered after randomization but before the first module of the intervention) served as the baseline scores. For the waitlist control group, the PHQ-9 and GAD-7 were administered at baseline and after a 42-day interval (this interval was chosen to be at par with the expected intervention duration of approximately 6 weeks for completing the 6 modules in the experimental group). After submitting the 42-day J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 3 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al assessments, the waitlisted participants were also given access to the intervention. Blinding All the participants followed the same recruitment procedure. Consequently, the participants were unaware of which version of TreadWill they were assigned to (they did not even know that 2 different versions existed). Therefore, we expected placebo effects in the 2 groups to be similar. In addition, the PHQ-9 and GAD-7 data were self-reported on the website; therefore, there was no scope for evaluator bias. Data Security and Privacy All the participants agreed to allow their data to be used for research purposes and to be reported in a deidentified format. All participant data were transferred over Secure Sockets Layer. The only personal identifiers provided by the participants were their email IDs and phone numbers. Before analyzing the data, all the participants’ email IDs and phone numbers were removed from the data set. Primary, Secondary, and Exploratory End Points We performed a completer’s analysis (Discussion section). The primary end point was the final PHQ-9 score in participants who completed at least half of the intervention (3 out of 6 modules). The primary end point was decided after the trial completion but before any data analysis. We decided the cutoff point at 3 modules to ensure that all the participants were exposed to the cognitive aspect of CBT, which we introduced in the third module. We also analyzed the data of participants who had completed all the modules. A similar analysis approach based on module completion in web-based studies has been used by Christensen et al [44], Keefe et al [45], and Rollman et al [46] (the Discussion section elaborates on the rationale for using this analysis approach). TreadWill was primarily designed to help individuals with depressive symptoms. Therefore, PHQ-9 was our primary outcome measure. However, as anxiety and depression are highly comorbid, we wanted to check whether the techniques presented in TreadWill also helped in the reduction of anxiety symptoms. Thus, for the experimental and the active control groups, the secondary end point was the GAD-7 score in participants who completed at least half of the intervention (3 out of 6 modules). Other secondary end points included PHQ-9 and GAD-7 scores at the 90-day follow-up. The intermediate PHQ-9 and GAD-7 scores (after every module) and 2 surveys conducted after the module 3 and the module 6 were used as exploratory end points. Statistical Analyses Owing to the high dropout rate, we did not assume the PHQ-9 and GAD-7 scores to be normally distributed; therefore, we used nonparametric statistical the effectiveness of the intervention. We used the Kruskal-Wallis test for comparing the reduction in depression or anxiety symptom severity from baseline to the primary end point among the 3 groups. All the tests were 2 tailed unless otherwise mentioned. For post hoc analysis between the groups, we used tests for analyzing https://www.jmir.org/2023/1/e41005 XSL•FO RenderX the Mann-Whitney U test. The tests were conducted using MATLAB (MathWorks) and Python (Python Software Foundation). Because this was the first trial of TreadWill, we did not have a prior estimate of the dropout rate and could not perform power calculations. We chose the sample size of 600 participants based on the previous studies of similar nature [16,47]. Results Approach Taken for Developing the Intervention We aimed to develop and evaluate a fully automated intervention, TreadWill, that would be engaging and effective without any expert guidance or contact. We reviewed the existing cCBT interventions before starting the development process and considered factors that may be responsible for the high dropout rates. The common shortcomings that we identified included the lack of interactive content, lack of tailoring of the content to different users, lack of peer support for users, and lack of engaging games. Different interventions addressed some of these shortcomings by including the corresponding features; however, none of them included all the features. We developed TreadWill the development process, we used the inputs on initial prototypes from the institute counselors and psychiatrists and from 13 pilot users (not included in the eventual trial), before finalizing the content and user experience in TreadWill. We hypothesized that TreadWill would lead to a high adherence rate and a significant reduction in depressive and anxiety symptom severity. As we did not plan to charge the users, we also expected TreadWill to be more accessible, especially in LMICs, compared with paid interventions. these shortcomings. During to address Design of TreadWill We designed the therapeutic content of TreadWill based on CBT, using the book by Beck [48] as the primary reference. TreadWill delivered the core concepts of CBT in a structured format with 6 modules (Table S2 in Multimedia Appendix 1 shows the details) in an easily understandable language. Each module consisted of 4 sections: Introduction, Learn, Discuss, and Practice. In the Introduction section, an automated virtual therapist explained the importance of the module through interactive text-based dialogue. The Learn section included psychoeducation in the form of slides and videos. Slides consisted of multiple infographics that were presented sequentially (Figure S1 in Multimedia Appendix 1). Videos consisted of animated content with a voiceover explaining the concepts that were visible on the screen. In the Discuss section, the participants learned to apply the psychoeducation to real-life situations through conversations. These conversations were text-based dialogues with an automated virtual patient (Figure S2 in Multimedia Appendix 1), presented in an interactive format designed to simulate human chat. Although the conversations were preprogrammed, in many instances, the participants could choose from >1 response, thus providing some control to the user in steering the dialogue. The Practice section included interactive quizzes on the material covered in each module. J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 4 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al To ensure sequential progression through the intervention, only the first module was initially accessible to the user and the later modules were locked. After the completion of all sections in a module and a 4-day gap since its unlocking (to prevent rushing through the modules), the next module was unlocked. Steps within a module were also unlocked gradually upon the completion of the preceding steps. Each participant had a maximum of 90 days to complete the 6 modules starting from their first log-in. After 90 days, they could continue to use the modules that were already unlocked until then but could not unlock new modules. We did this to restrict participants’ exposure to new therapeutic content after 90 days and provide a clear deadline, as recommended by several studies [49-51]. Interactive Games and CBT Forms in TreadWill TreadWill included 2 interactive games. The Identifying thinking errors game was aimed at training the participants in spotting thinking errors in their negative automatic thoughts. The gameplay involved the presentation of a situation, a related negative automatic thought, and a list of 10 thinking errors from which the participant had to select one or more thinking errors present in the thought. Selecting the correct option allowed the participant to move to the next level. When an incorrect option was selected, feedback was provided along with an opportunity to try again. The Identify the friendly face game is based on the training paradigm developed by Dandeneau and Baldwin [52] to train participants to overcome the negative attention bias and improve their self-esteem, thereby reducing the risk of depression [53,54]. The game presented 4 images in a 2×2 grid with 3 faces showing a negative emotion and 1 face showing a positive emotion. The participant was allowed 5 seconds to find the positive image and thus increase their score. If the participant responded or if 5 seconds elapsed, a new set of images was displayed. The gameplay incentivized quick attention to positive emotions. The difficulty of the game continuously adapted to the participants’ competence: incorrect responses increased the frequency of faces with obvious emotions, and correct responses increased the frequency of faces with subtle emotions. TreadWill provided an interactive interface to fill in the forms commonly used in CBT: Thought record worksheet, Core belief worksheet, Behavioral experiment worksheet, Problem-solving worksheet, Prepare for setback worksheet, and Schedule activity worksheet (Table S3 in Multimedia Appendix 1). The forms allowed participants to apply CBT techniques to their situations and save the information for future reference. Peer and Family Support in TreadWill Individuals looking for support on the internet might have low social support in real life [55]. In such cases, web-based peer-based support has been shown to be effective in reducing depressive symptoms [56,57]. Keeping this in mind, we designed the SupportGroup and PeerGroup features in TreadWill to provide a social space where participants could connect with other TreadWill users and potentially help each other in solving their problems. Posts in the SupportGroup were visible to all the TreadWill participants. The participants could upvote or downvote posts, add comments, and send thank you messages to each other. PeerGroups were smaller groups of 10 members each, designed in such a way that the posts in a PeerGroup were visible only to the members of that PeerGroup. We provided the participants with the option to invite a family member or a friend as their buddy who would receive weekly updates about the participant’s activities in TreadWill. We hypothesized that the involvement of the buddy would motivate the participants to complete the program. We sent an email to this buddy if the participant failed to use TreadWill regularly and requested them to nudge the participant. Content Tailoring in TreadWill Content tailoring has the potential to increase adherence to cCBT interventions, as participants are more likely to stick with a program if they find the content relatable [50,58,59]. In TreadWill, we implemented tailoring by selecting examples in the conversations based on the participant’s occupation (high school students, college students, or working professionals). In addition, we tailored the conversations based on participants’ thoughts, beliefs, and situations in the following manner. First, we asked the participants to select relatable intermediate and core beliefs from the Dysfunctional Attitude Scale [60], negative automatic thoughts from the Automatic Thoughts Questionnaire [61], and stressful situations from a curated list. Then, we made the simulated virtual patients in the subsequent conversations identify with similar beliefs, thoughts, and situations, and the participant’s goal was to help the simulated patient by using the CBT techniques learned in that module. The automated email and SMS text messaging notifications received by the users were also tailored according to their preferences (Methods section). Participants Recruitment commenced on February 14, 2018, with a planned enrollment of 600 participants. The primary completion date was March 2, 2019, after full enrollment, and the secondary completion date was May 31, 2019. Of the 5188 individuals who started the registration process for the study, 598 (11.53%) participants completed all the steps and met the study inclusion criteria (2 other participants who did not meet the inclusion criteria were initially included owing to a software bug but were excluded when we cross-checked the data during data analysis). The 598 participants were randomly assigned to the 3 study arms with equal probability (Methods section), resulting in 204 experimental, 189 active control, and 205 waitlist control participants (Figure 1). The participants in the 3 groups were found to be balanced in terms of age, sex, the severity of depressive symptoms, occupation, the use of other interventions, motivation for joining, and the occurrence of recent traumatic events (Table 1). The baseline PHQ-9 scores in the 3 groups were not significantly different: Kruskal-Wallis H(2)=2.04 (P=.36). However, a sex bias (478/598, 79.9% male) was observed because participants in our study were recruited mainly from Indian engineering colleges where the students were predominantly male [62]. In addition, in India, there is a 56% gender gap in mobile internet use [63]. https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 5 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Figure 1. The flow of participants in the trial. In the experimental and the active control groups, the follow-up scores of only those participants who had completed at least 3 modules were analyzed. In the waitlist group, the 42-day interval scores of only those participants who had also submitted the baseline scores were analyzed. GAD-7: Generalized Anxiety Disorder-7; PHQ-9: Patient Health Questionnaire-9. https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 6 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Table 1. Baseline and demographic characteristics of the participants recruited in the study.a Groups Experimental (n=204) Active control (n=189) Waitlist con- trol (n=205) All (n=598) Group comparison—test result H(2) Chi-square (df; n=598) P value Age (years), mean (SE) 23.76 (0.30) 23.42 (0.28) 23.48 (0.29) 23.56 (0.17) 0.205 N/Ab Sex, n (%) Male Female 160 (78.4) 44 (21.6) 151 (79.9) 167 (81.5) 478 (79.9) 38 (20.1) 38 (18.5) 120 (20.1) N/A 0.586 (2) Traumatic event or death of a loved one, n (%) N/A 0.199 (2) Yes No Joining for help, n (%) Yes No 22 (10.8) 182 (89.2) 180 (88.2) 24 (11.8) 18 (9.5) 20 (9.8) 60 (10) 171 (90.5) 185 (90.2) 538 (90) 157 (83.1) 180 (87.8) 517 (86.5) 32 (16.9) 25 (12.2) 81 (13.5) N/A 2.722 (2) Secondary help, n (%) N/A 4.968 (6) .90 .75 .91 .26 .55 185 (90.7) 178 (94.2) 193 (94.1) 556 (93) None Counseling Medication Both 5 (2.5) 12 (5.9) 2 (1) Occupation, n (%) High school Student 1 (0.5) Between school and college 8 (3.9) 4 (2.1) 4 (2.1) 3 (1.6) 2 (1.1) 4 (2.1) 4 (2) 6 (2.9) 2 (1) 4 (2) 6 (2.9) 13 (2.2) 22 (3.7) 7 (1.2) 7 (1.2) 18 (3) N/A 7.620 (14) .91 College student 113 (55.4) 103 (54.5) 120 (58.5) 336 (56.2) Coaching after college 33 (16.2) 33 (17.5) 30 (14.6) 96 (16.1) Working professionals 39 (19.1) 40 (21.2) 37 (18) 116 (19.4) Self-employed Freelancers Volunteers PHQ-9c, mean (SE) 7 (3.4) 2 (1) 1 (0.5) 4 (2.1) 3 (1.6) 0 (0) 4 (2) 2 (1) 2 (1) 15 (2.5) 7 (1.2) 3 (0.5) 10.76 (0.26) 10.81 (0.25) 10.42 (0.27) 10.66 (0.15) 2.04 N/A .36 aThe 3 groups were not statistically different in these characteristics, as indicated by the statistical tests reported in the last column. bN/A: not applicable. cPHQ-9: Patient Health Questionnaire-9. Effectiveness of TreadWill In the primary analysis, we included the participants who completed at least 3 modules in the experimental group or the active control group. For this analysis, we used the last PHQ-9 scores submitted by these participants, excluding the follow-up questionnaire. Henceforth, we refer to the time of these last scores as the primary end point. In the waitlist control group, all users who submitted the questionnaires after the waiting period were included in the analysis. We compared the reductions in the PHQ-9 scores from the baseline to the primary end point between the 3 groups (Figures 2A and 2B; Table 2). The 3 groups showed significant differences in the reductions in the PHQ-9 score (Kruskal-Wallis test H(2)=8.93; P=.01); a post hoc test with Bonferroni correction revealed that the experimental group showed a larger reduction than the waitlist control group (2.73 vs 1.12; Mann-Whitney U=1027; experimental group: n=22; waitlist control group: n=139; P=.04). The differences in PHQ-9 reductions between the experimental and the active control groups were not significant (U=96; experimental group: n=22; active control group: n=7; P=.34). in the reductions In secondary analysis, the 3 groups showed significant differences score (Kruskal-Wallis test H(2)=8.02; P=.02); a post hoc test with Bonferroni correction showed a larger reduction in the experimental group than in the waitlist control group (3.27 vs 0.89; Mann-Whitney U=637.50; experimental group: n=22; the GAD-7 in https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 7 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al waitlist control group: n=94; P=.02). The differences in the GAD-7 reductions between the experimental and the active control groups were not significant (U=52.5; experimental group: n=22; active control group: n=7; P=.22). We also checked the reduction in the PHQ-9 and GAD-7 scores for the smaller set of the experimental group participants who completed all 6 modules (Figures 2C and 2D; Table 2); this analysis could not be performed for the active control group because only 1 participant from that group completed all 6 modules. This analysis also showed that the experimental group had a significantly larger reduction in PHQ-9 scores compared with the waitlist control group (4.20 vs 1.12; Mann-Whitney U=368.5; experimental group: n=10; waitlist control group: n=139; P=.01) and GAD-7 scores (3.40 vs 0.89; Mann-Whitney U=260.5; experimental group: n=10; waitlist control group: n=94; P=.02). The participants who completed all modules in the experimental and the active control groups did not differ demographically or in their baseline PHQ-9 scores from the rest of the participants (Table S4 in Multimedia Appendix 1). The reductions observed in the PHQ-9 and GAD-7 scores in the experimental and the active control groups at the primary end point were maintained at the 90-day follow-up period (Figures 2A and 2B). Thus, both the full-featured version of TreadWill (experimental) and the plain-text version of TreadWill (active control) were effective in reducing depression- and anxiety-related symptoms in participants who completed all or at least 3 modules. We checked whether the novel features of the experimental version of TreadWill were able to increase engagement compared with the active control version. Every module was completed by more participants in the experimental version than in the active control version (Figure 3A). The odds of completing at least 3 modules were 3 times higher for a participant in the experimental group compared with a participant in the active control group (odds ratio 3.004, 95% CI 1.247-7.237; P=.01). The experimental group participants used TreadWill for an average of 79.8 minutes and the active control group participants for 26.1 minutes; the difference was statistically significant (Mann-Whitney U=10,290; experimental group, n=181; active control group, n=159; P<.001; Figure 3B). Thus, the full-featured version of TreadWill had higher engagement and less attrition than the plain-text version. Furthermore, we checked whether the level of engagement with TreadWill was related to the reductions in depressive and anxiety symptoms. We found that the reduction in the PHQ-9 score was positively correlated with the number of modules completed within each group (experimental group: Spearman ρ=0.38; P=.003; n=61; Figure 3C; active control group: ρ=0.51; P<.001; n=41; Figure 3D) and with the total use time (experimental group: ρ=0.39; P=.002; n=61; Figure 3E; active control group: ρ=0.47; P=.002; n=41; Figure 3F). The reduction in the GAD-7 score was also moderately correlated with the number of modules completed (experimental group: ρ=0.27; P=.04; n=57; Figure S3A in Multimedia Appendix 1; active control group: ρ=0.43; P=.009; n=37; Figure S3B in Multimedia Appendix 1) and total use time (experimental group: ρ=0.25; P=.07; n=57; Figure S3C in Multimedia Appendix 1; active control group: ρ=0.35; P=.04; n=37; Figure S3D in Multimedia Appendix 1). Figure 2. Changes in Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7) scores after using TreadWill. (A) and (B) Violin plots show PHQ-9 (A) or GAD-7 (B) scores at baseline, primary end point, and follow-up for the experimental group, the active control group, and the waitlist group participants. Primary end point is defined as the latest PHQ-9 or GAD-7 score submitted after completing at least 3 modules. For PHQ-9, experimental group: n=22, active control group: n=7, waitlist group: n=139; for GAD-7, experimental group: n=22, active control group: n=7, waitlist group: n=94. (C) and (D) Violin plots show the change from baseline to program completion in PHQ-9 (C) or GAD-7 (D) score for the experimental group participants who completed all 6 modules (blue violin). For waitlist group participants (orange violin), the plots show the change from the score at the baseline to the score after the 42-day waiting interval (considered as the primary end point for the waitlist group). Red horizontal lines: median; black: mean. Error bars represent SE. https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 8 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Table 2. Average changes (SE) in Patient Health Questionnaire-9 (PHQ-9) and Generalized Anxiety Disorder-7 (GAD-7) scores for the 3 groups from the baseline to the primary end point or to the completion of all modules.a Groups Experimental (n=204) Active control (n=189) Waitlist control (n=205) Average change (SE) Values, n (%) Average change (SE) Values, n (%) Average change (SE) Values, n (%) PHQ-9 Primary end point −2.73 (1.27) 22 (10.8) −5.14 (2.28) 7 (3.7) −1.12 (0.37) 139 (67.8) Completion of all modules −4.20 (0.83) 10 (4.9) —b — — — GAD-7 Primary end point −3.27 (0.97) 22 (10.8) −1.43 (0.92) 7 (3.7) −0.89 (0.42) 94 (45.9) Completion of all modules −3.40 (0.82) 10 (4.9) — — — — aAs only 1 participant in the active control group completed all modules, the corresponding values were not analyzed. bNot available. Figure 3. Adherence with TreadWill and the relationship between intervention use and symptom reductions. (A) The graph shows the number of participants in the experimental (blue) and the active control (red) groups who completed the indicated number of modules. (B) Violin plots show the total use times of the experimental and the active control group participants. Red horizontal lines: median, black: mean. Error bars represent SE. (C) and (D) The reduction in Patient Health Questionnaire-9 (PHQ-9) scores versus the number of modules completed by the experimental group participants (C) and the active control group participants (D). (E) and (F) The reduction in PHQ-9 scores versus the total use time in hours for the experimental group participants (E) and the active control group participants (F). In all cases, the reduction in PHQ-9 scores was calculated by subtracting the last PHQ-9 score (excluding follow-up) from the baseline score; a positive value indicates improvement. Some points in the graphs are overlapping. https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 9 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Evaluating the Role of Possible Confounding Factors Differences in time and motivation can act as confounding factors in the performance of an intervention. To test whether the observed differences between the experimental and the waitlist groups were affected by these factors, we performed additional analyses. We had planned to take the PHQ-9 and GAD-7 scores in the waitlist group at 42 days, as the experimental group participants were also expected to take 42 days to submit the final questionnaire (6 modules at the rate of 1 module per week). However, variability in the actual timing of score submission was inevitable in a fully unguided and remote study. In our data set, we found that the actual timing of the final questionnaire was 63.7 (SD 26.8) days for the experimental group and 47.4 (SD 10.5) days for the waitlist group participants. To check if this difference in timing can explain the difference in the performance, we split the waitlist group participants into 2 subgroups depending on when they submitted the PHQ-9 questionnaires: the first subgroup included participants who had submitted before 46 days (mean 43.56, SD 1.10 days; nWL1=99), and the second subgroup included participants who submitted after 46 days (mean 57.05, SD 15.91 days; nWL2=40); by design, the mean number of days for the 2 subgroups were significantly different (Mann-Whitney U=3960; P<.001). However, the mean reductions in PHQ-9 scores for these 2 subgroups were not significantly different (U=1987.5; P=.97). This indicates that for the waitlist group, the difference in the number of days in the observed range did not affect the PHQ-9 scores significantly. To check whether the higher reduction in the PHQ-9 scores in the experimental group than in the waitlist group can be explained by motivation, we performed the following analysis. In our study, the waitlist group participants were given the option to sign up for the experimental version of the intervention once the waitlist period was over (ie, when their formal participation in the study had ended, they were not considered as experimental group participants). It is reasonable to expect that the waitlist group participants who actually signed up for this option, despite the long gap of at least 42 days, were more motivated than the rest. We created a subgroup of these more motivated waitlist group participants and compared their performances with that of the remaining participants. These 2 subgroups did not show a significant difference in the reductions in the PHQ-9 scores (U=2160; motivated: n=64; unmotivated: n=75; P=.31). Another potential concern is that the users who happened to improve spontaneously may be likely to complete more modules; by performing a completer’s analysis, we may be selecting for such spontaneous improvers. We performed an additional analysis to check whether this was the case in our data. On the basis of this argument, the participants who went on to complete module 3 after completing module 2 would have seen more improvement in their PHQ-9 scores at the end of module 2 compared with the participants who dropped out just after completing module 2. We compared the reductions in PHQ-9 scores (from baseline to the end of module 2) of these 2 subgroups and found no significant difference (U=93; dropout: https://www.jmir.org/2023/1/e41005 XSL•FO RenderX n=9; continued: n=22; P=.81). Similarly, we compared the reductions in PHQ-9 scores (from baseline to the end of module 1) of participants who dropped out after completing module 1 and those who went on to complete the next module, and we did not find any significant difference (U=488; dropout: n=30; continued: n=31; P=.74). Thus, the idea that (spontaneous) improvement in performance encourages the participants to complete more modules is not supported by our data. On the basis of these analyses, we conclude that the higher reduction in PHQ-9 scores observed in the experimental group can be attributed to the effect of completion of the modules, rather than differences in the timing of questionnaires or in motivation. Feedback on the Features of TreadWill We programmed TreadWill to present surveys containing 15 questions using a 5-point Likert scale to quantify the participants’ feedback on various aspects of TreadWill. For example, one of the questions stated I found the email reminders helpful, to which the participant responded by selecting one of the following options: strongly agree, somewhat agree, neither agree nor disagree, somewhat disagree, and strongly disagree, which were mapped to a score of 2, 1, 0, −1, and −2, respectively (Table S5 in Multimedia Appendix 1 lists all questions). The surveys were conducted at 2 time points: after completing 3 modules and at the end of the intervention. In the experimental group, of the 22 participants who completed at least 3 modules, the first survey was submitted by 22 (100%) participants and the second survey was submitted by 18 (82%) participants. The participants reported positive feedback on most aspects of TreadWill (Figure 4A): mean feedback scores over all questions were significantly >0 for both the first survey (mean 1.16, SE 0.12; n=15 questions; t21=9.20; P<.001; 2-tailed t test) and the second survey (mean 1.29, SE 0.10; n=15 questions; t17=12.56; P<.001; 2-tailed t test). The scores remained largely consistent between the 2 surveys (Pearson r=0.87; P<.001; n=15). The strongest positive feedback was received for questions related to the ease of English used (mean 1.86, SE 0.10 in the first survey and mean 1.83, SE 0.12 in the second survey), the relatability of the examples (mean 1.23, SE 0.25 and mean 1.72, SE 0.13, respectively), the ease of using the CBT forms (mean 1.45, SE 0.18 and mean 1.22, SE 0.17), the engaging nature of the conversations (mean 1.36, SE 0.21 and mean 1.50, SE 0.20), the helpfulness of the Learning slides (mean 1.73, SE 0.10 and mean 1.67, SE 0.14), and the helpfulness of the Learning videos (mean 1.55, SE 0.13 and mean 1.67, SE 0.14). The features with the lowest ratings included the PeerGroup, which received weak positive feedback (mean 0.73, SE 0.23 and mean 0.61, SE 0.28) and the buddy feature, which received neutral feedback in both surveys (mean 0.09, SE 0.22 and mean 0.39, SE 0.20). In the active control group, of the 7 participants who completed at least 3 modules, 7 (100%) and 5 (71%) participants submitted their first and second surveys, respectively. The survey questions were slightly different in the active control group (Table S5 in Multimedia Appendix 1); the first 6 questions judged their opinion on aspects they experienced directly, and the next 9 J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 10 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al questions were asked in a prospective manner, for example, I would prefer to have email reminders. In the first 6 questions, the participants reported overall positive feedback (Figure 4B). In the 9 prospective questions, they showed interest in having only some of the proposed features, including conversations and game elements. Curiously, many features that the active they would not control group participants thought prefer—including SMS text messaging reminders, videos, and slides—were actually found to be useful by the experimental group participants who experienced the features (Figures 4A and 4B). No participant reported any adverse events through the contact form on the website. Figure 4. Feedback on the features of TreadWill. Violin plots show survey responses by the experimental group (A) and the active control group participants (B). Y-axis labels: 2=“strongly agree,” 1=“somewhat agree,” 0=“neither agree nor disagree,” −1=“somewhat disagree,” and −2=“strongly disagree.” Black lines indicate mean (SE). Exploratory Analysis To check if the content provided was engaging, we provided the experimental group participants with the option to provide feedback on the slides, videos, and conversations using like and dislike buttons. The slides, videos, and conversations were viewed 467, 205, and 1479 times, respectively, over all modules, of which nearly 17.1% (80/467), 19.5% (40/205), and 20.14% (298/1479) instances resulted in likes or dislikes feedback (Figures S4A, S4B, and S4C in Multimedia Appendix 1). We found that the feedback included more likes than dislikes for slides (mean 8.0 SE 2.30 likes vs mean 0, SE 0 dislikes; Wilcoxon W=45; n=10 slides; P<.001; Figure S4D in Multimedia Appendix 1); videos (mean 7.4, SE 1.51 likes vs mean 0.60, SE 0.54 dislikes; W=15; n=5 videos; P=.06; Figure S4E in Multimedia Appendix 1); and conversations (mean 1.88, SE 0.24 likes vs mean 0.12, SE 0.033 dislikes; W=6015; n=149 conversations; P<.001; Figure S4F in Multimedia Appendix 1). The participants also had the option of providing descriptive feedback on these elements. The subjective feedback was mostly positive, with participants frequently mentioning that they liked the given examples. One participant mentioned that they would have preferred to type their own answers in conversations (instead of choosing from prewritten text options). A word cloud created from the collated subjective feedback showed that the most frequently used words in feedback included given, example, liked, and idea (Figure S4G in Multimedia Appendix 1). TreadWill allowed participants to revisit previously completed conversations to refresh their memory; this option was used 17 times by the participants. TreadWill allowed participants to attach one or more word tags from a list of 44 tags to posts in the SupportGroup. A word cloud of the tags used during the study revealed the topics that were most commonly discussed by the participants: wasting https://www.jmir.org/2023/1/e41005 XSL•FO RenderX time, loneliness, guilt, self-esteem, and trust (Figure S5A in Multimedia Appendix 1). We also analyzed the entries made by the participants in the CBT forms (worksheets) to identify the common themes in their activities and concerns (Figures S5B and S5C in Multimedia Appendix 1). We checked the most commonly selected situations, thoughts, and beliefs from the lists presented to the experimental group participants. The most selected situation, thought, and belief were I am concerned about my career,I should be doing something better, and If I don’t work very hard, I’ll fail, respectively. (Figure S6 in Multimedia Appendix 1 presents the 10 most frequently selected situations, thoughts, and beliefs.) All waitlist group participants had the option to use the experimental intervention once their participation in the waitlist group was complete. Of the 205 waitlist group participants, 70 (34.1%) signed up to use the experimental group (of which 64/70, 91.4% submitted the follow-up). Of these 70 participants, 7 (10%) completed at least 3 modules and 5 (7.14%) completed all 6 modules. These values were comparable with the completion rates in the experimental group participants. In addition, we calculated the reduction in PHQ-9 and GAD-7 scores from waitlist posttreatment time point to the primary end point for the participants who completed at least 3 modules. The reductions in PHQ-9 (mean 3.14, SE 1.14; n=7) and GAD-7 (mean 4.28, SE 0.75; n=7) scores were statistically similar to those of the experimental group participants. Discussion Principal Findings We have presented the design of an unguided cCBT–based multicomponent intervention, TreadWill, aimed at high user engagement and universal accessibility. A fully remote RCT with 598 participants was performed to test the effectiveness J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 11 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al of TreadWill in reducing depression- and anxiety-related symptoms. The results of the trial show that the full-featured (experimental) and the plain-text (active control) versions of TreadWill effectively reduced both PHQ-9 and GAD-7 scores for the participants who completed at least 3 modules compared with the waitlist control group. The number of participants who completed at least 3 modules in the experimental group was nearly 3 times more than in the active control group. The extra features included in the experimental version increased adherence compared with the active control version in terms of both the time of engagement and the number of modules completed. The results also showed that the number of modules completed correlated with the reduction in the symptom severity of a participant. Two automated surveys presented during the intervention for taking participant feedback showed that the participants perceived TreadWill as useful and easy to use and found most of the interactive features helpful. In addition, the feedback provided by the participants using like and dislike buttons on different elements of the modules indicated that the participants found the content relatable and useful. Our target population was tech savvy and educated individuals (high school students, college students, and working young adults). We expected this target demographic to be comfortable with English to understand the material. We kept the language used in TreadWill simple enough for nonnative speakers to understand. The survey results confirmed that Easy English was one of the highest-rated features of TreadWill (Figure 4). Completer’s Analysis An intention-to-treat analysis allows one to assess whether assigning a particular intervention helps the participant. In an intention-to-treat analysis, all participants assigned to the interventions are analyzed, regardless of their completion status; the missing data are imputed or carried forward from earlier observations. The missing data problem is manageable in most studies, in which participants are recruited and monitored by experimenters, and the participants generally have high intrinsic motivation or perceived psychological pressure (owing to the involvement of others) or receive compensation for participating in the study. However, in a web-based, remote intervention, the intention-to-treat analysis might not be suitable, as previously noted by Christensen et al [44]. The problem becomes even more severe when a study, such as ours, is completely unguided; there is no compensation for the participants, and there are no psychological barriers to joining or leaving the study at any time, just by using a smartphone. Although such open designs pose a problem for the intention-to-treat analysis, they are desirable in other aspects, as they mimic the real-life use patterns of smartphone-based self-help interventions. An alternative analysis approach is to perform a completer’s analysis, in which the data of only those participants are analyzed who actually use the intervention. A completer’s analysis allows one to assess whether completing a particular intervention helps a participant. This is a more restricted claim compared with what can be made with an intention-to-treat analysis, especially from the perspective of a public health agency that has to decide which interventions to recommend to https://www.jmir.org/2023/1/e41005 XSL•FO RenderX people. However, in emerging cases of smartphone-based self-help interventions for which an intention-to-treat analysis is not ideal, a completer’s analysis can be a reasonable alternative. This approach has also been used in previous studies either in isolation or in combination with an intention-to-treat analysis [30,44-46,64-68]. Another rationale for using an intention-to-treat analysis is that, in the presence of dropouts, including all participants in the analysis maintains the equivalence established among the different groups at the baseline. Although we performed a completer’s analysis, we found that the baseline equivalence was also maintained in our data. The baseline PHQ-9 scores for all the participants who were included in our primary analysis after removing dropouts remained similar (Kruskal-Wallis H(2)=1.11; P=.57; Figure 2A). Limitations We did not require a clinical diagnosis of depression for including participants in the study because our goal was to create an accessible tool catering to both clinical and subclinical populations. Given that the prevalence of subclinical depression, defined as a score in the range of 5 to 9 on the PHQ-9, is fairly high at 15% to 20% [40,69,70], an unguided intervention can be immensely beneficial. We used only self-reported assessments for measuring symptom severity. Although self-reported assessments have their drawbacks [71], it was essential given the pragmatic nature of the study with an unguided intervention. For the same reason, we also included participants undergoing other treatments (42/598, 7% of our participants; Table 1). We used only 1 questionnaire each for assessing severity of depression and anxiety symptom. This decision was made keeping in mind that filling long questionnaires on the web is not a pleasant experience for users and might increase dropout rates [34]. In addition, while including multiple questionnaires for assessing the same disorder might improve validity, it also increases the risk of obtaining false-positive results by chance. Owing to the high dropout rate, we were unable to perform an intention-to-treat analysis. Although a completer’s analysis might be justified for a fully remote RCT, future work can evaluate TreadWill in a more traditional trial setting to assess intention-to-treat effects. Finally, our participants were young, mostly male, and tech-savvy college students, which reduces the generalizability of our results to the wider and much diverse population of India. Adherence Rates in cCBT Interventions Deprexis, a well-evaluated intervention, reported a full adherence rate of 7.5% in a fully unguided evaluation [50]. The high adherence rates observed in trial settings often fail to translate into the real world [35]. In real-world studies, adherence can be very low: 5.6% in the study by Lara et al [72], 13.11% in the study by Morgan et al [73], and 5% in the study by March et al [74]. In a fully remote trial of an app-based intervention, Arean et al [47] reported that 57.9% of the participants did not even download their assigned apps. Similarly, in another study involving no human contact, Morriss et al [34] reported that only 57.3% of participants randomized to the experimental group signed up for the intervention and only 42.5% accessed it more than once. Morriss et al [34] further J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 12 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al reported an attrition rate of 84.9% at the 3-week follow-up, with the attrition rate increasing at later follow-up points. In another recent study, Oehler et al [75] observed that the minimal dose was received by only 2.10% of the participants for the unguided version of iFightDepression. Guarino et al [76] also reported in a recent study that out of 2484 participants who signed up, only 562 started one of the modules, and the module completion rates ranged from 1% to 13% even by liberal definitions of module completion. The completion rates of web-based, self-help, and unguided cCBT interventions are comparable with the completion rates of massive open web-based courses, which have been reported to range from 3% to 6% [77,78]. The adherence rate observed for TreadWill, 12.1% (22/181) for moderate use and 5.5% (10/181) for full completion, is comparable with that reported in previous real-world studies. At this adherence level, TreadWill can benefit a significant number of people from the general population as a fully automated and scalable intervention. A web-based self-help intervention has a low opportunity cost; joining and dropping out of one intervention usually does not prevent a user from using another intervention; and trying out multiple apps before settling on one is a common behavioral pattern observed with smartphone apps. In addition, it is possible that the existing cCBT interventions and other digital mental health interventions do not provide help in the format that users expect on the web. The 12.1% adherence rate for moderate use was observed in our study despite additional challenges compared with other studies. Every step in our study, from participant recruitment to assessment, was fully automated; the lack of human contact is known to affect the commitment of participants [34]. Christensen et al [64] evaluated the cCBT intervention MoodGYM in 2 different settings: in a trial setting in which participants were called every week by human guides and provided instructions on completing the intervention, the completion rate for all 5 modules was 22.5%, but in an open setting (with no human contact), only 0.49% (97/19,607) of the participants completed the intervention. The intervention used in both cases was identical; the only difference was the interaction between participants and experimenters in the trial setting. This study shows that although it is possible to obtain higher completion rates in standard trial settings, these rates do not translate into real-world settings. Therefore, we used a pragmatic trial with no human contact, and even though the adherence rates are low, they are expected to be a more faithful representation of the real-world completion rates. Furthermore, it has been reported that male sex and young age significantly increase the chance of dropout [79]. The average age of our participant group was 23.6 (SE 0.17) years, and 79.9% (478/598) were male, which could have contributed to the dropout rate. Contrary to the practice of giving money or gift cards to participants [47,80-82], we did not reward participants for submitting assessments or for participation. In several studies [47,80-82], participants were paid even after submitting the baseline assessments. The practice of paying participants is likely to influence adherence to the intervention owing to the rule of reciprocity [83] and influence the assessment responses. Participants getting paid might feel that they owe it to the researchers to use the program and try to give answers in the assessments that they think the researchers expect. Not giving a reward also supports our pragmatic trial design; as in the real world, paying participants to use the intervention will be unsustainable. The generally positive comments that we received from the participants on the content (Figure S4 in Multimedia appendix 1) and various features of the intervention in the 2 surveys (Figure 4) suggested that the user dropout was not because of the unacceptability of the intervention. To check this further, we compared the survey responses of the experimental group participants who had completed all 6 modules with those who dropped out before completing 3 modules but completed a survey. The average feedback score was not lower in the dropout group than that in the completer group (Figure S7 in Multimedia Appendix 1). Implications Our study shows that even in low-resource settings, a cCBT-based intervention without expert support can help users who at least partially complete the intervention. This implication is immensely encouraging, as the number of mental health professionals is extremely low in India [84,85]. The reduction in PHQ-9 scores in our study was 2.73 for users who completed at least 3 modules and 4.20 for users who completed all modules. This level of reduction in a low-threshold intervention, with the potential to have a population-level impact, can be considered clinically significant [32]. Our study established TreadWill as a potential population-level intervention. This is among the largest studies conducted in India on digital mental health [86-89]. Our study is also the first fully web-based trial conducted in India and provides a template for conducting web-based trials for other mental health conditions in the country. Future work should focus on strategies, such as using gamification, serious games, or chatbots to build therapeutic alliance, to improve adherence to self-help interventions. Future work can also focus on making the intervention more similar to general web-based apps, so that users receive help in the formats with which they are familiar. In this study, we created tailored content for high school and college students and working professionals. Future work should also target the unemployed population and other susceptible groups. Acknowledgments The authors thank Romit Chaudhary, Divya Chauhan, Vinay Agarwal, Sahars Kumar, Pearl Sikka, Rahul Gupta, Nikhil Vanjani, and Sandarsh Pandey for their help in development of TreadWill. The authors thank Pranjul Singh, Aditya Patil, and Pearl Sikka for their help in developing the videos of TreadWill. The authors thank Prof. Braj Bhushan, Shoukkathali K, Rita Singh, Akanksha Awasthy, and Dr. Gitanjali Narayanan for helpful discussions, and they thank Dr Shikha Jain, Mrityunjay Bhargava, Pratibha Mishra, Jagriti Agnihotri, Swastika Tandon, Akash A, Harsh Agarwal, and the Indian Institute of Technology Kanpur media cell for promoting the visibility of the trial. The authors thank Silky Gupta and Aarush Mohit Mittal for helping with data analysis https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 13 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al and feedback on the manuscript. The authors thank many members of the Indian Institute of Technology Kanpur community for helping as pilot users of the intervention during its development, members of the counseling and psychiatry team for their feedback, and members of the Lab of Neural Systems for testing the beta version of TreadWill and giving feedback on the content and user experience. The authors thank Arun Shankar and Ranjeet Kumar for labeling images in “Identify the friendly face” game. This work was supported by the Cognitive Science Research Initiative of the Department of Science & Technology (grant DST/CSRI/2018/102). The funding agency had no role in the design or implementation of the study and in the interpretation of the results. Data Availability The deidentified data analyzed in this study are available from the corresponding author upon reasonable request. Authors' Contributions AG and NG conceptualized the project; AG, RJC, SW, AB, and NG designed the research; AG, RJC, SW, PS, and KRK developed the intervention; AG, AB, and NG recruited participants; AG and NG analyzed data; AG and NG wrote the paper with inputs from all coauthors. Conflicts of Interest None declared. Multimedia Appendix 1 Supplementary figures and tables. [DOCX File , 885 KB-Multimedia Appendix 1] Multimedia Appendix 2 CONSORT-EHEALTH (V 1.6.1) Checklist. [PDF File (Adobe PDF File), 728 KB-Multimedia Appendix 2] References 1. GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018 Nov 10;392(10159):1789-1858 [FREE Full text] [doi: 10.1016/S0140-6736(18)32279-7] [Medline: 30496104] 3. 2. Wang PS, Aguilar-Gaxiola S, Alonso J, Angermeyer MC, Borges G, Bromet EJ, et al. Use of mental health services for anxiety, mood, and substance disorders in 17 countries in the WHO world mental health surveys. Lancet 2007 Sep 08;370(9590):841-850 [FREE Full text] [doi: 10.1016/S0140-6736(07)61414-7] [Medline: 17826169] Andrade LH, Alonso J, Mneimneh Z, Wells JE, Al-Hamzawi A, Borges G, et al. Barriers to mental health treatment: results from the WHO world mental health surveys. Psychol Med 2014 Apr;44(6):1303-1317 [FREE Full text] [doi: 10.1017/S0033291713001943] [Medline: 23931656] Patel V, Saxena S, Lund C, Thornicroft G, Baingana F, Bolton P, et al. The Lancet Commission on global mental health and sustainable development. Lancet 2018 Oct 27;392(10157):1553-1598. [doi: 10.1016/S0140-6736(18)31612-X] [Medline: 30314863] Saxena S, Thornicroft G, Knapp M, Whiteford H. Resources for mental health: scarcity, inequity, and inefficiency. Lancet 2007 Sep 08;370(9590):878-889. [doi: 10.1016/S0140-6736(07)61239-2] [Medline: 17804062] 4. 5. 6. Murthy RS. National mental health survey of India 2015-2016. Indian J Psychiatry 2017;59(1):21-26 [FREE Full text] [doi: 10.4103/psychiatry.IndianJPsychiatry_102_17] [Medline: 28529357] 7. Weizenbaum J. ELIZA—a computer program for the study of natural language communication between man and machine. 8. 9. Commun ACM 1966 Jan;9(1):36-45 [FREE Full text] [doi: 10.1145/357980.357991] Selmi PM, Klein MH, Greist JH, Johnson JH, Harris WG. An investigation of computer-assisted cognitive-behavior therapy in the treatment of depression. Behav Res Methods Instrum 1982 Mar;14(2):181-185 [FREE Full text] [doi: 10.3758/bf03202150] Andersson G, Carlbring P. Cognitive behavioral therapy delivered using the internet. In: Wenzel A, editor. Handbook of Cognitive Behavioral Therapy: Applications. Volume 2. Washington, DC, USA: American Psychological Association; 2021:607-631. 10. Cuijpers P, Noma H, Karyotaki E, Cipriani A, Furukawa TA. Effectiveness and acceptability of cognitive behavior therapy delivery formats in adults with depression: a network meta-analysis. JAMA Psychiatry 2019 Jul 01;76(7):700-707 [FREE Full text] [doi: 10.1001/jamapsychiatry.2019.0268] [Medline: 30994877] https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 14 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al 11. Gilbody S, Brabyn S, Lovell K, Kessler D, Devlin T, Smith L, REEACT collaborative. Telephone-supported computerised cognitive-behavioural therapy: REEACT-2 large-scale pragmatic randomised controlled trial. Br J Psychiatry 2017 May;210(5):362-367 [FREE Full text] [doi: 10.1192/bjp.bp.116.192435] [Medline: 28254959] Johansson R, Andersson G. Internet-based psychological treatments for depression. Expert Rev Neurother 2012 Jul;12(7):861-870. [doi: 10.1586/ern.12.63] [Medline: 22853793] 12. 13. Karyotaki E, Efthimiou O, Miguel C, Bermpohl FM, Furukawa TA, Cuijpers P, Individual Patient Data Meta-Analyses for Depression (IPDMA-DE) Collaboration, et al. Internet-based cognitive behavioral therapy for depression: a systematic review and individual patient data network meta-analysis. JAMA Psychiatry 2021 Apr 01;78(4):361-371 [FREE Full text] [doi: 10.1001/jamapsychiatry.2020.4364] [Medline: 33471111] 14. Wright JH, Owen JJ, Richards D, Eells TD, Richardson T, Brown GK, et al. Computer-assisted cognitive-behavior therapy for depression: a systematic review and meta-analysis. J Clin Psychiatry 2019 Mar 19;80(2):18r12188 [FREE Full text] [doi: 10.4088/JCP.18r12188] [Medline: 30900849] 15. Carlbring P, Andersson G, Cuijpers P, Riper H, Hedman-Lagerlöf E. Internet-based vs. face-to-face cognitive behavior therapy for psychiatric and somatic disorders: an updated systematic review and meta-analysis. Cogn Behav Ther 2018 Jan;47(1):1-18. [doi: 10.1080/16506073.2017.1401115] [Medline: 29215315] 16. Gilbody S, Littlewood E, Hewitt C, Brierley G, Tharmanathan P, Araya R, REEACT Team. Computerised cognitive behaviour therapy (cCBT) as treatment for depression in primary care (REEACT trial): large scale pragmatic randomised controlled trial. BMJ 2015 Nov 11;351:h5627 [FREE Full text] [doi: 10.1136/bmj.h5627] [Medline: 26559241] 17. Andrews G, Basu A, Cuijpers P, Craske MG, McEvoy P, English CL, et al. Computer therapy for the anxiety and depression disorders is effective, acceptable and practical health care: an updated meta-analysis. J Anxiety Disord 2018 Apr;55:70-78 [FREE Full text] [doi: 10.1016/j.janxdis.2018.01.001] [Medline: 29422409] 18. Bruckner TA, Scheffler RM, Shen G, Yoon J, Chisholm D, Morris J, et al. The mental health workforce gap in low- and middle-income countries: a needs-based approach. Bull World Health Organ 2011 Mar 01;89(3):184-194 [FREE Full text] [doi: 10.2471/BLT.10.082784] [Medline: 21379414] 19. Karyotaki E, Riper H, Twisk J, Hoogendoorn A, Kleiboer A, Mira A, et al. Efficacy of self-guided internet-based cognitive behavioral therapy in the treatment of depressive symptoms: a meta-analysis of individual participant data. JAMA Psychiatry 2017 Apr 01;74(4):351-359 [FREE Full text] [doi: 10.1001/jamapsychiatry.2017.0044] [Medline: 28241179] 20. Twomey C, O'Reilly G, Bültmann O, Meyer B. Effectiveness of a tailored, integrative internet intervention (Deprexis) for depression: updated meta-analysis. PLoS One 2020 Jan 30;15(1):e0228100 [FREE Full text] [doi: 10.1371/journal.pone.0228100] [Medline: 31999743] 21. Garrido S, Millington C, Cheers D, Boydell K, Schubert E, Meade T, et al. What works and what doesn't work? A systematic review of digital mental health interventions for depression and anxiety in young people. Front Psychiatry 2019 Nov 13;10:759 [FREE Full text] [doi: 10.3389/fpsyt.2019.00759] [Medline: 31798468] 22. Harrer M, Adam SH, Baumeister H, Cuijpers P, Karyotaki E, Auerbach RP, et al. Internet interventions for mental health in university students: a systematic review and meta-analysis. Int J Methods Psychiatr Res 2019 Jun;28(2):e1759 [FREE Full text] [doi: 10.1002/mpr.1759] [Medline: 30585363] 23. Välimäki M, Anttila K, Anttila M, Lahti M. Web-based interventions supporting adolescents and young people with depressive symptoms: systematic review and meta-analysis. JMIR Mhealth Uhealth 2017 Dec 08;5(12):e180 [FREE Full text] [doi: 10.2196/mhealth.8624] [Medline: 29222079] 25. 24. Berger T, Hämmerli K, Gubser N, Andersson G, Caspar F. Internet-based treatment of depression: a randomized controlled trial comparing guided with unguided self-help. Cogn Behav Ther 2011;40(4):251-266. [doi: 10.1080/16506073.2011.616531] [Medline: 22060248] de Graaf LE, Gerhards SA, Arntz A, Riper H, Metsemakers JF, Evers SM, et al. Clinical effectiveness of online computerised cognitive-behavioural therapy without support for depression in primary care: randomised trial. Br J Psychiatry 2009 Jul;195(1):73-80 [FREE Full text] [doi: 10.1192/bjp.bp.108.054429] [Medline: 19567900] Farrer L, Christensen H, Griffiths KM, Mackinnon A. Internet-based CBT for depression with and without telephone tracking in a national helpline: randomised controlled trial. PLoS One 2011;6(11):e28099 [FREE Full text] [doi: 10.1371/journal.pone.0028099] [Medline: 22140514] 26. 27. Meyer B, Bierbrodt J, Schröder J, Berger T, Beevers CG, Weiss M, et al. Effects of an internet intervention (Deprexis) on severe depression symptoms: randomized controlled trial. Internet Interv 2015 Mar;2(1):48-59 [FREE Full text] [doi: 10.1016/j.invent.2014.12.003] 28. Mira A, Bretón-López J, García-Palacios A, Quero S, Baños RM, Botella C. An internet-based program for depressive symptoms using human and automated support: a randomized controlled trial. Neuropsychiatr Dis Treat 2017 Mar 31;13:987-1006 [FREE Full text] [doi: 10.2147/NDT.S130994] [Medline: 28408833] Spek V, Nyklícek I, Smits N, Cuijpers P, Riper H, Keyzer J, et al. Internet-based cognitive behavioural therapy for subthreshold depression in people over 50 years old: a randomized controlled clinical trial. Psychol Med 2007 Dec;37(12):1797-1806. [doi: 10.1017/S0033291707000542] [Medline: 17466110] 29. 30. Christensen H, Griffiths KM, Jorm AF. Delivering interventions for depression by using the internet: randomised controlled trial. BMJ 2004 Jan 31;328(7434):265 [FREE Full text] [doi: 10.1136/bmj.37945.566632.EE] [Medline: 14742346] https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 15 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al 31. Phillips R, Schneider J, Molosankwe I, Leese M, Foroushani PS, Grime P, et al. Randomized controlled trial of computerized cognitive behavioural therapy for depressive symptoms: effectiveness and costs of a workplace intervention. Psychol Med 2014 Mar;44(4):741-752 [FREE Full text] [doi: 10.1017/S0033291713001323] [Medline: 23795621] 32. Klein JP, Berger T, Schröder J, Späth C, Meyer B, Caspar F, et al. Effects of a psychological internet intervention in the treatment of mild to moderate depressive symptoms: results of the EVIDENT study, a randomized controlled trial. Psychother Psychosom 2016;85(4):218-228 [FREE Full text] [doi: 10.1159/000445355] [Medline: 27230863] 33. Moritz S, Schilling L, Hauschildt M, Schröder J, Treszl A. A randomized controlled trial of internet-based therapy in depression. Behav Res Ther 2012 Aug;50(7-8):513-521. [doi: 10.1016/j.brat.2012.04.006] [Medline: 22677231] 34. Morriss R, Kaylor-Hughes C, Rawsthorne M, Coulson N, Simpson S, Guo B, et al. A Direct-to-public peer support program (big white wall) versus web-based information to aid the self-management of depression and anxiety: results and challenges of an automated randomized controlled trial. J Med Internet Res 2021 Apr 23;23(4):e23487 [FREE Full text] [doi: 10.2196/23487] [Medline: 33890858] Fleming T, Bavin L, Lucassen M, Stasiak K, Hopkins S, Merry S. Beyond the trial: systematic review of real-world uptake and engagement with digital self-help interventions for depression, low mood, or anxiety. J Med Internet Res 2018 Jun 06;20(6):e199 [FREE Full text] [doi: 10.2196/jmir.9275] [Medline: 29875089] 35. 36. Lehtimaki S, Martic J, Wahl B, Foster KT, Schwalbe N. Evidence on digital mental health interventions for adolescents and young people: systematic overview. JMIR Ment Health 2021 Apr 29;8(4):e25847 [FREE Full text] [doi: 10.2196/25847] [Medline: 33913817] 37. Martínez P, Rojas G, Martínez V, Lara MA, Pérez JC. Internet-based interventions for the prevention and treatment of depression in people living in developing countries: a systematic review. J Affect Disord 2018 Jul;234:193-200. [doi: 10.1016/j.jad.2018.02.079] [Medline: 29529553] Fu Z, Burger H, Arjadi R, Bockting CL. Effectiveness of digital psychological interventions for mental health problems in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Psychiatry 2020 Oct;7(10):851-864 [FREE Full text] [doi: 10.1016/S2215-0366(20)30256-X] [Medline: 32866459] 38. 39. Köhnen M, Dreier M, Seeralan T, Kriston L, Härter M, Baumeister H, et al. Evidence on technology-based psychological interventions in diagnosed depression: systematic review. JMIR Ment Health 2021 Feb 10;8(2):e21700 [FREE Full text] [doi: 10.2196/21700] [Medline: 33565981] 40. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001 41. Sep;16(9):606-613 [FREE Full text] [doi: 10.1046/j.1525-1497.2001.016009606.x] [Medline: 11556941] Spitzer RL, Kroenke K, Williams JB, Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006 May 22;166(10):1092-1097. [doi: 10.1001/archinte.166.10.1092] [Medline: 16717171] 42. Gupta SC, Anand R, Trivedi JK. Development of a suicidal intent questionnaire. Indian J Psychiatry 1983 Jan;25(1):57-62 [FREE Full text] [Medline: 21847253] Psychological Image Collection at Stirling (PICS). URL: http://pics.stir.ac.uk/ [accessed 2023-04-10] 43. 44. Christensen H, Griffiths KM, Korten A. Web-based cognitive behavior therapy: analysis of site usage and changes in depression and anxiety scores. J Med Internet Res 2002 Jan;4(1):e3 [FREE Full text] [doi: 10.2196/jmir.4.1.e3] [Medline: 11956035] 45. Keefe RS, Cañadas E, Farlow D, Etkin A. Digital intervention for cognitive deficits in major depression: a randomized controlled trial to assess efficacy and safety in adults. Am J Psychiatry 2022 Jul;179(7):482-489. [doi: 10.1176/appi.ajp.21020125] [Medline: 35410496] 46. Rollman BL, Herbeck Belnap B, Abebe KZ, Spring MB, Rotondi AJ, Rothenberger SD, et al. Effectiveness of online collaborative care for treating mood and anxiety disorders in primary care: a randomized clinical trial. JAMA Psychiatry 2018 Jan 01;75(1):56-64 [FREE Full text] [doi: 10.1001/jamapsychiatry.2017.3379] [Medline: 29117275] 47. Arean PA, Hallgren KA, Jordan JT, Gazzaley A, Atkins DC, Heagerty PJ, et al. The use and effectiveness of mobile apps for depression: results from a fully remote clinical trial. J Med Internet Res 2016 Dec 20;18(12):e330 [FREE Full text] [doi: 10.2196/jmir.6482] [Medline: 27998876] 48. Beck JS. Cognitive Behavior Therapy: Basics and Beyond. 2nd edition. New York, NY, USA: Guilford Press; Aug 17, 2011. 49. Andersson G, Carlbring P, Berger T, Almlöv J, Cuijpers P. What makes internet therapy work? Cogn Behav Ther 2009;38 Suppl 1:55-60. [doi: 10.1080/16506070902916400] [Medline: 19675956] 50. Meyer B, Berger T, Caspar F, Beevers CG, Andersson G, Weiss M. Effectiveness of a novel integrative online treatment for depression (Deprexis): randomized controlled trial. J Med Internet Res 2009 May 11;11(2):e15 [FREE Full text] [doi: 10.2196/jmir.1151] [Medline: 19632969] 51. Nordin S, Carlbring P, Cuijpers P, Andersson G. Expanding the limits of bibliotherapy for panic disorder: randomized trial of self-help without support but with a clear deadline. Behav Ther 2010 Sep;41(3):267-276. [doi: 10.1016/j.beth.2009.06.001] [Medline: 20569776] 52. Dandeneau SD, Baldwin MW. The buffering effects of rejection-inhibiting attentional training on social and performance threat among adult students. Contemp Educ Psychol 2009 Jan;34(1):42-50 [FREE Full text] [doi: 10.1016/j.cedpsych.2008.05.004] https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 16 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al 53. Orth U, Robins RW, Meier LL, Conger RD. Refining the vulnerability model of low self-esteem and depression: disentangling the effects of genuine self-esteem and narcissism. J Pers Soc Psychol 2016 Jan;110(1):133-149 [FREE Full text] [doi: 10.1037/pspp0000038] [Medline: 25915133] Sowislo JF, Orth U. Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies. Psychol Bull 2013 Jan;139(1):213-240. [doi: 10.1037/a0028931] [Medline: 22730921] 54. 55. Houston TK, Cooper LA, Ford DE. Internet support groups for depression: a 1-year prospective cohort study. Am J Psychiatry 2002 Dec;159(12):2062-2068. [doi: 10.1176/appi.ajp.159.12.2062] [Medline: 12450957] 56. Griffiths KM, Mackinnon AJ, Crisp DA, Christensen H, Bennett K, Farrer L. The effectiveness of an online support group for members of the community with depression: a randomised controlled trial. PLoS One 2012;7(12):e53244 [FREE Full text] [doi: 10.1371/journal.pone.0053244] [Medline: 23285271] 57. Tomasino KN, Lattie EG, Ho J, Palac HL, Kaiser SM, Mohr DC. Harnessing peer support in an online intervention for older adults with depression. Am J Geriatr Psychiatry 2017 Oct;25(10):1109-1119 [FREE Full text] [doi: 10.1016/j.jagp.2017.04.015] [Medline: 28571785] 58. Carlbring P, Maurin L, Törngren C, Linna E, Eriksson T, Sparthan E, et al. Individually-tailored, internet-based treatment for anxiety disorders: a randomized controlled trial. Behav Res Ther 2011 Jan;49(1):18-24. [doi: 10.1016/j.brat.2010.10.002] [Medline: 21047620] Johansson R, Sjöberg E, Sjögren M, Johnsson E, Carlbring P, Andersson T, et al. Tailored vs. standardized internet-based cognitive behavior therapy for depression and comorbid symptoms: a randomized controlled trial. PLoS One 2012;7(5):e36905 [FREE Full text] [doi: 10.1371/journal.pone.0036905] [Medline: 22615841] 59. 60. Weissman AN, Beck AT. Development and validation of the dysfunctional attitude scale: a preliminary investigation. In: Proceedings of the 62nd Annual Meeting of the American Educational Research Association. 1978 Mar Presented at: AERA '78; March 27-31, 1978; Toronto, Canada p. 1-33 URL: https://files.eric.ed.gov/fulltext/ED167619.pdf 61. Hollon SD, Kendall PC. Cognitive self-statements in depression: development of an automatic thoughts questionnaire. Cognit Ther Res 1980 Dec;4(4):383-395 [FREE Full text] [doi: 10.1007/bf01178214] 62. All India survey on higher education 2018-19. Department of Higher Education, India. 2021. URL: https://aishe.gov.in/ aishe/gotoAisheReports [accessed 2022-02-17] 63. The mobile gender gap report 2019. GSMA. 2021. URL: https://www.gsma.com/mobilefordevelopment/resources/ mobile-gender-gap-report-2019/ [accessed 2022-02-17] 64. Christensen H, Griffiths KM, Korten AE, Brittliffe K, Groves C. A comparison of changes in anxiety and depression symptoms of spontaneous users and trial participants of a cognitive behavior therapy website. J Med Internet Res 2004 Dec 22;6(4):e46 [FREE Full text] [doi: 10.2196/jmir.6.4.e46] [Medline: 15631970] 65. Christensen H, Griffiths KM, Mackinnon AJ, Brittliffe K. Online randomized controlled trial of brief and full cognitive behaviour therapy for depression. Psychol Med 2006 Dec;36(12):1737-1746. [doi: 10.1017/S0033291706008695] [Medline: 16938144] 66. Harrer M, Apolinário-Hagen J, Fritsche L, Salewski C, Zarski AC, Lehr D, et al. Effect of an internet- and app-based stress intervention compared to online psychoeducation in university students with depressive symptoms: results of a randomized controlled trial. Internet Interv 2021 Apr;24:100374 [FREE Full text] [doi: 10.1016/j.invent.2021.100374] [Medline: 33718001] 67. Twomey C, O'Reilly G, Byrne M, Bury M, White A, Kissane S, et al. A randomized controlled trial of the computerized CBT programme, MoodGYM, for public mental health service users waiting for interventions. Br J Clin Psychol 2014 Nov;53(4):433-450. [doi: 10.1111/bjc.12055] [Medline: 24831119] 68. Wright B, Tindall L, Littlewood E, Allgar V, Abeles P, Trépel D, et al. Computerised cognitive-behavioural therapy for depression in adolescents: feasibility results and 4-month outcomes of a UK randomised controlled trial. BMJ Open 2017 Jan 27;7(1):e012834 [FREE Full text] [doi: 10.1136/bmjopen-2016-012834] [Medline: 28132000] 69. Khaled SM. Prevalence and potential determinants of subthreshold and major depression in the general population of Qatar. J Affect Disord 2019 Jun 01;252:382-393. [doi: 10.1016/j.jad.2019.04.056] [Medline: 31003107] 70. Kroenke K, Strine TW, Spitzer RL, Williams JB, Berry JT, Mokdad AH. The PHQ-8 as a measure of current depression 71. in the general population. J Affect Disord 2009 Apr;114(1-3):163-173. [doi: 10.1016/j.jad.2008.06.026] [Medline: 18752852] Sato H, Kawahara J. Selective bias in retrospective self-reports of negative mood states. Anxiety Stress Coping 2011 Jul;24(4):359-367. [doi: 10.1080/10615806.2010.543132] [Medline: 21253957] 72. Lara MA, Tiburcio M, Aguilar Abrego A, Sánchez-Solís A. A four-year experience with a web-based self-help intervention for depressive symptoms in Mexico. Rev Panam Salud Publica 2014 May;35(5-6):399-406. [Medline: 25211568] 73. Morgan C, Mason E, Newby JM, Mahoney AE, Hobbs MJ, McAloon J, et al. The effectiveness of unguided internet cognitive behavioural therapy for mixed anxiety and depression. Internet Interv 2017 Oct 24;10:47-53 [FREE Full text] [doi: 10.1016/j.invent.2017.10.003] [Medline: 30135752] 74. March S, Batterham PJ, Rowe A, Donovan C, Calear AL, Spence SH. Trajectories of change in an open-access internet-based cognitive behavior program for childhood and adolescent anxiety: open trial. JMIR Ment Health 2021 Jun 18;8(6):e27981 [FREE Full text] [doi: 10.2196/27981] [Medline: 34142971] https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 17 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al 75. Oehler C, Scholze K, Reich H, Sander C, Hegerl U. Intervention use and symptom change with unguided internet-based cognitive behavioral therapy for depression during the COVID-19 pandemic: log data analysis of a convenience sample. JMIR Ment Health 2021 Jul 16;8(7):e28321 [FREE Full text] [doi: 10.2196/28321] [Medline: 34115604] 76. Detweiler Guarino I, Cowan DR, Fellows AM, Buckey JC. Use of a self-guided computerized cognitive behavioral tool during COVID-19: evaluation study. JMIR Form Res 2021 May 31;5(5):e26989 [FREE Full text] [doi: 10.2196/26989] [Medline: 33973856] 77. Breslow L, Pritchard DE, DeBoer J, Stump GS, Ho AD, Seaton DT. Studying learning in the worldwide classroom research into edX's first MOOC. Res Pract Assess 2013;8:13-25 [FREE Full text] 78. Reich J, Ruipérez-Valiente JA. The MOOC pivot. Sci Educ 2019 Jan 11;363(6423):130-131. [doi: 10.1126/science.aav7958] [Medline: 30630920] 79. Karyotaki E, Kleiboer A, Smit F, Turner DT, Pastor AM, Andersson G, et al. Predictors of treatment dropout in self-guided web-based interventions for depression: an 'individual patient data' meta-analysis. Psychol Med 2015 Oct;45(13):2717-2726 [FREE Full text] [doi: 10.1017/S0033291715000665] [Medline: 25881626] 81. 80. Clarke G, Kelleher C, Hornbrook M, Debar L, Dickerson J, Gullion C. Randomized effectiveness trial of an internet, pure self-help, cognitive behavioral intervention for depressive symptoms in young adults. Cogn Behav Ther 2009;38(4):222-234 [FREE Full text] [doi: 10.1080/16506070802675353] [Medline: 19440896] Pratap A, Renn BN, Volponi J, Mooney SD, Gazzaley A, Arean PA, et al. Using mobile apps to assess and treat depression in Hispanic and Latino populations: fully remote randomized clinical trial. J Med Internet Res 2018 Aug 09;20(8):e10130 [FREE Full text] [doi: 10.2196/10130] [Medline: 30093372] Schure MB, Lindow JC, Greist JH, Nakonezny PA, Bailey SJ, Bryan WL, et al. Use of a fully automated internet-based cognitive behavior therapy intervention in a community population of adults with depression symptoms: randomized controlled trial. J Med Internet Res 2019 Nov 18;21(11):e14754 [FREE Full text] [doi: 10.2196/14754] [Medline: 31738173] 82. 83. Cialdini RB. Influence: The Psychology of Persuasion. New York, NY, USA: William Morrow; Dec 26, 2006. 84. Garg K, Kumar CN, Chandra PS. Number of psychiatrists in India: baby steps forward, but a long way to go. Indian J 85. Psychiatry 2019 Jan;61(1):104-105 [FREE Full text] [doi: 10.4103/psychiatry.IndianJPsychiatry_7_18] [Medline: 30745666] Patel V, Xiao S, Chen H, Hanna F, Jotheeswaran AT, Luo D, et al. The magnitude of and health system responses to the mental health treatment gap in adults in India and China. Lancet 2016 Dec 17;388(10063):3074-3084. [doi: 10.1016/S0140-6736(16)00160-4] [Medline: 27209149] 86. Kanuri N, Arora P, Talluru S, Colaco B, Dutta R, Rawat A, et al. Examining the initial usability, acceptability and feasibility of a digital mental health intervention for college students in India. Int J Psychol 2020 Aug;55(4):657-673 [FREE Full text] [doi: 10.1002/ijop.12640] [Medline: 31867730] 87. Mehrotra S, Sudhir P, Rao G, Thirthalli J, Srikanth TK. Development and pilot testing of an internet-based self-help intervention for depression for Indian users. Behav Sci (Basel) 2018 Mar 22;8(4):36 [FREE Full text] [doi: 10.3390/bs8040036] [Medline: 29565278] 88. Rodriguez-Villa E, Rozatkar AR, Kumar M, Patel V, Bondre A, Naik SS, et al. Cross cultural and global uses of a digital mental health app: results of focus groups with clinicians, patients and family members in India and the United States. Glob Ment Health (Camb) 2021 Aug 24;8:e30 [FREE Full text] [doi: 10.1017/gmh.2021.28] [Medline: 34512999] Srivastava P, Mehta M, Sagar R, Ambekar A. Smartteen- a computer assisted cognitive behavior therapy for Indian adolescents with depression- a pilot study. Asian J Psychiatr 2020 Apr;50:101970. [doi: 10.1016/j.ajp.2020.101970] [Medline: 32114331] 89. Abbreviations CBT: cognitive behavioral therapy cCBT: computerized cognitive behavioral therapy GAD-7: Generalized Anxiety Disorder-7 LMICs: low- and middle-income countries PHQ-9: Patient Health Questionnaire-9 RCT: randomized controlled trial https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 18 (page number not for citation purposes) JOURNAL OF MEDICAL INTERNET RESEARCH Ghosh et al Edited by A Mavragani; submitted 12.07.22; peer-reviewed by D Ramanathan, Q Liao, JC Buckey; comments to author 01.11.22; revised version received 23.02.23; accepted 08.03.23; published 26.04.23 Please cite as: Ghosh A, Cherian RJ, Wagle S, Sharma P, Kannan KR, Bajpai A, Gupta N An Unguided, Computerized Cognitive Behavioral Therapy Intervention (TreadWill) in a Lower Middle-Income Country: Pragmatic Randomized Controlled Trial J Med Internet Res 2023;25:e41005 URL: https://www.jmir.org/2023/1/e41005 doi: 10.2196/41005 PMID: 37099376 ©Arka Ghosh, Rithwik J Cherian, Surbhit Wagle, Parth Sharma, Karthikeyan R Kannan, Alok Bajpai, Nitin Gupta. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 26.04.2023. This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on https://www.jmir.org/, as well as this copyright and license information must be included. https://www.jmir.org/2023/1/e41005 XSL•FO RenderX J Med Internet Res 2023 \| vol. 25 \| e41005 \| p. 19 (page number not for citation purposes)
10.3390_ijerph17124311	Article Eﬀect of Trade Openness on Food Security in the EU: A Dynamic Panel Analysis Giulio Fusco , Benedetta Coluccia and Federica De Leo Department of Economics and Management, University of Salento, 73100 Lecce LE, Italy; [email protected] * Correspondence: [email protected] (G.F.); [email protected] (B.C.) Received: 22 May 2020; Accepted: 12 June 2020; Published: 16 June 2020 Abstract: The problem of food insecurity is growing across the world, including economically developed countries. In Europe, the question is not just about the total supply of foods, but it includes even the accessibility of prices and their nutritional and qualitative adequacy. In this context many countries recognize the importance of trade policies to ensure adequate levels of food security. The aim of this work was to analyze the impact of trade openness on the level of food security in European countries, using a dynamic panel analysis with the generalized method of moments (GMM) approach. We selected two diﬀerent indicators of food security (average protein supply, average dietary energy supply adequacy) capable of oﬀering information both on the quantity and on the nutritional quality of the food supply. In order to improve the robustness of the empirical results, we developed three diﬀerent regressions, with three trade openness indicators (trade openness, tariﬀ, globalization) for each food security indicator. The results showed that commercial opening has, on average, a statistically signiﬁcant net positive impact on the food security of European countries. Additional results indicate that also economic development, together with the importance of the agricultural sector, can improve food security levels. Keywords: food security; trade openness; Common Agricultural Policy (CAP); European countries; dynamic panel 1. Introduction Problems of food insecurity are growing across the world, including economically developed countries [1]. In Europe, around half a million people do not have regular and suﬃcient access to food and about 20 million families cannot regularly aﬀord high quality meals (i.e., ﬁsh, meat, or their vegetarian equivalent) [2,3]. These numbers are expected to rise due to the Covid-19 pandemic. According to the Food and Agriculture Organization (FAO) deﬁnition [4], food security is referred to as a condition in which all the people, at all times, have physical, social and economic access to suﬃcient, safe, and nutritious food to meet their dietary needs and food preferences for an active and healthy life. This phenomenon has historically been a challenge conﬁned primarily to the developing world [5], where, due to food insecurity, millions of people are still suﬀering from chronic undernourishment. In industrialized countries, food security has a wider dimension and includes economic access to food that people want to eat, without compromising other needs such as rent, fuel, debt repayments, etc. [6,7]. Therefore, even in Europe, concerns over food security have emerged due to the growing volatility of food prices since 2008, which was not followed by political measures aimed at adjusting wages and maintaining social welfare [8]. In this direction, the European problem is not just about the limited total supply of foods, but also the inaccessibility of prices, which makes some lack of food. However, global food reserves have also decreased signiﬁcantly compared to the past, due to the continuous increase in Int. J. Environ. Res. Public Health 2020, 17, 4311; doi:10.3390/ijerph17124311 www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2020, 17, 4311 2 of 13 population [9,10]. In addition to economic reasons, natural disasters (i.e., ﬂoods, droughts, earthquakes) have a negative impact on the stability of agricultural production, endangering the livelihood and food supply of millions of people around the world [11]. Over the past 50 years, the frequency and intensity of natural disasters associated with climate change have grown signiﬁcantly. In particular, data show that the economic impact of a natural disaster on the agricultural sector has increased by about 20 times [12]. Moreover, the FAO [13] deﬁnition outlines the connection between food security and food safety, requiring that available and accessible foods are also healthy and nutritionally adequate. According to Carvalho [14], food safety is an indefectible component of food security, as the prevention and control of diseases of plants and animals contribute to favoring the constant availability of agricultural products over time. Attention to hygiene and health standards, moreover, could open up international commercial outlets and could favor the ability to purchase food, making it less expensive [15]. However, it should be noted that the sanitary and phytosanitary rules can also represent an obstacle to exports, especially for some less advanced countries, with a negative impact on their food security [16]. In this context, we should note that food security indirectly depends on the economic, sanitary, social, and political system, and could have direct consequences for human health. In this context economic globalization has a fundamental role in order to generate imbalances of wealth between countries; in fact the globalization process eliminated the boundaries to a large extent among countries and has raised the integration of economies in terms of goods, services, and capital ﬂows, which are key factors in order to measure the food security level [17]. Therefore, it is one of the main objectives of the European Union, which counts it among the principal purposes of the Common Agricultural Policy (CAP). This policy provides economic aid to farmers to improve agricultural productivity and it ensures stable and inexpensive food supply [18]. In particular, it provides income support through direct payments to farmers who implement environmentally friendly agriculture, adopt market measures to cope with diﬃcult economic times, and implement rural development measures. Moreover, in recent times, in order to achieve adequate levels of food security, many countries have recognized the importance of trade policies, developing reforms to reduce taxes on incoming goods and contributing to the growth of the international market aimed at eradicating poverty and improving the availability of food [19,20]. In particular, trade openness plays a crucial role in ensuring the continuity of supply, as it allows to produce products in the most suitable areas and to move them to countries with insuﬃcient food supplies. In this way, supply and demand are smoothed out, price ﬂuctuations are reduced, and each country can increase the quantity and the variety of products available to the national population, guaranteeing a good level of food security [21,22]. Besides, through imports, each country can decide to obtain the food resources that it needs at a lower cost than it would sustain by producing it domestically [21]. Moreover, according to Wacziarg and Welch [23], trade openness allows access to larger markets which give the opportunity to beneﬁt from economies of scale, technological transfers, and knowledge spillovers. Despite the relevance of the topic, studies analyzing the link between trade openness and food security are lacking and many of them have focused on developing countries, where hunger still persists [21,24,25]. An additional problem comes from the fact that previous studies largely used poverty indicators instead of direct food security indicators. On the contrary, this study is part of an emerging literature that examines the problem of food security in economically developed countries such as Italy [26], Australia [5], United Kingdom [27], Ireland [28], and Canada [29]. Moreover, it contributes to another strand of literature on the analogies and diﬀerences between the European countries [30]. However, we considered the problem of food security in the entire European context for the ﬁrst time, evaluating the impact of trade policies on both quantity and quality of supplies. In particular, the aim of this work was to analyze the impact of trade openness on the level of food security in European countries (see Appendix A). Using two diﬀerent indicators of food security, a dynamic panel analysis was adopted, which is suitable for assessing the eﬀects of a long-term Int. J. Environ. Res. Public Health 2020, 17, 4311 3 of 13 policy reform [21,31]. A dynamic panel analysis with the generalized method of moments (GMM) approach was employed to the account for unobserved heterogeneity and potential endogeneity of the explanatory variables. This approach enables us to take a broader perspective and to focus on the overall level of food security of the European population. The paper is organized as follows: Section 2 explains the methodology applied in the empirical analysis; Section 3 describes the data used for the study; Section 4 presents and discusses the achieved results; Section 5 presents the main conclusions and the policy implications. 2. Materials and Methods The aim of this work was to analyze the impact of trade openness on the level of food security in European countries. According to previous studies [21,32], in this work we adopt a dynamic model approach to examine the eﬀects of trade openness on food security for a cross-section of countries. In the analysis of the economic aspects, it becomes fundamental to analyze the dynamic aspect of the phenomenon, as the eﬀects of economic policies could only be evident with the passage of time. According to previous studies, the suitable methodology for analyzing the dynamic eﬀect of a phenomenon of time is the “dynamic panel”. One simple way of allowing dynamic eﬀects in panel data models is including a lagged dependent variable. It is well known that the introduction of the lagged dependent variable will generally mean that standard estimators are inconsistent. Consistent estimators can be found using the GMM estimator proposed by Arellano and Bond (1991). In the following lines all the passages are described. This particular methodology allows to capture the dynamic aspects of the commercial reforms and to face the problem of the potential endogeneity that could derive from this speciﬁcation. In fact, the continuous evolution of economic processes means that the eﬀect of economic and trade policies is completely evident only in the long run [33]. Therefore, the dynamic model allows one to consider the eﬀects of explanatory variables over time. Therefore, we considered current food safety levels as a function of previous levels and we built the following regression models: FSi,t = a + βFSi,t−1 + γTOi.t + δCVi,t + µi + λt + εi,ti = 1, . . . 33, t = 1, . . . 18 (1) where FS, which stands for food security, was our dependent variable that indicates the level of food security, in our analysis we used two diﬀerent variables i.e., average protein supply and average dietary energy supply adequacy. Where average protein supply indicates national average protein expressed in grams per capita per day, while dietary energy supply adequacy is a percentage of the average dietary energy requirement in each country. We decided to use these indicators because they both oﬀer a quantitative information on the caloric energy input of foods available for human consumption and a qualitative information on the nutritional value of foods, since the protein component represents the major macronutrient group [24,34]. TO, which stands for trade openness, was an independent variable that indicates the level of trade openness in each country. In this study, in order to test the robustness of the result, we chose to use three diﬀerent indicators: trade openness [35], tariﬀ [36], and globalization [37]. CV is a set of control variables used to determine the potential level of food safety in each country. Finally, µi, e, and λt are respectively countries ﬁxed eﬀects and time ﬁxed eﬀects, while εi,t is the error term. The use of the delayed dependent variable in the model causes the phenomenon called “dynamic panel bias” [38], because the lagged dependent variable is endogenous to the ﬁxed eﬀects in the error term, which leads to estimation problems. Normally, this estimation problem cannot be eliminated with ﬁxed or random eﬀects regressions, and the estimation with the ordinary least squares (OLS) method is distorted, because the lagged dependent variable is correlated with the error term εi,t. The common approach to dealing with non-stationary data is to apply the diﬀerence operator in order to achieve a dynamic speciﬁcation in raw diﬀerences. ∆FSi,t = a + β∆FSi,t−1 + γ∆TOi.t + δ∆CVi,t + µi + λt + εi,ti = 1, . . . 33, t = 1, . . . 18 (2) Int. J. Environ. Res. Public Health 2020, 17, 4311 4 of 13 However, this approach is capable of removing the potential distortion, as it eliminated individual eﬀect, because it doesn’t remove the temporal eﬀect. In order to solve this problem Holtz-Eakin et al. in 1988 [39] and Arellano and Bond in 1991 [40] developed an estimator for linear dynamic panel data models, called the generalized method of moments (GMM). Despite the superiority of the diﬀerence-GMM (ﬁrst order condition and GMM) estimator over the simpler panel data estimations, if the series are very repeated the lagged levels have been demonstrated to be ineﬀective tools for ﬁrst-diﬀerences [41]. Then the performance of the diﬀerence-GMM estimator can be distorted for the small sample [42]. According to Arellano-Bover [40] and Blundell-Bond [43] the estimator performance can be increased by adding the original equation in levels to the system, which is known as the “system-GMM”. The peculiarity of the system-GMM estimator is that it weighs the moments in inverse proportion respecting their variances and covariances, for this reason, it reduces the weight in the estimation process of the instruments highly correlated. 3. Data Description We used panel data composed by the European countries over the period 2000–2017 for the dependent variable (average dietary energy supply adequacy). Regarding the dependent variable, average protein supply, we analyze the period 2000–2012 due to a lack of data for the remaining years. The variables were selected based on the FAO [44] and through the analysis of the previous empirical literature [18,42,43]. Most part of the data used in this study can be extracted from world development indicators and Food and Agriculture Organization Corporate Statistical Database (FAOSTAT). Moreover, according to Dithmer and Abdulai [21], we considered four groups of food security determinants: the ﬁrst group describes the general context of the country; the second group captures the economic and demographic development; in the third group there were control variables that measure domestic macroeconomic policies and conditions; ﬁnally in the last group we considered non-economic events such as natural disasters. In regards the ﬁrst group, we took into consideration the total amount of economic resources, the availability of resources for agricultural production, and the importance of agriculture [45]. In particular, we used the gross domestic product (GDP) per capita as the principle variable to measure the quantity of ﬁnal goods and services on the territory of a country. Rural population shares the variable which indicates the importance of agriculture and refers to the share of people living in rural areas out of the total population. The availability of resources for agricultural production is measured by the arable land variable, which includes land under temporary crops, temporary meadows for mowing or for pasture, land under market or kitchen gardens, and land temporarily fallow. With regard to the second group of food security determinants, the model included three diﬀerent variables that capture the agricultural, economic, and demographic development. In particular, in order to capture the agricultural development, we use Cereal yield (kg per hectare; FAOSTAT), as a proxy for agricultural productivity. The economic development is measured by the gross domestic product (GDP) growth rate per capita variable. Finally, the population growth variable captures the demographic development. In regards the third group, the inﬂation variable, measured by the consumer price index inﬂation rate, expresses the domestic macroeconomic policy quality; in particular, according to Loayza et al. [46], high inﬂation being associated with bad macroeconomic policies. In the four group we used natural disaster variables; this value indicates the intensity of natural disasters and it is computed through a ratio between the number of populations aﬀected by natural disasters and the total population for each country. Finally, regarding the last group, we selected three diﬀerent variables of trade openness, in order to test the research question. Int. J. Environ. Res. Public Health 2020, 17, 4311 5 of 13 The trade openness variable, according to Heston et al. [35], is a ratio between trade (real export and import) and GDP. The second variable was tariﬀ [36], which indicates ad-valorem tariﬀ, measured as import duties. Globalization [37] was the last variable, whereby we used the KOF (KOF is an acronym for the German word “Konjunkturforschungsstelle”, meaning: “economic cycle research institute”.) globalization index (0–100); The KOF index attempts to measure the degree to which a nation exchanges goods, capital, people, ideas, and information. It is a composite index that uses three dimensions: economic, social, and political, where a value close to 100 indicates a high level of globalization. Tables 1 and 2 present the variables, source of data, and their summary statistics. Table 1. Variables and data sources (FAOSTAT stands for Food and Agriculture Organization Corporate Statistical Database; GDP stands for gross domestic product; EM-DAT stands for Emergency Events Database). Variables Average protein supply Average dietary energy supply adequacy Trade openness Tariﬀ Globalization GDP per capita GDP growth Arable land Agricultural Production Rural population Population growth Natural disaster Inﬂation Unit (g/cap/day) Percentage % ◦ N ◦ N ◦ N US$ Percentage % (hectares per person) (kg per hectare) Percentage % Annual Percentage % ◦ N Annual Percentage % Data Source Time Period FAOSTAT FAOSTAT World Development Indicators World Development Indicators Swiss Economic Institute World Development Indicators World Development Indicators FAOSTAT World Development Indicators World Development Indicators World Development Indicators EM-DAT World Development Indicators 2000–2012 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 2000–2017 Table 2. Summary Statistics. Variables Mean Average protein supply Average dietary energy supply adequacy Trade openness Tariﬀ Globalization GDP per capita GDP growth Arable land Agricultural Production Rural population Population growth Natural Disaster Inﬂation 97.780 129.304% 1.009 3.20 78.6 27,028.053 2.734% 8,741,911.189 2,227,506,805 30.08% 0.225% 0.0049 4.598% Standard Deviation 12.399 11.546% 0.503 1.777 9.382 2,3952.970 3.078% 21,322,651.81 3,535,306,006 13.053% 0.802% 0.0224 10.085% Maximum Minimum 118 158% 3.785 11.9 91.313 118,823.648 25.162% 124,374,000 21,419,375,209 58.259% 2.890% 0.442 168.620% 67 101% 0.408 1.22 47.509 354.003 −14.758% 60,000 1,010,275 2.039% −3.847% 0 −4.478% As previously mentioned, the aim of this study was to estimate the impact of the trade openness on food security through a dynamic panel model approach. According to prior studies [21,31], we used the econometric structure described in the Equation (1). In order to improve the robustness of the empirical Int. J. Environ. Res. Public Health 2020, 17, 4311 6 of 13 results developed three regressions, with three diﬀerent trade openness indicators (trade openness, tariﬀ, and globalization). FSi,t = a + βFSi,t−1 +γTradeOpennessi.t + δLnGDPpercapitai,t +ϑGDPgrowthi,t + θLnArablelandi,t +πlnAgriculturalproductioni,t + ρRuralpopi,t +ϕLnPopgrowthi,t + τNaturaldisasteri,t + ωLnInﬂationi,t + µi + λt + εi,t FSi,t = a + βFSi,t−1 +γTari f fi.t + δLnGDPpercapitai,t + ϑGDPgrowthi,t +θLnArablelandi,t+πlnAgriculturalproductioni,t +ρRuralpopi,t + ϕLnPopgrowthi,t +τNaturaldisasteri,t + ωLnInﬂationi,t + µi + λt + εi,t FSi,t = a + βFSi,t−1 +γGlobalizationi.t + δLnGDPpercapitai,t +ϑGDPgrowthi,t + θLnArablelandi,t +πlnAgriculturalproductioni,t + ρRuralpopi,t +ϕLnPopgrowthi,t + τNaturaldisasteri,t +ωLnInﬂationi,t + µi + λt + εi,t (3) (4) (5) 4. Results and Discussions In Europe, the levels of food security, assessed through the average dietary energy supply adequacy and the average protein supply variables, were lower in Eastern area (Figure 1). In particular, from an initial exploratory analysis, it emerged that the countries characterized by greater economic development were also characterized by a certain stability in the food supply. Figure 1. Geographical distribution of food security level in European countries. Considering the possibility of collinearity in the model, we computed a correlation analysis between the independent variables. The correlation results are summarized in Table 3 [47]. From the results of the correlation analysis, we can aﬃrm that there was no collinearity in the model, and for this reason we could preserve the regression model. Table 4 shows the results of the three separate dynamic panel model regressions. The values in the table are coeﬃcients, their p-value, the standard errors (in parentheses), and summary statistics. 𝐹𝑆(cid:3036),(cid:3047)=𝑎+𝛽𝐹𝑆(cid:3036),(cid:3047)(cid:2879)(cid:2869)+𝛾𝑇𝑟𝑎𝑑𝑒𝑂𝑝𝑒𝑛𝑛𝑒𝑠𝑠(cid:3036).(cid:3047)+𝛿𝐿𝑛𝐺𝐷𝑃𝑝𝑒𝑟𝑐𝑎𝑝𝑖𝑡𝑎(cid:3036),(cid:3047)+ϑ𝐺𝐷𝑃𝑔𝑟𝑜𝑤𝑡ℎ(cid:3036),(cid:3047)+θ𝐿𝑛𝐴𝑟𝑎𝑏𝑙𝑒𝑙𝑎𝑛𝑑(cid:3036),(cid:3047)+π𝑙𝑛𝐴𝑔𝑟𝑖𝑐𝑢𝑙𝑡𝑢𝑟𝑎𝑙𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛(cid:3036),(cid:3047)+ρRuralpop(cid:3036),(cid:3047)+φLnPopgrowth(cid:3036),(cid:3047) +τ𝑁𝑎𝑡𝑢𝑟𝑎𝑙𝑑𝑖𝑠𝑎𝑠𝑡𝑒𝑟(cid:3036),(cid:3047)+ ωLnInflation(cid:3036),(cid:3047)+ μ(cid:2919)+λ(cid:2930)+ε(cid:2919),(cid:2930) (3) 𝐹𝑆(cid:3036),(cid:3047)=𝑎+𝛽𝐹𝑆(cid:3036),(cid:3047)(cid:2879)(cid:2869)+𝛾𝑇𝑎𝑟𝑖𝑓𝑓(cid:3036).(cid:3047)+𝛿𝐿𝑛𝐺𝐷𝑃𝑝𝑒𝑟𝑐𝑎𝑝𝑖𝑡𝑎(cid:3036),(cid:3047)+ϑ𝐺𝐷𝑃𝑔𝑟𝑜𝑤𝑡ℎ(cid:3036),(cid:3047)+θ𝐿𝑛𝐴𝑟𝑎𝑏𝑙𝑒𝑙𝑎𝑛𝑑(cid:3036),(cid:3047)+π𝑙𝑛𝐴𝑔𝑟𝑖𝑐𝑢𝑙𝑡𝑢𝑟𝑎𝑙𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛(cid:3036),(cid:3047)+ρRuralpop(cid:3036),(cid:3047)+φLnPopgrowth(cid:3036),(cid:3047) +τ𝑁𝑎𝑡𝑢𝑟𝑎𝑙𝑑𝑖𝑠𝑎𝑠𝑡𝑒𝑟(cid:3036),(cid:3047)+ωLnInflation(cid:3036),(cid:3047)+ μ(cid:2919)+λ(cid:2930)+ε(cid:2919),(cid:2930) (4) 𝐹𝑆(cid:3036),(cid:3047)=𝑎+𝛽𝐹𝑆(cid:3036),(cid:3047)(cid:2879)(cid:2869)+𝛾𝐺𝑙𝑜𝑏𝑎𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛(cid:3036).(cid:3047)+𝛿𝐿𝑛𝐺𝐷𝑃𝑝𝑒𝑟𝑐𝑎𝑝𝑖𝑡𝑎(cid:3036),(cid:3047)+ϑ𝐺𝐷𝑃𝑔𝑟𝑜𝑤𝑡ℎ(cid:3036),(cid:3047)+θ𝐿𝑛𝐴𝑟𝑎𝑏𝑙𝑒𝑙𝑎𝑛𝑑(cid:3036),(cid:3047)+π𝑙𝑛𝐴𝑔𝑟𝑖𝑐𝑢𝑙𝑡𝑢𝑟𝑎𝑙𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑖𝑜𝑛(cid:3036),(cid:3047)+ρRuralpop(cid:3036),(cid:3047)+φLnPopgrowth(cid:3036),(cid:3047) +τ𝑁𝑎𝑡𝑢𝑟𝑎𝑙𝑑𝑖𝑠𝑎𝑠𝑡𝑒𝑟(cid:3036),(cid:3047)+ωLnInflation(cid:3036),(cid:3047)+ μ(cid:2919)+λ(cid:2930)+ε(cid:2919),(cid:2930) (5) 4. Results and discussions In Europe, the levels of food security, assessed through the average dietary energy supply adequacy and the average protein supply variables, were lower in Eastern area (Figure 1). In particular, from an initial exploratory analysis, it emerged that the countries characterized by greater economic development were also characterized by a certain stability in the food supply. Figure 1. Geographical distribution of food security level in European countries. Considering the possibility of collinearity in the model, we computed a correlation analysis between the independent variables. The correlation results are summarized in Table 3 [47]. From the results of the correlation analysis, we can affirm that there was no collinearity in the model, and for this reason we could preserve the regression model. Table 4 shows the results of the three separate dynamic panel model regressions. The values in the table are coefficients, their p-value, the standard errors (in parentheses), and summary statistics.Int. J. Environ. Res. Public Health 2020, 17, 4311 7 of 13 Table 3. Correlation analysis. Glob GDP Percap GDP Growth Arable Land Agr. Prod. 0.167 −0.579 1 0.393 −0.276 0.708 1 0.119 −0.248 −0.335 −0.203 1 −0.271 0.497 −0.226 −0.213 0.054 1 −0.365 0.089 0.007 −0.090 −0.067 0.612 1 Tariﬀ 0.116 1 Variables Trade Op. 1 Trade openness Tariﬀ Glob GDPpercap GDPgrowth Arable land Agr.Prod. Rur.pop. Pop.growth Nat.disaster Inﬂation Rur. Pop. −0.139 0.166 −0.624 −0.601 0.192 −0.065 −0.123 1 Pop. Growth Nat. Dis. 0.241 −0.205 0.422 0.637 −0.073 −0.079 0.030 −0.399 1 0.001 −0.121 −0.168 −0.080 −0.002 −0.023 −0.040 0.133 −0.061 1 Inf. −0.035 0.225 −0.462 −0.274 0.102 0.190 0.077 0.093 −0.111 0.21 1 Note: Trade op. stands for trade openness; Glob stands for globalization; GDPpercap stands for gross domestic product per capita; GDPgrowth stands for gross domestic product growth; Agr.Prod. stands for agricultural production; Rur.pop. stands for rural population; Nat.dis. stands for natural disasters; Inf. stands for inﬂation. Int. J. Environ. Res. Public Health 2020, 17, 4311 8 of 13 Variables Av. Protein Supply (1) Av. Protein Supply (2) Av. Protein Supply (3) Av. Dietary Energy Supply Adequacy (4) Av. Dietary Energy Supply Adequacy (5) Av. Dietary Energy Supply Adequacy (6) Table 4. Dynamic panel model regressions. Av. Protein supply adequacy t-1 Av. Dietary energy supply adequacy t-1 Trade openness Tariﬀ Globalization Ln GDP per capita GDP growth Ln Arable land Ln Agricultural production Rural Population Ln Population growth Natural Disaster Ln Inﬂation AR (1) errors test AR (2) errors test 0.06774250 * (0.0163096) 0.0969006 * (0.0146225) 0.0779083 * (0.0144783) 0.0754607 * (0.0142629) 0.269481 * (0.00748153) 0.00088624 (0.00219032) 0.113109 * (0.00647071) −0.0204668 * (0.00516753) 0.00766082 * (0.000398240) −0.00294469 (0.00394025) 2.40071 (2.30896) 0.0660458 * (0.00784140) −1.62376 (0.1044) 1.51318 (0.1302) −0.0123728 * (0.00362975) 0.281922 * (0.00534777) 0.000487233 (0.00200030) 0.0888971 * (0.00604065) −0.0144307 (0.00595875) 0.00742802 * (0.000503302) −0.00180546 (0.00359173) 2.15373 (2.03266) 0.0747960 * (0.00900465) −1.60499 (0.1085) 1.53077 (0.1258) −0.000958626 (0.00143931) 0.298147 * (0.0121872) 0.00137128 (0.00202637) 0.0952195 * (0.00531122) −0.0180365 * (0.00579867) 0.00766486 *** (0.000500103) −0.00586717 * (0.00349110) 3.27440 (2.55524) 0.0798373 * (0.00939306) −1.65661 (0.0976) 1.57954 (0.1142) 0.944329 * (0.0123442) 0.293482 * (0.104098) 0.213976 * (0.0657183) 0.0104900 (0.0102529) 0.210299 * (0.0701103) 0.0788502 (0.0385403) 0.0131988 * (0.00343404) 0.0664214 (0.0460942) −6.90303 * (2.64627) −0.0112384 (0.0376728) −3.61083 (0.0003) −0.890439 (0.3732) 0.934857 * (0.0126720) 0.941284* (0.0122497) −0.0742521 (0.0325249) 0.379137 * (0.0855967) 0.0160616 (0.0107119) 0.158903 * (0.0597781) 0.0956332 (0.0416021) 0.0137986 * (0.00355062) 0.0626977 (0.0444805) −6.55824 * (2.75441) 0.00051853 (0.0402414) −3.29418 (0.0010) −1.19194 (0.2333) 0.00528695 (0.0113695) 0.290284 (0.136305) 0.104712 (0.0304211) 0.156196 * (0.0598957) 0.0896892 (0.0396425) 0.0138899 * (0.00351928) 0.476948 (0.103866) −6.48437 ** (2.74300) 0.00852784 (0.0394147) −3.59656 (0.0003) −0.890439 (0.3732) Note: , , ** stands for 10%, 5%, and 1% signiﬁcant level, respectively. Av. stands for average; Ln stands for natural logarithm; AR stands for autoregressive. Int. J. Environ. Res. Public Health 2020, 17, 4311 9 of 13 The ﬁrst column of Table 4 shows the relationship between our dependent variable (average protein supply) and the independent variables. In the ﬁrst model we used Trade Openness (TO) as a variable to represent the level of commercial openness for each country. The lagged dependent variable was strongly signiﬁcant, indicating that the level of food security changes only slowly over time and it depends on the past levels. This result also justiﬁes the dynamic model speciﬁcation and the employment of the system-GMM approach. From the results, we can aﬃrm, according to previous studies [21,45], that when the volume of trade increases, a country’s food security level has prospects for improvement. According to previous studies, the increase of food supply should generate a reduction in consumer prices, facilitating the purchase of food products, in particular for developed countries [44]. In addition, our empirical results highlighted the importance of the economic aspect on food security level; the coeﬃcient between our dependent variable (average protein supply) and the independent variable (GDP per capita) was positive and signiﬁcant. Therefore, in countries with higher incomes, citizens have access to good quality food [48]. However, this value should not be understood simply as purchasing power, but also as the ability to adopt better technologies and to improve the level of food security [49]. The impact of the primary sector on food security was conﬁrmed by our results, in fact, the relationship between average protein supply and arable land was positive and signiﬁcant, and the same result was valid for the independent variable of rural population. In particular, the coeﬃcient of arable land captures an important aspect of domestic resource endowments, and its value indicates that the households with larger arable land having a higher level of production are more likely to be food secure [49]. However, one of the weak points of the European agricultural sector is represented by the growing impermeabilization and the constant loss of soil productivity. It has been estimated that 18% of the cultivated land undergoes a decrease in productivity every year and that urbanized areas have grown by 78% in the last 50 years, increasingly limiting areas for cultivation [50]. The phenomenon is particularly felt in the area overlooking the Black Sea, but the Mediterranean regions are also very aﬀected, due to intensive land use and expansion of urbanized areas. In the long run, this phenomenon could negatively aﬀect the food security levels [51]. Finally, the value of the coeﬃcient of the inﬂation variable was positive and signiﬁcant, showing the importance of the macroeconomic policies in ensuring good levels of food security. The column labelled (2) and (3) shows the results of the additional analysis to assess the robustness of our empirical results, in fact, we use alternative trade openness measures as tariﬀ and globalization. As shown in column (2), the ad-valorem tariﬀ measure was signiﬁcantly negatively related to average protein supply, implying that trade restrictions, on average, have detrimental eﬀects on overall food security level. The relationship with the other independent variables was the same of the column (1). Column (3) shows the results with the globalization variable, demonstrating that the relationship between globalization and the average protein supply variable was not signiﬁcant between the European countries. The column labelled (4) of Table 4 shows the analysis carried out with an alternative food security indicator, in particular it describes the relationship between the dependent variable (average dietary energy supply adequacy) and the other selected independent variables. In column (4) we used trade openness as the variable that expresses the level of commercial openness. Also in this case, the lagged dependent variable was strongly signiﬁcant. From an empirical point of view, according to previous studies [21,44], when the volume of trade increases, in general thanks to a trade liberalization policy, food security levels also have the potential to improve. In fact, policies aimed at encouraging trade allow countries to access the global market, increasing the overall quantity of food and raw materials for agriculture [20]. Indeed, each state can use export earnings to import at an aﬀordable price all those goods whose domestic production is scarce or too expensive [23]. Furthermore, according to Jaﬀe et al. [52], commercial opening is beneﬁcial to food security levels because it can positively aﬀect the employment of citizens, especially in the case of the least developed countries. The latter, thanks to Int. J. Environ. Res. Public Health 2020, 17, 4311 10 of 13 the trade openness, can import products made with the relatively abundant factor and with low-skilled labor, thereby creating employment opportunities and raising workers’ incomes [48]. In addition, this alternative indicator also conﬁrms the importance of the economic and agricultural aspects to guarantee good levels of food security. Indeed, the relationship between the dependent variable (average Energy Supply Adequacy) and the dependent variables (GDP pro capita, Arable Land, Rural Population, and Coeﬃcient of Agricultural production) was positive and signiﬁcant. In addition, the empirical result conﬁrms the importance of the non-economic evidence for the food security levels: the relationship between the average energy supply adequacy and the natural disaster independent variable was positive and signiﬁcant, in particular, when the number of natural disaster decreases, the level of food security increases. Moreover, from the results showed in the column labeled (5), we can conﬁrm that the decrease in customs duties has a positive impact on improving food security levels. Finally, as shown in the column (6), the relationship between globalization and our food security indicator was not signiﬁcant in European countries. 5. Conclusions and Policy Implications In the present study, through the use of a dynamic modeling approach, we have shown that commercial opening has, on average, a statistically signiﬁcant net positive impact on the food security of European countries both from an energy and nutritional point of view. This implies that commercial openness, in an economically advanced context, can have a positive impact both on security of supply, but also on the nutritional quality of the same, demonstrating the eﬀectiveness of the commercial model proposed by the European Union, where the food sector represents a key resource from an economic, social and cultural point of view. In addition, our analysis conﬁrmed that the most resilient countries are those characterized by higher per capita incomes. Furthermore, the results showed that economic development, together with the importance of the agricultural sector in terms of production level, extension of agricultural land, and percentage of inhabitants in rural areas, have a signiﬁcant positive impact. From the results of the analysis it is evident the eﬀect of the single market established by the European Union (EU), in fact, the EU trade reform has been producing advantages in terms of increasing the level of food security. Moreover, the policies implemented by the EU on the agriculture ﬁeld can also have an important impact on the quantity and quality of food supplies. The last reform was the CAP (2014–2020), which aims to help farmers to produce enough food for Europe, ensuring safety and quality at aﬀordable prices. The objectives of the CAP aim to ensure a fair standard of living for farmers, protecting them from excessive volatility of prices, market crises, and imbalances within the food supply chain [18,53]. However, in order to continue guaranteeing adequate quantities of food in the long run, considering the continuous growth of the European population and the continuous pressures that agriculture exerts on the environment, it is necessary to act by promoting sustainable agricultural practices, that meet the current demand for food without compromising the ability of future generations to meet their needs [54]. Moreover, based on the results obtained from this study, in order to ensure quantity and quality of food supplies, it would be desirable for the European Union to adopt a liberal trade policy, which should represent a complement and not a substitute for domestic development policies. In addition, our analysis revealed that natural disasters also have a negative eﬀect on food security levels. Agriculture, in fact, is a sector particularly exposed to natural risks (i.e., droughts, ﬂoods, earthquakes), which cause big economic losses every year. The growing frequency and complexity of these phenomena, and the projection that climate change will lead to an increase in extreme weather events [55], making it necessary to implement prevention policies that deal with agricultural losses and that ensure food supply ﬂows [56]. In this context, it becomes crucial to put in place public policies aimed at ensuring the spread of the insurance instrument, in order to increase farmers’ productivity, limit damage from natural disasters, and strengthen resilience to food and nutritional insecurity [57,58]. Int. J. Environ. Res. Public Health 2020, 17, 4311 11 of 13 Although this document has assessed and conﬁrmed the advantages deriving from the trade openness, it is essential to specify that they could generate signiﬁcant negative externalities. Currently, due to the globalization process, the seed market is controlled by just a handful of corporations, with damage in terms of biodiversity and healthiness of production, as well as negative economic consequences for small farmers [59]. In this context, CAP plays a key role thanks to the disbursement of funds that ensure income stability for 22 million small European farmers, which provide a large variety of accessible, safe, and good quality products [18]. Further research is needed to explore how the food security discourse is likely to evolve in the future, considering more indicators and increasing the analysis sample. Author Contributions: The following authors contributed in full to this work. G.F. and B.C. conceived the study and analyzed the sources, the literature, drafted the manuscript, and contributed to data collection and analysis. F.D.L. supervised the research project and carried out a detailed revision. All the authors wrote the body of the paper and All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest. Appendix A See Table A1. Table A1. List of European countries used in the analysis (33), because the data are available for this countries. Table A1. List of European countries used in the analysis. Albania Bulgaria Denmark Germany Latvia Poland Russian Federation Sweden United Kingdom Belarus Croatia Estonia Greece Luxembourg Portugal Slovakia Switzerland Belgium Cyprus Finland Ireland Netherlands Republic of Moldova Slovenia Turkey Bosnia and Herzegovina Czechia France Italy Norway Romania Spain Ukraine References 1. 2. 3. 4. 5. 6. 7. 8. 9. Long, M.A.; Gonçalves, L.; Stretesky, P.B.; Defeyter, M.A. Food insecurity in advanced capitalist nations: A review. Sustainability 2020, 12, 3654. [CrossRef] Food Security Information Network (FSIN). In Global Report on Food Crises; World Food Program: Rome, Italy, 2020; Available online: https://www.fsinplatform.org/sites/default/ﬁles/resources/ﬁles/GRFC%20ONLINE% 20FINAL%202020.pdf (accessed on 12 June 2020). Eurostat. Income and Living Conditions in Europe (EU-SILC); European Union: Luxembourg, 2013. FAO. Declaration on world food security. World Food Summit; FAO: Rome, Italy, 1996; Available online: http://www.fao.org/3/w3548e/w3548e00.htm (accessed on 12 June 2020). Richards, C.; Kjærnes, U.; Vik, J. Food security in welfare capitalism: Comparing social entitlements to food in Australia and Norway. J. Rural Stud. 2016, 43, 61–70. [CrossRef] Garratt, E. Food insecurity in Europe: Who is at risk, and how successful are social beneﬁts in protecting against food insecurity. J. Soc. Policy 2019, 1–25. [CrossRef] Fischer, C.; Miglietta, P.P. The links between human diets and health and climate outcomes in the world’s macro-regions during the last 50 years. Int. J. Environ. Res. Public Health 2020, 17, 1219. [CrossRef] [PubMed] Davis, O.; Geiger, B.B. Did Food Insecurity rise across Europe after the 2008 Crisis? An analysis across welfare regimes. Soc. Policy Soc. 2017, 16, 343–360. Borch, A.; Kjærnes, U. Food security and food insecurity in Europe: An analysis of the academic discourse (1975–2013). Appetite 2016, 103, 137–147. [CrossRef] [PubMed] Int. J. Environ. Res. Public Health 2020, 17, 4311 12 of 13 10. Carolan, M.S. Reclaiming Food Security; Routledge: Abingdon, UK, 2013. 11. Prosekov, A.Y.; Ivanova, S.A. Food security: The challenge of the present. Geoforum 2018, 91, 73–77. [CrossRef] 12. Coronese, M.; Lamperti, F.; Keller, K.; Chiaromonte, F.; Roventini, A. Evidence for sharp increase in the economic damages of extreme natural disasters. Proc. Natl. Acad. Sci. USA 2019, 116, 21450–21455. [CrossRef] [PubMed] FAO. The State of Food Security and Nutrition in the World 2017; FAO: Rome, Italy, 2017. 13. 14. Carvalho, F.P. Agriculture, pesticides, food security and food safety. Environ. Sci. Policy 2006, 9, 685–692. [CrossRef] 15. Walls, H.; Baker, P.; Chirwa, E.; Hawkins, B. Food security, food safety healthy nutrition: Are they compatible. Glob. Food Sec. 2019, 21, 69–71. [CrossRef] 16. Curtis, T.; Halford, N.G. Food security: The challenge of increasing wheat yield and the importance of not compromising food safety. Ann. Appl. Boil. 2014, 164, 354–372. [CrossRef] 17. Vogli, R.D.; Kouvonen, A.; Elovainio, M.; Marmot, M. Economic globalization, inequality and body mass index: A cross-national analysis of 127 countries. Crit. Public Health 2014, 24, 7–21. [CrossRef] 18. European Commission. Overview of CAP Reform 2014–2020. Agricultural Policy Perspectives Brief No. 5, DG Agriculture and Rural Development; Unit for Agricultural Policy Analysis and Perspectives: Brussels, Belgium, 2013. 19. Anderson, K. Krueger, Schiﬀ, and Valdes revisited: Agricultural price and trade policy reform in developing countries since 1960. Appl. Econ. Perspect. Policy 2010, 32, 195–231. [CrossRef] 20. Alesandro, O.; Daniel, C.; Swinnen, J. Trade Liberalization and Child Mortality: A Synthethic Control Method; Working Papers Department of Economics 567787, KU Leuven; Faculty of Business and Economics, Department of Economics KU Leuven: Leuven, Belgium, 2017. 21. Dithmer, J.; Abdulai, A. Does trade openness contribute to food security? A dynamic panel analysis. Food Policy 2017, 69, 218–230. [CrossRef] FAO. Multilateral Trade Negotiations on Agriculture: A Resource Manual; FAO: Rome, Italy, 2000. 22. 23. Wacziarg, R.; Welch, K.H. Trade liberalization and growth: New evidence. World Bank Econ. Rev. 2008, 22, 187–231. [CrossRef] 24. Dorosh, P.A.; Rashid, S.; van Asselt, J. Enhancing food security in South Sudan: The role of markets and 25. regional trade. Agric. Econ. 2016, 47, 697–707. [CrossRef] Feleke, S.T.; Kilmer, R.L.; Gladwin, C.H. Determinants of food security in Southern Ethiopia at the household level. Agric. Econ. 2005, 33, 351–363. [CrossRef] 26. Brunori, G.; Malandrin, V.; Rossi, A. Trade-oﬀ or convergence? The role of food security in the evolution of food discourse in Italy. J. Rural Stud. 2013, 29, 19–29. [CrossRef] 27. Kirwan, J.; Maye, D. Food security framings within the UK and the integration of local food systems. J. Rural Stud. 2013, 29, 91–100. [CrossRef] 28. Dowler, E.A.; O’Connor, D. Rights-based approaches to addressing food poverty and food insecurity in Ireland and UK. Soc. Sci. Med. 2012, 74, 44–51. [CrossRef] 29. Miewald, C.; McCann, E. Foodscapes and the geographies of poverty: Sustenance, strategy, and politics in 30. an urban neighborhood. Antipode 2014, 46, 537–556. [CrossRef] Fanelli, R.M. The interactions between the structure of the food supply and the impact of livestock production on the environment. A multivariate analysis for understanding the diﬀerences and the analogies across European Union countries. Qual. Access Success 2018, 19, 131–139. FAO. Trade Reforms and Food Security: Conceptualizing the Linkages; FAO: Rome, Italy, 2003. 31. 32. Headey, D.D. Developmental drivers of nutritional change: A cross-country analysis. World Dev. 2013, 42, 33. 76–88. [CrossRef] Shen, J.H.; Liu, X.J.; Zhang, J. Toward a uniﬁed theory of economic reform. Struct. Chang. Econ. Dyn. 2019, 51, 318–333. [CrossRef] FAO. The State of Food Insecurity in the World 2013; FAO: Rome, Italy, 2013. 34. 35. Heston, A.; Summers, R.; Aten, B. Penn World Table. Vers. 6.3. In Center for International Comparisons of Production, Income and Prices; University of Pennsylvania: Philadelphia, PA, USA, August 2009. 36. DeJong, D.N.; Ripoll, M. Tariﬀs and growth: An empirical exploration of contingent relationships. Rev. Econ. Stat. 2006, 88, 625–640. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 4311 13 of 13 37. Dreher, A.; Gaston, N.; Martens, P. Measuring Globalisation. Gauging Its Consequences; Springer: New York, NY, USA, 2008. 38. Nickell, S. Biases in dynamic models with ﬁxed eﬀects. Econometrica 1981, 49, 1417–1426. [CrossRef] 39. Holtz-Eakin, D.; Newey, W.; Rosen, H.S. Estimating vector autoregressions with panel data. Econometrica 1988, 56, 1371–1395. [CrossRef] 40. Arellano, M.; Bond, S. Some tests of speciﬁcation for panel data: Monte Carlo evidence and an application to employment equations. Rev. Econ. Stud. 1991, 58, 277–297. [CrossRef] 41. Bound, J.; Jaeger, D.A.; Baker, R.M. Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak. J. Am. Stat. Assoc. 1995, 90, 443–450. [CrossRef] 42. Baltagi, B. Econometric Analysis of Panel Data; John Wiley Sons: Hoboken, NJ, USA, 2008. 43. Blundell, R.; Bond, S. Initial conditions and moment restrictions in dynamic panel data models. J. Econom. 44. 45. 1998, 87, 115–143. [CrossRef] FAO. Trade Reforms and Food Security: Country Case Studies and Synthesis; FAO: Rome, Italy, 2006. Fanelli, R.M. The (un)sustainability of the land use practices and agricultural production in EU countries. Int. J. Environ. Stud. 2019, 76, 273–294. [CrossRef] 46. Loayza, N.V.; Olaberria, E.; Rigolini, J.; Christiaensen, L. Natural disasters and growth: Going beyond the averages. World Dev. 2012, 40, 1317–1336. [CrossRef] 47. Mason, C.H.; Perreault, W.D., Jr. Collinearity, power, and interpretation of multiple regression analysis. J. Mark. Res. 1991, 28, 268–280. [CrossRef] 48. Mary, S. Hungry for free trade? Food trade and extreme hunger in developing countries. Food Secur. 2019, 11, 461–477. [CrossRef] 49. Porrini, D.; Fusco, G.; Miglietta, P.P. Post-adversities recovery and proﬁtability: The case of Italian farmers. Int. J. Environ. Res. 2019, 16, 3189. [CrossRef] [PubMed] 51. 50. Kaplan, J.O.; Krumhardt, K.M.; Gaillard, M.J.; Sugita, S.; Trondman, A.K.; Fyfe, R.; Marquer, L.; Mazier, F.; Nielsen, A.B. Constraining the deforestation history of Europe: Evaluation of historical land use scenarios with pollen-based land cover reconstructions. Land 2017, 6, 91. [CrossRef] Fanelli, R.M. The spatial and temporal variability of the eﬀects of agricultural practices on the environment. Environments 2020, 7, 33. [CrossRef] Jaﬀee, S.; Henson, S.; Diaz Rios, L. Making the Grade: Smallholder Farmers, Emerging Standards, and Development Assistance Programs in Africa—A Research Program Synthesis; No 2823, World Bank Other Operational Studies; The World Bank: Washington, DC, USA, 2011. Sarker, R.; Jayasinghe, S. Regional trade agreements and trade in agri-food products: Evidence for the European Union from gravity modeling using disaggregated data. Agric. Econ. 2007, 37, 93–104. [CrossRef] 54. Coluccia, B.; Valente, D.; Fusco, G.; De Leo, F.; Porrini, D. Assessing agricultural eco-eﬃciency in Italian 52. 53. 55. Regions. Ecol. Indic. 2020, 116, 106483. [CrossRef] Suk, J.E.; Vaughan, E.C.; Cook, R.G.; Semenza, J.C. Natural disasters and infectious disease in Europe: A literature review to identify cascading risk pathways. Eur. J. Public Health 2019. [CrossRef] 56. Palmi, P.; Morrone, D.; Miglietta, P.P.; Fusco, G. How did organizational resilience work before and after the ﬁnancial crisis? An empirical study. Publ. Fac. Sci. Technol. 2018. [CrossRef] 57. Boscia, V.; Stefanelli, V.; Coluccia, B.; De Leo, F. The role of ﬁnance in environmental protection: A report on regulators’ perspective. Risk. Gov. Control Financ. Mark. Inst. 2019, 9, 30–40. [CrossRef] 58. Tabeau, A.; van Meijl, H.; Overmars, K.P.; Stehfest, E. REDD policy impacts on the agri-food sector and food security. Food Policy 2017, 66, 73–87. [CrossRef] 59. Robin, M.M. The World According to Monsanto: Pollution, Corruption, and the Control of Our Food Supply; The New Press: New York, NY, USA, 2014. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.14814_phy2.15750	Received: 17 March 2023 DOI: 10.14814/phy2.15750 \| Revised: 12 May 2023 \| Accepted: 28 May 2023 O R I G I N A L A R T I C L E Force sensor reduced measurement error compared with verbal command during sit- to- stand assessment of cerebral autoregulation Alicen A. Whitaker1,2,3 Kailee Carter1 \| Katelyn Struckle1 \| Sandra A. Billinger4,5,7,8 \| Eric D. Vidoni4,5 \| Robert N. Montgomery6 \| 1Department of Physical Therapy, Rehabilitation Science, and Athletic Training, University of Kansas Medical Center, Kansas City, Kansas, USA 2Department of Physical Medicine and Rehabilitation, Medical College of Wisconsin, Milwaukee, Wisconsin, USA 3Cardiovascular Center, Medical College of Wisconsin, Milwaukee, Wisconsin, USA 4University of Kansas Alzheimer's Disease Research Center, Fairway, Kansas, USA 5Department of Neurology, University of Kansas Medical Center, Kansas City, Kansas, USA 6Department of Biostatistics & Data Science, University of Kansas Medical Center, Kansas City, Kansas, USA 7Department of Physical Medicine and Rehabilitation, University of Kansas Medical Center, Kansas City, Kansas, USA 8Department of Cell Biology and Physiology, University of Kansas Medical Center, Kansas City, Kansas, USA Correspondence Sandra A. Billinger, Department of Neurology, University of Kansas Medical Center, 3901 Rainbow Blvd, Mail Stop 3051, Kansas City, KS 66160, USA. Email: [email protected] Funding information American Heart Association (AHA), Grant/Award Number: 898190; Cardiovascular Center's A.O. Smith Fellowship Scholars Program; HHS\|NIH\|Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Grant/Award Number: T32HD057850; HHS\|NIH\|National Center for Research Resources (NCRR), Grant/ Award Number: ULTR000001; HHS\|NIH\|National Heart, Lung, and Blood Institute (NHLBI), Grant/ Award Number: T32HL134643; HHS\|NIH\|National Institute on Aging (NIA), Grant/Award Number: P30 AG072973 Abstract Current methods estimate the time delay (TD) before the onset of dynamic cerebral autoregulation (dCA) from verbal command to stand. A force sensor used during a sit- to- stand dCA measure provides an objective moment an individual stands (arise- and- off, AO). We hypothesized that the detection of AO would improve the accuracy of TD compared with estimation. We measured middle cerebral artery blood velocity (MCAv) and mean arterial pressure (MAP) for 60 s sitting followed by 2- min standing, three times separated by 20 min. TD was calculated as the time from: (1) verbal command and (2) AO, until an increase in cerebrovascular con- ductance index (CVCi = MCAv/MAP). Sixty- five participants were enrolled: young adults (n = 25), older adults (n = 20), and individuals post- stroke (n = 20). The TD calculated from AO (x = 2.98 ± 1.64 s) was shorter than TD estimated from verbal command (x = 3.35 ± 1.72 s, η2 = 0.49, p < 0.001), improving measurement error by ~17%. TD measurement error was not related to age or stroke. Therefore, the force sensor provided an objective method to improve the calculation of TD compared with current methods. Our data support using a force sensor during sit- to- stand dCA measures in adults across the lifespan and post- stroke. K E Y W O R D S cerebral hemodynamics, dCA, dynamic cerebral autoregulation, MCAv, middle cerebral artery blood velocity, stroke This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Physiological Reports published by Wiley Periodicals LLC on behalf of The Physiological Society and the American Physiological Society. Physiological Reports. 2023;11:e15750. https://doi.org/10.14814/phy2.15750 wileyonlinelibrary.com/journal/phy2 \| 1 of 8 2 of 8 \| 1 \| INTRODU CT ION Dynamic cerebral autoregulation (dCA) is the ability of the brain to independently react to increases or decreases in pe- ripheral mean arterial pressure (MAP) and maintain cere- brovascular stability (Aaslid et al., 1989; Diehl et al., 1995; Newell et al., 1994; van Beek et al., 2008). The dCA response can be measured during a sit- to- stand procedure, which represents a meaningful, everyday activity and could be an important assessment in older adults and those with chronic health conditions (Labrecque et al., 2017; Sorond et al., 2009; van Beek et al., 2008). During a sit- to- stand procedure, MAP rapidly decreases due to gravity and peripheral vasodila- tion. Therefore, the dCA response must quickly increase the cerebrovascular conductance index (CVCi) of middle cere- bral artery blood velocity (MCAv) to maintain homeostasis (Labrecque et al., 2017; Sorond et al., 2009). One physiological metric of the sit- to- stand dCA mea- surement is the time delay (TD) before the onset of the regulation, which measures how quickly the cerebrovascu- lar system responds to a drop in MAP by increasing CVCi (Labrecque et al., 2017, 2019; Sorond et al., 2009). The TD before the onset of the regulation response typically oc- curs within ~10 s of standing (Labrecque et al., 2017, 2019; Serrador et al., 2005; Sorond et al., 2009) but can occur on av- erage ~1.5 s after standing in healthy young men (Labrecque et al., 2017). Therefore, the TD before the onset of the regu- lation response is an important temporal measure of cere- brovascular function and requires accuracy and precision. The current method to identify the cerebrovascular TD response has been calculated from the time in which indi- viduals are verbally told to stand up from a seated position. To improve standardization in the assessment of dCA, pre- vious studies have implemented a goniometer to detect the angle and speed of the sit- to- stand (Barnes et al., 2018, 2021; Barnes, Ball, Haunton, et al., 2017; Barnes, Ball, Panerai, et al., 2017; Batterham et al., 2020; Klein et al., 2020; Panerai et al., 2021). Our laboratory has focused on older adults and those with physical limitations and wanted to further ex- plore an objective measure of stance time. Our initial work in this line of scientific inquiry described the construction of the force sensor, the method used, and established a “proof of principle” that the force sensor could identify the pre- cise moment of stance and be successfully integrated into our cerebrovascular physiology methods to provide a more standardized approach of the TD response of dCA mea- sures (Whitaker et al., 2022). Furthermore, the code devel- oped for use with the sensor identified the exact moment of transition between sitting and standing (arise- and- off, AO; Whitaker et al., 2022). For example, we showed that one in- dividual post- stroke stood prior to the completion of the ver- bal command, which potentially introduces measurement error into the dCA response (Whitaker et al., 2022). While our initial publication described the development of the force sensor and provided preliminary support for its use, the next logical step to advance in this line of research was to rigorously test whether measurement error was system- atically reduced with the force sensor compared with verbal command across a group of young adults, older adults, and individuals post- stroke. Therefore, the objective of this study was to determine whether implementing a force sensor to detect the exact moment of AO would significantly improve the accuracy of the calculation of the TD response compared with the current TD method from the verbal command to stand. We hypothesized that using a force sensor to measure AO to calculate the TD response during a sit- to- stand measure of dCA would be significantly more accurate than the es- timated TD response from the verbal command to stand in healthy young adults, older adults, and individuals with stroke. We also hypothesized that the measurement error of the TD response would be significantly associated with the following participant demographics: age, body mass index (BMI), fast stance time, and history of stroke. 2 \| MATERIALS A ND M ET HODS This study reports on the accuracy of the current method used in an ongoing study (NCT04673994). The inclusion and exclusion criteria of this study have been published previously (Whitaker et al., 2022). Briefly, we enrolled healthy young adults 18– 30 years old with low cardio- vascular risk (Thompson et al., 2013). Older adults and individuals post- stroke (6 months— 5 years ago) were (1) age 40– 80 years old, (2) sedentary (<150 min brisk exer- cise/week; Thompson et al., 2013), (3) able to answer con- senting questions and follow a 2- step command, (4) able to stand up from a chair without physical assistance, and (5) not diagnosed with another underlying neurological disease. Individuals with chronic stroke were included within this study as they are a clinical population that pre- sents with hemiplegia and a slower sit- to- stand response that may be detected by the force sensor. The Human Subjects Committee within the University of Kansas Medical Center's Institutional Review Board ap- proved the study. Prior to starting the study, all individuals were informed of study procedures, benefits, and risks, and asked to provide voluntary written consent. We collected de- mographic information following written informed consent. For all procedures, we maintained a constant temperature (22– 24°C) and the room dimly lit. On the day of the visit, par- ticipants were asked to take their medications as prescribed, not to have caffeine for at least 8 h (Addicott et al., 2009; Institute of Medicine, 2001; Perod et al., 2000), not to per- form vigorous exercise for 24 h (Burma et al., 2020), and to WHITAKER et al. abstain from alcohol for 24 h (Mathew & Wilson, 1986). For premenopausal women, we collected data on Days 1– 6 of the menstrual cycle (Billinger et al., 2017; Whitaker et al., 2022). Participants were seated with their feet flat on the ground and an upright trunk posture. To determine whether individuals had the leg strength and balance to perform a sit- to- stand independently, a standardized 5× sit- to- stand was performed. Participants were asked to stand up and sit down five times as quickly as they could without the use of their arms and the time it took to perform the 5× sit- to- stand was measured in seconds (Mong et al., 2010; Ng, 2010; Tiedemann et al., 2008). Equipment was then donned, which included (1) bilateral TCD probes (2- MHz, Multigon Industries Inc.) to measure MCAv, (2) a finger cuff was placed on the left middle finger (or upper extremity without spastic- ity for individuals post- stroke) to measure beat- to- beat MAP (Finometer, Finapres Medical Systems), (3) a 5- lead electro- cardiogram (ECG; Cardiocard, Nasiff Associates) to measure heart rate, and (4) a nasal cannula attached to a capnograph (BCI Capnocheck Sleep 9004 Smiths Medical) to measure end- tidal carbon dioxide (PETCO2). As reported in our prior work (Whitaker et al., 2022), participants were instructed to place their hand with the finger plethysmograph flat on their chest at heart level and were fitted with an arm sling to hold the Finometer in place (Labrecque et al., 2017; Lipsitz et al., 2000; Sorond et al., 2009). Our custom force sensor was then placed underneath the participant at the level of their right ischial tuberosity. For individuals with stroke, the force sensor was placed underneath the nonaffected lower extremity (Brunt et al., 2002; Whitaker et al., 2022). Participants performed seated rest for 60 s. The participant was then given a 3- s countdown and asked to stand at the 60- s mark. The par- ticipant continued standing for an additional 2 min for he- modynamic stability (Drapeau et al., 2019). All measures were recorded at 500 Hz using a custom written software within MATLab, implementing the Data Acquisition Toolbox (R2019a; The Mathworks Inc.). Participants per- formed three sit- to- stand procedures (T1, T2, and T3) during the study visit, each separated by 20 min. \| 3 of 8 significant difference between the right and left MCAv (Billinger et al., 2017). However, the right MCAv signal was used if the left was not obtainable (Billinger et al., 2017). In individuals post- stroke, the ipsilesional hemisphere's MCAv signal was used. As previously published, AO was identified as the minimum of the second derivative of the recorded force sensor voltage upon standing (Whitaker et al., 2022). The manual identification of TD before the onset of the reg- ulation response was completed by two trained researchers (A.W. and K.C.) and evaluated as the physiological beat after standing where there was a continuous increase in CVCi and is defined as CVCi = MCAv/MAP (Labrecque et al., 2017; Lind- Holst et al., 2011; Whitaker et al., 2022). The primary aim was to determine whether the force sen- sor improved the calculation of the cerebrovascular TD re- sponse when using AO compared with the estimated time of stance (from verbal command at 60 s) with a two- way repeated measures ANOVA with a within- subjects effect for time (T1, T2, and T3) and type of calculation (AO or Estimated). Across time points (T1, T2, and T3), post hoc Wilcoxon signed rank t- tests were used to differences between AO and Estimated. We performed a mixed model ANOVA to test differences in the AO response using a within- subjects effect for time (T1, T2, and T3) and between- subjects effect for group (young adult, older adult, and individuals post- stroke). The force sensor measures the true time of stance with AO. To analyze measurement error, we calculated the dif- ference between the AO TD response and the estimated TD response from 60 s, which is the current method used in sit- to- stand procedures. The average TD measurement error was then plotted within a histogram graph to show the fre- quency distribution. To further explore measurement error, we calculated the coefficient of variation (CoV = (standard deviation/mean) × 100). Finally, Spearman correlations were then used to determine whether measurement error was significantly related to age and the 5× sit- to- stand. Pearson correlations were used to determine whether measurement error was significantly related to BMI and whether a participant had a history of stroke. Normality was checked using a Shapiro– Wilk test. 2.1 \| Data analysis 3 \| RESULTS Offline processing of the collected data was done using custom written software within MATLab. Sixty- five individuals enrolled into the study. Participant characteristics are shown in Table 1. To analyze, the data were divided by R- to- R cardiac in- terval. For each cardiac cycle, mean finger arterial pres- sure and MCAv were calculated as the area under the curve, as described in previous work (Billinger et al., 2017; Ward et al., 2018). The left MCAv signal was used for both healthy young adults and older adults as our prior work showed no Due to noise in MAP and TCD with standing, 59 indi- viduals had complete data sets with a TD response at all three timepoints. All individuals complied with abstaining from caffeine, vigorous exercise, and alcohol, as instructed. Baseline lower extremity function and hemodynamics be- tween groups are shown in Table 2. Individuals post- stroke had a significantly slower five times sit- to- stand compared WHITAKER et al. 4 of 8 \| with older adults (p < 0.001) and young healthy adults (p < 0.001). Young adults had a significantly higher resting MCAv compared with individuals post- stroke (p < 0.001) and older adults (p < 0.001). There was no difference be- tween groups in resting MAP, HR, or PETCO2. \| Comparing the TD response using 3.1 AO versus estimating Our key finding showed a significant difference in TD when using AO compared with estimating from the verbal command to stand at 60 s (η2 = 0.49, p < 0.001), shown in Figure 1. Across the three trials within the study visit, the TD response was not significantly different (p = 0.34). When performing post hoc comparisons, the TD response at each separate time point was significantly shorter when using AO compared with estimating from 60 s (p < 0.001). The AO TD response for T1 was x = 3.2 ± 2.3 s, T2 was x = 2.9 ± 1.9 s, and T3 was x = 2.9 ± 2.2 s compared with the estimated TD response of T1 was x = 3.7 ± 2.4 s, T2 was x = 3.2 ± 2.1 s, and T3 was x = 3.2 ± 2.3 s. The data showed no group differences for the measured AO response (p = 0.86). \| Measurement error compared with 3.2 verbal command A histogram in Figure 2 demonstrates the frequency of how predicted values using the current method dif- fer from the true, measured value. The distribution of measurement error is negatively skewed and revealed T A B L E 1 Participant characteristics. T A B L E 2 Baseline hemodynamics. Young adults (n = 25) Age (years) Female, n (%) Body mass index (kg/m2) 25 ± 2 12 (48%) 23.9 ± 3.8 Race, n (%) White/Caucasian 18 (72%) Black/African American 0 Asian Native American Ethnicity, n (%) 6 (24%) 1 (4%) Hispanic/Latino 1 (4%) Older adults (n = 20) 61 ± 13a 6 (30%) 28.1 ± 6.2a Stroke (n = 20) 60 ± 13a 6 (30%) 30.7 ± 4.9a p- Value <0.001* 0.38 <0.001* 0.02* 17 (85%) 2 (10%) 1 (5%) 0 0 15 (75%) 5 (25%) 0 1 (5%) 0 1.00 Note: Means ± standard deviations. Two individuals identified as multiple races/ethnicities. Participant demographics were compared between groups using a one- way ANOVA or Kruskal– Wallis ANOVA, with post hoc t- tests. For categorical variables, a Fisher's exact or chi- squared test was used to compare differences between groups. aSignificantly different from young adults. Significantly different between groups (p < 0.05). Young adults (n = 22) Older adults (n = 19) 8.3 ± 2.1 8.9 ± 2.4 65.58 ± 10.32 74.91 ± 12.61 71 ± 11 47.70 ± 13.21a 80.64 ± 9.91 64 ± 9 Stroke (n = 18) 22.9 ± 14.1a,b 42.70 ± 10.33a 81.32 ± 12.66 68 ± 11 37.89 ± 3.58 35.35 ± 3.56 36.56 ± 4.21 p- Value <0.001 <0.001* 0.17 0.10 0.11 5 Time sit- to- stand (s) MCAv (cm/s) MAP (mmHg) HR (bpm) PETCO2 Note: Means ± standard deviations. Comparisons between groups used a one- way ANOVA or Kruskal– Wallis ANOVA, with post hoc t- tests. Abbreviations: HR, heart rate; MAP, mean arterial pressure; MCAv, middle cerebral artery blood velocity; PETCO2, end- tidal carbon dioxide. aSignificantly different from young adults. bSignificantly different from older adults. *Significantly different between groups (p < 0.05). WHITAKER et al. \| 5 of 8 F I G U R E 2 Histogram of the TD average measurement error. TD = time delay. The average measurement error of the TD response was not significantly related to age (p = 0.40), BMI (p = 0.73), 5× sit- to- stand (p = 0.35), or history of stroke (p = 0.82). Data presented in Table 3 show the CoV of AO is small across all groups and suggests low variability across time points and groups. 4 \| DISCUSSIO N Using a force sensor to objectively identify AO during a sit- to- stand resulted in a more standardized approach during the sit- to- stand maneuver when compared to the current method of estimating TD from verbal command to stand. The main findings of this study include (1) the TD response occurred significantly faster when calculated using AO compared with estimating from 60 s, (2) the force sensor reduced measurement error for all individuals compared with verbal command, and (3) the AO response showed small variability across time within the same day and across groups consisting of healthy young adults, older adults, and individuals post- stroke. This study extends our prior work (Whitaker et al., 2022) by showing the force sensor statis- tically reduced error compared with verbal command dur- ing the sit- to- stand measure of dCA in not only individuals post- stroke but also young and older adults. The reduction in measurement error using an objective identification of stance ensures more accurate reporting of the transient time course changes in MCAv and MAP, which is critical in the field of cerebrovascular physiology and may have implications for dCA. Prior studies have identified two physiological phases that occur in response to the sit- to- stand maneuver (Labrecque et al., 2017; Skow et al., 2021). Phase 1 is the initial response and the time point where MCAv responses are “independent of arte- rial baroreflex input and occur within 1– 7 s.” (Deegan F I G U R E 1 T1 sit- to- stand response. Young healthy adults = solid triangles. Older adults = solid circles. Individuals post- stroke = open circles. AO, arise- and- off measured via the force sensor; CVCi, cerebrovascular conductance index (MCAv/MAP); MAP, mean arterial pressure; MCAv, middle cerebral artery blood velocity. Resampled to 1 Hz for group averages. that stance occurred ~0.5 s after the verbal command. We report the average TD response was 3.0 s with ~17% measurement error when following the current method of estimating the TD response from the verbal command. The maximal error calculated between the AO TD and the TD estimated from 60 s was −2.52 s measured in an older healthy adult. WHITAKER et al. 6 of 8 \| t n e m e r u s a e m D T t n e m e r u s a e m D T t n e m e r u s a e m D T V o C O A ) s ( r o r r e ) s ( O A V o C O A ) s ( r o r r e ) s ( O A V o C O A ) s ( r o r r e ) s ( O A 3 T 2 T 1 T . O A f o n o i t a i r a v f o t n e i c i f f e o C 3 E L B A T % 7 8 0 . % 8 7 0 . % 8 6 0 . . 2 5 0 ± 9 2 0 − . . 7 4 0 ± 6 2 0 − . . 1 4 0 ± 4 3 0 − . . 2 5 0 ± 9 2 0 6 . . 7 4 0 ± 6 2 0 6 . . 1 4 0 ± 4 3 0 6 . % 2 8 0 . % 5 2 1 . % 4 8 0 . . 9 4 0 ± 7 1 0 − . . 9 4 0 ± 7 1 0 6 . . 5 7 0 ± 7 3 0 − . . 1 5 0 ± 1 4 0 − . . 5 7 0 ± 7 3 0 6 . . 1 5 0 ± 1 4 0 6 . % 5 8 0 . % 7 6 0 . % 8 0 1 . . 2 5 0 ± 5 4 0 − . . 2 5 0 ± 5 4 0 6 . . 1 4 0 ± 4 5 0 − . . 1 4 0 ± 4 5 0 6 . . 5 6 0 ± 5 3 0 − . . 5 6 0 ± 5 3 0 6 . s t l u d a g n u o Y s t l u d a r e d l O e k o r t S et al., 2009; Labrecque et al., 2017; Ogoh et al., 2008; Skow et al., 2021; Sorond et al., 2009; van Beek et al., 2008) Phase 2 continues past Phase 1 and involves the influence of the arterial baroreflex response (Labrecque et al., 2017; Ogoh et al., 2008; Skow et al., 2021). Given the Phase 1 time sensitivity, reducing measurement error in TD would have significant implications in reporting these data. The results reported in this study demonstrate that the force sensor reduced measurement error by ~17%, which would have significant implication for clinical and research ap- plications and may affect dCA results. Considerable interest exists in understanding cerebro- vascular function and brain aging in older adults and clin- ical populations such as traumatic brain injury, Alzheimer's disease, dementia, and stroke. The sit- to- stand maneuver mimics a common and very important daily activity that provides a physiologic challenge. While our intention was to employ a quick sit- to- stand response (0– 3 s) aligned with the literature (Labrecque et al., 2017), our prior work showed that different stance strategies may be used in older adults and people with stroke. We found some participants, espe- cially those post- stroke, stood prior to the verbal command (Whitaker et al., 2022), which introduces inherent variability in the sit- to- stand maneuver. Furthermore, data in Table 2 reveal a significantly slower 5× sit- to- stand time in people with stroke compared with the young and older adults. The implementation of the force sensor to identify the precise moment of stance in clinical populations may have even greater impact on reducing measurement error during a sit- to- stand maneuver. Finally, our data support previous find- ings of low within- subject variability (Sorond et al., 2009). For all participants regardless of age or stroke, we showed no significant differences in the TD across the three time points, suggesting low variability in this participant sample. While the TD only measures the temporal onset of the reg- ulation response, there are other metrics calculated during the sit- to- stand dCA response, such as the rate of regula- tion, which provides information about the change in CVCi after the TD. Many studies also report dCA in the temporal- domain alongside measures of dCA in the spectral- domain (i.e., transfer function analysis) as both techniques provide unique information about cerebrovascular regulation (Klein et al., 2020; Labrecque et al., 2017, 2019). We acknowledge that the present study did not imple- ment an accelerometer to normalize the dCA response to the speed of the sit- to- stand as others have previously done (Barnes et al., 2018, 2021; Barnes, Ball, Haunton, et al., 2017; Batterham et al., 2020; Panerai et al., 2021). Our ongoing work is currently implementing both the force sensor and an accelerometer to identify the precise moment of AO and the speed of stance during the dCA measure. Future studies should also implement the use of a force sensor to deter- mine whether measurement error is reduced and improves . y a l e d e m i t , D T ; n o i t a i r a v f o t n e i c i f f e o c , V o C ; ) e c n a t s f o t n e m o m t c a x e ( f f o - d n a - e s i r a , O A : s n o i t a i v e r b b A . s 0 6 m o r f d e t a m i t s e D T − O A D T = e c n e r e f f i d D T . s t n i o p e m i t = 3 – 1 T : e t o N WHITAKER et al. the accuracy of the dCA response through transfer function analysis (Burma et al., 2020; Drapeau et al., 2019; Labrecque et al., 2022). Lastly, we report differences in resting MCAv (Table 2) between young adults and older adults and in- dividuals post- stroke. However, to the extent which the resting MCAv influences AO and the TD presented here is unknown and not within the scope of this project. In conclusion, our results support the use of a force sensor to reduce TD measurement error during a sit- to- stand maneuver. The force sensor improves upon current methods and provides a standardized, objective approach to ensure rigor and reproducibility during a common daily activity, sit- to- stand, to assess dCA. AUTHOR CONT RIBUTIO NS Alicen A. Whitaker, Eric D. Vidoni, and Sandra A. Billinger conceived and designed research; Alicen A. Whitaker, Kailee Carter, and Katelyn Struckle performed experiments; Alicen A. Whitaker, Kailee Carter, Katelyn Struckle, and Sandra A. Billinger analyzed data; Alicen A. Whitaker, Eric D. Vidoni, Robert N. Montgomery, and Sandra A. Billinger interpreted results of experiments; Alicen A. Whitaker, Eric D. Vidoni, Robert N. Montgomery, Kailee Carter, Katelyn Struckle, and Sandra A. Billinger drafted manuscript; Alicen A. Whitaker, Eric D. Vidoni, Robert N. Montgomery, Kailee Carter, Katelyn Struckle, and Sandra A. Billinger edited and revised manuscript; Alicen A. Whitaker, Eric D. Vidoni, Robert N. Montgomery, Kailee Carter, Katelyn Struckle, and Sandra A. Billinger approved final version of manuscript; Alicen A. Whitaker prepared figures. FUNDING INFORMATIO N AW was supported by the National Heart, Lung and Blood Institute (T32HL134643), Cardiovascular Center's A.O. Smith Fellowship Scholars Program, Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health (T32HD057850), and the American Heart Association Predoctoral Fellowship Grant (898190). EV and SB were supported in part by the National Institute on Aging for the KU Alzheimer's Disease Research Center (P30 AG072973). REDCap at the University of Kansas Medical Center is supported by Clinical and Translational Science Awards (CTSA) Award # ULTR000001 from National Center for Research Resources (NCRR). The content is solely the re- sponsibility of the authors and does not necessarily repre- sent the official views of the National Institutes of Health. CONFLICT OF INT ER EST STATE ME N T The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. \| 7 of 8 ET HICS STAT EM ENT The Human subjects Committee and Institutional Review Board at the University of Kansas Medical Center granted ethical approval of this study. All participants were pro- vided verbal and written explanation of the experimental protocol and the associated risks prior to providing writ- ten informed consent. DATA AVAILABILIT Y STAT EM EN T The datasets analyzed for this study are available upon re- quest to the corresponding author. ORCID Alicen A. Whitaker org/0000-0002-8701-9227 Eric D. Vidoni Robert N. Montgomery org/0000-0002-1423-0590 Sandra A. Billinger org/0000-0002-1618-7207 https://orcid. https://orcid. https://orcid. https://orcid.org/0000-0001-5181-7131 R E F E R E N C E S Aaslid, R., Lindegaard, K. F., Sorteberg, W., & Nornes, H. (1989). Cerebral autoregulation dynamics in humans. Stroke, 20(1), 45– 52. Addicott, M. A., Yang, L. L., Peiffer, A. M., Burnett, L. R., Burdette, J. H., Chen, M. Y., Hayasaka, S., Kraft, R. A., Maldjian, J. A., & Laurienti, P. J. (2009). The effect of daily caffeine use on cere- bral blood flow: How much caffeine can we tolerate? Human Brain Mapping, 30(10), 3102– 3114. Barnes, S. C., Ball, N., Haunton, V. J., Robinson, T. G., & Panerai, R. B. (2017). The cerebrocardiovascular response to periodic squat- stand maneuvers in healthy subjects: A time- domain analysis. American Journal of Physiology. Heart and Circulatory Physiology, 313(6), H1240– H1248. Barnes, S. C., Ball, N., Haunton, V. J., Robinson, T. G., & Panerai, R. B. (2018). How many squat- stand manoeuvres to assess dy- namic cerebral autoregulation? European Journal of Applied Physiology, 118(11), 2377– 2384. Barnes, S. C., Ball, N., Panerai, R. B., Robinson, T. G., & Haunton, V. J. (2017). Random squat/stand maneuvers: A novel approach for assessment of dynamic cerebral autoregulation? Journal of Applied Physiology, 123(3), 558– 566. Barnes, S. C., Haunton, V. J., Beishon, L., Llwyd, O., Robinson, T. G., & Panerai, R. B. (2021). Extremes of cerebral blood flow during hypercapnic squat- stand maneuvers. Physiological Reports, 9(19), e15021. Batterham, A. P., Panerai, R. B., Robinson, T. G., & Haunton, V. J. (2020). Does depth of squat- stand maneuver affect estimates of dynamic cerebral autoregulation? Physiological Reports, 8(16), e14549. Billinger, S. A., Craig, J. C., Kwapiszeski, S. J., Sisante, J. V., Vidoni, E. D., Maletsky, R., & Poole, D. C. (2017). Dynamics of middle cerebral artery blood flow velocity during moderate- intensity exercise. Journal of Applied Physiology, 122(5), 1125– 1133. Brunt, D., Greenberg, B., Wankadia, S., Trimble, M. A., & Shechtman, O. (2002). The effect of foot placement on sit to stand in healthy WHITAKER et al. 8 of 8 \| young subjects and patients with hemiplegia. Archives of Physical Medicine and Rehabilitation, 83(7), 924– 929. Burma, J. S., Copeland, P., Macaulay, A., Khatra, O., Wright, A. D., & Smirl, J. D. (2020). Dynamic cerebral autoregulation across the cardiac cycle during 8 hr of recovery from acute exercise. Physiological Reports, 8(5), e14367. Deegan, B. M., Sorond, F. A., Lipsitz, L. A., Olaighin, G., & Serrador, J. M. (2009). Gender related differences in cerebral autoregula- tion in older healthy subjects. Annual International Conference of the IEEE Engineering in Medicine and Biology Society (pp. 2859– 2862). IEEE. Diehl, R. R., Linden, D., Lücke, D., & Berlit, P. (1995). Phase relation- ship between cerebral blood flow velocity and blood pressure. A clinical test of autoregulation. Stroke, 26(10), 1801– 1804. Drapeau, A., Labrecque, L., Imhoff, S., Paquette, M., Le Blanc, O., Malenfant, S., & Brassard, P. (2019). Six weeks of high- intensity interval training to exhaustion attenuates dynamic cerebral au- toregulation without influencing resting cerebral blood velocity in young fit men. Physiological Reports, 7(15), e14185. Institute of Medicine. (2001). Caffeine for the sustainment of men- tal task performance: Formulations for military operations. National Academy Press. Klein, T., Bailey, T. G., Wollseiffen, P., Schneider, S., & Askew, C. D. (2020). The effect of age on cerebral blood flow responses during repeated and sustained stand to sit transitions. Physiological Reports, 8(9), e14421. Labrecque, L., Burma, J. S., Roy, M. A., Smirl, J. D., & Brassard, P. (2022). Reproducibility and diurnal variation of the directional sensitivity of the cerebral pressure- flow relationship in men and women. Journal of Applied Physiology, 132(1), 154– 166. Labrecque, L., Rahimaly, K., Imhoff, S., Paquette, M., Le Blanc, O., Malenfant, S., Drapeau, A., Smirl, J. D., Bailey, D. M., & Brassard, P. (2019). Dynamic cerebral autoregulation is atten- uated in young fit women. Physiological Reports, 7(2), e13984. Labrecque, L., Rahimaly, K., Imhoff, S., Paquette, M., Le Blanc, O., Malenfant, S., Lucas, S. J., Bailey, D. M., Smirl, J. D., & Brassard, P. (2017). Diminished dynamic cerebral autoregulatory capacity with forced oscillations in mean arterial pressure with elevated cardiorespiratory fitness. Physiological Reports, 5(21), e13486. Lind- Holst, M., Cotter, J. D., Helge, J. W., Boushel, R., Augustesen, H., Van Lieshout, J. J., & Pott, F. C. (2011). Cerebral autoreg- ulation dynamics in endurance- trained individuals. Journal of Applied Physiology, 110(5), 1327– 1333. Lipsitz, L. A., Mukai, S., Hamner, J., Gagnon, M., & Babikian, V. (2000). Dynamic regulation of middle cerebral artery blood flow velocity in aging and hypertension. Stroke, 31(8), 1897– 1903. Mathew, R. J., & Wilson, W. H. (1986). Regional cerebral blood flow changes associated with ethanol intoxication. Stroke, 17(6), 1156– 1159. Mong, Y., Teo, T. W., & Ng, S. S. (2010). 5- repetition sit- to- stand test in subjects with chronic stroke: Reliability and validity. Archives of Physical Medicine and Rehabilitation, 91(3), 407– 413. Newell, D. W., Aaslid, R., Lam, A., Mayberg, T. S., & Winn, H. R. (1994). Comparison of flow and velocity during dynamic auto- regulation testing in humans. Stroke, 25(4), 793– 797. Ng, S. (2010). Balance ability, not muscle strength and exercise en- durance, determines the performance of hemiparetic subjects on the timed- sit- to- stand test. American Journal of Physical Medicine & Rehabilitation, 89(6), 497– 504. Ogoh, S., Brothers, R. M., Eubank, W. L., & Raven, P. B. (2008). Autonomic neural control of the cerebral vasculature: Acute hypotension. Stroke, 39(7), 1979– 1987. Panerai, R. B., Batterham, A., Robinson, T. G., & Haunton, V. J. (2021). Determinants of cerebral blood flow velocity change during squat- stand maneuvers. American Journal of Physiology Regulatory, Integrative and Comparative Physiology, 320(4), R452– R466. Perod, A. L., Roberts, A. E., & McKinney, W. M. (2000). Caffeine can affect velocity in the middle cerebral artery during hyperventi- lation, hypoventilation, and thinking: A transcranial Doppler study. Journal of Neuroimaging, 10(1), 33– 38. Serrador, J. M., Sorond, F. A., Vyas, M., Gagnon, M., Iloputaife, I. D., & Lipsitz, L. A. (2005). Cerebral pressure- flow relations in hypertensive elderly humans: Transfer gain in different frequency domains. Journal of Applied Physiology, 98(1), 151– 159. Skow, R. J., Labrecque, L., Rosenberger, J. A., Brassard, P., Steinback, C. D., & Davenport, M. H. (2021). Prenatal exercise and cardiovascu- lar health (PEACH) study: Impact of acute and chronic exercise on cerebrovascular hemodynamics and dynamic cerebral auto- regulation. Journal of Applied Physiology, 132, 247– 260. Sorond, F. A., Serrador, J. M., Jones, R. N., Shaffer, M. L., & Lipsitz, L. A. (2009). The sit- to- stand technique for the measurement of dynamic cerebral autoregulation. Ultrasound in Medicine & Biology, 35(1), 21– 29. Thompson, P. D., Arena, R., Riebe, D., Pescatello, L. S., & American College of Sports Medicine. (2013). ACSM's new prepartic- ipation health screening recommendations from ACSM's guidelines for exercise testing and prescription. Current Sports Medicine Reports, 12(4), 215– 217. Tiedemann, A., Shimada, H., Sherrington, C., Murray, S., & Lord, S. (2008). The comparative ability of eight functional mobility tests for predicting falls in community- dwelling older people. Age and Ageing, 37(4), 430– 435. van Beek, A. H., Claassen, J. A., Rikkert, M. G., & Jansen, R. W. (2008). Cerebral autoregulation: An overview of current concepts and methodology with special focus on the elderly. Journal of Cerebral Blood Flow and Metabolism, 28(6), 1071– 1085. Ward, J. L., Craig, J. C., Liu, Y., Vidoni, E. D., Maletsky, R., Poole, D. C., & Billinger, S. A. (2018). Effect of healthy aging and sex on middle cerebral artery blood velocity dynamics during moderate- intensity exercise. American Journal of Physiology. Heart and Circulatory Physiology, 315(3), H492– H501. Whitaker, A. A., Aaron, S. E., Kaufman, C. S., Kurtz, B. K., Bai, S. X., Vidoni, E. D., Montgomeryand, R. N., & Billinger, S. A. (2022). Cerebrovascular response to an acute bout of low- volume high- intensity interval exercise and recovery in young healthy adults. Journal of Applied Physiology, 132(1), 236– 246. Whitaker, A. A., Vidoni, E. D., Aaron, S. E., Rouse, A. G., & Billinger, S. A. (2022). Novel application of a force sensor during sit- to- stands to measure dynamic cerebral autoregulation onset. Physiological Reports, 10(7), e15244. How to cite this article: Whitaker, A. A., Vidoni, E. D., Montgomery, R. N., Carter, K., Struckle, K., & Billinger, S. A. (2023). Force sensor reduced measurement error compared with verbal command during sit- to- stand assessment of cerebral autoregulation. Physiological Reports, 11, e15750. https://doi.org/10.14814/phy2.15750 WHITAKER et al.
10.3390_genes14061231	Article The Use of Xpert MTB/RIF Ultra Testing for Early Diagnosis of Tuberculosis: A Retrospective Study from a Single-Center Database Cristian Sava 1,2 Cristian Phillip Marinău 1,2 and Andreea Bianca Balmos, , Mihaela Sava 2, Ana-Maria Drăgan 1,2, Alin Iuhas 1,2,* 1,2 , Larisa Niulas, 1,2 , 1 Faculty of Medicine and Pharmacy, University of Oradea, 410087 Oradea, Romania 2 Clinical Emergency Bihor County Hospital, 410167 Oradea, Romania * Correspondence: [email protected] Abstract: Tuberculosis (TB) is a multisystemic contagious disease produced by Mycobacterium tuberculosis complex bacteria (MTBC), with a prevalence of 65:100,000 inhabitants in Romania (six times higher than the European average). The diagnosis usually relies on the detection of MTBC in culture. Although this is a sensitive method of detection and remains the “gold standard”, the results are obtained after several weeks. Nucleic acid ampliﬁcation tests (NAATs), being a quick and sensitive method, represent progress in the diagnosis of TB. The aim of this study is to assess the assumption that NAAT using Xpert MTB/RIF is an efﬁcient method of TB diagnosis and has the capacity to reduce false-positive results. Pathological samples from 862 patients with TB suspicion were tested using microscopic examination, molecular testing and bacterial culture. The results show that the Xpert MTB/RIF Ultra test has a sensitivity of 95% and a speciﬁcity of 96.4% compared with 54.8% sensitivity and 99.5% speciﬁcity for Ziehl–Neelsen stain microscopy, and an average of 30 days gained in the diagnosis of TB compared with bacterial culture. The implementation of molecular testing in TB laboratories leads to an important increase in early diagnostics of the disease and the prompter isolation and treatment of infected patients. Keywords: tuberculosis; molecular testing; Xpert MTB/RIF Ultra; Ziehl–Neelsen staining; Lowenstein– Jensen medium culture 1. Introduction Tuberculosis (TB) represents a serious global health issue, and is one of the leading morbidity and mortality factors. Every year, several million people worldwide become infected with tuberculosis and lose their lives due to the disease. TB is a disease linked with poverty and economic stress; vulnerability, marginalization, stigma and discrimination are often problems that people with TB must confront [1]. TB is a multisystemic contagious disease caused by Mycobacterium tuberculosis complex bacteria (MTBC). It is estimated that over 1.7 billion people (over 25% of world population) are infected with MTBC. The global incidence had a peak in 2003, and has slowly been decreasing since then. According to the latest World Health Organization (WHO, Geneva, Switzerland) report, the estimated number of deaths from TB experienced a decline between 2005 and 2019, with over 10 million people contracting TB and 1.4 million dying in 2019 and 1.5 million in 2020; however, the estimates for 2020 and 2021 indicate that this trend has been reversed, with an increase in the number of deaths [1,2]. Poverty, HIV infection and drug resistance are the principal factors that contribute to the re-emergence of the global TB epidemic [3]. It is projected that in 2020 and 2021, tuberculosis (TB) will be the second most common cause of death attributed to a single infectious agent, following COVID- 19 [1]. About 95% of the cases are recorded in developing countries; one in every nine new cases affects HIV-infected people; and 75% of all cases occur in Africa. It is estimated Citation: Sava, C.; Sava, M.; Dr˘agan, A.-M.; Iuhas, A.; Niulas, , L.; Marin˘au, C.P.; Balmos, , A.B. The Use of Xpert MTB/RIF Ultra Testing for Early Diagnosis of Tuberculosis: A Retrospective Study from a Single-Center Database. Genes 2023, 14, 1231. https://doi.org/10.3390/ genes14061231 Academic Editor: Nathalie Bissonnette Received: 28 April 2023 Revised: 2 June 2023 Accepted: 6 June 2023 Published: 7 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Genes 2023, 14, 1231. https://doi.org/10.3390/genes14061231 https://www.mdpi.com/journal/genes genesG C A TT A C GG C A TGenes 2023, 14, 1231 2 of 10 that 500,000 new multi-drug-resistant TB (MDR-TB) or rifampicin-resistant TB cases occur annually [1]. TB epidemiology varies substantially around the world. The highest prevalence (over 100:100,000 inhabitants) can be observed in Sub-Saharan Africa, India and South-East insular Asia and Micronesia. Intermediary rates (25–99 cases per 100,000 inhabitants) are found in China, Central and South America, Eastern Europe and North Africa. Lower prevalence (under 25 cases per 100,000 inhabitants) can be observed in North America, Western Europe, Japan and Australia [1]. In 2018, there were 52,862 cases reported in the European Union and the European Economic Space (EU/EES), resulting a prevalence of 10.2 cases per 100,000 inhabitants. The prevalence and the incidence in EU/EES countries had declined over the last ﬁve years [4]. Unfortunately, Romania remains the country with the highest prevalence from the EU/EES—64.6 cases per 100,000 inhabitants in 2017, which is four times higher than the EU mean, with one of the lowest recovery rates and, at the same time, an annual increase in the infectious reservoir. Romania has a mortality rate due to TB of 4.2 per 100,000 inhabitants, more than six times higher than the EU mean and 1.9 times higher than the WHO European Region’s mean, according to the latest report from the INSP—CNSISP (Romania’s National Public Health Institute, Bucharest, Romania) [5,6]. During the 2014 World Health Summit in Geneva, the WHO proposed a global strategy and targets for tuberculosis prevention, care and control, aiming to stop the global TB epidemic [7]. The proposed objectives were a 95%reduction in TB-related death by 2030, a 90% reduction in disease incidence in the 2015–2035 period and the elimination of associated catastrophic costs for tuberculosis-affected households. In addition to targets for 2030, the End TB Strategy deﬁnes 2020 and 2025 milestones for reductions in TB incidence and in the number of TB deaths. The 2020 milestones are a 20% reduction in TB incidence and a 35% reduction in the number of TB deaths, compared with levels in 2015 [8,9]. Reaching these objectives requires the early diagnosis of TB, including through the improvement of diagnostic methods, complete treatment of all people with TB, and the diagnosis and treatment of latent TB infection. The COVID-19 pandemic hugely affected patients’ access to proper medical services. TB care and prevention were particularly affected by the redirection of human, ﬁnancial and other resources to the COVID-19 response. Furthermore, public health measures resulted in reducing access to TB diagnosis and treatment services [10]. The early diagnosis of tuberculosis enables the prompt initiation of treatment and has the potential to restrict the transmission of this infectious disease. Its diagnosis usually relies on the detection of MTBC in culture. Although this is a sensitive method of detection and remains the “gold standard”, the results are obtained after several weeks. Microscopic examination is an inexpensive and quick test, but is also a rather insensitive test and cannot distinguish between non-tuberculosis mycobacteria and MTBC or between susceptible and resistant strains. Nucleic acid ampliﬁcation tests (NAATs), being a quick and sensitive method, represent progress in the diagnosis of TB [11]. The WHO recommends replacing microscopic examinations, as the initial diagnostic method, with molecular tests capable of identifying MTBC, in certain epidemiological and geographical settings. The newer, more rapid and more sensitive molecular tests recommended for the initial detection of MTBC and drug resistance are designated as mWRDs (molecular WHO-recommended rapid diagnostics tests); these include Xpert MTB/RIF Ultra and Xpert MTB/RIF (Cepheid, Sunnyvale, CA, USA); Truenat MTB, MTB Plus and MTB-RIF Dx tests (Molbio Diagnostics, Goa, India); and loop-mediated isothermal ampliﬁcation (TB-LAMP; Eiken hemical, Tokyo, Japan) [12]. The Xpert MTB/RIF method is a molecular test that has the capacity to detect the MTBC and the rpoB gene variant associated with rifampicin resistance [13]. Molecular tests are becoming increasingly pertinent in the diagnosis of various diseases as their accessibility and performance capabilities continue to improve [14]. The primary objective of this study was to investigate several hypotheses regarding the efﬁciency of nucleic acid ampliﬁcation tests (NAATs) using the Xpert MTB/RIF Ultra Genes 2023, 14, 1231 3 of 10 method in the early diagnosis of tuberculosis (TB) and its impact on prompt treatment initi- ation in positive cases. Additionally, the study aimed to assess the ability of this diagnostic approach to reduce false-positive results in suspected TB cases and avoid unnecessary administration of antituberculosis treatment. Moreover, the Xpert MTB/RIF Ultra test was evaluated for its capacity to identify mutations in the rpoB gene associated with rifampicin resistance in samples where Mycobacterium tuberculosis complex (MTBC) was detected. The speciﬁc objectives of the study were as follows: (i) evaluating the sensitivity and speciﬁcity of molecular tests compared to microscopic examination and mycobacterial culture for TB diagnosis, (ii) estimating the time saved in initiating tuberculostatic treatment utilizing molecular tests, (iii) analyzing molecular tests’ ability to identify non-tuberculosis mycobac- terial infections and reduce false-positive results, and (iv) detecting rifampicin resistance. 2. Materials and Methods During the period of 1 January 2018–31 December 2020, in the TB Bacteriology Labora- tory of the “Dr. Gavril Curteanu” Municipal Clinical Hospital (currently Clinical Emergency Bihor County Hospital) in Oradea, Bihor County, Romania, 13,916 biological specimens were analyzed with the purpose of identifying MTBC. All these samples were tested using microscopic examination and bacterial culture. In 862 cases, the specimens were also tested using the Xpert MTB/RIF Ultra method. In this study, 862 patients with a high suspicion of TB infection were included. The suspicion of the disease was determined in accordance with the guidelines provided by the Romanian National Guideline for the prevention, surveillance and control of tuberculosis criteria (epidemiological, clinical and/or imagistic), whose samples were also analyzed using the Xpert MTB/RIF Ultra method, in the mentioned period [15]. The samples consisted of sputum obtained via direct matinal sampling, induced sputum, bronchial aspirate, gastric aspirate, pleural puncture or lumbar puncture (CSF). The quality of the biological samples was essential in obtaining a trustworthy result. Sputum samples deemed inadequate (thin, clear sputum; improper sampling) were excluded from the study. The collected data were analyzed using IBM SPSS Statistics version 26. 2.1. Microscopic Examination Technique Microscopic examination was performed for all the samples. Sputum was the elective pathological product. The sputum smear for the microscopic examination was prepared using a bacteriologic wire loop, choosing the spots with purulent, opaque sputum and spreading it on the central portion of the slide, uniformly, in a thin layer, on a surface area of approximately 1 × 2 cm, avoiding the edge of the slide. The slides were left to dry under the hood, at room temperature, and then, heat-ﬁxated using a Bunsen burner (3 times). Ziehl–Neelsen staining was used for acid-fast bacilli (AFB) detection. The slide’s surface was ﬂooded with 0.3% Fucsina fenica and heated until steaming. The process was repeated 3–4 times. After 10 min, the slides were rinsed under a gentle ﬂow of water until all free stain was washed away. Decolorization was performed by ﬂooding the slides with 3% acid-alcohol for 3 min, and rinsing them thoroughly with water afterwards. Re-colorization was performed by covering the slides with 0.3% methylene blue for 30 s. The technique for the other specimens was similar, the only difference being the processing method of the pathological product (prior centrifugation). After washing and drying the slides, microscopic examination was performed using an optic microscope with an immersion lens (100×) and an ocular lens (10×). The slide was examined over the entire length of the smear. A minimum of 100 ﬁelds were examined before the smear was reported as negative (Table 1). 2.2. Bacterial Culture Technique A bacterial culture examination was conducted for all the samples using NaOH method, without centrifugation (dripping method). Genes 2023, 14, 1231 4 of 10 Table 1. Semiquantitative expression of the microscopic examination results. Number of AFB under Ziehl–Neelsen Staining Result 0 AFB 1–9 AFB/100 ﬁelds 10–99 AFB/100 ﬁelds 1–10 AFB/ﬁeld >10 AFB/ﬁeld Negative Positive, scanty (exact value) Positive 1+ Positive 2+ Positive 3+ From the pathological product, 2–3 mL of purulent particles were extracted using a Pasteur pipette and put into a sterile tube with a threaded cap. An equal amount of 4% NaOH with pH indicator was added. The capped tube was put in a mechanical agitator for 10–15 s. Then, the tube was left at room temperature for 15 min. Neutralization of the sample is performed using 8% HCl until the color turned greenish yellow (neutral pH). The culturing was performed using a single-use pipette. The used culture medium was Lowenstein–Jensen; for every sample, 3 medium tubes were used. After culturing, the tubes were left in a temperature-controlled room at 37 ◦C, with the cap half closed, at a 25–30◦ angle, for 2–5 days. The ﬁrst reading was taken after 48 h, leaving the tubes vertical afterwards, and eliminating the contaminated tubes. The cultures were monitored weekly until the end of the 8-week period (60 days) of incubation (Table 2). Table 2. Semiquantitative expression of the Lowenstein–Jensen solid medium culture results. Mycobacterium Growth Result Absence of colonies Under 30 colonies 30–100 colonies Over 100 colonies Uncountable conﬂated colonies 3 or 2 tubes contaminated and a tube without Bacterial growth Negative Positive, scanty (exact value) Positive 1+ Positive 2+ Positive 3+ Contaminated 2.3. Xpert MTB/RIF Ultra Test Technique Xpert MTB/RIF Ultra (Cepheid AB Röntgenvägen 5 171 54, Solna, Sweden) is an automatized molecular test using nested real-time PCR for the qualitative detection of M complex and rifampicin resistance, simultaneously. The primers of this test amplify a region of the rpoB gene containing 81 base-pairs in the core region. The probes are designed to distinguish between wild-type sequences and mutations in the core region, which are associated with rifampicin resistance. The tests were performed using Cepheid GeneXpert® Systems equipment (Cepheid, 904 Caribbean Drive, Sunnyvale, CA, USA), which automatizes and integrates the sample puriﬁcation, ampliﬁes the nucleic acids and detects the targeted sequence using RT-PCR. The system consisted of apparatus, a computer and dedicated software, and it was used for the execution of the test and visualization of the results. The system uses single- use GeneXpert® cartilages which contain the reactive, the RT-PCR process, a sample processing control (SPC) and a probe check control (PCC). Due to the autonomic nature of these cartridges, and the automatic processes, the likelihood of cross-contamination between samples is low. SPC has the role of controlling the bacterial processing and of monitoring the presence of the inhibitor in the PCR reaction. PCC checks the reactive rehydration, the PCR tube feeling, the probe integrity and the colorant stability. Xpert MTB/RIF Ultra simultaneously detects the presence of the M. tuberculosis (MTB) complex and rifampicin (RIF) resistance by amplifying the speciﬁc sequence form the rpoB gene, which is marked with ﬁve signaling molecules (probes A to E) for the mutations of the rifampicin resistance determining region (RRDR). Each signaling molecule was marked with a different ﬂuorophore. The cycle threshold (Ct) was set at 39.0 for the A, B and C probes and at 36.0 for the D and E probes [16]. Genes 2023, 14, 1231 5 of 10 The Xpert MTB/RIF Ultra test was performed for 536 samples during the duration of the study. For each test, 1 mL of sputum was used, which was sampled with a sterile pipette and transferred into a sealed sterile tube. A total of 2 mL of reactive was added with bactericide and mucus lysis properties. After 10 s of vigorous agitation and 10 min rest at room temperature, followed by further vigorous agitation and 5 min rest, a uniformly homogenized solution was obtained. The content of the tube was transferred to the reaction cartilage using the producer-provided pipette. The test took 90 min, and the results were displayed. GeneXpert® Instrument Systems generates results using preestablished algorithms. The interpretation of the measurements is found in Table 3. Table 3. Possible results of the Xpert MTB/RIF Ultra test. MTB detected/rifampicin resistance detected MTB detected/rifampicin resistance not detected MTB detected/rifampicin resistance indeterminate MTB not detected Invalid result 3. Results In the observed period a total of 862 patients suspected of TB infection were tested with Xpert MTB/RIF Ultra, Ziehl–Neelsen stain and culture on Lowenstein–Jensen medium. In 2018, 320 (37.1%) tests were performed, in 2019, 289 (33.5%) tests were performed and in 2020, 253 (29.4%) tests were performed. From the study sample, 643 (74.6%) were adults and 219 (25.4%) were pediatric patients. The collected pathological samples were as follows: 353 (41%)—sputum, 384 (44.5%)— induced sputum, 24 (2.8%)—bronchial aspirate, 73 (8.5%)—gastric aspirate, 7 (0.8%)— pleural ﬂuid, 11 (1.3%)—cerebral spinal ﬂuid and 10 (1.2%)—other pathological products (examples of such ﬂuids include synovial ﬂuid from joints and pus from abscesses located in various regions). Out of the 862 tested molecular samples, 306 (35.5%) were positive—MTB detected (121 positive samples in 2018, 132 in 2019 and 53 in 2020), and 556 (64.5%) were negative. In the microscopy test, 694 (80.5%) samples were negative and only 168 (19.5%) were positive. Regarding the culture, 299 (34.7%) had a positive culture, 560 (65%) were negative and 3 samples (0.3%) were contaminated (Table 4). Table 4. Distribution of negative and positive results in the three TB tests studied. Test Positive Negative Total Xpert MTB/RIF Ultra test Ziehl–Neelsen stain microscopy Culture on Lowenstein Jensen medium 306 (35.5%) 168 (19.5%) 299 (34.7%) 556 (64.5%) 694 (80.5%) 560 (65%) 862 (100%) 862 (100%) 859 (99.7%) * * In 3 cases, the culture was contaminated. Rifampicin resistance was encountered in 27 cases out of the 306 positive tests (8.82%); in two cases, indeterminate rifampicin resistance was found (0.65%). Out of the 306 patients with detected MTB in the molecular test, 284 (92.81%) had a positive result in the bacterial culture, 20 had a negative culture and 2 samples were contaminated. Of the 556 negative results in the molecular test, 15 had a positive culture. Based on these data, the sensitivity of the Xpert MTB/RIF Ultra test, when compared to the “gold standard” culture, was calculated to be 95%, while the speciﬁcity was determined to be 96.4% (Figure 1). Of the 168 positive result in the Ziehl–Neelsen stain microscopy, 164 (97.6%) had a positive result in the culture, 3 (1.8%) had a negative culture, and 1 (0.6%) sample was contaminated. Of the 694 negative microscopy result, 135 (19.5%) had a positive culture and 2 (0.3%) were contaminated. Based on these data, the microscopy (Ziehl-Neelsen stain) demonstrates a calculated sensitivity of 54.8% and a speciﬁcity of 99.5% (Figure 1). Genes 2023, 14, 1231 6 of 10 Figure 1. Crosstabulation of molecular and microscopy test results compared with the bacterial culture results. Out of the 168 positive results of the microscopy, 166 had a positive molecular test, and 2 samples were negative. In both cases, the culture was positive for mycobacteria other than tuberculosis (MOTT). In the 302 cases where the culture was not negative (299 positive samples and 3 contaminated samples), the median time at which the samples were declared positive was 30 days (mean: 34.07 days, min: 21 days, max: 60 days). The majority (135, 44.7%) of the samples were declared positive at the 21-day reading; 53 (17.5%) samples were declared positive after 30 days; 65 (21.5%) were declared positive after 45 days; and 49 (16.2%) sam- ples were declared positive at the 60-day reading. There is a statistically relevant correlation (p < 0.0001), inversely related, between the duration of the positive determination and the number of colonies isolated in the culture (Figure 2). Figure 2. Number of days elapsed from the suspicion of TB until the diagnosis established by the positive culture: indicators of the central tendency (mean: 34.07 days, median: 30 days, min: 21 days, max: 60 days); these values also represent, as all the patients with positive molecular test were immediately started on treatment, the days gained in the early treatment of TB using molecular tests for the diagnosis. Genes 2023, 14, x FOR PEER REVIEW 6 of 10 Rifampicin resistance was encountered in 27 cases out of the 306 positive tests (8.82%); in two cases, indeterminate rifampicin resistance was found (0.65%). Out of the 306 patients with detected MTB in the molecular test, 284 (92.81%) had a positive result in the bacterial culture, 20 had a negative culture and 2 samples were con-taminated. Of the 556 negative results in the molecular test, 15 had a positive culture. Based on these data, the sensitivity of the Xpert MTB/RIF Ultra test, when compared to the “gold standard” culture, was calculated to be 95%, while the specificity was deter-mined to be 96.4% (Figure 1). Figure 1. Crosstabulation of molecular and microscopy test results compared with the bacterial cul-ture results. Of the 168 positive result in the Ziehl–Neelsen stain microscopy, 164 (97.6%) had a positive result in the culture, 3 (1.8%) had a negative culture, and 1 (0.6%) sample was contaminated. Of the 694 negative microscopy result, 135 (19.5%) had a positive culture and 2 (0.3%) were contaminated. Based on these data, the microscopy (Ziehl-Neelsen stain) demonstrates a calculated sensitivity of 54.8% and a specificity of 99.5% (Figure 1). Out of the 168 positive results of the microscopy, 166 had a positive molecular test, and 2 samples were negative. In both cases, the culture was positive for mycobacteria other than tuberculosis (MOTT). In the 302 cases where the culture was not negative (299 positive samples and 3 con-taminated samples), the median time at which the samples were declared positive was 30 days (mean: 34.07 days, min: 21 days, max: 60 days). The majority (135, 44.7%) of the sam-ples were declared positive at the 21-day reading; 53 (17.5%) samples were declared pos-itive after 30 days; 65 (21.5%) were declared positive after 45 days; and 49 (16.2%) samples were declared positive at the 60-day reading. There is a statistically relevant correlation (p < 0.0001), inversely related, between the duration of the positive determination and the number of colonies isolated in the culture (Figure 2). Molecular positivetestsMolecularnegative testsMicroscopypositive testsMicroscopynegative testsPositive bacterial culture28415164135Negative bacterial culture205413557Contaminated sample20120100200300400500600Number of casesGenes 2023, 14, x FOR PEER REVIEW 7 of 10 Figure 2. Number of days elapsed from the suspicion of TB until the diagnosis established by the positive culture: indicators of the central tendency (mean: 34.07 days, median: 30 days, min: 21 days, max: 60 days); these values also represent, as all the patients with positive molecular test were im-mediately started on treatment, the days gained in the early treatment of TB using molecular tests for the diagnosis. 4. Discussion The early detection and prompt treatment of positive cases are the most effective measures in controlling the spread of tuberculosis [15]. The most reliable method of TB diagnostics is bacteriological culture, which is per-formed, in most cases, using sputum sampled directly, but other pathological products may also be used. The sampling process is essential in ensuring the quality of the result. The microscopic examination of the pathologic product is extremely relevant in the control of tuberculosis, helping to identify the patients with the highest contagion rate. This method aims to identify AFB in the pathologic product; the test is later confirmed via bacteriological culture. However, microscopic examination using the Ziehl–Neelsen stain-ing technique, although it is a fast, cheap method, has a low sensitivity, and it is not able to distinguish between MTBC and other non-tuberculosis mycobacteria [17]. For AFB to be detected, at least 104 CFU/mL must exist in the pathologic product [18]. Culture confir-mation of a TB infection may take 21 to 60 days. Furthermore, neither microscopic exam-ination nor culture can distinguish drug-susceptible TB strains from drug-resistant ones [12]. The testing using nucleic acid amplification tests offered quick and precise diagnosis of tuberculosis, with a sensitivity rate of 95% and a specificity rate of 96.4%. Using this test shortens the isolation period of suspected patients and prevents useless treatment [17,19]. The sputum samples with negative microscopic examination results but with a later positive culture had a lower bacterial load compared with the samples with positive mi-croscopic examination results. With high sensitivity, the NAAT method can detect MTB even in microscopic-negative samples. TB patients coinfected with HIV are known to have a low bacterial load compared with the patients without HIV, even though these patients, untreated, have a more ag-gressive form of the disease [16]. This study cohort did not include any HIV patients. The utilization of NAAT was initially approved in 1995 for patients with positive microscopic examination and clinical signs suggestive of TB [19]. The recent progress in molecular testing for MTBC includes the Xpert MTB/RIF Ultra test, which allows for the simultaneous detection of tuberculous bacilli and rifampicin resistance. Patients with neg-ative result following this test can avoid isolation, and those with positive results may benefit from early treatment [20]. Genes 2023, 14, 1231 7 of 10 4. Discussion The early detection and prompt treatment of positive cases are the most effective measures in controlling the spread of tuberculosis [15]. The most reliable method of TB diagnostics is bacteriological culture, which is per- formed, in most cases, using sputum sampled directly, but other pathological products may also be used. The sampling process is essential in ensuring the quality of the result. The microscopic examination of the pathologic product is extremely relevant in the control of tuberculosis, helping to identify the patients with the highest contagion rate. This method aims to identify AFB in the pathologic product; the test is later conﬁrmed via bacteriological culture. However, microscopic examination using the Ziehl–Neelsen staining technique, although it is a fast, cheap method, has a low sensitivity, and it is not able to distinguish between MTBC and other non-tuberculosis mycobacteria [17]. For AFB to be detected, at least 104 CFU/mL must exist in the pathologic product [18]. Culture conﬁrmation of a TB infection may take 21 to 60 days. Furthermore, neither microscopic examination nor culture can distinguish drug-susceptible TB strains from drug-resistant ones [12]. The testing using nucleic acid ampliﬁcation tests offered quick and precise diagnosis of tuberculosis, with a sensitivity rate of 95% and a speciﬁcity rate of 96.4%. Using this test shortens the isolation period of suspected patients and prevents useless treatment [17,19]. The sputum samples with negative microscopic examination results but with a later positive culture had a lower bacterial load compared with the samples with positive microscopic examination results. With high sensitivity, the NAAT method can detect MTB even in microscopic-negative samples. TB patients coinfected with HIV are known to have a low bacterial load compared with the patients without HIV, even though these patients, untreated, have a more aggressive form of the disease [16]. This study cohort did not include any HIV patients. The utilization of NAAT was initially approved in 1995 for patients with positive microscopic examination and clinical signs suggestive of TB [19]. The recent progress in molecular testing for MTBC includes the Xpert MTB/RIF Ultra test, which allows for the simultaneous detection of tuberculous bacilli and rifampicin resistance. Patients with negative result following this test can avoid isolation, and those with positive results may beneﬁt from early treatment [20]. The advantages of NAAT include the possibility of early diagnosis and the prompt initiation of treatment, resulting a shorter period of contagion. Moreover, the quick dif- ferentiation of patients with MTBC from those infected with non-tuberculosis mycobac- teria prevents inadequate and useless treatments and useless investigations of patients’ families [21]. However, there are some limitations to molecular testing interpretations: these meth- ods can have slightly lower sensitivity than bacterial cultures; a negative molecular result does not absolutely exclude the diagnosis of tuberculosis. Furthermore, some sporadic errors in the system may lead to false-positive results in molecular testing [21]. Conven- tional microscopy and culture remain essential in the evaluation of disease response to treatment [12]. In this study, we reported several situations in which we had a positive molecular test using the Xpert MTB/RIF Ultra method that had a negative microscopic examination and negative culture. We also reported situations with negative molecular testing and negative microscopy but with positive culture. As can be seen in Figure 3, molecular testing has superior sensitivity compared with microscopic examination (95% compared with 54.8%). Regarding speciﬁcity, the two methods (molecular and microscopic) had similar results (96.4% and 99.5%, respectively). From this, it can be concluded that molecular testing has an important role in the early diagnosis of TB. In cases in which the Xpert MTB/RIF Ultra test was positive and the initial microscopy was negative, the initialization of the treatment would have been delayed by an average Genes 2023, 14, 1231 8 of 10 of 34.07 days. In these situations, molecular testing enables the prompt initialization of treatment, which has an impact on the evolution of the disease and the spreading of the disease. Similar results were reported in previous studies, such as Laraque et al. or Luetkemeyer et al. [22,23], but they are slightly different from the CDC reports that cite a 50–80% detection rate in the case of molecular tests performed on negative samples following microscopic examination [24]. Figure 3. Sensitivity and speciﬁcity comparison between molecular testing and microscopic examina- tion in the diagnosis of TB. 5. Conclusions The results of the present study validate the recent WHO recommendations; the implementation of molecular testing in TB laboratories leads to important increases in early diagnostics, and has superior sensitivity and similar speciﬁcity to microscopic examination. Molecular testing allows for a quicker diagnosis compared with bacterial culture (90 min vs. weeks), which leads to prompter isolation and treatment of infected patients. The capacity to distinguish between M. tuberculosis and non-tuberculosis mycobacteria shortens the isolation period and prevents the unnecessary treatment of suspected patients. Molecular testing can identify rifampicin-resistant strains (and other resistances), allowing for a personalized approach to the treatment of TB patients. Author Contributions: Conceptualization, C.S. and M.S.; methodology, A.B.B.; software, A.I.; val- idation, C.S., M.S. and A.B.B.; formal analysis, C.P.M.; investigation, M.S. and A.-M.D.; resources, C.S. and L.N.; data curation, A.I. and A.B.B.; writing—original draft preparation, C.S. and A.I.; writing—review and editing, A.B.B. and L.N.; visualization, A.I. and C.P.M.; supervision, C.S.; project administration, A.I. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Oradea County Emergency Clinical Hospital (14613/27 April 2023). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: Not applicable. Acknowledgments: The article publishing fees were funded by the University of Oradea. Conﬂicts of Interest: The authors declare no conﬂict of interest. Genes 2023, 14, x FOR PEER REVIEW 8 of 10 The advantages of NAAT include the possibility of early diagnosis and the prompt initiation of treatment, resulting a shorter period of contagion. Moreover, the quick differ-entiation of patients with MTBC from those infected with non-tuberculosis mycobacteria prevents inadequate and useless treatments and useless investigations of patients’ fami-lies [21]. However, there are some limitations to molecular testing interpretations: these meth-ods can have slightly lower sensitivity than bacterial cultures; a negative molecular result does not absolutely exclude the diagnosis of tuberculosis. Furthermore, some sporadic errors in the system may lead to false-positive results in molecular testing [21]. Conven-tional microscopy and culture remain essential in the evaluation of disease response to treatment [12]. In this study, we reported several situations in which we had a positive molecular test using the Xpert MTB/RIF Ultra method that had a negative microscopic examination and negative culture. We also reported situations with negative molecular testing and negative microscopy but with positive culture. As can be seen in Figure 3, molecular testing has superior sensitivity compared with microscopic examination (95% compared with 54.8%). Regarding specificity, the two methods (molecular and microscopic) had similar results (96.4% and 99.5%, respectively). From this, it can be concluded that molecular testing has an important role in the early diagnosis of TB. Figure 3. Sensitivity and specificity comparison between molecular testing and microscopic exami-nation in the diagnosis of TB. In cases in which the Xpert MTB/RIF Ultra test was positive and the initial micros-copy was negative, the initialization of the treatment would have been delayed by an av-erage of 34.07 days. In these situations, molecular testing enables the prompt initialization of treatment, which has an impact on the evolution of the disease and the spreading of the disease. Similar results were reported in previous studies, such as Laraque et al. or Luet-kemeyer et al. [22,23], but they are slightly different from the CDC reports that cite a 50–80% detection rate in the case of molecular tests performed on negative samples following microscopic examination [24]. Genes 2023, 14, 1231 References 9 of 10 1. World Health Organization. Global Tuberculosis Report 2020. Available online: https://www.who.int/publications/i/item/9789 240013131 (accessed on 4 March 2022). 2. Houben, R.M.; Dodd, P.J. The Global Burden of Latent Tuberculosis Infection: A Re-estimation Using Mathematical Modelling. PLoS Med. 2016, 13, e1002152. [CrossRef] [PubMed] 4. 5. 3. Wright, A.; Zignol, M.; Van Deun, A.; Falzon, D.; Gerdes, S.R.; Feldman, K.; Hoffner, S.; Drobniewski, F.; Barrera, L.; van Soolingen, D.; et al. Epidemiology of antituberculosis drug resistance 2002–07: An updated analysis of the Global Project on Anti-Tuberculosis Drug Resistance Surveillance. Lancet 2009, 373, 1861–1873. [CrossRef] [PubMed] European Centre for Disease Prevention and Control/WHO Regional Ofﬁce for Europe. Tuberculosis Surveillance and Monitoring in Europe 2020–2018 Data. Available online: https://www.ecdc.europa.eu/en/publications-data/tuberculosis-surveillance-and- monitoring-europe-2020-2018-data (accessed on 4 March 2022). Analiza de Situatie TB 2019. Available online: https://insp.gov.ro/sites/cnepss/wp-content/uploads/2019/04/Analiza-de- situatie-tb2019.pdf (accessed on 4 November 2021). Chiotan, D.; Popa, C.G. Ghidul Pentru Identiﬁcarea Suspect,ilor TB s, i Referire a Pacient,ilor cu TB Pentru Tot,i Furnizorii de Servicii (Inclusiv Identiﬁcarea Activă a Cazurilor de TB în Rândul Populat,iilor Vulnerabile. Available online: https://www.raa. ro/wp-content/uploads/2018/05/Ghid-de-identiﬁcare-si-referire-a-persoanelor-suspecte-de-TB-2017-RO.pdf (accessed on 4 November 2021). Uplekar, M.; Raviglione, M. WHO’s End TB Strategy: From stopping to ending the global TB epidemic. Indian J. Tuberc. 2015, 62, 196–199. [CrossRef] [PubMed] Lönnroth, K.; Raviglione, M. The WHO’s new End TB Strategy in the post-2015 era of the Sustainable Development Goals. Trans. R. Soc. Trop. Med. Hyg. 2016, 110, 148–150. [CrossRef] [PubMed] 7. 6. 8. 9. World Health Organization. Implementing the End TB Strategy: The Essentials; World Health Organization: Geneva, 10. Switzerland, 2022. Sava, C.N.; Bodog, T.M.; Niulas, L.R.; Iuhas, A.R.; Marinau, C.P.; Negrut, N.; Balmos, A.B.; Pasca, B.; Roman, N.A.; Nistor-Cseppento, C.D. Biomarker Changes in Pediatric Patients With COVID-19: A Retrospective Study from a Single Center Database. Vivo 2022, 36, 2813–2822. [CrossRef] [PubMed] 11. Wu, C.W.; Wu, Y.K.; Lan, C.C.; Yang, M.C.; Dong, T.Q.; Tzeng, I.S.; Hsiao, S.S. Impact of nucleic acid ampliﬁcation test on pulmonary tuberculosis notiﬁcations and treatments in Taiwan: A 7-year single-center cohort study. BMC Infect. Dis. 2019, 19, 726. [CrossRef] [PubMed] 12. World Health Organization. Module 3: Diagnosis-rapid diagnostics for tuberculosis detention. In WHO Operational Handbook on 13. 14. Tuberculosis; World Health Organization: Geneva, Switzerland, 2021. Steingart, K.R.; Schiller, I.; Horne, D.J.; Pai, M.; Boehme, C.C.; Dendukuri, N. Xpert® MTB/RIF assay for pulmonary tuberculosis and rifampicin resistance in adults. Cochrane Database Syst. Rev. 2014, 2014, CD009593. [CrossRef] [PubMed] Iuhas, A.; Jurca, C.; Kozma, K.; Riza, A.-L.; Streat,ă, I.; Petches, i, C.; Dan, A.; Sava, C.; Balmos, , A.; Marinău, C.; et al. PAH Pathogenic Variants and Clinical Correlations in a Group of Hyperphenylalaninemia Patients from North-Western Romania. Diagnostics 2023, 13, 1483. [CrossRef] [PubMed] 15. Ministerul Sănătăt,ii din România. Ghidul Metodologic de Implementare a Programului Na¸tional de Prevenire, Supraveghere ¸si Control al Tuberculozei. Available online: http://old.ms.ro/documente/GHID%20Metodologic%20de%20implementare%20a% 20Programului%20national%20de%20prevenire,%20supraveghere%20si%20control%20al%20tuberculozei%202015_15424_183 33.pdf (accessed on 4 November 2021). 16. Xpert®MTB/RIF Assay. Available online: https://www.cepheid.com/en_US/package-inserts/1608 (accessed on 5 March 2022). 17. Bourgi, K.; Patel, J.; Samuel, L.; Kieca, A.; Johnson, L.; Alangaden, G. Clinical Impact of Nucleic Acid Ampliﬁcation Testing in the Diagnosis of Mycobacterium Tuberculosis: A 10-Year Longitudinal Study. Open Forum Infect. Dis. 2017, 4, ofx045. [CrossRef] [PubMed] 18. Allen, B.W.; Mitchison, D.A. Counts of viable tubercle bacilli in sputum related to smear and culture gradings. Med. Lab. Sci. 1992, 49, 94–98. [PubMed] 19. Dinnes, J.; Deeks, J.; Kunst, H.; Gibson, A.; Cummins, E.; Waugh, N.; Drobniewski, F.; Lalvani, A. A systematic review of rapid diagnostic tests for the detection of tuberculosis infection. Health Technol. Assess. 2007, 11, 1–196. [CrossRef] [PubMed] 20. World Health Organization. Automated real-time nucleic acid ampliﬁcation technology for rapid and simultaneous detection of tuberculosis and rifampicin resistance. In Xpert MTB/RIF Assay for the Diagnosis of Pulmonary and Extrapulmonary TB in Adults and Children: Policy Update; World Health Organization: Geneva, Swizterland, 2013; Available online: http://apps.who.int/iris/ handle/10665/112472?locale=zh (accessed on 5 March 2022). 21. CDC. Report of an Expert Consultation on the Uses of Nucleic Acid Ampliﬁcation Tests for the Diagnosis of Tuberculosis. 2012. Available online: https://www.cdc.gov/tb/publications/guidelines/ampliﬁcation_tests/considerations.htm (accessed on 5 March 2022). 22. Laraque, F.; Griggs, A.; Slopen, M.; Munsiff, S.S. Performance of nucleic acid ampliﬁcation tests for diagnosis of tuberculosis in a large urban setting. Clin. Infect. Dis. 2009, 49, 46–54. [CrossRef] [PubMed] Genes 2023, 14, 1231 10 of 10 23. Luetkemeyer, A.F.; Firnhaber, C.; Kendall, M.A.; Wu, X.; Mazurek, G.H.; Benator, D.A.; Arduino, R.; Fernandez, M.; Guy, E.; Johnson, P.; et al. Evaluation of Xpert MTB/RIF Versus AFB Smear and Culture to Identify Pulmonary Tuberculosis in Patients with Suspected Tuberculosis from Low and Higher Prevalence Settings. Clin. Infect. Dis. 2016, 62, 1081–1088. [CrossRef] [PubMed] 24. Centers for Disease Control and Prevention (CDC). Updated guidelines for the use of nucleic acid ampliﬁcation tests in the diagnosis of tuberculosis. Morb. Mortal. Wkly. Rep. 2009, 58, 7–10. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijms241210104	Article An In Vitro Alveolar Model Allows for the Rapid Assessment of Particles for Respiratory Sensitization Potential Matthew Gibb 1 and Christie M. Sayes 1,2,* 1 Institute of Biomedical Studies, Baylor University, Waco, TX 76798, USA; [email protected] 2 Department of Environmental Science, Baylor University, Waco, TX 76798, USA * Correspondence: [email protected]; Tel.: +1-254-710-3469 Abstract: Dust, both industrial and household, contains particulates that can reach the most distal aspects of the lung. Silica and nickel compounds are two such particulates and have known proﬁles of poor health outcomes. While silica is well-characterized, nickel compounds still need to be fully understood for their potential to cause long-term immune responses in the lungs. To assess these hazards and decrease animal numbers used in testing, investigations that lead to veriﬁable in vitro methods are needed. To understand the implications of these two compounds reaching the distal aspect of the lungs, the alveoli, an architecturally relevant alveolar model consisting of epithelial cells, macrophages, and dendritic cells in a maintained submerged system, was utilized for high throughput testing. Exposures include crystalline silica (SiO2) and nickel oxide (NiO). The endpoints measured included mitochondrial reactive oxygen species and cytostructural changes assessed via confocal laser scanning microscopy; cell morphology evaluated via scanning electron microscopy; biochemical reactions assessed via protein arrays; transcriptome assessed via gene arrays, and cell surface activation markers evaluated via ﬂow cytometry. The results showed that, compared to untreated cultures, NiO increased markers for dendritic cell activation, trafﬁcking, and antigen presentation; oxidative stress and cytoskeletal changes, and gene and cytokine expression of neutrophil and other leukocyte chemoattractants. The chemokines and cytokines CCL3, CCL7, CXCL5, IL-6, and IL-8 were identiﬁed as potential biomarkers of respiratory sensitization. Keywords: cellular activation sensitization; pulmonary exposure; immunotoxicology; in vitro; dendritic cells; 1. Introduction The lungs are a complex network of cell types involving cellular crosstalk, commu- nication, and varying motions (e.g., mucociliary ladder and surfactants). Because of this heterogeneity of cellular structure, one of the most critical aspects of pulmonary in vitro study is the ability to adequately maintain relevant cellular architecture in selected models [1]. With the primary function of the lungs being gas exchange, it is critically important to test and assess the potential for poor health outcomes associated with inhaled air. Inhaled air can consist of chemicals and particulates that, depending on various physicochemical properties, can deposit on cells and affect cellular responses throughout the respiratory system [2]. Of the known potential health outcomes, allergic-type reactions are of primary concern, as they can lead to life-long issues or be severe enough to cause anaphylaxis and possibly death. Respiratory sensitization refers to the onset of inﬂammatory responses, including airway hypersensitivity, asthma, bronchiolitis, and more [3]. Sensitization, whether in the skin or lungs, involves two consecutive steps: (i) In- duction, where an exposure leads to a cascade of innate and adaptive cells activating and maturating to provide a speciﬁc elevated immune response on secondary exposure; and (ii) Elicitation, where an exacerbated immune response occurs on second exposure, leading to a variety of inﬂammation, as seen in acute and chronic asthma, as well as anaphylaxis [4]. Citation: Gibb, M.; Sayes, C.M. An In Vitro Alveolar Model Allows for the Rapid Assessment of Particles for Respiratory Sensitization Potential. Int. J. Mol. Sci. 2023, 24, 10104. https://doi.org/10.3390/ ijms241210104 Academic Editors: Daniel P. Potaczek, Malgorzata Wygrecka, Stefan Hadži´c and Bilal Alashkar Alhamwe Received: 7 April 2023 Revised: 1 May 2023 Accepted: 9 May 2023 Published: 14 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Mol. Sci. 2023, 24, 10104. https://doi.org/10.3390/ijms241210104 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10104 2 of 19 Typically, the order of events requires a minimum of two exposures for any allergic reaction to occur [5]. Currently, most research into respiratory sensitization has focused on low molecular weight (LMW) and high molecular weight (HMW) chemicals [6]. Most known sensitizers, including LMW chemicals, are too small to create an immune response independently and require protein binding to elicit an immune response. The sensitizer, a hapten, and a protein bind are needed to form a hapten–protein complex recognizable by the immune system [5,7]. Importantly, alveolar macrophages and surrounding epithelial cells can provide the proteins necessary to form these complexes [8]. Identifying and understanding the mechanisms associated with respiratory sensiti- zation has primarily focused on rodent studies or gathered from population-level studies in humans [9–11]. These kinds of studies are costly, time-consuming, and need more translatability to humans [9–13]. The ability to utilize human cells in vitro has helped to recapitulate human responses better. Furthermore, the ability to better mimic in vivo archi- tecture while working with human-derived cells allows for increased capacity for direct translation from in vitro to in vivo outcomes without dealing with the dynamic nature of in vivo studies [1,14,15]. Crystalline silica (SiO2) is ubiquitous in the earth’s crust and is known to lead to adverse pulmonary health through silicosis, where trapped silica lead to inﬂammation; scarring; lung cancer; chronic obstructive pulmonary disease (COPD), and kidney dis- ease [16]. Common exposures include industries that involve sand, mortar, stone, and concrete, where the respirable form of SiO2 is created from sawing, drilling, crushing, grinding, and cutting [17]. SiO2 is not known to lead to respiratory sensitization despite decades of research on human populations after exposure; however, it is known to be a respiratory irritant that leads to oxidative stress for all cell types within the lungs on exposure [18]. Nanomaterials have been shown to target immune cells to varying degrees, and with nanometals being produced in vast quantities, understanding their effect on human health is imperative [19–23]. Nickel compounds, speciﬁcally nickel oxide (NiO), have been shown to induce adverse respiratory effects such as asthma and eosinophilic inﬂammation [24]. IgE antibody tests are frequently utilized to assess for a portion of sensitizing reactions. NiO has been shown to increase serum IgE levels when using bulk and nano-scale material [24]. Animal modeling can provide insights into possible human responses, but difﬁculties arise when investigating the respiratory sensitizing potential. For instance, rats require a much higher level of the test compound to elicit a broncho-restrictive response, and guinea pigs will produce IgG1 rather than IgE to known respiratory allergens [6,25,26]. Sensitization can occur anywhere within the lungs; however, a single model is cur- rently incapable of recapitulating the lungs due to the complexity of the lung cellular architecture. Because gas exchange occurs at the alveolar space, understanding immune responses in this compartment is crucial to potential preventatives, interventions, and treatments. Within the alveolar region, there are three main cell types: epithelial cells (both Type I and II) and immune cells, speciﬁcally alveolar macrophages (AMs) in the luminal space and dendritic cells (DCs), which are scattered among the basement membrane [27]. Previously, an easy, reliable, and veriﬁed cell culture model that can be adopted by any lab capable of performing molecular toxicology studies was used to study a known chemical respiratory sensitizer, isophorone diisocyanate (IPDI), and a known cell activator, phorbal 12-myristate 13-acetate (PMA) and ionomycin [6]. Here, the same model and endpoints (morphology, biochemical perturbations, and transcriptome) were chosen to assess if the model can differentiate between a known irritating respiratory particulate (SiO2) and a suspected sensitizing respiratory particulate (NiO). Like the previous study, the results suggest that multiple techniques and endpoints can show objective distinctions in immune responses after different particulate exposures. Int. J. Mol. Sci. 2023, 24, 10104 3 of 19 2. Results The model setup is based on in vivo alveolar cellular architecture, which contains epithelial cells (ECs), alveolar macrophages (AMs), and dendritic cells (DCs). Figure 1 shows the developmental process from aerosol exposure to in vitro recapitulation of cel- lular components and location in the Transwell®. The characteristics of toxicology and immunology assessments, as well as basal properties, have been studied. This includes evaluating transepithelial electrical resistance (TEER) [28–32]. Figure 1. Model development. The model is based on real-world exposure to aerosol, where the ﬁnal deposition is in the alveolar space. The model depicts in vivo architecture. Cultured cells are arranged in a Transwell® and include differentiated U937 cells (as alveolar macrophages, AMs), A549 cells (as type 1 epithelial cells, ECs), and JawsII cells (as dendritic cells, DCs). Figure 2 shows the scanning electron micrographs of SiO2 and NiO along with the quantitative physicochemical properties of each material listed in the table. SiO2 had an av- erage size of approximately 3 µm, with NiO having an average size of approximately 80 nm. The surface charge of SiO2 averaged −56.8 and NiO −9.05, with the hydrodynamic diame- ter at 1.807 µm for SiO2 and 0.963 µm for NiO. Figure 3 shows scanning electron micrographs of EC and AM cells in the apical cham- ber (Figure 3A–C) and DCs in the basolateral chamber (Figure 3D–F). A normal unperturbed epithelial cell structure is seen by conﬂuent monolayers with ﬂattened morphology within untreated cultures (Figure 3A). In contrast, disruption and increased size of epithelial cells within the monolayer (an indication of apoptotic cells) are seen in SiO2- and NiO-treated cultures, respectively (Figure 3B,C). Increases in microvillar protrusions on the membrane surface are also visible in SiO2- and NiO-treated cultures. Dendritic cell size and dendrite length increased in the basal compartments (Figure 3D–F) of treated versus untreated cultures. Compared to treated cells, untreated cultures show DCs appearing smaller in size with fewer and shorter dendrites per cell. Confocal laser scanning microscopy (CLSM) was performed to measure reactive oxygen species (ROS), nuclear binding activity, and cytoskeletal structure. Figure 4 shows micrographs imaging DNA via NucBlue live cell stain, cytoskeleton (F-actin) via ActinGreen 488® ReadyProbes, and mitochondrial ROS via MitoTracker Red CMXRos. Quantiﬁcation of mean ﬂuorescence intensity (MFI) was assessed using the Olympus CellSens software V4.2. Resultant MFI calculations were compared across untreated (54.71, untreated) vs. SiO2-treated (74.87, SiO2) vs. NiO-treated (66.61, NiO) cultures. While both treated cultures showed increased nuclear binding activity, no statistical signiﬁcance was seen. Only NiO treatment induced signiﬁcant increases in ROS. ROS from all exposures were as follows: untreated, 19.02; SiO2-treated, 24.49; NiO-treated, 60.94. F-actin, a measure of proliferation, increased signiﬁcantly in both SiO2- and NiO-treated cultures compared to untreated Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 3 of 21 2. Results The model setup is based on in vivo alveolar cellular architecture, which contains epithelial cells (ECs), alveolar macrophages (AMs), and dendritic cells (DCs). Figure 1 shows the developmental process from aerosol exposure to in vitro recapitulation of cel-lular components and location in the Transwell®. The characteristics of toxicology and immunology assessments, as well as basal properties, have been studied. This includes evaluating transepithelial electrical resistance (TEER) [28–32]. Figure 1. Model development. The model is based on real-world exposure to aerosol, where the ﬁnal deposition is in the alveolar space. The model depicts in vivo architecture. Cultured cells are arranged in a Transwell® and include diﬀerentiated U937 cells (as alveolar macrophages, AMs), A549 cells (as type 1 epithelial cells, ECs), and JawsII cells (as dendritic cells, DCs). Figure 2 shows the scanning electron micrographs of SiO2 and NiO along with the quantitative physicochemical properties of each material listed in the table. SiO2 had an average size of approximately 3 µm, with NiO having an average size of approximately 80 nm. The surface charge of SiO2 averaged −56.8 and NiO −9.05, with the hydrodynamic diameter at 1.807 µm for SiO2 and 0.963 µm for NiO. Figure 3 shows scanning electron micrographs of EC and AM cells in the apical cham-ber (Figure 3A–C) and DCs in the basolateral chamber (Figure 3D–F). A normal unper-turbed epithelial cell structure is seen by conﬂuent monolayers with ﬂattened morphology within untreated cultures (Figure 3A). In contrast, disruption and increased size of epithe-lial cells within the monolayer (an indication of apoptotic cells) are seen in SiO2- and NiO-treated cultures, respectively (Figure 3B,C). Increases in microvillar protrusions on the membrane surface are also visible in SiO2- and NiO-treated cultures. Dendritic cell size and dendrite length increased in the basal compartments (Figure 3D–F) of treated versus untreated cultures. Compared to treated cells, untreated cultures show DCs appearing smaller in size with fewer and shorter dendrites per cell. Confocal laser scanning microscopy (CLSM) was performed to measure reactive ox-ygen species (ROS), nuclear binding activity, and cytoskeletal structure. Figure 4 shows micrographs imaging DNA via NucBlue live cell stain, cytoskeleton (F-actin) via Act-inGreen 488® ReadyProbes, and mitochondrial ROS via MitoTracker Red CMXRos. Quan-tiﬁcation of mean ﬂuorescence intensity (MFI) was assessed using the Olympus CellSens software V4.2. Resultant MFI calculations were compared across untreated (54.71, un-treated) vs. SiO2-treated (74.87, SiO2) vs. NiO-treated (66.61, NiO) cultures. While both treated cultures showed increased nuclear binding activity, no statistical signiﬁcance was seen. Only NiO treatment induced signiﬁcant increases in ROS. ROS from all exposures were as follows: untreated, 19.02; SiO2-treated, 24.49; NiO-treated, 60.94. F-actin, a meas-ure of proliferation, increased signiﬁcantly in both SiO2- and NiO-treated cultures com-pared to untreated cultures (untreated, 43.02; SiO2, 78.37; NiO, 74.09). NiO treatment showed signiﬁcant increases in ROS and F-actin compared to untreated, while SiO2-treated cultures only showed signiﬁcant increases in F-actin. Int. J. Mol. Sci. 2023, 24, 10104 4 of 19 cultures (untreated, 43.02; SiO2, 78.37; NiO, 74.09). NiO treatment showed signiﬁcant increases in ROS and F-actin compared to untreated, while SiO2-treated cultures only showed signiﬁcant increases in F-actin. Figure 2. Physicochemical characterization of the materials used in the study. (A) Scanning electron microscopy (SEM) image of irritating crystalline silica (SiO2), (B) SEM image of suspected sensi- tizer nickel oxide (NiO). Scale bars represent 500 nm in both micrographs. The table below the images lists the quantitative analyses of SiO2’s and NiO’s physicochemical properties. The table includes properties as dry powders (e.g., surface area and density as provided by the manufacturer); properties after suspension in ultrapure deionized water (e.g., hydrodynamic diameter and zeta potential and dispersity index), and properties after suspension in cAMEM (cell culture media) (e.g., hydrodynamic diameter, zeta potential, and disperity index. These data were collected using dynamic light scattering. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of 21 Figure 2. Physicochemical characterization of the materials used in the study. (A) Scanning electron microscopy (SEM) image of irritating crystalline silica (SiO2), (B) SEM image of suspected sensitizer nickel oxide (NiO). Scale bars represent 500 nm in both micrographs. The table below the images lists the quantitative analyses of SiO2’s and NiO’s physicochemical properties. The table includes properties as dry powders (e.g., surface area and density as provided by the manufacturer); prop-erties after suspension in ultrapure deionized water (e.g., hydrodynamic diameter and zeta poten-tial and dispersity index), and properties after suspension in cAMEM (cell culture media) (e.g., hy-drodynamic diameter, zeta potential, and disperity index. These data were collected using dynamic light scattering. A. Irritating Crystalline Silica (SiO2)NiOSiO2PropertyCharacterization as a dry powder8.0 m2/g0.648 m2/gSurface area6.67 g/cm32.69 g/cm3DensityCharacterization after suspension and sonication in water963 nm1807 nmHydrodynamic diameter−9.05 mV−56.8 mVZeta potential0.5870.762Dispersity indexCharacterization after suspension and sonication in cell culture media1651 nm2558 nmHydrodynamic diameter−7.03 mV−8.92 mVZeta potential0.7870.877Dispersity indexB. Sensitizing Nickel Oxide (NiO)Int. J. Mol. Sci. 2023, 24, 10104 5 of 19 Figure 3. Cell morphology is indicative of cell activation. (Above) Scanning electron micrographs of (A) naïve culture, (B) SiO2-treated culture, and (C) NiO-treated culture in an apical chamber. Scanning electron micrographs of (D) naïve culture, (E) crystalline silica-treated culture, and (F) nickel oxide-treated culture in the basolateral chamber. Alveolar macrophages are seen with yellow arrowheads. The scale bar denotes 50 µm. All images were taken at 1200× magniﬁcation. Scale bars in large images are 50 µm, while scale bars in inset images are 4 µm. Figure 4. Biochemical analyses via confocal laser scanning micrographs of cells. The nucleus is stained with DAPI (blue), mitochondrial ROS with MitoTracker™ Red CMXRos, and F-actin with ActinGreen™ 488 ReadyProbes™ Reagent. Images were taken at 60× magniﬁcation. The scale bar denotes 20 µm. Quantiﬁcation of ﬂuorescence was performed with CellSens software V4.2. The inset letters of bar graphs (in panels M–O) correspond to each micrograph label (in panels A–L). Signiﬁcance is noted: * p ≤ 0.001, and p ≤ 0.0001. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 5 of 21 Figure 3. Cell morphology is indicative of cell activation. (Above) Scanning electron micrographs of (A) naïve culture, (B) SiO2-treated culture, and (C) NiO-treated culture in an apical chamber. Scan-ning electron micrographs of (D) naïve culture, (E) crystalline silica-treated culture, and (F) nickel oxide-treated culture in the basolateral chamber. Alveolar macrophages are seen with yellow arrow-heads. The scale bar denotes 50 µm. All images were taken at 1200× magniﬁcation. Scale bars in large images are 50 µm, while scale bars in inset images are 4 µm. Figure 4. Biochemical analyses via confocal laser scanning micrographs of cells. The nucleus is stained with DAPI (blue), mitochondrial ROS with MitoTracker™ Red CMXRos, and F-actin with ActinGreen™ 488 ReadyProbes™ Reagent. Images were taken at 60× magniﬁcation. The scale bar denotes 20 µm. Quantiﬁcation of ﬂuorescence was performed with CellSens software V4.2. The inset letters of bar graphs (in panels M–O) correspond to each micrograph label (in panels A–L). Signiﬁ-cance is noted: * p ≤ 0.001, and ** p ≤ 0.0001. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 5 of 21 Figure 3. Cell morphology is indicative of cell activation. (Above) Scanning electron micrographs of (A) naïve culture, (B) SiO2-treated culture, and (C) NiO-treated culture in an apical chamber. Scan-ning electron micrographs of (D) naïve culture, (E) crystalline silica-treated culture, and (F) nickel oxide-treated culture in the basolateral chamber. Alveolar macrophages are seen with yellow arrow-heads. The scale bar denotes 50 µm. All images were taken at 1200× magniﬁcation. Scale bars in large images are 50 µm, while scale bars in inset images are 4 µm. Figure 4. Biochemical analyses via confocal laser scanning micrographs of cells. The nucleus is stained with DAPI (blue), mitochondrial ROS with MitoTracker™ Red CMXRos, and F-actin with ActinGreen™ 488 ReadyProbes™ Reagent. Images were taken at 60× magniﬁcation. The scale bar denotes 20 µm. Quantiﬁcation of ﬂuorescence was performed with CellSens software V4.2. The inset letters of bar graphs (in panels M–O) correspond to each micrograph label (in panels A–L). Signiﬁ-cance is noted: * p ≤ 0.001, and **** p ≤ 0.0001. Int. J. Mol. Sci. 2023, 24, 10104 6 of 19 Transcriptome related to innate and adaptive cytokines was performed on delta Ct values and normalized to the reference gene ubiquitin C (UBC) by subtracting gene(s) of interest from the reference gene. Using delta Ct values where higher values represent increased expression, heatmaps were created to compare transcriptomics across the array of cytokine-related genes. A comparison between SiO2- and NiO-treated cultures revealed several inﬂammatory genes associated with inﬂammatory responses and associated ex- plicitly with cell activation and recruitment (CCL1, CCL3, CNTF, CSF2, FASLG, IL-5, IL-8, OSM, IL-12b, IL-17, LIF, and TNF) were upregulated in NiO treatment relative to untreated and SiO2-treated cultures, indicating possible sensitizing potential. Figure 5 shows the transcriptome heatmaps for each respective treatment. For ECs and AMs in the apical chamber, the following genes were upregulated in NiO compared to SiO2 and untreated cultures: BMP6, CCL1, CCL2, CCL3, CCL17, CCL18, CCL19, CCL20, CNTF, CXCL1, CXCL2, CXCL5, CXCL9, CXCL10, CXCL13, IL-1RN, IL-1α, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12β, IL-15, IL-16, IL-17α, IL-17F, IL-22, MSTN, OSM, TGFβ2, THPO, TNF, TNFRSF11, TNFSF10, TNFSF11, VEGFa, ADIPOQ, NODAL. Figure 5. Transcriptome among macrophage and epithelial cells or dendritic cells. Red indi- cates upregulation and green indicates downregulation. ∆Ct values were calculated as follows: ∆Ct = Ctref − Ctgoi, where ref = reference gene and goi = gene of interest. As the ∆Ct value decreases, the goi expression also decreases. Downregulated genes from ECs and AMs for NiO compared to SiO2 and untreated cultures include: C5, CSF3, CD40LG, CXCL16, CXCL10, IFNa2, IL-1b, IL-23, XCL1, BMP4, IL-27, and CCL21. DCs in the basolateral chamber showed that the following genes were upregulated in NiO-treated cultures compared to SiO2-treated and untreated cultures: CCL3, CCL20, CCL24, CSF2, IL-5, IL-11, IL-12b, IL-17F, OSM, TNFSF10, TNFSF11, BMP4, and CX3CL1. Genes downregulated in DCs in NiO-treated cultures include ADIPOQ, BMP7, CD70, CXCL3, IL-4, IL-15, IL-21, IL-22, LIF, and CXCL12. To better understand which biological pathways may be perturbed, genes were subse- quently loaded to the david.ncifcrf.gov database, and KEGG pathways were investigated to examine potential biological consequences. Tables 1 and 2 show speciﬁed pathways from KEGG analyses, which genes were up- or downregulated, and the possible biological outcomes from the perturbed genes within the pathway analyzed. Genes from AMs and ECs following NiO treatment compared to untreated cultures corresponded to pathways associated with chemokine signaling, cytosolic DNA sensing, rheumatoid arthritis, Toll-like receptor signaling, Jak-STAT signaling, inﬂammatory bowel disease, RIG-I-like receptor signaling, type I diabetes mellitus, asthma, PI3K-Akt signaling, T cell receptor signaling, NF-κB signaling, TGF-β signaling, NOD-like receptor signaling, natural killer cell-mediated cytotoxicity, and TNF signaling (Table 1). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 6 of 21 Transcriptome related to innate and adaptive cytokines was performed on delta Ct values and normalized to the reference gene ubiquitin C (UBC) by subtracting gene(s) of interest from the reference gene. Using delta Ct values where higher values represent in-creased expression, heatmaps were created to compare transcriptomics across the array of cytokine-related genes. A comparison between SiO2- and NiO-treated cultures revealed several inﬂammatory genes associated with inﬂammatory responses and associated ex-plicitly with cell activation and recruitment (CCL1, CCL3, CNTF, CSF2, FASLG, IL-5, IL-8, OSM, IL-12b, IL-17, LIF, and TNF) were upregulated in NiO treatment relative to un-treated and SiO2-treated cultures, indicating possible sensitizing potential. Figure 5 shows the transcriptome heatmaps for each respective treatment. For ECs and AMs in the apical chamber, the following genes were upregulated in NiO compared to SiO2 and untreated cultures: BMP6, CCL1, CCL2, CCL3, CCL17, CCL18, CCL19, CCL20, CNTF, CXCL1, CXCL2, CXCL5, CXCL9, CXCL10, CXCL13, IL-1RN, IL-1α, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-11, IL-12β, IL-15, IL-16, IL-17α, IL-17F, IL-22, MSTN, OSM, TGFβ2, THPO, TNF, TNFRSF11, TNFSF10, TNFSF11, VEGFa, ADIPOQ, NODAL. Downregulated genes from ECs and AMs for NiO compared to SiO2 and untreated cultures include: C5, CSF3, CD40LG, CXCL16, CXCL10, IFNa2, IL-1b, IL-23, XCL1, BMP4, IL-27, and CCL21. DCs in the basolateral chamber showed that the following genes were upregulated in NiO-treated cultures compared to SiO2-treated and untreated cultures: CCL3, CCL20, CCL24, CSF2, IL-5, IL-11, IL-12b, IL-17F, OSM, TNFSF10, TNFSF11, BMP4, and CX3CL1. Genes downregulated in DCs in NiO-treated cultures include ADIPOQ, BMP7, CD70, CXCL3, IL-4, IL-15, IL-21, IL-22, LIF, and CXCL12. Figure 5. Transcriptome among macrophage and epithelial cells or dendritic cells. Red indicates upregulation and green indicates downregulation. ΔCt values were calculated as follows: ΔCt = Ctref − Ctgoi, where ref = reference gene and goi = gene of interest. As the ΔCt value decreases, the goi expres-sion also decreases. To better understand which biological pathways may be perturbed, genes were sub-sequently loaded to the david.ncifcrf.gov database, and KEGG pathways were investi-gated to examine potential biological consequences. Tables 1 and 2 show speciﬁed path-ways from KEGG analyses, which genes were up- or downregulated, and the possible biological outcomes from the perturbed genes within the pathway analyzed. Genes from AMs and ECs following NiO treatment compared to untreated cultures corresponded to pathways associated with chemokine signaling, cytosolic DNA sensing, rheumatoid ar-thritis, Toll-like receptor signaling, Jak-STAT signaling, inﬂammatory bowel disease, RIG-I-like receptor signaling, type I diabetes mellitus, asthma, PI3K-Akt signaling, T cell re-ceptor signaling, NF-𝜅B signaling, TGF-𝛽 signaling, NOD-like receptor signaling, natural killer cell-mediated cytotoxicity, and TNF signaling (Table 1). Int. J. Mol. Sci. 2023, 24, 10104 7 of 19 Table 1. DAVID pathway analysis for epithelial and macrophage cells in the apical compartment. The table includes the pathways of up- and down-regulated genes and possible biological consequences of regulation. Only genes with ∆Ct values > 0.5 for comparisons of naïve vs. SiO2 vs. NiO treatments were considered for analysis. Speciﬁed Pathway Regulation Cytokines Chemokine Signaling Cytosolic DNA-sensing pathway Rheumatoid arthritis Toll-like receptor signaling pathway Jak-STAT signaling pathway Inﬂammatory bowel disease RIG-I-like receptor signaling pathway Type 1 diabetes mellitus Asthma PI3-Akt signaling pathway T cell receptor signaling pathway NF-kappa B signaling pathway TGF-beta signaling pathway NOD-like receptor signaling pathway UP DOWN CCL1, CCL2, CCL3, CCL5, CCL7, CCL17, CCL18, CCL19, CCL20, CCL22, CXCL1, CXCL2, CXCL5, CXCL9, CXCL10, CXCL13, PPBP CCL8, CCL11, CCL13, CCL21, CCL24, CXCL11, CXCL12, CXCL16, CX3CL1, XCL1, UP CXCL5, CXCL10, IL-1b, IL-6, IL-18 IFNa2 CCL2, CCL20, CCL3, CCL5, CXCL1, CXCL2, CXCL5, CSF1/2, IFNg, IL-1a, IL-1b, IL-6, IL-11, IL-15, IL-17a, IL-18, LTB, TGFb2, TNF, VEGFa, TNFSF11, TNFSF13b Biological Consequence Related to Sensitization Cell inﬁltration, growth, survival, differentiation, ROS production, cytoskeletal changes, leukocyte migration Inhibition of cell cycling Production of pro-inﬂammatory cytokines, type I interferons, NK cell activation Decreased NK cell activation, improved cell survival Fibroblast activation, angiogenesis, VEGFa signaling, leukocyte migration, inﬂammatory cell inﬁltration DOWN IFNa2, CSF3, IL-13, IL-23a, IL-27 Decreased cell cycling CXCL12, IL-23a Decreased inﬂammatory cell responses, decreased vasculature permeability CCL3, CCL5, CXCL9, CXCL10, IL-1b, IL-6, IL-12a, IL-12b, SPP1, TNF Production of inﬂammatory cytokines, T cell stimulation and recruitment INFa2, CXCL11 decreased TH2 response LIF, CNTF, CSF2, IFNg, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, IL-11, IL-12a, IL-12b, IL-15, IL-21, IL-22, IL-24, OSM, THPO Cell proliferation, differentiation, survival IFNg, IL-1a, IL-1b, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12a, IL-12b, IL-17a, IL-17f, IL-18, IL-21, IL-22, TGFb2, TNF Inﬂammatory pathways and autoimmune responses, T helper (Th) 1, 2, 17 differentiation IL-13, IL-23a CXCL10, IFNa2, IL-12a, IL-12b, TNF FASLG, IFNg, IL-1a, IL-b, IL-2, IL-12a, IL-12b, LTA, TNF CD40lg, IL-3, IL-4, IL-5, IL-9, IL-10, TNF CCL11, IL-13 Decrease in T helper (TH) 1 and 17 effector cells, regulatory T cells, and NKT cells Inﬂammatory cytokines, type 1 interferons, protein synthesis, dendritic cell activation, NK cell activation, cytotoxic T lymphocyte (CTL) differentiation, antibody production Upregulation of MHCII, macrophage activation, cytotoxic T lymphocyte (CTL) differentiation, CD4 T cell activation Lung epithelial cell and ﬁbroblast activation, T helper cell 2 differentiation and B cell interactions, mast cell activation, eosinophil recruitment and activation Decrease in smooth muscle cell recruitment and repair, decrease in eosinophil recruitment FASLG, CSF1, IFNa2, IL-2, IL-3, IL-4, IL-6, IL-7, OSM, SPP1, VEGFa Cell proliferation, DNA repair, angiogenesis, cell survival CSF3 Decreases in cell survival CD40lg, CSF2, IFNg, IL-2, IL-4, IL-5, IL-10, TNF CCL19, CXCL1, CXCL2, CD40lg, TNFSF11, TNFSF13b, IL-1b, LTA, LTB, TNF Proliferation, differentiation, immune response, PI3-Akt and Nf-kappa B pathway activation Auto-ubiquitination, cell survival Decreased CD8 T-cell homing, decreased epithelial cell repair after lung injury DOWN CCL13, CCL21, CXCL12 UP UP BMP2, BMP6, BMP7, IFNg, TGFb2, TNF, NODAL Iron metabolism, transcription factor activation, ubiquitin-mediated proteolysis BMP4 Decreased T cell differentiation, decreased iron metabolism CCL2, CCL5, CXCL1, CXCL2, IFNa2, IL-1b, IL-6, IL-18, TNF Proinﬂammatory cytokine release, NLRP3 inﬂammasome activation DOWN UP DOWN UP DOWN UP UP DOWN UP UP UP DOWN UP DOWN UP UP Int. J. Mol. Sci. 2023, 24, 10104 8 of 19 Table 1. Cont. Speciﬁed Pathway Regulation Cytokines Biological Consequence Related to Sensitization Inﬂammatory cytokine release, release of granules from granulocytes Natural killer cell mediated cytotoxicity TNF signaling UP UP FASLG, TNFSF10, CSF2, IFNa2, IFNg, TNF CCL2, CCL5, CCL20, CXCL1, CXCL2, CXCL5, CXCL10, LIF, CSF1/2, IL-1b, IL-6, IL-15, LTA, TNF Leukocyte recruitment and activation, inﬂammatory cytokine release, cell survival DOWN CX3CL1 Decreased leukocyte recruitment and activation Table 2. DAVID pathway analysis for dendritic cells in the basolateral compartment. The table includes the pathways of up- and down-regulated genes and possible biological consequences of regulation. Only genes with ∆Ct values > 0.5 for comparisons of naïve vs. SiO2 vs. NiO treatments were analyzed. Speciﬁed Pathway Regulation Cytokines Chemokine signaling UP XCL1, CCL1, CCL12, CCL17, CCL19, CCL2, CCL20, CCL22, CCL24, CCL3, CCl4, CCL5, CCL7, CXCL1, CXCL10, CXCL11, CXCL13, CXCL16, CXCL5, CXCL9, CX3CL1, PF4, PPBP Biological Consequence Related to Sensitization Cell inﬁltration, growth, survival, differentiation, ROS production, cytoskeletal changes, leukocyte migration Cytosolic DNA-sensing pathway Rheumatoid arthritis Toll-like receptor signaling pathway Jak-STAT signaling pathway Inﬂammatory bowel disease RIG-I-like receptor signaling pathway Type 1 diabetes mellitus Asthma PI3-Akt signaling pathway T cell receptor signaling pathway DOWN CCL11, CXCL12, CXCL3 Inhibition of activated granulocytes UP UP CCL4, CCL5, CXCL10, IFNa2, IL-1b, IL-18, IL-6 Production of pro-inﬂammatory cytokines, type I interferons, NK cell activation CCL12, CCL2, CCL20, CCL3, CCL5, CXCL1, CXCL5, CSF1/2, IFNg, IL-1a, IL-1b, IL-11, IL-17a, IL-18, IL-23a, IL-6, LTB, TGFB2, TNFSF11, TNFSF13b, TNF, VEGFa Fibroblast activation, angiogenesis, VEGFa signaling, leukocyte migration, inﬂammatory cell inﬁltration DOWN CXCL12, CXCL3, IL-15, UP UP CCL3, CCL4, CCL5, CXCL10, CXCL11, CXCL9, IFNa2, IL-1b, IL-12a, IL-12b, IL-6, SPP1, TNF CTF1, CNTF, CSF2/3, IFNa2, IFNg, IL-10, IL-11, IL-12a, IL-12b, IL-13, IL-2, IL-23a, IL-24, IL-27, IL-3, IL-5, IL-6, IL7, IL-9, OSM, THPO Decreases in autocrine function of self-reactive Th1 cells, decreases in Th17 differentiation, decreases in blood vessel permeability Chemotaxis of leukocytes, T cell stimulation and recruitment Cell proliferation, differentiation, survival DOWN IL-15, IL-21, IL-22, IL-4, LIF Decreases in cell cycling, proliferation, differentiation, and survival IFNg, IL-1a, IL-1b, IL-10, IL-12a, IL-12b, IL-13, IL-17a, IL-17f, IL-18, IL-2, IL-23a, IL-5, IL-6, TGFB2, TNF Inﬂammatory pathways and autoimmune responses IL-21, IL-22, IL-4, UP CXCL10, IFNa2, IL-12a, IL-12b, TNF FASL, IFNg, IL-1a, IL-1b, IL-12a, IL-12b, IL-2, LTA, TNF CD40lg, IL-10, IL-13, IL-3, IL-4, IL-5, IL-9, TNF CCL11 Decrease in T helper (TH) 1 and 17 effector cells, regulatory T cells, and NKT cells Protein synthesis, dendritic cell activation, NK cell activation, cytotoxic T lymphocyte (CTL) differentiation, antibody production Decreases in cytotoxic CD8+ T cells Decreases in mast cell activation Decrease in smooth muscle cell recruitment and repair, decrease in eosinophil recruitment FASL, CSF1, CSF3, IFNa2, IL-2, IL-3, IL-6, IL-7, OSM, SPP1, VEGFa Cell proliferation, DNA repair, angiogenesis, cell survival IL-4 Decreases in cell survival CD40LG, CSF2, IFNg, IL-10, IL-2, IL-4, IL-5, TNF Proliferation, differentiation, immune response, PI3-Akt and Nf-kappa B pathway activation UP DOWN DOWN UP DOWN UP Down UP Int. J. Mol. Sci. 2023, 24, 10104 9 of 19 Genes in DCs following NiO treatment corresponded to the following pathways: chemokine signaling, cytosolic DNA-sensing, rheumatoid arthritis, Toll-like receptor sig- naling, Jak-STAT signaling, inﬂammatory bowel disease, RIG-I-like receptor signaling, type I diabetes mellitus, asthma, PI3K-Akt signaling, and T cell receptor signaling (Table 2). Luminex was performed on both culture supernatant and cell lysate at 24 h post- exposure to measure an array of cytokines associated with inﬂammation. The results indicate increases in protein expression common to inﬂammation and related to cell inﬁl- tration, activation, and maturation (Figure 6). The cell supernatant and lysate of ECs and AMs showed signiﬁcant increases in IL-8 for NiO-treated cultures. At the same time, SiO2 also showed signiﬁcant increases in RANTES from the cell lysate compared to NiO and untreated cultures. The cell lysates of DCs showed signiﬁcant increases in IL-6 and MIP-1a in NiO- and SiO2-treated cultures compared to untreated cultures. Signiﬁcant decreases were seen in IL-5 from NiO- and SiO2-treated cultures compared to untreated cultures. Supernatants from DCs showed only a signiﬁcant increase in IL-8 for NiO compared to both SiO2 and untreated cultures, indicating a prolonged recruitment of neutrophils (Figure 5). Figure 6. Luminex data for epithelial and macrophage cells (A,B) and dendritic cells (C,D). All cytokines are reported in pg/mL. Signiﬁcance is to untreated samples and is denoted with an * (, p ≤ 0.05; , p ≤ 0.01; , p ≤ 0.001; *, p ≤ 0.0001). Flow cytometry was performed to identify speciﬁc DC markers related to activation and antigen presentation (CD40, MHCII, CD80) and migration (CCR7) (Figure 7). MHCII expression was signiﬁcantly upregulated for both SiO2- and NiO-treated cultures compared to untreated cultures (29.07% for untreated; 38.97% for SiO2-treated; 58.57% for NiO- treated). CD40 expression was increased in both SiO2- (2.16%) and NiO-treated cultures (2.53%) compared to untreated cultures (0.66%), but not signiﬁcantly so. CD80 expression was signiﬁcantly increased in both SiO2- and NiO-treated cultures compared to untreated cultures (41.57%, 30.4%, and 6.18%, respectively). CCR7 expression was increased in both SiO2- (2.98%) and NiO-treated cultures (1.51%) when compared to untreated cultures (0.46%), but the results did not reach statistical signiﬁcance. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 11 of 21 indicate increases in protein expression common to inﬂammation and related to cell inﬁl-tration, activation, and maturation (Figure 6). The cell supernatant and lysate of ECs and AMs showed signiﬁcant increases in IL-8 for NiO-treated cultures. At the same time, SiO2 also showed signiﬁcant increases in RANTES from the cell lysate compared to NiO and untreated cultures. The cell lysates of DCs showed signiﬁcant increases in IL-6 and MIP-1a in NiO- and SiO2-treated cultures compared to untreated cultures. Signiﬁcant decreases were seen in IL-5 from NiO- and SiO2-treated cultures compared to untreated cultures. Supernatants from DCs showed only a signiﬁcant increase in IL-8 for NiO compared to both SiO2 and untreated cultures, indicating a prolonged recruitment of neutrophils (Fig-ure 5). Figure 6. Luminex data for epithelial and macrophage cells (A,B) and dendritic cells (C,D). All cy-tokines are reported in pg/mL. Signiﬁcance is to untreated samples and is denoted with an (, p ≤ 0.05; , p ≤ 0.01; , p ≤ 0.001; , p ≤ 0.0001). Flow cytometry was performed to identify speciﬁc DC markers related to activation and antigen presentation (CD40, MHCII, CD80) and migration (CCR7) (Figure 7). MHCII expression was signiﬁcantly upregulated for both SiO2- and NiO-treated cultures com-pared to untreated cultures (29.07% for untreated; 38.97% for SiO2-treated; 58.57% for NiO-treated). CD40 expression was increased in both SiO2- (2.16%) and NiO-treated cul-tures (2.53%) compared to untreated cultures (0.66%), but not signiﬁcantly so. CD80 ex-pression was signiﬁcantly increased in both SiO2- and NiO-treated cultures compared to untreated cultures (41.57%, 30.4%, and 6.18%, respectively). CCR7 expression was in-creased in both SiO2- (2.98%) and NiO-treated cultures (1.51%) when compared to un-treated cultures (0.46%), but the results did not reach statistical signiﬁcance. (A) (B) (C) (D) Int. J. Mol. Sci. 2023, 24, 10104 10 of 19 Figure 7. Dendritic cell activation markers. Flow cytometry data for surface markers of MHCII, CD80, CD40, and CCR7 were measured. All samples were analyzed in the live population only. Signiﬁcance is to untreated cultures, denoted with * p ≤ 0.001; ** p ≤ 0.0001. 3. Discussion The local milieu of the lungs is designed to be anti-inﬂammatory to prevent excessive inﬂammation and exacerbated immune responses to every exogenous material inhaled. Speciﬁcally, alveolar macrophages (AMs) phagocytose and continually patrol the lumen of the alveolar spaces where they engulf and dispose of foreign materials. In the steady state, AMs are suppressive by secreting immunosuppressive cytokines to surrounding cells [33–35]. Dendritic cells (DCs) are the primary antigen-presenting cells throughout the human system. When activation occurs, they can extend their dendrites through the tight junctions of the epithelial barrier and into the luminal space, where they recognize, capture, and process antigens [36,37]. Once activated, DCs will upregulate co-stimulatory markers and migratory receptors, which are necessary for traveling to local lymph nodes and eliciting an activating and sustained response from T and B cells to form a lasting immune response [38]. The formation of antigen-speciﬁc T and B cells can ultimately lead to sensitization to any xenobiotic. Therefore, cell activation and maturation mechanisms can potentially lead to detecting early biomarkers of respiratory sensitizing potential. While there are no current biosignatures of respiratory sensitization common to all known respiratory sensitizers, there are general principles of sensitization that appear to hold for most known sensitizers at the respiratory junction: neutrophil inﬂux and general cell activation of recruited cells [39]. The cytokine milieu within the lungs determines the effector function of immune cells, speciﬁcally regarding allergy and sensitization. Because neutrophils are commonly recruited as a ﬁrst-line defense against various cell and tissue assaults, their use as biomarkers is currently limited without additional endpoints simultaneously measured. This study examined the effects of particulates on cells, and the data obtained can assist in identifying biosignatures linked to respiratory sensitization. The observations made can be useful for future studies with differing experimental approaches. During respiratory sensitizing reactions to chemical sensitization, our previous study found that speciﬁc cytokine-related genes, including CXCL5, IL-6, IL-8, and CCL7, were expressed in a perturbed manner [6]. The CXC chemokine ligand 5 (CXCL5) is known to be a potent neutrophil attractant both in vivo and in vitro and is known to be secreted by both innate (e.g., ECs) and adaptive (CD4 T cells) immune cells [40–42]. Several known pathologies are associated with increased expression of CXCL5, including COPD from Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 12 of 21 Figure 7. Dendritic cell activation markers. Flow cytometry data for surface markers of MHCII, CD80, CD40, and CCR7 were measured. All samples were analyzed in the live population only. Signiﬁcance is to untreated cultures, denoted with * p ≤ 0.001; **** p ≤ 0.0001. 3. Discussion The local milieu of the lungs is designed to be anti-inﬂammatory to prevent excessive inﬂammation and exacerbated immune responses to every exogenous material inhaled. Speciﬁcally, alveolar macrophages (AMs) phagocytose and continually patrol the lumen of the alveolar spaces where they engulf and dispose of foreign materials. In the steady state, AMs are suppressive by secreting immunosuppressive cytokines to surrounding cells [33–35]. Dendritic cells (DCs) are the primary antigen-presenting cells throughout the human system. When activation occurs, they can extend their dendrites through the tight junctions of the epithelial barrier and into the luminal space, where they recognize, capture, and process antigens [36,37]. Once activated, DCs will upregulate co-stimulatory markers and migratory receptors, which are necessary for traveling to local lymph nodes and eliciting an activating and sustained response from T and B cells to form a lasting immune response [38]. The formation of antigen-speciﬁc T and B cells can ultimately lead to sensitization to any xenobiotic. Therefore, cell activation and maturation mechanisms can potentially lead to detecting early biomarkers of respiratory sensitizing potential. While there are no current biosignatures of respiratory sensitization common to all known respiratory sensitizers, there are general principles of sensitization that appear to hold for most known sensitizers at the respiratory junction: neutrophil inﬂux and general cell activation of recruited cells [39]. The cytokine milieu within the lungs determines the eﬀector function of immune cells, speciﬁcally regarding allergy and sensitization. Because neutrophils are commonly recruited as a ﬁrst-line defense against various cell and tissue assaults, their use as biomarkers is currently limited without additional endpoints simul-taneously measured. This study examined the eﬀects of particulates on cells, and the data obtained can assist in identifying biosignatures linked to respiratory sensitization. The observations made can be useful for future studies with diﬀering experimental ap-proaches. During respiratory sensitizing reactions to chemical sensitization, our previous study found that speciﬁc cytokine-related genes, including CXCL5, IL-6, IL-8, and CCL7, were expressed in a perturbed manner [6]. The CXC chemokine ligand 5 (CXCL5) is known to be a potent neutrophil attractant both in vivo and in vitro and is known to be secreted by Int. J. Mol. Sci. 2023, 24, 10104 11 of 19 cigarette smoking, infections, and allergy [41–47]. Interleukin 6 (IL-6), a pleiotropic cy- tokine capable of both inﬂammatory and anti-inﬂammatory responses, can elicit chronic inﬂammation and allergy in the lungs [48,49]. While various cell types can secrete IL-6 at the onset of insult or injury, it has recently been revealed that pulmonary DCs and AMs are speciﬁc cytokine sources for inﬂammatory conditions such as sensitization and allergic airway inﬂammation [50]. Another potent neutrophil attractant, interleukin 8 (IL-8), is secreted early during the inﬂammatory process by both ECs and AMs [51–53]. Importantly, IL-8 has been shown to increase various respiratory diseases in both in vivo and in vitro studies [39,54–56]. C-C chemokine ligand 7 (CCL7) is a powerful attractant for eosinophils and affects neutrophils and epithelial cells. Its expression increases in respiratory allergy, airway hyperresponsiveness, and sensitization. Furthermore, exposure to particulates, especially suspected respiratory-sensitizing particulates, leads to increased levels of the inﬂammatory protein chemokine ligand 3 (CCL3) [57,58]. This protein is secreted by dif- ferent cell types, such as ECs, AMs, and DCs, and has been observed to release cytokines previously seen with chemical sensitizers [59]. Mast cells and eosinophils are activated by CCL3, which is a potent trigger. These cells contribute signiﬁcantly to lung inﬂammation in conditions such as allergies and airway hyperresponsiveness [59–61]. Studies conducted in living organisms have demonstrated that exposure to NiO nanoparticles can lead to an increase in neutrophil and eosinophil counts [23,24]. Although the current study did not measure cellular inﬂux, it did evaluate the rise in transcripts and cytokines related to cellular inﬂux and activation. The results of this study are consistent with those observed in in vivo studies. Similarly, in vivo studies examining SiO2 have revealed an increase in inﬂammatory cytokines, such as IL-6, with little to no change in total IFN-gamma, which is in line with the ﬁndings of the current study [62]. Overall, the data suggest that the alveolar model used is comparable to animal models using similar exposure materials. Unlike skin sensitization, the lungs lack a validated model that accurately identiﬁes known or potential sensitizers [63]. The current gold standard uses animal models where the local lymph node assay (LLNA) and serum cytokine levels are the primary methods for assessing sensitization. Still, the LLNA is the only universally approved technique for der- mal testing. It is important to note that the cytokine levels are unreliable as different animal models have different immune systems and subsequent responses and poor translation to human immune responses [8,63–67]. Recent investigations into respiratory sensitization have attempted to use the skin sensitization assays of the direct peptide reactivity assay (DPRA) and the peroxidase peptide reactivity assay (PPRA). However, while these meth- ods show promise, they need to be more accurate on their own (accuracy ~80%) for the utilization [68,69]. As such, it is necessary to develop methods for identifying and assessing the respiratory sensitizing potential of both current and novel materials. As an alternative to animal testing and to circumvent many of the issues associated with failure to translate to humans, human-derived cells can and should be a current method of investigation [1,14,15]. Studies have used cells that closely mimic DCs or single-cell types (DCs) rather than multi-cell models capable of introducing intercellular communication and responses [38]. While promising results have been shown, a lack of high sensitivity, speciﬁcity, and accuracy in predicting outcomes precludes the use of single-cell systems for now. Including multiple cell types and various techniques designed to probe multiple endpoints may improve the speciﬁcity, sensitivity, and accuracy of any lung models in development. This study showed an alveolar cell culture model mimicking in vivo architecture to differentiate responses induced by a known respiratory irritant (SiO2) and a suspected res- piratory sensitizer (NiO). Endpoint measurements included: (1) Cell morphology measured by microscopy; (2) Transcriptomics measured by real-time polymerase chain reactions (rt-PCR); (3) Cytokine proﬁling and expression measured by a Luminex multiplex assay; (4) Expression of cell surface markers measured via ﬂow cytometry; and (5) Biological pathway analyses probed via the Database for Annotation Visualization and Integrated Discovery (DAVID). Respiratory sensitization typically requires an initial exposure, induc- Int. J. Mol. Sci. 2023, 24, 10104 12 of 19 tion, subsequent re-exposure, and elicitation, for typical symptomatic responses. However, speciﬁc biochemical (surface marker, cytokine, and gene) responses are required for innate cells to recruit and activate immune-speciﬁc adaptive cells (e.g., T and B cells). Because of this requirement, it is hypothetically possible to identify respiratory sensitizers before the elicitation phase by examining innate cells at the exposure site. This would allow for the development of a rapid assay capable of predicting sensitizing potential before exposure, preventing the implementation of novel materials that may lead to poor health outcomes. Some cytokines, such as IL-8, peak 24 h post-exposure [70]. Additionally, it has been shown that activation markers of DCs increase in expression as a function of time after exposure [71]. To best account for changes in cell marker expression, transcriptome, and cytokine release associated with known sensitization potential (i.e., cell recruitment, initiation, and activation), a timepoint measurement of 24 h post-exposure was chosen. Dendritic cells (DCs), the primary antigen-presenting cells, are critical to immune responses throughout the body and are essential to eliciting long-term immune responses. On activation, these cells will readily take up and process exogenous material, increase the surface expression of MHCII, where the foreign antigen is presented, and migrate to local draining lymph nodes to train and activate T and B cells [72–76]. Furthermore, DCs will increase the biosynthesis of costimulatory molecules (CD40 and CD80), which bind to T cells for effector phenotyping in lymph nodes [77,78]. Results from this study show signiﬁcant increases in MHCII and CD80, as well as trends towards increased expression levels of CD40 and the migratory receptor CCR7 after exposure to a suspected respiratory sensitizer (NiO). In ECs and AMs, perturbations in the transcriptome are related to biological pathways, which can affect immune cell recruitment, proliferation, differentiation, and survival; increases in ROS production and cytoskeletal component changes; cellular migration, and activation of ﬁbroblasts. Biologically, these pathways affect acute and chronic inﬂammatory responses, the ability of lymphoid cells to home relevant tissues of interest, and cell signaling. Downregulated transcriptomic proﬁles in ECs and AMs lead to perturbed biological pathways, which can cause decreases in cell cycling and cell-effector functionality. In the lungs, it has been shown that excessive increases in oxidative stress can hinder AM functionality, leading to pathologic inﬂammation [79]. While many regulatory mecha- nisms prevent undue oxidative stress, one of the most common methods to assess recovery or continued insult is to measure glutathione levels. Glutathione (GSH) concentrations are relatively high in the extracellular ﬂuid within the lung compartment, purportedly to reduce oxidative stress [80]. GSH levels tend to peak at 24 h post-exposure [81]. Results showed signiﬁcant increases in ROS after exposure to the suspected sensitizer NiO at 24 h post-exposure, indicating that not only had cellular mechanisms not compensated for the injury, but that an increased likelihood for severe pathologic inﬂammation exists. In dendritic cells, it has been shown that a known sensitizer will cause upregulation of the major histocompatibility complex (MHC) class II, co-stimulatory molecules (e.g., CD40, CD80, and CCR7), and inﬂammatory cytokines [82]. Results from this study showed several morphological changes after exposure to either a known irritant or suspected respiratory sensitizer (SiO2 and NiO, respectively). Further changes were seen when comparing the suspected sensitizer NiO to the irritant SiO2 and untreated cultures. Signiﬁcant increases in cytokine production, perturbations in the transcriptome, and surface marker expression related to inﬂammation, allergy, and sensitization were all noted after exposure to NiO. Taken together with previous studies, this study helps further the idea that various endpoint readouts (morphology, transcriptomics, cytokine production, and cell surface markers) can help establish a high throughput assay capable of assessing the sensitizing potential of new and existing substances. Limitations and Suggestions for Model Design The lungs are highly complex and constantly changing. Current models do not account for important processes such as cell turnover, activation, and communication between Int. J. Mol. Sci. 2023, 24, 10104 13 of 19 cells. Additionally, there are barriers such as ﬂuids, such as surfactant and mucus, that capture and eradicate foreign substances but are not present in current models. To improve accuracy, studies should use ﬂuidic devices and cell migration assays and include T and B cells to assess the activation of immune cells. While in vitro models have limitations compared to live organs or organisms, they offer the ability to study speciﬁc mechanisms and test new materials and contaminants. This can lead to signiﬁcant advancements in understanding and treating lung-related issues. 4. Materials and Methods Experimental premise. To better understand the utility of this model, a model comparing a suspected respiratory sensitizing particulate (e.g., nickel oxide, NiO) and a known respiratory irritant (e.g., crystalline silica, SiO2) were utilized. SiO2 and NiO were made in-house with physicochemical properties shown in Figure 2. Reagents. A commercially available engineered nickel oxide particle was purchased from Nanoshel, LLC (County Cavan, Ireland; Product No. NS6130-03-337). Similarly, a commercially available engineered silicon dioxide particle was purchased from Sigma- Aldrich (St. Louis, MO, USA; Product No. S5631). Physicochemical characterization of the materials. Hydrodynamic diameter, polydisper- sity, and zeta potential measurements were taken using a Zetasizer Nano ZS (Malvern Instruments Ltd., Westborough, MA, USA). All measurements were performed in trip- licate with sample parameters for absorbance and refractive indices set to 0.01 nm and 1.580, respectively. Nanoparticle preparation for cell culture studies. Samples were diluted to 0.002 wt% in triplicate. Dilutions were performed in phenol-free cell culture media. Cell culture. A549 epithelial cells and U937 monocytes were grown in complete RPMI (cRPMI) 1640 (Thermo Fisher Scientiﬁc Inc., Waltham, MA, USA) and supplemented with 10% FBS and 1% penicillin-streptomycin. JAWSII cells were cultured in complete Alpha minimum essential medium (cAMEM) with nucleosides (ThermoFisher Scientiﬁc Inc., Waltham, MA, USA) and supplemented with 5 ng/mL murine GM-CSF (BioLegend, San Diego, CA, USA); 20% fetal bovine serum, and 1% penicillin-streptomycin. All cells were maintained at 37 ◦C in a humidiﬁed 5% CO2 atmosphere until ready for use. Cells were plated as previously described [83]. A549 epithelial cells were added to 12-well plates ﬁtted with polyethylene terephthalate (PET) Transwell® membranes (Corning, Tewksbury, MA, USA) at 28 × 104 cells/cm2. Cells were allowed to adhere for 2–3 days until a conﬂuent monolayer was formed. Media were removed, and inserts were inverted and placed into sterile glass dishes. JAWSII cells were resuspended in 500 µL of cAMEM and plated on the basal surface of the membrane at 7 × 104 cells/cm2 and allowed to adhere for hours. After excess media were removed, inserts were reverted into the well plate, and 1 mL of cAMEM was added to the basolateral chamber. U937 macrophages were added at a 1:9 ratio of U937:A549 in cRPMI, and the apical chamber was replenished to 500 µL [84]. The model was then placed in a 37 ◦C humidiﬁed incubator at a 5% CO2 atmosphere to allow cells to rest for 24 h before exposure. Macrophage differentiation. U937 monocytic cells were incubated with 100 ng/mL phorbol 12-myristate-13-acetate (PMA) for 24 to 48 h, as previously described [85]. The cells were washed two times in sterile 1X PBS, and fresh media were added. Cells rested in the 37 ◦C humidiﬁed incubator at a 5% CO2 atmosphere for 72 h before use. The adherent cells were dissociated using trypsin, resuspended in cRPMI, counted, and plated according to use. Chemical exposure. All exposure materials were added to the apical chamber of the Transwell® membrane. SiO2 was added at 50 ppm and was a positive control for cellular irritation. NiO was added at 50 ppm as a test compound for suspected sensitization. The post-exposure period was 24 h to assess early markers of respiratory sensitizing potential. Cell imaging. Imaging occurred as previously described [6]. Brieﬂy, Transwell® plates were removed from the incubator, and cells from both chambers were washed twice with Int. J. Mol. Sci. 2023, 24, 10104 14 of 19 1X phosphate-buffered saline (PBS) solution. Glutaraldehyde, at a concentration of 1:10 in 1X PBS, was added to both chambers for 10 min and followed by three washes of 1X PBS at 10 min intervals. PBS was replaced with 4% osmium tetroxide in PBS for 2.5 h at 4 ◦C. Three consecutive wash steps were repeated, followed by a series of dehydration steps that occurred twice, each at 10 min intervals: 50% ethanol (EtOH); 70% EtOH; 90% EtOH; 100% EtOH. The well inserts were removed and submerged in 100% EtOH in sequence. The membranes were carefully excised with a razor blade, placed into sterile buckets, and dried in a critical point dryer (CPD300, Leica, Buffalo Grove, IL, USA). Imaging was performed on a focused ion beam scanning electron microscope (FIB-SEM, Versa 3D, FEI ThermoFisher Scientiﬁc, Hillsboro, OR, USA) at 5 kV with a spot size of 5.0 and a working distance of 10 mm using an Everhart–Thornley detector. Cells were stained with NucBlueTM live cell stain (ReadyProbes, Thermo Fisher Scien- tiﬁc Inc.), MitoTracker™ Red CM-H2Xros (Thermo Fisher Scientiﬁc Inc.), and ActinGreenTM 488 ReadyProbes reagent (Thermo Fisher Scientiﬁc Inc.) for analysis of the nucleus, reactive oxygen species (ROS), and F-actin cytoskeleton, respectively. Images were captured using a confocal laser scanning microscope (FV-3000, Olympus Corp., Center Valley, PA, USA). Quantifying ﬂuorescence was performed with CellSens software V4.2 on a Wacom Cintiq 22HD workstation (Olympus Corp.). Transcriptomics. Polymerase chain reaction (PCR) plates for a panel of innate and adaptive cytokine were purchased (AB Applied Biosystems TaqMan® Array 96-well plates) for mouse and human (catalog # 4391524). The Transwell® compartments were evalu- ated separately by collecting cells and supernatant from both the apical and basal sides. Macrophage and epithelial cells were analyzed from the former, while the latter was used for dendritic cells. RNA collection, cDNA formation, and plating protocols followed manufacturer instructions. Plates were assessed on a QuantStudio 6 Flex RealTime PCR system (ThermoFisher Scientiﬁc), delta Ct values were calculated (i.e., ∆Ct = Ct (gene of interest)—Ct (housekeeping gene)), and heatmaps were created. Only ∆Ct values greater than 0.5 were considered for statistical analyses. Analysis of PCR data was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID v6.8) [86]. Brieﬂy, gene lists from each assay were sorted based on the ofﬁcial gene symbols. Once sorted, pathways were identiﬁed utilizing the Kyoto Encyclopedia of Genes and Genomes (KEGG, Kenoisha laboratories, Tokyo, Japan) (https://www.genome.jp/kegg/) pathway analyses. Cytokine and chemokine multiplex analysis. Using the manufacturer’s instruction, cy- tokines and chemokines were measured using Milliplex MAP cytokine/chemokine mag- netic bead panels for both mouse and human (Millipore, Burlington, MA, USA). Flu- orescence was analyzed using the Bio-Plex Luminex 100 XYP (Bio-Rad, Hercules, CA, USA) with the Bio-Plex Manager 4.1 software. Subsequently, a 5-parameter curve-ﬁtting algorithm was applied for standard curve calculations. Flow cytometry. Fluorochrome-conjugated antibodies to CD40 (3/23), I-A/I-E (M5/114.15.2), CCR7 (4B12), and CD80 (B7-1) were purchased from BioLegend (San Diego, California, USA). TruStain FcX™ (anti-mouse CD16/32) was used to block against non-specific Fc binding, and 7-amino-actinomycin D (7-AAD) was used to measure live/dead cells. A live/dead cell gating was obtained, and the analysis was performed on a FACSVerse (BD Biosciences, Franklin Lakes, NJ, USA) with a subsequent analysis performed utilizing FlowJo v10. Cells were prepared as follows: after 24 h, the media were removed from the basal chamber, and 0.25% trypsin-EDTA (Fisher Scientific) was added for 5 min in a 37 ◦C humidified incubator at a 5% CO2 atmosphere. Equal parts of complete media were added to each well and mixed to resuspend cells. The samples were then spun in a temperature-controlled incubator at 4 ◦C, and the supernatant was removed. After removing the supernatant, cells were washed and spun before cell staining. Staining was performed as previously described [87]. Briefly, cells were washed with FACS buffer (2% FBS, 0.1% NaN3 in PBS), blocked with anti-mouse CD16/32 (93), and placed on ice for 10 min. Consecutively, the cells were stained on ice for 45 min using anti-CD40, I-A/I-E, Int. J. Mol. Sci. 2023, 24, 10104 15 of 19 CD80, and CCR7 at 1:200 dilutions in FACS buffer. Lastly, the cells were washed three times in a FACS buffer and resuspended to a final volume of 0.2 mL before a FACS analysis. Dynamic light scattering (DLS) analysis. Particles were suspended in phenol-free cell culture media. Hydrodynamic diameter, dispersity, and zeta potential measurements were taken using a Zetasizer Nano ZS (Malvern Instruments Ltd., Westborough, MA, USA). All measurements were performed in triplicate with sample parameters for absorbance and refractive index set to 0.01 nm and 1.580, respectively. Statistical Analyses. Unless otherwise noted, all samples were performed in triplicate with three replicates for each methodology for nine samples in each experimental setup. Data were analyzed using analysis of variance (ANOVA) followed by a t-test using Mi- crosoft Excel v16.72 and GraphPad Prism 9.4.1. Signiﬁcance is noted in the ﬁgure caption where applicable, with data presented as mean with ± standard deviation. 5. Conclusions While it is possible that any chemical or particulate can lead to sensitization within the respiratory system leading to lifelong allergies, hypersensitivity, and other complications, the incidence rate is still low within the more signiﬁcant population. However, to prevent poor health outcomes, especially in areas of lower development, assessing for respiratory sensitization is a continued focus in immunotoxicology. To provide preventative, protective, or curative responses, it is critical to understand the processes that promote long-term immune reactions before, during, and/or after toxicant exposure. The model utilized in this study can rapidly adjust cell types to mimic the area of the lung (i.e., upper or lower lung) to be studied. The simplicity of design, low cost of setup, ability to switch to an air–liquid interface if needed, and the ability to modify the endpoints measured are all strengths in using the model herein. While this model is static and does not include adaptive immune cells, subsequent studies are needed. They are underway to assess the ability of the cells in this model to activate and recruit T and B cells after exposure to known respiratory sensitizers and novel materials. Although submerged conditions are still commonly used in most studies, air–liquid interface (ALI) cultures have proven increasingly successful in recent years [88]. However, many labs are still unable to use ALI due to the high cost and limited availability of the necessary equipment and aerosol technology. It is crucial for labs to have easy access to the equipment and aerosol technology needed to make a universal assay capable of assessing respiratory sensitizing potential. Submerged systems are the optimal choice until ALI technology and equipment become more readily available. If a submerged system can differentiate between known respirable sensitizers and non-sensitizers, it is preferred due to its ease of use. This study used a submerged system and two particulates to achieve this task by successfully differentiating a known non-sensitizing particulate from a suspected sensitizer with supporting evidence in the clinical literature. When selecting and optimizing a co-culture system, the type of cell used is crucial. Although some human dendritic cell lines are available, they are not well-established in the literature and can be difﬁcult to obtain. Obtaining human PBMCs that have differentiated into dendritic cells is also challenging, expensive, and can vary signiﬁcantly between individuals. Therefore, researchers often use cell lines to ensure response consistency and simplify the validation process. Studies have demonstrated that murine dendritic cells exhibit similar responses to human dendritic cells, with JAWSII cells being a good example [1,89,90]. Instead of using human monocytic cell lines, immature dendritic cell lines that do not require differentiation or marker validation can streamline the assay development process and make it easier to validate results across multiple laboratories. Author Contributions: M.G.: Conceptualization, Methodology, Investigation, Writing—original draft. C.M.S.: Methodology, Writing—original draft, Writing—review and editing, Funding acquisi- tion. All authors have read and agreed to the published version of the manuscript. Int. J. Mol. Sci. 2023, 24, 10104 16 of 19 Funding: This research was funded by the C. Gus Glasscock, Jr. Endowed Fund for Excellence in Environmental Sciences in the College of Arts and Sciences at Baylor University, USA, and the Henry F. Jackson Foundation (Agreement 5055/PO 979338/Award 64695). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data are available upon request. Conﬂicts of Interest: The authors declare that they have no known competing ﬁnancial interest or personal relationships that could have appeared to inﬂuence the work reported in this paper. References 1. 2. Rothen-Rutishauser, B.; Gibb, M.; He, R.; Petri-Fink, A.; Sayes, C.M. Human lung cell models to study aerosol delivery– Considerations for model design and development. Eur. J. Pharm. Sci. 2022, 180, 106337. [CrossRef] [PubMed] Costa, D.L. Alternative test methods in inhalation toxicology: Challenges and opportunities. Exp. Toxicol. Pathol. 2008, 60, 105–109. [CrossRef] [PubMed] Blackwell, M. Respiratory sensitization. In Inhalation Toxicology; CRC Press: Boca Raton, FL, USA, 2006; pp. 243–267. 3. 4. Michaels, D.D. RE: Occupational Safety and Health Administration (OSHA) Draft Weight of Evidence Guidance Document (OSHA-2016- 5. 6. 7. 8. 9. 10. 0004); CDC OSHA: Washington, DC, USA, 2016. Boverhof, D.R.; Billington, R.; Gollapudi, B.B.; Hotchkiss, J.A.; Krieger, S.M.; Poole, A.; Wiescinski, C.M.; Woolhiser, M.R. Respiratory sensitization and allergy: Current research approaches and needs. Toxicol. Appl. Pharmacol. 2008, 226, 1–13. [CrossRef] [PubMed] Gibb, M.; Sayes, C. An in vitro alveolar model allows for the rapid assessment of chemical respiratory sensitization with modiﬁable biomarker endpoints. Chem.-Biol. Interact. 2022, 368, 110232. [CrossRef] Kimber, I.; Basketter, D.A.; Gerberick, G.F.; Ryan, C.A.; Dearman, R.J. Chemical allergy: Translating biology into hazard characterization. Toxicol. Sci. 2011, 120 (Suppl. S1), S238–S268. [CrossRef] Chary, A.; Hennen, J.; Klein, S.G.; Serchi, T.; Gutleb, A.C.; Blömeke, B. Respiratory sensitization: Toxicological point of view on the available assays. Arch. Toxicol. 2018, 92, 803–822. [CrossRef] Stoccoro, A.; Karlsson, H.L.; Coppedè, F.; Migliore, L. Epigenetic effects of nano-sized materials. Toxicology 2013, 313, 3–14. [CrossRef] Fröhlich, E.; Salar-Behzadi, S. Toxicological assessment of inhaled nanoparticles: Role of in vivo, ex vivo, in vitro, and in silico studies. Int. J. Mol. Sci. 2014, 15, 4795–4822. [CrossRef] 11. Hoymann, H.G. Lung function measurements in rodents in safety pharmacology studies. Front. Pharmacol. 2012, 3, 156. [CrossRef] 12. Bracken, M.B. Why animal studies are often poor predictors of human reactions to exposure. J. R. Soc. Med. 2009, 102, 120–122. [CrossRef] 13. Hackam, D.G.; Redelmeier, D.A. Translation of research evidence from animals to humans. JAMA 2006, 296, 1727–1732. [CrossRef] [PubMed] 14. Perlman, R.L. Mouse models of human diseaseAn evolutionary perspective. Evol. Med. Public Health 2016, 2016, 170–176. [PubMed] 15. Powell, K. Technology Feature\| Replacing the replacements: Animal model alternatives. Science 2018, 362, 246. [CrossRef] 16. Zarei, F.; Azari, M.R.; Salehpour, S.; Khodakarim, S.; Omidi, L.; Tavakol, E. Respiratory effects of simultaneous exposure to respirable crystalline silica dust, formaldehyde, and triethylamine of a group of foundry workers. J. Res. Health Sci. 2017, 17, 371. Silica, Crystalline. 2023. Available online: https://www.osha.gov/silica-crystalline (accessed on 15 January 2023). 17. 18. Berg, J.M.; Romoser, A.A.; Figueroa, D.E.; West, C.S.; Sayes, C.M. Comparative cytological responses of lung epithelial and pleural mesothelial cells following in vitro exposure to nanoscale SiO2. Toxicol. Vitr. 2013, 27, 24–33. [CrossRef] 19. Vance, M.E.; Kuiken, T.; Vejerano, E.P.; McGinnis, S.P.; Hochella, M.F., Jr.; Rejeski, D.; Hull, M.S. Nanotechnology in the real world: Redeveloping the nanomaterial consumer products inventory. Beilstein J. Nanotechnol. 2015, 6, 1769–1780. [CrossRef] 20. Cho, W.-S.; Dufﬁn, R.; Bradley, M.; Megson, I.L.; MacNee, W.; Howie, S.E.; Donaldson, K. NiO and Co3O4 nanoparticles induce lung DTH-like responses and alveolar lipoproteinosis. Eur. Respir. J. 2012, 39, 546–557. [CrossRef] 21. Park, E.-J.; Bae, E.; Yi, J.; Kim, Y.; Choi, K.; Lee, S.H.; Yoon, J.; Lee, B.C.; Park, K. Repeated-dose toxicity and inﬂammatory responses in mice by oral administration of silver nanoparticles. Environ. Toxicol. Pharmacol. 2010, 30, 162–168. [CrossRef] 22. De Jong, W.H.; Van Der Ven, L.T.; Sleijffers, A.; Park, M.V.; Jansen, E.H.; Van Loveren, H.; Vandebriel, R.J. Systemic and immunotoxicity of silver nanoparticles in an intravenous 28 days repeated dose toxicity study in rats. Biomaterials 2013, 34, 8333–8343. [CrossRef] 23. Lee, S.; Hwang, S.-H.; Jeong, J.; Han, Y.; Kim, S.-H.; Lee, D.-K.; Lee, H.-S.; Chung, S.-T.; Jeong, J.; Roh, C. Nickel oxide nanoparticles can recruit eosinophils in the lungs of rats by the direct release of intracellular eotaxin. Part. Fibre Toxicol. 2015, 13, 30. [CrossRef] 24. Roach, K.A.; Anderson, S.E.; Stefaniak, A.B.; Shane, H.L.; Kodali, V.; Kashon, M.; Roberts, J.R. Surface area-and mass-based comparison of ﬁne and ultraﬁne nickel oxide lung toxicity and augmentation of allergic response in an ovalbumin asthma model. Inhal. Toxicol. 2019, 31, 299–324. [CrossRef] [PubMed] Int. J. Mol. Sci. 2023, 24, 10104 17 of 19 25. Arts, J.H.; Kuper, C.F. Animal models to test respiratory allergy of low molecular weight chemicals: A guidance. Methods 2007, 41, 61–71. [CrossRef] [PubMed] 26. Pauluhn, J.; Mohr, U. Experimental approaches to evaluate respiratory allergy in animal models. Exp. Toxicol. Pathol. 2005, 56, 203–234. [CrossRef] 27. Lambrecht, B.N.; Hammad, H. The role of dendritic and epithelial cells as master regulators of allergic airway inﬂammation. Lancet 2010, 376, 835–843. [CrossRef] [PubMed] 28. Blank, F.; Rothen-Rutishauser, B.; Gehr, P. Dendritic cells and macrophages form a transepithelial network against foreign particulate antigens. Am. J. Respir. Cell Mol. Biol. 2007, 36, 669–677. [CrossRef] [PubMed] 29. Blank, F.; Wehrli, M.; Lehmann, A.; Baum, O.; Gehr, P.; von Garnier, C.; Rothen-Rutishauser, B.M. Macrophages and dendritic cells express tight junction proteins and exchange particles in an in vitro model of the human airway wall. Immunobiology 2011, 216, 86–95. [CrossRef] Fytianos, K.; Chortarea, S.; Rodriguez-Lorenzo, L.; Blank, F.; Von Garnier, C.; Petri-Fink, A.; Rothen-Rutishauser, B. Aerosol delivery of functionalized gold nanoparticles target and activate dendritic cells in a 3D lung cellular model. ACS Nano 2017, 11, 375–383. [CrossRef] 30. 31. Wang, G.; Zhang, X.; Liu, X.; Zheng, J. Co-culture of human alveolar epithelial (A549) and macrophage (THP-1) cells to study the potential toxicity of ambient PM2. 5: A comparison of growth under ALI and submerged conditions. Toxicol. Res. 2020, 9, 636–651. [CrossRef] 32. Wang, G.; Zhang, X.; Liu, X.; Zheng, J.; Chen, R.; Kan, H. Ambient ﬁne particulate matter induce toxicity in lung epithelial- endothelial co-culture models. Toxicol. Lett. 2019, 301, 133–145. [CrossRef] 33. Bedoret, D.; Wallemacq, H.; Marichal, T.; Desmet, C.; Calvo, F.Q.; Henry, E.; Closset, R.; Dewals, B.; Thielen, C.; Gustin, P. Lung interstitial macrophages alter dendritic cell functions to prevent airway allergy in mice. J. Clin. Investig. 2009, 119, 3723–3738. [CrossRef] Sibille, Y.; Reynolds, H.Y. Macrophages and Polymorphonuclear neutrophils in lung defense and Injury1-2. Am. Rev. Respir. Dis. 1990, 141, 471–501. [CrossRef] [PubMed] 34. 35. Toussaint, M.; Fievez, L.; Drion, P.; Cataldo, D.; Bureau, F.; Lekeux, P.; Desmet, C. Myeloid hypoxia-inducible factor 1α prevents airway allergy in mice through macrophage-mediated immunoregulation. Mucosal Immunol. 2013, 6, 485–497. [CrossRef] [PubMed] 36. Abbas, A.K.; Lichtman, A.H.; Pillai, S. Major histocompatibility complex molecules and antigen presentation to T lymphocytes. In Cellular and Molecular Immunology, 7th ed.; Elsevier/Saunders: Philadelphia, PA, USA, 2010; pp. 109–138. 37. Holt, P.G.; Haining, S.; Nelson, D.J.; Sedgwick, J.D. Origin and steady-state turnover of class II MHC-bearing dendritic cells in the 38. epithelium of the conducting airways. J. Immunol. 1994, 153, 256–261. [CrossRef] [PubMed] Forreryd, A.; Johansson, H.; Albrekt, A.-S.; Borrebaeck, C.A.; Lindstedt, M. Prediction of chemical respiratory sensitizers using GARD, a novel in vitro assay based on a genomic biomarker signature. PLoS ONE 2015, 10, e0118808. [CrossRef] [PubMed] 39. Hosoki, K.; Boldogh, I.; Sur, S. Neutrophil recruitment by allergens contribute to allergic sensitization and allergic inﬂammation. Curr. Opin. Allergy Clin. Immunol. 2016, 16, 45. [CrossRef] [PubMed] 40. Chandrasekar, B.; Smith, J.B.; Freeman, G.L. Ischemia-reperfusion of rat myocardium activates nuclear factor-κB and induces neutrophil inﬁltration via lipopolysaccharide-induced CXC chemokine. Circulation 2001, 103, 2296–2302. [CrossRef] 41. Hewitt, R.J.; Lloyd, C.M. Regulation of immune responses by the airway epithelial cell landscape. Nat. Rev. Immunol. 2021, 21, 347–362. [CrossRef] 42. Wuyts, A.; D’Haese, A.; Cremers, V.; Menten, P.; Lenaerts, J.-P.; De Loof, A.; Heremans, H.; Proost, P.; Van Damme, J. NH2-and COOH-terminal truncations of murine granulocyte chemotactic protein-2 augment the in vitro and in vivo neutrophil chemotactic potency. J. Immunol. 1999, 163, 6155–6163. [CrossRef] 43. Balamayooran, G.; Batra, S.; Cai, S.; Mei, J.; Worthen, G.S.; Penn, A.L.; Jeyaseelan, S. Role of CXCL5 in leukocyte recruitment to the lungs during secondhand smoke exposure. Am. J. Respir. Cell Mol. Biol. 2012, 47, 104–111. [CrossRef] 45. 44. de Souza, A.R.; Zago, M.; Eidelman, D.H.; Hamid, Q.; Baglole, C.J. Aryl hydrocarbon receptor (AhR) attenuation of subchronic cigarette smoke-induced pulmonary neutrophilia is associated with retention of nuclear RelB and suppression of intercellular adhesion molecule-1 (ICAM-1). Toxicol. Sci. 2014, 140, 204–223. [CrossRef] Foronjy, R.F.; Salathe, M.A.; Dabo, A.J.; Baumlin, N.; Cummins, N.; Eden, E.; Geraghty, P. TLR9 expression is required for the development of cigarette smoke-induced emphysema in mice. Am. J. Physiol.-Lung Cell. Mol. Physiol. 2016, 311, L154–L166. [CrossRef] [PubMed] Jeyaseelan, S.; Manzer, R.; Young, S.K.; Yamamoto, M.; Akira, S.; Mason, R.J.; Worthen, G.S. Induction of CXCL5 during inﬂammation in the rodent lung involves activation of alveolar epithelium. Am. J. Respir. Cell Mol. Biol. 2005, 32, 531–539. [CrossRef] [PubMed] 46. 47. Nikota, J.K.; Shen, P.; Morissette, M.C.; Fernandes, K.; Roos, A.; Chu, D.K.; Barra, N.G.; Iwakura, Y.; Kolbeck, R.; Humbles, A.A. Cigarette smoke primes the pulmonary environment to IL-1α/CXCR-2–dependent nontypeable Haemophilus inﬂuenzae– exacerbated neutrophilia in mice. J. Immunol. 2014, 193, 3134–3145. [CrossRef] [PubMed] 48. Bethin, K.E.; Vogt, S.K.; Muglia, L.J. Interleukin-6 is an essential, corticotropin-releasing hormone-independent stimulator of the adrenal axis during immune system activation. Proc. Natl. Acad. Sci. USA 2000, 97, 9317–9322. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10104 18 of 19 49. Waage, A.; Brandtzaeg, P.; Halstensen, A.; Kierulf, P.; Espevik, T. The complex pattern of cytokines in serum from patients with meningococcal septic shock. Association between interleukin 6, interleukin 1, and fatal outcome. J. Exp. Med. 1989, 169, 333–338. [CrossRef] 50. Gubernatorova, E.O.; Gorshkova, E.A.; Namakanova, O.A.; Zvartsev, R.V.; Hidalgo, J.; Drutskaya, M.S.; Tumanov, A.V.; Nedospasov, S.A. Non-redundant functions of IL-6 produced by macrophages and dendritic cells in allergic airway inﬂammation. Front. Immunol. 2018, 9, 2718. [CrossRef] 51. Bickel, M. The role of interleukin-8 in inﬂammation and mechanisms of regulation. J. Periodontol. 1993, 64 (Suppl. S5), 456–460. 52. Milara, J.; Mata, M.; Mauricio, M.D.; Donet, E.; Morcillo, E.J.; Cortijo, J. Sphingosine-1-phosphate increases human alveolar epithelial IL-8 secretion, proliferation and neutrophil chemotaxis. Eur. J. Pharmacol. 2009, 609, 132–139. [CrossRef] 53. Pantelidis, P.; Southcott, A.; Black, C.; Du Bois, R. Up-regulation of IL-8 secretion by alveolar macrophages from patients with 54. ﬁbrosing alveolitis: A subpopulation analysis. Clin. Exp. Immunol. 1997, 108, 95–104. [CrossRef] Jorens, P.G.; Van Damme, J.; De Backer, W.; Bossaert, L.; De Jongh, R.F.; Herman, A.G.; Rampart, M. Interleukin 8 (IL-8) in the bronchoalveolar lavage ﬂuid from patients with the adult respiratory distress syndrome (ARDS) and patients at risk for ARDS. Cytokine 1992, 4, 592–597. [CrossRef] 55. Nocker, R.E.; Schoonbrood, D.F.; van de Graaf, E.A.; Hack, E.; Lutter, R.; Jansen, H.M.; Out, T.A. lnterleukin-8 in airway inﬂammation in patients with asthma and chronic obstructive pulmonary disease. Int. Arch. Allergy Immunol. 1996, 109, 183–191. [CrossRef] [PubMed] 56. Ordonez, C.L.; Shaughnessy, T.E.; Matthay, M.A.; Fahy, J.V. Increased neutrophil numbers and IL-8 levels in airway secretions in acute severe asthma: Clinical and biologic signiﬁcance. Am. J. Respir. Crit. Care Med. 2000, 161, 1185–1190. [CrossRef] [PubMed] Smit, J.J.; Lukacs, N.W. A closer look at chemokines and their role in asthmatic responses. Eur. J. Pharmacol. 2006, 533, 277–288. [CrossRef] [PubMed] 57. 58. Zhang, Y.-L.; Han, D.H.; Kim, D.-Y.; Lee, C.H.; Rhee, C.-S. Role of interleukin-17A on the chemotactic responses to ccl7 in a murine allergic rhinitis model. PLoS ONE 2017, 12, e0169353. [CrossRef] 59. Ovidiu, B.; Mihai, D.; Ramona, C.; Catalin, T.; Anca, S.-P.; Roxana, S.-C.; Calin, G. The Relationship between Chemokine Ligand 3 and Allergic Rhinitis. Cureus 2020, 12, e7783. 60. Lloyd, C. Chemokines in allergic lung inﬂammation. Immunology 2002, 105, 144. [CrossRef] 61. Castan, L.; Magnan, A.; Bouchaud, G. Chemokine receptors in allergic diseases. Allergy 2017, 72, 682–690. [CrossRef] 62. Ray, J.L.; Holian, A. Sex differences in the inﬂammatory immune response to multi-walled carbon nanotubes and crystalline silica. Inhal. Toxicol. 2019, 31, 285–297. [CrossRef] 63. Golden, E.; Maertens, M.; Hartung, T.; Maertens, A. Mapping chemical respiratory sensitization: How useful are our current computational tools? Chem. Res. Toxicol. 2020, 34, 473–482. [CrossRef] 64. Alves, V.M.; Capuzzi, S.J.; Muratov, E.N.; Braga, R.C.; Thornton, T.E.; Fourches, D.; Strickland, J.; Kleinstreuer, N.; Andrade, C.H.; Tropsha, A. QSAR models of human data can enrich or replace LLNA testing for human skin sensitization. Green Chem. 2016, 18, 6501–6515. [CrossRef] 65. Council, N.R. Toxicity Testing in the 21st Century: A Vision and a Strategy; National Academies Press: Washington, DC, USA, 2007. 66. Hartung, T. Food for thought . . . on animal tests. ALTEX-Altern. Anim. Exp. 2008, 25, 3–16. [CrossRef] [PubMed] 67. Leist, M.; Hartung, T. Inﬂammatory ﬁndings on species extrapolations: Humans are deﬁnitely no 70-kg mice. Arch. Toxicol. 2013, 87, 563–567. [CrossRef] [PubMed] 68. Dik, S.; Rorije, E.; Schwillens, P.; van Loveren, H.; Ezendam, J. Can the direct peptide reactivity assay be used for the identiﬁcation of respiratory sensitization potential of chemicals? Toxicol. Sci. 2016, 153, 361–371. [CrossRef] [PubMed] 69. Lalko, J.F.; Dearman, R.J.; Gerberick, G.F.; Troutman, J.; Api, A.; Kimber, I. Reactivity of chemical respiratory allergens in the Peroxidase Peptide Reactivity Assay. Toxicol. Vitr. 2013, 27, 651–661. [CrossRef] 70. Palmberg, L.; Larsson, B.-M.; Malmberg, P.; Larsson, K. Induction of IL-8 production in human alveolar macrophages and human bronchial epithelial cells in vitro by swine dust. Thorax 1998, 53, 260–264. [CrossRef] 71. Hellman, P.; Eriksson, H. Early activation markers of human peripheral dendritic cells. Hum. Immunol. 2007, 68, 324–333. [CrossRef] 72. Banchereau, J.; Briere, F.; Caux, C.; Davoust, J.; Lebecque, S.; Liu, Y.-J.; Pulendran, B.; Palucka, K. Immunobiology of dendritic cells. Annu. Rev. Immunol. 2000, 18, 767–811. [CrossRef] 73. Banchereau, J.; Steinman, R.M. Dendritic cells and the control of immunity. Nature 1998, 392, 245–252. [CrossRef] 74. Cella, M.; Engering, A.; Pinet, V.; Pieters, J.; Lanzavecchia, A. Inﬂammatory stimuli induce accumulation of MHC class II complexes on dendritic cells. Nature 1997, 388, 782–787. [CrossRef] 75. Pierre, P.; Turley, S.J.; Gatti, E.; Hull, M.; Meltzer, J.; Mirza, A.; Inaba, K.; Steinman, R.M.; Mellman, I. Developmental regulation of 76. MHC class II transport in mouse dendritic cells. Nature 1997, 388, 787–792. [CrossRef] Sallusto, F.; Schaerli, P.; Loetscher, P.; Schaniel, C.; Lenig, D.; Mackay, C.R.; Qin, S.; Lanzavecchia, A. Rapid and coordinated switch in chemokine receptor expression during dendritic cell maturation. Eur. J. Immunol. 1998, 28, 2760–2769. [CrossRef] 77. Mbongue, J.C.; Nieves, H.A.; Torrez, T.W.; Langridge, W.H. The role of dendritic cell maturation in the induction of insulin- 78. dependent diabetes mellitus. Front. Immunol. 2017, 8, 327. [CrossRef] [PubMed] Schmidt, S.V.; Nino-Castro, A.C.; Schultze, J.L. Regulatory dendritic cells: There is more than just immune activation. Front. Immunol. 2012, 3, 274. [CrossRef] [PubMed] Int. J. Mol. Sci. 2023, 24, 10104 19 of 19 79. Liang, Y.; Harris, F.L.; Brown, L.A.S. Alcohol induced mitochondrial oxidative stress and alveolar macrophage dysfunction. BioMed Res. Int. 2014, 2014, 371593. [CrossRef] [PubMed] 80. Han, J.E.; Alvarez, J.A.; Staitieh, B.; Tangpricha, V.; Hao, L.; Ziegler, T.R.; Martin, G.S.; Brown, L.A.S. Oxidative stress in critically ill ventilated adults: Effects of vitamin D3 and associations with alveolar macrophage function. Eur. J. Clin. Nutr. 2018, 72, 744–751. [CrossRef] [PubMed] 81. Al-Humadi, N.H.; Siegel, P.D.; Lewis, D.M.; Barger, M.W.; Ma, J.Y.; Weissman, D.N.; Ma, J.K. Alteration of intracellular cysteine and glutathione levels in alveolar macrophages and lymphocytes by diesel exhaust particle exposure. Environ. Health Perspect. 2002, 110, 349–353. [CrossRef] [PubMed] 82. Galbiati, V.; Marinovich, M.; Corsini, E. Mechanistic understanding of dendritic cell activation in skin sensitization: Additional evidences to support potency classiﬁcation. Toxicol. Lett. 2020, 322, 50–57. [CrossRef] 83. Drasler, B.; Karakocak, B.B.; Tankus, E.B.; Barosova, H.; Abe, J.; Sousa de Almeida, M.; Petri-Fink, A.; Rothen-Rutishauser, B. An inﬂamed human alveolar model for testing the efﬁciency of anti-inﬂammatory drugs in vitro. Front. Bioeng. Biotechnol. 2020, 8, 987. [CrossRef] 84. Pollmächer, J.; Figge, M.T. Agent-based model of human alveoli predicts chemotactic signaling by epithelial cells during early Aspergillus fumigatus infection. PLoS ONE 2014, 9, e111630. [CrossRef] 85. Prasad, A.; Sedláˇrová, M.; Balukova, A.; Ovsii, A.; Rác, M.; Kˇrupka, M.; Kasai, S.; Pospíšil, P. Reactive oxygen species imaging in U937 cells. Front. Physiol. 2020, 11, 552569. [CrossRef] 86. Huang, D.W.; Sherman, B.T.; Lempicki, R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics 87. resources. Nat. Protoc. 2009, 4, 44–57. [CrossRef] [PubMed] Jiang, X.; Shen, C.; Rey-Ladino, J.; Yu, H.; Brunham, R.C. Characterization of murine dendritic cell line JAWS II and primary bone marrow-derived dendritic cells in Chlamydia muridarum antigen presentation and induction of protective immunity. Infect. Immun. 2008, 76, 2392–2401. [CrossRef] [PubMed] 88. Bessa, M.J.; Brandão, F.; Fokkens, P.H.; Leseman, D.L.; Boere, A.J.F.; Cassee, F.R.; Salmatonidis, A.; Viana, M.; Vulpoi, A.; Simon, S. In vitro toxicity of industrially relevant engineered nanoparticles in human alveolar epithelial cells: Air–liquid interface versus submerged cultures. Nanomaterials 2021, 11, 3225. [CrossRef] [PubMed] 89. Egger, M.; Jürets, A.; Wallner, M.; Briza, P.; Ruzek, S.; Hainzl, S.; Pichler, U.; Kitzmüller, C.; Bohle, B.; Huber, C.G. Assessing protein immunogenicity with a dendritic cell line-derived endolysosomal degradome. PLoS ONE 2011, 6, e17278. [CrossRef] 90. Paardekooper, L.M.; Dingjan, I.; Linders, P.T.; Staal, A.H.; Cristescu, S.M.; Verberk, W.C.; Van den Bogaart, G. Human monocyte- derived dendritic cells produce millimolar concentrations of ROS in phagosomes per second. Front. Immunol. 2019, 10, 1216. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_genes14061303	Article Characterization and Genome Study of a Newly Isolated Temperate Phage Belonging to a New Genus Targeting Alicyclobacillus acidoterrestris Dziyana Shymialevich 1 and Barbara Sokołowska 4,* , Michał Wójcicki 2 , Olga ´Swider 3 , Paulina ´Srednicka 2 1 Culture Collection of Industrial Microorganisms—Microbiological Resources Center, Department of 2 Microbiology, Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute, Rakowiecka 36 Str., 02-532 Warsaw, Poland; [email protected] Laboratory of Biotechnology and Molecular Engineering, Department of Microbiology, Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute, Rakowiecka 36 Str., 02-532 Warsaw, Poland; [email protected] (M.W.); [email protected] (P.´S.) 3 Department of Food Safety and Chemical Analysis, Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute, Rakowiecka 36 Str., 02-532 Warsaw, Poland; [email protected] 4 Department of Microbiology, Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute, Rakowiecka 36 Str., 02-532 Warsaw, Poland * Correspondence: [email protected] Abstract: The spoilage of juices by Alicyclobacillus spp. remains a serious problem in industry and leads to economic losses. Compounds such as guaiacol and halophenols, which are produced by Alicyclobacillus, create undesirable ﬂavors and odors and, thus, decrease the quality of juices. The inactivation of Alicyclobacillus spp. constitutes a challenge because it is resistant to environmental factors, such as high temperatures, and active acidity. However, the use of bacteriophages seems to be a promising approach. In this study, we aimed to isolate and comprehensively characterize a novel bacteriophage targeting Alicyclobacillus spp. The Alicyclobacillus phage strain KKP 3916 was isolated from orchard soil against the Alicyclobacillus acidoterrestris strain KKP 3133. The bacterial host’s range and the effect of phage addition at different rates of multiplicity of infections (MOIs) on the host’s growth kinetics were determined using a Bioscreen C Pro growth analyzer. The Alicyclobacillus phage strain KKP 3916, retained its activity in a wide range of temperatures (from 4 ◦C to 30 ◦C) and active acidity values (pH from 3 to 11). At 70 ◦C, the activity of the phage decreased by 99.9%. In turn, at 80 ◦C, no activity against the bacterial host was observed. Thirty minutes of exposure to UV reduced the activity of the phages by almost 99.99%. Based on transmission-electron microscopy (TEM) and whole-genome sequencing (WGS) analyses, the Alicyclobacillus phage strain KKP 3916 was classiﬁed as a tailed bacteriophage. The genomic sequencing revealed that the newly isolated phage had linear double-stranded DNA (dsDNA) with sizes of 120 bp and 131 bp and 40.3% G+C content. Of the 204 predicted proteins, 134 were of unknown function, while the remainder were annotated as structural, replication, and lysis proteins. No genes associated with antibiotic resistance were found in the genome of the newly isolated phage. However, several regions, including four associated with integration into the bacterial host genome and excisionase, were identiﬁed, which indicates the temperate (lysogenic) life cycle of the bacteriophage. Due to the risk of its potential involvement in horizontal gene transfer, this phage is not an appropriate candidate for further research on its use in food biocontrol. To the best of our knowledge, this is the ﬁrst article on the isolation and whole-genome analysis of the Alicyclobacillus-speciﬁc phage. Keywords: bacteriophages; Alicyclobacillus acidoterrestris; whole-genome sequencing; genomic analysis; functional annotation; transmission-electron microscopy; food biocontrol Citation: Shymialevich, D.; Wójcicki, M.; ´Swider, O.; ´Srednicka, P.; Sokołowska, B. Characterization and Genome Study of a Newly Isolated Temperate Phage Belonging to a New Genus Targeting Alicyclobacillus acidoterrestris. Genes 2023, 14, 1303. https://doi.org/10.3390/ genes14061303 Academic Editor: Silvia Turroni Received: 29 May 2023 Revised: 14 June 2023 Accepted: 19 June 2023 Published: 20 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Genes 2023, 14, 1303. https://doi.org/10.3390/genes14061303 https://www.mdpi.com/journal/genes genesG C A TT A C GG C A TGenes 2023, 14, 1303 2 of 21 1. Introduction Fruit juices are among the most popular beverages and play an important role in human health due to their nutrient content and bioactive compounds [1,2]. They are low-calorie drinks rich in nutrients and bioactive compounds, such as vitamins, proteins, carbohydrates, polyphenols, minerals, enzymes, ﬁber, and antioxidants [3–5]. According to their method of preservation, juices are divided into fresh-squeezed, chilled, frozen, pasteurized, and concentrated [6,7]. Currently, consumers prefer to opt for fruit juices as an easy way to consume the ﬁve servings of fruits and vegetables recommended by the World Health Organization (WHO) for a healthy diet [8]. The low pH value of fruit juices reduces the growth of pathogenic microorganisms, making these products safe and attractive to consumers [9]. In recent years, many cases related to microbiologically spoiled fruit juices, leading to heavy economic losses for industry, were reported. These spoilages were mainly re- lated to the ﬂavor-producing bacteria, which impoverished the sensory characteristics of the products and made them unacceptable to consumers [10–12]. The main source of microorganisms in this type of food is the environment, primarily through contaminated soil, water, process machinery, and filling lines with inadequate hygiene protocols [13–15]. The microorganisms found on the surfaces of fruits and vegetables are an abundant and diverse group [6,16]. This group includes bacteria, as well as yeasts and molds, and there is a clear relationship between the composition of the microﬂora and the chemical com- position of fruits and vegetables, especially the ratio of carbohydrates to proteins and acidity [17,18]. In the juicing industry, the high hydrostatic pressure (HHP) technique effec- tively eliminates both saprophytic and pathogenic bacteria [19–21]. However, the challenge lies with spore-forming bacteria, which survive at high temperatures and during pressur- ization, and then germinate and grow in the food matrix [22,23]. The most widespread species responsible for juice spoilage is A. acidoterrestris [20,24]. The bacteria from the Alicyclobacillus genus include Gram-positive, aerobic, and spore-forming bacilli [25,26]. The contamination of the product with this bacterium is difﬁcult to detect because there is no gas production or packaging swelling [26]. A noticeable sign of spoilage is a taste described as medicinal, phenolic, and antiseptic, which is mainly associated with the production of guaiacol (2-methoxyphenol) [27,28], but also halophenols, such us 2,6-dibromophenol, and 2,6-dichlorophenol [25,29]. The use of specific, obligately lytic bacteriophages is one of the methods that is increas- ingly considered to eradicate bacteria from environments in the food industry [30,31]. Phage cocktails can be used to disinfect production surfaces, during raw material washing, or they can be added directly to the ﬁnished product [31–33]. This biological method of food preservation has been approved for use in several countries outside the European Union, such as the USA, Canada, and Switzerland [33–35]. Recently, the use of bacteriophages in the food industry and therapeutics has become highly popular, and many scientiﬁc centers around the world research the development of effective phage biopreparations [36,37]. Bacteriophages can infect and replicate only in vegetative bacterial cells [38], making it difﬁcult to use this method to eliminate spore-forming bacteria [39]. In an article by Butala and Dragoš [39], the relationship between phages and hosts that form spores was described. The authors found that phages speciﬁc to spore-forming bacteria may have evolved molec- ular mechanisms, enabling them to cope with various stages of host development. For example, prophages have eliminated the mechanisms of translocation of their genomes into the pre-spore. Instead, phage DNA is protected from environmental conditions until the spore germinates [36]. After spore germination, the continuation of the phage’s lytic cycle occurs. Prophages can alter the frequency of sporulation and modulate the timely induction of sporulation and spore germination [36]. Recently, it was shown that prophage phi3T targeting Bacillus subtilis increases the rate of sporulation and affects the spore-germination time [40]. The simultaneous application of high-pressure techniques and phage cocktails rep- resents a potentially effective solution to the above problem. This combined method can Genes 2023, 14, 1303 3 of 21 reduce the growth of vegetative bacterial cells and increase the germination of spores, which can be subsequently eliminated by bacteriophages [41–43]. The necessary condition for selecting the appropriate phage is to determine its ability to infect the widest possible range of bacteria in the food environment and its multiplication cycle. Therefore, this study aimed at the genomic and functional characterization of a newly isolated bacteriophage tar- geting the guaiacol-producing strain of A. acidoterrestris, with an emphasis on its potential use in food biopreservation. 2. Materials and Methods 2.1. Source of Bacterial Host Strains Forty-six bacterial strains used in this research were deposited in the Culture Col- lection of Industrial Microorganisms—Microbiological Resources Center of the Depart- ment of Microbiology, at the Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute (IAFB; Warsaw, Poland). All strains were iso- lated from food products. The ability to produce guaiacol was determined for each of the strains [44,45]. The belonging of the strains to the Alicyclobacillus genus was conﬁrmed by ampliﬁcation of the 16S rRNA gene region. Sequencing was outsourced to Genomed S.A. company (Warsaw, Poland). Raw sequences were analyzed using BLASTn (NCBI) and deposited in the GenBank database. Guaiacol-producing A. acidoterrestris strain KKP 3133 was isolated in 2002 from apple-juice concentrate, and deposited as TO-117/02 [23,41]. 2.2. Bacteriophage Isolation Alicyclobacillus phage strain KKP 3916 was isolated from soil sample originating in an apple orchard near Warsaw, Poland. In total, 200 mL of saline solution was poured over the 20 g of soil, which was left overnight on a shaker set at 180 rpm, 23 ◦C (Innova 4000 Incubator Shaker, New Brunswick Scientiﬁc Co., Inc. Edison, NJ, USA). Subsequently, the mixture with soil was centrifuged at 8000× g for 8 min (ultracentrifuge Sorvall LYNX 6000, Thermo Fisher Scientiﬁc, Watertown, MA, USA) to separate bacteria and soil from the solution with bacteriophages. The supernatant was ﬁltered through a 0.45 µm syringe ﬁlter (Minisart® NML Cellulose Acetate; Sartorius, Goettingen, Germany). Aiming at the isolation of phages targeting Alicyclobacillus sp., 20 mL of double-concentrated BAT broth (Merck, Darmstadt, Germany), 20 mL of ﬁltered soil solution, and 1 mL of an overnight Alicyclobacillus culture were added to 50 mL falcons. The phage-culture medium was incubated at 45 ◦C for 24 h. Next, the culture was centrifuged at 8000× g for 10 min to separate bacteria from the proliferated bacteriophages. The supernatant was ﬁltered through a 0.45 µm syringe ﬁlter [46]. The concentration of the phages (phage titer; in PFU mL−1) in the obtained lysates was determined using the double-layer agar-plate method in triplicate [47,48]. For this purpose, a series of dilutions of phage lysate was prepared. In total, 100 µL of fresh bacterial suspension in BAT broth was added to 500 µL of appropriately diluted lysate. The suspension was gently stirred and left at 20 ◦C for 20 min to allow the phages to adsorb to the host-cell surface. After adsorption, the suspen- sion was poured onto a solid BAT agar (Merck, Germany) and 4 mL of liquid and cooled to 50 ◦C soft BAT agar (BAT broth with 0.75% agar-agar, pH 4.0–4.2) was added. The whole mixture was stirred and spread evenly over the surface of the solid medium. After drying, the dishes were kept upside-down overnight at 45 ◦C. 2.3. Puriﬁcation of Bacteriophages In order to purify the isolated phages, the resulting individual plaque was cut with an injection needle and transferred to the tubes supplemented with 1 mL of SM buffer (5.8 g L−1 of sodium chloride, 2.0 g L−1 of magnesium sulfate seven-hydrate, 50 mL of 1 M Tris-HCl (pH 7.5), 5 mL of 2% gelatin solution), and 10 µL of chloroform [49]. The tubes were left at 20 ◦C on a laboratory cradle for 2 h to diffuse the viral particles from the agar into the buffer and to inactivate the remaining bacteria in the lysate. One mL of fresh bacterial culture and the resulting puriﬁed phage suspension were transferred into tubes Genes 2023, 14, 1303 4 of 21 containing 40 mL of BAT broth. After incubation overnight at 45 ◦C, the suspension was centrifuged at 8000× g for 10 min and ﬁltered through a 0.45 µm ﬁlter. In the lysate obtained in this way, phage titers were determined using the double-layer agar-plate method [47,48]. Puriﬁcation was performed in four rounds of single-plaque passage. The resulting lysates were stored in refrigerator (at 4 ◦C) and frozen with the addition of 20% glycerol at −80 ◦C and −150 ◦C. 2.4. Determination of the Bacterial Hosts Range In this study, ﬁfty-three Alicyclobacillus strains were used to examine the range of bacterial hosts of the bacteriophage (of which forty-six strains were from the IAFB, and seven strains were from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures). Puriﬁed and ampliﬁed phage lysates (phage titers ~106 PFU mL−1) were used and the double-layer agar-plate method was applied to determine the phage’s ability to infect selected bacteria strains [47,48]. Plates were incubated at 45 ◦C for 24 h in triplicate. After incubation, clearing zones were determined on any bacterial digests recorded according to the assays: “++”—transparent plaques; “+”—cloudy plaques; “−”—no plaques (insensitive bacterial strain). 2.5. Changes in the Growth Kinetics of A. acidoterrestris Strain KKP 3133 after Phage Infection The kinetic growth of A. acidoterrestris strain KKP 3133 infected with newly isolated bacteriophages was measured in ten replicates using a Bioscreen C Pro automated growth analyzer (Yo AB Ltd., Growth Curves, Helsinki, Finland). A growth curve was prepared for bacterial strain (data unpublished in this paper). For this purpose, the bacterial culture was inoculated every hour onto BAT agar, incubated at 45 ◦C for 24 h, and optical density was measured simultaneously (DU® 640 Spectrophotometer, Beckman Instruments, Inc., Fullerton, CA, USA). The dependence of optical density on the number of bacterial cells was determined (performed in triplicate). The overnight bacterial culture was diluted with BAT broth until the optical density was appropriate for the phage titer. Phage lysates were prepared so that the values of MOI (multiplicity of infections) were 1.0. and 0.1. Next, 180 µL of BAT broth was pipetted into multi-well plates and 10 µL of appropriately diluted bacterial culture and phage lysate was added. The plate was placed in the Bioscreen C Pro for 24 h with an average stirring intensity of 15 s before measurement. Optical density measurement was performed every 30 min using a wide band of wavelengths, ranging from 400 nm to 600 nm (OD400–600). The effectiveness of bacterial growth inhibition resulting from phage application was assessed with regard to a control culture that contained only bacteria. Based on the obtained curves illustrating the dependence of the change in optical density and the duration of the culture, the speciﬁc growth-rate coefﬁcient (µ) was calculated according to the following formula: µ = (ln ODmax − ln ODmin)/∆t (1) where ln ODmax—is the natural logarithm of the maximum value of the exponential growth of the culture, ln ODmin—is the natural logarithm of the minimal value of exponential growth of the culture, and ∆t—duration of the exponential growth of culture, (h). 2.6. Effect of Selected Factors on the Preservation of the Phage Activity In this stage of the experiment, the stability of bacteriophages after exposure to a wide range of temperatures, pH values, and UV radiation times was examined. To determine the activity of phage lysates at various levels of active acidity, 100 µL of phage lysate was added to test tubes containing 9.9 mL of sterile saline (0.85% NaCl) with a ﬁxed active acidity (pH 2–12). The mixture was maintained at 20 ◦C for 1 h. To determine phage activity at different temperatures (−20 ◦C, 4 ◦C, 20 ◦C, 30 ◦C, 40 ◦C, 50 ◦C, 60 ◦C, 70 ◦C, and 80 ◦C), 100 µL of the phage lysate was added to 9.9 mL of saline with pH 7.0. The mixture was maintained for 1 h at the speciﬁed temperatures. To determine the effect of UV radiation, phage lysate was exposed to UV for 0, 5, 10, 15, 25, 30, and 60 min. The experiments were Genes 2023, 14, 1303 5 of 21 carried out in three independent replicates. After incubation under the selected physical or chemical conditions, the activity of phages was determined through the double-layer agar-plate method. 2.7. Determination of Morphological Characteristics of Phage Transmission-electron microscopy (TEM) was used to determine the morphological characteristics of the isolated bacteriophage. Phage lysate in a volume of 1 mL was cen- trifuged at 20 ◦C at 14,500 rpm for 40 min (MiniSpin® plus centrifuge, Eppendorf, Hamburg, Germany). The supernatant was removed, and the precipitate was suspended in 2 mL of 100 mM cold ammonium acetate (ﬁltered through a 0.22 µm syringe ﬁlter). The precipitate was dissociated by repeated pipetting and centrifuged again. The whole procedure was repeated four times. After centrifugation, the precipitate was rinsed off the wall of the Eppendorf tube with 50 µL of ammonium acetate according to the procedure of Acker- mann [49], with modiﬁcation. Two µL of phage suspension in ammonium acetate was coated onto carbon-sputtered copper–wolfram mesh grids. After drying, the preparation was stained for 1 min in a 2% uranyl acetate solution (Warchem, Warsaw, Poland). Prepared samples were dried for 12 h at ambient temperature under sterile conditions [50,51] and visualized under JEM 1400 PLUS transmission-electron microscope (Japan Electron Optics Laboratory Co., Ltd., Tokyo, Japan) at 100,000–200,000× magniﬁcation, at a voltage of 80 kV [52]. 2.8. Extraction of Bacteriophage Genomic DNA Bacteriophage genomic DNA from lysates was isolated using the PureLinkTM RNA/DNA Mini Kit (Thermo Fisher Scientiﬁc Inc., Carlsbad, CA, USA) according to the manufacturer’s protocol, with modiﬁcations by Wójcicki et al. [53]. The ﬁrst step was to concentrate the phage lysate by ultracentrifugation. For this purpose, 40 mL of phage lysate and 8 mL of precipitating solution (PEG-NaCl: 2.5 M NaCl, 20% PEG 8000) were transferred into bottle for ultracentrifugation. The resulting solution was incubated overnight on ice. The lysate was then centrifuged at 27,000 rpm for 1.5 h at 4 ◦C (Sorvall LYNX 6000 ultracen- trifuge, Thermo Fisher Scientiﬁc, Watertown, MA, USA). The supernatant was carefully removed, and the phage pellet was resuspended in 400 µL Lysis Buffer (containing 5.6 µg Carrier RNA) and vortexed. Subsequently, 50 µL of the proteinase K was added and the prepared solution was incubated for 1 h at 56 ◦C with shaking at 900 rpm (ThermoMixer C, Eppendorf, Hamburg, Germany). After incubation, the tube was brieﬂy centrifuged, and 300 µL of 100% ice-cold ethanol (molecular biology grade) was added. The tube was vortexed for 10 s and left at room temperature for 5 min. After incubation, sample was brieﬂy vortexed to remove droplets from the tube walls. The content of the tube was trans- ferred to a viral spin and centrifuged for 1 min at 10,000 rpm (MiniSpin® plus centrifuge, Eppendorf, Hamburg, Germany). Next, 500 µL of wash buffer was added and centrifuga- tion for 1 min at 10,000 rpm was carried out. The ﬁltrate was removed, and the column was centrifuged for 3 min at 14,500 rpm. The next step was the elution of the genetic material. The viral spin column was transferred to a new tube and 20 µL of RNase-free water was added. The sample was incubated for 1 min at 20 ◦C and then centrifuged for 1 min at 14,500 rpm. The ﬁltrate was applied to the same column and centrifuged again. The purity of the obtained DNA was evaluated with a Nanodrop ND-1000 spectrophotometer (Thermo Fisher Scientiﬁc, Watertown, MA, USA), and DNA concentration was determined with a Qubit 4.0 ﬂuorometer using the Qubit dsDNA BR Assay Kit (Invitrogen, Carlsbad, CA, USA). The DNA was stored at 4 ◦C until whole-genome sequencing (WGS) analysis. 2.9. Genome Sequencing and Bioinformatics Analysis The company genXone SA (Złotniki, Poland) was commissioned to sequence the entire phage genome. The DNA library was prepared using Rapid Barcoding Kit reagents (Oxford Nanopore Technologies, Oxford, UK), according to the manufacturer’s protocol. A sequencing depth of at least 50× genome coverage was assumed. The NGS sequencing was Genes 2023, 14, 1303 6 of 21 performed using the nanopore technology on the GridION X5 sequencing device (Oxford Nanopore Technologies, Oxford, UK) under the control of MinKnow v22.10.5. Guppy v6.3.8 (Oxford Nanopore Technologies, Oxford, UK), which was applied to call bases and to perform barcode demultiplexing, generating a .fastq ﬁle for each barcode. The de novo assembly of genome was performed in Flye v2.8.1 software [54], while annotation of phage genomes was performed in Phanotate v1.5.0 [55] and PhaGAA [56] software. Proksee [57] software was used to visualize phage genome. The viral proteomic tree of phage genome was calculated by BIONJ based on genomic distance matrixes and the mid-point rooted, and was represented in the circular view. Branch length was log-scaled. The sequence and taxonomic data were based on Virus-Host DB [58]. The tree was generated using the ViPTree server [59]. Phage similarity on scatter 2D plot computationally was predicted and rendered through PhageAI software [60]. The phage genome was deposited in the GenBank database. 2.10. Statistical Analysis All experiments were repeated at least three times. Data presented graphically or in tables were subjected to statistical analysis using Graph Prism 8.02 software (GraphPad Software Inc., San Diego, CA, USA). One-way ANOVA followed by Tukey’s test with a 95% conﬁdence interval (α = 0.05) was used to evaluate the effect of selected physical and chemical factors on phage activity. 3. Results and Discussion 3.1. Determination of Plaque and TEM Morphology of Bacteriophage Plaques usually result from a combination of phage propagation, viral diffusion, and the lysis of bacterial hosts. Plaques can also result from partial bacterial growth suppression in connection with phage propagation, even without direct bactericidal action [61,62]. The morphology of phage plaques is important for initial identiﬁcation, as transparent plaques indicate lytic (virulent) bacteriophages and cloudy plaques indicate lysogenic (temperate) bacteriophages [62]. The size of the bacteriophage is inversely proportional to the size of the plaque, as a small bacteriophage diffuses more easily in soft agar, causing a large plaque diameter [63,64]. In addition, the morphologies of lysates are linked to such ﬁtness characteristics as the size of the burst, the adsorption rate and lysis time, and the diffusion rate of the phage in a particular medium [65]. It is also possible to identify phage mutants based on changes in plaque morphology compared to the characterized wild-type phage, including molecularly engineered phages [66,67]. Figure 1A shows the plaque obtained for the Alicyclobacillus phage strain KKP 3916. The plaques were characterized by a transparent center with cloudy edges. Electronograms obtained using TEM allowed the visualization of the morphology of the bacteriophage (Figure 1B,C). The bacteriophage has a complex structure (tailed phages), containing an icosahedral symmetrical head (capsid) and a long, contractile tail. Figure 1. Morphology of plaques on double-layer agar plate (A) and electron micrographs from the TEM (B,C) of Alicyclobacillus phage strain KKP 3916 targeting A. acidoterrestris strain KKP 3133. Genes 2023, 14, x FOR PEER REVIEW 7 of 20 Figure 1. Morphology of plaques on double-layer agar plate (A) and electron micrographs from the TEM (B,C) of Alicyclobacillus phage strain KKP 3916 targeting A. acidoterrestris strain KKP 3133. 3.2. Range of Bacterial Hosts Forty-six strains isolated from food products were used to deﬁne the bacterial host range (Table 1). Moreover, in our study, we used seven strains from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures isolated from diﬀerent environments (Table 1). Table 1. Range of bacterial hosts of the genus Alicyclobacillus for the newly isolated Alicyclobacillus phage strain KKP 3916. Bacterial Host Strain (n = 53) Year of Isolation Source of Isolation Guaiacol Production GenBank Accession Number Infect with Phage Strains from the Culture Collection of Industrial Microorganisms—Microbiological Resources Center, IAFB (n = 46) A. acidoterrestris strain KKP 377 2011 Concentrated cherry juice + OQ285871 ++ A. acidoterrestris strain KKP 381 2005 Concentrated apple juice + OQ284561 − A. acidoterrestris strain KKP 382 2005 Concentrated apple juice + OQ284562 − A. acidoterrestris strain KKP 383 2006 Concentrated apple juice + OQ285847 − A. acidoterrestris strain KKP 394 2005 Concentrated apple juice + OQ284103 − A. acidoterrestris strain KKP 395 2002 Concentrated apple juice + OQ284104 + A. acidoterrestris strain KKP 400 2002 Concentrated apple juice + OQ285872 − A. acidoterrestris strain KKP 404 2012 Concentrated black currant juice + OQ284075 − A. acidoterrestris strain KKP 564 2006 Concentrated apple juice + OQ285856 − A. acidoterrestris strain KKP 596 2006 Concentrated apple juice + OQ285855 − A. acidoterrestris strain KKP 609 2013 Concentrated strawberry juice + OQ285862 − A. acidoterrestris strain KKP 610 2006 Concentrated apple juice + OQ285860 − A. acidoterrestris strain KKP 611 2006 Concentrated apple juice − OQ285857 − A. acidoterrestris strain KKP 613 2006 Concentrated apple juice + OQ318516 + Alicyclobacillus fastidiosus strain KKP 3000 2019 Soil from a pear orchard + KY088044 − A. fastidiosus strain KKP 3001 2019 Soil from a pear orchard + KY088045 − A. fastidiosus strain KKP 3002 2002 Soil from a pear orchard + KY088046 + A. acidoterrestris strain KKP 3133 2002 Concentrated apple juice + OQ261766 ++ A. acidoterrestris strain KKP 3134 2004 Carbonated ingredient + OQ263350 − Alicyclobacillus acidocaldarius strain KKP 3135 2004 Beverage powder − OQ285905 ++ A. acidoterrestris strain KKP 3136 2006 Spoiled orange drink + OQ263355 − A. acidoterrestris strain KKP 3137 2006 Concentrated apple juice + OQ263363 − A. acidoterrestris strain KKP 3138 2006 Banana nectar + OQ262932 − A. acidoterrestris strain KKP 3139 2007 Concentrated apple juice + OQ263354 − A. acidoterrestris strain KKP 3140 2007 Concentrated apple juice + OQ262861 − Genes 2023, 14, 1303 7 of 21 3.2. Range of Bacterial Hosts Forty-six strains isolated from food products were used to deﬁne the bacterial host range (Table 1). Moreover, in our study, we used seven strains from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures isolated from different environments (Table 1). Table 1. Range of bacterial hosts of the genus Alicyclobacillus for the newly isolated Alicyclobacillus phage strain KKP 3916. Bacterial Host Strain (n = 53) Year of Isolation Source of Isolation Guaiacol Production GenBank Accession Number Infect with Phage Strains from the Culture Collection of Industrial Microorganisms—Microbiological Resources Center, IAFB (n = 46) A. acidoterrestris strain KKP 377 A. acidoterrestris strain KKP 381 A. acidoterrestris strain KKP 382 A. acidoterrestris strain KKP 383 A. acidoterrestris strain KKP 394 A. acidoterrestris strain KKP 395 A. acidoterrestris strain KKP 400 A. acidoterrestris strain KKP 404 A. acidoterrestris strain KKP 564 A. acidoterrestris strain KKP 596 A. acidoterrestris strain KKP 609 A. acidoterrestris strain KKP 610 A. acidoterrestris strain KKP 611 A. acidoterrestris strain KKP 613 Alicyclobacillus fastidiosus strain KKP 3000 A. fastidiosus strain KKP 3001 A. fastidiosus strain KKP 3002 A. acidoterrestris strain KKP 3133 A. acidoterrestris strain KKP 3134 Alicyclobacillus acidocaldarius strain KKP 3135 A. acidoterrestris strain KKP 3136 A. acidoterrestris strain KKP 3137 A. acidoterrestris strain KKP 3138 A. acidoterrestris strain KKP 3139 A. acidoterrestris strain KKP 3140 A. acidoterrestris strain KKP 3141 A. acidoterrestris strain KKP 3142 A. acidoterrestris strain KKP 3143 A. acidocaldarius strain KKP 3144 A. acidoterrestris strain KKP 3145 2011 2005 2005 2006 2005 2002 2002 2012 2006 2006 2013 2006 2006 2006 2019 2019 2002 2002 2004 2004 2006 2006 2006 2007 2007 2008 2008 2008 2009 2009 Concentrated cherry juice Concentrated apple juice Concentrated apple juice Concentrated apple juice Concentrated apple juice Concentrated apple juice Concentrated apple juice Concentrated black currant juice Concentrated apple juice Concentrated apple juice Concentrated strawberry juice Concentrated apple juice Concentrated apple juice Concentrated apple juice Soil from a pear orchard Soil from a pear orchard Soil from a pear orchard Concentrated apple juice Carbonated ingredient Beverage powder Spoiled orange drink Concentrated apple juice Banana nectar Concentrated apple juice Concentrated apple juice Concentrated blackcurrant juice Concentrated apple juice Fresh apple Sugar syrup Concentrated apple juice + + + + + + + + + + + + − + + + + + + − + + + + + + + + + + OQ285871 OQ284561 OQ284562 OQ285847 OQ284103 OQ284104 OQ285872 OQ284075 OQ285856 OQ285855 OQ285862 OQ285860 OQ285857 OQ318516 KY088044 KY088045 KY088046 OQ261766 OQ263350 OQ285905 OQ263355 OQ263363 OQ262932 OQ263354 OQ262861 KX371237 KX371238 KX371239 OQ285906 KX371240 ++ − − − − + − − − − − − − + − − + ++ − ++ − − − − − ++ − − − − Genes 2023, 14, 1303 Table 1. Cont. Bacterial Host Strain (n = 53) Year of Isolation Source of Isolation Guaiacol Production A. acidoterrestris strain KKP 3146 A. acidoterrestris strain KKP 3147 A. acidoterrestris strain KKP 3148 A. acidocaldarius strain KKP 3149 A. acidoterrestris strain KKP 3150 A. acidoterrestris strain KKP 3151 A. acidoterrestris strain KKP 3152 A. acidoterrestris strain KKP 3153 A. acidoterrestris strain KKP 3154 A. acidoterrestris strain KKP 3156 A. acidocaldarius strain KKP 3157 A. acidoterrestris strain KKP 3194 A. acidoterrestris strain KKP 3195 A. acidoterrestris strain KKP 3347 A. acidoterrestris strain KKP 3348 A. acidoterrestris strain KKP 3349 2009 2009 2009 2010 2010 2011 2011 2011 2011 2012 2013 2020 2020 2020 2020 2020 Spoiled apple drink Concentrated apple juice Spoiled apple juice Concentrated strawberry juice Concentrated red beet juice Concentrated cherry juice Concentrated cherry juice Concentrated raspberry juice Tomato juice Tomato juice Concentrated apple juice Soil from an apple orchard Soil from an apple orchard Soil from an apple orchard Soil from an apple orchard Soil from an apple orchard + + − − + + + + + + − + + + + + GenBank Accession Number OQ261717 OQ263351 OQ261718 KX371241 KX371242 KX371243 KX371245 KX371246 OQ285874 OQ263364 OQ285907 KY088041 KY088042 KY088043 KY088047 MW332524 Strains from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures (n = 7) Alicyclobacillus acidocaldarius subsp. acidocaldarius strain DSM 446 Alicyclobacillus hesperidum strain DSM 12489 Alicyclobacillus herbarius strain DSM 13609 Alicyclobacillus acidophilus strain DSM 14558 A. fastidiosus strain DSM 17978 A. acidoterrestris strain DSM 2498 A. acidoterrestris strain DSM 3922 1990 1998 2000 2001 2003 1982 1986 Acid hot spring Solfataric soils Dry ﬂower of Hibiscus Acidic beverage, which was off-ﬂavored Apple juice Juice Garden soil − + + + − + + CP001727 FNOJ00000000 AUMH00000000 AB076660 NR_041471 AB059675 AURB00000000 8 of 21 Infect with Phage − − ++ − − ++ − − − − ++ − + − − − − − − − − ++ − Infected with phage: “++”—transparent plaques; “+”—cloudy plaques; “−”—no plaques (insensitive bacterial strain). The Alicyclobacillus phage strain KKP 3916 infected 22.6% (12/53) of the tested strains from the Alicyclobacillus genus. Most of the phage lysis zones observed were transparent plaques with cloudy edges. Examples of plaques on BAT agar are shown in Figure 2. Of the 41 strains of the A. acidoterrestris species, the newly isolated bacteriophage had the ability to infect nine. No lysis zone was observed against the strains of A. hesperidum, A. herbarius, or A. acidophilus. Out of four strains from A. fastidiosus species and ﬁve strains from A. acidocaldarius species, the Alicyclobacillus phage strain KKP 3916 had the ability to infect one (KKP 3002) and two (KKP 3135 and KKP 3157) strains, respectively. Of the twelve bacterial strains infected by Alicyclobacillus phage strain KKP 3916, nine were guaiacol-producing. Genes 2023, 14, 1303 9 of 21 Figure 2. Plaque morphology (on double-layer agar plate) of Alicyclobacillus phage strain KKP 3916 infecting A. acidoterrestris strain KKP 3148 (A), A. acidoterrestris strain KKP 3151 (B), A. acidocaldarius strain KKP 3157 (C), and A. acidoterrestris strain KKP 3195 (D). The host range plays an important role in the use of bacteriophage in therapy or food biocontrol. The elimination of only speciﬁc bacterial strains requires a narrower host range [68,69], to prevent the other microorganisms from being compromised (which is important in some industries, e.g., dairy). To eliminate pathogenic bacteria within a species, it is desirable to search for bacteriophages with wide host ranges [70]. This is analogous to the use of broad-spectrum antibiotics without the need to identify the pathogen or sensitivity to the antibiotic. In the absence of broad-spectrum phage isolation, a preparation containing several bacteriophages can be used [68,70,71]. 3.3. Evaluation of the Activity of Phage against Bacterial Hosts The effectiveness of a phage in inactivating bacteria is one of the critical characteristics that should be considered for potential candidates in phage therapy or food biocontrol. To evaluate this feature in relation to the tested bacteriophage, growth curves were plotted for each bacterial host strain using the Bioscreen C Pro automated growth analyzer. The growth curves showed that the Alicyclobacillus phage strain KKP 3916 was highly effective against the A. acidoterrestris strain KKP 3133 at both tested MOIs (Figure 3). Genes 2023, 14, x FOR PEER REVIEW 9 of 20 Figure 2. Plaque morphology (on double-layer agar plate) of Alicyclobacillus phage strain KKP 3916 infecting A. acidoterrestris strain KKP 3148 (A), A. acidoterrestris strain KKP 3151 (B), A. acidocaldarius strain KKP 3157 (C), and A. acidoterrestris strain KKP 3195 (D). Of the 41 strains of the A. acidoterrestris species, the newly isolated bacteriophage had the ability to infect nine. No lysis zone was observed against the strains of A. hesperidum, A. herbarius, or A. acidophilus. Out of four strains from A. fastidiosus species and ﬁve strains from A. acidocaldarius species, the Alicyclobacillus phage strain KKP 3916 had the ability to infect one (KKP 3002) and two (KKP 3135 and KKP 3157) strains, respectively. Of the twelve bacterial strains infected by Alicyclobacillus phage strain KKP 3916, nine were guai-acol-producing. The host range plays an important role in the use of bacteriophage in therapy or food biocontrol. The elimination of only speciﬁc bacterial strains requires a narrower host range [68,69], to prevent the other microorganisms from being compromised (which is im-portant in some industries, e.g., dairy). To eliminate pathogenic bacteria within a species, it is desirable to search for bacteriophages with wide host ranges [70]. This is analogous to the use of broad-spectrum antibiotics without the need to identify the pathogen or sen-sitivity to the antibiotic. In the absence of broad-spectrum phage isolation, a preparation containing several bacteriophages can be used [68,70,71]. 3.3. Evaluation of the Activity of Phage against Bacterial Hosts The eﬀectiveness of a phage in inactivating bacteria is one of the critical characteris-tics that should be considered for potential candidates in phage therapy or food biocon-trol. To evaluate this feature in relation to the tested bacteriophage, growth curves were plotted for each bacterial host strain using the Bioscreen C Pro automated growth ana-lyzer. The growth curves showed that the Alicyclobacillus phage strain KKP 3916 was Genes 2023, 14, 1303 10 of 21 Figure 3. Growth curves of A. acidoterrestris strain KKP 3133 treated with Alicyclobacillus phage strain KKP 3916 at infection coefﬁcients of MOI = 1.0 (red line) and MOI = 0.1 (green line) compared to the control culture (blue line). Points represent the mean (n = 10); error bars represent the standard deviation (± SD) of the optical density. Moreover, in these samples, there were no differences in inhibition effect during the ﬁrst 6 h after infection and after 20 h of incubation. The addition of phages to the culture of the A. acidoterrestris strain KKP 3133 caused a shorter logarithmic phase of host growth compared to the control culture. A slight increase in optical density may indicate the acquisition of resistance to the phage used, a transition to the lysogenic cycle, or an increase in the number of cells [72,73]. In the course of their evolution, bacteria have developed a number of defense mechanisms to protect cells from phage infection. The primary defense mechanism of bacteria is the physical masking of the receptor through changes in its conformation or its complete disappearance from the cell envelope. The physical masking of receptors involves the production of an additional envelope that limits the interaction of the bacteriophage with the bacterial cell [74,75]. Genetic mutations result in the development of mutants that are completely insensitive to phage adsorption [76]. Often, once a host cell is infected with a phage, resistant bacteria can destroy the foreign genome without reducing their own viability or damaging their own genetic material. In this case, the phage DNA is degraded in the cytoplasm as a result of endonucleolytic digestion. A key role in this process is played by the restriction/modiﬁcation system. This system consists of restrictionases that recognize and cut speciﬁc DNA sequences and modify them by adding methyl groups [75,77]. Cells with this restriction system have their chromosomal DNA modiﬁed by a methyltransferase enzyme that adds speciﬁc sequences. This protects against the insertion of cuts into the bacterial DNA. The phage DNA does not have a methylation pattern, so it is not degraded by restriction enzymes once it enters the cytoplasm [78]. A key role in the defense mechanisms of prokaryotic organisms is played by the CRISPR/Cas system. This is a sequence of clustered, regularly spaced, short palindromic sequences, which detects and destroys phage DNA [79]. This provides a type of immune-defense system in bacteria and archaeons. The system works by “remembering” a speciﬁc nucleic acid sequence inserted into the cytosol. During phage infection, the initial step in the response the recognition of phage DNA and its cutting by Cas proteins [80]. The changes in the optical density of the tested bacterial strains after the incubation with the phage are shown in Table 2. The lower values of the speciﬁc growth-rate coefﬁcient (µ) determined for the phage-infected cultures indicate the signiﬁcant inhibition of cell divisions in the log phase. Genes 2023, 14, x FOR PEER REVIEW 10 of 20 highly eﬀective against the A. acidoterrestris strain KKP 3133 at both tested MOIs (Figure 3). Figure 3. Growth curves of A. acidoterrestris strain KKP 3133 treated with Alicyclobacillus phage strain KKP 3916 at infection coeﬃcients of MOI = 1.0 (red line) and MOI = 0.1 (green line) compared to the control culture (blue line). Points represent the mean (n = 10); error bars represent the standard de-viation (± SD) of the optical density. Moreover, in these samples, there were no diﬀerences in inhibition eﬀect during the ﬁrst 6 h after infection and after 20 h of incubation. The addition of phages to the culture of the A. acidoterrestris strain KKP 3133 caused a shorter logarithmic phase of host growth compared to the control culture. A slight increase in optical density may indicate the ac-quisition of resistance to the phage used, a transition to the lysogenic cycle, or an increase in the number of cells [72,73]. In the course of their evolution, bacteria have developed a number of defense mechanisms to protect cells from phage infection. The primary defense mechanism of bacteria is the physical masking of the receptor through changes in its con-formation or its complete disappearance from the cell envelope. The physical masking of receptors involves the production of an additional envelope that limits the interaction of the bacteriophage with the bacterial cell [74,75]. Genetic mutations result in the develop-ment of mutants that are completely insensitive to phage adsorption [76]. Often, once a host cell is infected with a phage, resistant bacteria can destroy the foreign genome with-out reducing their own viability or damaging their own genetic material. In this case, the phage DNA is degraded in the cytoplasm as a result of endonucleolytic digestion. A key role in this process is played by the restriction/modiﬁcation system. This system consists of restrictionases that recognize and cut speciﬁc DNA sequences and modify them by adding methyl groups [75,77]. Cells with this restriction system have their chromosomal DNA modiﬁed by a methyltransferase enzyme that adds speciﬁc sequences. This protects against the insertion of cuts into the bacterial DNA. The phage DNA does not have a methylation pattern, so it is not degraded by restriction enzymes once it enters the cyto-plasm [78]. A key role in the defense mechanisms of prokaryotic organisms is played by the CRISPR/Cas system. This is a sequence of clustered, regularly spaced, short palin-dromic sequences, which detects and destroys phage DNA [79]. This provides a type of immune-defense system in bacteria and archaeons. The system works by “remembering” a speciﬁc nucleic acid sequence inserted into the cytosol. During phage infection, the ini-tial step in the response the recognition of phage DNA and its cutting by Cas proteins [80]. The changes in the optical density of the tested bacterial strains after the incubation with the phage are shown in Table 2. The lower values of the speciﬁc growth-rate Genes 2023, 14, 1303 11 of 21 Table 2. Changes in the optical densities of bacterial cultures after the addition of speciﬁc phages and values of the speciﬁc growth rate coefﬁcient (µ). Control Culture MOI = 1.0 MOI = 0.1 ∆OD µ [h—1] 0.066 0.064 0.038 0.031 0.007 0.032 Similar results were obtained in the study conducted by Kozantsev et al. [73], where the growth kinetics of Bacillus cereus strain VKM B-370 after infection with the lysogenic phage B13 at different MOIs was determined. It was shown that the higher the MOI, the faster the partial growth inhibition was achieved. It is noteworthy that the difﬁculty in eliminating spore-forming bacteria lies not only in the resistance of bacterial spores to factors such as UV, heating, low pH, and pressure, but also in phage infection. This has also been conﬁrmed in studies on the elimination of B. subtilis and B. cereus using the phages PBSC1 and PBSC2 [81]. The use of phages caused a signiﬁcant inhibition of bacterial culture growth, but due to the development of phage-resistant bacterial spores, the optical density increased at the end of the culture. The use of phage biocontrol combined with other antimicrobial agents to eliminate bacteria may be an effective solution [81–83]. 3.4. Effect of Environmental Conditions on the Phage Activity The determination of the stability of phages helps to predict how bacteriophages may behave in the environment. In this study, the activity of isolated phages exposed to a wide range of temperatures (from −20 to 80 ◦C), levels of active acidity (pH from 2 to 12), and times of UV exposure (0, 5, 10, 15, 20, 25, 30, and 60 min) was evaluated. The activity was expressed as phage titer (in log PFU mL−1). The effect of temperature on the activity of the phages is presented in Figure 4. Figure 4. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after exposure to a wide range of temperatures. Letters a, b, c, d, and e indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. The Alicyclobacillus phage strain KKP 3916 retained its activity at both low and high temperatures. Temperatures from 4 ◦C to 30 ◦C slightly affected the activity of the bac- teriophages, while freezing temperatures (−20 ◦C) reduced their activity by about 99% in relation to the control culture (20 ◦C). Temperature increases to 40 ◦C, 50 ◦C, and 60 ◦C caused a denaturation of the phage virions due to stress conditions and, consequently, reduced their activity by about 99% compared to the control. The incubation at 70 ◦C resulted in the almost complete inactivation of the bacteriophages (99.9%), whereas the Genes 2023, 14, x FOR PEER REVIEW 11 of 20 coeﬃcient (µ) determined for the phage-infected cultures indicate the signiﬁcant inhibi-tion of cell divisions in the log phase. Table 2. Changes in the optical densities of bacterial cultures after the addition of speciﬁc phages and values of the speciﬁc growth rate coeﬃcient (µ). Control Culture MOI = 1.0 MOI = 0.1 ΔOD 0.066 0.038 0.007 µ [h—1] 0.064 0.031 0.032 Similar results were obtained in the study conducted by Kozantsev et al. [73], where the growth kinetics of Bacillus cereus strain VKM B-370 after infection with the lysogenic phage B13 at diﬀerent MOIs was determined. It was shown that the higher the MOI, the faster the partial growth inhibition was achieved. It is noteworthy that the diﬃculty in eliminating spore-forming bacteria lies not only in the resistance of bacterial spores to factors such as UV, heating, low pH, and pressure, but also in phage infection. This has also been conﬁrmed in studies on the elimination of B. subtilis and B. cereus using the phages PBSC1 and PBSC2 [81]. The use of phages caused a signiﬁcant inhibition of bacte-rial culture growth, but due to the development of phage-resistant bacterial spores, the optical density increased at the end of the culture. The use of phage biocontrol combined with other antimicrobial agents to eliminate bacteria may be an eﬀective solution [81–83]. 3.4. Eﬀect of Environmental Conditions on the Phage Activity The determination of the stability of phages helps to predict how bacteriophages may behave in the environment. In this study, the activity of isolated phages exposed to a wide range of temperatures (from −20 to 80 °C), levels of active acidity (pH from 2 to 12), and times of UV exposure (0, 5, 10, 15, 20, 25, 30, and 60 min) was evaluated. The activity was expressed as phage titer (in log PFU mL—1). The eﬀect of temperature on the activity of the phages is presented in Figure 4. Figure 4. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after exposure to a wide range of temperatures. Letters a, b, c, d, and e indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. The Alicyclobacillus phage strain KKP 3916 retained its activity at both low and high temperatures. Temperatures from 4 °C to 30 °C slightly aﬀected the activity of the bacte-riophages, while freezing temperatures (−20 °C) reduced their activity by about 99% in relation to the control culture (20 °C). Temperature increases to 40 °C, 50 °C, and 60 °C Genes 2023, 14, 1303 12 of 21 application of 80 ◦C resulted in PFU reduction to undetectable levels, probably due to the denaturation of the bacteriophage protein. In most cases, bacteriophages specific to thermophilic bacteria are resistant to high temperatures [62]. In a study by Morozov et al. [84], the thermophilic bacterium Aeribacillus sp. was the bacterial host of the tested AP45 phage. The optimal temperature for the growth of this bacterium is 55 ◦C, while the optimal temperature for maintaining the speciﬁc phage viability was below 75 ◦C. The AP45 bacteriophage showed stability after incubation for 24 h at 85 ◦C and 1 h at 95 ◦C. Another example is the effect of temperature on phage activity against Campylobacter jejuni [85]. The incubation at 42 ◦C had the best effect on the phage stability, while temperatures of 55 ◦C and 4 ◦C reduced its activity by about 99.9%. The effect of a wide range of levels of active acidity on the activity of the phages is presented in Figure 5. The isolated bacteriophage had a high tolerance to active acidity. No statistically signiﬁcant differences were observed after the incubation of the lysate in solutions with pH ranging from 3–11. An alkaline environment (pH = 12) reduced the activity of the phage by 99.9%. The solution at pH = 2 caused the complete denaturation of the phage virions. Figure 5. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after exposition to a wide range of pH values. Letters a, b, and c indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. In the study by Capra et al. [86], it was shown that the effect of pH on the activity of Lacticaseibacillus casei and Lacticaseibacillus paracasei (former names Lactobacillus casei and Lactobacillus paracasei) phages was dependent on the type of phage. Most of the bacteriophages showed high activity after incubation for 30 min in the pH range between 4 and 11. Similar to the results obtained in our study, the incubation at pH 2 resulted in the complete elimination of bacteriophages due to the denaturation of the virion head [86]. In other studies, B. subtilis bacteriophages showed activity in the pH range of 6.0–8.0 after incubation for 60 min. Incubation at pH extremes of 2–3 and 10–11 often resulted in a complete loss of infectivity [87]. Figure 6 shows the effect of UV-exposure time on the phage activity. Irradiation for 5 min resulted in an insigniﬁcant decrease in phage stability; however, a 10-minute exposure to UV resulted in a reduction in its activity by almost 99% compared to the control sample. A phage reduction of almost 99.9% was observed after exposure to UV for 15 min and 20 min. In contrast, exposures for 25 min and 30 min reduced the number of phages by 99.99% compared to the control. No plaques were observed after 60 min of UV exposure. Genes 2023, 14, x FOR PEER REVIEW 12 of 20 caused a denaturation of the phage virions due to stress conditions and, consequently, reduced their activity by about 99% compared to the control. The incubation at 70 °C re-sulted in the almost complete inactivation of the bacteriophages (99.9%), whereas the ap-plication of 80 °C resulted in PFU reduction to undetectable levels, probably due to the denaturation of the bacteriophage protein. In most cases, bacteriophages speciﬁc to thermophilic bacteria are resistant to high temperatures [62]. In a study by Morozov et al. [84], the thermophilic bacterium Aeribacil-lus sp. was the bacterial host of the tested AP45 phage. The optimal temperature for the growth of this bacterium is 55 °C, while the optimal temperature for maintaining the spe-ciﬁc phage viability was below 75 °C. The AP45 bacteriophage showed stability after in-cubation for 24 h at 85 °C and 1 h at 95 °C. Another example is the eﬀect of temperature on phage activity against Campylobacter jejuni [85]. The incubation at 42 °C had the best eﬀect on the phage stability, while temperatures of 55 °C and 4 °C reduced its activity by about 99.9%. The eﬀect of a wide range of levels of active acidity on the activity of the phages is presented in Figure 5. The isolated bacteriophage had a high tolerance to active acidity. No statistically signiﬁcant diﬀerences were observed after the incubation of the lysate in solutions with pH ranging from 3–11. An alkaline environment (pH = 12) reduced the activity of the phage by 99.9%. The solution at pH = 2 caused the complete denaturation of the phage virions. Figure 5. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after exposition to a wide range of pH values. Letters a, b, and c indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. In the study by Capra et al. [86], it was shown that the eﬀect of pH on the activity of Lacticaseibacillus casei and Lacticaseibacillus paracasei (former names Lactobacillus casei and Lactobacillus paracasei) phages was dependent on the type of phage. Most of the bacterio-phages showed high activity after incubation for 30 min in the pH range between 4 and 11. Similar to the results obtained in our study, the incubation at pH 2 resulted in the complete elimination of bacteriophages due to the denaturation of the virion head [86]. In other studies, B. subtilis bacteriophages showed activity in the pH range of 6.0–8.0 after incubation for 60 min. Incubation at pH extremes of 2–3 and 10–11 often resulted in a complete loss of infectivity [87]. Figure 6 shows the eﬀect of UV-exposure time on the phage activity. Irradiation for 5 min resulted in an insigniﬁcant decrease in phage stability; however, a 10-minute expo-sure to UV resulted in a reduction in its activity by almost 99% compared to the control Genes 2023, 14, 1303 13 of 21 Figure 6. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after different UV-exposure times. Letters a, b, c, d, e, and f indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. Research data concerning the stability of bacteriophages subjected to UV expo- sure are scarce. According to Hazem [88], thermophilic Bacillus-speciﬁc bacteriophages 46 and 50 were stable after 13 min and 20 min of UV exposure, respectively, while the lytic bacteriophage against Salmonella SS3e lost 50% viability after 1 min, 90% of phage particles were inactivated after 5 min, and no phage particles were detected after 15 min of UV exposure [89]. In addition to environmental factors, the presence of metal salts has a signiﬁcant impact on the adsorption process during phage–bacterial-host interactions. Numerous studies indicate that most divalent metal cations can increase the efﬁciency of bacterial lysis and, thus, the invasiveness of bacteriophages [90]. Divalent metal cations are believed to enhance the structures of bacterial viruses, such as phage T4 [90]. The formation of a complex between the phage structure and metal ion inhibits the inactivation of the phage [91]. Metal sorption on the surfaces of viruses can not only contribute to nanoparticle-metal transport, but also increase phage infectivity. In a study by Carey-Smith et al. [92], T5 bacteriophages were examined, and the obtained results indicated that viruses are inactivated much more rapidly by heat when suspended in 0.1 M sodium salt solutions than in broth. The inactivation rate of the T5 phage in 0.1 N NaCl at 37 ◦C can be signiﬁcantly reduced by adding 10−8 M of the following divalent cations: Ca, Mg, Ba, Sr, Mn, Co, Ni, Zn, Cd, and Cu. Increases in the concentration of cations in the suspending medium increase the resistance of the T5 phage to the inactivating effects of temperature, and the stability of virions in the presence of various cations results from the formation of a phage complex with metal ions. The results obtained by Chhibber et al. [93] indicate that the MR-10 MRSA phage required the presence of calcium ions for primary attachment to bacterial cells (Staphylococcus aureus strain 43300). It is likely that calcium is used to introduce the phage’s genetic material into the cytoplasm of host cells. 3.5. Analysis of Phage Genome The complete genome of the Alicyclobacillus phage strain KKP 3916 was sequ- enced and deposited in the GenBank database under accession number OQ846916. In addition, the phage was deposited in the Culture Collection of Industrial Microorgan- isms—Microbiological Resources Center of the Department of Microbiology, at the Prof. Wacław D ˛abrowski Institute of Agricultural and Food Biotechnology—State Research Institute. Genes 2023, 14, x FOR PEER REVIEW 13 of 20 sample. A phage reduction of almost 99.9% was observed after exposure to UV for 15 min and 20 min. In contrast, exposures for 25 min and 30 min reduced the number of phages by 99.99% compared to the control. No plaques were observed after 60 min of UV expo-sure. Figure 6. The activity of Alicyclobacillus phage strain KKP 3916 against A. acidoterrestris strain KKP 3133 after diﬀerent UV-exposure times. Letters a, b, c, d, e, and f indicate homogenous groups at a signiﬁcance level of p ≤ 0.05, n = 3. Research data concerning the stability of bacteriophages subjected to UV exposure are scarce. According to Hazem [88], thermophilic Bacillus-speciﬁc bacteriophages 46 and 50 were stable after 13 min and 20 min of UV exposure, respectively, while the lytic bac-teriophage against Salmonella SS3e lost 50% viability after 1 min, 90% of phage particles were inactivated after 5 min, and no phage particles were detected after 15 min of UV exposure [89]. In addition to environmental factors, the presence of metal salts has a signiﬁcant im-pact on the adsorption process during phage–bacterial-host interactions. Numerous stud-ies indicate that most divalent metal cations can increase the eﬃciency of bacterial lysis and, thus, the invasiveness of bacteriophages [90]. Divalent metal cations are believed to enhance the structures of bacterial viruses, such as phage T4 [90]. The formation of a com-plex between the phage structure and metal ion inhibits the inactivation of the phage [91]. Metal sorption on the surfaces of viruses can not only contribute to nanoparticle-metal transport, but also increase phage infectivity. In a study by Carey-Smith et al. [92], T5 bacteriophages were examined, and the obtained results indicated that viruses are inacti-vated much more rapidly by heat when suspended in 0.1 M sodium salt solutions than in broth. The inactivation rate of the T5 phage in 0.1 N NaCl at 37 °C can be signiﬁcantly reduced by adding 10−8 M of the following divalent cations: Ca, Mg, Ba, Sr, Mn, Co, Ni, Zn, Cd, and Cu. Increases in the concentration of cations in the suspending medium in-crease the resistance of the T5 phage to the inactivating eﬀects of temperature, and the stability of virions in the presence of various cations results from the formation of a phage complex with metal ions. The results obtained by Chhibber et al. [93] indicate that the MR-10 MRSA phage required the presence of calcium ions for primary attachment to bacterial cells (Staphylococcus aureus strain 43300). It is likely that calcium is used to introduce the phage’s genetic material into the cytoplasm of host cells. 3.5. Analysis of Phage Genome Genes 2023, 14, 1303 14 of 21 As with the TEM, the genome analysis confirmed that the phage belongs to the Caudoviricetes class. Figure 7 shows the proteomic tree generated using the ViPTree. It indicates that the isolated bacteriophage is a distant relative of phages from the Herelleviridae family. Figure 7. The viral proteomic tree of the Alicyclobacillus phage strain KKP 3916 is shown in a circular view. The branch represented by the phage under study is marked with an asterisk. The colored rings indicate the virus family (inner rings) and host groups (at the phylum level; outer rings). The tree was calculated by BIONJ based on the genomic distance matrix and rooted at the midpoint. Branch lengths are logoscaled. Sequence and taxonomic data were based on the Virus-Host DB [58]. The trees shown were generated using the ViPTree server [59]. The sequence of the Alicyclobacillus phage strain KKP 3916 genome in the form of linear dsDNA is 120,131 bp long with 40.3% G+C pair content. Out of the 204 predicted open reading frames (ORFs), 70 ORFs are associated with genes encoding proteins with known functions and 134 ORFs encode hypothetical proteins with unknown functions (Figure 8). The annotated functional proteins were divided into several groups, depending on their function: activities related to metabolism and replication, genome packaging, structure, lysis, and lysogeny. The main group of proteins whose function was pre- dicted comprised those related to metabolism and replication. Among genes related to replication and metabolism, the following proteins were found in the phage genome: DNA helicase (DnaB-like replicative helicase), DNA primase, RNA ligase, DNA poly- merase, DNA methyltransferase, transposase, exonuclease, and other proteins (Figure 7 and Table S1). The group of other proteins included those associated with phage structure, including: portal protein, capsid, tail, and lipoprotein. The proteins associated with lysis include, among others, holin, endolysin, and spanin. These enzymes are capable of dis- rupting bacterial peptidoglycan from within, leading to bacterial lysis and the release of Genes 2023, 14, x FOR PEER REVIEW 14 of 20 The complete genome of the Alicyclobacillus phage strain KKP 3916 was sequenced and deposited in the GenBank database under accession number OQ846916. In addition, the phage was deposited in the Culture Collection of Industrial Microorganisms—Micro-biological Resources Center of the Department of Microbiology, at the Prof. Wacław Dąbrowski Institute of Agricultural and Food Biotechnology—State Research Institute. As with the TEM, the genome analysis conﬁrmed that the phage belongs to the Cau-doviricetes class. Figure 7 shows the proteomic tree generated using the ViPTree. It indi-cates that the isolated bacteriophage is a distant relative of phages from the Herelleviridae family. Figure 7. The viral proteomic tree of the Alicyclobacillus phage strain KKP 3916 is shown in a circular view. The branch represented by the phage under study is marked with an asterisk. The colored rings indicate the virus family (inner rings) and host groups (at the phylum level; outer rings). The tree was calculated by BIONJ based on the genomic distance matrix and rooted at the midpoint. Branch lengths are logoscaled. Sequence and taxonomic data were based on the Virus-Host DB [58]. The trees shown were generated using the ViPTree server [59]. The sequence of the Alicyclobacillus phage strain KKP 3916 genome in the form of linear dsDNA is 120,131 bp long with 40.3% G+C pair content. Out of the 204 predicted open reading frames (ORFs), 70 ORFs are associated with genes encoding proteins with known functions and 134 ORFs encode hypothetical proteins with unknown functions (Figure 8). Genes 2023, 14, 1303 15 of 21 new phages [94,95]. Small hydrophobic proteins, called holins, accumulate in the inner cell membrane, oligomerize, and form pores in the membrane, causing the activation of endolysin, which accumulates in the cytoplasm or periplasm [94–96]. Endolysins, known as peptidoglycan hydrolases, are a class of enzymes encoded by genes of bacteriophages that degrade the bacterial cell wall at the end of the lytic cycle [94,95]. In Gram-negative hosts, there is often a third protein whose action is required for complete cell lysis, namely spanin, which breaks down the last barrier—the outer membrane [97,98]. Spanins are lysing proteins, essential for disrupting the outer membrane of the bacterial host at the ﬁnal stage of bacterial lysis [96,99]. Most phages produce a two-component spanin complex, consisting of an outer-membrane lipoprotein (o-spanin) and an inner-membrane protein (i-spanin) with a periplasmic domain [94,99,100]. In the Alicyclobacillus phage strain KKP 3916, one region encoding Rz-like spanin, two regions encoding holin, and three regions encoding endolysin were predicted. The analysis of the Alicyclobacillus phage strain KKP 3916 genome showed ﬁve tRNAs. The presence of tRNA, especially in virulent phage genomes, is a common phenomenon. Canchaya et al. [101] suggested that the presence of tRNA in the phage genome can be correlated with the better integration of virulent phages with the bacterial host chromosome. Bailly-Bechet et al. [102] reported that there is a positive association between the size of the phage genome and the number of tRNA genes it contains. No genes related to antibiotic resistance were found in the genome of the newly isolated phage, but four integration-related regions were identiﬁed. Integrase integrates the phage’s nucleic acid into the bacterial host genome, which does not directly lead to bacterial destruction, but indicates a moderate (lysogenic) bacteriophage life cycle [103]. The integrated or extra-chromosomal form of the phage is called a prophage, while a bacterial cell with an integrated prophage genome is called a lysogen. External stressors (especially those that pose a threat to the bacterial host cell) can initiate the transition from the lysogenic to the lytic cycle [104,105]. In addition, other regions associated with lysogeny were noted in the Alicyclobacillus phage strain KKP 3916 genome: the toxin gene and excisionase. Bacteriophages encoding virulence factors can convert their bacterial hosts through a process known as the lysogenic conversion of the phage from a non-pathogenic strain to a virulent strain or a strain with enhanced virulence. An example of the provision of bacteria with a toxin gene is a group of E. coli strains that produce Shiga toxins [106]. It has been suggested that prophages increase the ecological adaptation of many bacteria, regardless of their characteristics. Prophages can, among other activities, modulate bacterial resistance to various environmental factors (such as low pH), increase antibiotic tolerance, or facilitate bioﬁlm formation [102,107,108]. It is also known that lysogeny requires the expression of bacteriophage integrase, while the excision of prophage DNA from the host genome requires an additional phage-encoded protein, called excisionase. Although this process is often accompanied by the lysis of host cells, the curing of prophages from host bacteria has been reported in some cases. It is worth noting that integrase and excisionase proteins are considered markers of lysogenic infection [109]. Based on the morphology and the comparison of its protein regions, Alicyclobacillus phage strain KKP 3916 was assigned to viruses with complex structures (tailed phages from the Caudoviricetes class). However, analyses of the phylogenetic relationship prevented its unambiguous assignment to a speciﬁc family and genus. The weak similarity with other phage genomes deposited in the databases suggests that the isolated bacteriophage may be representative of a new genus of tailed bacteriophages. Genes 2023, 14, 1303 16 of 21 Figure 8. Map of the genomic organization of Alicyclobacillus phage strain KKP 3916 generated using the Proksee program [57]. 4. Conclusions The application of bacteriophages in food biocontrol is gaining increasing interest among scientists around the world, including the European Union. Therefore, in this study, we isolated a phage active against guaiacol-producing A. acidoterrestris and investigated its biological properties. To the best of our knowledge, this is the ﬁrst study on the isolation and characterization of a phage speciﬁc to bacteria from the Alicyclobacillus genus. Arguably, a new genus of tailed bacteriophages was discovered. It was shown that the newly isolated phage exhibits a narrow host spectrum and broad tolerance to environmental factors, such as temperature, pH, and UV radiation. The bioinformatics analyses demonstrated that the genome of the Alicyclobacillus phage strain KKP 3916 contains sequences corresponding to integration into the host genome, suggesting that this virus is not strictly lytic. Due to the risk of its potential involvement in horizontal gene transfer, it is not an appropriate candidate for food biocontrol. Further research will address the characterization of newly isolated bacteriophages against spore-forming bacteria and their effectiveness in eliminating food-spoilage-causing strains either individually or combined with high pressure. The effectiveness of bacteriophages will be examined in plant-based food matrices. Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/genes14061303/s1. Table S1: Genome annotation of functional ORFs of the Alicyclobacillus phage strain KKP 3916. Author Contributions: Conceptualization, D.S.; methodology, D.S. and M.W.; formal analysis, D.S., M.W. and P.´S.; investigation, D.S., M.W., O.´S., P.´S. and B.S.; writing—original draft preparation, D.S.; writing—review and editing, D.S., M.W. and O.´S.; visualization, D.S. and M.W.; supervision, M.W. and B.S. All authors have read and agreed to the published version of the manuscript. Funding: The research was supported by internal statutory project no. ZM-810-05, “Isolation and characterization of Alicyclobacillus-speciﬁc bacteriophages.” Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Genes 2023, 14, x FOR PEER REVIEW 15 of 20 Figure 8. Map of the genomic organization of Alicyclobacillus phage strain KKP 3916 generated using the Proksee program [57]. The annotated functional proteins were divided into several groups, depending on their function: activities related to metabolism and replication, genome packaging, struc-ture, lysis, and lysogeny. The main group of proteins whose function was predicted com-prised those related to metabolism and replication. Among genes related to replication and metabolism, the following proteins were found in the phage genome: DNA helicase (DnaB-like replicative helicase), DNA primase, RNA ligase, DNA polymerase, DNA me-thyltransferase, transposase, exonuclease, and other proteins (Figure 7 and Table S1). The group of other proteins included those associated with phage structure, including: portal protein, capsid, tail, and lipoprotein. The proteins associated with lysis include, among others, holin, endolysin, and spanin. These enzymes are capable of disrupting bacterial peptidoglycan from within, leading to bacterial lysis and the release of new phages [94,95]. Small hydrophobic proteins, called holins, accumulate in the inner cell membrane, oligomerize, and form pores in the membrane, causing the activation of endolysin, which accumulates in the cytoplasm or periplasm [94–96]. Endolysins, known as peptidoglycan hydrolases, are a class of enzymes encoded by genes of bacteriophages that degrade the bacterial cell wall at the end of the lytic cycle [94,95]. In Gram-negative hosts, there is often a third protein whose action is required for complete cell lysis, namely spanin, which breaks down the last barrier—the outer membrane [97,98]. Spanins are lysing proteins, essential for disrupting the outer membrane of the bacterial host at the ﬁnal stage of bac-terial lysis [96,99]. Most phages produce a two-component spanin complex, consisting of an outer-membrane lipoprotein (o-spanin) and an inner-membrane protein (i-spanin) with a periplasmic domain [94,99,100]. In the Alicyclobacillus phage strain KKP 3916, one region encoding Rz-like spanin, two regions encoding holin, and three regions encoding endolysin were predicted. The analysis of the Alicyclobacillus phage strain KKP 3916 ge-nome showed ﬁve tRNAs. The presence of tRNA, especially in virulent phage genomes, is a common phenomenon. Canchaya et al. [101] suggested that the presence of tRNA in the phage genome can be correlated with the better integration of virulent phages with the bacterial host chromosome. Bailly-Bechet et al. [102] reported that there is a positive association between the size of the phage genome and the number of tRNA genes it con-tains. No genes related to antibiotic resistance were found in the genome of the newly isolated phage, but four integration-related regions were identiﬁed. Integrase integrates the phage’s nucleic acid into the bacterial host genome, which does not directly lead to Genes 2023, 14, 1303 17 of 21 Acknowledgments: We thank Grzegorz Nowicki and Jakub Grabowski from genXone SA for helping in the sequence analysis of whole phage genomes, and Julita Nowakowska from Laboratory of Electron and Confocal Microscopy, Faculty of Biology, University of Warsaw for helping in the preparation and visualization of phage preparations in the transmission-electron microscope. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. 2. Bhardwaj, R.L.; Nandal, U.; Pal, A.; Jain, S. Bioactive Compounds and Medicinal Properties of Fruit Juices. Fruits 2014, 69, 5. [CrossRef] Fernandes, F.A.N.; Rodrigues, S. Cold Plasma Processing on Fruits and Fruit Juices: A Review on the Effects of Plasma on Nutritional Quality. Processes 2021, 9, 2098. [CrossRef] 3. Nowak, D.; Go´sli ´nski, M.; Kł˛ebukowska, L. Antioxidant and Antimicrobial Properties of Selected Fruit Juices. Plant Foods Hum. 4. 5. 6. Nutr. 2022, 77, 427–435. [CrossRef] [PubMed] Zhu, Y.; Zhang, M.; Mujumdar, A.S.; Liu, Y. Application Advantages of New Non-Thermal Technology in Juice Browning Control: A Comprehensive Review. Food Rev. Int. 2022, 1–22. [CrossRef] Santarelli, G.A.; Migliorati, G.; Pomilio, F.; Marfoglia, C.; Centorame, P.; D’Agostino, A.; D’Aurelio, R.; Scarpone, R.; Battistelli, N.; Di Simone, F.; et al. Assessment of Pesticide Residues and Microbial Contamination in Raw Leafy Green Vegetables Marketed in Italy. Food Control 2018, 85, 350–358. [CrossRef] Pérez-Cacho, P.R.; Galan-Soldevilla, H.; Mahattanatawee, K.; Elston, A.; Rouseff, R.L. Sensory lexicon for fresh squeezed and processed orange juices. Food Sci. Technol. Int. 2008, 14, 131–141. [CrossRef] 7. Mphahlele, R.R.; Fawole, O.A.; Mokwena, L.M.; Opara, U.L. Effect of Extraction Method on Chemical, Volatile Composition and Antioxidant Properties of Pomegranate Juice. S. Afr. J. Bot. 2016, 103, 135–144. [CrossRef] 8. World Health Organization. Measuring the Intake of Fruit and Vegetables. Available online: http://www.who.int/ 9. dietphysicalactivity/fruit/en/ (accessed on 1 March 2023). Aneja, K.R.; Dhiman, R.; Aggarwal, N.K.; Kumar, V.; Kaur, M. Microbes Associated with Freshly Prepared Juices of Citrus and Carrots. Int. J. Food Sci. 2014, 2014, 408085. [CrossRef] 10. Kincal, D.; Hill, W.S.; Balaban, M.O.; Portier, K.M.; Wei, C.I.; Marshall, M.R. A Continuous High Pressure Carbon Dioxide System for Microbial Reduction in Orange Juice. J. Food Sci. 2005, 70, M249–M254. [CrossRef] 11. Yang, Y.; Yang, Q.; Ma, X.; Zhang, Y.; Zhang, X.; Zhang, W. A Novel Developed Method Based on Single Primer Isothermal Ampliﬁcation for Rapid Detection of Alicyclobacillus acidoterrestris in Apple Juice. Food Control 2017, 75, 187–195. [CrossRef] Silva, L.P.; Gonzales-Barron, U.; Cadavez, V.; Sant’Ana, A.S. Modeling the Effects of Temperature and PH on the Resistance of Alicyclobacillus acidoterrestris in Conventional Heat-Treated Fruit Beverages through a Meta-Analysis Approach. Food Microbiol. 2015, 46, 541–552. [CrossRef] [PubMed] 12. 13. Lee, S.; Han, A.; Jo, S.; Cheon, H.; Song, H.; Jang, A.R.; Kim, D.; Lee, S.Y. Microbiological Quality and Safety of Commercial Fresh Fruit and Vegetable Juices in Korea. LWT 2021, 152, 112432. [CrossRef] 14. Dewanti-Hariyadi, R.; Maria, A.; Sandoval, P. Microbiological Quality and Safety of Fruit Juices. Food Rev. Int. 2013, 1, 54–57. 15. Yeni, F.; Yava¸s, S.; Alpas, H.; Soyer, Y. Most Common Foodborne Pathogens and Mycotoxins on Fresh Produce: A Review of Recent Outbreaks. Crit. Rev. Food Sci. Nutr. 2016, 56, 1532–1544. [CrossRef] 16. Balali, G.I.; Yar, D.D.; Afua Dela, V.G.; Adjei-Kusi, P. Microbial Contamination, an Increasing Threat to the Consumption of Fresh Fruits and Vegetables in Today’s World. Int. J. Microbiol. 2020, 2020, 3029295. [CrossRef] 17. Bhilwadikar, T.; Pounraj, S.; Manivannan, S.; Rastogi, N.K.; Negi, P.S. Decontamination of Microorganisms and Pesticides from Fresh Fruits and Vegetables: A Comprehensive Review from Common Household Processes to Modern Techniques. Compr. Rev. Food Sci. Food Saf. 2019, 18, 1003–1038. [CrossRef] 18. Lykov, I.N.; Kakharova, M.A.; Kureber, V.S.; Yurova, A.E. Research of Antibiotic Resistance of Microorganisms Isolated from Fruits and Vegetables. IOP Conf. Ser. Earth Environ. Sci. 2021, 839, 042003. [CrossRef] 19. Misra, N.N. The Contribution of Non-Thermal and Advanced Oxidation Technologies towards Dissipation of Pesticide Residues. 20. Trends Food Sci. Technol. 2015, 45, 229–244. [CrossRef] Silva, F.V.M.; Tan, E.K.; Farid, M. Bacterial Spore Inactivation at 45–65 ◦C Using High Pressure Processing: Study of Alicyclobacillus acidoterrestris in Orange Juice. Food Microbiol. 2012, 32, 206–211. [CrossRef] 21. Nasiłowska, J.; Sokołowska, B.; Fonberg-Broczek, M. Long term storage of vegetable juices treated by high hydrostatic pressure– assurance of the microbial safety. BioMed Res. Int. 2018, 12, 7389381. [CrossRef] [PubMed] 22. Zhang, H.; Mittal, G.S. Effects of High-Pressure Processing (HPP) on Bacterial Spores: An Overview. Food Rev. Int. 2008, 23. 24, 330–351. [CrossRef] Sokołowska, B.; Sk ˛apska, S.; Fonberg-Broczek, M.; Niezgoda, J.; Chotkiewicz, M.; Dekowska, A.; Rzoska, S.J. Factors inﬂuencing the inactivation of Alicyclobacillus acidoterrestris spores exposed to high hydrostatic pressure in apple juice. High Press. Res. 2013, 33, 73–82. [CrossRef] Genes 2023, 14, 1303 18 of 21 24. Sokołowska, B.; Połaska, M.; Dekowska, A. Alicyclobacillus–still current issues in the beverage industry. In Safety Issues in Beverage Production; Academic Press: Cambridge, MA, USA, 2020; Volume 18, pp. 105–146. 25. Molva, C.; Baysal, A.H. Evaluation of Bioactivity of Pomegranate Fruit Extract against Alicyclobacillus acidoterrestris DSM 3922 26. 27. Vegetative Cells and Spores in Apple Juice. LWT 2015, 62, 989–995. [CrossRef] Sen Chang, S.; Kang, D.H. Alicyclobacillus spp. in the Fruit Juice Industry: History, Characteristics, and Current Isolation/Detection Procedures. Crit. Rev. Microbiol. 2004, 30, 55–74. [CrossRef] [PubMed] Smit, Y.; Cameron, M.; Venter, P.; Witthuhn, R.C. Alicyclobacillus Spoilage and Isolation-A Review. Food Microbiol. 2011, 28, 331–349. [CrossRef] [PubMed] 28. Uchida, R.; Silva, F.V.M. Alicyclobacillus acidoterrestris Spore Inactivation by High Pressure Combined with Mild Heat: Modeling the Effects of Temperature and Soluble Solids. Food Control 2017, 73, 426–432. [CrossRef] 29. Durak, M.Z.; Churey, J.J.; Danyluk, M.D.; Worobo, R.W. Identiﬁcation and Haplotype Distribution of Alicyclobacillus Spp. from Different Juices and Beverages. Int. J. Food Microbiol. 2010, 142, 286–291. [CrossRef] 30. Gutiérrez, D.; Rodríguez-Rubio, L.; Martínez, B.; Rodríguez, A.; García, P. Bacteriophages as Weapons against Bacterial Bioﬁlms in the Food Industry. Front. Microbiol. 2016, 7, 825. [CrossRef] 31. Li, J.; Zhao, F.; Zhan, W.; Li, Z.; Zou, L.; Zhao, Q. Challenges for the Application of Bacteriophages as Effective Antibacterial Agents in the Food Industry. J. Sci. Food Agric. 2022, 102, 461–471. [CrossRef] 32. Połaska, M.; Sokołowska, B. Review Bacteriophages—A New Hope or a Huge Problem in the Food Industry. AIMS Microbiol. 2019, 5, 324. [CrossRef] 33. Wang, Z.; Zhao, X. The Application and Research Progress of Bacteriophages in Food Safety. J. Appl. Microbiol. 2022, 133, 2137–2147. [CrossRef] 34. Osei, E.K.; Mahony, J.; Kenny, J.G. From Farm to Fork: Streptococcus suis as a Model for the Development of Novel Phage-Based Biocontrol Agents. Viruses 2022, 14, 1996. [CrossRef] [PubMed] 35. Moye, Z.D.; Woolston, J.; Sulakvelidze, A. Bacteriophage application for food production and processing. Viruses 2018, 10, 205. 36. [CrossRef] [PubMed] Fernández, L.; Gutiérrez, D.; Rodríguez, A.; García, P. Application of Bacteriophages in the Agro-Food Sector: A Long Way toward Approval. Front. Cell. Infect Microbiol. 2018, 8, 296. [CrossRef] [PubMed] 37. Ye, M.; Sun, M.; Huang, D.; Zhang, Z.; Zhang, H.; Zhang, S.; Hu, F.; Jiang, X.; Jiao, W. A Review of Bacteriophage Therapy for Pathogenic Bacteria Inactivation in the Soil Environment. Environ. Int. 2019, 129, 488–496. [CrossRef] 38. Harada, L.K.; Silva, E.C.; Campos, W.F.; Del Fiol, F.S.; Vila, M.; D ˛abrowska, K.; Krylov, V.N.; Balcão, V.M. Biotechnological Applications of Bacteriophages: State of the Art. Microbiol. Res. 2018, 212, 38–58. [CrossRef] [PubMed] 39. Butala, M.; Dragoš, A. Unique Relationships between Phages and Endospore-Forming Hosts. Trends Microbiol. 2022, 31, 498–510. [CrossRef] 40. Mok, J.H.; Sun, Y.; Pyatkovskyy, T.; Hu, X.; Sastry, S.K. Mechanisms of Bacillus subtilis Spore Inactivation by Single- and Multi-Pulse High Hydrostatic Pressure (MP-HHP). Innov. Food Sci. Emerg. Technol. 2022, 81, 103147. [CrossRef] 41. Por˛ebska, I.; Sokołowska, B.; Sk ˛apska, S.; Rzoska, S.J. Treatment with High Hydrostatic Pressure and Supercritical Carbon Dioxide 42. to Control Alicyclobacillus acidoterrestris Spores in Apple Juice. Food Control 2017, 73, 24–30. [CrossRef] Sarker, M.R.; Akhtar, S.; Torres, J.A.; Paredes-Sabja, D. High Hydrostatic Pressure-Induced Inactivation of Bacterial Spores. Crit. Rev. Microbiol. 2015, 41, 18–26. [CrossRef] 43. Mirzaei, M.K.; Nilsson, A.S. Isolation of Phages for Phage Therapy: A Comparison of Spot Tests and Efﬁciency of Plating Analyses for Determination of Host Range and Efﬁcacy. PLoS ONE 2015, 10, e0118557. [CrossRef] [PubMed] 44. Dekowska, A.; Niezgoda, J.; Sokołowska, B. Genetic heterogeneity of Alicyclobacillus strains revealed by RFLP analysis of vdc region and rpoB gene. Bio Med. Res. Intern. 2018, 2018, 9608756. 45. Połaska, M.; Dekowska, A.; Sokołowska, B. Isolation and identiﬁcation of guaiacol-producing Alicyclobacillus fastidiosus strains from orchards in Poland. Acta Bioch. Pol. 2021, 68, 301–307. 46. Alves, D.; Cerqueira, M.A.; Pastrana, L.M.; Sillankorva, S. Entrapment of a Phage Cocktail and Cinnamaldehyde on Sodium Alginate Emulsion-Based Films to Fight Food Contamination by Escherichia coli and Salmonella Enteritidis. Food Res. Int. 2020, 128, 108791. [CrossRef] [PubMed] 47. Echeverría-Vega, A.; Morales-Vicencio, P.; Saez-Saavedra, C.; Gordillo-Fuenzalida, F.; Araya, R. A Rapid and Simple Protocol for 48. the Isolation of Bacteriophages from Coastal Organisms. Methods X 2019, 6, 2614–2619. [CrossRef] Jamal, M.; Andleeb, S.; Jalil, F.; Imran, M.; Nawaz, M.A.; Hussain, T.; Ali, M.; Das, C.R. Isolation and Characterization of a Bacteriophage and Its Utilization against Multi-Drug Resistant Pseudomonas aeruginosa-2995. Life Sci. 2017, 190, 21–28. [CrossRef] [PubMed] 49. Ackermann, H.W. 5500 Phages examined in the electron microscope. Arch. Virol. 2007, 151, 227–243. [CrossRef] 50. Mahmoud, M.; Askora, A.; Barakat, A.B.; Rabie, O.E.F.; Hassan, S.E. Isolation and Characterization of Polyvalent Bacteriophages Infecting Multi Drug Resistant Salmonella Serovars Isolated from Broilers in Egypt. Int. J. Food Microbiol. 2018, 266, 8–13. [CrossRef] Genes 2023, 14, 1303 19 of 21 51. Amarillas, L.; Lightbourn-Rojas, L.; Angulo-Gaxiola, A.K.; Basilio Heredia, J.; González-Robles, A.; León-Félix, J. The Antibacterial Effect of Chitosan-Based Edible Coating Incorporated with a Lytic Bacteriophage against Escherichia coli O157:H7 on the Surface of Tomatoes. J. Food Saf. 2018, 38, e12571. [CrossRef] Strathdee, S.A.; Hatfull, G.F.; Mutalik, V.K.; Schooley, R.T. Phage Therapy: From Biological Mechanisms to Future Directions. Cell 2023, 186, 17–31. [CrossRef] 52. 53. Wójcicki, M.; ´Swider, O.; ´Srednicka, P.; Shymialevich, D.; Ilczuk, T.; Koperski, Ł.; Cie´slak, H.; Sokołowska, B.; Juszczuk-Kubiak, E. Newly Isolated Virulent Salmophages for Biocontrol of Multidrug-Resistant Salmonella in Ready-to-Eat Plant-Based Food. Int. J. Mol. Sci. 2023, 24, 10134. [CrossRef] 54. Kolmogorov, M.; Yuan, J.; Lin, Y.; Pevzner, P.A. Assembly of Long, Error-Prone Reads Using Repeat Graphs. Nat. Biotechnol. 2019, 37, 540–546. [CrossRef] [PubMed] 55. Mcnair, K.; Zhou, C.; Dinsdale, E.A.; Souza, B.; Edwards, R.A. PHANOTATE: A Novel Approach to Gene Identiﬁcation in Phage Genomes. Bioinformatics 2019, 35, 4537–4542. [CrossRef] [PubMed] 56. Wu, J.; Liu, Q.; Li, M.; Xu, J.; Wang, C.; Zhang, J.; Xiao, M.; Bin, Y.; Xia, J. PhaGAA: An Integrated Web Server Platform for Phage Genome Annotation and Analysis. Bioinformatics 2023, 39, btad120. [CrossRef] 57. Proksee Software. Available online: https://proksee.ca/ (accessed on 2 March 2023). 58. Mihara, T.; Nishimura, Y.; Shimizu, Y.; Nishiyama, H.; Yoshikawa, G.; Uehara, H.; Hingamp, P.; Goto, S.; Ogata, H. Linking Virus Genomes with Host Taxonomy. Viruses 2016, 8, 66. [CrossRef] [PubMed] 59. Nishimura, Y.; Yoshida, T.; Kuronishi, M.; Uehara, H.; Ogata, H.; Goto, S. ViPTree: The Viral Proteomic Tree Server. Bioinformatics 2017, 33, 2379–2380. [CrossRef] 60. Tynecki, P.; Guzi ´nski, A.; Kazimierczak, J.; Jadczuk, M.; Dastych, J.; Onisko, A. PhageAI-Bacteriophage Life Cycle Recognition with Machine Learning and Natural Language Processing. bioRxiv 2020, 198606. 61. Abedon, S.T.; Yin, J. Bacteriophage Plaques: Theory and Analysis. Methods Mol. Biol. 2009, 501, 161–174. 62. Zheng, X.-F.; Yang, Z.; Zhangl, H.; Jinl, W.-X.; Xul, C.-W.; Gaol, L.; Raol, S.-Q.; Jiao, X. Isolation of virulent phages infecting dominant mesophilic aerobic bacteria in cucumber pickle fermentation. Food Microbiol. 2020, 86, 103330. [CrossRef] 63. Cornelissen, A.; Ceyssens, P.J.; T’Syen, J.; van Praet, H.; Noben, J.P.; Shaburova, O.V.; Krylov, V.N.; Volckaert, G.; Lavigne, R. The T7-Related Pseudomonas Putida Phage Φ15 Displays Virion-Associated Bioﬁlm Degradation Properties. PLoS ONE 2011, 6, e18597. [CrossRef] 64. Krone, S.M.; Abedon, S.T. Modeling phage plaque growth. Bacteriophage Ecol. 2008, 15, 415–438. 65. Glonti, T.; Pirnay, J.P. In Vitro Techniques and Measurements of Phage Characteristics That Are Important for Phage Therapy 66. Success. Viruses 2022, 14, 1490. [CrossRef] [PubMed] Jaschke, P.R.; Dotson, G.A.; Hung, K.S.; Liu, D.; Endy, D. Deﬁnitive Demonstration by Synthesis of Genome Annotation Completeness. Proc. Natl. Acad. Sci. USA 2019, 116, 24206–24213. [CrossRef] 67. Wright, B.W.; Ruan, J.; Molloy, M.P.; Jaschke, P.R. Genome Modularization Reveals Overlapped Gene Topology Is Necessary for Efﬁcient Viral Reproduction. ACS Synth. Biol. 2020, 9, 3079–3090. [CrossRef] 68. Hyman, P. Phages for Phage Therapy: Isolation, Characterization, and Host Range Breadth. Pharmaceuticals 2019, 12, 3079–3090. [CrossRef] 69. Leibovici, L.; Shraga, I.; Drucker, M.; Konigsberger, H.; Samra, Z.; Pitlik, S.D. The Beneﬁt of Appropriate Empirical Antibiotic Treatment in Patients with Bloodstream Infection. J. Intern. Med. 1998, 244, 379–386. [CrossRef] 70. Ross, A.; Ward, S.; Hyman, P. More Is Better: Selecting for Broad Host Range Bacteriophages. Front. Microbiol. 2016, 7, 1352. [CrossRef] [PubMed] 71. Gientka, I.; Wójcicki, M.; ˙Zuwalski, A.W.; Bła ˙zejak, S. Use of phage cocktail for improving the overall microbiological quality of sprouts—Two methods of application. Appl. Microbiol. 2021, 1, 21. [CrossRef] 72. Lu, Z.; Breidt, F. Escherichia coli O157:H7 Bacteriophage F241 Isolated from an Industrial Cucumber Fermentation at High Acidity and Salinity. Front. Microbiol. 2015, 6, 67. [CrossRef] 73. Kazantseva, O.A.; Piligrimova, E.G.; Shadrin, A.M. Novel Bacillus-Infecting Bacteriophage B13—The Founding Member of the Proposed New Genus Bunatrivirus. Viruses 2022, 14, 2300. [CrossRef] 74. Christensen, L.L.; Josephsen, J. The methyltransferase from the LlaDII restriction-modiﬁcation system inﬂuences the level of expression of its own gene. J. Bacteriol. 2004, 186, 287–295. [CrossRef] [PubMed] 75. Egido, J.E.; Costa, A.R.; Aparicio-Maldonado, C.; Haas, P.J.; Brouns, S.J. Mechanisms and clinical importance of bacteriophage resistance. FEMS Microbiol. Rev. 2022, 46, fuab048. [CrossRef] [PubMed] 76. Mills, S.; McAuliffe, O.E.; Coffey, A.; Fitzgerald, G.F.; Ross, R.P. Plasmids of lactococci–genetic accessories or genetic necessities? FEMS Microbiol. Rev. 2006, 30, 243–273. [CrossRef] 77. Dupuis, M.È.; Villion, M.; Magadán, A.H.; Moineau, S. CRISPR-Cas and restriction–modiﬁcation systems are compatible and increase phage resistance. Nat. Commun. 2013, 4, 2087. [CrossRef] [PubMed] 78. Kim, J.W.; Dutta, V.; Elhanaﬁ, D.; Lee, S.; Osborne, J.A.; Kathariou, S. A novel restriction-modiﬁcation system is responsible for temperature-dependent phage resistance in Listeria monocytogenes ECII. Appl. Environ. Microbiol. 2012, 78, 1995–2004. [CrossRef] 79. Watson, B.N.; Steens, J.A.; Staals, R.H.; Westra, E.R.; van Houte, S. Coevolution between bacterial CRISPR-Cas systems and their bacteriophages. Cell Host Microbe 2021, 29, 715–725. [CrossRef] Genes 2023, 14, 1303 20 of 21 80. Pilosof, S.; Alcala-Corona, S.A.; Wang, T.; Kim, T.; Maslov, S.; Whitaker, R.; Pascual, M. The network structure and eco-evolutionary dynamics of CRISPR-induced immune diversiﬁcation. Nat. Commun. 2020, 4, 1650–1660. [CrossRef] 81. Gabiatti, N.; Yu, P.; Mathieu, J.; Lu, G.W.; Wang, X.; Zhang, H.; Soares, H.M.; Alvarez, P.J.J. Bacterial Endospores as Phage Genome Carriers and Protective Shells. Appl. Environ. Microbiol. 2018, 84, e01186-18. [CrossRef] 82. Pires, D.P.; Melo, L.D.R.; Vilas Boas, D.; Sillankorva, S.; Azeredo, J. Phage Therapy as an Alternative or Complementary Strategy to Prevent and Control Bioﬁlm-Related Infections. Curr. Opin. Microbiol. 2017, 39, 48–56. [CrossRef] 83. Oechslin, F.; Piccardi, P.; Mancini, S.; Gabard, J.; Moreillon, P.; Entenza, J.M.; Resch, G.; Que, Y.A. Synergistic Interaction between Phage Therapy and Antibiotics Clears Pseudomonas aeruginosa Infection in Endocarditis and Reduces Virulence. J. Infect. Dis. 2017, 215, 703–712. [CrossRef] 84. Morozova, V.; Bokovaya, O.; Kozlova, Y.; Kurilshikov, A.; Babkin, I.; Tupikin, A.; Bondar, A.; Ryabchikova, E.; Brayanskaya, A.; Peltek, S.; et al. A Novel Thermophilic Aeribacillus Bacteriophage AP45 Isolated from the Valley of Geysers, Kamchatka: Genome Analysis Suggests the Existence of a New Genus within the Siphoviridae Family. Extremophiles 2019, 23, 599–612. [CrossRef] 85. Thung, T.Y.; Lee, E.; Mahyudin, N.A.; Wan Mohamed Radzi, C.W.J.; Mazlan, N.; Tan, C.W.; Radu, S. Partial Characterization and in Vitro Evaluation of a Lytic Bacteriophage for Biocontrol of Campylobacter jejuni in Mutton and Chicken Meat. J. Food Saf. 2020, 40, e12770. [CrossRef] 86. Capra, M.L.; Quiberoni, A.; Reinheimer, J. Phages of Lactobacillus casei/paracasei: Response to Environmental Factors and Interaction with Collection and Commercial Strains. J. Appl. Microbiol. 2006, 100, 334–342. [CrossRef] 87. Krasowska, A.; Biegalska, A.; Augustyniak, D.; Ło´s, M.; Richert, M.; Łukaszewicz, M. Isolation and Characterization of Phages Infecting Bacillus subtilis. Biomed. Res. Int. 2015, 2015, 179597. [CrossRef] [PubMed] 88. Hazem, A. Effects of Temperatures, Ph-Values, Ultra-Violet Light, Ethanol and Chloroform on the Growth of Isolated Thermophilic Bacillus Phages. New Microbiol. 2002, 25, 469–476. 89. Kim, S.; Kim, S.H.; Rahman, M.; Kim, J. Characterization of a Salmonella Enteritidis Bacteriophage Showing Broad Lytic Activity against Gram-Negative Enteric Bacteria. J. Microbiol. 2018, 56, 917–925. [CrossRef] [PubMed] 90. Li, X.; Chen, Y.; Wang, S.; Duan, X.; Zhang, F.; Guo, A.; Qian, P. Exploring the beneﬁts of metal ions in phage cocktail for the treatment of methicillin-resistant Staphylococcus aureus (MRSA) infection. Infect. Drug Resist. 2022, 15, 2689–2702. [CrossRef] [PubMed] 91. Adams, M.H. The stability of bacterial viruses in solutions of salts. J. Gen. Physiol. 1949, 32, 579–594. [CrossRef] [PubMed] 92. Carey-Smith, G.V.; Billington, C.; Cornelius, A.J.; Hudson, J.A.; Heinemann, J.A. Isolation and characterization of bacteriophages infecting Salmonella spp. FEMS Microbiol. Lett. 2006, 258, 182–186. [CrossRef] 93. Chhibber, S.; Kaur, T.; Kaur, S. Essential role of calcium in the infection process of broad-spectrum methicillin-resistant Staphylo- coccus aureus bacteriophage. J. Microbiol. 2014, 54, 775–780. 94. Grabowski, Ł.; Łepek, K.; Stasiłoj´c, M.; Kosznik-Kwa´snicka, K.; Zdrojewska, K.; Maci ˛ag-Dorszy ´nska, M.; W˛egrzyn, G.; W˛egrzyn, A. Bacteriophage-Encoded Enzymes Destroying Bacterial Cell Membranes and Walls, and Their Potential Use as Antimicrobial Agents. Microbiol. Res. 2021, 248, 126746. [CrossRef] 95. Bujak, K.; Decewicz, P.; Kami ´nski, J.; Radli ´nska, M. Identiﬁcation, characterization, and genomic analysis of novel Serratia 96. temperate phages from a gold mine. Int. J. Mol. Sci. 2020, 21, 6709. [CrossRef] Jamal, M.; Bukhari, S.M.A.U.S.; Andleeb, S.; Ali, M.; Raza, S.; Nawaz, M.A.; Hussain, T.; Rahman, S.U.; Shah, S.S.A. Bacteriophages: An overview of the control strategies against multiple bacterial infections in different ﬁelds. J. Basic Microbiol. 2019, 59, 123–133. [CrossRef] [PubMed] 97. Chamakura, K.; Young, R. Phage Single-Gene Lysis: Finding the Weak Spot in the Bacterial Cell Wall. J. Biolog. Chem. 2019, 294, 3350–3358. [CrossRef] 98. Wang, I.N. Lysis Timing and Bacteriophage Fitness. Genetics 2006, 172, 17–26. [CrossRef] 99. Berry, J.D.; Rajaure, M.; Young, R.Y. Spanin Function Requires Subunit Homodimerization through Intermolecular Disulﬁde Bonds. Mol. Microbiol. 2013, 88, 35–47. [CrossRef] 100. Holt, A.; Cahill, J.; Ramsey, J.; Martin, C.; O’Leary, C.; Moreland, R.; Maddox, L.T.; Galbadage, T.; Sharan, R.; Sule, P.; et al. Phage-Encoded Cationic Antimicrobial Peptide Required for Lysis. J. Bacter. 2022, 204, e00214–21. [CrossRef] [PubMed] 101. Canchaya, C.; Fournous, G.; Brüssow, H. The impact of prophages on bacterial chromosomes. Mol. Microbiol. 2004, 53, 9–18. [CrossRef] [PubMed] 102. Bailly-Bechet, M.; Vergassola, M.; Rocha, E. Causes for the intriguing presence of tRNAs in phages. Genome Res. 2007, 17, 1486–1495. [CrossRef] [PubMed] 103. Cie´slik, M.; Bagi ´nska, N.; Jo ´nczyk-Matysiak, E.; W˛egrzyn, A.; W˛egrzyn, G.; Górski, A. Temperate Bacteriophages—The Powerful Indirect Modulators of Eukaryotic Cells and Immune Functions. Viruses 2021, 13, 1013. [CrossRef] 104. Howard-Varona, C.; Hargreaves, K.R.; Abedon, S.T.; Sullivan, M.B. Lysogeny in nature: Mechanisms, impact and ecology of temperate phages. ISME J. 2017, 11, 1511–1520. [CrossRef] 105. Chen, Y.; Yang, L.; Yang, D.; Song, J.; Wang, C.; Sun, E.; Gu, C.; Chen, H.; Tong, Y.; Tao, P.; et al. Speciﬁc Integration of Temperate Phage Decreases the Pathogenicity of Host Bacteria. Front. Cell. Infect. Microbiol. 2020, 10, 14. [CrossRef] [PubMed] 106. Krger, A.; Lucchesi, P.M.A. Shiga toxins and stx phages: Highly diverse entities. Microbiology 2015, 161, 451–462. [CrossRef] [PubMed] Genes 2023, 14, 1303 21 of 21 107. Davis, B.M.; Waldor, M.K. Mobile Genetic Elements and Bacterial Pathogenesis; Craig, N.L., Craigie, R., Gellert, M., Lambowitz, A.M., Eds.; Mobile DNA II; ASM Press: Washington, DC, USA, 2007; pp. 1040–1059. 108. Boyd, E.F.; Brüssow, H. Common themes among bacteriophage-encoded virulence factors and diversity among the bacteriophages involved. Trends Microbiol. 2002, 10, 521–529. [CrossRef] 109. Knowles, B.; Silveira, C.B.; Bailey, B.A.; Barott, K.; Cantu, V.A.; Cobián-Güemes, A.G.; Coutinho, F.H.; Dinsdale, E.A.; Felts, B.; Furby, K.A.; et al. Erratum: Corrigendum: Lytic to temperate switching of viral communities. Nat. Cell Biol. 2016, 539, 123. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.11604_pamj.2020.37.308.25829	Article Research Déterminants de la survie des enfants âgés de 6 mois à 15 ans, infectés par le VIH et suivis dans la ville d’Ebolowa au Cameroun de 2008 à 2018 Ginette Claude Mireille Kalla, Valery-Gustave Mve Mve, Nelly Kamgaing Noubi, Marcelle Nina Ehouzou Mandeng, Marie Claire Okomo Assoumou, Francois Xavier Mbopi-Keou, Francisca Monebenimp Corresponding author: Ginette Claude Mireille Kalla, Faculté de Médecine et des Sciences Biomédicales, Université de Yaoundé I, Yaoundé, Cameroun. [email protected] Received: 30 Aug 2020 - Accepted: 19 Oct 2020 - Published: 03 Dec 2020 Keywords: VIH, enfant, survie, Ebolowa, Cameroun Copyright: Ginette Claude Mireille Kalla et al. Pan African Medical Journal (ISSN: 1937-8688). This is an Open Access International 4.0 License article distributed under (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. the Creative Commons Attribution terms of the Cite this article: Ginette Claude Mireille Kalla et al. Déterminants de la survie des enfants âgés de 6 mois à 15 ans, infectés par le VIH et suivis dans la ville d’Ebolowa au Cameroun de 2008 à 2018. Pan African Medical Journal. 2020;37(308). 10.11604/pamj.2020.37.308.25829 Available online at: https://www.panafrican-med-journal.com//content/article/37/308/full Déterminants de la survie des enfants âgés de 6 mois à 15 ans, infectés par le VIH et suivis dans la ville d’Ebolowa au Cameroun de 2008 à 2018 Determinants of survival of HIV-infected children aged 6 months to 15 years on follow-up in the town of Ebolowa, Cameroon from 2008 to 2018 Ginette Claude Mireille Kalla1,2,&, Valery-Gustave Mve Mve1, Nelly Kamgaing Noubi1,2, Marcelle Nina Ehouzou Mandeng1,2, Marie Claire Okomo Assoumou1, Xavier Mbopi-Keou1, Francois Francisca Monebenimp1,2 et de Médecine 1Faculté Sciences Biomédicales, Université de Yaoundé I, Yaoundé, Cameroun, Centre Hospitalier et Universitaire de Yaoundé, Yaoundé, Cameroun Pédiatrie, 2Service des de &Auteur correspondant Ginette Claude Mireille Kalla, Faculté de Médecine et des Sciences Biomédicales, Université de Yaoundé I, Yaoundé, Cameroun Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 1 Article Résumé Introduction: la survie des enfants infectés par le VIH demeure un défi dans les pays en voie de développement. Au Cameroun, la mortalité liée au VIH chez les enfants de moins de 15 ans en 2018 était de 20%. Paradoxalement, la région du Sud Cameroun, malgré une séroprévalence élevée chez les enfants de 4,1% et une couverture en traitement antirétroviral faible de l´ordre de 64%, ne fait pas partie des régions du Cameroun les plus touchées par la mortalité pédiatrique liée au travail était de VIH/SIDA. L´objectif de ce déterminer le taux de survie et identifier ses déterminants chez les enfants âgés de 6 mois à 15 ans, infectés par le VIH. Méthodes: une étude de cohorte à collecte de données rétrospective et prospective a été menée de janvier 2008 à décembre 2018 dans trois formations sanitaires prenant en charge les enfants VIH positifs, à Ebolowa dans la région du Sud Cameroun. L´étude s´est faite en deux temps, une phase de collecte rétrospective pour la sélection des dossiers médicaux des enfants VIH positifs répondant aux critères d´inclusion dans registres de consultation, et une phase de collecte prospective qui nous a permis d´avoir auprès des parents, les informations sur le devenir des enfants. Un consentement éclairé parental a été obtenu au cours de cette deuxième phase. Les données cliniques, paracliniques, sociodémographiques, thérapeutiques, ainsi que le devenir des enfants ont été collectées. Les temps moyens de survie, ainsi que les facteurs associés à la survie ont été déterminés à l´aide du modèle de Kaplan Meier. La régression à risque proportionnel de Cox, nous a permis d´identifier les déterminants de la survie. Notre critère de jugement était le décès. Le niveau de significativité a été fixé à 5%. Résultats: au total, 186 enfants ont été enrôlés. La durée médiane de suivi était de 18,5 mois. Le taux de survie était de 66,7%. La majorité des décès (67%) est survenue avant le sixième mois de suivi. Après inférieur à 2 ans l´âge analyse multivariée, [aHR: 18,6 (6,48-53,59); p=0,001], l´anémie sévère [aHR: 7,69 (1,02-57,9); p=0,04], et la présence les étaient indépendamment à d´infections opportunistes [aHR: 4,52 (2,51-8,14); et p=0,05] significativement survie. associés Conclusion: en plus du traitement antirétroviral précoce, un bon suivi clinique et paraclinique est nécessaire pour améliorer la survie des enfants infectés par le VIH. la English abstract region, despite Cameroon Introduction: survival of HIV-infected children is a challenge in developing countries. In Cameroon, HIV-related mortality among children under the age of 15 in 2018 was 20%. Paradoxically, the Southern high seroprevalence among children (4.1%) and low antiretroviral therapy coverage (around 64%), is not among the regions of Cameroon most affected by HIV/AIDS-related pediatric mortality. The purpose of this study was to calculate survival rate and to identify its determinants in HIV-infected children aged 6 months-15 years. Methods: we conducted a retrospective, prospective cohort study data-collection in three health care facilities specialized in treating HIV-positive children in Ebolowa, South Cameroon from January 2008 to December 2018. The study was conducted in two phases, a retrospective collection phase for the selection of medical records of HIV-positive children that met inclusion criteria in consultation registries and a prospective collection phase in which we collected information from parents about the future of children. Informed parental consent was obtained during this second phase. Socio-demographic, paramedical, therapeutic data as well as data about the future of children were collected. Mean survival time and factors associated with survival were determined using the Kaplan Meier model. Cox proportional hazards regression allowed for the identification of survival determinants. Evaluation criterion was the death. Significance level was set at 5%. Results: a total of 186 patients were enrolled in the study: the average follow-up period was 18.5 months. Survival rate was 66.7%. The majority of deaths clinical, Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 2 Article (67%) occurred before the sixth month of follow- up. After multivariate analysis, an age less than 2 years [aHR: 18.6 (6.48-53.59); p=0.001), severe anemia [aHR: 7.69 (1.02-57.9); p=0.04) and the presence of opportunistic infections [aHR: 4.52 (2.51-8.14); p=0.05] were independently and significantly associated with survival. Conclusion: in addition to early antiretroviral therapy, good clinical and paraclinical monitoring is needed to improve the survival of HIV-infected children. Key words: HIV, Cameroon child, survival, Ebolowa, Introduction La survie des enfants infectés par le virus de l´immunodéficience humaine (VIH) demeure un problème de santé publique à l´échelle mondiale, notamment dans voie de Afrique développement subsaharienne [1]. En 2008, 91% de nouvelles infections à VIH chez les enfants ont été répertoriées dans cette région [2]. En 2016, il a été noté 91% de décès liés au VIH/SIDA chez les adolescents [3]. les pays en surtout en et En 2017, selon le Comité National de Lutte contre le Sida (CNLS), la prévalence générale du VIH/SIDA était estimée à 3,1% [4]. L´ONUSIDA estime d´ailleurs, que le Cameroun reste l´un des pays d´Afrique subsaharienne les plus touchés et que les enfants de moins 15 ans représenteraient 7,96% de personnes vivant avec le VIH. En 2018, 15,000 à 21,000 décès étaient liés au VIH parmi lesquels, 2400 à 4600 enfants de moins de 15 ans soit 20% de décès [1]. la prévention, particulièrement Des stratégies de riposte y ont été mises en place pour réduire cette mortalité. Ces mesures étaient axées sur la prévention de la transmission mère enfant (PTME) initiée en 1996 [5]. Le traitement antirétroviral (TAR) a été gratuit depuis 2007 [6]. Le diagnostic précoce par polymerase chain reaction (PCR) a été effectif depuis 2008 [5]. En outre, le TAR immédiat a débuté en 2011 [7]. Grâce à ces efforts, l´on a observé un recul de l´épidémie chez les enfants de 2010 à 2018 avec une diminution de nouvelles infections et une baisse de la mortalité respectivement de 44% et 45% [1]. depuis plusieurs Cette amélioration globale, est cependant inégalement répartie dans les différentes régions du pays. En effet, la région du Sud-Cameroun présente les séroprévalences au VIH les plus élevées. Estimée dans la population générale à 5,2% en 2017 [4], chez les enfants nés de mères séropositives au VIH, la séroprévalence était d´environ 4,1% avec une couverture en TAR d´environ 64% [8], bien loin de l´atteinte de l´objectif 90-90-90 prévus à l´horizon 2020 [9]. années la fondamentale qu´occupe De nombreuses études faites reconnaissent la place thérapie antirétrovirale pour la survie des enfants infectés le VIH. Une méta-analyse réalisée dans par plusieurs pays africains en 2011 montrait le bénéfice du traitement antirétroviral sur la survie des enfants infectés par le VIH comparativement à celle des enfants ne recevant pas de traitement [10]. Nyunt et al. en Asie avaient trouvé qu´entre 2006 et 2016, la survie globale des enfants sous TAR était de 86% [11]. Malgré le faible taux de couverture en ARV, la région du Sud Cameroun ne figure pas parmi les régions pour le plus de décès lesquelles on enregistre pédiatriques liés au VIH/SIDA au Cameroun [5,7]. Fort de ce constat, nous nous sommes posé la question de savoir ce qui, en dehors du TAR, expliquerait la survie des enfants infectés par le VIH dans la ville d´Ebolowa? D´où notre objectif qui était de déterminer le taux de survie et identifier ses déterminants chez ces enfants. Méthodes Une étude de cohorte à collecte rétrospective (dans les dossiers médicaux) et prospective (suite à une rencontre avec les parents) a été menée dans trois formations sanitaires à file active dans la ville d´Ebolowa, chef-lieu de la région du Sud Cameroun sur une période allant de Janvier 2008 à Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 3 Article Décembre 2018. La population d´étude était constituée d´enfants âgés de 6 mois à 15 ans révolus, infectés par le VIH et suivis dans ladite ville. Dans chaque formation sanitaire, les registres ont été consultés et les dossiers médicaux des enfants répondant à nos critères d´inclusion ont été retenus. Dans ces dossiers, les données sur les variables cliniques, sociodémographiques, paracliniques et thérapeutiques ont été recueillies. A partir du contact téléphonique des parents ou tuteurs retrouvé dans les dossiers, ces derniers ont leur consentement éclairé et avoir des informations sur le devenir des enfants. contactés d´obtenir afin été les enfants dont Procédures: après approbation du Comité d´Ethique et de la Recherche de la Faculté de Médecine et des Sciences Biomédicales de l´Université de Yaoundé I, une autorisation de recherche auprès du délégué régional de la santé publique du sud a été obtenue. Les registres des infection à VIH ont été enfants suivis pour consultés. Tous l´âge était compris entre 6 mois et 15 ans révolus dont le diagnostic avait été posé entre le 1er janvier 2008 et le 31 décembre 2018 ont été retenus. Les variables cliniques, sociodémographiques, les paracliniques, informations sur le devenir des enfants obtenues auprès des parents après consentement parental, ont été collectées sur des fiches techniques conçues à cet effet. Les registres de laboratoire, ainsi que les registres de ravitaillement en ARV ont également été consultés. Était considéré comme perdu de vue, tout enfant non retrouvé après recherche active dans la communauté. thérapeutiques, ainsi que les des variables données: Analyse sociodémographiques, cliniques, paracliniques, thérapeutiques, ainsi que le devenir des enfants VIH positifs et suivis à Ebolowa ont été étudiées. Les aspects descriptifs ont été présentés sous forme de tableaux ou figures avec, les effectifs, pourcentages, médianes, moyennes et écart- types. Le modèle de Kaplan Meier a permis de déterminer les temps moyens de survie des enfants infectés par le VIH. Les courbes de Kaplan de régression à le modèle de Meier nous ont permis d´identifier les facteurs associés à la survie. Les variables dignes d´intérêt avec p<0,1 à l´analyse bivariée ont été introduites risque dans proportionnel déterminants Cox. indépendants de la survie ont été identifiés en utilisant les rapports de risques ajustés à un intervalle de confiance de 95% et une valeur p<0,05. Le critère de jugement était le décès. L´analyse statistique des données a été effectuée en utilisant les logiciels Microsoft Excel et SPSS version 17. Les Considérations éthiques: un consentement éclairé des parents a été obtenu pour la phase de collecte de données prospective. L´anonymat a été conservé. Résultats sociodémographiques des Caractéristiques enfants parents et tuteurs: du 1er janvier 2008 au 31 décembre 2018, 186 enfants ont été retenus dont 124 (66,7%) vivants et sous TAR, 45 (24,2%) décédés et 17 (9,1%) perdus de vue. Quatre-vingt- douze (49,5%) étaient de sexe masculin et 94 (50,5%) de sexe féminin soit un sexe ratio de 0,98. La médiane d´âge était de 84,5 mois avec des extrêmes allant de 7 à 191 mois. La tranche d´âge la plus représentée était celle de 5-10 ans (38,2%). La majorité des enfants 117 (63%) vivaient en zone urbaine. Soixante et seize d´entre eux (40,9%) étaient orphelins d´au moins un parent. La majorité 132 (71%) était scolarisée (Table 1). La majorité des mères 130 (69,9%) étaient vivantes et avaient un âge compris entre 16 et 61 la plus tranche d´âge des mères ans. La représentée était celle de 17-35 ans (67%). La sérologie VIH des mères était positive pour 112 d´entre elles (86,1%) et 93 (83%) étaient sous TAR. Seulement 12 mères avaient un travail rémunéré (9,2%), 62 d´entre elles vivaient en couple (47,7%), 51 mères (39,2%) étaient célibataires et 17 étaient veuves (13,1%). La majorité des pères étaient vivants (n=84, 45%), avec un âge variant entre 22 et 71 ans. La tranche d´âge la plus représentée Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 4 Article celles de 35-55 ans (75%). La sérologie VIH était connue positive pour 23 (27,3%) d´entre eux et seulement 15 sous TAR. Vingt-et-un avaient un travail rémunéré (25%), 60 soit 61,4% vivaient en couple, 13 (15,5%) étaient célibataires et 11 (13,1%) étaient veufs. (71,4%) étaient Caractéristiques cliniques, paracliniques et thérapeutiques: le diagnostic de l´infection à VIH a été fait par sérologie pour la majorité 155 (83,3%) et seulement 31 (16,7%) ont été diagnostiqués précocement par PCR. Pour 143 (76,9%) d´entre eux, le diagnostic s´est fait au stade de Sida maladie. Plus de la moitié était déjà à un stade clinique avancé de la maladie au moment du diagnostic avec respectivement 81 (43,5%) au stade III et 19 (10,2%) au stade IV. La charge virale initiale avait été faite chez 35 (18,8%) enfants. Le dosage du taux de CD4 au début de la prise en charge n´a été disponible que pour 89 enfants sur les 186 soit 47,8% et 50 (56,2%) d´entre eux étaient en immunodépression sévère. Cent-deux enfants ont bénéficié d´une NFS initiale et la majorité était anémiée 88 (86,3%), dont 28 (27,5%) étaient en anémie sévère. Tous les enfants ont été mis sous ARV, 174 (93,5%) étaient sous prophylaxie à base de Le ravitaillement en ARV était de 100%, 50-99% et <50% respectivement pour 117 (62,9%), 52 (30%) et 17 (9,1%) enfants (Table 2). cotrimoxazole. Suivi: la durée du suivi était comprise entre 1-157 mois avec une médiane de 18,5 mois. Un changement de protocole est survenu chez 12 (6,5%) enfants. Les raisons de ce changement étaient l´échec clinique pour 5 (41,7%) enfants et la rupture de stock des ARV pour 4 (33,3%) enfants. Seuls 32 (17,2%) ont présenté au moins une infection opportuniste au cours du suivi. Les contrôles de charge virale et de dosage de CD4 au cours du suivi n'ont été faits que chez 8 (4,3%) et 18 (9,8%) enfants respectivement. Devenir: sur les 186 enfants retenus, 124 étaient en vie au moment de l´étude soit un taux brut de survie de 66,7% (Table 3). Des 45 (24,2%) enfants décédés, deux tiers l´ont été avant le 6e mois de suivi et la courbe globale de survie a montré une probabilité de survie meilleure après le 22e mois de suivi (Figure 1). Facteurs associés à la survie: l´âge était associé significativement à la survie (p<0,0001) avec des temps moyens de survie de 8,5 ± 1,8 mois, 32,4 ± 3,2 mois, 66,3 ± 6,9 mois et 105,8 ± 16,1 mois pour les enfants de <2 ans, 2-5 ans, 5-10 ans et >10 ans respectivement. La scolarisation était associée significativement à la survie (p=0,004) avec un temps moyen de survie de 79,6 ± 5,4 mois pour les enfants scolarisés et de 60,7 ± 14,7 mois pour ceux non scolarisés. La présence de comorbidités liées au VIH était associée significativement à la survie (p=0,005). Le temps moyen de survie des enfants qui ne présentaient pas de comorbidités était de 79,0 ± 5,3 mois et de 57,4 ± 12,9 mois pour ceux qui en présentaient. Une association significative a été retrouvée entre le stade clinique au moment du diagnostic et la survie (p<0,0001). Les temps moyens de survie étaient de 84,1 ± 9,7 mois, 51,0 ± 4,5 mois, 78,9 ± 14,5 mois et 26,2 ± 8,1 pour les stades cliniques I, II, III et IV respectivement. L´anémie était un facteur associé à la survie (p=0,003). Les temps moyens de survie étaient de 41,4 ± 9,5 mois pour les enfants avec anémie sévère, 95,6 ± 16,6 mois pour ceux avec anémie modérée et 75,5 ± 9,3 mois pour les enfants avec anémie légère. était en ARV ravitaillement Le associé significativement à la survie (p<0,0001) avec un temps moyen de survie de 84,0 ± 8,7 mois pour les enfants qui avaient un pourcentage de ravitaillement de 100%, 112,7 ± 12,2 mois pour des ravitaillements compris entre 50-99% et 30,0 ± 6,6 mois pour ceux dont le ravitaillement était < à 50%. Une association significative a été retrouvée entre les maladies opportunistes et la survie (p<0,0001) avec un temps moyen de survie de 119,7 ± 7,6 mois pour les enfants qui n´en avaient pas et 31,9 ± 7,3 mois pour ceux qui en avaient (Figure 2). Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 5 Article Analyse multi variée: l´âge <2 ans [HR: 15,62 (3,1- 79,7), p=0,001], l´anémie sévère [HR: 11,5 (1,1- 125,3) p=0,04], la présence de maladies opportunistes [HR: 2,33 (0,98-5,5), p=0,05] ont été retrouvés indépendants significativement associés à la survie (Table 4 and Figure 3). facteurs comme Discussion les registres répertorié Au total, sur la période allant de 2008 à 2018 nous avons de dans consultation, 186 enfants infectés par le VIH et suivis à Ebolowa. Le taux brut de survie au moment de l´étude était de 66,7%. Ce taux est inférieur au taux global de survie de 86% retrouvé par Nyunt et al. en 2018 au Myanmar. Cette différence pourrait s´expliquer par le fait que la majorité (83%) des enfants de leur étude résidait en zone urbaine, avaient plus de 200 CD4/mm³ avant la mise sous TAR et étaient sous prophylaxie au cotrimoxazole [11]. Gesesew et al. en Ethiopie relevaient également le fait que la proximité aux les moyens de méthodes diagnostiques et communication favorisent la mise précoce sous TAR et donc la survie [12]. Des 45 (24,2%) enfants qui étaient décédés au moment de l´étude. Deux tiers sont décédés avant le 6e mois de suivi. Ce résultat est proche de celui retrouvé par Ebissa et Njom respectivement en Ethiopie et au Cameroun (70%) [13,14] et pourrait s´expliquer par le fait que les enfants dans notre étude ont été diagnostiqués et mis sous traitement tardivement. Toutefois, on note que la survie était meilleure après le 22e mois de suivi. D´autres études ont également mentionné que, la mortalité souvent élevée au début du suivi baisse par la suite ceci s´expliquant par la survenue décalée des avantages du TAR [15]. L´âge était associé significativement à la survie (p<0,0001). En effet, les enfants de moins de 2 ans avaient moins de chances de survie que les autres. Ce résultat est différent de celui de Nyunt et Ebonyi qui n´ont trouvé aucune association entre l´âge et la survie [11,16]. Dans l´étude de Nyunt les enfants dans et al. l´âge médian était de six mois, ceci montre que son étude étaient majoritairement plus jeunes et précocement mis sous TAR [11]. Par contre, Judd A et al. dans une méta-analyse réalisée dans plusieurs pays d´Europe et d´Asie du Sud Est ont retrouvé une association entre l´âge et la survie. Dans leur étude, être dans la première année de vie était prédicteur de décès précoce [17]. La scolarisation la survie était associée significativement à (p=0,004). Cela pourrait s´expliquer par le fait que les enfants scolarisés étant plus âgés comprennent mieux l´importance du suivi et du traitement dans leur prise en charge. La présence de comorbidités liées au VIH était associée significativement à la survie (p=0,005) avec un temps moyen de survie meilleur pour les enfants qui ne présentaient pas de comorbidités. Ceci est en accord avec la littérature [18]. Une association significative a été retrouvée entre le stade clinique au moment du diagnostic et la survie (p<0,0001). La probabilité de survie diminuait au fur et à mesure que le stade clinique était avancé. Ce résultat est similaire à ceux trouvés par Ebissa et al. en Ethiopie [13] et Njom et Loussikila au Cameroun [14]. L´anémie était un facteur associé à la survie (p=0,003). En effet, la sévérité de l´anémie diminuait la probabilité de survie. Ce résultat est semblable à celui de Vermund et al. au Mozambique en 2014 et Ebissa et al. en Ethiopie en 2015 [13,19]. Notre résultat est cependant différent de celui de Njom et Loussikila au Cameroun [14]. Cette différence pourrait s´expliquer par le fait que la majorité de leurs patients étant issus de familles de classe sociale moyenne et élevée, la proportion d´enfants avec taux d´hémoglobine <8gr/dl était faible (9%) dans leur étude [14]. était en ARV ravitaillement littérature nous rapporte que associé Le significativement à la survie (p<0,0001). En effet, la la bonne adhérence au TAR diminue la mortalité et la [20]. Notre résultat morbidité corrobore avec celui d´Ebissa et al. en Ethiopie qui retrouvait que la mauvaise adhérence au TAR était liée au VIH Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 6 Article un facteur prédictif indépendant de décès [13]. Une association significative entre les maladies opportunistes et la survie a été retrouvée (p<0,0001). La probabilité de survie était meilleure pour les enfants qui n´avaient pas de maladies opportunistes. Ce résultat est comparable à celui trouvé par Traisathit et al. en Thaïlande dans une étude portant sur les évènements définissant le sida et décès sous antirétroviraux chez les enfants et les adolescents infectés par le VIH [21]. des étaient p=0,05] L´âge <2 ans [HR= 15,62 (3,1-79,7), p=0,001], l´anémie sévère [HR= 11,5 (1,1-125,3), p=0,04] et la présence des infections opportunistes [HR= 2,33 (0,98-5,5), facteurs indépendants significativement associés à la survie des enfants de 6 mois à 15 ans révolus infectés par le VIH à Ebolowa. Njom et Loussikila au Cameroun en 2017 a également retrouvé que l´âge ≤1 an à l´initiation du TAR était un facteur indépendant la diminution des significativement associé à chances de survie (HR= 2,1 [1,1-5,08], p=0,01), par retrouvé d´association contre, elle n´a pas [14]. significative Gebremedhin et al. en Ethiopie en 2013 a retrouvé comme diminuant significativement les chances de survie: l´âge <18 mois (HR= 4,39 [1,15-17,41], p=0,036), la diarrhée chronique (HR= 4,63 [1,50-14,31, p=0,008] et un taux d´hémoglobine <8g/dl (HR= 3,77 [29-10,98], p=0,015) [22]. Ebissa et al. en 2015 en Ethiopie a quant à lui retrouvé comme déterminant de la survie le taux d´hémoglobine <7g/dl (HR: 4,08 [1,33-12,56], p=0,014) [13]. concernant l´anémie associés facteurs Conclusion l´anémie sévère et Au terme de cette étude, le taux de survie était de 66,7%. L´âge <2 ans, la présence d´affections opportunistes au début du traitement antirétroviral se sont révélés être des déterminants de la survie des enfants âgés de 6 mois à 15 ans, infectés par le VIH et suivis dans la ville d´Ebolowa de 2008 à 2018. Cette étude a montré qu´en plus du traitement antirétroviral débuté précocement, un bon suivi clinique pour s´avère et paraclinique nécessaire l´amélioration de la survie des enfants infectés par le VIH. Etat des connaissances sur le sujet • Le traitement antirétroviral immédiat débuté en 2011 a permis de diminuer l´incidence et la mortalité liée au VIH/SIDA et a amélioré la survie des enfants infectés; • Très peu d´études se sont intéressées à la survie des enfants infectés par le VIH au Cameroun. Contribution de notre étude à la connaissance faibles chez • L´âge <2 ans, • Cette étude montre que les taux de survie traitement la disponibilité du malgré antirétroviral les restent enfants infectés par le VIH au Cameroun; l´anémie sévère et la présence d´affections opportunistes au début du traitement antirétroviral sont significativement et indépendamment associés à la survie des enfants infectés par le VIH au Cameroun; • En plus du traitement antirétroviral débuté précocement, un bon suivi clinique et paraclinique s´avère nécessaire pour l´amélioration de la survie des enfants infectés par le VIH. Conflits d'intérêts Les auteurs ne déclarent aucun conflit d´intérêts. Contributions des auteurs Ginette Claude Mireille Kalla: conception de l'étude, analyse et interprétation des résultats, rédaction du manuscrit; Gustave Mve Mve: analyse, interprétation des résultats, rédaction du manuscrit; Nelly Kamgaing Noubi: analyse, interprétation des résultats et rédaction du manuscrit; Marcelle Nina Ehouzou: analyse, interprétation des résultats et rédaction du manuscrit; Marie Claire Okomo Assoumou: analyse, interprétation des résultats et rédaction du manuscrit; Francisca Monebenimp: assure la Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 7 Article validité scientifique de l´étude, notamment, la conception de l´étude, l´analyse, l´interprétation des résultats et la rédaction du manuscrit; Francois Xavier Mbopi-Keou: assure la validité scientifique de l´étude, notamment, la conception de l´étude, l´analyse, la rédaction du manuscrit. Tous les auteurs ont lu et approuvé la version finale du manuscrit. l´interprétation des résultats et Remerciements Nos remerciements s´adressent à tous les patients et parents pour leur participation à l´étude, ainsi sanitaires qu´au personnel des retenues pour l´étude. formations Tableaux et figures à l’initiation du Table 1: caractéristiques sociodémographiques des enfants infectés par le VIH et suivis à Ebolowa Table 2: caractéristiques cliniques, paraclinique et thérapeutique traitement antirétroviral Table 3: devenir des enfants infectés par le VIH et suivis à Ebolowa Table 4: déterminants de la survie des enfants infectés par le VIH et suivis à Ebolowa Figure 1: courbe globale de survie des enfants infectés par le VIH et suivis à Ebolowa Figure 2: facteurs associés à la survie des enfants infectés par le VIH et suivis à Ebolowa Figure 3: courbes de survie ajustées des enfants selon les tranches d’âge, classes d’hémoglobine et maladies opportunistes Références 1. Joint United Nations Programme on HIV and AIDS. UNAIDS data 2019. 2019. 2. Programme commun des Nations Unies sur le VIH/Sida (ONUSIDA), Organisation Mondiale de la Santé (OMS). Le point sur l´épidémie de SIDA, Décembre 2009. Genève: ONUSIDA, OMS. 2009;9: 100. 3. United Nations Children's Fund (UNICEF). Les enfants et le sida: les chiffres de 2017. UNICEF France. 2017. 4. CNLS. Epidémiologique de l´infection à VIH au Cameroun: quatrième trimestre 2017. Bulletin Epidémiologique. 2018;3: 3. 5. Mossus-Etounou T, Essi M-JM, Isseini A, Souore-Sanda J, Pa´ana-Elemzo SB, Assala L-CB et al. Evolution des programmes nationaux de lutte contre l´infection à VIH et le Sida au Cameroun, de 2000 à 2015. HEALTH SCIENCES AND DISEASE. 29 févr 2016;17(1). Google Scholar 6. CNLS. Epidémiologique de l´infection à VIH au Cameroun: Premier trimestre 2018. Bulletin Epidémiologique. 2018;4. 7. Ministère de la Santé Publique, Organisation Mondiale de la Santé. Directives nationales de prévention et prise en charge du VIH au Cameroun. 2014;193. 8. CNLS. Rapport de progrès PTME N°12. 2017. 9. CNLS. Plan stratégique national de lutte contre le VIH, le Sida et les IST 2014-2017 Yaoundé, Cameroun. 2013. 10. Ndondoki C, Dabis F, Namale L, Becquet R, Ekouevi D, Bosse-Amani C et al. Survie et évolution clinique et biologique des enfants les infectés par antirétroviraux de littérature, 2004-2009. Presse Med. Jul-Aug 2011;40(7-8): e338-57. PubMed\| Google Scholar le VIH en Afrique: traités par revue 11. Nyunt KKK, Han WW, Satyanarayana S, Isaakidis P, Hone S, Khaing AA et al. Factors associated with death and loss to follow-up in children on antiretroviral care in Mingalardon Specialist Hospital, Myanmar, 2006-2016. PLOS ONE. 2018;13(4): e0195435. PubMed\| Google Scholar 12. Gesesew HA, Ward P, Woldemichael K, Mwanri L. Early mortality among children and adults in antiretroviral therapy programs in Southwest Ethiopia, 2003-15. PLoS One. 18 juin 2018;13(6): e0198815. PubMed\| Google Scholar Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 8 Article 13. Ebissa G, Deyessa N, Biadgilign S. Predictors of early mortality in a cohort of HIV-infected children receiving high active antiretroviral treatment in public hospitals in Ethiopia. AIDS Care. 2015;27(6): 723-30. PubMed\| Google Scholar 14. Njom Nlend AE, Loussikila AB. Predictors of mortality children among HIV-infected receiving highly active antiretroviral therapy. Med Mal Infect. 2017;47(1): 32-7. PubMed\| Google Scholar 15. Anigilaje EA, Aderibigbe SA. Mortality in a cohort of HIV-infected children: a 12-month outcome of antiretroviral therapy in Makurdi, Nigeria. Adv Med. 2018;2018: 6409134. PubMed\| Google Scholar 16. Ebonyi AO, Oguche S, Meloni ST, Sagay SA, Kyriacou DN, Achenbach CJ et al. Predictors of mortality in a clinic cohort of HIV-1 infected children initiated on antiretroviral therapy in Jos, Nigeria. J AIDS Clin Res. 2014;5(12): 403. PubMed\| Google Scholar study group 17. European Pregnancy and Paediatric HIV Cohort in (EPPICC) Collaboration EuroCoord, Judd A, Chappell E, Turkova A, Le Coeur S, Noguera-Julian A et al. Long-term trends in mortality and AIDS-defining events after combination ART initiation among children and adolescents with perinatal HIV infection in 17 middle- and high-income countries in Europe and Thailand: a cohort study. PLoS Med. 2018;15(1): e1002491. PubMed\| Google Scholar 18. Modi S, Chiu A, Ng'eno B, Kellerman SE, Sugandhi N, Muhe L et al. Understanding the contribution of common childhood illnesses and opportunistic infections to morbidity and mortality in resource-limited settings. AIDS. 2013;27 Suppl 2(2): S159-67. PubMed\| Google Scholar living with HIV in children for HIV 19. Vermund SH, Blevins M, Moon TD, José E, Moiane L, Tique JA et al. Poor clinical infected children on outcomes antiretroviral therapy in rural Mozambique: need for program quality improvement and community One. PubMed\| Google 2014;9(10): Scholar engagement. e110116. PLoS 20. Vreeman RC, Ayaya SO, Musick BS, Yiannoutsos CT, Cohen CR, Nash D et al. Adherence to antiretroviral therapy in a clinical cohort of HIV-infected children in East Africa. PLoS ONE. 2018;13(2): e0191848. PubMed\| Google Scholar R, Techakunakorn 21. Traisathit P, Delory T, Ngo-Giang-Huong N, Somsamai P, Theansavettrakul S et al. AIDS-defining events and deaths in HIV-infected children and adolescents on antiretrovirals: a 14-year study in Thailand. J Acquir Immune Defic Syndr. 2018;77(1): 17-22. PubMed\| Google Scholar 22. Gebremedhin A, Gebremariam S, Haile F, Weldearegawi B, Decotelli C. Predictors of mortality among HIV infected children on anti- retroviral in Mekelle hospital, Northern Ethiopia: a retrospective cohort study. BMC Public Health. 2013;13: 1047. PubMed\| Google Scholar therapy Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 9 Article 22 45 71 48 Pourcentages (%) 11,8 24,2 38,2 25,8 Table 1 : caractéristiques sociodémographiques des enfants infectés par le VIH et suivis à Ebolowa Variables Effectifs (n) Tranches d’âge (années) ≤2 (2-5) (5-10) (10-16) Sexe Masculin Féminin Résidence Urbain Rural Statut d’orphelin Oui Non Scolarisation Oui Non Pas encore 71 10,3 17,7 132 21 33 40,9 59,1 49,5 50,5 76 110 118 69 92 94 63 37 Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 10 Article 31 155 16,7 83,3 Effectifs (n) 42 44 81 19 Pourcentages (%) 22,6 23,6 43,5 10,2 Table 2: caractéristiques cliniques, paraclinique et thérapeutique à l'initiation du traitement antirétroviral Variables Méthodes de diagnostic PCR Sérologie Classification clinique OMS Stade I Stade II Stade III Stade IV Comorbidités liées au VIH Oui Non Stade immunologique Pas d'ID ID modérée ID avancée ID sévère Anémie Pas d'anémie Légère Modérée Sévère Ravitaillement en ARV (%) 100 50-99 0-49 ID: immunodépression; ARV: antirétroviraux 13,7 28,4 30,4 27,5 22,5 7,9 13,5 56,2 62,9 30 9,1 117 52 17 14 29 31 28 20 7 12 50 34,4 65,6 64 122 Effectifs (n) Table 3: devenir des enfants infectés par le VIH et suivis à Ebolowa Variables Devenir Vivants Décédés Perdus de vue Total 66,7 24,2 9,1 100 124 45 17 186 Pourcentages (%) Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 11 Article Table 4: déterminants de la survie des enfants infectés par le VIH et suivis à Ebolowa Effets ajustés HR (IC â 95%) p value 0,09 0,3 (0,07-1,13) 0,73 (0,2-2,64) 0,63 p value <0,0001 <0,0001 0,034 0,152 1 0,006 0,007 <0,001 0,001 0,002 0,002 1 0,08 0,008 0,048 0,277 0,539 Variables Effets non ajustés HR (IC â 95%) 2,26 (1,23-4,1) 1,78 (0,91-3,47) 6 (0,7-35,2) 2,23 (1,09-16,58) 3,3 (0,94-11,64) 1 18,6 (6,48-53,59) 2,9 (1,08-7,78) 1,92 (0,79-4,66) 1 Tranches d'âge (années) ≤2 (2-5) (5-10) (10-16) Scolarisation (Oui/Non) 0,43 (0,24-0,78) Méthodes diagnostiques (PCR/Sérologie) Comorbidités liées au VIH (Oui/Non) Stade clinique OMS Stade IV Stade III Stade II Stade I Comorbidités non liées au VIH (Oui/Non) Anémie Sévère Modérée Légère Pas d'anémie Ravitaillement en ARV (%) 0-49 50-99 100 Maladies opportunistes (Oui/Non) 7,69 (1,02-57,9) 3,15 (0,4-24,94) 1,96 (0,23-16,9) 1 11 (3,12-29,3) 5,21 (0,19-37,48) 1 0,56 (0,29-1,07) 4,52 (2,51-8,14) 15,62 (3,1-79,7) 1,96 (0,37-10,5) 2,48 (0,74-8,33) 1 0,96 (0,33-2,8) 0,53 4,57 (0,12-33,76) 1,67 (0,13-13,5) 0,88 (0,17-17,29) 1 0,68 (0,17-2,6) 11,5 (1,1-125,3) 5,04 (0,47-53,3) 1,89 (0,19-19,1) 0,001 0,432 0,143 1 0,94 0,08 0,062 0,16 0,09 1 0,587 0,043 0,04 0,18 0,59 0,07 0,08 1 0,05 <0,0001 0,19 <0,0001 <0,0001 1 1,95 (0,79-17,5) 0,3 (0,54-7,64) 1 <0,0001 2,33 (0,98-5,5) Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 12 Article Figure 1: courbe globale de survie des enfants infectés par le VIH et suivis à Ebolowa Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 13 Article Figure 2: facteurs associés à la survie des enfants infectés par le VIH et suivis à Ebolowa Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 14 Article Figure 3: courbes de survie ajustées des enfants selon les tranches d'âge, classes d'hémoglobine et maladies opportunistes Ginette Claude Mireille Kalla et al. PAMJ - 37(308). 03 Dec 2020. - Page numbers not for citation purposes. 15
10.11604_pamj.2019.33.277.18711	Research Open Access Survey of antiretroviral therapy adherence and predictors of poor adherence among HIV patients in a tertiary institution in Nigeria Adekunle Olatayo Adeoti1,&, Mobolaji Dada2, Tobiloba Elebiyo3, Joseph Fadare1, Opeyemi Ojo1 1Department of Medicine, Ekiti State University Teaching Hospital, Ado-Ekiti, Ekiti State, Nigeria, 2Department of Psychiatry, Ekiti State University Teaching Hospital, Ado-Ekiti, Ekiti State, Nigeria, 3Department of Biochemistry, University of Ibadan, Oyo State, Nigeria &Corresponding author: Adekunle Olatayo Adeoti, Department of Medicine, Ekiti State University Teaching Hospital, Ado-Ekiti, Ekiti State, Nigeria Key words: Adherence, predictors, HIV/AIDS, CD4 count, mental health Received: 22/03/2019 - Accepted: 23/06/2019 - Published: 31/07/2019 Abstract Introduction: adherence is vital to effective antiretroviral therapy (ART) for reducing viral load and HIV/AIDS-related morbidity and mortality. This study was aimed at evaluating the adherence of HIV seropositive patients to ART in a tertiary institution in Nigeria. Methods: a cross sectional observational study was conducted among 400 HIV seropositive patients. The study was carried out between December 2016 and February 2017 at the HIV clinic of the Ekiti State University Teaching Hospital, Ado-Ekiti, Nigeria. Results: the mean age of the HIV patients was 42.2±9.5 years with a predominant female gender (Male:Female = 1:2.8). The median CD4 counts increased from 302.1±15.0cells/mm3 at diagnosis to 430.8±13.3cells/mm3 at the time of the study. Majority of participants were unaware of their spouses' HIV status (59.3%) while 32.5% of participants had a serodiscordant spouse. Poverty was a major challenge as 73.3% earned less than 140 dollars per month. Depressive symptoms, anxiety disorder and insomnia were also reported in 40.7%, 33.2% and 47.2% respectively. Poor adherence to ART was observed in almost 20% of the patients. Logistic regression indicated that predictors of poor adherence were depression, anxiety and low CD4 counts. Conclusion: adherence to anti-retroviral therapy was good amongst the majority of HIV seropositive patients. Depression, anxiety disorder and low CD4 count were however associated with poor adherence. This emphasizes the role of the psychology units as integral part of the HIV clinic to assist patients' adherence to anti-retroviral regimens. Pan African Medical Journal. 2019;33:277. doi:10.11604/pamj.2019.33.277.18711 This article is available online at: http://www.panafrican-med-journal.com/content/article/33/277/full/ © Adekunle Olatayo Adeoti et al. The Pan African Medical Journal - ISSN 1937-8688. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Pan African Medical Journal – ISSN: 1937- 8688 (www.panafrican-med-journal.com) Published in partnership with the African Field Epidemiology Network (AFENET). (www.afenet.net) Page number not for citation purposes 1 Introduction HIV remains a major pandemic that has claimed more than 35 million lives over the last three decades [1]. In 2017, approximately a million HIV related death and 1.8 million newly infected were reported by the World Health Organization with a major burden in sub-Saharan Africa [2]. Globally, nearly 40 million people are living with the disease, with a decline in the annual death from AIDS-related causes by nearly half in the past ten years, but this is still higher than the proposed UN target of 500,000 deaths in 2020 [3]. Since the first two AIDS cases were diagnosed in Nigeria in 1985, there has been an increase in the number of new cases with 9% of the global HIV burden coming from this region [4]. Antiretroviral therapy (ART) transformed this potentially incurable disease to a manageable chronic illness by suppressing the viral load and reducing the risk of transmission of the disease [5, 6]. Nevertheless, the continued success of ART is highly dependent on early initiation of therapy, continuity in care and high treatment adherence [7]. Drug adherence is the key factor in disease control, as ART adherence of ≥95.0% can achieve suppression of viral load to undetectable levels, improve immune system function and reduce AIDS-related morbidity and mortality [8, 9]. In addition, achieving the UNAIDS 90-90-90 targets (90% of all people living with HIV will know their HIV status, 90% of those people will be on ART and 90% of them will be virally suppressed) issues of adherence to ART and viral suppression need to be given more attention [10]. An observational study indicated that just 62% of HIV infected patients take at least 90% of their prescribed ART doses [11]. Therefore, drug adherence is a major challenge to effective ART patient management and the development of ART adherence intervention research is crucial for effective HIV management with the aim of achieving the 90-90-90 goal of the UNAIDS/WHO [7]. This study was aimed at evaluating the adherence of HIV seropositive patients to ART treatment regimens in a tertiary institution in Nigeria. Methods capital serving a population of over 2 million people in the state. In 2014, the sentinel HIV survey reported the prevalence of HIV/AIDS in Ekiti State as 2.9%, which was one of the lowest prevalence in the country [12]. Sampling and sample size: Raosoft incorporation software was used to calculate the sample size of 327 using a 95% confidence level [13]. Furthermore, this was increased to 400 for absolute precision of 5% points to have sufficient variation in the population characteristics (age and sex) that may influence adherence as well as power of the study. The list of 1,128 adult HIV patients from HIV clinic register was used to select 400 patients by simple random sampling method. Patients who were aged 18 years or older and had been on anti-retroviral therapy for a minimum period of 6 months were recruited for the study. Patients below 18 years, drug addicts, patients on chronic use of other medications aside anti-retroviral drugs or patients whose treatments were interrupted due to adverse effects were excluded from the study. Data collection tools: the modified 8-item Morisky Medication Adherence Scale (MMAS-8), a validated self-reported questionnaire, was used to assess the adherence of patients to medication [14]. It had eight questions to assess the knowledge and motivation levels of participants regarding adherence. Each question was used in determining a specific type of adherence behaviour. In order for a participant to achieve an optimum result, seven of the questions must have had a negative response while one of them must have had a positive response. One of the questions was answered using a scale of five options: never; almost never; sometimes; often and always. Scoring was graded as high adherence (0 points); medium adherence (1 and 2 points); poor adherence (above 2 points). A closed ended self-assessment questionnaire was employed in the collection of data regarding number of medications which have been taken, number of doses missed by each of the participants, socio- demographic information, information regarding family support and reasons for not taking medications as directed by their doctor. In addition, the Hospital Anxiety Depression Scale (HADS) was also used Study design and setting: this study was a cross sectional to assess the prevalence of anxiety and depression among the observational study which was carried out at the HIV clinic of Ekiti respondents [15]. Insomnia among the patients was assessed with State University Teaching Hospital (EKSUTH), Ado-Ekiti, Nigeria insomnia severity index scale. The questionnaires were administered between December 2016 and February 2017. This HIV clinic runs in to consenting patients who met the inclusion criteria [16]. All recruited conjunction with Ekiti State Agency for Control for HIV/AIDS and the patients were on same antiretroviral therapy combination (Atripla- implementing partners under the support of the Federal Ministry of Efavirenz 600mg/Emtricitabine 200mg/Tenofovir 300mg) for at least Health, Nigeria. EKSUTH is the major tertiary institution in the state six months. The lymphocyte CD4 cell counts were measured for Page number not for citation purposes 2 eligible patients who had been on ART for the stipulated time period load of half of the patients (51.3%) was undetectable (<50 copies). using flow cytometer (Cyflow Partec Counter 2) likewise the viral load About forty percent of the patients were clinically depressed, 47.2% was determined using polymerase chain reaction. had insomnia and 33.2% had anxiety disorder while 81% had good adherence to their medications as shown in Table 2. Data analysis: data collected were analyzed using SPSS 20. Categorical variables were expressed in proportions and compared ART adherence and associated factors: multivariate analysis on using chi-square while continuous variable were expressed in means the association of ART adherence against predictors such as; gender, and compared using T-test. In order to explore the factors associated age, CD4 count, viral load, insomnia index, depression score and with lower adherences scores, a logistic regression analysis was anxiety score were carried out using binary logistic regression. Table 3 employed. P-value <0.05 will be considered as statistically significant. shows that the predictors of poor adherence were depression, anxiety For ease of analysis, the adherence categories comprising of low, and CD4 count below 200 cells/mm3. As presented in Table 3, the medium and high adherence were merged into two groups. The odds of depressed patients adhering to ART regimens were 0.2 times medium and high adherences were merged together as “good lower than patients who were not depressed (0.001). The result also adherence” while low adherence remained as “poor adherence”. indicates that seropositive patients suffering from anxiety are less adherent to ART (0.25 times, p=0.006). The current CD4 count of the Ethical approval: approval was obtained from the Ethical and patients was also a predicator of non-adherence as seropositive Research committee of Ekiti State University Teaching Hospital, Ado- patients with CD4 below 200 were 15 times less adherent to ART Ekiti. Informed consent was obtained from all participants and regimens in comparison to patients whose CD4 count was between confidentiality was ensured in the obtained information. 200-350 cells/m2 (p=0.0005). As indicated by the results presented in Results the table, patients with CD4 below 200 cells/m2 also had ART adherence levels that were 3.75 times lower than those of patients with CD4 above 500 cells/m2 (p=0.001). Socio-demographic characteristics of HIV patients: four hundred HIV seropositive patients were randomly selected for the study. The socio-demographics of the patients presented in Table 1, Discussion indicated that their mean age was 42.18±9.5 years with almost three In this study, majority of the HIV seropositive patients were young quarter of the respondents being females (74%). A large proportion females in their reproductive age. Nigeria has the second largest HIV of the participants were married (76%) and self-employed (60.7%). epidemic in the world [17]. It is estimated that 58% of people living Most of the patients had formal education as a negligible percentage with HIV and AIDS in Nigeria are women [18]. The underlying reason of 1% of the participants had no formal education. It was observed why a larger proportion of women are infected with HIV in comparison from the study that a large proportion of the patients were low income to men in Nigeria has been attributed to deep roots of gender earners with monthly pay below 140 dollars (less than 5 dollars per inequality in Nigeria society, culture and law [19]. The predominantly day). The majority of the HIV infected patients were unaware of their young, female population in our study could be due to more sexually partners HIV status (59.3%) while 32.5% had serodiscordant spouse. active among this younger age group and the receptive sexual anatomic design in females [20]. On the contrary, in countries with Clinical characteristics: the mean CD4 count of the HIV higher prevalence of HIV infection amongst homosexual men, more seropositive patients at diagnosis was an average of males were reported to be HIV positive [21]. A preponderance of the 302.06±15.82cells/mm3 (mean ± SEM). It also showed that about half patients showed good adherence to their ART, however the minority of the HIV infected patients (44.8%) had an initial CD4 count below with poor adherence had associations with depression, anxiety 200cells/mm3. An increase in the CD4 count of the patients to an disorder and low CD4 count. Good adherence to ART has average of 430.8±13.30 cells/mm3 (mean ± SEM) was observed, with revolutionized HIV medicine as it leads to suppressed viral load and a high proportion of the seropositive patients (34.9%) having a CD4 repopulation of diminished CD4 T-lymphocytes and the resultant count above 500 cells/mm3 at the time of the study and also the viral decreases mortality as well as risk of opportunistic infection [22]. Our Page number not for citation purposes 3 study shows an increase in the average CD4 count from diagnosis to Conclusion the time of commencement of this study. There was a significant increase in the CD4 count of the seropositive patients. Also, a high proportion of the seropositive patients had undetectable viral load which indicates minimized risk of transmitting HIV infection to their partners, by viral load suppression, as a third of the respondents had serodiscordant sexual partners [23]. Similar to other studies, significant percentage of the HIV seropositive patients in our study had depression, anxiety disorder and insomnia [24, 25]. These co-existing manifestations have also been reported as sleep disturbances was said to be prevalent amongst the patients with underlying depression [26]. Furthermore, anxiety could be associated to the thought of the illness, fear of the future and financial concerns, as the majority of the study participants earned This study indicates a good adherence in majority of the patients but a small proportion with predictors of poor adherence to anti-retroviral therapy which were depression, anxiety disorder and low CD4 count. An integration of a psychology unit in the HIV programme with regular and active monitoring of patients will be crucial for improved adherence. What is known about this topic • • Adherence to antiretroviral medication is crucial to medication effectiveness; Viral load suppression is necessary in achieving WHO/UNAIDS 90-90-90 goal; less than 140 dollars per month [27]. Most of the HIV seropositive • Drug adherence reduces the risk of the development of patients were adherent to anti-retroviral therapy which was similar to resistant strains. other studies in sub-Saharan Africa with pooled estimate of 77% What this study adds adherence but higher than adherence of 55% in North America [28]. This discrepancy could be due to the effectiveness of the HIV programme in this region and the free HIV treatment funded by the partners, which most patients could consider as a new opportunity to live. In our study, predictors of poor adherence to ART drugs were depression, anxiety disorder and low CD4 count which have been reported in similar studies [29, 30]. Unidentified mental health disorders among people living with HIV may prevent the actualization of the last 90 in the UNAIDS-90-90-90-target. This has significant impact on their quality of life, disease progression and mortality [30]. Thus, regular and active psychological assessment for all HIV patients • • • The predictors of poor antiretroviral adherence in Ekiti State University Teaching Hospital (EKSUTH); Identified the key role of mental health as an integral component of HIV medicine; Estimated the level of adherence of HIV seropositive patients in EKSUTH. Competing interests becomes imperative for a holistic care in order to achieve this target. The authors declare no competing interests. Poor adherence, accompanying sub-optimal antiretroviral effect of medication and its attendant development of drug resistant strains of HIV, low CD4 count and unsuppressed viral load could be responsible Authors’ contributions for the low CD4 count as a predictor in our study [31, 32]. Furthermore, the majority of the patients in the study presented in advanced stage of the disease with low CD4 count. This late presentation could account for slower increase in CD4 cell count especially in patients with virological failure [29]. Limitation of the study: a multi-center study and community based study would have been more representative compared to our study which was hospital based. Also, the data on adherence was based on responses from patients which may have some level of bias. Adekunle Olatayo Adeoti: concept, write up and final draft. Mobolaji Dada: concept, write up and final draft. Tobiloba Elebiyo: statistical analysis, final draft. Joseph Fadare: statistical analysis, final draft. Opeyemi Ojo: final draft. All the authors have read and agreed to the final manuscript. Page number not for citation purposes 4 Acknowledgments The authors would like to acknowledge the contributions of Dr Kukoyi Oladipo (M.D., M.S, Birmingham VA Medical Center, IOWA, USA) and clinical psychologists in the department of psychiatry, EKSUTH, Ado- Ekiti. We also acknowledge the support of the Equitable Health Access Initiative (EHAI) during this study. Tables Table 1: socio-demographics of the HIV patients 7. Chaiyachatia KH, Ogbuojib O, Priceb M, Sutharc AB, Negussiec EK, Ba¨rnighausenb T. Interventions to improve adherence to antiretroviral therapy: a rapid systematic review. AIDS. 2014 Mar;28 Suppl 2:S187-204. PubMed \| Google Scholar 8. Chesney MA. Factors affecting adherence to antiretroviral therapy. Clin Infect Dis. 2000 Jun;30 Suppl 2:S171-6. PubMed \| Google Scholar 9. Bangsberg DR, Moss AR, Deeks SG. Paradoxes of adherence and drug resistance to HIV antiretroviral therapy. J Antimicrob Chemother. 2004;53(5):696-9. PubMed \| Google Scholar Table 2: clinical characteristics of the HIV patients 10. UNAIDS. 90-90-90: An ambitious treatment target to help end Table 3: logistic regression analysis showing associations between the AIDS epidemic. Accessed 18th Feburary 2019. adherence and potential predictors References 11. Ortego C, Huedo-Medina TB, Llorca J, Sevilla L, Santos P, Rodriguez E et al. Adherence to highly active antiretroviral therapy (HAART): a meta-analysis. AIDS Behav. 2011; 15(7): 1381-96. PubMed \| Google Scholar 1. WHO. Global action plan on HIV drug resistance 2017-2021. Accessed 18th Feburary 2019. 12. National agency for the control of AIDS. Nigeria GARPR 2015 report. Accessed 4th March 2019. 2. UNAIDS. UNAIDS data 2018. Accessed 30th January 2019. 3. HIV Gov. Global Statistics. Accessed 4th March 2019. 4. USAID. Orphans and Vulnerable Children Affected by HIV and AIDS. Accessed 18thFeburary 2019. 5. Palella FJ, Delaney KM, Moorman AC, Loveless MO, Fuhrer J, Satten GA et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection HIV outpatients study investigations. New England Journal of Medicine. 1998; 338: 853-860. Google Scholar 6. Ware NC, Idoko J, Kaaya S, Biraro IA, Wyatt MA, Agbaji O et al. Explaining adherence success in sub-Saharan Africa: an ethnographic study. PLoS Med. 2009;6(1):e11. PubMed \| Google Scholar 13. Raosoft. Database web survey software for gathering information. Accessed 24thDecember 2017. 14. Morisky DE, Ang A, Krousel-Wood M, Ward HJ. Predictive validity of a medication adherence measure in an outpatient setting. J Clin Hypertens (Greenwich). 2008;10(5):348-54. PubMed \| Google Scholar 15. Stern AF. The hospital anxiety and depression scale. Occup Med (Lond). 2014;64(5):393-4. PubMed \|Google Scholar 16. Morin CM, Belleville G, Belanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatment response. Sleep. 2011;34(5):601-8. PubMed \| Google Scholar 17. National agency for the control of AIDS. National HIV and AIDS Strategic Plan 2017-2021. Accessed 4th March 2019. 18. National agency for the control of AIDS. FACT SHEET: HIV Prevention program. Accessed 15thJanuary 2019. Page number not for citation purposes 5 19. UNAIDS. Prevention Gap Report. Accessed 4th March 2019. 27. Low Y, Preud'homme X, Goforth HW, Omonuwa T, Krystal AD. 20. Gyar SD, Reuben CR, Haruna MS. Study on the Distribution of seropositive patients: a pilot study. Sleep. 2011;34(12):1723- HIV/AIDS Infections among Age Groups Attending General 1726. PubMed \|Google Scholar Hospital Toto, Central Nigeria. International Journal of Microbiology and Immunology Research. 2014;3(3):038-042. 28. Mills EJ, Nachega JB, Buchan I, Orbinski J, Attaran A, Singh S The association of fatigue with depression and insomnia in HIV Google Scholar et al. Adherence to antiretroviral therapy in sub-Saharan Africa and North America: a meta-analysis. JAMA. 2006;296(6):679-90. 21. Miyada S, Garbin AJI, Gatto RCJ, Garbin CAS. Treatment PubMed \|Google Scholar adherence in patients living with HIV/AIDS assisted at a specialized facility in Brazil. Rev Soc Bras Med Trop. 29. Trotta MP, Cozzi-Lepri A, Ammassari A, Vecchiet J, Cassola G, 2017;50(5):607-612. PubMed \|Google Scholar Caramello P et al. Rate of CD4+ cell count increase over periods of viral load suppression: relationship with the number of 22. Detels R1, Muñoz A, McFarlane G, Kingsley LA, Margolick JB, previous virological failures. Clin Infect Dis. 2010;51(4):456- Giorgi J et al. Effectiveness of potent Antiretroviral Therapy on 64. PubMed \| Google Scholar time to AIDS and DEATH in men with known HIV infection. JAMA. 1998;280(17):1497-1499. PubMed \| Google Scholar 30. Ammassari A, Murri R, Pezzotti P, Trotta MP, Ravasio L, De Longis P et al. Self reported symptoms and medication side effects 23. Kumi SM, jewell m, Hallett T, Cohen M. Treatment of HIV for the influence adherence to highly active anti-retroviral therapy in prevention of transmission in discordant couples and at the persons with HIV infection. Journal of acquired immune population level. Adv Exp Med Biol. 2018;1075:125-162. deficiency syndromes. 2011;28(5):445-9. PubMed \| Google PubMed \| Google Scholar Scholar 24. Eller LS, Rivero-Mendez M, Voss J, Chen WT, Chaiphibalsarisdi P, 31. Abdulraheem IS, Onajole AT, Jimoh AAG, Oladipo AR. Reasons Iipinge S et al. Depressive symptoms, self-esteem, HIV symptom for incomplete vaccination and factors for missed opportunities management self-efficacy and self-compassion in people living among rural Nigerian children. Journal of Public Health and with HIV. AIDS Care. 2014;26(7):795-803. PubMed \| Google Epidemiology. 2011;3(4):194-203. Google Scholar Scholar 25. Ramirez-Avila L, Regan S, Giddy J, Chetty S, Ross D, Katz JN et M, Sheiner L et al. Adherence to protease inhibitors, HIV-1 viral al. Depressive symptoms and their impact on health-seeking load, and development of drug resistance in an indigent behaviors in newly-diagnosed HIV-infected patients in Durban, population. AIDS. 2000; 14(4): 357-366. PubMed \| Google 32. Bangsberg DR1, Hecht FM, Charlebois ED, Zolopa AR, Holodniy South Africa. AIDS Behav. 2012;16(8):2226-2235. PubMed \| Scholar Google Scholar 26. Rubinstein ML, Selwyn PA. High prevalence of insomnia in an outpatient population with HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;19(3):260-5. PubMed \| Google Scholar Page number not for citation purposes 6 Frequency 104 296 31 284 85 Table 1: socio-demographics of the HIV patients Variable Age in years (42.2 ± 9.5 years) Below 30 30-50 Above 50 Sex Male Female Marital status Single Married Divorced Widowed Education None Primary Secondary Tertiary Average income per month < 140 dollars 140–280 dollars > 280 dollars HIV Status of spouse/partners Unknown Positive Negative Exchange rate at 357 Naira to 1 dollar Values are expressed in median ± SD 4 68 144 197 42 304 24 30 237 33 130 289 86 12 Percentage (%) 7.8 71.0 21.2 26.0 74.0 10.5 76.0 6.0 7.5 1.0 16.5 34.9 47.7 73.3 23.4 3.3 59.3 8.2 32.5 Table 2: clinical characteristics of the HIV patients Variables Initial CD4 (cells/mm3) Below 200 200-350 351-500 Above 500 Current CD4 (cells/mm3) Below 200 200-350 351-500 Above 500 Viral load (number of copies) Below 50 (undetectable) 50 -1000 Above 1000 Depression (Median depression score = 7 ± 4.4) Non case Depression Insomnia (Median insomnia index = 6 ± 5.5) Clinically insignificant Insomnia Anxiety (6 ± 3.9) Non case Anxiety Adherence (1 ± 1.2) Poor Good Values are expressed in median ± SD P is statistically significant at p< 0.05 Frequency Percentage (%) 179 89 68 63 70 81 105 137 98 61 32 273 163 211 189 267 132 78 323 44.9 22.3 17 15.8 17.8 20.6 26.7 34.9 51.3 31.9 16.8 59.3 40.7 52.8 47.2 66.8 33.2 19 81 Page number not for citation purposes 7 Table 3: logistic regression analysis showing associations between adherence and potential predictors Variables OR (95% Cl) Adherence p-value Good n (%) Poor n (%) 78 (74.3) 246 (83.7) 205 (86.9) 119 (73) 23 (67.6) 230 (80.1) 71 (91) 27 (62.8) 245 (81.7) 24 (96) 28 (90.3) Age Below 30 30 -50 Above 50 Sex Male Female Marital status Single Married Divorced Widowed Depression Not depressed Depressed Insomnia Absent Present Anxiety Absent Present Viral load Suppressed Unsuppressed CD4 at diagnosis Below 200 200-350 351-500 Above 500 Current CD4 Below 200 200-350 351-500 Above 500 P is statistically significant at p< 0.05 37 (57.8) 68 (91.9) 82 (81.2) 133 (86.9) 133 (79.6) 67 (80.7) 54 (85.7) 54 (84.4) 177 (70) 147 (77.8) 303 (81.7) 21 (75) 235 (82.7) 89 (78.1) 11 (34.4) 57 (19.9) 7 (9) 27 (25.7) 48 (16.3) 16 (37.2) 55 (18.3) 1 (4) 3 (9.7) 31 (13.1) 44 (27) 33 (30) 42 (22.2) 49 (17.3) 25 (21.9) 68 (18.3) 7 (25) 34 (20.4) 16 (19.3) 9 (14.3) 10 (15.6) 27 (42.2) 6 (8.1) 19 (18.8) 17 (13.7) 1 1.09 (0.18 – 6.73) 0.99 (0.30 – 3.22) 1 2.64 (0.99 – 6.96) 1 2.47 (0.34 – 18.09) 0.87 (0.16 – 4.58) 0.09 (0.01 – 1.48) 1 0.20 (0.08 – 0.52) 1 0.78 (0.34 – 1.78) 1 0.25 (0.09 – 0.67) 1 0.36 (0.09 – 1.58) 1 0.29 ( 0.08 – 1.03) 0.97 (0.28 – 3.37) 0.56 (0.15 – 2.09) 0.991 0.925 0.985 0.050 0.050 0.101 0.374 0.865 0.092 0.001 0.001 0.561 0.561 0.006 0.006 0.163 0.163 0.147 0.056 0.963 0.388 1 15.15 (4.24 – 54.04) 1.45 (0.38 – 5.58) 3.75 (1.38 – 10.19) 0.0005 0.0005 0.592 0.001 Page number not for citation purposes 8
10.1371_journal.pone.0287008	RESEARCH ARTICLE Sex differences in hepatitis A incidence rates– a multi-year pooled-analysis based on national data from nine high-income countries Manfred S. GreenID, Naama Schwartz, Victoria PeerID School of Public Health, University of Haifa, Haifa, Israel [email protected] a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Abstract Background OPEN ACCESS Citation: Green MS, Schwartz N, Peer V (2023) Sex differences in hepatitis A incidence rates–a multi-year pooled-analysis based on national data from nine high-income countries. PLoS ONE 18(6): e0287008. https://doi.org/10.1371/journal. pone.0287008 Editor: Inge Roggen, Universitair Kinderziekenhuis Koningin Fabiola: Hopital Universitaire des Enfants Reine Fabiola, BELGIUM Received: April 30, 2022 Accepted: May 28, 2023 Published: June 13, 2023 Copyright: © 2023 Green et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: The data underlying the results presented in the study are available from Table 1. Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist. Possible sex differences in hepatitis A virus (HAV) incidence rates in different age groups are not well documented. We aimed to obtain stable pooled estimates of such differences based on data from a number of high-income countries. Methods We obtained data on incident cases of HAV by sex and age group over a period of 6–25 years from nine countries: Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand and Spain. Male to female incidence rate ratios (IRR) were com- puted for each year, by country and age group. For each age group, we used meta-analytic methods to combine the IRRs. Meta-regression was conducted to estimate the effects of age, country, and time period on the IRR. Results A male excess in incidence rates was consistently observed in all age groups, although in the youngest and oldest age groups, where the numbers tended to be lower, the lower bounds of the 95% confidence intervals for the IRRs were less than one. In the age groups <1, 1–4, 5–9, 10–14, 15–44, 45–64 and 65+, the pooled IRRs (with 95% CI) over countries and time periods were 1.18 (0.94,1.48), 1.22 (1.16,1.29), 1.07 (1.03,1.11), 1.09 (1.04,1.14), 1.46 (1.30,1.64), 1.32 (1.15,1.51) and 1.10 (0.99,1.23) respectively. Conclusions The excess HAV incidence rates in young males, pooled over a number of countries, sug- gest that the sex differences are likely to be due at least in part to physiological and biologi- cal differences and not just behavioral factors. At older ages, differential exposure plays an important role. These findings, seen in the context of the excess incidence rates in young PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 1 / 18 PLOS ONEMale excess in hepatitis A incidence rates males for many other infectious diseases, can provide further keys to the mechanisms of the infection. Introduction There is an expanding literature on sex differences in the incidence rates of various infectious diseases [1–3]. The type and extent of the differences frequently vary by disease and age group. The mechanisms underlying these differences have not been fully elucidated and cannot be explained entirely by differences in exposure. The pattern of male to female ratios in the inci- dence rates of different infectious diseases can make an important contribution to understand- ing the underlying mechanisms of the diseases. Despite the availability of an effective vaccine, hepatitis A virus (HAV) infection remains a common disease, particularly in low-income countries with overcrowding and poor sanitation, where the incidence rates of the disease are particularly high in infancy and childhood [4, 5]. In countries with high hepatitis A vaccine coverage, the incidence of cases and outbreaks have decreased in children and the infection has shifted significantly to other risk groups, such as men who have sex with men (MSM) [6–9]. There are reports in the literature on sex differences in the incidence rates of hepatitis A, but they are inconsistent and poorly documented by age group [10–13]. While some report higher incidence rates of viral hepatitis A in males [6, 10], there are inconsistencies. For exam- ple, in a report from Germany, during 2018–2020, no sex differences were observed in the inci- dence of the disease [11]. One report from South Korea found a change in the sex differences, possibly due to increased immunization in the military [13]. In this study, we aimed to obtain pooled estimates of the age-specific male to female ratios in the incidence rates of HAV infection based on data from a number of developed countries over extended time periods. Methods Source of data National surveillance data on reported cases of HAV infection, by age, sex and year, were obtained from relevant government institutions for nine countries from Czech Republic, Fin- land, Germany, Netherland, Spain, Australia, New Zealand, Canada and Israel. The data for Australia, for years 2001–2016, was extracted from the National Notifiable Diseases Surveil- lance System (NNDSS), [14] for Canada for the years 1991–2015, from the Public Health Agency of Canada (PHAC) [15], for the Czech Republic, for 2008–2013, from the Institute of Health Information and Statistics [16], for Finland, for years1995-2016 from the National Institute for Health and Welfare (THL) [17], for Germany for the years 2001–2016, from the German Federal Health Monitoring System [18], for Israel from the Department of Epidemiol- ogy in the Ministry of Health for years 1998–2016, for the Netherland (2003–2017), directly from the official representative of RIVM, for New Zealand for years 1997–2015 from the Insti- tute of Environmental Science and Research (ESR) [19] and for Spain from the Spanish Epide- miological Surveillance for years 2005–2015 [20]. Information about the population size by age, sex and year was obtained for Australia from ABS.Stat [21] (Australia’s Bureau of statistics), for Canada from Statistics Canada CANSIM database [22], for the Czech Republic from the Czech Statistical Office [23], for Finland from Statistics Finland’s PX-Web databases [24], for Germany from the German Federal Health PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 2 / 18 PLOS ONEMale excess in hepatitis A incidence rates Monitoring System [25], for Israel from the Central Bureau of Statistics [26], for Netherland from Netherlands’ database (StatLine) [27], for New Zealand from Statistics New Zealand [28] and for the Spain from the Department of Economic and Social Affairs, Population Division [29]. Ethics and informed consent National, open access, sex-and-age disaggregated, anonymous data were used and there was no need for ethics committee approval. Statistical analyses Data analysis. HAV incidence rates (IR) per 100,000 were calculated by age group and sex, for each country and calendar year using the number of reported cases divided by the respective population size and multiplied by 100,000. The age groups considered were <1 years (infants), 1–4 (early childhood), 5–9 (late childhood), 10–14 (puberty), 15–44 (young adulthood), 45–64 (middle adulthood) and 65+ (senior adulthood). Surveillance systems in Canada and New Zealand used similar age-groups except for 15–39, 40–59 and 60+. For Aus- tralia, data for infants and age 1–4, disaggregated by sex and age, are missing. The male to female incidence rate ratio (IRR) was calculated by dividing the incidence rate in males by that of females, by age group, country and time period. Pooled analysis. As in previous studies of sex differences in infectious diseases [1–3], we used meta-analytic methods to establish the magnitude of the pooled sex differences in the incidence of HAV infection, by age group, across different countries and over a number of years. The outcome variable was the male to female IRR. For each age group, the IRRs for each country were pooled over time periods and then the pooled IRRs for each country were com- bined. Forest plots with the pooled IRRs, over countries and years of reporting, were prepared separately for the seven age groups. Heterogeneity was evaluated using the Q statistic and I2 was calculated as an estimate of the percentage of between-study variance. If the p-value for the Q statistic was less than 0.05, or I2 exceeded 50%, the random effects models was used to estimate pooled IRRs and 95% confidence intervals (CI). Otherwise, the fixed effects model was considered, although due to the low power of the Q statistic, the more conservative ran- dom effects model was preferred. In order to explore the contribution of countries and the reported years to the variability in the IRRs, meta-regression analyses were performed. To eval- uate the effect of individual countries and years on the male to female incidence risk ratio, we performed leave-one-out sensitivity analysis and recomputed the pooled IRRs. The meta-ana- lytic methods and meta-regressions were carried out using STATA software version 12.1 (Stata Corp., College Station, TX). Results Descriptive statistics The summary of the male to female IRRs per 100,000 populations in different countries for each age group is presented in Table 1. Significant differences in incidence rates were observed between the countries, with the high- est incidence rates in all ages and both sexes in Czech Republic. Higher incidence rates were observed in Israel and Spain up to age 44 and in Germany in the group of adults (age 45–64). Forest plots. The forest plots for the IRRs by age group, are shown in Figs 1–7. The forest plot for infants is shown in Fig 1. The pooled male to female IRR was 1.18 (95% CI 0.94–1.48) with I2 = 0.0% and varied between 0.86 in Canada and 2.26 in Spain. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 3 / 18 PLOS ONETable 1. Details of the countries included in the study, by sex and age group—descriptive data. Male excess in hepatitis A incidence rates Age <1 1–4 5–9 10–14 15–44 45–64 Country Canada Czech Republic Germany Israel Netherland New Zealand Spain Canada Czech Republic Germany Israel Netherland New Zealand Spain Australia Canada Czech Republic Finland Germany Israel Netherlands New Zealand Spain Australia Canada Czech Republic Finland Germany Israel Netherlands New Zealand Spain Australia Canada Czech Republic Finland Germany Israel Netherlands New Zealand Spain Australia Canada Czech Republic Finland Germany Years 1991–2015 2008–2013 2001–2016 1998–2016 2003–2017 1997–2015 2005–2015 1991–2015 2008–2013 2001–2016 1998–2016 2003–2017 1997–2015 2005–2015 2001–2016 1991–2015 2008–2013 1995–2016 2001–2016 1998–2016 2003–2017 1997–2015 2005–2005 2001–2016 1991–2015 2008–2013 1995–2016 2001–2016 1998–2016 2003–2017 1997–2015 2005–2005 2001–2016 1991–2015 2008–2013 1995–2016 2001–2016 1998–2016 2003–2017 1997–2015 2005–2005 2001–2016 1991–2015 2008–2013 1995–2016 2001–2016 Males n/N 29/4682619 25/349195 27/5740478 38/1486100 3/1616870 3/576900 41/2679186 679/19156418 267/1410748 615/23509315 778/5731500 94/5811264 80/2308880 587/10880587 258/11398585 1556/24668602 256/1532669 42/3440956 1361/30760941 1263/6616300 208/7478265 126/2899540 983/13017097 190/11377822 954/25685783 157/1416001 45/3522497 908/33455166 522/6106400 182/7658243 72/2919850 535/12301238 1462/73591102 7148/143987472 1428/13725818 562/18898064 3840/257895408 1103/29586200 774/39930903 401/13546700 4962/110542308 420/41988401 2419/110461323 356/8403729 220/16513241 Females IR 0.62 7.16 0.47 2.56 0.19 0.52 1.53 3.54 18.93 2.62 13.57 1.62 3.46 5.39 2.26 6.31 16.70 1.22 4.42 19.09 2.78 4.35 7.55 1.67 3.71 11.09 1.28 2.71 8.55 2.38 2.47 4.35 1.99 4.96 10.40 2.97 1.49 3.73 1.94 2.96 4.49 1.00 2.19 4.24 1.33 n/N 32/4446799 23/332712 23/5448550 34/1410400 3/1540059 1/548520 17/2514548 560/18225737 222/1343670 511/22311030 550/5443300 88/5543773 55/2191980 419/10233932 212/10814642 1506/23469919 220/1450621 42/3297629 1196/29187252 1012/6287700 243/7139402 139/2752910 821/12287011 144/10797396 854/24391864 148/1339518 42/3375446 846/31724889 430/5807300 151/7312854 80/2776650 409/11627137 990/72741755 3620/140453550 918/12978912 332/18050351 2753/247590330 878/29264100 414/39138712 295/13976900 2425/105413400 330/42573071 1390/109655649 347/8624880 177/16307550 IR 0.72 6.91 0.42 2.41 0.19 0.18 0.68 3.07 16.52 2.92 10.10 1.59 2.51 4.09 1.96 6.42 15.17 1.27 4.10 16.09 3.40 5.05 6.68 1.33 3.50 11.05 1.24 2.67 7.40 2.06 2.88 3.52 1.36 2.58 7.07 1.84 1.11 3.00 1.06 2.11 2.30 0.78 1.27 4.02 1.09 1742/181698132 15.55 1694/181849520 15.08 PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 RR 0.86 1.04 1.11 1.06 0.95 2.85 2.26 1.15 1.15 1.14 1.34 1.02 1.38 1.32 1.15 0.98 1.10 0.96 1.08 1.19 0.82 0.86 1.13 1.25 1.06 1.00 1.03 1.02 1.15 1.15 0.86 1.24 1.46 1.93 1.47 1.62 1.34 1.24 1.83 1.40 1.95 1.29 1.73 1.05 1.23 1.03 (Continued ) 4 / 18 PLOS ONEMale excess in hepatitis A incidence rates Table 1. (Continued) Age 65+ Country Israel Netherlands New Zealand Spain Australia Canada Czech Republic Finland Germany Israel Netherlands New Zealand Spain Years 1998–2016 2003–2017 1997–2015 2005–2005 2001–2016 1991–2015 2008–2013 1995–2016 2001–2016 1998–2016 2003–2017 1997–2015 2005–2005 Males n/N 140/12368500 400/36361477 186/10201030 601/63103755 120/21417772 693/64590224 50/4087584 75/11159619 948/108019284 65/6010700 95/24551738 75/6302700 131/37127234 Females IR 1.13 1.10 1.82 0.95 0.56 1.07 1.22 0.67 0.88 1.08 0.39 1.19 0.35 n/N 129/13327000 211/35859166 126/10685350 432/64340310 151/25538457 816/78346403 93/5999018 74/15066114 1435/149862231 54/7903600 89/29339281 70/7386000 133/49879431 IR 0.97 0.59 1.18 0.67 0.59 1.04 1.55 0.49 0.96 0.68 0.30 0.95 0.27 RR 1.17 1.87 1.55 1.42 0.95 1.03 0.79 1.37 0.92 1.58 1.28 1.26 1.32 IR = incidence rate, IR per 100 000 Male or Female population, incidence RR = female: male incidence Rate Ratio n- Cumulative total of cases for given years. N- Cumulative total of the population for given years. Infants = age<1 year; early childhood = 1–4 years; late childhood = 5–9 years; puberty = 10–14 years; young adulthood = 15–44 or 15–39 years; middle adulthood = 40– 59 or 45–64 years; senior adulthood = 60+ or 65+ years. https://doi.org/10.1371/journal.pone.0287008.t001 The forest plot for the age 1–4 is shown in Fig 2. The pooled IRR was 1.22 (95% CI 1.16– 1.29) with I2 = 23.3% and varied from 1.02 in Netherland 1.38 in New Zealand. The forest plot for age 5–9 is given in Fig 3. The pooled IRR was 1.07 (95% CI 1.03–1.11) with I2 = 30.3% and varied from 0.82 in the Netherlands to 1.19 in Israel. The forest plot for age 10–14 is given in Fig 4. The pooled IRR was 1.09 (95% CI 1.04–1.14) with I2 = 0.0% and varied between 0.86 in New Zealand to 1.25 in Australia. The forest plot for age 15–44 is given in Fig 5. The pooled IRR was 1.46 (95% CI 1.30–1.64), I2 = 95.9% and varied between 1.33 in Israel to 1.86 in the Netherlands. The forest plot for age 45–64 is shown in Fig 6. The pooled IRR = 1.32 (95% CI 1.15–1.51), I2 = 90.8%, and varied from 1.03 in Germany to 1.87 in the Netherlands. The forest plot for age 65+ is given in Fig 7. The pooled IRR was 1.10 (95% CI 0.99–1.23) I2 = 57.0% and varied from 0.80 in Czech Republic to 1.50 in Israel. Other analyses. Meta-regression analysis showed that almost all the variance in the inci- dence RRs was contributed by the age groups, with small differences between countries and time periods. To evaluate the effect of individual countries on the male to female incidence ratios, we performed leave-one-out sensitivity analysis and recomputed the pooled IRRs (pre- sented in Tables 2 and 3). After omitting each country (one country at a time, Table 2) or a group of years at a time (Table 3), the pooled IRR’s remained very similar. Thus, no single country or group of years substantially affected the pooled IRRs. This con- firms that the results of this pooled analysis are stable and robust. Discussion In this study, we found that the incidence rates of clinically manifested HAV, pooled over a number of years, for various high-income countries, are consistently higher in males in all age PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 5 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 1. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) in infancy for different years in Canada, Czech Republic, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g001 groups. In the youngest and oldest age groups, where the numbers were small, the confidence intervals included unity. Based on the pooled analysis of national data from nine countries, over a period of 6–25 years, we found that the incidence rates of clinical hepatitis A were higher in males by 22%, 7%, 9%, 46%, 32%, and 10% in the age groups 1–4, 5–9, 10–14, 15–44, 45–64 and 65+ respectively. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 6 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 2. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 1–4, for different years in Canada, Czech Republic, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g002 While sex differences in the incidence of HAV have been examined in a number of studies, they have usually been conducted in individual countries or selected groups of patients. For example, in a national study in Israel in 1992, there was a male predominance of HAV inci- dence rates [30]. This sex differential was especially pronounced among infants. In a 15-year nationwide epidemiological study in Taiwan, there were higher hospitalization rates in males while male sex and age over 40 years were significant factors associated with mortality [31]. In PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 7 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 3. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 5–9, for different years in Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g003 the study of HAV patients in Saudi Arabia, no sex differences were among hospitalized patients [12]. In a hepatitis A outbreak in Chiba, Japan, in 2011, 40.7% of the 27 patients were male [32], and in another, 65% of the 60 patients were male [33]. However, these figures may simply represent gender differences in exposure to the virus. In addition, the impact of vac- cines on sex differences in HAV incidence rates is not clear. There is evidence that females may respond with up to 2–3 times higher anti-HAV antibody levels than males after the prim- ing and after the booster dose and has been observed at different ages [34–37]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 8 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 4. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 10–14, for different years in Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g004 The incidence of both viral and bacterial diseases have frequently been reported to be higher in males [1–3]. In addition, there are reported sex differences in the severity of different infections, suggesting that males are more prone to suffer from clinical manifestations of infec- tions than females [38, 39]. While in excess morbidity in males is most common for infectious diseases [1–3], pertussis is a prominent exception, where there is a female excess in morbidity [40]. It is of interest that in the COVID-19 pandemic, there has been no clear evidence of sex PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 9 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 5. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 15–44 (15–39), for different years in Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g005 differences in incidence rates, although case-fatality rates have consistently been reported to be higher in males [41, 42], even after controlling for other variables. It has been shown that the male to female IRRs differential will be most evident where there is a low proportion of clinical disease [30]. Since children more commonly suffer from asymp- tomatic HAV infection [43, 44] and the clinical to subclinical ratio for HAV increases with age, one might expect that the male excess in disease would be less evident at older ages. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 10 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 6. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 45-64(40–59), for different years in Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g006 However, the higher male to female IRRs in the older age groups is most likely due to larger differences in exposure in high risk groups such as in the men who have sex with men (MSM) or people who are HIV positive [7, 8, 45–49]. Thus, behavioral factors can partially explain sex differences in HAV incidence rates in the older age groups. For the youngets age group, there may be protection from maternal HAV antibodies on short-term immunity [50]. However, we have not found evidence that it impacts male and female infants differently. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 11 / 18 PLOS ONEMale excess in hepatitis A incidence rates Fig 7. Forest plot of the male to female hepatitis A incidence rate ratios (IRR) at age 65+ (60+), for different years in Australia, Canada, Czech Republic, Finland, Germany, Israel, Netherland, New Zealand, and Spain. https://doi.org/10.1371/journal.pone.0287008.g007 The exact mechanisms underlying the excess HAV incidence rates in males found in this study are not clear and probably multi-factorial. This study was not designed to address the mechanisms. In addition to behavioral differences, genetic and hormonal factors could be important. In infants and early childhood, and based on the seroprevalence studies, it is unlikely that the sex differences in incidence rates are due to differences in exposure [51]. A study of kindergarten children showed that females had higher anti-HAV antibodies than PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 12 / 18 PLOS ONEMale excess in hepatitis A incidence rates Table 2. Sensitivity analysis, by age group and country. Sensitivity by country Countries Removed Infants Early childhood Late childhood Puberty Young adulthood Middle adulthood Senior adulthood Australia Canada - - 1.04 (0.96–1.13) 1.08 (1.03–1.14) 1.58 (1.37–1.82) 1.35 (1.11–1.64) 1.13 (0.98–1.3) 1.29 (0.99–1.66) 1.24 (1.17–1.32) 1.06 (0.98–1.15) 1.1 (1.04–1.16) 1.52 (1.33–1.74) 1.3 (1.11–1.51) 1.13 (0.96–1.34) Czech Republic 1.21 (0.95–1.56) 1.23 (1.16–1.3) 1.04 (0.96–1.13) 1.09 (1.04–1.15) 1.58 (1.37–1.81) 1.39 (1.15–1.67) 1.14 (1–1.29) Finland Germany Israel Netherland New Zealand Spain - - 1.05 (0.97–1.14) 1.09 (1.04–1.14) 1.56 (1.36–1.79) 1.36 (1.12–1.64) 1.08 (0.95–1.23) 1.2 (0.93–1.54) 1.24 (1.17–1.32) 1.04 (0.95–1.14) 1.12 (1.06–1.18) 1.6 (1.41–1.81) 1.4 (1.2–1.62) 1.15 (1–1.32) 1.23 (0.94–1.59) 1.19 (1.12–1.26) 1.03 (0.95–1.11) 1.08 (1.02–1.14) 1.61 (1.42–1.83) 1.36 (1.13–1.65) 1.07 (0.95–1.2) 1.19 (0.94–1.5) 1.23 (1.17–1.3) 1.08 (1.01–1.15) 1.09 (1.03–1.14) 1.54 (1.34–1.76) 1.29 (1.07–1.55) 1.09 (0.95–1.24) 1.17 (0.93–1.47) 1.22 (1.16–1.29) 1.08 (1.01–1.15) 1.1 (1.04–1.15) 1.59 (1.38–1.81) 1.32 (1.09–1.6) 1.09 (0.96–1.25) 1.02 (0.8–1.32) 1.2 (1.13–1.28) 1.08 (1.01–1.15) 1.07 (1.01–1.12) 1.52 (1.33–1.74) 1.33 (1.09–1.63) 1.08 (0.95–1.22) IRR = Incidence rate ratio; CI = confidence interval https://doi.org/10.1371/journal.pone.0287008.t002 males [52]. In adults, the results are varied. In a study of blood donors in the US in 2015, [53] no sex differences were observed in the prevalence of anti-HAV IgG antibodies (61% and 60% for males and females, respectively). In a study of ambulatory patients in Portugal between 2002 and 2012, no significant differences between sexes were observed [54]. In a study of refu- gees and asylum seekers in Germany, HAV seroprevalence rates were higher in adult males than females [55]. Although liver injury in hepatitis A is known to be caused by immune-mediated events, the exact biological mechanisms are not clarified. It is plausible that immune-related mechanisms of liver injury are common to the pathogenesis of all types of hepatitis [56]. Virus-specific CD8 + T cells from hepatitis A patients are considered as a major cause of liver damage. Natural killer cells are also involved and contribute to liver damage [57, 58]. In hepatitis A patients, serum levels of cytokines and chemokines, including interleukin (IL)-6, IL-8, IL-18, IL-22, CXC-chemokine ligand (CXCL)9, and CXCL10 are increased [59] and contribute to liver injury. Many studies have shown that the overall inflammatory response, innate and adaptive immune systems are stronger in females than males, with greater CD4+ T-cell counts a higher CD4+ /CD8+ ratio in females but higher CD8+ T and NK frequencies in males [60]. Sex differences in the clinical expression of hepatitis A may be related to the imbalance in the expression of genes encoded on the X and Y-chromosomes of a host. X chromosome-asso- ciated biological processes and X-linked genes are responsible for the immunological advan- tage of females due to the X-linked microRNAs related processes. The phenomenon of X chromosome inheritance and expression is a cause of immune disadvantage of males and the enhanced survival of females following immunological challenges [61]. The increase in sex hormone levels in infancy that mimics sex steroid levels during puberty (‘minipuberty’) could affect immune cells differently in boys and girls. Testosterone levels Table 3. Sensitivity analysis, by age group and years. Sensitivity by years Years Removed Infants Early childhood Late childhood Puberty Young adulthood Middle adulthood Senior adulthood 1991–1999 2000–2009 2010–2017 1.2 (0.92–1.58) 1.21 (1.13–1.29) 1.08 (1.03–1.13) 1.11 (1.05–1.17) 1.49 (1.3–1.72) 1.19 (1.14–1.24) 0.97 (0.87–1.09) 1.07 (0.79–1.45) 1.21 (1.12–1.29) 1.04 (0.99–1.09) 1.07 (1–1.14) 1.68 (1.16–2.43) 1.51 (0.92–2.48) 1.13 (0.94–1.35) 1.26 (0.96–1.64) 1.25 (1.18–1.33) 1.07 (1.03–1.12) 1.09 (1.03–1.15) 1.8 (1.43–2.27) 1.52 (0.94–2.46) 1.06 (0.79–1.42) IRR = Incidence rate ratio; CI = confidence interval https://doi.org/10.1371/journal.pone.0287008.t003 PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 13 / 18 PLOS ONEMale excess in hepatitis A incidence rates predominate in boys at 1–3 months of age and decline at 6–9 months of age, whereas in girls, estradiol levels remain elevated longer [62]. This phenomenon of ’’mini-puberty’’ with sex dif- ferences in gonadal hormone levels could influence the maturation of the immune system [63]. This transient rise in sex steroid levels may also influence immune cells differently between boys and girls at later ages [64]. Before any physical signs of puberty, girls had higher levels of estrogens than boys at age 5–9. These higher estradiol levels or lower testosterone lev- els in young girls may play a part in protection against clinical disease and should be investi- gated further. Strengths and limitations This current study has several strengths and limitations. The inclusion of nine countries, each evaluated over a number of years, allowed us to evaluate the consistency of the findings over different populations and many years. The analyses are based on national data where both the numbers of cases and denominators are large. Selection bias has been minimized by using national data, which should be representative of each country. However, the countries evalu- ated in this study are classified as high-income, so the results may not be directly generalizable to low- and middle-income countries. Differential underreporting between countries is likely and may contribute to the variability in the incidence of reported cases of HAV. However, there does not appear to be any reason to believe that the reporting differs between males and females. In the countries examined, there is no evidence that male infants and children are more likely to receive health care. Thus any information bias in the underreporting of inci- dence rates will most likely be non-differential by sex and the IRRs should not be materially affected. In adults, there could be gender differences in the utilization of medical care, although reports suggest that females in some countries tend to make greater use of health services [65], which would operate in the opposite direction of our observations. Conclusions This study provides stable estimates of the excess male incidence rates in hepatitis A incidence rates in most age groups. While much of the excess in older males may be attributed to differ- ential exposure, the excess in young males, while not large, is remarkably consistent over a number of high-income countries and for extended periods of time. The mechanism is largely unknown. A better understanding of the gender differences can help to elucidate genetic and hormonal determinants of HAV infection and contribute to the role of sex as a biological variable. Acknowledgments We thank the official representative of RIVM, Netherlands, and to all the official institutions of all other countries for the providing their national data on hepatitis A incidence. Author Contributions Conceptualization: Manfred S. Green. Data curation: Manfred S. Green, Naama Schwartz, Victoria Peer. Formal analysis: Naama Schwartz. Methodology: Manfred S. Green, Naama Schwartz, Victoria Peer. Project administration: Manfred S. Green. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 14 / 18 PLOS ONEMale excess in hepatitis A incidence rates Supervision: Manfred S. Green. Writing – original draft: Manfred S. Green. Writing – review & editing: Manfred S. Green, Victoria Peer. References 1. Green MS, Schwartz N, Peer V. Sex differences in campylobacteriosis incidence rates at different ages —a seven country, multi-year, meta-analysis. A potential mechanism for the infection. BMC Infect Dis. 2020; 20:625. https://doi.org/10.1186/s12879-020-05351-6 PMID: 32842973 2. Peer V, Schwartz N, Green MS. Consistent, excess viral meningitis incidence rates in young males: A multi-country, multi-year, meta-analysis of national data. The importance of sex as a biological variable. EClinicalMedicine. 2019; 15:62–71. https://doi.org/10.1016/j.eclinm.2019.08.006 PMID: 31709415 3. Peer V, Schwartz N, Green MS. A pooled analysis of sex differences in rotaviral enteritis incidence rates in three countries over different time periods. Womens Health Rep (New Rochelle). 2022; 3:228– 237 https://doi.org/10.1089/whr.2021.0096 PMID: 35262061 4. 5. Jacobsen KH. Globalization and the changing epidemiology of hepatitis A virus. Cold Spring Harb Per- spect Med. 2018; 8:a031716. https://doi.org/10.1101/cshperspect.a031716 PMID: 29500305 Zeng Dan-Yi, Li Jing-Mao, Lin Su, et al. Global burden of acute viral hepatitis and its association with socioeconomic development status, 1990–2019. J Hepatol. 2021. 75:547–556. https://doi.org/10. 1016/j.jhep.2021.04.035 PMID: 33961940 6. Green MS, Block C, Slater PE. Rise in the incidence of viral hepatitis in Israel despite improved socio- economic conditions. Rev Infect Dis. 1989; 11:464–469. https://doi.org/10.1093/clinids/11.3.464 PMID: 2749104 7. Alberts CJ, Boyd A, Bruisten SM, et al. Hepatitis A incidence, seroprevalence, and vaccination decision among MSM in Amsterdam, the Netherlands. Vaccine. 2019; 37:2849–2856 https://doi.org/10.1016/j. vaccine.2019.03.048 PMID: 30992222 8. Honda M, Asakura H, Kanda T, et. al. Male-dominant hepatitis A outbreak observed among non-HIV- infected persons in the northern part of Tokyo, Japan.Viruses. 2021; 13:207. 9. Andani A, Bunge E, Kassianos G, et.al. Hepatitis A occurrence and outbreaks in Europe over the past two decades: a systematic review. J Viral Hepat. 2023 10. Bauer D, Farthofer A, Chromy D, et al. Recent outbreaks of severe hepatitis A virus infections in Vienna. Eur J Clin Microbiol Infect Dis. 2021; 40:335–344. https://doi.org/10.1007/s10096-020-04028-x PMID: 32940811 11. Dudareva S, Faber M, Zimmermann R, et.al. Epidemiology of viral hepatitis A to E in Germany. Bundes- gesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2022; 65:149–158. 12. Al-Tawfiq JA, Anani A. Profile of viral hepatitis A, B, and C in a Saudi Arabian hospital. Med Sci Monit. 2008; 14:CR52–56. PMID: 18160946 13. Choe YJ, Son H. The changing gender differences in hepatitis A incidence in South Korea. Vaccine. 2020; 38: 712–714. https://doi.org/10.1016/j.vaccine.2019.11.048 PMID: 31787416 14. National Notifiable Diseases Surveillance System (NNDSS), Department of Health. Available at: http:// www9.health.gov.au/cda/source/rpt_5_sel.cfm. Accessed 1 April 2018. 15. Public Health Agency of Canada. Available at: https://www.canada.ca/en/public-health.html. Accessed on 1 June 2018. 16. Institute of Health Information and Statistics. Available at: https://www.uzis.cz/en/catalogue/infectious- diseases. Accessed 1 March 2018. 17. National institute for health and welfare (THL): https://www.thl.fi/ttr/gen/rpt/tilastot.html. Accessed on May 1, 2018. 18. German Federal Health Monitoring System. Available at: http://www.gbe-bund.de/ gbe10/pkg_isgbe5. prc_isgbe? p_uid¼gast&p_aid¼0&p_sprache¼D (1 February 2018, date last accessed). 19. Environmental Science and Research (ESR) for the Ministry of Health. Available at: https://surv.esr.cri. nz/surveillance/annual_surveillance.php. Accessed 30 March 2018. 20. Instituto de Salud Carlos III. Available: http://www.eng.isciii.es/ISCIII/es/contenidos/fd-servicios- cientifico-tecnicos/fd-vigilancias-alertas/fd-enfermedades/enfermedades-declaracion-obligatoria- informes-anuales.shtml. Accessed 1 March 2018. 21. ABS.Stat (Australian Bureau of Statistics). Available at: http://stat.data.abs.gov.au/Index.aspx? DatasetCode=ABS_ERP_ASGS2016.Accessed 15 May 2018. PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 15 / 18 PLOS ONEMale excess in hepatitis A incidence rates 22. Statistics, Canada, CANSIM database: Available at: https://www150.statcan.gc.ca/t1/tbl1/en/cv.action? pid=1710010201. Accessed 1 June 2018. 23. Czech Statistical Office. Available at: https://www.czso.cz/csu/czso/population. Accessed 1 March 2018. 24. Statistics Finland’s PX-Web databases. Available at: http://pxnet2.stat.fi/PXWeb/pxweb/en/StatFin/ StatFin__vrm__vaerak/statfin_vaerak_pxt_021.px/?rxid=2f968705-bdaa-48b1-9d5a-d4985ead7d40. Accessed 15 April 2018. 25. German Federal Health Monitoring System: http://www.gbe-bund.de/gbe10/abrechnung.prc_abr_test_ logon?p_uid=gast&p_aid=46300054&p_knoten=VR&p_sprache=E&p_suchstring=population. Accessed on February 1, 2018 26. Central Bureau of Statistics: http://www.cbs.gov.il/reader/shnatonhnew_site.htm?sss=%E4%EE%F9% EA&shnaton_scan=45. Accessed on March 1, 2018. 27. Statistics Netherlands’ database (StatLine). Available at: https://opendata.cbs.nl/statline/#/CBS/en/ dataset/37325eng/table?ts=1528798782913. Accessed 15 May 2018. 28. Stats NZ, Infoshare. Available at: http://archive.stats.govt.nz/infoshare/SelectVariables.aspx?pxID= b854d8a2-3fdf-402c-af69-604112e80baa. Accessed 15 May 2018. 29. Demographic Statistics Database (United Nations Statistics: Division). Available at: http://data.un.org/ Data.aspx?d=POP&f=tableCode%3A22. Accessed 1 April 2018. 30. Green MS. The male predominance in the incidence of infectious diseases in children: a postulated explanation for disparities in the literature. Int J Epidemiol. 1992; 21:381–386 https://doi.org/10.1093/ ije/21.2.381 PMID: 1428496 31. Chen CM, Chen SC, Yang HY, Yang ST, Wang CM. Hospitalization and mortality due to hepatitis A in Taiwan: a 15-year nationwide cohort study. J Viral Hepat. 2016; 23:940–945. https://doi.org/10.1111/ jvh.12564 PMID: 27386835 32. 33. Tominaga A, Kanda T, Akiike T, et.al. Hepatitis A outbreak associated with a revolving sushi bar in Chiba, Japan: Application of molecular epidemiology. Hepatol Res. 2012; 42: 828–834. https://doi.org/ 10.1111/j.1872-034X.2012.00988.x PMID: 22776552 Takahashi H, Yotsuyanagi H, Yasuda K, et.al. Molecular epidemiology of hepatitis A virus in metropoli- tan areas in Japan. J Gastroenterol. 2006; 41:981–986. https://doi.org/10.1007/s00535-006-1888-9 PMID: 17096067 34. Bovier PA, Bock J, Loutan L, Farinelli T, Glueck R, Herzog C. Long-term immunogenicity of an inacti- vated virosome hepatitis A vaccine. J Med Virol. 2002; 68:489–493. https://doi.org/10.1002/jmv.10244 PMID: 12376955 35. Ho¨hler T, Groeger-Bicanic G, Hoet B, Stoffel M. Antibody persistence and immune memory elicited by combined hepatitis A and B vaccination in older adults. Vaccine. 2007; 25:1503–1508. https://doi.org/ 10.1016/j.vaccine.2006.10.024 PMID: 17097774 36. Spradling PR, Bulkow LR, Negus SE, Homan C, Bruce MG, McMahon BJ. Persistence of seropositivity among persons vaccinated for hepatitis A during infancy by maternal antibody status: 15-year follow- up. Hepatology. 2016; 63:703–711. https://doi.org/10.1002/hep.28375 PMID: 26637987 37. Van Herck K, Hens A, De Coster I, et.al. Long-term antibody persistence in children after vaccination with the pediatric formulation of an aluminum-free virosomal hepatitis A vaccine. Pediatr Infect Dis J. 2015; 34:e85–91 https://doi.org/10.1097/INF.0000000000000616 PMID: 25389920 38. Klein SL, Flanagan KL. Sex differences in immune responses. Nat Rev Immunol. 2016; 16:626–638 https://doi.org/10.1038/nri.2016.90 PMID: 27546235 39. Klein SL, Jedlicka A, Pekosz A. The Xs and Y of immune responses to viral vaccines. Lancet Infect Dis. 2010; 10:338–349. https://doi.org/10.1016/S1473-3099(10)70049-9 PMID: 20417416 40. Peer V, Schwartz N, Green MS. A multi-country, multi-year, meta-analytic evaluation of the sex differ- ences in age-specific pertussis incidence rates. PLoS One. 2020; 15:e0231570 https://doi.org/10.1371/ journal.pone.0231570 PMID: 32324790 41. Jacobsen H, Klein SL. Sex differences in immunity to viral Infections. Front Immunol. 2021; 12:720952 https://doi.org/10.3389/fimmu.2021.720952 PMID: 34531867 42. Green MS, Nitzan D, Schwartz N, Niv Y, Peer V. Sex differences in the case-fatality rates for COVID- 19-A comparison of the age-related differences and consistency over seven countries. PLoS One. 2021; 16(4):e0250523 https://doi.org/10.1371/journal.pone.0250523 PMID: 33914806 43. Wensley A, Smout E, Ngui SL, et.al. An outbreak of hepatitis A virus infection in a secondary school in England with no undetected asymptomatic transmission among students. Epidemiol Infect. 2022; 151: e6 https://doi.org/10.1017/S095026882200190X PMID: 36502811 PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 16 / 18 PLOS ONEMale excess in hepatitis A incidence rates 44. Abutaleb A, Kottilil S. Hepatitis A: epidemiology, natural history, unusual clinical manifestations, and prevention. Gastroenterol Clin North Am. 2020; 49:191–199. https://doi.org/10.1016/j.gtc.2020.01.002 PMID: 32389358 45. 46. Lin KY, Chen GJ, Lee YL, et.al. Hepatitis A virus infection and hepatitis A vaccination in human immuno- deficiency virus-positive patients: A review. World J Gastroenterol. 2017; 23: 3589–3606. https://doi. org/10.3748/wjg.v23.i20.3589 PMID: 28611512 Franco E, Giambi C, Ialacci R, Coppola RC, Zanetti AR. Risk groups for hepatitis A virus infection. Vac- cine. 2003; 21:2224–2233 https://doi.org/10.1016/s0264-410x(03)00137-3 PMID: 12744847 47. Migueres M, Lhomme S, Izopet J. Hepatitis A: epidemiology, high-risk groups, prevention and research on antiviral treatment. Viruses. 2021; 13:1900 https://doi.org/10.3390/v13101900 PMID: 34696330 48. Yoshimura Y, Horiuchi H, Sawaki K, et.al. Hepatitis A Outbreak Among Men Who Have Sex With Men, Yokohama, Japan, January to May 2018. Sex Transm Dis. 2019; 46: e26–e27. https://doi.org/10.1097/ OLQ.0000000000000937 PMID: 30395105 49. Bazzardi R, Dore E, Ciccozzi M, et.al. Outbreak of acute hepatitis A associated with men who have sex with men (MSM) in North Sardinia 2017–2018. J Infect Dev Ctries. 2020; 14:1065–1070. https://doi.org/ 10.3855/jidc.12184 PMID: 33031097 50. Bell BP, Negus S, Fiore AE, et.al. Immunogenicity of an inactivated hepatitis A vaccine in infants and young children. J.Pediatr Infect Dis J. 2007; 26: 116–122. https://doi.org/10.1097/01.inf.0000253253. 85640.cc PMID: 17259872 51. Walter F, Ott JJ, Claus H, Krause G. Sex- and age patterns in incidence of infectious diseases in Ger- many: analyses of surveillance records over a 13-year period (2001–2013). Epidemiol Infect. 2018; 146: 372–378. https://doi.org/10.1017/S0950268817002771 PMID: 29357958 52. 53. Lin DB, Tsai TP, Yang CC, et.al. Association between seropositivity of antibodies against hepatitis a virus and Helicobacter pylori. Am J Trop Med Hyg. 2000; 63:189–191. https://doi.org/10.4269/ajtmh. 2000.63.189 PMID: 11388513 Tejada-Strop A, Zafrullah M, Kamili S, Stramer SL, Purdy MA. Distribution of hepatitis A antibodies in US blood donors. Transfusion. 2018; 58: 2761–2765. https://doi.org/10.1111/trf.14916 PMID: 30284286 54. Pereira S, Linhares I, Neves AF, Almeida A. Hepatitis A immunity in the District of Aveiro (Portugal): an eleven-year surveillance study (2002–2012). Viruses. 2014; 6: 133613–45. https://doi.org/10.3390/ v6031336 PMID: 24638206 55. Jablonka A, Solbach P, Wo¨bse M, et.al. Seroprevalence of antibodies and antigens against hepatitis A- E viruses in refugees and asylum seekers in Germany in 2015. Eur J Gastroenterol Hepatol. 2017; 29: 939–945. https://doi.org/10.1097/MEG.0000000000000889 PMID: 28492419 56. Shin EC, Sung PS, Park SH. 2016a. Immune responses and immunopathology in acute and chronic viral hepatitis. Nat Rev Immunol. 2016: 509–523. https://doi.org/10.1038/nri.2016.69 PMID: 27374637 57. Wang M, Feng Z. Mechanisms of Hepatocellular Injury in Hepatitis A. Viruses. 2021; 13:861. https://doi. org/10.3390/v13050861 PMID: 34066709 58. Vallbracht A, Maier K, Stierhof YD, Wiedmann KH, Flehmig B, Fleischer B. Liver-derived cytotoxic T cells in hepatitis A virus infection. J Infect Dis. 1989; 160:209–217. https://doi.org/10.1093/infdis/160.2. 209 PMID: 2503564 59. Shin SY, Jeong SH, Sung PS, et.al. Comparative analysis of liver injury-associated cytokines in acute hepatitis A and B. Yonsei Med J. 2016; 57:652–657. https://doi.org/10.3349/ymj.2016.57.3.652 PMID: 26996565 60. Abdullah M, Chai P-S, Chong M-Y, et al. Gender effect on in vitro lymphocyte subset levels of healthy individuals. Cell Immunol. 2012; 272:214–219. https://doi.org/10.1016/j.cellimm.2011.10.009 PMID: 22078320 61. Schurz H, Salie M, Tromp G, Hoal EG, Kinnear CJ, Mo¨ ller M. The X chromosome and sex-specific effects in infectious disease susceptibility. Hum Genomics. 2019; 13:2. https://doi.org/10.1186/s40246- 018-0185-z PMID: 30621780 62. Lanciotti L, Cofini M, Leonardi A, Penta L, Esposito S. Up-To-Date review about minipuberty and over- view on hypothalamic-pituitary-gonadal axis activation in fetal and neonatal life. Front Endocrinol (Lau- sanne). 2018; 9:410. https://doi.org/10.3389/fendo.2018.00410 PMID: 30093882 63. Moreira-Filho CA, Bando SY, Bertonha FB, et.al Minipuberty and sexual dimorphism in the infant human thymus. Sci Rep. 2018; 8: 13169. https://doi.org/10.1038/s41598-018-31583-3 PMID: 30177771 64. Courant F, Aksglaede L, Antignac JP, et.al. Assessment of circulating sex steroid levels in prepubertal and pubertal boys and girls by a novel ultrasensitive gas chromatography-tandem mass spectrometry PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 17 / 18 PLOS ONEMale excess in hepatitis A incidence rates method. J Clin Endocrinol Metab. 2010; 95:82–92. https://doi.org/10.1210/jc.2009-1140 PMID: 19933393 65. Bertakis KD, Azari R, Helms LJ, Callahan EJ, Robbins JA. Gender differences in the utilization of health care services. J Fam Pract. 2000; 49:147–152. PMID: 10718692 PLOS ONE \| https://doi.org/10.1371/journal.pone.0287008 June 13, 2023 18 / 18 PLOS ONE
10.3390_ijms22031055	Article Physiological and Proteomic Analyses of Two Acanthus Species to Tidal Flooding Stress Yi-ling Liu and Hai-lei Zheng * Key Laboratory of the Ministry of Education for Coastal and Wetland Ecosystems, College of the Environment and Ecology, Xiamen University, Xiamen 361005, Fujian, China; [email protected] * Correspondence: [email protected] Abstract: The mangrove plant Acanthus ilicifolius and its relative, Acanthus mollis, have been previ- ously proved to possess diverse pharmacological effects. Therefore, evaluating the differentially expressed proteins of these species under tidal ﬂooding stress is essential to fully exploit and beneﬁt from their medicinal values. The roots of A. ilicifolius and A. mollis were exposed to 6 h of ﬂooding stress per day for 10 days. The dry weight, hydrogen peroxide (H2O2) content, anatomical characteris- tics, carbon and energy levels, and two-dimensional electrophoresis coupled with MALDI-TOF/TOF MS technology were used to reveal the divergent ﬂooding resistant strategies. A. ilicifolius performed better under tidal ﬂooding stress, which was reﬂected in the integrity of the morphological structure, more efﬁcient use of carbon and energy, and a higher percentage of up-regulated proteins associated with carbon and energy metabolism. A. mollis could not survive in ﬂooding conditions for a long time, as revealed by disrupting cell structures of the roots, less efﬁcient use of carbon and energy, and a higher percentage of down-regulated proteins associated with carbon and energy metabolism. Energy provision and ﬂux balance played a role in the ﬂooding tolerance of A. ilicifolius and A. mollis. Keywords: Acanthus species; ﬂooding stress; physiological; comparative proteomics analyses; carbon; energy metabolism Citation: Liu, Y.-l.; Zheng, H.-l. Physiological and Proteomic Analyses of Two Acanthus Species to Tidal Flooding Stress. Int. J. Mol. Sci. 1. Introduction 2021, 22, 1055. https://doi.org/ 10.3390/ijms22031055 Received: 3 December 2020 Accepted: 18 January 2021 Published: 21 January 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). The physical characteristics of soil inﬂuence a variety of physiological and biochemical processes of plants. The leaves and roots of terrestrial plants absorb molecular oxygen from air and land, respectively [1]. Previous studies have shown that ﬂooding stress is a widespread phenomenon that inhibits plant growth and production [2]. Continuous and heavy rainfall causes soil pores saturated with excess water, inducing oxygen deﬁciency in plant roots [1,2]. Meanwhile, the roots that are subjected to ﬂooding stress may inhibit photosynthesis, including a decrease of photosynthetic electron transport chain and an increase in the level of reactive oxygen species (ROS) [3]. A shift of aerobic respiration to anaerobic respiration reduced the availability of the adenosine triphosphate (ATP) in plants [4] and increased the content of ethanol [5]. Some plants have evolved morphological, physiological, and metabolic adaptation strategies to ensure survival under ﬂooding stress [6,7]. For example, maize develops an extensive aerenchyma system to facilitate gas transport apart from adventitious roots [6]. Rice retains a gas-associated ﬁlm to facilitate oxygen uptake to survive under ﬂooding [7]. Mangrove plants, such as Kandelia obovata, Sonneratia apetala, Aegiceras corniculatum, and Rhizophora stylosa develop specialized roots for gaseous exchange [8]. Soybean (PELBR10- 6000) increased the level of CO2 assimilation rate and readily responded to the lack of energy by activating the fermentative enzymes and alanine aminotransferase, when it was allowed to recover for additional seven days after ﬂooding treatment [9], indicating that the capacity to quickly resume the normal energy level is crucial in tolerating ﬂooding stress [10]. Int. J. Mol. Sci. 2021, 22, 1055. https://doi.org/10.3390/ijms22031055 https://www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2021, 22, 1055 2 of 16 A comparative study of species is one of the important methods to determine the mechanism of stress-resistance. The difference in fermentative enzymes and alanine aminotransferase activity resulted in different responses to energy deﬁciency between two soybean genotypes under ﬂooding conditions [9]. A comparison of Alternanthera philoxeroides with Hemarthria altissima showed that plants could adapt to wetland habitats, in which water levels ﬂuctuate, by maintaining the functionality of the photosynthetic apparatus [11]. Waterlogged Phalaris aquatica and Festuca arundinacea regained growth during the recovery period compared with Dactylis glomerata and Bromus catharticus [12]. In addition, the favorable alleles of related species are more comfortable to introduce to improve crops [13]. The transfer of resistance genes between Sinapis alba and Brassica species by somatic and sexual hybridization has been accomplished [14]. The mangrove plant, Acanthus ilicifolius, has remarkable morphology and physiol- ogy [15]. A. ilicifolius is mainly distributed in Australia, Australasia, and the southeast- ern Asia intertidal zone and has numerous medicinal properties [16]. Previous ﬁndings showed that untreated or submerged A. ilicifolius over 3 h per day was not conducive to the growth [17]. Meanwhile, the effect of ﬂooding stress on A. ilicifolius at the molecular level is not well elucidated. Acanthus belongs to the Acanthaceae family and is the only genus that comprises of both terrestrial and aquatic species [15]. As the Acanthus model plant, A. mollis is native to the Mediterranean region from central Europe and northwest Africa [18]. A. mollis was recently introduced into China and used as a medicinal plant in traditional medicine [18,19]. The extracts of A. mollis tissues have been used for the treatment of inﬂammation and cancer problems [20]. However, the ﬂooding tolerance of A. mollis has not yet been described. Since non-model plants lack a genetic transformation system to elucidate the metabolic mechanism, proteome and transcriptome are useful to provide powerful information about the metabolic pathways of non-model plants. The protein, as the functional executor, is closely related to physiological changes. In our previous study, we have reported the ﬂooding tolerance of the leaves of Avicennia marina using comparative proteomic analyses [21]. Hence, evaluating the differentially expressed proteins (DEPs) of A. ilicifolius and A. mollis under tidal ﬂooding stress is essential to fully exploit and beneﬁt from their medicinal values. Parafﬁn sections, physiological index measurements, and two-dimensional elec- trophoresis (2-DE) technique were performed on the leaves and roots of A. ilicifolius and A. mollis under tidal ﬂooding stress. Our results ﬁrst provide the anatomical characteristics, carbon and energy levels, and proteomic information about A. ilicifolius and its relative, A. mollis, under tidal ﬂooding stress. 2. Results 2.1. Relative Dry Weight and H2O2 Content of Acanthus Species under Tidal Flooding Stress The two species changed the relative dry weight and H2O2 content to differing extents under tidal ﬂooding stress. The relative dry weight of A. ilicifolius had no change in both the leaf and root tissues (Figure 1A). The relative H2O2 content was signiﬁcantly decreased on the tenth day in A. ilicifolius leaves and showed no signiﬁcant differences from day four to 12 in A. ilicifolius roots (Figure 1C). Tidal ﬂooding treatment signiﬁcantly decreased the relative dry weight (Figure 1B) and increased the relative H2O2 content of A. mollis in both tissues (Figure 1D). Overall, the relative dry weight (expressed as a percentage of the control) of A. ilicifolius was signiﬁcantly higher compared to that of A. mollis from day eight to 12 (Supplementary Table S1). 2.2. Effect of Tidal Flooding on the Phenotype and Anatomical Characteristics of Acanthus Species The phenotype and anatomical characteristics of the two species are shown in Figure 2. After 10 days of tidal ﬂooding treatment, there was no signiﬁcant change in A. ilicifolius (Figure 2(A1,B1)), while the length of A. mollis roots became shorter (Figure 2(C1,D1)). Int. J. Mol. Sci. 2021, 22, 1055 3 of 16 Figure 1. (A) Relative plant dry weight of A. ilicifolius. (B) Relative plant dry weight of A. mollis. (C) Relative H2O2 content of A. ilicifolius. (D) Relative H2O2 content of A. mollis. Gray represents leaves; black represents roots. Data are shown as means ± SD from three independent biological replicates. Means marked with the same letter were not different from each other, but were different from means marked with a different letter; p < 0.05. The different bars indicate the different tissue of plant. Figure 2. Phenotypic and anatomical changes of A. ilicifolius and A. mollis exposed to tidal ﬂooding for 10 days. (A1–A5) A. ilicifolius plants on control treatment, (B1–B5) A. ilicifolius plants under tidal ﬂooding stress, (C1–C5) A. mollis plants on control treatment, (D1–D5) A. mollis plants under tidal ﬂooding stress. Row 1: The phenotypic of Acanthus species, Row 2: The cross-section of the leaf blade, Row 3: The main vein of the leaf, Row 4: Stele of root, Row 5: Epidermis of root. Root sections, about 5 cm from root tip, photos of optical microscopes. Cross-sections with thickness of 10 mm were made and stained with safranine and fast green. cc: Cork cambium; co: Collenchyma; ct: Cortex; en: Endodermis; Ep: Epidermis; gt: Glandular trichome; le: Lenticel; pe: Periderm; ph: Phloem; pi: Pith; pp: Palisade parenchyma; sa: Schizogenous aerenchyma; st: Stomata; sp: Lacunar parenchyma; xy: Xylem. Bars = 100 µm. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 17 Figure 1. (A) Relative plant dry weight of A. ilicifolius. (B) Relative plant dry weight of A. mollis. (C) Relative H2O2 content of A. ilicifolius. (D) Relative H2O2 content of A. mollis. Gray represents leaves; black represents roots. Data are shown as means ± SD from three independent biological replicates. Means marked with the same letter were not different from each other, but were differ-ent from means marked with a different letter; p < 0.05. The different bars indicate the different tissue of plant. 2.2. Effect of Tidal Flooding on the Phenotype and Anatomical Characteristics of Acanthus Species The phenotype and anatomical characteristics of the two species are shown in Figure 2. After 10 days of tidal flooding treatment, there was no significant change in A. ilicifolius (Figure 2(A1,B1)), while the length of A. mollis roots became shorter (Figure 2(C1,D1)). The leaf blade of A. ilicifolius consisted of the upper epidermis, upper multiple epi-dermises, palisade parenchyma, spongy parenchyma, lower epidermis, and salt gland (Figure 2(A2)), whereas that of A. mollis showed a different structure. The upper and lower epidermises of A. mollis were all single-layered and had glandular trichome (Figure 2(C2)). A. ilicifolius roots consisted of the endodermis, epidermis, xylem, phloem, pith, cortex, lenticel, and periderm (Figure 2(A4,A5)), whereas the roots of A. mollis contained cork cambium (Figure 2(D5)). In the leaf blade of A. ilicifolius, the vein phloem possessed a hollow cavity that was enlarged after tidal flooding treatment (Figure 2(B3)). The control group of A. mollis ex-hibited schizogenous aerenchyma in the leaf vein, which disappeared after tidal flooding treatment. The periderm is a secondary protective tissue that protects plant roots from bacterial infections [22]. In A. mollis roots, the pith parenchyma cells were damaged and the periderm cells were ruptured under tidal flooding stress (Figure 2(D4,D5)). Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 4 of 17 Figure 2. Phenotypic and anatomical changes of A. ilicifolius and A. mollis exposed to tidal flooding for 10 days. (A1–A5) A. ilicifolius plants on control treatment, (B1–B5) A. ilicifolius plants under tidal flooding stress, (C1–C5) A. mollis plants on control treatment, (D1–D5) A. mollis plants under tidal flooding stress. Row 1: The phenotypic of Acanthus species, Row 2: The cross-section of the leaf blade, Row 3: The main vein of the leaf, Row 4: Stele of root, Row 5: Epidermis of root. Root sections, about 5 cm from root tip, photos of optical microscopes. Cross-sections with thickness of 10 mm were made and stained with safranine and fast green. cc: Cork cambium; co: Collenchyma; ct: Cortex; en: Endodermis; Ep: Epidermis; gt: Glandular trichome; le: Lenticel; pe: Periderm; ph: Phloem; pi: Pith; pp: Palisade parenchyma; sa: Schizogenous aerenchyma; st: Stomata; sp: Lacunar parenchyma; xy: Xylem. Bars = 100 μm. 2.3. Identification and Quantification of Tidal Flooding-Responsive Proteins Representative 2-DE gels of the leaves and roots of the two species are shown in Figure 3. In A. ilicifolius, approximately 78 and 40 spots were identified from leaves and roots, respectively (Figure 3A,B; Supplemental Tables S2, S3 and S6). In A. mollis, about 67 and 45 spots were identified from leaves and roots, respectively (Figure 3C,D; Supple-mental Tables S4–S6). To understand the global relationship between samples, PCA was performed to eval-uate the similarity between the samples (Figure 4) [23]. In A. ilicifolius leaves, the control group and the tidal flooding group were well separated from each other in the dimension of the second component. In A. mollis tissues, the control group and the tidal flooding group were well separated from each other in the dimension of the first component. Leaves and roots were well separated from each other in the dimension of the second component. However, the two species were not well separated from each other. PC1 ex-plained 21.0% and PC2 19.4% of total variance. Int. J. Mol. Sci. 2021, 22, 1055 4 of 16 The leaf blade of A. ilicifolius consisted of the upper epidermis, upper multiple epi- dermises, palisade parenchyma, spongy parenchyma, lower epidermis, and salt gland (Figure 2(A2)), whereas that of A. mollis showed a different structure. The upper and lower epidermises of A. mollis were all single-layered and had glandular trichome (Figure 2(C2)). A. ilicifolius roots consisted of the endodermis, epidermis, xylem, phloem, pith, cortex, lenticel, and periderm (Figure 2(A4,A5)), whereas the roots of A. mollis contained cork cambium (Figure 2(D5)). In the leaf blade of A. ilicifolius, the vein phloem possessed a hollow cavity that was enlarged after tidal ﬂooding treatment (Figure 2(B3)). The control group of A. mollis exhibited schizogenous aerenchyma in the leaf vein, which disappeared after tidal ﬂooding treatment. The periderm is a secondary protective tissue that protects plant roots from bacterial infections [22]. In A. mollis roots, the pith parenchyma cells were damaged and the periderm cells were ruptured under tidal ﬂooding stress (Figure 2(D4,D5)). 2.3. Identiﬁcation and Quantiﬁcation of Tidal Flooding-Responsive Proteins Representative 2-DE gels of the leaves and roots of the two species are shown in Figure 3. In A. ilicifolius, approximately 78 and 40 spots were identiﬁed from leaves and roots, respectively (Figure 3A,B; Supplemental Tables S2, S3 and S6). In A. mollis, about 67 and 45 spots were identiﬁed from leaves and roots, respectively (Figure 3C,D; Supplemental Tables S4–S6). To understand the global relationship between samples, PCA was performed to evaluate the similarity between the samples (Figure 4) [23]. In A. ilicifolius leaves, the control group and the tidal ﬂooding group were well separated from each other in the dimension of the second component. In A. mollis tissues, the control group and the tidal ﬂooding group were well separated from each other in the dimension of the ﬁrst component. Leaves and roots were well separated from each other in the dimension of the second component. However, the two species were not well separated from each other. PC1 explained 21.0% and PC2 19.4% of total variance. 2.4. Functional Classiﬁcation of DEPs More proteins were up-regulated in A. ilicifolius than in A. mollis tissues (Figure 5, Supplemental Table S7). A higher percentage of up-regulated proteins were found in carbon and energy metabolism and amino acid and protein metabolism in A. mollis leaves, while in transcription and signal transduction in A. ilicifolius leaves. In addition, A. ilicifolius leaves had a lower percentage of up-regulated proteins associated with stress and defense. Overall, A. ilicifolius tissues had a high percentage of up-regulated proteins and a low percentage of down-regulated proteins associated with carbon and energy metabolism. Meanwhile, a higher percentage of down-regulated proteins of A. mollis leaves were associated with carbon and energy metabolism, stress and defense, and transcription and signal transduction (Figure 5, Supplemental Table S7). Compared with A. mollis roots, a higher percentage of up-regulated proteins of A. ilicifolius roots were associated with carbon and energy metabolism, amino acid and protein metabolism, stress and defense, and transcription and signal transduction. In A. mollis roots, there was a higher percentage of down-regulated proteins in all pathways (Figure 5, Supplemental Table S7). 2.5. Identiﬁcation of Hub Proteins in Acanthus Species Because of the lack of genome information on A. ilicifolius and A. mollis, the DEPs of the two species were annotated based on the existing non-redundant protein sequence database (NR). Based on our previous studies [21] and homologous protein distribution analysis (Supplementary Figure S3), Arabidopsis thaliana was used to assemble the protein- protein interaction (PPI) network of A. ilicifolius and A. mollis. The top-ten hub proteins were identiﬁed with a degree score of CytoHubba and displayed in Figure 6. The hub proteins of A. ilicifolius tissues were mostly associated with carbon and energy metabolism Int. J. Mol. Sci. 2021, 22, 1055 5 of 16 (Figure 6A,B), whereas those of the A. mollis tissues were mostly associated with photosyn- thesis and photorespiration and the TCA cycle (Figure 6C,D). Figure 3. Two-dimensional (2-DE) analysis of proteins extracted from (A) A. ilicifolius leaves, (B) A. ilicifolius roots, (C) A. mollis leaves, and (D) A. ilicifolius roots. The numbers correspond with the spot ID, mentioned in Supplementary Tables S2–S5. The isoelectric point (pI) and molecular weight (MW) in kilodaltons are indicated on the top and left of the gel, respectively. CK and SF represent the control group and soil tidal ﬂooding stress, respectively. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 5 of 17 Figure 3. Two-dimensional (2-DE) analysis of proteins extracted from (A) A. ilicifolius leaves, (B) A. ilicifolius roots, (C) A. mollis leaves, and (D) A. ilicifolius roots. The numbers correspond with the spot ID, mentioned in Supplementary Table S2–S5. The isoelectric point (pI) and molecular weight (MW) in kilodaltons are indicated on the top and left of the gel, respectively. CK and SF represent the control group and soil tidal flooding stress, respectively. Int. J. Mol. Sci. 2021, 22, 1055 6 of 16 Figure 4. Principal component analysis (PCA) of total proteome data for the tissues of A. ilicifolius and A. mollis. Percentage variance for each principal component is given. Figure 5. Functional classiﬁcation analysis of the differentially expressed proteins (DEPs) of A. ilicifolius and A. mollis. The detailed information for each spot is shown in Supplementary Tables S2–S5. 2.6. Tidal Flooding Stress Inﬂuences the Energy Status Level of A. ilicifolius and A. mollis The further comparison demonstrated that A. mollis had a higher adenosine monophos- phate (AMP) content, adenosine diphosphate (ADP) content, and ATP content than A. ilicifolius in the control group (Supplementary Figure S4). A. ilicifolius promptly re- sponded to ﬂooding stress by signiﬁcantly increasing ADP and ATP contents in the leaves (Figure 7B,C). However, AMP and ATP contents were signiﬁcantly decreased in A. mollis roots (Figure 7B,C) under tidal ﬂooding stress. The energy charge represents the energy status of biological cells [24]. Whereas the energy charge of A. ilicifolius roots was signif- icantly increased under tidal ﬂooding stress, it was signiﬁcantly decreased in A. mollis tissues (Figure 7D). Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 6 of 17 Figure 4. Principal component analysis (PCA) of total proteome data for the tissues of A. ilicifolius and A. mollis. Percentage variance for each principal component is given. 2.4. Functional Classification of DEPs More proteins were up-regulated in A. ilicifolius than in A. mollis tissues (Figure 5, Supplemental Table S7). A higher percentage of up-regulated proteins were found in car-bon and energy metabolism and amino acid and protein metabolism in A. mollis leaves, while in transcription and signal transduction in A. ilicifolius leaves. In addition, A. ilicifo-lius leaves had a lower percentage of up-regulated proteins associated with stress and defense. Overall, A. ilicifolius tissues had a high percentage of up-regulated proteins and a low percentage of down-regulated proteins associated with carbon and energy metabo-lism. Meanwhile, a higher percentage of down-regulated proteins of A. mollis leaves were associated with carbon and energy metabolism, stress and defense, and transcription and signal transduction (Figure 5, Supplemental Table S7). Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 6 of 17 Figure 4. Principal component analysis (PCA) of total proteome data for the tissues of A. ilicifolius and A. mollis. Percentage variance for each principal component is given. 2.4. Functional Classification of DEPs More proteins were up-regulated in A. ilicifolius than in A. mollis tissues (Figure 5, Supplemental Table S7). A higher percentage of up-regulated proteins were found in car-bon and energy metabolism and amino acid and protein metabolism in A. mollis leaves, while in transcription and signal transduction in A. ilicifolius leaves. In addition, A. ilicifo-lius leaves had a lower percentage of up-regulated proteins associated with stress and defense. Overall, A. ilicifolius tissues had a high percentage of up-regulated proteins and a low percentage of down-regulated proteins associated with carbon and energy metabo-lism. Meanwhile, a higher percentage of down-regulated proteins of A. mollis leaves were associated with carbon and energy metabolism, stress and defense, and transcription and signal transduction (Figure 5, Supplemental Table S7). Int. J. Mol. Sci. 2021, 22, 1055 7 of 16 Figure 6. Top 10 hub proteins in network of (A) A. ilicifolius leaves, (B) A. ilicifolius roots, (C) A. mollis leaves, and (D) A. mollis roots ranked by Matthews correlation coefﬁcient (MCC) method. R represents the root tissue. Underlined numbers represent A. mollis tissues. Figure 7. Effects of tidal ﬂooding stress on (A) adenosine monophosphate content, (B) adenosine diphosphate content, (C) adenosine triphosphate content, and (D) energy charge of A. ilicifolius and A. mollis. * and ** indicate signiﬁcant difference at the 0.05 level and the 0.01 level, respectively. CK and SF represent the control group and soil tidal ﬂooding stress, respectively. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 7 of 17 Figure 5. Functional classification analysis of the differentially expressed proteins (DEPs) of A. ilicifolius and A. mollis. The detailed information for each spot is shown in Supplementary Tables S2–S5. Compared with A. mollis roots, a higher percentage of up-regulated proteins of A. ilicifolius roots were associated with carbon and energy metabolism, amino acid and pro-tein metabolism, stress and defense, and transcription and signal transduction. In A. mollis roots, there was a higher percentage of down-regulated proteins in all pathways (Figure 5, Supplemental Table S7). 2.5. Identification of Hub Proteins in Acanthus Species Because of the lack of genome information on A. ilicifolius and A. mollis, the DEPs of the two species were annotated based on the existing non-redundant protein sequence database (NR). Based on our previous studies [21] and homologous protein distribution analysis (Supplementary Figure S3), Arabidopsis thaliana was used to assemble the protein-protein interaction (PPI) network of A. ilicifolius and A. mollis. The top-ten hub proteins were identified with a degree score of CytoHubba and displayed in Figure 6. The hub proteins of A. ilicifolius tissues were mostly associated with carbon and energy metabolism (Figure 6A,B), whereas those of the A. mollis tissues were mostly associated with photo-synthesis and photorespiration and the TCA cycle (Figure 6C,D). Figure 6. Top 10 hub proteins in network of (A) A. ilicifolius leaves, (B) A. ilicifolius roots, (C) A. mollis leaves, and (D) A. mollis roots ranked by Matthews correlation coefficient (MCC) method. R represents the root tissue. Underlined numbers represent A. mollis tissues. 2.6. Tidal Flooding Stress Influences the Energy Status Level of A. ilicifolius and A. mollis The further comparison demonstrated that A. mollis had a higher adenosine mono-phosphate (AMP) content, adenosine diphosphate (ADP) content, and ATP content than A. ilicifolius in the control group (Supplementary Figure S4). A. ilicifolius promptly re-sponded to flooding stress by significantly increasing ADP and ATP contents in the leaves (Figure 7B,C). However, AMP and ATP contents were significantly decreased in A. mollis roots (Figure 7B,C) under tidal flooding stress. The energy charge represents the energy status of biological cells [24]. Whereas the energy charge of A. ilicifolius roots was signifi-cantly increased under tidal flooding stress, it was significantly decreased in A. mollis tis-sues (Figure 7D). Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 8 of 17 Figure 7. Effects of tidal flooding stress on (A) adenosine monophosphate content, (B) adenosine diphosphate content, (C) adenosine triphosphate content, and (D) energy charge of A. ilicifolius and A. mollis. * and ** indicate significant difference at the 0.05 level and the 0.01 level, respec-tively. CK and SF represent the control group and soil tidal flooding stress, respectively. 2.7. Tidal Flooding Stress Influences the Total Soluble Sugar and Starch Contents of Acanthus Species There was no significant change in the content of total soluble sugar and starch of A. ilicifolius tissues under tidal flooding stress (Figure 8A,B). Nevertheless, there was signif-icant tidal flooding tolerance in the ratio of soluble sugar to starch in A. ilicifolius tissues Int. J. Mol. Sci. 2021, 22, 1055 8 of 16 2.7. Tidal Flooding Stress Influences the Total Soluble Sugar and Starch Contents of Acanthus Species There was no signiﬁcant change in the content of total soluble sugar and starch of A. ilicifolius tissues under tidal ﬂooding stress (Figure 8A,B). Nevertheless, there was signiﬁcant tidal ﬂooding tolerance in the ratio of soluble sugar to starch in A. ilicifolius tissues (Figure 8C). The total soluble sugar content, starch content, and the ratio of soluble sugar to starch were lower in A. mollis tissues than in the control group, except for the ratio of soluble sugar to starch in the leaves (Figure 8A–C). Figure 8. The concentration of (A) soluble sugar, (B) starch and (C) the ratio of soluble sugar to starch for A. ilicifolius and A. mollis under tenth day tidal ﬂooding stress. * and ** indicate signiﬁcant difference at the 0.05 level and the 0.01 level, respectively. CK and SF represent control and soil tidal ﬂooding stress, respectively. 3. Discussion 3.1. Differences in Tissue Tolerance between Acanthus Species Unlike previous ﬁndings in the mangrove, A. marina, seedlings [25], the upper and lower epidermises of A. ilicifolius showed no change with prolonged waterlogging duration in the present study (Figure 2(B2,B3)). The leaf anatomical features of A. mollis were also relatively less susceptible to tidal ﬂooding stress within a short span. The leaf anatomy plays an important role in determining photosynthetic capacity. Herbaceous plants with high photosynthetic capacity usually have thinner epidermis, leading to high values of Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 9 of 17 (Figure 8C). The total soluble sugar content, starch content, and the ratio of soluble sugar to starch were lower in A. mollis tissues than in the control group, except for the ratio of soluble sugar to starch in the leaves (Figure 8A–C). Figure 8. The concentration of (A) soluble sugar, (B) starch and (C) the ratio of soluble sugar to starch for A. ilicifolius and A. mollis under tenth day tidal flooding stress. * and ** indicate signifi-cant difference at the 0.05 level and the 0.01 level, respectively. CK and SF represent control and soil tidal flooding stress, respectively. 3. Discussion 3.1. Differences in Tissue Tolerance between Acanthus Species Unlike previous findings in the mangrove, A. marina, seedlings [25], the upper and lower epidermises of A. ilicifolius showed no change with prolonged waterlogging dura-tion in the present study (Figure 2(B2,B3)). The leaf anatomical features of A. mollis were also relatively less susceptible to tidal flooding stress within a short span. The leaf anat-omy plays an important role in determining photosynthetic capacity. Herbaceous plants with high photosynthetic capacity usually have thinner epidermis, leading to high values of mesophyll conductance [26]. The mangrove leaf exhibited a range of xeromorphic fea-tures, including thick epidermis and wax coatings [25]. Therefore, like other mangrove plants, A. ilicifolius leaves are likely to regulate the Calvin cycle to resist the tidal flooding Int. J. Mol. Sci. 2021, 22, 1055 9 of 16 mesophyll conductance [26]. The mangrove leaf exhibited a range of xeromorphic features, including thick epidermis and wax coatings [25]. Therefore, like other mangrove plants, A. ilicifolius leaves are likely to regulate the Calvin cycle to resist the tidal ﬂooding stress (Supplementary Table S2). A comparative analysis showed that palisade and spongy tissue that were loosely arranged with large spaces and epidermis were thinner in A. mollis leaf, making CO2 entry easier. A. ilicifolius, mainly distributed in the foreshore seaward region, was found to de- velop a high percentage of schizogenous aerenchyma to facilitate efﬁcient internal oxygen transfer [27]. According to previous study, the mangrove species appeared to higher waterlogging tolerance when the aerenchyma formation was induced [28]. The aortic root anatomy of A. ilicifolius was not affected by the tidal ﬂooding stress. The special anatomical structure of roots was not the main reason for A. ilicifolius to tolerance tidal ﬂooding stress at the early stage. Water and minerals transport from the root system to the aerial portions via the xylem tissue. The phloem translocates photosynthetic products from mature leaves to roots and redistributes water and various compounds throughout the plant body [29]. In A. mollis, the aortic root anatomy exhibited broken xylem, phloem, and periderm tissues, indicating a negative inﬂuence on the allocation and partitioning of photosynthetic products. 3.2. Effect of Tidal Flooding on the Photosynthesis of Acanthus Species The proportion of photosynthesis-related proteins within the total DEPs of A. ilicifolius leaves was close to that of A. mollis leaves (Figure 5). The abundance of oxygen-evolving enhancer protein (OEE, spot 18) showed an increasing trend in A. ilicifolius leaves under tidal ﬂooding stress. Oxygen-evolving complex (OEC) proteins are degraded and release OEE as a degradation product to promote the plant to adapt to the adverse conditions [30]. OEE is a subunit of the OEC of photosystem II in the chloroplast [31] considered to be directly involved in photosynthesis. It is suggested that decreased OEE abundance (spot 20, 33) might negatively affect A. mollis leaves. Most of chlorophyll a/b-binding proteins increased in A. ilicifolius leaves (spot 12, 13, 58) but decreased in A. mollis leaves (spot 43) under tidal ﬂooding stress. However, electron transport chain proteins, such as ferredoxin- nicotinamide adenine dinucleotide phosphate (NADP) reductase (spot 52), cytochrome b6-f complex iron-sulfur subunit 1 (spot 63), and photosystem I reaction center subunit IV b (spot 64), increased in A. mollis leaves, promoting photosynthetic electron transport under tidal ﬂooding stress [31]. One fraction of the captured light energy is used to reduce NADP+ to reduced nicoti- namide adenine dinucleotide phosphate (NADPH) and the other fraction is used for light-dependent ATP synthesis. The proteomic data showed that chloroplast ATP synthesis was decreased in A. mollis leaves (spot 23, 29, 30, 31). ATP-dependent zinc metalloprotease FTSH2 (FtsH2, spot 41, 42), which is involved in the turnover of the ΦPSII reaction center D1 protein [32], was increased in maize to protect chloroplast photosynthesis under heat stress [33]. Herein, increased FtsH2 abundance had a positive effect on tidal ﬂooding tolerance of A. ilicifolius leaves. 3.3. Effect of Tidal Flooding on Carbon and Energy Metabolism of Acanthus Species Plants change their energy metabolism pathways to meet the increased demands for survival [34]. CO2 ﬁxation is performed through the Calvin cycle to drive sugar production and energy storage in plants [33]. The activation state of RuBisCO, a key enzyme in the Calvin cycle, is regulated by RuBisCO activase [35]. RuBisCO activase increased in both A. ilicifolius (spot 5 and 19) and A. mollis (spot 14, 15, 21, 28, 35, and 36). The abundance of RuBisCO large subunits increased in A. ilicifolius (spots 7, 9, and 15) but decreased in A. mollis (spots 3, 5, 22, 25, and 40). A similar result was observed in Trifolium species, the waterlogging-sensitive species exhibited a higher reduction of RuBisCO large subunits expression [34]. Int. J. Mol. Sci. 2021, 22, 1055 10 of 16 A higher soluble sugar concentration (Figure 8A) in ﬂooded A. ilicifolius plant is not solely due to photosynthesis but also the conversion of carbohydrates from starch to sugar (Figure 8C). Fructokinase (spot R32) regulates starch synthesis coordinately with sucrose synthase and plays a key role in starch accumulation in tomato fruit [36], which exhibited a three-fold increase in A. ilicifolius roots under tidal ﬂooding treatment. Decreased alpha- galactosidase (spot R7), beta-galactosidase (spot R26), beta-glucosidase (spot R10), and UDP-glucose 4-epimerase family protein (spot R35) abundance had adverse effects on A. mollis roots under tidal ﬂooding stress, indicating the inhibition of polysaccharide catabolism and the interconversion of hexoses (glucose/galactose) in A. mollis roots under tidal ﬂooding stress [37–39]. Glycolysis and the TCA cycle mainly provide energy for plant growth and devel- opment [40]. Pyruvate produced via the glycolytic pathway into the TCA cycle and the consequent electrons is transferred along an electron transport chain and then return to the mitochondrial matrix via ATP synthase [41]. Similar to the proteomic data of a previous study [21], increased glycolysis and TCA cycle-related protein abundances contributed to the defense system of A. ilicifolius tissues tidal ﬂooding stress. Meanwhile, increased V-ATPase abundance was helpful to maintain the cytosolic pH homeostasis and provide free energy to establish a proton motive force across membranes in A. ilicifolius tissues [42]. Mitochondrial ATP synthase was decreased in A. ilicifolius leaves (spot 24, 32) but increased in A. ilicifolius roots (spot R16, R22, R26, R29), indicating that the requirement of ATP decreased accordingly. This implied that high levels of ATP synthase abundance might not be needed in A. ilicifolius leaves. In A. mollis, the abundance of glycolysis and TCA cycle-related proteins mostly showed an increase in leaves and a decrease in the roots under tidal ﬂooding stress. Due to the low efﬁciency of energy conservation under fermentation, an increased rate of glycolysis is required to sustain ATP production necessary for cell sur- vival [43]. The abundance of malate dehydrogenase (ADH; spot R36, R44, R45) increased with a concomitant reduction in glycolysis and TCA cycle-related proteins (spot R28, R37), indicating the low efﬁciency of energy conservation in A. mollis roots under tidal ﬂooding stress [44]. Physiological data also showed the ATP level and energy charge of A. mollis roots became too low to sustain the basal metabolic requirement of roots (Figure 7C,D). 3.4. Effect of Tidal Flooding on Nutrient Assimilation and Protein Metabolism of Acanthus Species Sugar metabolism provides sufﬁcient energy to amino acid metabolism and interme- diates from glycolysis can be utilized as precursors for the synthesis of amino acids [45]. Increased abundances of glutamine synthetase (spot 53), glutamate-ammonia ligase family protein (spot 66, R34, R40), glycine dehydrogenase (spot 75, 76), MTA/SAH nucleosidase (spot 37), and S-adenosyl methionine synthetase (spot R33) were considered to contribute to the production of amino acids in the cell of A. ilicifolius tissues [45–47]. In A. mollis tis- sues, nitrogen metabolism-related enzymes, such as glutamine synthetase (spot 10, 44) and arginase (spot 56), showed an increase. However, decreased sulfate adenylyl transferase (spot 48), cysteine synthase (spot 34) and adenosyl homocysteinase (spot 50) abundances indicated adverse effects on producing cysteine from sulfate in A. mollis leaves [12,48]. Plant tissue exposure to abiotic stress induces protein damage in cells. Increased proteasome subunit alpha type (spot 17) and proteasome subunit beta type (spot 59) abun- dances were observed in A. ilicifolius to break down damaged proteins under tidal ﬂooding stress [49]. Furthermore, heat shock proteins (spots 10, 11, 26, 28, 29, and R13), RNA- binding protein NOB1 (spot 47), protein disulﬁde isomerase (spot 27), which participate in plant ﬁtness, the biogenesis of 40S ribosomal subunits and the formation of disulﬁde bonds during protein folding [50,51], mostly increased in A. ilicifolius tissues under tidal ﬂooding stress. In A. mollis, the abundance of proteasome subunit alpha type (spot 11, R14) and polyubiquitin (spot R13) were decreased, but 26S protease regulatory subunit 6A homolog (spot R15) and 20S proteasome alpha subunit (spot R25) increased during the ﬂooding stage. In addition, protein synthesis essentially requires ribosomes, which play a distinct role in Int. J. Mol. Sci. 2021, 22, 1055 11 of 16 all living cells [52]. The tidal ﬂooding stress decreased 60S ribosomal-related proteins (spot R1, R38) abundance, suggesting inhibition of protein synthesis in A. mollis roots. 14-3-3 protein, which takes part in the regulation of carbon and nitrogen metabolism [53], showed an increase in A. ilicifolius (spot 3, 14, R2, and R3) but a decrease in A. mollis (spot 9, R4, and R8), indicating tidal ﬂooding stress damage of the signal pathway of carbon and nitrogen metabolism in A. mollis tissues. 3.5. Effect of Tidal Flooding on Antioxidative Defense System of Acanthus Species The antioxidative defense system has a stronger reactive oxygen scavenging capacity to mitigate oxidative damage under ﬂooding stress [54]. Our data suggested that most of the antioxidant enzyme-related proteins, including l-ascorbate peroxidase (spot 36, R17), thioredoxin (spot R15, R31), monodehydroascorbate reductase (spot R20), and catalase (spot R36), were increased in A. ilicifolius under tidal ﬂooding stress. The ascorbate-glutathione cycle together with 2-Cys peroxiredoxin are relevant systems in detoxifying reactive oxygen in stressed plants [55]. It has been reported that most of the enzymes (spot 2, R7, 24, and 49) involved in this process decrease in A. mollis. In addition, annexin genes have been reported to have peroxidase activity [56]; it has been hypothesized that elevated abundance of annexin D5-like (spot 42) together with glutathione S-transferase (spot 60) modulate endogenous ROS levels in A. mollis leaves under tidal ﬂooding stress. Moreover, glutathione S-transferase in protein regulation via S-glutathionylation, as a post-translational modiﬁcation, have been reported in plants [57]. 4. Materials and Methods 4.1. Plant Material and Experimental Setup Experiment materials were obtained from vegetative propagation. The stems of A. ilicifolius (9–12 mm in diameter and 10–20 cm in length) were collected from the Zini mangrove forest (117◦91(cid:48) E, 24◦45(cid:48) N), south of the Jiulong River Estuary, Fujian Province, China. The roots of A. mollis were collected from a mother plant that was planted in the greenhouse for one year. The explants of A. ilicifolius and A. mollis were placed in pots (19 cm in diameter and 20 cm in depth) with soil plus vermiculite in a ratio of 3:1. The growth of cuttings is shown in Supplementary Figure S1. Hoagland nutrient solution of 1/8 strength with rooting hormone powder was used to promote the growth of stems and adventitious roots. In the ﬁrst two months, the cuttings were grown under controlled conditions: Temperature (28 ± 2 ◦C), weak light, and relative humidity (60 ± 5%). The pots were then transferred to a new condition with 1000 µmol·m−2·s−1 light intensity, 12 h light period day−1, and 28 ± 2 ◦C temperature. After growth for ﬁve months, uniform and healthy plants were selected for further analysis. The plants were randomly divided into two groups. The soil water content of the control group was kept at 65 ± 5% and regulated by the oven drying method. The pots were placed inside 50 L plastic containers maintaining a 1–2 cm water layer above the soil surface (Supplementary Figure S2) and treated with ﬂooding stress for 6 h per day. All the pots perforated at the bottom to ensure proper drainage. The dry weight of the leaves and roots samples were taken every two days. The extraction of protein and RNA, parafﬁn sectioning, and the measurement of energy AMP, ADP, ATP, and hydrogen peroxide (H2O2) content were performed on the tenth day. 4.2. Determination of the Dry Weight Five plants were randomly selected, washed with distilled water, and divided into two parts (leaves and roots). The oven-dried (at 70 ◦C for 72 h) leaves and roots samples were measured for dry weight (DW) in grams (g) using an electric weight balance. Relative plant DW = Wt/Wo × 100%, where Wt is the dry weight under tidal ﬂooding treatment and Wo is the dry weight under control conditions [58]. Each measurement was repeated three times with ﬁve replications per treatment. Int. J. Mol. Sci. 2021, 22, 1055 12 of 16 4.3. Anatomical Features of Leaves and Roots The leaf center and mature root were collected from A. ilicifolius and A. mollis for parafﬁn sectioning. The fresh tissues were ﬁxed in formalin-acetic acid-alcohol (FAA) solution for 72 h, followed by 6 h in an ethanol series (50–100%) to being embedded in parafﬁn, sectioned, and stained with 1% aqueous safranin and 0.5% fast green. Tissues sections were photographed under a light microscope (Leica DM4 P, Germany) to determine anatomical parameters. 4.4. Protein Extraction and Quantiﬁcation Protein extraction was according to the method described by He and Wang [59] with some modiﬁcations. The 2–4 g of treated tissue from A. ilicifolius and A. mollis were ground in liquid nitrogen and extracted using the tricarboxylic acid (TCA)-acetone/phenol- methanol combined extraction method. The tissue powder was transferred to a centrifuge tube and precipitated by adding cold acetone solution containing 0.2% dithiothreitol (DTT), then the supernatant was discarded and the pellet was suspended in the 2× extraction buffer (20 mmol·L−1 Tris-HCl (pH 8.0), 250 mmol·L−1 sucrose, 10 mmol·L−1 ethylene glycol tetraacetic acid, 1 mmol·L−1 phenylmethylsulfonyl ﬂuoride, 1% Triton X-100, 2% β-mercaptoethanol) at 4 ◦C for 15 min. The homogenate was adding an equal volume of saturated phenol (pH 7.5) to obtain the upper phenol phase, then mixed with three volumes of ice-cold methanol (containing 0.1 mol·L−1 ammonium acetate) overnight at −20 ◦C. Then the pellet was washed with ammonium acetate/methanol (0.1 mol·L−1) and acetone (containing 0.2% DTT). Protein quantiﬁcation was performed using the Bradford [60] method with bicinchoninic acid as the standard. Each treatment was performed for three biological replications for quantitative analysis. An immobiline dry strip gel (17 cm, pH 4–7; Bio-Rad Laboratories, Inc., Hercules, CA, USA) and 12.5% polyacrylamide gels were used to separate the prepared samples in the ﬁrst and second dimension, respectively. The sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels were stained with Coomassie Brilliant Blue R-250 and scanned with Uniscan M3600 (China) at 600 dpi. Gel alignment, spots detection and quantiﬁcation were done using PDQuest software (Version 8.0, Bio-Rad). The DEPs were obtained by pairwise comparison with a fold change ≥ 2.0 and a Student’s t-test (p < 0.05). In-gel tryptic digestion and protein identiﬁcation were performed according to the method of Liu et al. [61]. 4.5. Determinations of AMP, ADP, ATP, and Sugar Content The method of ATP, ADP, and AMP extraction was according to Chen et al. [62] with some modiﬁcations. The powder was obtained from 2 g tissue and then homogenized with 10 mL perchloric acid (0.6 mol·L−1) at 4 ◦C for 30 min, then centrifuged at 6000 rpm for 20 min. The resulting supernatant (6 mL) was quickly neutralized to pH 6.5–6.8 with 1 mol·L−1 potassium hydroxide solution and passed through a 0.22 µm syringe ﬁlter. The supernatant was diluted to 10 mL before measuring. Shimadzu® LC-20A Prominence high-performance liquid chromatography (HPLC) equipped with Syncronis C18 column (4.6 mm × 250 mm, Thermo Fisher Scientiﬁc) was used to measure ATP, ADP, and AMP contents. The mobile phase was 0.1% phosphoric acid and the ﬂow rate was 0.8 mL·min−1. The ultraviolet detection wavelength was 254 nm. The energy charge (EC) was calculated using the following formula: EC = ([ATP] + 1/2 [ADP])/([ATP] + [ADP] + [AMP]) [62]. Data were expressed as means of the ﬁve replicates. The starch and total soluble sugar contents were measured with the starch content kit and plant soluble sugar content test kit, respectively (Nanjing Jian Cheng Institute, Nanjing, China). 4.6. Data Analysis The experimental data were evaluated with IBM SPSS Statistics for Mac (Version 23.0). Principal component analysis (PCA) was performed using OriginPro 2021 (OriginLab Corp, Int. J. Mol. Sci. 2021, 22, 1055 13 of 16 Northampton, MA, USA). The classiﬁcation of identiﬁed proteins was performed using the UniProt Knowledgebase (http://www.uniprot.org) and the NCBI (https://www.ncbi.nlm. nih.gov). The PPI network analysis was acquired using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) version 11.0 (https://string-db.org). Cytoscape software of the PPI network can visualize signiﬁcant protein-protein associations [63]. The Cytoscape plugin (cytoHubba) was utilized to evaluate hub proteins from the PPI network by Matthews correlation coefﬁcient (MCC) method [63]. We used the degree score to identify hub proteins. 5. Conclusions Physiological and proteomic analyses have greatly enriched the current knowledge of ﬂooding resistance in Acanthus species. A. ilicifolius performed better under tidal ﬂooding stress, which was reﬂected in the integrity of the morphological structure, a high level of energy charge, and an increase in the ratio of soluble sugar to starch. A higher percentage of up-regulated proteins associated with carbon and energy metabolism were found in A. ilicifolius tissues under tidal ﬂooding stress. However, the change in the root structure was not responsible for adaption to ﬂooding conditions at the early stage and the maintenance of physiological homeostasis had higher demands for essential supply of energy. A. mollis leaves remained structurally intact even after tidal ﬂooding stress, which might be due to partially enhanced ROS scavenging capacity and carbon and energy metabolism. The disruption of energy provision and ﬂux balance in A. mollis roots demonstrates that maintenance of an energy balance under abiotic stress is critical for cell survival. As shown in Figure 9, we propose a working model to illustrate the detailed mechanism of A. ilicifolius and A. mollis under tidal ﬂooding stress. Figure 9. A schematic representation of A. ilicifolius and A. mollis response and adaptation strategies to tidal ﬂooding stress. Red arrows, up-regulated; blue arrows, down-regulated; pink, the energy- producing pathway and storage substance; orange, biological metabolism that requires energy; green, the pathways in chloroplast. Pink circle, the transport of energy and sugar from leaves to roots; cyan circle, the transport of H2O2. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 14 of 17 4.6. Data Analysis The experimental data were evaluated with IBM SPSS Statistics for Mac (Version 23.0). Principal component analysis (PCA) was performed using OriginPro 2021 (OriginLab Corp, Northampton, MA, USA). The classification of identified proteins was performed using the UniProt Knowledgebase (http://www.uniprot.org) and the NCBI (https://www.ncbi.nlm.nih.gov). The PPI network analysis was acquired using the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) version 11.0 (https://string-db.org). Cytoscape software of the PPI network can visualize significant protein-protein associations [63]. The Cytoscape plugin (cytoHubba) was utilized to evaluate hub proteins from the PPI network by Matthews correlation coefficient (MCC) method [63]. We used the degree score to identify hub proteins. 5. Conclusions Physiological and proteomic analyses have greatly enriched the current knowledge of flooding resistance in Acanthus species. A. ilicifolius performed better under tidal flood-ing stress, which was reflected in the integrity of the morphological structure, a high level of energy charge, and an increase in the ratio of soluble sugar to starch. A higher percent-age of up-regulated proteins associated with carbon and energy metabolism were found in A. ilicifolius tissues under tidal flooding stress. However, the change in the root struc-ture was not responsible for adaption to flooding conditions at the early stage and the maintenance of physiological homeostasis had higher demands for essential supply of energy. A. mollis leaves remained structurally intact even after tidal flooding stress, which might be due to partially enhanced ROS scavenging capacity and carbon and energy me-tabolism. The disruption of energy provision and flux balance in A. mollis roots demon-strates that maintenance of an energy balance under abiotic stress is critical for cell sur-vival. As shown in Figure 9, we propose a working model to illustrate the detailed mech-anism of A. ilicifolius and A. mollis under tidal flooding stress. Figure 9. A schematic representation of A. ilicifolius and A. mollis response and adaptation strate-gies to tidal flooding stress. Red arrows, up-regulated; blue arrows, down-regulated; pink, the energy-producing pathway and storage substance; orange, biological metabolism that requires energy; green, the pathways in chloroplast. Pink circle, the transport of energy and sugar from leaves to roots; cyan circle, the transport of H2O2. Supplementary Materials: Supplementary Materials can be found at www.mdpi.com/xxx/s1. Int. J. Mol. Sci. 2021, 22, 1055 14 of 16 Supplementary Materials: Supplementary Materials can be found at https://www.mdpi.com/1422 -0067/22/3/1055/s1. Author Contributions: Conceptualization, methodology, software, validation, formal analysis, in- vestigation, resources, data curation, original draft preparation, writing—review and editing: Y.-l.L.; project administration and funding acquisition: H.-l.Z. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Key Research and Development Program of China (2017YFC0506102), and the Natural Science Foundation of China (NSFC 31570586 and 31870581). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. 2. 3. 4. 5. 6. 7. 8. 9. Arbona, V.; Hossain, Z.; López-Climent, M.F.; Pérez-Clemente, R.M.; Gómez-Cadenas, A. Antioxidant enzymatic activity is linked to waterlogging stress tolerance in citrus. Physiol. Plant. 2008, 132, 452–466. [CrossRef] [PubMed] Vandoorne, B.; Descamps, C.; Mathieu, A.S.; Van den Ende, W.; Vergauwen, R.; Javaux, M.; Lutts, S. Long term intermittent ﬂooding stress affects plant growth and inulin synthesis of Cichorium intybus (var. sativum). Plant Soil 2014, 376, 291–305. [CrossRef] García-Sánchez, F.; Syvertsen, J.P.; Gimeno, V.; Botía, P.; Perez-Perez, J.G. Responses to ﬂooding and drought stress by two citrus rootstock seedlings with different water-use efﬁciency. Physiol. Plant 2007, 130, 532–542. [CrossRef] Gibbs, J.; Greenway, H. Mechanisms of anoxia tolerance in plants. I. Growth, survival and anaerobic catabolism. Funct. Plant Biol. 2003, 30, 1–47. [CrossRef] Kotula, L.; Clode, P.L.; Striker, G.G.; Pedersen, O.; Läuchli, A.; Shabala, S.; Colmer, T.D. Oxygen deﬁciency and salinity affect cell-speciﬁc ion concentrations in adventitious roots of barley (H ordeum vulgare). New Phytol. 2015, 208, 1114–1125. [CrossRef] Azahar, I.; Ghosh, S.; Adhikari, A.; Adhikari, S.; Roy, D.; Shaw, A.K.; Singh, K.; Hossain, Z. Comparative analysis of maize root sRNA transcriptome unveils the regulatory roles of miRNAs in submergence stress response mechanism. Environ. Exp. Bot. 2020, 171, 103924. [CrossRef] Pedersen, O.; Rich, S.M.; Colmer, T.D. Surviving ﬂoods: Leaf gas ﬁlms improve O2 and CO2 exchange, root aeration, and growth of completely submerged rice. Plant J. 2009, 58, 147–156. [CrossRef] Cheng, H.; Wang, Y.S.; Fei, J.; Jiang, Z.Y.; Ye, Z.H. Differences in root aeration, iron plaque formation and waterlogging tolerance in six mangroves along a continues tidal gradient. Ecotoxicology 2015, 24, 1659–1667. [CrossRef] Garcia, N.; Da-Silva, C.J.; Cocco, K.L.T.; Pomagualli, D.; de Oliveira, F.K.; da Silva, J.V.L.; de Oliveira, A.C.B.; do Amarante, L. Waterlogging tolerance of ﬁve soybean genotypes through different physiological and biochemical mechanisms. Environ. Exp. Bot. 2020, 172, 103975. [CrossRef] 10. Van Dongen, J.T.; Licausi, F. Oxygen Sensing and Signaling. Annu. Rev. Plant Biol. 2015, 66, 345–367. [CrossRef] 11. Luo, F.L.; Nagel, K.A.; Zeng, B.; Schurr, U.; Matsubara, S. Photosynthetic acclimation is important for post-submergence recovery of photosynthesis and growth in two riparian species. Ann. Bot. 2009, 104, 1435–1444. [CrossRef] [PubMed] 12. Ploschuk, R.A.; Grimoldi, A.A.; Ploschuk, E.L.; Striker, G.G. Growth during recovery evidences the waterlogging tolerance of forage grasses. Crop Pasture Sci. 2017, 68, 574–582. [CrossRef] 13. Li, A.M.; Wei, C.X.; Jiang, J.J.; Zhang, Y.T.; Snowdon, R.J.; Wang, Y.P. Phenotypic variation in progenies from somatic hybrids between Brassica napus and Sinapis alba. Euphytica 2009, 170, 289–296. [CrossRef] 14. Hansen, L.N.; Earle, E.D. Somatic hybrids between Brassica oleracea L. and Sinapis alba L. with resistance to Alternaria brassicae (Berk.) Sacc. Theor. Appl. Genet 1997, 94, 1078–1085. [CrossRef] 15. Yang, Y.C.; Yang, S.H.; Li, J.F.; Deng, Y.F.; Zhang, Z.; Xu, S.H.; Guo, W.X.; Zhong, C.R.; Zhou, R.C.; Shi, S.H. Transcriptome analysis of the holly mangrove Acanthus ilicifolius and its terrestrial relative, Acanthus leucostachyus, provides insights into adaptation to intertidal zones. BMC Genom. 2015, 16, 605. [CrossRef] Shackira, A.M.; Puthur, J.T. Cd2+ inﬂuences metabolism and elemental distribution in roots of Acanthus ilicifolius L. Int. J. Phytoremediat. 2019, 21, 866–877. 16. 17. Zhang, L.E.; Liao, B.W.; Guan, W. Effects of simulated tide inundation on seed germination and seedling growth of mangrove 18. species Acanthus ilicifolius. Chin. J. Ecol. 2011, 30, 2165–2172. ˇRezanka, T.; ˇRezanka, P.; Sigler, K. Glycosides of arylnaphthalene lignans from Acanthus mollis having axial chirality. Phytochemistry 2009, 70, 1049–1054. [CrossRef] 19. Matos, P.; Figueirinha, A.; Ferreira, I.; Cruz, M.T.; Batista, M.T. Acanthus mollis L. leaves as source of anti-inﬂammatory and antioxidant phytoconstituents. Nat. Prod. Res. 2019, 33, 1824–1827. [CrossRef] Int. J. Mol. Sci. 2021, 22, 1055 15 of 16 20. Bader, A.; Martini, F.; Schinella, G.R.; Rios, J.L.; Prieto, J.M. Modulation of cox-1, 5-, 12- and 15-lox by popular herbal remedies used in southern Italy against psoriasis and other skin diseases. Phytother. Res. 2015, 29, 108–113. [CrossRef] 21. Li, H.; Li, Z.; Shen, Z.J.; Luo, M.R.; Liu, Y.L.; Wei, M.Y.; Wang, W.H.; Qin, Y.Y.; Gao, C.H.; Li, K.K.; et al. Physiological and proteomic responses of mangrove plant Avicennia marina seedlings to simulated periodical inundation. Plant Soil 2020, 450, 231–254. [CrossRef] 22. Campilho, A.; Nieminen, K.; Ragni, L. The development of the periderm: The ﬁnal frontier between a plant and its environment. Curr. Opin. Plant Biol. 2020, 53, 10–14. [CrossRef] [PubMed] 23. Yi, T.; Zhu, L.; Peng, W.L.; He, X.C.; Chen, H.L.; Li, J.; Yu, T.; Liang, Z.T.; Zhao, Z.Z.; Chen, H.B. Comparison of ten major constituents in seven types of processed tea using HPLC-DAD-MS followed by principal component and hierarchical cluster analysis. LWT Food Sci. Technol. 2015, 62, 194–201. [CrossRef] 24. Chen, Y.R.; Chen, C.L.; Zhang, L.W.; Green-Church, K.B.; Zweier, J.L. Superoxide generation from mitochondrial NADH dehydrogenase induces self-inactivation with speciﬁc protein radical formation. J. Biol. Chem. 2005, 280, 37339–37348. [CrossRef] 25. Xiao, Y.; Jie, Z.L.; Wang, M.; Lin, G.H.; Wang, W.Q. Leaf and stem anatomical responses to periodical waterlogging in simulated tidal ﬂoods in mangrove Avicennia marina seedlings. Aquat. Bot. 2009, 91, 231–237. [CrossRef] 26. Hassiotou, F.; Ludwig, M.; Renton, M.; Veneklaas, E.J.; Evans, J.R. Inﬂuence of leaf dry mass per area, CO2, and irradiance on mesophyll conductance in sclerophylls. J. Exp. Bot. 2009, 60, 2303–2314. [CrossRef] 27. Pi, N.; Tam, N.F.Y.; Wu, Y.; Wong, M.H. Root anatomy and spatial pattern of radial oxygen loss of eight true mangrove species. Aquat. Bot. 2009, 90, 222–230. [CrossRef] 28. Cheng, H.; Wu, M.L.; Li, C.D.; Sun, F.L.; Sun, C.C.; Wang, Y.S. Dynamics of radial oxygen loss in mangroves subjected to waterlogging. Ecotoxicology 2020, 29, 684–690. [CrossRef] 29. McGarry, R.C.; Kragler, F. Phloem-mobile signals affecting ﬂowers: Applications for crop breeding. Trends Plant Sci. 2013, 18, 198–206. [CrossRef] 30. Downton, W.J.S.; Loveys, B.R.; Grant, W.J.R. Non-uniform stomatal closure induced by water stress causes putative non-stomatal 31. inhibition of photosynthesis. New Phytol. 2006, 110, 503–509. [CrossRef] James, H.E.; Bartling, D.; Musgrove, J.E.; Kirwin, P.M.; Herrmann, R.G.; Robinson, C. Transport of proteins into chloroplasts. Import and maturation of precursors to the 33-, 23-, and 16-kDa proteins of the photosynthetic oxygen-evolving complex. J. Biol. Chem. 1989, 264, 19573–19576. [CrossRef] 32. Kato, Y.; Miura, E.; Ido, K.; Ifuku, K.; Sakamoto, W. The variegated mutants lacking chloroplastic FtsHs are defective in D1 degradation and accumulate reactive oxygen species. Plant Physiol. 2009, 151, 1790–1801. [CrossRef] [PubMed] 33. Zhao, F.Y.; Zhang, D.Y.; Zhao, Y.L.; Wang, W.; Yang, H.; Tai, F.J.; Li, C.H.; Hu, X.L. The difference of physiological and proteomic changes in maize leaves adaptation to drought, heat, and combined both Stresses. Front Plant Sci. 2016, 7, 1471. [CrossRef] [PubMed] Stoychev, V.; Simovastoilova, L.; Vaseva, I.; Kostadinova, A.; Nenkova, R.; Feller, U.; Demirevska, K. Protein changes and proteolytic degradation in red and white clover plants subjected to waterlogging. Acta Physiol. Plant 2013, 35, 1925–1932. [CrossRef] 34. 35. Takahashi, H.; Takahara, K.; Hashida, S.; Hirabayashi, T.; Fujimori, T.; Kawai-Yamada, M.; Yamaya, T.; Yanagisawa, S.; Uchimiya, H. Pleiotropic modulation of carbon and nitrogen metabolism in Arabidopsis plants overexpressing the NAD kinase2 gene. Plant Physiol. 2009, 151, 100–113. [CrossRef] 36. Odanaka, S.; Bennett, A.B.; Kanayama, Y. Distinct physiological roles of fructokinase isozymes revealed by gene-speciﬁc suppression of Frk1 and Frk2 expression in tomato. Plant Physiol. 2002, 129, 1119–1126. [CrossRef] [PubMed] 37. Eda, M.; Ishimaru, M.; Tada, T.; Sakamoto, T.; Kotake, T.; Tsumuraya, Y.; Mort, A.J.; Gross, K.T. Enzymatic activity and substrate speciﬁcity of the recombinant tomato β-galactosidase 1. J. Plant Physiol. 2014, 171, 1454–1460. [CrossRef] 38. Dai, N.; Petreikov, M.; Portnoy, V.; Katzir, N.; Pharr, D.M.; Schaffer, A.A. Cloning and expression analysis of a UDP- Galactose/Glucose pyrophosphorylase from melon fruit provides evidence for the major metabolic pathway of galactose metabolism in rafﬁnose oligosaccharide metabolizing plants. Plant Physiol. 2006, 142, 294–304. [CrossRef] Ishiyama, N.; Creuzenet, C.; Lam, J.S.; Berghuis, A.M. Crystal Structure of WbpP, a genuine UDP-N-acetylglucosamine 4- epimerase from Pseudomonas aeruginosa Substrate speciﬁcity in UDP- hexose 4-epimerases. J. Biol. Chem. 2004, 279, 22635–22642. [CrossRef] 39. 40. Rocha, M.; Licausi, F.; Araujo, W.L.; Nunes-Nesi, A.; Sodek, L.; Fernie, A.R.; van Dongen, J.T. Glycolysis and the tricarboxylic acid cycle are linked by alanine aminotransferase during hypoxia induced by waterlogging of Lotus japonicus. Plant Physiol. 2010, 152, 1501–1513. [CrossRef] 41. Andre, C.; Froehlich, J.E.; Moll, M.R.; Benning, C.A. Heteromeric plastidic pyruvate kinase complex involved in seed oil biosynthesis in Arabidopsis. Plant Cell 2007, 19, 2006–2022. [CrossRef] [PubMed] 42. Arino, J.; Ramos, J.; Sychrova, H. Alkali metal cation transport and homeostasis in yeasts. Microbiol. Mol. Biol. Rev. 2010, 74, 95–120. [CrossRef] [PubMed] 43. Zabalza, A.; van Dongen, J.T.; Froehlich, A.; Oliver, S.N.; Faix, B.; Gupta, K.J.; Schmälzlin, E.; Igal, M.; Orcaray, L.; Royuela, M.; et al. Regulation of respiration and fermentation to control the plant internal oxygen concentration. Plant Physiol. 2009, 149, 1087–1098. [CrossRef] [PubMed] Int. J. Mol. Sci. 2021, 22, 1055 16 of 16 44. Saika, H.; Matsumura, H.; Takano, T.; Tsutsumi, N.; Nakazono, M. A point mutation of Adh1 gene is involved in the repression of coleoptile elongation under submergence in rice. Breed. Sci. 2006, 56, 69–74. [CrossRef] 45. Wang, L.; Fu, J.L.; Li, M.; Fragner, L.; Weckwerth, W.; Yang, P.F. Metabolomic and proteomic proﬁles reveal the dynamics of primary metabolism during seed development of Lotus (Nelumbo nucifera). Front. Plant Sci. 2016, 7, 750. [CrossRef] [PubMed] 46. Li, L.L.; Zhao, J.Y.; Zhao, Y.N.; Lu, X.; Zhou, Z.H.; Zhao, C.X.; Xu, G.W. Comprehensive investigation of tobacco leaves during natural early senescence via multi-platform metabolomics analyses. Sci. Rep. 2016, 6, 37976. [CrossRef] 47. Tedder, M.E.; Nie, Z.; Margosiak, S.; Chu, S.; Feher, V.A.; Almassy, R.; Appelt, K.; Yager, K.M. Structure-based design, synthesis, and antimicrobial activity of purine derived SAH/MTA nucleosidase inhibitors. Bioorg. Med. Chem. Lett. 2004, 14, 3165–3168. [CrossRef] 48. Vauclare, P.; Suter, M.; Sticher, L.; Ballmoos, P.V.; Krähenbühl, U.; Camp, R.O.D.; Brunold, C. Flux control of sulphate assimilation in Arabidopsis thaliana: Adenosine 5(cid:48)-phosphosulphate reductase is more susceptible than ATP sulphurylase to negative control by thiols. Plant J. 2010, 31, 729–740. [CrossRef] 49. Kim, M.H.; Jeon, J.; Lee, S.; Lee, J.H.; Gao, L.; Lee, B.H.; Park, J.M.; Kim, Y.J.; Kwak, J.M. Proteasome subunit RPT2a promotes 50. PTGS through repressing RNA quality control in Arabidopsis. Nat. Plants 2019, 5, 1273–1282. [CrossRef] Schramm, F.; Ganguli, A.; Kiehlmann, E.; Englich, G.; Walch, D.; von Koskull-Döring, P. The heat stress transcription factor HsfA2 serves as a regulatory ampliﬁer of a subset of genes in the heat stress response in Arabidopsis. Plant Mol. Biol. 2006, 60, 759–772. [CrossRef] 51. Gruber, C.W.; ˇCemažar, M.; Heras, B.; Martin, J.L.; Craik, D.J. Protein disulﬁde isomerase: The structure of oxidative folding. Trends Biochem. Sci. 2006, 31, 455–464. [CrossRef] [PubMed] 52. Boni, I.V.; Isaeva, D.M.; Musychenko, M.L.; Tzareva, N.V. Ribosome-messenger recognition: mRNA target sites for ribosomal protein S1. Nucleic Acids Res. 1991, 19, 155–162. [CrossRef] [PubMed] 53. Roberts, M.R.; Salinas, J.; Collinge, D.B. 14-3-3 proteins and the response to abiotic and biotic stress. Plant Mol. Biol. 2002, 50, 54. 55. 1031–1039. [CrossRef] [PubMed] Ji, X.L.; Gai, Y.P.; Zheng, C.C.; Mu, Z.M. Comparative proteomic analysis provides new insights into mulberry dwarf responses in mulberry (Morus alba L.). Proteomics 2010, 9, 5328–5339. [CrossRef] [PubMed] Sun, Y.K.; Jang, H.H.; Lee, J.R.; Sung, N.R.; Lee, H.B.; Lee, D.H.; Park, D.J.; Kang, C.H.; Chung, W.S.; Lim, C.O.; et al. Oligomerization and chaperone activity of a plant 2-Cys peroxiredoxin in response to oxidative stress. Plant Sci. 2009, 177, 227–232. 56. Góngora-Castillo, E.; Ibarra-Laclette, E.; Trejo-Saavedra, D.L.; Rivera-Bustamante, R.F. Transcriptome analysis of symptomatic and recovered leaves of geminivirus-infected pepper (Capsicum annuum). Virol. J. 2012, 9, 295. [CrossRef] [PubMed] 57. Dixon, D.P.; Skipsey, M.; Edwards, R. Roles for glutathione transferases in plant secondary metabolism. Phytochemistry 2010, 71, 338–350. [CrossRef] [PubMed] 58. Bora, L.S.; Thomaz, S.M.; Padial, A.A. Evidence of rapid evolution of an invasive poaceae in response to salinity. Aquat. Sci. 2020, 82, 76. [CrossRef] 59. He, C.F.; Wang, Y.M. Protein extraction from leaves of Aloe vera L., a succulent and recalcitrant plant for proteomic analysis. Plant Mol. Biol. Rep. 2008, 26, 292–300. [CrossRef] 60. Bradford, M.M.; Bradford, M. A rapid and sensitive method for the quantiﬁcation of microgram quantities of proteins utilizing the principle-dye binding. Anal. Biochem. 1976, 72, 248–254. [CrossRef] 61. Liu, Y.L.; Shen, Z.J.; Simon, M.; Li, H.; Ma, D.N.; Zhu, X.Y.; Zheng, H.L. Comparative proteomic analysis reveals the regulatory effects of H2S on salt tolerance of mangrove plant Kandelia obovata. Int. J. Mol. Sci. 2020, 21, 118. [CrossRef] [PubMed] 62. Chen, Y.; Lin, H.T.; Jiang, Y.M.; Zhang, S.; Lin, Y.F.; Wang, Z.H. Phomopsis longanae Chi-induced pericarp browning and disease development of harvested longan fruit in association with energy status. Postharvest Biol. Technol. 2014, 93, 24–28. [CrossRef] 63. Chin, C.H.; Chen, S.H.; Wu, H.H.; Ho, C.W.; Ko, M.T.; Lin, C.Y. CytoHubba: Identifying hub objects and sub-networks from complex interactome. BMC Syst. Biol. 2014, 8, S11. [CrossRef] [PubMed]
10.3390_ijerph18052223	Article Promoting Self-Care in Nursing Encounters with Persons Affected by Long-Term Conditions—A Proposed Model to Guide Clinical Care Carina Hellqvist 1,2 1 Department of Neurology, University Hospital Linköping, 58185 Linköping, Sweden; [email protected] 2 Department of Health, Medicine and Caring Sciences, Linköping University, 58183 Linköping, Sweden Abstract: Background: Nursing interventions for persons affected by long-term conditions should focus on providing support to enhance the ability to manage disease in everyday life. Many clinical nurses feel they have inadequate training or experience to provide self-management support in a beneﬁcial and structured way. This study explores the process towards independent self-care and management of disease in persons affected by Parkinson’s disease and the support required from healthcare to achieve this. It presents a nursing model to guide nurses in providing self-management support in the clinical care encounter. Methods: The results from three previously published articles investigating a self-management support program for persons with Parkinson’s disease were combined to form a new data set, and analyzed using qualitative thematic analysis. Results: Three separate, but interrelated, themes were identiﬁed, which described the process towards self- management of disease as expressed by the participants of the self-management program. Themes describe the factors important for developing and improving self-management abilities and actions. The results were applied to Orem’s Self-care deﬁcit theory to suggest a model of self-management support in the clinical nursing encounter. Conclusion: This study investigated factors important for self-management and highlighted the unique contribution and focus of nursing support to promote independent self-care. Keywords: self-care; self-management; nursing support; long-term condition; Parkinson’s disease Citation: Hellqvist, C. Promoting Self-Care in Nursing Encounters with Persons Affected by Long-Term Conditions—A Proposed Model to Guide Clinical Care. Int. J. Environ. Res. Public Health 2021, 18, 2223. https://doi.org/10.3390/ ijerph18052223 Academic Editors: Tiny Jaarsma and Anna E. Strömberg 1. Introduction Received: 18 January 2021 Accepted: 17 February 2021 Published: 24 February 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Long-term conditions, sometimes referred to as non-communicable diseases, largely affect the resources available in healthcare and are responsible for approximately 70% of all deaths worldwide [1]. A longer life expectancy is rapidly changing population demograph- ics leading to a larger number of persons above the age of 60 years. With increasing age, the likelihood of being affected by a long-term condition is also higher. These conditions often require major adjustments and adaptations in life for the person affected, and re- current and regular monitoring and contacts with healthcare services [2]. Persons with long-term conditions are encouraged and expected to be active participants in their own care and well-being. Skills and strategies to handle the impact of disease in everyday life are vital for persons to undertake adequate self-care actions, and thereby, maintain a sense of control, to manage symptoms and maintain satisfaction in their lives. For nurses working in outpatient settings, the provision of self-management support to persons with long-term conditions is considered a major part of professional care [3]. Nursing care is considered holistic in nature, focusing on the entire life-situation and wellbeing of the person, not only the medical aspects of disease [4]. Nurses are recognized as having the appropriate medical knowledge, and are often easily available for persons seeking guidance and support on how to handle the impact of disease on everyday life. These pre-requisites place nurses in the optimal position to provide self-management support. Nevertheless, many nurses feel Int. J. Environ. Res. Public Health 2021, 18, 2223. https://doi.org/10.3390/ijerph18052223 https://www.mdpi.com/journal/ijerph International Journal ofEnvironmental Researchand Public Health(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Environ. Res. Public Health 2021, 18, 2223 2 of 15 uncertain about how to provide self-management support and do not feel that they have received sufﬁcient education and training in skills needed to provide this [3,5]. Strengthening patients’ abilities for self-care actions is the main goal of nursing care according to Dorothea Orem´s Self-care deﬁcit theory. This general and well-known theory of nursing is based on the assumption that all nursing interventions should be aimed at strengthening patients’ own ability to perform self-management activities to manage their own self-care “in order to maintain life, health and well-being” [6]. Orem acknowledges self-care to be a broad concept concerning all aspects of life and not just management of disease, although nursing support primarily focuses on support of self-care deﬁcits due to the impact of disease or ill health. The self-care deﬁcit theory is partly concerned with the more philosophical underlying assumptions of nursing, but above all, it is an action-theory intended to be used in clinical work. The theory has clear speciﬁcations for the nurse and patient roles in order to enhance self-care [6,7]. Parkinson´s disease (PD) is an example of a long-term condition primarily present in the older population. PS is a slowly progressing neurological condition diagnosed in approximately 1% of persons above the age of 60 years, meaning that about 6.1 million persons in the world are living with this long-term condition [8]. In PD the loss of dopamin- ergic neurons in the brain lead to symptoms affecting movement and tremors as well as psychiatric symptoms like anxiety and depression. Although good medical treatment is available to manage symptoms, it will not cure or halt disease progression [9]. Cognitive decline and dementia are common in the later stages of disease. Persons with PD will live for decades with slowly progressing symptoms increasingly affecting everyday life for the person, as well as for close family members and spouses. Persons with PD are primarily cared for in outpatient clinics and supported by multidisciplinary teams to monitor symp- toms, adjust treatment, and provide practical and emotional support [10]. Admissions to hospitals are rare. Specialized nurses, often called Parkinson’s Disease Nurse Specialists (PDNS), work in outpatient care to support persons with PD in many countries across the world. Guidelines describing the PDNS work have been developed in several countries, and the PDNS function is considered an important part of providing high-quality care for this group of patients [11–13]. Nevertheless, even with the regular support of a multidisci- plinary team of specialized healthcare professionals, the management of medication and symptoms in everyday life are the concerns of persons with PD and close family members. To be able to live a good and fulﬁlling life in spite of PD, both the persons affected, as well as their care partners need good skills in managing and monitoring the disease and they need the ability to perform self-care actions [14,15]. Several self-management support programs, speciﬁcally tailored to persons with PD, are provided and tested in clinical care. Most of these programs are designed as small group interventions where participants meet in person for a number of sessions and provided on site at the outpatient clinic. Although the programs have been evaluated scientiﬁcally, many studies are using small samples recruiting participants from only one study-site and included different outcome measures. The contents of the programs also show substantial diversity, some focusing more on provision of medical information, whilst others include one or more strategies for problem-solving, action planning, self-monitoring and behavioural change. A majority of studies also adopted a research design that is not suitable for comparisons and the effectiveness and outcomes. Although self-management support is provided to patients with PD in clinical care, there are still large knowledge gaps concerning what these inter- ventions should include in terms of their contents, the best way of delivery, as well as the expected outcomes for participants and for healthcare [16,17]. In recent years, technological advances have provided new opportunities in clinical care [18]. Technical devices are now used for symptom monitoring, initiation and titration of treatments for and provision of therapies for example cognitive behavioural therapy in Parkinson’s disease [19,20]. In the coming years technology will also be more integrated in the way self-management support is carried out. Technical apps and platforms provide alternative ways of self-management support provision for both group and individual interventions [21,22]. Int. J. Environ. Res. Public Health 2021, 18, 2223 3 of 15 Previous studies addressing PDNS function have mentioned self-management support as one important part of a multi-dimensional role together with several other important tasks [11,12]. This study is the ﬁrst one to explore and describe in depth the contents and interactions present in PDNS self- management support, using a general nursing framework to promote understanding of how this can be performed in the clinical encounter with patient and care partner. The aim of this study was two-fold. Firstly, to explore self-management of disease as it was expressed by persons with Parkinson´s disease and their care partners when participating in the self-management intervention, the Swedish National Parkinson School. A better understanding of what the persons themselves ﬁnd important to be able to manage disease in everyday life can provide insight for nurses into the preferred focus of their nursing interventions and guide how these interventions can be performed in clinical care. The second aim was to apply the ﬁndings to a general nursing theory of self-care, Dorothea Orem’s self-care deﬁcit theory, to elaborate on how the results can be used in clinical care, as well as how the nursing actions to support self-management can be understood and carried out in the clinical encounter. 2. Materials and Methods This study used qualitative and explorative design with a two-step analysis applying both inductive and deductive approaches. The ﬁrst inductive step of analysis in this study was conducted as a secondary analysis of data [23]. Three previously published articles were included [24–26] (Table 1), all of which investigated the outcomes of a dyadic self- management support program: Swedish National Parkinson School (NPS) for persons with PD and their care partners. The results sections of these articles together formed the new set of data analysed in this study. There are only three studies available investigating the NPS. Therefore, sampling included all three published studies. The secondary analysis was performed in order to move focus from merely evaluating the outcomes of the educational intervention and instead aim to learn about the self-management process, and facilitators of and barriers to self-management from the perspective of the participants. The author was involved in the parent studies and is well familiar with the data collection and research methods that was used to achieve the results in these. In the second step of analysis, in the current study, the results from step one were deductively applied to Dorothea Orem´s Self- care Deﬁcit Nursing theory [6]. This was done to interpret the results within a framework that could be used to guide nurses engaging in self-care support in clinical settings. 2.1. Intervention The Swedish National Parkinson School (NPS) is a dyadic self-management inter- vention for persons with PD and their care partners. The program was developed in 2013, and has been provided in clinical practice since 2014 in neurologic and geriatric outpatient clinics across Sweden [27,28]. The focus of the NPS is to handle symptoms of Parkinson´s disease in everyday life, by introducing techniques and strategies for self- monitoring, planning ahead, taking action, positive thinking, communication and resource utilisation. Participants of the NPS meet in a small group once a week for total of seven sessions, each session being two hours long. Every session provides information about a topic relevant to everyday life with Parkinson’s disease followed by a group discussion stimulating interaction and peer-support between the participants. Between the sessions, participants have the opportunity to reﬂect on, and try out, what has been discussed during the NPS in their own everyday lives through home-assignments. The contents of the NPS were inﬂuenced by a previous European project that resulted in a model of patient education speciﬁcally tailored to this group of patients with PD called PEPP [29]. (For more information about the NPS see supplementary material) Int. J. Environ. Res. Public Health 2021, 18, 2223 4 of 15 Study Design Data Collection Analysis I II III Qualitative group discussions. Audio recordings. Inductive qualitative thematic analysis. Data collected Aug. 2015- June 2016 Quantitative quasi- experimental case–control study in clinical care. Self-reported questionnaires before/after NPS intervention or 7 weeks’ standard care. Descriptive statistical analysis within and between groups. Data collected Jan 2015- April 2017. Qualitative observational study with follow-up interviews. Audio recordings and ﬁeldnotes. Inductive constant comparative analysis. Data collected April 2016-Jan 2018. Table 1. Summary of design and participants for studies included in analysis. Number of Participants (PwPD/CP) 42 (25/17) Study Location Five outpatient clinics in different parts of Sweden. (3 county and two university hospitals) Age Range in Years (Median) 68–73 (71) Gender (n) Male/Female PwPD: 11/14 CP: 9/8 Years Since Diagnosis (Median) 3–7 years (4.5) Married or Living Together n (%) Educational Level (n) 41 (98%) Total: 147 (92/55) Intervention Group: 78 (48/30) Control Group: 69 (44/25) Five outpatient clinics in different parts of Sweden. (3 county and two university hospitals) Intervention Group: 65–75 (71) 19/29 2–7 years (5) 52 (88%) CP 68–77 (72) Control Group: PwPD 64–75 (68) CP 67–74 (69) 19/11 —– 34 (97%) 30/14 3–8 years (7) 41 (85%) 7/18 —– 28 (97%) 13 (10/3) One Swedish university hospital 68–79 (75.5) PwPD: 5/5 CP: 1/2 2–7 years (4.5) 10 (77%) PS: 13 HS: 12 UD: 17 PS: 15 HS: 19 UD:25 PS: 8 HS: 18 UD: 22 PS: 12 HS: 8 UD: 15 PS: 5 HS: 7 UD: 16 PS: 4 HS: 6 UD:3 PwPD: Persons with PD, CP: Care partners, PS: Primary school, HS: High school, UD: University degree. Study I: Hellqvist C, Dizdar N, Hagell P, Berterö C, Sund-Levander M. Improving self-management for persons with Parkinson’s disease through education focusing on management of daily life: Patients’ and relatives’ experience of the Swedish National Parkinson School. J Clin Nurs. 2018 Oct;27(19–20):3719–3728. doi:10.1111/jocn.14522. Epub 2018 Jul 30. PMID: 29782061. Study II: Hellqvist C, Berterö C, Dizdar N, Sund-Levander M, Hagell P. Self-Management Education for Persons with Parkinson’s Disease and Their Care Partners: A Quasi-Experimental Case-Control Study in Clinical Practice. Parkinsons Dis. 2020 Apr 30;2020:6920943. doi:10.1155/2020/6920943. PMID: 32399171; PMCID: PMC7210533. Study III: Hellqvist C, Berterö C, Hagell P, Dizdar N, Sund-Levander M. Effects of self-management education for persons with Parkinson’s disease and their care partners: A qualitative observational study in clinical care. Nurs Health Sci. 2020 Sep;22(3):741–748. doi:10.1111/nhs.12721. Epub 2020 Apr 28. PMID: 32270898. Int. J. Environ. Res. Public Health 2021, 18, 2223 5 of 15 2.2. Participants and Settings In total, 127 persons with PD and 75 care partners from ﬁve separate outpatient clinics across Sweden were included in the material used for this study. The three studies included used different perspectives and methodologies to explore the outcomes of the NPS. Study I is a qualitative study with data consisting of audio recordings of the last group session of the NPS educational program, analysed with qualitative thematic analysis [30,31]. Study II is a quantitative quasi-experimental case-control study using self-reported questionnaires administered before and after participation in the NPS program or 7 weeks of standard care. Statistic methods comparing results within and between groups were used to analyse data [32] Study III is a qualitative study with data consisting of observations and follow-up interviews of a clinical encounter and analysed with constant comparative analysis [33,34]. (for an overview of methods and participants see Table 1). All studies were conducted with the permission of the regional ethical review board and participants gave written and verbal consent following the declaration of Helsinki [35]. 2.3. Data Analysis 2.3.1. Inductive Secondary Analysis of Data Qualitative thematic analysis, as described by [23], was chosen for compiling and performing secondary analysis of data. The method is not tied to any speciﬁc theory or underlying philosophy, and could be applied to several types of data and data sources. It allows for comparison and analysis of both qualitative and quantitative data together as parts to form a new whole. The method has also previously been used in studies carried out in clinical care settings [31]. An inductive approach to data was used with no previous theory or framework to guide analysis. A realist semantic standpoint was adopted for the analysis to remain close to the data and the actual words and experiences as expressed by the participants in the studies. This was important as the focus of the study was to explore participants’ views. The analysis was performed by the author of this study who is a registered nurse and has a PhD with previous experience of using a qualitative thematic method to analyse data. The author is also working clinically in an outpatient care setting to support persons affected by PD and their care partners. During the ﬁrst step of the analysis all three studies were read several times to get a sense of the whole body of data. Initial thoughts about the contents were written down. After this, each study was read more thoroughly and codes relevant to the research question were extracted. Codes from all studies were then compared and related codes were sorted into broader themes. Initially six themes were identiﬁed but some themes were found to describe the same phenomenon from another angle, and were, thus, incorporated and combined. The ﬁnal result of this ﬁrst step using inductive and qualitative analysis consists of three related but separate themes describing the process and activities of self-management, including the need and wishes for self-management support from the perspective of the participants (see Figure 1) The ﬁnal themes were checked against the initial text to conﬁrm and identify them in the initial text. The 15-step checklist for good quality thematic analysis as provided by Braun and Clarke was followed throughout the analysis to ensure structure and quality [30]. 2.3.2. Deductive Application of Results To explore how the results retrieved in step one could be understood and applied in the clinical practice of nurses working to support and enhance self-management in persons with long-term conditions, the second step of analysis applied the ﬁndings to Dorothea Orem’s Self-care deﬁcit nursing theory [6]. The application was undertaken to evaluate if the theory could be used to guide nurses in their work to provide self-management support in clinical care settings. The three themes obtained in the ﬁrst step of inductive analysis resulted in suggestions for nursing actions appropriate to support patients’ and care partners’ abilities to engage in self-care. These were checked for their ﬁt into the framework of Orem’s Self-care deﬁcit theory. This second part of analysis generated a model describing the nurse/patient/caregiver interaction to support self-care during Int. J. Environ. Res. Public Health 2021, 18, 2223 6 of 15 a clinical encounter and is presented in the results section below. The model obtained through analysis was distributed to other nurses working in clinical care to support persons with PD and care partners to gather their views on the usefulness and applicability in clinical practice. Figure 1. The over-all results of the inductive qualitative thematic analysis. Three interrelated themes describing the building blocks and process towards self-management of PD for persons affected and their care partners as described by participants of the Swedish National Parkinson School. 3. Results 3.1. Inductive Thematic Analysis The ﬁrst inductive step of secondary analysis of the three studies resulted in three distinct but related themes. These themes are describing the process of self-management of disease and the important conditions for enhancing self-management abilities and actions in everyday life, as expressed by persons with PD and their care partners during participation in the Swedish National Parkinson School (NPS). (Figure 1) The contents of each theme is described and supported with quotations from the original studies below. 3.1.1. Theme 1 “A Changed Reality” Persons with PD and care partners both acknowledged the time of diagnosis as the starting-point of a new era in life, as well as the starting-point in the process towards self-management of PD in everyday life. To be informed of a lifelong chronic illness was experienced by many as a shock and brought about emotional reactions like anger, denial and sadness. “When you are ﬁrst diagnosed with the disease you feel very lonely. You have a lot of questions about the future and maybe see the worst scenario in front of you” (Paper III, Participant 1, page 3). Receiving a diagnosis appeared to change the very foundations of life and also affected relationships in the family and between husbands and wives. Persons affected by PD also felt that their perceptions about themselves, i.e., their personal identity, was altered. Persons with PD, as well as the care partners expressed worries about the future, how rapidly the disease would progress and affect mental and physical abilities and how this would impact life ahead. For many, the initial period after the diagnosis was primarily dominated by strong feelings and emotional reactions and a loss of control over life. The ﬁrst attempts Int. J. Environ. Res. Public Health 2020, 17, x 6 of 17 analysis to remain close to the data and the actual words and experiences as expressed by the participants in the studies. This was important as the focus of the study was to explore participants’ views. The analysis was performed by the author of this study who is a reg-istered nurse and has a PhD with previous experience of using a qualitative thematic method to analyse data. The author is also working clinically in an outpatient care setting to support persons affected by PD and their care partners. During the first step of the analysis all three studies were read several times to get a sense of the whole body of data. Initial thoughts about the contents were written down. After this, each study was read more thoroughly and codes relevant to the research question were extracted. Codes from all studies were then compared and related codes were sorted into broader themes. Ini-tially six themes were identified but some themes were found to describe the same phe-nomenon from another angle, and were, thus, incorporated and combined. The final result of this first step using inductive and qualitative analysis consists of three related but sep-arate themes describing the process and activities of self-management, including the need and wishes for self-management support from the perspective of the participants (see Fig-ure 1) The final themes were checked against the initial text to confirm and identify them in the initial text. The 15-step checklist for good quality thematic analysis as provided by Braun and Clarke was followed throughout the analysis to ensure structure and quality [30]. Figure 1: The over-all results of the inductive qualitative thematic analysis. Three interrelated themes describing the building blocks and process towards self-management of PD for persons affected and their care partners as described by participants of the Swedish National Parkinson School. 2.3.2. Deductive Application of Results To explore how the results retrieved in step one could be understood and applied in the clinical practice of nurses working to support and enhance self-management in per-sons with long-term conditions, the second step of analysis applied the findings to Dor-othea Orem’s Self-care deficit nursing theory [6]. The application was undertaken to eval-uate if the theory could be used to guide nurses in their work to provide self-management support in clinical care settings. The three themes obtained in the first step of inductive analysis resulted in suggestions for nursing actions appropriate to support patients’ and care partners’ abilities to engage in self-care. These were checked for their fit into the A changed realityThe companionsFinding a new pathInt. J. Environ. Res. Public Health 2021, 18, 2223 7 of 15 to accept and manage the changed reality included speaking openly to others about the diagnosis. Also upward and downward social comparison was used to relieve stress in their own life. Statements like “there is always someone worse off”, “ it is not a fatal disease “ or “ there are good medical treatments available now to manage symptoms of disease” were common. “I still think that I’ve been fortunate. Because there are others that are much worse off than I am. I don’t have any problems walking. I can walk without aids!” (Paper III, Participant 8, page 2). 3.1.2. Theme 2 “Finding a New Path” After the initial period of strong emotions at diagnosis, most participants found a more balanced way to deal with the new circumstances of life. They started to come to terms with PD as being a part of life and with this new outlook they were planning for life ahead. When participants, both persons with PD and their care partners, reached this mental state of acknowledging the present situation, they were also ready and willing to ﬁnd new ways to deal with the physical, as well as emotional symptoms of disease. Many felt that acquiring new knowledge about the disease itself, including medical treatments available, was one way to feel more in control of the situation. Also thinking difﬁcult situations through, and make action-plans for how to handle them, was helpful. Learning techniques for self-monitoring symptoms was helpful in managing them when they occurred in everyday life and also helped to alert patients to contact healthcare professionals when new symptoms occurred and to communicate the current health status to care providers. These skills were achieved to some extent by the participants in the NPS program. “Results of participation in the NPS could be measured as a shift in how persons view the impact of disease in their everyday lives and are connected to a mind-set of not allowing the disease to control life . . . [This mind-set reﬂects] better knowledge of skills and techniques to manage and cope with the impact of PD.” (Study II, results page 10) Trying to keep a positive mindset and to be active participants in their own lives, prioritizing activities that brought feelings of happiness and joy were important to deal with both the physical and emotional impact of PD in daily life. “I sing in a choir. Singing makes me feel good, even if you felt exhausted before going you feel really good afterward.” (Study III, Participant 6, page 3). Engaging in physical activities was a common way to maintain physical abilities as well as to enhance emotional well-being. Participants also adopted many self-care strategies to deal with the more practical impact of PD; these included, for example, strategies to facilitate dressing, managing personal hygiene and intake of medication. 3.1.3. Theme 3 “The Companions” Adjusting to PD clearly involves more people than the person affected by the disease. The new skills and abilities needed to manage PD were developed in a social context includ- ing interactions with spouses, other members of the family and friends. Also healthcare professionals were important in providing advice and guidance. These people surrounding the person affected by PD can be viewed as “companions” to provide support and help them on the path of self-management. Their actions most often support and encourage self-care but can occasionally hinder self-care abilities. If persons close to them were per- ceived as being supportive and tolerant this would ease the psychological strain of being affected by the disease. The person closest is often a spouse. Many spouses expressed a view that managing PD in everyday life is a shared mission, and not just a concern of the person affected by the disease. Based on this view, spouses would also act as care partners and the approach of a joint concern was found to be beneﬁcial, strengthening the relationship and promoting well-being for both. There were several examples of support from spouses in managing PD in everyday life, which included providing motivation and support in physical activities and activities of daily living, support in following medication regimens and accompanying patients at medical appointments. The amount and type of Int. J. Environ. Res. Public Health 2021, 18, 2223 8 of 15 support provided by the care partner differed according to the progression of the disease. Many care partners expressed a wish for healthcare to teach them methods and strategies to provide further support for their partner in managing several aspects of the disease in order to maintain healthy behaviours and physical and psychological well-being. The NPS provided an opportunity to meet others in the same situation. This was found to be valu- able and it was much appreciated by the participants in discussing with others how to handle situations in life brought on by PD and to give each other advice on how to handle these instances made them feel less isolated and lonely in their life situation. “My feeling is that this group has been a helping hand because it has been very difﬁcult trying to carry it all by myself . . . It’s been a lifesaver . . . I feel that my husband and I have a future to look forward to” (Study I, Group 1 care partner, row 1110–1115). Even though the social surroundings, and support offered by the persons close by and available in everyday life, undoubtedly were most important for persons with PD and care partners, the participants also expressed that the support given by health care professionals was important in guiding them towards self-management. Persons with PD would seek healthcare professionals for general information and support concerning the disease itself including pathophysiology, symptoms, progression and medical treatments. They would also seek advice on how to handle speciﬁc symptoms of the disease, i.e., obstipation or hal- lucinations when they occurred in everyday life as a result of PD. Healthcare professionals were also involved in providing knowledge of strategies to monitor symptoms and to grasp which signs could indicate progression of disease, and to explain important information to communicate in encounters with healthcare to adjust care and medical treatment; they also wanted information about which symptoms or situations should trigger persons with PD, and care partners themselves to actively seek medical help if they occurred in everyday life. A good relationship with an easily accessible health care provider could in itself ease the emotional burden of the disease for persons affected by PD and their care partners. “I feel she cares. She listens, looks up information, shows me how things work and tells me why she changes medications, for example.” (Study III, Participant 4, Page 1). 3.2. Deductive Analysis: Applying the Results on Orem’s Self-Care Deﬁcit Theory The inductive analysis clearly showed that the social context of people close to the person affected by PD was most important for their ability to handle the disease in everyday life; even though participants also described the importance of support of healthcare professionals to gain new knowledge and develop self-care strategies to manage the impact of the disease. Providing self-management support to patients with long-term conditions has been identiﬁed as one of the main tasks of nursing care, the results of the ﬁrst step of the inductive analysis presented above was applied to a general nursing theory focusing on self-care. This was done to bring together theory and practical reality and to serve as a guide for nurses in their work providing self-management support. Dorothea Orem’s grand nursing theory, The Self-care Deﬁcit theory, was found to be suitable as a theoretical basis. The theory clearly speciﬁes the aim of all nursing care is to help and support patients to develop the abilities needed to manage self-care independently in order to promote and preserve life, health and well-being. The deductive application of the results to the Self-care deﬁcit theory produced a model of the interactions in a clinical encounter between nurse, person with PD and care partner presented and described below (Figure 2 Self-management support nursing model). This model acknowledges every encounter between nurse, person with PD and care partner as being a unique encounter. It is a complex encounter between three unique persons. Every encounter will differ from another due to the characteristics that each person brings to the encounter, i.e., personality, previous experiences, emotional state, educational level, ability to relate to other persons, the speciﬁc needs and wishes of the par- ticipants. Nurses’ profession-speciﬁc skills, i.e., factual knowledge of disease/medication, experience in consultation and patient education, ability to understand other person’s emotions—inﬂuence their ability to provide nursing support. Orem calls this nursing Int. J. Environ. Res. Public Health 2021, 18, 2223 9 of 15 agency. The encounter will always have its starting point in the personal narrative of what life is like with PD as it is experienced by the person with PD and the care partner. This per- sonal story is the foundation for a mutual understanding of the situation and a prerequisite for the contents and collaboration between nurse, person with PD and care partners in the encounter. The intra-relational encounter between nurse, patient with PD and care partner is represented by the purple box in the middle of the model. All interactions within the encounter take place here. Although every encounter is unique, there are still some general assumptions that can be applied more broadly to describe the nature and structure of the care encounters. Figure 2. Model of self-management support for persons with long-term conditions in the clinical nursing encounter. According to Orem, an encounter in a health care setting always includes the meeting of three unique persons but in their situation-speciﬁc roles as nurse, patient and care partner. The three persons all have their speciﬁc reasons for entering into the encounter. In the model, this is presented by the green arrows pointing from each person into the encounter. The nurse’s aim is to provide nursing care to support self-management abilities, the patient and care partners seek information and advice on how to manage self-care demands brought on by the disease. All persons need to perform activities of self-care in many areas of life to maintain life and health. This is considered a person’s self-care agency, and it is a learned ability developed gradually during life. When being affected by long-term disease, this brings new demands, self-care requisites, that the person needs to recognize and learn how to manage. For persons affected by a progressing disease like PD, the support of a care partner is of great importance. For many, the self-care activities Int. J. Environ. Res. Public Health 2020, 17, x 10 of 17 Figure 2. Model of self-management support for persons with long-term conditions in the clinical nursing encounter. This model acknowledges every encounter between nurse, person with PD and care partner as being a unique encounter. It is a complex encounter between three unique per-sons. Every encounter will differ from another due to the characteristics that each person brings to the encounter, i.e., personality, previous experiences, emotional state, educa-tional level, ability to relate to other persons, the specific needs and wishes of the partici-pants. Nurses’ profession-specific skills, i.e., factual knowledge of disease/medication, ex-perience in consultation and patient education, ability to understand other person’s emo-tions—influence their ability to provide nursing support. Orem calls this nursing agency. The encounter will always have its starting point in the personal narrative of what life is like with PD as it is experienced by the person with PD and the care partner. This personal story is the foundation for a mutual understanding of the situation and a prerequisite for the contents and collaboration between nurse, person with PD and care partners in the encounter. The intra-relational encounter between nurse, patient with PD and care partner is represented by the purple box in the middle of the model. All interactions within the encounter take place here. Although every encounter is unique, there are still some gen-eral assumptions that can be applied more broadly to describe the nature and structure of the care encounters. Int. J. Environ. Res. Public Health 2021, 18, 2223 10 of 15 of daily life are seen as a joint concern particularly for activities that might be difﬁcult to perform independently with advanced disease. These were often taken over by the care partner. The care partner can compensate and fulﬁl the self-care requisite brought on by PD if the person affected by disease can no longer manage it by themselves. The nurse should consider and evaluate this joint or composite self-care ability in every encounter as it is crucial for many patients. In the model, joint self-care ability is represented by the yellow circle in the right corner. In the encounter between nurse, person with PD and care partner the new requisites brought on by PD are explored and discussed. They include a learning process of how the symptoms of disease can present themselves, as well as learning strategies on how to manage them. The type of intervention to support self-care ability performed for persons with PD primarily takes place in outpatient care and is focused on providing information and strategies on how to handle situations and symptoms occurring in everyday life due to PD. According to the terminology used by Orem, this type of self-management support is called the supporting and educational nursing system. The goal of the encounter is to form a mutual understanding of the situation and together make a plan of action on how to manage the symptoms of disease in everyday life. In the model, this plan of action is represented by the orange circle in the right-hand corner. The plan should be documented by the nurse, and tested by the person affected with PD and care partner in their everyday lives. It is the nurse’s responsibility to evaluate the outcomes of the action plan. If the plan was successful the overall goal of the nursing intervention was met, by providing the support needed for patient and care partners to be independent in performing the self-care actions needed to address the impact of PD. If the plan did not work as intended to resolve the self-care deﬁcit in the situation, nursing intervention is still needed to address the situation. A new encounter between nurse, person with PD and care partner should take place and a new plan of action should be negotiated and again tested in everyday life to address the self-care requisite brought on by PD. The results of the ﬁrst step of inductive analysis, described in the three themes above, will be incorporated as new knowledge to improve the nursing agency. The information about the process of acceptance and factors inﬂuencing self-management described by the participants will guide the nursing actions in knowing what is important to incorporate in the support to persons with PD and care partners in the clinical encounter (see Figure 3). During consultation, the nurse should ﬁrst evaluate where persons with PD and their care partners are in the emotional and psychological process towards accepting diagnosis. If, through the personal narrative being conveyed in the clinical encounter, the nurse understands that the patient and/or care partner is still in the ﬁrst stage of emotionally reacting to receiving notice of PD, the support needs to focus on providing emotional support. The aim is to encourage speaking openly about the diagnosis and the feelings surrounding it, and to acknowledge the strong feelings of sadness and anger and justify that it is okay to express feelings of unfairness and a fear of the future. The nurse should also focus on providing hope that there is a good future to come and that it is possible to live a fulﬁlling life even in the presence of PD. If through the personal narrative the nurse feels that the person with PD and/or care partner has accepted or at least acknowledged the presence of PD in their life and is receptive and looking for ways and suggestions on how to handle their new situation, the support should be directed towards providing strategies to enhance self-management. The nursing consultation can include both situation-speciﬁc strategies, addressing how to deal with certain symptoms occurring due to PD and more general strategies used to enhance their overall ability to manage life with PD. These general strategies can include providing medical knowledge of disease including symptoms, available treatment and effects/side effects of medications. Self-management support should also introduce cogni- tive strategies like planning ahead, making plans of action to handle difﬁcult situations, self-monitoring and registration of health status and symptoms. Int. J. Environ. Res. Public Health 2021, 18, 2223 11 of 15 Figure 3. Nursing interventions connected to the themes. As support from other people close to the person with PD is important, nursing assessment should also include exploring the social network of the patient and caregiver including extended family, close friends, other social contexts, i.e., through leisure activities. Nursing support should always include support also for the care partner if there is one, and special consideration should be given to patients without a care partner or with a small social network. 4. Discussion The aim of this study was two-fold. Firstly, it explored self-management and ap- proaches that can be helpful in the process towards independence in self-management of PD in everyday life as experienced by participants of the patient educational intervention NPS. Secondly, it further investigated how the new ﬁndings could be understood in relation to Orem’s self-care deﬁcit theory and be applied in the clinical encounter between nurse, patient and care partner. Three distinct themes were identiﬁed describing the process towards self-management of PD in everyday life. The ﬁrst theme describes the life altering experience of being diagnosed with a long-term condition and the strong emotional reactions often surrounding diagnosis. This event was considered the starting point of a new chapter in life and also the starting point towards self-management of the disease. This initial phase of strong emotional reactions following diagnosis has also been described in other groups of patients [36,37] and should be recognized by nurses working to support persons with long-term disorders. This phase can often be described in terms of a personal crisis and a shock for both the person receiving diagnosis and the persons close to them. During crisis, the susceptibility to new information is often reduced and support should mainly address the strong emotional reactions [12,37]. With the gradual acceptance of the disease, participants felt a need to acquire new knowledge to be able to handle the new circumstances brought on by PD. The strategies ar- Int. J. Environ. Res. Public Health 2020, 17, x 12 of 17 Figure 3. Nursing interventions connected to the themes. During consultation, the nurse should first evaluate where persons with PD and their care partners are in the emotional and psychological process towards accepting diagnosis. If, through the personal narrative being conveyed in the clinical encounter, the nurse un-derstands that the patient and/or care partner is still in the first stage of emotionally react-ing to receiving notice of PD, the support needs to focus on providing emotional support. The aim is to encourage speaking openly about the diagnosis and the feelings surrounding it, and to acknowledge the strong feelings of sadness and anger and justify that it is okay to express feelings of unfairness and a fear of the future. The nurse should also focus on providing hope that there is a good future to come and that it is possible to live a fulfilling life even in the presence of PD. If through the personal narrative the nurse feels that the person with PD and/or care partner has accepted or at least acknowledged the presence of PD in their life and is re-ceptive and looking for ways and suggestions on how to handle their new situation, the support should be directed towards providing strategies to enhance self-management. The nursing consultation can include both situation-specific strategies, addressing how to deal with certain symptoms occurring due to PD and more general strategies used to en-hance their overall ability to manage life with PD. These general strategies can include providing medical knowledge of disease including symptoms, available treatment and effects/side effects of medications. Self-management support should also introduce cog-nitive strategies like planning ahead, making plans of action to handle difficult situations, self-monitoring and registration of health status and symptoms. As support from other people close to the person with PD is important, nursing as-sessment should also include exploring the social network of the patient and caregiver including extended family, close friends, other social contexts, i.e., through leisure activi-ties. Nursing support should always include support also for the care partner if there is Int. J. Environ. Res. Public Health 2021, 18, 2223 12 of 15 ticulated by persons with PD and their care partners in this study also show a large overlap with the strategies used by persons with other long-term conditions [38,39]. The support of social networks, peers living in the same situation and support from family for persons living with long-term conditions have been described as essential, especially for children and persons with signiﬁcant physical or cognitive symptoms [40,41]. This means that in order to support the person with the disease, nursing support must also include support for care partners and other persons close by. Participants in this study described thinking difﬁcult situations through and making action-plans helpful to manage disease and maintain a sense of control. They also expressed a wish for guidance from healthcare in to grasp which signs could indicate progression of disease. The opportunity to discuss self-management strategies and to discuss the future including possible scenarios ahead with healthcare professionals has previously been identiﬁed as unmet needs by PD patients and care partners [42]. Nursing support for persons with PD and their care partners should provide opportunities to discuss future and offer support in making and evaluating action plans to promote independent self-care. A recent study proposed how this could be achieved in clinical care by introducing the concept of a “road-map” allowing patients and care partners the opportunity to discuss this from their personal wishes and needs [43]. The inductive results presented within the context of the three themes above provide important ﬁrst-hand knowledge of what persons with PD and care partners experience as important factors towards successful management of the disease. This information should be considered and incorporated into practice when providing nursing support for self-management. In a recent review of published qualitative studies investigating self-management components expressed by persons with PD also recognized knowledge and information, self-monitoring strategies, psychological strategies, social engagement, physical exercise and as factors important for self-management [44]. These components were also expressed by persons with PD and care partners in the current study. Nurses working in clinical care to support patients with long-term disorders some- times feel unprepared and lacking in adequate training to provide self-management sup- port [5,45]. This study provides a framework that is based on Orem’s grand nursing theory about self-care to guide nurses working in outpatient care and to enhance their understand- ing of self-management support in the clinical encounter. Recognizing the overall goal of nursing care as helping persons to gain independence in self-care activities brought on by disease can also help nurses to understand their unique role and contribution to patient care. This study is the ﬁrst to investigate, in depth, how the PD nurse’s self-management support, including the content of nursing actions and interactions with patient and care partners are carried out in the clinical encounter. The examples in this study are retrieved from persons affected by PD and their care- partners, but it might be reasonable to believe that there are many similarities in the process towards independent self-management also present in other groups of patients affected by long-term conditions. Future studies are needed to explore this. Future studies are also needed to test the applicability of the nursing self-management support model in clinical care encounters presented in this paper also with other groups of patients. Strengths and Limitations This study is important for nurses working in clinical care as it presents a way of thinking about self-management support and provides a model of self-care support that can be applied in the clinical encounter. The model presented is based on a well-known nursing theory, but is still directly applicable in clinical care. This study can also help nurses to understand better their unique contribution to care for persons with long-term conditions. The data used to explore self-management from the patients’ and care partners’ views used in this study consisted of the three studies exploring a self-management intervention, the Swedish national Parkinson school (NPS). The NPS is a dyadic intervention performed in a small group setting. Although the total number of participants in the three studies Int. J. Environ. Res. Public Health 2021, 18, 2223 13 of 15 included 127 persons with PD and 75 care partners, most of the ﬁndings representing the participants’ voices are derived from the two qualitative studies (study I and III) including merely 35 people with PD and 20 care partners. As study II is using self-reported questionnaires with standardized questions and alternatives for answers, and as it was analysed with quantitative methods, there are very limited ﬁndings of individual voices in this study. However, even if the individual voices were not apparent in this quantitative study, there were important ﬁndings in two domains of one scale showing improved knowledge of self-management strategies, as well as a shift in mind-set to not let the disease control their life, following the NPS self-management support intervention. Qualitative thematic analysis was used as a method of analysing data in this study. This method is suitable and has previously been used to analyse many types of data and allow for synthesis of different types of data sources and data. It has also been found suitable for and used to inform practice in clinical care [31]. Even though the method allow synthesis of both qualitative and quantitative data, it was obvious during the analysis that the qualitative data were heavily affecting the contents of the themes for the reasons explained above. The analysis was performed by the author of this study who has previous experience with using this method. The author also works part-time as a clinical nurse and therefore has a previous understanding of meeting persons with PD and care partners in clinical consultations. This could have inﬂuenced the analysis, for example, in interpreting data and forming themes. The advantage of this pre-understanding is that the author knows what can be of use to other nurses in clinical practice. The model of self-management support presented in this study was presented to other nurses from other hospitals and clinics working to support persons with PD and care partners to collect their views on its value in guiding clinical care and the process. Based on their evaluations, a joint care-plan was made, and evaluating it was further clariﬁed in the model as a result of their input. The limitations and risk of researcher bias presented above should be taken into account when reading the results of this study. Nevertheless, this study is a valuable contribution to the research literature, and the results can be easily accessed by nurses to improve their abilities to provide self-management support to persons with long-term illness and their care partners. The general model of self-management support in the clinical encounter presented in this paper could be applied to patients affected by various types of long-term conditions, and is not restricted to persons with PD. 5. Conclusions Nursing care for individuals affected by long-term conditions was found to be a unique aspect of their care. The overall goal of nursing care is to provide the information and strategies needed to manage the impact of disease independently in everyday life. This paper contributes to understanding how self-management support can be performed in clinical care. The nursing model, presented in this paper, is based upon Dorothea Orem’s theory of nursing, and describes the interactions between nurse, patient and care partner taking part in the interpersonal meeting. The model can guide nursing interventions and serve as a way of thinking about self-management support in the clinical nursing encounter. In this paper, the model is presented using examples of nursing interventions retrieved from data of persons with PD and their care partners, but the model of self-care support in the nursing encounter can also be useful for nurses supporting persons affected by other long-term conditions. Supplementary Materials: The following are available online at https://www.mdpi.com/1660-460 1/18/5/2223/s1, Supporting information 1: Swedish National Parkinson School. Funding: This research received no external funding. Institutional Review Board Statement: The protocol was approved by the Ethics Committee of Linköping, Sweden (Reg. no. 2014/497–31, 2015/458–32, 2016/166–32, 2017/264–32). Int. J. Environ. Res. Public Health 2021, 18, 2223 14 of 15 Informed Consent Statement: All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki. Data Availability Statement: The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy of participants. Conﬂicts of Interest: The author declares no conﬂict of interest. References 1. World Health Organization. Global Status Report on Non Communicable Diseases 2014; World Health Organization: Geneva, Switzerland, 2014; Available online: https://www.who.int/nmh/publications/ncd-status-report-2014/en/ (accessed on 21 January 2018). 2. World Health Organization. Integrated Care for Older Persons. Guidelines on Community-Level Interventions to Manage Declines in Intrinsic Capacity; World Health Organization: Geneva, Switzerland, 2017; Available online: https://apps.who.int/iris/handle/ 10665/258981 (accessed on 21 January 2018). 3. Westland, H.; Schröder, C.D.; de Wit, J.; Frings, J.; Trappenburg, J.C.A.; Schuurmans, M.J. Self-management support in routine primary care by nurses. Br. J. Health Psychol. 2018, 23, 88–107. [CrossRef] 6. 7. 8. 4. McCormack, B.; McCance, T. Person-Centered Nursing: Theory and Practice; Wiley-Blackwell: Chichester, UK, 2010. 5. Duprez, V.; Beeckman, D.; Verhaeghe, S.; Van Hecke, A. Are person-related and socio-structural factors associated with nurses’ self-management support behavior? A correlational study. Patient Educ. Couns. 2018, 101, 276–284. [CrossRef] Orem, D.; Taylor, S.; Renpenning, K.M. Nursing Concepts of Practice, 3rd ed.; Mosby: St. Louis, MO, USA, 2001. Orem, D. Nursing Concepts of Practice, 4th ed.; Mosby: St. Louis, MO, USA, 1991. GBD 2016 Parkinson’s Disease Collaborators. Global, regional, and national burden of Parkinson’s disease 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2018, 17, 939–953. [CrossRef] Kalia, L.V.; Lang, A.E. Parkinson’s disease. Lancet 2015, 386, 896–912. [CrossRef] 9. 10. Radder, D.L.M.; Nonnekes, J.; van Nimwegen, M.; Eggers, C.; Abbruzzese, G.; Alves, G.; Browner, N.; Chaudhuri, K.R.; Ebersbach, G.; Ferreira, J.J.; et al. Recommendations for the Organization of Multidisciplinary Clinical Care Teams in Parkinson’s Disease. J. Parkinson’s Dis. 2020, 10, 1087–1098. [CrossRef] [PubMed] 11. Lennaerts, H.; Groot, M.; Rood, B.; Gilissen, K.; Tulp, H.; van Wensen, E.; Munneke, M.; van Laar, T.; Bloem, R.B. A Guideline for Parkinson’s Disease Nurse Specialists, with Recommendations for Clinical Practice. J. Parkinson’s Dis. 2017, 7, 749–754. [CrossRef] 12. Hellqvist, C.; Berterö, C. Support supplied by Parkinson’s disease specialist nurses to Parkinson’s disease patients and their spouses. Appl. Nurs. Res. 2015, 28, 86–91. [CrossRef] [PubMed] 13. Parkinsons UK. Competencies: A Competency Framework for Nurses Working in Parkinson’s Disease Management; Parkinsons UK: London, UK, 2016; Available online: https://www.rcn.org.uk/professional-development/publications/pub-005584 (accessed on 22 March 2020). 14. Thordardottir, B.; Nilsson, M.; Iwarsson, S.; Haak, M. “You plan, but you never know”—Participation among people with 15. different levels of severity of Parkinson’s disease. Disabil. Rehabil. 2014, 36, 2216–2224. [CrossRef] [PubMed] Sjödahl-Hammarlund, C.; Westergren, A.; Åström, I.; Edberg, A.K.; Hagell, P. The Impact of Living with Parkinson’s Disease: Balancing within a Web of Needs and Demands. Parkinsons Dis. 2018, 2018, 4598651. [CrossRef] [PubMed] 16. Kessler, D.; Liddy, C. Self-management support programs for persons with Parkinson’s disease: An integrative review. Patient Educ. Couns. 2017, 100, 1787–1795. [CrossRef] 17. Tennigkeit, J.; Feige, T.; Haak, M.; Hellqvist, C.; Seven, Ü.S.; Kalbe, E.; Schwarz, J.; Warnecke, T.; Tönges, L.; Eggers, C.; et al. Structured Care and Self-Management Education for Persons with Parkinson’s Disease: Why the First Does Not Go without the Second-Systematic Review, Experiences and Implementation Concepts from Sweden and Germany. J. Clin. Med. 2020, 9, 2787. [CrossRef] [PubMed] 18. Hansen, C.; Sanchez-Ferro, A.; Maetzler, W. How Mobile Health Technology and Electronic Health Records Will Change Care of Patients with Parkinson’s Disease. J. Parkinson’s Dis. 2018, 8 (Suppl. 1), S41–S45. [CrossRef] 19. Kraepelien, M.; Schibbye, R.; Månsson, K.; Sundström, C.; Riggare, S.; Andersson, G.; Lindefors, N.; Svenningsson, P.; Kaldo, V. Individually Tailored Internet-Based Cognitive-Behavioral Therapy for Daily Functioning in Patients with Parkinson’s Disease: A Randomized Controlled Trial. J. Parkinson’s Dis. 2020, 10, 653–664. [CrossRef] 20. Evans, L.; Mohamed, B.; Thomas, E.C. Using telemedicine and wearable technology to establish a virtual clinic for people with 21. Parkinson’s disease. BMJ Open Qual. 2020, 9, e001000. [CrossRef] [PubMed] Isernia, S.; Di Tella, S.; Pagliari, C.; Jonsdottir, J.; Castiglioni, C.; Gindri, P.; Salza, M.; Gramigna, C.; Palumbo, G.; Molteni, F.; et al. Effects of an Innovative Telerehabilitation Intervention for People with Parkinson’s Disease on Quality of Life, Motor, and Non-motor Abilities. Front. Neurol. 2020, 11, 846. [CrossRef] [PubMed] 22. Wannheden, C.; Revenäs, Å. How People with Parkinson’s Disease and Health Care Professionals Wish to Partner in Care Using eHealth: Co-Design Study. J. Med. Internet Res. 2020, 22, e19195. [CrossRef] 23. Ruggiano, N.; Perry, T.E. Conducting secondary analysis of qualitative data: Should we, can we, and how? Qual. Soc. Work 2019, 18, 81–97. [CrossRef] Int. J. Environ. Res. Public Health 2021, 18, 2223 15 of 15 24. Hellqvist, C.; Dizdar, N.; Hagell, P.; Berterö, C.; Sund-Levander, M. Improving self-management for persons with Parkinson’s disease through education focusing on management of daily life: Patients’ and relatives’ experience of the Swedish National Parkinson School. J. Clin. Nurs. 2018, 27, 3719–3728. [CrossRef] 25. Hellqvist, C.; Berterö, C.; Dizdar, N.; Sund-Levander, M.; Hagell, P. Self-Management Education for Persons with Parkinson’s Disease and Their Care Partners: A Quasi-Experimental Case-Control Study in Clinical Practice. Parkinsons Dis. 2020, 2020, 6920943. [CrossRef] 26. Hellqvist, C.; Berterö, C.; Hagell, P.; Dizdar, N.; Sund-Levande, M. Effects of self-management education for persons with Parkinson’s disease and their care partners: A qualitative observational study in clinical care. Nurs. Health Sci. 2020, 22, 741–748. [CrossRef] 27. Carlborg, C. The new Swedish national Parkinson School can make life simpler and more enjoyable. Parkinsonjournalen 2013, 3, 18–20. Available online: https://www.parkinsonforbundet.se/meny2/Information/Parkinsonjournalen. (accessed on 22 January 2020). (In Swedish) 28. Carlborg, C. Pilottesting of the Swedish National Parkinson School—“Meeting others equally important as the knowledge itself”. Parkinsonjournalen 2013, 4, 18–20. Available online: https://www.parkinsonforbundet.se/meny2/Information/ Parkinsonjournalen (accessed on 22 January 2020). (In Swedish) 29. Pasqualini, M.S.; Simons, G. Patient Education for People with Parkinson’s Disease and Their Carers; John Wiley & Sons: Chichester, UK, 2006. 30. Braun, V.; Clarke, V. Using thematic analysis in psychology. Qual. Res. Psychol. 2006, 3, 77–101. [CrossRef] 31. Braun, V.; Clarke, V. What can “thematic analysis” offer health and wellbeing researchers? Int. J. Qual. Stud. Health Well-being 2014, 9, 26152. [CrossRef] [PubMed] Fowler, J.; Jarvis, P.; Chevannes, M. Practical Statistics for Nursing and Health Care; John Wiley & Sons: Chichester, UK, 2013. 32. 33. Glaser, B.G. The constant comparative method of qualitative analysis. Soc. Probl. 1965, 12, 436–445. [CrossRef] 34. Glaser, B.G. Theoretical Sensitivity: Advances in Methodology of Grounded Theory; Sociological Press: Mill Valley, CA, USA, 1978. 35. World Medical Association. WMA Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects; Canary Publications: Guildford, UK, 2013; Available online: http://www.wma.net/en/30publications/10policies/b3/ (accessed on 22 April 2020). 36. Derksen, E.; Vernooij-Dassen, M.; Gillissen, F.; Olde Rikkert, M.; Scheltens, P. Impact of diagnostic disclosure in dementia on patients and carers: Qualitative case series analysis. Aging Ment. Health 2006, 10, 525–531. [CrossRef] [PubMed] 37. Hegge, M.; Dinndorf-Hogenson, G. Elders adapting to a chronic diagnosis within a nursing refuge. Nurs. Forum 2012, 47, 131–137. [CrossRef] 38. McGilton, K.S.; Vellani, S.; Yeung, L.; Chishtie, J.; Commisso, E.; Ploeg, J.; Andrew, M.K.; Ayala, A.P.; Gray, M.; Morgan, D.; et al. Identifying and understanding the health and social care needs of older adults with multiple chronic conditions and their caregivers: A scoping review. BMC Geriatr. 2018, 18, 231. [CrossRef] 39. Rees, S.; Williams, A. Promoting and supporting self-management for adults living in the community with physical chronic illness: A systematic review of the effectiveness and meaningfulness of the patient-practitioner encounter. JBI Libr. Syst. Rev. 2009, 7, 492–582. [CrossRef] 40. Whitehead, L.; Jacob, E.; Towell, A.; Abu-Qamar, M.; Cole-Heath, A. The role of the family in supporting the self-management of chronic conditions: A qualitative systematic review. J. Clin. Nurs. 2018, 27, 22–30. [CrossRef] [PubMed] 41. Bennett, P.N.; Wang, W.; Moore, M.; Nagle, C. Care partner: A concept analysis. Nurs. Outlook 2017, 65, 184–194. [CrossRef] [PubMed] 42. Vlaanderen, F.P.; Rompen, L.; Munneke, M.; Stoffer, M.; Bloem, B.R.; Faber, M.J. The Voice of the Parkinson Customer. J. Parkinson’s 43. Dis. 2019, 9, 197–201. [CrossRef] [PubMed] Jordan, S.R.; Kluger, B.; Ayele, R.; Brungardt, A.; Hall, A.; Jones, J.; Katz, M.; Miyasaki, J.M.; Lum, H.D. Optimizing future planning in Parkinson disease: Suggestions for a comprehensive roadmap from patients and care partners. Ann. Palliat. Med. 2020, 9 (Suppl. 1), S63–S74. [CrossRef] 44. Tuijt, R.; Tan, A.; Armstrong, M.; Pigott, J.; Read, J.; Davies, N.; Walters, K.; Schrag, A. Self-Management Components as Experienced by People with Parkinson’s Disease and Their Carers: A Systematic Review and Synthesis of the Qualitative Literature. Parkinsons Dis. 2020, 2020, 8857385. [CrossRef] 45. Verkaik, R.; van Antwerpen-Hoogenraad, P.; de Veer, A.; Francke, A.; Huis, I.H.V.J. Self-management-support in dementia care: A mixed methods study among nursing staff. Dementia 2017, 16, 1032–1044. [CrossRef] [PubMed]
10.3390_ijms20071662	Article Ameliorative Effect of Spinach on Non-Alcoholic Fatty Liver Disease Induced in Rats by a High-Fat Diet Laura Inés Elvira-Torales 1,2,, Gala Martín-Pozuelo 1, Rocío González-Barrio 1, , Marina Santaella 1, Inmaculada Navarro-González 1, Francisco-José Pallarés 3 Javier García-Alonso 1, Ángel Sevilla 4 and María Jesús Periago-Castón 1, 1 Department of Food Technology, Food Science and Nutrition, Faculty of Veterinary Sciences, Regional Campus of International Excellence “Campus Mare Nostrum”, Biomedical Research Institute of Murcia (IMIB-Arrixaca-UMU), University Clinical Hospital “Virgen de la Arrixaca”, University of Murcia, Espinardo, 30071 Murcia, Spain; [email protected] (G.M.-P.); [email protected] (R.G.-B.); [email protected] (I.N.-G.); [email protected] (M.S.); [email protected] (J.G.-A.) 2 Department of Food Engineering, Tierra Blanca Superior Technological Institute, 95180 Tierra Blanca, Veracruz, Mexico 3 Department of Anatomy and Comparative Pathological Anatomy, Faculty of Veterinary Sciences, Regional Campus of International Excellence “Campus Mare Nostrum”, University of Murcia, Espinardo, 30071 Murcia, Spain; [email protected] 4 Anchormen, Pedro de Medinalaan 11, 1086 XK Amsterdam, the Netherlands; [email protected] * Correspondence: [email protected] (L.I.E.-T.); [email protected] (M.J.P.-C.); Tel: +34-868-884-793 (M.J.P.-C.) Received: 19 February 2019; Accepted: 1 April 2019; Published: 3 April 2019 Abstract: The purpose of this work was to evaluate the effect of dietary carotenoids from spinach on the inﬂammation and oxidative stress biomarkers, liver lipid proﬁle, and liver transcriptomic and metabolomics proﬁles in Sprague–Dawley rats with steatosis induced by a high-fat diet. Two concentrations of spinach powder (2.5 and 5%) were used in two types of diet: high-fat (H) and standard (N). Although rats fed diet H showed an accumulation of fat in hepatocytes, they did not show differences in the values of adiponectin, tumor necrosis factor alpha (TNF-α), and oxygen radical absorption (ORAC) in plasma or of isoprostanes in urine compared with animals fed diet N. The consumption of spinach and the accumulation of α and β carotenes and lutein in the liver was inversely correlated with serum total cholesterol and glucose and the content of hepatic cholesterol, increasing monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA) and reducing cholesterol in the livers of rats fed diet H and spinach. In addition, changes in the expression of genes related to the fatty liver condition occurred, and the expression of genes involved in the metabolism of fatty acids and cholesterol increased, mainly through the overexpression of peroxisome proliferator activated receptors (PPARs). Related to liver metabolites, animals fed with diet H showed hypoaminoacidemia, mainly for the glucogenic aminoacids. Although no changes were observed in inﬂammation and oxidative stress biomarkers, the consumption of spinach modulated the lipid metabolism in liver, which must be taken into consideration during the dietary treatment of steatosis. Keywords: spinach; carotenoids; steatosis; gene expression; metabolomic; lipid metabolism 1. Introduction Non-alcoholic fatty liver disease (NAFLD) is the liver disease which is disseminated most widely around the world due to genetic, dietary, and lifestyle factors [1]. Day et al. [2] suggested a two-stage development of NAFLD: ﬁrstly, an accumulation of triglycerides and free fatty acids in the hepatocytes, Int. J. Mol. Sci. 2019, 20, 1662; doi:10.3390/ijms20071662 www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2019, 20, 1662 2 of 24 and secondly, lipid peroxidation, mitochondrial dysfunction, and liver inﬂammation. These processes result in an increase in fatty acid synthesis and a decline in β-oxidation and in the exportation of triglycerides from the liver as very low-density lipoprotein (VLDL) [3]. For green leafy vegetables, such as spinach, a number of functional properties have been found regarding their nutrients and bioactive compounds, such as antioxidant, anti-inﬂammatory, anti-proliferative, anti-obesity, hypoglycemic, and hypolipidemic activity [4]. Among the commonly consumed leafy vegetables, spinach can be considered a source of bioactive compounds such as phenolic compounds and carotenoids. In relation to phenolic compounds, spinach only contributes approximately 0.8 mg of gallic acid equivalents/day/person, due to the low daily intake [4]. However, spinach is considered one of the richest plant sources of carotenoids, contributing to the intake of lutein, zeaxanthin, and carotene. Carotenoids can contribute positively to liver health [5,6], and their consumption has been associated with decreased fat accumulation in the liver in patients with NAFLD [7], being effective in the prevention and treatment of this liver pathology [6]. However, the mechanisms have not yet been elucidated, and further investigations must be conducted. In animal studies, it has been observed that lycopene reduces fat accumulation and inﬂammation of the liver through the activation of the antioxidant and anti-inﬂammatory response, increased transport of cholesterol and fatty acids, improvement of β-oxidation, and regulation of mRNA translation [8,9]. Lutein (the most abundant carotenoid in spinach) reduced cholesterol, malondialdehyde (MDA), and tumor necrosis factor alpha (TNF-α) levels in the liver of guinea pigs when administered in hypercholesterolemic diets [10], but this effect depends on the form in which lutein is administrated [11]. The supplementation of the diet with astaxanthin could reduce the expression of the peroxisome proliferator-activated receptor gamma (PPARG) and DNA damage inducible-transcript 3 (CHOP-10) genes, thereby diminishing hepatic lipid transport and fatty acid synthesis and avoiding the development of hepatic steatosis [12]. The administration of β-cryptoxanthin signiﬁcantly reduced steatosis in mice by decreasing oxidative stress and the inﬂammatory response by inhibiting the expression of related genes [13]. Although the inﬂuence of several pure carotenoids on liver health has been reported, information about the preventive effect of dietary carotenoids on NAFLD is still rare. Because spinach is a natural source of carotenoids, its utilization in a whole-food intervention approach provides multiple nutrients with a broad range of biological activities, creating the potential for complementary, additive, or synergistic activities that are lacking when supplementation involves only a single nutrient. Moreover, given that the prevention of a disease at an early stage is the principle of dietary treatment, the purpose of this work was to evaluate whether supplementation of the diet with spinach, as a dietary source of carotenoids, has an effect on biomarkers of the steatosis of Sprague–Dawley rats fed a high-fat diet. To achieve this general objective, we have evaluated the changes in the plasmatic parameters, inﬂammation and oxidative stress markers, liver lipid content, and transcriptomic and metabolomic proﬁles of rats after supplementation of their feed with spinach powder. 2. Results 2.1. Feed Composition, Weight Gain and Volume of Feed Consumption Table 1 shows the proximate composition, energy values, total phenolics and carotenoids of the six experimental diets—NC (standard diet), N2.5 (standard diet + 2.5% spinach), N5 (standard diet + 5% spinach), HC (high-fat diet), H2.5 (high-fat diet + 2.5% spinach) and H5 (high-fat diet + 5% spinach)—which were administered to the rats for ﬁve weeks. The feed of H groups provided a higher content of protein and fat, resulting in a mean energetic value of around 450 kcal/100 g. In addition, this diet showed a lower proportion of total dietary ﬁber (TDF) and total phenolic compounds (TPC) than N diets due to the low content of unreﬁned agricultural commodities. In general, the incorporation of spinach in both diets did not led to signiﬁcant differences in the composition of the diets, only regarding carotenoid intake. The administration of 5% spinach in the feed provided a mean content Int. J. Mol. Sci. 2019, 20, 1662 3 of 24 of 9.1 µg of total carotenoids/g of feed, while the supplementation with 2.5% spinach gave 3.1 µg of carotenoids/g of feed. According to the carotenoid proﬁle of spinach, lutein and α-carotene were the predominant carotenoids in the feed, followed by β-carotene, whereas neoxanthin and violaxanthin were not detected in the feed. Table 1. Proximate composition, energy values and total phenolic compounds in the experimental diets 1. Parameters Protein (g/100 g) Fat (g/100 g) Total dietary ﬁber (TDF) (g/100 g) Carbohydrate (g/100 g) Starch (g/100 g) Ash (g/100 g) Energetic value (kcal/100 g) Calories from protein (%) Calories from fat (%) Calories from carbohydrate (%) Total phenolics (TPC) (mg GAE/100 g) NC 14.5 4.0 21.2 55.6 34.4 4.7 316.3 18.3 11.4 70.3 N2.5 15.0 4.0 21.3 55.0 33.7 4.7 316.0 19.0 11.4 69.6 N5 HC 15.5 4.0 22.6 53.2 30.5 4.7 310.8 19.9 11.7 68.4 17.3 21.2 6.9 51.1 44.2 3.5 464.4 14.9 41.1 44.0 H2.5 17.7 20.8 7.4 50.6 43.1 3.5 460.1 15.4 40.6 44.0 188.4 192.2 196.1 20.4 21.2 H5 18.1 20.4 8.0 49.9 41.9 3.6 455.4 15.9 40.3 43.8 22.4 1 Values are expressed as mean. Table 2 shows the consumption of food and water, changes in body and liver weight, feces and urine excretion, and daily carotenoid intake during the intervention period. The initial mean body weights did not exhibit signiﬁcant differences among the six experimental groups. In contrast, at the end of the experimental period, the body weight increases and liver weight differed signiﬁcantly between the animals fed the N diets and those fed the H diet, due to the higher caloric value of the latter diet. In addition, the liver weight was signiﬁcantly lower in H2.5 and H5 than in the H group. Excreted feces values were signiﬁcantly higher in N groups in comparison with H groups, which could be explained by the daily intake of TDF. Despite the higher content of TFD in H2.5 and H5 diets, no differences were observed in the amount of excreted feces with HC. According to the proportion of spinach and the content of carotenoids in the feed, the consumption of these compounds was signiﬁcantly higher (p < 0.05) in the groups that received 5% spinach (55.5 and 53.2 µg/day for N5 and H5, respectively) than in groups N2.5 and H2.5 (Table 2). Differences in the daily intake of total TPC were also observed between the N and H diet. Table 2. Food and drink intake, excreted feces and urine, and carotenoid intake of the six experimental groups in the 5-week intervention period 1. Parameters NC N2.5 N5 HC H2.5 H5 Initial body weight (g) Final body weight (g) Body weight increase (g) Liver weight (g) Food intake (g/day) Water intake (mL/day) Excreted feces (g/day) Excreted urine (mL/day) Carotenoids intake (µg/day) TDF intake (g/day) TPC intake (mg GAE/day) 371.9 ± 33.9 459.4 ± 43.7 87.55 ± 18.85 * 15.32 ± 2.16 * 8.98 ± 3.65 22.50 ± 7.21 b 4.70 ± 1.46 * 12.67 ± 7.63 - 1.90 ± 0.77 * 19.30 ± 4.87 * 380.4 ± 15.2 451.4 ± 22.5 71.02 ± 12.71 * 13.92 ± 1.80 * 6.76 ± 2.27 33.52 ± 9.10 ab 4.28 ± 2.23 15.20 ± 7.51 * 20.63 ± 6.91 b 1.65 ± 0.35 * 14.85 ± 3.13 * 383.3 ± 19.9 442.0 ± 16.6 68.21 ± 9.32 * 13.65 ± 1.90 * 7.21 ± 3.89 36.46 ± 7.58 a 5.32 ± 1.66 * 16.02 ± 6.68 55.52 ± 35.41 a 1.64 ± 0.18 * 14.26 ± 1.56 * 377.8 ± 20.5 500.1 ± 37.3 b 122.28 ± 21.83 ab 25.36 ± 3.06 a 7.23 ± 2.09 29.33 ± 7.77 2.82 ± 0.92 7.64 ± 4.97 - 0.54 ± 0.12 1.45 ± 0.42 372.3 ± 28.2 475.1 ± 34.4 b 102.85 ± 22.09 b 21.44 ± 2.95 b 7.84 ± 2.69 25.42 ± 7.06 3.01 ± 0.85 5.40 ± 3.17 * 23.92 ± 8.20 b 0.46 ± 0.12 1.41 ± 0.41 387.6 ± 5.7 552.5 ± 29.9 a 154.5 ± 29.2 a 19.31 ± 1.37 b 5.85 ± 1.91 26.96 ± 10.15 2.54 ± 1.05 10.59 ± 4.62 53.24 ± 17.33 a 0.42 ± 0.09 1.29 ± 0.42 1 Data are expressed as mean ± SD. a,b Different letters show signiﬁcant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high fat diet (HC, H2.5, H5), after performing a one-way ANOVA. * Signiﬁcant statistical difference (p < 0.05), after carrying out a two-sample t test, between the members of the NC–HC, N2.5–H2.5, and N5–H5 pairings. Int. J. Mol. Sci. 2019, 20, 1662 4 of 24 2.2. Histopathological Examination and Biochemical Parameters Considering the anatomical and pathological examination (Figure 1), the presence of steatosis in rats of the H groups can be observed in both the macroscopic and microscopic images. Macroscopically, the liver was enlarged, yellow, and greasy (pictures not shown). Microscopically, the hepatocytes contained small and large vesicles due to the abnormal accumulation of lipids, particularly triglycerides. The accumulation of fat was conﬁrmed using Sudan III, which stains triglycerides and other intracellular lipid droplets, providing an orange color. According to the number of vacuoles, the steatosis was classiﬁed as grade 3, with 50–75% of the hepatocytes showing vacuolar degeneration. However, after the consumption of spinach (H2.5 and H5 rats), the vacuoles were slightly smaller in comparison with those of animals that had received the high fat diet. Figure 1. Microscopic photographs of liver tissue. Microscopic images with H&E (a–c and g–i) and Sudan III (d–f and j–l) visualized by light microscopy (×40) for the control and experimental groups. Arrows show the vacuolar degeneration of the hepatocyte (V). In addition, the inﬁltration of mononuclear cells and the degeneration and necrosis of hepatocytes were evaluated to determine the inﬂammation level. Only a low grade of inﬂammation was detected (grade 1), with less than 20% of the examined area affected. The steatosis was conﬁrmed by the analysis Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 5 of 25 Figure 1. Microscopic photographs of liver tissue. Microscopic images with H&E (a–c and g–i) and Sudan III (d–f and j–l) visualized by light microscopy (×40) for the control and experimental groups. Arrows show the vacuolar degeneration of the hepatocyte (V). Table 3. Biochemical parameters of plasma, inflammation and oxidative stress biomarkers analyzed in the six experimental groups at the end of the 5-week intervention period 1. Parameters NC N2.5 N5 HC H2.5 H5 Glucose (mg/dL) 199.4 ± 42.7 * 147.1 ± 18 132.17 ± 4.38 274.1 ± 31.1 a 177.9 ± 3.89 b 153.6 ± 20.5 b Proteins (g/dL) 5.42 ± 0.33 * 5.79 ± 0.18 5.67 ± 0.55 6.32 ± 0.26 6.04 ± 0.43 6.25 ± 0.12 Final ALT (U/L) 32.6 ± 5.09 * 28.95 ± 0.35 * 34.40 ± 4.16 * 47.95 ± 9.55 45.43 ± 9.83 44.30 ± 4.85 Final AST (U/L) 75.07 ± 10.14 b* 88.28 ± 10.95 b* 106.1 ± 6.93 a* 151.10 ± 5.20 139.2 ± 43.7 142.7 ± 32.5 Adiponectin (pg/mL) 0.95 ± 0.38 0.62 ± 0.16 1.09 ± 0.84 0.55 ± 0.10 0.69 ± 0.13 0.54 ± 0.13 TNF-α (pg/mL) 15.6 ± 1.13 14.49 ± 1.29 14.09 ± 1.13 14.46 ± 1.55 14.68 ± 1.33 14.16 ± 1.62 ORAC (mmoles equiv trolox/L) 8.29 ± 1.62 9.06 ± 0.66 9.28 ± 1.17 9.79 ± 0.72 9.91 ± 1.48 9.81 ± 0.75 Urine isoprostanes (ng/mg creatinine) 0.95 ± 0.17 1.01 ± 0.18 1.1 ± 0.14 1.13 ± 0.004 0.97 ± 0.26 0.96 ± 0.09 1 Data are expressed as mean ± SD. a,b Different letters show significant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high fat diet (HC, H2.5, H5), after performing a one-way ANOVA. Significant statistical difference (p < 0.05), after carrying out a Int. J. Mol. Sci. 2019, 20, 1662 5 of 24 of the plasmatic transaminase enzymes alanine aminotransferase (ALT) and aspartate aminotransferase (AST), whose activities showed an increase at the end of the experimental period (Table 3). It can be observed that rats fed diet N showed a normal histopathological liver (Figure 1). Table 3. Biochemical parameters of plasma, inﬂammation and oxidative stress biomarkers analyzed in the six experimental groups at the end of the 5-week intervention period 1. Parameters NC N2.5 N5 Glucose (mg/dL) Proteins (g/dL) Final ALT (U/L) Final AST (U/L) Adiponectin (pg/mL) TNF-α (pg/mL) ORAC (mmoles equiv trolox/L) Urine isoprostanes (ng/mg creatinine) 199.4 ± 42.7 5.42 ± 0.33 * 32.6 ± 5.09 * 75.07 ± 10.14 b* 0.95 ± 0.38 15.6 ± 1.13 8.29 ± 1.62 0.95 ± 0.17 147.1 ± 18 5.79 ± 0.18 28.95 ± 0.35 * 88.28 ± 10.95 b* 0.62 ± 0.16 14.49 ± 1.29 9.06 ± 0.66 1.01 ± 0.18 132.17 ± 4.38 5.67 ± 0.55 34.40 ± 4.16 * 106.1 ± 6.93 a* 1.09 ± 0.84 14.09 ± 1.13 9.28 ± 1.17 1.1 ± 0.14 HC 274.1 ± 31.1 a 6.32 ± 0.26 47.95 ± 9.55 151.10 ± 5.20 0.55 ± 0.10 14.46 ± 1.55 9.79 ± 0.72 1.13 ± 0.004 H2.5 177.9 ± 3.89 b 6.04 ± 0.43 45.43 ± 9.83 139.2 ± 43.7 0.69 ± 0.13 14.68 ± 1.33 9.91 ± 1.48 0.97 ± 0.26 H5 153.6 ± 20.5 b 6.25 ± 0.12 44.30 ± 4.85 142.7 ± 32.5 0.54 ± 0.13 14.16 ± 1.62 9.81 ± 0.75 0.96 ± 0.09 1 Data are expressed as mean ± SD. a,b Different letters show signiﬁcant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high fat diet (HC, H2.5, H5), after performing a one-way ANOVA. Signiﬁcant statistical difference (p < 0.05), after carrying out a two-samples t test, between the members of the NC-HC, N2.5-H2.5, and N5-H5 pairings. ALT: alanine aminotransferase; AST: aspartate aminotransferase; TNF: tumor necrosis factor; ORAC: oxygen radial absorption capacity. The biochemical parameters of the plasma (levels of glucose, protein and hepatic enzymes) are shown in Table 3. The glucose concentration did not differ signiﬁcantly among the N groups, but did among the H groups, being signiﬁcantly reduced in rats of group H5 (153.6 mg/dL), in contrast to the total protein level, which remained unchanged at the end of the study. As mentioned above, rats of the H groups showed a signiﬁcantly higher level of hepatic enzymes than animals of N groups, indicating disturbances in the liver functionality. Regarding the plasma lipid levels, Figure 2 represents the changes between initial and ﬁnal values. No signiﬁcant changes were observed for total cholesterol, low-density lipoprotein (LDL), high-density lipoprotein (HDL), VLDL, or triglycerides (TG) in groups NC, N2.5, and N5 during the intervention period. In contrast, important changes were observed between the initial and ﬁnal parameters for the H groups, showing a signiﬁcant drop in total cholesterol, LDL and VLDL and a signiﬁcant increase in plasmatic TG (Figure 2). This trend was due to the steatosis, since the metabolism of lipoprotein is altered, signiﬁcantly increasing the content of plasmatic TG. The consumption of spinach only leads to a signiﬁcant reduction in ﬁnal cholesterol (108 in HC, 91 in H2.5 and 75 in H5), and triglycerides (123 in HC, 102 in H2.5 and 103 in H5), showing a hipocholestrelomic effect (data not shown). Other parameters measured were the inﬂammation and oxidative stress biomarkers. For oxygen radical absorption capacity (ORAC) in plasma and for urinary isoprostanes, there were no signiﬁcant differences between the initial (data not shown) and ﬁnal values or among the different conditions (diet and spinach supplementation, Table 3). The levels of adiponectin and TNF-α did not change either; no signiﬁcant difference was found according to the diet or the consumption of spinach. 2.3. Content of Carotenoids, Total Fat, Fatty Acids and Cholesterol in the Liver The bioavailability of carotenoids was measured considering their accumulation in the liver. For groups N5, H2.5, and H5, the total carotenoid accumulation in the liver was 0.20, 0.29 and 1.45 µg/g, respectively. The carotenoid with the highest concentration in the liver was β-carotene, whereas the lowest concentration was observed for lutein. In N2.5, NC, and HC rats, no carotenoids were detected (Table 4). The accumulation of total fat in the liver of animals of the H groups was ﬁve times greater than in the N groups (mean values: 25.35% vs. 5.51%, data not shown), and the consumption of spinach had no effect on the liver total fat content. Nevertheless, spinach consumption and the accumulation of carotenoids in the liver appeared to have an effect on the accumulation of cholesterol, since there were signiﬁcant reductions in hepatic cholesterol in groups H2.5 and H5, which reached healthier values with respect to the control (HC) (Table 5). In addition, a signiﬁcant reduction of cholesterol Int. J. Mol. Sci. 2019, 20, 1662 6 of 24 was detected in the liver of animals of group N5. The analysis of total fatty acids in the liver showed differences in the quantities of speciﬁc fatty acids between N and H groups (Table 5). A higher proportion of monounsaturated fatty acids (MUFA) was observed in animals fed the H diet compared to those fed the N diet, which showed a high proportion of saturated (SAFA) and polyunsaturated fatty acids (PUFA). The addition of spinach positively inﬂuenced the fatty acid proﬁle of the liver, signiﬁcantly reducing the contents of SAFA as well as signiﬁcantly increasing the content of PUFA in N and H groups. Also, an increase of MUFA was observed, but only in rats fed H diets (Figure 3); although n-3 (linolenic acid (ALA), eicosapentaenoic acid (EPA), docosahexaenoic (DHA)) and n-6 (linoleic acid (LA), eicosadienoic acid (EDA), and arachidonic acid (AA)) increased in animals fed the H diet supplemented with spinach (Table 5), the n-6/n-3 ratio decreased signiﬁcantly (Figure 3). Table 4. Carotenoids content (µg/g) in the liver of rats of four experimental groups, at the end of the 5-week intervention period 1. Carotenoids N2.5 N5 H2.5 H5 Lutein α-carotene β-carotene Total nd nd 0.20 ± 0.09 b 0.20 ± 0.09 b 1 Data are expressed as mean ± SD. a,b Different letters shown signiﬁcant statistical differences (p < 0.05) among groups fed the standard diet (N2.5, N5) or the high fat diet (H2.5, H5), after performing a one-way ANOVA.nd: not detected. 0.03 ± 0.06 0.04 ± 0.03 b 0.22 ± 0.08 b 0.29 ± 0.12 b 0.03 ± 0.06 0.15 ± 0.10 a 1.28 ± 0.47 a 1.45 ± 0.51 a nd nd nd nd Figure 2. Changes in lipid parameters measured in plasma at the beginning and at the end of the intervention period of 5-weeks for the six experimental groups. LDL: low-density lipoprotein; HDL: high-density lipoprotein; VLDL: very low-density lipoprotein; TG: triglycerides. -100-80-60-40-200204060NCN2.5N5HCH2.5H5TOTAL CHOLESTEROLLDLHDLVLDLTGInt. J. Mol. Sci. 2019, 20, 1662 7 of 24 Table 5. Total fat (mg/100 g), cholesterol content (mg/g) and fatty acid concentrations (mg/g) in the liver of rats in the six experimental groups at the end of the 5-week intervention period 1. Parameters NC N2.5 N5 HC H2.5 H5 Total fat Total cholesterol Caprylic acid (C8:0) Capric acid (C10:0) Undecanoic acid (C11:0) Lauric acid (C12:0) Tridecanoic acid (C13:0) Myristic acid (C14:0) Pentadecanoic acid (C15:0) Palmitic acid (C16:0) Margaric acid (C17:0) Stearic acid (C18:0) Arachidic acid (C20:0) Myristoleic acid (C14:1) Cis Pentadecanoic acid (C15:1) Palmitoleic acid (C16:1) Cis Heptadecenoic acid (C17:1) Oleic cid (C18:1n9c) Eicosenoic acid (C20:1n9) Nervonic acid (C24:1n9) Linolelaidic acid (C18:2tn-6) Linoleic acid (C18:2cn-6) γ-Linolenic acid (C18:3n-6) α- Linolenic acid (C18:3n-3) Eicosadienoic acid (C20:2n-6) Dihomo-γ-linolenic acid (C20:3n-6) Arachidonic acid (C20:4n-6) Eicosapentaenoic acid (C20:5n-3) Docosahexaenoic acid (C22:6n-3) 4.77 ± 1.33 223.9 ± 46 ab* 0.15 ± 0.03 0.33 ± 0.07 a nd 0.21 ± 0.06 a 0.14 ± 0.05 a 0.66 ± 0.34 a 0.21 ± 0.10 a 7.79 ± 0.41 a 0.39 ± 0.21 a 5.00 ± 0.42 0.80 ± 0.13 a 0.08 ± 0.01 0.16 ± 0.07 b 0.70 ± 0.04 a 0.07 ± 0.03 3.27 ± 0.11 0.06 ± 0.01 0.33 ± 0.03 b nd 5.36 ± 0.85 b 0.21 ± 0.10 a 0.32 ± 0.09 a 0.11 ± 0.05 b 0.17 ± 0.01 b 5.76 ± 0.82 nd 1.19 ± 0.13 b 6.37 ± 0.90 * 275.0 ± 48 a* 0.15 ± 0.06 0.14 ± 0.01 b nd 0.18 ± 0.03 a 0.11 ± 0.03 a 0.18 ± 0.03 b 0.08 ± 0.008 b 7.83 ± 0.98 ab 0.16 ± 0.03 b 4.42 ± 0.70 0.10 ± 0.02 b 0.10 ± 0.04 0.18 ± 0.06 b 0.51 ± 0.19 b 0.06 ± 0.02 3.47 ± 0.24 0.06 ± 0.02 0.39 ± 0.08 b nd 6.97 ± 1.20 a 0.08 ± 0.02 b 0.18 ± 0.07 b 0.17 ± 0.04 a 0.19 ± 0.03 b 5.81 ± 0.61 nd 1.37 ± 0.01 a 5.41 ± 2.32 * 192.5 ± 45 b nd 0.12 ± 0.03 b nd 0.10 ± 0.03 b 0.06 ± 0.007 b 0.13 ± 0.02 b 0.08 ± 0.009 b 6.78 ± 0.85 b 0.16 ± 0.02 b 4.95 ± 0.81 0.10 ± 0.03 b 0.10 ± 0.03 0.35 ± 0.05 a 0.51 ± 0.09 b 0.06 ± 0.009 3.25 ± 0.41 0.08 ± 0.04 0.55 ± 0.09 a nd 6.58 ± 0.61 a 0.06 ± 0.03 b 0.28 ± 0.05 ab 0.15 ± 0.03 ab 0.32 ± 0.07 a 6.27 ± 0.97 0.19 ± 0.03 1.53 ± 0.16 a 25.58 ± 4.98 6048 ± 2801 a* 0.27 ± 0.0.08 a 0.35 ± 0.09 a 0.32 ± 0.09 a 0.12 ± 0.03 0.14 ± 0.01 a 1.67 ± 0.08 a 0.37 ± 0.02 22.63 ± 2.98 0.46 ± 0.006 a 8.08 ± 0.23 a 0.40 ± 0.13 a 0.12 ± 0.03 0.10 ± 0.03 2.98 ± 0.64 0.32 ± 0.05 30.99 ± 2.68 0.31 ± 0.02 0.32 ± 0.04 a nd 13.87 ± 1.28 0.14 ± 0.04 0.51 ± 0.06 b 0.28 ± 0.11 b 0.73 ± 0.06 4.40 ± 0.42 nd 0.84 ± 0.10 b 25.09 ± 0.89 647.8 ± 229 c* 0.13 ± 0.02 a,b 0.14 ± 0.07 b 0.15 ± 0.06 b 0.07 ± 0.02 0.12 ± 0.02 a 1.28 ± 0.16 b 0.36 ± 0.05 20.46 ± 2.17 0.36 ± 0.04 b 6.35 ± 0.88 b 0.25 ± 0.16 ab 0.11 ± 0.08 0.11 ± 0.04 3.63 ± 0.70 0.39 ± 0.07 39.58 ± 5.81 0.38 ± 0.06 0.46 ± 0.10 ab 0.11 ± 0.04 14.44 ± 1.73 0.17 ± 0.05 1.19 ± 0.27 a 0.43 ± 0.07 a 0.84 ± 0.23 5.00 ± 1.04 0.14 ± 0.09 b 1.04 ± 0.23 ab 25.73 ± 0.62 233.4 ± 50 b 0.10 ± 0.06 b 0.11 ± 0.05 b nd 0.07 ± 0.01 0.09 ± 0.03 b 1.29 ± 0.27 b 0.34 ± 0.01 19.13 ± 1.46 0.36 ± 0.03 b 6.73 ± 0.88 b 0.13 ± 0.02 b 0.12 ± 0.07 0.11 ± 0.03 3.63 ± 0.68 0.43 ± 0.09 40.94 ± 8.92 0.36 ± 0.09 0.58 ± 0.13 b 0.11 ± 0.02 18.05 ± 4.21 0.18 ± 0.04 1.39 ± 0.3 a 0.42 ± 0.08 a 0.81 ± 0.06 5.40 ± 0.70 0.28 ± 0.05 a 1.28 ± 0.16 a 1 Data are expressed as mean ± SD. a–c Different letters show signiﬁcant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high fat diet (HC, H2.5, H5), after performing a one-way ANOVA. nd: Not detected. * Signiﬁcant statistical difference (p < 0.05), after carrying out a two-sample t test, between the members of the NC-HC, N2.5-H2.5, and N5-H5 pairings. Figure 3. Saturated fatty acid (SAFA)/total fatty acid (TFA), monounsaturated fatty acid (MUFA)/TFA, polyunsaturated fatty acid (PUFA)/TFA, and ω-6/ω-3 ratios in the liver of rats of the six experimental groups. a,b Different letters show signiﬁcant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high-fat diet (HC, H2.5, H5) after performing a one-way ANOVA. * Signiﬁcant statistical difference (p < 0.05), after carrying out a two-sample t test, between the members of the NC-HC, N2.5-H2.5, and N5-H5 pairings. Result are expressed as mean ± SD. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 8 of 25 Figure 3. Saturated fatty acid (SAFA)/total fatty acid (TFA), monounsaturated fatty acid (MUFA)/TFA, polyunsaturated fatty acid (PUFA)/TFA, and ω-6/ω-3 ratios in the liver of rats of the six experimental groups. a,b Different letters show significant statistical differences (p < 0.05) among groups fed the standard diet (NC, N2.5, N5) or the high-fat diet (HC, H2.5, H5) after performing a one-way ANOVA. * Significant statistical difference (p < 0.05), after carrying out a two-sample t test, between the members of the NC-HC, N2.5-H2.5, and N5-H5 pairings. Result are expressed as mean ± SD. 2.4. Gene Expression Related to NAFLD Twenty-seven differentially expressed genes were selected from the fatty liver array (Table 6), according to the criteria indicated above. All genes with differential expression showed an overexpression of the mRNA; none was down-expressed. In general terms, the group in which the most major changes in the transcriptome were observed was N5, showing changes in genes related to β-oxidation (3 genes), cholesterol and other lipid transport and metabolism (14 genes), and the inflammatory response and apoptosis (6 genes). For animals of the HC and H5 group, 5 and 11 genes were overexpressed, respectively. According to these results, a fat diet led to changes in gene expression, but higher changes were observed in animals that took in spinach and accumulated carotenoids in their livers. 2.5. Metabolites in Liver Regarding the liver metabolites, the principal component analysis (PCA) shows a great influence of the type of diet (standard or high fat) on the metabolites (represented by PC1), accounting for 74% of the total variance in the amino acids and 69% of the total variance in the antioxidant and nucleotide components (Figure 4). The spinach supplementation of the diet (represented by PC2) accounted for 10.7% of the total variance in the amino acids and 13% for the other components analyzed. Hence, the total variance explained jointly by PC1 and PC2 was 84% for the amino acids and more than 82% for the other components. In global terms, a standard diet had a positive influence on all the metabolites; however, the effect of spinach (PC2) mainly altered redox molecules such as L-glutathione (GSH), L-glutathione oxidised form (GSSG), L-homocysteine (Homo-Cys), nicotinamide adenine dinucleotide oxidised form (NAD), and nicotinamide adenine dinucleotide reduced form (NADH), showing a modulation of the redox responses, which could be Int. J. Mol. Sci. 2019, 20, 1662 8 of 24 2.4. Gene Expression Related to NAFLD Twenty-seven differentially expressed genes were selected from the fatty liver array (Table 6), according to the criteria indicated above. All genes with differential expression showed an overexpression of the mRNA; none was down-expressed. In general terms, the group in which the most major changes in the transcriptome were observed was N5, showing changes in genes related to β-oxidation (3 genes), cholesterol and other lipid transport and metabolism (14 genes), and the inﬂammatory response and apoptosis (6 genes). For animals of the HC and H5 group, 5 and 11 genes were overexpressed, respectively. According to these results, a fat diet led to changes in gene expression, but higher changes were observed in animals that took in spinach and accumulated carotenoids in their livers. Table 6. Gene symbol, gene name, and relative fold change of the genes that showed an over or down-expression value higher than 2 (p < 0.05) in the rat livers 1. Gene Name NC-HC NC-N5 NC-H5 Symbol ACADL CPT1A CPT2 PPARA ABCG1 APOA1 APOB APOE CNBP CYP2E1 CYP7A1 LDLR NR1H3 NR1H4 PPARD PPARG SREBF2 β-oxidation Acyl-coenzyme A dehydrogenase, long-chain Carnitine palmitoyltransferase 1A, liver Carnitine palmitoyltransferase 2 Peroxisome proliferator activated receptor alpha - - - - Cholesterol transport and metabolism ATP-binding cassette, subfamily G (WHITE), member 1 Apolipoprotein A-I Apolipoprotein B Apolipoprotein E CCHC-type zinc ﬁnger, nucleic acid binding protein Cytochrome P450, family 2, subfamily E, polypeptide 1 Cytochrome P450, family 7, subfamily A, polypeptide 1 Low density lipoprotein receptor Nuclear receptor subfamily 1, group H, member 3 Nuclear receptor subfamily 1, group H, member 4 Peroxisome proliferator-activated receptor delta Peroxisome proliferator-activated receptor gamma Sterol regulatory element binding transcription factor 2 ACSM3 LPL Acyl-CoA synthetase medium-chain family member 3 Lipoprotein lipase Other lipid transport and metabolism Inﬂammatory response and apoptosis ADIPOR1 FAS IL1B NFKB1 CASP3 MAPK8 SOCS3 Adiponectin receptor 1 Fas (TNF receptor superfamily, member 6) Interleukin 1 beta Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 Caspase 3 Mitogen-activated protein kinase 8 Suppressor of cytokine signaling 3 - 2.15 7.04 - - - - 5.89 - - - - - - - - - - 3.36 - - 2.98 3.03 - 7.97 2.75 2.26 7.86 2.68 36.06 4.70 2.94 - 4.76 2.23 11.29 2.93 6.81 3.66 2.27 - 6.96 - 3.23 3.05 2.62 2.77 5.77 - 2.25 2.24 2.44 - - - 3.17 - 2.95 2.47 - - 2.88 - 3.85 - - - 2.63 - - - 3.02 - - 1 The fold change for each gene in groups N5, H5, and HC was calculated taking as reference a value of 1 for the group NC. ATP: adenosine triphosphate. 2.5. Metabolites in Liver Regarding the liver metabolites, the principal component analysis (PCA) shows a great inﬂuence of the type of diet (standard or high fat) on the metabolites (represented by PC1), accounting for 74% of the total variance in the amino acids and 69% of the total variance in the antioxidant and nucleotide components (Figure 4). The spinach supplementation of the diet (represented by PC2) accounted for 10.7% of the total variance in the amino acids and 13% for the other components analyzed. Hence, Int. J. Mol. Sci. 2019, 20, 1662 9 of 24 the total variance explained jointly by PC1 and PC2 was 84% for the amino acids and more than 82% for the other components. In global terms, a standard diet had a positive inﬂuence on all the metabolites; however, the effect of spinach (PC2) mainly altered redox molecules such as L-glutathione (GSH), L-glutathione oxidised form (GSSG), L-homocysteine (Homo-Cys), nicotinamide adenine dinucleotide oxidised form (NAD), and nicotinamide adenine dinucleotide reduced form (NADH), showing a modulation of the redox responses, which could be associated with the antioxidant capacity of the carotenoids. In addition to the above, a change in nucleotides (adenosine monophosphate (AMP), adenosine triphosphate (ATP), and inosine triphosphate (ITP)) and some amino acids (L-proline (Pro), L-asparagine (Asn), L-cysteine (Cys), L-arginine (Arg), L-histidine (His), L-alanine (Ala), L-glutamic acid (Glu), and taurine) occurred, which indicates an effect on the redox metabolism and some other key points in the rat metabolism. Figure 4. (A) PCA scores and (B) eigenvector plots for the amino acids, antioxidants, and nucleotides compounds, according to the metabolites found to differ signiﬁcantly (ANOVA p < 0.05) among the different diets: group NC: standard diet, N5: standard diet + 5% spinach, HC: high fat diet and H5: high fat diet + 5% spinach. Conﬁdence regions are marked with different ellipses. In general, a signiﬁcant reduction in the amino acid content was observed in the animals of HC and H5 groups, in comparison with the NC and N5 groups. When the diet of the healthy animals was Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 9 of 25 associated with the antioxidant capacity of the carotenoids. In addition to the above, a change in nucleotides (adenosine monophosphate (AMP), adenosine triphosphate (ATP), and inosine triphosphate (ITP)) and some amino acids (L-proline (Pro), L-asparagine (Asn), L-cysteine (Cys), L-arginine (Arg), L-histidine (His), L-alanine (Ala), L-glutamic acid (Glu), and taurine) occurred, which indicates an effect on the redox metabolism and some other key points in the rat metabolism. Table 6. Gene symbol, gene name, and relative fold change of the genes that showed an over or down-expression value higher than 2 (p < 0.05) in the rat livers 1. Symbol Gene Name NC-HC NC-N5 NC-H5 β-oxidation ACADL Acyl-coenzyme A dehydrogenase, long-chain - 3.03 - CPT1A Carnitine palmitoyltransferase 1A, liver - - 2.25 CPT2 Carnitine palmitoyltransferase 2 - 7.97 2.24 PPARA Peroxisome proliferator activated receptor alpha - 2.75 2.44 Cholesterol transport and metabolism ABCG1 ATP-binding cassette, subfamily G (WHITE), member 1 5.89 2.26 - APOA1 Apolipoprotein A-I - 7.86 - APOB Apolipoprotein B - 2.68 - APOE Apolipoprotein E - 36.06 3.17 CNBP CCHC-type zinc finger, nucleic acid binding protein - 4.70 - CYP2E1 Cytochrome P450, family 2, subfamily E, polypeptide 1 - 2.94 2.95 CYP7A1 Cytochrome P450, family 7, subfamily A, polypeptide 1 - - 2.47 LDLR Low density lipoprotein receptor - 4.76 - NR1H3 Nuclear receptor subfamily 1, group H, member 3 - 2.23 - NR1H4 Nuclear receptor subfamily 1, group H, member 4 - 11.29 2.88 PPARD Peroxisome proliferator-activated receptor delta - 2.93 - PPARG Peroxisome proliferator-activated receptor gamma 3.36 6.81 3.85 SREBF2 Sterol regulatory element binding transcription factor 2 - 3.66 - Other lipid transport and metabolism ACSM3 Acyl-CoA synthetase medium-chain family member 3 - 2.27 - LPL Lipoprotein lipase 2.98 - - Inflammatory response and apoptosis ADIPOR1 Adiponectin receptor 1 - 6.96 2.63 FAS Fas (TNF receptor superfamily, member 6) 2.15 - - IL1B Interleukin 1 beta 7.04 3.23 - NFKB1 Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 - 3.05 - CASP3 Caspase 3 - 2.62 3.02 MAPK8 Mitogen-activated protein kinase 8 - 2.77 - SOCS3 Suppressor of cytokine signaling 3 - 5.77 - 1 The fold change for each gene in groups N5, H5, and HC was calculated taking as reference a value of 1 for the group NC. ATP: adenosine triphosphate. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 10 of 25 Figure 4. (A) PCA scores and (B) eigenvector plots for the amino acids, antioxidants, and nucleotides compounds, according to the metabolites found to differ significantly (ANOVA p < 0.05) among the different diets: group NC: standard diet, N5: standard diet + 5% spinach, HC: high fat diet and H5: high fat diet + 5% spinach. Confidence regions are marked with different ellipses. In general, a significant reduction in the amino acid content was observed in the animals of HC and H5 groups, in comparison with the NC and N5 groups. When the diet of the healthy animals was supplemented with spinach, the concentrations of some amino acids increased significantly (L-serine (Ser), Pro, Cys, Asn, L-lysine/L-glutamine (Lys/Gln), Homo-Cys, L-tryptophan (Trp)), whereas those of Glu and taurine declined (Figure 5). The antioxidant and nucleotide compounds also decreased under the fatty diet (Figure 6). However, in rats fed diet N, the consumption of spinach significantly reduced the concentrations of redox compounds (GSH, GSSG, NAD, NADH) and of some nucleotides (uridine diphosphate (UDP) and cytidine triphosphate (CTP)). The H diet and the consumption of spinach influenced the GSH/GSSG ratio of the liver, decreasing in rats fed H diets in comparison with the NC group, and, similarly, a significant reduction was observed in the N5 group (Table 7). The ratios of NAD/NADH and nicotinamide adenine dinucleotide phosphate oxidised form/nicotinamide adenine dinucleotide phosphate reduced form (NADP/NADPH) remained unchanged among the four experimental groups (Table 7). Int. J. Mol. Sci. 2019, 20, 1662 10 of 24 supplemented with spinach, the concentrations of some amino acids increased signiﬁcantly (L-serine (Ser), Pro, Cys, Asn, L-lysine/L-glutamine (Lys/Gln), Homo-Cys, L-tryptophan (Trp)), whereas those of Glu and taurine declined (Figure 5). The antioxidant and nucleotide compounds also decreased under the fatty diet (Figure 6). However, in rats fed diet N, the consumption of spinach signiﬁcantly reduced the concentrations of redox compounds (GSH, GSSG, NAD, NADH) and of some nucleotides (uridine diphosphate (UDP) and cytidine triphosphate (CTP)). The H diet and the consumption of spinach inﬂuenced the GSH/GSSG ratio of the liver, decreasing in rats fed H diets in comparison with the NC group, and, similarly, a signiﬁcant reduction was observed in the N5 group (Table 7). The ratios of NAD/NADH and nicotinamide adenine dinucleotide phosphate oxidised form/nicotinamide adenine dinucleotide phosphate reduced form (NADP/NADPH) remained unchanged among the four experimental groups (Table 7). Table 7. Redox ratios in the liver of the rats of four experimental groups at the end of the intervention period 1. Ratio NC N5 HC H5 1.39 ± 0.38 * GSH/GSSG 0.33 ± 0.09 NAD/NADH NADP/NADPH 1.0 ± 0.26 0.79 ± 0.16 0.31 ± 0.06 1.29 ± 0.49 1 Data are expressed as mean ± SD. * Signiﬁcant statistical difference (p < 0.05), after carrying out a two-sample t test comparing the NC–N5 and HC–H5 groups. 0.97 ± 0.30 0.36 ± 0.04 1.27 ± 0.59 0.97 ± 0.21 0.36 ± 0.10 1.27 ± 0.23 Int. J. Mol. Sci. 2019, 20, 1662 11 of 24 Figure 5. Amino acid content in the liver of rats of the four experimental groups (HC: high fat diet, H5: high fat diet + 5% spinach, NC: standard diet and N5: standard diet + 5% spinach). The bar height indicates the mean value of each feed condition and the error bar indicates the standard deviation. a–c Different letters indicate signiﬁcant statistical differences (p < 0.05). Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 11 of 25 1 Figure 5. Amino acid content in the liver of rats of the four experimental groups (HC: high fat diet, H5: high fat diet + 5% spinach, NC: standard diet and N5: standard diet 2 + 5% spinach). The bar height indicates the mean value of each feed condition and the error bar indicates the standard deviation. a–c Different letters indicate significant 3 statistical differences (p < 0.05). 4 020004000600080001000012000AspAlaSerProValHCHANCNA0100020003000400050006000700080009000ThrCysGluIle/Leu/OH-ProMetHCHANCNA050001000015000200002500030000350004000045000GlyHomo-CysLys/GlnTaurineAsnHCHANCNA05001000150020002500HisCarnitinePheArgTyrTrpCystineHCHANCNAbbaaabbaaccbabbabbbaaabbababbbabbabbbaabcbacabbaabbbaccabcbabbbbaabaabbaabbaabbaabcacbaccbabbaaHCH5NCN5HCH5NCN5HCH5NCN5HCH5NCN5Int. J. Mol. Sci. 2019, 20, 1662 12 of 24 Figure 6. Antioxidant and nucleotide compounds in the liver of rats of the four experimental groups (HC: high fat diet, H5: high fat diet + 5% spinach, NC: standard diet and N5: standard diet + 5% spinach). The bar height indicates the mean value of each feed condition and the error bar indicates a–c Different letters indicate signiﬁcant statistical differences (p < 0.05). the standard deviation. GSH: L-glutathione; GSSG: L-glutathione oxidized form; NAD: nicotinamide adenine dinucleotide; NADH: nicotinamide adenine dinucleotide reduced form; NADP: nicotinamide adenine dinucleotide phosphate; NADPH: nicotinamide adenine dinucleotide phosphate reduced form; CMP: cytidine monophosphate; AMP: adenosine monophosphate; GMP: guanosine monophosphate; UDP: uridine diphosphate; GDO: guanosine diphosphate; CTP: cytidine triphosphate; ATP: adenosine triphosphate; ITP: inosine triphosphate. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 12 of 25 Figure 6. Antioxidant and nucleotide compounds in the liver of rats of the four experimental groups (HC: high fat diet, H5: high fat diet + 5% spinach, NC: standard diet and N5: standard diet + 5% spinach). The bar height indicates the mean value of each feed condition and the error bar indicates the standard deviation. a–c Different letters indicate significant statistical differences (p < 0.05). GSH: L-glutathione; GSSG: L-glutathione oxidized form; NAD: nicotinamide adenine dinucleotide; NADH: nicotinamide adenine dinucleotide reduced form; NADP: nicotinamide adenine dinucleotide phosphate; NADPH: nicotinamide adenine dinucleotide phosphate reduced form; CMP: cytidine monophosphate; AMP: adenosine monophosphate; GMP: guanosine monophosphate; UDP: uridine diphosphate; GDO: guanosine diphosphate; CTP: cytidine triphosphate; ATP: adenosine triphosphate; ITP: inosine triphosphate. 050100150200250GSHGSSGNADNADHNADPNADPHHCHANCNAbbabccabbbabbbabbbaabbbaaHCH5NCN50200400600800100012001400160018002000HypoxanCMPAMPGMPUDPGDPCTPATPITPHCHANCNA3000400050006000babababbaaccabbbaabccabbbaaccabbbaabbaaHCHANCNAHCH5NCN5Int. J. Mol. Sci. 2019, 20, 1662 13 of 24 3. Discussion 3.1. Carotenoid Supplementation and Biomarkers of Stetaosis In the present study, the intake of high fat provoked steatosis in rats in the H groups, as described by different authors [9,14,15]. Animals fed the high-fat diet showed a signiﬁcant increase in the activities of hepatic enzymes and microvesicular steatosis. However, this hepatic disturbance was not severe, because there were no clear symptoms of lipotoxicity and steatohepatitis, as revealed by the inﬂammation and oxidative stress biomarkers. Although the consumption of spinach had no clear effect in the inﬂammation and stress biomarkers, we observed in the histological examination a reduction of the lipid size of vacuoles and a signiﬁcant reduction in the total weight of the liver and plasmatic glucose levels in H2.5 and H5 groups. This effect of carotenoids on glucose metabolism has been reported in previous in vivo studies using dietary carotenoids [16,17], showing a beneﬁcial effect on steatosis features by decreasing insulin resistance. Although some plasmatic parameters remained within the normal range [18,19], from the beginning to the end of the experiment, it was expected that the most signiﬁcant changes would occur in the lipid proﬁle of the plasma due to the alteration of lipid metabolism by the hepatic steatosis, increasing total cholesterol and TG. In contrast, a decrease in total cholesterol and lipoprotein LDL and VLDL could be explained by a decline in the synthesis of lipoprotein in the liver, so that lipids accumulated in the hepatocytes instead of being liberated into the peripheral circulation [20], whereas TG increased in plasma. It is remarkable that the consumption of spinach signiﬁcantly reduced the proportion of plasmatic TG, indicating a role in lipid metabolism. The role of carotenoids in the regulation of speciﬁc functions and in the prevention of disease is determined by their bioavailability. Carotenoids are absorbed through the mucosa of the small intestine by passive diffusion, lodging inside the chylomicrons, which are rich in triglycerides, due to their lipophilic nature, and are transported in the lymph to the liver [21]. Later, carotenoids are transported by the LDL and are incorporated into the inner body of the lipoproteins, whereas xanthophylls become attached to their surface; for this reason, xanthophylls are transported at a greater rate to other organs [22]. The accumulation of total carotenoids in the liver (carotenes and lutein) was signiﬁcantly correlated with the amount of spinach provided and the type of diet, since the fat content of the feed facilitated the absorption of carotenoids, as has been mentioned in other studies [9]; for this reason, in rats in group N2.5, carotenoids were not detected in the liver. Although carotenoids were absorbed and accumulated in the liver of rats of N5, H2.5 and H5 groups, these antioxidants did not appear to have a signiﬁcant effect on the biomarkers of oxidative stress and inﬂammation in plasma. Ko et al. [23] reported that the oxidative stress caused by hyperlipidemia can be partly prevented by the antioxidant activities of spinach administrated at 5%, together with a fat-rich diet. However, these authors described a decline in thiobarbituric acid reactive substances (TBARs) in the liver, but not in plasma. Other researchers have reported that several dietary carotenoids—such as lycopene from tomato juice administered ad libitum, lutein (100 mg/g of diet), zeaxanthin (0.25 mg/g), and astaxanthin (0.2 mg/g)—can diminish the oxidative stress and/or levels of biomarkers of cellular inﬂammation in rats [6,9,10,16]. These concentrations are higher than those assayed in this study. Hence, to evaluate the beneﬁcial effect of spinach on these biomarkers, a higher concentration in the diet could be required. Although no changes were observed for plasma adiponectin and TNF-α, the modulation of the inﬂammatory response by carotenoids depends on different factors, such as speciﬁc compounds and their concentrations, but also on the level of oxidative stress [24]. Carotenoids have been used to reduce the inﬂammatory response through effects on the transcription system of nuclear factor-κB (NF-κB). However, the results are contradictory: in different cell cultures, lycopene was found to repress NF-κB, while β-carotene stimulated it [25]. Changes in the genes related to the inﬂammatory response were studied for the NC, H5, and N5 groups, and will be discussed later. The depletion of PUFA in the H groups indicates a decline in fatty acid oxidation and triglyceride release from the liver, with a consequent increase in triglyceride synthesis that may have contributed Int. J. Mol. Sci. 2019, 20, 1662 14 of 24 signiﬁcantly to the triglyceride accumulation in hepatocytes [26]. The consumption of spinach led to signiﬁcant changes in the liver fatty acid proﬁle, which became healthier, with increases in the proportions of MUFA and PUFA and decreases in SAFA and the n-6/n-3 ratio. These changes could be considered beneﬁcial with respect to the inhibition of stress-related kinases and apoptosis [27], since it has been reported that MUFA decreases intracellular lipid levels and the markers of inﬂammation, increasing fatty acid oxidation and triglycerides synthesis [28]. Moreover, this behavior suggests that the accumulation of carotenoids stimulated the conversion of fatty acids into long-chain unsaturated products, yielding α-linolenic acid (C18:3 n-3, ALA) and eicosadienoic acid (C20:2 n-6, EDA), as described by Bell et al. [29] for the ﬂesh of salmon. In our study, when the high-fat diet was supplemented with spinach, the MUFA and PUFA concentrations increased signiﬁcantly, reducing the SAFA contribution (Figure 2). The most plausible explanation is a higher activity of both desaturases and elongase enzymes, as described by other authors [30]. These changes reduce or inhibit the de novo synthesis of fatty acids and activate their β-oxidation by the stimulation of MUFA and PUFA. The decrease in the n-6/n-3 ratio indicates an increase in the concentration of n-3 fatty acids, which limits the storage of triglycerides in the liver, reducing the risk of the development of NAFLD [31]. In addition, n-3 PUFAs are precursors of anti-inﬂammatory eicosanoids, as opposed to n-6 PUFAs, which produce pro-inﬂammatory eicosanoids [32]. When carotenoids were present in the liver, the concentrations of the ALA (C18:3 n-3 α), EDA, EPA (C20:5 n-3), and DHA (C22:6 n-3) fatty acids increased, improving the liver fatty acid proﬁle in health terms. In addition, carotenoids also inﬂuenced the reduction of the cholesterol level in the liver of rats with steatosis, as has been described in the scientiﬁc literature for different carotenoids. Kim et al. [10] reported a signiﬁcant reduction in the percentage distribution of free cholesterol in the liver of rats after the administration of 0.1 g/100 g of lutein. Nicolle et al. [33] reported that supplementation of the diet with 0.25% cholesterol and 20% lyophilized carrots signiﬁcantly reduced the cholesterol levels in plasma and in the liver of C57BL/6J mice. Qiu et al. [34] fed Sprague–Dawley rats with a high-fat diet supplemented with 50 mg of lutein/kg body weight/day and found a reduction in the cholesterol content in the liver, which correlated with the concentration of lutein in the diet. In other animal studies involving a high-fat diet supplemented with 20 mg of lycopene/kg body weight/day or 30 mg of astaxanthin/kg body weight/day, the administration of these carotenoids reduced the concentration of cholesterol in the liver to levels similar to the control [35,36]. In general, these authors used carotenoid-rich plant material or pure compounds at a dose 4–8 times higher than that used in our experiment. Although we used a lower dose, liver cholesterol decreased in rats with steatosis after the intake of spinach, as well in healthy animals when taking in around 50 µg/day of total carotenoids. Different mechanisms have been proposed for the reduction of cholesterol accumulation in the liver; one is the inhibition of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR) activity, as has been described for other carotenoids such as lycopene [37] and β-carotene [38]. 3.2. Bioactivity of Diets Supplemented with Spinach: Modulation of Gene Expression Although the available evidence regarding the potential use of non-provitamin A carotenoids in the prevention and treatment of NAFLD suggests that these compounds are effective in decreasing lipid accumulation, insulin resistance, oxidative stress, and inﬂammation in hepatic tissue, more complex pathways related to gene expression have not been elucidated completely [6]. In this research, we observed that the intake of a fatty diet and spinach led to changes in the expression of genes related to fatty liver disease. In the N5 group—the healthy animals that consumed spinach—there was an over-expression of a great number of genes in comparison with the control group (NC). The over-expression of genes related to β-oxidation, Acyl-Coenzyme A dehydrogenase, long-chain (ACADL), Carnitine palmitoyltransferase 2 (CPT2), and peroxisome proliferator activated receptor alpha (PPARA) was observed. ACADL encodes a dehydrogenase enzyme that catalyzes the initial step in each cycle of fatty acid β-oxidation; this is one of a class of enzymes that are important due to their role in the metabolism of fatty acids present in the diet [39]. In addition, CPT2, together with carnitine Int. J. Mol. Sci. 2019, 20, 1662 15 of 24 palmitoyltransferase 1A, liver (CPT1A), oxidizes long-chain fatty acids in the mitochondria, which allows the linkage of acyl-CoA derivatives to a polar molecule of carnitine, resulting in the formation of acylcarnitine molecules that are transported into the mitochondria [40]. In this study, only CPT2 was overexpressed, but in a previous study, we reported that the intake of lycopene from tomato juice increased the mRNA abundance of CPT1A, together with the carnitine content, in the liver of rats with steatosis [8,9]. The peroxisome proliferator activated receptors (PPARs) consist of three members—PPARA, peroxisome proliferator-activated receptor gamma (PPARG), and peroxisome proliferator-activated receptor delta (PPARD)—which form obligate heterodimers with the retinoid X receptor (RXR). Carotenoids and their metabolites are activators of these receptors [25,41], which are involved in the transcriptional regulation of several pathways of lipid metabolism. PPARA is expressed in the liver and has been considered to exert a critical role in the prevention of fat-related oxidative stress, inﬂammation and NAFLD [42]. PPARD and PPARG were also over-expressed in N5 rats, showing the effect of carotenoids from spinach on the activation of other types of PPAR receptors, expressed in other tissues [43]. In addition, the acyl-CoA synthetase medium-chain family member 3 gene (ACSM3), which encodes a protein that participates in the synthesis of medium-chain fatty acids, was over-expressed, showing an enhancement of lipid metabolism. The activity of this protein leads to a higher inﬂux of fatty acids into the mitochondria, and in particular facilitates the oxidation of medium-chain fatty acids (from C4 to C11), since small- and medium-chain acyl-CoA derivatives have the ability to cross the inner mitochondrial membrane by diffusion [40]. For these reasons, rats fed with spinach (groups N5 and H5) had a lower SAFA/TFA ratio in the liver (Table 6 and Figure 2). Related to cholesterol transport and metabolism, the over-expression of apolipoproteins (APOA1, APOB, and APOE) and membrane receptors such as low-density lipoprotein receptor (LDLR) and ABCG1 was observed, which could have led to an increase in the activity of proteins involved in cholesterol transport. ABCG1 was over-expressed, increasing the efﬂux of cholesterol and phospholipids to lipid-poor apolipoproteins. This transporter is a major regulator of cellular cholesterol and phospholipid homeostasis, since cholesterol taken up by the liver can be recycled back through the ABCG1 pathway, being secreted into bile—as either free cholesterol or bile acids—or assembled into lipoprotein particles that are secreted back into the circulation [44]. The overexpression of ABCG1 was accompanied by the overexpression of the mRNA of apolipoproteins A, B, and E, and also that of the LDL receptor, which indicates an increase in lipoprotein metabolism. We also found an over-expression of the mRNA of genes involved in the repression of the synthesis of different lipids, such as nuclear receptor subfamily 1, group H, member 3 (NR1H3), nuclear receptor subfamily 1, group H, member 4 (NR1H4), CCHC-type zinc ﬁnger, nucleic acid binding protein (CNBP) and sterol regulatory element binding transcription factor 2 (SREBF2), which are responsible for the inhibition of the synthesis of cholesterol and bile acids. LXR (NR1H3) is involved in liver lipid metabolism [45] and exhibits a homeostatic effect at the transcriptional level. This receptor regulates the synthesis of lipids via SREBF2 and the excretion of bile acids, by activating cytochrome P450, family 7, subfamily A, polypeptide 1 (CYP7A1), thereby reaching a balance in the content of hepatic cholesterol [46]. This effect has been described after the consumption of other carotenoids, such as lycopene from tomato, by rats with NAFLD induced by a fatty diet [47]. Additionally, the genes involved in the inﬂammatory response increased their activity, as demonstrated by overexpression of the genes suppressor of cytokine signaling 3 (SOCS3), caspase 3 (CASP3), and mitogen-activated protein kinase 8 (MAPK8), which participate in anti-inﬂammatory and apoptotic processes. Other carotenoids, such as lycopene, can prevent oxidative stress in hepatocytes and also modulate the transcriptome response of genes related to apoptosis and regulation of the cell cycle, particularly by the over-expression of the tumor-suppressor protein (TP53), which acts as a major defense against cancer through the differential activation of target genes [48]. However, it is important to highlight that the adiponectin receptor 1 (ADIPOR1) and nuclear factor (NFKB1) genes were over-expressed, which also could be considered beneﬁcial for the liver. Adiponectin Int. J. Mol. Sci. 2019, 20, 1662 16 of 24 has been shown to have cytoprotective properties, improving both hepatic and peripheral insulin sensitivity and preventing steatosis, inﬂammation, and necrosis, whereas the inhibition of NFKB1 induces non-alcoholic steatohepatitis (NASH) and hepatocellular carcinoma (HCC) by sensitizing hepatocytes to undergo spontaneous apoptosis [49]. Also, in the N5 group, the over-expression of IL1B, considered an inﬂammatory cytokine, was observed, but no negative effect was detected in the rats. In the HC group (rats with steatosis induced by the fatty diet), genes such as ABCG1 and LPL were over-expressed. These genes encode the proteins that facilitate the extracellular transport of lipids and the hydrolysis of TG in free fatty acids, respectively. The gene expression of LPL is higher in obese subjects with NAFLD than in subjects without NAFLD, suggesting that free fatty acids released by the lipolysis of circulating triglycerides also contribute to hepatocellular fatty acid accumulation and steatosis [50]. In addition, there was an increase in the expression of the transcription factor PPARG; this is responsible for increasing insulin sensitivity and for promoting the entry of fatty acids and storage of triglycerides in adipocytes [51]. Additionally, the IL1B and TNFR (FAS) genes were overexpressed, revealing inﬂammation in the adipocytes—a result of the accumulation of fat. These changes are in concordance with the disturbance of lipid metabolism associated with NAFLD and with lipotoxicity [51]. Unlike the HC group, the H5 group showed some similarities with the N5 group, which indicates that spinach consumption had a positive effect on the amelioration of the fatty liver condition. For this reason, a positive modulation of the genes involved in the β-oxidation of fatty acids was observed, associated with the overexpression of CPT1A, CPT2, and the nuclear factor PPARA. With regard to cholesterol transport and metabolism, the APOE gene, which encodes lipoproteins, was also over-expressed, but to a lesser extent in comparison to the N5 group. Hence, spinach intake promoted the expression of these genes, so that liver cholesterol declined due to its efﬂux from the liver through the more abundant VLDL and LDL apoliproteins. The cholesterol catabolic activity was enhanced by the over-expression of the cytochrome P450, family 2, subfamily E, polypeptide 1 (CYP2E1) and CYP7A1 genes, responsible for the augmented synthesis of bile acids through the overexpression of NR1H4. FXR (NR1H4) stimulated the expression of CYP7A1 in H5 rats, keeping the synthesis of bile acids active and thus leading to a decline in hepatic cholesterol [45]. PPARG is required for adipocyte differentiation and for the maintenance of differentiated adipocyte, and is considered as an adipogenic factor, since it is an activator of fatty acid synthesis and storage. However, it plays divergent roles in the metabolism, and these effects in the metabolism are controversial. Thus, PPARG’s activation by thiazolidinediones, the most investigated synthetic agonist, results in the increased production of adipokines, including adiponectin, which enhances hepatic fatty acid oxidation. In addition, it also promotes fat storage in adipocytes and decreases adipose tissue lipolysis, thereby decreasing the concentration of fatty acids stored in the liver. Moreover, its activation also exhibits an anti-inﬂammatory role. This effect is associated with the increase of insulin sensitivity, improving the resistance of insulin associated with the steatosis and metabolic syndrome [52]. In this study, we do not analyze the resistance of insulin, but the supplementation of spinach and the accumulation of carotenoids in the liver (H5 group) exhibited a signiﬁcant effect on the reduction of plasmatic glucose, total cholesterol and TG, as well as cholesterol in the liver. Finally, the ADIPOR1 and CASP3 genes were over-expressed, as described above for rats of the N5 group. 3.3. Changes in Liver Metabolites The amino acids analyzed in the metabolomic study had lower concentrations in rats fed the high-fat diet because of the high caloric content of this diet, which produced a hypoaminoacidemic effect [53], mainly for the glucogenic amino acids, conﬁrming the results obtained in studies performed with hyperlipidemic diets. The concentrations of the intermediaries of the redox process (Met and taurine) were decreased in the H5 group, which may be due to the spinach intake and the antioxidant effect of the accumulated carotenoids in the liver [54]. A plausible explanation is that the level of endogenous antioxidant molecules was reduced, and hence the GSH/GSSG ratio showed this Int. J. Mol. Sci. 2019, 20, 1662 17 of 24 same reduction. Although different authors have considered that a high level of Homo-Cys in plasma is related to NAFLD, being a critical factor in the pathogenesis groups [55,56], in this study, the liver content of Homo-Cys was highest in the animals of the N5 group; however, this cannot be considered a negative effect, since the animals of this group also had a high liver content of methionine. The GSH/GSSG ratio was altered by the treatments, type of diet, and supplementation with spinach, being lowered by the antioxidant effect of spinach. A plausible explanation could be that the high antioxidant content in the diet determined a lower in vivo response for maintaining the redox balance, and hence the use of reduced glutathione signiﬁcantly decreased. This effect agrees with the ﬁndings of other authors who used tomato extract, lycopene, and astaxanthin and obtained a reduction in the redox ratio [57,58]. 4. Materials and Methods 4.1. Spinach and Preparation of Diets Spinach (Spinacia oleracea) was obtained from a local supermarket as a fresh-cut product. The edible part was boiled for 10 minutes to remove oxalic acid, the water was discarded, and then the cooked spinach was lyophilized and ground. The powdered samples were stored at 4 ◦C until their use. The total content of carotenoids in spinach powder was determined by HPLC and was 1750 µg/g, showing the following amounts of the individual compounds: 228 µg of neoxanthin/g, 292 µg of violaxanthin/g, 944 µg of lutein/g, 46 µg of α-carotene/g, and 225 µg of β-carotenene/g. The spinach-enriched diets were prepared by mixing the pulverized pellets of the standard diet (Teklad Global 14% Protein Rodent Maintenance Diet TD-2014; Harland Laboratories, Indianapolis, IN, USA) or the high-fat diet (Atherogenic rodent diet TD-02028; Harland Laboratories) with 2.5% and 5% freeze-dried spinach powder. Water was added to each of the mixtures until a mass was formed which was not sticky. The pellets were then prepared using a pastry bag and dried in a tray dryer at 60 ◦C for 21 h. Dried pellets with spinach were packed in polythene bags and stored in the refrigerator until they were used. 4.2. Animals and Experimental Design The experimental protocol of this work was approved by the Ethical Committee of Animal Experimentation of the University of Murcia and by the General Directorate of Livestock and Fisheries of the C.A.R.M. (No. A1320140701, permitted on 23 July 2014). Based on preliminary studies, the sample size (n) was estimated by comparing two proportions using the following formula [59]: (cid:104) z1−α/2 (cid:112){2p(1 − p (cid:111) + z1−β n = (cid:112){PA(1 − PA) + PB(1 − PB)} δ2 (cid:105)2 where z1−α/2 is the zα value corresponding to the desired risk, z1−β is the zβ value corresponding to the statistical power, PA is the value of the proportion in the control group, PB is the value of the proportion in the group of the treatment, p is the average of the two proportions PA and PB, and δ2 is the PA − PB. The sample size obtained was adjusted for 10% attrition. It was estimated using the following formula [60]: Sample adjusted to the attrition = n (cid:18) 1 (cid:19) 1 − R where n is the number of subjects without attrition and R is the expected proportion of attrition. Forty-four male adult Sprague-awley rats (8 weeks of age) were grouped into two groups (n = 22) according to their diet: standard diet (diet N) or a high-fat diet (diet H). These diets were administered for two weeks; after this time, the animals were classiﬁed into six experimental groups. There were two control groups (n = 6 rats/group), a standard diet (NC) and a high fat diet (HC), and four experimental groups (n = 8 rats/group): N5 (standard diet + 5% spinach), N2.5 (standard diet + 2.5% spinach), Int. J. Mol. Sci. 2019, 20, 1662 18 of 24 H5 (high fat diet + 5% spinach), and H2.5 (high fat diet + 2.5% spinach). The experimental period was ﬁve weeks (Figure 7). Body weight was registered weekly and feed intake and urinary and fecal excretions were recorded in the initial, middle, and ﬁnal parts of the experimental period, using metabolic cages for data collection. At the end of this period, the rats were sacriﬁced and the different biological samples (plasma, feces, urine, and liver) were obtained. All the samples were stored at −80 ◦C until the analytical processes were performed. 4.3. Histopathological Examination Figure 7. Experimental design of the study. Histological examinations were carried out in the Pathological Anatomy Service of the Veterinary Hospital of the University of Murcia. Samples of the liver from each animal were taken and each sample was divided into two parts. One was ﬁxed in formalin and parafﬁn-embedded, and 4-µm-thick sections were obtained and stained with hematoxylin–eosin. The other was snap-frozen in 2-methylbutane cooled in liquid nitrogen and stored at −70 ◦C until use. Frozen 5-µm-thick sections were cut with a cryostat at −20 ◦C and stained with Sudan III for lipids detection. The liver sections were examined (without knowledge of their experimental group) and given an estimated score for the severity of interstitial hepatitis: 0 = no microscopic lesions; 1 = mild interstitial hepatitis; 2 = moderate multifocal interstitial hepatitis; 3 = severe multifocal interstitial hepatitis. The degree of steatosis was evaluated as follows: 0 = no vacuolar degeneration; 1 = less than 25% of hepatocytes affected; 2 = 25–50% of hepatocytes affected; 3 = 50–75% of hepatocytes affected; 4 = more than 75% of hepatocytes affected. 4.4. Plasma Biochemical Parameters Glucose, total protein, insulin, total cholesterol, fractions of HDL-cholesterol, LDL-cholesterol, and VLDL-cholesterol, TG, and activity of the enzymes AST and ALT were analyzed in plasma samples, using an automatic analyzer (AU 600 Olympus Life, Hamburg, Germany) in the Veterinary Hospital of the University of Murcia. 4.5. Determination of Biomarkers of Inﬂammation and Oxidative Stress The TNF-α and adiponectin levels in plasma were determined using a commercial ELISA kit (Single Analyte ElisarrayTM kits for Rat; QIAGEN, SA Biosciences, Frederick, MD, USA). The plasma antioxidant capacity, expressed as mmol of Trolox equivalents (TE)/L, was determined by the ORAC technique, using a multimodal microplate reader (Synergy HT BioTek, Winooski, VT, USA) [61]. Urinary excretion of 15-F2t-isoprostane (8-epi-PGF2α) was determined with an ELISA kit (OxySelectTm-epi-PGF2α Elisa Kit, Cell Biolabs), and the creatinine concentration was used to normalize the constituents [62]. These parameters were measured at the beginning and end of the intervention period. Int. J. Mol. Sci. 2019, 20, x FOR PEER REVIEW 18 of 25 where 𝑛 is the number of subjects without attrition and 𝑅 is the expected proportion of attrition. Forty-four male adult Sprague-awley rats (8 weeks of age) were grouped into two groups (n = 22) according to their diet: standard diet (diet N) or a high-fat diet (diet H). These diets were administered for two weeks; after this time, the animals were classified into six experimental groups. There were two control groups (n = 6 rats/group), a standard diet (NC) and a high fat diet (HC), and four experimental groups (n = 8 rats/group): N5 (standard diet + 5% spinach), N2.5 (standard diet + 2.5% spinach), H5 (high fat diet + 5% spinach), and H2.5 (high fat diet + 2.5% spinach). The experimental period was five weeks (Figure 7). Body weight was registered weekly and feed intake and urinary and fecal excretions were recorded in the initial, middle, and final parts of the experimental period, using metabolic cages for data collection. At the end of this period, the rats were sacrificed and the different biological samples (plasma, feces, urine, and liver) were obtained. All the samples were stored at −80 °C until the analytical processes were performed. 4.3. Histopathological Examination Histological examinations were carried out in the Pathological Anatomy Service of the Veterinary Hospital of the University of Murcia. Samples of the liver from each animal were taken and each sample was divided into two parts. One was fixed in formalin and paraffin-embedded, and 4-μm-thick sections were obtained and stained with hematoxylin–eosin. The other was snap-frozen in 2-methylbutane cooled in liquid nitrogen and stored at −70 °C until use. Frozen 5-μm-thick sections were cut with a cryostat at −20 °C and stained with Sudan III for lipids detection. The liver sections were examined (without knowledge of their experimental group) and given an estimated score for the severity of interstitial hepatitis: 0 = no microscopic lesions; 1 = mild interstitial hepatitis; 2 = moderate multifocal interstitial hepatitis; 3 = severe multifocal interstitial hepatitis. The degree of steatosis was evaluated as follows: 0 = no vacuolar degeneration; 1 = less than 25% of hepatocytes affected; 2 = 25–50% of hepatocytes affected; 3 = 50–75% of hepatocytes affected; 4 = more than 75% of hepatocytes affected. Figure 7. Experimental design of the study. 4.4. Plasma Biochemical Parameters Glucose, total protein, insulin, total cholesterol, fractions of HDL-cholesterol, LDL-cholesterol, and VLDL-cholesterol, TG, and activity of the enzymes AST and ALT were analyzed in plasma samples, using an automatic analyzer (AU 600 Olympus Life, Hamburg, Germany) in the Veterinary Hospital of the University of Murcia. 4.5. Determination of Biomarkers of Inflammation and Oxidative Stress The TNF-α and adiponectin levels in plasma were determined using a commercial ELISA kit (Single Analyte ElisarrayTM kits for Rat; QIAGEN, SA Biosciences, Frederick, MD, USA). The plasma antioxidant capacity, expressed as mmol of Trolox equivalents (TE)/L, was determined by the ORAC technique, using a multimodal microplate reader (Synergy HT BioTek, Winooski, VT, USA) [61]. Urinary excretion of 15-F2t-isoprostane (8-epi-PGF2α) was determined with an ELISA kit Int. J. Mol. Sci. 2019, 20, 1662 19 of 24 4.6. Analysis of Carotenoids in the Spinach and Liver The analysis of carotenoids was carried out according to the procedure described previously by our research group [9]. Carotenoids were extracted twice with tetrahydrofuran/methanol (1/1, v/v) containing 0.1% butylhydroxytoluene. The combined extracts were brought to dryness in a rotary evaporator and the residues were suspended in 5 mL of (TBME/MeOH). The carotenoids were analyzed by HPLC (Agilent 1200, Waldbronn, Germany), with a C30 column (250 × 4.6 mm, 5 µm i.d.) (Trentec, Gerlingen, Germany) at 17 ◦C, using a TBME (A) and MeOH (B) mobile phase with a ﬂow rate of 1 mL/min. The gradient used began with 2% A in B, reaching 35% A at 35 min, 60% A at 45 min, and 60% A at 56 min, before returning to the initial conditions for 4 min before the next injection. Detection of the carotenoids was carried out using a diode array detector (DAD) system at 450 nm. Standard curves were prepared using reference standards for quantiﬁcation. 4.7. Analysis of Total Dietary Fiber (TDF) and Total Phenolic Compounds (TPC) in Feed The TDF was determined according to the AOAC procedure (985.29) (1990) described by Prosky et al. [63]. The TPC were determined, using Folin Ciocalteu’s Phenol reagent, according to Hirawan et al. [64]. 4.8. Analysis of Lipids in the Liver The contents of total fat, fatty acids, and cholesterol were determined in the rat livers. Total fat was analyzed by the Soxhlet method [65], using ethyl ether as solvent. Fatty acids and cholesterol were analyzed using a Sigma Aldrich Lipid Extraction Kit (MAK174, St. Louis, MO, USA) and Cholesterol Extraction Kit (MAK175), respectively, following the extraction method of Folch et al. [66] and the procedures described by the manufacturer. The quantiﬁcation of fatty acids and cholesterol was performed on a GC (Agilent GC 7890A, Palo Alto, CA, USA) equipped with a ﬂame ionization detector (FID), as reported by Martin-Pozuelo [9]. 4.9. Study of the Expression of Genes involved in Fatty Liver Disease Liver samples were used for the analysis of gene expression, following the procedure previously described by Martin-Pozuelo et al. [9]. Real-time PCR analyses were carried out according to the manufacturer’s instructions, using a 96-well PCR array for the evaluation of fatty liver disease genes (PARN-157ZD-24, Qiagen, SABiosciences, Frederick, MD, USA). Relative gene expression was determined according to the comparative Ct method. The gene expression was only investigated in rats of the control groups (NC and HC) and in those that had ingested high levels of spinach (N5 and H5). 4.10. Analysis of Liver Metabolites by HPLC-MS The extraction of metabolites was based on the work described previously by our research group [8]. After extraction, the samples were injected into an Agilent 1200 series HPLC instrument (Agilent Technologies, California, USA) coupled to an Agilent 6120 single quadrupole mass spectrometer with an orthogonal ESI source. In order to avoid potential degradation of metabolites, the samples were prepared shortly before chromatographic analysis. The analysis of metabolites was only carried out with the liver of animals belonging to the control groups (NC and HC) and the groups with high intake of spinach (N5 and H5) 4.11. Statistical Analysis All analytical determinations were performed in triplicate and the data were expressed as the mean ± standard deviation of the results obtained. In all cases, the normality and equality of variances were tested. A one-way ANOVA with repeated measures was applied, with a post-hoc test to determine differences among the means of all the analytical determinations: Tukey’s test or the Games–Howell test Int. J. Mol. Sci. 2019, 20, 1662 20 of 24 according to the case. For the data of food and drink intake, weight gain and excreta, and cholesterol and fatty acids in the liver, the two-sample Student’s t test was performed to compare groups N and H. In addition, for each biochemical parameter and biomarker of inﬂammation or oxidative stress analyzed, a paired Student’s t test was performed to compare the values at the beginning and end of the experiment. The level of signiﬁcance was p < 0.05. For the analysis of liver metabolites, the concentrations were normalized according to the weight of the tissues and a two-way ANOVA was carried out, followed by a post-hoc Tukey HSD test; its familywise error rate was corrected using the Benjamini–Hochberg false discovery rate (FDR) with a 5% proportion of false discovery [67]. In addition, a principal componentaAnalysis was applied. The statistical analysis was carried out with the IBM Statistical Package for the Social Sciences (SPSS), version 24.0. The signiﬁcance of the relative gene expression, taking NC as the control group, was determined with Partek®Genomics Suite 6.6., considering the genes with differential expression to be those satisfying the following criteria: a fold change > 2 or < −2 and p < 0.05. 5. Conclusions We can summarize that, in rats with steatosis, the consumption of spinach and the accumulation of carotenoids in the liver reduced the content of SAFA, the ω-6/ω-3 fatty acid ratio and cholesterol, and increased MUFA and PUFA, as well as modiﬁed the expression of genes related to the fatty liver condition, increasing the abundance of proteins involved in the metabolism of fatty acids and cholesterol, mainly through the overexpression of PPARs. Moreover, the supplementation of spinach improved the expression of genes related to inﬂammatory response. Therefore, based on the results obtained, spinach can be considered as part of a dietary strategy in the control and treatment of NAFLD as a natural and safe source of carotenoids. However, this research underlines the need for further investigation to elucidate the effect on the content of speciﬁc enzymes and to determine the doses that could regulate lipid metabolism in the treatment of this pathology. Author Contributions: L.I.E.-T. has made substantial contributions to the acquisition and analysis of the samples and interpretation of the data, in addition to writing the manuscript; G.M.-P., R.G.-B., I.N.-G., and M.S. have collaborated in the animal study, collected the biological samples and collaborated in the analysis; F.-J.P. has contributed to the histopathological examinations of the liver. A.S. has helped with the analysis of the data of metabolomics study. J.G.-A. and M.J.P.-C. designed the experiment, checked all data, and revised the manuscript. All authors read and approved the ﬁnal manuscript. Funding: This study was supported by the projects MINECO (Spanish)/FEDER-EU BIO2012-38103 and CONSOLIDER Fun-C-Food CSD2007. Acknowledgments: Laura Inés Elvira-Torales thanks the Mexican Public Education Secretary for a Doctoral Scholarship (ITESTB-003 PRODEP Program). Conﬂicts of Interest: The authors declare no competing ﬁnancial interest. All authors have agreed to its publication in Nutrients. Abbreviations GC HPLC-MS DAD PCR Asp Ala Ser Pro Val Thr Cys Glu Ile/Leu/OH-Pro Gas chromatography High-performance liquid chromatography–mass spectrometry Diode array detector Polymerase chain reaction L-aspartic acid L-alanine L-serine L-proline L-valine L-threonine L-cysteine L-glutamic acid L-isoleucine/L-leucine/L-hydroxyproline Int. J. Mol. Sci. 2019, 20, 1662 21 of 24 Met Gly Homo-Cys Lys/Gln Asn His Phe Arg Tyr Trp GSH GSSG NAD NADH NADP NADPH Hypoxan CMP AMP GMP UDP GDP CTP ATP ITP References L-methionine Glycine L-homocysteine L-lysine/ L-glutamine L-asparagine L-histidine L-phenylalanine L-arginine L-tyrosine L-tryptophan L-glutathione L-glutathione oxidised form Nicotinamide adenine dinucleotide oxidised form Nicotinamide adenine dinucleotide reduced form Nicotinamide adenine dinucleotide phosphate oxidised form Nicotinamide adenine dinucleotide phosphate reduced form Hypoxanthine Cytidine monophosphate Adenosine monophosphate Guanosine monophosphate Uridine diphosphate Guanosine diphosphate Cytidine triphosphate Adenosine triphosphate Inosine triphosphate 1. 2. 3. 4. 5. Bellentani, S.; Scaglioni, F.; Marino, M.; Bedogni, G. Epidemiology of non-alcoholic fatty liver disease. Dig. Dis. 2010, 28, 155–161. [CrossRef] [PubMed] Day, C.P.; James, O.F. Steatohepatitis: A tale of two “hits”? Gastroenterology 1998, 114, 842–845. [CrossRef] Charatcharoenwitthya, P.; Lindor, K.D. Lipid metabolism and control in nonalcoholic fatty liver disease. In Nutrition, Diet Therapy, and the Liver, 1st ed.; Preedy, V.R., Lakshman, R., Srirajaskanthan, R., Watson, R.R., Eds.; CRC Press: Boca Raton, FL, USA, 2010; pp. 67–80. ISBN 9781138111790. Roberts, J.L.; Moreau, R. Functional properties of spinach (Spinacia oleracea L.) phytochemicals and bioactives. Food Funct. 2016, 7, 3337–3353. [CrossRef] [PubMed] Vitaglione, P.; Morisco, F.; Caporaso, N.; Fogliano, V. Dietary antioxidant compounds and liver health. Crit. Rev. Food. Sci. Nutr. 2010, 44, 575–586. [CrossRef] 6. Murillo, A.G.; DiMarco, D.M.; Fernandez, M.L. The potential of non-provitamin A carotenoids for the 7. 8. prevention and treatment of non-alcoholic fatty liver disease. Biology 2016, 5, 42. [CrossRef] Yilmaz, B.; Sahin, K.; Bilen, H.; Bahcecioglu, I.H.; Bilir, B.; Ashraf, S.; Halazun, K.J.; Kucuk, O. Carotenoids and non-alcoholic fatty liver disease. Hepatobiliary Surg. Nutr. 2015, 4, 161–171. [CrossRef] [PubMed] Bernal, C.; Martín-Pozuelo, G.; Lozano, A.B.; Sevilla, A.; García-Alonso, J.; Canovas, M.; Periago, M.J. Lipid biomarkers and metabolic effects of lycopene from tomato juice on liver of rats with induced hepatic steatosis. J. Nutr. Biochem. 2013, 24, 1870–1881. [CrossRef] 9. Martín-Pozuelo, G.; Navarro-González, I.; González-Barrio, R.; Santaella, M.; García-Alonso, J.; Hidalgo, N.; Gómez-Gallego, C.; Ros, G.; Periago, M.J. The effect of tomato juice supplementation on biomarkers and gene expression related to lipid metabolism in rats with induced hepatic steatosis. Eur. J. Nutr. 2015, 54, 933–944. [CrossRef] 10. Kim, J.E.; Clark, R.M.; Park, Y.; Lee, J.; Fernandez, M.L. Lutein decreases oxidative stress and inﬂammation in liver and eyes of guinea pigs fed a hypercholesterolemic diet. Nutr. Res. Pract. 2012, 6, 113–119. [CrossRef] 11. Murillo, A.G.; Aguilar, D.; Norris, G.H.; DiMarco, D.M.; Missimer, A.; Hu, S.; Smyth, J.A.; Gannon, S.; Blesso, C.N.; Luo, Y.; et al. Compared with powdered lutein, a lutein nanoemulsion increases plasma and liver lutein, protects against hepatic steatosis, and affects lipoprotein metabolism in guinea pigs. J. Nutr. 2016, 146, 1961–1969. [CrossRef] Int. J. Mol. Sci. 2019, 20, 1662 22 of 24 12. Her, G.M.; Pai, W.Y.; Lai, C.Y.; Hsieh, Y.W.; Pang, H.W. Ubiquitous transcription factor YY1 promotes zebrafish liver steatosis and lipotoxicity by inhibiting CHOP-10 expression. Biochim. Biophys. Acta 2013, 1831, 1037–1051. [CrossRef] [PubMed] 13. Kobori, M.; Ni, Y.; Takahashi, Y.; Watanabe, N.; Sugiura, M.; Ogawa, K.; Nagashimada, M.; Kaneko, S.; Naito, S.; Ota, T. β-Cryptoxanthin alleviates diet-induced nonalcoholic steatohepatitis by suppressing inﬂammatory gene expression in mice. PLoS ONE 2014, 9, e98294. [CrossRef] 14. Han, H.; Cui, W.; Wang, L.; Xiong, Y.; Liu, L.; Sun, X.; Hao, L. Lutein prevents high fat diet-induced atherosclerosis in ApoE-deﬁcient mice by inhibiting NADPH oxidase and increasing PPAR expression. Lipids 2015, 50, 261–273. [CrossRef] [PubMed] 15. Goossens, N.; Jornayvaz, F.R. Translational Aspects of Diet and Non-Alcoholic Fatty Liver Disease. Nutrients 2017, 9, 1077. [CrossRef] [PubMed] 16. Ni, Y.; Nagashimada, M.; Zhuge, F.; Zhan, L.; Nagata, N.; Tsutsui, A.; Nakanuma, Y.; Kaneko, S.; Ota, T. Astaxanthin prevents and reverses diet-induced insulin resistance and steatohepatitis in mice: A comparison with vitamin E. Sci. Rep. 2015, 5, 17192. [CrossRef] 17. Murillo, A.G.; Fernandez, M.L. Potential of dietary non-provitamin A carotenoids in the prevention and treatment of diabetic microvascular complications. Adv. Nutr. 2016, 7, 14–24. [CrossRef] [PubMed] 18. León-Goñi, A.C.; Blanco, D.; Peña, A.; Ronda, M.; González, B.O.; Arteaga, M.E.; Bada, A.M.; González, Y.; Mancebo, A. Hematological and biochemical parameters in Sprague Dawley laboratory rats breed in CENPALAB, Cenp: SPRD. Rev. Electron. Vet. 2011, 12, 1–10. Ihedioha, J.I.; Noel-Uneke, O.A.; Ihedioha, T.E. Reference values for the serum lipid proﬁle of albino rats (Rattus norvegicus) of varied ages and sexes. Comp. Clin. Pathol. 2013, 22, 93–99. [CrossRef] 19. 20. Cabré-Gelada, E.; Peña-Quintana, L.; Virgili-Casas, N.; Tomo, V. Nutrición en las enfermedades hepatobiliares. In Tratado de Nutrición: Nutrición y enfermedad, 3rd ed.; Gil-Hernández, A., Ed.; Editorial Médica Panamericana: Madrid, España, 2017; p. 884. ISBN 9788491101949. 21. Maiani, G.; Periago Castón, M.J.; Catasta, G.; Toti, E.; Cambrodón, I.G.; Bysted, A.; Granado-Lorencio, F.; Olmedilla-Alonso, B.; Knuthsen, P.; Valoti, M.; et al. Carotenoids: Actual knowledge on food sources, intakes, stability and bioavailability and their protective role in humans. Mol. Nutr. Food Res. 2009, 53, S194–S218. [CrossRef] 22. Massey, J.B. Kinetics of transfer of alpha-tocopherol between model and native plasma lipoproteins. Biochim. Biophys. Acta 1984, 793, 387–392. [CrossRef] 23. Ko, S.H.; Park, J.H.; Kim, S.Y.; Lee, S.W.; Chun, S.S.; Park, E. Antioxidant effects of spinach (Spinacia oleracea L.) supplementation in hyperlipidemic rats. Prev. Nutr. Food Sci. 2014, 19, 19–26. [CrossRef] [PubMed] 24. Makon-Sébastien, N.; Francis, F.; Eric, S.; Henri, V.P.; François, L.J.; Laurent, P.; Yves, B.; Serge, C. Lycopene modulates THP1 and CaCO2 cells inﬂammatory state through transcriptional and nontranscriptional processes. Mediat. Inﬂamm. 2014, 2014, 507272. [CrossRef] Sharoni, Y.; Linnewiel-Hermoni, K.; Khanin, M.; Salman, H.; Veprik, A.; Danilenko, M.; Levy, J. Carotenoids and apocarotenoids in cellular signaling related to cancer: A review. Mol. Nutr. Food Res. 2016, 56, 259–269. [CrossRef] [PubMed] 25. 26. Videla, L.A.; Rodrigo, R.; Araya, J.; Poniachik, J. Oxidative stress and depletion of hepatic long-chain polyunsaturated fatty acids may contribute to nonalcoholic fatty liver disease. Free Radic. Biol. Med. 2004, 37, 1499–1507. [CrossRef] [PubMed] 27. Pardo, V.; Gonzalez-Rodriguez, A.; Muntane, J.; Kozma, S.C.; Valverde, A.M. Role of hepatocyte S6K1 in palmitic acid-induced endoplasmic reticulum stress, lipotoxicity, insulin resistance and in oleic acid-induced protection. Food Chem. Toxicol. 2015, 80, 298–309. [CrossRef] 28. Reyes-Quiroz, M.E.; Alba, G.; Saenz, J.; Santa-María, C.; Geniz, I.; Jiménez, J.; Ramírez, R.; Martín-Nieto, J.; Pintado, E.; Sobrino, F. Oleic acid modulates mRNA expression of liver X receptor (LXR) and its target genes ABCA1 and SREBP1c in human neutrophils. Eur. J. Nutr. 2014, 53, 1707–1717. [CrossRef] [PubMed] 29. Bell, J.G.; McEvoy, J.; Tocher, D.R.; Sargent, J.R. Depletion of alpha-tocopherol and astaxanthin in Atlantic salmon (Salmo salar) affects autoxidative defense and fatty acid metabolism. J. Nutr. 2000, 130, 1800–1808. [CrossRef] [PubMed] 30. Da Silva-Santi, L.; Antunes, M.; Caparroz-Assef, S.; Carbonera, F.; Masi, L.; Curi, R.; Visentainer, J.; Bazotte, R. Liver fatty acid composition and inﬂammation in mice fed with high-carbohydrate diet or high-fat diet. Nutrients 2016, 8, 682. [CrossRef] Int. J. Mol. Sci. 2019, 20, 1662 23 of 24 31. Ferramosca, A.; Zara, V. Modulation of hepatic steatosis by dietary fatty acids. World J. Gastroenterol. 2014, 20, 1746–1755. [CrossRef] [PubMed] 32. Monteiro, J.; Leslie, M.; Moghadasian, M.H.; Arendt, B.M.; Allard, J.P.; Ma, D.W. The role of n-6 and n-3 polyunsaturated fatty acids in the manifestation of the metabolic syndrome in cardiovascular disease and non-alcoholic fatty liver disease. Food Funct. 2014, 5, 426–435. [CrossRef] 33. Nicolle, C.; Cardinault, N.; Aprikian, O.; Busserolles, J.; Grolier, P.; Rock, E.; Demigné, C.; Mazur, A.; Scalbert, A.; Amouroux, P.; et al. Effect of carrot intake on cholesterol metabolism and on antioxidant status in cholesterol-fed rat. Eur. J. Nutr. 2003, 42, 254–261. [CrossRef] [PubMed] 35. 34. Qiu, X.; Gao, D.H.; Xiang, X.; Xiong, Y.F.; Zhu, T.S.; Liu, L.G.; Sun, X.F.; Hao, L.P. Ameliorative effects of lutein on non-alcoholic fatty liver disease in rats. World J. Gastroenterol. 2015, 21, 8061–8072. [CrossRef] Jia, Y.; Wu, C.; Kim, J.; Kim, B.; Lee, S.J. Astaxanthin reduces hepatic lipid accumulations in high-fat fed C57BL/6J mice via activation of peroxisome proliferator-activated receptor (PPAR) alpha and inhibition of PPAR gamma and Akt. J. Nutr. Biochem. 2016, 28, 9–18. [CrossRef] 36. Piña-Zentella, R.M.; Rosado, J.L.; Gallegos-Corona, M.A.; Madrigal-Pérez, L.A.; García, O.P.; Ramos-Gomez, M. Lycopene improves diet-mediated recuperation in rat model of nonalcoholic fatty liver disease. J. Med. Food. 2016, 19, 607–614. [CrossRef] [PubMed] 37. Periago, M.J.; Martín-Pozuelo, G.; Gonzalez-Barrio, R.; Santaella, M.; Gómez, V.; Vazquez, N.; Navarro-González, I.; García-Alonso, J. Effect of tomato juice consumption on the plasmatic lipid proﬁle, hepatic HMGCR activity, and fecal short chain fatty acids content of rats. Food Funct. 2016, 7, 4460–4467. [CrossRef] [PubMed] Fuhrman, B.; Elis, A.; Aviram, M. Hypocholesterolemic effect of lycopene and β-carotene is related to suppression of cholesterol synthesis and augmentation of LDL receptor activity in macrophages. Biochem. Biophys. Res. Commun. 1997, 233, 658–662. [CrossRef] [PubMed] 38. 39. Thorpe, C.; Kim, J.J. Structure and mechanism of action of the acyl-CoA dehydrogenases. FASEB J. 1995, 9, 718–725. [CrossRef] 40. Gyamﬁ, D.; Patel, V. Liver metabolism: Biochemical and molecular regulations. In Nutrition, Diet Therapy, and the Liver, 1st ed.; Preedy, V.R., Lakshman, R., Srirajaskanthan, R., Watson, R.R., Eds.; CRC Press: Boca Raton, FL, USA, 2010; pp. 3–15. ISBN 9781138111790. 41. Rühl, R.; Landrier, J.F. Dietary regulation of adiponectin by direct and indirect lipid activators of nuclear hormone receptors. Mol. Nutr. Food Res. 2016, 60, 175–184. [CrossRef] 42. Abdelmegeed, M.A.; Yoo, S.H.; Henderson, L.E.; Gonzalez, F.J.; Woodcroft, K.J.; Song, B.J. PPARα Expression Protects Male Mice from High Fat–Induced Nonalcoholic Fatty Liver. J. Nutr. 2011, 141, 603–610. [CrossRef] [PubMed] Sharoni, Y.; Agbaria, R.; Amir, H.; Ben-Dor, A.; Bobilev, I.; Doubi, N.; Giat, Y.; Hirsh, K.; Izumchenko, G.; Khanin, M.; et al. Modulation of transcriptional activity by antioxidant carotenoids. Mol. Asp. Med. 2003, 24, 371–384. [CrossRef] 43. 44. Wiersma, H.; Nijstad, N.; de Boer, J.F.; Out, R.; Hogewerf, W.; Van Berkel, T.J.; Kuipers, F.; Tietge, U.J. Lack of Abcg1 results in decreased plasma HDL cholesterol levels and increased biliary cholesterol secretion in mice fed a high cholesterol diet. Atherosclerosis 2009, 206, 141–147. [CrossRef] 45. Musso, G.; Gambino, R.; Cassader, M. Recent insights into hepatic lipid metabolism in non-alcoholic fatty liver disease (NAFLD). Prog. Lipid Res. 2009, 48, 1–26. [CrossRef] [PubMed] 46. Tainaka, T.; Shimada, Y.; Kuroyanagi, J.; Zang, L.; Oka, T.; Nishimura, Y.; Nishimura, N.; Tanaka, T. Transcriptome analysis of anti-fatty liver action by Campari tomato using a zebraﬁsh diet-induced obesity model. Nutr. Metab. 2011, 8, 88. [CrossRef] [PubMed] 47. Elvira-Torales, L.I.; Navarro-González, I.; González-Barrio, R.; Martín-Pozuelo, G.; Doménech, G.; Seva, J.; García-Alonso, J.; Periago-Castón, M.J. Tomato juice supplementation inﬂuences the gene expression related to steatosis in rats. Nutrients 2018, 10, 1215. [CrossRef] [PubMed] 48. Navarro-González, I.; García-Alonso, J.; Periago, M.J. Bioactive compounds of tomato: Cancer chemopreventive 49. effects and influence on the transcriptome in hepatocytes. J. Funct. Foods 2018, 42, 271–280. [CrossRef] Streba, L.A.M.; Vere, C.C.; Rogoveanu, I.; Streba, C.T. Nonalcoholic fatty liver disease, metabolic risk factors, and hepatocellular carcinoma: An open question. World J. Gastroenterol. 2015, 21, 4103–4110. [CrossRef] [PubMed] Int. J. Mol. Sci. 2019, 20, 1662 24 of 24 50. Fabbrini, E.; Sullivan, S.; Klein, S. Obesity and nonalcoholic fatty liver disease: Biochemical, metabolic and clinical implications. Hepatology 2010, 51, 679–689. [CrossRef] [PubMed] 51. Gavrilova, O.; Haluzik, M.; Matsusue, K.; Cutson, J.J.; Johnson, L.; Dietz, K.R.; Nicol, C.J.; Vinson, C.; Gonzalez, F.J.; Reitman, M.L. Liver peroxisome proliferator-activated receptor gamma contributes to hepatic steatosis, triglyceride clearance, and regulation of body fat mass. J. Biol. Chem. 2003, 278, 34268–34276. [CrossRef] [PubMed] 52. Kim, H.H.; Finck, B.N. PPARs and Nonalcoholic Fatty liver disease. Biochimie 2017, 136, 65–74. [CrossRef] 53. Xie, Z.; Li, H.; Wang, K.; Lin, J.; Wang, Q.; Zhao, G.; Jia, W.; Zhang, Q. Analysis of transcriptome and metabolome proﬁles alterations in fatty liver induced by high-fat diet in rat. Metabolism 2010, 59, 554–560. [CrossRef] [PubMed] Ip, B.C.; Liu, C.; Lichtenstein, A.H.; von Lintig, J.; Wang, X.D. Lycopene and apo-10(cid:48)-lycopenoic acid have differential mechanisms of protection against hepatic steatosis in β-carotene-9(cid:48),10(cid:48)-oxygenase knockout male mice. J. Nutr. 2015, 145, 268–276. [CrossRef] 54. 55. Bravo, E.; Palleschi, S.; Aspichueta, P.; Buqué, X.; Rossi, B.; Cano, A.; Napolitano, M.; Ochoa, B.; Botham, K.M. High fat diet-induced non alcoholic fatty liver disease in rats is associated with hyperhomocysteinemia caused by down regulation of the transsulphuration pathway. Lipids Health Dis. 2011, 10, 60. [CrossRef] [PubMed] 56. Ai, Y.; Sun, Z.; Peng, C.; Liu, L.; Xiao, X.; Li, J. Homocysteine Induces Hepatic Steatosis Involving ER Stress Response in High Methionine Diet-Fed Mice. Nutrients 2017, 9, 346. [CrossRef] 57. Leal, M.; Shimada, A.; Ruíz, F.; González de Mejía, E. Effect of lycopene on lipid peroxidation and glutathione-dependent enzymes induced by T-2 toxin in vivo. Toxicol. Lett. 1999, 109, 1–10. [CrossRef] 58. Anuradha, C.V. Astaxanthin, a marine carotenoid against hepatic oxidative stress: A systematic review. In The Liver: Oxidative Stress and Dietary Antioxidants, 1st ed.; Patel, V.B., Rajendram, R., Preedy, V.R., Eds.; Academic Press: London, UK, 2018; pp. 216–217. ISBN 978-3-319-56438-8. 59. Campbell, M.J.; Julious, S.A.; Altman, D.G. Estimating sample sizes for binary, ordered categorical, and continuous outcomes in two group comparisons. BMJ 1995, 311, 1145–1148. [CrossRef] [PubMed] 60. García-García, J.A.; Rending-Bernal, A.; López-Alvarenga, J.C. Cálculo del tamaño de la muestra en investigación en educación médica. Investigación en Educación Médica 2013, 2, 217–224. [CrossRef] 61. Prior, R.L.; Hoang, H.A.; Gu, L.; Wu, X.; Bacchiocca, M.; Howard, L.; Hampsch-Woodill, M.; Huang, D.; Ou, B.; Jacob, R. Assays for hydrophilic and lipophilic antioxidant capacity (oxygen radical absorbance capacity (ORACFL)) of plasma and other biological and food samples. J. Agric. Food Chem. 2003, 51, 3273–3279. [CrossRef] [PubMed] 62. Helger, R.; Rindfrey, H.; Hilgenfeldt, J. Direct estimation of creatinine in serum and in urine without deproteinization using a modiﬁed Jaffé method. Z. Klin. Chem. Klin. Biochem. 1974, 12, 344–349. [PubMed] 63. Prosky, L.; Asp, N.G.; Furda, I.; DeVries, J.W.; Schweizer, T.F.; Harland, B.F. Determination of total dietary ﬁber in foods and food products: Collaborative study. J. Assoc. Off. Anal. Chem. 1985, 68, 677–679. 64. Hirawan, R.; Diehl-Jones, W.; Beta, T. Comparative evaluation of the antioxidant potential of infant cereals produced from purple wheat and red rice grains and LC-MS analysis of their anthocyanins. J. Agric. Food Chem. 2011, 59, 12330–12341. [CrossRef] [PubMed] 65. Hijona, E.; Hijona, L.; Larzabal, M.; Sarasqueta, C.; Aldazabal, P.; Arenas, J.; Bujanda, L. Biochemical determination of lipid content in hepatic steatosis by the Soxtec method. World J. Gastroenterol. 2010, 16, 1495–1499. [CrossRef] [PubMed] Folch, J.; Lees, M.; Sloane-Stanley, G.H. A simple method for the isolation and puriﬁcation of total lipids from animal tissues. J. Biol. Chem. 1957, 226, 497–509. [PubMed] 66. 67. Benjamini, Y.; Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B-Methodol. 1995, 57, 289–300. [CrossRef] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.7150_ijbs.74123	Int. J. Biol. Sci. 2023, Vol. 19 Ivyspring International Publisher Research Paper International Journal of Biological Sciences 2023; 19(7): 2053-2066. doi: 10.7150/ijbs.74123 2053 Gli1 promotes the phenotypic transformation of valve interstitial cells through Hedgehog pathway activation exacerbating calcific aortic valve disease Yuming Huang1#, Chen Jiang2#, Liang Chen1#, Juanjuan Han3, Ming Liu2, Tingwen Zhou2, Nianguo Dong2, Kang Xu3 1. Department of Thoracic Surgery, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, China. 2. Department of Cardiovascular Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430022, China. 3. Hubei Engineering Technology Research Center of Chinese Materia Medica Processing, College of Pharmacy, Hubei University of Chinese Medicine, Wuhan 430065, China. # These authors contributed equally to this work  Corresponding authors: Prof. Nianguo Dong, Dean, Department of Cardiovascular Surgery, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China. (Email: [email protected]); Prof. Kang Xu, Hubei Engineering Technology Research Center of Chinese Materia Medica Processing, College of Pharmacy, Hubei University of Chinese Medicine, Wuhan 430065, China. (Email: [email protected]) © The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/). See http://ivyspring.com/terms for full terms and conditions. Received: 2022.04.18; Accepted: 2023.03.03; Published: 2023.04.09 Abstract Calcific aortic valve disease (CAVD) is the most prevalent human valve disease worldwide. Multiple factors induce "irreversible" pathological changes in the aortic valve leaflets, resulting in changes in cardiac hemodynamics, eventually leading to heart failure. However, no effective pharmaceutical interventions have been found and prosthetic valve replacement is the only curative approach. Glioma-associated oncogene 1 (Gli1) exerts a regulatory role on cardiovascular diseases, and it is already a therapeutic target to combat tumors. Our research aimed to explore the role and basic mechanism of Gli1 in CAVD, to pave the way for the discovery of effective drugs in the treatment of CAVD. Human aortic valve tissues were obtained to evaluate Gli1 expression and primary valve interstitial cells (VICs) were used to perform related experiments. The results showed that Gli1 promoted cell proliferation and significantly accelerated cell osteogenic transformation through the up-regulation of the osteogenic factors Runx2 and Alp, in turn through the AKT signaling pathway by targeting P130cas expression. Furthermore, Gli1 was activated by TGF-β and sonic hedgehog through the canonical and non-canonical Hedgehog signaling pathways in VICs. Our results indicated that Gli1 promoted cell proliferation and accelerated cell osteogenic transformation in VICs, providing a new strategy for the therapy of CAVD by targeting Gli1. Keywords: Calcific aortic valve disease (CAVD), glioma-associated oncogene 1 (Gli1), Hedgehog signaling pathway, AKT signaling pathway, osteogenic transformation Introduction its prevalence it is only 0.2% increases Calcified aortic valve disease (CAVD) is, by far, the most prevalent human valve disease worldwide. the Although middle-aged people, in octogenarians [1-3]. CAVD is characterized by a slow progressive fibro-calcific remodeling of the valve leaflets. As a consequence of that, hemodynamic changes happen the cardiovascular system contributing to heart failure [4]. So far, no effective pharmaceutical interventions have been found to stop in to 9.8% in or just slow down the progression of CAVD; thus when this disease reaches the “point of no return”, only aortic valve replacement is the solution to cure [5]. Recently, the use of transcatheter aortic valve replacement therapy has increased. However, despite the good results achieved, the way to success is still far. Therefore, it is of utmost importance to explore the mechanisms regulating CAVD initiation and progression to identify promising therapeutic targets. factor glioma-associated transcription The https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2054 oncogene 1 (Gli1) has been recognized as a key nuclear executor at the distal end of the hedgehog signaling pathway. It plays a crucial role in regulating many biological processes, such as differentiation, proliferation, and apoptosis [6]. Previous studies on the hedgehog pathway mostly focus on the embryonic development and neural tissue homeostasis, or cancer stemness, but its regulatory role in cardiovascular disease has been continually ignored [7]. One research group in Germany identified the hedgehog signaling pathway as a specific process for activating adventitial fibroblasts during vasculogenesis, and promoting angiogenesis in the adult [8]. Another group previously found that the activation of the induces the proliferation of hedgehog pathway vascular intimal smooth muscle cells during hyperplasia [9]. French researchers also found that the in hedgehog pathway plays an cardio-protection ischemia-reperfusion injuries [10]. More functions of the Gli1 molecule in cardiovascular diseases have been explored. important role against cells, include: The pathogenesis of CAVD is now considered a complex pathophysiological process involving network modulation between multiple cells and multiple mechanisms. The predisposing and endothelial causative mechanisms dysfunction, lipid accumulation, sterile inflammation, excessive production or degradation of the extracellular matrix, excessive proliferation of myofibroblastic/osteoblastic immune differentiation of valvular interstitial cells (VICs) and subsequent calcification [11-13]. Therefore, our hypothesis is that Gli1 inevitably plays an important role in this disease due to the above characteristics of the valvular calcification pathological process and the role of the hedgehog signaling pathway found in cardiovascular disease. Gli1 might function to induce VICs proliferation and accelerate VIC calcification. The present research showed for the first time that hedgehog signaling pathway was activated during CAVD. The the proliferation of VICs and promoted the expression of Runx2 as the key factor in calcification. factor Gli1 induced target their proliferation. This work aimed to investigate the role of Gli1 in CAVD disease. Our results indicated that Gli1 significantly promoted the calcification of VICs and Furthermore, accelerated Gli1-AKT-P130cas axis played an important role in the osteogenic differentiation of VICs. The mechanism of Gli1 activation was also explored, revealing that and non-canonical Hedgehog both signaling pathways were involved in the upregulation of the Gli1 transcription factor in VICs. Thus, the inhibition of the expression of Gli1 might be a new and effective strategy for delaying the osteogenic canonical differentiation of human aortic VICs. Materials and Methods VIC isolation and culture Human aortic valves were obtained from patients undergoing heart valve surgery at the cardiovascular surgeon department of the Union hospital. Patients provided a written informed consent, and the study was approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology in Wuhan, China. Aortic valve leaflets were excised and rinsed according to our previous protocol. Then tissues were minced and placed in collagenase (150 units/mL) in Dulbecco’s Modified Eagle’s Medium (Hyclone, Logan, UT, USA) for 6–8 h at 37 °C. The cell suspension was obtained at the end of collagenase digestion by removing the undigested tissue with a 70 μm cell strainer. The cells were cultured in standard DMEM supplemented with 10% heat-inactivated FBS (Thermo Fisher Scientific, Waltham, MA, USA) and 150 U/mL penicillin/streptomycin (Hyclone). VICs were seeded at a density of 10,000 cells/cm2 in tissue culture flasks in a complete medium, which was changed every 3 days until VICs reached 90% confluence. Cells at the third or fourth passage were used in all the experiments. Adenovirus plasmid overexpression vehicles, Small-interfering RNA and cell transfection, and small molecule inhibitors The overexpression adenovirus plasmid of GLI family zinc finger 1 (GLI1) and control adenovirus were purchased from Vigene Biosciences (Shandong, China). Valve interstitial cells were transfected with small-interfering RNA (Ribobio bio, Guangzhou, China) targeting SMAD2, SM AD3, SMAD4, SHH, lipofectamine 3000 kits GLI1, BCAR1, using (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions after cells were grown to 60% confluence. All small molecule inhibitors were bought from Selleck Chemicals (Houston, TX, USA): Gli1 selective inhibitor GANT61 (S8075), AKT1 selective inhibitor MK-2206 (S1078), TGF-β type I receptor-selective inhibitor SB431542 (S1067), PTCH1 selective inhibitor GDC-0449(S1082). RNA extraction and qPCR analysis Total RNA was extracted with Trizol reagent reverse (Invitrogen, Carlsbad, CA) transcription product was used as a template to perform real-time polymerase chain reaction (PCR) on a StepOne Plus thermal cycler (Applied Biosystems, Foster City, CA) using a 2x SYBR Green qPCR Master Mix (High ROX) (Bimake, Houston, TX) following the and the https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2055 manufacturer’s guide. All the primers were referenced from the previous study, and primer sequences were shown in Supplementary Table 1. The final data were analyzed by the 2-ΔΔct method [14-16]. Western blotting For western blotting, cell samples were extracted and quantified then boiled at 95℃, 5min. The protein sample was separated on an 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel then transferred on a polyvinylidene fluoride membrane. Incubating primary antibodies overnight at 4℃, followed by the corresponding secondary antibodies for protein expression visualization. Primary antibodies were shown in Supplementary Table 2 [17]. For Immunofluorescence Staining tissue: Tissues were in 4% paraformaldehyde overnight, then dehydrated by 15% and 30% sucrose solution. Six microns frozen sections of formalin-fixed human valve tissues were cut and used for staining. fixed For cell: VICs seeded into 24-well plates at a density of 10000 cells/well were washed twice with PBS and fixed in 4% paraformaldehyde for 15 min. The fixative solution was removed by rinsing three times with PBS. Cells were permeabilized with 0.2% Triton X-100 for 10 min, washed three times with PBS, and blocked for 30 min with goat serum albumin (Boster, Wuhan, China). Immediately after blocking, cells were incubated with primary antibodies at 4°C overnight. After washing three times with PBS, samples were incubated with secondary antibodies (CST) in PBS for 60 min at room temperature. Then samples were washed twice with PBS and incubated with DAPI (Biofroxx GmbH, Einhausen, Germany) for 4 min to stain the nuclei. Samples were washed twice with PBS and then imaged on the Axio Observer Z1 microscope (Zeiss, Oberkochen, Germany) [18]. Chromatin Immunoprecipitation (ChIP) and ChIP-seq analysis Cells were cross-linked with 1% formaldehyde for 10 min at room temperature and the reaction was subsequently stopped by adding glycine to a final concentration of 0.125 M for another 10 min at room temperature. Then, the cells were washed twice in cold PBS and harvested in lysis buffer (1% Triton X-100, 0.1% SDS, 150 mM NaCl, 1 mM EDTA, 20 mM Tris, pH 8.0, and complete protease inhibitor mixture). The samples were sonicated 20 times (30 s on/off, 260 W) at 4 °C using a Diagenode Bioruptor. After pre-clearing overnight, samples were incubated with IgG (sigma, #R2655) previously conjugated with Dynabeads Protein G (Invitrogen, #10009D) for 3 h at 4 °C, and the immunocomplexes were washed twice with buffer 1 (20 mM Tris-HCl pH 8.0, 2 mM EDTA, 150 mM NaCl, 0.1% SDS, 1% Triton X-100), once with buffer 2 (20 mM Tris-HCl pH 8.0, 2 mM EDTA, 500 mM NaCl, 0.1% SDS, 1% Triton X-100), once with buffer 3 (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 250 mM LiCl, 1% Sodium deoxycholate, 1% NP-40), and twice with buffer 4 (10 mM Tris-HCl pH 8.0, 1 mM EDTA). Each wash was performed for 5 min at 4 °C rotation. DNA-protein complexes were eluted with 200 µl elution buffer (1% SDS and 0.1 M NaHCO3) and de-cross-linked by adding 0.2 M NaCl and shaking overnight at 65 °C. Then, digestion was performed using proteinase K, and the enriched DNA was purified by QIAquick PCR purification kit (Qiagen, #28104). The ChIP DNA libraries of VICs were sequenced on an Illumina HisSeq-3000. Reads were aligned to Ensembl human genome (hg19) using the Bowtie software. After format conversion and sorting, duplicate reads were removed by the rod up tool from the SAMtools package. The mapped sequence reads were processed using MACS version 1.4.2 against their matching control samples to identify the significant binding site of Gli1, and only peaks with P values < 10-5 were kept for further analyses. A Perl script annotatePeaks.pl in the HOMER package was used to associate ChIP peaks with nearby genes, and the visualization of the location and the shape of the called peaks was performed using Integrative Genomics Viewer (IGV). FACS for cell cycle and Cell Viability Assay VICs (passage 3) were cultured in 60 mm dishes until 80% confluency. Then, the medium was changed to DMEM with 2% FBS for 8 h before performing the different treatments. The samples were treated with trypsin and re-suspended in PBS at a density of 5×105/mL, followed by fixation in 70% precooled ethanol overnight at 4 °C, centrifugation, washing, and staining with PI/RNase staining buffer (BD Biosciences) for 30 min at 4 °C. Cell count at different phases of the cell cycle was performed using FCM as previously described [19]. Cell Viability Assay to according Cell viability was assessed with the Cell Counting Kit-8 (CCK-8) assay (Bimake.com, Houston, the manufacturer’s TX, USA) instructions. The cells were seeded at a density of 5000 cells/well in 24-well plates and cultured for 1–6 days. At the end of each time interval, cell samples were washed with PBS and incubated with a serum-free medium containing 10% CCK-8 reagent. After 4 h of incubation at 37°C in an atmosphere of 5% CO2, aliquots were pipetted into a 96-well plate and measured at 450 nm using an enzyme-labeling https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2056 instrument (Thermo Fisher Scientific) [20]. Edu assay Cell proliferation ability was tested by Edu assay (Ribobio bio, Guangzhou, China) according to the manufacturer’s instructions. The cells were seeded at a density of 10000 cells/well in 12-well plates. After different treatments, VICs were cultured in a medium containing Edu for 2h at 37°C in an atmosphere of 5% CO2. The fixed by 4% samples were paraformaldehyde for 15 min then washed several times, stained with apollo staining solution for 30min, observed under the fluorescent microscope [21]. cell identify differentially expressed genes (DEGs). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were also performed [23]. Statistical analysis RNA-seq results were analyzed using the R (version 3.5.1) according to a previous study and all other data were analyzed and expressed as the mean ± standard deviation (SD). Statistical comparisons were made by analysis of variance to evaluate differences among groups. A p-value less than 0.05 was considered statistically significant. Calcification analysis Results labeling instrument Cells were seeded into 12-well plates and grown for three days until 90% confluency and further cultured in an osteogenic differentiation medium (OM) containing 10mM β-glycerophosphate, 100nM dexamethasone, 50μg/ml vitamin C, 1% FBS, 100 IU/ml penicillin/streptomycin under different culture conditions for 15 days. The degree of cell calcification was measured using Alizarin Red S staining. In brief, cells were incubated in ddH20 and the amount of Alizarin Res S dye released from the extracellular matrix was measured at 450 nm using an [22]. For ex vivo enzyme osteogenic differentiation, human aortic valve leaflets were harvested from patients undergoing Bentall surgery due to acute type I aortic dissection. (or can be heart transplantation samples, should be normal via visibility). The samples (avoid the marginal area) were cut into 1mm1mm (or can be bigger, no more than 1 cm1cm) small pieces and treated with OM culture medium for at least 21 days. (Note: the small pieces were floaty culturing without attachment to the culture dishes). After OM induction, the small pieces can be transfected with re-comb adenovirus or chemical factors for one to two months (The treated time should be tested according to specific condition, no more than two months). The small pieces were fixed with 4% PFA for 30 min, then embedded in paraffin and 8-μm-thick sections were made. The specimens were stained with hematoxylin and eosin (H&E), histological immunofluorescence staining for specific antibodies. kossa, Von and RNA sequencing of VICs RNA sequencing (RNA-seq) was utilized to compare the mRNA profiles between different treatments for VICs. Isolated RNA was sent to BGI Tech Solutions Co., Ltd. (Shenzhen, China) for RNA-seq, which was performed on the BGISEQ-500 sequencer; all samples were sequenced in triplicate for confirmation purposes. Sequencing results were analyzed using the “R Project (version 3.5.1)” to Increased expression of Gli1 in human calcific aortic valve tissues Tissues from 20 pairs of calcific aortic valves and adjacent normal tissues were used to analyze Gli1 expression in CAVD conditions. Gli1 significantly increased in calcific valves as well as Runt-related transcription factor 2 (Runx2) and Alkaline Phospha- tase (ALP) which are markers of osteogenesis, as demonstrated by western blot (Fig. 1A and Fig. 1B). The qRT-PCR results were consistent with those of protein expression (Fig. 1C). HE, Von Kossa staining along with Gli1 immunohistochemical staining showed the presence of calcification nodules in calcified aortic valves and Gli1 was high expressed in the lesions (Fig. 1D). Dual immunofluorescence staining demonstrated that Gli1 was colocalized with α-smooth muscle actin (α-SMA), Vimentin, Runx2 in calcified aortic valves (Fig. 1E). These results suggested that Gli1 accumulation in aortic valves might play a role in VIC phenotypic transition and aortic valve calcification. Total RNA-seq analysis after the regulation of Gli1 expression in valve interstitial cells Gant61, a selective Gli1 inhibitor that inhibits the DNA binding function of Gli1, was used to block the function of Gli1. An Adenovirus containing Gli1 expression cassette was used to induce Gli1 factor overexpression. The box plot and correlation heat map showed the distribution of the gene expression among different groups. Gli1 enrichment group was differently scattered compared with the Gant61 group and control group (Fig. 2A and Fig. 2B). When Gli1 was overexpressed, 1224 genes were upregulated while 1264 genes were down-regulated (P<0.05). When Gli1 was blocked by Gant61, 1148 genes were upregulated while 1354 genes were down-regulated (P<0.05) (Fig. 2C). The Venn diagram of DEGs in the Gli1 group versus control, Gant61 group versus control, and Gli1 group versus Gant61 group, revealed https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2057 the presence of 308 differentially expressed genes (DEGs), which were used for further analysis (Fig. 2D). All these DEGs were analyzed by the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and GO function annotation (Fig. 2E and Fig. 2F). The top ten significant pathways and biological functions were listed. Genes related to cell differentiation, inflamma- tory response, extracellular matrix organization, and positive regulation of cell proliferation were modified. Gli1 promoted the proliferation of VICs and regulated the cell cycle of VICs The molecular structure of Gant61 is presented in Fig. 3A. IC50 was calculated to assess the toxicity of Gant61 on VICs, and the result indicated that Gant61 showed clear signs of toxicity at concentrations above 5 μM (Fig. 3B). Therefore, all subsequent in vitro experiments were performed using 5 μM Gant61. CCK-8 assay, ki-67, and Edu staining were performed to analyze the proliferation of VICs under Gli1 treatment with or without Gant61. Cell growth was almost completely inhabited when incubated in the presence of Gant61 (Fig. 3C). Ki-67 and Edu staining indicated that the proliferation of VICs was reduced (Fig. 3D). The statistical analysis showed that the proportion of positive cells was significantly different when performed with and without Gant61 (Fig. 3E). In addition, the overexpression of Gli1 in VICs rescued the inhibition of cell proliferation in the presence of Gant61, Ki-67 and Edu staining showed that the proportion of positive cells in the Gli1 group was the highest among the three groups and statistically significant (Fig. 3F and Fig. 3G). The cell cycle of VICs by flow cytometry showed that the treatment with Gli1 alone resulted in a unnormal cell growth (Fig. 3H). Our results indicated that the proportion of cells at the G0/G1 phase and S phase was significantly (Fig. 3I). qRT-PCR demonstrated that MKI67, CCNB1, CCNE1, and cell cycle-related genes, were significantly increased (Fig. 3J). Western blotting showed consistent results, since cyclinB1 protein significantly increased in the group with Gli1 overexpression and this effect was blocked by the Gli1 inhibitor Gant61 (Fig. 3K and Fig. 3L). increased Figure 1. Analysis of Gli1 expression in aortic valve tissues from CAVD and healthy samples. (A) and (B) western blotting analysis showed that Gli1, Runx2, ALP proteins increased significantly in CAVD samples compared with normal ones. (C) RT-PCR was applied to detect the mRNA expression level of Gli1 and Runx2 in 10 pairs of aortic valve samples from CAVD and normal patients. Data are expressed as the mean ± SD, n=10. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. (D) HE, Von Kossa staining along with Gli1 Immunohistochemical staining were performed in calcified aortic valves and normal ones. (E) Dual immunofluorescence staining of Gli1, α-SMA (α-smooth muscle actin), Vimentin, Runx2, and DAPI were performed in calcified aortic valves and normal ones. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2058 Figure 2. Total RNA-seq analysis after regulating the expression of Gli1 in VICs. (A) and (B) The box plot and correlation heat map the in the control group, Gli1 group, and Gant61 group. (C) volcano map of differentially expressed genes (DEGs) in these three different groups. Gli1 group versus control group: up-regulation 1224 genes and down-regulation 1264 genes; Gant61 group versus control group: up-regulation 1148 genes and down-regulation 1354 genes. FC (fold change) > 1 was accepted as positive DEGs. (D) Venn interaction of DEGs of the three groups compared with each other and 308 common DEGs were found. (E) KEGG pathway enrichment of common DEGs, bubble colors (deep) indicate the degree of enrichment (-Log10(P-value)), bubble size indicates gene counts matched the pathway enrichment, and rich factor indicates the matched gene counts in the integrated pathway background genes. (F) GO biological function enrichment of common DEGs, blue bars represent the degree of enrichment (-Log10(P-value)), red polyline showed the gene numbers involved. Gli1 accelerated the osteogenic differentiation of VICs The osteogenic differentiation medium (OM) was used in vitro to induce the calcification of VICs. In addition, cells with Gli1 overexpression alone or combined with Gant61 were subjected to Alizarin Red S staining to evaluate the formation of the calcified nodules. Gli1 treatment group and Gli1+Gant61 group showed a positive staining and were signifi- cantly different than the control group (Fig. 4A). The semi-quantification of calcification demonstrated a more than five times increase in the Gli1 group compared to the control group, while the calcification in the Gli1+Gant61 group was half that in the control (Fig. 4B). Runx2 and Gli family genes Gli2, Gli3 were measured in the three groups Gli1, Gli1+Gant61, and Control. Our results showed that Runx2 increased more than 1.5-fold in the Gli1 group compared with the control group, and similar result was observed for Gli2 gene expression, indicating a synergy with Gli1, while Gli3 which works opposite to Gli1, was decreased in the Gli1 group (Fig. 4C). Next, Runx2 and ALP protein expression was measured after the treatment with Gli1 overexpression for 72 h. The results the osteogenesis-related proteins Runx2 and ALP significantly increased in the indicated that https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2059 Gli1 group (Fig. 4D and Fig. 4E). Moreover, an ex-vivo osteogenic differentiation model was performed. Human aortic valve tissues were cut into small pieces and cultured in OM conditions. The tissues were subjected to Gli1 overexpression alone by transfection with adenovirus or combined with Gant61 treatment for two months (Fig. 4F). HE, Von Kossa, and Alizarin Red S staining showed a significant difference between the three groups (Fig. 4G). In addition, Gli1 and Runx2 interacted with each other as revealed by co-immunoprecipitation (Co-IP) experiments (Fig. 4H). Cellular immunofluorescence staining showed that Gli1 and Runx2 were co-localized in the control group and Gant61 group (Fig. 4I). Identification of Gli1 candidate target genes related to osteogenic differentiation Three sets of sequencing data were used for comprehensive analysis to explore the possible targets induce calcification of VICs: RNA- of Gli1 to sequencing data of Gli1-treated cells, ChIP- sequencing data of Gli1 in VICs, and RNA-sequencing data of different valve tissues. The TF binding motifs of Gli1 in VICs were CCACCC and TGGGTGG (Fig. 5A) and the Gli1 binding sites within 2kb from the transcriptional starting site (TSS) were evaluated to discover the candidate genes (Fig. 5B). Peak calling analysis of Gli1 analyzed peak positions relative to gene locations and revealed the distribution of the binding sites (Fig. 5C). GO biological process enrichment and KEGG pathway enrichment analysis showed top five significant processes and pathways for Gli1 binding genes (Fig. 5D). Three pairs of aortic valves were chosen to perform RNA-seq. The entire transcriptome profile of these samples was performed by Principal Component Analysis (PCA). The results showed that the calcific valves and the normal ones were clustered in their respective groups (Fig. 5E). KEGG signaling pathway enrichment analysis showed the top ten significant the Figure 3. Gli1 promotes the proliferation of VICs and regulates the cell cycle. (A) Molecular structure of Gant61. (B) IC50 of Gant61 on VICs, concentrations were transferred to Log(c). (C) CCK-8 assay was performed in the control group and Gant61 group. (D) Ki-67 and Edu staining in VICs were performed in both control and Gant61 groups. (E) Semi-quantitative statistics on the percentage of Ki-67 positive cells and Edu positive cells. (F) Ki-67 and Edu staining in VICs under different treatments: control, Gli1, and Gli1+Gant61. (G) Semi-quantitative statistics on the percentage of Ki-67 positive cells and Edu positive cells. (H) and (I) Flow cytometry cell cycle of VICs in a different group and semi-quantitative statistics of each phase ratio. (J) and (K) RT-PCR and western blotting analysis of cell cycle-related factors in different groups. (L) Semi-quantitative statistics on the cyclin B protein. Data are expressed as the mean ± SD. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2060 identified: pathways (Fig. 5F). After Venn’s interaction of the top ten significant activated signaling pathways between tissue RNA-seq and cellular RNA-seq, two common pathways were focal adhesion and PI3K-AKT signaling pathway (Fig. 5G). The Venn plot showed 15 candidate genes that may play a critical role in calcification after integration of the DEGs related to the two common pathways PI3K- AKT signaling and focal adhesion with ChIP-seq data (Fig. 5H). Finally, three potential direct binding genes such as FLNB, ITGB5, and BCAR1 were chosen according to the fold enrichment and p-value (Fig. 5I). BCAR1 was selected as the final target gene which encodes the protein P130cas. The high-resolution peak identified from the reading density of the ChIP-Seq data revealed a binding region in the BCAR1 gene (Fig. 5I). Gli1-P130cas-AKT axis promoted the osteogenic differentiation of VICs MK2206, one small molecule compound that specifically inhibits AKT1 phosphorylation, was used to block the osteogenic differentiation of VICs promoted by Gli1. Western blotting showed that P-AKT, ALP, and Runx2 expression were increased in the Gli1 overexpression group while they were significantly decreased in the Gli1+MK2206 group (Fig. 6A). The statistical analysis of the western blotting bands showed that the difference in protein expression was statistically significant (Fig. 6B). After culturing the cells in vitro under OM condition for 15 days, the calcification nodules were evaluated by Alizarin Red S. Different groups showed different (Fig. 6C). Semi- results by Alizarin Red S quantification of the calcification demonstrated a significant change between the Gli1 group and the Gli1+MK2206 group (Fig. 6D). Furthermore, the expression of the BCAR1 gene was silenced by si-RNA to reduce P130cas. Western blotting indicated that P-AKT, ALP, and Runx2 expression was significantly reduced due to the down-regulation of P130cas (Fig. 6E and Fig. 6F). After being cultured in OM condition for 15 days, the calcification nodules tested by Alizarin Red S were also different (Fig. 6G). The semi-quantification of Alizarin Red S indicated the the silencing of P130cas attenuated that calcification induced by Gli1 (Fig. 6H). Figure 4. Gli1 accelerates the osteogenic differentiation of VICs. (A) Alizarin Red S staining of the cells under different conditioned cultures. (B) Semi-quantitative statistics of Alizarin Red S stain. (C) RT-PCR of Runx2, Gli2, Gli3 for VICs stimulated with Gli1 and then treated with Gant61 or not for 48h. Data were analyzed using one-way ANOVA, (#) versus control, p < 0.05 indicates a significant difference. (D) and (E) Western blotting and semi-quantification of ALP, Runx2 for VICs stimulated with Gli1 and then treated with Gant61 or not for 72h. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. (F) and (G) Ex-vivo osteogenic differentiation of aortic valve tissues in different conditions. Observed under the fluorescent microscope and Representative HE, Alizarin Red S, Von Kossa staining figures of each tissue. (H) and (I) Co-IP and cellular immunofluorescence staining of Gli1 and Runx2 proteins in VICs. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2061 Figure 5. Identification of Gli1 candidate target genes related to osteogenic differentiation. (A) and (B) The TF binding motifs of Gli1 and the heatmap of the Gli1 binding site within 2kb from the transcriptional starting site. (C) Peak calling analysis of Gli1. (D) GO biological process enrichment and KEGG pathway enrichment analysis for Gli1 binding genes. (E) Principal Component Analysis (PCA) analysis for three pairs of aortic valves. (F) KEGG signaling pathway enrichment analysis of DEGs between samples from the CAVD group and the control group. (G) Venn’s interaction of KEGG signaling pathway enrichment analysis between tissue RNA-seq and cellular RNA-seq. (H) Venn’s interaction of the two common pathways with ChIP-seq data. (I) Fold enrichment and p-value of the final target genes according to the ChIP-seq data. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2062 Figure 6. Gli1-P130cas-AKT axis promotes the osteogenic differentiation of VICs. (A) Western blotting for p-AKT, AKT, ALP, p-p130cas, p130cas under the Gli1 treatment with or without MK2206. (B) Statistical analysis of Western Blotting gray values of p-AKT/AKT, p-p130cas/p130cas, ALP, and RUNX2. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. (C) Calcification nodules were tested by Alizarin Red S after being cultured under different treatments. (D) Semi-quantitation of calcification. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. (E) Western blotting for p-AKT, AKT, ALP, p-p130cas, p130cas under the Gli1 treatment with or without si-BCAR treatment. (F) Statistical analysis of Western Blotting gray values of p-AKT/AKT, p-p130cas/p130cas, ALP, and RUNX2. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. (G) Calcification nodules were tested by Alizarin Red S after being cultured under different treatments. (H) Semi-quantitation of calcification. Data were analyzed using one-way ANOVA, () p < 0.05 indicates a significant difference. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2063 Figure 7. Gli1 activates through canonical and non-canonical Hedgehog signaling pathways in VICs. (A) PCR test of Gli1, Gli2, Gli3, Ptch1 genes in VICs after SHH treatment with or without GDC-0449. Data were analyzed using one-way ANOVA, (#) p < 0.05 indicates a significant difference. (B and C) Western blotting and immunofluorescence staining of Gli1 in VICs were consistent with gene expression. (D) PCR test of Gli1 and Runx2 gene expression in VICs treated by common calcification-inducing stimulators for 48h. Data were analyzed using one-way ANOVA, ( versus control) p < 0.05 indicates a significant difference. (E) Changes in Gli1 and Gli2 gene expression induced by tgf-β1 within 24 hours. (F) PCR test of Gli1, Gli2, Runx2 genes in VICs stimulated by tgf-β1 with or without SB431542. (G) Western blotting of p-SMAD2, p-SMAD3, SMAD2/3, and Gli1 protein in VICs stimulated by tgf-β1 with or without SB431542. (H) Immunofluorescence staining of Gli1 in VICs under different treatments. (I and J) Western blotting of Gli1 in VICs stimulated by tgf-β1 with si-SMAD2/3 or si-SMAD4. Gli1 was activated through the canonical and non-canonical Hedgehog signaling pathways in VICs the The mechanism of up-regulation of Gli1 expression was also explored. Sonic hedgehog ligand is triggered canonical hedgehog signaling pathway. The SHH gene was overexpressed in VICs. PCR analysis showed that Gli1, and PTCH1 gene increased in SHH overexpression group, while they were down-regulated in the SHH+GDC-0449 group, in which GDC-0449 is one specific small molecular hedgehog inhibitor (Fig. 7A). Western blotting and immunofluorescence staining also confirmed that SHH induced Gli1 expression (Fig. 7B and Fig. 7C). A non-canonical hedgehog signaling pathway is another possible mechanism that may happen in VICs. First, common calcification-inducing stimulators were used on VICs, and then, Gli1 and Runx2 gene expression was measured after 48 h by PCR. Gli1 gene along with RUNX2 gene was significantly increased after the induction with TGF-β1 (Fig. 7D). Then, the changes in Gli1 and Gli2 gene expression induced by TGF-β1 were evaluated within 24 hours, Gli1 gene increase started at 12 hours (Fig. 7E). SB431542 is a classical inhibitor of the TGF-β type I receptor. PCR analysis that Gli1, Gli2, and Runx2 genes indicated significantly increased in the TGF-β1 group, while SB431542 could reverse this up-regulation induced by TGF-β1 (Fig. 7F). Western blotting and immunofluo- rescence staining also confirmed that Gli1 protein increased after the stimulation with TGF-β, while it was inhibited by SB431542 (Fig. 7G and Fig. 7H). Smad 2/3 factor was found phosphorylated during https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2064 this biological process (Fig. 7G). Next, smad2/3 factor was silenced by si-RNA, and western blotting indicated that Gli1 did not increase after the stimula- tion with TGF-β1 (Fig. 7I). Furthermore, smad4, which plays a key role in nuclear transportation in the TGF-β1 signaling pathway was down-regulated using si-RNA. Western blotting indicated that the silencing of smad4 did not influence Gli1 expression (Fig. 7J). Discussion Our research demonstrated that Gli1 promoted the calcification of VICs. Transcriptome sequencing analysis was used to explore the mechanism of this biological process. The transcriptome of VICs under different treatments was first sequenced. GO function annotation indicated that Gli1 enrichment regulated genes related to cell differentiation, inflammatory response, extracellular matrix organization, and positive regulation of cell proliferation. All these biological processes play critical roles during the pathological process of aortic valve calcification. The present study separated the process of valve calcification into two different phases according to its histological features: fibrosis and biomineralization [24]. The histological analysis revealed a significant thickening of the fibrotic layer, accompanied by a large degree of cell proliferation within the interstitial layer of the valve, resulting in a reduced elasticity and increased hardness of the valve leaflets. In contrast, normal human aortic valve interstitial cells were apparently in a resting situation, with a low viability, low proliferation, and low activity, to keep the valve soft and thin [25, 26]. The overexpression of Gli1 significantly increased the viability and proliferation of VICs as revealed by CCK-8 assay, Edu, and Ki-67 staining. Flow cytometry cell cycle of VICs also indicated the regulation caused by the Gli1 factor. Western blotting and PCR showed that the cell cycle-related factor CCNB1 may play a key role in these changes. Most studies indicated that Gli1 directly increases the expression of CCND [27], but our study found a different result in VICs. Therefore, it is important to understand its deep mechanism. Runx2 is one of the important factors that regulate the osteogenic differentiation of VICs [28, 29]. When cells were cultured in OM, Gli1 accelerated the calcification of VICs by increasing the expression of Runx2. Further ex-vivo osteogenic differentiation model was performed. Tiny human aortic valve tissues were transfected with adenovirus. After two months under OM condition, various types of chemical staining were performed to show the calcification of the tissues. All these experiments showed the osteogenic function of the Gli1 factor. Interestingly, Co-IP experiments showed that Runx2 and Gli1 interacted with each other and cellular indicated this immunofluorescence staining also discovery. When the cells were treated with GANT61, Gli1 did not enter the nucleus, and the Gli1 and Runx2 complex also accumulated in the cytoplasm. Some research found that Runx2 directly regulates Gli1 and Ptch1 expression in osteoblast progenitors and osteoblasts [30]. But our results indicated that Gli1 may also regulate Runx2 expression. As far as we know, this is the first time this phenomenon has been demonstrated in valve interstitial cells. It seems that Runx2 and hedgehog signaling regulates each other and induce osteogenic differentiation. Its internal mechanism is worthy of our in-depth exploration. Gli1 is an important transcription factor in different biological processes. Therefore, ChIP-seq technology was performed in valve interstitial cells to explore the downstream target genes of the GLI1 transcription factor. Our results demonstrated that its TF binding motifs in VICs were CCACCC and TGGGTGG and 1478 genes related to the Gli1 transcription factor were up-regulated. The GO and KEGG enrichment analysis revealed that actin cytoskeleton, focal adhesion, and axon guidance were significantly activated. Furthermore, RNA-transcrip- tome sequencing was performed on three pairs of aortic valve tissues and the top ten significant signaling pathways were obtained by KEGG enrichment analysis. The interaction with the cellular transcriptome data revealed that the PI3K-AKT signaling pathway and focal adhesion signaling pathway were involved. Thus, an interactive analysis of two target gene sets was performed, revealing 15 genes as the final candidate genes. The protein P130cas encoded by BCAR1 is a member of the Crk-associated substrate (CAS) family of scaffold proteins, characterized by the presence of multiple protein-protein interaction domains. It has also many serine and tyrosine phosphorylation sites [31-33]. Therefore, BCAR1 was selected as the final target gene that leads to the calcification of VICs caused by Gli1. However, the relationship between Gli1, AKT1, P130cas, and Runx2 needs further study. AKT1 is the key intersection molecule of the two signaling pathways PI3K-AKT and focal adhesion signal pathway; thus, MK2206 and siRNA were used to regulate the expression of P-AKT1, P-P130cas, and P130cas. Our results showed the down-regulation of the osteogenic related proteins Runx2 and ALP in the Gli1 overexpression group when AKT1 phospho- rylation was inhibited by MK2206. The osteogenic model in vitro also confirmed the osteogenic inhibi- tory effect, and the expression of P-P130cas was also reduced. These results revealed that the phospho- rylation of AKT1 was the trigger of osteogenesis and https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2065 phosphorylation of P130cas caused by Gli1. Our study found that P130cas was one of the target genes regulated by Gli1, which increased when the Gli1 factor was overexpressed. When P130cas expression was silenced by siRNA, the osteogenic related proteins ALP and Runx2 were also reduced although the phosphorylation of AKT1 was not affected. This result suggested that P130cas might play a key role in Gli1-induced calcification instead of AKT signaling pathway activation, which was involved in the extracellular matrix secretion when Gli1 was overexpressed in VICs. Consistent with this result, the cell morphology was significantly different in the Gli1 enrichment group compared to the control group., The pathophysiological process of valve tissue in CAVD involves the tissue remodeling: the cells of the valve tissue change, their number increases, and the extracellular matrix is remodeled [34]. Our results showed that Gli1 was the transcription factor that changed all these three aspects of VICs. The last part of our research was dedicated to evaluate the reason of Gli1 up-regulation in CAVD. Our focus was on the Hedgehog signaling pathway, since the core of this pathway lies in the Gli1 factor. Based on studies on this pathway over the past few decades, the mechanism of Gli1 upregulation in VICs was explored in two ways, such as through the activation of the canonical and non-canonical hedge- hog signaling pathway [35]. Interestingly, our results showed that both the two pathways activated the expression of the Gli factor in human aortic valve interstitial cells. TGF-β is one of the most significant activating factors of Gli transcription factors in VICs especially in the non-canonical pathway. TGF-β is one of the upstream molecules that are widely explored in the basic research of CAVD [12]. A large number of studies showed that the increased expression of TGF-β in the aortic valve leads to the transformation of valve endothelial cells into VICs, remodels the extracellular matrix in valve tissues, and allows the osteogenic differentiation of VICs. The TGF-β-smad4 axis regulates the expression of the Gli1 factor, but our results revealed that smad4 did not affect the expression of Gli1 factor induced by TGF-β in VICs, while smad 2/3 factor was the regulator of Gli1 expression [36, 37]. This is another interesting point that needs to be further explored. Supplementary Material Supplementary tables. https://www.ijbs.com/v19p2053s1.pdf Young Scholars Fostering Fund of the First Affiliated Hospital of Nanjing Medical University (PY2022011), Natural Science Foundation of Hubei Province, China (2021CFB227), Hubei University of Chinese Medicine (2021ZZX003), “Young Crops Program” Project China Program National (2016YFA0101100). R&D Key of Data Availability Statement Generated Statement: The datasets analyzed in this manuscript are not publicly available. Requests to access the datasets should be directed to Yuming Huang, [email protected]. Ethics Approval and Consent to Participate The study protocol was approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology. Written informed consent was obtained from each patient. Author Contributions Yuming Huang and Kang Xu contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Yuming Huang, Jiang Chen, Ming Liu, and Tingwen Zhou. Liang Chen and Nianguo Dong contributed to the data curation and resources. The first draft of the manuscript was written by Yuming Huang and Kang Xu. Competing Interests The authors have declared that no competing interest exists. References 1. Pawade TA, Newby DE, Dweck MR. Calcification in Aortic Stenosis: The Skeleton Key. Journal of the American College of Cardiology. 2015; 66: 561-77. 2. Lindman BR, Clavel MA, Mathieu P, Iung B, Lancellotti P, Otto CM, et al. Calcific aortic stenosis. Nature reviews Disease primers. 2016; 2: 16006. 3. Coffey S, Cairns BJ, Iung B. The modern epidemiology of heart valve disease. Heart. 2016; 102: 75-85. 4. Dweck MR, Boon NA, Newby DE. Calcific aortic stenosis: a disease of the valve and the myocardium. Journal of the American College of Cardiology. 2012; 60: 1854-63. 5. Hutcheson JD, Aikawa E, Merryman WD. Potential drug targets for calcific aortic valve disease. Nature reviews Cardiology. 2014; 11: 218-31. 6. Robbins DJ, Fei DL, Riobo NA. The Hedgehog signal transduction network. Science signaling. 2012; 5: re6. 7. Guerrero I, Rohatgi R. Frontiers in hedgehog signal transduction. Seminars in cell & developmental biology. 2014; 33: 50-1. 8. Dutzmann J, Koch A, Weisheit S, Sonnenschein K, Korte L, Haertle M, et al. Sonic hedgehog-dependent activation of adventitial fibroblasts promotes neointima formation. Cardiovascular research. 2017; 113: 1653-63. 9. Li F, Duman-Scheel M, Yang D, Du W, Zhang J, Zhao C, et al. Sonic hedgehog signaling induces vascular smooth muscle cell proliferation via induction of the G1 cyclin-retinoblastoma axis. Arteriosclerosis, thrombosis, and vascular biology. 2010; 30: 1787-94. 10. Roncalli J, Renault MA, Tongers J, Misener S, Thorne T, Kamide C, et al. Sonic hedgehog-induced is enhanced by AMD3100-mediated progenitor-cell mobilization. Journal of the American College of Cardiology. 2011; 57: 2444-52. functional recovery after myocardial infarction Acknowledgments This work was supported by the National Natural Science Foundation of China (82204659), 11. Mathieu P, Boulanger MC. Basic mechanisms of calcific aortic valve disease. The Canadian journal of cardiology. 2014; 30: 982-93. 12. Rutkovskiy A, Malashicheva A, Sullivan G, Bogdanova M, Kostareva A, Stenslokken KO, et al. Valve Interstitial Cells: The Key to Understanding the Pathophysiology of Heart Valve Calcification. Journal of the American Heart Association. 2017; 6. https://www.ijbs.com Int. J. Biol. Sci. 2023, Vol. 19 2066 13. Xu K, Xie S, Huang Y, Zhou T, Liu M, Zhu P, et al. Cell-Type Transcriptome Atlas of Human Aortic Valves Reveal Cell Heterogeneity and Endothelial to Mesenchymal Transition in Calcific Aortic Valve Disease. Involved Arteriosclerosis, thrombosis, and vascular biology. 2020; 40: 2910-21. 14. Wang C, Huang Y, Liu X, Li L, Xu H, Dong N, et al. Andrographolide ameliorates aortic valve calcification by regulation of lipid biosynthesis and glycerolipid metabolism targeting MGLL expression in vitro and in vivo. Cell calcium. 2021; 100: 102495. 15. Huang Y, Zhou X, Liu M, Zhou T, Shi J, Dong N, et al. The natural compound andrographolide inhibits human aortic valve interstitial cell calcification via the NF-kappa B/Akt/ERK pathway. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie. 2020; 125: 109985. SMAD3/beta-catenin cooperation. The Journal of biological chemistry. 2009; 284: 31523-31. 16. Wang C, Xia Y, Qu L, Liu Y, Liu X, Xu K. Cardamonin inhibits osteogenic differentiation of human valve interstitial cells and ameliorates aortic valve calcification via inflammasome pathway. Food & function. 2021; 12: 11808-18. the NF-kappaB/NLRP3 interfering in 17. Qu L, Shi K, Xu J, Liu C, Ke C, Zhan X, et al. Atractylenolide-1 targets SPHK1 and B4GALT2 to regulate intestinal metabolism and flora composition to improve inflammation in mice with colitis. Phytomedicine. 2022; 98: 153945. 18. Qu L, Wang C, Xu H, Li L, Liu Y, Wan Q, et al. Atractylodin targets GLA to regulate D-mannose metabolism to inhibit osteogenic differentiation of human valve interstitial cells and ameliorate aortic valve calcification. Phytother Res. 2023; 37: 477-89. 19. Huang Y, Xu K, Zhou T, Zhu P, Dong N, Shi J. Comparison of Rapidly Proliferating, Multipotent Aortic Valve-Derived Stromal Cells and Valve Interstitial Cells in the Human Aortic Valve. Stem Cells Int. 2019; 2019: 7671638. 20. Liu M, Li F, Huang Y, Zhou T, Chen S, Li G, et al. Caffeic Acid Phenethyl Ester the Ameliorates AKT/NF-kappaB/NLRP3 Inflammasome Pathway in Human Aortic Valve Interstitial Cells. Front Pharmacol. 2020; 11: 826. Calcification Activation Inhibiting by of 21. Pan L, Feng F, Wu J, Fan S, Han J, Wang S, et al. Demethylzeylasteral targets lactate by inhibiting histone lactylation to suppress the tumorigenicity of liver cancer stem cells. Pharmacol Res. 2022; 181: 106270. 22. Zhou T, Wang Y, Liu M, Huang Y, Shi J, Dong N, et al. Curcumin inhibits calcification of human aortic valve interstitial cells by interfering NF-kappaB, AKT, and ERK pathways. Phytother Res. 2020; 34: 2074-81. 23. Wang C, Han J, Liu M, Huang Y, Zhou T, Jiang N, et al. RNA-sequencing of human aortic valves identifies that miR-629-3p and TAGLN miRNA-mRNA pair involving in calcified aortic valve disease. J Physiol Biochem. 2022; 78: 819-31. 24. Sritharen Y, Enriquez-Sarano M, Schaff HV, Casaclang-Verzosa G, Miller JD. Pathophysiology of Aortic Valve Stenosis: Is It Both Fibrocalcific and Sex Specific? Physiology. 2017; 32: 182-96. 25. Wu B, Wang Y, Xiao F, Butcher JT, Yutzey KE, Zhou B. Developmental Mechanisms of Aortic Valve Malformation and Disease. Annual review of physiology. 2017; 79: 21-41. 26. Huang Y, Liu M, Liu C, Dong N, Chen L. The Natural Product Andrographolide Ameliorates Calcific Aortic Valve Disease by Regulating the Proliferation of Valve Interstitial Cells via the MAPK-ERK Pathway. Front Pharmacol. 2022; 13: 871748. 27. Sun Q, Bai J, Lv R. Hedgehog/Gli1 signal pathway facilitates proliferation, invasion, and migration of cutaneous SCC through regulating VEGF. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine. 2016. 28. Xu K, Huang Y, Zhou T, Wang C, Chi Q, Shi J, et al. Nobiletin exhibits potent inhibition on tumor necrosis factor alpha-induced calcification of human aortic valve interstitial cells via targeting ABCG2 and AKR1B1. Phytother Res. 2019; 33: 1717-25. 29. Xu K, Al-Ani MK, Wang C, Qiu X, Chi Q, Zhu P, et al. Emodin as a selective proliferative inhibitor of vascular smooth muscle cells versus endothelial cells suppress arterial intima formation. Life sciences. 2018; 207: 9-14. 30. Komori T. Regulation of Proliferation, Differentiation and Functions of Osteoblasts by Runx2. Int J Mol Sci. 2019; 20. 31. Barrett A, Pellet-Many C, Zachary IC, Evans IM, Frankel P. p130Cas: a key signalling node in health and disease. Cellular signalling. 2013; 25: 766-77. 32. Camacho Leal Mdel P, Sciortino M, Tornillo G, Colombo S, Defilippi P, Cabodi S. p130Cas/BCAR1 scaffold protein in tissue homeostasis and pathogenesis. Gene. 2015; 562: 1-7. 33. Tikhmyanova N, Little JL, Golemis EA. CAS proteins in normal and pathological cell growth control. Cellular and molecular life sciences : CMLS. 2010; 67: 1025-48. 34. Goody PR, Hosen MR, Christmann D, Niepmann ST, Zietzer A, Adam M, et al. Aortic Valve Stenosis: From Basic Mechanisms to Novel Therapeutic Targets. Arteriosclerosis, thrombosis, and vascular biology. 2020; 40: 885-900. 35. Briscoe J, Therond PP. The mechanisms of Hedgehog signalling and its roles in development and disease. Nature reviews Molecular cell biology. 2013; 14: 416-29. 36. Nye MD, Almada LL, Fernandez-Barrena MG, Marks DL, Elsawa SF, Vrabel A, et al. The transcription factor GLI1 interacts with SMAD proteins to modulate transforming growth factor beta-induced gene expression in a p300/CREB-binding protein-associated factor (PCAF)-dependent manner. The Journal of biological chemistry. 2014; 289: 15495-506. 37. Dennler S, Andre J, Verrecchia F, Mauviel A. Cloning of the human GLI2 Promoter: transcriptional activation by transforming growth factor-beta via https://www.ijbs.com
10.1371_journal.ppat.1008185	RESEARCH ARTICLE KSHV activates unfolded protein response sensors but suppresses downstream transcriptional responses to support lytic replication Benjamin P. JohnstonID 1,2, Eric S. PringleID 1,2, Craig McCormickID 1,2* 1 Department of Microbiology & Immunology, Dalhousie University, Halifax, Nova Scotia, Canada, 2 Beatrice Hunter Cancer Research Institute, Halifax, Nova Scotia, Canada a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * [email protected] Abstract Herpesviruses usurp host cell protein synthesis machinery to convert viral mRNAs into pro- teins, and the endoplasmic reticulum (ER) to ensure proper folding, post-translational modi- fication and trafficking of secreted and transmembrane viral proteins. Overloading ER folding capacity activates the unfolded protein response (UPR), whereby sensor proteins ATF6, PERK and IRE1 initiate a stress-mitigating transcription program that accelerates catabolism of misfolded proteins while increasing ER folding capacity. Kaposi’s sarcoma- associated herpesvirus (KSHV) can be reactivated from latency by chemical induction of ER stress, which causes accumulation of the XBP1s transcription factor that transactivates the viral RTA lytic switch gene. The presence of XBP1s-responsive elements in the RTA promoter suggests that KSHV evolved a mechanism to respond to ER stress. Here, we report that ATF6, PERK and IRE1 were activated upon reactivation from latency and required for efficient KSHV lytic replication; genetic or pharmacologic inhibition of each UPR sensor diminished virion production. Despite UPR sensor activation during KSHV lytic repli- cation, downstream UPR transcriptional responses were restricted; 1) ATF6 was cleaved to activate the ATF6(N) transcription factor but ATF6(N)-responsive genes were not tran- scribed; 2) PERK phosphorylated eIF2α but ATF4 did not accumulate; 3) IRE1 caused XBP1 mRNA splicing, but XBP1s protein did not accumulate and XBP1s-responsive genes were not transcribed. Ectopic expression of the KSHV host shutoff protein SOX did not affect UPR gene expression, suggesting that alternative viral mechanisms likely mediate UPR suppression during lytic replication. Complementation of XBP1s deficiency during KSHV lytic replication inhibited virion production in a dose-dependent manner in iSLK.219 cells but not in TREx-BCBL1-RTA cells. However, genetically distinct KSHV virions har- vested from these two cell lines were equally susceptible to XBP1s restriction following infection of naïve iSLK cells. This suggests that cell-intrinsic properties of BCBL1 cells may circumvent the antiviral effect of ectopic XBP1s expression. Taken together, these findings indicate that while XBP1s plays an important role in reactivation from latency, it can inhibit virus replication at a later step, which the virus overcomes by preventing its synthesis. OPEN ACCESS Citation: Johnston BP, Pringle ES, McCormick C (2019) KSHV activates unfolded protein response sensors but suppresses downstream transcriptional responses to support lytic replication. PLoS Pathog 15(12): e1008185. https://doi.org/10.1371/journal.ppat.1008185 Editor: Sankar Swaminathan, University of Utah, UNITED STATES Received: July 2, 2019 Accepted: November 2, 2019 Published: December 2, 2019 Peer Review History: PLOS recognizes the benefits of transparency in the peer review process; therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. The editorial history of this article is available here: https://doi.org/10.1371/journal.ppat.1008185 Copyright: © 2019 Johnston et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 1 / 34 Funding: This work was funded by an operating grant to CM from the Canadian Institutes for Health Research (MOP-84554) (http://www.cihr-irsc.gc. ca/e/193.html). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. KSHV modulates the host unfolded protein response to support lytic replication These findings suggest that KSHV hijacks UPR sensors to promote efficient viral replication while sustaining ER stress. Author summary Like all viruses, Kaposi’s sarcoma-associated herpesvirus (KSHV) uses cellular machinery to create viral proteins. Some of these proteins are folded and modified in the endoplasmic reticulum (ER) and traverse the cellular secretory apparatus. Exceeding ER protein folding capacity activates the unfolded protein response (UPR), which resolves ER stress by put- ting the brakes on protein synthesis and turning on stress-mitigating genes. We show that KSHV replication activates the three cellular proteins that sense ER stress, which are each required to support efficient viral replication. By contrast, KSHV blocks the UPR gene expression program downstream from each of these activated sensor proteins. The failure to resolve ER stress might normally be expected to put the virus at a disadvantage, but we demonstrate that reversal of this scenario is worse; when we supplement infected epithelial cells with the UPR transcription factor XBP1s to artificially stimulate the production of UPR-responsive gene products, virus replication is blocked at a late stage and very few viruses are released from infected cells. Taken together, these observations suggest that KSHV requires UPR sensor protein activation to replicate but has dramatically altered the outcome to prevent the synthesis of new UPR proteins and sustain stress in the ER compartment. Introduction Secreted and transmembrane proteins are synthesized in the endoplasmic reticulum (ER), where they are folded by chaperone proteins and modified by glycosyltransferases and protein disulfide isomerases. Demands on the protein folding machinery that exceed ER folding capac- ity cause the accumulation of misfolded proteins and trigger ER stress [1]. This accrual of mis- folded proteins activates the unfolded protein response (UPR) to mitigate the stress [2–4]. The UPR resolves ER stress by transiently attenuating translation, increasing synthesis of folding machinery, increasing lipid biogenesis to expand ER surface area, and degrading misfolded proteins in a process called ER-associated degradation (ERAD). Thus, the UPR adapts the lev- els of ER-associated biosynthetic machinery to meet demands on the system; however, if pro- teostasis is not re-established, the UPR switches from an adaptive to an apoptotic response. The UPR is coordinated by three transmembrane sensor proteins that sample the ER lumen; activated transcription factor-6 (ATF6), protein kinase R (PKR)-like endoplasmic reticulum kinase (PERK) and inositol-requiring enzyme 1 (IRE1). These sensor proteins are maintained in an inactive state by association of their luminal domains with the ER chaperone BiP [5]. In response to ER stress, BiP is mobilized to participate in re-folding reactions in the ER, releasing UPR sensors from their repressed state [6]. Together these three UPR sensors coordinate complementary aspects of an ER stress-mitigating gene expression program. ATF6 is an ER-localized type II transmembrane protein. Detection of unfolded proteins in the ER lumen causes ATF6 to traffic to the Golgi apparatus, where it is cleaved by Golgi-resi- dent site-1 protease (S1P) and site-2 protease (S2P) enzymes [7,8], which releases the amino- terminal ATF6(N) fragment into the cytosol. ATF6(N) is a basic leucine zipper (bZIP) tran- scription factor that translocates to the nucleus and transactivates genes encoding chaperones, foldases and lipogenesis factors. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 2 / 34 KSHV modulates the host unfolded protein response to support lytic replication PERK is an ER-localized type I transmembrane kinase. ER stress causes displacement of inhibitory BiP proteins from PERK, which triggers dimerization and trans-autophosphoryla- tion [9]. Active PERK phosphorylates serine 51 of eIF2α, which increases eIF2α affinity for its guanine exchange factor eIF2B [10,11]. This binding depletes the small pool of eIF2B, thereby inhibiting replenishment of the eIF2-GTP-Met-tRNAMeti ternary complex required for trans- lation initiation [12]. Bulk cap-dependent translation is attenuated, while a subset of uORF- containing mRNAs encoding stress response proteins are preferentially translated [13]. Acti- vating transcription factor 4 (ATF4) is chief among these stress response proteins [14]; this bZIP transcription factor translocates to the nucleus and drives the synthesis of gene products that mitigate ER stress by increasing the antioxidant response and a catabolic process known as autophagy [15,16]. ATF4 upregulates protein-phosphatase 1 α (PP1α) cofactor GADD34 (PPP1R15A), which dephosphorylates eIF2α and allows recharging of the ternary complex and resumption of translation [17,18]. This direct control of protein synthesis by stress is known as the integrated stress response (ISR) [19,20]. During chronic or severe ER stress, ATF4 also upregulates the bZIP transcription factor CHOP, which promotes stress-induced cell death by coordinating the expression of genes that promote apoptosis [21–23]. IRE1 is an ER-localized type I transmembrane kinase and endoribonuclease. ER stress causes release of repressive BiP proteins from IRE1, which triggers IRE1 dimerization and trans-autophosphorylation, and stimulates RNase activity. On the cytosolic face of the ER, active IRE1 cleaves a conserved nucleotide sequence in two stem loops in the X-box binding protein-1 (Xbp1) mRNA, which excises a 26-nucleotide intron [24–26]. The tRNA ligase RTCB completes this cytosolic mRNA splicing event by re-ligating cleaved Xbp1 mRNA, which generates a shifted open reading frame that can be translated to produce the active bZIP transcription factor XBP1s [27,28]. XBP1s translocates to the nucleus and drives production of chaperones, lipid synthesis proteins and proteins involved in ERAD [29]. The combined action of these gene products simultaneously increases ER folding capacity and decreases the load via ERAD. In a process called regulated IRE1-dependent decay (RIDD), IRE1 can also cleave select ER-targeted mRNAs with a stem loop that resembles that of XBP1, which may serve to attenuate translation [30,31]. One of the first roles ascribed to XBP1 was its requirement for terminal B cell differentiation into antibody-secreting plasma cells [32], where it expands ER capacity to support antibody secretion [33]. KSHV is a gammaherpesvirus that causes Kaposi’s sarcoma (KS), primary effusion lym- phoma (PEL) and multicentric Castleman’s disease (MCD) [34–36]. Like all herpesviruses, KSHV can establish a quiescent form of infection known as latency in which viral gene expres- sion is severely restricted and the genome is maintained as a nuclear episome. The virus can latently infect a variety of cell types, but it is thought that life-long infection of human hosts is primarily enabled by latent infection of immature B lymphocytes [37], followed by viral repro- gramming into an intermediate cell type that resembles a plasma cell precursor [38–40]. Indeed, PEL cells display elevated UPR gene expression consistent with a plasma cell-like phe- notype [40]. An essential feature of latency is reversibility, which is required for viral replication and production of viral progeny. The true physiologic cues for KSHV lytic reactivation remain obscure, but in vitro studies have implicated ER stress-activated signalling pathways [41–43]. Reactivation from latency requires the immediate-early lytic switch protein replication and transcriptional activator (RTA), a transcription factor that initiates a temporal cascade of gene expression [44,45]. RTA recruits host cell co-factors to transactivate viral early genes required for genome replication [46]. KSHV has usurped the IRE1/XBP1s pathway to regulate the latent/lytic switch. XBP1s binds to canonical response elements in the RTA promoter and drives synthesis of the RTA lytic switch protein required for lytic reactivation [42]. Functional PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 3 / 34 KSHV modulates the host unfolded protein response to support lytic replication XBP1s binding motifs have also been found in the promoter for v-IL6 [47], the KSHV homo- log of human IL6, which plays key roles in PEL and MCD [48–50]. Thus, UPR activation causes reactivation from latency and initiation of lytic gene expression. Interestingly, because the XBP1 transcription factor is required for normal B cell differentiation into mature plasma cells [51,52], it is likely that the viral acquisition of XBP1s target sequences hardwires KSHV reactivation from latency to terminal B cell differentiation [42], although the physiological consequences of this link are not known. In sharp contrast to these mechanistic linkages, noth- ing is known about how ATF6 and PERK impact latent KSHV infection or reactivation from latency. KSHV encodes structural and non-structural proteins that are folded in the ER and traverse the secretory apparatus. UPR activation during herpesvirus lytic replication has been reported, and there is evidence for UPR sensor engagement by specific gene products, rather than simply by exceeding ER folding capacity [53–55]. For example, herpes simplex virus type 1 (HSV-1) glycoprotein B can bind and suppress PERK activation to promote virus protein production [56]. To better understand how KSHV usurps the UPR, we investigated UPR activation follow- ing reactivation from latency in B cell- and epithelial cell-based models. We report that all three proximal UPR sensors are activated following reactivation from latency; ATF6 is cleaved, PERK and eIF2α are phosphorylated, and IRE1 is phosphorylated and Xbp1 mRNA is spliced. Furthermore, we determined that UPR sensor activation is pro-viral; pharmacologic or genetic inhibition of each UPR sensor diminished virion yield from infected cells. Surprisingly, viral proteins accumulated despite sustained phosphorylation of eIF2α throughout the lytic cycle, suggesting that viral messenger ribonucleoproteins (mRNPs) may have unique properties that ensure priority access to translation machinery during stress. Remarkably, activation of proxi- mal UPR sensors during lytic replication failed to elicit any of the expected downstream effects: ATF6(N) and XBP1-targets genes were not upregulated and ATF4 was not translated. Indeed, despite clear evidence of IRE1 activation and XBP1 mRNA splicing, XBP1s protein failed to accumulate during KSHV lytic replication. This suggests that the virus requires proximal acti- vation of UPR sensors but prevents downstream UPR transcription required to mitigate stress and restore ER homeostasis. Despite its important role in host shutoff, the KSHV RNA endo- nuclease SOX did not affect UPR gene expression in ectopic expression models, suggesting that alternative viral mechanisms likely mediate UPR suppression during lytic replication. Remarkably, complementation of XBP1s deficiency during KSHV lytic replication by ectopic expression inhibited the production of infectious virions in the iSLK.219 epithelial cell model, but not in the TREx BCBL1-RTA PEL cell model. Therefore, while XBP1s plays an important role in reactivation from latency, it inhibits later steps in lytic replication in some cells, which the virus overcomes by tempering its synthesis. Taken together, these findings suggest that KSHV hijacks UPR sensors to promote efficient viral replication instead of resolving ER stress. Results KSHV lytic replication activates IRE1 and PERK but downstream UPR transcription factors XBP1s and ATF4 do not accumulate ER stress can induce KSHV lytic replication via IRE1 activation and XBP1s-mediated transac- tivation of the viral RTA latent/lytic switch gene [41,42]. However, it is not known whether the burden of synthesizing secreted and transmembrane KSHV proteins causes ER stress over the course of the lytic replication cycle. To test UPR activation during the lytic cycle we used the TREx BCBL1-RTA cell line that expresses RTA from a doxycycline (dox)-regulated promoter to reactivate KSHV from latency [57]. We treated cells with dox for 0, 24, and 48 hours (h) and immunoblotted for UPR proteins to determine their activation state and KSHV proteins to PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 4 / 34 KSHV modulates the host unfolded protein response to support lytic replication monitor the progression of the lytic cycle from early (ORF45) to late (ORF65) stages (Fig 1A). As a positive control to ensure that UPR sensors were intact in our system, we also treated latent and lytic cells with the SERCA (sarco/endoplasmic reticulum Ca2+-ATPase) inhibitor thapsigargin (Tg) to pharmacologically induce ER stress [58]. After 2 h of Tg treatment of latently infected cells, IRE1α was phosphorylated, as determined by slower electrophoretic mobility, and a semiquantitative XBP1 RT-PCR splicing assay revealed that the majority of Xbp1 mRNA was spliced, which enabled translation of XBP1s protein (Fig 1A). Tg activated the ISR in latent cells, as determined by phosphorylation of PERK (also revealed by slower electrophoretic mobility of PERK) and its downstream target eIF2α-Ser51, which promoted translation of ATF4 from a uORF-containing mRNA. Lysates collected at 24 and 48 h post- dox addition displayed strong accumulation of ORF45 and ORF65, which indicated progres- sion through early and late lytic replication, respectively. By 24 h post-dox, IRE1α and PERK pathways were both activated: IRE1 and PERK were phosphorylated, which corresponded to increased spliced XBP1 mRNA and phospho-eIF2α, respectively. Activation of the IRE1 and PERK pathways persisted until 48 h post-dox. However, PERK activation and eIF2α phosphor- ylation did not activate the ISR in lytic cells; ATF4 mRNA levels in latent and lytic cells were identical (Fig 1B), there was negligible accumulation of ATF4 protein (Fig 1A), and levels of ATF4-dependent CHOP mRNA did not increase (Fig 1B). Tg treatment of lytic cells at 24 h post-dox caused a strong increase in PERK activation and eIF2α phosphorylation, but once again, ATF4 protein did not accumulate and mRNA levels of the ATF4-target gene CHOP were significantly less than elicited by Tg-treatment of latently-infected cells (Fig 1A and 1B). Taken together, these observations suggest that KSHV lytic replication prevents downstream execution of the ISR when eIF2α is phosphorylated. By 48 h post-dox, Tg treatment was no longer able to stimulate accumulation of phospho-eIF2α, and total PERK levels were dimin- ished compared to earlier stages of replication (Fig 1A). Tg-treated lytic cells also displayed reduced total IRE1α and XBP1s levels, even though Xbp1 mRNA splicing was similar to Tg- treated latent cells. These data demonstrate that KSHV lytic replication activates IRE1α and PERK UPR sensor proteins but prevents the accumulation of XBP1s and ATF4 transcription factors required for downstream UPR transcriptional responses. KSHV lytic replication activates ATF6 but downstream UPR transcriptional responses are inhibited The third UPR sensor ATF6 is cleaved in the Golgi during ER stress to release its active N-ter- minal fragment ATF6(N) that translocates to the nucleus and transactivates UPR genes [8]. We observed elevated total endogenous ATF6 levels at 24 h post-dox, which diminished over the following 24 h (Fig 1A). However, the ATF6 antibody that we employed in this study could not detect endogenous ATF6(N) in any samples, including Tg-treated positive controls; this is consistent with previous reports of rapid degradation of the labile ATF6(N) protein [8,59]. To further investigate ATF6 cleavage in this system, we transduced TREx BCBL1-RTA cells with a lentiviral vector encoding HA-epitope-tagged full-length ATF6 (HA-ATF6α-FL) [60]. This ectopic expression system allowed us to monitor cleavage of the ~100 kDa HA-ATF6α-FL pre- cursor into the ~60 kDa HA-ATF6α N-terminal fragment, which was revealed after Tg treat- ment (Fig 1C, lane 2). Consistent with endogenous ATF6 in lytic cells (Fig 1A), we observed the accumulation of full-length HA-ATF6 at 24 h post-dox followed by return to basal levels by 48 h post-dox. During lytic replication, we also observed the cleavage of HA-ATF6α into the active 60 kDa fragment (Fig 1C, lanes 3–6). Indeed, by 24 h post-dox the levels of HA-ATF6α-N were higher than latently infected cells treated with Tg. Interestingly, ATF6-tar- get genes BiP and HERPUD1 [61–63] were not transactivated during lytic replication, even in PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 5 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 1. KSHV lytic replication activates IRE1, PERK and ATF6 but downstream UPR transcription factors are inhibited. (A) TREx BCBL1-RTA cells were treated with 1 μg/mL dox for 0, 24 and 48 h to induce lytic replication followed by 75 nM Thapsisgargin (Tg) for 2 h prior to harvesting for protein and total RNA. Whole cell lysates were analyzed by immunoblots for UPR markers (IRE1α, XBP1s, PERK, phospho- and total eIF2α, ATF4, and full length ATF6α). Migration shift in PERK and IRE1α immunoblots correspond to phosphorylation. KSHV proteins ORF45 and ORF65 were probed for to indicate induction of early lytic and late lytic, respectively. β-actin was used as a loading control. ns corresponds to an unknown protein species that cross-reacted with ATF6 anti- sera. Xbp1 mRNA was amplified by RT-PCR, digested with PstI (cleaves unspliced XBP1 isoform only), and separated on SYBR Safe-stained agarose gel. The asterisk (�) corresponds to xbp1u-xbp1s hybrid cDNA. Representative immunoblot and agarose gel of two independent experiments are shown. (B) TREx BCBL1-RTA cells were treated with 1 μg/mL dox for 0 or 24 h to induce lytic replication and 4 h prior to harvesting for total RNA, cells were treated with or without 75 nM Tg. Relative changes in mRNA levels of ATF4 and CHOP were measured by qPCR and calculated using the ΔΔCt method using 18S rRNA as a reference gene. (�, p value < 0.05, ��, p value < 0.01) (C) TREx BCBL1-RTA cells were transduced with lentiviral expression vector encoding HA-ATF6α and selected for with 1 μg/mL puromycin. Following selection, cells were treated with 1 μg/mL dox for 0, 24 and 48 h and treated with 75 nM Tg for 4 h prior to harvest. Whole cell lysates were analyzed by immunoblots for HA epitope tag, PERK, ORF45 and β-actin (loading control). Immunoblot shown is representative of two independent experiments. (D) As in (C), HA-ATF6a-transduced TREx BCBL1-RTA were treated with dox for 0 and 24 h and then treated with 75nM Tg for 4 h prior to total RNA isolation. mRNA levels of ATF6α target genes BiP and HERPUD1 were measured by qPCR. Changes in mRNA levels were calculated by the ΔΔCt method and normalized using 18S rRNA as a reference gene. An average of 3 independent experiments are graphed and error bars denote SEM. Two-way ANOVA and a post-hoc multiple comparisons tests were done to determine statistical significance (��, p value < 0.0001). https://doi.org/10.1371/journal.ppat.1008185.g001 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 6 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 2. PERK-dependent eIF2α phosphorylation during the KSHV lytic cycle. TREx BCBL1-RTA cells were treated with 1 μg/mL dox -/+ 500 nM of the PERK inhibitor GSK2606414 (PERKi) for 0, 24 and 48 h and treated with or without 75 nM Thapsisgargin (Tg) for 2 h prior to harvest. Whole cell lysates were analyzed by immunoblots for PERK, phospho and total eIF2α, ATF4, IRE1α, and ORF45. Migration shift in PERK and IRE1α immunoblots correspond to phosphorylation. β-actin was used as a loading control. Immunoblots shown are representative of two independent experiments. https://doi.org/10.1371/journal.ppat.1008185.g002 cells treated with Tg (Fig 1D). These data show that ATF6 is proteolytically cleaved during lytic replication but the transcription factor cannot transactivate canonical target genes. PERK-dependent eIF2α phosphorylation during the KSHV lytic cycle Because eIF2α can be phosphorylated on serine 51 by four possible eIF2α kinases activated by different stresses (ER stress activates PERK, dsRNA activates PKR, nutrient stress activates GCN2, oxidative stress activates HRI [9,64–66]), we investigated the specific contribution of PERK to eIF2α phosphorylation during KSHV lytic replication with the PERK inhibitor GSK2606414 (PERKi) [67,68], which fully inhibited PERK and eIF2α phosphorylation follow- ing 2 h Tg treatment (Fig 2, lane 3 vs 4). TREx BCBL1-RTA cells were treated with PERKi con- current with dox addition. After 24 h and 48 h treatment with dox and PERKi, there was a reduction in PERK and eIF2α phosphorylation compared to cells treated with dox alone (Fig 2, lanes 5 vs 6 & 9 vs 10), indicating that PERK is involved in phosphorylating eIF2α during lytic replication. Importantly, PERKi had no appreciable effect on IRE1α activation, indicating that ER stress is still manifested during KSHV lytic replication when PERK is inhibited, and PERK inhibition does not heighten this response. Furthermore, lytic cells treated with PERKi in the presence or absence of Tg also showed increased levels of ORF45 (Fig 2, lanes 9 vs 10 & lanes 11 vs 12), which suggests that even though ATF4 fails to accumulate, PERK-mediated phosphorylation of eIF2α may still impede viral protein synthesis to some extent. Finally, the level of eIF2α phosphorylation following PERKi treatment did not return to baseline levels observed in latently infected TREx BCBL1-RTA cells, suggesting that other eIF2α kinases besides PERK may be activated during lytic replication. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 7 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 3. The type or duration of ER stress does not account for the lack of ATF4 and XBP1s. TREx BCBL1-RTA cells were pretreated with dox for 0 or 24 h and treated with 75 nM Thapsisgargin (Tg) or 5 μg/mL Tunicamycin (Tm) for 1, 4, or 8 h prior to harvest for either protein or RNA. Whole cell lysates were analyzed by immunoblots for UPR markers (IRE1α, XBP1s, phospho- and total eIF2α, and ATF4). Migration shift in IRE1α immunoblot corresponds to phosphorylation. KSHV protein ORF45 was probed to show lytic replication and β-actin was used as a loading control. Xbp1 RT-PCR splicing assay was performed as previously indicated. (�) corresponds to xbp1u-xbp1s hybrid cDNA. Immunoblots and agarose gels are representative of two independent experiments. https://doi.org/10.1371/journal.ppat.1008185.g003 KSHV lytic replication suppresses XBP1s and ATF4 accumulation irrespective of the type or duration of ER stress We previously observed UPR sensor activation during the lytic cycle but failure to accumulate active UPR transcription factors XBP1s and ATF4 following a brief 2 h pulse of Tg (Fig 1A). We confirmed these observations by treating latent and lytic TREx BCBL1-RTA cells with Tg or tunicamycin (Tm), which induces ER stress by blocking protein N-linked glycosylation in the ER [69,70], over a range of incubation times (1, 4 and 8 h). The majority of Xbp1 mRNA splicing was observed after latent cells were Tg-treated for 1 h or Tm-treated for 4 h (Fig 3); these slower kinetics are consistent with known properties of Tm, which relies on new protein synthesis to trigger ER stress [71]. As expected, in latently infected cells, Xbp1 mRNA splicing and XBP1s and ATF4 protein accumulation were observed throughout the 8 h course of Tg or Tm treatment. By contrast, during KSHV lytic replication, regardless of duration of Tg or Tm treatment, comparable levels of spliced Xbp1 mRNA and phospho-eIF2α were detected but XBP1s and ATF4 proteins failed to accumulate. Consistent with previous results, there were slightly higher levels of unspliced Xbp1 mRNA observed during the lytic cycle that correlated with reduced total IRE1α protein levels, but this was only observed in Tg-treated cells; Tm treatment converted the bulk of the Xbp1 mRNA pool to the spliced form by 4 h post-treat- ment, even though IRE1α accumulation was still blocked. Taken together, these observations confirm that XBP1s and ATF4 transcription factors do not accumulate in the KSHV lytic cycle despite robust activation of UPR sensors in response to chemically induced ER stress. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 8 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 4. Low levels of XBP1s induced in lytic are insufficient to upregulate XBP1s-target genes. (A) TREx BCBL1-RTA cells were treated with 1 μg/mL dox for the indicated times or treated with 75 nM Thapsisgargin (Tg) for 4 h (as a positive control) and harvested for either protein or RNA. Whole cell lysates were analyzed by immunoblots for IRE1α, XBP1s, ORF45 and β-actin (loading control). Xbp1 RT-PCR splicing assay was performed as previously indicated. (�) corresponds to xbp1u-xbp1s hybrid cDNA. A representative immunoblot and agarose gel for three independent experiments are shown. (B) Densitometry analysis from semi-quantitative XBP1 RT-PCR splicing assay in (A) was used to calculate the percentage of XBP1 in the spliced isoform. The mean of three independent experiments are graphed and error bars represent the standard deviation of the mean. (C) TREx BCBL1-RTA cells were treated with 1μg/mL dox for 0 or 24 h and treated with 75 nM Tg or 5 μg/mL Tunicamycin (Tm) for 4 h prior to RNA isolation. XBP1 mRNA splicing was determined by semi- quantitative RT-PCR splicing assay as previously described. The gel shown is representative of two independent experiments. (D) Total RNA samples from (C) were used to measure the mRNA levels of XBP1s target genes EDEM1 and ERdj4 by qPCR. Changes in mRNA levels were calculated by the ΔΔCt method and normalized using 18S rRNA as a reference gene. An average of 4 independent experiments are graphed and error bars denote SEM. Two-way ANOVA and a post-hoc multiple comparisons tests were done to determine statistical significance (��, p value < 0.0001). https://doi.org/10.1371/journal.ppat.1008185.g004 XBP1s target genes are not transactivated by XBP1s during lytic replication To better understand the kinetics of IRE1 activity during KSHV lytic replication we reactivated TREx BCBL1-RTA cells from latency using dox and monitored IRE1 activity over a 24 h time course. Spliced XBP1 mRNA and XBP1s protein were detected by 6 h post-dox addition, con- comitant with a modest increase in IRE1 phosphorylation, as determined by reduced electro- phoretic mobility (Fig 4A). This induction of Xbp1 splicing corresponded with increased accumulation of the early viral protein ORF45. By 18 h post-dox addition, which coincides PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 9 / 34 KSHV modulates the host unfolded protein response to support lytic replication with the beginning of viral genome replication in this model [57], IRE1 phosphorylation, Xbp1 splicing and XBP1s protein accumulation had peaked, and almost 40% of Xbp1 mRNA was spliced (Fig 4A and 4B). By contrast, Tg treatment converted nearly the entire pool of Xbp1 mRNA to the spliced form and caused strong accumulation of XBP1s protein. To determine whether the low levels of XBP1s observed during lytic replication were sufficient to induce syn- thesis of XBP1s target genes, RNA was harvested from latent and lytic (24 h post-dox addition) TREx BCBL1-RTA cells treated with Tg, Tm or vehicle control; Semiquantitative RT-PCR analysis revealed that Xbp1 mRNA was efficiently spliced following Tg or Tm treatment, both in latent and lytic samples, while moderate splicing was observed in samples from vehicle- treated control lytic cells (Fig 4C). RT-qPCR was performed to measure relative levels of canonical XBP1s target genes EDEM1 and ERdj4 [72] (Fig 4D); Tg and Tm treatments caused dramatic accumulation of EDEM1 and ERdj4 transcripts in latently infected cells, whereas these transcripts remained at low levels in lytic cells treated with Tg, Tm or mock-treated. Thus, low expression of XBP1s during lytic replication was insufficient to induce synthesis of canonical XBP1s target gene products involved in ER stress mitigation. KSHV SOX host shutoff protein is not sufficient to block UPR transcriptional responses Our observations indicate that KSHV activates UPR sensor proteins while simultaneously sup- pressing downstream XBP1s, ATF6(N), and ATF4 transcriptional responses. During KSHV lytic replication the viral host shutoff endonuclease SOX targets the majority of host mRNAs for degradation [73,74]. This indirectly causes accumulation of host RNA-binding proteins in the nucleus [75] and reduces the recruitment of RNA polymerase II to host promoters, causing global repression of host transcription [76]. To test whether SOX can prevent accumulation of UPR gene products, we engineered a dox-inducible myc-SOX 293A cell line. Cells were trans- duced with lentiviral vectors encoding the reverse tetracycline-controlled transactivator rtTA3 and myc-SOX under the control of seven tandem Tet operator (TetO) elements and stable cells were selected with antibiotics. Dox was added to induce SOX expression for 48, 72, and 96 h; cells were treated with 500 nM Tg for 4 h prior to harvesting total RNA at each time point. GAPDH, a known substrate of SOX, was included as a positive control [77]. SOX reduced GAPDH mRNA levels ~ 2-fold, both in the presence and absence of Tg (Fig 5A). However, SOX had little effect on the low steady-state levels of UPR transcripts ERdj4, BiP, or CHOP. Tg-treatment dramatically increased UPR gene expression at each stage of the time- course, which was also largely unaffected by SOX, suggesting that SOX does not impede UPR gene expression. To corroborate these findings, cells bearing dox-inducible wild type SOX or a P176S endonuclease-defective SOX mutant [78] were induced over a 48 h time-course and treated with 500 nM Tg or vehicle control for 4 h prior to harvesting protein lysates. Immuno- blotting analysis revealed no change in the accumulation of BiP (ATF6-dependent), XBP1s (IRE1-dependent) or CHOP (PERK-dependent) in response to Tg treatment in the presence of WT or mutant SOX (Fig 5B). Likewise, Tm treatment elicited PERK activation, eIF2α phos- phorylation and BiP, XBP1s, ATF4 and CHOP accumulation to a similar extent in the pres- ence of either WT or mutant SOX (Fig 5C). These findings indicate that SOX expression had little impact on UPR signaling irrespective of the mode of ER stress induction and is unlikely to play a significant role in inhibiting UPR transcriptional responses during lytic replication. UPR sensor activation supports efficient KSHV replication We next used a combination of genetic and pharmacologic approaches to inhibit each UPR sensor protein and measure effects on KSHV lytic replication. To investigate the role of ATF6, PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 10 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 5. The KSHV early gene SOX is not sufficient to inhibit the UPR. (A) 293A Tet-On cells (see Materials and Methods for generation) were transduced with lentiviral expression vectors encoding myc-SOX and selected with blasticidin. SOX expression was induced with 1 μg/ml dox for 0, 48, 72, and 96 h; cells were treated with 500 nM Thapsisgargin (Tg) for 4 h prior to harvesting total RNA. Changes in mRNA levels of ERdj4, BiP, CHOP, and GAPDH were calculated with the ΔΔCt method using 18S rRNA as a reference gene. (B) Dox-inducible myc-SOX wildtype (WT) and myc-SOX P176S point-mutant 293A cells were treated with 1 μg/ml dox for 0, 24, and 48 h and 500 nM Tg was added 4 h prior to harvesting whole cell lysates. Lysates were analyzed by immunoblotting for myc-Sox, BiP, XBP1s, PERK and CHOP. β-actin was used as a loading control. (C) Similar to (B), myc-SOX WT or myc-SOX P176S expression was induced with dox for 48 h and treated with increasing concentrations of Tm at 4 h prior to harvesting total cell lysates. Immunoblot analysis of myc-SOX and UPR substrates BiP, XBP1s, PERK, p-eIF2α, ATF4 (ns, denotes non-specific protein bands), and CHOP. β-actin was used as a loading control. Experiments in A, B, and C were conducted once. https://doi.org/10.1371/journal.ppat.1008185.g005 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 11 / 34 KSHV modulates the host unfolded protein response to support lytic replication we silenced ATF6α expression in TREx BCBL-RTA cells with shRNAs (Fig 6A) and collected cell supernatants from ATF6-silenced or control cells at 48 h post-dox addition. Cell superna- tants were processed to measure relative levels of released capsid-protected viral genomes by qPCR. ATF6 knockdown reduced virus titer by ~50% compared to cells transduced with non- targeting shRNA (Fig 6B). To further confirm a positive role for ATF6 cleavage in KSHV virion production, we treated TREx BCBL1-RTA cells with the ATF6 inhibitor Ceapin-A7 or the S1P inhibitor PF 429242, which selectively inhibited Tg-induced ATF6-target gene BiP and had no effect on XBP1s-target gene ERdj4 (Fig 6C and 6D). Inhibition of ATF6 with Cea- pin-A7 and S1P inhibitor inhibited KSHV production in TREx-BCBL1-RTA cells almost 2-fold (Fig 6E). We corroborated these findings in the dox-inducible iSLK.219 cell model that produces recombinant KSHV virions harbouring a GFP transgene. iSLK.219 cells were treated with Ceapin-A7 or S1P inhibitor at the time of dox addition and cell supernatants were har- vested 96 h later, serially diluted, and titered on naive monolayers of 293A cells by flow cytom- etry. Like in TREx BCBL1-RTA cells, inhibition of ATF6 reduced virus titers from iSLK.219 cells (Fig 6F). Interestingly, while Ceapin-A7 had a similar effect on titer in the two cell models, S1P inhibition in iSLK.219 cells caused an ~100-fold decrease in virus titre, indicating that S1P may have an ATF6-independent role during lytic replication in iSLK.219 cells. To determine whether activation of PERK or downstream engagement of the ISR are important for viral replication TREx BCBL1-RTA cells were treated with the selective PERK inhibitor GSK2606414 (PERKi) or the ISR inhibitor ISRIB, a small molecule that blocks phos- pho-eIF2α-mediated inhibition of translation by maintaining active eIF2B [79,80]. PERKi and ISRIB each inhibited viral particle release by ~50% (Fig 6G). To determine if IRE1 RNase activity is required for efficient viral replication TREx BCBL1-RTA cells were treated with the IRE1 inhibitor 4μ8c [81], which inhibited virus release in a dose-dependent manner (Fig 7A). We corroborated these findings in the dox-inducible iSLK.219 cell model [82]; as in the TREx BCBL1-RTA cell model, higher doses of 4μ8c inhib- ited virion production from iSLK.219 cells, with statistically significant inhibition achieved at the 25 μM dose (Fig 7B). To confirm a role for IRE1 in viral replication, we inhibited IRE1α expression in iSLK.219 cells via RNA silencing. Cells transduced with IRE1α-targeting shRNAs or non-targeting controls were treated with dox for 96 h, and cell supernatants were once again collected to titer GFP-expressing KSHV virions by flow cytometry. iSLK.219 cells bearing IRE1α shRNAs inhibited release of infectious virions by more than two-fold compared to cells transduced with non-targeting shRNAs (Fig 7C and 7D). Taken together, these data suggest that all three sensors of the UPR are important for robust virus replication. Ectopic XBP1s expression inhibits release of infectious KSHV virions in a cell type- and dose-dependent manner Our studies to this point suggest that efficient lytic replication depends on activation of all three UPR sensor proteins, but the cell fails to produce UPR transcription factors and down- stream transcriptional responses required to mitigate ER stress. This suggests that KSHV may re-dedicate UPR sensors for a new purpose rather than resolving ER stress, and that sustained, low-level ER stress may not impede viral replication. We also found it puzzling that IRE1 RNase activity was required for efficient lytic replication but XBP1s could not transactivate XBP1s-target genes, including RTA, during the lytic cycle. For these reasons, we hypothesized that XBP1s accumulation may negatively impact KSHV replication. To complement the XBP1s deficiency in the KSHV lytic cycle we overexpressed XBP1s using dox-inducible lenti- viral vectors. We engineered myc-tagged XBP1s to be weakly expressed under the control of a single tet operator (TetO) element, or strongly expressed under the control of seven tandem PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 12 / 34 KSHV modulates the host unfolded protein response to support lytic replication PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 13 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 6. ATF6α and PERK support robust KSHV replication. (A) TREx BCBL1-RTA cells were transduced with lentivirus expressing either non-targeting shRNA (NS) or shRNA against ATF6 and selected for with 1 μg/mL puromycin. Following selection, cells were treated with 5 ug/mL Tm for 6 h and whole cell lysates were analyzed by immunoblots for ATF6 and BiP, a transcriptional target of ATF6. (B) As in (A), cells were treated with 1 μg/mL dox for 48 h and virus-protected genomic DNA from cell supernatants were column purified for determining virus titer by qPCR using primers against ORF26. Firefly luciferase plasmid DNA was added during DNA purification for normalization using the ΔΔCt method. (C and D) TREx BCBL1-RTA cells were pre-treated with 10 μM Ceapin-A7 (ATF6 inhibitor) or 10 μM S1P inhibitor (PF 429242) for 20 h followed by 75 nM Thapsisgargin (Tg) for 4 h and then harvested for total RNA. Relative mRNA levels of (C) ATF6-target gene BiP and (D) XBP1-target gene ERdj4 were measured by qPCR, respectively. 18S rRNA was used as a reference gene and changes in mRNA levels were calculated by the ΔΔCt method. (E) TREx BCBL1-RTA cells were treated with 1 μg/mL dox and 10 μM of either Ceapin-A7 or S1P inhibitor PF 429242 for 48 h and virus-protected genomic DNA from cell supernatants was measured by qPCR to determine changes in virus titer using the ΔΔCt method by normalizing to luc2 DNA levels. (F) iSLK.219 cells were treated with 1 μg/mL dox with or without 10 μΜ Ceapin-A7 or S1P inhibitor for 96 h and virus-containing supernatants were serially diluted and spinfected onto a monolayer of 293A cells. GFP-positive cells (infected) were quantified by flow cytometry the following day and virus titer (IU/mL) was calculated as described in the Materials & Methods. (G) TREx BCBL1-RTA cells were treated with 1 μg/mL dox for 48 h with or without 500 nM PERKi GSK2606414 or 250 nM ISRIB, and virus titer was measured by qPCR as in (B). Data are represented as 3 (B, C, D, and F) or 4 (E and G) independent experiments and error bars denote SEM. One-way ANOVA and a post-hoc multiple comparisons tests were done to determine statistical significance. (�, p value < 0.05; ��, p value < 0.01; ��, p value < 0.001; ��, p value < 0.0001). https://doi.org/10.1371/journal.ppat.1008185.g006 TetO elements (7xTetO). We transduced iSLK.219 cells with these constructs or an empty con- trol vector and selected stable cells with blasticidin. With the addition of dox, XBP1s and RTA were concurrently expressed from dox-inducible promoters. To demonstrate that the ectopic XBP1s was functional we measured mRNA levels of the XBP1s-target gene ERdj4 and observed that it is upregulated in cells expressing XBP1s from the 7xTetO compared to empty vector control and peaks at 24 h post-dox addition (Fig 8A). These cells also displayed higher levels of mRNA and protein for RTA and the RTA target gene ORF45 by 24 h post-dox treatment; and by 48 h, mRNA encoding the late viral protein K8.1 was markedly increased compared to con- trols (Fig 8A and 8B), suggesting accelerated viral genome replication. Indeed, intracellular levels of viral genomes at 96 h post-dox were two-fold higher in XBP1s-overexpressing cells compared to empty vector (Fig 8C). We harvested cell supernatants at 48, 72, and 96 h post- dox and measured virion titer as previously described. Surprisingly, despite accelerated viral gene expression and genome replication in XBP1s-overexpressing cells, there was a dramatic, dose-dependent reduction in virion production by 72 and 96 h post-dox compared to empty vector control (Fig 8D). There was also a corresponding 20-fold decrease in release of viral par- ticles by XBP1s-overexpressing cells compared to controls, as measured by qPCR for capsid- protected viral genomic DNA (Fig 8E). Thus, while ectopic XBP1s expression promotes KSHV lytic gene expression and genome replication, it prevents efficient release of infectious progeny. Since XBP1s transactivates the RTA promoter, we hypothesized that this defect in virion production could be a negative consequence of RTA hyper-activation. We observed a decrease in the accumulation of capsid proteins ORF26 and ORF65 (Fig 8B), which we speculated could result from RTA hyper-activation and negatively impact late stages of replication [83]. However, experiments with the weaker 1xTetO-XBP1s construct revealed a 2-fold decrease in virion release at 96 h post-dox (Fig 8D) without affecting RTA, ORF26 and ORF65 protein accumulation (Fig 8B). To confirm that the diminished production of infectious virions from XBP1s-overexpressing cells is not due to enhanced RTA expression, we also overexpressed RTA in parallel from a 7xTetO in iSLK.219 cells, such that dox addition causes RTA expression by two dox-responsive promoters. KSHV from iSLK.219 cells also express monomeric red fluorescent protein (mRFP) from the viral lytic PAN promoter and can be used to monitor virus reactivation [84]. At 48 h post-dox addition, the levels of mRFP were similar between PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 14 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 7. IRE1 is required for efficient KSHV replication. (A) TREx BCBL1-RTA cells were treated with 1 μg/mL dox for 48 h with or without increasing concentrations of IRE1 inhibitor 4μ8c and DNase I-protected genomic DNA from cell supernatants was measured by qPCR to determine changes in virus titer using the ΔΔCt method by normalizing to luc2 DNA levels. The mean of four independent experiments are graphed -/+ SEM. (B) iSLK.219 cells were treated with 1 μg/mL dox with or without 10 or 25 μΜ 4μ8c for 96 h and virus-containing supernatants were serially diluted and spinfected onto a monolayer of 293A cells. GFP-positive cells (infected) were quantified by flow cytometry the following day and virus titer (IU/mL) was calculated as described in the Materials & Methods. The mean of four independent experiments are graphed -/+ SEM. (C) iSLK.219 cells were transduced with two different pLKO.1-blast shRNA lentiviruses targeting IRE1α or a non-targeting control and selected for with blasticidin. Following selection, cells were treated with 150 nM Thapsisgargin (Tg) for 4 h and harvested for immunoblot analysis to confirm IRE1α knockdown. The immunoblot shown is representative of two independent experiments performed. (D) Lentivirus transduced iSLK.219 cells from (C), were treated with dox for 96 h and virus titer was determined by flow cytometry as previously described. The data are represented as the change in virus titer relative to the non-targeting shRNA control sample and the mean of three independent experiments are graphed -/+ SEM. One-way ANOVA and multiple comparisons test were done to determine statistical significance in (A), (B), and (D). (ns, not statistically significant; �, p value < 0.05; ��, p value < 0.01; ��, p value < 0.001). https://doi.org/10.1371/journal.ppat.1008185.g007 XBP1s and RTA-expressing cells and noticeably greater than that of the empty vector control (Fig 8F). We harvested virus-containing supernatants at 24, 48, 72, and 96 h post-dox and PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 15 / 34 KSHV modulates the host unfolded protein response to support lytic replication PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 16 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 8. XBP1s overexpression inhibits KSHV replication at a late stage. (A) iSLK.219 cells were transduced with lentiviral dox-inducible expression vectors encoding myc-XBP1s whose expression is driven from 7x tandem tet operator (TetO) for robust gene expression. Following blasticidin expression, lytic replication was induced with dox for 0, 24, 48, 72, and 96 h or 96 h with 500 nM PAA (to inhibit genome replication) and harvested for total RNA. mRNA levels of XBP1s-target gene ERdj4 and viral genes RTA (immediate early gene [IE]), ORF45 (early gene [E]), and K8.1 (late gene [L]) were measured by qPCR using 18S rRNA as a reference gene for normalization. An average of 3 independent experiments are graphed and error bars denote SEM. (B) iSLK.219 cells were transduced with lentiviral dox- inducible expression vectors encoding myc-XBP1s whose expression is driven from either 1x TetO (weak expression) or 7x TetO (strong expression) and treated with dox for the indicated times and harvested for total cell lysates. Immunoblots were done for myc-epitope tag (XBP1s), and the viral proteins RTA (IE), ORF45 (E), ORF65 (L), and ORF26 (L). β-actin was used as a loading control. The presented immunoblots are representative of 2 independent experiments. (C) 7xTetO-myc-XBP1s and vector transduced iSLK.219 cells were treated with dox for 96 h and intracellular DNA purified. qPCR against ORF26 DNA was done to measure the relative change in viral genome replication using the ΔΔCt method and normalized to β-actin DNA. Values are the mean of 3 independent experiments -/+ SEM. (D) Virus- containing supernatants from (C) were serially diluted and spinfected onto a monolayer of 293A cells. GFP-positive cells were quantified by flow cytometry the following day and used to calculate virus titer (IU/mL). The values are the mean virus titer of 4 independent experiments -/+ SEM. (E) 7xTetO-myc-XBP1s and vector transduced iSLK.219 cells were treated with dox for 96 h and DNase-protected genomic DNA from supernatants were column purified. Firefly luciferase plasmid DNA was added during DNA purification to allow for normalization. The relative change in virus titer (infectious and non-infectious) was quantified by qPCR using ORF26 primers and normalized to luciferase DNA using the ΔΔCt method. Values are the mean of 4 independent experiments -/+ SEM. (F and G) iSLK.219 cells were transduced with lentiviral expression vectors encoding either dox-inducible 7xTetO-myc-XBP1s or 7xTetO-FLAG-RTA. Following blasticidin selection, lytic replication was induced with dox for (F) 48 h and fluorescence microscopy was used to image RFP-positive cells (cells undergoing lytic replication) or for (G) 24, 48, 72, and 96 h and virus supernatants were harvested to measure titer by flow cytometry. Values are an average of 3 independent experiments -/+ SEM. https://doi.org/10.1371/journal.ppat.1008185.g008 measured virion release by flow cytometry following infection of a naïve 293A monolayer (Fig 8G). At 24h, when virion production is negligible in empty vector- and XBP1s-expressing cells, there is a significantly higher level of virions produced by 7xTetO RTA-expressing cells which continues up until 72 h. At 96 h post-dox, virion production from the 7xTetO RTA- expressing cells hits a plateau. Here again, in agreement with our previous observations, virion production from 7xTetO XBP1s-expressing cells is comparable to empty vector control after 48 h but by 96 h post-dox virion production is ~20-fold lower. These data demonstrate that the dramatic reduction in virions produced by XBP1-overexpressing cells is not due to RTA hyper-activation. Our findings indicate that despite the important role that XBP1s plays in reactivation from latency, ectopic expression of XBP1s suppresses virion production in iSLK.219 cells. To cor- roborate these observations, we transduced iSLK.219 cells with increasing concentrations of lentiviral vectors that constitutively express myc-tagged XBP1s or FLAG-tagged RTA from a CMV promoter. To ensure that there were would be an equivalent level of reactivation between cells constitutively expressing RTA or XBP1s, we chose virus dilutions that resulted in a similar range of mRFP levels (indicating lytic cycle initiation) as monitored by fluorescence microscopy (Fig 9A). Six days after cell transduction, virus-containing supernatants were har- vested, and infectious virions were enumerated by infecting naive 293A cells and detecting GFP positive cells by flow cytometry; as a positive control, we harvested virus from untrans- duced iSLK.219 cells treated with dox for 72h (Fig 9B). Here, we observed an expected dose- dependent increase in virion release from RTA-transduced iSLK.219 cells. By contrast, increasing levels of XBP1s did not increase virion production by these cells, and the highest virion yield was almost 100-fold lower than the corresponding highest yield from RTA-trans- duced cells or dox-treated iSLK.219 cells. In RTA-transduced cells, RTA and ORF45 protein levels increase in step with increasing fluorescent signal of the mRFP reporter, but surprisingly, in XBP1-expressing cells, increasing XBP1s expression had a nominal impact on RTA and ORF45 protein accumulation (Fig 9C). However, we observed that escalating doses of ectopic XBP1s altered the electrophoretic mobility of ORF45 and prevented accumulation of the late protein ORF65. These findings confirm that XBP1s prevents release of KSHV virions in the iSLK.219 cell model, but do not pinpoint the defect in replication. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 17 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 9. XBP1s ectopic expression inhibits KSHV replication in iSLK.219 cells. (A,B, and C) iSLK.219 cells were transduced with increasing concentrations of CMV-driven lentiviral expression vectors encoding myc-XBP1s (LV dilutions: 1:400, 1:200, 1:100, 1:50, 1:25) or FLAG-RTA (LV dilutions: 1:32, 1:16, 1:8, 1:4, 1:2) for 6 days or treated with dox for 72 h (as a positive control) and (A) a subset of the lentiviral dilutions, cells were imaged for RFP-positive cells by fluorescence microscopy; (B) virus-containing supernatants were harvested for measuring virus titer by flow cytometry; and (C) cell lysates were analyzed by immunoblots for myc-epitope tag (XBP1s), and the viral proteins RTA, ORF45, and ORF65. β-actin was used as a loading control. https://doi.org/10.1371/journal.ppat.1008185.g009 To determine if this inhibition of virion production by XBP1s is observed in other cell lines, we also transduced TREx BCBL1-RTA cells with increasing concentrations of lentiviral vectors expressing either CMV-driven myc-tagged XBP1s or FLAG-tagged RTA. 48 h after transduc- tion, immunoblot analysis revealed that contrary to the situation in iSLK.219 cells, ectopic expression of XBP1s in TREx BCBL1-RTA cells caused dose-dependent increase in RTA, ORF45, and ORF65 (Fig 10A). Likewise, XBP1s caused a dose-dependent increase in virus production at 72 h post-transduction (Fig 10B). Indeed, cells transduced with the highest con- centration of myc-XBP1s lentiviral vector produced as much virus as cells transduced with the highest concentration of FLAG-RTA lentiviral vector or control TREx BCBL1-RTA cells reac- tivated with dox alone for 48 h. Taken together, these observations suggest that XBP1s induc- tion of the lytic cycle can stimulate a productive infection in TREx BCBL1-RTA cells or an abortive infection in iSLK.219 cells, and that the outcome is likely determined during later stages of replication. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 18 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 10. XBP1s ectopic expression does not inhibit KSHV replication in TREx BCBL1-RTA cells. (A) TREx BCBL1-RTA cells were treated with 1 μg/mL dox or transduced with increasing concentrations of lentiviral vectors encoding CMV-driven expression of FLAG-RTA (lentivirus dilutions: 1:16, 1:8, 1:4, and 1:2) or myc-XBP1s (lentivirus dilutions: 1:32, 1:16, 1:8, and 1:4) for 48 h. Cell lysates were harvested for immunoblot analysis for myc-epitope tag (XBP1s) and the viral proteins RTA, ORF45, and ORF65. β-actin was used as a loading control. The immunoblots shown are representative of two independent experiments. (B) TREx BCBL1-RTA cells were treated with dox or transduced with the same lentivirus concentrations of FLAG-RTA or myc-XBP1s as in (A). 72 h post-transduction or post-dox, the supernatant was harvested and DNase I- protected viral genomic DNA was measured by qPCR using primers against ORF26. Firefly luciferase plasmid DNA was added during DNA purification for normalization using the ΔΔCt method. The mean of 3 independent experiments is shown and the error bars correspond to the SEM. https://doi.org/10.1371/journal.ppat.1008185.g010 The iSLK.219 cells harbour a genetically-modified KSHV (rKSHV.219) derived from the JSC-1 parental strain [84], whereas the TREx BCBL1-RTA cells carry the BCBL1 KSHV strain. These KSHV strains are highly similar, with minor differences evident in internal repeat regions and the K1 gene. To determine whether these subtle differences in KSHV strains deter- mine susceptibility to XBP1s restriction, we transduced naïve iSLK cells with lentiviruses expressing myc-tagged XBP1s or empty vector control, and then infected them with virus-con- taining supernatant harvested from lytic iSLK.219 or TREx BCBL1-RTA cells. These cells were immediately treated with dox to stimulate RTA transgene expression and bypass latency estab- lishment, spurring lytic KSHV replication. By 4 days post-infection (dpi) we observed KSHV replication and plaque formation in cells transduced with the empty lentiviral control vector which was enhanced by dox treatment (Fig 11A). Plaque formation was most evident in cell monolayers infected with the TREx-BCBL1-RTA-derived virus. By contrast, iSLK cells that expressed myc-XBP1s showed signs of increased lytic reactivation (higher number of RFP-pos- itive cells in rKSHV.219-infected cell monolayers and increased cell detachment in BCBL1-in- fected cell monolayers), but plaques did not form. This suggests that XBP1s can block plaque formation by both KSHV strains. To quantify this effect, we measured levels of intact progeny viral particles in cell supernatants harvested at 4 dpi by qPCR amplification of capsid-protected genomes, as described above. We observed that ectopic XBP1s expression dramatically reduced the production of both KSHV strains (Fig 11B). This decrease in viral particle produc- tion by myc-XBP1s-expressing cells following de novo infection correlated with decreased lev- els of RTA and the early protein ORF57 at 4 dpi, which likely results from a blockade in multi- round replication (Fig 11C). These findings indicate that the JSC-1 and BCBL1 strains of KSHV are equally susceptible to the antiviral effects of XBP1s during lytic replication in epi- thelial cells. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 19 / 34 KSHV modulates the host unfolded protein response to support lytic replication Fig 11. XBP1s ectopic expression inhibits virion production of both TREx-BCBL1-RTA-derived virus and rKSHV.219 in de novo infection. (A, B, and C) iSLK cells were transduced with CMV-driven lentiviral expression vectors encoding myc-XBP1s then infected with virus-containing supernatant from lytic iSLK.219 or TREx-BCBL1-RTA cultures the following day. After infection, the medium was replaced with fresh medium containing 1 μg/mL doxycycline. Supernatant was removed at 96 h post-infection and (A) the cell monolayer was fixed with paraformaldehyde and imaged. (B) Supernatant from de novo infected iSLK cells was harvested at 96 h post-infection and DNase I-protected viral genomic DNA was measured by qPCR as described above. The mean of four independent experiments is shown and the error bars correspond to the SEM. (��, p value < 0.0001) (C) Cell lysates were harvested at 96 h post-infection for immunoblot analysis for myc-epitope (XBP1s) and the viral proteins RTA (immediate early) and ORF57 (early). β- actin was used as a loading control. The immunoblots shown are representative of at least two independent experiments. https://doi.org/10.1371/journal.ppat.1008185.g011 Discussion XBP1s-mediated transactivation of the RTA promoter causes KSHV reactivation from latency, but little is known about how ER stress and the UPR affect the ensuing lytic replication cycle. Here, we report that activation of all three UPR sensors (PERK, IRE1, ATF6) is required for PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 20 / 34 KSHV modulates the host unfolded protein response to support lytic replication efficient KSHV lytic replication, because genetic or pharmacologic inhibition of each UPR sen- sor diminishes virion production. Despite UPR sensor activation during KSHV lytic replica- tion, downstream UPR transcriptional responses were restricted. We do not yet know precisely how KSHV blocks the accumulation of ATF4 despite PERK activation and sustained eIF2α phosphorylation, nor do we understand why XBP1s protein does not accumulate despite IRE1 activation and Xbp1 mRNA cleavage, or why ATF6(N) fails to transactivate target genes. The failure of SOX to suppress UPR transcription in our hands suggests that KSHV regulation of UPR is complex and likely represents the collective action of multiple viral gene products. Our ongoing studies are focused on identifying and characterizing these UPR-modulating viral gene products. How might KSHV activate UPR sensors? All three UPR sensors are activated concurrently during the lytic cycle, which suggests that they may respond to canonical protein misfolding events in the ER. Such misfolding could result from biogenesis of viral glycoproteins, but we think this is unlikely because UPR sensor activation in the TREx BCBL1-RTA cell model pre- cedes viral genome replication and bulk synthesis of structural proteins. We favour an alterna- tive hypothesis whereby lytic replication disrupts ER chaperone function in some way, perhaps through the direct action of viral proteins or indirect activation of signal transduction pathways. Indeed, early in KSHV lytic replication, production of a viral G protein-coupled receptor (vGPCR) activates stress signalling and increases cytosolic calcium at least in part by inhibiting the SERCA calcium pump [85]. In this scenario, vGPCR may act similarly to Tg, inducing broad activation of all three UPR sensors by depleting ER calcium stores essential for protein folding. It seems counterintuitive that an enveloped virus would actively trigger ER stress, but we speculate that UPR activation in the earliest stages of lytic replication could help prime the ER for the impending burden of modifying secreted viral proteins and viral trans- membrane glycoproteins. A more comprehensive accounting of changes in gene expression during early lytic replication is required to determine whether there is indeed an activation of an adaptive UPR to prepare the cell for late gene expression. Why would a virus elicit ER stress but ultimately block the downstream UPR stress-mitigat- ing transcription? UPR signalling is etiologically-linked to inflammatory diseases including Crohn’s Disease and type 2 diabetes [86–88]. Furthermore, maximal IFN production following Toll-like receptor (TLR) engagement has been shown to require UPR activation and XBP1s activity [89]. Thus, KSHV suppression of UPR transcription may dampen antiviral inflamma- tory responses. Furthermore, because UPR transcription factors transactivate genes involved in catabolism, we speculate that blockade of their activity may allow newly synthesized viral gene products to evade degradation. For example, ATF6 and XBP1s transactivate genes involved in ERAD [29,61], so suppression of ATF6/XBP1s transcription could prevent ERAD- mediated degradation of newly-synthesized viral glycoproteins. Similarly, because ATF4 and its downstream target CHOP transactivate autophagy [16] and autophagy restricts herpesvirus replication [90,91], suppression of ATF4 and CHOP accumulation may allow KSHV to evade negative catabolic effects of autophagy. ATF4 and CHOP also promote apoptosis [92] and viral suppression of CHOP may thereby extend the survival of cells enduring the later stages of lytic replication. The consequences of allowing unchecked accumulation of misfolded proteins during lytic replication remain unclear. Autophagy has been shown to compensate and assist with bulk protein degradation when ERAD is defective or inadequate for degrading misfolded proteins like aggregation-prone membrane proteins [93,94]. Since ERAD genes are not upregulated during lytic replication, it is possible that autophagy is required to clear misfolded proteins. However, there are multiple KSHV proteins that have been shown to directly inhibit autop- hagy including vFLIP, K-survivin, and vBCL2 [95–97], as well as multiple KSHV proteins that PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 21 / 34 KSHV modulates the host unfolded protein response to support lytic replication stimulate the pro-growth PI3K-Akt-mTOR signalling pathway [98,99]. Simultaneous inhibi- tion of ERAD and autophagy likely would cause the accumulation of the misfolded proteins, including aggregated membrane proteins, in KSHV infected cells. Mechanisms that KSHV employs (if any) to circumvent a possible accrual of misfolded proteins remain unknown. The accumulating evidence indicates that herpesviruses have evolved distinct mechanisms to regulate the UPR to promote viral replication. For example, HSV-1 has been shown to dys- regulate IRE1 activity [100]; HSV-1-infected cells displayed increased IRE1 protein levels but diminished IRE1 RNase activity, and treatment with the IRE1 inhibitor STF-083010 [101] diminished viral replication [102]. Moreover, HSV-1 glycoprotein B (gB) binds and inhibits PERK to support robust viral protein synthesis [56]. Interestingly, ectopic expression of FLAG-tagged XBP1s inhibits HSV-1 replication [102], consistent with our findings in the KSHV iSLK.219 infection model. By contrast, human cytomegalovirus (HCMV) lytic replica- tion in fibroblast cells selectively activates PERK and IRE1, but not ATF6 [54]. Consistent with our model, IRE1 activation causes normal Xbp1 mRNA splicing but fails to upregulate the XBP1s target gene EDEM (XBP1s protein levels were not evaluated in this study). However, unlike our observations in KSHV lytic infection, PERK causes ISR activation and normal ATF4 accumulation during HCMV lytic replication [54]. A recent study linked HCMV PERK and IRE1 activation to the UL148 glycoprotein [55]. In cells infected with a UL148-deficient HCMV, XBP1s protein levels were diminished compared to wildtype but levels of the XBP1s- target gene EDEM1 are unaffected, suggesting that XBP1s activity may also be inhibited in CMV-infected cells. Another study showed that HCMV UL38 induces eIF2α phosphorylation and ATF4 synthesis and promotes cell survival and virus replication during drug-induced ER stress by repressing JNK phosphorylation [103]. HCMV UL50 protein and the murine cyto- megalovirus (MCMV) ortholog M50 bind IRE1 and promote its degradation, thereby inhibit- ing Xbp1 mRNA splicing [104]. We suspect that many more viral regulators of the UPR will be discovered in the coming years. Such studies will benefit from the development of appropriate platforms to screen hundreds of viral ORFs for UPR and ISR-modulating activity. The activation of IRE1 while simultaneously inhibiting XBP1s protein accumulation during KSHV lytic replication is curious. IRE1 can cleave a subset of ER-targeted mRNAs for degra- dation in a process called RIDD [31]. The physiological role of RIDD is unclear but it is thought that like the attenuation of translation by PERK, RIDD also helps to reduce the trans- lation load in the ER. It is possible that certain viruses may hijack RIDD as a form of host shut- off to ensure preferential translation of secreted viral proteins. Conversely, if viral mRNAs contain XBP1-like cleavage sites, then the virus would likely want to suppress IRE1 RNase activity [105]. The molecular events that direct IRE1 toward RIDD or Xbp1 mRNA splicing are not well understood but one study suggests that the higher-order oligomerization of IRE1 can dictate the response, whereby oligomeric IRE1 prefers Xbp1 as a substrate, whereas dimeric IRE1 preferentially cleaves mRNAs through RIDD [106]. Since the levels of IRE1 are impacted by many of these viruses, including CMV, HSV-1, and KSHV, it will be interesting to determine if IRE1 specificity of mRNAs is impacted during infection. We do not yet understand why XBP1s ectopic expression blocked KSHV virion production by iSLK.219 epithelial cells but not the PEL-derived (and engineered) TREx BCBL1-RTA cells. The rKSHV.219 (JSC-1 parental strain) and BCBL-1 strain are highly similar, with only signifi- cant differences evident in internal repeat regions and the K1 gene. We observed that these two virus strains, when used to infect naïve iSLK cells, were inhibited to a similar extent by ectopic XBP1s expression. This finding indicates that prior establishment of latency is not required to license XBP1s antiviral effects in epithelial cells, and also suggests that the failure of XBP1s to restrict KSHV virion production in the TREx BCBL1-RTA cell model may be due to cell intrinsic differences between epithelial cells and B lymphocytes. Additional mechanistic PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 22 / 34 KSHV modulates the host unfolded protein response to support lytic replication studies will be required to fully explore these differences. Our work showed that ectopic XBP1s expression enabled RTA accumulation and KSHV reactivation from latency in both TREx BCBL1-RTA cells and iSLK.219 cells to a similar extent, and for the most part did not disrupt viral gene expression or genome replication. However, the diminished accumulation of struc- tural proteins like ORF26 and ORF65 in iSLK.219 cells in the presence of high levels of XBP1s suggest a defect in the late stages of replication, which may give some clues about its antiviral mechanism of action. Thus, although XBP1s restriction of KSHV replication in the iSLK.219 cell model would seem to provide a tidy explanation for viral inhibition of XBP1s protein accu- mulation and UPR transcription, further mechanistic studies will be required to fully charac- terize the antiviral nature of XBP1s in certain cell types. Materials and methods Cell culture and chemicals 293A (ThermoFisher), HEK293T (ATCC), Hela Tet-Off (Clontech), iSLK and iSLK.219 cells (iSLK and iSLK.219 cells were kind gifts from Don Ganem) were cultured in DMEM (Ther- moFisher) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 Units/mL penicillin, 100 μg/mL streptomycin, and 2mM L-Glutamine. iSLK.219 cells were also passaged in the presence of 10 μg/mL of puromycin (ThermoFisher) to maintain the rKSHV.219 episomal DNA. TREx BCBL1-RTA cells (a kind gift from Jae Jung) were cultured in RPMI- 1640 supplemented with 10% heat-inactivated FBS, 500 μM β-mercaptoethanol and the same concentrations of penicillin, streptomycin, and L-Glutamine as the adherent cell lines. All cells were maintained at 37˚C with 5% CO2. 293A Tet-On cells were generated by transducing 293A cells with an MOI of < 1 of a lenti- viral expression vector (see below for methods to generate lentiviral vectors and transducing cells) that encodes rtTA3 and the antibiotic selection marker for hygromycin B from the plas- mid pLenti CMV rtTA3 Hygro (w785-1), which was a gift from Eric Campeau (Addgene plas- mid # 26730). 24 h post-transduction fresh medium was added containing 200 μg/mL Hygromycin B (ThermoFisher) and positive transductants were selected for approximately 2 weeks. To induce lytic replication via expression of the RTA transgene in TREx BCBL1-RTA and iSLK.219 cells, 1 μg/mL of doxycycline (dox; Sigma) was added to the cells. iSLK.219 cells were seeded at a density of 2x105 cells/well of a 6-well plate one day prior to lytic induction and TREx BCBL1-RTA cells were seeded at a concentration of 2.5x105 cells/mL immediately prior to lytic induction. 4μ8c (Axon Medchem and Sigma), PERKi (GSK2606414; Tocris), ISRIB (Sigma), Ceapin- A7 (Sigma), thapsigargin (Sigma), and tunicamycin (Sigma) were dissolved in DMSO (Sigma) and S1P inhibitor PF 429242 (Tocris) was dissolved in nuclease-free water to stock concentra- tions and diluted to the indicated concentrations in cell media. Plasmids To generate the plasmids pLJM1 B� Puro and pLJM1 B� BSD, the genes encoding puromycin N-acetyltransferase and blasticidin S deaminase were amplified from pBMN-IRES-Puro and pBMN-IRES-Blast, respectively (previously generated in the lab by switching GFP with PuroR or BlastR from pBMN-I-GFP that was generated by the Nolan Lab [Stanford]), with forward and reverse primers containing BglII and KpnI RE sites. The antibiotic selection cassettes were cut-and-pasted into pLJM1.D (previously generated in the lab from modifying the MCS of pLJM1 plasmid that was generated by the Sabatini Lab [MIT]), deleting the BamHI RE site. A new MCS was generated by annealing overlapping oligos containing NheI and MfeI forward PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 23 / 34 KSHV modulates the host unfolded protein response to support lytic replication and reverse RE sites and inserted into the plasmid digested with NheI and EcoRI to replace the existing MCS with unique RE sites in the following order: NheI, AgeI, BamHI, EcoRI, PstI, XbaI, MluI, SalI, EcoRV. 1x TO (tet operator) and 7x TO promoters were generated by anneal- ing overlapping oligos containing one or seven copies of the tetracycline operator (5’-TCCCT ATCAGTGATAGAGA-3’) and a minimal CMV promoter and using PCR extension to amplify a blunt oligo containing NdeI and NheI RE sites. The amplicon was digested with NdeI and NheI and pasted into the pLJM1 B� BSD replacing the CMV promoter. To generate pLJM1 B� Puro HA-ATF6, HA-ATF6 was PCR amplified from pCGN-ATF6 (a gift from Ron Prywes, Addgene plasmid # 11974) and cut and pasted into pLJM1 B� Puro with NheI and AgeI. pCMV2B-XBP1s was generated by PCR amplifying XBP1 from total RNA isolated from TREx BCBL1-RTA cells treated with Tg for 4h. After PCR, DNA was digested with PstI (cleaves unspliced XBP1 DNA only) to enrich for the spliced isoform of XBP1 and cut and pasted into pCMV2B with BamHI and XhoI and in-frame of the N-terminal FLAG-tag. pLJM1 BSD CMV, 1x TetO, and 7x TetO myc-XBP1s were generated by including the myc- tag nucleotide sequence in the 5’ primer and amplifying XBP1s from pCMV2B-XBP1s. Cloned myc-XBP1s was subsequently cut and pasted into their corresponding lentiviral plasmids with BamHI and SalI. FLAG-RTA was cut from pcDNA3-FLAG-RTA (previously generated in the lab) with RE sites EcoRI and XhoI and pasted into pLJM1 B� BSD CMV and 7x TetO with RE sites EcoRI and SalI (via compatible sticky ends between XhoI and SalI). pLJM1 BSD 7x TetO myc-SOX WT and P176S plasmids were generated by including the myc-tag nucleotide sequence in the 5’ primer and amplifying wildtype SOX and the P176S point mutant from iSLK.219 and iSLK.BAC16 SOX P176S mutant genomic DNA [76], respec- tively. Amplified myc-SOX was cut and pasted into pLJM1 BSD 7x TetO via the RE sites NheI and SalI. Lentiviral vectors HEK293T cells were seeded on 10 cm plates for 60–70% cell density the following day. Cells were transfected with polyethylenimine (PEI; Sigma) and the following plasmids for lentiviral generation: pLJM1 transfer plasmid, pMD2.G, and psPAX2. pMD2.G and psPAX2 are gifts from Didier Trono (Addgene plasmids # 12259 and # 12260). 48 h post-transfection, lentivirus containing supernatants were passed through a 0.45 μM filter and frozen at -80˚C. For transducing iSLK.219, iSLK.219 cells were seeded at a density of 5x104 cells/well of a 6-well plate and the following day were resuspended in media containing 4 μg/mL polybrene (Sigma). Lentivirus was serially diluted dropwise onto cells and incubated at 37˚C for 24 h. Following infection, the media was refreshed containing 10 μg/mL blasticidin. For de novo infection of iSLK cells, cells were seeded at a density of 2x105 cells/well and transduced with a low dilution of lentivirus. These cells were infected one day after transduction and were not selected for lentivirus integration. For transducing TREx BCBL1-RTA cells, cells were seeded at a concentration of 2.5x105 cells/mL in media containing 4 μg/mL polybrene and lentivirus was serially diluted onto cells. 24 h post-infection, the media was replaced with media contain- ing 1 μg/mL puromycin. The first lentivirus dilution that resulted in minimal cell death follow- ing antibiotic selection was used for subsequent experiments. shRNA lentivirus cloning and knockdown The pLKO.1 -TRC control plasmid was a gift from David Root (Addgene plasmid # 10879) and the RNAi Consortium was used to design the following oligos for generating shRNA lenti- viral vectors against ATF6α and IRE1α with the targeting sequences underlined (Fwd/Rev): PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 24 / 34 KSHV modulates the host unfolded protein response to support lytic replication shAFT6α (TRCN0000416318): 5’- CCGGACAGAGTCTCTCAGGTTAAATCTCGAGAT TTAACCTGAGAGACTCTGTTTTTTG/5’- AATTCAAAAAACAGAGTCTCTCAGGTTA AATCTCGAGATTTAACCTGAGAGACTCTGT-3’ shIRE1α-1 (TRCN0000356305): 5’- CCGGTCAACGCTGGATGGAAGTTTGCTCGAGC AAACTTCCATCCAGCGTTGATTTTTG -3’/5’- AATTCAAAAATCAACGCTGGATGGA AGTTTGCTCGAGCAAACTTCCATCCAGCGTTGA-3’ shIRE1α-2 (TRCN0000235529): 5’- CCGGAGAGGAGGGAATCGTACATTTCTCGAGA AATGTACGATTCCCTCCTCTTTTTTG-3’/5’- AATTCAAAAAAGAGGAGGGAATCGTA CATTTCTCGAGAAATGTACGATTCCCTCCTCT-3’ The cloning strategy on Addgene (http://www.addgene.org/tools/protocols/plko/) was used to clone the oligos into either pLKO.1-Puro or pLKO.1-Blast and lentivirus generation was completed as described above. Fluorescent imaging For Fig 8, iSLK.219 cells were seeded at a density of 2x105 cells/well of a 6-well plate and 48 h post-dox, brightfield images and fluorescent images were captured with Olympus CKX41 microscope fitted with a QImaging QICAM Fast 1394 digital camera and Lumencor Mira light engine and using the 10x objective. For Fig 9, iSLK.219 cells were seeded at density of 5x104 cells/well of a 6-well plate and transduced with the indicated lentiviral vectors. Fluo- rescent images were captured with an EVOS FL Cell Imaging System with 10x objective (ThermoFisher) following 5 days post-transduction or 48 h post-dox. Images of lytic foci (Fig 11A) were captured on an Axiovert 200M with a 5x objective using a Orca R2 mono- chrome camera Hamamatsu). All greyscale images were false coloured red or green for presentation. Immunoblotting Cells were washed once with ice-cold PBS and lysed in 2x Laemmli Buffer (120mM Tris-HCl pH 6.8, 20% glycerol, 4% SDS). Lysates were passed through a 21-gauge needle 5–7 times to minimize viscosity and protein concentration was quantified with DC protein assay (Bio- Rad). DTT and bromophenol blue were added to samples, boiled, and 10–20 μg of protein per sample were loaded on 6–12% polyacrylamide gels and resolved by SDS-PAGE. Protein was transferred to PVDF membranes using the Trans-Blot Turbo Transfer System (Bio-Rad) and blocked in 5% skim milk diluted in Tris-buffered-saline (TBS) supplemented with 0.1% Tween 20 (Fisher Bio) (TBS-T) for 1 h at room temperature followed by overnight incubation at 4˚C with primary antibody diluted in 5% bovine serum albumin in TBS-T. Following washing of primary antibody, membranes were incubated with IgG HRP-linked antibodies followed by exposure to ECL-2 western blotting substrate (Thermo Scientific, Pierce) and imaged by che- mifluorescence on a Carestream Image Station 400mm Pro (Carestream) or using Clarity ECL luminescent reagent (Bio-Rad) and a ChemiDoc imaging system (Bio-Rad). The following antibodies were used: PERK (Cell Signaling Technology (CST); #5683); IRE1α (CST; #3294); ATF6α (Abcam; ab122897); XBP1 (CST; #12782); Phospho-eIF2α (Ser51; Abcam; ab32157); Total eIF2α (CST; #5324); ATF4 (Santa Cruz; sc-200); Myc-tag (CST; #2278); FLAG-tag (DYKDDDDK tag; CST; #8146); ORF26 (ThermoFisher; MA5-15742) ORF45 (ThermoFisher; MA5-14769); ORF57 (Santa Cruz; sc-135746); ORF65 (a kind gift from Shou-Jiang Gao); RTA (a kind gift from David Lukac); β-actin HRP conjugate (CST; #5125); anti-rabbit IgG HRP- linked (CST; #7074); anti-mouse IgG HRP-linked (CST; #7076). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 25 / 34 KSHV modulates the host unfolded protein response to support lytic replication XBP1 RT-PCR splicing assay RNA was isolated from TREx BCBL1-RTA cells with the RNeasy Plus Kit (Qiagen) and 500 ng total RNA was reverse transcribed with qScript cDNA SuperMix (Quanta) according to manu- facturers’ protocols. Based on the Ron Lab protocol (http://ron.cimr.cam.ac.uk/protocols/ XBP-1.splicing.06.03.15.pdf) and minor modifications, a 473 bp PCR product overlapping the IRE1 splice site was amplified with XBP1 Fwd primer (5’-AAACAGAGTAGCAGCTCA- GACTGC-3’) and XBP1 Rev primer (5’-TCCTTCTGGGTAGACCTCTGGGAG-3’). The amplified PCR product was digested overnight with 40 units of High Fidelity PstI (New England Biolabs) to cleave unspliced XBP1 cDNA. The PCR products were resolved on a 2.5% agarose gel made with 1x TAE (Tris-acetate-EDTA) buffer and stained with SYBR Safe (Ther- moFisher) and visualized on a ChemiDoc Imaging Station (Bio-Rad). Percent Xbp1 mRNA splicing was calculated by densitometry analysis with Image Lab software ver. 6.0.0 (Bio-Rad) and calculated using the following formula: % xbp1 mRNA splicing ¼ 0:5 � xbp1hybrid þ xbp1s xbp1hybrid þ xbp1s þ xbp1u1 þ xbp1u2 � 100 Quantitative Reverse-Transcription PCR (RT-qPCR) RNA was isolated from TREx BCBL1-RTA cells with the RNeasy Plus Kit (Qiagen) and 500 ng total RNA was reverse transcribed with qScript cDNA SuperMix (Quanta) according to manu- facturers’ protocols. A CFX96 Touch Real-Time PCR Detection System (Bio-Rad) and GoTaq qPCR MasterMix (Promega) was used to perform Real-Time PCR. Changes in mRNA levels were calculated by the ΔΔCt method [107] and normalized using 18S rRNA as a reference gene. The following primer sets were used in the study: 18S F: 5’-TTCGAACGTCTGCCCTATCAA-3’; R: 5’-GATGTGGTAGCCGTTTCTCAGG-3’ ATF4: 5’-CCACCATGGCGTATTAGGGG-3’; R: 5’-TAAATCGCTTCCCCCTTGGC-3’ BiP F: 5’-GCCTGTATTTCTAGACCTGCC-3’; R: 5’-TTCATCTTGCCAGCCAGTTG-3’ CHOP: 5’-ATGAACGGCTCAAGCAGGAA-3’; R: 5’-GGGAAAGGTGGGTAGTGTGG-3’ EDEM1 F: 5’-TTGACAAAGATTCCACCGTCC-3’; R: 5’-TGTGAGCAGAAAG- GAGGCTTC-3’ ERdj4 F: 5’-CGCCAAATCAAGAAGGCCT-3’; R: 5’-CAGCATCCGGGCTCTTATTTT-3’ HERPUD1 F: 5’-AACGGCATGTTTTGCATCTG-3’; R: 5’-GGGGAAGAAAGGTTCCGA AG-3’ K8.1 F: 5’-AGATACGTCTGCCTCTGGGT-3’; R: 5’-AAAGTCACGTGGGAGGTCAC-3’ ORF26 F: 5’-CAGTTGAGCGTCCCAGATGA-3’; R: 5’-GGAATACCAACAGGAGGCCG-3’ ORF45 F: 5’-TGATGAAATCGAGTGGGCGG-3’; R: 5’-CTTAAGCCGCAAAGCAGTGG-3’ RTA F: 5’-GATTACTGCGACAACGGTGC-3’; R: 5’-TCTGCGACAAAACATGCAGC Viral genome amplification DNA was harvested from iSLK.219 cells with DNeasy Blood & Tissue Kit (Qiagen) according to manufacturer’s protocol. RT-PCR was carried out as described previously with primers spe- cific to KSHV ORF26 (as listed previously) and β-actin (F: 5’-CTTCCAGCAGATGTGGAT CA-3’; R: 5’-AAAGCCATGCCAATCTCATC-3’). Changes in KSHV genome copy number was calculated by the ΔΔCt method and normalized to β-actin. Titering DNAse-protected viral genomes Virus containing cell supernatants were processed by first pelleting floating cells and debris and then 180 μL cleared supernatant were treated with 20 μL of 3 mg/mL DNase I (Sigma) for PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 26 / 34 KSHV modulates the host unfolded protein response to support lytic replication 30 minutes at 37˚C. Viral genomic DNA was then purified from the supernatant with DNeasy Blood & Tissue Kit (Qiagen) according to the manufacturer’s protocol with the following mod- ifications: 10 μg of salmon sperm DNA (Invitrogen) and 1 ng of luciferase plasmid pGL4.26 [luc2/minP/Hygro] (Promega) were added to the lysis buffer. RT-PCR was performed as described previously with primers specific to KSHV ORF26 (as listed previously) and luc2 (F: 5’-TTCGGCAACCAGATCATCCC-3’; R: 5’-TGCGCAAGAATAGCTCCTCC-3’). Changes in virus titer was calculated by the ΔΔCt method and normalized to luc2. rKSHV.219 infection and titering Virus-containing supernatant was harvested from iSLK.219 cells at the indicated times by pel- leting cellular debris at 3300 x g for 5 minutes and then stored at -80˚C until ready to titer the virus. 105 293A cells/well were seeded in 12-well plates to obtain a confluent monolayer two days later. The thawed viral inoculum was briefly vortexed and centrifuged again at 3300 x g for 5 minutes. Two-fold serial dilutions of viral supernatants were applied to the monolayer containing 4 μg/mL polybrene and 25 mM HEPES (Gibco) and centrifuged at 800 x g for 2 h at 30˚C. The total cell count per well was also determined from an uninfected well. Fresh media was applied immediately after spinoculation. 20–24 h post-infection, two dilutions that resulted in less than 30% GFP-positive cells (the linear range for infection) were trypsinized, washed once with PBS and fixed with 1% paraformaldehyde in PBS. GFP-positivity was mea- sured on either FACSCalibur or FACSCanto cytometers (BD) by gating on FSC/SSC and counting 10000–15000 “live” events. Gating and % GFP positive events were determined with FCS Express 6 Flow Cytometry Software (ver.6.0; De Novo). Virus titer was calculated as IU/ mL with the following formula: Virus titre � � IU mL ¼ % GFP positive events � dilution factor � cell count=100 The virus titer of the two dilutions were averaged for the final titer value. De novo infection rKSHV.219 was harvested from the supernatant of iSLK.219 cells four days after lytic reactiva- tion with 1 μg/mL doxycycline. Virus from TREx-BCBL1-RTA cells was isolated from cultures 48 h after reactivation. Cell debris was removed, and aliquots of supernatant were processed and stored as described above. 2x105 iSLK cells/well were seeded in 6 well plates and the fol- lowing day cells were transduced with lentiviral vectors as described above. The next day a 1:10 dilution of viral supernatant supplemented with 4 μg/mL polybrene was added to the cells, which were then centrifuged as described above. The inoculum was removed after spino- culation and medium was replaced with fresh media, or media containing 1 μg/mL doxycy- cline. Supernatant from the de novo infected cells were harvested and analysed for DNase I- protected genomes as described above. Statistical analysis Prism7 (GraphPad) was used for generating graphs and performing statistical analysis. Unpaired Student’s t-tests were used to determine significance between two groups. One-way or two-way ANOVA was used to compare multiple samples or between grouped samples respectively, and an appropriate post-hoc test was done to determine differences between groups. p-values <0.05 were considered significant and denoted as the following: <0.05 (�), <0.01 (��), <0.001 (��), <0.0001 (��) PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 27 / 34 KSHV modulates the host unfolded protein response to support lytic replication Acknowledgments We thank members of the McCormick lab for critical reading of this manuscript. We would like to thank Don Ganem (UCSF), Shou-Jiang Gao (USC), Jae Jung (USC), and David Lukac (Rutgers) for providing reagents. Author Contributions Conceptualization: Benjamin P. Johnston, Craig McCormick. Funding acquisition: Craig McCormick. Investigation: Benjamin P. Johnston, Eric S. Pringle, Craig McCormick. Methodology: Benjamin P. Johnston, Eric S. Pringle. Project administration: Craig McCormick. Supervision: Craig McCormick. Writing – original draft: Benjamin P. Johnston, Craig McCormick. Writing – review & editing: Benjamin P. Johnston, Craig McCormick. References 1. Ron D, Walter P. Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol. England; 2007; 8: 519–529. https://doi.org/10.1038/nrm2199 PMID: 17565364 2. Walter P, Ron D. The unfolded protein response: from stress pathway to homeostatic regulation. Sci- ence. United States; 2011; 334: 1081–1086. https://doi.org/10.1126/science.1209038 PMID: 22116877 3. Wang M, Kaufman RJ. Protein misfolding in the endoplasmic reticulum as a conduit to human disease. Nature. England; 2016; 529: 326–335. https://doi.org/10.1038/nature17041 PMID: 26791723 4. Hetz C. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nat Rev Mol Cell Biol. England; 2012; 13: 89–102. https://doi.org/10.1038/nrm3270 PMID: 22251901 5. Gardner BM, Pincus D, Gotthardt K, Gallagher CM, Walter P. Endoplasmic reticulum stress sensing in the unfolded protein response. Cold Spring Harb Perspect Biol. United States; 2013; 5: a013169. https://doi.org/10.1101/cshperspect.a013169 PMID: 23388626 6. Zhang K, Kaufman RJ. Signaling the unfolded protein response from the endoplasmic reticulum. J Biol Chem. United States; 2004; 279: 25935–25938. https://doi.org/10.1074/jbc.R400008200 PMID: 15070890 7. Okada T, Haze K, Nadanaka S, Yoshida H, Seidah NG, Hirano Y, et al. A serine protease inhibitor pre- vents endoplasmic reticulum stress-induced cleavage but not transport of the membrane-bound tran- scription factor ATF6. J Biol Chem. United States; 2003; 278: 31024–31032. https://doi.org/10.1074/ jbc.M300923200 PMID: 12782636 8. Ye J, Rawson RB, Komuro R, Chen X, Dave´ UP, Prywes R, et al. ER stress induces cleavage of mem- brane-bound ATF6 by the same proteases that process SREBPs. Mol Cell. Elsevier; 2000; 6: 1355– 64. https://doi.org/10.1016/s1097-2765(00)00133-7 PMID: 11163209 9. Harding HP, Zhang Y, Ron D. Protein translation and folding are coupled by an endoplasmic-reticu- lum-resident kinase. Nature. England; 1999; 397: 271–274. https://doi.org/10.1038/16729 PMID: 9930704 10. Clemens MJ, Pain VM, Wong ST, Henshaw EC. Phosphorylation inhibits guanine nucleotide exchange on eukaryotic initiation factor 2. Nature. England; 1982; 296: 93–95. 11. Krishnamoorthy T, Pavitt GD, Zhang F, Dever TE, Hinnebusch AG. Tight binding of the phosphory- lated alpha subunit of initiation factor 2 (eIF2alpha) to the regulatory subunits of guanine nucleotide exchange factor eIF2B is required for inhibition of translation initiation. Mol Cell Biol. United States; 2001; 21: 5018–5030. https://doi.org/10.1128/MCB.21.15.5018-5030.2001 PMID: 11438658 12. Merrick WC. Mechanism and regulation of eukaryotic protein synthesis. Microbiol Rev. United States; 1992; 56: 291–315. PMID: 1620067 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 28 / 34 KSHV modulates the host unfolded protein response to support lytic replication 13. Spriggs KA, Bushell M, Willis AE. Translational regulation of gene expression during conditions of cell stress. Mol Cell. United States; 2010; 40: 228–237. https://doi.org/10.1016/j.molcel.2010.09.028 PMID: 20965418 14. Vattem KM, Wek RC. Reinitiation involving upstream ORFs regulates ATF4 mRNA translation in mammalian cells. Proc Natl Acad Sci U S A. United States; 2004; 101: 11269–11274. https://doi.org/ 10.1073/pnas.0400541101 PMID: 15277680 15. Rzymski T, Milani M, Pike L, Buffa F, Mellor HR, Winchester L, et al. Regulation of autophagy by ATF4 in response to severe hypoxia. Oncogene. England; 2010; 29: 4424–4435. https://doi.org/10.1038/ onc.2010.191 PMID: 20514020 16. B’chir W, Maurin A-C, Carraro V, Averous J, Jousse C, Muranishi Y, et al. The eIF2alpha/ATF4 path- way is essential for stress-induced autophagy gene expression. Nucleic Acids Res. England; 2013; 41: 7683–7699. https://doi.org/10.1093/nar/gkt563 PMID: 23804767 17. Ma Y, Hendershot LM. Delineation of a negative feedback regulatory loop that controls protein transla- tion during endoplasmic reticulum stress. J Biol Chem. United States; 2003; 278: 34864–34873. https://doi.org/10.1074/jbc.M301107200 PMID: 12840028 18. Novoa I, Zeng H, Harding HP, Ron D. Feedback inhibition of the unfolded protein response by GADD34-mediated dephosphorylation of eIF2alpha. J Cell Biol. Rockefeller University Press; 2001; 153: 1011–22. https://doi.org/10.1083/jcb.153.5.1011 PMID: 11381086 19. Harding HP, Zhang Y, Zeng H, Novoa I, Lu PD, Calfon M, et al. An integrated stress response regu- lates amino acid metabolism and resistance to oxidative stress. Mol Cell. United States; 2003; 11: 619–633. 20. Lu PD, Harding HP, Ron D. Translation reinitiation at alternative open reading frames regulates gene expression in an integrated stress response. J Cell Biol. United States; 2004; 167: 27–33. https://doi. org/10.1083/jcb.200408003 PMID: 15479734 21. Han J, Back SH, Hur J, Lin Y-H, Gildersleeve R, Shan J, et al. ER-stress-induced transcriptional regu- lation increases protein synthesis leading to cell death. Nat Cell Biol. England; 2013; 15: 481–490. https://doi.org/10.1038/ncb2738 PMID: 23624402 22. Ma Y, Brewer JW, Diehl JA, Hendershot LM. Two distinct stress signaling pathways converge upon the CHOP promoter during the mammalian unfolded protein response. J Mol Biol. England; 2002; 318: 1351–1365. https://doi.org/10.1016/s0022-2836(02)00234-6 PMID: 12083523 23. Oyadomari S, Mori M. Roles of CHOP/GADD153 in endoplasmic reticulum stress. Cell Death Differ. England; 2004; 11: 381–389. https://doi.org/10.1038/sj.cdd.4401373 PMID: 14685163 24. Cox JS, Walter P. A novel mechanism for regulating activity of a transcription factor that controls the unfolded protein response. Cell. United States; 1996; 87: 391–404. https://doi.org/10.1016/s0092- 8674(00)81360-4 PMID: 8898193 25. Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell. United States; 2001; 107: 881–891. https://doi.org/10.1016/s0092-8674(01)00611-0 PMID: 11779464 26. 27. 28. Lee K, Tirasophon W, Shen X, Michalak M, Prywes R, Okada T, et al. IRE1-mediated unconventional mRNA splicing and S2P-mediated ATF6 cleavage merge to regulate XBP1 in signaling the unfolded protein response. Genes Dev. United States; 2002; 16: 452–466. https://doi.org/10.1101/gad.964702 PMID: 11850408 Jurkin J, Henkel T, Nielsen AF, Minnich M, Popow J, Kaufmann T, et al. The mammalian tRNA ligase complex mediates splicing of XBP1 mRNA and controls antibody secretion in plasma cells. EMBO J. England; 2014; 33: 2922–2936. https://doi.org/10.15252/embj.201490332 PMID: 25378478 Lu Y, Liang F-X, Wang X. A synthetic biology approach identifies the mammalian UPR RNA ligase RtcB. Mol Cell. United States; 2014; 55: 758–770. https://doi.org/10.1016/j.molcel.2014.06.032 PMID: 25087875 29. Acosta-Alvear D, Zhou Y, Blais A, Tsikitis M, Lents NH, Arias C, et al. XBP1 controls diverse cell type- and condition-specific transcriptional regulatory networks. Mol Cell. United States; 2007; 27: 53–66. https://doi.org/10.1016/j.molcel.2007.06.011 PMID: 17612490 30. Hollien J, Lin JH, Li H, Stevens N, Walter P, Weissman JS. Regulated Ire1-dependent decay of mes- senger RNAs in mammalian cells. J Cell Biol. United States; 2009; 186: 323–331. https://doi.org/10. 1083/jcb.200903014 PMID: 19651891 31. Hollien J, Weissman JS. Decay of endoplasmic reticulum-localized mRNAs during the unfolded pro- tein response. Science. United States; 2006; 313: 104–107. https://doi.org/10.1126/science.1129631 PMID: 16825573 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 29 / 34 KSHV modulates the host unfolded protein response to support lytic replication 32. Reimold AM, Iwakoshi NN, Manis J, Vallabhajosyula P, Szomolanyi-Tsuda E, Gravallese EM, et al. Plasma cell differentiation requires the transcription factor XBP-1. Nature. 2001; 412: 300–307. https://doi.org/10.1038/35085509 PMID: 11460154 33. Iwakoshi NN, Lee A-H, Vallabhajosyula P, Otipoby KL, Rajewsky K, Glimcher LH. Plasma cell differen- tiation and the unfolded protein response intersect at the transcription factor XBP-1. Nat Immunol. Nature Publishing Group; 2003; 4: 321–329. https://doi.org/10.1038/ni907 PMID: 12612580 34. Chang Y, Cesarman E, Pessin MS, Lee F, Culpepper J, Knowles DM, et al. Identification of herpesvi- rus-like DNA sequences in AIDS-associated Kaposi’s sarcoma. Science. United States; 1994; 266: 1865–1869. https://doi.org/10.1126/science.7997879 PMID: 7997879 35. Cesarman E, Chang Y, Moore PS, Said JW, Knowles DM. Kaposi’s sarcoma-associated herpesvirus- like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med. United States; 1995; 332: 1186–1191. https://doi.org/10.1056/NEJM199505043321802 PMID: 7700311 36. Soulier J, Grollet L, Oksenhendler E, Cacoub P, Cazals-Hatem D, Babinet P, et al. Kaposi’s sarcoma- associated herpesvirus-like DNA sequences in multicentric Castleman’s disease. Blood. United States; 1995; 86: 1276–1280. PMID: 7632932 37. Coleman CB, Nealy MS, Tibbetts SA. Immature and transitional B cells are latency reservoirs for a gammaherpesvirus. J Virol. United States; 2010; 84: 13045–13052. https://doi.org/10.1128/JVI. 01455-10 PMID: 20926565 38. Klein U, Gloghini A, Gaidano G, Chadburn A, Cesarman E, Dalla-Favera R, et al. Gene expression profile analysis of AIDS-related primary effusion lymphoma (PEL) suggests a plasmablastic derivation and identifies PEL-specific transcripts. Blood. United States; 2003; 101: 4115–4121. https://doi.org/ 10.1182/blood-2002-10-3090 PMID: 12531789 39. Du MQ, Liu H, Diss TC, Ye H, Hamoudi RA, Dupin N, et al. Kaposi sarcoma-associated herpesvirus infects monotypic (IgM lambda) but polyclonal naive B cells in Castleman disease and associated lym- phoproliferative disorders. Blood. United States; 2001; 97: 2130–2136. 40. Jenner RG, Maillard K, Cattini N, Weiss RA, Boshoff C, Wooster R, et al. Kaposi’s sarcoma-associated herpesvirus-infected primary effusion lymphoma has a plasma cell gene expression profile. Proc Natl Acad Sci U S A. United States; 2003; 100: 10399–10404. https://doi.org/10.1073/pnas.1630810100 PMID: 12925741 41. Yu F, Feng J, Harada JN, Chanda SK, Kenney SC, Sun R. B cell terminal differentiation factor XBP-1 induces reactivation of Kaposi’s sarcoma-associated herpesvirus. FEBS Lett. John Wiley & Sons, Ltd; 2007; 581: 3485–3488. https://doi.org/10.1016/j.febslet.2007.06.056 PMID: 17617410 42. Wilson SJ, Tsao EH, Webb BLJ, Ye H, Dalton-Griffin L, Tsantoulas C, et al. X Box Binding Protein XBP-1s Transactivates the Kaposi’s Sarcoma-Associated Herpesvirus (KSHV) ORF50 Promoter, Linking Plasma Cell Differentiation to KSHV Reactivation from Latency. J Virol. 2007; 81: 13578– 13586. https://doi.org/10.1128/JVI.01663-07 PMID: 17928342 43. Dalton-Griffin L, Wilson SJ, Kellam P. X-box binding protein 1 contributes to induction of the Kaposi’s sarcoma-associated herpesvirus lytic cycle under hypoxic conditions. J Virol. United States; 2009; 83: 7202–7209. https://doi.org/10.1128/JVI.00076-09 PMID: 19403667 44. Sun R, Lin SF, Gradoville L, Yuan Y, Zhu F, Miller G. A viral gene that activates lytic cycle expression of Kaposi’s sarcoma-associated herpesvirus. Proc Natl Acad Sci U S A. United States; 1998; 95: 10866–10871. https://doi.org/10.1073/pnas.95.18.10866 PMID: 9724796 45. 46. Lukac DM, Renne R, Kirshner JR, Ganem D. Reactivation of Kaposi’s sarcoma-associated herpesvi- rus infection from latency by expression of the ORF 50 transactivator, a homolog of the EBV R protein. Virology. United States; 1998; 252: 304–312. https://doi.org/10.1006/viro.1998.9486 PMID: 9878608 Liang Y, Chang J, Lynch SJ, Lukac DM, Ganem D. The lytic switch protein of KSHV activates gene expression via functional interaction with RBP-Jkappa (CSL), the target of the Notch signaling path- way. Genes Dev. United States; 2002; 16: 1977–1989. https://doi.org/10.1101/gad.996502 PMID: 12154127 47. Hu D, Wang V, Yang M, Abdullah S, Davis DA, Uldrick TS, et al. Induction of Kaposi’s Sarcoma-Asso- ciated Herpesvirus-Encoded Viral Interleukin-6 by X-Box Binding Protein 1. J Virol. United States; 2016; 90: 368–378. https://doi.org/10.1128/JVI.01192-15 PMID: 26491160 48. Aoki Y, Tosato G, Fonville TW, Pittaluga S. Serum viral interleukin-6 in AIDS-related multicentric Cas- tleman disease. Blood. United States; 2001. pp. 2526–2527. 49. Aoki Y, Yarchoan R, Braun J, Iwamoto A, Tosato G. Viral and cellular cytokines in AIDS-related malig- nant lymphomatous effusions. Blood. United States; 2000; 96: 1599–1601. PMID: 10942415 50. Polizzotto MN, Uldrick TS, Wang V, Aleman K, Wyvill KM, Marshall V, et al. Human and viral interleu- kin-6 and other cytokines in Kaposi sarcoma herpesvirus-associated multicentric Castleman disease. Blood. United States; 2013; 122: 4189–4198. https://doi.org/10.1182/blood-2013-08-519959 PMID: 24174627 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 30 / 34 KSHV modulates the host unfolded protein response to support lytic replication 51. Shaffer AL, Shapiro-Shelef M, Iwakoshi NN, Lee A-H, Qian S-B, Zhao H, et al. XBP1, downstream of Blimp-1, expands the secretory apparatus and other organelles, and increases protein synthesis in plasma cell differentiation. Immunity. United States; 2004; 21: 81–93. https://doi.org/10.1016/j. immuni.2004.06.010 PMID: 15345222 52. Iwakoshi NN, Lee A-H, Vallabhajosyula P, Otipoby KL, Rajewsky K, Glimcher LH. Plasma cell differen- tiation and the unfolded protein response intersect at the transcription factor XBP-1. Nat Immunol. United States; 2003; 4: 321–329. https://doi.org/10.1038/ni907 PMID: 12612580 53. Cheng G, Feng Z, He B. Herpes simplex virus 1 infection activates the endoplasmic reticulum resident kinase PERK and mediates eIF-2alpha dephosphorylation by the gamma(1)34.5 protein. J Virol. United States; 2005; 79: 1379–1388. https://doi.org/10.1128/JVI.79.3.1379-1388.2005 PMID: 15650164 54. Isler JA, Skalet AH, Alwine JC. Human cytomegalovirus infection activates and regulates the unfolded protein response. J Virol. United States; 2005; 79: 6890–6899. https://doi.org/10.1128/JVI.79.11. 6890-6899.2005 PMID: 15890928 55. Siddiquey MNA, Zhang H, Nguyen CC, Domma AJ, Kamil JP. The Human Cytomegalovirus Endoplas- mic Reticulum-Resident Glycoprotein UL148 Activates the Unfolded Protein Response. J Virol. United States; 2018;92. https://doi.org/10.1128/JVI.00896-18 PMID: 30045994 56. Mulvey M, Arias C, Mohr I. Maintenance of endoplasmic reticulum (ER) homeostasis in herpes sim- plex virus type 1-infected cells through the association of a viral glycoprotein with PERK, a cellular ER stress sensor. J Virol. United States; 2007; 81: 3377–3390. https://doi.org/10.1128/JVI.02191-06 PMID: 17229688 57. Nakamura H, Lu M, Gwack Y, Souvlis J, Zeichner SL, Jung JU. Global changes in Kaposi’s sarcoma- associated virus gene expression patterns following expression of a tetracycline-inducible Rta transac- tivator. J Virol. United States; 2003; 77: 4205–4220. https://doi.org/10.1128/JVI.77.7.4205-4220.2003 PMID: 12634378 58. Lytton J, Westlin M, Hanley MR. Thapsigargin inhibits the sarcoplasmic or endoplasmic reticulum Ca- ATPase family of calcium pumps. J Biol Chem. United States; 1991; 266: 17067–17071. PMID: 1832668 59. Gallagher CM, Garri C, Cain EL, Ang KK-H, Wilson CG, Chen S, et al. Ceapins are a new class of unfolded protein response inhibitors, selectively targeting the ATF6alpha branch. Elife. England; 2016;5. https://doi.org/10.7554/eLife.11878 PMID: 27435960 60. Wang Y, Shen J, Arenzana N, Tirasophon W, Kaufman RJ, Prywes R. Activation of ATF6 and an ATF6 DNA binding site by the endoplasmic reticulum stress response. J Biol Chem. United States; 2000; 275: 27013–27020. https://doi.org/10.1074/jbc.M003322200 PMID: 10856300 61. Shoulders MD, Ryno LM, Genereux JC, Moresco JJ, Tu PG, Wu C, et al. Stress-independent activa- tion of XBP1s and/or ATF6 reveals three functionally diverse ER proteostasis environments. Cell Rep. United States; 2013; 3: 1279–1292. https://doi.org/10.1016/j.celrep.2013.03.024 PMID: 23583182 62. Li M, Baumeister P, Roy B, Phan T, Foti D, Luo S, et al. ATF6 as a transcription activator of the endo- plasmic reticulum stress element: thapsigargin stress-induced changes and synergistic interactions with NF-Y and YY1. Mol Cell Biol. United States; 2000; 20: 5096–5106. https://doi.org/10.1128/mcb. 20.14.5096-5106.2000 PMID: 10866666 63. Wu J, Rutkowski DT, Dubois M, Swathirajan J, Saunders T, Wang J, et al. ATF6alpha optimizes long- term endoplasmic reticulum function to protect cells from chronic stress. Dev Cell. United States; 2007; 13: 351–364. https://doi.org/10.1016/j.devcel.2007.07.005 PMID: 17765679 64. Meurs E, Chong K, Galabru J, Thomas NS, Kerr IM, Williams BR, et al. Molecular cloning and charac- terization of the human double-stranded RNA-activated protein kinase induced by interferon. Cell. United States; 1990; 62: 379–390. https://doi.org/10.1016/0092-8674(90)90374-n PMID: 1695551 65. Dever TE, Feng L, Wek RC, Cigan AM, Donahue TF, Hinnebusch AG. Phosphorylation of initiation factor 2 alpha by protein kinase GCN2 mediates gene-specific translational control of GCN4 in yeast. Cell. United States; 1992; 68: 585–596. https://doi.org/10.1016/0092-8674(92)90193-g PMID: 1739968 66. Hurst R, Schatz JR, Matts RL. Inhibition of rabbit reticulocyte lysate protein synthesis by heavy metal ions involves the phosphorylation of the alpha-subunit of the eukaryotic initiation factor 2. J Biol Chem. United States; 1987; 262: 15939–15945. PMID: 3680235 67. Axten JM, Medina JR, Feng Y, Shu A, Romeril SP, Grant SW, et al. Discovery of 7-methyl-5-(1-{[3-(tri- fluoromethyl)phenyl]acetyl}-2,3-dihydro-1H-indol-5-yl)-7H-pyrrolo[2,3-d]pyrimidin-4-amine (GSK2606414), a potent and selective first-in-class inhibitor of protein kinase R (PKR)-like endoplas- mic reticulum kinase (PERK). J Med Chem. United States; 2012; 55: 7193–7207. https://doi.org/10. 1021/jm300713s PMID: 22827572 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 31 / 34 KSHV modulates the host unfolded protein response to support lytic replication 68. Harding HP, Zyryanova AF, Ron D. Uncoupling proteostasis and development in vitro with a small mol- ecule inhibitor of the pancreatic endoplasmic reticulum kinase, PERK. J Biol Chem. United States; 2012; 287: 44338–44344. https://doi.org/10.1074/jbc.M112.428987 PMID: 23148209 69. Merlie JP, Sebbane R, Tzartos S, Lindstrom J. Inhibition of glycosylation with tunicamycin blocks assembly of newly synthesized acetylcholine receptor subunits in muscle cells. J Biol Chem. United States; 1982; 257: 2694–2701. PMID: 7061443 70. Kaufman RJ. Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene tran- scriptional and translational controls. Genes Dev. United States; 1999; 13: 1211–1233. https://doi.org/ 10.1101/gad.13.10.1211 PMID: 10346810 71. 72. Li B, Yi P, Zhang B, Xu C, Liu Q, Pi Z, et al. Differences in endoplasmic reticulum stress signalling kinetics determine cell survival outcome through activation of MKP-1. Cell Signal. England; 2011; 23: 35–45. https://doi.org/10.1016/j.cellsig.2010.07.019 PMID: 20727407 Lee A-H, Iwakoshi NN, Glimcher LH. XBP-1 regulates a subset of endoplasmic reticulum resident chaperone genes in the unfolded protein response. Mol Cell Biol. United States; 2003; 23: 7448–7459. https://doi.org/10.1128/MCB.23.21.7448-7459.2003 PMID: 14559994 73. Glaunsinger B, Ganem D. Lytic KSHV infection inhibits host gene expression by accelerating global mRNA turnover. Mol Cell. United States; 2004; 13: 713–723. https://doi.org/10.1016/s1097-2765(04) 00091-7 PMID: 15023341 74. Chandriani S, Ganem D. Host Transcript Accumulation during Lytic KSHV Infection Reveals Several Classes of Host Responses. PLoS One. 2007; 2: e811. https://doi.org/10.1371/journal.pone.0000811 PMID: 17726541 75. Lee YJ, Glaunsinger BA. Aberrant Herpesvirus-Induced Polyadenylation Correlates With Cellular Messenger RNA Destruction. PLoS Biol. 2009; 7: e1000107. https://doi.org/10.1371/journal.pbio. 1000107 PMID: 19468299 76. Abernathy E, Gilbertson S, Alla R, Glaunsinger B. Viral Nucleases Induce an mRNA Degradation- Transcription Feedback Loop in Mammalian Cells. Cell Host Microbe. Elsevier; 2015; 18: 243–253. https://doi.org/10.1016/j.chom.2015.06.019 PMID: 26211836 77. Clyde K, Glaunsinger BA. Deep Sequencing Reveals Direct Targets of Gammaherpesvirus-Induced mRNA Decay and Suggests That Multiple Mechanisms Govern Cellular Transcript Escape. PLoS One. 2011; 6: e19655. https://doi.org/10.1371/journal.pone.0019655 PMID: 21573023 78. Glaunsinger B, Chavez L, Ganem D. The exonuclease and host shutoff functions of the SOX protein of Kaposi’s sarcoma-associated herpesvirus are genetically separable. J Virol. 2005; 79: 7396–401. https://doi.org/10.1128/JVI.79.12.7396-7401.2005 PMID: 15919895 79. Sidrauski C, Tsai JC, Kampmann M, Hearn BR, Vedantham P, Jaishankar P, et al. Pharmacological dimerization and activation of the exchange factor eIF2B antagonizes the integrated stress response. Elife. England; 2015; 4: e07314. https://doi.org/10.7554/eLife.07314 PMID: 25875391 80. Sekine Y, Zyryanova A, Crespillo-Casado A, Fischer PM, Harding HP, Ron D. Stress responses. Mutations in a translation initiation factor identify the target of a memory-enhancing compound. Sci- ence. United States; 2015; 348: 1027–1030. https://doi.org/10.1126/science.aaa6986 PMID: 25858979 81. Cross BCS, Bond PJ, Sadowski PG, Jha BK, Zak J, Goodman JM, et al. The molecular basis for selec- tive inhibition of unconventional mRNA splicing by an IRE1-binding small molecule. Proc Natl Acad Sci U S A. United States; 2012; 109: E869–78. https://doi.org/10.1073/pnas.1115623109 PMID: 22315414 82. Myoung J, Ganem D. Generation of a doxycycline-inducible KSHV producer cell line of endothelial ori- gin: maintenance of tight latency with efficient reactivation upon induction. J Virol Methods. Nether- lands; 2011; 174: 12–21. https://doi.org/10.1016/j.jviromet.2011.03.012 PMID: 21419799 83. Nishimura K, Ueda K, Sakakibara S, Do E, Ohsaki E, Okuno T, et al. A viral transcriptional activator of Kaposi’s sarcoma-associated herpesvirus (KSHV) induces apoptosis, which is blocked in KSHV- infected cells. Virology. United States; 2003; 316: 64–74. https://doi.org/10.1016/s0042-6822(03) 00582-8 PMID: 14599791 84. Vieira J, O’Hearn PM. Use of the red fluorescent protein as a marker of Kaposi’s sarcoma-associated herpesvirus lytic gene expression. Virology. United States; 2004; 325: 225–240. https://doi.org/10. 1016/j.virol.2004.03.049 PMID: 15246263 85. Zhang J, He S, Wang Y, Brulois K, Lan K, Jung JU, et al. Herpesviral G protein-coupled receptors acti- vate NFAT to induce tumor formation via inhibiting the SERCA calcium ATPase. PLoS Pathog. United States; 2015; 11: e1004768. https://doi.org/10.1371/journal.ppat.1004768 PMID: 25811856 86. Kaser A, Lee A-H, Franke A, Glickman JN, Zeissig S, Tilg H, et al. XBP1 links ER stress to intestinal inflammation and confers genetic risk for human inflammatory bowel disease. Cell. United States; 2008; 134: 743–756. https://doi.org/10.1016/j.cell.2008.07.021 PMID: 18775308 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 32 / 34 KSHV modulates the host unfolded protein response to support lytic replication 87. Nakatani Y, Kaneto H, Kawamori D, Yoshiuchi K, Hatazaki M, Matsuoka T, et al. Involvement of endo- plasmic reticulum stress in insulin resistance and diabetes. J Biol Chem. United States; 2005; 280: 847–851. https://doi.org/10.1074/jbc.M411860200 PMID: 15509553 88. Wang S, Kaufman RJ. The impact of the unfolded protein response on human disease. J Cell Biol. United States; 2012; 197: 857–867. https://doi.org/10.1083/jcb.201110131 PMID: 22733998 89. Smith JA, Turner MJ, DeLay ML, Klenk EI, Sowders DP, Colbert RA. Endoplasmic reticulum stress and the unfolded protein response are linked to synergistic IFN-beta induction via X-box binding pro- tein 1. Eur J Immunol. Germany; 2008; 38: 1194–1203. https://doi.org/10.1002/eji.200737882 PMID: 18412159 90. 91. Talloczy Z, Virgin HW 4th, Levine B. PKR-dependent autophagic degradation of herpes simplex virus type 1. Autophagy. United States; 2006; 2: 24–29. https://doi.org/10.4161/auto.2176 PMID: 16874088 Taylor GS, Mautner J, Munz C. Autophagy in herpesvirus immune control and immune escape. Her- pesviridae. England; 2011; 2: 2. https://doi.org/10.1186/2042-4280-2-2 PMID: 21429245 92. Yamaguchi H, Wang H-G. CHOP is involved in endoplasmic reticulum stress-induced apoptosis by enhancing DR5 expression in human carcinoma cells. J Biol Chem. United States; 2004; 279: 45495– 45502. https://doi.org/10.1074/jbc.M406933200 PMID: 15322075 93. Houck SA, Ren HY, Madden VJ, Bonner JN, Conlin MP, Janovick JA, et al. Quality control autophagy degrades soluble ERAD-resistant conformers of the misfolded membrane protein GnRHR. Mol Cell. United States; 2014; 54: 166–179. https://doi.org/10.1016/j.molcel.2014.02.025 PMID: 24685158 94. Kroeger H, Miranda E, MacLeod I, Perez J, Crowther DC, Marciniak SJ, et al. Endoplasmic reticulum- associated degradation (ERAD) and autophagy cooperate to degrade polymerogenic mutant serpins. J Biol Chem. United States; 2009; 284: 22793–22802. https://doi.org/10.1074/jbc.M109.027102 PMID: 19549782 95. 96. Lee J-S, Li Q, Lee J-Y, Lee S-H, Jeong JH, Lee H-R, et al. FLIP-mediated autophagy regulation in cell death control. Nat Cell Biol. England; 2009; 11: 1355–1362. https://doi.org/10.1038/ncb1980 PMID: 19838173 Liang Q, Chang B, Brulois KF, Castro K, Min C-K, Rodgers MA, et al. Kaposi’s sarcoma-associated herpesvirus K7 modulates Rubicon-mediated inhibition of autophagosome maturation. J Virol. United States; 2013; 87: 12499–12503. https://doi.org/10.1128/JVI.01898-13 PMID: 24027317 97. Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N, et al. Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell. United States; 2005; 122: 927–939. https://doi.org/10.1016/j.cell. 2005.07.002 PMID: 16179260 98. Tomlinson CC, Damania B. The K1 protein of Kaposi’s sarcoma-associated herpesvirus activates the Akt signaling pathway. J Virol. United States; 2004; 78: 1918–1927. https://doi.org/10.1128/JVI.78.4. 1918-1927.2004 PMID: 14747556 99. Montaner S, Sodhi A, Pece S, Mesri EA, Gutkind JS. The Kaposi’s sarcoma-associated herpesvirus G protein-coupled receptor promotes endothelial cell survival through the activation of Akt/protein kinase B. Cancer Res. United States; 2001; 61: 2641–2648. PMID: 11289142 100. Zhang P, Su C, Jiang Z, Zheng C. Herpes Simplex Virus 1 UL41 Protein Suppresses the IRE1/XBP1 Signal Pathway of the Unfolded Protein Response via Its RNase Activity. J Virol. United States; 2017; 91. https://doi.org/10.1128/JVI.02056-16 PMID: 27928013 101. Papandreou I, Denko NC, Olson M, Van Melckebeke H, Lust S, Tam A, et al. Identification of an Ire1al- pha endonuclease specific inhibitor with cytotoxic activity against human multiple myeloma. Blood. United States; 2011; 117: 1311–1314. https://doi.org/10.1182/blood-2010-08-303099 PMID: 21081713 102. Su A, Wang H, Li Y, Wang X, Chen D, Wu Z. Opposite Roles of RNase and Kinase Activities of Inosi- tol-Requiring Enzyme 1 (IRE1) on HSV-1 Replication. Viruses. Switzerland; 2017; 9. https://doi.org/ 10.3390/v9090235 PMID: 28832521 103. Xuan B, Qian Z, Torigoi E, Yu D. Human cytomegalovirus protein pUL38 induces ATF4 expression, inhibits persistent JNK phosphorylation, and suppresses endoplasmic reticulum stress-induced cell death. J Virol. United States; 2009; 83: 3463–3474. https://doi.org/10.1128/JVI.02307-08 PMID: 19193809 104. Stahl S, Burkhart JM, Hinte F, Tirosh B, Mohr H, Zahedi RP, et al. Cytomegalovirus downregulates IRE1 to repress the unfolded protein response. PLoS Pathog. United States; 2013; 9: e1003544. https://doi.org/10.1371/journal.ppat.1003544 PMID: 23950715 105. Jheng J-R, Ho J-Y, Horng J-T. ER stress, autophagy, and RNA viruses. Front Microbiol. Frontiers Media SA; 2014; 5: 388. https://doi.org/10.3389/fmicb.2014.00388 PMID: 25140166 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 33 / 34 KSHV modulates the host unfolded protein response to support lytic replication 106. 107. Tam AB, Koong AC, Niwa M. Ire1 has distinct catalytic mechanisms for XBP1/HAC1 splicing and RIDD. Cell Rep. Elsevier; 2014; 9: 850–8. https://doi.org/10.1016/j.celrep.2014.09.016 PMID: 25437541 Livak KJ, Schmittgen TD. Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2−ΔΔCT Method. Methods. Academic Press; 2001; 25: 402–408. https://doi.org/10. 1006/meth.2001.1262 PMID: 11846609 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1008185 December 2, 2019 34 / 34
10.3390_genes14061279	Article Sequencing and Characterization of M. morganii Strain UM869: A Comprehensive Comparative Genomic Analysis of Virulence, Antibiotic Resistance, and Functional Pathways Dibyajyoti Uttameswar Behera 1, Sangita Dixit 1 Rajesh Kumar Sahoo 1 Sutar Suhas Bharat 4 and Enketeswara Subudhi 1,* , Maheswata Sahoo 1 , Mahendra Gaur 2,3 , Bijay Kumar Behera 5 , Rukmini Mishra 4, , Bharat Bhusan Subudhi 2, 1 Centre for Biotechnology, School of Pharmaceutical Sciences, Siksha ‘O’ Anusandhan (Deemed to be University), Bhubaneswar 751003, Odisha, India; [email protected] (D.U.B.); [email protected] (S.D.) 2 Drug Development and Analysis Laboratory, School of Pharmaceutical Sciences, Siksha ‘O’ Anusandhan (Deemed to be University), Bhubaneswar 751003, Odisha, India; [email protected] (M.G.) 3 Department of Biotechnology & Food Technology, Punjabi University, Patiala 147002, Punjab, India 4 Department of Botany, School of Applied Sciences, Centurion University of Technology and Management, Bhubaneswar 761211, Odisha, India 5 College of Fisheries, Rani Lakshmi Bai Central Agricultural University, Gwalior Road, Jhansi 284003, Uttar Pradesh, India; [email protected] * Correspondence: [email protected]; Tel.: +91-9861075829 Abstract: Morganella morganii is a Gram-negative opportunistic Enterobacteriaceae pathogen inherently resistant to colistin. This species causes various clinical and community-acquired infections. This study investigated the virulence factors, resistance mechanisms, functional pathways, and com- parative genomic analysis of M. morganii strain UM869 with 79 publicly available genomes. The multidrug resistance strain UM869 harbored 65 genes associated with 30 virulence factors, including efﬂux pump, hemolysin, urease, adherence, toxin, and endotoxin. Additionally, this strain contained 11 genes related to target alteration, antibiotic inactivation, and efﬂux resistance mechanisms. Further, the comparative genomic study revealed a high genetic relatedness (98.37%) among the genomes, possibly due to the dissemination of genes between adjoining countries. The core proteome of 79 genomes contains the 2692 core, including 2447 single-copy orthologues. Among them, six were associated with resistance to major antibiotic classes manifested through antibiotic target alteration (PBP3, gyrB) and antibiotic efﬂux (kpnH, rsmA, qacG; rsmA; CRP). Similarly, 47 core orthologues were annotated to 27 virulence factors. Moreover, mostly core orthologues were mapped to transporters (n = 576), two-component systems (n = 148), transcription factors (n = 117), ribosomes (n = 114), and quorum sensing (n = 77). The presence of diversity in serotypes (type 2, 3, 6, 8, and 11) and variation in gene content adds to the pathogenicity, making them more difﬁcult to treat. This study highlights the genetic similarity among the genomes of M. morganii and their restricted emergence, mostly in Asian countries, in addition to their growing pathogenicity and resistance. However, steps must be taken to undertake large-scale molecular surveillance and to direct suitable therapeutic interventions. Keywords: M. morganii; pathogens; comparative genomics; AMR; virulence; serotype 1. Introduction Morganella morganii is a Gram-negative facultative anaerobic, rod-shaped enteric bac- terium of the Enterobacteriaceae family. It was initially categorized under the Proteus genus but subsequently reclassiﬁed as a distinct genus through DNA–DNA hybridization analy- sis [1,2]. This genus is distinguished by its capability to perform trehalose fermentation, generate lysine ornithine decarboxylase and is recognized as the type genus of the newly Citation: Behera, D.U.; Dixit, S.; Gaur, M.; Mishra, R.; Sahoo, R.K.; Sahoo, M.; Behera, B.K.; Subudhi, B.B.; Bharat, S.S.; Subudhi, E. Sequencing and Characterization of M. morganii Strain UM869: A Comprehensive Comparative Genomic Analysis of Virulence, Antibiotic Resistance, and Functional Pathways. Genes 2023, 14, 1279. https://doi.org/10.3390/ genes14061279 Academic Editor: Silvia Turroni Received: 23 March 2023 Revised: 10 June 2023 Accepted: 14 June 2023 Published: 16 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Genes 2023, 14, 1279. https://doi.org/10.3390/genes14061279 https://www.mdpi.com/journal/genes genesG C A TT A C GG C A TGenes 2023, 14, 1279 2 of 16 classiﬁed family Morganellaceae [3]. This family comprises eight genera: Arsenophonus, Cosenzaea, Moellerella, Morganella, Photorhabdus, Proteus, Providencia, and Xenorhabdus [1]. These bacterial species are detected in various ecological niches, including the environment, animals, and human microbiota. This organism is a crucial opportunistic pathogen because it can cause many clinical and community-acquired infections [4]. M. morganii has been implicated in various clinical infections, such as urinary tract infections (UTIs) due to long-term urinary catheters, septicemia, and wound infections [5,6], which have more fatal consequences compared to those caused by Escherichia coli [7]. Invasive infections caused by M. morganii are commonly associated with a considerable mortality rate due to a lack of suitable empirical antibiotic interventions [8]. It has also been associated with various pathological conditions, including brain abscess, liver abscess, chorioamnionitis, peritonitis, pericarditis, septic arthritis, rhabdomyolysis, necrotizing fasciitis following snakebites, bilateral keratitis, neonatal sulfhemoglobinemia, and non-clostridial gas gangrene [8]. M. morganii is naturally resistant to ampicillin, amoxicillin, and most of the ﬁrst- and second-generation cephalosporins due to the presence of the ampC resistance gene [4,9,10] as well the last-resort drug, colistin [4]. The use of broad-spectrum antibiotics resulted in the emergence of multidrug-resistant (MDR) or even extensively drug-resistant (XDR) M. morganii, leading to the failure of therapy in clinical settings [11]. Various plasmids and transposons or integrons, such as the IncP6 plasmid carrying blaKPC-2, the IncN plasmid car- rying blaOXA-181, the IncC plasmid carrying blaNDM-1, the IncX3 plasmid carrying blaNDM-5, the Tn7 transposon carrying blaIMP-27, the Tn6741 transposon carrying blaCTX-M-3, the Tn7 transposon carrying cfr, and the In1390 integron carrying blaGES-5, are also reported to be acquired [12–14]. Resistance acquired in M. morganii through these integrative and con- jugative elements (ICEs) and mobilizable genomic islands (MGIs) poses a clinical treatment challenge [15]. Whole genome sequencing of bacteria is the most suitable option for understanding the genetic processes behind this organism’s antibiotic resistance and pathogenicity and monitoring the spread of infections. Oxford nanopore sequencing (ONS) is a nucleic acid sequencing technology that uses protein nanopores to read genome sequences [16,17]. This technology can produce long DNA reads, making it useful for genome assembly, identifying structural variations, and detecting repeats. It has high accuracy in detecting small genetic variations and is essential for identifying antibiotic resistance and virulence in bacteria [17]. ONS also allows real-time sequencing, enabling the rapid identiﬁcation of pathogens, monitoring bacterial populations, and detecting outbreaks at a reasonable cost within the laboratory setup [18]. This study aimed to analyze the genome of M. morganii strain UM869, isolated from a urine sample of a patient with a urinary tract infection in Bhubaneswar, India. The research investigated the genome’s virulence factors, drug-resistance mechanisms, prediction of genomic islands, and COGs functional pathways through a comparative genomic analysis using published M. morganii genomes. 2. Materials and Methods 2.1. Identiﬁcation of Bacteria The strain UM869 was isolated from a 75-year-old female patient diagnosed with community-acquired urinary tract infections (UTI), who was hospitalized in a super- specialty hospital at Bhubaneswar City, India, in October 2021. The antibiotic susceptibility and species identiﬁcation of the strain UM869 was performed by an automated VITEK 2 system (bioMerieux, Inc., Hazelwood, Portland, OR, USA). The result was interpreted following CLSI guidelines (CLSI, 2018) [19]. The E. coli ATCC 25922 was taken as a reference strain for antimicrobial susceptibility testing analysis. Further, the species of the strain was conﬁrmed by ampliﬁcation and sequencing of the 16S rRNA. The strain UM869 was cultured overnight in Luria–Bertani broth medium (LB medium) at 37 ◦C in a shaking incubator (Remi orbital shaking incubator). The genomic DNA was then extracted from 2 mL of overnight incubated bacterial culture, as described by Genes 2023, 14, 1279 3 of 16 Sahoo et al. [20]. The quality and quantity of extracted DNA were evaluated using elec- trophoresis on a 1% agarose gel. The extracted high-quality DNA was subjected to 16S rRNA gene PCR ampliﬁcation using universal primers [21]. The quality of PCR products was examined using electrophoresis in 1% agarose gel. The PCR-ampliﬁed product was outsourced for sequencing at AgriGenome Labs Pvt. Ltd., Cochin, India. The quality- trimmed 16S rRNA sequence was submitted to NCBI’s GeneBank under the accession number ON533444. 2.2. Whole Genome Sequencing, De Novo Assembly, and Functional Annotation The whole genome sequencing of the UM869 strain was accomplished using the Oxford nanopore technology platform at Centurion University of Technology and Manage- ment, Bhubaneswar, India. Oxford nanopore technology (ONT) is a sequencing technology that produces long-read sequences (tens of thousands of bases) compared to traditional sequencing methods. These long reads beneﬁt genome assembly, structural variant identiﬁ- cation, and repeat detection. ONS is also portable and can be used in remote locations. The ﬂow-cell FLO-MIN106 vR9 containing the prepared genomic DNA library was inserted into the MinION set, followed by sequencing using MinKNOW v1.7.14 [22]. Base-calling was performed using the ONT base-caller Guppy tool [23] and the fastq ﬁles were gen- erated from the fast5 ﬁle using Poretools [24]. Porechop v0.2.1 [25] was used for adaptor trimming and NanoFilt v2.2.0 [26] was used to remove the reads having quality scores ≤ 20. The QC-passed high-quality long reads were assembled using Flye v2.9 [27] with default parameters, and the assembly ﬁles were assessed for quality using QUAST v5.0.2 [28]. The assembled sequences were deposited in the NCBI’s GenBank under the accession CP104700.1. The genome of the UM869 was annotated using the NCBI’s prokaryotic genomes annotation pipeline (PGAP) [29]. The genomic assembly of strain UM869 was explored to identify virulence factors and resistance genes determinants through the Vir- ulence Factors Database (VFDB) and the Comprehensive Antibiotic Resistance Database (CARD), respectively. The genome was further screened for mobile genetic elements (in- sertion sequence, transposon elements, plasmid signature sequence, and phage elements) using IsFinder [30], TnCentral [31], PlasmidFinder [32], and Phaster [33]. 2.3. Comparative, Phylogenetic, and Core Orthologues Analysis The genome assemblies of 81 publicly available M. morganii (subspecies morganii) were retrieved from the NCBI genome database on November 30, 2022. The complete- ness, and contamination of all the assemblies were evaluated by CheckM v1.2.2 [34], and BUSCO v5.4.4 [35]. The assembly with completeness ≥90%, and zero contamination (79 genomes) was taken for further downstream analysis. To evaluate the genetic relatedness among the genomes, average nucleotide identity (ANI) was calculated using the ‘ANI’ module of PGCGAP v1.0.28 [36]. The generated ANI distance matrix was plotted into a heat map using the gplots [37] package in R Studio v4.1.3. The maximum likelihood phylogenetic tree of single-copy core protein was reconstructed using the “CoreTree” module of PGCGAP v1.0.21 [38], and inferences were performed by plotting the tree using iTol [39]. Brieﬂy, the sequence of single-copy core orthologues was extracted using perl scripts [36], and aligned using MAFFT [40], followed by a concatenation of each protein’s alignment. Further, the concatenated alignment of each protein was converted into the corresponding codon alignment using PAL2NAL v14 [41], followed by the calling of core SNPs using SNP sites [42]. Then, a phylogenetic tree was construed based on the best model of evolution using IQ-TREE [43]. OrthoFinder v2.5.4 [44] was used to identify core orthologue proteins among the genomes of M. morganii species. The consensus sequence of each core orthologue was generated by multiple sequence alignment using the CIAlign tool [45]. Genes 2023, 14, 1279 4 of 16 2.4. Functional Annotation of Core Orthologues To annotate the core orthologues, the consensus sequences of all the core orthologues underwent BLASTing against KOfam, which is an HMM database of KEGG orthologues, using kofamKOALA [46] with an e-value threshold of ≥1 × 10−5. Subsequently, the eggNOG-mapper tool [47] with the EggNOG database [48] was employed to classify all the core orthologues sequences into clusters of orthologous groups of proteins (COGs). Antimicrobial resistance genes were identiﬁed by BLASTing against the Comprehensive Antibiotic Resistance Database (CARD) using RGI v5.1.1 [49,50]. Similarly, the virulence factors were identiﬁed by BLASTing core proteins against the Virulence Factors Database (VFDB) [51]. 2.5. Comparative O-Antigen Gene Cluster (O-AGC) Analysis The assembled sequences of UM869, and 78 genomes of M. morganii were BLAST against serotypes (type 1 to type 11) of M. morganii available in the NCBI database [52]. The island map of identiﬁed most similar O-AGC was created by gggenes v0.4.1 R package. Ad- ditionally, all the serotype genes were clustered at 97% similarity using CD-HIT [53]. Then, the genes were BLAST against the selected serotype sequence. Genes with 90% query cover- age and 100% identity were selected to generate the heatmap through OriginPro v2021. A detailed workﬂow presentation depicting all the steps in the above methodology is shown in Figure S1. 3. Results 3.1. Bacterial Identiﬁcation, and Antibiogram Study The UM869 strain was isolated from a 75-year-old female patient with urinary tract in- fections in a super-specialty hospital in Bhubaneswar, Odisha, India. From the VITEK 2 anal- ysis, UM869 was identiﬁed as M. morganii. The strain showed resistance to major antibiotics such as amoxicillin/clavulanic acid, piperacillin/tazobactam, cefoperazone/sulbactam, cefuroxime, cefepime, imipenem, ertapenem, amikacin, gentamicin, levoﬂoxacin, minocy- cline, fosfomycin, trimethoprim/sulfamethoxazole„ and colistin. The resistance phenotype was multidrug resistance (MDR), as interpreted using CLSI guidelines [19]. From 16S rRNA gene sequencing, the strain UM869 showed 99.77% identity at 99% query coverage with M. morganii NBRC 3848 (accession no. AB680150) through BLASTn analysis [54]. Further, whole genome sequencing using Oxford nanopore technology conﬁrmed the strain as M. morganii. 3.2. Genome Sequencing of UM869 The de novo assembly of high-quality reads obtained from the nanopore sequencing technology resulted in one contig of 3,761,991 bp size, and GC content of 51%. UM869 had a genome fraction of 49.752%, and a genome completeness, and contamination level of 97.01% and 0.27%, respectively. The genome UM869 (NCBI assembly accession. GCA_025398975) comprised 2870 protein-coding sequences (CDS), 718 pseudogenes, 79 tRNAs, 22 rRNAs, and 1 tmRNA. The M. morganii strain UM869 was assembled into a single circular genome. 3.3. Resistance Genes, Virulence Factors, and Mobile Genetic Elements of UM869 UM869’s genome comprises 65 genes associated with 30 virulence factors, including efﬂux pump, hemolysin, urease, serum resistance, iron uptake, adherence factors, toxin, and endotoxin (Table S1A). The genome also contains 11 resistance genes, including Escherichia coli EF-Tu mutants, conferring resistance to puromycin; DHA-17, Hemophilus inﬂuenzae PBP3, conferring resistance to β-lactam antibiotics; and catII from Escherichia coli K-12, qacG, fosA8, KpnH, rsmA, CRP, and gyrB. These genes confer resistance to various classes of antibiotics, including cephalosporins, cephamycins, penams, phenols, macrolides, ﬂuoroquinolones, aminoglycosides, diaminopyrimidines, and phosphonic acid antibiotics (Table S1B). They are also associated with alterations in antibiotic targets (PBP3, gyrB), the inactivation of antibiotics (DHA-17, fosA8), and the efﬂux of antibiotics (qacG, kpnH, rsmA, CRP). The Genes 2023, 14, 1279 5 of 16 genome also contained the insertion sequences IS200G, In36/37, and In6 with sequence identities of 84%, 99%, and 95%, respectively. IS200G is a salmonella-speciﬁc insertion sequence and contains the transposon gene (tnpA) [55]. This gene was located at position 1,945,287–1,945,977 bp, whereas, In36/37 and In6 were found at positions 580,797–582,017 and 1,049,772–1,050,897 bp, respectively [55]. These two insertion sequences are of E. coli plasmid origin (AY259086/5 and L06822) and carry the genes hypA (metallo-chaperon), ampR (transcriptional activator) and catA (chloramphenicol acetyltransferase). However, we obtained no positive results for the plasmid signature sequence and phage elements. 3.4. Comparative Phylogenomic Analysis of M. morganii Strains In this study, we performed a comparative genomics analysis of 82 M. morganii genomes retrieved from the NCBI GenBank database “https://www.ncbi.nlm.nih.gov (accessed on 30 November 2022)” from six countries, including the UM869 strain from this study (Table S2). The completeness and contamination levels of all the genomes ranged from 97.01–100% and 0–8.66%, respectively. UM869 showed 97.01% completeness and 0.27% contamination (Table S2). From the genome reannotation, the genome size of all strains ranged from 3,618,144 to 4,575,834 bps with varied N50 values (Table S2). Based on the contamination and completeness of the genomes, we excluded three genomes (GCF_026341575, GCF_018802465, and GCF_003852695) and performed a comparative genomics analysis of the remaining 79 M. morganii genomes. ANI between strains was calculated and subjected to hierarchical clustering into major groups determined among the 79 genomes to identify the closest strains based on their genome similarities. Genomes with ANI values greater than 95% were considered the same species. Among the strains, the ANI values ranged from 91.79% to 100%, with the highest ANI (100%) observed between GCF_018475065 and GCF_018475585, whereas the lowest (91.79%) was observed between GCF_018456225 and GCF_018475185 (Figure S2). Similarly, the ANI value of the UM869 strain with all other strains ranged between 91.96 and 99.82%. Because the average ANI percentage of all the genomes was 97.92%, which is greater than the ANI cutoff of 95%, all the strains belong to the same species. The ANI tree (Figure S2) is divided into three clusters, namely 1, 2, and 3, containing 64, 2, and 6 strains, respectively. The UM869 strain (GCA_025398975) is clustered (99.82% ANI value) with GCF_018474645. 3.5. M. morganii Phylogeny and Genetic Diversity The core SNP-based phylogenetic analysis exhibited diversity among the genomes of M. morganii (Figure 1). Phylogenetic analysis of the 79 M. morganii genomes revealed four major clusters and seven singlet nodes, as highlighted in Figure 1. We identiﬁed a close phylogenetic relationship between UM869 (GCA_025398975), isolated from urine, and GCF_018474645, isolated from sputum in China. However, it was found that the animal sample, i.e., GCF_018475125 and GCF_018475285, are closely related to the human samples (GCF_018475645 and GCF_018474925), although all four samples are from the same country (China) (Figure 1). The results unequivocally refute the hypothesis that host- speciﬁc lineages share a common evolutionary background with the host species under consideration [56]. The close sequence similarity between clinical and zoonotic strains demonstrates that food and the environment signiﬁcantly transmit the strain from animals to humans and between countries [57]. 3.6. Identiﬁcation and Analysis of Orthologues Genes From the OrthoFinder, we could assign 290,050 genes to 6069 orthogroups, which included 2447 genes belonging to single-copy orthologues, while 1577 genes remained unas- signed to any orthogroups. Out of the 6069 orthogroups identiﬁed from the 79 genomes, 2692 (44%) were core orthologues (99% ≥ strains ≤ 100%), 306 (5%) were soft-core ortho- logues (95% ≥ strains < 99%), 1005 (17%) were shell orthologues (15% ≥ strains < 95%), and 2066 (34%) were cloud genes (0% ≥ strains < 15%) as predicted by OrthoFinder. Further, Genes 2023, 14, 1279 6 of 16 2692 core orthologous groups were subjected to multiple sequence alignment to extract the consensus sequences for subsequent annotations. Figure 1. Core SNP-based phylogenetic analysis among 79 strains of M. morganii. Metadata such as isolation sources, host, and country are marked with different colors. The year of isolation for each genome is labeled after its accession number. Strains associated with the four clusters are delineated by blue, green, light purple, and light brown branches. The tree scale is presented as the estimated branching length. 3.7. Functional Annotation of Core Genomes From the functional annotation, 2692 core orthologues were assigned to 7 pathways, 48 super pathways, and 254 sub-pathways (Table S3). The most commonly identiﬁed super pathways in all core orthologues were transporters (468 orthologues), two-component systems (148 orthologues), transcription factors (117 orthologues), ribosomes (114 ortho- logues), ABC transporters (108 orthologues) with EC numbers (98 orthologues), transfer RNA biogenesis, DNA repair and recombination proteins (88 orthologues), and quorum sensing (77 orthologues) (Table S3). The top 20 key super pathways (≥15 counts) are shown Genes 2023, 14, x FOR PEER REVIEW 6 of 17 country (China) (Figure 1). The results unequivocally refute the hypothesis that host-spe-cific lineages share a common evolutionary background with the host species under con-sideration [56]. The close sequence similarity between clinical and zoonotic strains demonstrates that food and the environment significantly transmit the strain from ani-mals to humans and between countries [57]. Figure 1. Core SNP-based phylogenetic analysis among 79 strains of M. morganii. Metadata such as isolation sources, host, and country are marked with different colors. The year of isolation for each genome is labeled after its accession number. Strains associated with the four clusters are delineated by blue, green, light purple, and light brown branches. The tree scale is presented as the estimated branching length. 3.6. Identification and Analysis of Orthologues Genes From the OrthoFinder, we could assign 290,050 genes to 6069 orthogroups, which included 2447 genes belonging to single-copy orthologues, while 1577 genes remained unassigned to any orthogroups. Out of the 6069 orthogroups identified from the 79 ge-nomes, 2692 (44%) were core orthologues (99% ≥ strains ≤ 100%), 306 (5%) were soft-core orthologues (95% ≥ strains < 99%), 1005 (17%) were shell orthologues (15% ≥ strains < 95%), Genes 2023, 14, 1279 7 of 16 in Figure 2A. About 95 core orthologues were mapped to the “function unknown” category, suggesting that many aspects of M. morganii still require exploration. Figure 2. Distribution of core orthologues mapped to KEGG orthologues pathways and clusters of orthologous groups of proteins (COGs). Each bar represents the number of genes in their respective pathways/categories. (A) Top 20 KEGG pathways (≥15 counts). (B) Distribution of core orthologues in 20 COG categories mapped using eggNOG mapper. The identiﬁed core orthologues of M. morganii mapped to 2627 distinct clusters of orthologous groups (COGs) were divided into 21 unique COG categories, as listed in Table S4A. The highest number of COGs (453) belonged to the ‘function unknown’ category [S], followed by [E] amino acid transport and metabolism (269), [K] transcription (225), and [C] energy production and conversion (201). However, the lowest COGs were observed in [A] RNA processing and modiﬁcation (4), as shown in Figure 2B, while UM869 exhibits 73 cloud orthogroups belonging to 14 COG categories. Out of all COG categories, [S] ‘function unknown’ has the highest number of cloud orthologues (15), followed by [E] amino acid transport and metabolism (9), [P] inorganic ion transport and metabolism (9), and [K] transcription (7). Details of the COG categories with their descriptions are presented in Table S4B. 3.8. Identiﬁcation of AMR and Virulence Genes The annotation of core orthologues revealed that multiple antimicrobial resistance genes belong to different resistance mechanisms. Speciﬁcally, KpnH, PBP3, rsmA, CRP, Genes 2023, 14, x FOR PEER REVIEW 7 of 17 and 2066 (34%) were cloud genes (0% ≥ strains < 15%) as predicted by OrthoFinder. Fur-ther, 2692 core orthologous groups were subjected to multiple sequence alignment to ex-tract the consensus sequences for subsequent annotations. 3.7. Functional Annotation of Core Genomes From the functional annotation, 2692 core orthologues were assigned to 7 pathways, 48 super pathways, and 254 sub-pathways (Table S3). The most commonly identified su-per pathways in all core orthologues were transporters (468 orthologues), two-component systems (148 orthologues), transcription factors (117 orthologues), ribosomes (114 orthologues), ABC transporters (108 orthologues) with EC numbers (98 orthologues), transfer RNA biogenesis, DNA repair and recombination proteins (88 orthologues), and quorum sensing (77 orthologues) (Table S3). The top 20 key super pathways (≥15 counts) are shown in Figure 2A. About 95 core orthologues were mapped to the “function un-known” category, suggesting that many aspects of M. morganii still require exploration. Figure 2. Distribution of core orthologues mapped to KEGG orthologues pathways and clusters of orthologous groups of proteins (COGs). Each bar represents the number of genes in their respective pathways/categories. (A) Top 20 KEGG pathways (≥15 counts). (B) Distribution of core orthologues in 20 COG categories mapped using eggNOG mapper. The identified core orthologues of M. morganii mapped to 2627 distinct clusters of orthologous groups (COGs) were divided into 21 unique COG categories, as listed in Genes 2023, 14, 1279 8 of 16 and gyrB genes were identiﬁed in all genomes conferring resistance to ﬂuoroquinolone, aminoglycoside, carbapenem, cephalosporin, diaminopyrimidine, phenicol, cephamycin, and macrolide antibiotics, as shown in Table 1. The presence of qacG in all the genomes confers resistance to disinfecting agents and antiseptics. Further, four antibiotic efﬂux resistance mechanisms, including major facilitator superfamily (MFS), small multidrug re- sistance (SMR), resistance-nodulation-cell division (RND), antibiotic efﬂux pump, and two antibiotic target alteration resistance mechanisms, were predicted among all the genomes. Table 1. Antimicrobial resistance genes identiﬁed in core orthologues of 79 M. morganii strains. The AMR genes were identiﬁed using the CARD database in strict mode. Core Orthologues Gene Drug Class Resistance Mechanism AMR Gene Family OG0000319 KpnH Macrolide antibiotic; Fluoroquinolone antibiotic; Aminoglycoside antibiotic; Carbapenem; Cephalosporin; Penam; Peptide antibiotic; Penem Antibiotic efﬂux Major facilitator superfamily (MFS) antibiotic efﬂux pump OG0000873 PBP3 Cephalosporin; Cephamycin; Penam Antibiotic target alteration OG0001323 qacG Disinfecting agents and antiseptics Antibiotic efﬂux OG0002043 rsmA OG0002548 OG0002685 CRP gyrB Fluoroquinolone antibiotic; Diaminopyrimidine antibiotic; Phenicol antibiotic Macrolide antibiotic; Fluoroquinolone antibiotic; Penam Fluoroquinolone antibiotic Antibiotic efﬂux Antibiotic efﬂux Antibiotic target alteration Penicillin-binding protein mutations conferring resistance to β-lactam antibiotics Small multidrug resistance (SMR) antibiotic efﬂux pump Resistance-nodulation-cell division (RND) antibiotic efﬂux pump Resistance-nodulation-cell division (RND) antibiotic efﬂux pump Fluoroquinolone-resistant gyrB From the VFDB database annotation, only 47 core orthologues annotated to 15 VF classes, 27 virulence factors, and 38 associated genes across all the genomes (Table 2). The most frequently identiﬁed virulence factors included the type III secretion system (T3SS), type I ﬁmbriae, endotoxins, and toxins. The secretion system virulence factors class, such as T3SS, T4SS, and TTSS, were found to be particularly prevalent in pathogenic strains of the species, with several genes associated with T3SS, T4SS, and TTSS. Toxins, such as hemolysins, and endotoxins, such as lipooligosaccharide (LOS), were also identiﬁed, as were outer-membrane proteins involved in the adhesion and invasion of host cells. Autotransporters and ﬂagella virulence factor classes involved in diverse functions such as adhesion, invasion, toxin secretion, and host colonization were detected less frequently among the strains (Table 2). 3.9. Serotype In this study, we analyzed the serotype content of all 79 strains of M. morganii, as reported by Liu et al. [52]. The reported 11 serotypes were BLASTed against all the genomes. In thirty-one genomes, ﬁve serotypes were mapped with 100% coverage (Table S5). The type 8 O-antigen serotype was predicted in the genome UM869 at position 3,495,232 to 3,509,433 bp. The serotype region in UM869 was characterized by ten genes, including tarF, gt1, wzy, tagD, gt2, wzx, gnu, trmL, cysE, and gpsA, and ﬁve unannotated genes (ORFs) (Figure 3). These reported genes are involved in the biosynthesis and transport of the O-antigen component of the bacterial lipopolysaccharide. The presence of the wzx gene, responsible for encoding the O-unit ﬂippase and the wzy gene, responsible for encoding the O-antigen polymerase, suggest that M. morganii is likely to synthesize its O-antigen via the wzx/wzy-dependent pathway (Figure 3). In addition, the genes present in all the mapped Genes 2023, 14, 1279 9 of 16 serotypes were clustered and BLASTed against 31 genomes. The result was visualized by plotting the presence/absence of genes versus the genome, as depicted in Figure 4. Table 2. Details of identified putative virulence factors in the core orthologues of 79 M. morganii strains. Core Orthologues Virulence Gene Virulence Factors VF Class OG0000167 OG0000540 OG0001077 OG0002806 OG0001650 OG0001651 OG0001256 OG0001254 OG0000956 OG0000280 OG0001230 OG0001229 OG0000279 OG0000533 OG0002794 OG0002449 OG0001080 OG0000220 OG0000295 OG0000933 OG0000857 OG0000142 OG0000221 OG0000109 OG0000422 OG0000833 OG0001671 OG0002647 OG0002313 OG0000783 OG0001828 OG0002643 OG0000370 OG0001781 OG0001782 OG0001783 OG0001347 OG0001468 OG0001659 OG0001656 OG0001655 OG0002110 OG0001526 OG0000371 OG0000253 OG0000285 OG0001161 ﬁmD cheB cheR chuS chuU ireA sitA sitB sitC sitD basG feoA hemG phoQ spaP ﬂhB exsA - invC ysaS ysaV hlyA farB htrB lgtF lpxA lpxH lpxK opsX/rfaC wecA ﬁmC ﬁmD ﬁmH - mgtB mgtC motA motB - msbB2 - galE - katA Type I ﬁmbriae Adherence Flagella (Burkholderia) Autotransporter Heme uptake Iron-regulated element Iron/manganese transport Iron uptake Acinetobactin (Acinetobacter) Ferrous iron transport (Legionella) Heme biosynthesis (Hemophilus) PhoPQ (Salmonella) Regulation Bsa T3SS (Burkholderia) Flagella (cluster I) T3SS (Aeromonas) T4SS effectors (Coxiella) TTSS (SPI-1 encode) Ysa TTSS (Yersinia) Ysa TTSS (Yersinia) Secretion system HemolysinHlyA (Aeromonas) FarAB (Neisseria) Toxin Efﬂux pump LOS (Hemophilus) Endotoxin Fim (Salmonella) Capsule (Acinetobacter) Mg2+ transport (Salmonella) Flagella (Bordetella) Cysteine acquisition MsbB2 (Shigella) O-antigen (Yersinia) LPS rfb locus Catalase Fimbrial adherence determinants Immune evasion Magnesium uptake Motility Others Serum resistance Stress adaptation Genes 2023, 14, 1279 10 of 16 Figure 3. The gene content of serotype cassettes present in 31 out of 79 strains of M. morganii was used in this study. The genes and unannotated ORFs are drawn as arrows with orientations (forward and reverse). The image was created with gggenes v0.4.1. The details of mapping are presented in Supplementary Table S5. Figure 4. The heatmap illustrates the presence and absence of genes, where yellow indicates gene presence and blue indicates gene absence in their respective strains. The numbers in square brackets represents the cluster variants of the gene at 97% identity. The serotype genes were clustered at 97% similarity using CD-HIT. Genes with query coverage of ≥90% and identity of ≥99% were selected for generating the heatmap. Genes 2023, 14, x FOR PEER REVIEW 10 of 17 OG0001655 motB OG0002110 - Cysteine acquisition OG0001526 msbB2 MsbB2 (Shigella) Others OG0000371 - O-antigen (Yersinia) OG0000253 galE OG0000285 - LPS rfb locus Serum resistance OG0001161 katA Catalase Stress adaptation 3.9. Serotype In this study, we analyzed the serotype content of all 79 strains of M. morganii, as reported by Liu et al. [52]. The reported 11 serotypes were BLASTed against all the ge-nomes. In thirty-one genomes, five serotypes were mapped with 100% coverage (Table S5). The type 8 O-antigen serotype was predicted in the genome UM869 at position 3,495,232 to 3,509,433 bp. The serotype region in UM869 was characterized by ten genes, including tarF, gt1, wzy, tagD, gt2, wzx, gnu, trmL, cysE, and gpsA, and five unannotated genes (ORFs) (Figure 3). These reported genes are involved in the biosynthesis and transport of the O-antigen component of the bacterial lipopolysaccharide. The presence of the wzx gene, responsible for encoding the O-unit flippase and the wzy gene, responsible for encoding the O-antigen polymerase, suggest that M. morganii is likely to synthesize its O-antigen via the wzx/wzy-dependent pathway (Figure 3). In addition, the genes present in all the mapped serotypes were clustered and BLASTed against 31 genomes. The result was visualized by plotting the presence/absence of genes versus the genome, as depicted in Figure 4. Figure 3. The gene content of serotype cassettes present in 31 out of 79 strains of M. morganii was used in this study. The genes and unannotated ORFs are drawn as arrows with orientations (for-ward and reverse). The image was created with gggenes v0.4.1. The details of mapping are pre-sented in Supplementary Table S5. Genes 2023, 14, x FOR PEER REVIEW 11 of 17 Figure 4. The heatmap illustrates the presence and absence of genes, where yellow indicates gene presence and blue indicates gene absence in their respective strains. The numbers in square brackets represents the cluster variants of the gene at 97% identity. The serotype genes were clustered at 97% similarity using CD-HIT. Genes with query coverage of ≥90% and identity of ≥99% were selected for generating the heatmap. 4. Discussion Epidemiological investigations have consistently identified M. morganii as a frequent causative agent of nosocomial bacterial infections worldwide [58–62]. Repeated reports on acquired resistance in M. morganii reveal more about their life-threatening actions as it further complicates existing treatment options [4]. Despite the serious clinical threat posed by the intrinsic and acquired resistance of M. morganii, it has received less attention so far. However, few studies have explored the evolutionary relationships and intricate internal genome structure of multiple genomes of M. morganii using genetic information from pub-lic databases [3,10,63]. Comparative genomic analysis, combined with a geographical re-gion, isolation source, host, and antibiotic resistance gene content, is valuable for conduct-ing genomic epidemiological analysis. In this study, the MDR M. morganii UM869 strain genome, obtained from patients with urinary tract infection (UTI), was compared with 78 publicly available M. morganii genomes through ANI and core SNP-based phylogenetic analysis. Comparative phylo-genomic analysis revealed that the UM869 genome was closely related to the GCF_018474645 (strain FS112720; isolated from sputum) and GCF_018474565 (strain E89; isolated from secretions) genomes from China (Figures 1 and S2). The observed clustering of M. morganii strains from different geographic locations and isolation sources in a short timeline over a period from 2015 to 2021 suggests that these strains may be highly clonal Genes 2023, 14, 1279 11 of 16 4. Discussion Epidemiological investigations have consistently identiﬁed M. morganii as a frequent causative agent of nosocomial bacterial infections worldwide [58–62]. Repeated reports on acquired resistance in M. morganii reveal more about their life-threatening actions as it further complicates existing treatment options [4]. Despite the serious clinical threat posed by the intrinsic and acquired resistance of M. morganii, it has received less attention so far. However, few studies have explored the evolutionary relationships and intricate internal genome structure of multiple genomes of M. morganii using genetic information from public databases [3,10,63]. Comparative genomic analysis, combined with a geographical region, isolation source, host, and antibiotic resistance gene content, is valuable for conducting genomic epidemiological analysis. In this study, the MDR M. morganii UM869 strain genome, obtained from patients with urinary tract infection (UTI), was compared with 78 publicly available M. morganii genomes through ANI and core SNP-based phylogenetic analysis. Comparative phylogenomic analysis revealed that the UM869 genome was closely related to the GCF_018474645 (strain FS112720; isolated from sputum) and GCF_018474565 (strain E89; isolated from secretions) genomes from China (Figures 1 and S2). The observed clustering of M. morganii strains from different geographic locations and isolation sources in a short timeline over a period from 2015 to 2021 suggests that these strains may be highly clonal [64] and might have spread due to their close geographical location in the map or dissemination due to frequent trade and tourism. From the analysis of core orthologues, 290,050 genes were grouped into 6069 orthologues, of which 2692 were core orthologues. These core orthologues proteins usually retain their orig- inal function during microorganism evolution and help determine the relationships between genome structure, gene function, and taxonomic classification [65]. However, 2066 cloud genes might reflect the phenotypical traits specific to the group of M. morganii [66]. Therefore, it is important to classify these core orthologues into COGs and predict their functions, par- ticularly in emerging pathogens with newly sequenced genomes [65]. This study mapped most core orthologues to transporters and two-component system pathways (Table S3). These transporters utilize ATP hydrolysis or proton gradient to transport a wide range of substrates across the membrane, including nutrients, toxins, and antibiotics [67,68]. Similarly, 2631 core orthologues were assigned to 2627 distinct COGs belonging to 21 categories. In the UM869 strain, 73 cloud orthologues were predicted and mapped to 14 COG categories. Moreover, most core and cloud orthologues were mapped to the [S] function unknown COG category, which might contain novel functional genes. Several TCSs have been identiﬁed and characterized in M. morganii, including the PhoP/PhoQ, ArcAB, CpxAR, and PmrAB systems responsible for antimicrobial resistance in bacteria. Studies have shown that the PhoP/PhoQ system regulates phosphate home- ostasis, virulence, and antimicrobial resistance [69], suggesting that these strains might be highly resistant to antibiotics. This further complies with the resistance proﬁling as analyzed through VITEK 2 system. In this study, WGS analysis revealed that UM869 harbored KpnH, qacG, rsmA, and CRP efﬂux pumps, which may confer resistance to most routinely used classes of antibiotics such as macrolide, ﬂuoroquinolone, aminoglycoside, cephalosporin, carbapenem, and colistin antibiotics (Table 1). The rsmA gene belongs to the resistance-nodulation-cell division (RND) efﬂux pump, which regulates quorum sensing, a communication system in bacteria, and the mutated rsmA is linked to increased production of bioﬁlm, elastase, and antibiotic resistance [70]. Similarly, CRP is a regulatory gene that codes for the cAMP receptor protein, which regulates bacteria’s virulence genes and carbon metabolism [71]. Another multidrug efﬂux pump qacG gene that confers resistance to various antimicrobial agents, identiﬁed in Gram- positive and Gram-negative bacteria, is associated with increased resistance to commonly used healthcare disinfectants [72]. Fluoroquinolone resistance could be mediated by a point mutation in gyrB, which encodes the β-subunit of DNA gyrase [73]. The point mutation was also observed in PBP3 (penicillin-binding protein), which results in resistance to β-lactam Genes 2023, 14, 1279 12 of 16 antibiotics, such as penicillins, cephalosporins, and carbapenems across various bacterial species, including Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa [74,75]. However, only the UM869 strain exhibited unique resistant genes such as fosA8, ArnT, and qacG. Previous studies have reported that fosA8 expression in chromosomal fosA genes of E. coli signiﬁcantly confers resistance to fosfomycin [76]. Similarly, ArnT, a glycosyltransferase, is essential for bacterial resistance against antimicrobial peptides as it adds 4-amino-4-deoxy-l-arabinose (l-Ara4N) to the lipid A component of lipopolysaccharide, enabling the evasion of antimicrobial effects [77]. The identiﬁcation of multiple antibiotic resistance mechanisms in M. morganii emphasizes the potential threats associated with it. These mechanisms, including efﬂux pumps and gene mutations, enable bacteria to survive exposure to commonly used antibiotics, complicating treatment and increasing the risk of untreatable infections. Several virulence factors have been identiﬁed in the M. morganii genome, including type III secretion system (T3SS), type I ﬁmbriae, endotoxins, and toxins. In the UM869 strain, ﬁmCDH genes are predicted as type I ﬁmbriae virulence factors and play an important role in the colonization and pathogenicity of M. morganii as these are the most common virulence factors responsible for adherence to surfaces or other cells [4]. The iron acquisition and secretion system (T3SS, T4SS, and TTSS) was the most abundant virulence factor encoded in the UM869 genome that functions in immune evasion (IgA protease) and hemolysins [78]. The type-3 secretion system (T3SS) is a highly conserved virulence factor in disease-causing Gram-negative bacteria and is responsible for injecting bacterial effector proteins directly into the host cell cytoplasm [79]. Similarly, iron acquisition genes sitABCD mediate manganese–iron transfer and are essential for bacterial survival in iron-deﬁcient environments [80]. The gene hlyA was also encoded in the UM869 genome, which is homologous to the α-hemolysin gene of E. coli. It binds to the cell surface and matures into a β-barrel transmembrane pore, creating an aqueous channel that permits the transport of small molecules such as K+ and Ca2+ ions, which causes the necrotic death of the target cell [81]. The UM869 genome also contains the efﬂux pump (farB), which contributes to resistance by pumping out molecules, such as bile salts and antimicrobial peptides, which can help the bacteria, evade the immune system and disease-causing agents [82–84]. The mobile genetic elements play a relevant evolutionary role that drives genome plasticity. The insertion sequence and transposon elements in the UM869 genome imply the possibility of disseminating resistant determinants via horizontal gene transfer [85]. The varied structure of O-antigen in bacteria differentiated the M. morganii species at the strain level. Liu et al. 2021 [52] developed a molecular serotyping based on diverse O-antigen gene clusters (O-AGC) in M. morganii. Based on serotype sequences reported by Liu et al., 2021 [52], 31 genomes were classiﬁed into ﬁve different serotypes (Figure 4) in our study. The gene structure of serotype cassettes was present in varied combinations in different strains of M. morganii, suggesting that the serotyping of M. morganii may be complex and require various genotypic and phenotypic methods to be understood [74]. In Gram-negative bacteria, the wzx/wzy-dependent pathway is predominant for producing O- antigen and differentiates the O-antigen clusters [86]. In this study, the serotypes predicted both wzx and wzy genes in the M. morganii genomes. 5. Conclusions This is the ﬁrst report from India that provides a genomic insight into the diversity and emergence of resistant determinants in M. morganii through a comparative genomic study. However, the episodes of the outbreak of M. morganii clones are less frequent and restricted to the eastern part of the globe (Asia). Therefore, studying the faster dissemination rate, the acquisition of resistance determinants, and the comprehensive surveillance of the M. morganii infection are all highly desirable before this genus potentially causes an uncontrollable epidemic. Further, the enrichment of the M. morganii public database will help better understand the bacteria’s origin, evolution, and transmission and also toward designing suitable therapeutics to overcome infections. Genes 2023, 14, 1279 13 of 16 Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/genes14061279/s1, Figure S1: Workflow of the genome assembly and annotation of M. morganii UM869 strain isolated from UTI, and comparative genomics analysis with other M. morganii genomes; Figure S2: Heatmap and dendrogram illustrating the ANI among the 79 M. morganii genomes generated by FastANI are organized according to the dendrograms (left/below) obtained with the neighbor-joining clustering method. The color represents the highest and lowest genetic similarity between the genomes. Genomes containing the ≥99% identity are shown in red, and identity containing ≤93% are shown in blue; Table S1A: Virulence factor and virulence class predicted in M. Morganii UM869; Table S1B: Presence of Antimicrobial resistance genes, resistant drug class and their presence in M. morganii UM869; Table S2: Metadata and summary statistics of 82 M. morganii genome; Table S3: Details of mapping core orthologues of 79 M. morganii strains to KEGG orthologues (KO) and KEGG pathways; Table S4A: Mapping of core orthologues gene of 79 M. morganii strain to the COG categories. Forty-three core orthologues did not map to any COG categories; Table S4B: Mapping cloud orthologues gene of UM869 M. morganii strain to the COG categories; Table S5: Mapping of 11 putative O-Antigen Gene Clusters (serotype) in 79 M. morganii strains using BLAST. Only 5 serotypes mapped to 31 strains at 100% query coverage. Author Contributions: Conceptualization, E.S., D.U.B. and M.G.; methodology, D.U.B., S.D., M.G., R.M., B.B.S. and R.K.S.; formal analysis, D.U.B., S.D. and M.G.; data curation, D.U.B. and S.D.; visualization, D.U.B., S.D., M.G., M.S. and B.K.B.; writing—original draft preparation, D.U.B., S.D. and E.S.; writing—review and editing, D.U.B., S.D., M.G., R.K.S., R.M., M.S., S.S.B., B.K.B., B.B.S. and E.S.; supervision, E.S. All authors have read and agreed to the published version of the manuscript. Funding: This research did not receive any speciﬁc grant from funding agencies in the public, commercial, or not-for-proﬁt sectors. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: WGS sequence reads were submitted to the NCBI’s Bioproject database with the accession ID: PRJNA598939. Acknowledgments: We gratefully acknowledge the infrastructure facility provided by the president of Siksha ‘O’ Anusandhan (deemed to be university), Bhubaneswar, and the computing facility at the Drug Development and Analysis Lab, School of Pharmaceutical Sciences, Siksha ‘O’ Anusandhan (deemed to be university) sponsored by ICMR (grant no. AMR/DHR/GIA/4/ECD-II-2020), India. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. 2. Adeolu, M.; Alnajar, S.; Naushad, S.; Gupta, R.S. Genome-Based Phylogeny and Taxonomy of the ‘Enterobacteriales’: Proposal for Enterobacterales Ord. Nov. Divided into the Families Enterobacteriaceae, Erwiniaceae Fam. Nov., Pectobacteriaceae Fam. Nov., Yersiniaceae Fam. Nov., Hafniaceae Fam. Nov., Morgane. Int. J. Syst. Evol. Microbiol. 2016, 66, 5575–5599. [CrossRef] O’Hara, C.M.; Brenner, F.W.; Miller, J.M. Classiﬁcation, Identiﬁcation, and Clinical Signiﬁcance of Proteus, Providencia, and Morganella. Clin. Microbiol. Rev. 2000, 13, 534–546. [CrossRef] 3. Minnullina, L.; Pudova, D.; Shagimardanova, E.; Shigapova, L.; Sharipova, M.; Mardanova, A. Comparative Genome Analysis of 4. 5. 6. 7. 8. 9. Uropathogenic Morganella Morganii Strains. Front. Cell. Infect. Microbiol. 2019, 9, 167. [CrossRef] Liu, H.; Zhu, J.; Hu, Q.; Rao, X. Morganella Morganii, a Non-Negligent Opportunistic Pathogen. Int. J. Infect. Dis. 2016, 50, 10–17. [CrossRef] van Bentum, J.S.; Sijbrandij, M.; Kerkhof, A.J.F.M.; Huisman, A.; Arntz, A.R.; Holmes, E.A.; Franx, G.; Mokkenstorm, J.; Huibers, M.J.H. Treating Repetitive Suicidal Intrusions Using Eye Movements: Study Protocol for a Multicenter Randomized Clinical Trial. BMC Psychiatry 2019, 19, 143. [CrossRef] Zhang, B.; Pan, F.; Zhu, K. Bilateral Morganella Morganii Keratitis in a Patient with Facial Topical Corticosteroid-Induced Rosacea-like Dermatitis: A Case Report. BMC Ophthalmol. 2017, 17, 106. [CrossRef] Erlanger, D.; Assous, M.V.; Wiener-Well, Y.; Yinnon, A.M.; Ben-Chetrit, E. Clinical Manifestations, Risk Factors and Prognosis of Patients with Morganella Morganii Sepsis. J. Microbiol. Immunol. Infect. 2019, 52, 443–448. [CrossRef] van Bentum, R.; Nieken, J.; de Waal, E.; Hoogendoorn, M. Native Aortic Valve Endocarditis with Morganella Morganii in a Patient with Multiple Myeloma and Valvular Amyloidosis: A Case Report. BMC Infect. Dis. 2019, 19, 957. [CrossRef] Kohlmann, R.; Bähr, T.; Gatermann, S.G. Species-Speciﬁc Mutation Rates for AmpC Derepression in Enterobacterales with Chromosomally Encoded Inducible AmpC β-Lactamase. J. Antimicrob. Chemother. 2018, 73, 1530–1536. [CrossRef] Genes 2023, 14, 1279 14 of 16 10. Ryser, L.T.; Arias-Roth, E.; Perreten, V.; Irmler, S.; Bruggmann, R. Genetic and Phenotypic Diversity of Morganella Morganii Isolated From Cheese. Front. Microbiol. 2021, 12, 738492. [CrossRef] 11. Karaiskos, I.; Giamarellou, H. Multidrug-Resistant and Extensively Drug-Resistant Gram-Negative Pathogens: Current and Emerging Therapeutic Approaches. Expert Opin. Pharmacother. 2014, 15, 1351–1370. [CrossRef] 12. Luo, X.; Zhai, Y.; He, D.; Cui, X.; Yang, Y.; Yuan, L.; Liu, J.; Hu, G. Molecular Characterization of a Novel Bla CTX-M-3-Carrying Tn6741 Transposon in Morganella Morganii Isolated from Swine. J. Med. Microbiol. 2020, 69, 1089–1094. [CrossRef] 13. Moura, Q.; Cerdeira, L.; Fernandes, M.R.; Vianello, M.A.; Lincopan, N. Novel Class 1 Integron (In 1390) Harboring Bla GES-5 in a Morganella Morganii Strain Recovered from a Remote Community. Diagn. Microbiol. Infect. Dis. 2018, 91, 345–347. [CrossRef] 14. Chen, Y.; Lei, C.; Zuo, L.; Kong, L.; Kang, Z.; Zeng, J.; Zhang, X.; Wang, H. A Novel Cfr-Carrying Tn7 Transposon Derivative Characterized in Morganella Morganii of Swine Origin in China. J. Antimicrob. Chemother. 2019, 74, 603–606. [CrossRef] 15. Xiang, G.; Lan, K.; Cai, Y.; Liao, K.; Zhao, M.; Tao, J.; Ma, Y.; Zeng, J.; Zhang, W.; Wu, Z.; et al. Clinical Molecular and Genomic Epidemiology of Morganella Morganii in China. Front. Microbiol. 2021, 12, 744291. [CrossRef] 16. Mikheyev, A.S.; Tin, M.M.Y. A First Look at the Oxford Nanopore MinION Sequencer. Mol. Ecol. Resour. 2014, 14, 1097–1102. 17. [CrossRef] Jain, M.; Olsen, H.E.; Paten, B.; Akeson, M. The Oxford Nanopore MinION: Delivery of Nanopore Sequencing to the Genomics Community. Genome Biol. 2016, 17, 239. [CrossRef] 18. Quick, J.; Ashton, P.; Calus, S.; Chatt, C.; Gossain, S.; Hawker, J.; Nair, S.; Neal, K.; Nye, K.; Peters, T.; et al. Rapid Draft Sequencing and Real-Time Nanopore Sequencing in a Hospital Outbreak of Salmonella. Genome Biol. 2015, 16, 114. [CrossRef] 19. Clinical and Laboratory Standards Institute. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically. Standard, Approval CDM-A. In M07 Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Clinical and Laboratory Standards Institute: Wayne, PA, USA, 2018; p. 91. Sahoo, R.K.; Subudhi, E.; Kumar, M. Quantitative Approach to Track Lipase Producing Pseudomonas Sp. S1 in Nonsterilized Solid State Fermentation. Lett. Appl. Microbiol. 2014, 58, 610–616. [CrossRef] 20. 21. Weisburg, W.G.; Barns, S.M.; Pelletier, D.A.; Lane, D.J. 16S Ribosomal DNA Ampliﬁcation for Phylogenetic Study. J. Bacteriol. 22. 1991, 173, 697–703. [CrossRef] Jain, M.; Koren, S.; Miga, K.H.; Quick, J.; Rand, A.C.; Sasani, T.A.; Tyson, J.R.; Beggs, A.D.; Dilthey, A.T.; Fiddes, I.T.; et al. Nanopore Sequencing and Assembly of a Human Genome with Ultra-Long Reads. Nat. Biotechnol. 2018, 36, 338–345. [CrossRef] [PubMed] 23. Wick, R.R.; Judd, L.M.; Holt, K.E. Performance of Neural Network Basecalling Tools for Oxford Nanopore Sequencing. Genome Biol. 2019, 20, 129. [CrossRef] [PubMed] 24. Loman, N.J.; Quinlan, A.R. Poretools: A Toolkit for Analyzing Nanopore Sequence Data. Bioinformatics 2014, 30, 3399–3401. [CrossRef] [PubMed] 25. Wick, R.R.; Judd, L.M.; Gorrie, C.L.; Holt, K.E. Unicycler: Resolving Bacterial Genome Assemblies from Short and Long Sequencing Reads. PLOS Comput. Biol. 2017, 13, e1005595. [CrossRef] [PubMed] 26. De Coster, W.; D’Hert, S.; Schultz, D.T.; Cruts, M.; Van Broeckhoven, C. NanoPack: Visualizing and Processing Long-Read Sequencing Data. Bioinformatics 2018, 34, 2666–2669. [CrossRef] 27. Kolmogorov, M.; Yuan, J.; Lin, Y.; Pevzner, P.A. Assembly of Long, Error-Prone Reads Using Repeat Graphs. Nat. Biotechnol. 2019, 37, 540–546. [CrossRef] 28. Gurevich, A.; Saveliev, V.; Vyahhi, N.; Tesler, G. QUAST: Quality Assessment Tool for Genome Assemblies. Bioinformatics 2013, 29, 1072–1075. [CrossRef] 29. Tatusova, T.; Dicuccio, M.; Badretdin, A.; Chetvernin, V.; Nawrocki, E.P.; Zaslavsky, L.; Lomsadze, A.; Pruitt, K.D.; Borodovsky, 30. M.; Ostell, J. NCBI Prokaryotic Genome Annotation Pipeline. Nucleic Acids Res. 2016, 44, 6614–6624. [CrossRef] Siguier, P.; Perochon, J.; Lestrade, L.; Mahillon, J.; Chandler, M. ISﬁnder: The Reference Centre for Bacterial Insertion Sequences. Nucleic Acids Res. 2006, 34, D32-6. [CrossRef] 31. Ross, K.; Varani, A.M.; Snesrud, E.; Huang, H.; Alvarenga, D.O.; Zhang, J.; Wu, C.; McGann, P.; Chandlere, M. TnCentral: A Prokaryotic Transposable Element Database and Web Portal for Transposon Analysis. MBio 2021, 12, e0206021. [CrossRef] 32. Rozwandowicz, M.; Brouwer, M.S.M.; Fischer, J.; Wagenaar, J.A.; Gonzalez-Zorn, B.; Guerra, B.; Mevius, D.J.; Hordijk, J.; Lefebvre, B.; Lévesque, S.; et al. In Silico Detection and Typing of Plasmids Using Plasmidﬁnder and Plasmid Multilocus Sequence Typing. Antimicrob. Agents Chemother. 2018, 10, 1–14. 33. Arndt, D.; Grant, J.R.; Marcu, A.; Sajed, T.; Pon, A.; Liang, Y.; Wishart, D.S. PHASTER: A Better, Faster Version of the PHAST Phage Search Tool. Nucleic Acids Res. 2016, 44, W16–W21. [CrossRef] [PubMed] 34. Parks, D.H.; Imelfort, M.; Skennerton, C.T.; Hugenholtz, P.; Tyson, G.W. CheckM: Assessing the Quality of Microbial Genomes Recovered from Isolates, Single Cells, and Metagenomes. Genome Res. 2015, 25, 1043–1055. [CrossRef] [PubMed] 35. Manni, M.; Berkeley, M.R.; Seppey, M.; Zdobnov, E.M. BUSCO: Assessing Genomic Data Quality and Beyond. Curr. Protoc. 2021, 1, e323. [CrossRef] [PubMed] 36. Liu, H.; Xin, B.; Zheng, J.; Zhong, H.; Yu, Y.; Peng, D.; Sun, M. Build a Bioinformatic Analysis Platform and Apply It to Routine Analysis of Microbial Genomics and Comparative Genomics. Protoc. Exch. 2022, 4, 88–100. Genes 2023, 14, 1279 15 of 16 37. Warnes, G.R.; Bolker, B.; Bonebakker, L.; Gentleman, R.; Liaw, W.H.A.; Lumley, T.; Maechler, M.; Magnusson, A.; Moeller, S.; Schwartz, M.; et al. Package “Gplots”: Various R Programming Tools for Plotting Data. R Package. Version 2.17.0. 2016, pp. 1–68. Available online: https://rdrr.io/cran/gplots/ (accessed on 5 March 2023). 38. Emms, D.M.; Kelly, S. OrthoFinder: Solving Fundamental Biases in Whole Genome Comparisons Dramatically Improves Orthogroup Inference Accuracy. Genome Biol. 2015, 16, 157. [CrossRef] 39. Letunic, I.; Bork, P. Interactive Tree Of Life (ITOL) v5: An Online Tool for Phylogenetic Tree Display and Annotation. Nucleic Acids Res. 2021, 49, W293–W296. [CrossRef] 40. Katoh, K.; Misawa, K.; Kuma, K.I.; Miyata, T. MAFFT: A Novel Method for Rapid Multiple Sequence Alignment Based on Fast Fourier Transform. Nucleic Acids Res. 2002, 30, 3059–3066. [CrossRef] 41. Mikita, S.; David, T.; Peer, B. PAL2NAL: Robust Conversion of Protein Sequence Alignments into the Corresponding Codon Alignments. Nucleic Acids Res. 2006, 34, W609–W612. 42. Page, A.J.; Taylor, B.; Delaney, A.J.; Soares, J.; Seemann, T.; Keane, J.A.; Harris, S.R. SNP-Sites: Rapid Efﬁcient Extraction of SNPs from Multi-FASTA Alignments. Microb. Genom. 2016, 2, e000056. [CrossRef] 43. Minh, B.Q.; Schmidt, H.A.; Chernomor, O.; Schrempf, D.; Woodhams, M.D.; Von Haeseler, A.; Lanfear, R.; Teeling, E. IQ-TREE 2: New Models and Efﬁcient Methods for Phylogenetic Inference in the Genomic Era. Mol. Biol. Evol. 2020, 37, 1530–1534. [CrossRef] [PubMed] 44. Emms, D.M.; Kelly, S. OrthoFinder: Phylogenetic Orthology Inference for Comparative Genomics. Genome Biol. 2019, 20, 238. [CrossRef] [PubMed] 45. Tumescheit, C.; Firth, A.E.; Brown, K. CIAlign: A Highly Customisable Command Line Tool to Clean, Interpret and Visualise Multiple Sequence Alignments. PeerJ 2022, 10, e12983. [CrossRef] 46. Aramaki, T.; Blanc-Mathieu, R.; Endo, H.; Ohkubo, K.; Kanehisa, M.; Goto, S.; Ogata, H. KofamKOALA: KEGG Ortholog Assignment Based on Proﬁle HMM and Adaptive Score Threshold. Bioinformatics 2020, 36, 2251–2252. [CrossRef] [PubMed] 47. Cantalapiedra, C.P.; Hernández-Plaza, A.; Letunic, I.; Bork, P.; Huerta-Cepas, J. EggNOG-Mapper v2: Functional Annotation, Orthology Assignments, and Domain Prediction at the Metagenomic Scale. Mol. Biol. Evol. 2021, 38, 5825–5829. [CrossRef] 48. Huerta-Cepas, J.; Szklarczyk, D.; Heller, D.; Hernández-Plaza, A.; Forslund, S.K.; Cook, H.; Mende, D.R.; Letunic, I.; Rattei, T.; Jensen, L.J.; et al. EggNOG 5.0: A Hierarchical, Functionally and Phylogenetically Annotated Orthology Resource Based on 5090 Organisms and 2502 Viruses. Nucleic Acids Res. 2019, 47, D309–D314. [CrossRef] 49. Alcock, B.P.; Raphenya, A.R.; Lau, T.T.Y.; Tsang, K.K.; Bouchard, M.; Edalatmand, A.; Huynh, W.; Nguyen, A.L.V.; Cheng, A.A.; Liu, S.; et al. CARD 2020: Antibiotic Resistome Surveillance with the Comprehensive Antibiotic Resistance Database. Nucleic Acids Res. 2020, 48, D517–D525. [CrossRef] Feldgarden, M.; Brover, V.; Gonzalez-Escalona, N.; Frye, J.G.; Haendiges, J.; Haft, D.H.; Hoffmann, M.; Pettengill, J.B.; Prasad, A.B.; Tillman, G.E.; et al. AMRFinderPlus and the Reference Gene Catalog Facilitate Examination of the Genomic Links among Antimicrobial Resistance, Stress Response, and Virulence. Sci. Rep. 2021, 11, 12728. [CrossRef] 50. 51. Liu, B.; Zheng, D.; Jin, Q.; Chen, L.; Yang, J. VFDB 2019: A Comparative Pathogenomic Platform with an Interactive Web Interface. Nucleic Acids Res. 2019, 47, D687–D692. [CrossRef] 52. Liu, B.; Guo, X.; Wang, J.; Wu, P.; Li, S.; Feng, L.; Liu, B.; Wang, L. Development of a Molecular Serotyping Scheme for Morganella Morganii. Front. Microbiol. 2021, 12, 791165. [CrossRef] 53. Li, W.; Godzik, A. Cd-Hit: A Fast Program for Clustering and Comparing Large Sets of Protein or Nucleotide Sequences. Bioinformatics 2006, 22, 1658–1659. [CrossRef] [PubMed] 54. Boratyn, G.M.; Camacho, C.; Cooper, P.S.; Coulouris, G.; Fong, A.; Ma, N.; Madden, T.L.; Matten, W.T.; McGinnis, S.D.; Merezhuk, Y.; et al. BLAST: A More Efﬁcient Report with Usability Improvements. Nucleic Acids Res. 2013, 41, W29–W33. [CrossRef] [PubMed] 55. Lam, S.; Roth, J.R. IS200: A Salmonella-Speciﬁc Insertion Sequence. Cell 1983, 34, 951–960. [CrossRef] [PubMed] 56. Palmieri, N.; Hess, C.; Hess, M.; Alispahic, M. Sequencing of Five Poultry Strains Elucidates Phylogenetic Relationships and 57. Divergence in Virulence Genes in Morganella Morganii. BMC Genom. 2020, 21, 579. [CrossRef] Sahoo, R.K.; Gaur, M.; Dey, S.; Sahoo, S.; Das, A.; Subudhi, E. Genomic Insight of Extremely Drug-Resistant Klebsiella Pneumoniae ST5378 from a Pediatric Bloodstream Infection. J. Glob. Antimicrob. Resist. 2023, 33, 227–230. [CrossRef] [PubMed] 58. Lee, I.-K.; Liu, J.-W. Clinical Characteristics and Risk Factors for Mortality in Morganella Morganii Bacteremia. J. Microbiol. Immunol. Infect. 2006, 39, 328–334. [PubMed] 59. Gebhart-Mueller, E.Y.; Mueller, P.; Nixon, B. Unusual Case of Postoperative Infection Caused by Morganella Morganii. J. Foot Ankle Surg. 1998, 37, 145–147. [CrossRef] 60. Kim, J.H.; Cho, C.R.; Um, T.H.; Rhu, J.Y.; Kim, E.S.; Jeong, J.W.; Lee, H.R. Morganella Morganii Sepsis with Massive Hemolysis. 61. 62. J. Korean Med. Sci. 2007, 22, 1082. [CrossRef] Sinha, A.K.; Kempley, S.T.; Price, E.; Sharma, B.K.; Livermore, D.M. Early onset morganella morganii sepsis in a newborn infant with emergence of cephalosporin resistance caused by derepression of ampc?-lactamase production. Pediatr. Infect. Dis. J. 2006, 25, 376–377. [CrossRef] Falagas, M.E.; Kavvadia, P.K.; Mantadakis, E.; Kofteridis, D.P.; Bliziotis, I.A.; Saloustros, E.; Maraki, S.; Samonis, G. Morganella Morganii Infections in a General Tertiary Hospital. Infection 2006, 34, 315–321. [CrossRef] Genes 2023, 14, 1279 16 of 16 63. Guo, X.; Rao, Y.; Guo, L.; Xu, H.; Lv, T.; Yu, X.; Chen, Y.; Liu, N.; Han, H.; Zheng, B. Detection and Genomic Characterization of a Morganella Morganii Isolate From China That Produces NDM-5. Front. Microbiol. 2019, 10, 1156. [CrossRef] [PubMed] 64. Chen, Y.T.; Peng, H.L.; Shia, W.C.; Hsu, F.R.; Ken, C.F.; Tsao, Y.M.; Chen, C.H.; Liu, C.E.; Hsieh, M.F.; Chen, H.C.; et al. Whole-Genome Sequencing and Identiﬁcation of Morganella Morganii KT Pathogenicity-Related Genes. BMC Genom. 2012, 13 (Suppl. S7), S4. [CrossRef] [PubMed] 65. Tekaia, F. Inferring Orthologs: Open Questions and Perspectives. Genom. Insights 2016, 9, 17–28. [CrossRef] 66. Kahlke, T.; Goesmann, A.; Hjerde, E.; Willassen, N.; Haugen, P. Unique Core Genomes of the Bacterial Family Vibrionaceae: Insights into Niche Adaptation and Speciation. BMC Genom. 2012, 13, 179. [CrossRef] [PubMed] 67. Drew, D.; North, R.A.; Nagarathinam, K.; Tanabe, M. Structures and General Transport Mechanisms by the Major Facilitator 68. Superfamily (MFS). Chem. Rev. 2021, 121, 5289–5335. [CrossRef] [PubMed] Sarathy, J.P.; Dartois, V.; Lee, E.J.D. The Role of Transport Mechanisms in Mycobacterium Tuberculosis Drug Resistance and Tolerance. Pharmaceuticals 2012, 5, 1210–1235. [CrossRef] 69. Lee, J.; Zhang, L. The Hierarchy Quorum Sensing Network in Pseudomonas Aeruginosa. Protein Cell 2015, 6, 26–41. [CrossRef] 70. Li, X.Z.; Nikaido, H. Efﬂux-Mediated Drug Resistance in Bacteria: An Update. Drugs 2009, 69, 1555–1623. [CrossRef] 71. Soberón-Chávez, G.; Alcaraz, L.D.; Morales, E.; Ponce-Soto, G.Y.; Servín-González, L. The Transcriptional Regulators of the CRP Family Regulate Different Essential Bacterial Functions and Can Be Inherited Vertically and Horizontally. Front. Microbiol. 2017, 8, 959. [CrossRef] 72. Dashtbani-Roozbehani, A.; Brown, M.H. Efﬂux Pump Mediated Antimicrobial Resistance by Staphylococci in Health-Related Environments: Challenges and the Quest for Inhibition. Antibiotics 2021, 10, 1502. [CrossRef] 73. Han, J.; Wang, Y.; Sahin, O.; Shen, Z.; Guo, B.; Shen, J.; Zhang, Q. A Fluoroquinolone Resistance Associated Mutation in GyrA Affects DNA Supercoiling in Campylobacter Jejuni. Front. Cell. Infect. Microbiol. 2012, 2, 21. [CrossRef] 74. Macheboeuf, P.; Contreras-Martel, C.; Job, V.; Dideberg, O.; Dessen, A. Penicillin Binding Proteins: Key Players in Bacterial Cell Cycle and Drug Resistance Processes. FEMS Microbiol. Rev. 2006, 30, 673–691. [CrossRef] [PubMed] 75. Papp-Wallace, K.M.; Endimiani, A.; Taracila, M.A.; Bonomo, R.A. Carbapenems: Past, Present, and Future. Antimicrob. Agents Chemother. 2011, 55, 4943–4960. [CrossRef] [PubMed] 76. Zurﬂuh, K.; Treier, A.; Schmitt, K.; Stephan, R. Mobile Fosfomycin Resistance Genes in Enterobacteriaceae—An Increasing Threat. Microbiologyopen 2020, 9, e1135. [CrossRef] [PubMed] 77. Tavares-Carreón, F.; Fathy Mohamed, Y.; Andrade, A.; Valvano, M.A. ArnT Proteins That Catalyze the Glycosylation of Lipopolysaccharide Share Common Features with Bacterial N -Oligosaccharyltransferases. Glycobiology 2016, 26, 286–300. [CrossRef] [PubMed] 78. Nielubowicz, G.R.; Mobley, H.L.T. Host–Pathogen Interactions in Urinary Tract Infection. Nat. Rev. Urol. 2010, 7, 430–441. [CrossRef] [PubMed] 79. Coburn, B.; Sekirov, I.; Finlay, B.B. Type III Secretion Systems and Disease. Clin. Microbiol. Rev. 2007, 20, 535–549. [CrossRef] 80. Nunes, P.H.S.; Valiatti, T.B.; Santos, A.C.M.; Nascimento, J.A.D.S.; Santos-Neto, J.F.; Rocchetti, T.T.; Yu, M.C.Z.; Hoﬂing-Lima, A.L.; Gomes, T.A.T. Evaluation of the Pathogenic Potential of Escherichia Coli Strains Isolated from Eye Infections. Microorganisms 2022, 10, 1084. [CrossRef] 81. Vandenesch, F.; Lina, G.; Henry, T. Staphylococcus Aureus Hemolysins, Bi-Component Leukocidins, and Cytolytic Peptides: A Redundant Arsenal of Membrane-Damaging Virulence Factors? Front. Cell. Infect. Microbiol. 2012, 2, 12. [CrossRef] 82. Nikaido, H.; Pagès, J.-M. Broad-Speciﬁcity Efﬂux Pumps and Their Role in Multidrug Resistance of Gram-Negative Bacteria. FEMS Microbiol. Rev. 2012, 36, 340–363. [CrossRef] 83. Li, X.Z.; Plésiat, P.; Nikaido, H. The Challenge of Efﬂux-Mediated Antibiotic Resistance in Gram-Negative Bacteria. Clin. Microbiol. Rev. 2015, 28, 337–418. [CrossRef] 84. Poole, K. Stress Responses as Determinants of Antimicrobial Resistance in Gram-Negative Bacteria. Trends Microbiol. 2012, 85. 20, 227–234. [CrossRef] [PubMed] Sadler, M.; Mormile, M.R.; Frank, R.L. Characterization of the IS200/IS605 Insertion Sequence Family in Halanaerobium Hydrogeniformans. Genes 2020, 11, 484. [CrossRef] [PubMed] 86. Hong, Y.; Morcilla, V.A.; Liu, M.A.; Russell, E.L.M.; Reeves, P.R. Three Wzy Polymerases Are Speciﬁc for Particular Forms of an Internal Linkage in Otherwise Identical O Units. Microbiology 2015, 161, 1639–1647. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_biom13060952	Article Ca2+ Inﬂux through TRPC Channels Is Regulated by Homocysteine–Copper Complexes Gui-Lan Chen 1, Bo Zeng 1 and Shang-Zhong Xu 1,2,* , Hongni Jiang 1, Nikoleta Daskoulidou 1 , Rahul Saurabh 1, Rumbidzai J. Chitando 1 1 Centre for Atherothrombosis and Metabolic Disease, Hull York Medical School, University of Hull, Hull HU6 7RX, UK; [email protected] (G.-L.C.); [email protected] (B.Z.) 2 Diabetes, Endocrinology and Metabolism, Hull York Medical School, University of Hull, Hull HU6 7RX, UK * Correspondence: [email protected]; Tel.: +44-1482-465372 Abstract: An elevated level of circulating homocysteine (Hcy) has been regarded as an independent risk factor for cardiovascular disease; however, the clinical beneﬁt of Hcy lowering-therapy is not satisfying. To explore potential unrevealed mechanisms, we investigated the roles of Ca2+ inﬂux through TRPC channels and regulation by Hcy–copper complexes. Using primary cultured human aortic endothelial cells and HEK-293 T-REx cells with inducible TRPC gene expression, we found that Hcy increased the Ca2+ inﬂux in vascular endothelial cells through the activation of TRPC4 and TRPC5. The activity of TRPC4 and TRPC5 was regulated by extracellular divalent copper (Cu2+) and Hcy. Hcy prevented channel activation by divalent copper, but monovalent copper (Cu+) had no effect on the TRPC channels. The glutamic acids (E542/E543) and the cysteine residue (C554) in the extracellular pore region of the TRPC4 channel mediated the effect of Hcy–copper complexes. The interaction of Hcy–copper signiﬁcantly regulated endothelial proliferation, migration, and angiogenesis. Our results suggest that Hcy–copper complexes function as a new pair of endogenous regulators for TRPC channel activity. This ﬁnding gives a new understanding of the pathogenesis of hyperhomocysteinemia and may explain the unsatisfying clinical outcome of Hcy-lowering therapy and the potential beneﬁt of copper-chelating therapy. Keywords: homocysteine; calcium channel; TRPC; TRPM2; copper; endothelial cells; angiogenesis; 2-aminoethoxydiphenyl borate 1. Introduction Cardiovascular disease (CVD) is the leading cause of death in developed nations and is increasing rapidly in developing countries. The well-described risk factors include high blood pressure, dyslipidemia, smoking, diabetes mellitus, obesity, and new independent risk factors, such as C-reactive protein, lipoprotein (a), ﬁbrinogen, and homocysteine (Hcy). The association between elevated Hcy levels and atherosclerosis was ﬁrst demonstrated in patients with hyperhomocysteinemia in 1969 [1]; however, the importance of Hcy as a risk factor has been especially acknowledged during the last two decades in that even a mild or moderate increase in Hcy level (>15 µmol/L) in serum or plasma is closely associated with the morbidity and mortality of coronary heart diseases [2–6], stroke [7,8], peripheral vascular disease [9], venous thrombosis [10], dementia or Alzheimer’s disease [11], nerve degeneration [12], diabetes [13], osteoporotic fractures [14], end-stage renal disease [15], and other conditions, such as adverse pregnancy outcome (early abortion, placental vas- culopathy, and birth defects) [16] and liver ﬁbrosis [17]. In patients with genetic enzyme defects including cystathionine β-synthase (CBS), methylenetetrahydrofolate reductase (MTHFR), and methionine synthase (MS) in the Hcy metabolic pathway, the concentration of Hcy is much higher and accompanied with more severe cardiovascular damage [8,18]. The MTHFR (T677C point mutation) variant is the most common enzyme defect associated Citation: Chen, G.-L.; Zeng, B.; Jiang, H.; Daskoulidou, N.; Saurabh, R.; Chitando, R.J.; Xu, S.-Z. Ca2+ Inﬂux through TRPC Channels Is Regulated by Homocysteine–Copper Complexes. Biomolecules 2023, 13, 952. https://doi.org/10.3390/ biom13060952 Academic Editor: Fabrice Antigny Received: 25 April 2023 Revised: 15 May 2023 Accepted: 17 May 2023 Published: 6 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Biomolecules 2023, 13, 952. https://doi.org/10.3390/biom13060952 https://www.mdpi.com/journal/biomolecules biomoleculesBiomolecules 2023, 13, 952 2 of 16 with high Hcy and its prevalence is 5~15% in Caucasian and Asian populations. The mech- anisms of how Hcy causes diseases or becomes a risk for diseases are still unknown [19,20]; in particular, the intervention for lowering plasma Hcy levels in patients did not show any preventive effects against cardiovascular diseases [21,22], suggesting unrecognised mechanisms or interactions with Hcy may exist in vivo. Since Hcy is involved in the patho- genesis of many diseases and is associated with all-cause mortality [23], it is reasonable to hypothesise that Hcy may target some ubiquitously expressed proteins or key signalling molecules in the body. Calcium is a key signalling messenger in the cell and several studies have suggested that Hcy may interfere with Ca2+ signalling pathways. For example, Ca2+ inﬂux and intracellular Ca2+ release were enhanced by Hcy [24], and the ligand-gated Ca2+ channel NMDA receptor was stimulated by Hcy [25]. Interestingly, it has been shown that the up-regulation of Ca2+ permeable channels, such as TRPC1 and TRPC5, is related to vascular neointimal growth and cell mobility [26,27], while neointimal growth was also observed in the blood vessels from patients with hyperhomocysteinemia [1]. TRPC channels are ubiqui- tously expressed in the cardiovascular system and mediate the common pathway of Ca2+ entry via G-protein coupled receptor activation and/or the depletion of the endoplasmic reticulum (ER) Ca2+ store [28,29]. Therefore, we hypothesised that TRPC channels could be involved in the pathophysiology of hyperhomocysteinemia. On the other hand, the correlation between Hcy and copper in cardiovascular disease has been demonstrated in clinical surveys [30–33], and copper-lowering therapy with a chelator could be beneﬁcial for cardiac hypertrophy [34]. We, therefore, aimed to investigate the effects of Hcy on TRPC channels and its regulatory mechanisms with copper ions in causing endothelial dysfunction and subsequent atherogenicity. 2. Materials and Methods 2.1. Cell Culture and Transfection Human TRPC4α (NM_016179), TRPC4β1 (NM_001135955, but the β1 isoform was cloned from the endothelial cell with one glutamic acid deletion at E785), and TRPC5 (AF054568) in the tetracycline-regulatory vector pcDNA4/TO (Invitrogen, Paisley, UK) were transfected into HEK-293 T-REx cells using the LipofectamineTM 2000 transfection reagent (Invitrogen, Paisley, UK). TRPC4 was tagged with an enhanced yellow ﬂuorescent protein (EYFP) at the N-terminus. Expression was induced by 1 µg·mL−1 tetracycline for 48–72 h before recording. The non-induced cells without the addition of tetracycline were used as a control. Cells were grown in DMEM-F12 medium (Invitrogen, Paisley, UK) containing 10% foetal calf serum (FCS), 100 units·mL−1 penicillin, and 100 µg·mL−1 streptomycin at 37 ◦C under 95% air and 5% CO2. Cells were seeded on coverslips prior to experiments. Human aortic endothelial cells (HAECs) were purchased from PromoCell (Heidel- berg, Germany) and cultured in an endothelial cell growth medium as we described previously [35,36]. The medium was supplemented with 2% foetal calf serum, 5.0 µg·L−1 epidermal growth factor, 0.5 µg·L−1 vascular endothelial growth factor, 10 µg·L−1 basic ﬁbroblast factor, 20 µg·L−1 R3 IGF-1, and 22.5 mg·L−1 heparin. Cells in passages 2 to 4 were used in the experiment to avoid age-dependent variations. 2.2. Electrophysiological Recordings and Ca2+ Measurements A whole-cell clamp was performed at room temperature (23–26 ◦C) as described before [37,38]. Brieﬂy, the signal was ampliﬁed with an Axopatch B200 ampliﬁer and controlled with pClamp software 10. A 1 s ramp voltage protocol from −100 mV to +100 mV was applied at a frequency of 0.2 Hz from a holding potential of 0 mV. Signals were sampled at 3 kHz and ﬁltered at 1 kHz. A glass microelectrode with a resistance of 3–5 MΩ was used. The 200 nM Ca2+ buffered pipette solution contained 115 CsCl, 10 EGTA, 2 MgCl2, 10 HEPES, and 5.7 CaCl2 in mM. The pH was adjusted to 7.2 with CsOH and the osmolarity was adjusted to ~290 mOsm with mannitol. The calculated free Ca2+ was Biomolecules 2023, 13, 952 3 of 16 200 nM using EQCAL (Biosoft, Cambridge, UK). The standard bath solution contained (mM): 130 NaCl, 5 KCl, 8 D-glucose, 10 HEPES, 1.2 MgCl2, and 1.5 CaCl2. The pH was adjusted to 7.4 with NaOH. For excised patch recordings, the procedures were similar to our previous reports [39,40]. Intracellular Ca2+ was measured using a cuvette-based system as we described previously [35,41]. Brieﬂy, HAECs were loaded with Fluo3-AM (5 µM) in a Ca2+ free standard bath solution (130 NaCl, 5 KCl, 8 D-glucose, 10 HEPES, and 1.2 MgCl2 in mM), then washed and resuspended in the standard bath solution. A total volume of 2 mL of standard bath solution with suspended cells was pipetted into a cuvette and the ﬂuores- cence was measured using a Perkin–Elmer LS50B ﬂuorimeter. All electrophysiological recordings and Ca2+ measurements were performed at room temperature (25 ◦C). 2.3. RT-PCR Total RNA was extracted from the cultured endothelial cells using TRI Reagent (Sigma- Aldrich, Poole, UK) and reverse transcribed with the Moloney murine leukaemia virus (M-MLV) reverse transcriptase using random primers (Promega, Southampton, UK). The PCR primer sequences used in this study and the detailed procedures were described in our previous report [42]. PCR products were conﬁrmed by 2% agarose gel electrophoresis or direct sequencing. 2.4. Cell Proliferation, Migration, and Angiogenesis Assays Endothelial cells were grown to conﬂuence in 24-well plates in an endothelial cell medium. Cell proliferation was assayed by a WST-1 kit (Roche) as we reported [42,43]. For the cell migration assay, a linear scrape of ~0.3 mm width was made through a pipette tip [26]. The cells were cultured in an endothelial cell medium with or without Hcy. After 24 h of culture, the cells were ﬁxed with 4% paraformaldehyde, and cells across the edge of the wound were counted. For the angiogenesis experiment, bovine skin collagen (Sigma, Hertfordshire, UK) was diluted to 1.5 mg/mL with extracellular matrix (ECM) (Sigma) at 2–8 ◦C as a working solution. The pH and osmolarity were adjusted by 1 M NaOH and 10× phosphate-buffered saline, respectively. Human vascular endothelial growth factor (Sigma, UK) was added to a ﬁnal concentration of 20 ng/mL. Collagen working solution at a volume of 120 µL was added to each well of a 48-well plate and allowed to gelatinise for 30 min at 37 ◦C. EA.hy926 cells were resuspended in the ECM solution and added to each well at a volume of 300 µL (~3 × 104 cells/well) and incubated at 37 ◦C for 30 min under 95% air and 5% CO2. After 24 h of culture with Hcy or the vehicle, cells were ﬁxed with 4% paraformaldehyde, stained with 0.025% crystal violet, and photographed. The angiogenesis score was calculated by a semi-quantitative method as reported previously [44]. The BD MatrigelTM (BD Bioscience, Chester, UK) was also used to see the effects of Hcy and Cu2+ on endothelial cell tube formation. The angiogenesis was analysed with Wim Tube software (Wimasis, Munich, Germany). 2.5. Reagents and Drugs All general salts and reagents were purchased from Sigma-Aldrich (Poole, UK). L- homocysteine, lanthanum chloride (La3+), CuSO4 (Cu2+), gadolinium chloride (Gd3+), 2-aminoethoxydiphenyl borate (2-APB), trypsin, thapsigargin (TG), D-(−)-2-amino-5- phosphonopentanoic acid (D-AP5), verapamil, A23187, (1,10-phenanthroline)bis (triphenylphosphine)copper(I) nitrate dichloromethane adduct, and foetal calf serum were purchased from Sigma-Aldrich. Matrigel was purchased from BD Biosciences (UK) and Fluo-3 AM from Invitrogen (Paisley, UK). Fluo-3 AM (5 mM), TG (1 mM), and 2-APB (100 mM) were made up as stock solutions in 100% dimethyl sulphoxide (DMSO). 2.6. Statistics Data are expressed as mean ± s.e.m. where n is the cell number for electrophysiological recordings and Ca2+ imaging. Data sets were compared using a paired t-test for the results Biomolecules 2023, 13, 952 4 of 16 before and after treatment, or the ANOVA Bonferroni’s post-hoc analysis for comparing more than two groups with signiﬁcance indicated if p < 0.05. 3. Results 3.1. Ca2+ Inﬂux Induced by Hcy in HAECs The effect of Hcy on Ca2+ inﬂux was measured in the primary cultured HAECs using Fluo-3 AM Ca2+ dye. Hcy at 1–100 µM increased the intracellular [Ca2+]i, which accounted for 33.1 ± 1.1% of the amplitude of the Ca2+ signal induced by calcium ionophore A231872 (Figure 1A,B). Blocking the voltage-gated Ca2+ channels with verapamil or using 100 mM K+ in the bath solution (equal molar substitution of Na+) to clamp the membrane potential did not prevent the effect of Hcy (Figure 1C,D), suggesting that Hcy-induced Ca2+ increase is mediated by non-voltage gated Ca2+-permeable channels. We also examined the Ca2+ release using the sarco/endoplasmic reticulum Ca2+-ATPase (SERCA) inhibitor thapsigargin (TG). Depletion of the ER Ca2+ store showed no signiﬁcant blocking effect on Hcy-induced intracellular Ca2+ increase (Figure 1E). Hcy has been reported to induce Ca2+ transient through NMDA receptor activation in cultured neurons [24], therefore, we tested the effect of Hcy in cells treated with the NMDA antagonist D-(−)-2-amino-5- phosphonopentanoic acid (D-AP5). D-AP5 at 50 µM was unable to prevent the Hcy-induced Ca2+ inﬂux (Figure 1F), suggesting that other Ca2+ entry pathways exist in endothelial cells. These results suggest that Hcy increases Ca2+ inﬂux mainly through non-voltage gated channels, rather than the Ca2+ release or NMDA receptors in vascular endothelial cells. Figure 1. Effect of Hcy on Ca2+ inﬂux in HAECs. Ca2+ inﬂux was measured using Fluo-3 AM. (A) Example of Hcy on Ca2+ inﬂux. Hcy was added accumulatedly and followed by calcium ionophore A23187 (2 µM). (B) The mean ± s.e.m. for the effect of Hcy. (C) Effect of Hcy under the bath solution with 100 mM K+. (D) Response to Hcy after blocking the voltage-gated Ca2+ channel with 10 µM verapamil. (E) Thapsigargin (2 µM) was added to block the SERCA. (F) NMDA antagonist 5-AP (50 µM) added. The ANOVA test was used and n = 6–8 for each experiment. * p < 0.001. Biomolecules 2023, 13, x FOR PEER REVIEW 4 of 17 2.6. Statistics Data are expressed as mean ± s.e.m. where n is the cell number for electrophysiolog-ical recordings and Ca2+ imaging. Data sets were compared using a paired t-test for the results before and after treatment, or the ANOVA Bonferroni’s post-hoc analysis for com-paring more than two groups with significance indicated if p < 0.05. 3. Results 3.1. Ca2+ Influx Induced by Hcy in HAECs The effect of Hcy on Ca2+ influx was measured in the primary cultured HAECs using Fluo-3 AM Ca2+ dye. Hcy at 1–100 µM increased the intracellular [Ca2+]i, which accounted for 33.1 ± 1.1% of the amplitude of the Ca2+ signal induced by calcium ionophore A231872 (Figure 1A,B). Blocking the voltage-gated Ca2+ channels with verapamil or using 100 mM K+ in the bath solution (equal molar substitution of Na+) to clamp the membrane potential did not prevent the effect of Hcy (Figure 1C,D), suggesting that Hcy-induced Ca2+ increase is mediated by non-voltage gated Ca2+-permeable channels. We also examined the Ca2+ release using the sarco/endoplasmic reticulum Ca2⁺-ATPase (SERCA) inhibitor thapsigar-gin (TG). Depletion of the ER Ca2+ store showed no significant blocking effect on Hcy-induced intracellular Ca2+ increase (Figure 1E). Hcy has been reported to induce Ca2+ tran-sient through NMDA receptor activation in cultured neurons [24], therefore, we tested the effect of Hcy in cells treated with the NMDA antagonist D-(−)-2-amino-5-phosphono-pentanoic acid (D-AP5). D-AP5 at 50 µM was unable to prevent the Hcy-induced Ca2+ influx (Figure 1F), suggesting that other Ca2+ entry pathways exist in endothelial cells. These results suggest that Hcy increases Ca2+ influx mainly through non-voltage gated channels, rather than the Ca2+ release or NMDA receptors in vascular endothelial cells. Figure 1. Effect of Hcy on Ca2+ influx in HAECs. Ca2+ influx was measured using Fluo-3 AM. (A) Example of Hcy on Ca2+ influx. Hcy was added accumulatedly and followed by calcium ionophore Biomolecules 2023, 13, 952 5 of 16 3.2. Hcy-Induced Ca2+ Inﬂux through TRPC4 and TRPC5 Channels To explore which pathway is involved in Hcy-induced Ca2+ entry, we examined the expression and function of TRPC channels in endothelial cells. The mRNAs of TRPC1, 3, 4, and 6 were detected in the HAECs using RT-PCR. TRPC1 and TRPC4 were more abundant in HUVEC, but TRPC5 was low and TRPC3, TRPC6, and TRPC7 seemed to be absent in HUVEC (Figure 2A). The spliced isoforms of TRPC1E9del, TRPC4β1, and TRPC4ε1 were also identiﬁed in the HAECs using the primer sets we reported previously [42] (Figure 2B). Figure 2. Hcy-induced Ca2+ inﬂux through TRPC4 and TRPC5 channels in endothelial cells. (A) mRNAs of TRPCs in vascular endothelial cells (HAECs and HUVECs). The plasmid cDNAs for TRPC3, 6, and 7 were used as positive controls. (B) Detection of TRPC1 and TRPC4 spliced variants in HAECs. The PCR primers and the corresponding size of amplicons were given in our previous reports [42]. (C) TRPC4 current recorded in HEK293 T-REx cells inducibly overexpressing TRPC4α channels and the effect of Hcy (100 µM). (D) Current for induced TRPC5 cells. (E) Non-induced T-REx cell as control. (F) The mean ± s.e.m. measured at ±80 mV after exposure to each compound. n = 5–6 for each group. * p < 0.001 compared with La3+ treatment measured at ±80 mV. Using whole-cell patch recordings, the effects of Hcy on TRPC4 and TRPC5 currents were examined in the HEK293 T-REx cells inducibly expressing TRPC channels [38]. Lan- thanides (La3+ or Gd3+) were used as channel activators in our experiment as we used before [41,45]. After perfusion with Hcy, the currents of TRPC4 and TRPC5 were signiﬁ- cantly stimulated (Figure 2C,D) while no effects were observed on the non-induced cells (Figure 2E,F), suggesting that Hcy induced Ca2+ inﬂux via the activation of TRPC4 and TRPC5 channels. Biomolecules 2023, 13, x FOR PEER REVIEW 5 of 17 A23187 (2 µM). (B) The mean ± s.e.m. for the effect of Hcy. (C) Effect of Hcy under the bath solution with 100 mM K+. (D) Response to Hcy after blocking the voltage-gated Ca2+ channel with 10 µM verapamil. (E) Thapsigargin (2 µM) was added to block the SERCA. (F) NMDA antagonist 5-AP (50 µM) added. The ANOVA test was used and n = 6–8 for each experiment. * p < 0.001. 3.2. Hcy-Induced Ca2+ Influx through TRPC4 and TRPC5 Channels To explore which pathway is involved in Hcy-induced Ca2+ entry, we examined the expression and function of TRPC channels in endothelial cells. The mRNAs of TRPC1, 3, 4, and 6 were detected in the HAECs using RT-PCR. TRPC1 and TRPC4 were more abun-dant in HUVEC, but TRPC5 was low and TRPC3, TRPC6, and TRPC7 seemed to be absent in HUVEC (Figure 2A). The spliced isoforms of TRPC1E9del, TRPC4β1, and TRPC4Ɛ1 were also identified in the HAECs using the primer sets we reported previously [42] (Figure 2B). Using whole-cell patch recordings, the effects of Hcy on TRPC4 and TRPC5 currents were examined in the HEK293 T-REx cells inducibly expressing TRPC channels [38]. Lan-thanides (La3+ or Gd3+) were used as channel activators in our experiment as we used be-fore [41,45]. After perfusion with Hcy, the currents of TRPC4 and TRPC5 were signifi-cantly stimulated (Figure 2C,D) while no effects were observed on the non-induced cells (Figure 2E,F), suggesting that Hcy induced Ca2+ influx via the activation of TRPC4 and TRPC5 channels. Figure 2. Hcy-induced Ca2+ influx through TRPC4 and TRPC5 channels in endothelial cells. (A) mRNAs of TRPCs in vascular endothelial cells (HAECs and HUVECs). The plasmid cDNAs for TRPC3, 6, and 7 were used as positive controls. (B) Detection of TRPC1 and TRPC4 spliced variants Biomolecules 2023, 13, 952 6 of 16 3.3. Activation of TRPC4 and TRPC5 by Divalent Cu2+ and the Interference by Hcy Hcy and copper are two important regulators of cellular oxidative stress and both are involved in atherogenicity, however, their mechanisms are unclear [30]. We found that divalent Cu2+ showed an initial transient inhibition and then a gradual activation of TRPC4α and TRPC5 currents after perfusion with 10 µM Cu2+ (Figure 3A,B). The current of TRPC4β1 was also activated by Cu2+ (Figure S1A). The EC50 of Cu2+ for TRPC4α channel activation was 6.8 µM (Figure S1B). The Cu2+-induced currents were also sensitive to the non-selective TRPC blocker 2-APB as the currents of TRPC4 and TRPC5 induced by lanthanides [41,45]. Interestingly, perfusion with Hcy (100 µM) completely prevented the TRPC4 and TRPC5 channel activation by Cu2+ (Figure 3C,D), suggesting that the interaction of Hcy and copper is critical for regulating TRPC channel activity. We also examined the interaction on TRPM2 channels, since the channel is expressed in endothelial cells and inhibited by Cu2+ [35,46]. Hcy had no signiﬁcant effect on TRPM2, but it prevented the inhibitory effect of Cu2+ (Figure S2). These data indicate that the complexes of Hcy–copper or the charge of copper ions may be the determinant for their effects on ion channels. Figure 3. TRPC channel activated by Cu2+ and counteracted by Hcy. (A,B) Representative time course and IV curve for TRPC4 and TRPC5 activated by Cu2+. 2-APB (100 µM) as a control channel blocker. (C,D) TRPC4 and TRPC5 currents after perfusion with 100 µM Hcy, the addition of 10 µM Cu2+, and the washout of Hcy. (E) The mean ± s.e.m. data for the effect of Cu2+ (n = 6–8. * p < 0.01). (F) The mean ± s.e.m. data for Hcy plus Cu2+ (n = 5–6. * p < 0.001). 3.4. No Effect of Monovalent Cu+ on TRPC Channel To test the role of copper ion charges, we examined the effects of monovalent copper (I) compounds. As shown in Figure 4, the copper (I), (1,10-phenanthroline)bis(triphenylphosphine) copper (I) nitrate dichloromethane adduct, had no effect on TRPC4α and TRPC5 chan- nel activity, but the divalent Cu2+ activated them (Figure 4A–C). Similarly, no effects of the monovalent copper, copper (I) 1-butanethiolate), and copper (I) tetrakis(acetonitrile) copper(I) tetraﬂuoroborate) were observed on TRPC4α channels (Figure S3). These data suggest that the divalent copper ions are essential for TRPC channel activation, but there Biomolecules 2023, 13, x FOR PEER REVIEW 6 of 17 in HAECs. The PCR primers and the corresponding size of amplicons were given in our previous reports [42]. (C) TRPC4 current recorded in HEK293 T-REx cells inducibly overexpressing TRPC4α channels and the effect of Hcy (100 µM). (D) Current for induced TRPC5 cells. (E) Non-induced T-REx cell as control. (F) The mean ± s.e.m. measured at ±80 mV after exposure to each compound. n = 5–6 for each group. * p < 0.001 compared with La3+ treatment measured at ±80 mV. 3.3. Activation of TRPC4 and TRPC5 by Divalent Cu2+ and the Interference by Hcy Hcy and copper are two important regulators of cellular oxidative stress and both are involved in atherogenicity, however, their mechanisms are unclear [30]. We found that divalent Cu2+ showed an initial transient inhibition and then a gradual activation of TRPC4α and TRPC5 currents after perfusion with 10 µM Cu2+ (Figure 3A,B). The current of TRPC4β1 was also activated by Cu2+ (Figure S1A). The EC50 of Cu2+ for TRPC4α channel activation was 6.8 µM (Figure S1B). The Cu2+-induced currents were also sensitive to the non-selective TRPC blocker 2-APB as the currents of TRPC4 and TRPC5 induced by lan-thanides [41,45]. Interestingly, perfusion with Hcy (100 µM) completely prevented the TRPC4 and TRPC5 channel activation by Cu2+ (Figure 3C,D), suggesting that the interac-tion of Hcy and copper is critical for regulating TRPC channel activity. We also examined the interaction on TRPM2 channels, since the channel is expressed in endothelial cells and inhibited by Cu2+[35,46]. Hcy had no significant effect on TRPM2, but it prevented the inhibitory effect of Cu2+ (Figure S2). These data indicate that the complexes of Hcy–copper or the charge of copper ions may be the determinant for their effects on ion channels. Figure 3. TRPC channel activated by Cu2+ and counteracted by Hcy. (A,B) Representative time course and IV curve for TRPC4 and TRPC5 activated by Cu2+. 2-APB (100 µM) as a control channel blocker. (C,D) TRPC4 and TRPC5 currents after perfusion with 100 µM Hcy, the addition of 10 µM Biomolecules 2023, 13, 952 7 of 16 are no effects for monovalent Cu+ ions. In addition, Se2+ with antioxidant properties had no effect on TRPC4α channels (Figure 4D–F), suggesting that the TRPC channel has metal ion speciﬁcity. The conversion from divalent to monovalent copper ions under oxidative stress conditions could be an important part of endogenous regulators for TRPC4 and TRPC5 channel activity. Figure 4. Monovalent copper (Cu+) had no effect on TRPC channels. (A) TRPC4 cells were perfused with 10 µM monovalent copper ((1, 10-phenanthroline), bis (triphenylphosphine) copper (I) nitrate dichloromethane adduct), and then 10 µM divalent Cu2+. (B) Similar to (A) but TRPC5 cells were used. (C) The mean ± s.e.m. data measured at ± 80 mV after perfusion with Cu+ and Cu2+. n = 5–7 for each group, p < 0.01 and * p < 0.001. (D) Effect of sodium selenite on TRPC4 current. (E) IV curves for (D). (F) The mean ± s.e.m. data for the effect of Se2+ and Cu2+ on TRPC4 current. 3.5. Extracellular Activation of Cu2+ on TRPC4 and 5 Channels Whole-cell patch recordings were performed using a pipette solution containing 10 µM Cu2+. The activation of the TRPC4 current by the intracellular Cu2+ application did not happen after the whole-cell conﬁguration was formed for more than 5 min; however, bath perfusion with 10 µM Cu2+ signiﬁcantly activated the current of TRPC4α with typical IV curves (Figure 5A). A similar effect on TRPC5 was observed (Figure 5B). We also performed outside-out excised membrane patches and the stimulating effects on TRPC4 and TRPC5 currents by Cu2+ were signiﬁcant after the external surface exposure to Cu2+ by bath perfu- sion (Figure 5C,D). These data suggest that the action site for Cu2+ is extracellularly located. Biomolecules 2023, 13, x FOR PEER REVIEW 7 of 17 Cu2+, and the washout of Hcy. (E) The mean ± s.e.m. data for the effect of Cu2+ (n = 6–8. * p < 0.01). (F) The mean ± s.e.m. data for Hcy plus Cu2+ (n = 5–6. * p < 0.001). 3.4. No Effect of Monovalent Cu+ on TRPC Channel To test the role of copper ion charges, we examined the effects of monovalent copper (I) compounds. As shown in Figure 4, the copper (I), (1,10-phenanthroline)bis(tri-phenylphosphine) copper (I) nitrate dichloromethane adduct, had no effect on TRPC4α and TRPC5 channel activity, but the divalent Cu2+ activated them (Figure 4A–C). Similarly, no effects of the monovalent copper, copper (I) 1-butanethiolate), and copper (I) tetrakis(acetonitrile) copper(I) tetrafluoroborate) were observed on TRPC4α channels (Figure S3). These data suggest that the divalent copper ions are essential for TRPC chan-nel activation, but there are no effects for monovalent Cu+ ions. In addition, Se2+ with an-tioxidant properties had no effect on TRPC4α channels (Figure 4D–F), suggesting that the TRPC channel has metal ion specificity. The conversion from divalent to monovalent cop-per ions under oxidative stress conditions could be an important part of endogenous reg-ulators for TRPC4 and TRPC5 channel activity. Figure 4. Monovalent copper (Cu+) had no effect on TRPC channels. (A) TRPC4 cells were perfused with 10 µM monovalent copper ((1, 10-phenanthroline), bis (triphenylphosphine) copper (I) nitrate dichloromethane adduct), and then 10 µM divalent Cu2+. (B) Similar to (A) but TRPC5 cells were used. (C) The mean ± s.e.m. data measured at ± 80 mV after perfusion with Cu+ and Cu2+. n = 5–7 for each group, p < 0.01 and * p < 0.001. (D) Effect of sodium selenite on TRPC4 current. (E) IV curves for (D). (F) The mean ± s.e.m. data for the effect of Se2+ and Cu2+ on TRPC4 current. Biomolecules 2023, 13, 952 8 of 16 Figure 5. Extracellular effect of Cu2+ on TRPC4 and TRPC5 channels. (A) A whole-cell patch was recorded in the HEK293 T-REx cells overexpressing TRPC4α with a pipette solution containing 10 µM Cu2+ (n = 4 for each group). (B) Same as (A) but cells overexpressing TRPC5 cells were used. (C) Example of outside-out patches showing the effect of Cu2+ on TRPC4α. (D) Outside-out patches for TRPC5 channels. (E) The mean ± s.e.m. for (A) and (B) (n = 4). (F) The mean ± s.e.m. for (C,D) (n = 4). * p < 0.05, p < 0.01, and * p < 0.001. 3.6. Amino acid Residues of TRPC4 Involved in Copper Activation To identify the action site of channel activation by Cu2+, we substituted the negatively charged glutamic acids (E) at the position of E542, E543, and E555 with the uncharged amino acid glutamine (Q); the cysteine (C554) with tryptophan (W); and the positively charged lysine (K) with the negatively charged glutamic acid (E) in the putative extracellular loops between the S5 and S6 domain of TRPC4α (Figure 6). The mutants of E542Q/E543Q, E555Q, C554W, and K556E did not affect the membrane trafﬁcking of the channel proteins; however, the mutants of E542Q/E543Q and C554W caused resistance to Cu2+, but these mutants did not alter the sensitivity to trypsin, since trypsin is assumed to be an intracellular signalling process through GPCR activation (Figure 6). The mutants E555Q and K556E did not signiﬁcantly change the effect of copper activation. These data indicate that negatively charged glutamic acids and the cysteine residue in the third extracellular loop are functional targets for divalent copper. Biomolecules 2023, 13, x FOR PEER REVIEW 8 of 17 3.5. Extracellular Activation of Cu2+ on TRPC4 and 5 Channels Whole-cell patch recordings were performed using a pipette solution containing 10 µM Cu2+. The activation of the TRPC4 current by the intracellular Cu2+ application did not happen after the whole-cell configuration was formed for more than 5 min; however, bath perfusion with 10 µM Cu2+ significantly activated the current of TRPC4α with typical IV curves (Figure 5A). A similar effect on TRPC5 was observed (Figure 5B). We also per-formed outside-out excised membrane patches and the stimulating effects on TRPC4 and TRPC5 currents by Cu2+ were significant after the external surface exposure to Cu2+ by bath perfusion (Figure 5C,D). These data suggest that the action site for Cu2+ is extracellu-larly located. Figure 5. Extracellular effect of Cu2+ on TRPC4 and TRPC5 channels. (A) A whole-cell patch was recorded in the HEK293 T-REx cells overexpressing TRPC4α with a pipette solution containing 10 µM Cu2+ (n = 4 for each group). (B) Same as (A) but cells overexpressing TRPC5 cells were used. (C) Example of outside-out patches showing the effect of Cu2+ on TRPC4α. (D) Outside-out patches for TRPC5 channels. (E) The mean ± s.e.m. for (A) and (B) (n = 4). (F) The mean ± s.e.m. for (C,D) (n = 4). * p < 0.05, p < 0.01, and * p < 0.001. Biomolecules 2023, 13, 952 9 of 16 Figure 6. Identiﬁcation of amino acids involved in channel activation by Cu2+. The mutants of TRPC4α tagged with EYFP were made by site-mutagenesis and membrane localisation was examined using a ﬂuorescent microscope. (A) The double glutamic acid mutants (TRPC4-E542Q/E543Q) showed the loss of channel activation by Cu2+, but the robust current through the mutant channel can also be activated by trypsin (2 nM). (B) The TRPC4-E555Q mutant was activated by Cu2+. (C) Less sensitivity to Cu2+ for the cysteine mutant (TRPC4-C555W). (D) Glysine at the position of 556 substituted with glutamic acid (TRPC4-K556E). (E) Amino acid alignment of the transmembrane region (S5-S6) for TRPCs (red asterisks indicate residues subject to mutagenesis) and the mean ± s.e.m. data showing the amplitude of currents corresponding to the mutants and the wild-type control after perfusion with Cu2+ (n = 8). *** p < 0.001. 3.7. TRPC and Homocysteine-Copper Complexes in the Regulation of Endothelial Cell Proliferation The blocking of TRPC channels has been shown to inhibit cell proliferation by us and others [27,42,47]. Here we further demonstrated the roles of TRPCs in the endothelial cells from macrovasculature. The proliferation of HAECs was signiﬁcantly inhibited by speciﬁc pore-blocking TRPC antibodies (Figure 7A), which was consistent with the nonselective blocker 2-APB (Figure 7B). The over-expression of TRPC1 or TRPC4 promoted proliferation (Figure 7C), suggesting the signiﬁcant contribution of TRPC channel activity to endothelial cell proliferation. However, Hcy inhibited the proliferation of HAECs but increased the proliferation of HUVECs. The pro-proliferative effect was more pronounced in the culture medium omitting cysteine and methionine (Figure 7D), or in the T-REx cells overexpressing Hcy-sensitive TRPC5 channels (Figure S4). On the other hand, divalent Biomolecules 2023, 13, x FOR PEER REVIEW 9 of 17 3.6. Amino acid Residues of TRPC4 Involved in Copper Activation To identify the action site of channel activation by Cu2+, we substituted the negatively charged glutamic acids (E) at the position of E542, E543, and E555 with the uncharged amino acid glutamine (Q); the cysteine (C554) with tryptophan (W); and the positively charged lysine (K) with the negatively charged glutamic acid (E) in the putative extracel-lular loops between the S5 and S6 domain of TRPC4α (Figure 6). The mutants of E542Q/E543Q, E555Q, C554W, and K556E did not affect the membrane trafficking of the channel proteins; however, the mutants of E542Q/E543Q and C554W caused resistance to Cu2+, but these mutants did not alter the sensitivity to trypsin, since trypsin is assumed to be an intracellular signalling process through GPCR activation (Figure 6). The mutants E555Q and K556E did not significantly change the effect of copper activation. These data indicate that negatively charged glutamic acids and the cysteine residue in the third ex-tracellular loop are functional targets for divalent copper. Figure 6. Identification of amino acids involved in channel activation by Cu2+. The mutants of TRPC4α tagged with EYFP were made by site-mutagenesis and membrane localisation was exam-ined using a fluorescent microscope. (A) The double glutamic acid mutants (TRPC4-E542Q/E543Q) Biomolecules 2023, 13, 952 10 of 16 copper had no signiﬁcant effect on the proliferation of HAECs but signiﬁcantly reduced the proliferation of HUVECs and the HUVEC-derived cell line EA.hy926 (Figure 7E). Combined incubation with Hcy and Cu2+ showed inhibitory effects at low concentrations of copper but stimulatory effects at a high concentration (100 µM Cu2+) (Figure 7F), which exhibited signiﬁcant differences from the groups treated with Hcy alone. These data suggest that the sensitivity to Hcy and Cu2+ may rely on the types of vascular endothelial cells and the ratio of Hcy and copper complexes. Figure 7. Endothelial cell proliferation regulated by TRPC channels and the effects of Hcy and copper. Cell proliferation was assayed by a WST-1 kit and absorbance was measured at a wavelength of 450 nm. (A) Endothelial cells were incubated with the pore-blocking TRPC antibodies [28,42,48] for 24 h. The TRPC5 antibody targeting the C-terminal (T5C3) and preimmune serum (Preimmune) were used as controls. (B) 2-APB. (C) HAEC cells transfected with plasmid cDNAs for TRPC1 and TRPC4 using the electroporation method [49]. (D) Effect of Hcy on HAECs and HUVECs. (E) Effect of Cu2+ on HAEC, HUVEC, and the HUVEC-derived cell line Eahy926. (F) Combined effect of Hcy (10 µM) and Cu2+. n = 8 for each group, * p < 0.05, p < 0.01, and * p < 0.001, ##, not signiﬁcant. Biomolecules 2023, 13, x FOR PEER REVIEW 11 of 17 Figure 7. Endothelial cell proliferation regulated by TRPC channels and the effects of Hcy and cop-per. Cell proliferation was assayed by a WST-1 kit and absorbance was measured at a wavelength of 450 nm. (A) Endothelial cells were incubated with the pore-blocking TRPC antibodies [28,42,48] for 24 h. The TRPC5 antibody targeting the C-terminal (T5C3) and preimmune serum (Preimmune) were used as controls. (B) 2-APB. (C) HAEC cells transfected with plasmid cDNAs for TRPC1 and TRPC4 using the electroporation method [49]. (D) Effect of Hcy on HAECs and HUVECs. (E) Effect of Cu2+ on HAEC, HUVEC, and the HUVEC-derived cell line Eahy926. (F) Combined effect of Hcy (10 µM) and Cu2+. n = 8 for each group, * p < 0.05, p < 0.01, and * p < 0.001, ##, not significant. 3.8. Hcy–Copper Complexes in the Regulation of Cell Migration and Angiogenesis TRPC channels are involved in cell migration and angiogenesis [26,50,51], so we ob-served the effects of Hcy and copper on endothelial cell migration and angiogenesis. Us-ing a linear wound assay, the number of migrated cells was seen to be significantly re-duced after treatment with Hcy (Figure 8A,B). Angiogenesis was examined using the ex-tracellular matrix (ECM) gel and Matrigel assays. The score of angiogenesis in the ECM gel and the tube formation in the Matrigel were significantly inhibited by Hcy (Figure 8C–G). However, the addition of Cu2+ in the culture medium alleviated the inhibitory effects of Hcy on endothelial cell tube formation and angiogenesis, suggesting that endothelial Biomolecules 2023, 13, 952 11 of 16 3.8. Hcy–Copper Complexes in the Regulation of Cell Migration and Angiogenesis TRPC channels are involved in cell migration and angiogenesis [26,50,51], so we ob- served the effects of Hcy and copper on endothelial cell migration and angiogenesis. Using a linear wound assay, the number of migrated cells was seen to be signiﬁcantly reduced after treatment with Hcy (Figure 8A,B). Angiogenesis was examined using the extracel- lular matrix (ECM) gel and Matrigel assays. The score of angiogenesis in the ECM gel and the tube formation in the Matrigel were signiﬁcantly inhibited by Hcy (Figure 8C–G). However, the addition of Cu2+ in the culture medium alleviated the inhibitory effects of Hcy on endothelial cell tube formation and angiogenesis, suggesting that endothelial cell mobility and angiogenesis are regulated by the complexes of homocysteine and copper. Taken together, regulation by Hcy and copper complexes via TRPC4/TRPC5 channels could be regarded as a new mechanism to control endothelial function. Figure 8. Endothelial cell migration and angiogenesis regulated by Hcy and Cu2+ complexes. (A) Example of endothelial cell migration using a linear wound assay. (B) Effect of Hcy on cell migration after 24 h of incubation. (C) Example of angiogenesis using ECM gel (i) and Matrigel for HUVEC (ii) and Eahy926 cells (iii). (D) The mean ± s.e.m. showing the effect of Hcy on angiogenesis (n = 40–60 imaging ﬁelds from six cell culture dishes for each group). (E–G) Effect of Hcy (10 µM) and Cu2+ (10 µM) on endothelial cell tube formation. n = 6 for each group. The number of loops, branching, and total length of tubes were analysed by software. * p < 0.001. Biomolecules 2023, 13, x FOR PEER REVIEW 12 of 17 cell mobility and angiogenesis are regulated by the complexes of homocysteine and cop-per. Taken together, regulation by Hcy and copper complexes via TRPC4/TRPC5 channels could be regarded as a new mechanism to control endothelial function. Figure 8. Endothelial cell migration and angiogenesis regulated by Hcy and Cu2+ complexes. (A) Example of endothelial cell migration using a linear wound assay. (B) Effect of Hcy on cell migration after 24 h of incubation. (C) Example of angiogenesis using ECM gel (i) and Matrigel for HUVEC (ii) and Eahy926 cells (iii). (D) The mean ± s.e.m. showing the effect of Hcy on angiogenesis (n = 40–60 imaging fields from six cell culture dishes for each group). (E–G) Effect of Hcy (10 µM) and Cu2+ (10 µM) on endothelial cell tube formation. n = 6 for each group. The number of loops, branching, and total length of tubes were analysed by software. * p < 0.001. Biomolecules 2023, 13, 952 12 of 16 4. Discussion Our data show that Hcy can increase Ca2+ inﬂux in HAECs. The increase is mediated by the opening of TRPC4 and TRPC5 channels. Divalent copper acts as a non-selective activator of TRPC4/5 channels. The channel activation by divalent copper is regulated by Hcy. The charge of copper ions is critical for TRPC channel opening because monovalent copper (I) shows no signiﬁcant effect on TRPC channel activity. We also explored the action site for divalent copper using excised membrane patches and site mutagenesis. The cysteine (C554) and glutamic acids (E542 and E543) in the third extracellular loop of TRPC4α are responsible for copper activation. Moreover, we showed that copper and Hcy are essential regulators for endothelial cell proliferation, migration, and angiogenesis. Divalent copper seems to counteract the effect of Hcy on proliferation and angiogenesis which suggests the importance of the Hcy–copper interaction in causing endothelium dysfunction and atherosclerosis. The regulation of TRPC channels is the sought-after underlying mechanism for the pathogenesis of patients with hyperhomocysteinemia. The effect of Hcy on intracellular [Ca2+]i is still unclear in endothelial cells, although there are several reports showing that Hcy increases the Ca2+ inﬂux in human platelets [52], cultured vascular smooth muscle cells [23], podocytes [53], and neurons [24,54]. Here, we found that Hcy increased the Ca2+ inﬂux in HAECs which is mediated by the activation of TRPC4 and TRPC5. The blocking of voltage-gated Ca2+ channels and NMDA receptors was unable to prevent the Hcy-induced Ca2+ inﬂux, suggesting that the Hcy-induced Ca2+ entry pathway is not through the voltage-gated channel or the ligand-gated NMDA receptor channel in vascular endothelial cells. In addition, the Hcy-induced intracellular Ca2+ increase has been linked to ER calcium release via the homocysteine-inducible ER stress protein [55]; however, Hcy-induced Ca2+ inﬂux also happened in the cells acutely treated with SERCA blocker TG which suggests that main pathways of Ca2+ inﬂux are across the plasma membrane rather than the intracellular Ca2+ release from the ER. The effect of Hcy on store-operated channels or ORAI channels is unknown, but high concentrations (≥100 µM) of Hcy may inhibit the store-operated Ca2+ inﬂux [56]. Hcy also inhibits BKCa and thus depolarises the membrane potential and increases the vascular tone [57]. This action may explain the diverse responses in vascular tone or [Ca2+]i observed in some cell types [58,59]. The N-methyl-D-aspartate (NMDA) receptor activation by Hcy could also be a mechanism for Ca2+ inﬂux in the nervous system [24] but this mechanism may be less signiﬁcant in vascular endothelial cells. Homocysteine contains sulphuric residues so its toxic effect has been attributed to redox homeostasis, such as the production of different reactive oxygen species (ROS), thus leading to the oxidation of low-density lipoprotein [20]. Cellular oxidative stress including ER stress has also been proposed for Hcy pathophysiology [19]; the increased ROS production activates ROS-sensitive Ca2+ channels. In addition, we demonstrated that the TRPC5 channel is a redox-sensitive channel that can be activated by thioredoxin and reducing agents [37] and mercury compounds [41]. Here, we found that TRPC4 and TRPC5 channel activities can be enhanced by Hcy, especially when the channels are opened by lanthanides. TRPM2 is also a redox-sensitive channel; however, Hcy itself had no acute effect on TRPM2 but signiﬁcantly regulated the effect of Cu2+ on the TRPM2 channel. Chronic exposure to Hcy may change gene expressions, through Ca2+ channels and ROS signalling molecules [53,60], but we did not observe such gene expression in this study. The total Hcy level in the blood is determined by both genetic and environmental factors and is typically maintained at a normal range (2–14 µM). Vitamin deﬁciencies, in particular folate acid and vitamins B6 and B12, appear to be the most common causes of elevated Hcy [61]. A supplement of folic acid alone or with vitamin B12 or B6 can help to lower Hcy levels, but it is still uncertain how effective this will be in the prevention of cardiovascular disease or Hcy-related diseases. It has been demonstrated that both Hcy and copper are increased in diseased vessels and diabetic patients; however, the question of how Hcy interacts with copper and causes occlusive diseases remains unanswered. Here, we show for the ﬁrst time that copper can interact with Hcy, controlling TRPC channel Biomolecules 2023, 13, 952 13 of 16 activity, thus changing intracellular Ca2+ signalling, and subsequently the endothelial function. This mechanism gives a new understanding of the two factors in the pathogenesis of cardiovascular diseases. Too low or too high concentrations of copper are detrimental, but we have demonstrated that the charge of copper ions could be more important than the copper concentration. Although treatment with a divalent-copper-selective chelator, triethylenetetramine (TETA), to lower the copper in the body may improve the cardiac structure and function in patients and rats with diabetic cardiomyopathy [34], a more pre- cise clinical trial is needed, especially regarding the charge of copper ions and consideration of the redox environment in the body. The inhibition of TRPCs shows anti-proliferative effects while the activation of TRPC channels shows proliferative effects in vascular endothelial cells, which is consistent with the observations in other cell types [26,42]. However, different types of endothelial cells may show differences, such as the HAECs showing inhibitory characteristics and the HUVECs showing pro-proliferative characteristics. This could be related to the predominance of Hcy-sensitive channels. In patients with hyperhomocysteinemia, neointimal hyperplasia in small vessels is evident [1]. In summary, we revealed a new mechanism of Hcy and copper and their interplay with TRPC channels in endothelial cells. This new concept could be extended to other cell types since many diseases are related to Hcy and copper and Hcy is associated with all-cause mortality. The ﬁndings suggest the importance of copper ion charges in the pathogenesis of vascular disorders, particularly in patients with increased homocysteine levels, and may also provide an alternative explanation for why Hcy-lowering therapy is not very signiﬁcant in clinical trials and how Hcy-copper complexes could be the determinants. Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/biom13060952/s1, Figure S1. Activation of TRPC4β1 by Cu2+ and the dose-response of Cu2+ on TRPC4α; Figure S2. Effect of Hcy and copper on TRPM2 current; Figure S3. Example of monovalent copper (I) 1-butanethiolate on TRPC4α current; and Figure S4. Hcy increased cell proliferation of T-Rex cells overexpressing TRPC5. Author Contributions: Conceptualisation, G.-L.C., B.Z. and S.-Z.X.; methodology, G.-L.C., B.Z., N.D., H.J. and R.J.C.; validation, G.-L.C., B.Z., N.D. and S.-Z.X.; formal analysis, G.-L.C., B.Z., N.D. and R.J.C.; investigation, G.-L.C., B.Z., H.J., N.D., R.S. and R.J.C.; writing—original draft preparation, G.- L.C., B.Z. and S.-Z.X.; writing—review and editing, S.-Z.X.; supervision, S.-Z.X.; funding acquisition, S.-Z.X. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the British Heart Foundation (PG/08/071/25473) (to S.-Z.X.). B.Z. received a Scholarship from the China Scholarship Council. H.J. was supported by a Leverhulme Trust fellowship. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data supporting this study are available from the corresponding authors upon reasonable request. Acknowledgments: We thank Neil Watson and Sahar Avazzadeh for their technical assistance. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. McCully, K.S. Vascular pathology of homocysteinemia: Implications for the pathogenesis of arteriosclerosis. Am. J. Pathol. 1969, 2. 3. 56, 111–128. [PubMed] Jensen, M.K.; Bertoia, M.L.; Cahill, L.E.; Agarwal, I.; Rimm, E.B.; Mukamal, K.J. Novel metabolic biomarkers of cardiovascular disease. Nat. Rev. Endocrinol. 2014, 10, 659–672. [CrossRef] [PubMed] Chen, S.-C.; Su, H.-M.; Chang, J.-M.; Liu, W.-C.; Tsai, J.-C.; Tsai, Y.-C.; Lin, M.-Y.; Hwang, S.-J.; Chen, H.-C. Increasing prevalence of peripheral artery occlusive disease in hemodialysis patients: A 2-year follow-up. Am. J. Med. Sci. 2012, 343, 440–445. [CrossRef] [PubMed] Biomolecules 2023, 13, 952 14 of 16 4. 5. 6. 7. 8. 9. Schaffer, A.; Verdoia, M.; Cassetti, E.; Marino, P.; Suryapranata, H.; De Luca, G.; Novara Atherosclerosis Study Group (NAS). Relationship between homocysteine and coronary artery disease. Results from a large prospective cohort study. Thromb. Res. 2014, 134, 288–293. [CrossRef] Zylberstein, D.E.; Bengtsson, C.; Björkelund, C.; Landaas, S.; Sundh, V.; Thelle, D.; Lissner, L. Serum homocysteine in relation to mortality and morbidity from coronary heart disease: A 24-year follow-up of the population study of women in Gothenburg. Circulation 2004, 109, 601–606. [CrossRef] Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: A meta-analysis. JAMA 2002, 288, 2015–2022. [CrossRef] Zylberstein, D.E.; Skoog, I.; Björkelund, C.; Guo, X.; Hultén, B.; Andreasson, L.-A.; Palmertz, B.; Thelle, D.S.; Lissner, L. Homocysteine levels and lacunar brain infarcts in elderly women: The prospective population study of women in Gothenburg. J. Am. Geriatr. Soc. 2008, 56, 1087–1091. [CrossRef] Casas, J.P.; Bautista, L.E.; Smeeth, L.; Sharma, P.; Hingorani, A.D. Homocysteine and stroke: Evidence on a causal link from mendelian randomisation. Lancet 2005, 365, 224–232. [CrossRef] Kuan, Y.M.; Dear, A.E.; Grigg, M.J. Homocysteine: An aetiological contributor to peripheral vascular arterial disease. ANZ J. Surg. 2002, 72, 668–671. [CrossRef] 10. Den Heijer, M.; Lewington, S.; Clarke, R. Homocysteine, MTHFR and risk of venous thrombosis: A meta-analysis of published epidemiological studies. J. Thromb. Haemost. 2005, 3, 292–299. [CrossRef] 11. Loscalzo, J. Homocysteine and dementias. N. Engl. J. Med. 2002, 346, 466–468. [CrossRef] 12. Rozycka, A.; Jagodzinski, P.P.; Kozubski, W.; Lianeri, M.; Dorszewska, J. Homocysteine Level and Mechanisms of Injury in Parkinson’s Disease as Related to MTHFR, MTR, and MTHFD1 Genes Polymorphisms and L-Dopa Treatment. Curr. Genom. 2013, 14, 534–542. [CrossRef] 13. Elias, A.N.; Eng, S. Homocysteine concentrations in patients with diabetes mellitus-relationship to microvascular and macrovas- cular disease. Diabetes Obes. Metab. 2005, 7, 117–121. [CrossRef] 14. Van Meurs, J.B.; Dhonukshe-Rutten, R.A.; Pluijm, S.M.; Van Der Klift, M.; De Jonge, R.; Lindemans, J.; De Groot, L.C.; Hofman, A.; Witteman, J.C.; Van Leeuwen, J.P.; et al. Homocysteine levels and the risk of osteoporotic fracture. N. Engl. J. Med. 2004, 350, 2033–2041. [CrossRef] 15. Yi, F.; Li, P.L. Mechanisms of homocysteine-induced glomerular injury and sclerosis. Am. J. Nephrol. 2008, 28, 254–264. [CrossRef] 16. Mills, J.L.; Lee, Y.J.; Conley, M.R.; Kirke, P.N.; McPartlin, J.M.; Weir, D.G.; Scott, J.M. Homocysteine metabolism in pregnancies complicated by neural-tube defects. Lancet 1995, 345, 149–151. [CrossRef] 17. Adinolﬁ, L.E.; Ingrosso, D.; Cesaro, G.; Cimmino, A.; D’Antò, M.; Capasso, R.; Zappia, V.; Ruggiero, G. Hyperhomocysteinemia and the MTHFR C677T polymorphism promote steatosis and ﬁbrosis in chronic hepatitis C patients. Hepatology 2005, 41, 995–1003. [CrossRef] 18. Wang, L.; Chen, X.; Tang, B.; Hua, X.; Klein-Szanto, A.; Kruger, W.D. Expression of mutant human cystathionine {beta}-synthase rescues neonatal lethality but not homocystinuria in a mouse model. Hum. Mol. Genet. 2005, 14, 2201–2208. [CrossRef] 19. Austin, R.C.; Lentz, S.R.; Werstuck, G.H. Role of hyperhomocysteinemia in endothelial dysfunction and atherothrombotic disease. Cell Death Differ. 2004, 11 (Suppl. S1), S56–S64. [CrossRef] 20. Becker, J.S.; Adler, A.; Schneeberger, A.; Huang, H.; Wang, Z.; Walsh, E.; Koller, A.; Hintze, T.H. Hyperhomocysteinemia, a cardiac metabolic disease: Role of nitric oxide and the p22phox subunit of NADPH oxidase. Circulation 2005, 111, 2112–2118. [CrossRef] 21. Toole, J.; Malinow, M.; Chambless, L. Lowering homocysteine in patients with ischemic stroke to prevent recurrent stroke, myocardial infarction, and death: The Vitamin Intervention for Stroke Prevention (VISP) randomized controlled trial. JAMA 2004, 291, 565–575. [CrossRef] [PubMed] Spence, J.D.; Stampfer, M.J. Understanding the complexity of homocysteine lowering with vitamins: The potential role of subgroup analyses. JAMA 2011, 306, 2610–2611. [CrossRef] [PubMed] 22. 23. Mujumdar, V.S.; Hayden, M.R.; Tyagi, S.C. Homocyst(e)ine induces calcium second messenger in vascular smooth muscle cells. J. Cell Physiol. 2000, 183, 28–36. [CrossRef] 24. Abushik, P.A.; Niittykoski, M.; Giniatullina, R.; Shakirzyanova, A.; Bart, G.; Fayuk, D.; Sibarov, D.A.; Antonov, S.M.; Giniatullin, R. The role of NMDA and mGluR5 receptors in calcium mobilization and neurotoxicity of homocysteine in trigeminal and cortical neurons and glial cells. J. Neurochem. 2014, 129, 264–274. [CrossRef] 25. Ganapathy, P.S.; White, R.E.; Ha, Y.; Bozard, B.R.; McNeil, P.L.; Caldwell, R.W.; Kumar, S.; Black, S.M.; Smith, S.B. The role of N-methyl-D-aspartate receptor activation in homocysteine-induced death of retinal ganglion cells. Investig. Ophthalmol. Vis. Sci. 2011, 52, 5515–5524. [CrossRef] 26. Xu, S.-Z.; Muraki, K.; Zeng, F.; Li, J.; Sukumar, P.; Shah, S.; Dedman, A.M.; Flemming, P.K.; McHugh, D.; Naylor, J.; et al. A sphingosine-1-phosphate-activated calcium channel controlling vascular smooth muscle cell motility. Circ. Res. 2006, 98, 1381–1389. [CrossRef] 27. Kumar, B.; Dreja, K.; Shah, S.S.; Cheong, A.; Xu, S.Z.; Sukumar, P.; Naylor, J.; Forte, A.; Cipollaro, M.; McHugh, D.; et al. Upregulated TRPC1 channel in vascular injury in vivo and its role in human neointimal hyperplasia. Circ. Res. 2006, 98, 557–563. [CrossRef] 28. Xu, S.Z.; Beech, D.J. TrpC1 is a membrane-spanning subunit of store-operated Ca2+ channels in native vascular smooth muscle cells. Circ. Res. 2001, 88, 84–87. [CrossRef] Biomolecules 2023, 13, 952 15 of 16 29. Beech, D.J.; Muraki, K.; Flemming, R. Non-selective cationic channels of smooth muscle and the mammalian homologues of Drosophila TRP. J. Physiol. 2004, 559 Pt 3, 685–706. [CrossRef] 30. Kang, Y.J. Copper and homocysteine in cardiovascular diseases. Pharmacol. Ther. 2010, 129, 321–331. [CrossRef] 31. Mansoor, M.A.; Bergmark, C.; Haswell, S.J.; Savage, I.F.; Evans, P.H.; Berge, R.K.; Svardal, A.M.; Kristensen, O. Correlation between plasma total homocysteine and copper in patients with peripheral vascular disease. Clin. Chem. 2000, 46, 385–391. [CrossRef] 32. Dudman, N.P.; Wilcken, D.E. Increased plasma copper in patients with homocystinuria due to cystathionine beta-synthase deﬁciency. Clin. Chim. Acta 1983, 127, 105–113. [CrossRef] 33. Gromadzka, G.; Rudnicka, M.; Chabik, G.; Przybyłkowski, A.; Członkowska, A. Genetic variability in the methylenetetrahydrofo- late reductase gene (MTHFR) affects clinical expression of Wilson’s disease. J. Hepatol. 2011, 55, 913–919. [CrossRef] 34. Zhang, L.; Ward, M.-L.; Phillips, A.R.; Zhang, S.; Kennedy, J.; Barry, B.; Cannell, M.B.; Cooper, G.J. Protection of the heart by treatment with a divalent-copper-selective chelator reveals a novel mechanism underlying cardiomyopathy in diabetic rats. Cardiovasc. Diabetol. 2013, 12, 123. [CrossRef] 35. Xu, S.Z.; Zhong, W.; Watson, N.M.; Dickerson, E.; Wake, J.D.; Lindow, S.W.; Newton, C.J.; Atkin, S.L. Fluvastatin reduces oxidative damage in human vascular endothelial cells by upregulating Bcl-2. J. Thromb. Haemost. 2008, 6, 692–700. [CrossRef] 36. Daskoulidou, N.; Zeng, B.; Berglund, L.M.; Jiang, H.; Chen, G.-L.; Kotova, O.; Bhandari, S.; Ayoola, J.; Grifﬁn, S.; Atkin, S.L.; et al. High glucose enhances store-operated calcium entry by upregulating ORAI/STIM via calcineurin-NFAT signalling. J. Mol. Med. 2015, 93, 511–521. [CrossRef] 37. Xu, S.-Z.; Sukumar, P.; Zeng, F.; Li, J.; Jairaman, A.; English, A.; Naylor, J.; Ciurtin, C.; Majeed, Y.; Milligan, C.J.; et al. TRPC 38. channel activation by extracellular thioredoxin. Nature 2008, 451, 69–72. [CrossRef] Jiang, H.; Zeng, B.; Chen, G.-L.; Bot, D.; Eastmond, S.; Elsenussi, S.E.; Atkin, S.L.; Boa, A.N.; Xu, S.-Z. Effect of non-steroidal anti-inﬂammatory drugs and new fenamate analogues on TRPC4 and TRPC5 channels. Biochem. Pharmacol. 2012, 83, 923–931. [CrossRef] 39. Chen, G.-L.; Zeng, B.; Eastmond, S.; Elsenussi, S.E.; Boa, A.; Xu, S.-Z. Pharmacological comparison of novel synthetic fenamate analogues with econazole and 2-APB on the inhibition of TRPM2 channels. Br. J. Pharmacol. 2012, 167, 1232–1243. [CrossRef] 40. Li, P.; Rubaiy, H.N.; Chen, G.L.; Hallett, T.; Zaibi, N.; Zeng, B.; Saurabh, R.; Xu, S.Z. Mibefradil, a T-type Ca(2+) channel blocker also blocks Orai channels by action at the extracellular surface. Br. J. Pharmacol. 2019, 176, 3845–3856. [CrossRef] 41. Xu, S.-Z.; Zeng, B.; Daskoulidou, N.; Chen, G.-L.; Atkin, S.L.; Lukhele, B. Activation of TRPC cationic channels by mercurial compounds confers the cytotoxicity of mercury exposure. Toxicol. Sci. 2012, 125, 56–68. [CrossRef] [PubMed] 42. Zeng, B.; Yuan, C.; Yang, X.; Atkin, S.L.; Xu, S.-Z. TRPC channels and their splice variants are essential for promoting human ovarian cancer cell proliferation and tumorigenesis. Curr. Cancer Drug Targets 2013, 13, 103–116. [CrossRef] [PubMed] 43. Zaibi, N.; Li, P.; Xu, S.Z. Protective effects of dapagliﬂozin against oxidative stress-induced cell injury in human proximal tubular cells. PLoS ONE 2021, 16, e0247234. [CrossRef] [PubMed] 44. Aranda, E.; Owen, G.I. A semi-quantitative assay to screen for angiogenic compounds and compounds with angiogenic potential using the EA.hy926 endothelial cell line. Biol. Res. 2009, 42, 377–389. [CrossRef] 45. Xu, S.-Z.; Zeng, F.; Boulay, G.; Grimm, C.; Harteneck, C.; Beech, D.J. Block of TRPC5 channels by 2-aminoethoxydiphenyl borate: A differential, extracellular and voltage-dependent effect. Br. J. Pharmacol. 2005, 145, 405–414. [CrossRef] 46. Zeng, B.; Chen, G.L.; Xu, S.Z. Divalent copper is a potent extracellular blocker for TRPM2 channel. Biochem. Biophys. Res. Commun. 2012, 424, 279–284. [CrossRef] 47. Kuang, C.-Y.; Yü, Y.; Wang, K.; Qian, D.-H.; Den, M.-Y.; Huang, L. Knockdown of transient receptor potential canonical-1 reduces the proliferation and migration of endothelial progenitor cells. Stem. Cells Dev. 2012, 21, 487–496. [CrossRef] 48. Xu, S.-Z.; Zeng, F.; Lei, M.; Li, J.; Gao, B.; Xiong, C.; Sivaprasadarao, A.; Beech, D. Generation of functional ion-channel tools by E3 targeting. Nat. Biotechnol. 2005, 23, 1289–1293. [CrossRef] 49. Zeng, B.; Chen, G.-L.; Garcia-Vaz, E.; Bhandari, S.; Daskoulidou, N.; Berglund, L.M.; Jiang, H.; Hallett, T.; Zhou, L.-P.; Huang, L.; et al. ORAI channels are critical for receptor-mediated endocytosis of albumin. Nat. Commun. 2017, 8, 1920. [CrossRef] 50. Yu, P.-C.; Gu, S.-Y.; Bu, J.-W.; Du, J.-L. TRPC1 is essential for in vivo angiogenesis in zebraﬁsh. Circ. Res. 2010, 106, 1221–1232. [CrossRef] 51. Antigny, F.; Girardin, N.; Frieden, M. Transient receptor potential canonical channels are required for in vitro endothelial tube formation. J. Biol. Chem. 2012, 287, 5917–5927. [CrossRef] 52. Alexandru, N.; Jardín, I.; Popov, D.; Simionescu, M.; García-Estañ, J.; Salido, G.M.; Rosado, J.A. Effect of homocysteine on calcium mobilization and platelet function in type 2 diabetes mellitus. J. Cell Mol. Med. 2008, 12, 2586–2597. [CrossRef] 53. Han, H.; Wang, Y.; Li, X.; Wang, P.A.; Wei, X.; Liang, W.; Ding, G.; Yu, X.; Bao, C.; Zhang, Y.; et al. Novel role of NOD2 in mediating Ca2+ signaling: Evidence from NOD2-regulated podocyte TRPC6 channels in hyperhomocysteinemia. Hypertension 2013, 62, 506–511. [CrossRef] 54. Ovey, I.S.; Naziroglu, M. Homocysteine and cytosolic GSH depletion induce apoptosis and oxidative toxicity through cytosolic calcium overload in the hippocampus of aged mice: Involvement of TRPM2 and TRPV1 channels. Neuroscience 2015, 284, 225–233. [CrossRef] Biomolecules 2023, 13, 952 16 of 16 55. Chigurupati, S.; Wei, Z.; Belal, C.; Vandermey, M.; Kyriazis, G.; Arumugam, T.; Chan, S.L. The homocysteine-inducible endoplasmic reticulum stress protein counteracts calcium store depletion and induction of CCAAT enhancer-binding protein homologous protein in a neurotoxin model of Parkinson disease. J. Biol. Chem. 2009, 284, 18323–18333. [CrossRef] 56. Zhang, H.-S.; Xiao, J.-H.; Cao, E.-H.; Qin, J.-F. Homocysteine inhibits store-mediated calcium entry in human endothelial cells: Evidence for involvement of membrane potential and actin cytoskeleton. Mol. Cell Biochem. 2005, 269, 37–47. [CrossRef] 57. Cai, B.; Gong, D.; Pan, Z.; Liu, Y.; Qian, H.; Zhang, Y.; Jiao, J.; Lu, Y.; Yang, B. Large-conductance Ca2+-activated K+ currents blocked and impaired by homocysteine in human and rat mesenteric artery smooth muscle cells. Life Sci. 2007, 80, 2060–2066. [CrossRef] 58. Cortés, M.P.; Becerra, J.P.; Vinet, R.; Álvarez, R.; Quintana, I. Inhibition of ATP-induced calcium inﬂux by homocysteine in human umbilical vein endothelial cells. Cell Biol. Int. 2013, 37, 600–607. [CrossRef] 59. Cai, B.; Gong, D.; Chen, N.; Li, J.; Wang, G.; Lu, Y.; Yang, B. The negative inotropic effects of homocysteine were prevented by matrine via the regulating intracellular calcium level. Int. J. Cardiol. 2011, 150, 113–115. [CrossRef] 60. Thilo, F.; Liu, Y.; Krueger, K.; Förste, N.; Wittstock, A.; Scholze, A.; Tepel, M. Do cysteine residues regulate transient receptor potential canonical type 6 channel protein expression? Antioxid. Redox Signal 2012, 16, 452–457. [CrossRef] 61. Lucock, M.; Yates, Z. Folic acid-vitamin and panacea or genetic time bomb? Nat. Rev. Genet. 2005, 6, 235–240. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijms22126253	Article S-Palmitoylation of Synaptic Proteins as a Novel Mechanism Underlying Sex-Dependent Differences in Neuronal Plasticity Monika Zar˛eba-Kozioł 1,,† Izabela Figiel 1 Weiqi Zhang 5, Evgeni Ponimaskin 4 , Anup Kumar Halder 3 , Anna Bartkowiak-Kaczmarek 1,†, Matylda Roszkowska 1, Krystian Bijata 1,2, , , Paulina Kami ´nska 1, Franziska E. Müller 4, Subhadip Basu 3 and Jakub Włodarczyk 1, 1 2 Laboratory of Cell Biophysics, Nencki Institute of Experimental Biology, Polish Academy of Science, Pasteur Str. 3, 02-093 Warsaw, Poland; [email protected] (A.B.-K.); [email protected] (M.R.); [email protected] (K.B.); i.ﬁ[email protected] (I.F.); [email protected] (P.K.) Faculty of Chemistry, University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland 3 Department of Computer Science and Engineering, Jadvapur University, Kolkata 700032, India; [email protected] (A.K.H.); [email protected] (S.B.) 4 Cellular Neurophysiology, Hannover Medical School, Carl-Neuberg Str. 1, 30625 Hannover, Germany; [email protected] (F.E.M.); [email protected] (E.P.) 5 Department of Mental Health, University of Münster, Albert-Schweitzer-Campus 1/A9, 48149 Munster, Germany; [email protected] * Correspondence: [email protected] (M.Z.-K.); [email protected] (J.W.) † These authors contributed equally. Abstract: Although sex differences in the brain are prevalent, the knowledge about mechanisms underlying sex-related effects on normal and pathological brain functioning is rather poor. It is known that female and male brains differ in size and connectivity. Moreover, those differences are related to neuronal morphology, synaptic plasticity, and molecular signaling pathways. Among different processes assuring proper synapse functions are posttranslational modiﬁcations, and among them, S-palmitoylation (S-PALM) emerges as a crucial mechanism regulating synaptic integrity. Protein S-PALM is governed by a family of palmitoyl acyltransferases, also known as DHHC proteins. Here we focused on the sex-related functional importance of DHHC7 acyltransferase because of its S-PALM action over different synaptic proteins as well as sex steroid receptors. Using the mass spectrometry-based PANIMoni method, we identiﬁed sex-dependent differences in the S-PALM of synaptic proteins potentially involved in the regulation of membrane excitability and synaptic transmission as well as in the signaling of proteins involved in the structural plasticity of dendritic spines. To determine a mechanistic source for obtained sex-dependent changes in protein S-PALM, we analyzed synaptoneurosomes isolated from DHHC7-/- (DHHC7KO) female and male mice. Our data showed sex-dependent action of DHHC7 acyltransferase. Furthermore, we revealed that different S-PALM proteins control the same biological processes in male and female synapses. Keywords: posttranslational modiﬁcations; palmitoylation; sexes; proteomics; synapses; synaptic plasticity; DHHC7 Citation: Zar˛eba-Kozioł, M.; Bartkowiak-Kaczmarek, A.; Roszkowska, M.; Bijata, K.; Figiel, I.; Halder, A.K.; Kami ´nska, P.; Müller, F.E.; Basu, S.; Zhang, W.; et al. S-Palmitoylation of Synaptic Proteins as a Novel Mechanism Underlying Sex-Dependent Differences in Neuronal Plasticity. Int. J. Mol. Sci. 2021, 22, 6253. https://doi.org/ 10.3390/ijms22126253 Academic Editor: Seung Ho Jung Received: 9 April 2021 Accepted: 8 June 2021 Published: 10 June 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. 1. Introduction Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Synaptic plasticity plays a fundamental role in the brain since it is essential for learning and memory. Changes in synapse strength are expressed at the level of different synaptic proteins (i.e., receptors, cytoskeleton elements, signaling molecules) and translated into structural and functional modiﬁcations of neuronal functions. Many multiple and coor- dinated signaling pathways that control memory formation at the molecular level have been identiﬁed. However, the substantial knowledge of how plastic changes of neurons govern the information processing in the brain comes from the research conducted mainly on the male population [1,2]. Sex should be considered as an important biological variable in neuroscience since sex-dependent differences in the brain are prevalent and can be Int. J. Mol. Sci. 2021, 22, 6253. https://doi.org/10.3390/ijms22126253 https://www.mdpi.com/journal/ijms International Journal of Molecular Sciences(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Int. J. Mol. Sci. 2021, 22, 6253 2 of 21 detected even at the level of single synaptic connections. Divergent synapse molecular organization [3] and signaling pathways [4,5] together with sex-speciﬁc changes in plastic- ity [6,7] may contribute to sex differences in neuronal function and account for sex-related differences in learning and memory [8], emotional responses [9], fear and anxiety. Biologi- cal sex has serious clinical consequences and manifests in the existence of gender bias in neuropsychiatric disorders [10]. Females are more vulnerable to develop major depressive disorder [11], while males are at higher risk to suffer from autism spectrum disorder [12]. Moreover, the same disorders exhibit sex-differences in symptoms of severity [13,14]. De- spite the obvious importance, current knowledge regarding sex-speciﬁc neuroplasticity is poor and mostly focused on the role of estrogen [7,15], calcium/calmodulin-dependent protein kinase [16] in the structural and functional plasticity of dendritic spines and nitric oxide synthase (NOS) [17] impact on synaptic potentiation in both sexes. It is possible that other subtle processes controlling function and activity of synaptic proteins exist and may inﬂuence downstream signaling at the synapses, switching alternative signal transduction pathways in females and males. Identiﬁcation of intracellular mechanisms regulating infor- mation processing and storage by neurons in both sexes is critical for the development of sex-speciﬁc therapies addressing numerous memory disorders and psychiatric conditions. Posttranslational modiﬁcations (PTMs) are known to be critically involved in assuring proper synapse function. Among multiple PTMs, S-palmitoylation (S-PALM) emerges as a crucial mechanism underlying synaptic integrity and its dysfunction relates to neuropsy- chiatric disorders [18]. This reversible modiﬁcation modulates properties of target proteins including neurotransmitter receptors, synaptic scaffolding proteins and secreted signaling molecules what allows quick and precise regulation of synaptic plasticity [19–22]. One of the palmitoylating enzymes, palmitoyl acyltransferase DHHC7, is of particular interest in relation to the sex-dependent neuroplasticity because of its S-PALM-mediating actions mainly by sex steroid receptors [23]. Functional consequences arising from palmitoylation of estrogen (ER) and progesterone (PR) receptors are expressed in sex-speciﬁc changes of synaptic function, plasticity and connectivity of different brain regions [24,25]. Further- more, DHHC7 is also engaged in the modiﬁcation of various synaptic proteins [26–28] thus regulating their membrane attachment, sorting, and function relevant for the proper function of synaptic connections. However, the distinct and sex-speciﬁc mechanism of DHHC7 action is currently unknown. Since S-PALM affects a wide range of synaptic proteins some of them could be im- plicated in sex-speciﬁc neuroplasticity, the present study investigated the differences in S-PALM of synaptic proteins between sexes. We have applied a high-throughput proteomic approach—Acyl Biotin Exchange (ABE) method which not only allows pinpointing the S-palmitoylated proteins but also precise sites of S-PALM modiﬁcations in those proteins. Using synaptoneurosomal fraction from whole-brain, we showed different S-PALM proﬁles in female and male wild-type mice (WT). We identiﬁed 2458 peptides assigned to 1239 proteins and among all identiﬁed S-PALM synaptoneurosomal proteins, 200 were present only in female brains and 271 were identiﬁed exclusively in male mice brains. Furthermore, to determine a mechanistic source for sex-dependent changes in S-PALM pattern and examine S-PALM role as a crucial intracellular mechanism governing sex-speciﬁc variances in synapse structure and function, we studied S-PALM of synaptic proteins in DHHC-7 knock-out mice of both sexes. We strictly deﬁned synaptic targets for DHHC7 proteins in female and male brains. Our data showed the sex-dependent action of DHHC7 acyltrans- ferase. We identiﬁed a total of 150 uniquely S-PALM by DHHC7 proteins in female brains compared with 125 exclusively S-PALM proteins in male brains. Finally, to gain better insight into the biological and functional relevance of the discovered S-PALM synaptic targets, we incorporated bioinformatics analysis. We demonstrated that S-PALM proteins modulate crucial processes for neuronal functioning. Interestingly, the same signaling pathways/biological processes appeared to be regulated by different S-PALM proteins in both sexes. Int. J. Mol. Sci. 2021, 22, 6253 3 of 21 2. Results 2.1. Sex-Dependent Differences in S-Palmitoylation of Synaptic Proteins Several studies demonstrate differences between female and male brains at the level of molecular and structural synaptic plasticity, but the underlying mechanisms are not fully understood [1,29]. Looking for the possible source of these differences, we analyzed the S-PALM proﬁle of synaptoneurosomal proteins isolated from three-month-old C57BL/6J lit- termates, male and female wild type mice (WT). First, we applied the Acyl-Biotin Exchange (ABE) method to check the sex-dependent pattern of synaptic proteins S-PALM. This method is based on the selective cleavage of thioester bonds between cysteines and palmi- tate by hydroxylamine (NH2OH/HA) after blocking free-SH groups by N-ethylmaleimide (NEM) (Figure 1a). Cleavage of the ester linkage allows the speciﬁc incorporation of bi- otin (biotin HPDP) to the newly available thiol group and enables detection by Western blotting. We used non-deacylated samples as controls for the speciﬁcity of the reaction. The ABE experiments revealed slight differences in the S-PALM pattern between the sexes (Figure 1b). Figure 1. Analysis of S-PALM pattern in the female and male WT synaptoneurosomes. (a) Scheme of the applied ABE method. (b) Western blot analysis of the pattern of biotinylated proteins obtained with streptavidin-HRP antibody. Controls were prepared without selective cleavage of S-PALM thioester bonds by HA. Ponceau S staining was used as a loading control. To identify differentially S-PALM proteins in both sexes, we used the mass spectrometry- based approach developed by us earlier, the PANIMoni method (see Materials and Methods for details) [30]. Similarly to the ABE, S-PALM proteins are labeled with biotin. This is followed by proteolytic digestion prior to the capture by avidin. This step allows the selective isolation of previously S-PALM peptides but does not detect intact S-PALM proteins. The PANIMoni method allows ﬁnding changes in S-PALM at the level of speciﬁc proteins and cysteine residues (Figure 2a). A standard control that omits the HA-driven cleavage of acyl bonds allows the separation of false identiﬁcations. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 3 of 21 ronal functioning. Interestingly, the same signaling pathways/biological processes ap-peared to be regulated by different S-PALM proteins in both sexes. 2. Results 2.1. Sex-Dependent Differences in S-Palmitoylation of Synaptic Proteins Several studies demonstrate differences between female and male brains at the level of molecular and structural synaptic plasticity, but the underlying mechanisms are not fully understood [1,29]. Looking for the possible source of these differences, we analyzed the S-PALM profile of synaptoneurosomal proteins isolated from three-month-old C57BL/6J littermates, male and female wild type mice (WT). First, we applied the Acyl-Biotin Exchange (ABE) method to check the sex-dependent pattern of synaptic proteins S-PALM. This method is based on the selective cleavage of thioester bonds be-tween cysteines and palmitate by hydroxylamine (NH2OH/HA) after blocking free-SH groups by N-ethylmaleimide (NEM) (Figure 1a). Cleavage of the ester linkage allows the specific incorporation of biotin (biotin HPDP) to the newly available thiol group and enables detection by Western blotting. We used non-deacylated samples as controls for the specificity of the reaction. The ABE experiments revealed slight differences in the S-PALM pattern between the sexes (Figure 1b). Figure 1. Analysis of S-PALM pattern in the female and male WT synaptoneurosomes. (a) Scheme of the applied ABE method. (b) Western blot analysis of the pattern of biotinylated proteins obtained with streptavidin-HRP antibody. Con-trols were prepared without selective cleavage of S-PALM thioester bonds by HA. Ponceau S staining was used as a loading control. To identify differentially S-PALM proteins in both sexes, we used the mass spec-trometry-based approach developed by us earlier, the PANIMoni method (see Materials and Methods for details) [30]. Similarly to the ABE, S-PALM proteins are labeled with biotin. This is followed by proteolytic digestion prior to the capture by avidin. This step allows the selective isolation of previously S-PALM peptides but does not detect intact S-PALM proteins. The PANIMoni method allows finding changes in S-PALM at the level of specific proteins and cysteine residues (Figure 2a). A standard control that omits the HA-driven cleavage of acyl bonds allows the separation of false identifications. Int. J. Mol. Sci. 2021, 22, 6253 4 of 21 Figure 2. Sex-dependent analysis of protein S-PALM in synaptoneurosomes isolated from female and male WT mice. (a) Scheme of PANIMoni mass spectrometry-based method. (b) Scheme of S-PALM analysis showing differential and sequential S-PALM classiﬁcation. (c) Venn diagram analysis of differential S-PALM in female and male synaptoneurosomes (Nmice = 3/group). Using this method combined with peptide identiﬁcation by mass spectrometry, we recognized 2017 peptides in the synaptoneurosomal fraction that are assigned to 1217 S-PALM proteins with less than 1% false discovery rate (FDR) (Table S1). Besides, the high similarity of S-PALM identiﬁcation was conﬁrmed in three biological replicates (Nmice = 3/per group) by mass spectrometry. The overall scheme of differential S-PALM analysis and classiﬁcation of synaptoneurosomal proteins into different sets of proteins is presented in Figure 2b. Among all identiﬁed S-PALM synaptoneurosomal proteins, 116 were present exclu- sively in female brains, while 164 were identiﬁed only in male mouse brains. We also observed differences at the level of speciﬁc sites of S-PALM. Altogether, we identiﬁed 200 S-PALM synaptic proteins upregulated in female synaptoneurosomes and 271 S-PALM synaptic proteins that were upregulated in male derived synaptoneurosomes (Figure 2c). To gain functional insight into the S-PALM-proteins which differentiate between fe- male and male WT mice, we applied the ClueGO, the widely used Cytoscape plugin [31,32]. In WT female synaptoneurosomes, 113 Gene Ontology biological processes (GO_BP) terms were signiﬁcantly enriched (p-value < 0.01) among WT female speciﬁc proteins and catego- rized into 11 GO groups (networks) as shown in Figure 3a. The most presented GO_BP signiﬁcant functional groups included: receptor localization to synapse (p-value = 0.00001, e.g., Dlg4, Git1, Gphn), synapse organization (p-value = 2.01 × 10−7, e.g., Dlg4, Nectin1, Septin7, Shank3), pyruvate dehydrogenase activity (p-value = 5.34 × 10−8, e.g., Dld, Pdha1, Pdk3), cellular respiration (p-value = 3.85 × 10−8, e.g., Aco2, Ndufb9, Ndufv1), neuro- transmitter transport (p-value = 7.77 × 10−7, e.g., Nrxn1, Septin5, Slc25a22), regulation of transmembrane transporter activity (p-value = 0.00002, e.g., Gja1, Homer1, Park7, Tcaf1), and other (Figure 3b). All GO_BP terms (p < 0.01) along with the percentage of genes associated with upregulated S-PALM synaptic proteins in the female WT are presented in Figure S1. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 4 of 21 Figure 2. Sex-dependent analysis of protein S-PALM in synaptoneurosomes isolated from female and male WT mice. (a) Scheme of PANIMoni mass spectrometry-based method. (b) Scheme of S-PALM analysis showing differential and se-quential S-PALM classification. (c) Venn diagram analysis of differential S-PALM in female and male synaptoneuro-somes (Nmice = 3/group). Using this method combined with peptide identification by mass spectrometry, we recognized 2017 peptides in the synaptoneurosomal fraction that are assigned to 1217 S-PALM proteins with less than 1% false discovery rate (FDR) (Table S1). Besides, the high similarity of S-PALM identification was confirmed in three biological replicates (Nmice = 3/per group) by mass spectrometry. The overall scheme of differential S-PALM analysis and classification of synaptoneurosomal proteins into different sets of proteins is presented in Figure 2b. Among all identified S-PALM synaptoneurosomal proteins, 116 were present ex-clusively in female brains, while 164 were identified only in male mouse brains. We also observed differences at the level of specific sites of S-PALM. Altogether, we identified 200 S-PALM synaptic proteins upregulated in female synaptoneurosomes and 271 S-PALM synaptic proteins that were upregulated in male derived synaptoneurosomes (Figure 2c). To gain functional insight into the S-PALM-proteins which differentiate between female and male WT mice, we applied the ClueGO, the widely used Cytoscape plugin [31,32]. In WT female synaptoneurosomes, 113 Gene Ontology biological processes (GO_BP) terms were significantly enriched (p-value < 0.01) among WT female specific proteins and categorized into 11 GO groups (networks) as shown in Figure 3a. The most presented GO_BP significant functional groups included: receptor localization to synapse (p-value = 0.00001, e.g., Dlg4, Git1, Gphn), synapse organization (p-value = 2.01 × 10−7, e.g., Dlg4, Nectin1, Septin7, Shank3), pyruvate dehydrogenase activity (p-value = 5.34 × 10−8, e.g., Dld, Pdha1, Pdk3), cellular respiration (p-value = 3.85 × 10−8, e.g., Aco2, Ndufb9, Ndufv1), neurotransmitter transport (p-value = 7.77 × 10−7, e.g., Nrxn1, Septin5, Slc25a22), regulation of transmembrane transporter activity (p-value = 0.00002, e.g., Gja1, Homer1, Park7, Tcaf1), and other (Figure 3b). All GO_BP terms (p < 0.01) along with the percentage of genes associated with upregulated S-PALM synaptic proteins in the female WT are presented in Figure S1. Int. J. Mol. Sci. 2021, 22, 6253 5 of 21 Figure 3. Functional enrichment analysis of proteins speciﬁc for female WT mice using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Network depicting interactions between enriched functional classes. Each circle represents a biological term consisting of various related proteins/genes. Terms that belong to the same pathway are shown with the same color, and terms associated with two different pathways are marked with two colors. The size of the circles relates to the statistical signiﬁcance of the term enrichment. The connectivity (edges) between the terms in the network is derived from kappa score, (indicates the similarity of associated genes shared by different terms). Thicker edges indicate stronger similarity. Diamonds represent directed edges which link parent terms to child terms. Only the name of the most signiﬁcant term in each group is shown to reduce overlay. (b) A diagram showing the percentage of terms per group of enriched protein classes. Similar analysis of speciﬁc S-PALM synaptic proteins in WT male mice is presented in Figure 4. In contrast to females, in male synaptoneurosomes 74 GO_BP functional groups were signiﬁcantly enriched (p < 0.01) and grouped into 17 GO terms networks, as presented in Figure 4a and Figure S2. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 5 of 21 Figure 3. Functional enrichment analysis of proteins specific for female WT mice using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Network depicting interactions between enriched functional classes. Each circle represents a biological term consisting of various related proteins/genes. Terms that belong to the same pathway are shown with the same color, and terms associated with two different pathways are marked with two colors. The size of the circles relates to the statistical significance of the term enrichment. The connectivity (edges) between the terms in the network is derived from kappa score, (indicates the similarity of associated genes shared by different terms). Thicker edges indicate stronger similarity. Diamonds represent directed edges which link parent terms to child terms. Only the name of the most significant term in each group is shown to reduce overlay. (b) A diagram showing the per-centage of terms per group of enriched protein classes. Similar analysis of specific S-PALM synaptic proteins in WT male mice is presented in Figure 4. In contrast to females, in male synaptoneurosomes 74 GO_BP functional groups were significantly enriched (p < 0.01) and grouped into 17 GO terms networks, as presented in Figures 4a and S2. Int. J. Mol. Sci. 2021, 22, 6253 6 of 21 Figure 4. Functional enrichment analysis of proteins speciﬁc for male WT mice using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Network depicting interactions between enriched functional classes. (b) A diagram showing the percentage of terms per group analysis of enriched protein classes. (c) Comparison of semantic similarity (SS) of protein pairs from male and female WT mice for each type of Gene Ontology term: cellular components (CC), molecular functions (MF) and biological processes (BP), respectively. The X-axis represents the SS score (ranges—0, 1) and the Y-axis represents the frequency of protein pairs. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 6 of 21 Figure 4. Functional enrichment analysis of proteins specific for male WT mice using the Gene Ontology biological pro-cesses database and the Clue Go algorithm. (a) Network depicting interactions between enriched functional classes. (b) A diagram showing the percentage of terms per group analysis of enriched protein classes. (c) Comparison of semantic similarity (SS) of protein pairs from male and female WT mice for each type of Gene Ontology term: cellular components (CC), molecular functions (MF) and biological processes (BP), respectively. The X-axis represents the SS score (ranges—0, 1) and the Y-axis represents the frequency of protein pairs. Int. J. Mol. Sci. 2021, 22, 6253 7 of 21 The most signiﬁcant GO_BP pathways enriched (p-value < 0.01) in the 155 male- speciﬁc S-PALM proteins set were related to: neuron projection morphogenesis (p-value = 2.12 × 10−7, e.g., Bcan, Cdc42, Flot1, Nptn), neuron projection development (p-value = 1.51 × 10−9, e.g., Camk1, Camk2g, Nptn), modulation of chemical synapse transmission (p-value = 4.07 × 10−6, e.g., Gria4, Grik2, Grin2b, Snap25), but also synap- tic vesicle cycle (p-value = 1.71 × 10−8, e.g., Prkaca, Rap1b, Rims1, Slc32a1, Snap25), synapse organization (p-value = 0.0058, e.g., Camk1, Cdc42, Grin2b, Srcin1), and others (Figure 4a,b). Additionally, we quantiﬁed pairwise similarities between protein annotations based on semantic similarity measure for Gene Ontology terms proposed by Dutta et al. [33]. The ontological annotations of each protein pair were incorporated into a graph-theoretic approach for assessing the SS score. Our analysis shows that the distribution of semantic similarities between protein pairs of the male WT are not from the same distribution as the female WT (Figure 4c). To verify that the observed changes were not due to a difference in protein expression, we compared the expression levels of synaptoneurosomal proteins from the brains of female and male WT mice. Our analyses showed slight changes in protein expression between the compared groups (Table S2). Overall, in the proteomics experiments, we identiﬁed a total of 6250 peptides which are assigned to 2592 proteins with FDR 1%. Among 2592 identiﬁed and quantiﬁed synaptoneurosomal proteins, a minority of them showed differential expression, with 17 proteins found to be signiﬁcantly upregulated in female WT (e.g., Slc6a1, Sept10, Asap2,) and 8 in male WT (e.g., Reep2, Skp1a or Crat) (Figure 5a, Table S2). Figure 5. Sex-dependent differences in synaptic protein expression. Volcano plots display differentially regulated synaptic proteins of female and male WT mice (a), male WT and male DHHC7 KO (b), and female WT and female DHHC7 KO (c). Proteins with statistically signiﬁcant differential expression (p < 0.05) are highlighted in the top right and left quadrants. 2.2. DHHC7–Dependent Synaptic Proteins S-Palmitoylation in Male and Female Mice To better understand the mechanisms underlying the sex-dependent S-PALM, we used DHHC7 KO [34,35]. DHHC7 is not only involved in S-PALM of different synaptic substrates but is also responsible for modifying sex steroid receptors [36,37]. Moreover, it was shown to be developmentally regulated in a sex-dependent manner [23]. To determine a mechanistic source for the sex-dependent activity of DHHC7, we examined S-PALM of synaptic proteins in female and male DHHC7 KO mice. In the ﬁrst step, we found that the vast majority of proteins did not show signiﬁcant changes in expression between WT and DHHC7 KO synaptoneurosomal brain tissues in females and males (Figure 5b,c). However, differential analysis revealed that a small handful of proteins exhibit differential expression, with one protein (Rmnd5b) which was signiﬁcantly increased in male DHHC7 KO. In contrast, 15 proteins (e.g., Prkcd, Slc18a2) were found to be signiﬁcantly increased in male WT synaptoneurosomes (Figure 5b and Table S2). Similarly, when we compared the synaptic proteins of female WT and DHHC7 Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 7 of 21 The most significant GO_BP pathways enriched (p-value < 0.01) in the 155 male-specific S-PALM proteins set were related to: neuron projection morphogenesis (p-value = 2.12 × 10−7, e.g., Bcan, Cdc42, Flot1, Nptn), neuron projection development (p-value = 1.51 × 10−9, e.g., Camk1, Camk2g, Nptn), modulation of chemical synapse transmission (p-value = 4.07 × 10−6, e.g., Gria4, Grik2, Grin2b, Snap25), but also synaptic vesicle cycle (p-value = 1.71 × 10−8, e.g., Prkaca, Rap1b, Rims1, Slc32a1, Snap25), synapse organization (p-value = 0.0058, e.g., Camk1, Cdc42, Grin2b, Srcin1), and others (Figure 4a,b). Additionally, we quantified pairwise similarities between protein annotations based on semantic similarity measure for Gene Ontology terms proposed by Dutta et al. [33]. The ontological annotations of each protein pair were incorporated into a graph-theoretic approach for assessing the SS score. Our analysis shows that the distribution of semantic similarities between protein pairs of the male WT are not from the same distribution as the female WT (Figure 4c). To verify that the observed changes were not due to a difference in protein expres-sion, we compared the expression levels of synaptoneurosomal proteins from the brains of female and male WT mice. Our analyses showed slight changes in protein expression between the compared groups (Table S2). Overall, in the proteomics experiments, we identified a total of 6250 peptides which are assigned to 2592 proteins with FDR 1%. Among 2592 identified and quantified synaptoneurosomal proteins, a minority of them showed differential expression, with 17 proteins found to be significantly upregulated in female WT (e.g., Slc6a1, Sept10, Asap2,) and 8 in male WT (e.g., Reep2, Skp1a or Crat) (Figure 5a, Table S2). Figure 5. Sex-dependent differences in synaptic protein expression. Volcano plots display differentially regulated synap-tic proteins of female and male WT mice (a), male WT and male DHHC7 KO (b), and female WT and female DHHC7 KO (c). Proteins with statistically significant differential expression (p < 0.05) are highlighted in the top right and left quad-rants. 2.2. DHHC7–Dependent Synaptic Proteins S-Palmitoylation in Male and Female Mice To better understand the mechanisms underlying the sex-dependent S-PALM, we used DHHC7 KO [34,35]. DHHC7 is not only involved in S-PALM of different synaptic substrates but is also responsible for modifying sex steroid receptors [36,37]. Moreover, it was shown to be developmentally regulated in a sex-dependent manner [23]. To deter-mine a mechanistic source for the sex-dependent activity of DHHC7, we examined S-PALM of synaptic proteins in female and male DHHC7 KO mice. In the first step, we found that the vast majority of proteins did not show significant changes in expression between WT and DHHC7 KO synaptoneurosomal brain tissues in females and males (Figure 5b,c). However, differential analysis revealed that a small handful of proteins exhibit differential expression, with one protein (Rmnd5b) which was significantly increased in male DHHC7 KO. In contrast, 15 proteins (e.g., Prkcd, Slc18a2) were found to be significantly increased in male WT synaptoneurosomes (Figure 5b and Table S2). Similarly, when we compared the synaptic proteins of female WT and DHHC7 Int. J. Mol. Sci. 2021, 22, 6253 8 of 21 KO, we noticed 17 proteins upregulated in WT female mice (e.g., Kcnma1) and 6 proteins upregulated in female DHHC7 KO (e.g., Atp1a1, Dld, uba1a) (Figure 5c and Table S2). To assess the role of DHHC7 in the palmitoylation of synaptic proteins, we applied the PANIMoni proteomic method to synaptoneurosomes isolated from DHHC7 KO male brains (Figure 6a, Table S3). In total, we identiﬁed 1896 peptides that correspond to 1117 proteins. To ﬁnd the degree of similarity between the identiﬁed protein groups (male WT and male DHHC7 KO) we used Venn diagram analysis. Proteins relying on DHHC7 for their palmitoylation were expected to be absent from the palmitoyl proteomes of DHHC7 KO. Among all identiﬁed S-PALM synaptosomal proteins, 106 proteins were present only in male WT, while 122 proteins were found only in male DHHC7 KO (Figure 6a). Additionally, we detected the differences at the level of speciﬁc S-PALM sites. Overall, combining the protein and site identiﬁcation data, we indicated 148 proteins with S-PALM regulated by DHHC7 in male synaptoneurosomes (Figure 6a). Proteins identiﬁed only in male DHHC7 KO may represent proteins that are modiﬁed by some undeﬁned compensation mechanisms. Figure 6. Analysis of male synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in male WT and male DHHC7 KO synaptoneurosomes. (b) Network depicting interactions between enriched functional classes. (c) List of all enriched GO_BP functional classes (p < 0.01) showed on the network. (d) Percentage of terms per group analysis of enriched protein classes. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 8 of 21 KO, we noticed 17 proteins upregulated in WT female mice (e.g., Kcnma1) and 6 proteins upregulated in female DHHC7 KO (e.g., Atp1a1, Dld, uba1a) (Figure 5c and Table S2). To assess the role of DHHC7 in the palmitoylation of synaptic proteins, we applied the PANIMoni proteomic method to synaptoneurosomes isolated from DHHC7 KO male brains (Figure 6a, Table S3). In total, we identified 1896 peptides that correspond to 1117 proteins. To find the degree of similarity between the identified protein groups (male WT and male DHHC7 KO) we used Venn diagram analysis. Proteins relying on DHHC7 for their palmitoylation were expected to be absent from the palmitoyl proteomes of DHHC7 KO. Among all identified S-PALM synaptosomal proteins, 106 proteins were present only in male WT, while 122 proteins were found only in male DHHC7 KO (Figure 6a). Additionally, we detected the differences at the level of specific S-PALM sites. Overall, combining the protein and site identification data, we indicated 148 proteins with S-PALM regulated by DHHC7 in male synaptoneurosomes (Figure 6a). Proteins identi-fied only in male DHHC7 KO may represent proteins that are modified by some unde-fined compensation mechanisms. Figure 6. Analysis of male synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in male WT and male DHHC7 KO synaptoneurosomes. (b) Network depicting interactions between enriched functional classes. (c) List of all enriched GO_BP functional classes (p < 0.01) showed on the network. (d) Percentage of terms per group analysis of enriched protein classes. In order to decipher the molecular mechanisms at the synapse in which DHHC7-dependent S-PALM in males might play a distinct role, we performed func-tional enrichment analyses of terms from the GO_BP using the ClueGO algorithm. Int. J. Mol. Sci. 2021, 22, 6253 9 of 21 In order to decipher the molecular mechanisms at the synapse in which DHHC7- dependent S-PALM in males might play a distinct role, we performed functional enrichment analyses of terms from the GO_BP using the ClueGO algorithm. Analyses of this set of pro- tein comprising 148 proteins revealed multiple enriched functional categories (Figure 6b,c). Terms related to: phosphatidylinositol phosphorylation (p-value = 0.00034, e.g., Cdc42, Erbb4, Ptk2b), acyl-CoA metabolic process (p-value = 0.0039, e.g., Acot1, Dld, Pdha1, Su- clg1), actomyosin structure organization (p-value = 0.00024, e.g., Cdc42, Csrp1, Epb41l1, Pdgfra, Zyx), cellular component assembly involved in morphogenesis (p-value = 0.00003, e.g., Ckap5, Clasp2, Csrp1, Nfasc, Ttn), but also cellular respiration (p-value = 0.0022, e.g., Dld, Mtch2, Ndufs1, Ndufv1), dendrite morphogenesis (p-value = 0.0041, e.g., Cdc42, Cdkl3, Shank1), relaxation of cardiac muscle (p-value = 0.0082, e.g., Atp2a2, Camk2g, Ttn), and negative regulation of actin ﬁlament bundle assembly (p-value = 0.0091, e.g., Clasp2, Coro2b, Dbn1, Shank1) were signiﬁcantly (p-value < 0.01) statistically enriched (Figure 6c). Similarly, we analyzed S-PALM in synaptoneurosomes isolated from DHHC7 KO female brains. A total of 1894 distinct peptides assigned to 1149 S-PALM proteins were identiﬁed and quantiﬁed with less than 1%FDR (Table S4). The S-PALM proﬁles of female DHHC7 KO and female WT synaptoneurosomes were comparatively analyzed to identiﬁed differentially modiﬁed proteins (Figure 7a). Combining the protein and site identiﬁcation data, we revealed that 173 proteins from female synaptoneurosomes are dependent on DHHC7 for their palmitoylation. Additionally, in the case of female synaptoneurosomal proteins, we observed that the lack of DHHC7 leads to an increased number of S-PALM proteins in synaptoneurosomes isolated from DHHC7 KO. To classify proteins regulated by DHHC7 in female brains, we analyzed a set of previously distinguished 173 synaptic proteins, again using the Gene Ontology and ClueGO algorithm (Figure 7a). Our analysis revealed that the proteins modulated by DHHC7 in female mice belong to some different functional categories Figure 7b,c. Importantly, several well-deﬁned categories related to brain plasticity were signiﬁcantly enriched, such as: maintenance synapse structures (p- value = 2.38 × 10−8, Bsn, Dlg2, Dlg4, Syngap1), tricarboxylic acid cycle (p-value = 0.00004, e.g., Dlat, Idh2, Ogdh, Pdha1), positive regulation of ion transport (p-value = 0.0002, e.g., Atp2b2, Cacna2d1, Dpysl2, Kcna1), regulation of synaptic plasticity (p-value = 0.00008, e.g., Dlg4, Grik2, Kras, Shank1), synaptic vesicle transport (p-value = 0.0005, e.g., Dnm1, Dnm3, Dpysl2, Rab3a), and other (Figure 7b,c). Additionally, we used the semantic similarity score to measure the redundancy of the identiﬁed terms within each dataset. The histogram distribution plots for the SS scores for all three GO terms graphs for male and female DHHC7 speciﬁc proteins are presented in Figure 7e. In conclusion, we revealed the substrate speciﬁcity of the DHHC7 protein using the PANIMoni mass spectrometry-based approach in female and male mice synapses. Int. J. Mol. Sci. 2021, 22, 6253 10 of 21 Figure 7. Analysis of female synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in WT and DHHC7 KO female synaptoneurosomes. (b) Network depicting interactions between enriched functional classes. (c) List of all enriched GO_BP functional classes (p < 0.01). (d) Percentage of terms per group analysis of enriched protein classes. (e) Comparison of semantic similarity (SS) of protein pairs from DHHC7 male and female-speciﬁc proteins for each type of Gene ontology terms, cellular components (CC), molecular functions (MF) and biological processes (BP) respectively. X-axis represents the SS score (ranges [0, 1]) and the Y-axis represents the frequency of protein pairs. Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 10 of 21 Figure 7. Analysis of female synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes data-base and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in WT and DHHC7 KO female synaptoneuro-somes. (b) Network depicting interactions between enriched functional classes. (c) List of all enriched GO_BP functional classes (p < 0.01). (d) Percentage of terms per group analysis of enriched protein classes. (e) Comparison of semantic sim-ilarity (SS) of protein pairs from DHHC7 male and female-specific proteins for each type of Gene ontology terms, cellular components (CC), molecular functions (MF) and biological processes (BP) respectively. X-axis represents the SS score (ranges [0, 1]) and the Y-axis represents the frequency of protein pairs. Additionally, we used the semantic similarity score to measure the redundancy of the identified terms within each dataset. The histogram distribution plots for the SS scores for all three GO terms graphs for male and female DHHC7 specific proteins are presented in Figure 7e. Int. J. Mol. Sci. 2021, 22, 6253 11 of 21 2.3. DHHC7 Operates Differently in Male and Female Mice One of the main topics of the present study was to resolve sex-speciﬁc differences in the synaptic S-palmitoylome. Thus, we compared the sets of proteins identiﬁed as regulated exclusively by DHHC7 in male (148 proteins) and female (173 proteins) synaptoneurosomes. Interestingly, we observed slight overlap of DHHC7-dependent synaptic proteins between males and females. We found that only 23 S-PALM proteins were commonly regulated by DHHC7 in female and male synaptoneurosomes, while 125 were exclusively modiﬁed in males and 150 in females (Figure 8a). Figure 8. Analysis of female synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes database and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in female WT and male DHHC7 KO synaptoneurosomes. (b) Networks depicting interactions between enriched functional classes for male and female DHHC7 speciﬁc proteins. (c) Summary of all enriched GO_BP functional classes for male and female DHHC7 speciﬁc proteins. (p < 0.05) with depicted fold change and signiﬁcance of enrichment. We then analyzed which pathways are regulated by the identiﬁed DHHC7 dependent S-PALM proteins in male and female synaptoneurosomes (Figure 8b). Regarding proteins regulated by DHHC7 in females, the GO analysis of biological processes showed that proteins involved in maintenance synapse structures (p-value = 2.812 × 10−7, e.g., Bsn, Dlg2, Dlg4, Dlgap1, Gphn, Rab3a, Syngap1), cellular respiration (p-value = 0.0006, e.g., Cyct, Dlat, Hif1a, Idh2, Ndufb9, Ogdh, Park7, Sdha, Slc25a18), and synaptic vesicle cycle Int. J. Mol. Sci. 2021, 22, x FOR PEER REVIEW 11 of 21 In conclusion, we revealed the substrate specificity of the DHHC7 protein using the PANIMoni mass spectrometry-based approach in female and male mice synapses. 2.3. DHHC7 Operates Differently in Male and Female Mice One of the main topics of the present study was to resolve sex-specific differences in the synaptic S-palmitoylome. Thus, we compared the sets of proteins identified as regu-lated exclusively by DHHC7 in male (148 proteins) and female (173 proteins) synap-toneurosomes. Interestingly, we observed slight overlap of DHHC7-dependent synaptic proteins between males and females. We found that only 23 S-PALM proteins were commonly regulated by DHHC7 in female and male synaptoneurosomes, while 125 were exclusively modified in males and 150 in females (Figure 8a). Figure 8. Analysis of female synaptic proteins regulated by DHHC7 using the Gene Ontology biological processes data-base and the Clue Go algorithm. (a) Venn diagram analysis of S-PALM in female WT and male DHHC7 KO synap-toneurosomes. (b) Networks depicting interactions between enriched functional classes for male and female DHHC7 specific proteins. (c) Summary of all enriched GO_BP functional classes for male and female DHHC7 specific proteins. (p < 0.05) with depicted fold change and significance of enrichment. We then analyzed which pathways are regulated by the identified DHHC7 de-pendent S-PALM proteins in male and female synaptoneurosomes (Figure 8b). Regard-ing proteins regulated by DHHC7 in females, the GO analysis of biological processes showed that proteins involved in maintenance synapse structures (p-value = 2.812 × 10−7, Int. J. Mol. Sci. 2021, 22, 6253 12 of 21 and transport (p-value = 0.00138, e.g., Bsn, Cadps2, Ctnnb1, Dnajc6, Dnm1, Dnm3, Napb, Ptpn11, Rab3a, Rap1a) are the most enriched at the synapse. Contrary to females, in males the most enriched proteins dependent on DHHC7 are pro- teins involved in processes related to: actomyosin structure organization (p-value = 0.00022, e.g., Cdc42, Clasp2, Csrp1, Epb41l1, Pdgfra, Ptk2b, Ttn, Wdr1, Zyx), membrane repolarization (p-value = 0.0037, e.g., Atp1b3, Dlg1, Kcnd3, Wdr1), and other. Next, we summarized all of the enriched terms of GO_BP (p-value < 0.05) from each set. Interestingly, we found that these sex-speciﬁc sets of proteins are involved in the regulation of the same biological processes but differ in their enrichment levels (Figure 8c). In conclusion, we demonstrated that the common signaling pathways of both sexes appear to be regulated by different S-PALM proteins. 3. Discussion Research over the last few decades has provided evidence that sex differences are more widespread than previously supposed. The interest in the differences between males and females concerns not only brain morphology and neurocognitive functions, but also epidemiology and clinical expression of the neurological and psychiatric disor- ders [2,8,12,13,38,39]. In our work, we used sophisticated proteomics and bioinformatics tools to study sex-dependent differences in biological processes related to protein S-PALM in synaptoneurosomes. We focused our research on the sex-dependent functions of one of the enzymes governing the S-PALM, the palmitoyl acyltransferase DHHC7. Despite the exceptional sensitivity in identifying S-PALM modiﬁcation sites in proteins, MS-based indirect approaches have some limitations. Rigorously reproducible sample preparation is the basis of all differential proteomic studies and is especially important in the differential analysis of unstable PTMs. It is well established that the protein S-PALM analyzed in this work possesses unique reactivity. It is also known that transpalmitoylation and depalmitoy- lation reactions, which lead to artefacts in the identiﬁcation of S-PALM proteins may occur in the presence of an activated fatty acyl thioester so that the various substrate cysteine nucleophiles can attack. False-positive results are often observed but rarely reported due to deletion at the level of data analysis. To eliminate false positives, we performed a negative control and implemented very stringent analysis criteria for selection of S-PALM proteins. It is well-known that S-PALM modulates the functions of synaptic proteins involved in neuronal development, plasticity, and also those related to synaptic dysfunction, thus leading to neurological diseases [18,20,35,40,41]. Our results showed, for the ﬁrst time, that synaptic proteins are differentially regulated by S-PALM in male and female synapses, which may be the source of sex differences in signaling pathways in the brain. Moreover, we demonstrated that DHHC7 acyltransferase acts in a sex-dependent manner and modu- lates different proteins in the brains of female and male mice. We showed that 150 proteins were exclusively S-palmitoylated by DHHC7 in females, while 125 in male synapses. Our bioinformatics analysis clearly showed that S-PALM-dependent signaling pathways found in both sexes are modulated by different S-PALM proteins. At the same time, we observed slight changes in protein expression between sexes. However, these changes did not affect S-PALM results. Sex-dependent differences have been described in the neuronal structure, dendritic branching, as well as in the morphology and density of dendritic spines [2,29,42]. Sex- related changes have been also investigated at the level of brain proteome [43–46]. Distler et al. reported a proﬁle of sex-speciﬁc synaptic proteins for different regions of the adult mouse brain, namely the hippocampus, cerebellum, prefrontal cortex, and striatum [43]. It was also shown that the intersexual alterations in hippocampal protein expression pattern are related to those observed in behavioral tests [44]. Moreover, the differences in the proteomic proﬁling of the mouse hippocampus were shown to depend not only on the sex but also on the age of animals [45]. PTMs that modulate protein function are also involved in the control and regulation of synaptic processes [47–49]. One important aspect of the research on brain sexual di- Int. J. Mol. Sci. 2021, 22, 6253 13 of 21 morphism was ﬁnding the changes in S-nitrosylation (S-NO) of mouse cortical proteins. Mass spectrometry analysis revealed that female mice showed elevated levels of S-NO proteins involved in synaptic processes, while males exhibited higher enrichment of the S-NO-dependent cytoskeletal pathways [50]. S-NO is engaged in a variety of cellular pro- cesses and, importantly, this PTM appears to compete with S-PALM for cysteine residues in proteins [51,52]. In our previous study, we reported that proteins crucial in proper synapse functioning can undergo atypical crosstalk between the S-PALM and S-NO, which can result in the development of chronic stress disorder [30]. Hence, both aforementioned PTMs of synaptic proteins appeared to be an important mechanism for controlling normal brain function in both sexes. Although it was reported that DHHC7 modulates S-PALM of several synaptic proteins, its role in the regulation of synaptic plasticity is still elusive. Recently, DHHC7 deﬁciency has been shown to impair excitatory transmission, synaptic plasticity, and hippocampal structural connectivity [34]. Additionally, the authors reported sex-related differences in the hippocampal microstructure as well as in synaptic transmission in the medial prefrontal cortex of DHHC7 KO mice [34]. Here, using the same mouse model, we showed that the observed changes might arise from divergent substrate speciﬁcity of DHHC7 acyltransferase in male and female synapses. We found that proteins speciﬁcally regulated by DHHC7 in females show signiﬁcantly higher enrichment in processes such as synapse structure maintenance, synaptic vesicle cycle and transport, cellular respiration, or learning. The molecular basis of sex-dependent changes in neuronal function remains elusive. Several data have shown that sex-differences in synaptic plasticity are regulated by both direct and indirect mechanisms of steroid hormone action [53]. A growing body of evidence shows that sex steroids can alter these processes [54–57]. It has been repeatedly shown that estrogen enhances synaptogenesis and modulates synaptic transmission [55,57,58]. Alterations in spine density in response to sex steroid hormone ﬂuctuations across the estrous cycle in female rodents have been observed [59–61]. Female neurons in general are characterized by a higher dendritic spine density in com- parison to males, suggesting that the sex difference may be due to the effects of sex steroids [62,63]. Sex-speciﬁc changes in spine density were observed in different brain regions, such as the hippocampus, nucleus accumbens, or the prefrontal cortex [63,64]. Moreover, it was shown that estradiol can potentiate excitatory synapses and attenuate inhibitory synapses exclusively in females [58,59,65–67]. In contrast, testosterone appears to inhibit long term potentiation (LTP) and dendritic sprouting in the male hippocam- pus [68]. Taken together, all these data provide evidence that sex steroid hormones regulate synaptic plasticity in a sex-speciﬁc manner. Here, we postulate that DHHC7-mediated palmitoylation of synaptic proteins underlies some of the sex-related differences in neu- ronal plasticity. Recent studies have shown that S-PALM is an important regulator of synaptic transmission, synaptic vesicle cycle and activity of synaptic receptors [18,69,70]. In our study, we identiﬁed a group of proteins palmitoylated by DHHC7 acyltransferase that are upregulated in female synapses. This group includes proteins associated with the maintenance of synapse structure, such as gephyrin, Rab3a, and Syngap1. It is widely known that the S-PALM of these proteins dynamically regulates interactions with other proteins, and thus participates in synaptic stability and trafﬁcking [71–73]. Gephyrin is an essential scaffolding protein that forms post-synaptic clusters at in- hibitory synapses [74]. S-PALM has been found to be critical for both the stable aggregation of gephyrin and inhibitory synaptic transmission [71]. A recent study demonstrated that S-PALM of gephyrin potentiated GABAergic synaptic transmission by increased amplitude of miniature inhibitory postsynaptic currents. Moreover, it was shown that DHHC7 KO female mice exhibited increased inhibitory transmission while male DHHC7 KO displayed reduced inhibitory transmission [34,71]. Our study suggests that these sex-speciﬁc effects may be regulated by S-PALM of gephyrin. Ras-related protein Rab-3A (Rab3a) is a small GTP-binding protein that plays a crucial role in exocytosis and regulation of secretion. It was reported that the expression of Rab3a Int. J. Mol. Sci. 2021, 22, 6253 14 of 21 protein in the rat pituitary gland is regulated by long term estrogen therapy [75]. However, the effects of Rab3a S-PALM have never been studied. Therefore, our ﬁndings could be the starting point of the research into the palmitoylation role in the function of this important protein. Syngap1 is a Ras/Rap GTPase activating protein that is speciﬁcally expressed in neurons and highly abundant at glutamatergic synapses in the brain [76]. Interestingly, a decrease in SynGAP1 concentration correlates with changes in the postsynaptic density (PSD) composition exclusively in females [77]. Although S-PALM of this protein was previously reported, the inﬂuence of this PTM on protein activity is unknown. The second highly enriched biological process in female is cellular respiration. Mito- chondria exhibit sex-dependent activity at various levels, including oxidative capacities, calcium handling, and resistance to oxidative stress [78]. Accumulating evidence indicates that brain mitochondria are targets for steroids [79]. For example, it was shown that the NADH-related respiratory rate was higher in females than in males [80]. Additionally, a lower level of oxidative stress was found in mitochondria of young females compared with males. Nevertheless, the brain mitochondria of young male mice have a more efﬁcient glutathione cycle than female mice [80]. In contrast, mitochondria from the female brain have higher activity of the electron transport chain, increased ATP production, and higher functional capacities [81]. In our study, we conﬁrmed that the succinate dehydrogenase (Sdha) is speciﬁcally modiﬁed by DHHC7 in female mice. Sdha belongs to the oxidative phosphorylation system in mitochondria and connects the TCA cycle to the electron transport chain. Harish et al. demonstrated that the activity of Sdha protein in the human brain was signiﬁcantly higher in females compared with males [82]. Here, we also identiﬁed a pyruvate dehydrogenase Pdha1 that catalyzes the conversion of pyruvate to acetyl-CoA and CO2 and links the glycolytic pathway to the TCA. This protein is much more active in the brains of female mice as compared with those of male [83]. We also found that, unlike female synapses, the synapses of male mice were character- ized by signiﬁcant enrichment of actomyosin structure organization, morphogenesis, and enhanced phosphatidylinositol phosphorylation and membrane repolarization. The myosin-actin interaction is essential for regulating cell growth. Actin ﬁlaments support the structure of dendritic spines, and changes in actin dynamics are known to mediate synaptic plasticity [84]. Myosin activity is particularly important during synapto- genesis, but little is known about the role of motor proteins in mature synapses and synaptic plasticity. It has been reported that estradiol and progesterone promote actin cytoskeleton remodeling, causing morphological changes in dendritic spines [15,85]. Noteworthily, testosterone is also involved in the regulation of cytoskeletal proteins in the brain [85]. In our study, we identiﬁed proteins engaged in actomyosin structure organization, modiﬁed speciﬁcally in male mice, such as Cdc42, zyxin, and Casp2. The small GTPase Cdc42 is a major regulator of actin cytoskeleton, and thus modulates neuronal morphol- ogy [86,87]. A number of studies indicate that S-PALM regulates the activity of Cdc42, promoting dendritic spine stabilization [21,88,89]. It was shown that estrogen receptor β signaling may lead to activation of the Cdc42/Rac-dependent pathway that results in changes in the number of dendritic spines [90]. Additionally, it is known that testosterone triggers the activation of Cdc42 in cancer cells [91]. Another protein we identified was, a non-receptor protein-tyrosine kinase (Ptk2b/Pyk2), which regulates reorganization of the actin cytoskeleton, cell polarization, and cell mi- gration [92]. Ptk2b/Pyk2 modulates hippocampal excitatory synaptic transmission and contributes to cognitive deﬁcits [93]. However, the sex-dependent activity of this protein has never been studied. We also identiﬁed a group of proteins including Atp1b3, Dlg1 (SAP97), and Kcnd3 that govern membrane repolarization and are speciﬁcally regulated by DHHC7 in males. SAP97 is a member of MAGUK family of proteins that play a major role in the trafﬁcking and anchoring of potassium channels to the plasma membrane. These channels are essential Int. J. Mol. Sci. 2021, 22, 6253 15 of 21 for maintaining resting membrane potential, repolarizing action potential and mediating cell excitability [94]. Interestingly, S-PALM SAP97 is directed to the PSD, where it regulates the distribution of AMPA receptors and, hence, inﬂuences the synaptic strength [95]. Our research shows that, with the exception of highly enriched biological processes that are unique to females or males, there are several biological processes that are common to both sexes but regulated by different proteins. We can conclude that the same signaling pathways/biological processes appeared to be regulated by different palmitoylated pro- teins in both sexes. This is an interesting observation because, to the best of our knowledge, the role of S-PALM in the sexual dimorphism of synaptic processes in the brain has not been described previously. Taken together, the presented mass spectrometry-based high throughput analysis revealed signiﬁcant sex-dependent differences in the protein S-PALM and the biological processes controlled by this PTM. We unraveled unique protein substrates for DHHC7 along with their speciﬁc S-PALM sites in a sex-related manner. Moreover, we demonstrated that different S-PALM proteins control the same biological processes in male and female synapses. Our data provide a unique, mechanistic understanding of the role of S-PALM in females and males. The results of this study unravel sexual dimorphism in S-PALM regulation of brain functions, especially in synaptic signaling pathways. Furthermore, our ﬁndings underline the necessity of including both males and females in all experimental studies. In many cases, sex may inﬂuence the major outcome of the research. 4. Materials and Methods 4.1. Animals and Ethical Statement In experiments, we used 90-day-old mice C57BL/6J (females and males) wild type (WT) and DHHC7 KO housed in groups [34,35]. Experiments were approved by the local institutional animal care and research advisory committee and permitted by the Lower Saxony State Ofﬁce for Consumer Protection and Food Safety (LAVES; ﬁle number 16/2230) The study was performed in accordance with all relevant guidelines and regulations of German animal protection law and with the European Directive 2010/63/EU. 4.2. Synaptoneurosomes Synaptoneurosomes were prepared from the brains of WT and DHHC7 KO (males and females Nmice/group = 3) as previously described [30,96]. Brieﬂy, after euthanasia by cervical dislocation, the mice were decapitated. Brains were homogenized with Dounce homogenizer in 3 mL of buffer A (5 mM HEPES (pH 7.4), 0.32 M sucrose, 0.2 mM ethylene- diaminetetra acetic acid (EDTA), 50 mM N-ethylmaleimide (NEM), and protease inhibitor cocktail. Nuclei and cell debris were pelleted by 5 min centrifugation at 2500× g. Super- natant was then centrifuged at 12000× g for 5 min. The obtained pellet fraction was layered over a discontinuous Ficoll (Sigma Aldrich) gradient (4%, 6%, and 13%), and centrifuged at 70,000× g for 45 min. The synaptoneurosomal fraction was collected in buffer A and centrifuged at 20,000× g for 20 min. The pellet corresponded to the synaptoneurosomes fraction. The puriﬁed synaptoneurosomes were used in all experiments. 4.3. Acyl-Biotin Exchange (ABE) To the changes in the S-PALM of proteins, acyl-biotin exchange (ABE) was used. Synaptoneurosomes were dissolved in the buffer that contained 50 mM Tris HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 4% SDS and 1% Triton X-100. Next, to block free thiol groups samples were incubated with 50 mM N-ethylmaleimide at 4 ◦C for 16 h with agitation. The thioester bonds between SH and palmitic acid were decomposed using the selective agent 1 M hydroxylamine, and newly formed thiols were blocked with 400 µM cysteine- speciﬁc biotin-HPDP (N-[6-(biotinamido)hexyl]-3(cid:48)-(2(cid:48)-pyridyldithio)propionamide). The ABE technique combined with immunoblotting analysis was used for S-PALM pattern analysis. Int. J. Mol. Sci. 2021, 22, 6253 16 of 21 4.4. PANIMoni PANIMoni analysis was performed as described previously [30]. The biotin labeling of S-PALM proteins in lysates was performed based on ABE procedure. Protein fractions that contained biotinylated proteins were digested using sequencing-grade modiﬁed trypsin (Promega V 5111) for 16 h at 37 ◦C. Digestion was terminated using a protease inhibitor cocktail. The tryptic peptide mixture was incubated with 100 µL of NeutrAvidin beads at room temperature for 1 h. The NeutrAvidin beads were washed ﬁve times in 1 mL of wash buffer (50 mM Tris (pH 7.7), 600 mM NaCl 0.2 mM EDTA). Neutravidin-bound peptides were eluted with 150 µL of elution buffer and 5 mM TCEP and concentrated in a SpeedVac. Triﬂuoroacetic acid was added to the peptide solution to achieve a ﬁnal concentration of 0.1%. The samples were analyzed by nanoLC-MS and nanoLC-MS/MS. 4.5. Mass Spectrometry The S-PALM or all proteins peptide mixture (20 µL) was applied to the nanoACQUITY UPLC Trapping Column (Waters, 186003514) using water containing 0.1% formic acid as the mobile phase and transferred to the nanoACQUITY UPLC BEH C18 Column (75 µm inner diameter; 250 mm long, Waters 186003545) using an acetonitrile gradient in the presence of 0.1% formic acid with a ﬂow rate of 250 nL/min. The column outlet was directly coupled to the ion source of the Thermo Orbitrap Elite mass spectrometer (Thermo Electron Corp., San Jose, CA, USA) working in the regime of data-dependent MS to MS/MS switch. HCD fragmentation was used. All MS runs were separated by blank runs to reduce the carry-over of peptides from previous samples. Results of measurements were processed using Mascot-Distiller 2.7.1 software (MatrixScience, London, UK, on-site license). The Mascot search engine (ver- sion 2.7.1) was used to survey data against the UniProtKB/Swiss-Prot database (Swis- sprot 2020_02; 16,905 sequences). The search parameters were set to the following: taxonomy (Mus musculus), variable modiﬁcations (cysteine carbamidomethylation or N-malemideidation, methionine oxidation, peptide tolerance (5 ppm), fragment mass tolerance (5 ppm). Enzyme speciﬁcity: trypsin with one missed or nonspeciﬁc cleav- ages permitted. The mass calibration and data ﬁltering described above were also car- ried out. The lists of the peptide sequences (SPL) that were identiﬁed in all of the LC- MS/MS runs from females (WT and DHHC7 KO) and males (WT and DHHC7 KO) synaptoneurosomal fractions were merged into one peptide list using MascotScan software (http://proteom.ibb.waw.pl/mscan/, accessed on 9 April 2021). The SPL consists of se- quences of peptides with Mascot scores exceeding the threshold value corresponding to 5% expectation value and FDR1% calculated by Mascot procedure. For proteome quantitative analysis, peptides intensities were determined as the surface of the isotopic envelope of the tagged isotopic envelopes. Before the analysis, quantitative values were normalized with LOWESS as described previously [30,95]. 4.6. Functional Bioinformatics Analysis For integrative analysis, we used the ClueGO software to observe differential proteins involved in the GO terms. The input list of proteins for each GO analysis were distinguished at the basis of proteo-mic data analysis and Venn diagram analysis. The lists of speciﬁc pro- teins are grouped in Tables S1–S4. Proteins were analyzed with ClueGO v2.6.4/CluePedia v1.6.5 to achieve complete Gene Ontological terms (GO) from our datasets [31]. ClueGO integrates GO terms and creates an organized GO/pathway term network. The statisti- cal test used for the nodes enrichment was based on right-sided hypergeometric option with a Benjamini-Hochberg correction and kappa score of 0.5. As a reference set for term enrichment calculations we utilized genes from Mus musculus genome (NCBI unique Gene identiﬁers). Enrichment of GO was conducted for different sets of proteins, and p-values < 0.05 were considered to be signiﬁcant. All ClueGO results are grouped in the Table S5. Int. J. Mol. Sci. 2021, 22, 6253 17 of 21 4.7. Semantic Similarity (SS) Analysis SS analysis between the protein pairs was performed using three genes ontological (GO) relationship graphs (CC, MF and BP) based on the SS measure proposed by Dutta et al. [33]. The ontological annotations of each protein pair were incorporated into a graph-theoretic approach for assessing the SS score. GO terms were grouped into three independent direct acyclic graphs where nodes represent speciﬁc GO terms and the links among nodes represent different hierarchical relationships (‘is_a’, ‘part_of’, and ‘has_part’) between the GO terms. To compute the SS score between two proteins, the semantic similarity was estimated for all the GO term pairs associated with the two proteins. A greater number of similar GO annotations between the two proteins indicates a higher SS score between the proteins. Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/ijms22126253/s1, This article contains Supplemental Figures S1 and S2 and Tables S1–S5. Author Contributions: Conceptualization, M.Z.-K. and J.W.; methodology, M.Z.-K., A.B.-K., K.B., F.E.M., P.K., M.R. and A.K.H.; formal analysis, M.Z.-K., A.B.-K.; K.B., A.K.H. and S.B.; investigation, M.Z.-K., A.B.-K., K.B., F.E.M. and S.B.; resources J.W.; E.P., W.Z., S.B. and I.F., writing—original draft preparation, M.Z.-K., J.W., I.F. and M.R.; supervision, M.Z.-K., J.W. and E.P.; project administration, J.W.; funding acquisition, J.W. and E.P. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Science Centre grant number 2017/26/E/NZ4/00637 to J.W. In addition, this research was funded by Deutsche Forschung Gemeinschaft (DFG), grant number PO732 to E.P and by the National Science Centre grant number 2015/19/B/NZ3/01376 to I.F. K.B. was supported by Operational Project Knowledge Education Development 2014–2020 co-ﬁnanced by European Social Fund, Project No POWR.03.02.00-00-I007/16-00 (POWER 2014–2020). Institutional Review Board Statement: Experiments were approved by the local institutional animal care and research advisory committee and permitted by the Lower Saxony State Ofﬁce for Consumer Protection and Food Safety (LAVES; ﬁle number 16/2230) The study was performed in accordance with all relevant guidelines and regulations of German animal protection law and with the European Directive 2010/63/EU. Informed Consent Statement: Not applicable. Data Availability Statement: Data are available via ProteomeXchange with identiﬁer PXD025286. Acknowledgments: We would like to thank Mass Spectrometry Laboratory IBB PAN (http://mslab- ibb.pl/en/, accessed on 9 April 2021) where all MS measurements were performed. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Tronson, N.C. Focus on females: A less biased approach for studying strategies and mechanisms of memory. Curr. Opin. Behav. Sci. 2018, 23, 92–97. [CrossRef] Beery, A.; Zucker, I. Sex bias in neuroscience and biomedical research. Neurosci. Biobehav. Rev. 2011, 35, 565–572. [CrossRef] 2. 3. Mizuno, K.; Antunes-Martins, A.; Ris, L.; Peters, M.; Godaux, E.; Giese, K.P. Calcium/calmodulin kinase kinase beta has a male-speciﬁc role in memory formation. Neuroscience 2007, 145, 393–402. [CrossRef] 4. Waters, E.M.; Thompson, L.I.; Patel, P.; Gonzales, A.D.; Ye, H.Z.; Filardo, E.J.; Clegg, D.J.; Gorecka, J.; Akama, K.T.; McEwen, B.S.; et al. G-Protein-Coupled Estrogen Receptor 1 Is Anatomically Positioned to Modulate Synaptic Plasticity in the Mouse Hippocampus. J. Neurosci. 2015, 35, 2384–2397. [CrossRef] [PubMed] Nuñez, J.L.; McCarthy, M.M. Resting intracellular calcium concentration, depolarizing Gamma-Aminobutyric Acid and possible role of local estradiol synthesis in the developing male and female hippocampus. Neuroscience 2009, 158, 623–634. [CrossRef] [PubMed] 5. 6. Monfort, P.; Gómez-Giménez, B.; Llansola, M.; Felipo, V. Gender Differences in Spatial Learning, Synaptic Activity, and Long-Term Potentiation in the Hippocampus in Rats: Molecular Mechanisms. ACS Chem. Neurosci. 2015, 6, 1420–1427. [CrossRef] [PubMed] 7. Wang, W.; Le, A.A.; Hou, B.; Lauterborn, J.C.; Cox, C.D.; Levin, E.R.; Lynch, G.; Gall, C.M. Memory-Related Synaptic Plasticity Is Sexually Dimorphic in Rodent Hippocampus. J. Neurosci. 2018, 38, 7935–7951. [CrossRef] Andreano, J.M.; Cahill, L. Sex inﬂuences on the neurobiology of learning and memory. Learn. Mem. 2009, 16, 248–266. [CrossRef] 8. 9. Hamann, S. Sex Differences in the Responses of the Human Amygdala. Neuroscience 2005, 11, 288–293. [CrossRef] Int. J. Mol. Sci. 2021, 22, 6253 18 of 21 10. Wickens, M.M.; Bangasser, D.A.; Briand, L.A. Sex Differences in Psychiatric Disease: A Focus on the Glutamate System. Front. Mol. Neurosci. 2018, 11, 197. [CrossRef] 11. Altemus, M.; Sarvaiya, N.; Epperson, C.N. Sex differences in anxiety and depression clinical perspectives. Front. Neuroendocr. 2014, 35, 320–330. [CrossRef] 12. Ecker, C.; Andrews, D.S.; Gudbrandsen, C.M.; Marquand, A.F.; Ginestet, C.E.; Daly, E.M.; Murphy, C.M.; Lai, M.-C.; Lombardo, M.V.; Ruigrok, A.N.V.; et al. Association Between the Probability of Autism Spectrum Disorder and Normative Sex-Related Phenotypic Diversity in Brain Structure. JAMA Psychiatry 2017, 74, 329–338. [CrossRef] 13. Mendrek, A.; Mancini-Marïe, A. Sex/gender differences in the brain and cognition in schizophrenia. Neurosci. Biobehav. Rev. 2016, 14. 67, 57–78. [CrossRef] Sloan, D.M.; Kornstein, S.G. Gender differences in depression and response to antidepressant treatment. Psychiatr. Clin. N. Am. 2003, 26, 581–594. [CrossRef] 15. Vierk, R.; Bayer, J.; Freitag, S.; Muhia, M.; Kutsche, K.; Wolbers, T.; Kneussel, M.; Sommer, T.; Rune, G. Structure–function–behavior relationship in estrogen-induced synaptic plasticity. Horm. Behav. 2015, 74, 139–148. [CrossRef] [PubMed] 16. Mizuno, K.; Ris, L.; Sánchez-Capelo, A.; Godaux, E.; Giese, K.P. Ca2+/Calmodulin Kinase Kinase α Is Dispensable for Brain Development but Is Required for Distinct Memories in Male, though Not in Female, Mice. Mol. Cell. Biol. 2006, 26, 9094–9104. [CrossRef] [PubMed] 17. Dachtler, J.; Hardingham, N.R.; Fox, K. The Role of Nitric Oxide Synthase in Cortical Plasticity Is Sex Speciﬁc. J. Neurosci. 2012, 32, 14994–14999. [CrossRef] 18. Zar˛eba-Kozioł, M.; Figiel, I.; Bartkowiak-Kaczmarek, A.; Włodarczyk, J. Insights Into Protein S-Palmitoylation in Synaptic Plasticity and Neurological Disorders: Potential and Limitations of Methods for Detection and Analysis. Front. Mol. Neurosci. 2018, 11, 175. [CrossRef] [PubMed] Fukata, Y.; Fukata, M. Protein palmitoylation in neuronal development and synaptic plasticity. Nat. Rev. Neurosci. 2010, 11, 161–175. [CrossRef] 19. 20. Naumenko, V.S.; Ponimaskin, E. Palmitoylation as a Functional Regulator of Neurotransmitter Receptors. Neural Plast. 2018, 2018, 5701348. [CrossRef] 21. Kang, R.; Wan, J.; Arstikaitis, P.; Takahashi, H.; Huang, K.; Bailey, A.O.; Thompson, J.X.; Roth, A.F.; Drisdel, R.C.; Mastro, R.; et al. Neural palmitoyl-proteomics reveals dynamic synaptic palmitoylation. Nat. Cell Biol. 2008, 456, 904–909. [CrossRef] 22. Prescott, G.R.; Gorleku, O.A.; Greaves, J.; Chamberlain, L.H. Palmitoylation of the synaptic vesicle fusion machinery. J. Neurochem. 2009, 110, 1135–1149. [CrossRef] 23. Pedram, A.; Razandi, M.; Deschenes, R.J.; Levin, E.R. DHHC-7 and -21 are palmitoylacyltransferases for sex steroid receptors. Mol. Biol. Cell 2012, 23, 188–199. [CrossRef] 24. Balthazart, J.; Ball, G.F. Is brain estradiol a hormone or a neurotransmitter? Trends Neurosci. 2006, 29, 241–249. [CrossRef] [PubMed] 25. Baudry, M.; Bi, X.; Aguirre, C. Progesterone–estrogen interactions in synaptic plasticity and neuroprotection. Neuroscience 2013, 26. 239, 280–294. [CrossRef] Fukata, M.; Fukata, Y.; Adesnik, H.; Nicoll, R.A.; Bredt, D.S. Identiﬁcation of PSD-95 Palmitoylating Enzymes. Neuron 2004, 44, 987–996. [CrossRef] 27. Greaves, J.; Gorleku, O.A.; Salaun, C.; Chamberlain, L.H. Palmitoylation of the SNAP25 Protein Family: Speciﬁcity and regulation by dhhc palmitoyl transferases. J. Biol. Chem. 2010, 285, 24629–24638. [CrossRef] 28. Ponimaskin, E.; Dityateva, G.; Ruonala, M.O.; Fukata, M.; Fukata, Y.; Kobe, F.; Wouters, F.S.; Delling, M.; Bredt, D.S.; Schachner, M.; et al. Fibroblast growth factor-regulated palmitoylation of the neural cell adhesion molecule determines neuronal morphogenesis. J. Neurosci. 2008, 28, 8897–8907. [CrossRef] [PubMed] 29. Dachtler, J.; Fox, K. Do cortical plasticity mechanisms differ between males and females? J. Neurosci. Res. 2016, 95, 518–526. [CrossRef] 30. Zareba-Koziol, M.; Bartkowiak-Kaczmarek, A.; Figiel, I.; Krzystyniak, A.; Wojtowicz, T.; Bijata, M.; Wlodarczyk, J. Stress-induced Changes in the S-palmitoylation and S-nitrosylation of Synaptic Proteins. Mol. Cell. Proteom. 2019, 18, 1916–1938. [CrossRef] [PubMed] 31. Bindea, G.; Mlecnik, B.; Hackl, H.; Charoentong, P.; Tosolini, M.; Kirilovsky, A.; Fridman, W.-H.; Pagès, F.; Trajanoski, Z.; Galon, J. ClueGO: A Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics 2009, 25, 1091–1093. [CrossRef] 32. Ashburner, M.; Ball, C.A.; Blake, J.A.; Botstein, D.; Butler, H.; Cherry, J.M.; Davis, A.P.; Dolinski, K.; Dwight, S.S.; Eppig, J.T.; et al. Gene Ontology: Tool for the uniﬁcation of biology. Nat. Genet. 2000, 25, 25–29. [CrossRef] [PubMed] 33. Dutta, P.; Basu, S.; Kundu, M. Assessment of Semantic Similarity between Proteins Using Information Content and Topological Properties of the Gene Ontology Graph. IEEE/ACM Trans. Comput. Biol. Bioinform. 2017, 15, 839–849. [CrossRef] 34. Hohoff, C.; Zhang, M.; Ambrée, O.; Kravchenko, M.; Buschert, J.; Kerkenberg, N.; Gorinski, N.; Galil, D.A.; Schettler, C.; Werth, K.L.V.; et al. Deﬁciency of the palmitoyl acyltransferase ZDHHC7 impacts brain and behavior of mice in a sex-speciﬁc manner. Brain Struct. Funct. 2019, 224, 2213–2230. [CrossRef] Int. J. Mol. Sci. 2021, 22, 6253 19 of 21 35. Gorinski, N.; Wojciechowski, D.; Guseva, D.; Galil, D.A.; Mueller, F.E.; Wirth, A.; Thiemann, S.; Zeug, A.; Schmidt, S.; Zareba- Kozioł, M.; et al. DHHC7-mediated palmitoylation of the accessory protein barttin critically regulates the functions of ClC-K chloride channels. J. Biol. Chem. 2020, 295, 5970–5983. [CrossRef] 36. Greaves, J.; Chamberlain, L.H. DHHC palmitoyl transferases: Substrate interactions and (patho)physiology. Trends Biochem. Sci. 2011, 36, 245–253. [CrossRef] [PubMed] 37. De, I.; Sadhukhan, S. Emerging Roles of DHHC-mediated Protein S-palmitoylation in Physiological and Pathophysiological Context. Eur. J. Cell Biol. 2018, 97, 319–338. [CrossRef] [PubMed] 38. Duman, R.S.; Sanacora, G.; Krystal, J.H. Altered Connectivity in Depression: GABA and Glutamate Neurotransmitter Deﬁcits and Reversal by Novel Treatments. Neuron 2019, 102, 75–90. [CrossRef] 39. Conrad, C.; Jackson, J.; Wise, L. Chronic stress enhances ibotenic acid-induced damage selectively within the hippocampal CA3 region of male, but not female rats. Neuroscience 2004, 125, 759–767. [CrossRef] 40. Albanesi, J.P.; Barylko, B.; DeMartino, G.N.; Jameson, D.M. Palmitoylated Proteins in Dendritic Spine Remodeling. Front. Synaptic 41. Neurosci. 2020, 12, 22. [CrossRef] Sohn, H.; Park, M. Palmitoylation-mediated synaptic regulation of AMPA receptor trafﬁcking and function. Arch. Pharm. Res. 2019, 42, 426–435. [CrossRef] 42. Tu, X.; Yasuda, R.; Colgan, L.A. Rac1 is a downstream effector of PKCα in structural synaptic plasticity. Sci. Rep. 2020, 10, 1777. [CrossRef] 43. Distler, U.; Schumann, S.; Kesseler, H.-G.; Pielot, R.; Smalla, K.-H.; Sielaff, M.; Schmeisser, M.J.; Tenzer, S. Proteomic Analysis of Brain Region and Sex-Speciﬁc Synaptic Protein Expression in the Adult Mouse Brain. Cells 2020, 9, 313. [CrossRef] 44. Yang, J.; Hu, L.-L.; Liu, L.-Y.; Zhao, L.-Y.; Hou, N.; Ni, L.; Li, Z.-F.; Wang, A.-Y.; Song, T.-S.; Huang, C. Proteomics Reveals Intersexual Differences in the Rat Brain Hippocampus. Anat. Rec. Adv. Integr. Anat. Evol. Biol. 2013, 296, 462–469. [CrossRef] 45. Bundy, J.L.; Vied, C.; Nowakowski, R.S. Sex differences in the molecular signature of the developing mouse hippocampus. BMC Genom. 2017, 18, 237. [CrossRef] 46. Bian, C.; Zhu, K.; Guo, Q.; Xiong, Y.; Cai, W.; Zhang, J. Sex differences and synchronous development of steroid receptor coactivator-1 and synaptic proteins in the hippocampus of postnatal female and male C57BL/6 mice. Steroids 2012, 77, 149–156. [CrossRef] 47. Pascovici, D.; Wu, J.X.; McKay, M.J.; Joseph, C.; Noor, Z.; Kamath, K.; Wu, Y.; Ranganathan, S.; Gupta, V.; Mirzaie, M. Clinically Relevant Post-Translational Modiﬁcation Analyses—Maturing Workﬂows and Bioinformatics Tools. Int. J. Mol. Sci. 2019, 20, 16. [CrossRef] 48. Karve, T.M.; Cheema, A.K. Small Changes Huge Impact: The Role of Protein Posttranslational Modiﬁcations in Cellular 49. Homeostasis and Disease. J. Amino Acids 2011, 2011, 207691. [CrossRef] Sunyer, B.; Diao, W.; Lubec, G. The role of post-translational modiﬁcations for learning and memory formation. Electrophoresis 2008, 29, 2593–2602. [CrossRef] 50. Khaliulin, I.; Kartawy, M.; Amal, H. Sex Differences in Biological Processes and Nitrergic Signaling in Mouse Brain. Biomedicine 2020, 8, 124. [CrossRef] 51. Ho, G.P.H.; Selvakumar, B.; Mukai, J.; Hester, L.D.; Wang, Y.; Gogos, J.A.; Snyder, S.H. S-Nitrosylation and S-Palmitoylation 52. Reciprocally Regulate Synaptic Targeting of PSD-95. Neuron 2011, 71, 131–141. [CrossRef] [PubMed] Stamler, J.S.; Toone, E.J.; Lipton, S.A.; Sucher, N.J. (S) NO Signals: Translocation, Regulation, and a Consensus Motif. Neuron 1997, 18, 691–696. [CrossRef] 53. Hyer, M.; Phillips, L.L.; Neigh, G.N. Sex Differences in Synaptic Plasticity: Hormones and Beyond. Front. Mol. Neurosci. 2018, 11, 266. [CrossRef] 54. Garrett, J.; Wellman, C. Chronic stress effects on dendritic morphology in medial prefrontal cortex: Sex differences and estrogen dependence. Neuroscience 2009, 162, 195–207. [CrossRef] 55. Monteiro-Fernandes, D.; Sousa, N.; Almeida, O.; Sotiropoulos, I. Sex Hormone Depletion Augments Glucocorticoid Induction of 56. Tau Hyperphosphorylation in Male Rat Brain. Neuroscience 2021, 454, 140–150. [CrossRef] Farrell, M.R.; Gruene, T.M.; Shansky, R.M. The inﬂuence of stress and gonadal hormones on neuronal structure and function. Horm. Behav. 2015, 76, 118–124. [CrossRef] 57. Galea, L.A.M.; Leuner, B.; Slattery, D.A. Hippocampal Plasticity during the Peripartum Period: Inﬂuence of Sex Steroids, Stress and Ageing. J. Neuroendocr. 2014, 26, 641–648. [CrossRef] 58. Kramár, E.A.; Chen, L.Y.; Brandon, N.J.; Rex, C.S.; Liu, F.; Gall, C.M.; Lynch, G.; Christopher, S.R. Cytoskeletal Changes Underlie Estrogen’s Acute Effects on Synaptic Transmission and Plasticity. J. Neurosci. 2009, 29, 12982–12993. [CrossRef] 59. Rehbein, E.; Hornung, J.; Poromaa, I.S.; Derntl, B. Shaping of the female human brain by sex hormones—A review. Neuroen- docrinology 2020, 111, 183–206. [CrossRef] 60. Lisofsky, N.; Mårtensson, J.; Eckert, A.; Lindenberger, U.; Gallinat, J.; Kühn, S. Hippocampal volume and functional connectivity changes during the female menstrual cycle. Neuroimage 2015, 118, 154–162. [CrossRef] [PubMed] 61. Protopopescu, X.; Butler, T.; Pan, H.; Root, J.; Altemus, M.; Polanecsky, M.; McEwen, B.; Silbersweig, D.; Stern, E. Hippocampal 62. structural changes across the menstrual cycle. Hippocampus 2008, 18, 985–988. [CrossRef] Shors, T.J.; Falduto, J.; Leuner, B. The opposite effects of stress on dendritic spines in male vs. female rats are NMDA receptor- dependent. Eur. J. Neurosci. 2004, 19, 145–150. [CrossRef] Int. J. Mol. Sci. 2021, 22, 6253 20 of 21 63. 64. Shors, T.J.; Chua, C.; Falduto, J. Sex differences and opposite effects of stress on dendritic spine density in the male ver-sus female hippocampus. J. Neurosci. 2001, 21, 6292–6297. [CrossRef] Forlano, P.M.; Woolley, C.S. Quantitative analysis of pre-and postsynaptic sex differences in the nucleus accumbens. J. Comp. Neurol. 2009, 518, 1330–1348. [CrossRef] [PubMed] 65. Ren, B.; Dunaevsky, A. Modeling Neurodevelopmental and Neuropsychiatric Diseases with Astrocytes Derived from Human- Induced Pluripotent Stem Cells. Int. J. Mol. Sci. 2021, 22, 1692. [CrossRef] [PubMed] 66. Zhang, M.; Weiland, H.; Schöfbänker, M.; Zhang, W. Estrogen Receptors Alpha and Beta Mediate Synaptic Transmission in the PFC and Hippocampus of Mice. Int. J. Mol. Sci. 2021, 22, 1485. [CrossRef] 67. Rabiant, K.; Antol, J.; Naassila, M.; Pierreﬁche, O. Sex difference in the vulnerability to hippocampus plasticity impairment after 68. binge-like ethanol exposure in adolescent rat: Is estrogen the key? Addict. Biol. 2021, e13002. [CrossRef] Skucas, V.A.; Duffy, A.M.; Harte-Hargrove, L.; Magagna-Poveda, A.; Radman, T.; Chakraborty, G.; Schroeder, C.E.; MacLusky, N.J.; Scharfman, H.E. Testosterone Depletion in Adult Male Rats Increases Mossy Fiber Transmission, LTP, and Sprouting in Area CA3 of Hippocampus. J. Neurosci. 2013, 33, 2338–2355. [CrossRef] [PubMed] 69. Matt, L.; Kim, K.; Chowdhury, D.; Hell, J.W. Role of Palmitoylation of Postsynaptic Proteins in Promoting Synaptic Plasticity. Front. Mol. Neurosci. 2019, 12, 8. [CrossRef] 70. Han, J.; Wu, P.; Wang, F.; Chen, J. S-palmitoylation regulates ampa receptors trafﬁcking and function: A novel insight into synaptic regulation and therapeutics. Acta Pharm. Sin. B 2015, 1–7. [CrossRef] 71. Dejanovic, B.; Semtner, M.; Ebert, S.; Lamkemeyer, T.; Neuser, F.; Lüscher, B.; Meier, J.C.; Schwarz, G. Palmitoylation of Gephyrin 72. 73. Controls Receptor Clustering and Plasticity of GABAergic Synapses. PLoS Biol. 2014, 12, e1001908. [CrossRef] Jeyifous, O.; Lin, E.I.; Chen, X.; Antinone, S.E.; Mastro, R.; Drisdel, R.; Reese, T.S.; Green, W.N. Palmitoylation regulates glutamate receptor distributions in postsynaptic densities through control of PSD95 conformation and orientation. Proc. Natl. Acad. Sci. USA 2016, 113, E8482–E8491. [CrossRef] Fukata, Y.; Dimitrov, A.; Boncompain, G.; Vielemeyer, O.; Perez, F.; Fukata, M. Local palmitoylation cycles deﬁne activity-regulated postsynaptic subdomains. J. Cell Biol. 2013, 202, 145–161. [CrossRef] 74. Choii, G.; Ko, J. Gephyrin: A central GABAergic synapse organizer. Exp. Mol. Med. 2015, 47, e158. [CrossRef] 75. Majo, G.; Lorenzo, M.J.; Blasi, J.; Aguado, F. Exocytotic protein components in rat pituitary gland after long-term estrogen administration. J. Endocrinol. 1999, 161, 323–331. [CrossRef] 76. Verpelli, C.; Schmeisser, M.J.; Sala, C.; Boeckers, T.M. Scaffold Proteins at the Postsynaptic Density. Adv. Exp. Med. Biol. 2012, 970, 29–61. [CrossRef] 77. Mastro, T.L.; Preza, A.; Basu, S.; Chattarji, S.; Till, S.M.; Kind, P.C.; Kennedy, M.B. A sex difference in the response of the rodent postsynaptic density to synGAP haploinsufﬁciency. Elife 2020, 9, e52656. [CrossRef] 78. Ventura-Clapier, R.; Moulin, M.; Piquereau, J.; Lemaire, C.; Mericskay, M.; Veksler, V.; Garnier, A. Mitochondria: A central target for sex differences in pathologies. Clin. Sci. 2017, 131, 803–822. [CrossRef] 79. Velarde, M.C. Mitochondrial and sex steroid hormone crosstalk during aging. Longev. Healthspan 2014, 3, 2. [CrossRef] 80. Arias-Reyes, C.; Losantos-Ramos, K.M.; Gonzales, M.; Furrer, D.; Soliz, J.; Christian, A.-R.; Losantos, R.K.; Marcelino, G.; Daniela, F.; Jorge, S. NADH-linked mitochondrial respiration in the developing mouse brain is sex-, age- and tissue-dependent. Respir. Physiol. Neurobiol. 2019, 266, 156–162. [CrossRef] Irwin, R.W.; Yao, J.; Hamilton, R.T.; Cadenas, E.; Brinton, R.D.; Nilsen, J. Progesterone and Estrogen Regulate Oxidative Metabolism in Brain Mitochondria. Endocrinology 2008, 149, 3167–3175. [CrossRef] [PubMed] 81. 82. Harish, G.; Venkateshappa, C.; Mahadevan, A.; Pruthi, N.; Bharath, M.M.S.; Shankar, S.K. Mitochondrial function in human brains is affected bypre-andpost mortemfactors. Neuropathol. Appl. Neurobiol. 2013, 39, 298–315. [CrossRef] 83. Gaignard, P.; Savouroux, S.; Liere, P.; Pianos, A.; Thérond, P.; Schumacher, M.; Slama, A.; Guennoun, R. Effect of Sex Differences on Brain Mitochondrial Function and Its Suppression by Ovariectomy and in Aged Mice. Endocrinology 2015, 156, 2893–2904. [CrossRef] [PubMed] 84. Bosch, M.; Castro, J.; Saneyoshi, T.; Matsuno, H.; Sur, M.; Hayashi, Y. Structural and Molecular Remodeling of Dendritic Spine Substructures during Long-Term Potentiation. Neuron 2014, 82, 444–459. [CrossRef] 85. Ehansberg-Pastor, V.; Egonzález-Arenas, A.; Piña-Medina, A.G.; Ecamacho-Arroyo, I. Sex Hormones Regulate Cytoskeletal Proteins Involved in Brain Plasticity. Front. Psychiatry 2015, 6, 165. [CrossRef] 86. Kalpachidou, T.; Spiecker, L.; Kress, M.; Quarta, S. Rho GTPases in the Physiology and Pathophysiology of Peripheral Sensory Neurons. Cells 2019, 8, 591. [CrossRef] [PubMed] Spillane, M.; Gallo, G. Involvement of Rho-family GTPases in axon branching. Small GTPases 2014, 5, e27974. [CrossRef] 87. 88. Moutin, E.; Nikonenko, I.; Stefanelli, T.; Wirth, A.; Ponimaskin, E.; De Roo, M.; Muller, D. Palmitoylation of cdc42 Promotes Spine Stabilization and Rescues Spine Density Deﬁcit in a Mouse Model of 22q11.2 Deletion Syndrome. Cereb. Cortex 2016, 27, 3618–3629. [CrossRef] 89. Wirth, A.; Chen-Wacker, C.; Wu, Y.-W.; Gorinski, N.; Filippov, M.A.; Pandey, G.; Ponimaskin, E. Dual lipidation of the brain- 90. speciﬁc Cdc42 isoform regulates its functional properties. Biochem. J. 2013, 456, 311–322. [CrossRef] Srivastava, D.P.; Woolfrey, K.M.; Liu, F.; Brandon, N.; Penzes, P. Estrogen Receptor Activity Modulates Synaptic Signaling and Structure. J. Neurosci. 2010, 30, 13454–13460. [CrossRef] Int. J. Mol. Sci. 2021, 22, 6253 21 of 21 91. Papadopoulou, N.; Charalampopoulos, I.; Alevizopoulos, K.; Gravanis, A.; Stournaras, C. Rho/ROCK/actin signaling regulates membrane androgen receptor induced apoptosis in prostate cancer cells. Exp. Cell Res. 2008, 314, 3162–3174. [CrossRef] 92. De Pins, B.; Montalban, E.; Vanhoutte, P.; Giralt, A.; Girault, J.-A. The non-receptor tyrosine kinase Pyk2 modulates acute locomotor effects of cocaine in D1 receptor-expressing neurons of the nucleus accumbens. Sci. Rep. 2020, 10, 6619. [CrossRef] 93. Giralt, A.; Brito, V.; Chevy, Q.; Simonnet, C.; Otsu, Y.; Diaz, C.C.; De Pins, B.; Coura, R.; Alberch, J.; Ginés, S.; et al. Pyk2 modulates hippocampal excitatory synapses and contributes to cognitive deﬁcits in a Huntington’s disease model. Nat. Commun. 2017, 8, 15592. [CrossRef] [PubMed] Fourie, C.; Li, D.; Montgomery, J.M. The anchoring protein SAP97 inﬂuences the trafﬁcking and localisation of multiple membrane channels. Biochim. Biophys. Acta Biomembr. 2014, 1838, 589–594. [CrossRef] [PubMed] 94. 95. Waites, C.L.; Specht, C.G.; Härtel, K.; Leal-Ortiz, S.; Genoux, D.; Li, D.; Drisdel, R.C.; Jeyifous, O.; Cheyne, J.; Green, W.N.; et al. Synaptic SAP97 Isoforms Regulate AMPA Receptor Dynamics and Access to Presynaptic Glutamate. J. Neurosci. 2009, 29, 4332–4345. [CrossRef] [PubMed] 96. Zar˛eba-Kozioł, M.; Szwajda, A.; Dadlez, M.; Wysłouch-Cieszy ´nska, A.; Lalowski, M. Global Analysis of S-nitrosylation Sites in the Wild Type (APP) Transgenic Mouse Brain-Clues for Synaptic Pathology. Mol. Cell. Proteom. 2014, 13, 2288–2305. [CrossRef]
10.3390_cells10112855	Article A Microﬂuidic Flip-Chip Combining Hydrodynamic Trapping and Gravitational Sedimentation for Cell Pairing and Fusion Gaurav Pendharkar 1,†, Yen-Ta Lu 2,3,†, Chia-Ming Chang 3, Meng-Ping Lu 3, Chung-Huan Lu 1, Chih-Chen Chen 1,4 and Cheng-Hsien Liu 1,4,* 1 Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu 30044, Taiwan; [email protected] (G.P.); [email protected] (C.-H.L.); [email protected] (C.-C.C.) 2 Chest Department, MacKay Memorial Hospital, New Taipei City 10449, Taiwan; [email protected] 3 Department of Medical Research, MacKay Memorial Hospital, New Taipei City 10449, Taiwan; [email protected] (C.-M.C.); [email protected] (M.-P.L.) Institute of Nanoengineering and Microsystems, National Tsing Hua University, Hsinchu 30044, Taiwan 4 * Correspondence: [email protected] † These authors equally contributed to this work. Abstract: Cancer cell–immune cell hybrids and cancer immunotherapy have attracted much attention in recent years. The design of efﬁcient cell pairing and fusion chips for hybridoma generation has been, subsequently, a subject of great interest. Here, we report a three-layered integrated Microﬂuidic Flip-Chip (MFC) consisting of a thin through-hole membrane sandwiched between a mirrored array of microﬂuidic channels and saw-tooth shaped titanium electrodes on the glass. We discuss the design and operation of MFC and show its applicability for cell fusion. The proposed device combines passive hydrodynamic phenomenon and gravitational sedimentation, which allows the transportation and trapping of homotypic and heterotypic cells in large numbers with pairing efﬁciencies of 75~78% and fusion efﬁciencies of 73%. Additionally, we also report properties of fused cells from cell biology perspectives, including combined ﬂuorescence-labeled intracellular materials from THP1 and A549, mixed cell morphology, and cell viability. The MFC can be tuned for pairing and fusion of cells with a similar protocol for different cell types. The MFC can be easily disconnected from the test setup for further analysis. Citation: Pendharkar, G.; Lu, Y.-T.; Chang, C.-M.; Lu, M.-P.; Lu, C.-H.; Chen, C.-C.; Liu, C.-H. A Microﬂuidic Flip-Chip Combining Hydrodynamic Trapping and Gravitational Sedimentation for Cell Pairing and Fusion. Cells 2021, 10, 2855. https:// Keywords: hydrodynamic trapping; cell pairing; dielectrophoresis; cell fusion doi.org/10.3390/cells10112855 Academic Editor: Alessandro Poggi Received: 6 September 2021 Accepted: 21 October 2021 Published: 22 October 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1. Introduction It has been hypothesized that cell fusion contributes to tumor development and spreading behavior [1]. Therefore, cellular vaccines were produced and described a decade ago, based on the cell fusion of dendritic cells (DCs) and cancer cells to offer hybrid cells sharing a united cytoplasm but keeping the identity of dual nuclei [2–4]. Cellular fusion is a process in which two or more cells are merged in an asexual way producing a hybrid cell. The process of hybridoma formation is an essential step for the development of organisms as well as human beings. Extending the use of such hybridomas has led to the development of a tool called cancer immunotherapy [5]. Cancer cell–immune cell hybrids and cancer immunotherapy has attracted much attention in recent years. In addition to cancer immunotherapy [6], applications of cell fusion have long been discussed extensively in a variety of ﬁelds such as hybridoma generation [7,8], reprogramming of somatic cells [9], and mammal cloning [10,11]. Cell fusion can be categorized primarily into three types, viz, poly-ethylene glycol (PEG) based fusion (chemical fusion) [12], fusion via viruses (biological) [13], and electroporation (physical) [14]. In vitro techniques for cell fusion have been demonstrated using microﬂuidic sys- tems. Cell electrofusion in a microﬂuidic device is a two-step process, cell pairing being the ﬁrst, followed by cell electroporation. Improving fusion efﬁciency needs careful de- sign and implementation of the mechanism for bringing the cells together (cell pairing) Cells 2021, 10, 2855. https://doi.org/10.3390/cells10112855 https://www.mdpi.com/journal/cells cells(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Cells 2021, 10, 2855 2 of 16 and the triggering mechanism to initiate membrane fusion. It is necessary to avoid un- wanted hybridomas caused by multi-cell fusion or fusion among the same cells [15,16]. Microﬂuidic-based cell operation has many advantages, such as precise manipulation and high efﬁciency in cell pairing. There are primarily three cell pairing methods—a chemical method, the use of the electric ﬁeld, and that of incorporating microstructures. Higher fusion efﬁciency depends on perfectly paired cells instead of random cell pairing, a common scenario seen in conventional methods. Several microﬂuidic designs utilizing microchannel geometrical effects, hydrodynamic forces for cell pairing have been demon- strated. The controlled pairing of partner cells has been shown using a high throughput cell pairing and a combination of cell pairing and fusion using chemical conjugation [17], ﬁeld-free microstructure assisted [18], and electric-ﬁeld assisted cell pairing with better adaptability using hydrodynamic traps or constriction trapping [19–22]. Some researchers have demonstrated single cell block printing [23] and accurate one-to-one pairing between the tumor and ﬁbroblasts [24]. Conventional fusion approaches, such as PEG-based fusion, have demonstrated a high probability of fusion among the same cell type, resulting in extensive sample processing and, ultimately, low fusion efﬁciency. Microﬂuidic-based cell electrofusion with metal electrode integration has resulted in greater efﬁciency, lower sample contamination, and increased cell survival. The contraction of the local electric ﬁeld also minimizes the Joule heating effect and increases cell viability [25]. Several microﬂuidic designs, such as the use of protruding electrodes [18,21,26–30], and the inclusion of microstructures between electrodes [19,20,31–33], helps to modify spatial distribution, which in turn shows electric ﬁeld enhancement. The electric ﬁeld constriction ensures the ﬁeld is concentrated between paired cells. The last step in this process is the electrofusion of cells. In this step, a series of short interval direct current (DC) pulses are applied between the electrodes. The process is followed by applying the alternating current (AC) signal for a short duration to ensure cell–cell contact for complete hybridoma formation. Several new techniques carrying out cell electrofusion have been reported. The electrofusion process inside droplets has been demonstrated by using homogeneous cell types [34]. Hsiao et al. and Yang et al. reported the electrofusion based on optically induced local ﬁeld enhancement [35,36]. The use of optically induced local ﬁeld enhancement for electrofusion, although a novel idea, requires expensive instrumentation for the implementation. PEG-based fusion reported by Huang et al. demonstrates centrifugal microﬂuidics for single-cell trapping [37]. Despite the efforts in improving the cell fusion process, most of the microsystems heavily rely on specialized setups (optical tweezers, high voltage power supply, etc.) as well as on the skills of personnel. Secondly, some of the methods require cellular modiﬁcation or the use of hypo-osmolar treatment for increasing cell sizes. The current cell fusion microsystems are generally designed for a particular cell size. Integrated realization of ﬂow-through channels for heterogeneous cells types with large variations is another difﬁculty in optimizing such devices for high efﬁciency and high throughput. The newly proposed methods offer a limited throughput and cell pairing efﬁciency (e.g., optical fusion, droplet microﬂuidics). Several microﬂuidic chips required pre-treatment of cells and the use of high electric ﬁelds, leading to handling difﬁculties and longer exposure time for cell manipulation. Here, we propose a Microﬂuidic Flip-Chip (MFC) using a hydrodynamic approach. The MFC consists of the hook-shaped trapping structures placed in the ﬂow-through channel. The microarrayed trapping structures placed vertically allow sequential loading for cell capture. We have demonstrated cell pairing and fusion using THP1 and A549 cells. The cells are ﬁrst trapped in hydrodynamic traps and later transferred to fusion wells by ﬂipping the chip. The electrofusion was carried out inside fusion wells with the application of the DC pulse. Then the cells were successfully retrieved from the ﬂip-chip. Finally, the fused cells were successfully retrieved and transferred from the chip to the culture plate. Additionally, we have also analyzed the properties of fused cells from a cell biology Cells 2021, 10, 2855 3 of 16 perspective, including combined ﬂuorescence-labeled intracellular materials from THP1 and A549, cell morphology, and cell viability. 2. Materials and Methods 2.1. Device Design Figure 1 shows the fabricated MFC using a soft lithography technique [38] (Figure 1A) and a graphic illustration (Figure 1B). The proposed MFC is a three-layered structure consisting of PDMS microchannels as the top layer, a thin PDMS membrane as the middle layer, and titanium electrodes on the glass as the bottom layer (Figure 1C). The MFC consists of 1000 pairs of trapping structures carefully designed by placing them into a straight channel spanned along an area of 59 mm × 37 mm. A narrow hook-shaped channel connects individual trapping sites and is cascaded in series to form an array over a large area. The hook-shaped trapping structures are designed in such a way that only one cell is captured. This array is mirrored vertically to make another set of structures. The two structures operate independently, making it a serial cell loading process. The middle layer, PDMS through-hole membrane, consists of an array of holes called fusion wells with a diameter of dw = 240 µm and distance between adjacent wells daw = 150 µm arrayed over a large area. The bottom layer consists of saw-tooth-shaped electrodes designed to form a non-uniform electric ﬁeld, separated by a distance of d = 60 µm. The adjacent electrodes are placed at a distance of dae = 240 µm. The chip is biased using the AC and DC supply to pair and fuse the cells, respectively. 2.2. Device Fabrication The MFC was fabricated using a soft lithography process. The master mold for mi- crochannels was created using the negative photoresist (SU-8). The SU-8 2015 (MicroChem, Newton, MA, USA) photoresist was spun at 3500 rpm for 30 sec to get the desired feature height of 16~18 µm. The wafer was UV exposed through a photomask for the micro-channel patterns. The development and baking steps were performed as per the manufacturer’s protocol. Following development and baking, the wafer was hard-baked at 150 ◦C for 30 min. The elastomer base and the curing agent (Sylgard 184, Dow Corning Corporation, Midland, MI, USA) were mixed in the ratio of 10:1, degassed in a vacuum chamber to remove the bubbles inside, to make the applicable PDMS. The devices were then cast by pouring this PDMS on the master mold and cured overnight at 40 ◦C. The PDMS was peeled off, and individual devices were diced. Holes for ﬂuidic connections were punctured using hole punchers. A 4-inch glass wafer was piranha cleaned, followed by a titanium deposition of 2000 Å using E-Gun evaporation to fabricate electrodes. The wafer was coated with a positive photoresist, exposed under UV light, and further developed for the desired saw-tooth-shaped pattern of electrodes. The unwanted metal was etched using a metal etchant. The middle layer is a thin PDMS membrane (11 ± 2 µm). This is a two-step process. First, SU-8 2015 (SU-8 2015, MicroChem, Newton, MA, USA) photoresist was spun at 2000~2200 rpm for 30 s to get the desired feature height of 20~22 µm. The wafer was UV exposed through a photomask for the micro-channel pattern consisting of an array of pillars. The wafer was developed and baked, followed by a hard bake at 150 ◦C for 30 min. A thin layer of Teﬂon was coated on the SU-8 master. In the next step, the PDMS pre-polymer (10:1) was diluted in n-Hexane (Sigma Aldrich, St. Louis, MO, USA) at 1:2 ratios by weight. Dilution with n-hexane decreases the viscosity of PDMS, making the formation of a thin PDMS layer possible. The wafer was then baked at 65 ◦C for 45 min. Finally, the PDMS layer with micro-channels was bonded on the membrane, and the whole structure was peeled off. Isopropyl alcohol (2-Propanol, Sigma Aldrich, St. Louis, MO, USA) was used as a solvent during the membrane peeling process. This proposed fabrication technique eliminates the handling difﬁculties making chip reproducibility eas- ier. A pictorial depiction of fabrication steps has been included in the supplementary information (Figure S1). Cells 2021, 10, 2855 4 of 16 Figure 1. Illustration of Microﬂuidic Flip-Chip (A) Image shows the fabricated MFC using a soft lithography process along with the illustration. (B) A graphic illustration of the MFC shows the three-layered structure with PDMS channels as the top layer, through-hole membrane as the middle layer, and titanium electrodes as the third bottom layer. (C) The parameters affecting chip performance—the PDMS membrane thickness (tm), the diameter of fusion well (dw), the distance between adjacent wells (daw), the distance between electrodes (d), and the distance between adjacent electrodes (dae) are as shown. Scale bar: 200 µm. 2.3. PDMS Membrane Optimization We studied two critical processing parameters controlling the thickness of the PDMS membrane, namely, the n-hexane dilution ratio and spinning speed (Figure S5). It should be noted that the thickness of the SU-8 layer consisting of a pillar structure needs to be higher than the required PDMS thickness for the perfect PDMS membrane. We observed an exponential decrease in PDMS thickness with an increase in spin speeds. Another vital parameter affecting PDMS thickness is the n-hexane dilution ratio. Among two different dilution ratios tested (1:1 and 1:2), a steeper slope was observed for a dilution ratio of 1:1. 2.4. Modeling and Simulation Cell damage can occur over time by being exposed to an electric ﬁeld. Using passive hydrodynamics, MFC allows efﬁcient cell entrapment. Researchers have studied several hydrodynamic-based trapping techniques, including passive approaches, for single-cell studies. One of the earliest and still popular ideas was published by Tan et al. [39], who envisioned employing a dynamic microarray system to capture and then release Cells 2021, 10, 2855 4 of 18 reproducibility easier. A pictorial depiction of fabrication steps has been included in the supplementary information (Figure S1). Figure 1. Illustration of Microfluidic Flip-Chip (A) Image shows the fabricated MFC using a soft lithography process along with the illustration. (B) A graphic illustration of the MFC shows the three-layered structure with PDMS channels as the top layer, through-hole membrane as the middle layer, and titanium electrodes as the third bottom layer. (C) The param-eters affecting chip performance—the PDMS membrane thickness (tm), the diameter of fusion well (dw), the distance be-tween adjacent wells (daw), the distance between electrodes (d), and the distance between adjacent electrodes (dae) are as shown. Scale bar: 200 µm. 2.3. PDMS Membrane Optimization We studied two critical processing parameters controlling the thickness of the PDMS membrane, namely, the n-hexane dilution ratio and spinning speed (Figure S5). It should be noted that the thickness of the SU-8 layer consisting of a pillar structure needs to be higher than the required PDMS thickness for the perfect PDMS membrane. We observed an exponential decrease in PDMS thickness with an increase in spin speeds. Another vital parameter affecting PDMS thickness is the n-hexane dilution ratio. Among two different dilution ratios tested (1:1 and 1:2), a steeper slope was observed for a dilution ratio of 1:1. Cells 2021, 10, 2855 5 of 16 polystyrene beads. An advantage of systems such as these is that the ﬂuid velocity is not dependent on trapping efﬁciency. For the trapping structure to work, it is necessary that the trapping structure provides a ﬂow of ﬂuid via the path to keep the volume ﬂow rate above that of the bypass path. The proposed MFC is a three-layered structure with a through-hole membrane as the middle layer. The mirrored array design with a through-hole membrane in between has not been previously reported. The original design proposed by Tan et al. [39] is independent of the ﬂow rate. However, the presence of the membrane makes ﬂuid velocity a parameter of consideration. At higher ﬂow rates (>3 µL/min), the ﬂuid from one channel ﬂowed across the mirrored channel due to the presence of a membrane. At a lower ﬂow rate (<3 µL/min), we observed a very little cross-channel ﬂuid ﬂow, and hence the leakage effects could be ignored. A ﬂuidic resistance-based model has been designed based on the volumetric ﬂow rate assuming leakage due to the added thin membrane are negligible. Once the dimensions were ﬁxed, we experimentally calculated the exact ﬂow rate needed for the chip to work at its peak efﬁciency. Figure 2A shows a schematic design of trapping sites that are symmetrical in design. Each trapping structure is made up of a trapping site (path 1: ABC) with volumetric ﬂow rate Q1 and a bypass channel (path 2: AC) through resistance R3 with a volumetric ﬂow rate Q2. The ﬂow channels are designed such that the volumetric ﬂow rate Q1 in path 1 is higher than the volumetric ﬂow rate Q2 in path 2. The majority volume of the medium-containing cells passes through path 1 as the ﬂow resistance is lower than path 2. While ﬂowing through the channels, the individual cells are trapped at the trapping sites, eventually increasing the resistance in path 1. Under this condition, most of the ﬂuid passes through path 2 as the resistance in this path is lower. In this chip, Q1/Q2 equals 1.35, a critical value for this design (detailed explanation is included in the supplementary material). The decrease in this ratio’s value would not capture the cells; however, a higher value will trap multiple cells. Detailed numerical modeling and functioning are explained in the supplementary material (Figure S2). The MFC uses the dielectrophoresis phenomenon for cell pairing. A neutral particle suspended in a dielectric medium is exposed to a non-uniform electric ﬁeld. The force exerted on the particle depends primarily on the magnitude and polarity of the charges induced on the particle under the non-uniform electric ﬁeld. We performed a Finite Element Method (FEM) simulation using COMSOL Multiphysics for the ideal electrode design for our MFC (Figure 2B and Figure S3). The normalized electric ﬁeld gradient along the path between the two electrodes was analyzed. The electric ﬁeld lines concentration for the saw-tooth shaped electrode pattern was observed to be more suitable for cell pairing and electrofusion as the highest electric ﬁeld gradient was observed at the tip of an electrode showing the non-uniform nature of the ﬁeld. A detailed comparison between the simulation of two electrode patterns has been discussed in the supplementary material. Figure 2C shows the zoom-in view of the y-component of the electric ﬁeld after applying the AC signal with a frequency of 1 MHz. Figure 2. Cont. Cells 2021, 10, 2855 6 of 18 Figure 2. (A) Fluidic resistance model for the cell trapping process. The first cell is trapped when there is less flow re-sistance. After the cell is trapped, the fluidic resistance across this short path is increased, and the cells inside the fluid follow the bypass path. (B) FEM-based simulation of saw-tooth shaped electrodes for studying the electric field for maxi-mum fusion efficiencies. (C) Zoomed-in view of the simulated electric field near the electrode tip. The cells are attracted to the tip of triangular electrodes for positive-DEP (pDEP). 2.5. Cell Preparation The A549 (ATCC® CCL185™) is a human lung carcinoma cell line. The A549 cells were cultured in 90% Ham’s F12K medium supplemented with 10% Fetal Bovine Serum (FBS; Invitrogen). The pH of the culture medium containing 2 mM L-glutamine was ad-justed to 7.2 by NaOH and HCl. The THP-1 (ATCC® TIB202™) is a human peripheral blood acute monocytic leukemia cell line. The THP-1 cells were maintained in a standard cell culture incubator (5% CO2, 95% humidity, 37 °C). Cells were cultured in the Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% Fetal Bovine Se-rum (FBS), 4.5 g/L glucose, 10 mM HEPES, and 1.0 mM sodium pyruvate, supplemented with 0.05 mM 2-mercaptoethanol. A suspension of THP-1 cells (104 cells/mL) stained in red (CellTracker™ Red CMTPX) and A549 cells (104 cells/mL) stained in green (CellTracker™ Green CMFDA) were used for a better understanding of the pairing and fusion process. Once the operating parameters were optimized, the experiments were re-peated using non-labelled cells for long-term cell culture studies. 2.6. Experimental Setup The setup consists of a syringe pump (KDS230, KDScientifc), a function generator (33220A, Agilent Technology, Santa Clara, CA, USA), and a fluorescent microscope (BX51, Olympus, Tokyo, Japan) fitted with a digital microscope camera (SPOT RT3, Diagnostic Instruments, Sterling Heights, MI, USA). The process of cell pairing and fusion was cap-tured using vendor proprietary software SPOT Advanced. The illustration for experi-mental setup and step-by-step operation has been shown in the supplementary infor-mation (Figure S4). The whole process is recorded as a series of images. The images were Cells 2021, 10, 2855 6 of 16 Figure 2. (A) Fluidic resistance model for the cell trapping process. The ﬁrst cell is trapped when there is less ﬂow resistance. After the cell is trapped, the ﬂuidic resistance across this short path is increased, and the cells inside the ﬂuid follow the bypass path. (B) FEM-based simulation of saw-tooth shaped electrodes for studying the electric ﬁeld for maximum fusion efﬁciencies. (C) Zoomed-in view of the simulated electric ﬁeld near the electrode tip. The cells are attracted to the tip of triangular electrodes for positive-DEP (pDEP). 2.5. Cell Preparation The A549 (ATCC® CCL185™) is a human lung carcinoma cell line. The A549 cells were cultured in 90% Ham’s F12K medium supplemented with 10% Fetal Bovine Serum (FBS; Invitrogen). The pH of the culture medium containing 2 mM L-glutamine was adjusted to 7.2 by NaOH and HCl. The THP-1 (ATCC® TIB202™) is a human peripheral blood acute monocytic leukemia cell line. The THP-1 cells were maintained in a standard cell culture incubator (5% CO2, 95% humidity, 37 ◦C). Cells were cultured in the Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% Fetal Bovine Serum (FBS), 4.5 g/L glucose, 10 mM HEPES, and 1.0 mM sodium pyruvate, supplemented with 0.05 mM 2-mercaptoethanol. A suspension of THP-1 cells (104 cells/mL) stained in red (CellTracker™ Red CMTPX) and A549 cells (104 cells/mL) stained in green (CellTracker™ Green CMFDA) were used for a better understanding of the pairing and fusion process. Once the operating parameters were optimized, the experiments were repeated using non-labelled cells for long-term cell culture studies. 2.6. Experimental Setup The setup consists of a syringe pump (KDS230, KDScientifc), a function generator (33220A, Agilent Technology, Santa Clara, CA, USA), and a ﬂuorescent microscope (BX51, Olympus, Tokyo, Japan) ﬁtted with a digital microscope camera (SPOT RT3, Diagnostic In- struments, Sterling Heights, MI, USA). The process of cell pairing and fusion was captured using vendor proprietary software SPOT Advanced. The illustration for experimental setup and step-by-step operation has been shown in the supplementary information (Figure S4). The whole process is recorded as a series of images. The images were then analyzed using ImageJ, a public domain image processing program for calculating cell trapping, pairing, and fusion efﬁciency. After recording the images, the images were converted to the grayscale. After adjusting the threshold, the cells were counted using the analyze particles feature in ImageJ. 2.7. Device Operation 2.7.1. Cell Trapping An essential step before cell loading is cleaning microﬂuidic devices to avoid bubble formation and bacterial contamination. The MFC was ﬁlled with deionized water (DI) and placed in a DI water-ﬁlled dish in a desiccator until the air bubbles were removed from the microchannel (approximately 20 min). The chip was later exposed to UV light to sterilize for 30 min. The MFC was conditioned with 1% Bovine Serum Albumin (BSA) and ddH2O solution to modify the surface properties of the microchannel as it prevents the adhering of cells to the microchannel wall. Cells 2021, 10, 2855 6 of 18 Figure 2. (A) Fluidic resistance model for the cell trapping process. The first cell is trapped when there is less flow re-sistance. After the cell is trapped, the fluidic resistance across this short path is increased, and the cells inside the fluid follow the bypass path. (B) FEM-based simulation of saw-tooth shaped electrodes for studying the electric field for maxi-mum fusion efficiencies. (C) Zoomed-in view of the simulated electric field near the electrode tip. The cells are attracted to the tip of triangular electrodes for positive-DEP (pDEP). 2.5. Cell Preparation The A549 (ATCC® CCL185™) is a human lung carcinoma cell line. The A549 cells were cultured in 90% Ham’s F12K medium supplemented with 10% Fetal Bovine Serum (FBS; Invitrogen). The pH of the culture medium containing 2 mM L-glutamine was ad-justed to 7.2 by NaOH and HCl. The THP-1 (ATCC® TIB202™) is a human peripheral blood acute monocytic leukemia cell line. The THP-1 cells were maintained in a standard cell culture incubator (5% CO2, 95% humidity, 37 °C). Cells were cultured in the Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 10% Fetal Bovine Se-rum (FBS), 4.5 g/L glucose, 10 mM HEPES, and 1.0 mM sodium pyruvate, supplemented with 0.05 mM 2-mercaptoethanol. A suspension of THP-1 cells (104 cells/mL) stained in red (CellTracker™ Red CMTPX) and A549 cells (104 cells/mL) stained in green (CellTracker™ Green CMFDA) were used for a better understanding of the pairing and fusion process. Once the operating parameters were optimized, the experiments were re-peated using non-labelled cells for long-term cell culture studies. 2.6. Experimental Setup The setup consists of a syringe pump (KDS230, KDScientifc), a function generator (33220A, Agilent Technology, Santa Clara, CA, USA), and a fluorescent microscope (BX51, Olympus, Tokyo, Japan) fitted with a digital microscope camera (SPOT RT3, Diagnostic Instruments, Sterling Heights, MI, USA). The process of cell pairing and fusion was cap-tured using vendor proprietary software SPOT Advanced. The illustration for experi-mental setup and step-by-step operation has been shown in the supplementary infor-mation (Figure S4). The whole process is recorded as a series of images. The images were Cells 2021, 10, 2855 7 of 16 The MFC has two inlets and two outlets for the independent loading and unloading operation of cells. The optimal design parameters ensure single-cell trapping and prevent channel clogging throughout the cell loading and unloading process. Figure 3 (Step 1) shows the cell trapping of THP-1 and A549 cells. The cell loading process of THP-1 was accomplished by injecting 1ml of fusion buffer containing cells from Inlet A and collecting the cells from the outlet at a ﬂow rate of 1.5 µL/min using a syringe pump. After most of the trapping sites were ﬁlled with cells, a cleaning step was performed. A fusion buffer was ﬂushed into the channels at a reduced ﬂow rate (0.5 µL/min). This process ensured the removal of excess cells while maintaining the already ﬁlled trapping sites as intact. Next, the second type of cell, i.e., A549, was introduced from Inlet B with a ﬂow rate of 1.5 µL/min. The cell loading process was continued until most of the trapping sites were ﬁlled. A washing step was performed for the removal of excess cells using a fusion buffer. No signiﬁcant effect concerning cell unloading from the traps was observed during the second step of cell loading. Figure 3. Illustration of MFC device operation for cell loading and pairing. Step 1: THP-1 cells and A549 are loaded from the respective inlets sequentially, followed by a washing step to ﬂush out excess cells. Step 2: The chip is ﬂipped, and the cells are transferred to the fusion wells. Step 3: The cell–cell contact is achieved by applying the AC signal (Frequency: 1MHz, Amplitude: 10Vpp). The illustration of an objective lens and an arrow indicates the direction from which the image was captured. Scale bar: 200 µm. 2.7.2. Cell Pairing and Fusion After successful cell trapping, the tubing was removed from the chip, and the holes were plugged. The MFC was ﬂipped gently in which the trapping sites are now on the channel’s ceiling, and the fusion wells are on the ﬂoor, allowing the captured cells to fall Cells 2021, 10, 2855 8 of 18 Figure 3. Illustration of MFC device operation for cell loading and pairing. Step 1: THP-1 cells and A549 are loaded from the respective inlets sequentially, followed by a washing step to flush out excess cells. Step 2: The chip is flipped, and the cells are transferred to the fusion wells. Step 3: The cell–cell contact is achieved by applying the AC signal (Frequency: 1MHz, Amplitude: 10Vpp). The illustration of an objective lens and an arrow indicates the direction from which the image was captured. Scale bar: 200 µm. 2.7.2. Cell Pairing and Fusion After successful cell trapping, the tubing was removed from the chip, and the holes were plugged. The MFC was flipped gently in which the trapping sites are now on the channel’s ceiling, and the fusion wells are on the floor, allowing the captured cells to fall off from the trapping sites to the fusion wells by gravity. After the cells were settled in a fusion well, the uncaptured cells were washed away from the device by slowly injecting a fusion buffer (Figure 3, Step 2). The cell trapping and transferring procedures require 15–18 min approximately. After flipping the chip, all the cells are transferred to the fusion well. An alignment signal (Amplitude: 10Vpp, Frequency: 1 MHz) was applied between the electrode array. (Figure 3, Step 3). A low conductivity buffer (<200 μS/cm) solution is necessary for the dielectrophoresis phenomenon to occur. It also helps to have better cell viability. The alignment signal induces positive DEP force on the cells aligning them as pairs with high efficiency. In the next step, a DC pulse (Duration: 100 μs, Number of pulses: 10) was applied to induce temporary cell membrane perforation. The optimum value of the DC field helps in cell membrane reconstruction because of the cell’s self-re-covering and resealing ability. In addition to this, better cell viability is achieved due to the cytoplasm exchange between paired cells. In the next step, the fused cells are removed by the flowing buffer solution through Outlet A and Inlet B, and the cells are collected from Inlet A and Outlet B. Cells 2021, 10, 2855 8 of 16 off from the trapping sites to the fusion wells by gravity. After the cells were settled in a fusion well, the uncaptured cells were washed away from the device by slowly injecting a fusion buffer (Figure 3, Step 2). The cell trapping and transferring procedures require 15–18 min approximately. After ﬂipping the chip, all the cells are transferred to the fusion well. An alignment signal (Amplitude: 10Vpp, Frequency: 1 MHz) was applied between the electrode array. (Figure 3, Step 3). A low conductivity buffer (<200 µS/cm) solution is necessary for the dielectrophoresis phenomenon to occur. It also helps to have better cell viability. The alignment signal induces positive DEP force on the cells aligning them as pairs with high efﬁciency. In the next step, a DC pulse (Duration: 100 µs, Number of pulses: 10) was applied to induce temporary cell membrane perforation. The optimum value of the DC ﬁeld helps in cell membrane reconstruction because of the cell’s self- recovering and resealing ability. In addition to this, better cell viability is achieved due to the cytoplasm exchange between paired cells. In the next step, the fused cells are removed by the ﬂowing buffer solution through Outlet A and Inlet B, and the cells are collected from Inlet A and Outlet B. 2.8. Image Acquisition and Analysis The cell pairing and fusion phenomenon can be observed by imaging ﬂuorescence exchange among the cells. The array of images were captured at different times, such as after ﬂipping the chip, after applying the AC signal, and after applying the DC pulse. The application of the DC pulse made the cell membrane unstable and formed reversible membrane pores, leading cells in physical contact to achieve electrofusion. 2.9. Statistical Analysis All experiments were performed in triplicate and the data are presented as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for the comparison of each group. The p-value has been represented in each ﬁgure. 3. Results 3.1. Cell Pairing The MFC uses a two-step protocol for the cell loading process. In the ﬁrst step, THP-1 cells are loaded from Inlet A, followed by a subsequent wash step to remove excess cells. The next step continues with the loading of A549 cells from Inlet B. The washing step is repeated to ﬂush out excess A549 cells in the ﬂow-through channels. The process of cell trapping and pairing takes about 18 min. 3.2. Cell Electrofusion The cell electrofusion process in our device was studied by analyzing ﬂuorescence signals at different timestamps (Figure 4). The complete cell fusion undergoes various stages such as cell–cell contact, electroporation of membranes by short DC pulse, and ﬁnally, the exchange of cytosols. The electrofusion process is analyzed by observing the ﬂuorescence dye exchange over properly paired cells using ImageJ. Fusion efﬁciency is deﬁned as the total number of cells showing ﬂuorescence exchange among properly paired cells (cell pairing consisting of one THP1 and one A549 cell). An RGB histogram for red (THP-1), green (A549) and yellow (fused cell) colors can be represented by RGB codes as (255, 0, 0), (0, 255, 0), and (255, 255, 0), respectively. At time t = 0, the ﬂuorescence image shows red and green colored stained cells distinctly. After applying the DC pulse, the color of the cell changes to yellow, indicating a complete cell fusion process, as shown in Figure 4. Cells 2021, 10, 2855 9 of 16 Figure 4. A magniﬁed image sequence for cell electrofusion process along with brightﬁeld image, ﬂuorescence image, and RGB histogram. Step 1: Cell loading process with trapped THP-1 and A549 cells. Step 2: Transfer of cells into the fusion well by ﬂipping the MFC. Step 3: Cell–cell contact is achieved by applying AC electric ﬁeld. Step 4: After applying the DC pulse, the color of the fused cell changed to yellow. The RGB histogram represents the position and respective color of cells. The white arrows in the ﬂuorescence images indicate the region across which the histogram is plotted. Scale bar: 50 µm. Effect of Electric Field on Fusion Efﬁciency A biological cell is modeled as a particle having a conductive interior (cytoplasm) encapsulated by a thin insulating layer (lipid bilayer membrane). Equation (1) deﬁnes the membrane electric voltage (Vm) of an isolated spherical cell under the inﬂuence of an electric ﬁeld (E0). Vm = aE0cos(θ[1 − exp(−τ/τm)]) (1) 3 2 As can be seen from the equation, the presence of the time component explains the exponential effect on the voltage generated across the plasma membrane. The angle between normal to the membrane and the electric ﬁeld vector is deﬁned as θ, and τm is the time constant of membrane charging given by τm = aCm (cid:18) 1 σcell + 1 σext (cid:19) (2) where, σcell and σext are the interior and exterior conductivity of the cell. The Cm is the membrane capacitance per unit area with a typical value of 10 mF/m2. When the Cells 2021, 10, 2855 10 of 18 Figure 4. A magnified image sequence for cell electrofusion process along with brightfield image, fluorescence image, and RGB histogram. Step 1: Cell loading process with trapped THP-1 and A549 cells. Step 2: Transfer of cells into the fusion well by flipping the MFC. Step 3: Cell–cell contact is achieved by applying AC electric field. Step 4: After applying the DC pulse, the color of the fused cell changed to yellow. The RGB histogram represents the position and respective color of cells. The white arrows in the fluorescence images indicate the region across which the histogram is plotted. Scale bar: 50 µm. Effect of Electric Field on Fusion Efficiency A biological cell is modeled as a particle having a conductive interior (cytoplasm) encapsulated by a thin insulating layer (lipid bilayer membrane). Equation (1) defines the membrane electric voltage (𝑉𝑚) of an isolated spherical cell under the influence of an elec-tric field (𝐸0). 𝑉𝑚=32𝑎𝐸0𝑐𝑜𝑠(𝜃[1−𝑒𝑥𝑝(−𝜏𝜏𝑚⁄)]) (1) As can be seen from the equation, the presence of the time component explains the exponential effect on the voltage generated across the plasma membrane. The angle be-tween normal to the membrane and the electric field vector is defined as 𝜃, and 𝜏𝑚 is the time constant of membrane charging given by Cells 2021, 10, 2855 10 of 16 pulse duration is equal to or higher than 5∗τm, the membrane gets charged entirely, and Equation (2) is reduced to Equation (3) given as Vm(θ) = 3 2 aEextcos(θ) (3) The temporary membrane perforation occurs if the value is 1V at room temperature. This critical value (Vc) is of particular interest since it helps to optimize the values of pulse strength and duration required for efﬁcient cell electrofusion. However, if the membrane voltage is higher than the critical value, the cell membrane cannot be recovered, and there is permanent damage. After the washing step, the MFC was ﬂipped and the AC signal was applied for performing the cell pairing process. We observed a pairing efﬁciency of around 87% (~610 cells/experiment). Next, we studied the effect of the electric ﬁeld on the fusion efﬁciency among perfectly paired single THP-1 and single A549 cells. We observed a critical value of 0.7 kV/cm (Figure 5A), at which maximum fusion efﬁciency of 72.8% (~445 cells/experiment) was achieved. The fusion efﬁciency followed a downward curve beyond this value, with an increase in the electric ﬁeld value. At an electric ﬁeld less than 0.7 kV/cm, we observed incomplete cell fusion as the electric ﬁeld was insufﬁcient for the membrane perforation. As the voltage increased, we observed a decrease in the fusion efﬁciency with an electric ﬁeld greater than 0.7 kV/cm. The decrease in the fusion efﬁciency was due to the incomplete cell fusion observed due to the permanent membrane perforation in which the damage is caused to the cell due to a high electric ﬁeld. We also studied the effect of different duration of DC pulse on fusion efﬁciency, as shown in Figure 5B. We chose a range of values from 50 µs to 150 µs in three equal intervals with the number of pulses ﬁxed to 10. The fusion efﬁciency was relatively lower with a 50 µs pulse duration as we observed incomplete cell fusion. We then increased the pulse duration to 150 µs and observed cell damage to a large extent due to prolonged pulse exposure. The maximum fusion efﬁciency was achieved when the pulse duration was 100 µs, as shown in Figure 5B. The higher time duration (greater than 100 µs) of the DC pulse resulted in permanent damage to the cells. Figure 5. Effect of electric ﬁeld on fusion efﬁciency (A) The maximum fusion efﬁciency of 72.8% was achieved when the electric ﬁeld strength was 0.7 kV/cm. (B) The effect of the duration of DC pulses applied was also studied. The DC pulse duration varied from 50 µs to 150 µs. The number of pulses was ﬁxed to 10. The highest fusion efﬁciency was observed when the duration of the pulse was 100 µs. Error bars, mean ± s.d., (n = 3) (p < 0.05). Cells 2021, 10, 2855 11 of 18 𝜏𝑚=𝑎𝐶𝑚(1𝜎𝑐𝑒𝑙𝑙+1𝜎𝑒𝑥𝑡) (2) where, 𝜎𝑐𝑒𝑙𝑙 and 𝜎𝑒𝑥𝑡 are the interior and exterior conductivity of the cell. The 𝐶𝑚 is the membrane capacitance per unit area with a typical value of 10 mF/m2. When the pulse duration is equal to or higher than 5∗𝜏𝑚, the membrane gets charged entirely, and Equa-tion (2) is reduced to Equation (3) given as 𝑉𝑚(𝜃)= 32 𝑎𝐸𝑒𝑥𝑡𝑐𝑜𝑠(𝜃) (3) The temporary membrane perforation occurs if the value is 1V at room temperature. This critical value (𝑉𝑐) is of particular interest since it helps to optimize the values of pulse strength and duration required for efficient cell electrofusion. However, if the membrane voltage is higher than the critical value, the cell membrane cannot be recovered, and there is permanent damage. After the washing step, the MFC was flipped and the AC signal was applied for per-forming the cell pairing process. We observed a pairing efficiency of around 87% (~610 cells/experiment). Next, we studied the effect of the electric field on the fusion efficiency among perfectly paired single THP-1 and single A549 cells. We observed a critical value of 0.7 kV/cm (Figure 5A), at which maximum fusion efficiency of 72.8% (~445 cells/exper-iment) was achieved. The fusion efficiency followed a downward curve beyond this value, with an increase in the electric field value. At an electric field less than 0.7 kV/cm, we observed incomplete cell fusion as the electric field was insufficient for the membrane perforation. As the voltage increased, we observed a decrease in the fusion efficiency with an electric field greater than 0.7 kV/cm. The decrease in the fusion efficiency was due to the incomplete cell fusion observed due to the permanent membrane perforation in which the damage is caused to the cell due to a high electric field. We also studied the effect of different duration of DC pulse on fusion efficiency, as shown in Figure 5B. We chose a range of values from 50 µs to 150 µs in three equal intervals with the number of pulses fixed to 10. The fusion efficiency was relatively lower with a 50 µs pulse duration as we observed incomplete cell fusion. We then increased the pulse duration to 150 µs and ob-served cell damage to a large extent due to prolonged pulse exposure. The maximum fu-sion efficiency was achieved when the pulse duration was 100 µs, as shown in Figure 5B. The higher time duration (greater than 100 µs) of the DC pulse resulted in permanent damage to the cells. Cells 2021, 10, 2855 11 of 16 3.3. Effect of Membrane Thickness For THP-1 and A549 cells to be collected inside the fusion well, the membrane and the channels should be precisely aligned. Several factors are also responsible for the process of cell trapping, cell retention, and cell transfer. We studied the effects of membrane thickness to optimize the operation of MFC. After ﬁxing the height of the channel (15 µm), two sets of devices with PDMS membrane thicknesses of 12 µm and 20 µm were fabricated. The devices were tested under different ﬂow conditions. We observed a considerable decrease in trapping efﬁciency with an increase in the membrane thickness for both cell types. At higher membrane thickness, the cells could not get trapped as they slipped through the trapping area due to excessive ﬂuid leakage (Figure 6A,B). After observations from the previous step, we ﬁxed the value of the PDMS membrane to 12 µm. Though we observed an increasing trend of trapping efﬁciencies, we needed to extract exact ﬂow rate values for maximum efﬁciency. The presence of a PDMS membrane makes the ﬂow rate a parameter of consideration. We tested our device for ﬂow rates between 0.5 µL/min and 3 µL/min. As seen from the graphs in Figure 6A,B, the membrane thickness of 20 µm leads to very poor trapping efﬁciencies as most of the cells were ﬂushed away from the trapping sites. At membrane thickness of 12 µm with higher ﬂow rates (>1.5 µL/min), cells do not get trapped at trapping sites and ﬂow through the membrane area. However, at lower ﬂow rates (<0.5 µL/min), we observed multiple cell trapping and some cell clogging too. An average trapping efﬁciency of 77% (~770 cells/experiment) was observed at a ﬂow rate of 1.5 µL/min for both THP-1 and A549 cells. Figure 6. Factors responsible for THP-1 and A549 cell trapping. (A) Effect of cell loading ﬂow rate on the THP-1 trapping efﬁciency. (B) Effect of cell loading ﬂow rate on the A549 trapping efﬁciency (C,D) Effect of washing ﬂow rate on cell retention of THP-1 and A549. An efﬁciency of 95% and 91% were achieved for THP-1 cells and A549 cells, respectively. Error bars, mean ± s.d., (n = 3) (p < 0.05). This optimization step was carried out on non-ﬂuorescence labelled cells. Cells 2021, 10, 2855. https://doi.org/10.3390/cells10112855 www.mdpi.com/journal/cells . Effect of electric field on fusion efficiency (A) The maximum fusion efficiency of 72.8% was achieved when the electric field strength was 0.7 kV/cm. (B) The effect of the duration of DC pulses applied was also studied. The DC pulse duration varied from 50 µs to 150 µs. The number of pulses was fixed to 10. The highest fusion efficiency was observed when the duration of the pulse was 100 µs. Error bars, mean ± s.d., (n = 3) (p < 0.05). 3.3. Effect of Membrane Thickness For THP-1 and A549 cells to be collected inside the fusion well, the membrane and the channels should be precisely aligned. Several factors are also responsible for the pro-cess of cell trapping, cell retention, and cell transfer. We studied the effects of membrane thickness to optimize the operation of MFC. After fixing the height of the channel (15 µm), two sets of devices with PDMS membrane thicknesses of 12 µm and 20 µm were fabri-cated. The devices were tested under different flow conditions. We observed a considerable decrease in trapping efficiency with an increase in the membrane thickness for both cell types. At higher membrane thickness, the cells could not get trapped as they slipped through the trapping area due to excessive fluid leakage (Figure 6A,B). After observations from the previous step, we fixed the value of the PDMS membrane to 12 µm. Though we observed an increasing trend of trapping efficiencies, we needed to extract exact flow rate values for maximum efficiency. The presence of a PDMS membrane makes the flow rate a parameter of consideration. We tested our device for flow rates between 0.5 µL/min and 3 µL/min. As seen from the graphs in Figure 6A,B, the membrane thickness of 20 µm leads to very poor trapping efficiencies as most of the cells were flushed away from the trapping sites. At membrane thickness of 12 µm with higher flow rates (>1.5 µL/min), cells do not get trapped at trapping sites and flow through the membrane area. However, at lower flow rates (<0.5 µL/min), we observed multiple cell trapping and some cell clogging too. An average trapping efficiency of 77% (~770 cells/ex-periment) was observed at a flow rate of 1.5 µL/min for both THP-1 and A549 cells. Cells 2021, 10, 2855 12 of 16 3.4. Effect of Washing Flow Rate Excess cells need to be washed away to avoid multi-cell trapping and channel clog- ging. The washing ﬂow rate is another crucial parameter; if chosen, a wrong value reduces efﬁciency. The cell retention value was calculated based on the percentage of cells which remained at trapped sites after the washing step. During the cell loading process, THP-1 cells were ﬁrst loaded, followed by A549. For the washing step, too, washing of THP-1 cells was performed ﬁrst. A relatively slower washing ﬂow rate ensured maxi- mum cell retention at trapping sites. As seen from the graph, cell retention for THP-1 cells was 95% (~731 cells/experiment) at a ﬂow rate of 0.9 µL/min, and that for A549 (~700 cells/experiment) was 91% at 0.5 µL/min (Figure 6C,D). 3.5. Characterization of Fused Cells A ﬂuorescence exchange within a few seconds after the application of the DC pulse was observed. A shift in color indicated that the ﬂuorescence dye had moved to the nucleus. An AC signal was applied after complete cell fusion for the membrane reorganization. The process of cell electrofusion required approximately 35 min. The fused cells were transferred to a single well plate for further culture. A live/dead assay was performed to investigate the viability of cells in a single well plate. The cells were stained (LIVE/DEAD™ Cell Imaging Kit, Thermo Fisher Scientiﬁc, USA) and were observed on day 0, day 2, and day 4. As seen from Figure 7, the cell viability increased from day 0 to day 4. Figure 7. Live/dead assay. The fused cells were transferred from the chip to a single well plate for long-term cell culture. The cells were treated with a live/dead assay kit on day 0, day 2, and day 4, respectively. The high cell viability was observed even on day 4. This study was performed using non-ﬂuorescence labelled cells. Scale bar: 100 µm. We also performed an experiment by seeding the THP-1 and A549 cells together for 4 days as a negative control for the cell electrofusion experiment. We did not observe any signiﬁcant ﬂuorescence exchange over the cells indicating that negligible fusion occurred without any fusing stimuli (Figure S6). Cells 2021, 10, 2855 2 of 18 Figure 6. Factors responsible for THP-1 and A549 cell trapping. (A) Effect of cell loading flow rate on the THP-1 trapping efficiency. (B) Effect of cell loading flow rate on the A549 trapping efficiency (C,D) Effect of washing flow rate on cell retention of THP-1 and A549. An efficiency of 95% and 91% were achieved for THP-1 cells and A549 cells, respectively. Error bars, mean ± s.d., (n = 3) (p < 0.05). This optimization step was carried out on non-fluorescence labelled cells. 3.4. Effect of Washing Flow Rate Excess cells need to be washed away to avoid multi-cell trapping and channel clog-ging. The washing flow rate is another crucial parameter; if chosen, a wrong value reduces efficiency. The cell retention value was calculated based on the percentage of cells which remained at trapped sites after the washing step. During the cell loading process, THP-1 cells were first loaded, followed by A549. For the washing step, too, washing of THP-1 cells was performed first. A relatively slower washing flow rate ensured maximum cell retention at trapping sites. As seen from the graph, cell retention for THP-1 cells was 95% (~731 cells/experiment) at a flow rate of 0.9 µL/min, and that for A549 (~700 cells/experi-ment) was 91% at 0.5 µL/min (Figure 6C,D). 3.5. Characterization of Fused Cells A fluorescence exchange within a few seconds after the application of the DC pulse was observed. A shift in color indicated that the fluorescence dye had moved to the nu-cleus. An AC signal was applied after complete cell fusion for the membrane reorganiza-tion. The process of cell electrofusion required approximately 35 min. The fused cells were transferred to a single well plate for further culture. A live/dead assay was performed to investigate the viability of cells in a single well plate. The cells were stained (LIVE/DEAD™ Cell Imaging Kit, Thermo Fisher Scientific, USA) and were observed on day 0, day 2, and day 4. As seen from Figure 7, the cell via-bility increased from day 0 to day 4. Figure 7. Live/dead assay. The fused cells were transferred from the chip to a single well plate for long-term cell culture. The cells were treated with a live/dead assay kit on day 0, day 2, and day 4, respectively. The high cell viability was observed even on day 4. This study was performed using non-fluorescence labelled cells. Scale bar: 100 µm. Cells 2021, 10, 2855 13 of 16 3.6. Cell Viability in 96-Well Plate In addition to the live/dead cell assay, we studied the viability of fused cells using PrestoBlue® reagent. We ﬁrst used 96 well plates to culture THP-1, A549, and fused cells. A similar cell density per well was seeded in 100 µL. We evaluated the cell viability of THP-1, A549, and fused cells using the PrestoBlue assay. Resazurin (λmax.abs = 600 nm) in the PrestoBlue® reagent, a nonﬂuorescent blue compound, can be reduced in live cells by metabolism to resoruﬁn (λmax.abs = 570 nm), which is red in color and highly ﬂuorescent. Since the number of metabolically active cells proportionally correlates with the reduction level, the absorbance readings can be converted and expressed as the percentage reduction of the PrestoBlue® reagent, indicating the relative cell viability. We added 10 µL of the PrestoBlue® reagent to each well and incubated the plate for 2 hrs. for better readability. The absorbance was observed at 570 nm with a reference wavelength of 600 nm using an ELISA reader (BioTek 800TS) as mentioned in the datasheet. As seen from Figure 8A to Figure 8B, the absorbance shows a steady increase from day 0 to day 4 for THP-1 and A549, respectively. The steady increase in absorbance values indicates that the cells are viable. On the other hand, for fused cells (Figure 8D), the absorbance curve also followed a steady path over 4 days. The absorbance value indicates metabolically active and viable cells. Figure 8. Cell viability using PrestoBlue® reagent (A) The graph showing cell density and absorbance was plotted for THP-1 (B) The graph showing cell density and absorbance was plotted for A549 from day 0 to day 4. (C,D) The graph showing cell density and absorbance was plotted for fused cells from day 0 to day 4. The absorbance values indicate that the cells are viable. Error bars, mean ± s.d., (n = 2) (p < 0.05). This study was performed using non-ﬂuorescence labelled cells. 4. Discussion New means of cell pairing and fusion in microﬂuidic systems are increasingly impor- tant since microﬂuidics is emerging as an important domain in cancer immunotherapy. The ability to obtain ideal conditions in a well-deﬁned microﬂuidic environment for precise cell manipulation is promising for cell-based studies. We have proposed a MicroFluidic Flip-Chip combining hydrodynamic trapping and gravitational sedimentation. The cell Cells 2021, 10, 2855 3 of 18 We also performed an experiment by seeding the THP-1 and A549 cells together for 4 days as a negative control for the cell electrofusion experiment. We did not observe any significant fluorescence exchange over the cells indicating that negligible fusion occurred without any fusing stimuli (Figure S6). 3.6. Cell Viability in 96-Well Plate In addition to the live/dead cell assay, we studied the viability of fused cells using PrestoBlue® reagent. We first used 96 well plates to culture THP-1, A549, and fused cells. A similar cell density per well was seeded in 100 µL. We evaluated the cell viability of THP-1, A549, and fused cells using the PrestoBlue assay. Resazurin (λmax.abs = 600 nm) in the PrestoBlue® reagent, a nonfluorescent blue compound, can be reduced in live cells by metabolism to resorufin (λmax.abs = 570 nm), which is red in color and highly fluorescent. Since the number of metabolically active cells proportionally correlates with the re-duction level, the absorbance readings can be converted and expressed as the percentage reduction of the PrestoBlue® reagent, indicating the relative cell viability. We added 10 µL of the PrestoBlue® reagent to each well and incubated the plate for 2 hrs. for better reada-bility. The absorbance was observed at 570 nm with a reference wavelength of 600 nm using an ELISA reader (BioTek 800TS) as mentioned in the datasheet. As seen from Figure 8A to Figure 8B, the absorbance shows a steady increase from day 0 to day 4 for THP-1 and A549, respectively. The steady increase in absorbance values indicates that the cells are viable. On the other hand, for fused cells (Figure 8D), the absorbance curve also fol-lowed a steady path over 4 days. The absorbance value indicates metabolically active and viable cells. Figure 8. Cell viability using PrestoBlue® reagent (A) The graph showing cell density and absorbance was plotted for THP-1 (B) The graph showing cell density and absorbance was plotted for A549 from day 0 to day 4. (C,D) The graph showing cell density and absorbance was plotted for fused cells from day 0 to day 4. The absorbance values indicate that the cells are viable. Error bars, mean ± s.d., (n = 2) (p < 0.05). This study was performed using non-fluorescence labelled cells. Cells 2021, 10, 2855 14 of 16 trapping process eliminates the need for any pre-treatment, e.g., hypo-osmolar treatment. The hydrodynamic trapping process is fast and efﬁcient, with trapping efﬁciencies of up to 77%. The MFC chip introduced an arrayed hydrodynamic pairing structure connected internally by a thin membrane. Although the presence of a thin membrane might look like an issue concerning leakage between the two trapping channels, a thorough experimental veriﬁcation has been carried out, and it was observed that at optimum ﬂow rates, as dis- cussed in the previous sections, there is a negligible cross channel ﬂow observed at lower ﬂow rates ensuring that the cell trapping process is unaffected. In this study, we have obtained up to 78% efﬁciency for pairing single THP-1 and single A549 cells, with up to 95% retention efﬁciencies after ﬂipping the chip. The process of fabricating a thin through-hole PDMS membrane is a crucial step and requires precise handling. The integration of electrodes for performing electrofusion gives an edge in pairing and fusion efﬁciencies compared with traditional bench-top and chemical-based fusion processes. The overall fusion efﬁciency of 73% is achieved using the proposed MFC. High cell viability was observed even on day 4. Although the working of the chip seems to be a lengthy process, the overall time required from beginning to cell fusion takes about 35 min. This has been a considerable improvement over the traditional PEG-based fusion process. A lot of other devices have been proposed utilizing droplet microﬂuidics. Although, a novel technique, the efﬁciency is quite low at this point. A few other devices have demonstrated cell fusion with much higher efﬁciencies and precision. However, the devices have less throughput and their compatibility with carrying out cell fusion among heterogeneous cells has not been reported. A detailed comparison among other designs has been included in Table S1. The proposed MFC can be designed for various cell sizes. The performance of the MFC can be improved by adjusting the diameter of the fusion well to reduce further leakage at higher ﬂow rates. The mass-scale production of hybrids has been of more importance in recent years, apart from high viability rates. We anticipate that the proposed MFC can be scaled; however, the channel dimensions need to be carefully modiﬁed to avoid high-pressure drops across channels after arraying the trapping structures. 5. Conclusions Microﬂuidics has enabled the precise manipulation of cells. The increasing demand for microﬂuidics-based devices is due to several advantages, such as a lower sample volume, ability to perform experiments at faster rates, and modiﬁcation of device as per the application and type of cell. Microﬂuidic devices have shown immense potential for fusion application. This research aims to develop an efﬁcient lab chip to pair single cells and carry out electrofusion efﬁciently. Our device design primarily focuses on two key parameters (i) single-cell trapping using passive hydrodynamics and transferring cells to fusion wells using gravitational sedimentation and (ii) performing cell electrofusion. Our hydrodynamic trapping and pairing approach is expected to cause lower cell damage due to reduced ﬂow rates for cell trapping and lower exposure to the electric ﬁeld. Our design leads us to observe the high cell viability of fused cells even after 4 days of culture. Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/cells10112855/s1, Figure S1. Step-by-step MFC fabrication process, Figure S2. COMSOL Multiphysics simulations of microﬂuidic channels, Figure S3. COMSOL Multiphysics simulations of designed electrodes, Figure S4. Step-by-step functioning of MFC, Figure S5. Effect of spin speed on PDMS thickness, Figure S6. THP-1 and A549 coculture in the absence of external stimuli, Table S1. Comparative analysis with previously reported works, Table S2. Cell trapping, pairing and efﬁciency calculations. Author Contributions: G.P. designed and implemented lab-chip, performed experiments, and prepared the manuscript; Y.-T.L. and C.-H.L. (Cheng-Hsien Liu), and C.-C.C. conceived this study and supervised this research on bio/immunotherapy and engineering parts, respectively. C.-M.C. and M.-P.L. advised on cell biology and provided cells; C.-M.C., M.-P.L. and C.-H.L. (Chung-Huan Lu) jointly performed bio experiments with G.P.; C.-H.L. (Chung-Huan Lu) jointly performed lab-chip Cells 2021, 10, 2855 15 of 16 microfabrication with G.P.; C.-H.L. (Cheng-Hsien Liu) edited the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This work was ﬁnancially supported partially by the NTHU/MMH collaboration funds (Grant No. MMH-TH-10303 and NTHU-103N2765E1) and the Ministry of Science and Technology, Taiwan, under the grants of MOST 108-2221-E-007-078 and 106-2221-E-007-043-MY3. The fabrication facilities were supported by the Centre for Nanotechnology, Material Science, and Microsystem (CNMM) of National Tsing Hua University and the National Nano Device Laboratory (NDL). We extend our gratitude to Mackay Memorial Hospital for research support. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. 2. 3. 4. Gast, C.E.; Silk, A.D.; Zarour, L.; Riegler, L.; Burkhart, J.G.; Gustafson, K.T.; Parappilly, M.S.; Roh-Johnson, M.; Goodman, J.R.; Olson, B.; et al. Cell fusion potentiates tumor heterogeneity and reveals circulating hybrid cells that correlate with stage and survival. Sci. Adv. 2018, 4, eaat7828. [CrossRef] Gong, J.; Chen, D.; Kashiwaba, M.; Kufe, D. Induction of antitumor activity by immunization with fusions of dendritic and carcinoma cells. Nat. Med. 1997, 3, 558–561. [CrossRef] Koido, S.; Hara, E.; Homma, S.; Torii, A.; Toyama, Y.; Kawahara, H.; Watanabe, M.; Yanaga, K.; Fujise, K.; Tajiri, H.; et al. Dendritic Cells Fused with Allogeneic Colorectal Cancer Cell Line Present Multiple Colorectal Cancer–Speciﬁc Antigens and Induce Antitumor Immunity against Autologous Tumor Cells. Clin. Cancer Res. 2005, 11, 7891. [CrossRef] [PubMed] Liu, W.-L.; Zou, M.-Z.; Liu, T.; Zeng, J.-Y.; Li, X.; Yu, W.-Y.; Li, C.-X.; Ye, J.-J.; Song, W.; Feng, J.; et al. Cytomembrane nanovaccines show therapeutic effects by mimicking tumor cells and antigen presenting cells. Nat. Commun. 2019, 10, 3199. [CrossRef] [PubMed] 5. Mellman, I.; Coukos, G.; Dranoff, G. Cancer immunotherapy comes of age. Nature 2011, 480, 480–489. [CrossRef] [PubMed] 6. Rosenblatt, J.; Kufe, D.; Avigan, D. Dendritic cell fusion vaccines for cancer immunotherapy. Expert Opin. Biol. Ther. 2005, 5, 703–715. [CrossRef] Kemna, E.W.M.; Wolbers, F.; Vermes, I.; van den Berg, A. On chip electrofusion of single human B cells and mouse myeloma cells for efﬁcient hybridoma generation. Electrophoresis 2011, 32, 3138–3146. [CrossRef] Tomita, M.; Tsumoto, K. Hybridoma technologies for antibody production. Immunotherapy 2011, 3, 371–380. [CrossRef] [PubMed] Cowan, C.A.; Atienza, J.; Melton, D.A.; Eggan, K. Nuclear Reprogramming of Somatic Cells After Fusion with Human Embryonic Stem Cells. Science 2005, 309, 1369. [CrossRef] 7. 8. 9. 10. Oback, B.; Wiersema, A.T.; Gaynor, P.; Laible, G.; Tucker, F.C.; Oliver, J.E.; Miller, A.L.; Troskie, H.E.; Wilson, K.L.; Forsyth, J.T.; et al. Cloned Cattle Derived from a Novel Zona-Free Embryo Reconstruction System. Cloning Stem Cells 2003, 5, 3–12. [CrossRef] [PubMed] 11. Oback, B.; Wells, D.N. Cloning Cattle. In Somatic Cell Nuclear Transfer; Sutovsky, P., Ed.; Springer: New York, NY, USA, 2007; pp. 30–57. 12. White, J.; Matlin, K.; Helenius, A. Cell fusion by Semliki Forest, inﬂuenza, and vesicular stomatitis viruses. J. Cell Biol. 1981, 89, 674–679. [CrossRef] 13. Zakai, N.; Kulka, R.G.; Loyter, A. Fusion of human erythrocyte ghosts promoted by the combined action of calcium and phosphate ions. Nature 1976, 263, 696–699. [CrossRef] [PubMed] 14. Ching, C.T.-S.; Sun, T.-P.; Huang, W.-T.; Huang, S.-H.; Hsiao, C.-S.; Chang, K.-M. A circuit design of a low-cost, portable and programmable electroporation device for biomedical applications. Sens. Actuators Chem. 2012, 166–167, 292–300. [CrossRef] 15. KÖHler, G.; Milstein, C. Continuous cultures of fused cells secreting antibody of predeﬁned speciﬁcity. Nature 1975, 256, 495–497. [CrossRef] 16. Dessain, S.K.; Adekar, S.P.; Stevens, J.B.; Carpenter, K.A.; Skorski, M.L.; Barnoski, B.L.; Goldsby, R.A.; Weinberg, R.A. High efﬁciency creation of human monoclonal antibody-producing hybridomas. J. Immunol. Methods 2004, 291, 109–122. [CrossRef] 17. Clow, A.L.; Gaynor, P.T.; Oback, B.J. A novel micropit device integrates automated cell positioning by dielectrophoresis and nuclear transfer by electrofusion. Biomed. Microdev. 2010, 12, 777–786. [CrossRef] 18. Tresset, G.; Takeuchi, S. A Microﬂuidic Device for Electrofusion of Biological Vesicles. Biomed Microdev. 2004, 6, 213–218. [CrossRef] 19. Gel, M.; Kimura, Y.; Kurosawa, O.; Oana, H.; Kotera, H.; Washizu, M. Dielectrophoretic cell trapping and parallel one-to-one fusion based on ﬁeld constriction created by a micro-oriﬁce array. Biomicroﬂuidics 2010, 4, 022808. [CrossRef] [PubMed] Cells 2021, 10, 2855 16 of 16 20. Kimura, Y.; Gel, M.; Techaumnat, B.; Oana, H.; Kotera, H.; Washizu, M. Dielectrophoresis-assisted massively parallel cell pairing and fusion based on ﬁeld constriction created by a micro-oriﬁce array sheet. Electrophoresis 2011, 32, 2496–2501. [CrossRef] [PubMed] 21. Hu, N.; Yang, J.; Qian, S.; Joo, S.W.; Zheng, X. A cell electrofusion microﬂuidic device integrated with 3D thin-ﬁlm microelectrode 22. arrays. Biomicroﬂuidics 2011, 5, 034121. [CrossRef] Skelley, A.M.; Kirak, O.; Suh, H.; Jaenisch, R.; Voldman, J. Microﬂuidic control of cell pairing and fusion. Nat. Methods 2009, 6, 147–152. [CrossRef] 23. Zhang, K.; Chou, C.-K.; Xia, X.; Hung, M.-C.; Qin, L. Block-Cell-Printing for live single-cell printing. Proc. Natl. Acad. Sci. USA 2014, 111, 2948. [CrossRef] 24. Zhao, L.; Liu, Y.; Liu, Y.; Zhang, M.; Zhang, X. Microﬂuidic Control of Tumor and Stromal Cell Spheroids Pairing and Merging for Three-Dimensional Metastasis Study. Anal. Chem. 2020, 92, 7638–7645. [CrossRef] [PubMed] 25. Hu, N.; Yang, J.; Joo, S.W.; Banerjee, A.N.; Qian, S. Cell electrofusion in microﬂuidic devices: A review. Sens. Actuators B Chem. 2013, 178, 63–85. [CrossRef] 26. Hu, N.; Yang, J.; Yin, Z.-Q.; Ai, Y.; Qian, S.; Svir, I.B.; Xia, B.; Yan, J.-W.; Hou, W.-S.; Zheng, X.-L. A high-throughput 27. dielectrophoresis-based cell electrofusion microﬂuidic device. Electrophoresis 2011, 32, 2488–2495. [CrossRef] Ju, J.; Ko, J.-M.; Cha, H.-C.; Park, J.Y.; Im, C.-H.; Lee, S.-H. An electrofusion chip with a cell delivery system driven by surface tension. J. Micromech. Microeng. 2008, 19, 015004. [CrossRef] 28. Cao, Y.; Yang, J.; Yin, Z.Q.; Luo, H.Y.; Yang, M.; Hu, N.; Yang, J.; Huo, D.Q.; Hou, C.J.; Jiang, Z.Z.; et al. Study of high-throughput cell electrofusion in a microelectrode-array chip. Microﬂuid. Nanoﬂuid. 2008, 5, 669–675. [CrossRef] 29. Hu, N.; Yang, J.; Qian, S.; Zhang, X.; Joo, S.W.; Zheng, X. A cell electrofusion microﬂuidic chip using discrete coplanar vertical sidewall microelectrodes. Electrophoresis 2012, 33, 1980–1986. [CrossRef] 30. Qu, Y.; Hu, N.; Xu, H.; Yang, J.; Xia, B.; Zheng, X.; Yin, Z.Q. Somatic and stem cell pairing and fusion using a microﬂuidic array device. Microﬂuid. Nanoﬂuid. 2011, 11, 633–641. [CrossRef] 31. Masuda, S.; Washizu, M.; Nanba, T. Novel method of cell fusion in ﬁeld constriction area in ﬂuid integration circuit. IEEE Trans. Ind. Appl. 1989, 25, 732–737. [CrossRef] 32. Techaumnat, B.; Tsuda, K.; Kurosawa, O.; Murat, G.; Oana, H.; Washizu, M. High-yield electrofusion of biological cells based on ﬁeld tailoring by microfabricated structures. IET Nanobiotechnol. 2008, 2, 93–99. [CrossRef] [PubMed] 33. Gel, M.; Suzuki, S.; Kimura, Y.; Kurosawa, O.; Techaumnat, B.; Oana, H.; Washizu, M. Microoriﬁce-Based High-Yield Cell Fusion on Microﬂuidic Chip: Electrofusion of Selected Pairs and Fusant Viability. IEEE Trans. NanoBiosci. 2009, 8, 300–305. [CrossRef] [PubMed] Schoeman, R.M.; van den Beld, W.T.E.; Kemna, E.W.M.; Wolbers, F.; Eijkel, J.C.T.; van den Berg, A. Electrofusion of single cells in picoliter droplets. Sci. Rep. 2018, 8, 3714. [CrossRef] [PubMed] 34. 35. Hsiao, Y.-C.; Wang, C.-H.; Lee, W.-B.; Lee, G.-B. Automatic cell fusion via optically-induced dielectrophoresis and optically- induced locally-enhanced electric ﬁeld on a microﬂuidic chip. Biomicroﬂuidics 2018, 12, 034108. [CrossRef] [PubMed] 36. Yang, P.-F.; Wang, C.-H.; Lee, G.-B. Optically-Induced Cell Fusion on Cell Pairing Microstructures. Sci. Rep. 2016, 6, 22036. [CrossRef] 37. Huang, L.; Chen, Y.; Huang, W.; Wu, H. Cell pairing and polyethylene glycol (PEG)-mediated cell fusion using two-step centrifugation-assisted single-cell trapping (CAScT). Lab. A Chip. 2018, 18, 1113–1120. [CrossRef] 38. Qin, D.; Xia, Y.; Whitesides, G.M. Soft lithography for micro- and nanoscale patterning. Nat. Protoc. 2010, 5, 491–502. [CrossRef] 39. Tan, W.-H.; Takeuchi, S. A trap-and-release integrated microﬂuidic system for dynamic microarray applications. Proc. Natl. Acad. Sci. USA 2007, 104, 1146. [CrossRef]
10.14814_phy2.15703	Received: 19 April 2023 DOI: 10.14814/phy2.15703 \| Revised: 21 April 2023 \| Accepted: 23 April 2023 O R I G I N A L A R T I C L E Calcium- and voltage- driven atrial alternans: Insight from [Ca]i and Vm asynchrony G. Kanaporis \| E. Martinez- Hernandez \| L. A. Blatter Department of Physiology & Biophysics, Rush University Medical Center, Chicago, Illinois, USA Correspondence L. A. Blatter, Department of Physiology & Biophysics, Rush University Medical Center, 1750 W. Harrison Street, Chicago, IL 60612, USA. Email: [email protected] Funding information HHS \| NIH \| National Heart, Lung, and Blood Institute (NHLBI), Grant/ Award Number: HL057832, HL062231, HL080101, HL101235, HL132871, HL134781, HL155762 and HL164453; Fondation Leducq 1 \| INTRODU CT ION Abstract Cardiac alternans is defined as beat- to- beat alternations in contraction strength, action potential duration (APD), and Ca transient (CaT) amplitude. Cardiac excitation– contraction coupling relies on the activity of two bidirectionally cou- pled excitable systems, membrane voltage (Vm) and Ca release. Alternans has been classified as Vm- or Ca- driven, depending whether a disturbance of Vm or [Ca]i regulation drives the alternans. We determined the primary driver of pac- ing induced alternans in rabbit atrial myocytes, using combined patch clamp and fluorescence [Ca]i and Vm measurements. APD and CaT alternans are typically synchronized; however, uncoupling between APD and CaT regulation can lead to CaT alternans in the absence of APD alternans, and APD alternans can fail to precipitate CaT alternans, suggesting a considerable degree of independence of CaT and APD alternans. Using alternans AP voltage clamp protocols with extra APs showed that most frequently the pre- existing CaT alternans pattern prevailed after the extra- beat, indicating that alternans is Ca- driven. In electrically cou- pled cell pairs, dyssynchrony of APD and CaT alternans points to autonomous regulation of CaT alternans. Thus, with three novel experimental protocols, we collected evidence for Ca- driven alternans; however, the intimately intertwined regulation of Vm and [Ca]i precludes entirely independent development of CaT and APD alternans. K E Y W O R D S atrial myocyte, cardiac alternans, excitation– contraction coupling, intracellular [Ca]i, membrane potential is a recognized risk Alternans for cardiac arrhythmia— including atrial fibrillation (AF), the most common cardiac arrhythmia— and sudden cardiac death (Comtois & Nattel, 2012; Franz et al., 2012; Walker & factor Rosenbaum, 2003). At the cellular level, cardiac alternans is defined as beat- to- beat alternations in contraction am- plitude (mechanical alternans), action potential duration (APD or electrical alternans), and Ca transient amplitude (CaT alternans) at constant stimulation frequency. Its cause is multifactorial (for reviews, see Blatter et al., 2003; This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. Physiological Reports published by Wiley Periodicals LLC on behalf of The Physiological Society and the American Physiological Society. Physiological Reports. 2023;11:e15703. https://doi.org/10.14814/phy2.15703 wileyonlinelibrary.com/journal/phy2 \| 1 of 17 2 of 17 \| Edwards & Blatter, 2014; Eisner et al., 2006; Kanaporis & Blatter, 2017a; Qu & Weiss, 2023; Weiss et al., 2006, 2011) which is the reason why hitherto a comprehen- sive and unifying paradigm of cardiac alternans has not been established. A key to the understanding of alternans mechanism is that the regulation of [Ca]i and Vm is bidi- rectionally coupled ([Ca]i↔Vm coupling) and governed by complex overlapping feedback pathways. Vm → [Ca]i cou- pling is determined by APD restitution and activity of volt- age dependent ion channels that affect Ca signaling and its disturbances result in Vm- driven alternans. [Ca]i → Vm coupling is determined by the effect of [Ca]i dynamics, Ca fluxes, and Ca- dependent ion currents and transporters on Vm and APD. Disturbances of [Ca]i → Vm coupling lead to Ca- driven alternans. Despite enormous progress made, it has remained a matter of debate whether a disturbance of [Ca]i signaling ([Ca]i → Vm coupling) or a disturbance of Vm and AP regulation (Vm → [Ca]i coupling) is the pri- mary cause of alternans (this question has indeed been referred to in the literature as the “chicken or egg” co- nundrum of cardiac alternans; Qu & Weiss, 2007). While there is evidence supporting both directions of coupling, recent progress (including our own: Banach & Blatter, 2023; Kanaporis & Blatter, 2015; Shkryl et al., 2012) and computational findings increasingly point to Ca signal- ing disturbances as the primary cause of alternans (Bien et al., 2006; Eisner et al., 2006; Goldhaber et al., 2005; Nivala & Qu, 2012; Qu et al., 2013; Rovetti et al., 2010). However, the debate is far from settled (Blatter et al., 2021; Kanaporis & Blatter, 2017a) and significant gaps in our understanding of the [Ca]i↔Vm interplay remain. In general, APD and CaT alternans are coupled and co- incide in time; however, we have shown previously with AP voltage clamp experiments that CaT alternans can occur in the absence of APD alternans, and pacing induced APD alternans disappeared when Ca release was inhibited (Kanaporis & Blatter, 2015), thus indicating that the tem- poral correlation is not absolute and the possibility of CaT and APD alternans dyssynchrony exists. Based on these observations, we set out to test whether conditions of dys- synchrony of CaT and APD alternans provide a window into the question of the primary disturbance of cellular signaling that causes atrial alternans. We conducted three sets of experiments: We investigated the onset, temporal development and termination of APD and CaT alternans in current- clamped myocytes, we applied “extra- beats” alternans AP voltage clamp protocols and tested their po- tential, efficacy or failure to disturb CaT alternans, and fi- nally, we investigated the synchrony/dyssynchrony of CaT and APD alternans in pairs of atrial myocytes. The results indicate that the observed patterns of CaT- APD alternans dyssynchrony support the hypothesis that atrial APD and CaT alternans can develop independently and in many in- stances are Ca- driven; however, the data do not exclude the possibility that Vm disturbances have the potential to precipitate CaT alternans. 2 \| METH ODS 2.1 \| Ethics statement All procedures and protocols were approved by the Institutional Animal Care and Use Committee of Rush University Chicago. 2.2 \| Myocyte isolation Atrial myocytes were isolated from male New Zealand White rabbits (2.5– 2.9 kg; 39 rabbits; Envigo). Rabbits are the animal model of choice because cellular calcium regu- lation and electrophysiology are similar to that found in human myocytes, and many cardiac pathologies resemble closely human disease (Hasenfuss, 1998; Milani- Nejad & Janssen, 2014; Panfilov, 2006; Zaragoza et al., 2011). Rabbits were anesthetized with an intravenous injection of sodium pentobarbital (Euthasol; 100 mg/kg) and heparin (1000 IU/kg). The depth of the anesthesia was evaluated by foot pinch or checking corneal reflexes. Hearts were excised, mounted on a Langendorff apparatus, and retro- gradely perfused via the aorta. After an initial 5– 10 min perfusion with oxygenated Ca- free Tyrode solution (in mM: 140 NaCl, 4 KCl, 10 D- Glucose, 5 Hepes, 1 MgCl2, 1000 IU/L Heparin; pH 7.4 with NaOH), the heart was per- fused with minimal essential medium Eagle (MEM) so- lution containing 20 μM Ca and 22.5 μg/mL Liberase TH (Sigma- Aldrich) for ~20 min at 37°C. The left atrium was dissected from the heart and minced, filtered and washed in MEM solution containing 50 μM Ca and 10 mg/mL bo- vine serum albumin. Isolated cells were washed and kept in MEM solution with 50 μM Ca at room temperature (20– 24°C) and were used within 1– 8 h after isolation. 2.3 \| Chemicals and solutions During experiments, cells were superfused continuously with external Tyrode solution composed of (mM): 135 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 Hepes, 10 D- glucose; pH 7.4 with NaOH. All chemicals and reagents were from Sigma- Aldrich unless otherwise stated. KANAPORIS et al. 2.4 \| Fluorescent [Ca]i measurements The fluorescent Ca- sensitive probes Indo- 1 and Fluo- 4 were used to measure dynamic changes of [Ca]i. In cur- rent- and voltage- clamp experiments Fluo- 4 or Indo- 1 pentapotassium salts were added to the pipette solution (see below). For simultaneous optical Vm and [Ca]i meas- urements in cell pairs [Ca]i was recorded with Rhod- 2 (see below). During the course of experiments, cells were continuously superfused with Tyrode solution. Fluo- 4 was excited at 485 nm with a Xe arc lamp and signals were collected at 515 nm using a photomultiplier tube. Background- subtracted fluorescence emission signals (F) were normalized to diastolic or resting fluorescence (F0), and changes of [Ca]i are presented as changes of F/F0. Indo- 1 fluorescence was excited at 357 nm (Xe arc lamp) and emitted cellular fluorescence was recorded simultaneously at 410 nm (F410) and 485 nm (F485) with photomultiplier tubes. F410 and F485 signals were back- ground subtracted and changes of [Ca]i were expressed as changes of the ratio R = F410/F485. Data recording and digitization were achieved using the Axon Digidata 1440A interface and pCLAMP 10.7 software (Molecular Devices). Fluorescence signals were low- pass filtered at 30 Hz. \| Simultaneous optical AP and [Ca]i 2.5 measurements Simultaneous Vm and [Ca]i measurements in atrial cell pairs were conducted with fluorescent indicators. For Vm measurements, the voltage- sensitive fluorescent probe FluoVolt (Martinez- Hernandez et al., 2022; McPheeters et al., 2017) was used and [Ca]i was recorded with the membrane- permeable Ca indicator Rhod- 2/AM (both Thermo Fisher Scientific). Confocal microscopy (Nikon A1R, Nikon Corporation) was used for APs and [Ca]i imaging. Cells were loaded for 15 min with FluoVolt in standard Tyrode solution using proprietary (Thermo Fisher Scientific) loading conditions and dye concen- tration, followed by 10 min wash. Both FluoVolt™ dye (component A of the Thermo Fisher loading kit) and FluoVolt™ Loading Solution (component B) had a final concentration of 1X (specific concentrations not disclosed by manufacturer). FluoVolt preloaded atrial myocytes were subsequently loaded for 15 min with 5 μM Rhod- 2/ AM in the presence of 0.25% Pluronic F- 127. Cells were washed with Tyrode solution and preincubated with blebbistatin (10 μM) for 5 min to minimize motion arti- facts. FluoVolt was excited at 488 nm and emission col- lected at wavelengths >515 nm while Rhod- 2 AM was excited at 543 nm and emission collected at wavelengths >600 nm. FluoVolt and Rhod- 2 signals are presented as \| 3 of 17 F/F0 ratios where F0 represents resting Vm and [Ca]i at the beginning of a recording. [Ca]i and optical AP meas- urements were conducted in the confocal line scan mode (512 lines/s) using a ×40 oil- immersion objective lens. The scan line was placed along the transverse axis of the cell avoiding the nucleus. CaTs and APs were elicited by electrical field stimulation of intact atrial myocytes using a pair of platinum electrodes (5 ms voltage pulses set at ∼50% above the threshold for triggering APs and CaTs). In this set of experiments, cell pairs were stimulated at 1.4 Hz. Electrical coupling between cells was confirmed by monitoring simultaneous spontaneous depolarizations of both cells during a 10- s period of rest immediately follow- ing cessation of field stimulation. While the exact degree of electrical coupling and the intercellular resistance are unknown in these experiments, the simultaneous occur- rence of Vm depolarization in both cells of the magnitude of an AP and the similarity of the time course of the spon- taneous depolarizations were taken as a strong indication of robust electrical coupling. Experiments were conducted at room temperature (22– 24°C). 2.6 \| Patch clamp experiments Patch clamp pipettes (1.5– 3 MΩ filled with internal solu- tion) were pulled from borosilicate glass capillaries (WPI) with a horizontal puller (model P- 97; Sutter Instruments). Pipettes were filled with an internal solution containing (mM): 130 K glutamate, 10 NaCl, 10 KCl, 0.33 MgCl2, 4 MgATP, and 10 Hepes with pH adjusted to 7.2 with KOH. For simultaneous Vm and [Ca]i measurements, 100 μM Fluo- 4 pentapotassium salt or 75 μM Indo- 1 pentapotas- sium salt (Thermo Fisher Scientific) was added to the in- ternal solution. The internal solution was filtered through 0.22 μm pore filters. Electrophysiological signals were re- corded from single atrial myocytes in the whole- cell rup- tured patch clamp configuration using an Axopatch 200A patch clamp amplifier, the Axon Digidata 1440A inter- face, and pCLAMP 10.7 software (Molecular Devices). AP recordings were low- pass filtered at 5 kHz and digitized at 10 kHz. All patch clamp experiments were performed at room temperature (20– 24°C). For AP measurements, the whole- cell “fast” current clamp mode of the Axopatch 200A was used and APs were evoked by 4 ms stimulation pulses with a magnitude ~1.5 times higher than AP activation threshold. Vm mea- surements were corrected for a junction potential error of −10 mV. For AP voltage clamp experiments, voltage commands in form of atrial APs were generated from averages of APs (three consecutive APs/cell) recorded in current- clamp experiments from three individual cells paced at 1.3 Hz KANAPORIS et al. 4 of 17 \| and exhibiting CaT alternans (Kanaporis & Blatter, 2015, 2017b; Figure 1a,b). These representative AP waveforms were used in all AP voltage clamp experiments. Thus, the AP voltage commands were not specific for an individual cell and therefore are expected to differ marginally from the endogenous AP of a specific cell. Two voltage commands were generated: APCaT_Large representing APs recorded during large amplitude alternans CaTs, here also referred to as APN (where “N” stands for “Narrow”). APCaT_Small re- fers to the AP- waveform observed during small amplitude alternans CaTs, also termed here APW (“W” for “Wide”). These two distinct AP waveforms were used to generate the following four pacing protocols (Figure 1c) (i) alternans AP clamp interrupted by two consecutive APN (…N- W- N- W- N- N- W- N- W- N …; referred to here as “NN- protocol”); (ii) alternans AP clamp interrupted by two consecutive APW (…W- N- W- N- W- W- N- W- N- W…; WW- protocol); (iii) alternans AP clamp interrupted by three consecutive APN (…N- W- N- W- N- N- N- W- N- W- N…; NNN- protocol); and (iv) alternans AP clamp interrupted by three consecutive APW (…W- N- W- N- W- W- W- N- W- N- W…; WWW- protocol). Stimulation frequency was modified by varying the cycle length (CL) between 840 and 390 ms by changing diastolic intervals between voltage commands (CLs applied: 390, 440, 490, 540, 640, and 840 ms). In alternans AP voltage clamp experiments, when APCaT_Small/APW elicited a small amplitude CaT and APCaT_Large/APN elicited a large am- plitude CaT (as expected from current clamp experiments where AP kinetics were determined entirely by the cell), CaT and APD alternans are termed “in- phase” (Kanaporis & Blatter, 2017b). This was the case for 244 of a total of 303 experimental observations (81%) in this study. Less frequently (19%), APW elicited a large amplitude CaT and APN elicited a small amplitude CaT. In this case, CaT and APD alternans are termed “out- of- phase” (Figure 1d). 2.7 \| CaT alternans APD and CaT alternans were induced by electrical pacing, either by electrical field stimulation or by a depolarizing pulse in current clamp and AP voltage clamp experi- ments. Typical range where stable CaT alternans was ob- served was 1.2– 2.5 Hz. In current clamp experiments, CaT alternans was induced by incrementally increasing the pacing frequency from a basal pacing frequency of 0.5 or 1 Hz until stable alternans was observed (stimulation frequency increments: 1.3, 1.6, 1.8, 2 and 2.5 Hz). The de- gree of CaT alternans was quantified as the alternans ratio (AR). AR = 1−[Ca]i,Small/[Ca]i,Large, where [Ca]i,Large and [Ca]i, Small are the amplitudes of the large and small CaTs of a pair of alternating CaTs. By this definition, AR values fall between 0 and 1, where AR = 0 indicates no CaT al- ternans and AR = 1 indicates a situation where sarcoplas- mic reticulum (SR) Ca release is completely abolished on every other beat. CaTs were considered alternating when the beat- to- beat difference in CaT amplitude exceeded 10% or AR >0.1 (Kanaporis & Blatter, 2015). For CaT alternans measurements in cell pairs, alter- nans are termed concordant when large and small ampli- tude CaTs, respectively, coincide in both cells (Figure 1e). Alternans is termed discordant when the small amplitude CaT in one cell coincides with the large amplitude CaT in the other cell, and vice versa. Analogously, APD alternans F I G U R E 1 Terminology and experimental protocols. (a) Simultaneous CaT and APD alternans measurement in current clamped rabbit atrial myocyte. APDN: APD recorded during the large amplitude alternans CaT; APDW: APD recorded during the small amplitude alternans CaT. (b) Average APDCaT_Small/APDW and average APDCaT_Large/APDN constructed from experiments as illustrated in (a). To construct the average AP, three consecutive APs recorded from three individual cells paced at 1.3 Hz and exhibiting CaT alternans were averaged. (c) Command voltages for four different AP alternans voltage clamp protocols to deliver extra beats. N: APDCaT_Large or APDN; W: APDCaT_Small or APDW. (d) AP alternans voltage clamp experiments result in either ‘in- phase’ or ‘out- of- phase’ CaT alternans. “In- phase” refers to the situation typically occurring in current clamped cells (i.e., when the voltage is not clamped and is controlled by the cell; cf. panel a). (e) CaT alternans in cell pairs. Concordant: small and large amplitude CaTs coincide in both cells; Discordant: small amplitude CaT in cell A coincides with large amplitude CaT in cell B and vice versa. KANAPORIS et al. is concordant when APW and APN coincide in both cells, while during discordant APD alternans APW in one cell coincides with APN in the other cell, and during the sub- sequent beat APN in the first cell coincides with APW in the second cell. 3 \| RESULT S \| Kinetics and synchrony/ 3.1 dyssynchrony of onset and termination of CaT and APD alternans In a first set of experiments, we investigated the onset and termination kinetics of pacing- induced CaT and APD al- ternans (N = 16 rabbits; n = 24 cells). Experiments were conducted in Fluo- 4 or Indo- 1 loaded cells under current clamp conditions to measure [Ca]i and Vm simultane- ously. The aim of this part of the study was to gain insight from the detailed investigation of onset and termination kinetics of atrial alternans whether alternans were Ca- or Vm- driven. Specifically, we were interested whether CaT and APD alternans started and terminated synchronously or whether there was a temporal uncoupling of CaT and APD alternans start and termination. Figure 2 shows an example of highly synchronized onset of CaT and APD alternans, that is, the onset occurred on the same beat, il- lustrated by the overlays of two consecutive APs recorded at the times indicated by the gray bar. In 76.9% of the observations, APD and CaT alternans started simultane- ously; thus, a synchronized onset was the most commonly observed scenario (alternans was defined to be present when AR >0.1). However, while common, the high degree of synchronization of alternans onset was not always pre- sent and dyssynchrony between CaT and APD alternans \| 5 of 17 lasting for one or more beats was observed. Figure 3a shows an example where begin of APD alternans preceded CaT alternans (9.6% of observations), whereas Figure 3b shows the opposite where CaT alternans started before APD alternans (13.5% of observations). These observa- tions suggest that both APD alternans and CaT alternans can develop independently. We gained further valuable insight into the synchrony/ dyssynchrony dynamics of CaT and APD alternans through closer examination of alternans termination ki- netics. As previously mentioned, overall APD and CaT alternans are synchronized, including termination of al- ternans, as illustrated in Figure 4a. However, synchrony between CaT and APD alternans termination was not an obligatory feature and variable degrees of dyssynchrony between APD and CaT alternans termination were ob- served. Figure 4b shows an example where APD alternans continued for >15 beats after cessation of CaT alternans, and Figure 4c shows an example where pronounced APD alternans failed to be accompanied by CaT alternans. Furthermore, we did not observe the situation where APD alternans terminated before cessation of CaT alternans, that is, CaT alternans terminated either simultaneously with APD alternans (36% of observations) or in the major- ity of the cases before (64%), but never after APD alternans termination. These observations support the hypothesis that alternans onset and termination are primarily (but not exclusively, Figure 3a) Ca- driven. \| Effect of extra- beats on synchrony/ 3.2 dyssynchrony of CaT and APD alternans In the next set of experiments, we tested the effect of extra- beat stimulations on alternans AP- clamped atrial Synchronized APD F I G U R E 2 and CaT alternans in single rabbit atrial myocyte. Simultaneous recording of APs (current clamp) and CaTs, showing initiation of pacing induced alternans. n = 40 observations of a total of 52 observations. Top: overlays of two consecutive APs recorded at time intervals marked by the gray bars. [Ca]i was measured with Indo- 1. AR, alternans ratio. KANAPORIS et al. 6 of 17 \| Simultaneous APD F I G U R E 3 (current clamp) and CaT alternans recording in single rabbit atrial myocyte. (a) APD alternans onset precedes CaT alternans. n = 5 observations of a total of 52 observations. (b) CaT alternans onset precedes APD alternans. n = 7 observations of a total of 52 observations. Top in both panels: overlay of two consecutive APs recorded at time intervals marked by the gray bars. [Ca]i was measured with Fluo- 4. AR, alternans ratio. myocytes (N = 12 rabbits). Atrial myocytes were voltage clamped with an AP alternans command voltage, using actual AP morphologies previously recorded during pac- ing induced CaT alternans in atrial myocytes (for details, see Methods section; Figure 1). Four different AP voltage clamp protocols were applied. With two protocols, the overall phase of APD alternans remained unchanged and only the shape of a single AP was changed from wide (W) to narrow (N) or from N to W (NNN- and WWW- protocol; Figure 1c). The other two protocols introduced a phase shift in the sequence of APD alternans command voltage by adding a single extra- beat (NN- and WW- protocol). A total of 151 recordings with the WWW- and the NNN- protocols were analyzed (WWW- protocol: n = 13 cells; NNN- protocol; n = 15 cells). The vast majority of recordings showed that CaT and APD alternans were in- phase before the intervention and remained in- phase (134 of 151 recordings; 89%; Figure 5a,b). In 9% of recordings (13/151), alternans was out- of- phase before the interven- tion and remained out- of- phase (Figure 5c). On rather rare occasions (4/151; <3%), the intervention caused a phase shift, either from in- phase to out- of- phase or from out- of- phase to in- phase. An example is shown in Figure 5d. Overall, these experiments (Figure 5) show that chang- ing the morphology of a single AP, while maintaining the overall phase of APD alternans, has very limited or no ef- fect on the course of concomitant CaT alternans, suggest- ing that CaT alternans are regulated rather autonomously. Clearly more profound effects on the course of CaT alternans could be achieved by the protocols that KANAPORIS et al. \| 7 of 17 Simultaneous APD F I G U R E 4 (current clamp) and CaT alternans recording in single rabbit atrial myocyte. (a) Simultaneous termination of APD and CaT alternans. n = 4 observations of a total of 11 observations. (b) Dyssynchronous termination of APD and CaT alternans. APD alternans continues after cessation of CaT alternans. n = 7 observations of a total of 11 observations. (c) Prolonged episode of APD alternans in the absence of CaT alternans. Top in all panels: overlay of two consecutive APs recorded at time intervals marked by the gray bars. [Ca]i was measured with Indo- 1 (panels a and c) or Fluo- 4 (panel b). AR, alternans ratio. introduced a phase shift of APD alternans (NN- protocol, Figure 6; WW- protocol, Figure 7). Of a total of 72 ex- periments where the NN- protocol was applied (n = 19 cells), in 59 (82%) experiments CaT alternans started in- phase (i), and 13 (18%) experiments started out- of- phase (o). During application of the NN- protocol, in 57% of recordings (41/72), the APD phase shift failed to shift the phase of the CaT alternans (Figure 6c), whereas in 43% of the experiments (31/72), the APD phase shift was accompanied by a phase shift in CaT alternans (i → o or o → i). When a CaT alternans phase shift occurred, it was more commonly from in- phase to out- of- phase (32% of all NN- experiments; Figure 6a) and less common from out- of- phase to in- phase (11% of all NN- experiments; Figure 6b). Of all experiments that started in- phase (59 experiments) 61% remained in- phase (36/59) and only 39% (23/59) shifted to out- of- phase, supporting the no- tion that in- phase is the preferred condition. Similarly, of all experiments that started out- of- phase (13 exper- iments) only 38% (5/13) stayed out- of- phase, while the majority (62%; 8/13) changed to the preferred in- phase condition. The effect of the APD phase shift induced by the NN- protocol is summarized in Figure 6d. Similar effects were achieved with the WW- protocol (n = 20 cells). In 58% of recordings (46/80), the APD phase shift failed to change the phase of CaT alternans (Figure 7a,d). While in these cases, a phase shift was ab- sent, occasionally a transient period of CaT amplitude irregularities that could last for several beats (Figure 7a) could be observed. These CaT disturbances typically consisted of irregular CaT amplitudes with no clear alter- nans pattern. A CaT alternans phase shift was observed in 42% of recordings (34/80) and was more common for out- of- phase to in- phase (25/80, 31%; Figure 7b) than for in- phase to out- of- phase (9/80, 11%; Figure 7c). Again, the data confirm that “in- phase” is the preferred condition. Of all experiments that started out- of- phase (31 experiments) only 19% (6/31) stayed out- of- phase, while 81% (25/31) changed to in- phase. Of all experiments that started in- phase (49 experiments) 82% (40/49) stayed in- phase and only 18% (9/49) changed to out- of- phase. Finally, in a small fraction of experiments (6 observations), applica- tion of the NN- or WW- protocol caused a cessation of CaT alternans after the intervention (Figure 8). In the exam- ple shown, application of the NN- protocol caused a phase shift from in- phase to out- of- phase followed by cessation of CaT alternans. In summary, these experiments showed that simply changing the shape of a single AP during APD alternans (NNN- and WWW- protocol) had little or no effect on CaT alternans; thus in these cases, the overall alternans pattern remained unchanged. More profound were the effects of a phase shift in APD alternans (NN- and WW- protocol) on the course of CaT alternans. The following trends were ob- served: (i) in most cases (108/152), CaT and APD alternans were in- phase before the intervention and remained in- phase after the intervention (76/108), that is, the in- phase pattern, which is physiologically observed, overall remains the preferred pattern also in these specific voltage clamp experiments. Considering all experiments, essentially an KANAPORIS et al. 8 of 17 \| F I G U R E 5 CaT alternans in AP alternans voltage- clamped atrial myocytes elicited with WWW- and NNN- protocols. (a) WWW- protocol applied to in- phase CaT alternans failed to induce a phase shift. (b) NNN- protocol applied to in- phase CaT alternans failed to induce a phase shift. (c) WWW- protocol applied to out- of- phase CaT alternans failed to induce a phase shift. (d) NNN- protocol applied to in- phase CaT alternans caused phase shift from in- phase to out- of- phase alternans. i: in- phase; o: out- of- phase. [Ca]i was measured with Fluo- 4. Number of recordings with WWW- protocol: 70. Number of recordings with NNN- protocol: 81. identical fraction of experiments (109/152) showed in- phase CaT alternans after the intervention, irrespective whether a phase shift had occurred or not. (ii) A phase shift of CaT alternans (i → o and o → i) was observed in 65 of a total of 152 recordings. For a phase shift of CaT to occur with the NN- and WW- protocols, requires that the large- small- large- small amplitude pattern of CaT alternans is maintained and thus argues for a largely autonomous regu- lation of CaT alternans and a high degree of independence from APD alternans. (iii) Interestingly, in all recordings with a CaT alternans phase shift 43% of the cells revealed an AR >0.5. In stark contrast, in experiments without a CaT alternans phase shift less than 8% of the cells had an AR >0.5. Thus, prominent CaT alternans is a factor that fa- cilitates a phase shift of CaT alternans (e.g., Figure 6a) and resulted in an undisturbed continuation of the CaT alter- nans pattern after the application of the extra beat. \| Synchrony/dyssynchrony of 3.3 CaT and APD alternans in cell pairs We next investigated the spatio- temporal organization of APD and CaT alternans in atrial cell pairs (N = 11 rabbits; KANAPORIS et al. \| 9 of 17 F I G U R E 6 CaT alternans in AP alternans voltage- clamped atrial myocytes elicited with NN- protocol. (a) NN- protocol applied to in- phase CaT alternans caused phase shift from in- phase to out- of- phase alternans. (b) NN- protocol applied to out- of- phase CaT alternans caused phase shift from out- of- phase to in- phase alternans. (c) NN- protocol applied to in- phase CaT alternans failed to induce a phase shift. (d) Frequency distribution of effect of NN- protocol on alternans phase. i: in- phase; o: out- of- phase. [Ca]i was measured with Fluo- 4. Number of recordings with NN- protocol: 72. n = 37 cell pairs). [Ca]i and Vm were measured simultane- ously using the fluorescent probes Rhod- 2 and FluoVolt, respectively. Fluorescence signals were recorded in confo- cal transverse line scan mode (Figure 9a). Alternans was induced by electrical pacing by field stimulation (1.4 Hz). Most commonly, both CaT and APD alternans were con- cordant in electrically coupled cell pairs, as illustrated in Figure 9. Figure 9a shows line scan (x- t) images of the two cells. Figure 9b shows [Ca]i and Vm signals derived from the line scan images and averaged over the width of the cell. To test for electrical coupling between the two cells, we took advantage of the fact that rabbit atrial myocytes showed spontaneous activity (spontaneous Ca release and spontaneous membrane depolarizations and APs) during a period of rest immediately following electrical pacing at rates that elicited alternans. As shown in Figure 9, ap- proximately 5 s after cessation of pacing both cells showed a simultaneous spontaneous Vm depolarization that was accompanied by a transient elevation of [Ca]i. The simul- taneous occurrence of a depolarization signal of similar magnitude and time course in both cells was taken as clear evidence of electrical coupling between the two cells of a pair. The high degree of synchronization between [Ca]i and Vm signals in form of concordant alternans was ob- served in the majority of experiments. However, several deviations from this general pattern was observed that provided new insight into the underlying [Ca]i and Vm disturbances causing alternans. Figures 10– 12 illustrate examples of dyssynchrony of APD and CaT alternans in atrial cell pairs. Figure 10 shows an example where one cell developed CaT alternans whereas in the other cell the CaT amplitudes remained constant, despite electrical cou- pling between the two cells. The CaT alternans in cell A was accompanied by APD alternans, that did not extend to cell B. Thus, the example illustrates dyssynchrony of CaT and APD alternans between the two cells despite electrical coupling. Figure 11 illustrates that CaT alternans can develop autonomously and essentially in the absence of APD al- ternans. Both cells of the pair developed CaT alternans however with different onset. The interval between onset KANAPORIS et al. 10 of 17 \| F I G U R E 7 CaT alternans in AP alternans voltage- clamped atrial myocytes elicited with WW- protocol. (a) WW- protocol applied to in- phase CaT alternans failed to induce a phase shift. (b) WW- protocol applied to out- of- phase CaT alternans caused phase shift from out- of- phase to in- phase alternans. (c) WW- protocol applied to in- phase CaT alternans caused phase shift from in- phase to out- of- phase alternans. (d) Frequency distribution of effect of WW- protocol on alternans phase. i, in- phase; o, out- of- phase. [Ca]i was measured with Fluo- 4. Number of recordings with WW- protocol: 80. F I G U R E 8 CaT alternans in AP alternans voltage- clamped atrial myocytes elicited with NN- protocol. NN- protocol applied to in- phase CaT alternans caused a transient phase shift from in- phase to out- of- phase, followed by cessation of CaT alternans. i: in- phase; o: out- of- phase. [Ca]i was measured with Fluo- 4. n = 6 observations. in the two cells lasted >7 beats. In addition, CaT alternans was discordant. Furthermore, CaT alternans developed in the apparent absence of APD alternans, indicative of dissociation of [Ca]i regulation from Vm control, despite clearly evident electrical coupling between the cells. Finally, Figure 12 shows an example of uncoupling of [Ca]i and Vm regulation with different effects on alternans development, compared to Figure 11. Here, both cells show concordant APD alternans; however, APD alternans was accompanied by CaT alternans only in cell B, but not cell A. KANAPORIS et al. \| 11 of 17 Simultaneous F I G U R E 9 measurements of CaT and APD alternans in atrial myocyte pairs. Vm and [Ca]i were recorded with fluorescence transverse line scan confocal imaging. Vm probe: FluoVolt; Ca probe: Rhod- 2. (a) Confocal line scan (x- t) [Ca]i (Rhod- 2 fluorescence) images. (b) [Ca]i and Vm traces recorded from line scan images in panel (a). Signals were spatially averaged over the width of the cell. Highly synchronized concordant CaT and APD alternans were recorded. Synchronized spontaneous Vm depolarization and Ca release during rest after pacing indicate electrical coupling between the two cells. Number of cell pairs with synchronized APD and CaT alternans: 40. In summary, the examples shown in Figures 10– 12 indicate that CaT alternans can develop with variable degrees of independence from APD alternans and can escape complete synchronization between the two cells, despite electrical coupling between the cells. These obser- vations again suggest a degree of autonomy of CaT alter- nans development and support the [Ca]i → Vm coupling hypothesis for alternans. 4 \| DISCUSSIO N In this study, we investigated the organization of dyssyn- chrony of APD and CaT alternans in order to gain insight whether atrial cellular alternans are primarily Vm- or Ca- driven. The main findings are: (i) in general, APD and CaT alternans are synchronized in time and magnitude, how- ever uncoupling between APD and CaT alternans onset, duration and termination was observed; (ii) simultaneous AP and CaT measurements in current- clamped myocytes revealed that CaT alternans can develop in the absence of APD alternans, and that APD alternans can fail to pre- cipitate CaT alternans, suggesting independence of CaT alternans development from Vm disturbances; (iii) using an alternans AP voltage clamp protocol that included ap- plication of one extra AP (NN- and WW- protocol) showed that a phase shift in the sequence of APs was accompanied by a phase shift in nearly half of the experiments, indicat- ing that the CaT alternans pattern prevailed undisturbed and thus alternans appear to be Ca- driven. Furthermore, in 50% of the experiments, CaT started and remained in- phase with APD alternans after the extra AP, suggesting that the physiological APD- CaT alternans relationship remains the preferred pattern also in AP voltage clamp experiments; (iv) while APD and CaT alternans in electri- cally coupled cell pairs are typically synchronized between cells, uncoupling of CaT alternans from APD alternans as well as dyssynchrony of CaT alternans between the two KANAPORIS et al. 12 of 17 \| Simultaneous F I G U R E 1 0 measurements of CaT and APD alternans in atrial myocyte pairs. Vm (FluoVolt) and [Ca]i (Rhod- 2) were recorded with fluorescence transverse line scan confocal imaging. Cell A reveals the development of synchronized CaT and APD alternans. Cell B failed to develop alternans. Two synchronized spontaneous Vm depolarizations and Ca release events during rest after pacing indicate electrical coupling between the two cells. Number of cell pairs showing dyssynchrony of APD and CaT alternans: 15. cells can occur, suggesting autonomous regulation of CaT alternans. In cardiac cells, the regulation of Vm and [Ca]i is in- timately linked and bi- directionally coupled. Complex feedback pathways between Vm and [Ca]i regulation are governed by transmembrane ion gradients and the activ- ity of surface membrane ion channels and transporters. Cardiac alternans is considered a proarrhythmic condi- tion and in the atria has been causally linked to AF. CaT and APD alternans are highly correlated in time and space, as well as in the degree of alternans quantified as APD and CaT alternans ratios (Kanaporis & Blatter, 2015). Disturbances of both Vm and [Ca]i regulation have been identified that facilitate the occurrence of alternans. This has led to the distinction between Ca- and voltage- driven alternans (Qu & Weiss, 2023). Two main elements of cel- lular Ca signaling have emerged as potential causes of Ca- driven alternans: SR Ca load and refractoriness of the SR Ca release channel, the ryanodine receptor (RyR). Ca alternans driven by SR Ca load alternans, by definition, entail a beat- to- beat alternating end- diastolic filling of the SR (Diaz et al., 2004; Kanaporis & Blatter, 2017b) and facilitates CaT alternans further through the steepness of the non- linear fractional SR Ca release function (Eisner et al., 2000; Xie et al., 2008) and limitations to the rate of diastolic refilling of the SR by SERCA. SR Ca load- driven alternans has been observed experimentally and formu- lated mathematically in computational studies. For this mechanism to be able to sustain alternans beat- to- beat alternans of SR Ca load is an obligatory feature. However, the observation that CaT alternans can occur in the ab- sence of SR Ca load alternans (Huser et al., 2000; Picht et al., 2006; Shkryl et al., 2012) argues in favor of addi- tional alternans mechanisms. Refractoriness of the RyR and the SR Ca release machinery has emerged as an alter- native key factor for Ca- driven alternans (Alvarez- Lacalle et al., 2013; Kornyeyev et al., 2012; Lugo et al., 2014; Shkryl et al., 2012). RyR refractoriness refers to the fact that SR Ca release is unavailable after a release for a lim- ited time interval, and therefore becomes increasingly more important at shorter diastolic intervals. In contrast, at low heart rates, when the diastolic interval exceeds RyR refractoriness, the SR Ca load mechanism becomes the predominant mechanism of Ca- driven alternans. Recently, an elegant conceptual model for cardiac alter- nans has been forwarded, termed “3R theory” of alternans (Cui et al., 2009; Nivala & Qu, 2012; Qu et al., 2013; Rovetti et al., 2010) that links Ca spark properties (i.e., the prop- erties of elementary Ca release events from individual Ca release unites of the SR) to whole- cell CaT alternans and forms the basis of a unifying theory of CaT alternans. The 3R theory predicts by numerical computations how KANAPORIS et al. \| 13 of 17 Simultaneous F I G U R E 1 1 measurements of CaT and APD alternans in atrial myocyte pairs. Vm (FluoVolt) and [Ca]i (Rhod- 2) were recorded with fluorescence transverse line scan confocal imaging. CaT alternans develop autonomously in the absence of APD alternans at different times in the two cells and are discordant. Synchronized spontaneous Vm depolarization and Ca release during rest after pacing indicate electrical coupling of the two cells. Number of cell pairs showing dyssynchrony of APD and CaT alternans: 15. instabilities in the relationship of three critical spark at- tributes (Randomness, Recruitment, Refractoriness) lead to alternans, and how excitation– contraction coupling Ca handling proteins (L- type Ca channels, RyR, SERCA, Na/Ca exchanger, Ca buffers, and IP3 receptor Ca release channel) and Ca handling organelles (SR, mitochondria) determine alternans probability. Alternatively, disturbances of Vm regulation have also been identified as cause of cardiac alternans (Vm → [Ca]i coupling). Three mechanisms of Vm- driven alternans have been proposed (Qu et al., 2010). First, a steep APD res- titution curve facilitates alternans (Martinez- Hernandez et al., 2022; Nolasco & Dahlen, 1968) which becomes particularly prominent at high heart rates or short dia- stolic intervals. Additional conditions that facilitate Vm- driven alternans involve AP prolongation (Kanaporis et al., 2019; Kanaporis & Blatter, 2023) and, related to it, early afterdepolarizations (EADs; Qu & Weiss, 2023). However, also very short APs, brought upon by K channel activity, have been linked to Vm- driven alternans (Fish & Antzelevitch, 2008). It is generally agreed that disturbances of both Vm and [Ca]i regulation can precipitate cardiac alternans. Voltage and Ca signaling can be considered as two excitable subsys- tems that are bidirectionally coupled ([Ca]i↔Vm coupling) and their respective regulatory mechanisms are intimately intertwined. Mediators of the interactions between the two systems are the Ca- dependence of membrane ion channels and transporters, and the voltage dependence of Ca signaling mechanisms (Weiss et al., 2006). For exam- ple, the Ca- dependent inactivation of voltage- gated L- type Ca channels (LCCs) has been implicated in the determi- nation of APD during alternans where a large CaT tends to shorten ADP through LCC inactivation. The electro- genic Na/Ca exchanger has been shown to modulate APD where during a large amplitude alternans CaT NCX activ- ity tends to prolong the AP. To mention a few additional examples, we have shown previously that Ca- dependent chloride channels and LCCs (Kanaporis & Blatter, 2017b) are involved in alternans (Kanaporis & Blatter, 2016a, 2016b). Also, modulation of AP morphology by several K channels has been shown to determine the develop- ment of alternans (Kanaporis et al., 2019; Kanaporis & Blatter, 2023). Thus, undoubtedly disturbance of one of the systems inadvertently leads to failures in the other. While this notion is generally accepted, the question has been raised whether there is a preference for one system to be primarily affected and to become the driver of alter- nans (Qu & Weiss, 2007). Our observation that CaT alter- nans can occur in the absence of APD alternans could be interpreted that the primary cause roots in a Ca signaling failure. In order to address this question further, we set out KANAPORIS et al. 14 of 17 \| Simultaneous F I G U R E 1 2 measurements of CaT and APD alternans in atrial myocyte pairs. Vm (FluoVolt) and [Ca]i (Rhod- 2) were recorded with fluorescence transverse line scan confocal imaging. The two cells show concordant APD alternans; however, only cell B developed CaT alternans. Number of cell pairs showing dyssynchrony of APD and CaT alternans: 15. in this study to investigate dyssynchronies between CaT and APD alternans using novel experimental protocols. Simultaneous recording of CaT and Vm in single current- clamped atrial myocytes showed that most commonly CaT and APD alternans coincide, are synchronized and in- phase (Figure 2). However, close examination of onset and termi- nation of alternans revealed subtle signs of dyssynchrony between CaT and APD alternans. Patterns of dyssynchrony consisted of APD alternans onset before CaT alternans, CaT alternans onset before APD alternans begin, and CaT alternans termination while APD alternans continued. These patterns of CaT and APD alternans uncoupling in- dicate a degree of independence of CaT and APD alter- nans development. Furthermore, while continued APD alternans in the absence of CaT alternans was observed, we did not record prolonged CaT alternans in the absence of APD alternans in the current study. The latter suggests that APD alternans is required for CaT alternans to occur; however, the presence of APD alternans does not guaran- tee the occurrence of CaT alternans (Figure 4b). (Note: the apparent absence of cell averaged CaT alternans could be mimicked by subcellular out- of- phase CaT alternans, as shown previously (Edwards & Blatter, 2014; Kockskamper & Blatter, 2002). However, this alternans pattern is rare and analysis of spatially resolved confocal CaT alternans data (n = 32 cells) failed to reveal subcellular out- of- phase alternans in the current study). Furthermore, as we have shown previously in AP voltage clamp experiments CaT alternans can develop in the absence of APD alternans. This is consistent with our earlier observation (Kanaporis & Blatter, 2015) in AP voltage- clamped experiments that an APD alternans voltage clamp protocol not automat- ically precipitates CaT alternans. Interestingly, while in current clamp experiments we did not observe long- lasting CaT alternans in the absence of APD alternans, previously we were able to induce CaT alternans with an AP voltage clamp protocol without APD alternans. This was also ob- served in rabbit ventricular myocytes (Chudin et al., 1999). The latter suggests that under conditions where the cell determines AP morphology (current clamp experiments) long- lasting CaT alternans requires concomitant APD alter- nans, whereas in AP voltage clamp experiments, when an AP shape is “forced” onto the cell, CaT alternans can occur in the absence of APD alternans. Taken together, these data are consistent with the hypothesis that CaT alternans de- velop with a noticeable degree of independence from Vm. KANAPORIS et al. We collected additional experimental evidence that CaT alternans developed independently and autonomously. In the extra- beat experiments (Figures 5– 8), the voltage clamp protocols where only a single AP was replaced by one of opposite duration (NNN- and WWW- protocol) the course of CaT alternans was typically not altered. Only in ~4% of experiments a phase- shift was observed, that is, only in a small fraction of experiments a single beat AP disturbance was able to permanently alter the course of CaT alternans (Figure 5). In contrast, alternans AP voltage clamp protocols that entailed an APD phase shift (NN- and WW- protocol), the consequences for CaT alternans were more diverse and complex. Two CaT alternans patterns stood out as most common: (i) in half of the experiments, CaT and APD alternans began in- phase and remained in- phase after the extra beat intervention. This indicates that the “physiological” pattern is the preferred alternans pat- tern that is difficult to be altered by an AP disturbance. (ii) The second most frequently case observed (Figure 6, NN- protocol: 43%; Figure 7, WW- protocol: 42%) was the AP phase- shift being accompanied by a CaT phase- shift, that is, the cell maintained its CaT alternans pattern and the small- large- small- large CaT pattern continued undis- turbed, arguing in favor of Ca- driven alternans. Further evidence for Ca- driven alternans came from simultaneous AP and CaT measurements in electrically coupled cell pairs. While undoubtedly CaT and APD al- ternans were highly synchronized in cells pairs (Figure 9), occurrence of uncoupling of CaT and APD provided in- teresting insight into the origin of alternans. We observed several patterns of uncoupling and dyssynchrony between the two cells: (i) CaT and APD alternans in one cell were not accompanied by CaT and APD alternans in the sec- ond cells, despite clear evidence of electrical coupling be- tween the two cells. In this case APD alternans was not capable of driving alternans in the other cell (Figure 10). (ii) Despite clear electrical coupling CaT alternans could be discordant and occur in the absence of APD alternans (Figure 11). (iii) Synchronized concordant APD alter- nans (Figure 12) precipitate CaT only in one cell. Taken together, the data collected from cell pairs emphasize autonomy of CaT alternans in situations where CaT and APD among a cell pair became dyssynchronized and APD alternans was either unable to drive CaT alternans or CaT alternans developed independently of APD. In addition to single- cell studies, electrical (and CaT) alternans have been investigated extensively at tissue, organ, and organ- ism level. The latter include numerous clinical studies on alternans in humans, and alternans was proposed to identify patients with AF substrate (Lalani et al., 2013; Narayan et al., 2011; Verrier et al., 2016) and serve as a tool for arrhythmia risk stratification. However, cardiac cell pairs have rarely been subject of alternans investigations \| 15 of 17 or have focused on cell– cell propagation of Ca waves (Li et al., 2012). This strikes as quite surprising given the fact that the cardiac cell pair represents the elementary struc- tural and functional unit of cell– cell communication in the heart. As the elementary unit of cardiac cell– cell inter- action insight from alternans study in cell pairs have the potential to serve as crucial linchpin between cellular and multicellular findings in the understanding of alternans mechanisms. This study provides a first account of APD and CaT alternans mechanism in cell pairs. In conclusion, we have initially discussed the fact that historically the alternans field has been occupied with the lingering question whether the primary de- fect that leads to alternans results from disturbances of Vm regulation or from Ca signaling failures. The ever- growing body of new data on alternans at cellular and organ level have still not generated an unequivocal an- swer to this question, and might indicate that the view of alternans being caused by a primary disturbance of one or the other system has become obsolete and requires a new perspective. As outlined above, cardiac excitation– contraction coupling is governed by two excitable systems (Qu & Weiss, 2023), membrane voltage, and in- tracellular Ca signaling. The key element here is that the two systems are bidirectionally coupled and their regula- tion is characterized by numerous overlapping and mu- tually influencing feedback pathways. When the control of both systems is entirely determined by the cell and not manipulated by experimental interventions such as interference by voltage- and current- clamp approaches or exogenous Ca buffers (inadvertently induced by the use of fluorescent Ca indicator dyes), the two systems never act independently and are always under mutual influence. In other words, alternans cannot be solely Ca- or solely Vm- driven, rather alternans are either predom- inantly Vm- driven or predominantly Ca- driven. In either case, however, disturbances of one of the two systems have immediate consequences for the proper function- ing of the other which often results in further enhance- ment and stabilization of alternans. AUT HOR CON TRIBUT IONS G.K, E.M.- H was involved in the collection of data. E.M.- H., G.K., L.A.B was involved in the conception and design of the work, data analysis and interpretation, and drafting the article. All authors have approved the final version of the manuscript. FUNDING INFORMAT ION This work was supported by the National Institutes of Health grants HL057832, HL062231, HL080101, HL101235, HL132871, HL134781, HL155762, and HL164453, and the Fondation Leducq. KANAPORIS et al. 16 of 17 \| CONFLICT OF INTEREST STATEME N T No conflicts of interest to declare. ORCID G. Kanaporis E. Martinez- Hernandez org/0000-0002-5004-2239 L. A. Blatter https://orcid.org/0000-0003-3593-1182 https://orcid. https://orcid.org/0000-0003-3409-8678 R E F E R E N C E S Alvarez- Lacalle, E., Cantalapiedra, I. R., Penaranda, A., Cinca, J., Hove- Madsen, L., & Echebarria, B. (2013). Dependency of cal- cium alternans on ryanodine receptor refractoriness. PLoS One, 8, e55042. Banach, K., & Blatter, L. A. (2023). The 'Reverse FDUF' mechanism of atrial excitation- contraction coupling sustains calcium al- ternans— A hypothesis. Biomolecules, 13(1), 7. https://doi. org/10.3390/biom1 3010007 Bien, H., Yin, L., & Entcheva, E. (2006). Calcium instabilities in mammalian cardiomyocyte networks. Biophysical Journal, 90, 2628– 2640. Blatter, L. A., Kanaporis, G., Martinez- Hernandez, E., Oropeza- Almazan, Y., & Banach, K. (2021). Excitation- contraction cou- pling and calcium release in atrial muscle. Pflügers Archiv, 473, 317– 329. Blatter, L. A., Kockskamper, J., Sheehan, K. A., Zima, A. V., Huser, J., & Lipsius, S. L. (2003). Local calcium gradients during excitation- contraction coupling and alternans in atrial myo- cytes. The Journal of Physiology, 546, 19– 31. Chudin, E., Goldhaber, J., Garfinkel, A., Weiss, J., & Kogan, B. (1999). Intracellular Ca2+ dynamics and the stability of ventric- ular tachycardia. Biophysical Journal, 77, 2930– 2941. Comtois, P., & Nattel, S. (2012). Atrial repolarization alternans as a path to atrial fibrillation. Journal of Cardiovascular Electrophysiology, 23, 1013– 1015. Cui, X., Rovetti, R. J., Yang, L., Garfinkel, A., Weiss, J. N., & Qu, Z. (2009). Period- doubling bifurcation in an array of coupled sto- chastically excitable elements subjected to global periodic forc- ing. Physical Review Letters, 103, 044102. Diaz, M. E., O'Neill, S. C., & Eisner, D. A. (2004). Sarcoplasmic re- ticulum calcium content fluctuation is the key to cardiac alter- nans. Circulation Research, 94, 650– 656. Edwards, J. N., & Blatter, L. A. (2014). Cardiac alternans and intracel- lular calcium cycling. Clinical and Experimental Pharmacology and Physiology, 41, 524– 532. Eisner, D. A., Choi, H. S., Diaz, M. E., O'Neill, S. C., & Trafford, A. W. (2000). Integrative analysis of calcium cycling in cardiac mus- cle. Circulation Research, 87, 1087– 1094. Eisner, D. A., Li, Y., & O'Neill, S. C. (2006). Alternans of intracel- lular calcium: Mechanism and significance. Heart Rhythm, 3, 743– 745. Fish, J. M., & Antzelevitch, C. (2008). Cellular mechanism and ar- rhythmogenic potential of T- wave alternans in the Brugada syn- drome. Journal of Cardiovascular Electrophysiology, 19, 301– 308. Franz, M. R., Jamal, S. M., & Narayan, S. M. (2012). The role of ac- tion potential alternans in the initiation of atrial fibrillation in humans: A review and future directions. Europace, 14(Suppl 5), v58– v64. Goldhaber, J. I., Xie, L. H., Duong, T., Motter, C., Khuu, K., & Weiss, J. N. (2005). Action potential duration restitution and alternans in rabbit ventricular myocytes: The key role of intracellular cal- cium cycling. Circulation Research, 96, 459– 466. Hasenfuss, G. (1998). Animal models of human cardiovascular dis- ease, heart failure and hypertrophy. Cardiovascular Research, 39, 60– 76. Huser, J., Wang, Y. G., Sheehan, K. A., Cifuentes, F., Lipsius, S. L., & Blatter, L. A. (2000). Functional coupling between glycolysis and excitation- contraction coupling underlies alternans in cat heart cells. The Journal of Physiology, 524(Pt 3), 795– 806. Kanaporis, G., & Blatter, L. A. (2015). The mechanisms of calcium cycling and action potential dynamics in cardiac alternans. Circulation Research, 116, 846– 856. Kanaporis, G., & Blatter, L. A. (2016a). Calcium- activated chloride current determines action potential morphology during cal- cium alternans in atrial myocytes. The Journal of Physiology, 594(3), 699– 714. Kanaporis, G., & Blatter, L. A. (2016b). Ca2+- activated chloride chan- nel activity during Ca2+ alternans in ventricular myocytes. Channels, 10, 507– 517. Kanaporis, G., & Blatter, L. A. (2017a). Alternans in atria: Mechanisms and clinical relevance. Medicina (Kaunas, Lithuania), 53, 139– 149. Kanaporis, G., & Blatter, L. A. (2017b). Membrane potential deter- mines calcium alternans through modulation of SR Ca2+ load and L- type Ca2+ current. Journal of Molecular and Cellular Cardiology, 105, 49– 58. Kanaporis, G., & Blatter, L. A. (2023). Activation of small conduc- tance Ca2+- activated K+ channels suppresses Ca2+ transient and action potential alternans in ventricular myocytes. The Journal of Physiology, 601, 51– 67. Kanaporis, G., Kalik, Z. M., & Blatter, L. A. (2019). Action poten- tial shortening rescues atrial calcium alternans. The Journal of Physiology, 597, 723– 740. Kockskamper, J., & Blatter, L. A. (2002). Subcellular Ca2+ alternans represents a novel mechanism for the generation of arrhyth- mogenic Ca2+ waves in cat atrial myocytes. The Journal of Physiology, 545, 65– 79. Kornyeyev, D., Petrosky, A. D., Zepeda, B., Ferreiro, M., Knollmann, B., & Escobar, A. L. (2012). Calsequestrin 2 deletion shortens the refractoriness of Ca2+ release and reduces rate- dependent Ca2+- alternans in intact mouse hearts. Journal of Molecular and Cellular Cardiology, 52, 21– 31. Lalani, G. G., Schricker, A. A., Clopton, P., Krummen, D. E., & Narayan, S. M. (2013). Frequency analysis of atrial action potential alternans: A sensitive clinical index of individual propensity to atrial fibrillation. Circulation: Arrhythmia and Electrophysiology, 6, 859– 867. Li, Y., Eisner, D. A., & O'Neill, S. C. (2012). Do calcium waves propa- gate between cells and synchronize alternating calcium release in rat ventricular myocytes? The Journal of Physiology, 590, 6353– 6361. Lugo, C. A., Cantalapiedra, I. R., Penaranda, A., Hove- Madsen, L., & Echebarria, B. (2014). Are SR Ca content fluctuations or SR re- fractoriness the key to atrial cardiac alternans? Insights from a human atrial model. American Journal of Physiology Heart and Circulatory Physiology, 306, H1540– H1552. Martinez- Hernandez, E., Kanaporis, G., & Blatter, L. A. (2022). Mechanism of carvedilol induced action potential and calcium alternans. Channels, 16, 97– 112. KANAPORIS et al. McPheeters, M. T., Wang, Y. T., Werdich, A. A., Jenkins, M. W., & Laurita, K. R. (2017). An infrared optical pacing system for screening cardiac electrophysiology in human cardiomyocytes. PLoS One, 12, e0183761. Milani- Nejad, N., & Janssen, P. M. (2014). Small and large animal models in cardiac contraction research: Advantages and disad- vantages. Pharmacology & Therapeutics, 141, 235– 249. Narayan, S. M., Franz, M. R., Clopton, P., Pruvot, E. J., & Krummen, D. E. (2011). Repolarization alternans reveals vulnerability to human atrial fibrillation. Circulation, 123, 2922– 2930. Nivala, M., & Qu, Z. (2012). Calcium alternans in a couplon network model of ventricular myocytes: Role of sarcoplasmic reticulum load. American Journal of Physiology Heart and Circulatory Physiology, 303, H341– H352. Nolasco, J. B., & Dahlen, R. W. (1968). A graphic method for the study of alternation in cardiac action potentials. Journal of Applied Physiology, 25, 191– 196. Panfilov, A. V. (2006). Is heart size a factor in ventricular fibrillation? Or how close are rabbit and human hearts? Heart Rhythm, 3, 862– 864. Picht, E., DeSantiago, J., Blatter, L. A., & Bers, D. M. (2006). Cardiac alternans do not rely on diastolic sarcoplasmic reticulum cal- cium content fluctuations. Circulation Research, 99, 740– 748. Qu, Z., Nivala, M., & Weiss, J. N. (2013). Calcium alternans in car- diac myocytes: Order from disorder. Journal of Molecular and Cellular Cardiology, 58, 100– 109. Qu, Z., & Weiss, J. N. (2007). The chicken or the egg? Voltage and calcium dynamics in the heart. American Journal of Physiology. Heart and Circulatory Physiology, 293, H2054– H2055. Qu, Z., & Weiss, J. N. (2023). Cardiac alternans: From bedside to bench and back. Circulation Research, 132, 127– 149. Qu, Z., Xie, Y., Garfinkel, A., & Weiss, J. N. (2010). T- wave alter- nans and arrhythmogenesis in cardiac diseases. Frontiers in Physiology, 1, 154. Rovetti, R., Cui, X., Garfinkel, A., Weiss, J. N., & Qu, Z. (2010). Spark- induced sparks as a mechanism of intracellular calcium \| 17 of 17 alternans in cardiac myocytes. Circulation Research, 106, 1582– 1591. Shkryl, V. M., Maxwell, J. T., Domeier, T. L., & Blatter, L. A. (2012). Refractoriness of sarcoplasmic reticulum Ca release determines Ca alternans in atrial myocytes. American Journal of Physiology. Heart and Circulatory Physiology, 302, H2310– H2320. Verrier, R. L., Fuller, H., Justo, F., Nearing, B. D., Rajamani, S., & Belardinelli, L. (2016). Unmasking atrial repolarization to as- sess alternans, spatiotemporal heterogeneity, and susceptibility to atrial fibrillation. Heart Rhythm, 13, 953– 961. Walker, M. L., & Rosenbaum, D. S. (2003). Repolarization alternans: Implications for the mechanism and prevention of sudden car- diac death. Cardiovascular Research, 57, 599– 614. Weiss, J. N., Karma, A., Shiferaw, Y., Chen, P. S., Garfinkel, A., & Qu, Z. (2006). From pulsus to pulseless: The saga of cardiac alter- nans. Circulation Research, 98, 1244– 1253. Weiss, J. N., Nivala, M., Garfinkel, A., & Qu, Z. (2011). Alternans and arrhythmias: From cell to heart. Circulation Research, 108, 98– 112. Xie, L. H., Sato, D., Garfinkel, A., Qu, Z., & Weiss, J. N. (2008). Intracellular Ca alternans: Coordinated regulation by sar- coplasmic reticulum release, uptake, and leak. Biophysical Journal, 95, 3100– 3110. Zaragoza, C., Gomez- Guerrero, C., Martin- Ventura, J. L., Blanco- Colio, L., Lavin, B., Mallavia, B., Tarin, C., Mas, S., Ortiz, A., & Egido, J. (2011). Animal models of cardiovascular diseases. Journal of Biomedicine & Biotechnology, 2011, 497841. How to cite this article: Kanaporis, G., Martinez- Hernandez, E., & Blatter, L. A. (2023). Calcium- and voltage- driven atrial alternans: Insight from [Ca]i and Vm asynchrony. Physiological Reports, 11, e15703. https://doi.org/10.14814/phy2.15703 KANAPORIS et al.
10.12688_f1000research.16224.3	F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 RESEARCH ARTICLE Molecular identification and phylogenetic analysis of GABA-producing lactic acid bacteria isolated from indigenous dadih of West Sumatera, Indonesia [version 3; peer review: 2 approved, 1 approved with reservations, 1 not approved] Lili Anggraini 1 Yetti Marlida , Wizna Wizna , Jamsari Jamsari , 4 5,6 5,6 2 2 Frederick Adzitey , Nurul Huda 3 2 , Mirzah Mirzah , 1 2 3 4 5 6 Graduate Program, Andalas University, Padang, West Sumatera, Indonesia Department of Nutrition and Feed Technology, Faculty of Animal Science, Andalas University, Padang, West Sumatera, Indonesia Department of Plant Breeding, Faculty of Agriculture, Andalas University, Padang, West Sumatera, Indonesia Department of Veterinary Science, University for Development Studies, Temale, Ghana School of Food Industry, Universiti Sultan Zainal Abidin, Kuala Nerus, Terengganu, 21300, Malaysia Faculty of Food Science and Nutrition, Universiti Malaysia Sabah, Kota Kinabalu, Sabah, 88400, Malaysia v3 First published: 19 Oct 2018, https://doi.org/10.12688/f1000research.16224.1 ) :1663 ( 7 Second version: 06 Feb 2019, https://doi.org/10.12688/f1000research.16224.2 ) :1663 ( 7 Latest published: https://doi.org/10.12688/f1000research.16224.3 ) 17 Oct 2019, :1663 ( 7 Dadih (fermented buffalo milk) is a traditional Indonesian Abstract Background: food originating from West Sumatra province. The fermentation process is carried out by lactic acid bacteria (LAB), which are naturally present in buffalo milk. Lactic acid bacteria have been reported as one of potential producers of γ-aminobutyric acid (GABA). GABA acts as a neurotransmitter inhibitor of the central nervous system. In this study, molecular identification and phylogenetic analysis Methods: of GABA producing LAB isolated from indigenous dadih of West Sumatera were determined. Identification of the GABA-producing LAB DS15 was based on conventional polymerase chain reaction. 16S rRNA gene sequence analysis was used to identify LAB DS15. Results: approximately 1400 bp amplicon. Phylogenetic analysis showed that LAB DS15 was query coverage to Conclusions: indigenous dadih was , with high similarity of 99% at 100% strain DSM 20284. PCR of the 16S rRNA gene sequence of LAB DS15 gave an It can be concluded that GABA producing LAB isolated from Pediococcus acidilactici Pediococcus acidilactici Pediococcus acidilactici . Keywords indigenous dadih, GABA, LAB, 16S rRNA gene, phylogenetic analysis Open Peer Review Reviewer Status Invited Reviewers 1 2 3 4 report report report report report version 3 (revision) 17 Oct 2019 version 2 (revision) 06 Feb 2019 version 1 19 Oct 2018 1 2 3 Qinglong Wu , Baylor College of Medicine, Houston, USA Jagadish Mahanta , Indian Council of Medical Research (ICMR), Dibrugarh, India Sahilah Abd Mutalib , Universiti Kebangsaan Malaysia (UKM), Selangor, Malaysia Page 1 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 4 Usman Pato , Riau University, Pekanbaru, Indonesia Any reports and responses or comments on the article can be found at the end of the article. Corresponding author: Yetti Marlida ( [email protected] ) Author roles: Anggraini L Conceptualization; Adzitey F : Investigation; Marlida Y : Writing – Review & Editing; : Supervision; Huda N Wizna W : Writing – Review & Editing : Conceptualization; Jamsari J : Conceptualization; Mirzah M : Competing interests: No competing interests were disclosed. This research was supported by Ministry of Research, Technology and Higher Education Republic of Indonesia through Grant information: Master of Education Towards Doctoral Scholarship Program for Excellence Undergraduate and the support through World Class Professor Program Scheme-B No. 123.57/D2.3/KP/2018. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Copyright: which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. . This is an open access article distributed under the terms of the © 2019 Anggraini L et al Creative Commons Attribution License , How to cite this article: lactic acid bacteria isolated from indigenous dadih of West Sumatera, Indonesia [version 3; peer review: 2 approved, 1 approved with reservations, 1 not approved] et al. Molecular identification and phylogenetic analysis of GABA-producing https://doi.org/10.12688/f1000research.16224.3 Anggraini L, Marlida Y, Wizna W F1000Research 2019, :1663 ( 7 ) First published: 19 Oct 2018, 7 :1663 ( https://doi.org/10.12688/f1000research.16224.1 ) Page 2 of 15 REVISED Amendments from Version 2 Additional information on the reference of forward primer 63F (5’- CAG GCC TAA CAC ATG CAA GTC-3’) and reverse primer 1387R (5’-GGG CGG GGT GTA CAA GGC-3’). Updating the sentence of 1% agarose electrophoresis to 1 % (w/v) agarose electrophoresis. Mentioning the marker used 1 Kb Plus DNA ladder (ThermoFisher Scientific). Any further responses from the reviewers can be found at the end of the article acid amino γ-aminobutyric Introduction The non-proteinogenic acid (GABA) is widely found in bacteria, animals, plants, and fungi (Dhakal et al., 2012; Nonaka et al., 2017). GABA acts as a neurotransmitter inhibitor of the central nervous system (Olsen & Li, 2012). It is formed by decarboxylation of L-glutamate, a reaction catalyzed by an enzyme that depends on the peridoxal phosphate of decarboxylated L-glutamate (Murray et al., 2003). Lactic acid bacteria (LAB) have been reported as a potential producer of GABA (Seo et al., 2013; Wu & Shah, 2017). LAB are generally regarded as safe and non-pathogenic microbes, and has been referred to as ‘generally recognized as safe’. Therefore, GABA-producing LAB can be used directly in functional foods (Zhao et al., 2017). Some LAB can be found in the dairy industry for the production of cheese, yogurt, and other fermented milk products (Yamada et al., 2018). Dadih (fermented buffalo milk) is an Indonesian traditional food originating from West Sumatra Province; it is an extremely popular dairy product in Bukittinggi, Padangpanjang, Solok, Lima Puluh Kota, and Tanah Datar, Indonesia (Surono, 2015). Dadih is made from buffalo milk which is fermented in bamboo for 24–48 hours. The fermentation process is carried out by LAB which are naturally present in buffalo milk (Rizqiati et al., 2015) and the environment (Wirawati et al., 2017). Studies have found that, the LAB strains present in dadih are generally Lactobacillus, Streptococcus, Leuconostoc and Lactococcus (Collado et al., 2007; Surono, 2003). Extraction of DNA is a basic principle in molecular analysis and it is one of the success factors in DNA amplification that is used in the analysis of genetic characters (Mustafa et al., 2016). Polymerase chain reaction (PCR) and phylogenetic analysis based on 16S rRNA gene sequences have been used for successful identification of isolates from various fermented food products (Malik et al., 2015). These molecular approaches have allowed Lactobacillus species to be reliably identified (Henry et al., 2015). This research was conducted to identify and isolated from indigenous dadih of West Sumatera based on 16 S rRNA gene sequence analysis. to characterize GABA producing LAB Methods Sample This study used lactic acid bacteria (LAB) DS15, a GABA- producing LAB isolated from dadih of West Sumatera origin. This bacterium was isolated previously according to the method described by Ali et al. (2009). The experiment was carried out at the Feed Technology Industry Laboratory, Faculty of Animal F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Science, Andalas University, West Sumatra, Indonesia. LAB DS15 was grown anaerobically in MRS medium (Merck, Darmstadt, Germany) at 30°C and stored for further analysis. Isolation of bacterial genomic DNA Isolation of the total genome of LAB DS15 was done using Genomic DNA Mini Kit purchased from Invitrogen (Pure- LinkTM, USA) by following the manufacturer’s instructions. We used Lysozyme (PureLinkTM, USA) at a concentration of 20 mg/ml to break down the bacterial cell wall to improve protein or nucleic acid extraction efficiency. 16S rRNA gene amplification Genomic DNA of LAB DS15 was used for amplification of 16S rRNA gene. Amplification was done using forward primer 63F (5’-CAG GCC TAA CAC ATG CAA GTC-3’) and reverse primer 1387R (5’-GGG CGG GGT GTA CAA GGC-3’). of Laboratory of Medical Molecular Biology and Diagnos- tic, Indonesian Institute of Sciences. The reaction was car- ried out in a volume of 50 μl. The PCR mixture contained 22 μl of MQ, 25 μl DreamTaq Green DNA Polymerase (Thermo Fisher Scientific, USA), 1 μl of each forward and reverse primer (10 μM each, IDT synthesized) and 1 μl template. Amplification conditions were 5 minutes of preheat- ing at 95°C, 30 seconds denaturation at 95°C, 30 seconds of primer annealing at 58°C, 1 minute extension step at 72°C and post cycling extension of 5 minutes at 72°C for 35 cycles. The reactions were carried out in a thermal cycler (Biometra’s T-Personal Thermal Cycler, USA). Electrophoresis PCR products were stored at 4°C for further examination using 1% (w/v) agarose electrophoresis in 1x TAE, 100 V for 30 minutes. The DNA bands formed from electrophoresis process was visualized using UV transluminator. The marker used was 1 Kb Plus DNA ladder (ThermoFisher Scientific). Sequence alignment of the 16S rRNA gene Sequencing of the 16S rRNA gene was performed at the Laboratory of Medical Molecular Biology and Diagnostic, Indonesian Institute of Sciences, Jakarta. Sequencing results were edited (contig and peak chromatogram verification) using the SeqManTM II program. Analysis of 16S rRNA sequences of LAB DS15 was performed using NCBI BLAST. Multiple alignment was done using the ClustalX 2.1 program. BioEdit version 7.2.5 in edit mode to see the absence of an inverted sequence and align the sequence length. Kinship visualization was done using the combined phylogenetic tree of the MEGA 7.0.20 program with the Neighbor-Joining hood method (Saitou & Nei, 1987). Results and discussion The identification of LAB DS15 to determine the strain was done based on 16S rRNA gene. The first step was amplification using PCR method, with the electrophoresis image shown in Supplementary File 1. The amplification process was carried out to obtain more copies of the 16S rRNA gene for the sequencing process. Analysis of sequencing results begun by aligning the base sequence of the 63F forward sequence and reverse 138R using the SegMan program. PCR of the 16S rRNA gene of LAB DS15 gave an approximately 1400 bp amplicon (Figure 1). Page 3 of 15 Saitou & Nei (1987) indicated that the evolutionary history of organisms can be known using the neighbour-joining method. Organisms within the same taxa are normally clustered together in the phylogenetic tree and have better bootstrap values (Felsenstein, 1985). In this study, we drew a phylogenetic tree to scale and determined the evolutionary distances using the p-distance method. A total of 26 nucleotide sequences and codon positions 1st + 2nd + and 3rd + noncoding were con- sidered, using MEGA 7.0 as reported by Kumar et al. (2016) for evolutionary analyses. Figure 1. Agarose gel (1%) electrophoresis showing amplified 16S rRNA gene of LAB DS18. M, DNA marker; 1, PCR product of LAB DS18. F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 DNA sequencing results were analyzed using NCBI BLAST. According to Willey et al. (2009), 16S rRNA sequencing looks at the similarity of isolates to those already available in GenBank; this is one molecular detection method that is ideal enough to know the kinship relationship between bacteria because the 16S rRNA sequence is a gene found in all microbes and is indispensable in maintain life. The 16S rRNA gene sequencing identified the LAB DS15 to belong to the genus Pediococcus, forming a well-defined cluster with Pediococcus acidilactici. This cluster was recovered in 100% of bootstrap analysis. Pediococcus spp. are widely described as probiotics (Porto et al., 2017). Abbasiliasi et al. (2012) also found Pediococcus acidilactici in fermented milk products. Pediococcus acidilactici are important LAB which have been used as starter cultures in meat, vegetable and dairy fermentation causing charac- teristic flavor changes, improving hygiene and extending the shelf life of these products (Mora et al., 1997; Porto et al., 2017). A phylogenetic tree (Figure 2) was constructed to determine the kinship relationship of LAB DS15. The phylogenetic tree is known to show a high consistency of the relationships between organisms. In this study, the isolate showed similarity of 99% at 100% query coverage to Pediococcus acidilactici strain DSM 20284. A value of 99% indicates that the isolate can be considered as the same species with Pediococcus acidilac- tici strain DSM 20284. The sequence of homology levels was high, as shown by the red color with a score of ≥200 (Figure 3). Figure 2. Phylogenetic tree of 16S rRNA gene of LAB DS18 using the neighbor-joining method. Page 4 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Figure 3. Graphic summary of DS18 and Pediococcus acidilactici strain DSM 20284. From the results of this homology it can be concluded that the two sequences are the same and have an evolutionary relationship. LAB DS15 was Pediococcus acidilactici, with 99% similarity to Pediococcus acidilactici strain DSM 20284. The next closest species for which a sequence alignment of at least 100% query coverage was observed were Pediococcus pentosaceus strain DSM 20336, Pediococcus acidilactici strain NGRI 0510Q and Pediococcus argentini strain CRL 776 at 98% similarity to the DS15 isolate. Pediococcus stilesi strain FAIR-E 180 showed 98% similarity with 99% query coverage. An alignment query result of 100% indicates a significant alignment, which means the search sequence in this study was identical with the species level. identified genus, even at the Conclusion The PCR of 16S rRNA gene sequence gave an approximately 1400 bp amplicon for LAB DS15, isolated from indigenous dadih of West Sumatera. Phylogenetic analysis showed that Data availability Pediococcus acidilactici strain DS32 16S ribosomal RNA gene, partial sequence, obtained during this study. GenBank accession http://identifiers.org/ncbigi/ GI:1481059229. number MH938236: Grant information This research was supported by Ministry of Research, Technology and Higher Education Republic of Indonesia through Master of Education Towards Doctoral Scholarship Program for Excellence Undergraduate and the support through World Class Professor Program Scheme-B No. 123.57/D2.3/KP/2018. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Supplementary material Supplementary File 1. Electrophoresis image of the PCR amplification product. Click here to access the data. Page 5 of 15 References Abbasiliasi S, Tan JS, Ibrahim TA, et al.: Isolation of Pediococcus acidilactici Kp10 with ability to secrete bacteriocin-like inhibitory substance from milk products for applications in food industry. BMC Microbiol. 2012; 12: 260. PubMed Abstract \| Publisher Full Text \| Free Full Text Ali FWO, Abdulamir AS, Mohammed AS, et al.: Novel, practical and cheap source for isolating beneficial γ-aminobutyric acid-producing leuconostoc NC5 bacteria. Res J Med Sci. 2009; 3(4): 146–153. Reference Source Collado CM, Surono IS, Meriluoto J, et al.: Potential probiotic characteristics of Lactobacillus and Enterococcus strains isolated from traditional dadih fermented milk against pathogen intestinal colonization. J Food Prot. 2007; 70(3): 700–705. PubMed Abstract \| Publisher Full Text Dhakal R, Baipai VK, Baek KH: Production of gaba (γ - Aminobutyric acid) by microorganisms: a review. Braz J Microbiol. 2012; 43(4): 1230–41. PubMed Abstract \| Publisher Full Text \| Free Full Text Felsenstein J: Confidence Limits On Phylogenies: An Approach Using The Bootstrap. Evolution. 1985; 39(4): 783–791. PubMed Abstract \| Publisher Full Text Henry DE, Halami PM, Prapulla SG: Lactobacillus plantarum mcc2034, a novel isolate from traditional Indian lactic fermented preparation: molecular identification and evaluation of its in vitro probiotic potential. J Microbiol Biotechnol Food Sci. 2015; 4(4): 328–331. Publisher Full Text Kumar S, Stecher G, Tamura K: MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Mol Biol Evol. 2016; 33(7): 1870–1874. PubMed Abstract \| Publisher Full Text Malik V, Devi U, Yadav RNS, et al.: 16s rRNA based phylogenetic analysis of lactobacillus plantarum isolated from various fermented food products of Assam. J Microbiol Biotechnol Food Sci. 2015; 5(1): 20–22. Publisher Full Text Mora D, Fortina MG, Parini C, et al.: Identification of Pediococcus acidilactici and Pediococcus pentosaceus based on 16s rRNA and ldhD gene-targeted multiplex PCR analysis. FEMS Microbiol Lett. 1997; 151(2): 231–236. PubMed Abstract \| Publisher Full Text Murray RK, Granner DK, Rodwell VW, et al.: Harper’s Illustrated Biochemistry 23th edn. McGraw-Hill Companies, Inc. USA. 2003. Mustafa H, Rachmawati I, Udin Y: Genomic DNA concentration and purity measurement of Anopheles barbirostris. Journal of Disease Vector. 2016; 1: 7–10. Publisher Full Text Nonaka S, Arai C, Takayama M, et al.: Efficient increase of γ-aminobutyric acid (GABA) content in tomato fruits by targeted mutagenesis. Sci Rep. 2017; 7(1): 7057. PubMed Abstract \| Publisher Full Text \| Free Full Text F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Olsen RW, Li GD: Gaba. In Brady ST, Siegel GJ, Albers LW, Price DL, (Editors). Basic Neurochemistry (Eight edition): Principles of molecular, cellular, and medical neurobiology. 2012; 367–376. Publisher Full Text Porto MC, Kuniyoshi TM, Azevedo PO, et al.: Pediococcus spp.: An important genus of lactic acid bacteria and pediocin producers. Biotechnol Adv. 2017; 35(3): 361–374. PubMed Abstract \| Publisher Full Text Rizqiati H, Sumantri C, Noor RR, et al.: Isolation and identification of indigenous lactic acid bacteria from North Sumatra river buffalo milk. Indonesian Journal of Animal and Veterinary Sciences. 2015; 20(2): 87–94. Publisher Full Text Saitou N, Nei M: The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol. 1987; 4(4): 406–425. PubMed Abstract \| Publisher Full Text Seo MJ, Lee JY, Nam YD, et al.: Production of γ-Aminobutyric Acid by Lactobacillus brevis 340G Isolated from Kimchi and Its Application to Skim Milk. Food Eng Prog. 2013; 17(4): 418–423. Publisher Full Text Surono IS: In vitro probiotic properties of indigenous dadih lactic acid bacteria. Asian-Australas J Anim Sci. 2003; 16(5): 726–731. Publisher Full Text Surono IS: Traditional Indonesian dairy foods. Asia Pac J Clin Nutr. 2015; 24 Suppl 1: S26–S30. PubMed Abstract \| Publisher Full Text Willey JM, Sherwood LM, Woolverton CJ: Prescott’s Principles of Microbiology. Boston: McGraw-Hill Higher Education. 2009. Reference Source Wirawati CU, Sudarwanto MB, Lukman DW, et al.: Characteristic and development of cow’s milk dadih as an alternate of buffalo’s milk dadih. WARTAZOA Indonesian Bulletin of Animal and Veterinary Sciences. 2017; 27(2): 95–103. Publisher Full Text Wu Q, Shah NP: High γ-aminobutyric acid production from lactic acid bacteria: Emphasis on Lactobacillus brevis as a functional dairy starter. Crit Rev Food Sci Nutr. 2017; 57(17): 3661–3672. PubMed Abstract \| Publisher Full Text Yamada Y, Endou M, Morikawa S, et al.: Lactic Acid Bacteria Isolated from Japanese Fermented Fish (Funa-Sushi) Inhibit Mesangial Proliferative Glomerulonephritis by Alcohol Intake with Stress. J Nutr Metab. 2018; 2018: 6491907. PubMed Abstract \| Publisher Full Text \| Free Full Text Zhao A, Hu X, Wang X: Metabolic engineering of Escherichia coli to produce gamma-aminobutyric acid using xylose. Appl Microbiol Biotechnol. 2017; 101(9): 3587–3603. PubMed Abstract \| Publisher Full Text Page 6 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Open Peer Review Current Peer Review Status: Version 3 Reviewer Report 12 March 2020 https://doi.org/10.5256/f1000research.22366.r56703 © 2020 Pato U. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License work is properly cited. , which permits unrestricted use, distribution, and reproduction in any medium, provided the original Usman Pato Faculty of Agriculture, Riau University, Pekanbaru, Indonesia INTRODUCTION 1. In general, the introduction is relatively good but needs to be added by the results of research from Hosono et al 1989 and Wirawati et al., 2019 about micflora in dadih 1 2 2. In the introduction, the author needs to explain in more detail the role of GABA produced by LAB and other organisms METHODS 1. An explanation should be added as to why only to choose the DS15 strain producing GABA in this study. 2. It is necessary to add the reference methods used in the 16S rRNA gene amplification analysis and electrophoresis process RESULTS AND DISCUSSION The results are well presented and discussed systematically because the authors used only one strain. REFERENCES The author needs to add references as a follow-up to suggestions for improvement in the introduction and method of this paper The strength of this paper The strength of study is the first research to report on GABA-producing LAB from dadih and local fermented milk products from Indonesia The weakness of this paper One of the weaknesses of this study is that the authors only used one LAB dadih isolate (strain DS15) so that no comparative data were produced and the discussion was relatively limited. References 1. Hosono A, Wardojo R, Otani H: Microbial flora in dadih a traditional fermented milk in indonesia. Life, Earth . Page 7 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 . Earth 2. Wirawati CU, Sudarwanto MB, Lukman DW, Wientarsih I, et al.: Diversity of lactic acid bacteria in dadih produced by either back-slopping or spontaneous fermentation from two different regions of West Sumatra, Indonesia. PubMed Abstract Publisher Full Text \| (6): 823-829 Vet World . 2019; 12 Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? No Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? No Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: Food Microbiology, Probiotic I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Reviewer Report 25 November 2019 https://doi.org/10.5256/f1000research.22366.r55312 © 2019 Wu Q. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Qinglong Wu Texas Children's Microbiome Center, Baylor College of Medicine, Houston, TX, USA I did not see any improvements of scientific value that have been made in the revision. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Page 8 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Yes Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests: No competing interests were disclosed. I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above. Version 2 Reviewer Report 11 June 2019 https://doi.org/10.5256/f1000research.19627.r48482 © 2019 Mutalib S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License work is properly cited. , which permits unrestricted use, distribution, and reproduction in any medium, provided the original Sahilah Abd Mutalib Centre for Biotechnology and Functional Food, Faculty of Science and Technology, Universiti Kebangsaan Malaysia (UKM), Selangor, Malaysia 1. Introduction - fairly good and can be improved i. Dadih from Indonesia has- Lactobacillus Streptococcus Leuconostoc identify the bacteria? Biochemical tests or using molecular approaches? ii. Is there any data on dadih from Malaysia as well for comparison. , , and Lactococcus - How did they 2. Methods - can be improved 2.1 Sample - Subtopic sample suggested to change - Bacterial strain The month and year of the bacterium should be mentioned for ex: in June 2009. Why too long to continue the partial sequence 16s rRNA analysis? 2. 2 Isolation of bacterial genomic DNA We used lysozyme-change to "Twenty(20) mg/ml of lysozyme was used to break down ...." Please state where did you keep the genomic DNA. Example in -20 C freezer or 4 C refrigerator prior o o analysis Page 9 of 15 analysis F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 2.3 16S rRNA gene amplification Please state the reference after forward and reverse primers is mentioned. 2.4 Electrophoresis 1% change to 1% (w/v) in 1x change to1x TAE, 100 V The marker should be mentioned in this section, 1 Kb ladder? What kind of dye did you used? Red dye, syber green, ethidium bromide? State their brand as well Results and discussion Good - due to a single strain/isolate was studied, thus, the explanation is straight forward. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? Yes Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? No Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? Yes Competing Interests: No competing interests were disclosed. Reviewer Expertise: Food microbiology, Halal Science, biomass degradation (EFB and POME) I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Author Response 14 Aug 2019 Nurul Huda , Universiti Malaysia Sabah, Malaysia, Malaysia 1. Introduction - fairly good and can be improved i. Dadih from Indonesia has- did they identify the bacteria? Biochemical tests or using molecular approaches? Lactobacillus Streptococcus Leuconostoc and , , Lactococcus - They identify bacteria with a molecular approach using the 16SsRNA technique ii. Is there any data on dadih from Malaysia as well for comparison. How Page 10 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 ii. Is there any data on dadih from Malaysia as well for comparison. No, we don’t have data on dadih from Malaysia. 2. Methods - can be improved 2.1 Sample - Subtopic sample suggested to change - Bacterial strain The month and year of the bacterium should be mentioned for ex: in June 2009. Why too long to continue the partial sequence 16s rRNA analysis? Bacterial strains isolated in July 2017. We did this isolation based on the method of Ali et al., (2009), not isolates from the author. 2. 2 Isolation of bacterial genomic DNA We used lysozyme-change to "Twenty (20) mg/ml of lysozyme was used to break down ...." Please state where did you keep the genomic DNA. Example in -20 C freezer or 4 C refrigerator prior Analysis We keep the genomic DNA in 4 C refrigerator. o 2.3 16S rRNA gene amplification Please state the reference after forward and reverse primers is mentioned. We got reference for forward primer 63F (5'-CAG GCC TAA CAC ATG CAA GTC-3') and reverse primer 1387R (5'-GGG CGG GGT GTA CAA GGC-3') from Laboratory of Medical Molecular Biology and Diagnostic, Indonesian Institute of Sciences, Jakarta, Indonesia. 2.4 Electrophoresis 1% change to 1% (w/v) Ok we will change in 1x change to1x TAE, 100 V Ok we will change The marker should be mentioned in this section, 1 Kb ladder? What kind of dye did you used? Red dye, syber green, ethidium bromide? State their brand as well We used 1 Kb Plus DNA Ladder (ThermoFisher Scientific) Results and discussion Good - due to a single strain/isolate was studied, thus, the explanation is straight forward. Thank You. Competing Interests: No competing interests were disclosed. Reviewer Report 08 April 2019 https://doi.org/10.5256/f1000research.19627.r46293 Page 11 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 © 2019 Mahanta J. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License work is properly cited. , which permits unrestricted use, distribution, and reproduction in any medium, provided the original Jagadish Mahanta Regional Medical Research Centre, Indian Council of Medical Research (ICMR), Dibrugarh, Assam, India Authors wanted to identify and characterize GABA producing LAB isolated from “Dadih”. 1. 2. 3. 4. However, authors have taken a strain already isolated and identified in 2009. Authors have not mentioned anything about the gap in the previous research that necessitated undertaking the present exercise. Authors may clarify the issue. Authors have done elaborate molecular testing and phylogenetic analysis of the bacteria taken from the stock. Authors should elaborate the achievement of this exercise. As emphasized by the authors, they should elaborate, about characterization and GABA production potential of the strain Authors should elaborate on the novelty of the study. Is the work clearly and accurately presented and does it cite the current literature? Yes Is the study design appropriate and is the work technically sound? No Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Not applicable Are all the source data underlying the results available to ensure full reproducibility? Partly Are the conclusions drawn adequately supported by the results? Yes Competing Interests: No competing interests were disclosed. I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Author Response 10 Apr 2019 Nurul Huda , Universiti Malaysia Sabah, Kota Kinabalu, Malaysia Authors wanted to identify and characterize GABA producing LAB isolated from “Dadih”. 1. However, authors have taken a strain already isolated and identified in 2009. Authors have not mentioned anything about the gap in the previous research that necessitated undertaking the Page 12 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 1. However, authors have taken a strain already isolated and identified in 2009. Authors have not mentioned anything about the gap in the previous research that necessitated undertaking the present exercise. Authors may clarify the issue. Exploration of isolates from dadih has been carried out, but no studies have used these isolates as GABA producers. In this study we obtained DS15 isolate from the isolation of various fermented foods which had the highest GABA production. We did this isolation based on the method of Ali et al., (2019), not isolates from the author. 2. Authors have done elaborate molecular testing and phylogenetic analysis of the bacteria taken from the stock. Authors should elaborate the achievement of this exercise. The result of BLAST at the NCBI GenBank site from the sequences showed that DS15 isolate were Pediococcus acidilactici with P. acidilactici P. acidilactici similarity. DSM 20284, with the difference of one base pair. The next closest species were . Based on the phylogenetic tree, DS15 has a 99% similarity or homology FAIR-E 180 shows 98% similarity with 99% query coverage. CRL 776 with 98% DSM 20336 and P. pentosaceus P. argentinicus NGRI 0510Q, P. stilesi 3. As emphasized by the authors, they should elaborate, about characterization and GABA production potential of the strain. We have carried out quantitative screening on some of the isolates we obtained from dadih, and we found that DS15 isolates produced the highest amount of GABA. Data and discussion are used in another publications. 4. Authors should elaborate on the novelty of the study. The novelty of this study was the use of bacterial isolates from dadih as a GABA producer. Competing Interests: No competing interests were disclosed. Reviewer Report 01 April 2019 https://doi.org/10.5256/f1000research.19627.r46425 © 2019 Wu Q. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Qinglong Wu Texas Children's Microbiome Center, Baylor College of Medicine, Houston, TX, USA The authors detailed the 16S rRNA gene-based to be a good lab protocol without demonstrating any scientific value. There is no experimental data to support the GABA production from isolate DS15. They have to present the GABA data in terms of GABA yield under defined fermentation conditions. Meanwhile, they have to demonstrate the pathway in isolate DS15 that is responsible for GABA biosynthesis and GABA export in this strain. Page 13 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 Secondly, the authors just use one isolate to achieve the claim "GABA producing LAB isolated from indigenous dadih was . This is not a rigorous way. Pediococcus acidilactici" Here are my questions: 1. 2. 3. What is the level of GABA in dadih? How is GABA production capacity of DS15? There is no massive bacterial isolation from dadih; neither no microbial community profiling for dadih, nor pathway identification of GABA production for microbial community of dadih; so one isolate from dadih does not mean anything. Is the work clearly and accurately presented and does it cite the current literature? Partly Is the study design appropriate and is the work technically sound? No Are sufficient details of methods and analysis provided to allow replication by others? Yes If applicable, is the statistical analysis and its interpretation appropriate? Not applicable Are all the source data underlying the results available to ensure full reproducibility? Partly Are the conclusions drawn adequately supported by the results? No Competing Interests: No competing interests were disclosed. Reviewer Expertise: Food microbiology, microbiome science, microbial genomics, functional genomics, microbial GABA biosynthesis, biochemistry I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above. Author Response 10 Apr 2019 Nurul Huda , Universiti Malaysia Sabah, Kota Kinabalu, Malaysia 1. What is the level of GABA in dadih? We don't count the amount of GABA on dadih. We did not count the amount of GABA produced in dadih. GABA is produced by lactic acid bacteria of dadih origin but not the dadih, so we don’t count or determine GABA level of dadih 1. How is GABA production capacity of DS15? GABA production capacity of DS15 was 49.365 mg/L 1. There is no massive bacterial isolation from dadih; neither no microbial community profiling Page 14 of 15 F1000Research 2019, 7:1663 Last updated: 12 MAR 2020 1. There is no massive bacterial isolation from dadih; neither no microbial community profiling for dadih, nor pathway identification of GABA production for microbial community of dadih; so, one isolate from dadih does not mean anything. In this study, we isolated bacteria from various fermented food products (dadih, ikan budu, asam durian and tape singkong), determined their GABA producing ability and selected the isolate with the highest GABA production for further identification. The results of isolation and characterization are explained in other articles. The distribution of LAB isolates from the indigenous West Sumatera fermented food (dadih only) is; 1. 2. 3. Origin from Aiadingin area. Number of isolate 131; Number of LAB isolate 125; Number of GABA producing LAB isolate 23. Origin from Sijunjung area. Number of isolate 166; Number of LAB isolate 93; Number of GABA producing LAB isolate 19. Origin from Solok area. Number of isolate 100; Number of LAB isolate 96; Number of GABA producing LAB isolate 19. In total, from 3 areas, number of isolate 397; number of LAB isolate 314; number of GABA producing LAB isolate 62. Competing Interests: No competing interests were disclosed. The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact [email protected] Page 15 of 15
10.3390_s19071571	Article Hand-held Colorimetry Sensor Platform for Determining Salivary α-Amylase Activity and Its Applications for Stress Assessment Hsien-Yi Hsiao 1, Richie L. C. Chen 1, Chih-Chi Chou 1 and Tzong-Jih Cheng 1,2,* 1 Department of Bio-industrial Mechatronics Engineering, College of Bio-Resources and Agriculture, National Taiwan University, Taipei 100617, Taiwan; [email protected] (H.-Y.H.); [email protected] (R.L.C.C.); [email protected] (C.-C.C.) 2 Department of Biomedical Engineering, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei 10002, Taiwan * Correspondence: [email protected]; Tel.: +886-2-3366-5345 Received: 28 January 2019; Accepted: 28 March 2019; Published: 1 April 2019 Abstract: This study develops a hand-held stress assessment meter with a chemically colorimetric strip for determining salivary α-amylase activity, using a 3,5 dinitrosalicylic acid (DNS) assay to quantify the reducing sugar released from soluble starch via α-amylase hydrolysis. The colorimetric reaction is produced by heating the strip with a mini polyester heater plate at boiling temperature to form a brick red colored product, which measured at 525 nm wavelength. This investigation describes in detail the design, construction, and performance evaluation of a hand-held α-amylase activity colorimeter with a light emitted diode (LED) and photo-detector with built-in ﬁlters. The dimensions and mass of the proposed prototype are only 120 × 60 × 60 mm3 and 200 g, respectively. This prototype has an excellent correlation coefﬁcient (>0.995), comparable with a commercial ultraviolet–visible spectroscope, and has a measurable α-amylase activity range of 0.1–1.0 U mL−1. The hand-held device can measure the salivary α-amylase activity with only 5 µL of saliva within 12 min of testing. This sensor platform effectively demonstrates that the level of salivary α-amylase activity increases more signiﬁcantly than serum cortisol, the other physiological stressor biomarker, under physiologically stressful exercise conditions. Thus, this work demonstrates that the hand-held α-amylase activity meter is an easy to use and cost-effective stress assessment tool for psychoneuroendocrinology research. Keywords: hand-held optical sensing meter; stress assessment; salivary α-amylase; colorimetry 1. Introduction Stress, stress-related diseases, and stress assessments are critical issues in clinical, psychological, biomedical, and sport medicine research. A stressor causes the two major biological stress systems, the Sympathetic–Adrenal–Medullary (SAM) system and the Hypothalamic–Pituitary–Adrenal system (HPA), to release hormones that control of most of the body’s internal organs, causing symptoms such as elevated heart and breathing rates [1]. Certain hormones in saliva, such as cortisol and catecholamines (norepinephrine, NE), that are released from the blood and are found in an unbound free state in saliva, are considered to be good stress biomarkers, and are widely monitored to measure endocrinological stress [2]. This is mainly because saliva sampling is non-invasive and not stressful, and does not require specialized personnel [1]. However, salivary catecholamines have concentrations several-fold lower than those of venous blood, and do not reﬂect the actuate changes in the blood catecholamines released by activation of SAM. Sensors 2019, 19, 1571; doi:10.3390/s19071571 www.mdpi.com/journal/sensors sensors(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Sensors 2019, 19, 1571 2 of 14 Several works have reported that salivary α-amylase (sAA) level is signiﬁcantly positively correlated with plasma norepinephrine [1,2], and increases in response to both physical (e.g., exercise, heat and cold stress) and psychological (e.g., written examinations) stress stimuli via the SAM system, which is controlled and ampliﬁed with low levels of norepinephrine release in salivary glands [3]. The SAM system has a much faster response time of sAA secretion, from 1 min to a few minutes, than the endocrine system [4]. Therefore, sAA has been recommended to replace norepinephrine as a sensitive biomarker for stress via the SAM system, because it is very easy to monitor [5]. Alpha-amylase (EC 3.2.1.1) is a major protein component of saliva, and plays a key role in carbohydrate metabolism by hydrolyzing the 1,4-α-D-glycosidic linkages of starch components, glycogen, and various oligosaccharides to form maltose [6]. This major function of sAA is very important for food digestion. Therefore, the increase in sAA activity in response to stressful conditions, such as exercise, is considered to apply to intense energy expenditure. Analytic approaches for assay of α-amylase activity fall into two main categories based on the carbohydrate hydrolysis mechanism. One group of methods measures the consumption of natural starch hydrolyzed by α-amylase. Some examples of this approach include viscometries by capillary viscometer [7], magnetoelastic cantilever [8] or piezoelectric transducer [9,10], electrochemically amperometry [11–16], potentiometry [17], colorimetry of starch-iodine [18,19] or a chromogenic substrate [20] color intensity, and liquid crystals-based sensing platform [21]. These methods generally have poor reproducibility, with large differences associated with distinct operators or slight protocol variations, because the attack speed of α-amylase is deeply inﬂuenced by varying molecular mass and branching degrees of heterogeneous starch macromolecules [22]. The other main category of methods overcomes those deﬁciencies by directly determining the amount of cleaved bonds in the substrate, the measure of which can be readily converted into International Units of enzymatic activity [23]. These methods are classiﬁed into two sub-categories, the chromogenic and saccharogenic approaches, according to analytical approach. In chromogenic methods, the release amounts of soluble dye are directly measured from covalently chemical-modiﬁed starch or maltosaccharide substrates, such as Cibacron Blue F3 G-A cross-linked starch [24] or nonreducing end-blocked 2-chloro-p-nitrophenyl maltohepotosides (Gal-G2-CNP) [25]. Chromogenic methods are simple, reliable, and sensitive for α-amylase determination, but are extremely expensive because they require a synthetic substrate and speciﬁc enzymes. Saccharogenic methods measure the release amount of reducing sugars from unmodiﬁed starch by colorimetric oxidant reagents, and include the Nelson–Somogyi assay [26] and 3,5-dinitrosalicyclic acid (DNS) assay [27]. These methods are reliable and low-cost, and are adopted extensively to measure α-amylase activity without any further modiﬁcation in the procedure. However, they are not suitable for daily frequent use, because they require analytic instruments, and involve complicated analytic processes [27]. To improve the reproducibility, convenience, and cost in analytical applications, many portable optical-sensing instruments have been developed for process the monitoring and medical diagnosis ﬁelds, because they can be operated by lay people, and their results are obtained in the ﬁeld and within a good time frame [28–32]. These hand-held colorimetric devices are generally miniaturized with light-emitting diodes (LEDs) as quasimonochromatic light sources for sensor-exciting and solid-state photodiodes for absorbance measurements. With appropriate bias and support circuits, the low-energy consumption, stability, and sensitivity of these optoelectronic devices have been shown to have the desired analytical performance, including accuracy. This study develops a compact hand-held colorimetric device with a homemade disposable strip for determining sAA activity, and illustrates its feasibility for stress assessment. The sensing principle is based on the saccharogenic method, in which a DNS oxidant reagent is utilized to monitor the amount of reducing sugars released from soluble starch within the sAA hydrolysis at 37 ◦C. The hydrolytic solutions and reagents were injected into a homemade disposable strip, mixed with the vibration unit, and heated by a polyimide heater to form color at high temperature. The color results were then directly measured using an optical-sensing platform, to obtain the sAA activity. The experimental procedure addressed in detail the issues associated with the implementation of key components, disposable strip Sensors 2019, 19, 1571 3 of 14 design, electronic and signal conditionings, and mechanism/optical chamber design. The analytical performance of the proposed device was compared with that of a commercial spectrophotometric instrument. Finally, the proposed hand-held device was then applied as a stress assessment platform to measure the sAA level in saliva collected under physiological exercise stress conditions and daily routine conditions. 2. Materials and Methods 2.1. Electronic Instruments and Machines for Developing a Hand-Hold Colorimeter A power supply (GW instek, GPC-3030DQ, New Taipei City, Taiwan), oscilloscope (Tektronix, TDS2002, Taipei, Taiwan), function generator (GW instek, GFG-8015G, New Taipei City, Taiwan), multimeter (DHA, DMM-93B, Taipei, Taiwan), and frequency counter (Hewlett-Packard, 53131A, Palo Alto, CA, USA) were employed as electronic instruments to develop and verify the electronic hardware of the prototype. An in-circuit emulator for an 80S51 microprocessor (Ocean Sky Technology, TE-S51, New Taipei City, Taiwan) was used as the development platform. An UV/Vis spectroscope (JASCO, F-530, Tokyo, Japan) was adopted as a reference instrument to evaluate the consistency of the developed prototype. A laser cutting machine (LLC EZLaser, LCR-200R, Hsinchu, Taiwan) was utilized in the design and construction of the reactor chips by cutting polycarbonate from compact discs. 2.2. Chemicals and Reagents Sodium hydroxide, potassium sodium tartrate, disodium hydrogenphosphate, and potassium dihydrogenphosphate were obtained from Nacalai Tesque, Kyoto, Japan. The 3,5-dinitrosalicylic acid was acquired from Sigma-Aldrich, St. Louis, MO, USA. α-amylase (EC 3.2.1.1, 20 U mg−1, from Bacillus subtilis) and soluble starch (solubility 20 mg mL−1 at 25 ◦C, pH 4.0–6.0, from potato) were purchased from Wako, Osaka, Japan. All chemicals were of analytical-reagent grade, and were used as received Starch stock solution (10 mg mL−1) and α-amylase stock solution (1 mg mL−1) were prepared by dissolving starch and α-Amylase powders in 10 mM phosphate buffer (pH 6.90), respectively, and stored at 4 ◦C. The DNS reagent was prepared by dissolving 0.250 g 3,5-dinitrosalicylic acid and 75.0 g potassium sodium tartrate in 50 mL, 2 M NaOH solution, and diluted with de-ionized (DI) water to total volume 250 mL. Buffers and solutions were prepared with deionized water with conductivity < 1 µS cm−1. 2.3. Measurement of α-Amylase Activity Amylase activity was determined by measuring the amount of reducing sugars released from soluble starch using alkaline DNS oxidant reagent [33]. Thirty microliters of 10 mg mL−1 soluble starch solution was mixed with 20 µL of α-amylase solution or salivary sample, and incubated for 1 min at 37 ◦C. The hydrolyzing solution was injected into a home-made disposable strip, and the reaction was stopped with the addition of 150 µL DNS reagent. The mixture was mixed uniformly with the micro-eccentric vibrator, and heated with ﬂexible polyester heater (above 80 ◦C) for 4 min to form color. The reducing groups were quantiﬁed at 525 nm with an optical-sensing platform for measuring α-amylase activity. Measurements were also performed on a blank without substrate but with α-amylase, and a control containing no α-amylase but with substrate, at the same time as the reaction mixtures. Each measurement procedure was performed several times, with measurements taken in duplicate. 2.4. Determination of Stress Levels The salivary α-amylase activity and serum cortisol concentration were measured to assess the stress level of a volleyball player following exercises of different intensity. The male was 25 years old, and had no oral diseases. The aim of the experiment was explained to the subject, and consent was Sensors 2019, 19, 1571 4 of 14 obtained after conﬁrmation that he fully understood the experiment. Table 1 shows the self-training exercise conditions and the collection times of the blood samples. At conditions III and V, the blood was sampled 1 h after termination of the exercise. The subject voluntarily provided serum cortisol data from clinical reports of laboratory chemistry tests. The experimental procedure is brieﬂy described as follows. All blood samples were accumulated in sterile tubes containing ethylenediaminetetraacetic acid (EDTA) and heparin. The total serum cortisol level of blood samples were measured in enzymatic mode with a clinical chemistry system (SIEMENS, Dimension RXL, Erlangen, Germany) in the UNION clinical laboratory, Taiwan. The saliva was collected 30 min before blood sampling, immediately after blood sampling, and at 30-min intervals thereafter. To minimize error variance and chances of saliva secretion, food or drink intake must be avoided at least 1 h prior to sampling. Before sampling, the participant should clean his mouth with warm water, at least 10 min before collection, to eliminate any residues. The whole saliva samples were gathered using the “passive drool” method [34], and diluted with 20-fold volume of 10 mM phosphate buffer (pH 6.90). The α-amylase activity of the diluted saliva sample was determined with the hand-held colorimeter based on DNS coloration. 3. Results and Discussions 3.1. System Realization The major objective of this prototype was to develop a low-cost portable colorimeter with a heat-resistant disposable strip, in order to obtain an easy-to-use piece of equipment intended for lay-personnel in analytical chemistry technology. Figure 1 illustrates a block diagram of the overall instrument design with a circuit. The two main components of the hand-held instrument are the reaction/optoelectronic and the electronic signal-processing units. The schematic diagram also describes the transformation of the signal from a light signal to a frequency-modulated signal and then to digital information feeding into the microprocessor (PC/PDA). Our previous work [35] revealed that the combination of a photo-detector with frequency-modulated digital output and a frequency-digital converter are cost-effective and easy to implement. The following sub-section describes the design and fabrication of disposable strips, as well as the heat/spatial conﬁguration to support the light pathway. The later paragraphs describe in detail the two principal components of the device, and elucidate all aspects of their operation. Figure 1. Function block diagram of hand-held photometer. PC: personal computer; PDA: personal digital assistant; MAX323: precision analog switches, MAXIM; TTL: transistor-transistor logic; LED: light-emitting diode. Sensors 2019, 19, x 4 of 14 data from clinical reports of laboratory chemistry tests. The experimental procedure is briefly described as follows. All blood samples were accumulated in sterile tubes containing ethylenediaminetetraacetic acid (EDTA) and heparin. The total serum cortisol level of blood samples were measured in enzymatic mode with a clinical chemistry system (SIEMENS, Dimension RXL, Erlangen, Germany) in the UNION clinical laboratory, Taiwan. The saliva was collected 30 min before blood sampling, immediately after blood sampling, and at 30-minute intervals thereafter. To minimize error variance and chances of saliva secretion, food or drink intake must be avoided at least 1 h prior to sampling. Before sampling, the participant should clean his mouth with warm water, at least 10 min before collection, to eliminate any residues. The whole saliva samples were gathered using the “passive drool” method [34], and diluted with 20-fold volume of 10 mM phosphate buffer (pH 6.90). The α-amylase activity of the diluted saliva sample was determined with the hand-held colorimeter based on DNS coloration. 3. Results and Discussions 3.1. System Realization The major objective of this prototype was to develop a low-cost portable colorimeter with a heat-resistant disposable strip, in order to obtain an easy-to-use piece of equipment intended for lay-personnel in analytical chemistry technology. Figure 1 illustrates a block diagram of the overall instrument design with a circuit. The two main components of the hand-held instrument are the reaction/optoelectronic and the electronic signal-processing units. The schematic diagram also describes the transformation of the signal from a light signal to a frequency-modulated signal and then to digital information feeding into the microprocessor (PC/PDA). Our previous work [35] revealed that the combination of a photo-detector with frequency-modulated digital output and a frequency-digital converter are cost-effective and easy to implement. The following sub-section describes the design and fabrication of disposable strips, as well as the heat/spatial configuration to support the light pathway. The later paragraphs describe in detail the two principal components of the device, and elucidate all aspects of their operation. Figure 1. Function block diagram of hand-held photometer. PC: personal computer; PDA: personal digital assistant; MAX323: precision analog switches, MAXIM; TTL: transistor-transistor logic; LED: light-emitting diode. Sensors 2019, 19, 1571 5 of 14 3.2. Manufacture of Disposable Strips To achieve the high temperature and alkaline conditions required for the DNS coloration reaction, the disposable strip was supported by a polycarbonate (PC) plate, because this has good chemical/heat resistance and visible-light transmittance. The polycarbonate plate was cut from a compact disc (thickness 1.20 mm) with a laser cutting machine (cutting rate: 1500 mm min−1; power: 40 W) into suitable schematic diagrams, shown in detail in Figure 2. The protective/reﬂective layer was then removed by tape. The upper and lower sides of transparent specimen were sealed with heat-resistant sealing ﬁlm (UC-500, Axygen Bioscience, Inc., Union, CA, USA) to fabricate the homemade disposable strip, and the upper sealing ﬁlm was punched with a syringe needle to form two pinholes at the terminal end of narrow channels to form the solution injection and gas exhaust sites. The spatial conﬁguration of narrow channels and pinholes efﬁciently reduced the measurement error by removing from the solution the small bubbles that were produced from the DNS coloration at high temperatures or mixing procedures. To remove the bubbles, small vibrations were generated to push them into narrow channels and exhaust them via pinholes. Therefore, the absorbance of mixtures was monitored in real time with the optoelectronic platform through the cut surface of the disposable strip. The red dashed line in Figure 2A depicts the optical sensing path. The sealed structural design of this disposable strip improved the measurement repeatability by lowering the sample evaporation. The disposable strip had a transparency of about 20% (CV < 10%, n = 8), and required a total reagent volume of only 200 µL to decrease the sample usage and heating time (4 min, from 20 ◦C to 80 ◦C). Figure 2. The schematic diagram (A) and outward view (B) of homemade disposable strip. The pinholes a and b are the reagent injection side and exhaust side, respectively. 3.3. Reaction/Optoelectronic Hardware Figure 3 shows a photograph of the established prototype and its sub-units. This hand-held colorimeter is assembled into a small plastic case (120 mm length, 60 mm width, and 60 mm height) to prevent ambient light from entering. The total net mass is approximately 200 g. To ensure thermal regulation and real-time detection for DNS coloration, the reaction/detection chamber(s) are fabricated from Bakelite (phenol formaldehyde resins), which has good heat resistance, optical insulation, and mechanical performance, using lathe processes. This chamber design ensures that the LED, Sensors 2019, 19, x 5 of 14 3.1. Manufacture of Disposable Strips To achieve the high temperature and alkaline conditions required for the DNS coloration reaction, the disposable strip was supported by a polycarbonate (PC) plate, because this has good chemical/heat resistance and visible-light transmittance. The polycarbonate plate was cut from a compact disc (thickness 1.20 mm) with a laser cutting machine (cutting rate: 1500 mm min−1; power: 40 W) into suitable schematic diagrams, shown in detail in Figure 2. The protective/reflective layer was then removed by tape. The upper and lower sides of transparent specimen were sealed with heat-resistant sealing film (UC-500, Axygen Bioscience, Inc., Union, CA, USA) to fabricate the homemade disposable strip, and the upper sealing film was punched with a syringe needle to form two pinholes at the terminal end of narrow channels to form the solution injection and gas exhaust sites. The spatial configuration of narrow channels and pinholes efficiently reduced the measurement error by removing from the solution the small bubbles that were produced from the DNS coloration at high temperatures or mixing procedures. To remove the bubbles, small vibrations were generated to push them into narrow channels and exhaust them via pinholes. Therefore, the absorbance of mixtures was monitored in real time with the optoelectronic platform through the cut surface of the disposable strip. The red dashed line in Figure 2A depicts the optical sensing path. The sealed structural design of this disposable strip improved the measurement repeatability by lowering the sample evaporation. The disposable strip had a transparency of about 20% (CV < 10%, n = 8), and required a total reagent volume of only 200 μl to decrease the sample usage and heating time (4 min, from 20 °C to 80 °C). Figure 2. The schematic diagram (A) and outward view (B) of homemade disposable strip. The pinholes a and b are the reagent injection side and exhaust side, respectively. 3.3. Reaction/Optoelectronic Hardware Figure 3 shows a photograph of the established prototype and its sub-units. This hand-held colorimeter is assembled into a small plastic case (120 mm length, 60 mm width, and 60 mm height) to prevent ambient light from entering. The total net mass is approximately 200 g. To ensure thermal regulation and real-time detection for DNS coloration, the reaction/detection chamber(s) are fabricated from Bakelite (phenol formaldehyde resins), which has good heat resistance, optical insulation, and mechanical performance, using lathe processes. This chamber design ensures that the LED, sample cell, polyester heater, motor-based mini-vibrator, and photo-detector are placed close Sensors 2019, 19, 1571 6 of 14 sample cell, polyester heater, motor-based mini-vibrator, and photo-detector are placed close to each other, as illustrated in Figure 3C–E, thus exempting lenses and mirrors from optical conditioning requirement. The disposable strip is ﬁlled with the DNS reagents and samples, and then guided into reaction/detection chamber through the slot design. The reaction solution is mixed uniformly with the mini-vibrator located on the back of the Bakelite plate (Figure 3E), and heated by a ﬂexible polyester heater to form the DNS coloration. This polyester-based heater with low power, higher watt density, and distributed wattage, closely located under the disposable strip, generates a uniform heat output to reduce dissipated heat. The absorbance of the mixture is measured from transmission at 525 nm with the LED and the photo-detector. Figure 3. The prototype of hand-held photometer. (A,B) Outward of hand-held photometer and its side view, respectively. (C,D) The inside view of the hand-held photometer and a zoom-in view of disposable strip with polyester heater, respectively. (E) The zoom-in view of vibrator which adhesive on the back of bakelite plate. The light source of the colorimeter is an LED with a peak wavelength of 525 nm and a spectral line half-width of 30 nm (AM2520ZGC09, Kingbright Elec. Co., Ltd., New Taipei City, Taiwan). A 20 mA pulsed-current LED driver IC (TPS60230, Texas Instruments, TX, USA) is adopted to minimize the peak wavelength shift resulting from thermal dissipation from the LED, and thereby prevent ﬂuctuation (short-term variation) and shift (long-term change) in light intensity. The transmission light is detected with a color/light sensor (TCS230, Texas Advanced Optoelectronic Solutions Inc., TX, USA). By combining silicon photodiodes and a current-to-frequency converter, the TCS230 sensor outputs a frequency-modulated signal (50% duty cycle) with a frequency value proportional to the intensity of light (irradiance). This frequency-modulated digital signal can be combined with a universal frequency-to-digital converter IC (UFDC-1M-16, International Frequency Sensor Association, Barcelona, Spain) to make a quasi-digital sensor that can interface with conventional digital circuits. The four types (R., G., B., and clear) of built-in ﬁlter in the TCS230 are responsible for the sensing selectivity of the wavelength windows. Owing to the transmission wavelength of the reagents (525 nm), the ﬁlter mode was set to green to minimize background interference. The selection of optical sensor is critical in developing the compact instrument due to considerations of hardware deployment and signal processes of both optical and electronic signals. The proposed device provides a simple and Sensors 2019, 19, x 6 of 14 to each other, as illustrated in Figure 2c,d,e, thus exempting lenses and mirrors from optical conditioning requirement. The disposable strip is filled with the DNS reagents and samples, and then guided into reaction/detection chamber through the slot design. The reaction solution is mixed uniformly with the mini-vibrator located on the back of the Bakelite plate (Figure 2e), and heated by a flexible polyester heater to form the DNS coloration. This polyester-based heater with low power, higher watt density, and distributed wattage, closely located under the disposable strip, generates a uniform heat output to reduce dissipated heat. The absorbance of the mixture is measured from transmission at 525 nm with the LED and the photo-detector. Figure 3. The prototype of hand-held photometer. (A and B) Outward of hand-held photometer and its side view, respectively. (C and D) The inside view of the hand-held photometer and a zoom-in view of disposable strip with polyester heater, respectively. (E) The zoom-in view of vibrator which adhesive on the back of bakelite plate. The light source of the colorimeter is an LED with a peak wavelength of 525 nm and a spectral line half-width of 30 nm (AM2520ZGC09, Kingbright Elec. Co., Ltd., New Taipei City, Taiwan). A 20 mA pulsed-current LED driver IC (TPS60230, Texas Instruments, TX, USA) is adopted to minimize the peak wavelength shift resulting from thermal dissipation from the LED, and thereby prevent ﬂuctuation (short-term variation) and shift (long-term change) in light intensity. The transmission light is detected with a color/light sensor (TCS230, Texas Advanced Optoelectronic Solutions Inc., TX, USA). By combining silicon photodiodes and a current-to-frequency converter, the TCS230 sensor outputs a frequency-modulated signal (50% duty cycle) with a frequency value proportional to the intensity of light (irradiance). This frequency-modulated digital signal can be combined with a universal frequency-to-digital converter IC (UFDC-1M-16, International Frequency Sensor Association, Barcelona, Spain) to make a quasi-digital sensor that can interface with conventional digital circuits. The four types (R., G., B., and clear) of built-in ﬁlter in the TCS230 are responsible for the sensing selectivity of the wavelength windows. Owing to the transmission wavelength of the reagents (525 nm), the ﬁlter mode was set to green to minimize background interference. The selection of optical sensor is critical in developing the compact instrument due to considerations of hardware deployment and signal processes of both optical and electronic signals. The proposed device provides a simple and convenient solution to the system requirements. These optical-sensing elements are tightly embedded into the vertical surface of thermal insulation Bakelite chamber to Sensors 2019, 19, 1571 7 of 14 convenient solution to the system requirements. These optical-sensing elements are tightly embedded into the vertical surface of thermal insulation Bakelite chamber to lower the thermal interference from heated samples, and the driving and measuring circuits are separated from reaction/detection chamber to reduce thermal noise in the heating process. 3.4. Electronics, Signal Conditioning and Data Processing As mentioned above, a UFDC was used as a simple converting interface between an optical sensor and a digital data processor. The communication protocol of the UFDC was set according to the Universal Asynchronous Receiver Transmitter (UART) convention by RS-232 protocol in binary-coded decimal (BCD) American standard code for information interchange (ASCII) format, and the data transfer rate (baud-rate) was set to 2400 bps. The UFDC outputs digital signals in transistor-transistor logic (TTL) evel (0/5 V) to communicate with a PC via a MAX232 IC (HIN 232CPZ, Intersil Americas Inc., Milpitas, CA, USA), which was adopted as a level converter from TTL to RS-232 level (±12 V). As well as simplifying the signal conditioning, UFDC improves the measurement performance, in terms of the detection range and the signal-to-noise ratio (SNR), beyond that of typical analog-to-digital (A/D) conversions in low-speed applications. The dynamic range of UFDC is not limited by the supply voltage of the converter integrated circuit (IC), and the signal is virtually immune to noise because it is in the frequency domain. All of the ICs, the optical components and polyester heater in this prototype were powered by two 5 V supplies. A future version of this colorimeter will thus be integrated into a single battery or USB-powered system. A program in the LabView software (LabView 8.6.1, National Instruments, Austin, TX, USA) was developed to acquire data at a sampling rate of 30 Hz via an RS-232 virtual protocol in a USB hardware port of a PC; the data can then be processed on the same operating platform. The program plots the real-time measurements on a chart. The accuracy and precision of the hand-held instrument in measuring frequency values, displayed on a PC, were evaluated and conﬁrmed with a universal counter (HP-53131A, Hewlett-Packard, Palo Alto, CA, USA). 3.5. Feasibility Veriﬁcation of Hand-held Colorimeter for Determining α-amylase Activity In biochemistry and analytical chemistry, samples are routinely cooled to improve measurement repeatability by reducing the solution evaporation or thermal interference from electronic elements. However, this cooling process requires additional cooling equipment, and is time-consuming and unsuitable for real-time monitors. To improve the measurement repeatability to monitor heated samples, this hand-held colorimeter incorporates an advanced mechanism with four main structural subunits: (1) Sealed disposable strip with pinholes to lower the sample evaporation; (2) Bakelite insulation material between heating source and electronic elements to reduce the thermal noise effects; (3) closely integrated disposable strip and heater to reduce the power dissipation; and (4) a mini-vibrator adhesive on the back of the bakelite plate to provide movement energy for small bubbles to dissipate via the pinholes in the heating process. The optical signal shift and variation coefﬁcient of the heated samples (DNS solution) are only 0.32 Hz min−1 and 0.20%, respectively, after the hand-held colorimeter has warmed up for about 20 s. On the basis of these values, the hand-held colorimeter is suitable for use as a real-time monitor for heated samples. To shorten the time taken to heat from room temperature to reaction temperature (above 80 ◦C), the disposable strips are designed with a ﬂat structure to raise the heating rate by increasing the heating surface and reducing the sample ﬁlled volume. Additionally, a low-power ﬂexible polyester heater is utilized to increase the density and uniformity of the heating output. The heating time of the 200 µL sample in the strip from 25 ◦C to 80 ◦C and 80 ◦C to 94 ◦C were 4 min and 3 min, respectively, and then the temperature was maintained at 94 ◦C with heat dissipation to prevent sample boiling. The DNS coloration for reducing sugars was fully formed above 80 ◦C within 4 min, and was monitored at 525 nm. The developed hand-held photometer yields results that correlate (R2 > 0.994) closely with those obtained using a commercialized ultraviolet–visible spectrometer, with a highly sensitive but Sensors 2019, 19, 1571 8 of 14 expensive photomultiplier tube (PMT), photo-detector, from samples diluted 10-fold with DI water, as shown in Figure 4. The experimental results reveal that the developed hand-held instrument can feasibly be applied in quantitative assay of photosensing (bio)chemistry. Figure 4. Correlation between the proposed hand-held photometer (vertical axis) and a commercial laboratory UV/Vis spectroscopy (horizontal axis) for 0.1–1.0 U mL−1 of α-amylase. Twenty microliters of α-amylase solution and 30 mL of 10 mg mL−1 starch solution were mixed and incubated for 1 min at 37 ◦C. The mixtures and 150 L DNS reagent were injected into the homemade disposable strip and heated above 80 ◦C for 4 min with a polyester heater. Figure 5 plots the frequency output of the photodetector for the absorbance response of DNS reagents to various concentrations of α-amylase. The linear detection range of the hand-held photometer for α-amylase activity is 0.1–1.0 U mL−1, and the response reaches saturation when the α-amylase activity reaches 2.0 U mL−1 (not shown here). The calibration curve is linear to α-amylase activity because it is located in the low activity range. The achieved detection limit and sensitivity values were 0.1 U mL−1 and 1703.4 Hz (log U mL−1)−1, respectively. The resolution of the proposed system with DNS reagent for α-amylase is as low as 5.9 × 10−4 log U ml−1, given a 1 Hz precision of the instrument adopted with the UFDC module. As shown in Table 1, the proposed α-amylase meter performs with an excellent sensitivity and good sample volume requirement, as well as a slightly better detection range than an iodine-based colorimetry [19]. Despite the meter’s detection range being lower than some electrochemical methods [11,16,17], its excellent sensitivity will overcome this weakness when it used in higher concentration conditions by dilution processes. The linear detection range of this prototype for α-amylase activity was below the diurnal range of sAA activity (10–300 U ml−1), but is still usable to quantitatively determine sAA levels in a sample to within the predetermined linear α-amylase activity range (0.1–1 U ml−1). Sensors 2019, 19, x 8 of 14 Figure 4. Correlation between the proposed hand-held photometer (vertical axis) and a commercial laboratory UV/Vis spectroscopy (horizontal axis) for 0.1–1.0 U mL−1 of α-amylase. Twenty microliters of α-amylase solution and 30 mL of 10 mg mL−1 starch solution were mixed and incubated for 1 min at 37 °C. The mixtures and 150 L DNS reagent were injected into the homemade disposable strip and heated above 80 °C for 4 min with a polyester heater. Figure 5 plots the frequency output of the photodetector for the absorbance response of DNS reagents to various concentrations of α-amylase. The linear detection range of the hand-held photometer for α-amylase activity is 0.1–1.0 U mL−1, and the response reaches saturation when the α-amylase activity reaches 2.0 U mL−1 (not shown here). The calibration curve is linear to α-amylase activity because it is located in the low activity range. The achieved detection limit and sensitivity values were 0.1 U mL−1 and 1703.4 Hz (log U mL−1)−1, respectively. The resolution of the proposed system with DNS reagent for α-amylase is as low as 5.9 × 10−4 log U ml−1, given a 1 Hz precision of the instrument adopted with the UFDC module. As shown in Table 1, the proposed α-amylase meter performs with an excellent sensitivity and good sample volume requirement, as well as a slightly better detection range than an iodine-based colorimetry [19]. Despite the meter’s detection range being lower than some electrochemical methods [11,16,17], its excellent sensitivity will overcome this weakness when it used in higher concentration conditions by dilution processes. The linear detection range of this prototype for α-amylase activity was below the diurnal range of sAA activity (10–300 U ml−1), but is still usable to quantitatively determine sAA levels in a sample to within the predetermined linear α-amylase activity range (0.1–1 U ml−1). Sensors 2019, 19, 1571 9 of 14 Figure 5. Calibration curve of 0.1–1.0 U mL−1 α-amylase hydrolyzed starch solution with 3,5 dinitrosalicylic acid (DNS) reagent. Other detailed operational conditions are shown in Figure 4. Table 1. Comparison between the analytical performances obtained in the current study and other methods reported in literature. QCM-D: quartz crystal microbalance with dissipation monitoring; BIA-AD: batch injection analysis system with amperometric detection; C: α-amylase activity (U mL−1). Reference [method] Yamaguchi et al., 2003 [amperometry] Yamaguchi et al., 2004 & 2006* [colorimetry] Wu et al., 2007 [magnetoelastic] Sasaki et al., 2008 [QCM-D] Mahosenaho et al., 2010 [amperometry] Zhang et al., 2014 [chronocoulometry] Zhang et al., 2015 [potentiometry] Wang et al., 2015 [amperometry] Dutta et al., 2016 [colorimetry, paper-based] Garcia et al., 2018 [amperometry, BIA-AD] Garcia et al., 2018 [amperometry] Ma et al., 2019 [liquid crystal] This work [colorimetry] Detection Limit (U mL−1) Detection Range (U mL−1) Sensitivity (Response / U mL−1) Reaction Time (min.) Sample Volume (mL) - - - - - 5 0.022 0.12 0.02 - 0.05 1.1 0.015 0.1 0~30 0~200 10~140* 75~125 - - - - 0.001~1 0.88 log C 5~250 0.03~3 30~1000 0.997 C 5.20 C 0.12 C 0~1 0.01276 C 0.01~0.11 0.003 C 0.5~10 48.8 C 100~1200 10.7 log C - - 0.1~1.0 1703.4 log C - 2.5 - - - - - 5 15 - 0.5 20 - 12 - 5 30* - - - - 25 - - 15 15 - 5 Sensors 2019, 19, x 9 of 14 Figure 5. Calibration curve of 0.1–1.0 U mL−1 α-amylase hydrolyzed starch solution with 3,5 dinitrosalicylic acid (DNS) reagent. Other detailed operational conditions are shown in Figure 4. Table 1. Comparison between the analytical performances obtained in the current study and other methods reported in literature. QCM-D: quartz crystal microbalance with dissipation monitoring; BIA-AD: batch injection analysis system with amperometric detection; C: α-amylase activity (U mL−1). Reference [method] Detection Limit (U mL−1) Detection Range (U mL−1) Sensitivity (Response / U mL−1) Reaction Time (min.) Sample Volume (mL) Yamaguchi et al., 2003 [amperometry] - 0~30 - - - Yamaguchi et al., 2004 & 2006* [colorimetry] - - 0~200 10~140* - - 2.5 - 5 30* Wu et al., 2007 [magnetoelastic] - 75~125 - - - Sasaki et al., 2008 [QCM-D] - 0.001~1 0.88 log C - - Mahosenaho et al., 2010 [amperometry] 5 5~250 0.997 C - - Zhang et al., 2014 [chronocoulometry] 0.022 0.03~3 5.20 C - - Zhang et al., 2015 [potentiometry] 0.12 30~1000 0.12 C 5 25 Wang et al., 2015 [amperometry] 0.02 0~1 0.01276 C 15 - Dutta et al., 2016 [colorimetry, paper-based] - 0.01~0.11 0.003 C - - Garcia et al., 2018 [amperometry, BIA-AD] 0.05 0.5~10 48.8 C 0.5 15 Garcia et al., 2018 [amperometry] 1.1 100~1200 10.7 log C 20 15 Ma et al., 2019 [liquid crystal] 0.015 - - - - This work [colorimetry] 0.1 0.1~1.0 1703.4 log C 12 5 Sensors 2019, 19, 1571 10 of 14 3.6. Application to Stress Evaluation Stress can be measured using several methods, including psychological tests, measurements of hormonal and cardiovascular responses, and other physiological parameters. Compared with measurements of physiological parameters such as heart rate and blood pressure, the monitors of stress-related indicators such as cortisol and norepinephrine are easy to quantify, and show signiﬁcant differences in response to eustress and distress [36]. This investigation selected serum cortisol (reference method) and sAA as stress-related biomarkers of HPA and SAM activity in response to daily lifestyle and stressful exercise. The experimental results demonstrate that the proposed sAA activity sensor platform performs better than serum cortisol for distinguishing stress levels. Table 2 lists the blood sample collection times, exercise training conditions, and serum cortisol levels. The serum cortisol level was higher in the morning (Exp II) than in the afternoon (Exp IV and VI) for the normal daily routine, because cortisol is a steroid hormone that is secreted diurnally [37] in response to pulsatile trophic hormone stimulation; cortisol levels peak early prior to awakening, and decreasing progressively during the day to reach low levels in the evening. This appearance is also conﬁrmed by the comparison between the morning conditions (Exp I and II) and afternoon conditions (Exp III ~ VI), despite the different intensity of exercises shown in Table 2. The serum cortisol levels sampled 1 h after exercise (Exp III, V) were signiﬁcantly higher than those in the normal daily routine (Exp IV, VI) at similar collection times (13:40), but the serum cortisol level collected in the morning with over 8 hr of rest after exercise (Exp I) was similar to that in the normal daily routine (Exp II). This ﬁnding indicates that the serum cortisol after rest was metabolized and self-regulated via the negative feedback of HPA axis whereby elevated circulating cortisol levels lead to diurnal concentration [37], and the serum cortisol level does not fully reﬂect the fatigue feeling of prolonged exercise. Therefore, the serum cortisol level is not a good reference indicator to consider the psychological or physiological stress of subjects. Consequently, this study seeks to demonstrate that the proposed sAA meter could be a valuable tool to assess fatigue levels. Table 2. The collection conditions and serum cortisol levels of six blood samples. Number Collection Time Collection Status Serum Cortisol (µg/dL) I II III IV V VI 09:25 09:50 13:40 13:30 13:40 12:45 More than ﬁve hours of continuous sport last night Normal daily routine 1 hour of low intensity race at noon Normal daily routine 1 hour of high intensity race at noon Normal daily routine 19.0 18.1 13.3 5.0 11.4 5.3 Unlike serum cortisol, the non-hormonal sAA activities sampled from the normal daily routine conditions (Exp. II, IV, and VI in Figure 6) were similar (<8 U mL−1) at different collection times (morning or afternoon) [5,38], and its activity could be a good reference indicator for assessing psychological or physiological stress of subjects. To evaluate the psychological stress of subjects during the blood collection procedure, the saliva was gathered 30 min before blood sampling, immediately after blood sampling, and at intervals of 30 min thereafter. The sAA results from these samples were ﬁlled in at the same intervals (Figure 6), indicating that the blood collection procedure did not signiﬁcantly inﬂuence the psychological stress of subjects. The sAA activity of salivary samples collected in one hour of low- and high-intensity tests (Exp. III and V, respectively) after stressful prolonged exercise were three times those under normal daily routine (Exp II, IV, and VI), even if the subject took an overnight rest after a super-loading exercise (Exp I). The sAA activity determined by the proposed system can effectively and quantitatively distinguish physiological stress levels, and avoid the diurnal effect found in serum cortisol. Sensors 2019, 19, 1571 11 of 14 Figure 6. Serum cortisol concentration (horizontal axis) and salivary α-amylase (vertical axis) response to physiological stressful exercise conditions (Exp I, III, and V) and daily routine conditions (Exp II, IV, and VI). The solid circles (Exp I, II) and open circles (Exp III–VI) are blood samples collected in the morning and afternoon, respectively. As indicated in Exp. I and II of Figure 6, two subjects who underwent signiﬁcantly different loading exercises had signiﬁcantly sAA levels, three-fold different as measured by the proposed meter, but had almost equal serum cortisol levels (value difference < 5%). The recovery time of sAA response for physiological exercise was much longer than the tens of minutes found for psychological stressful conditions, i.e., the Trier Social Stress Test [5,39]. This phenomenon indicates that the sAA secretion is released from the salivary glands by controlling the stimulation of plasma norepinephrine, which is a catecholamine with multiple roles, including neurotransmitter and hormone. In response to physiological and psychological stressors, this neurotransmitter is immediately released from the sympathetic neurons affecting the salivary glands to secrete sAA (about a few minutes) and recovers to normal concentration (about tens of minutes) by signal termination of norepinephrine reuptake. However, the performance of acute norepinephrine reuptake inhibition may decrease in a high ambient temperature whilst at rest and during exercise [40]. The exercise-induced increase of norepinephrine is delayed, or the onset of central fatigue is accelerated, to control body temperature during prolonged exercise (1–3 h) [41,42]. Therefore, the sAA is not only a sensitive biomarker for evaluating physiological and psychological stress, but can also be utilized to evaluate central fatigue after prolonged exercise [43]. This result implies that the proposed sAA meter might be directly useful for evaluating the physical resilience (capacity of physiological recovery) or accumulation effect of fatigue. 4. Conclusions In summary, this study develops a compact analytic system for α-amylase activity based on colorimetry, and shows its feasibility in stress assessment with salvia. The linear detection range of α-amylase activity is 0.1–1 U mL−1; it also shows good correlation with commercial ultraviolet–visible spectroscopy. The developed prototype has satisfactory instrumentation performance, and is therefore very attractive as a device that exploits photometry to determine quantities of α-amylase activity in the ﬁeld. The hand-held device has a small sample consumption (about 5 µL), low cost, and convenient measurement, and is applied to physiological exercise stress assessment by measuring the α-amylase Sensors 2019, 19, x 11 of 14 Figure 6. Serum cortisol concentration (horizontal axis) and salivary α-amylase (vertical axis) response to physiological stressful exercise conditions (Exp I, III, and V) and daily routine conditions (Exp II, IV, and VI). The solid circles (Exp I, II) and open circles (Exp III–VI) are blood samples collected in the morning and afternoon, respectively. As indicated in Exp. I and II of Figure 6, two subjects who underwent significantly different loading exercises had significantly sAA levels, three-fold different as measured by the proposed meter, but had almost equal serum cortisol levels (value difference < 5%). The recovery time of sAA response for physiological exercise was much longer than the tens of minutes found for psychological stressful conditions, i.e., the Trier Social Stress Test [5,39]. This phenomenon indicates that the sAA secretion is released from the salivary glands by controlling the stimulation of plasma norepinephrine, which is a catecholamine with multiple roles, including neurotransmitter and hormone. In response to physiological and psychological stressors, this neurotransmitter is immediately released from the sympathetic neurons affecting the salivary glands to secrete sAA (about a few minutes) and recovers to normal concentration (about tens of minutes) by signal termination of norepinephrine reuptake. However, the performance of acute norepinephrine reuptake inhibition may decrease in a high ambient temperature whilst at rest and during exercise [40]. The exercise-induced increase of norepinephrine is delayed, or the onset of central fatigue is accelerated, to control body temperature during prolonged exercise (1–3 h) [41,42]. Therefore, the sAA is not only a sensitive biomarker for evaluating physiological and psychological stress, but can also be utilized to evaluate central fatigue after prolonged exercise [43]. This result implies that the proposed sAA meter might be directly useful for evaluating the physical resilience (capacity of physiological recovery) or accumulation effect of fatigue. 4. Conclusions In summary, this study develops a compact analytic system for α-amylase activity based on colorimetry, and shows its feasibility in stress assessment with salvia. The linear detection range of α-amylase activity is 0.1–1 U mL−1; it also shows good correlation with commercial ultraviolet–visible spectroscopy. The developed prototype has satisfactory instrumentation performance, and is therefore very attractive as a device that exploits photometry to determine quantities of α-amylase activity in the ﬁeld. The hand-held device has a small sample consumption (about 5 μL), low cost, Sensors 2019, 19, 1571 12 of 14 level in the whole saliva. The sAA activity determined by the proposed system can effectively and quantitatively distinguish physiological stress levels, but does not exhibit the diurnal effect found in serum cortisol. The sAA activity of saliva in stressful high-intensity exercise was thrice that under the normal daily routine, even if the subject took an overnight rest. These results demonstrate that the proposed sAA meter considered is a sensitive biomarker meter not only for short-term stressful stimuli, but also for long-term fatigue following prolonged exercise. This convenient hand-held device will be a valuable tool in the evaluation of stress and central fatigue in the future. Author Contributions: Conceptualization, R.L.C.C. and T.-J.C.; methodology, H.-Y.H. and R.L.C.C.; investigation, C.-C.C. and H.-Y.H.; resources, R.L.C.C. and T.-J.C.; data curation, C.-C.C. and H.-Y.H.; writing—original draft preparation, H.-Y.H.; writing—review and editing, R.L.C.C. and T.-J.C.; supervision, T.-J.C.; project administration, H.-Y.H. and R.L.C.C.; funding acquisition, R.L.C.C. and T.-J.C. Funding: This research received no external funding. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Nater, U.M.; Rohleder, N. Salivary alpha-amylase as a non-invasive biomarker for the sympathetic nervous 2. 3. 4. 5. 6. 7. system: Current state of research. Psychoneuroendocrinology 2009, 34, 486–496. [CrossRef] Rohleder, N.; Nater, U.M. Determinants of salivary α-amylase in humans and methodological considerations. Psychoneuroendocrinology 2009, 34, 469–485. [CrossRef] [PubMed] Gatti, R.; De Palo, E.F. An update: Salivary hormones and physical exercise. Scand. J. Med. Sci. Sports 2011, 21, 157–169. [CrossRef] [PubMed] Granger, D.A.; Kivlighan, K.T.; El Sheikh, M.E.; Gordis, B.; Stroud, L.R. Salivary alpha-amylase in biobehavioral research: Recent developments and applications. Ann. N. Y. Acad. Sci. 2007, 1098, 122–144. [CrossRef] [PubMed] Rohleder, N.; Nater, U.M.; Wolf, J.M.; Ehlert, U.; Kirschbaum, C. Psychosocial stress-induced activation of salivary alpha-amylase: An indicator of sympathetic activity? Ann. N. Y. Acad. Sci. 2004, 1032, 258–263. [CrossRef] [PubMed] De Souza, P.M.; de Oliveria Magalhães, P. Application of microbial α-amylase in industry—A review. Braz. J. Microbiol. 2010, 41, 850–861. [CrossRef] González, C.F.; Fariña, J.I.; Figueroa, L.I.C. A critical assessment of a viscometric assay for measuring Saccharomycopsis ﬁbuligera α-amylase activity on gelatinised cassava starch. Enzyme Microb. Technol. 2002, 30, 169–175. [CrossRef] 8. Wu, S.; Zhu, Y.; Cai, Q.; Zeng, K.; Grimes, C.A. A wireless magnetoelastic α-amylase sensor. Sens. Actuators B 9. 2007, 121, 476–481. [CrossRef] Sasaki, T.; Noe, T.R.; Ring, S.G. Study on α-amylase hydrolysis of potato amylopectin by a quartz crystal microbalance. J. Agric. Food Chem. 2008, 56, 1091–1096. [CrossRef] 10. Bouchet-Spinelli, A.; Coche-Guérente, L.; Armand, S.; Lenouvel, F.; Labbé, P.; Fort, S. Functional characterization of starch-degrading enzymes using quartz crystal microbalance with dissipation monitoring (QCM-D). Sens. Actuators B 2013, 176, 1038–1043. [CrossRef] 11. Garcia, P.T.; Guimarães, L.N.; Dias, A.A.; Ulhoa, C.J.; Coltro, W.K.T. Amperometric detection of salivary α-amylase on screen-printed carbon electrodes as a simple and inexpensive alternative for point-of-care testing. Sens. Actuators B 2018, 258, 342–348. [CrossRef] 12. Zhang, J.; Cui, J.; Liu, Y.; Chen, Y.; Li, G. A novel electrochemical method to determine a-amylase activity. Analyst 2014, 139, 3429–3433. [CrossRef] 13. Wang, Q.; Wang, H.; Yang, X.; Wang, K.; Liu, R.; Li, Q.; Ou, J. A sensitive one-step method for quantitative detection of a-amylase in serum and urine using a personal glucose meter. Analyst 2015, 140, 1161–1165. [CrossRef] 14. Garcia, P.T.; Dias, A.A.; Souza, J.A.C.; Coltro, W.K.T. Batch injection analysis towards auxiliary diagnosis of periodontal diseases based on indirect amperometric detection of salivary α-amylase on a cupric oxide electrode. Anal. Chim. Acta 2018, 1041, 50–57. [CrossRef] Sensors 2019, 19, 1571 13 of 14 15. Yamaguchi, M.; Kanemaru, M.; Kanemori, T.; Mizuno, Y. Flow-injection-type biosensor system for salivary amylase activity. Biosens. Bioelectron. 2003, 18, 835–840. [CrossRef] 16. Mahosenaho, M.; Caprio, F.; Micheli, L.; Sesay, A.M.; Palleschi, G.; Virtanen, V. A disposable biosensor for the determination of alpha-amylase in human saliva. Microchim. Acta 2010, 170, 243–249. [CrossRef] 17. Zhang, L.; Yang, W.; Yang, Y.; Liu, H.; Gu, Z. Smartphone-based point-of-care testing of salivary alpha-amylase for personal psychological measurement. Analyst 2015, 140, 7399–7406. [CrossRef] 18. Xiao, Z.; Storms, R.; Tsang, A. A quantitative starch-iodine method for measuring alpha-amylase and glucoamylase activities. Anal. Biochem. 2006, 351, 146–148. [CrossRef] 19. Dutta, S.; Mandal, N.; Bandyopadhyay, D. Paper-based a-amylase detector for point-of-care diagnostics. Biosens. Bioelectron. 2016, 78, 447–453. [CrossRef] [PubMed] 20. Yamaguchi, M.; Kanemori, T.; Kanemaru, M.; Takai, N.; Mizuno, Y.; Yoshida, H. Performance evaluation of salivary amylase activity monitor. Biosens. Bioelectron. 2004, 20, 491–497. [CrossRef] 21. Ma, H.; Kang, Q.; Wang, T.; Yu, L. A liquid crystals-based sensing platform for detection of α-amylase coupled with destruction of host-guest interaction. Colloids Surfaces B Biointerfaces 2019, 173, 616–622. [CrossRef] 22. Bassinello, P.Z.; Cordenunsi, B.R.; Lajolo, F.M. Amylolytic activity in fruits: Comparison of different substrates and methods using banana as model. J. Agric. Food Chem. 2002, 50, 5781–5786. [CrossRef] 23. McCleary, B.V.; Sheehan, H. Measurement of cereal α-amylase: A new assay procedure. J. Cereal Sci. 1987, 6, 237–251. [CrossRef] 24. Dhawale, M.R.; Wilson, J.J.; Khachatourians, G.G.; Ingledew, W.M. Improved method for detection of starch hydrolysis. Appl. Environ. Microbiol. 1982, 44, 747–750. 25. Yamaguchi, M.; Deguchi, M.; Wakasugi, J.; Ono, S.; Takai, N.; Higashi, T.; Mizuno, Y. Hand-held monitor of sympathetic nervous system using salivary amylase activity and its validation by driver fatigue assessment. Biosens. Bioelectron. 2006, 21, 1007–1014. [CrossRef] 26. Najaﬁ, M.F.; Kembhavi, A. One step puriﬁcation and characterization of an extracellular α-amylase from marine Vibrio sp. Enzyme Microb. Technol. 2005, 36, 535–539. [CrossRef] 27. Van Staden, J.F.; Mulaudzi, L.V. Flow injection spectrophotometric assay of α-amylase activity. Anal. Chim. Acta 2000, 421, 19–25. [CrossRef] 28. Capitán-Vallvey, L.F.; Palma, A.J. Recent developments in hand-held and portable optosensing—A review. 29. Anal. Chim. Acta 2011, 696, 27–46. [CrossRef] Sorouraddin, M.-H.; Saadati, M. A simple homemade light emitting diode based photometer for chromium speciation. Sens. Actuators B 2013, 188, 73–77. [CrossRef] 30. Bueno, D.; Alonso, G.; Muñoz, R.; Marty, J.L. Low-cost and portable absorbance measuring system to carbamate and organophosphate pesticides. Sens. Actuators B 2014, 203, 81–88. [CrossRef] 31. Miller, G.L. Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal. Chem. 1959, 31, 32. 426–428. [CrossRef] Sumriddetchkajorn, S.; Chaitavon, K.; Intaravanne, Y. Mobile-platform based colorimeter for monitoring chlorine concentration in water. Sens. Actuators B 2014, 191, 561–566. [CrossRef] 33. Kanchi, S.; Sabela, M.I.; Mdluli, P.S.; Bisetty, K. Smartphone based bioanalytical and diagnosis applications: A review. Biosens. Bioelectron. 2018, 102, 136–149. [CrossRef] [PubMed] 34. Navazesh, M. Methods for collecting saliva. Ann. N. Y. Acad. Sci. 1993, 694, 72–77. [CrossRef] [PubMed] 35. Chang, K.H.; Chen, R.L.C.; Hsieh, B.C.; Chen, P.C.; Hsiao, H.Y.; Nieh, C.H.; Cheng, T.J. A hand-held electronic tongue based on ﬂuorometry for taste assessment of tea. Biosens. Bioelectron. 2010, 26, 1507–1513. [CrossRef] [PubMed] 36. Ali, N.; Pruessner, J.C. The salivary alpha amylase over cortisol ratio as a marker to assess dysregulations of the stress systems. Physiol. Behav. 2012, 106, 65–72. [CrossRef] [PubMed] 37. Crown, A.; Lightman, S. Why is the management of glucocorticoid deﬁciency still controversial: A review of the literature. Clin. Endocrinol. 2005, 3, 483–492. [CrossRef] [PubMed] 38. Nater, U.M.; Rohleder, N.; Schlotz, W.; Ehlert, U.; Kirschbaum, C. Determinants of the diurnal course of salivary alpha-amylase. Psychoneuroendocrinology 2007, 32, 392–401. [CrossRef] 39. Nater, U.M.; Rohleder, N.; Gaab, J.; Berger, S.; Jud, A.; Kirschbaum, C.; Ehlert, U. Human salivary alpha-amylase reactivity in a psychosocial stress paradigm. Int. J. Psychophysiol. 2005, 55, 333–342. [CrossRef] Sensors 2019, 19, 1571 14 of 14 40. Roelands, B.; Goekint, M.; Heyman, E.; Piacentini, M.F.; Watson, P.; Hasegawa, H.; Buyse, L.; Pauwels, F.; De Schutter, G.; Meeusen, R. Acute norepinephrine reuptake inhibition decreases performance in normal and high ambient temperature. J. Appl. Physiol. 2008, 105, 206–212. [CrossRef] 41. Meeusen, R.; Watson, P.; Hasegawa, H.; Roelands, B.; Piacentini, M.F. Central fatigue: The serotonin hypothesis and beyond. Sports Med. 2006, 36, 881–909. [CrossRef] [PubMed] 42. Meeusen, R.; Roelands, B. Central fatigue and neurotransmitters, can thermoregulation be manipulated? Scand. J. Med. Sci. Sports 2010, 20, 19–28. [CrossRef] [PubMed] 43. Li, T.L.; Gleeson, M. The effect of single and repeated bouts of prolonged cycling and circadian variation on saliva ﬂow rate, immunoglobulin A and α-amylase responses. J. Sports Sci. 2004, 22, 1015–1024. [CrossRef] [PubMed] © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_nu11061240	Article Multidisciplinary Integrated Metabolic Rehabilitation in Elderly Obese Patients: Eﬀects on Cardiovascular Risk Factors, Fatigue and Muscle Performance Antonello E. Rigamonti 1,* Diana Caroli 2, Roberta De Micheli 2, Gabriella Tringali 2, Laura Abbruzzese 2,3, Nicoletta Marazzi 2, Silvano G. Cella 1 and Alessandro Sartorio 2,3 1 Department of Clinical Sciences and Community Health, University of Milan, via Vanvitelli 32, 20129 Milan, , Alessandra De Col 2, Soﬁa Tamini 2, Sabrina Cicolini 2, 2 3 Italy; [email protected] Istituto Auxologico Italiano, IRCCS, Experimental Laboratory for Auxo-Endocrinological Research, strada L. Cadorna 90, 28824 Piancavallo (VB), Italy; [email protected] (A.D.C.); [email protected] (S.T.); [email protected] (S.C.); [email protected] (D.C.); [email protected] (R.D.M.); [email protected] (G.T.); [email protected] (L.A.); [email protected] (N.M.); [email protected] (A.S.) Istituto Auxologico Italiano, IRCCS, Division of Metabolic Diseases, strada L. Cadorna 90, 28824 Piancavallo (VB), Italy * Correspondence: [email protected]; Tel.: +39-02-503-17013; Fax: +39-02-503-17011 Received: 13 May 2019; Accepted: 28 May 2019; Published: 31 May 2019 Abstract: Background: Obesity is a widespread problem in the elderly, being associated with severe comorbidities negatively inﬂuencing life expectancy. Integrated multidisciplinary metabolic rehabilitation aimed to reduce body weight (BW) and fatigue, increase physical autonomy and introduce healthy life style changes has been proposed as a useful intervention to improve the general health status and quality of life of the obese geriatric population. Methods: Six hundred-eighty four severely obese subjects (F/M = 592/92; age range: 61–83 years; mean body mass index, BMI ± SD: 42.6 ± 5.6 kg/m2) were admitted to take part in a three-week in-hospital BW reduction program (BWRP), entailing energy restricted diet, psychological counselling, physical rehabilitation and nutritional education. Biochemical parameters, cardiovascular risk factors (throughout the Coronary Heart Disease Risk, CHD-R), fatigue (throughout the Fatigue Severity Scale, FSS) and lower limb muscle performance (throughout the Stair Climbing Test, SCT) were evaluated before and at the end of the BWRP. Results: A 4% BW reduction was achieved at the end of the BWRP. This ﬁnding was associated with a signiﬁcant improvement of the metabolic homeostasis (i.e., decrease in total cholesterol and glucose) and a reduction of systolic blood pressure in both females and males, thus resulting in a reduction of CHD-R in the male group. Total FSS score and SCT time decreased in female and male obese patients. The eﬀects of BWPR were comparable among all age-related subgroups (>60, 60–69 and >70 years), apart from ∆CHD-R, which was higher in male subgroups. Finally, age was negatively correlated with ∆BMI and ∆FSS. Conclusions: Though only a relatively limited number of outcomes were investigated, the present study shows that a 4% BW reduction in severely elderly obese patients is associated with positive multisystemic eﬀects, particularly, muscle-skeletal and cardiometabolic beneﬁts, which can favorably inﬂuence their general well-being and improve the autonomy level in performing more common daily activities. The maintenance of a healthy life style, including controlled food intake and regular physical activity, after a BWRP is obviously recommended in all elderly obese patients to further improve their clinical condition. Keywords: multidisciplinary integrated metabolic rehabilitation; diet; exercise; geriatric obesity Nutrients 2019, 11, 1240; doi:10.3390/nu11061240 www.mdpi.com/journal/nutrients nutrients(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Nutrients 2019, 11, 1240 1. Introduction 2 of 15 In the last decade, we have assisted a dramatic worldwide increase in geriatric population and, concomitantly, in obesity [1]. Thus, the elderly obese patient is becoming an increasingly prevalent phenotype in the general population from developed and also developing countries, with relevant socioeconomic implications for the public health system and political decision-making [2]. Geriatric obesity is associated with a worsening of the coronary heart disease risk (CHD-R), due to dyslipidemia, hypertension, type 2 diabetes mellitus and physical inactivity [3]. Furthermore, functional autonomy and, in general, the quality of life are consistently reduced in an elderly obese subject, so that obesity can be considered a determinant of “frailty” in geriatric practice and have a prominent causative role for several clinical conditions that require hospitalization or institutionalization [4]. Finally, elderly women seem to be at high risk to develop or maintain a pre-existing obesity due the well-known post-menopausal changes [5]. Body weight (BW) reduction programs (BWRPs) have been demonstrated to be a valid strategy to contrast the negative eﬀects associated with geriatric obesity, particularly when a multidisciplinary integrated approach is adopted. In this context, diet combined with physical activity and psychological counseling has been shown to be more eﬀective when administered in cohorts of elderly obese patients in comparison with the single interventions [6]. Nevertheless, some controversies regarding the short- and long-term beneﬁts and safety of BWRPs against geriatric obesity still persist [4]. For instance, elderly subjects have been reported to have lower resting energy expenditure and caloric requirements, with the consequence that a BWRP may induce an unsatisfactory negative energy balance and a negligible weight loss [7–10]. In addition, since sarcopenic obesity is more prevalent in elderly than young individuals, a protein-restricted dietetic regimen administered to an elderly obese patient could increase the catabolism of muscle proteins, a process ampliﬁed by the obesity-associated low-grade chronic inﬂammation and aging-related hypomotility [11,12]. Weight loss obtained in an elderly obese patient undergoing caloric restriction may derive from a relevant shrinkage of lean mass (more than 25%), with possible reduction in muscle strength, limitations in motor function and impairment of glucose tolerance, with muscle tissue being fundamental for glucose uptake and glucometabolic homeostasis [13–17]. Independently from age, weight loss reduces bone mass density, an eﬀect that, in osteopenic or osteoporotic post-menopausal women, could be deleterious due the risk of fractures [18]. Finally, some authors have proposed the “obesity paradox” to indicate the protective value of a high BMI in later life [19]. In particular, the progression of some clinical conditions, frequently diagnosed in geriatric patients, such as cancer, chronic heart failure and end-stage renal disease, is slowed down to wasting syndrome or cachexia in the obese subgroup compared to the lean counterpart [20]. Based on these conﬂicting results, well highlightened by recent reviews [4,6], there is the urgent need to evaluate adjunctive outcomes in order to deﬁne and quantify the eﬀects of an (integrated) BWRP on cardiometabolic status, muscle performance and quality of life in elderly obese patients. Understanding of the beneﬁcial eﬀects of any BWRP on speciﬁc outcomes in diﬀerent subgroups of elderly obese patients (e.g., for gender and age ranges) might allow us to deﬁne the demographic and clinical characteristics of the elderly obese patient who will favorably respond to a standardized BWRP and, additionally, to tailor each component of the BWRP (dietetic regimen, type/duration/intensity of physical activity, adherence to healthy life styles, concomitant pharmacological therapy etc.) to a particular elderly obese patient having speciﬁc demographic and clinical characteristics. Therefore, the present study was aimed at investigating the eﬀects of a three-week BWRP, administered to a large cohort of elderly obese females and males, on (1) CHD-R, a validated scoring system, including demographic, clinical, biochemical and cardiovascular parameters, which permits the calculation of CHD-R over the next 10 years and compare this value to that of others of the same age [21]; (2) stair climbing test (SCT), used to evaluate functional strength, balance and agility of lower Nutrients 2019, 11, 1240 3 of 15 limbs through ascending a set number of steps [22,23]; and (3) fatigue severity scale (FSS), a largely employed self-report questionnaires to evaluate fatigue in daily activities, which does not depend upon an underlying depressive condition [24]. 2. Material and Methods 2.1. Patients and Body Weight Reduction Program Six-hundred-eighty-four severely obese subjects (females, F/males, M = 592/92; age range: 61–83 years; body mass index, BMI: 42.6 ± 5.6 kg/m2) were recruited at the Division of Metabolic Diseases, Istituto Auxologico Italiano, Piancavallo (VB), where they were hospitalized for a three-week multidisciplinary integrated BWRP, including hypocaloric diet, nutritional education, psychological counselling and moderate physical activity (see below for details). The sample size was considered adequate, taking into account a power analysis in which a mean value of ∆BMI (%) after BWRP was supposed to be equal to 4.0 ± 4.0% with an α error of 0.05 at two tails and a power of 0.80. The consort ﬂow diagram is shown in Figure 1. The unique criterion of inclusion was a BMI > 35 kg/m2, while the main exclusion criteria were physical inability in performing SCT and cognitive impairments hampering the FSS execution. Figure 1. Consort ﬂow diagram of the study. BWRP, body weight reduction program. When considering the BWRP in detail, energy intake was restricted by imposing a diet (5023–7113 kJ/day, i.e., 1200–1700 kcal/day) containing about 21% proteins, 53% carbohydrates and 26% lipids. The calories to be given with the diet were calculated by subtracting approximately 25% from the value of resting energy expenditure as measured in each patient by indirect calorimetry (Vmax 29; SensorMedics Corporation, Yorba Linda, CA, USA) for a total duration of 20 min. Under the energy restriction, each patient was free to choose foods from a heterogeneous daily menu. Foods to which the patient declared to be allergic were removed from the menu. Five daily portions of fruits and vegetables were obligatory. A ﬂuid intake of at least 1500 mL/day was encouraged. Nutritional education consisted of lectures, demonstrations and group discussions with and without a supervisor, took place every day throughout the whole rehabilitation period. Sessions of psychological counselling were conducted by clinical psychologists 2–3 times/week and were based on individual or cognitive Nutrients 2019, 11, x FOR PEER REVIEW 3 of 17 Nutrients 2019, 11, x; doi: FOR PEER REVIEW www.mdpi.com/journal/nutrients Therefore, the present study was aimed at investigating the effects of a three‐week BWRP, administered to a large cohort of elderly obese females and males, on (1) CHD‐R, a validated scoring system, including demographic, clinical, biochemical and cardiovascular parameters, which permits the calculation of CHD‐R over the next 10 years and compare this value to that of others of the same age [21]; (2) stair climbing test (SCT), used to evaluate functional strength, balance and agility of lower limbs through ascending a set number of steps [22,23]; and (3) fatigue severity scale (FSS), a largely employed self‐report questionnaires to evaluate fatigue in daily activities, which does not depend upon an underlying depressive condition [24]. 2. Material and Methods 2.1. Patients and Body Weight Reduction Program Six‐hundred‐eighty‐four severely obese subjects (females, F/males, M = 592/92; age range: 61–83 years; body mass index, BMI: 42.6 ± 5.6 kg/m2) were recruited at the Division of Metabolic Diseases, Istituto Auxologico Italiano, Piancavallo (VB), where they were hospitalized for a three‐week multidisciplinary integrated BWRP, including hypocaloric diet, nutritional education, psychological counselling and moderate physical activity (see below for details). The sample size was considered adequate, taking into account a power analysis in which a mean value of ΔBMI (%) after BWRP was supposed to be equal to 4.0 ± 4.0% with an α error of 0.05 at two tails and a power of 0.80. The consort flow diagram is shown in Figure 1. The unique criterion of inclusion was a BMI > 35 kg/m2, while the main exclusion criteria were physical inability in performing SCT and cognitive impairments hampering the FSS execution. Figure 1. Consort flow diagram of the study. BWRP, body weight reduction program. Assessedfor eligibility(no. 805, 686F/119M) ExcludedDeclinedto participate: 35 (28F/7M)Medicalreasons: 15 (10F/5M)Enrollment(BWRP)755 (648F/107M)In‐hospital follow‐up(3 weeks)Lost to follow‐up: 71 (56F/15M)Analysis (684, 592F/92MNutrients 2019, 11, 1240 4 of 15 behavioral strategies. Physical activity consisted of ﬁve training sessions/week (about 1 h each session), including indoor light jogging, dynamic exercises of the upper and lower limbs (standing and ﬂoor gymnastics routines, focalized on muscle strength-power development) at moderate intensity under the guide of a therapist; furthermore, subjects underwent either 15–20 min aerobic exercise or 2 km outdoor walking on a predetermined track, according to individual capabilities and clinical status. Before and after BWRP, each subject underwent the following tests/evaluations, which are described in detail below: –SCT (stair climbing test); –FSS (fatigue severity score); –CHD-R (coronary heart disease risk); –Standard hematochemistry. The protocol was approved by the local Ethical Committee (research project code: 18A301, acronym: FUOBAUXO); all subjects gave their written consent after being fully informed of the nature and procedures of the study. 2.2. Stair Climbing Test SCT is a well-standardized procedure, which was readapted and validated by our group to measure the maximal anaerobic power [22,23,25]. Subjects were invited to climb up ordinary stairs (13 steps of 15.3 cm each, vertical distance: 1.99 mt) at the highest possible speed, according to their capabilities. The stairs were climbed one time by the participant. The time taken to perform the test was measured by an experimental investigator with a digital stopwatch. In line with the assumptions by Margaria et al. [26], anaerobic power (in W) was calculated by using the following formula: (body weight × 9:81 × vertical distance)/time where body weight, vertical distance and time are expressed in kg, m and sec, respectively, and 9.81 m/sec2 represents the acceleration of gravity. See also Reference [27] for further details. 2.3. Fatigue Severity Scale FSS is one of the most commonly used self-report questionnaires for the fatigue assessment in chronic diseases [28,29], already used and validated in Italian obese patients by our group [24]. FSS consists of nine statements (items) describing the negative eﬀects of fatigue on motivation, exercise, physical functioning, ability to carry out duties, work, family or social life. Responders are asked to rate each statement considering the previous week and using a Likert scale ranging from one (strong disagreement) to seven (strong agreement). The total score is computed by averaging the raw scores of each item. 2.4. Evaluation of Coronary Heart Disease Risk Selected CHD-R factors, including systolic and diastolic blood pressures (SBP and DBP, respectively), total and high-density lipoprotein cholesterol (T-C and HDL-C, respectively), cigarette smoking and diabetes mellitus, were assessed in each subject [30]. Two BP determinations were made after the participant had been sitting at least 5 min, being the mean value used for the subsequent analyses. BP and T-C/HDL-C levels were considered without regard to the use of anti-hypertensive or lipid-lowering drugs. The diagnosis of diabetes mellitus was deﬁned if the patient was under treatment with insulin or oral hypoglycemic agents or, alternatively, if fasting blood glucose levels exceeded 140 mg/dL. Individuals who smoked at least one cigarette per day during the previous 12 months were classiﬁed as smokers. Nutrients 2019, 11, 1240 5 of 15 The CHD-R scores were estimated using a simple CHD prediction model developed by Wilson et al. [30], which takes into account gender, age, diabetes, smoking, BP, T-C and HDL-C categories. This scoring scale, based on established, independent and biologically important risk factors for CHD, represents a simpliﬁed approach to predict risk for initial CHD events in disease-free outpatients at 10 years. CHD score sheets (for men and women) attribute diﬀerent ranks in the function of classes of age (nine subgroups, from 30 up to 74 years), T-C (ﬁve subgroups, from <160 to ≥280 mg/dL), HDL-C (ﬁve subgroups, from <35 up to ≥60 mg/dL) and BP (SBP: Five subgroups, from <120 up to ≥160 mmHg; DBP: Five subgroups, from <80 up to ≥100 mmHg). Diabetes and smoking were deﬁned into two categories (yes/no). When SBP and DBP fell into diﬀerent categories, the higher category was selected for the calculation of the CHD score. 2.5. Hematochemical Testing Serum glucose level was measured by the glucose oxidase enzymatic method (Roche Diagnostics, Monza, Italy). The sensitivity of the method was 2 mg/dL. Colorimetric enzymatic-assays (Roche Diagnostics, Monza, Italy) were used to determine serum T-C and HDL-C levels. The sensitivities of the assays were 3.86 mg/dL and 3.09 mg/dL, respectively. 2.6. Statistical Analysis The Sigma Stat 3.5 statistical software package (Systat Software, San Jose, CA, USA) was used for data analyses and the GraphPad Prisma 5.0 software (GraphPad Software, San Diego, CA, USA) for data plotting. The Shapiro-Wilk test showed that all parameters were normally distributed. Results are reported as mean ± SD (standard deviation). Each parameter, particularly BW, BMI, T-C, HDL-C, glucose, SBP, DBP, CHD-R, FSS score and SCT time, were evaluated not only as an absolute value, but also as a pre-post-BWRP diﬀerence (∆ in the corresponding unit of measurement for CHD-R or % for all remaining parameters). All parameters were compared among all subgroups (all, females/males, >60 years, 60–69 years and >70 years) before and after BWRP by using a t-Student test (for paired or unpaired data) or one-way ANOVA, followed by the post hoc Tukey’s test, if appropriate. A linear regression was applied to correlate each other parameter. A level of signiﬁcance of p < 0.05 was used for all data analyses. 3. Results As shown in Tables 1–3, WHR, height, BW and BMI were signiﬁcantly higher in males than females for each age range (>60 years, 60–69 years and >70 years) (p < 0.01) before BWRP, while signiﬁcantly lower HDL-C levels were found in males than females (p < 0.01). The CHD-R, FSS score and SCT time were signiﬁcantly higher in females than males (p < 0.01), with the exception of the subgroups aged >70 years. Males or females aged >70 years had a signiﬁcantly lower BW when compared to the other subgroups of the same gender (i.e., >60 years and 60–69 years) (p < 0.01). BMI was signiﬁcantly higher in females aged 60–69 years than in those aged >70 years (p < 0.01). Females aged >70 years had a signiﬁcantly higher SCT time than those belonging to the subgroup of the same gender, aged 60–69 years (p < 0.01). When pooling BW or BMI values of females and males (i.e., total), subjects aged >70 years had a signiﬁcant lower BW or BMI than the remaining subgroups (i.e., >60 years and 60–69 years) (p < 0.01). When pooling HDL-C levels of females and males (i.e., total), subjects aged >70 years had signiﬁcant lower HDL-C levels than the subgroup aged 60–69 years (p < 0.01). CHD-R was signiﬁcantly higher in subjects aged 60–69 years than in those aged >70 years, including both females and males (p < 0.01). Nutrients 2019, 11, 1240 6 of 15 Table 1. Demographic and anthropometric characteristics of the study population (before and after BWRP). Parameter >60 Year 60–69 Year >70 Year >60 Year 60–69 Year >70 Year >60 Year Total Males No. Age (year) WHR Height (m) 684 67.7 ± 4.7 1.0 ± 0.1 1.6 ± 0.1 500 64.7 ± 2.6 0.9 ± 0.1 1.6 ± 0.1 184 73.4 ± 3.1 1.0 ± 0.1 1.6 ± 0.1 92 67.0 ± 4.8 1.0 ± 0.1 a 1.7 ± 0.1 a 69 64.7 ± 2.5 1.0 ± 0.1 a 1.7 ± 0.1 a 23 74.1 ± 2.6 1.0 ± 0.1 a 1.6 ± 0.1 a 592 67.1 ± 4.7 0.9 ± 0.1 1.5 ± 0.1 PRE-BW (kg) POST-BW (kg) ∆BW (%) 104.3 ± 15.7 100.3 ± 15.2 c 4.7 ± 9.2 106.0 ± 16.4 101.8 ± 15.9 c 4.8 ± 8.7 99.6 ± 12.3 96.4 ± 12.2 c 4.4 ± 10.3 114.8 ± 16.8 a 110.1 ± 16.3 c 5.3 ± 10.1 118.1 ± 16.7 a 113.3 ± 16.3 c 5.7 ± 11.6 104.9 ± 12.6 a,b 98.8 ± 14.4 c 4.0 ± 1.5 102.6 ± 14.9 98.8 ± 14.4 c 4.6 ± 9.0 PRE-BMI (kg/m2) POST-BMI (kg/m2) ∆BMI (%) 42.6 ± 5.6 41.0 ± 5.5 c 4.0 ± 2.1 43.0 ± 5.9 41.3 ± 5.7 c 4.2 ± 1.8 41.4 ± 4.8 d 40.0 ± 4.8 c 3.5 ± 2.8 40.6 ± 4.8 e 38.9 ± 4.7 c 4.2 ± 1.7 41.1 ± 4.6 e 39.4 ± 4.6 c 4.3 ± 1.7 38.9 ± 5.0 e 37.4 ± 4.8 c 4.0 ± 1.6 42.9 ± 5.7 41.3 ± 5.6 c 3.9 ± 2.2 Females 60–69 Year 432 64.7 ± 2.6 0.9 ± 0.1 1.5 ± 0.1 104.0 ± 15.5 99.9 ± 15.0 c 4.6 ± 8.1 43.3 ± 6.0 41.6 ± 5.8 c 4.1 ± 1.8 >70 Year 160 73.4 ± 3.1 0.9 ± 0.1 1.5 ± 0.1 98.8 ± 12.1 b 95.7 ± 12.0 c 4.4 ± 11.1 41.8 ± 4.7 f 40.5 ± 4.7 c 3.4 ± 3.0 Values are expressed as mean ± SD. a p < 0.01 compared to the corresponding female subgroup of the same age; b p < 0.01 compared to the subgroups of the same gender, aged >60 years and 60–69 years; c p < 0.01 compared to the corresponding subgroup before BWRP; d p < 0.01 compared to the subgroups aged >60 years and 60–69 years, including both females and males (i.e., total); e p < 0.01 compared to the corresponding female subgroup of the same age range; f p < 0.01 compared to the subgroup of the same gender, aged 60–69 years. WHR, waist to hip circumference ratio; BW, body weight; BMI, body mass index; BWRP, body weight reduction program. Nutrients 2019, 11, 1240 7 of 15 Table 2. Biochemical and cardiovascular parameters of the study population (before and after BWRP). Parameter >60 Year 60–69 Year >70 Year >60 Year 60–69 Year >70 Year >60 Year Total Males Females 60–69 Year >70 Year PRE-GLU (mg/dL) POST-GLU (mg/dL) ∆GLU (%) 107.3 ± 33.9 88.0 ± 14.4 a 10.6 ± 14.5 108.3 ± 35.4 87.8 ± 14.9 a 10.9 ± 14.7 104.5 ± 29.0 88.5 ± 12.8 a 9.6 ± 13.9 105.4 ± 28.0 86.1 ± 12.8 a 11.7 ± 13.8 108.4 ± 30.3 86.2 ± 13.0 a 11.2 ± 15.6 96.1 ± 16.7 85.8 ± 12.6 a 12.9 ± 8.0 107.6 ± 34.7 88.5 ± 14.8 a 10.2 ± 14.8 108.2 ± 36.2 88.2 ± 15.4 a 10.8 ± 14.5 105.8 ± 30.3 89.3 ± 13.0 a 8.5 ± 15.6 PRE-T-C (mg/dL) POST-T-C (mg/dL) ∆T-C (%) 199.0 ± 38.8 169.7 ± 35.9 a 13.7 ± 14.3 199.2 ± 39.8 170.0 ± 36.2 a 13.5 ± 14.0 198.4 ± 36.1 168.8 ± 34.8 a 14.3 ± 15.1 193.7 ± 36.1 155.9 ± 36.8 a 17.9 ± 16.7 193.6 ± 36.0 156.4 ± 39.4 a 17.8 ± 17.9 194.0 ± 37.3 154.6 ± 28.0 a 18.2 ± 12.6 199.8 ± 39.2 171.8 ± 35.3 a 13.1 ± 13.8 200.1 ± 40.3 172.2 ± 35.3 a 12.8 ± 13.2 199.0 ± 36.1 170.6 ± 35.3 a 13.8 ± 15.4 PRE-HDL-C (mg/dL) POST-HDL-C (mg/dL) ∆HDL-C (%) PRE-DBP (mmHg) POST-DBP (mmHg) ∆DBP (%) PRE-SBP (mmHg) POST-SBP (mmHg) ∆SBP (%) 50.9 ± 13.5 42.7 ± 10.3 a 14.8 ± 13.7 77.6 ± 8.2 74.6 ± 6.7 a 3.2 ± 11.7 50.5 ± 13.9 42.6 ± 10.6 a 14.3 ± 13.2 77.8 ± 7.9 74.6 ± 7.0 a 3.5 ± 10.8 51.7 ± 12.1 b 43.0 ± 9.3 a 16.0 ± 15.0 76.9 ± 8.8 74.4 ± 5.9 a 2.6 ± 13.8 40.9 ± 9.2 c 35.6 ± 7.9 a 11.0 ± 13.3 76.8 ± 9.5 74.7 ± 7.5 a 2.7 ± 16.8 40.0 ± 9.0 c 34.8 ± 7.8 a 10.9 ± 13.8 78.0 ± 9.0 74.9 ± 7.9 a 3.3 ± 10.7 43.8 ± 9.6 c 38.1 ± 7.8 a 11.1 ± 11.8 73.3 ± 10.5 74.3 ± 6.2 1.0 ± 28.7 52.4 ± 13.4 43.8 ± 10.2 a 15.4 ± 13.7 77.7 ± 7.9 74.5 ± 6.6 a 3.3 ± 10.7 52.2 ± 13.9 43.8 ± 10.5 a 14.9 ± 13.0 77.8 ± 7.7 74.5 ± 6.8 a 3.5 ± 10.8 52.9 ± 12.1 43.8 ± 9.3 a 16.7 ± 15.3 77.4 ± 8.4 74.5 ± 5.9 a 2.9 ± 10.5 132.6 ± 14.6 123.8 ± 10.2 a 5.9 ± 9.9 132.3 ± 14.4 123.7 ± 10.3 a 5.7 ± 10.2 133.4 ± 15.4 124.0 ± 9.9 a 6.4 ± 9.1 131.5 ± 14.5 123.9 ± 11.1 a 5.0 ± 9.7 32.4 ± 14.0 124.0 ± 11.4 a 5.4 ± 10.0 128.6 ± 16.0 123.6 ± 10.4 a 3.7 ± 8.6 132.8 ± 14.7 123.8 ± 10.0 a 6.0 ± 9.9 132.2 ± 14.5 123.7 ± 10.1 a 5.7 ± 10.2 134.2 ± 15.1 124.1 ± 9.9 a 6.0 ± 9.1 Values are expressed as mean ± SD. a p < 0.01 compared to the corresponding subgroup before BWRP; b p < 0.01 compared to the subgroup aged 60–69 years, including both females and males (i.e., total); c p < 0.01 compared to the corresponding female subgroup of the same age. BWRP, body weight reduction program; GLU, glucose; T-C, total cholesterol; HDL-C, HDL cholesterol; DBP, diastolic blood pressure; SBP, systolic blood pressure. Nutrients 2019, 11, 1240 8 of 15 Table 3. CHD-R, FSS score and SCT time in the study population (before and after BWRP). Outcome >60 Year 60–69 Year >70 Year >60 Year 60–69 Year >70 Year >60 Year Total Males PRE-CHD-R (points) POST-CHD-R (points) ∆CHD-R (points) 10.2 ± 3.5 9.9 ± 3.5 d 0.4 ± 2.7 10.2 ± 3.4 9.9 ± 3.5 d 0.4 ± 2.7 10.1 ± 3.6 a 9.7 ± 3.5 d 0.6 ± 2.6 8.6 ± 2.9 b 7.3 ± 2.5 d 1.4 ± 2.9 b 8.4 ± 3.1 b 7.1 ± 2.7 d 1.4 ± 2.9 b 9.0 ± 2.3 7.9 ± 1.5 d 1.3 ± 2.7 c 10.4 ± 3.5 10.3 ± 3.5 0.3 ± 2.7 Females 60–69 Year 10.5 ± 3.4 10.4 ± 3.5 0.2 ± 2.7 >70 Year 10.3 ± 3.8 c 10.0 ± 3.6 0.5 ± 2.6 PRE-FSS score (points) POST-FSS score (points) ∆FSS score (%) 40.9 ± 13.5 34.1 ± 13.1 d 16.4 ± 18.5 40.1 ± 13.7 33.3 ± 13.2 d 16.6 ± 18.3 43.3 ± 12.4 36.2 ± 12.4 d 15.8 ± 18.9 36.2 ± 13.1 b 30.1 ± 11.2 d 15.0 ± 17.4 35.2 ± 13.2 b 29.9 ± 11.5 d 13.8 ± 16.0 39.7 ± 12.6 31.1 ± 10.3 d 19.3 ± 21.9 41.7 ± 13.4 34.7 ± 13.2 d 16.6 ± 18.6 40.8 ± 13.7 33.8 ± 13.4 d 17.1 ± 18.6 43.9 ± 12.3 36.9 ± 12.5 d 15.4 ± 18.7 PRE-SCT time (sec) POST-SCT time (sec) ∆SCT time (%) 7.7 ± 1.9 7.4 ± 1.9 d 4.3 ± 4.7 7.6 ± 1.8 7.2 ± 1.8 d 4.5 ± 4.9 8.2 ± 2.2 7.9 ± 2.1 d 4.0 ± 4.2 6.8 ± 1.5 b 6.5 ± 1.5 d 3.7 ± 3.3 6.6 ± 1.3 b 6.3 ± 1.3 d 3.8 ± 3.5 7.4 ± 2.0 7.2 ± 2.0 d 3.3 ± 2.9 7.9 ± 2.0 7.5 ± 1.9 d 4.5 ± 4.9 7.8 ± 1.9 7.4 ± 1.8 d 4.6 ± 5.1 8.4 ± 2.2 8.0 ± 2.1 d 4.1 ± 4.4 Values are expressed as mean ± SD. a p < 0.01 compared to the subgroup aged 60–69 years, including both females and males (i.e., total); b p < 0.01 compared to the corresponding female subgroup of the same age range; c p < 0.01 compared to the female subgroup aged 60–69 years. d p < 0.01 compared to the corresponding subgroup before BWRP; BWRP, body weight reduction program; CHD-R, coronary heart disease risk; FSS, fatigue severity scale; SCT, stair climbing test. Nutrients 2019, 11, 1240 9 of 15 BWRP signiﬁcantly reduced BW and BMI when considering all data, female and male groups and age- and gender-speciﬁc subgroups (p < 0.01) (Table 1; Figure 2). No signiﬁcant diﬀerences in ∆BW (%) and ∆BMI (%) were observed among all subgroups, indicating a similar eﬀectiveness of the intervention irrespectively from gender (females/males) and age (>60 years, 60–69 years and >70 years) (Table 1; Figure 3). In particular, a ∆BMI of 4.0 ± 2.1% was found for all data (Table 1; Figure 3). Figure 2. Body mass index (BMI, top panel), coronary heart disease risk (CHD RISK, top middle panel), fatigue severity scale (FSS) score (bottom middle panel) and stair climbing test (SCT) time (bottom panel) before and after a three-week body weight reduction program (BWRP) in elderly obese females and males. Data are expressed as mean ± SD. The values corresponding to each gender- and age-related subgroup (total, females/males, >60 years, 60–69 years and >70 years) are reported. * p < 0.01 compared to before body weight reduction program (BWRP). Nutrients 2019, 11, x FOR PEER REVIEW 8 of 17 Nutrients 2019, 11, x; doi: FOR PEER REVIEW www.mdpi.com/journal/nutrients BWRP significantly reduced BW and BMI when considering all data, female and male groups and age‐ and gender‐specific subgroups (p < 0.01) (Table 1; Figure 2). No significant differences in ΔBW (%) and ΔBMI (%) were observed among all subgroups, indicating a similar effectiveness of the intervention irrespectively from gender (females/males) and age (>60 years, 60–69 years and >70 years) (Table 1; Figure 3). In particular, a ΔBMI of 4.0 ± 2.1% was found for all data (Table 1; Figure 3). BMITOTAL (>60 yr)TOTAL(60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)0204060PRE-BWRPPOST-BWRP*******Patients GroupBMI (kg/m2)CHD RISKTOTAL (>60 yr)TOTAL(60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)051015PRE-BWRPPOST-BWRP**Patients GroupCHD RISKFSSTOTAL (>60 yr)TOTAL(60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)020406080PRE-BWRPPOST-BWRP*****Patients GroupFSSSCTTOTAL (>60 yr)TOTAL(60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)051015PRE-BWRPPOST-BWRP******Patients GroupSCT (sec)Nutrients 2019, 11, 1240 10 of 15 Figure 3. Changes of body mass index (∆BMI, top panel), coronary heart disease risk (∆CHD-R, top middle panel), fatigue severity scale (∆FSS) score (bottom middle panel) and stair climbing test (∆SCT) time (bottom panel) before and after a three-week body weight reduction program (POST-PRE-BWRP) in elderly obese females and males. Data are expressed as mean ± SD. The values corresponding to each gender- and age-related subgroup (total, females/males, >60 years, 60–69 years and >70 years) are reported. p < 0.01 compared to the corresponding female subgroup; p < 0.01 compared to females aged 60–69 years. ◦ BWRP signiﬁcantly reduced T-C and HDL-C in all subgroups (all, females/males, >60 years, 60–69 years and >70 years) (p < 0.01) (Table 2). Glucose was signiﬁcantly reduced by BWRP when comparing each subgroup before and after the intervention (p < 0.01) (Table 2). No signiﬁcant diﬀerences in ∆T-C, ∆HDL-C and ∆GLU (%) were detected among all subgroups (Table 2). In particular, a ∆T-C, ∆HDL-C and ∆GLU of 13.7 ± 14.3%, 14.8 ± 13.7% and 10.6 ± 14.5%, respectively, were found for all data (Table 2). SBP and DBP were signiﬁcantly reduced at the end of BWRP in all subgroups (p < 0.01), with the exception of DBP in males aged >70 years (Table 2). No signiﬁcant diﬀerences in ∆SBP and ∆DBP (%) were found among all subgroups (Table 2). In particular, a ∆SBP and ∆DBP of 5.9 ± 9.9% and 3.2 ± 11.7%, respectively, were found for all data (Table 2). Nutrients 2019, 11, x FOR PEER REVIEW 11 of 17 Nutrients 2019, 11, x; doi: FOR PEER REVIEW www.mdpi.com/journal/nutrients Figure 3. Changes of body mass index (ΔBMI, top panel), coronary heart disease risk (ΔCHD‐R, top middle panel), fatigue severity scale (ΔFSS) score (bottom middle panel) and stair climbing test (ΔSCT) time (bottom panel) before and after a three‐week body weight reduction program (POST‐PRE‐BWRP) in elderly obese females and males. Data are expressed as mean ± SD. The values corresponding to each gender‐ and age‐related subgroup (total, females/males, >60 years, 60–69 years and >70 years) are reported. * p < 0.01 compared to the corresponding female subgroup; ° p < 0.01 compared to females aged 60–69 years POST-PRE-BWRP BMI02468TOTAL (>60 yr)TOTAL (60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)OUTCOMECHANGEPOST-PRE-BWRP CHD-R012345TOTAL (>60 yr)TOTAL (60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)**°OUTCOMECHANGEPOST-PRE-BWRP FSS01020304050TOTAL (>60 yr)TOTAL (60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)OUTCOMECHANGEPOST-PRE-BWRP SCT051015TOTAL (>60 yr)TOTAL (60-69 yr)TOTAL (>70 yr)MALES (>60 yr)MALES (60-69 yr)MALES (>70 yr)FEMALES (>60 yr)FEMALES (60-69 yr)FEMALES (>70 yr)OUTCOMECHANGENutrients 2019, 11, 1240 11 of 15 Due to the favorable eﬀect of BWRP on BMI, T-C and BP, a signiﬁcant decrease in CHD-R was observed at the end of the intervention in all subgroups (p < 0.01), with the exception of females aged >60 years, 60–69 years and >70 years (Table 3; Figure 2). Noteworthy, when pooling data from males and females, due to the positive contribution of males aged >60 years, whose CHD-R signiﬁcantly decreased from 8.6 ± 2.9 to 7.3 ± 2.5, the pre-post-BWRP diﬀerence of CHD-R was signiﬁcant in the two genders considered together (aged >60 years, 60–69 years and >70 years) (p < 0.01) (Table 3; Figure 2). There were no signiﬁcant diﬀerences in ∆CHD-R (absolute value) among male or female subgroups irrespectively from age (>60 years, 60–69 years and >70 years); anyway, males aged >60 years or 60–69 years had a higher ∆CHD-R than that in the corresponding female subgroup (p < 0.01); furthermore, the ∆CHD-R in males aged >70 years was signiﬁcantly higher than that in females aged 60–69 years (p < 0.01), but not >70 years (Table 1; Figure 3). When pooling data from females and males, ∆CHD-R was 0.4 ± 2.7. BWRP signiﬁcantly reduced FSS score and SCT time in all subgroups (all, females/males, >60 years, 60–69 years and >70 years) (p < 0.01) (Table 3; Figure 2). There were no signiﬁcant diﬀerences in the ∆FSS score and ∆SCT time (%) among all subgroups (Table 3; Figure 3). In particular, a ∆FSS score and ∆SCT time of 16.4 ± 18.5% and 4.3 ± 4.7%, respectively, were found for all data (Table 3; Figure 3). Among all possible correlations, the most relevant were those between ∆BMI (%) and age (r = −0.153, p < 0.01) and between ∆FSS score (%) and age (r = −0.0763, p < 0.05) (Figure 4). Figure 4. Correlations of age with changes in body mass index (∆BMI, top panel) or fatigue severity scale (∆FSS,) score (bottom panel) before and after a three-week body weight reduction program (POST-PRE-BWRP) in elderly obese females and males. The results of the linear regression analysis are reported (i.e., the best ﬁtted line for all data together with 95% conﬁdence interval). 4. Discussion The main ﬁndings of the present study, carried out in a large cohort of elderly obese females and males, undergoing a three-week multidisciplinary BWRP, including hypocaloric diet, physical activity and psychological counseling, can be summarized in the following key points: (1) BWRP is capable of reducing BW and BMI in an obese geriatric population, including patients aged >70 years, the eﬀects being similar to those obtained in young obese adults (a ∆BMI of about 4%) [21,27,31,32]; (2) BWRP Nutrients 2019, 11, x FOR PEER REVIEW 13 of 17 Nutrients 2019, 11, x; doi: FOR PEER REVIEW www.mdpi.com/journal/nutrients Among all possible correlations, the most relevant were those between ΔBMI (%) and age (r = −0.153, p < 0.01) and between ΔFSS score (%) and age (r = −0.0763, p < 0.05) (Figure 4). Figure 4. Correlations of age with changes in body mass index (ΔBMI, top panel) or fatigue severity scale (ΔFSS,) score (bottom panel) before and after a three‐week body weight reduction program (POST‐PRE‐BWRP) in elderly obese females and males. The results of the linear regression analysis are reported (i.e., the best fitted line for all data together with 95% confidence interval). 4. Discussion The main findings of the present study, carried out in a large cohort of elderly obese females and males, undergoing a three‐week multidisciplinary BWRP, including hypocaloric diet, physical activity and psychological counseling, can be summarized in the following key points: (1) BWRP is capable of reducing BW and BMI in an obese geriatric population, including patients aged >70 years, the effects being similar to those obtained in young obese adults (a ΔBMI of about 4%) [21,27,31,32]; (2) BWRP improves some cardiovascular risk factors, such as T‐C, glucose, SBP and DBP, with the ensuing reduction of CHD‐R (a ΔCHD‐R of about 0.4 points), an effect more evident in all male subgroups irrespectively from age; (3) BWRP reduces the FSS score and SCT time, indicating beneficial effects on fatigue in daily activities and motor performance, which are requisites to obtain a better functional autonomy in the later life [33]; (4) an age‐dependent reduction in ΔBMI and ΔFSS score is present, the decline between 60 and 80 years being however relatively modest, indicating that BWRP is able to produce beneficial effects even in advanced age. As stated in recent recommendations by the most important scientific societies in clinical nutrition, geriatrics and obesiology [9], any BWRP administered in elderly obese patients, in addition 5560657075808590-505101520Age (yr)BMI (%)5560657075808590-100-50050100Age (yr) FSS (%)Nutrients 2019, 11, 1240 12 of 15 improves some cardiovascular risk factors, such as T-C, glucose, SBP and DBP, with the ensuing reduction of CHD-R (a ∆CHD-R of about 0.4 points), an eﬀect more evident in all male subgroups irrespectively from age; (3) BWRP reduces the FSS score and SCT time, indicating beneﬁcial eﬀects on fatigue in daily activities and motor performance, which are requisites to obtain a better functional autonomy in the later life [33]; (4) an age-dependent reduction in ∆BMI and ∆FSS score is present, the decline between 60 and 80 years being however relatively modest, indicating that BWRP is able to produce beneﬁcial eﬀects even in advanced age. As stated in recent recommendations by the most important scientiﬁc societies in clinical nutrition, geriatrics and obesiology [9], any BWRP administered in elderly obese patients, in addition to reduce obesity-associated medical complications, ought to focus on improving physical function and quality of life. Reportedly, fatigue is a highly prevalent symptom in obese subjects irrespectively from age, but the prevalence further increases in the geriatric population due to other comorbidities such as osteoarthritis and sarcopenia, which generally are not present in young subjects [24,34]. An obese patient complaining of fatigue can be trapped in a vicious circle because fatigue reduces physical activity and a sedentary state promotes weight gain, with a negative impact on the quality of life [35,36]. As shown in the present study, BWRP signiﬁcantly reduced both BW and FSS score in all subgroups of patients (male/female or >60 years, 60–69 years and >70 years); nevertheless, no correlation was found between the ∆BMI and ∆FSS score. These ﬁndings seem to suggest that the beneﬁcial eﬀects of BWRP on fatigue are not uniquely a consequence of the weight loss and the improved physical agility per se, but also due to a perceived psychological well being [37]. In this context, it is noteworthy to recall the mood-elevating eﬀect of exercise, even when practiced moderately, which can modulate monoaminergic neurotransmission, which governs physical symptoms, including fatigue [38]. SCT is a validated and simple test to assess the maximally attainable lower limbs muscle power output [22,23]. In the present study, the SCT time was shown to improve after BWRP. The relevance of this ﬁnding relies on the demographic characteristics of the recruited population, consisting in elderly obese subjects, for whom execution of the test may be more diﬃcult for many age- and not only BW-related reasons in comparison with a younger group [33]. Diﬀerent factors (i.e., reduction of BW, shift in the balance between parasympathetic and sympathetic activities, central and peripheral adaptations with more eﬃcient muscle contractions and a more favorable left shift of the muscle power-velocity curve, acquisition of motor skillfulness and coordination during the repeated trials, increased self-esteem and motivation) might be invoked, alone or in combination, to explain the positive eﬀects of BMRP on SCT-related motor performance [22,23]. Since a high level of physical activity has been shown to increase the odds of a healthy aging and protect the “frail” elderly patient independently from any eﬀect on BW [39], any eﬀort should be made to engage elderly obese subjects in a BWRP, appropriately tailored to patient’s clinical needs. The advantage of a (even partially) restored physical agility in an elderly obese subject undergoing a BWRP might overcome the cardiometabolic beneﬁts (see below) or contrast the so-called “obesity paradox”, a debated issue for which there would be a survival advantage in overweight/obese patients, calling into question the importance of weight loss particularly in geriatric population [19]. The present study conﬁrms the eﬀectiveness of our BWRP to reduce the CHD-R in elderly obese patients [40]. Though the duration of the BWRP was limited (only three weeks), the cardiometabolic beneﬁts were relevant as shown by a ∆CHD-R of about 0.4 points, which might correspond to a decreased (not uniquely cardiovascular) mortality over a long time interval, particularly in the subjects aged 60–69 years. Though the imposition of a healthy life style is fundamental to maintain the cardiometabolic beneﬁts produced by a (previous) BWRP over time [41], epidemiological studies in a large geriatric population are mandatory to detract the theory of the “obesity paradox”, including the elimination of methodological biases such as elderly obese patients for whom a catabolic state due to a BWPR would represent a precipitating factor of other pre-existing critical diseases (e.g., oncological or cardiological conditions) [42]. Nutrients 2019, 11, 1240 13 of 15 In the present study, ∆CHD-R was higher in males than females and similar among all male subgroups (>60 years, 60–69 years and >70 years); furthermore, BWRP did not signiﬁcantly reduce CHD-R in female subgroups. These ﬁndings are not surprising. In fact, as shown by other studies [43], there is a dramatic increase of CHD-R in females during the menopausal transition, for whom the well-known cardioprotective eﬀects of (endogenous) estrogens are missing [44]. Nevertheless, due to the multisystemic beneﬁts that are produced by a BWRP in both males and females, post-menopausal women, who epidemiologically have a longer life expectancy than men [45] and have a higher risk of weight gain because of the hypoestrogenism [5], should be encouraged to take part in any BWRP to obtain other beneﬁts, e.g., increased motor performance. Though statistically signiﬁcant, the correlations of age with the ∆BMI or ∆FSS score do not represent a valid reason to exclude the very elderly subject from a BWRP. In fact, the age-dependent decline of BWRP eﬀectiveness appears to be modest and, despite this fact, as reported above, the BWRP-induced beneﬁts are evident in all subgroups, including subjects aged >70 years. In our opinion, this age-dependent hyporesponsiveness should be further investigated to better optimize the BWRP for very elderly subjects, who could have diﬀerent clinical needs [7–10,14]. In conclusion, though only a limited number of outcomes was investigated, the present study shows that a 4% BW reduction following a multidisciplinary BWRP in severely elderly obese patients is associated with positive multisystemic eﬀects, particularly, muscle-skeletal and cardiometabolic beneﬁts, which can favorably inﬂuence their general well-being and improve the autonomy level in performing the more common daily activities. The maintenance of healthy life styles, including controlled food intake and regular physical activity, after a period of metabolic rehabilitation (i.e., BWRP), which can be repeated and/or personalized when needed, is strongly recommended in all elderly obese patients to improve their clinical condition and autonomy level [41]. Author Contributions: A.S. designed the study. L.A., A.D.C., S.T., D.C. and S.C. enrolled the subjects and administered the FSS; R.D.M. and G.T. performed SCT and calculated the CHD-R. N.M., S.T., A.D.C., L.A., D.C., S.C., G.T. and R.D.M. elaborated the database. A.E.R. analyzed the data and, together with A.S., wrote the manuscript. S.G.C. contributed to data interpretation and discussion writing. All authors contributed to the manuscript revision. Funding: The study was supported by Progetti di Ricerca Corrente, Istituto Auxologico Italiano, IRCCS, Milan, Italy. Acknowledgments: The authors thank the nursing staﬀ at the Division of Metabolic Diseases, Istituto Auxologico Italiano, Piancavallo, VB, Italy. Our special thanks go to the subjects for their willingness to participate in this research. Conﬂicts of Interest: The authors declare no conﬂict of interest. Availability of Data and Materials: The datasets used and/or analyzed in the present study are available from the corresponding author on reasonable request. References 1. Flegal, K.M.; Kruszon-Moran, D.; Carroll, M.D.; Fryar, C.D.; Ogden, C.L. Trends in obesity among adults in theUnited States, 2005 to 2014. JAMA 2016, 315, 2284–2291. [CrossRef] [PubMed] 3. 2. Ng, M.; Fleming, T.; Robinson, M.; Thomson, B.; Graetz, N.; Margono, C.; Mullany, E.C.; Biryukov, S.; Abbafati, C.; Abera, S.F.; et al. Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: A systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014, 384, 766–781. [CrossRef] Amarya, S.; Singh, K.; Sabharwal, M. Health consequences of obesity in the elderly. J. Clin. Gerontol. Geriatr. 2014, 5, 63–67. [CrossRef] Bales, C.W.; Porter Starr, K.N. Obesity interventions for older adults: Diet as a determinant of physical function. Adv. Nutr. 2018, 9, 151–159. [CrossRef] [PubMed] Brown, T.J. Health beneﬁts of weight reduction in postmenopausal women: A systematic review. J. Br. Menopause Soc. 2006, 12, 164–171. [CrossRef] 5. 4. Nutrients 2019, 11, 1240 14 of 15 6. 7. 8. 9. Locher, J.L.; Goldsby, T.U.; Goss, A.M.; Kilgore, M.L.; Gower, B.; Ard, J.D. Calorie restriction in overweight older adults: Do beneﬁts exceed potential risks? Exp. Gerontol. 2016, 86, 4–13. [CrossRef] [PubMed] Tzankoﬀ, S.P.; Norris, A.H. Eﬀect of muscle mass decrease on age-related BMR changes. J. Appl. Physiol. Respir. Environ. Exerc. Physiol. 1977, 43, 1001–1006. [CrossRef] Elia, M.; Ritz, P.; Stubbs, R.J. Total energy expenditure in the elderly. Eur. J. Clin. Nutr. 2000, 54 (Suppl. 3), S92–S103. [CrossRef] [PubMed] Villareal, D.T.; Apovian, C.M.; Kushner, R.F.; Klein, S. Obesity in older adults: Technical review and position statement of the American Society for Nutrition and NAASO, The Obesity Society. Am. J. Clin. Nutr. 2005, 82, 923–934. [CrossRef] 10. Ten Haaf, T.; Verreijen, A.M.; Memelink, R.G.; Tieland, M.; Weijs, P.J. Reduction in energy expenditure during weight loss is higher than predicted based on fat free mass and fat mass in older adults. Clin. Nutr. 2018, 37, 250–253. [CrossRef] 11. Porter Starr, K.N.; McDonald, S.R.; Bales, C.W. Obesity and physical frailty in older adults: A scoping review of lifestyle intervention trials. J. Am. Med. Dir. Assoc. 2014, 15, 240–250. [CrossRef] [PubMed] 12. Costamagna, D.; Costelli, P.; Sampaolesi, M.; Penna, F. Role of inﬂammation in muscle homeostasis and myogenesis. Mediat. Inﬂamm. 2015, 2015, 805172. [CrossRef] [PubMed] 13. Weinheimer, E.M.; Sands, L.P.; Campbell, W.W. A systematic review of the separate and combined eﬀects of energy restriction and exercise on fat-free mass in middle-aged and older adults: Implications for sarcopenic obesity. Nutr. Rev. 2010, 68, 375–388. [CrossRef] 14. Kalyani, R.R.; Metter, E.J.; Ramachandran, R.; Chia, C.W.; Saudek, C.D. Glucose and insulin measurements from the oral glucose tolerance test and relationship to muscle mass. J. Gerontol. Ser. A Biol. Sci. Med. Sci. 2012, 67, 74–81. [CrossRef] [PubMed] 16. 15. Giunta, M.; Rigamonti, A.E.; Agosti, F.; Patrizi, A.; Compri, E.; Cardinale, M.; Sartorio, A. Combination of external load and whole body vibration potentiates the GH-releasing eﬀect of squatting in healthy females. Horm. Metab. Res. 2013, 45, 611–666. [CrossRef] [PubMed] Sartorio, A.; Lafortuna, C.L.; Maﬃuletti, N.A.; Agosti, F.; Marazzi, N.; Rastelli, F.; Rigamonti, A.E.; Muller, E.E. GH responses to two consecutive bouts of whole body vibration, maximal voluntary contractions or vibration alternated with maximal voluntary contractions administered at 2-h intervals in healthy adults. Growth Horm. IGF Res. 2010, 20, 416–421. [CrossRef] [PubMed] 17. Rigamonti, A.E.; Locatelli, L.; Cella, S.G.; Bonomo, S.M.; Giunta, M.; Molinari, F.; Sartorio, A.; Müller, E.E. Muscle expressions of MGF, IGF-IEa, and myostatin in intact and hypophysectomized rats: Eﬀects of rhGH and testosterone alone or combined. Horm. Metab. Res. 2009, 41, 23–29. [CrossRef] [PubMed] Soltani, S.; Hunter, G.R.; Kazemi, A.; Shab-Bidar, S. The eﬀects of weight loss approaches on bone mineral density in adults: A systematic review and meta-analysis of randomized controlled trials. Osteoporos. Int. 2016, 27, 2655–2671. [CrossRef] 18. 19. Wang, S.; Ren, J. Obesity paradox in aging: From prevalence to pathophysiology. Prog. Cardiovasc. Dis. 2018, 61, 182–189. [CrossRef] [PubMed] 21. 22. 20. Evans, W.J.; Morley, J.E.; Argiles, J.; Bales, C.; Baracos, V.; Guttridge, D.; Jatoi, A.; Kalantar-Zadeh, K.; Lochs, H.; Mantovani, G.; et al. Cachexia: A new deﬁnition. Clin. Nutr. 2008, 27, 793–799. [CrossRef] Sartorio, A.; Lafortuna, C.L.; Marinone, P.G.; Tavani, A.; La Vecchia, C.; Bosetti, C. Short-term eﬀects of two integrated, non-pharmacological body weight reduction programs on coronary heart disease risk factors in young obese patients. Diab. Nutr. Metab. 2003, 16, 262–265. Sartorio, A.; Lafortuna, C.L.; Conte, G.; Faglia, G.; Narici, M.V. Changes in motor control and muscle performance after a short-term body mass reduction program in obese subjects. J. Endocrinol. Investig. 2001, 24, 393–398. [CrossRef] [PubMed] Sartorio, A.; Narici, M.V.; Fumagalli, E.; Faglia, G.; Lafortuna, C.L. Aerobic and anaerobic performance before and after a short-term body mass reduction program in obese subjects. Diab. Nutr. Metab. 2001, 14, 51–57. Impellizzeri, F.M.; Agosti, F.; De Col, A.; Sartorio, A. Psychometric properties of the Fatigue Severity Scale in obese patients. Health Qual. Life Outcomes 2013, 11, 32. [CrossRef] Sartorio, A.; Proietti, M.; Marinone, P.G.; Agosti, F.; Adorni, F.; Lafortuna, C.L. Inﬂuence of gender, age and BMI on lower limb muscular power output in a large population of obese men and women. Int. J. Obes. 2004, 28, 91–98. [CrossRef] [PubMed] 23. 24. 25. Nutrients 2019, 11, 1240 15 of 15 26. Margaria, R.; Aghemo, P.; Rovelli, E. Measurement of muscular power (anaerobic) in man. J. Appl. Physiol. 27. 1966, 21, 1662–1664. [CrossRef] Sartorio, A.; Fontana, P.; Trecate, L.; Lafortuna, C.L. Short-term changes of fatigue and muscle performance in severe obese patients after an integrated body mass reduction program. Diab. Nutr. Metab. 2003, 16, 88–93. 28. Hjollund, N.H.; Andersen, J.H.; Bech, P. Assessment of fatigue in chronic disease: A bibliographic study of fatigue measurement scales. Health Qual. Life Outcomes 2007, 5, 12. [CrossRef] 29. Elbers, R.G.; Rietberg, M.B.; van Wegen, E.E.; Verhoef, J.; Kramer, S.F.; Terwee, C.B.; Kwakkel, G. Self-report fatigue questionnaires in multiple sclerosis, Parkinson’s disease and stroke: A systematic review of measurement properties. Qual. Life Res. 2012, 21, 925–944. [CrossRef] 30. Wilson, P.W.F.; D’Agostino, R.B.; Levy, D.; Belanger, A.M.; Silbershats, H.; Kannel, W.B. Prediction of coronary hearth disease using risk factor categories. Circulation 1998, 97, 1837–1847. [CrossRef] 31. Rigamonti, A.E.; Piscitelli, F.; Aveta, T.; Agosti, F.; De Col, A.; Bini, S.; Cella, S.G.; Di Marzo, V.; Sartorio, A. Anticipatory and consummatory eﬀects of (hedonic) chocolate intake are associated with increased circulating levels of the orexigenic peptide ghrelin and endocannabinoids in obese adults. Food Nutr. Res. 2015, 59, 29678. [CrossRef] [PubMed] 32. Rigamonti, A.E.; Resnik, M.; Compri, E.; Agosti, F.; De Col, A.; Monteleone, P.; Marazzi, N.; Bonomo, S.M.; Müller, E.E.; Sartorio, A. The cholestyramine-induced decrease of PYY postprandial response is negatively correlated with fat mass in obese women. Horm. Metab. Res. 2011, 43, 569–573. [CrossRef] [PubMed] Jensen, G.L. Obesity and functional decline: Epidemiology and geriatric consequences. Clin. Geriatr. Med. 2005, 21, 677–687. [CrossRef] 33. 34. Moreh, E.; Jacobs, J.M.; Stessman, J. Fatigue, function, and mortality in older adults. J. Gerontol. Ser. A Biol. 35. 36. Sci. Med. Sci. 2010, 65, 887–895. [CrossRef] [PubMed] Fritschi, C.; Quinn, L. Fatigue in patients with diabetes: A review. J. Psychosom. Res. 2010, 69, 33–41. [CrossRef] Singh, R.; Teel, C.; Sabus, C.; McGinnis, P.; Kluding, P. Fatigue in Type 2 Diabetes: Impact on Quality of Life and Predictors. PLoS ONE 2016, 11, e0165652. [CrossRef] 37. Ten Hoor, G.A.; Kok, G.; Peters, G.Y.; Frissen, T.; Schols, A.M.; Plasqui, G. The psychological eﬀects of strength exercises in people who are overweight or obese: A systematic review. Sports Med. 2017, 47, 2069–2081. [CrossRef] 38. Lin, T.W.; Kuo, Y.M. Exercise beneﬁts brain function: The monoamine connection. Brain Sci. 2013, 3, 39–53. [CrossRef] [PubMed] 39. Liao, C.D.; Lee, P.H.; Hsiao, D.J.; Huang, S.W.; Tsauo, J.Y.; Chen, H.C.; Liou, T.H. Eﬀects of protein supplementation combined with exercise intervention on frailty indices, body composition, and physical function in frail older adults. Nutrients 2018, 10, 1916. [CrossRef] [PubMed] 40. Cottell, K.E.; Dorfman, L.R.; Straight, C.R.; Delmonico, M.J.; Lofgren, I.E. The eﬀects of diet education plus light resistance training on coronary heart disease risk factors in community-dwelling older adults. J. Nutr. Health Aging 2011, 15, 762–767. [CrossRef] [PubMed] 41. Wadden, T.A.; Butryn, M.L.; Byrne, K.J. Eﬃcacy of lifestyle modiﬁcation for long-term weight control. Obes. Res. 2004, 12 (Suppl. 12), 151S–162S. [CrossRef] 42. Braun, N.; Gomes, F.; Schütz, P. “The obesity paradox” in disease—Is the protective eﬀect of obesity true? Swiss Med. Wkly. 2015, 145, w14265. [CrossRef] [PubMed] 43. Wells, G.L. Cardiovascular risk factors: Does sex matter? Curr. Vasc. Pharmacol. 2016, 14, 452–457. [CrossRef] [PubMed] 44. Kallen, A.N.; Pal, L. Cardiovascular disease and ovarian function. Curr. Opin. Obstet. Gynecol. 2011, 23, 258–267. [CrossRef] [PubMed] 45. Barford, A.; Dorling, D.; Davey Smith, G.; Shaw, M. Life expectancy: Women now on top everywhere. BMJ 2006, 332, 808. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.15252_embj.2023113578	Article The Ebola virus VP40 matrix layer undergoes endosomal disassembly essential for membrane fusion Sophie L Winter1,2 Keerthihan Thiyagarajah1,2 Britta Brügger4, Thomas Hoenen8 , Samy Sid Ahmed6, Christian Lüchtenborg4, Oliver T Fackler6,7 , Melina Vallbracht1,2, , Fabio Lolicato4,5 , Gonen Golani2,3 , Walter Nickel4 , Ulrich S Schwarz2,3 & Petr Chlanda1,2,* , Abstract Ebola viruses (EBOVs) assemble into filamentous virions, whose shape and stability are determined by the matrix viral protein 40 (VP40). Virus entry into host cells occurs via membrane fusion in late endosomes; however, the mechanism of how the remarkably long virions undergo uncoating, including virion disassembly and nucleocapsid release into the cytosol, remains unknown. Here, we investigate the structural architecture of EBOVs entering host cells and discover that the VP40 matrix disassembles prior to membrane fusion. We reveal that VP40 disassembly is caused by the weaken- ing of VP40–lipid interactions driven by low endosomal pH that equilibrates passively across the viral envelope without a dedicated ion channel. We further show that viral membrane fusion depends on VP40 matrix integrity, and its disassembly reduces the energy barrier for fusion stalk formation. Thus, pH-driven structural remo- deling of the VP40 matrix acts as a molecular switch coupling viral matrix uncoating to membrane fusion during EBOV entry. Keywords Ebola virus; in situ cryo-ET; membrane fusion; virus entry and uncoating; membrane modeling and molecular dynamics simulations Subject Category Microbiology, Virology & Host Pathogen Interaction DOI 10.15252/embj.2023113578 \| Received 21 January 2023 \| Revised 9 March 2023 \| Accepted 22 March 2023 \| Published online 21 April 2023 The EMBO Journal (2023) 42: e113578 Introduction Ebola viruses (EBOVs) are highly pathogenic negative-sense RNA viruses causing severe outbreaks of viral hemorrhagic fever in humans with high case fatality rates (Feldmann & Klenk, 1996). They enter host cells by macropinocytosis and undergo cytosolic entry in late endosomal compartments, where the fusion of the viral and endosomal membranes leads to genome release into the cyto- plasm. EBOVs are characterized by their filamentous morphology, which is determined by the matrix composed of the viral protein 40 (VP40) that drives budding of virions reaching up to several micro- meters in length (Geisbert & Jahrling, 1995; Bharat et al, 2012). VP40 interacts with negatively charged lipids (Ruigrok et al, 2000; Jeevan et al, 2016; Johnson et al, 2016; Amiar et al, 2021) to assem- ble into a quasi-helical scaffold underneath the viral membrane (Noda et al, 2002; Wan et al, 2020) and is critical for the incorpora- tion of the viral nucleocapsid into the virions by so far unknown VP40–nucleocapsid interactions. The EBOV nucleocapsid is com- posed of the nucleoprotein (NP), which encapsidates the single- stranded RNA genome, as well as VP24 and VP35 (Bharat et al, 2012; Wan et al, 2017; Takamatsu et al, 2018). Upon host cell entry, the nucleocapsid needs to dissociate from the virus particle and viral genome to enable genome replication and transcription (Greber et al, 1994). These processes together are referred to as virus uncoating, which involves the weakening of protein–protein and protein–membrane interactions inside the virus lumen. The resulting changes in virion architecture allow the timely nucleocap- sid release upon membrane fusion (Yamauchi & Greber, 2016). It is well established that fusion of the viral and endosomal membrane relies on interactions with the EBOV fusion glycoprotein (GP), which is the only transmembrane protein that studs the Ebola viral envelope (Dube et al, 2009; Lee & Saphire, 2009; Nanbo et al, 2010). GP-mediated membrane fusion is triggered after proteolytic proces- sing of GP by host cell cathepsin proteases (Brecher et al, 2012) and depends on the interaction of the cleaved GP subunit GP1 with the late endosomal Niemann-Pick C1 (NPC1) receptor (Carette et al, 2011; C^ot(cid:2)e et al, 2011; Miller et al, 2012; Simmons 1 Schaller Research Groups, Department of Infectious Diseases, Virology, University Hospital Heidelberg, Heidelberg, Germany 2 BioQuant-Center for Quantitative Biology, Heidelberg University, Heidelberg, Germany 3 Institute for Theoretical Physics, Heidelberg University, Heidelberg, Germany 4 Heidelberg University Biochemistry Center, Heidelberg, Germany 5 Department of Physics, University of Helsinki, Helsinki, Finland 6 Department of Infectious Diseases, Integrative Virology, University Hospital Heidelberg, Heidelberg, Germany 7 German Centre for Infection Research (DZIF), Partner Site Heidelberg, Heidelberg, Germany 8 Institute of Molecular Virology and Cell Biology, Friedrich-Loeffler-Insitut, Greifswald-Insel Riems, Greifswald, Germany *Corresponding author. Tel: +49 6221 54 51231; E-mail: [email protected] (cid:1) 2023 The Authors. Published under the terms of the CC BY 4.0 license. The EMBO Journal 42: e113578 \| 2023 1 of 20 The EMBO Journal Sophie L Winter et al et al, 2015). However, the molecular mechanism of how the remark- ably long EBOV virions undergo uncoating during cytosolic entry remains enigmatic. A growing body of evidence shows that matrix disassembly during viral entry can trigger a cascade of events required for viral uncoating and efficient virus entry (Banerjee et al, 2014; Li et al, 2014). While the structure of isolated Ebola virions is well characterized, it is currently unknown whether the VP40 matrix undergoes conformational changes during virion entry and factors initiating EBOV disassembly remain to be elucidated. In addition, a mechanistic understanding of how interactions between the EBOV VP40 matrix, the viral membrane, and nucleocapsid are modulated during viral entry is still missing. Since EBOVs is a late- penetrating virus, which requires low endosomal pH for cytosolic entry (Lozach et al, 2011), the acidic environment may serve as one of the triggers for virion uncoating. Here, we investigate EBOV uncoating and the role of VP40 during virus entry into host cells by characterizing EBOVs in endosome- mimicking conditions in vitro, and in endo-lysosomal compartments in situ, by cryo-electron tomography (cryo-ET), which is comple- mented by membrane modeling approaches, lipidomics, and time- lapse fluorescence imaging. We find that the VP40 matrix and its interactions with lipids in the viral envelope are sensitive to low pH, which passively equilibrates across the viral envelope in acidic envi- ronments. This leads to the disassembly of the matrix layer allowing for fusion and genome release. Results The Ebola virus VP40 matrix undergoes disassembly in endosomal compartments To shed light on endosomal uncoating of EBOV virions at molecular resolution, we infected Huh7 cells cultured on electron microscopy grids with EBOVs (Zaire ebolavirus species, Mayinga strain) in bio- safety level 4 (BSL4) containment. Infected cells were chemically fixed using 4% paraformaldehyde (PFA) and 0.1% glutaraldehyde (GA) after multiple rounds of infection had occurred at 48 h post- infection (Fig 1A, at 22 h post-infection: Figs EV1 and EV2, Appen- dix Figs S1 and S2). After vitrification and cryo-focused ion beam (cryo-FIB) milling of the infected cells, we performed in situ cryo-ET of endosomal compartments containing EBOV virions (Fig 1B–G, Movie EV1). Late endosomal compartments were identified by the presence of vesicles and membrane fragments (white arrow, Fig 1F), which are likely products of lysosomal degradation. In addi- tion, we observed the accumulation of crystalline lipidic structures with a spacing of 3.2 nm (Fig EV1), consistent with the spacing found in cholesterol ester crystals previously described in lamellar bodies, lipid droplets, and isolated low-density lipoprotein particles (van Niel et al, 2015; Mahamid et al, 2019; Klein et al, 2020b). Inter- estingly, Ebola virions in late endosomes retained their filamentous morphology and displayed well-defined nucleocapsids of approxi- mately 20 nm in diameter (Fig 1, Appendix Fig S1). They appeared condensed and resembled nucleocapsid structures formed by trun- cated EBOV NP alone (Wan et al, 2017) but lacked the regular pro- trusions observed in nucleocapsids of isolated virions (Fig EV2). However, the VP40 matrix layer was detached from the envelope as apparent from the empty space adjacent to the EBOV membrane and disordered protein densities, which presumably represent disas- sembled VP40, surrounding the nucleocapsid in the EBOV lumen (Fig 1D–G). Importantly, none of the five EBOV virions captured in endosomes displayed ordered VP40 matrices, and two virions had engulfed intraluminal vesicles (Appendix Fig S1). In contrast, bud- ding virions and extracellular virions adjacent to the plasma mem- brane of infected cells displayed assembled VP40 layers with VP40 proteins visible as distinct densities lining the membrane (Fig 1H–K, Appendix Fig S2, n = 8), similar to the VP40 layer in isolated virions (Fig EV2). We were not able to capture virions residing in endo- somes in the process of fusing with the endosomal membrane, pre- sumably because virus membrane fusion is a rapid event. However, in a similar experiment using virus-like particles (VLPs) composed of VP40 and GP, we could confirm the absence of ordered VP40 matrix layers in VLPs inside endosomal compartments. Moreover, we were able to capture one fusion event and several intracellular membranes studded with luminal GPs, indicating that fusion had taken place (Fig EV3). Overall, these data indicate that EBOV uncoating involves VP40 disassembly in late endosomal compart- ments and suggest that endosomal VP40 disassembly occurs prior to GP-mediated membrane fusion. Low pH triggers disassembly of the EBOV matrix in vitro We next sought to identify factors driving VP40 disassembly. Since EBOVs enter host cells via late endosomes, which are characterized by low pH, we assessed the effect of external pH on the shape of EBOV VLPs and, in particular, on the structure of the VP40 matrix. VLPs composed of VP40 and GP were produced from HEK 293T cells and analyzed by cryo-ET (Fig 2A–D). At neutral pH, the organi- zation of VP40 proteins into a helical scaffold was apparent from transverse cross-sections as an additional profile adjacent to the inner membrane monolayer and as regular striations spanning the width of the particles when observed close to the VLP surface (Fig 2B and C). Individual VP40 proteins were visible as distinct densities lining the membrane (Fig 2D). To understand their organi- zation within the matrix, we applied subtomogram averaging of the VP40 matrix in purified VLPs. In accord with recently published data (Wan et al, 2020), the subtomogram average revealed the lin- ear arrangement of VP40 dimers via their C-terminal domains (CTDs), which are directly connected to the inner membrane mono- layer (Fig 2E). The available crystal structure of the VP40 dimer (pdb: 7jzj) fitted well into the average (Fig 2F) except for three short helical segments of one VP40 monomer (Fig EV4A). To assess whether the VP40 matrix undergoes disassembly at low pH, VLPs were then subjected to the late endosomal pH of 4.5 for 30 min. Consistent with Ebola virions found in late endosomes, the VLPs retained their overall filamentous morphology but did not show ordered VP40 matrix layers. Instead, they contained disor- dered protein aggregates accumulated at the VLP core (Fig 2G–J). Additionally, a lack of densities between the membrane and protein aggregates indicates that VP40 detaches from the membrane, as par- ticularly apparent from the cross-sections (Fig 2H and J), which was also reflected in a more variable particle diameter (Fig EV4B). To elucidate whether this phenotype depends on the presence of EBOV GP, VLPs composed of VP40 alone were analyzed by cryo-ET. The presence and absence of the ordered VP40 matrix at neutral and low pH, respectively, were clearly apparent as regular striations and 2 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal Figure 1. (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 3 of 20 The EMBO Journal Sophie L Winter et al ◀ Figure 1. In situ cryo-electron tomography of EBOV virions localized in endosomes of an infected cell. A B C Schematic of the in situ cryo-ET workflow, including infection of cells grown on electron microscopy grids and chemical fixation using 4% PFA and 0.1% GA for bio- safety reasons before removal from BSL4. Vitrification was performed prior to cell thinning by cryo-FIB milling and imaging by cryo-ET. Slice through a tomogram showing EBOV virions inside a late endosomal compartment. 3D segmentation of the delimiting endosomal membrane (yellow), cholesterol ester crystals (pink), viral membranes (shades of green) of three EBOV virions (numbered 1–3), and nucleocapsids (shades of light green) for visualization. D Magnified view of the area highlighted in (B) showing the transverse cross-section of a virion. A cholesterol ester crystal adjacent to the virion is marked by a white arrowhead. 3D segmentation of the viral membrane, nucleocapsid, and VP40 shown in (D). E F Magnified view of a different slice of the tomogram in (B) showing a longitudinal cross-section through a virion. A linear membrane fragment adjacent to the virion is marked with a white arrow. 3D segmentation of the viral membrane, nucleocapsid, and VP40 displayed in (F). G H Slices through a tomogram showing an extracellular EBOV adjacent to the plasma membrane of an infected cell. I, J Magnified areas highlighted in (F) and (H), respectively, showing the viral membrane and VP40 densities at the luminal side. For comparison, line profiles at 3 nm distance from the inner membrane monolayer, visualized by dotted profiles (magenta and blue, respectively), were determined. Line profiles adjacent to the inner viral membrane leaflet of a virion inside an endosome and a purified virion before infection. K Data information: Scale bars: (B), (H): 200 nm, (D), (F): 50 nm, (I), (J): 20 nm. Source data are available online for this figure. disordered protein accumulations at the particles’ cores (Figs 2K and EV4C). Accordingly, line density profiles proximal to the inner mem- brane monolayer of VLPs showed the 5–6 nm pitch of the assembled VP40 matrix at neutral pH, whereas no repeating densities were detected at low pH (Fig 2L). We further repeated the experiment using VLPs composed of VP40, GP and the nucleocapsid proteins NP, VP24, and VP35, and observed the same low pH-phenotype described above. These results show that nucleocapsid proteins do not influence the VP40 disassembly driven by low pH. Performing the experiments on unpurified VLPs harvested from the supernatant of transfected cells confirmed that the purification protocol applied did not influence the disassembly of the VP40 matrix (Appendix Fig S3). Hence, pH-mediated VP40 disassembly is independent of other viral proteins including the transmembrane protein GP. VP40 interactions with negatively charged lipids are weakened at low pH To further probe the specific VP40–lipid interactions at neutral and low pH, we performed all-atom molecular dynamics (MD) simula- tions and modeled the binding of VP40 dimers to membrane lipids at different pH levels. To this end, we emulated a simplified mem- brane containing 30% phosphatidylcholine, 40% cholesterol, and 30% phosphatidylserine mimicking the overall negative charge of the VLP inner membrane monolayer. We modeled missing C- terminal residues, which are inherently flexible and disordered, into the VP40 dimer structure (pdb: 7jzj, Wan et al, 2020) and simulated VP40–membrane interactions for a cumulative time of 10 μs for each pH using the CHARMM36m force field (Pastor & MacKerell, 2011; Huang & Mackerell, 2013; Huang et al, 2016). We show that after one CTD of the VP40 dimer established interactions with phosphati- dylserines, the second CTD is pulled toward the membrane, leading to the anchoring of the dimer into the membrane (Fig 3A). The membrane interactions were driven by positively charged residues decorating the C-termini of the VP40 dimer, including K224, K225, K274, and K275, which corroborates experimental data showing that these residues form a basic patch required for membrane associa- tion and budding of VLPs (Bornholdt et al, 2013). In the MD simula- tions, the basic patches strongly promote lipid interactions and localize in flexible loops at the CTDs, which penetrate into the inner membrane monolayer (Fig 3B and C) and correspond to the previ- ously unassigned densities (Wan et al, 2020) between the VP40 matrix and viral membrane in the subtomogram average (Fig 2E and F). Moreover, the MD simulations showed that the rotation angle of VP40 monomers oscillates around 1° (SD 9.5) along the N- terminal-dimerization domain and is in agreement with the subto- mogram average (Figs 3B and EV5C–E), such that only flexible loops protrude from the average (Fig 3D). Accordingly, when aligning the crystal structure of the VP40 dimer (pdb: 7jzj) with the VP40 structure obtained from the MD simulations, the membrane- proximal loops and short alpha-helices were mismatched while the core of the monomer aligned well (Fig 3B, highlighted in yellow, Fig EV5A and B). The second monomer displayed similar secondary structures, which were tilted with respect to the crystal structure by 17°, causing a mismatch when compared to the crystal structure (Fig 3B, blue monomer, Fig EV5E). Next, we simulated VP40–membrane interactions at pH 4.5 and observed a significantly decreased affinity toward the membrane, consistent with our tomography data (Fig 3E). The free energy pro- file determined from the MD simulations (Fig 3E) revealed an energy minimum that was 4.1 kBT weaker at low pH compared to pH 7.4. However, binding was not completely diminished since 10% of the phosphatidylserines used in the simulation are still charged (Tsui et al, 1986), and the membrane modeled here containing high levels of phosphatidylserine can still engage in elec- trostatic interactions. To identify which lipids are enriched in the VLP membrane and are thus likely involved in VP40 binding, we then determined the VLP lipid composition by mass spectrometry (Figs 3F and EV5, Table EV1). As expected for EBOV VLPs budding from microdomains in the plasma membrane (Panchal et al, 2003; Stahelin, 2014), the EBOV VLP envelope was rich in phosphatidyl- serine and cholesterol, phosphatidylcholine, and sphingomyelin (9, 39, 25, and 9%, respectively). Collectively, these data argue for low pH-mediated VP40 disassembly through neutralization of negatively charged phospholipids in the viral envelope and highlights electro- static interactions as the main driving forces of the VP40-membrane binding (Fig EV5F). These interactions are driven by basic patches of amino acids, which are highly conserved across all EBOV species further emphasizing their importance in adaptable (Fig EV5H), membrane binding. 4 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal Protons passively equilibrate across the EBOV membrane We next assessed the acidification kinetics to elucidate the mecha- nism of ion permeability across the viral membrane. EBOV VLPs composed of GP, VP40, and the pH-sensitive GFP variant pHluorin (Miesenbo¨ck et al, 1998) N-terminally fused to VP40 (pHluorin- VP40) were prepared to monitor pH changes in VLP lumina upon altering the pH of the surrounding buffer (Fig 4A). Pleomorphic VLPs containing VP40 in excess over VP40-pHluorin, including filamentous and spherical particles, were imaged by time-lapse microscopy (Fig 4B and C). At neutral pH, the VLPs showed a fluorescent signal, which gradually decayed over several minutes after lowering the Figure 2. (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 5 of 20 The EMBO Journal Sophie L Winter et al ◀ Figure 2. The VP40 matrix in EBOV VLPs disassembles at low pH. Slices of a tomogram showing a filamentous EBOV VLP composed of VP40 and GP at neutral pH (n = 37). A B, C Transverse cross-section and longitudinal near-to-surface slices of the tomogram shown in (A) displaying the densities for the outer and inner membrane mono- D E F G–J K L layer and an additional density of the VP40 matrix apparent as striations in (C) (white arrows). Longitudinal cross-section highlighting the VP40 densities adjacent to the membrane (white arrows). Subtomogram average of the VP40 matrix in EBOV VLPs composed of GP and VP40. A density representing a single VP40 dimer is indicated by a black dashed rectangle. Crystal structure of the VP40 dimer (pdb: 7jzj) fitted into the subtomogram average with the C-terminal domains (CTDs) indicated by arrows. Slices of a tomogram showing a filamentous EBOV VLP composed of VP40 and GP after incubation at low pH (n = 18). White arrows in (H) and (J) highlight areas adjacent to the VLP membrane devoid of protein densities in contrast to corresponding slices of VLPs at neutral pH. Slices of tomograms showing filamentous VLPs composed of VP40 after incubation at neutral (n = 22) and low pH (n = 8), respectively. Line density profiles determined adjacent to the inner membrane monolayer of VLPs incubated at neutral (blue) and low pH (magenta). At neutral pH, the VP40 matrix detectable as regular densities in (D) have a 5–6 nm pitch. Data information: Scale bars: (A–C), (G–I), and (K): 50 nm, (E): 2.5 nm, (D), (J): 20 nm. Source data are available online for this figure. external pH (Fig 4D). In contrast, when adding the detergent Triton X-100 (T-X100) before imaging to permeabilize the VLP membrane, the signal decayed to background fluorescence within the first 15 s (Fig 4D), indicating that protonation of pHluorin was slowed down by the membrane of the VLPs. To calculate the acidification kinetics of the VLPs’ lumen, we determined pH levels in the VLPs (Fig 4E) by correlating the pHluorin fluorescence intensity to pH using a calibra- tion curve (Appendix Fig S4A). We found that the luminal pH of fila- mentous VLPs decreased from 7.4 to 6.4 after 6 min, while for the spherical particles, this decay had already occurred after 3.5 min (Fig 4E). We next calculated the membrane proton permeability coef- ficient, Pm, based on the geometry of the VLPs measured by cryo-ET (Fig 2) and the fluorescence decay times (Appendix Fig S4B). Fila- mentous VLPs had a permeability coefficient of 1.2 (cid:1) 0.2 (cid:3)A/s, whereas the membrane of spherical VLPs was significantly more per- meable with a permeability coefficient of 33 (cid:1) 9 (cid:3)A/s. To compare the membrane permeability of the VLPs with the permeability of membranes containing a well-characterized viral ion channel, we used HEK 293T cells expressing VP40-pHluorin and the influenza virus ion channel M2. In line with previous measurements (Deamer & Bramhall, 1986; Deamer, 1987), the plasma membrane in cells displayed a permeability coefficient of 345 (cid:1) 71 (cid:3)A/s (n = 44) in the absence of M2. As expected, the permeability increased with increasing amounts of M2 present in the plasma membrane up to 1940 (cid:1) 562 (cid:3)A/s (n = 26) when M2 and VP40 were transfected at a 1:0.2 molar ratio (Fig 4E and F). Compared to the envelope of fila- mentous EBOV VLPs, the plasma membrane was more permeable to protons already in the absence of M2. Disassembly of the VP40 matrix is critical for membrane fusion Collectively, our experimental data and MD simulations indicate that low pH drives VP40 matrix disassembly and detachment from the viral envelope. We speculated that this influences the GP- mediated membrane fusion between the EBOV envelope and the endosomal membrane. To test this hypothesis, we numerically sim- ulated the membrane fusion pathway in the presence of the VP40 matrix and estimated the magnitude of the two major energy bar- riers to membrane fusion: stalk and fusion pore formation (Jahn & Grubmu¨ ller, 2002; Chernomordik et al, 2006; Chernomordik & Kozlov, 2008; Harrison, 2008). We applied a continuum approach to model the lipid membrane with the commonly used framework of the theory of splay-tilt deformations (Helfrich, 1973; Hamm & Kozlov, 2000) and the VP40 matrix layer as a uniform thin shell that interacts continuously with the virus envelope but can also locally detach from the membrane near the stalk and diaphragm rim (Fig 5A). Based on the VP40-membrane binding energy obtained from the MD simulations at pH 7.4 (Fig 3E) and the density of VP40 dimers on the viral envelope determined from the subtomogram average (Fig 2E), we estimated the VP40 matrix interaction energy density to be 0.38 (cid:1) 0.02 kBT/nm2 (with a dimer density of 0.036 −2 and free binding energy of 10.77 (cid:1) 0.47 kBT). Consistently nm with our cryo-ET data, we assume that the interaction energy den- sity vanishes at low pH due to the VP40 matrix disassembly. Impor- tantly, our calculations showed that the initiation of viral membrane fusion is more favorable after VP40 disassembly. The calculated stalk formation energy barrier drops from 89–79 kBT to 65 kBT due to the weakening of low pH, depending on the matrix layer rigidity (Fig 5B). Hence, the intact VP40 matrix can prevent or slow down hemifusion stalk formation, which is the first step of membrane fusion. the VP40–lipid interactions at Interestingly, our model predicts that fusion pore formation, which occurs after stalk formation, is facilitated in the presence of the assembled VP40 matrix because of increased stress in the hemi- fusion diaphragm. Simulation data showed that the interaction energy density and the rigidity of the VP40 matrix modulate the shape of the hemifusion diaphragm structure (Fig 5C), which deter- mines the energy barrier of pore formation. Strong interactions between the lipids and VP40 matrix (Fig 5C, “tight-bound” configu- thereby inhibiting ration) stabilize the hemifusion diaphragm, fusion pore formation. Conversely, in case of a weakly interacting or stiff VP40 matrix (Fig 5C, “loose” configuration), the hemifusion diaphragm is more unstable, which results in a lower energy barrier for fusion pore formation (Fig 5D). Our model showed that the mini- mal pore opening energy is at the phase boundary between “loose” and “tight-bound” configurations, where diaphragm stress is maxi- mal (Fig 5D). Given the VP40-membrane binding energy and VP40 dimer envelope density found in the MD simulations (Fig 3E), we could show that the VP40 matrix and the membrane preferably adopt the “loose” configuration at both neutral and acidic pH. Therefore, contrary to the stalk formation energy barrier, which is decreased upon VP40 matrix disassembly, the pore formation energy barrier is lower in the presence of the VP40 matrix layer by 16–33 kBT (Fig 5D), depending on the matrix layer rigidity (Fig 5E, 6 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal Figure 3. (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 7 of 20 The EMBO Journal Sophie L Winter et al ◀ Figure 3. VP40–lipid interactions at neutral and low pH. A Initial, intermediate, and final state of the VP40-membrane interaction pathway sampled with unbiased all-atom molecular dynamics (MD) simulations. The simu- lated membrane is composed of 30% phosphatidylcholine, 40% cholesterol, and 30% phosphatidylserine. VP40 is randomly oriented towards the membrane in the initial state. Lipid interactions are first mediated via one C-terminal domain (CTD1) (intermediate state) before the second CTD (CTD2) is ultimately pulled towards to membrane. B MD simulation frame of the VP40-membrane-bound state, with a rotation angle of VP40 monomers along the N-terminal-dimerization domain (Appendix Fig S1C and D) of 1°, fitted into the subtomogram average shown in (Fig 2E). Missing C-terminal residues in the crystal structure of the VP40 dimer (pdb: 7jzj) were computa- tionally modeled (magenta). VP40 conformational changes upon lipid-interaction resulted in a displacement of secondary structures (steel blue), while the core of the protein remained unaltered in comparison to the crystal structure (yellow). C The area highlighted in (B) shows a flexible, C-terminal loop (green) containing the residues K224 and K225 that interact with phosphatidylserines in the inner mem- brane monolayer. D Area highlighted in the rotated MD simulation in (B) showing a flexible loop (residues T58-G67) protruding from the subtomogram average. E Free energy profiles of VP40–lipid interactions at pH 7.4 and pH 4.5 determined from MD simulations. The plot shows free energy (in kBT) at increasing membrane- VP40 distances (nm) with indicated three states shown in (A). F EBOV VLP lipid composition showing highly abundant lipids determined by mass spectrometry in mol%. Lipid abbreviations: phosphatidylcholine (PC), phosphatidylserine (PS), phosphatidylethanolamine (PE), phosphatidylinositol (PI), lyso-phosphatidylcholine (LPC), sphingomyelin (SM). Prefix “a” indicates acyl-linked glycerophospholipids, prefix “e” indicates ether-linked (plasmanyl) or the presence of one odd and one even chain fatty acyl. Bars represent mean and error bars repre- sent standard error of the mean (red); n = 3 biological replicates. Source data are available online for this figure. To validate Appendix Fig S5A). However, it is important to note that hemifusion precedes pore formation in the membrane fusion pathway. Since the disassembly of the VP40 matrix is required for hemifusion and hence for the initiation of membrane fusion, it determines the out- come of the membrane fusion pathway. this experimentally, we theoretical model performed beta-lactamase entry assays using VLPs (Jones & Padilla-Parra, 2016). EBOV membrane fusion requires proteolytic cleavage of GP by low pH-activated cathepsin proteases (Brecher et al, 2012) and subsequent binding of the cleaved GP1 subunit to the endosomal receptor NPC1. To circumvent the need for low pH to activate cathepsin proteases, we substituted cathepsins with thermolysins which are active at neutral pH (Stauffer, 1971), thereby decoupling the low pH requirement from proteolytic GP processing. EBOV VLPs composed of GP, VP40, and beta- lactamase N-terminally fused to VP40 (BlaM-VP40) were purified and subjected to thermolysin treatment followed by incubation at neutral or low pH. We then incubated target Huh7 cells with the pretreated VLPs, loaded the cells with a fluorescent BlaM sub- strate, and assessed virus entry by FACS (Appendix Fig S5). Ther- molysin treatment significantly enhanced host cell entry, whereas the enhancement of entry by low pH treatment alone was less pronounced and not statistically significant (Fig 5F). To determine whether thermolysin-treated VLPs still require low pH for entry, we challenged host cells treated with ammonium chloride, which blocks endosomal acidification. Strikingly, entry of VLPs treated with thermolysin was completely inhibited by ammonium chlo- ride, which is in line with a previous study conducted with bafi- lomycin to inhibit endosomal acidification (Mingo et al, 2015b). This suggests that GP processing alone is insufficient to enable entry. Conversely, low pH treated VLPs were also unable to enter target cells treated with ammonium chloride since impaired endo- somal acidification prevents the activation of cathepsin proteases and hence GP priming. Combined thermolysin- and low pH- treatment of VLPs in vitro rescued entry into host cells with inhibited endosomal acidification, albeit to a lesser extent com- pared to entry into untreated cells. Since thermolysin-treated EBOV VLPs efficiently enter untreated host cells at neutral and low pH, we further conclude that low pH alone does not induce the GP2 post-fusion conformation, which would inhibit virus entry. Together, this suggests a role of low endosomal pH beyond proteolytic processing of EBOV GP, likely for the disassembly of the VP40 matrix. Overall, these data show that VP40 matrix integ- rity modulates GP-mediated membrane fusion, strongly supporting the notion that VP40 disassembly is required for and precedes membrane fusion. Discussion Ebola viruses form remarkably long, filamentous virions that enter the cytoplasm by fusion with late endosomal membranes. Similar to other enveloped viruses, the shape and stability of EBOVs are determined by a matrix layer forming a flexible scaffold under- neath the viral envelope, which is indispensable for particle forma- tion and protects the encapsidated genome during transmission. Here, we investigate the molecular architecture of the EBOV VP40 matrix in Ebola virions during host cell entry to elucidate whether and how it is released from the viral envelope to allow virion uncoating. Using in situ cryo-electron tomography, we directly visualize EBOVs entering host cells via the endosomal route. Virions inside endosomal compartments exclusively exhibited disassembled VP40 matrices and some had engulfed endosomal vesicles, suggesting that the membranes of these virions are suffi- ciently flexible to engage in membrane fusion (Fig 1). Considering that the nucleocapsids in all endosomal EBOVs were condensed, we propose that VP40 disassembly precedes membrane fusion, while nucleocapsid integrity is maintained until cytoplasmic entry is concluded. The VP40 aggregation surrounding the nucleocapsid may be involved in engaging cellular factors required to pull nucleocapsids out of the fusion site as has recently been suggested for influenza A virus, whose disassembled M1 matrix layer recruits the aggresome machinery by mimicking misfolded proteins (Bane- rjee et al, 2014). Supported by our functional data and computa- tional simulations, we propose that EBOV uncoating occurs in a cascade-like fashion. Tightly regulated by pH, uncoating starts with the disassembly of the VP40 layer, followed by GP-driven mem- brane fusion and release of the compact nucleocapsid into the 8 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal Figure 4. Time-lapse microscopy of EBOV VLPs at different pH. A Schematic showing the VLP membrane and pHluorin-VP40 facing the luminal side of the VLPs. Upon protonation, pHluorin loses its fluorescence properties and serves as a proxy for proton diffusion across the membrane. B Overview confocal fluorescence microscopy image showing pleomorphic pHluorin-labeled VLPs composed of VP40, pHluorin-VP40 (ratio 10:1) and GP. C Magnified images of representative VLPs acquired during time-lapse microscopy at neutral pH and after acidification to approximately pH 5. Frames are exemplarily shown at 0, 1, 3, 5 and 10 min after lowering the external pH. D Plot showing the mean relative fluorescence intensities and standard deviation of VLPs imaged at neutral pH, low pH and in the presence of T-X100 at low pH over time. Data was obtained from 3 independent VLP preparations. Number of technical replicates: n = 42 (pH 7.4); n = 19 (pH 4.5 and pH 5 + T-X100). E Plot showing the drop of pH inside VLPs over time after lowering the pH of the surrounding buffer to 5. The dots represent the mean values, and the dashed lines are the theoretical fit to Equation 3. F Membrane permeability of VLPs (red) and HEK 293T cells expressing different ratios of VP40 and M2 (blue). The mean permeability obtained from 3 independent VLP preparations and 3 biological replicates of cell transfections is displayed on a logarithmic scale, error bars represent standard errors of the mean. Permeability coefficients: filamentous VLPs 1.2 (cid:1) 0.2(cid:3)A/s (n = 154), spherical VLPs 33 (cid:1) 9(cid:3)A/s (n = 66), cells expressing no M2 345 (cid:1) 71(cid:3)A/s (n = 44), cells expressing pHluorin-VP40 and M2 at 1:0.002 molar ratio 409 (cid:1) 85(cid:3)A/s (n = 30), cells expressing pHluorin-VP40 and M2 at 1:0.02 molar ratio 683 (cid:1) 263(cid:3)A/s (n = 28) and cells expressing pHluorin-VP40 and M2 at 1:0.2 molar ratio 1940 (cid:1) 562(cid:3)A/s (n = 26). Data information: Scale bars: B: 10 μm, C: 2 μm. Source data are available online for this figure. cytoplasm. It remains to be elucidated when and how the nucleo- capsid undergoes de-condensation to allow viral genome replica- tion and transcription. The organization of VP40 proteins within the VP40 matrix, including their oligomeric state and orientation of C-termini towards the membrane, has long been subject of debate (Scianimanico (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 9 of 20 The EMBO Journal Sophie L Winter et al et al, 2000; Adu-Gyamfi et al, 2013; Soni & Stahelin, 2014; Stahelin, 2014; Del Vecchio et al, 2018; Pavadai et al, 2018). While the structure of VP40 in solution was revealed as a dimer (Bornholdt et al, 2013), structures of VP40 in the context of lipid environments were proposed based on purified VP40 proteins either truncated or characterized in the presence of lipid mimics. These revealed VP40 Figure 5. 10 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal ◀ Figure 5. Membrane fusion dynamics in the presence and absence of the VP40 layer. A Simulation result of a hemifusion stalk in the presence of a rigid matrix layer (VP40). The blue and red lines represent the averaged lipid director of the distal and proximal monolayers, respectively. The VP40 matrix layer is represented by the continuous thick brown strip. Parameters used in panels (A–D) for the lipid membrane: lipid monolayer bending rigidity 10 kBT, tilt decay length 1.5 nm, saddle splay modulus to bending modulus ratio −5 kBT, monolayer spontaneous curvature −0.22 nm−1, and monolayer width 1.5 nm. VP40 matrix layer: width 4 nm, Poisson’s ratio 0.5, and membrane mid-plane to VP40 mid-plane optimal distance 4 nm. In panel (A) the matrix layer Young’s modulus is 11.2 MPa, and the interaction energy density is 0.2 kBT/nm2. B Stalk formation energy as a function of VP40-membrane interaction energy. The stalk energy for non-interacting VP40 matrix (u0 ¼ 0Þ is 65 kBT. VP40 matrix layer Young’s modulus legend – red 5.6 MPa, blue 11.2 MPa, black 16.8 MPa, and orange is infinitely rigid. The bending rigidity ratio between the VP40 matrix layer and lipid monolayer are 1, 2, 3, and infinity, respectively. The line represents an infinitely rigid layer. C Hemifusion diaphragm configurations phase-diagram – above dotted red line: tight-bound solution and loose configuration below. The inserts are simulation results with layer Young’s modulus of 11.2 MPa. The interaction energy density is 0.2 kBT/nm2 in loose configuration and 0.85 kBT/nm2 in the tight-bound configuration. The scale bar is 10 nm. D Fusion pore formation energy as a function of VP40-membrane interaction energy. The discontinuity in the energy is located at the phase line between configurations (see C). The change in pore formation energy, ΔEpore is defined as the difference between fusion-pore formation energy at interaction energy density 0.2 kBT/nm2 (the value found using MD simulations) to the matrix-free case (no interaction energy). (A–D) Dotted lines serve as a guide to the eye. E Illustration of the effect of the matrix layer on the fusion pathway and the fusion intermediates in the absence of the matrix layer. As a result of the presence of the matrix layer, the stalk formation energy barrier increases while the pore formation energy barrier decreases and the hemifusion diaphragm intermediate is less stable. F Quantification of FACS data showing EBOV VLP entry as measured by a fluorescence shift of infected cells from emission at 510 nm (no entry) to 450 nm (entry). VLPs were treated prior to infection as indicated on the x-axis, with control: uninfected control cells, −: no treatment, T: thermolysin-treatment at neutral pH, Lp: low pH treatment. Target cells were treated with media or ammonium chloride (NH4Cl), n = 3 with 10,000 cells measured per sample. Source data are available online for this figure. hexamers as the building blocks of the VP40 matrix (Scianimanico et al, 2000; Nguyen et al, 2005; Bornholdt et al, 2013), in which the C-termini alternatingly face the viral membrane. Recently published data (Wan et al, 2020) and our subtomogram averaging (Fig 2) show that the VP40 matrix within VLPs is instead composed of line- arly arranged dimers, in which all C-termini are facing the VLP membrane and thus collectively contribute to the electrostatic inter- actions. Importantly, a combination of MD simulations and subto- mogram averaging allowed us to refine the structure of the VP40 dimer interacting with the membrane and to map the basic patch of lysine residues to flexible loops that extend into the inner membrane monolayer (Jeevan et al, 2017; Fig 3). Additionally, our MD simula- tions reveal lipid-induced conformational changes of the VP40 dimer that complement our subtomogram averaging data. The rotation of VP40 monomers along the N-terminal-dimerization domain is in line with the structural data and emphasizes the modularity of the VP40 dimer, which may contribute to the flexibility of the large filamen- tous particles (Booth et al, 2013; Wan et al, 2020). Using VLPs of minimal protein composition (VP40 and GP and VP40 alone), we show that VP40-disassembly, i.e. the detachment of the matrix from the viral envelope is triggered by low endosomal pH (Fig 2). This indicates that VP40 disassembly does not depend on structural changes of other viral proteins, including GP, and is driven solely by the acidic environment. Furthermore, we deduced VP40–lipid interaction strengths from the MD simulations, which are strongly diminished at pH 4.5 and thus support a dissociation of VP40 from the membrane in endosomal environments. Our data demonstrate that VP40 detachment from the membrane is driven by the neutralization of negatively charged phospholipids at endosomal pH. VP40 detachment from viral envelope is caused by a disruption of electrostatic interactions between VP40 and negatively charged lipids in the viral envelope, which have experimentally been demon- strated and attributed to a basic patch of lysine residues decorating the VP40 C-termini (Ruigrok et al, 2000; Bornholdt et al, 2013; Del Vecchio et al, 2018; Lee et al, 2021). Considering that matrix protein assembly of other RNA viruses relies on electrostatic interactions with negatively charged lipids (Norris et al, 2022), we propose that pH-mediated matrix disassembly is a general mechanism critical for viral uncoating. Notably, pH-driven structural remodeling of viruses has so far only been shown and extensively studied for influenza A virus (Fon- tana & Steven, 2013), which is known to encode the viral ion chan- nel M2 (reviewed here (Manzoor et al, 2017)). Since EBOVs do not encode a dedicated ion channel, we determined the permeability of the EBOV VLP membrane in comparison to the plasma membrane in the absence and presence of the M2 ion channel. We show that the proton permeability of the VLP membrane depends on particle morphology and is markedly lower in filamentous VLPs when com- pared to spherical VLPs (Fig 4). Since spherical EBOV virions are predominantly released at very late infection time-points (4 days post infection) and are less infectious than filamentous particles (Welsch et al, 2010), it is plausible that their membrane properties including proton permeability result from improper particle forma- tion due to cell exhaustion. The higher proton permeability of the plasma membrane already in the absence of M2 likely results from its complex composition comprising host cell ion channels (DeCoursey, 2008). While the membrane permeability of filamen- tous VLPs is low compared to values reported in the literature for protein-free liposomes (Deamer & Bramhall, 1986), pH equilibration inside filamentous virions is fast due to their small radius and takes place within minutes. This suggests that acidification occurs rapidly after EBOV uptake into late endosomes and is not rate-limiting dur- ing virus entry into host cells, in agreement with a previous report (Mingo et al, 2015a). Taken together, we show that protons diffuse passively across the EBOV envelope, independent of a dedicated ion channel. It remains to be elucidated whether virion acidification also occurs by passive diffusion in other late-penetrating viruses lacking a dedicated ion channel. We further show that the energy barriers of both the hemifusion stalk and fusion pore formation strongly depend on the VP40 matrix rigidity (Fig 5). The assembled VP40 matrix inhibits stalk formation, which precedes fusion pore formation during membrane fusion, (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 11 of 20 The EMBO Journal Sophie L Winter et al arguing for VP40 disassembly as a critical step required for mem- brane fusion and highlighting the role of the matrix as a modulator of membrane fusion. Together, the findings presented here reveal a yet unknown role of viral matrix proteins during viral entry and uncoating as membrane fusion modulators. We propose that low- pH-driven matrix protein disassembly is decisive for membrane fusion of other enveloped late-penetrating viruses, making the pro- cess a promising target for interventions by development of virus matrix-specific weak base inhibitors. transferred to the BSL4 laboratory after 4–5 h. Cells were infected with unpurified EBOVs at an MOI of 0.1 for either 22 h or 48 h before chemical fixation for 24 h with 4% paraformaldehyde and 0.1% glutaraldehyde in PHEM buffer (60 mM PIPES, 25 mM HEPES, 2 mM MgCl2, 10 mM EGTA, pH 6.9). After transfer of the samples out of BSL4, the grids were kept in PHEM buffer and plunge-frozen within three days. Sample preparation for cryo-electron tomography Materials and Methods Cell lines and EBOV VLP production Cell lines used in this work include HEK 293T cells for Ebola virus- like particle (VLP) production and Huh7 cells as target cells to assess VLP and EBOV entry. HEK293T (293T ECACC, 1202201) cells were purchased from Sigma-Aldrich. Huh7 cells were kindly pro- vided by Prof. Ralf Bartenschlager (Heidelberg University Hospital). Both cell lines were maintained in DMEM media (ThermoFisher Sci- entific) supplemented with 10% (v/v) FBS and 100 U/ml penicillin– streptomycin (ThermoFisher Scientific) at 37°C, 5% CO2. All cells were tested for Mycoplasma contamination every 3 months. Ebola virus VLPs were produced by transfecting HEK 293T cells with equal amounts of pCAGGS plasmids encoding EBOV GP, VP40, NP, VP35, and VP24 (species Zaire ebolavirus, Mayinga strain). Supernatants of transfected cells were harvested 48 h post transfec- tion and clarified by centrifugation at 398 g for 10 min and 2,168 g for 15 min (JA-10 rotor, Beckmann). Clarified supernatants were passed through a 30% sucrose cushion in HNE buffer (10 mM HEPES, 100 mM NaCl, 1 mM EDTA, pH 7.4) by centrifugation for 2.5 h at 15,960 g (SW32 Ti rotor, Optima L-90 K ultracentrifuge, Beckmann) at 4°C. Pellets were resuspended in HNE buffer and centrifuged at 12,817 g (TLA 120.2 rotor, Optima TLX ultracentrifuge (Beckmann)) at 4°C for 10 min to remove residual media and sucrose. Final pellets were resuspended in HNE buffer and protein concentrations were mea- sured using the Pierce BCA Protein Assay Kit (ThermoFisher Scien- tific) according to the manual provided by the manufacturer. To produce reporter VLPs, pHluorin was N-terminally cloned to VP40 and beta-lactamase-VP40 (BlaM-VP40) was a kind gift from Dr. Kartik Chandran. Reporter VLPs were produced by transfecting EBOV GP, VP40, and pHluorin-VP40 or BlaM-VP40 in a 10:10:1 ratio and purified as described above. Production of Ebola virus and infection of Huh7 cells Ebola virus (species Zaire ebolavirus, strain Mayinga) was produced in VeroE6 cells in the BSL4 facility at the Friedrich-Loeffler Institut (Insel Riems, Greifswald), following approved standard operating procedures. 5 days post-infection, supernatants of infected cells were harvested and purified as described for the VLPs above and then fixed by adding paraformaldehyde and glutaraldehyde in HNE buffer for a final concentration of 4% and 0.1%, respectively. For structural characterization of EBOV-infected cells, Huh7 cells were seeded on 200 mesh Au Quantifoil™ SiO2 R1.2/20 EM grids placed on 3D-printed grid holders in a 96-well plate (F€aßler et al, 2020). 0.0075 × 106 cells were seeded and the plates were Ebola virus VLPs and chemically fixed EBOV were plunge-frozen as previously described (Winter & Chlanda, 2021). Briefly, VLPs were diluted to approximately 10–20 ng/μl, mixed with 10 nm protein A- coated colloidal gold (Aurion), and applied onto glow-discharged EM grids (200 mesh, R 2/1, Quantifoil) prior to plunge-freezing with a Leica EM GP2 automatic plunge-freezer. Chemically fixed EBOV-infected Huh7 cells on EM grids were vit- rified using the GP2 plunge freezer (Leica) at an ethane temperature of −183°C, chamber temperature of 25°C, and 95% humidity. 5 μl PHEM buffer was added to the grids before blotting them from the back with a Whatman Type 1 paper for 3 s. For cryo-FIB milling, the grids were clipped into specifically designed AutoGrids™ (Thermo- Fisher Scientific). Cryo-FIB milling was performed as previously described (Klein et al, 2020b) using an Aquilos dual-beam FIB-SEM microscope (ThermoFisher Scientific). Briefly, cells were selected for milling and coated with an organometallic platinum layer for 5 s before milling in four successive steps using a gallium-ion beam at acceler- ation voltage 30 eV. Resulting lamellae were 200–250 nm thick. Tomogram acquisition, reconstruction, and volume rendering Cryo-ET of VLPs and lamellae of EBOV-infected Huh7 cells were performed as previously described (Klein et al, 2020a). Briefly, data were collected on a Titan Krios Transmission Electron Microscope (TEM, ThermoFisher Scientific) at Heidelberg University operated at 300 keV and equipped with a BioQuantum LS energy filter with a slit width of 20 eV and K3 direct electron detector (Gatan). Tilt series were acquired at 33,000 magnification (pixel size 2.671 (cid:3)A) using a dose- symmetric acquisition scheme (Hagen et al, 2017) with an electron /(cid:3)A2 per projection with tilt ranges from 60° − dose of approximately 3 e to −60° in 3° increments using SerialEM (Mastronarde, 2005) and a scripted dose-symmetric tilt-scheme (Hagen et al, 2017). ® For subtomogram averaging, tomograms were acquired at EMBL Heidelberg using a Titan Krios TEM (ThermoFisher Scientific) oper- ated at 300 keV and equipped with a Gatan Quantum 967 LS energy filter with a slit width of 20 eV and a Gatan K2xp detector. Tilt series were acquired at 81,000 magnification (pixel size 1.7005 (cid:3)A) at a defocus range of −3 to −1.5 μm using SerialEM (Mastronarde, 2005) and a scripted dose-symmetric tilt scheme (Hagen et al, 2017) from −60° to 60° with 3° steps. Tomograms were reconstructed using the IMOD software pack- age (Mastronarde & Held, 2017). Stacks of tomograms of VLPs were aligned using gold fiducials, and stacks of tomograms acquired on lamellae were aligned using patch tracking. After 3D contrast trans- fer function (CTF) correction and dose filtration implemented in the reconstruction was performed by weighted back- IMOD, projections with a simultaneous iterative reconstruction technique 12 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal (SIRT)-like filter equivalent to 10 iterations. Tomograms used for subtomogram averaging were reconstructed using 2D CTF correc- tion by phase flipping and weighted back- projection without a SIRT-like filter. For visualization, 10 slices of the final tomogram were averaged and low-pass filtered. 3D segmentation was performed using the Amira software and the implemented membrane enhancement filter. Membranes were automatically segmented using the Top-hat tool, and final segmenta- tions were manually refined. Subtomogram averaging Subtomogram averaging of the VP40 matrix was performed using the Dynamo software package (Casta~no-Dıez et al, 2012; Casta~no- Dıez, 2017). Particles were automatically picked using the filament model, and subtomograms were extracted with a cubic side length of 128 voxels from 23 tomograms. A reference template was obtained by iteratively aligning and averaging of 50 subtomograms using a mask permitting alignments only a membrane VP40 layer. The initial average was then used as a template for the final averag- ing of approximately 7,800 particles. Molecular dynamics simulations We used the truncated (residues 45–311) crystallographic structure of the VP40 dimer deposited by Norris et al (2022) (pdb: 7JZJ; Wan et al, 2020) for atomistic molecular dynamics simulations. The miss- ing CTD loops were modeled using the GalxyFill software (Coutsias et al, 2004) within the CHARMM-GUI web server (Jo et al, 2008). The protonation states of the proteins at pH 7.4 and 4.5 were calcu- lated through the PROPKA web server (Søndergaard et al, 2011), which indicated a change in the protonation state at pH 4.5 for the following residues: E76, E325, H61, H124, H210, H269, and H315. Importantly, these residues are located away from the interaction interface of VP40 with the membrane and their protonation accord- ingly does not influence membrane binding. However, protonation of these residues may contribute to conformational changes that facilitate the VP40 disassembly. First, the proteins were simulated in water with a 0.1 M NaCl for 1 μs. Next, the final structures were placed at a distance of 2 nm from a previously built model mem- brane surface containing POPC:POPS:CHOL (30:30:40) at 10 differ- ent random orientations. The model membrane was made using the CHARMM-GUI membrane builder (Jo et al, 2009). Since the percent- age of POPS charged molecules at pH 4.5 is 10% (Tsui et al, 1986), we modeled the membrane at pH 4.5 by randomly replacing 90% of POPS molecules with its protonated model (POPSH). Then, each of the 10 repeats was solvated with 40,913 water molecules and 0.1 M NaCl. Next, charges were neutralized by adding or removing the + needed amount of Na -ions. Finally, each system was simu- lated for 1 μs under NpT conditions. Four out of 10 simulations, at both pH conditions, showed VP40 dimer binding to the membrane with the experimentally known binding residues, K224, K225, K274, and K275. These simulations were used for the analysis. For the pro- duction run, we employed the Parrinello-Rahman barostat (Parri- nello & Rahman, 1981) with a semi-isotropic pressure coupling scheme and a time constant set to 5.0 ps to maintain the pressure constant. The pressure was set to 1.0 bar and the isothermal com- −1. The temperature was maintained at pressibility to 4.5 × 10 −5 bar - or CL − 310 K using the Nose-Hoover thermostat (Hoover, 1985) with a time constant of 1.0 ps. Electrostatic interactions were handled using the PME method (Essmann et al, 1995). The cutoff length of 1.2 nm was used for electrostatic (real space component) and van der Waals interactions. Hydrogen bonds were constrained using the LINCS algorithm (Hess et al, 1997). Finally, periodic boundary conditions were applied in all directions. The simulations were carried out using an integration time step of 2 fs with coordinates saved every 100 ps. All simulations have been carried out with the GROMACS- 2021 software (Abraham et al, 2015). Protein, lipids, and salt ions were described using the CHARMM36m force field (Pastor & MacK- erell, 2011; Huang & Mackerell, 2013; Huang et al, 2016). For water, we used the TIP3 model (Jorgensen et al, 1983). All pictures, snap- shots, and movies were rendered with the visual molecular dynam- ics (VMD) software (Humphrey et al, 1996). Free energy calculation The potential of mean force (PMF) for the VP40 dimer binding on a model membrane surface was calculated using an atomistic res- olution, employing the umbrella sampling protocol (Torrie & Valleau, 1974, 1977). The initial configuration for each umbrella window was taken directly from unbiased MD simulations. The center of the mass distance between the VP40 dimer and the phosphate atoms of one leaflet was used as the reaction coordinate. A total of 49 windows, 0.1 nm spaced, were generated and simulated with a harmonic restraint force constant of 2,000 kJ/mol/nm2 for 200 ns. The first 100 ns of the simulations were considered as an equilibration phase and discarded from the actual free energy calculation. The free energy profiles were reconstructed using the weighted histogram analysis method (Hub et al, 2010). The statistical error was estimated with 200 bootstrap analyses. Dihedral angle calculation The rotation angle of VP40 monomers along the NTD-dimerization domain is defined as the angle between the plane containing the vector connecting alpha carbon atoms of L75monomer1 and T112monomer1 and the vector connecting atoms T112monomer1 and T112monomer2 and the plane containing this second vector and the vector connecting atoms T112monomer2 and L75monomer2 as explained in Fig EV5C. The angle has been calculated rerunning the simula- tions trajectory with a GROMACS version patched with the open- source, (Bonomi et al, 2019), version 2.4 (Tribello et al, 2014). The angle measure- ment in water has been calculated using all simulation frames. For the angle calculation upon the binding to the membrane, the last 100 ns of the four simulations showing VP40-membrane interaction via the experimentally known critical residues (i.e., K224, K225, K274, and K275) have been used. community-developed PLUMED library Regardless of pH, VP40 monomers within the dimer are flexible with a rotation angle, defined as the torsional angle around the alpha carbons of residue T112 of the two monomers (Fig EV5), oscillating around 1° (SD 9.5) in water, which is 17° smaller of the one measured for the crystallographic structure (pdb: 7jzj; Fig EV5). VP40 dimer flexibility is not constrained upon binding to the mem- brane; however, after binding to the bilayer, the angle distribution (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 13 of 20 The EMBO Journal Sophie L Winter et al was significantly (P ≤ 0.0001) shifted to a value of 3.7° (SD 8.1) and 4.5° (SD 10.7) at pH 7.4 and 4.5, respectively. Sequence alignment Lipidomics of EBOV VLPs Ebola virus VLPs composed of GP, VP40, and the nucleocapsid pro- teins NP, VP24, and VP35 were produced from HEK 293T cells and purified as described above. They were used at a final protein con- centration of 880 ng/μl for lipidomics analysis. VLPs were subjected to lipid extractions using an acidic liquid–liquid extraction method (Blight & Dyer, 1959) as described in Malek et al (2021). In order to ensure that similar amounts of lipids were extracted, a test extraction was performed to determine the concentration of PC as a bulk membrane lipid. Quantification was achieved by adding 1–3 internal lipid standards for each lipid class, with the standards resembling the structure of the endogenous lipid species. Of note, sample volumes were adjusted to ensure that all lipid standard to lipid species ratios were in a linear range of quantification. Typi- cally, the standard to species ratios were within a range of > 0.1 to < 10. Following this approach, a relative quantification of lipid species was performed. Lipid standards were added prior to extrac- tions, using a master mix consisting of 50 pmol phosphatidylcho- line (PC, 13:0/13:0, 14:0/14:0, 20:0/20:0; 21:0/21:0, Avanti Polar Lipids), 50 pmol sphingomyelin (SM), d18:1 with N-acylated 13:0, 17:0, 25:0, semi-synthesized (O¨ zbalci et al, 2013), 100 pmol deuter- ated cholesterol (D7-cholesterol, Cambridge Isotope Laboratory), 30 pmol phosphatidylinositol (PI, 17:0/ 20:4, Avanti Polar Lipids), 25 pmol phosphatidylethanolamine (PE) and 25 pmol phosphatidyl- serine (PS); both 14:1/14:1, 20:1/20:1, 22:1/22:1, semi-synthesized (O¨ zbalci et al, 2013), 25 pmol diacylglycerol (DAG, 17:0/17:0, Larodan), 25 pmol cholesteryl ester (CE, 9:0, 19:0, 24:1, Sigma), and 24 pmol (TAG, LM-6000/D5-17:0,17:1,17:1, Avanti Polar Lipids), 5 pmol ceramide (Cer, d18:1 with N-acylated 14:0, 17:0, 25:0, semi-synthesized (O¨ zbalci et al, 2013) or Cer d18:1/18:0-D3, Matreya) and 5 pmol glucosylceramide (HexCer; d18:1 with N-acylated 14:0, 19:0, 27:0, semi-synthesized or GlcCer d18:1/17:0, Avanti lactosylceramide (Hex2Cer, d18:1 with N-acylated C17 fatty acid), 10 pmol phospha- tidic acid (PA, 17:0/20:4, Avanti Polar Lipids), 10 pmol phosphati- dylglycerol (PG, 14:1/14:1, 20:1/20:1, 22:1/22:1), semi-synthesized (O¨ zbalci et al, 2013), and 5 pmol lysophosphatidylcholine (LPC, 17:1, Avanti Polar Lipids). The final CHCl3 phase was evaporated under a gentle stream of nitrogen at 37°C. Samples were either directly subjected to mass spectrometric analysis or were stored at −20°C prior to analysis, which was typically done within 1–2 days after extraction. Lipid extracts were resuspended in 10 mM ammo- nium acetate in 60 μl methanol. Two μl aliquots of the resuspended lipids were diluted 1:10 in 10 mM ammonium acetate in methanol in 96-well plates (Eppendorf twin tec 96) prior to measurement. For cholesterol determinations, the remaining lipid extract was again evaporated and subjected to acetylation as previously described (Liebisch et al, 2006). Samples were analyzed on an QTRAP 6500+ mass spectrometer (Sciex) with chip-based (HD-D ESI Chip, Advion Biosciences) electrospray infusion and ionization via a Triversa Nanomate (Advion Biosciences). MS settings and scan procedures are listed in Table EV2. Data evaluation was done using LipidView (Sciex) and an in-house-developed software (ShinyLipids). triacylglycerol Lipids), 5 pmol Polar To evaluate the conservation of the basic amino acid patch decorat- ing the VP40 CTDs, we used Clustal Omega from EMBL-EBI (Sievers et al, 2011) and applied a color code to display hydrophobicity (Kyte & Doolittle, 1982). Calibration of pHluorin fluorescence HEK 293T cells were reverse transfected and seeded at a seeding density of 0.02 × 106 cells per well in a 96-well plate. Briefly, trans- fection mixtures were prepared containing a pCAGGS plasmid encoding pHluorin-VP40. After 15 min incubation at RT, HEK 293T cells were trypsinized and mixed with the transfection complexes before seeding on a fibronectin-coated 96 well plate. HNE buffers (10 mM HEPES, 100 mM NaCl, 1 mM EDTA) were prepared and cali- brated to a pH of 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.34, 7.52, and 8. Approximately 20 h post seeding, media was removed, the cells were washed once with HNE buffer at pH 7.34, and incubated for 45 min at 37°C, 5% CO2, in the different HNE buffers. Fluorescence intensities were measured at 488 nm excitation using a Tecan plate reader. To calibrate the fluorescence of pHluorin at different pH, fluorescence was plotted against the pH. Permeability of the HEK 293T cell plasma membrane HEK 293T cells were reverse transfected as described above using pCAGGS plasmids encoding pHluorin-VP40 and influenza virus M2 (A/Udorn/307/1972 (subtype H3N2)) at a molar ratio of 1:0, 1:0.0002, 1:0.002, and 1:0.2. Approximately 20 h post seeding, the media was removed, and cells were washed once with HNE buffer, pH 7.34. The buffer was then exchanged with HNE buffer calibrated to pH 4.5, and fluorescence was immediately measured in 15 s inter- vals using a Tecan plate-reader. Time-lapse microscopy of EBOV VLPs a 63 × oil objective. immersion Time-lapse microscopy was performed using a Leica SP8 confocal microscope with Purified pHluorin-labeled VLPs were added to a glow-discharged μ-Slide 8 well dish (ibidi) at a protein concentration of 10 ng/μl and were allowed to settle for 20 min at RT. They were then imaged using a 488 nm excitation laser and emission at 500–600 nm. Z-stacks were acquired in 15 s intervals for 30 min. To assess acidification kinetics, citric acid was added at a final concentration of 2.6 mM 2 min after starting the data acquisition. To assess acidification kinetics in the absence of the viral membrane, VLPs were incu- bated for 5 min with TX-100 at a final concentration of 0.1% before imaging. Membrane permeability theory We estimate the membrane permeability based on the geometry and pH equilibrium time of the VLPs. The membrane total proton flux, I, is proportional to the area of the VLPs, AVLP, ion con- centration difference between the buffer, CB, and VLPs, CVLP, ΔC tð Þ ¼ CB(cid:3)CVLP tð Þ, and membrane permeability coefficient Pm (Deamer & Bramhall, 1986), 14 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal I ¼ Pm∙AVLP∙ΔC tð Þ (1) It is easy to show that the protons concentration difference decays exponentially with time t, ΔC tð Þ ¼ ΔC0∙e(cid:3)t τ (2) with the decay time constant τ ¼ VVLP , VVLP the volume of the Pm∙AVLP VLPs and ΔC0 the initial concentration difference. The pH level is the logarithm of the protons concentration and can be related to the concentration difference as follows: pHVLP tð Þ ¼ pHBuffer(cid:3)log10 1(cid:3) (cid:1) (cid:3) ΔC0 CB ∙e(cid:3)t τ (3) ΔC0 CB Next, we use a least-squares minimization procedure to fit the measured pH to Equation 3. We find the three minimization param- , and τ. Since the VLPs are either spherical or fila- eters pHBuffer, mentous, we can derive the membrane permeability coefficient Pm ¼ R n∙τ, with R the respective radius and n is either 2 for filamtoues VLPs or 3 for spherical VLPs and cells. The fitted decay times τ are presented in Appendix Fig S4B, and the VLPs radii are found using cryo-ET. In line with previous measurements of EBOV VLPs and virions (Wan et al, 2020), the filamentous VLPs had an average radius of 34 (cid:1) 4.5 nm (n = 90), while spherical particles are more heterogeneous in size, with a radius of 426 (cid:1) 100 nm (n = 12). A similar analysis was also performed on HEK 293T cells. The cells had a round shape. The radius was estimated using fluorescence microscopy to be 17.5 (cid:1) 2.5 nm. Membrane fusion in the presence of a matrix layer The fusion process involves three players—the membrane, the matrix layer, and the fusion proteins. In the following section, we describe the physical properties of these three, their interaction, and the fusion pathway in the presence of a matrix layer. We determine the effect of the matrix layer on fusion rate by calculat- ing the magnitude of the two major energy barriers to membrane fusion (Jahn & Grubmu¨ ller, 2002; Chernomordik & Kozlov, 2008; Fuhrmans et al, 2015) stalk formation and fusion pore expansion in the presence of the matrix layer and compare it to the matrix- free state. Description of the fusion site and the fusion process The fusion reaction starts when the fusion proteins bring the EBOV and endosomal membranes to proximity and drive the merger of only the proximal monolayers. As a result, the membrane and matrix layer deform and locally detach. The fusion site is axially symmetric; its cross-section is illustrated in Fig 4B. The two fusing membranes form a junction in the center of the stalk, with the two membrane mid-planes forming a corner with a 45° angle. As a result, the lipid tails are sheared and splayed to prevent voids in the hydrocarbon tail moiety (Kozlovsky & Kozlov, 2002). The shear and splay magnitude decays within several nanometers from the stalk and smoothly connects to the flat surrounding membranes. After the stalk has formed, it radially expands to an equilibrium radius RD by bringing the two inner monolayers into contact along a joint mid-plane, a state called hemifusion diaphragm (Kozlovsky et al, 2002; Golani et al, 2021). The rim of the diaphragm is the three-way junction between the diaphragm and the two fusing membranes. The lipid monolayer deformations are continuous; therefore, we explicitly require that the magnitude of lipid splay, saddle-splay, and shear are continuous everywhere in our numerical calculations. The matrix layer adheres to the membrane by electro- static interaction, and it can locally detach from it at the vicinity of the stalk and the diaphragm rim junction to avoid substantial defor- mation there. Thus, the matrix is not necessarily parallel to the membrane and can bend independently. The deformation of both the membranes and the matrix layer vanishes at the edge of the fusion site and connects smoothly to a flat surrounding membrane and matrix layer. The membrane fluidity in the lateral direction allows the matrix layer to slide on it freely as the fusion process pro- gresses. The fusion reaction ends by opening and expending a mem- brane pore within the diaphragm, which must involve the detachment of the favorable bounds between the EBOV luminal monolayer and the VP40 matrix layer. The lipid membrane We model the lipid membrane using the well-established theory of lipid tilt, splay, and saddle-splay (Helfrich, 1973; Hamm & Kozlov, 2000). The membrane is composed of two monolayers that share a joint mid-plane. The orientation of the lipids in the two monolayers is independent and is given by the lipid director vector, ! (cid:3) bN, characterizes the shear magni- bn. The lipid tilt vector, t tude and its direction (Hamm & Kozlov, 1998), with bN the midplane normal. The monolayer dividing plane is parallel to the membrane ¼ bn bn∙bN midplane and is located at a distance of δ ¼ δ0 from it, with δ0 the length of the undeformed monolayer tails. The lipid splay and β eb α ¼ rαnβ, where saddle splay are derived from the splay tensor, the sub- and superscripts denote, respectively, the co- and contra- variant components in the local coordinate basis of the monolayer dividing plane (Hamm & Kozlov, 2000). Lipid splay is the trace of the splay tensor eJ ¼ eb β eb eK ¼ det α. The energy density with respect to the flat tilt-free con- figuration associated with these deformations is given by Hamm & Kozlov (2000) and Terzi et al (2019), α α, and lipid saddle-splay is its determinant it (cid:5) κ J 2(cid:3)2 J Jsm (cid:6) f m ¼ 1 2 þ κ eK þ 1 2 !2 κt t : (4) The bending rigidity of the monolayer, κm, has a typical value of 10 kBT (Dimova, 2014), saddle-splay modulus, κm, and tilt modulus, κt, cannot be directly measured and are indirectly estimated. The ratio between saddle-splay modulus and bending rigidity is between −1 and 0 (Templer et al, 1998; Terzi et al, 2019). The ratio between the bend- ﬃﬃﬃﬃﬃﬃﬃﬃﬃ ing rigidity to tilt modulus gives a typical tilt decay length of l ¼ κ=κt , typically between 1 and 2 nm (Terzi & Deserno, 2017). Here we use l ¼ 1:5 nm and κ=κ ¼ (cid:3)0:5. The monolayer spontaneous curvature, Jsm, is the averaged intrinsic curvature of its constituting lipids, p ∑i¼M i¼1 ζiφ i : (5) M is the total number of lipid components, and ζ i are the intrinsic curvature and mole fraction of the i lipid components, respectively. The lipid composition is found using lipidomic data of i and φ q ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ !2 1 þ t (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 15 of 20 The EMBO Journal Sophie L Winter et al PC the endosomal and viral membranes (Fig 1J). The intrinsic curva- ≈ (cid:3)0:1 nm(cid:3)1 for phosphati- ture of the most abundant lipids is ζ dylcholine (PC; Chen & Rand, 1997; Szule et al, 2002), cholesterol ≈ (cid:3)0:5 nm(cid:3)1 (Chen & Rand, 1997; Kollmitzer et al, 2013), ζ CHOL phosphatidylethanolamine (PE) with ζ (Leikin et al, and sphingomyelin ≈ (cid:3)0:1 nm(cid:3)1 (Leikin et al, 1996). We find that the endosomal ζ SM and Ebola virus both have monolayer spontaneous curvature of roughly Jsm ¼ (cid:3)0:22 nm(cid:3)1. 1996; Kollmitzer ≈ (cid:3)0:35 nm(cid:3)1 PE 2013), et al, The overall membrane deformation energy is given by the integra- tion of Equation 4 over the area of both monolayers independently, Z Z FMem ¼ f þdAþ þ f þdAþ (6) The first and second integral in Equation 6 is performed over the upper and lower monolayer area, respectively. The matrix layer We model the matrix layer as a thin, uniform rigid elastic shell with a flat resting configuration. The matrix can avoid the sharp corners in the vicinity of the stalk and diaphragm rim by local detachment from the membrane. These allow the matrix to avoid strong shear deformations. The elastic energy of matrix deformation up to qua- dratic order in the area strain, ϵ; and in principle curvatures, c1 and c2, is given by Landau & Lifshitz (1970), Z 1 2 F mat ¼ Yd Þ ð 2 1(cid:3)ν þ Yd3 ð 12 1(cid:3)ν2 Þ ϵ2dA0 (cid:1) Z 1 2 ð c1 þ c2 Þ2(cid:3) 1(cid:3)ν ð Þc1∙c2 dA (7) (cid:3) with dA0 and dA are the area elements of the undeformed and deformed states, d is the matrix thickness, Y is the Young’s modulus, and ν is the Poisson’s ratio. We consider only stretching and bending deformations and explicitly prohibit shear. The thickness of the VP40 matrix layer is estimated to be d ¼ 4 nm based on the cryo-ET data presented here (Fig 1A–H). The Young’s modulus and Poisson ratio of the VP40 matrix layer was never measured, but we estimate them to be within the same magnitude as other viruses with similar matrix layer structures, such as M1 of influenza virus. The VP40 matrix layer Poisson’s ratio is taken as ν ¼ 0:5, and the Youngs mod- ulus is in the range 5–22 MPa (Li et al, 2011; Schaap et al, 2012). With that, we estimate the stretching modulus of the VP40 matrix Þ ∼ 20 ! 80 mN=m, and the pure bending layer in the range of contribution with modulus in the range of Þ ∼ 8 ! 35 kBT. Yd ð 2 1(cid:3)ν Yd3 ð 24 1(cid:3)ν2 The matrix layer and the membrane can locally detach in the vicinity of the stalk and diaphragm rim to avoid substantial defor- mation there. Besides these regions, the matrix interacts continu- ously with the membrane since the VP40 matrix layer is tightly packed. Inspired by the MD simulations (Fig 2E), we describe the VP40-membrane interaction energy density with Lennard–Jones-like potential, Z (cid:1) Uint ¼ u0 (cid:5) (cid:6) z0 z 12 (cid:3)2 (cid:3) 6 (cid:5) (cid:6) z0 z dA (8) sub-tomogram averaging and the MD simulations (Fig 4H). The interaction energy density, u0, is estimated from the MD simulations as the free energy of a single VP40 dimer at z ¼ z0 (11 kBT for pH 7.4 and 6.5 kBT for pH 4.5, Fig 2E) divided by the density of VP40 dimers obtained from the cryo-ET data (Fig 1I), we find u0 ¼ 0:2 kbT∙nm(cid:3)2 at pH 7.4 and u0 ¼ 0:1 kbT∙nm(cid:3)2 at pH 4.5. Computational procedure GitHub Our computational approach is based on previous works (Golani et al, 2021; Zucker et al, 2021) and published as open-source code on (https://github.com/GonenGolani/Fusion_Solver), where further details can be found. The calculation involves three parts. We start by simulating the stalk shape and find its minimal energy configuration. Next, we allow the stalk to expand to hemifu- sion diaphragm, and finally, we calculate the energy barrier of pore formation based on the membrane stress and the interaction energy with the VP40 matrix layer in the diaphragm. The stalk energy barrier represents the minimal mechanical work needed to merge the proximal monolayers. We calculate the hemifu- sion stalk shape and its formation energy by setting the membrane in stalk configuration. Then, we minimize the sum of the membrane and matrix interaction deformation energies (Equations 7 and (8)) while requiring that RD ¼ 0, E(cid:4) stalk ¼ min F Mem þ F mat þ Uint ½ (cid:5) (9) after the stalk has formed, we release the constrain on RD and allow the system to spontaneously relaxes to a hemifusion dia- phragm. The matrix layer can remain attached to the diaphragm or detached. We calculate the fusion-pore formation energy barrier based on the stress in the hemifusion diaphragm. To facilitate our computa- tion, we assume that the pore formation is initiated at the center of the diaphragm and that the fast fluctuation in pore size does not change the hemifusion diaphragm and matrix layer equilibrium shapes. The pore must expand to the majority of the diaphragm before it overcomes the critical energy, so the initiation point is mainly irrelevant to the magnitude energy barrier. However, since the pores are more likely to form in the vicinity of the diaphragm rim, where stress is maximal, our estimation should be considered a slight overestimation of the actual energy barrier. With this assump- tion, the energy of pore opening to radius ρ is thus given by Epore ρð Þ ¼ 2πρλ(cid:3)π Z ρ0¼ρ ρ0¼0 γ ρ0ð Þρ0dρ0; (10) with λ, the pore rim line-tension magnitude, being independent of the matrix layer or the membrane shape. In our simulations, we take it to be λ ¼ 12 pN (Portet & Dimova, 2010). The second term in Equation 10 is the energy gained by removing lipids from the stressed diaphragm and relocating them to the surrounding mem- branes. The stress contains two contributions: the relaxation of the splay, saddle-splay, and shear of the lipids compared to the sur- rounding membranes and the detachment from the matrix layer, γ ρð Þ ¼ f þ ρð Þ þ f (cid:3) ρð Þ þ u ρð Þ: (11) with the integral performed on the area of the matrix layer, dA. z is the distance from the monolayer dividing plane to the mid-plane of the VP40 layer, and z0 ¼ 4 nm is the resting length obtained from with f þ and f (cid:3) the energy deformation density of the upper and lower monolayer (Equation 4), respectively, and u the interaction 16 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal energy density with the matrix (Equation 8). The pore formation energy barrier is the maxima of Epore ρð Þ, at the Heidelberg University (HD-cryoNET) for support and assistance; the Electron Microscopy Core Facility at EMBL and Wim Hagen for data acquisi- E(cid:4) pore ρ ¼ ρ(cid:4) ð (cid:7) Þ ¼ max Epore (cid:8) : (12) tion; Dimitrios Papagiannidis for critical reading of the manuscript. Plasmids encoding BlaM-VP40 and pHluorin were a kind gift from Kartik Chandran and Gero Miesenböck, respectively. CSC-IT Centre for Science Ltd. (Espoo, Finland) We find the stress (Equation 11) based on the equilibrium shape of the diaphragm, and Equation 12 is found by numerically integrat- ing Equation 10 and finding the maximum. Beta-lactamase assay Huh7 cells were seeded on a 96-well plate coated with 2 μg fibronectin in 1 × PBS at a density of 0.02 × 106 cells per well. 24 h post seeding, the media of inhibitor-treated cells were replaced with 25 mM NHCl4 in DMEM media (ThermoFisher Scientific) supplemented with 10% (v/v) FBS and 100 U/ml penicillin–streptomycin (ThermoFisher Scien- tific), and cells were incubated for 1.5 h at 37°C, 5% CO2. Same amounts of purified beta-lactamase (BlaM)-containing VLPs were either untreated, treated with low pH, thermolysin, or a combination of low pH and thermolysin. For the thermolysin- treatment, 500 μg/ml thermolysin (ThermoFisher Scientific), recon- stituted in H2O and filtered through a 0.22 μm membrane filter, was added to the VLPs for 30 min at 37°C. To quench the reaction, 300 μ g/ml phosphoramidon was added for 10 min at 37°C. For the low pH treatment, citric acid prepared in HNE buffer (10 mM HEPES, 100 mM NaCl, 1 mM EDTA) was added in a final concentration of 1.67 mM to the VLPs for 30 min at 37°C. The BlaM-VLPs were immediately placed on ice until infection. For infection, the media of all cells were removed, and 50 μl pretreated BlaM-VLP solution was added to each well and the plate was centrifuged for 30 min at 200 g, 20°C (Beckmann). BlaM-VLP solutions were immediately removed and replaced with 100 μl media with and without 25 mM NH4Cl. Cells were incubated for 1.5 h at 37°C, 5% CO2, before freshly preparing the BlaM dye from the LiveBLAzer™ FRET-B/G Loading Kit with CCF4-AM (ThermoFisher Scientific) supplemented with probenecid (Invitrogen) according to the protocol provided by the manufacturer. 20 μl of the BlaM mix was added per well. After incubation at 11°C for 12–14 h, the cells were briefly checked for viability using a Nikon microscope and detached for 5–10 min using trypsin–EDTA at 37°C. Cells were harvested and washed with 3× with PBS before performing FACS using a BD FACS Celesta Cell Analyzer (BD Biosciences). Data availability Electron tomography data were deposited to EMDB (EMD-15268, http://www.ebi.ac.uk/pdbe/entry/EMD-15268; EMD-15244, http:// www.ebi.ac.uk/pdbe/entry/EMD-15244). Additional data and mate- rial related to this publication may be obtained upon request. Atom- istic molecular dynamics simulations of initial structures and topology files were deposited to Zenodo (https://doi.org/10.5281/ zenodo.7652685). Expanded View for this article is available online. Acknowledgments We thank the Infectious Diseases Imaging Platform (IDIP) at the Center for Integrative Infectious Disease Research Heidelberg and the cryo-EM network is acknowledged for excellent computational resources. The authors gratefully acknowledge the data storage service SDS@hd supported by the Ministry of Science, Research, and the Arts Baden-Württemberg (MWK), the German Research Foundation (DFG) through grant INST 35/1314-1 FUGG and INST 35/ 1503-1 FUGG. We gratefully acknowledge funding from the Chica and Heinz Schaller Foundation to PC, the Minerva Stiftung to GG, Deutsche Forschungsge- meinschaft (DFG) project A5—SFB/TRR 83 to WN, FL, DFG project number 240245660—SFB 1129 to OTF, BB, USS, and PC, DFG project number 278001972 —TRR 186 and project number 112927078—TRR 83 to BB, DFG project VA 1570/ 1-1 to MV. Open Access funding enabled and organized by Projekt DEAL. Author contributions Sophie L Winter: Conceptualization; formal analysis; validation; investigation; visualization; methodology; writing – original draft; writing – review and editing. Gonen Golani: Conceptualization; software; validation; investigation; visualization; methodology; writing – original draft; writing – review and editing. Fabio Lolicato: Conceptualization; software; formal analysis; investigation; visualization; methodology; writing – review and editing. Melina Vallbracht: Investigation; methodology; writing – review and editing. Keerthihan Thiyagarajah: Formal analysis; methodology. Samy Sid Ahmed: Formal analysis; investigation; methodology. Christian Lüchtenborg: Methodology. Oliver T Fackler: Supervision; funding acquisition; writing – review and editing. Britta Brügger: Supervision; funding acquisition. Thomas Hoenen: Formal analysis; investigation; methodology; writing – review and editing. Walter Nickel: Supervision; funding acquisition; writing – review and editing. Ulrich S Schwarz: Supervision; funding acquisition; writing – review and editing. Petr Chlanda: Conceptualization; formal analysis; supervision; funding acquisition; validation; investigation; writing – original draft; project administration; writing – review and editing. Disclosure and competing interests statement The authors declare that they have no conflict of interest. References Abraham MJ, Murtola T, Schulz R, P(cid:2)all S, Smith JC, Hess B, Lindah E (2015) Gromacs: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1–2: 19 – 25 Adu-Gyamfi E, Soni SP, Xue Y, Digman MA, Gratton E, Stahelin RV (2013) The ebola virus matrix protein penetrates into the plasma membrane: a key step in viral protein 40 (VP40) oligomerization and viral egress. J Biol Chem 288: 5779 – 5789 Amiar S, Husby ML, Wijesinghe KJ, Angel S, Bhattarai N, Gerstman BS, Chapagain PP, Li S, Stahelin RV (2021) Lipid-specific oligomerization of the Marburg virus matrix protein VP40 is regulated by two distinct interfaces for virion assembly. J Biol Chem 296: 100796 Banerjee I, Miyake Y, Philip Nobs S, Schneider C, Horvath P, Kopf M, Matthias P, Helenius A, Yamauchi Y (2014) Influenza a virus uses the aggresome processing machinery for host cell entry. Science 346: 473 – 477 Bharat TAM, Noda T, Riches JD, Kraehling V, Kolesnikova L, Becker S, Kawaoka Y, Briggs JAG (2012) Structural dissection of Ebola virus and its assembly (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 17 of 20 The EMBO Journal Sophie L Winter et al determinants using cryo-electron tomography. Proc Natl Acad Sci U S A 109: 4275 – 4280 microscopy. J Struct Biol 212: 107633 https://doi.org/10.1016/j.jsb.2020. 107633 Blight EG, Dyer WJ (1959) A rapid method of total lipid extraction and Feldmann H, Klenk HD (1996) Marburg and Ebola viruses. Adv Virus Res 47: purification. Can J Biochem Physiol 37: 911 – 917 1 – 52 Bonomi M, Bussi G, Camilloni C, Tribello GA, Ban(cid:2)aˇs P, Barducci A, Bernetti M, Bolhuis PG, Bottaro S, Branduardi D et al (2019) (2019) promoting transparency and reproducibility in enhanced molecular simulations. Nat Methods 168: 670 – 673 Booth TF, Rabb MJ, Beniac DR (2013) How do filovirus filaments bend without breaking? Trends Microbiol 21: 583 – 593 Fontana J, Steven AC (2013) At low pH, influenza virus matrix protein M1 undergoes a conformational change prior to dissociating from the membrane. J Virol 87: 5621 – 5628 Fuhrmans M, Marelli G, Smirnova YG, Müller M (2015) Mechanics of membrane fusion/pore formation. Chem Phys Lipids 185: 109 – 128 Geisbert TW, Jahrling PB (1995) Differentiation of filoviruses by electron Bornholdt ZA, Noda T, Abelson DM, Halfmann P, Wood MR, Kawaoka Y, microscopy. Virus Res 39: 129 – 150 Saphire EO (2013) Structural rearrangement of ebola virus vp40 begets multiple functions in the virus life cycle. Cell 154: 763 – 774 Brecher M, Schornberg KL, Delos SE, Fusco ML, Saphire EO, White JM (2012) Cathepsin cleavage potentiates the Ebola virus glycoprotein to undergo a subsequent fusion-relevant conformational change. J Virol 86: 364 – 372 Golani G, Leikina E, Melikov K, Whitlock JM, Gamage DG, Luoma-Overstreet G, Millay DP, Kozlov MM, Chernomordik LV (2021) Myomerger promotes fusion pore by elastic coupling between proximal membrane leaflets and hemifusion diaphragm. Nat Commun 12: 1 – 18 Greber UF, Singh I, Helenius A (1994) Mechanisms of virus uncoating. Trends Carette JE, Raaben M, Wong AC, Herbert AS, Obernosterer G, Mulherkar N, Microbiol 2: 52 – 56 Kuehne AI, Kranzusch PJ, Griffin AM, Ruthel G et al (2011) Ebola virus entry requires the cholesterol transporter Niemann-pick C1. Nature 477: 340 – 343 Casta~no-D(cid:2)ıez D (2017) The Dynamo package for tomography and subtomogram averaging: components for MATLAB, GPU computing and EC2 Amazon web Services. Acta Crystallogr D Struct Biol 73: 478 – 487 Casta~no-D(cid:2)ıez D, Kudryashev M, Arheit M, Stahlberg H (2012) Dynamo: a flexible, user-friendly development tool for subtomogram averaging of Hagen WJH, Wan W, Briggs JAG (2017) Implementation of a cryo-electron tomography tilt-scheme optimized for high resolution subtomogram averaging. J Struct Biol 197: 191 – 198 Hamm M, Kozlov MM (1998) Tilt model of inverted amphiphilic mesophases. Eur Phys J B 6: 519 – 528 Hamm M, Kozlov MM (2000) Elastic energy of tilt and bending of fluid membranes. Eur Phys J E 3: 323 – 335 cryo-EM data in high-performance computing environments. J Struct Biol 178: 139 – 151 Harrison SC (2008) Viral membrane fusion. Nat Struct Mol Biol 15: 690 – 698 Helfrich W (1973) Elastic properties of lipid bilayers: theory and possible Chen Z, Rand RP (1997) The influence of cholesterol on phospholipid membrane curvature and bending elasticity. Biophys J 73: 267 – 276 Chernomordik LV, Kozlov MM (2008) Mechanics of membrane fusion. Nat Struct Mol Biol 15: 675 – 683 experiments. Z Naturforsch C 28: 693 – 703 Hess B, Bekker H, Berendsen HJC, Fraaije JGEM (1997) LINCS: a linear constraint solver for molecular simulations. J Comput Chem 18: 1463 – 1472 Chernomordik LV, Zimmerberg J, Kozlov MM (2006) Membranes of the world Hoover WG (1985) Canonical dynamics: equilibrium phase-space unite! J Cell Biol 175: 201 – 207 distributions. Phys Rev A 31: 1695 – 1697 C^ot(cid:2)e M, Misasi J, Ren T, Bruchez A, Lee K, Filone CM, Hensley L, Li Q, Ory D, Huang J, Mackerell AD (2013) CHARMM36 all-atom additive protein force Chandran K et al (2011) Small molecule inhibitors reveal Niemann-pick C1 is essential for Ebola virus infection. Nature 477: 344 – 348 Coutsias EA, Seok C, Jacobson MP, Dill KA (2004) A kinematic view of loop closure. J Comput Chem 25: 510 – 528 Deamer DW (1987) Proton permeation of lipid bilayers. J Bioenerg Biomembr 19: 457 – 479 Deamer DW, Bramhall J (1986) Permeability of lipid bilayers to water and ionic solutes. Chem Phys Lipids 40: 167 – 188 field: validation based on comparison to NMR data. J Comput Chem 34: 2135 – 2145 Huang J, Rauscher S, Nawrocki G, Ran T, Feig M, De Groot BL, Grubmüller H, MacKerell AD (2016) CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nat Methods 14: 71 – 73 Hub JS, De Groot BL, Van Der Spoel D (2010) G-whams-a free weighted histogram analysis implementation including robust error and autocorrelation estimates. J Chem Theory Comput 6: 3713 – 3720 DeCoursey TE (2008) Voltage-gated proton channels. Cell Mol Life Sci 65: Humphrey W, Dalke A, Schulten K (1996) VMD: visual molecular dynamics. J 2554 – 2573 Del Vecchio K, Frick CT, Jeevan Gc XB, Oda S, Bernard Gerstman XS, Erica Ollmann Saphire X, Prem Chapagain XP, Robert Stahelin XV (2018) A cationic, C-terminal patch and structural rearrangements in Ebola virus matrix VP40 protein control its interactions with phosphatidylserine. J Biol Chem 293: 3335 – 3349 Dimova R (2014) Recent developments in the field of bending rigidity measurements on membranes. Adv Colloid Interface Sci 208: 225 – 234 Dube D, Brecher MB, Delos SE, Rose SC, Park EW, Schornberg KL, Kuhn JH, White JM (2009) The primed ebolavirus glycoprotein (19-Kilodalton GP 1,2): sequence and residues critical for host cell binding. J Virol 83: 2883 – 2891 Essmann U, Perera L, Berkowitz ML, Darden T, Lee H, Pedersen LG (1995) A smooth particle mesh Ewald method. J Chem Phys 103: 8577 – 8593 Fäßler F, Zens B, Hauschild R, Schur FKM (2020) 3D printed cell culture grid holders for improved cellular specimen preparation in cryo-electron Mol Graph 14: 33 – 38 Jahn R, Grubmüller H (2002) Membrane fusion. Curr Opin Cell Biol 14: 488 – 495 Jeevan BG, Gerstman BS, Stahelin RV, Chapagain PP (2016) The Ebola virus protein VP40 hexamer enhances the clustering of PI(4,5)P2 lipids in the plasma membrane. Phys Chem Chem Phys 18: 28409 – 28417 Jeevan BG, Gerstman BS, Chapagain PP (2017) Membrane association and localization dynamics of the Ebola virus matrix protein VP40. Biochim Biophys Acta Biomembr 1859: 2012 – 2020 Jo S, Kim T, Iyer VG, Im W (2008) CHARMM-GUI: a web-based graphical user interface for CHARMM. J Comput Chem 29: 1859 – 1865 Jo S, Lim JB, Klauda JB, Im W (2009) CHARMM-GUI membrane builder for mixed bilayers and its application to yeast membranes. Biophys J 97: 50 – 58 Johnson KA, Taghon GJF, Scott JL, Stahelin RV (2016) The Ebola virus matrix protein, VP40, requires phosphatidylinositol 4,5-bisphosphate (PI(4,5)P 2) 18 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors Sophie L Winter et al The EMBO Journal for extensive oligomerization at the plasma membrane and viral egress. Sci Rep 6: 1 – 14 Mastronarde DN, Held SR (2017) Automated tilt series alignment and tomographic reconstruction in IMOD. J Struct Biol 197: 102 – 113 Jones DM, Padilla-Parra S (2016) The β-lactamase assay: harnessing a FRET Miesenböck G, De Angelis DA, Rothman JE (1998) Visualizing secretion and biosensor to analyse viral fusion mechanisms. Sensors 16: 950 Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML (1983) Comparison of simple potential functions for simulating liquid water. J Chem Phys 79: 926 – 935 Klein S, Cortese M, Winter S, Wachsmuth-Melm M, Neufeldt C, Cerikan B, Stanifer M, Boulant S, Bartenschlager R, Chlanda P (2020a) SARS-CoV-2 structure and replication characterized by in situ cryo-electron tomography. Nat Commun 11: 5885 Klein S, Wimmer B, Winter SL, Kolovou A, Laketa V, Chlanda P (2020b) Post- correlation on-lamella cryo-CLEM reveals the membrane architecture of lamellar bodies. Commun Biol 4: 137 Kollmitzer B, Heftberger P, Rappolt M, Pabst G (2013) Monolayer spontaneous curvature of raft-forming membrane lipids. Soft Matter 9: 10877 – 10884 Kozlovsky Y, Kozlov MM (2002) Stalk model of membrane fusion: solution of energy crisis. Biophys J 82: 882 – 895 Kozlovsky Y, Chernomordik LV, Kozlov MM (2002) Lipid intermediates in membrane fusion: formation, structure, and decay of hemifusion diaphragm. Biophys J 83: 2634 – 2651 Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character of a protein. J Mol Biol 157: 105 – 132 Landau LD, Lifshitz EM (1970) Theory of elasticity. New York, NY: Pergamon Press Lee JE, Saphire EO (2009) Ebolavirus glycoprotein structure and mechanism of entry. Future Virol 4: 621 – 635 Lee J, Kreutzberger AJB, Odongo L, Nelson EA, Nyenhuis DA, Kiessling V, Liang B, Cafiso DS, White JM, Tamm LK (2021) Ebola virus glycoprotein interacts with cholesterol to enhance membrane fusion and cell entry. Nat Struct Mol Biol 28: 181 – 189 Leikin S, Kozlov MM, Fuller NL, Rand RP (1996) Measured effects of diacylglycerol on structural and elastic properties of phospholipid membranes. Biophys J 71: 2623 – 2632 Li S, Eghiaian F, Sieben C, Herrmann A, Schaap IAT (2011) Bending and puncturing the influenza lipid envelope. Biophys J 100: 637 – 645 Li S, Sieben C, Ludwig K, Höfer CT, Chiantia S, Herrmann A, Eghiaian F, Schaap IAT (2014) PH-ontrolled two-step uncoating of influenza virus. Biophys J 106: 1447 – 1456 Liebisch G, Binder M, Schifferer R, Langmann T, Schulz B, Schmitz G (2006) High throughput quantification of cholesterol and cholesteryl ester by electrospray ionization tandem mass spectrometry (ESI-MS/MS). Biochim Biophys Acta 1761: 121 – 128 Lozach PY, Huotari J, Helenius A (2011) Late-penetrating viruses. Curr Opin Virol 1: 35 – 43 synaptic transmission with pH-sensitive green fluorescent proteins. Nature 394: 192 – 195 Miller EH, Obernosterer G, Raaben M, Herbert AS, Deffieu MS, Krishnan A, Ndungo E, Sandesara RG, Carette JE, Kuehne AI et al (2012) Ebola virus entry requires the host-programmed recognition of an intracellular receptor. EMBO J 31: 1947 – 1960 Mingo RM, Simmons JA, Shoemaker CJ, Nelson EA, Schornberg KL, D’Souza RS, Casanova JE, White JM (2015a) Ebola virus and severe acute respiratory syndrome coronavirus display late cell entry kinetics: evidence that transport to NPC1+ Endolysosomes is a rate-defining step. J Virol 89: 2931 – 2943 Mingo RM, Simmons JA, Shoemaker CJ, Nelson EA, Schornberg KL, D’Souza RS, Casanova JE, White JM (2015b) EBOV and SARS-CoV display late cell entry kinetics: evidence that transport to NPC1 + Endolysosomes is a rate- defining step. J Virol 89: 2931 – 2943 Nanbo A, Imai M, Watanabe S, Noda T, Takahashi K, Neumann G, Halfmann P, Kawaoka Y (2010) Ebolavirus is internalized into host cells via macropinocytosis in a viral glycoprotein-dependent manner. PLoS Pathog 6: e1001121 Nguyen TL, Schoehn G, Weissenhorn W, Hermone AR, Burnett JC, Panchal RG, McGrath C, Zaharevitz DW, Aman MJ, Gussio R et al (2005) An all- atom model of the pore-like structure of hexameric VP40 from Ebola: structural insights into the monomer-hexamer transition. J Struct Biol 151: 30 – 40 van Niel G, Bergam P, Di Cicco A, Hurbain I, Lo Cicero A, Dingli F, Palmulli R, Fort C, Potier MC, Schurgers LJ et al (2015) Apolipoprotein E regulates amyloid formation within endosomes of pigment cells. Cell Rep 13: 43 – 51 Noda T, Sagara H, Suzuki E, Takada A, Kida H, Kawaoka Y (2002) Ebola virus VP40 drives the formation of virus-like filamentous particles along with GP. J Virol 76: 4855 – 4865 Norris MJ, Husby ML, Kiosses WB, Yin J, Saxena R, Rennick LJ, Heiner A, Harkins SS, Pokhrel R, Schendel SL et al (2022) Measles and Nipah virus assembly: specific lipid binding drives matrix polymerization. Sci Adv 8: eabn1440 Özbalci C, Sachsenheimer T, Brügger B (2013) Quantitative analysis of cellular lipids by nano-electrospray ionization mass spectrometry. Methods Mol Biol 1033: 3 – 20 Panchal RG, Ruthel G, Kenny TA, Kallstrom GH, Lane D, Badie SS, Li L, Bavari S, Aman MJ (2003) In vivo oligomerization and raft localization of Ebola virus protein VP40 during vesicular budding. Proc Natl Acad Sci U S A 100: 15936 – 15941 Parrinello M, Rahman A (1981) Polymorphic transitions in single crystals: a Mahamid J, Tegunov D, Maiser A, Arnold J, Leonhardt H, Plitzko JM, new molecular dynamics method. J Appl Phys 52: 7182 – 7190 Baumeister W (2019) Liquid-crystalline phase transitions in lipid droplets are related to cellular states and specific organelle association. Proc Natl Acad Sci U S A 116: 16866 – 16871 Malek M, Wawrzyniak AM, Koch P, Lüchtenborg C, Hessenberger M, Sachsenheimer T, Jang W, Brügger B, Haucke V (2021) Inositol triphosphate-triggered calcium release blocks lipid exchange at endoplasmic reticulum-Golgi contact sites. Nat Commun 12: 2673 Manzoor R, Igarashi M, Takada A (2017) Influenza a virus M2 protein: roles from ingress to egress. Int J Mol Sci 18: 2649 Mastronarde D (2005) Automated electron microscope tomography using robust prediction of specimen movements. J Struct Biol 152: 36 – 51 Pastor RW, MacKerell AD (2011) Development of the CHARMM force field for lipids. J Phys Chem Lett 2: 1526 – 1532 Pavadai E, Gerstman BS, Chapagain PP (2018) A cylindrical assembly model and dynamics of the Ebola virus VP40 structural matrix. Sci Rep 8: 9776 Portet T, Dimova R (2010) A new method for measuring edge tensions and stability of lipid bilayers: effect of membrane composition. Biophys J 99: 3264 – 3273 Ruigrok RWH, Schoehn G, Dessen A, Forest E, Volchkov V, Dolnik O, Klenk HD, Weissenhorn W (2000) Structural characterization and membrane binding properties of the matrix protein VP40 of Ebola virus. J Mol Biol 300: 103 – 112 (cid:1) 2023 The Authors The EMBO Journal 42: e113578 \| 2023 19 of 20 The EMBO Journal Sophie L Winter et al Schaap IAT, Eghiaia F, Des George A, Veigel C (2012) Effect of envelope Terzi MM, Ergüder MF, Deserno M (2019) A consistent quadratic curvature- proteins on the mechanical properties of influenza virus. J Biol Chem 287: 41078 – 41088 Scianimanico S, Schoehn G, Timmins J, Ruigrok RHW, Klenk HD, Weissenhorn W (2000) Membrane association induces a conformational change in the Ebola virus matrix protein. EMBO J 19: 6732 – 6741 Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Söding J et al (2011) Fast, scalable generation of high- quality protein multiple sequence alignments using Clustal omega. Mol Syst Biol 7: 539 Simmons JA, D’Souza RS, Ruas M, Galione A, Casanova JE, White JM (2015) Ebolavirus glycoprotein directs fusion through NPC1- Endolysosomes. J Virol 90: 605 – 610 Søndergaard CR, Olsson MHM, Rostkowski M, Jensen JH (2011) Improved treatment of ligands and coupling effects in empirical calculation and rationalization of p K a values. J Chem Theory Comput 7: 2284 – 2295 Soni SP, Stahelin RV (2014) The Ebola virus matrix protein VP40 selectively induces Vesiculation from phosphatidylserine-enriched membranes. J Biol Chem 289: 33590 – 33597 tilt theory for fluid lipid membranes. J Chem Phys 151: 164108 Torrie GM, Valleau JP (1974) Monte Carlo free energy estimates using non- Boltzmann sampling: application to the sub-critical Lennard-Jones fluid. Chem Phys Lett 28: 578 – 581 Torrie GM, Valleau JP (1977) Nonphysical sampling distributions in Monte Carlo free-energy estimation: umbrella sampling. J Comput Phys 23: 187 – 199 Tribello GA, Bonomi M, Branduardi D, Camilloni C, Bussi G (2014) PLUMED 2: new feathers for an old bird. Comput Phys Commun 185: 604 – 613 Tsui FC, Ojcius DM, Hubbell WL (1986) The intrinsic pKa values for phosphatidylserine and phosphatidylethanolamine in phosphatidylcholine host bilayers. Biophys J 49: 459 – 468 Wan W, Kolesnikova L, Clarke M, Koehler A, Noda T, Becker S, Briggs JAG (2017) Structure and assembly of the Ebola virus nucleocapsid. Nature 551: 394 – 397 Wan W, Clarke M, Norris MJ, Kolesnikova L, Koehler A, Bornholdt ZA, Becker S, Saphire EO, Briggs JAG (2020) Ebola and Marburg virus matrix layers are locally ordered assemblies of VP40 dimers. Elife 9: e59225 Welsch S, Kolesnikova L, Kr€ahling V, Riches JD, Becker S (2010) Electron tomography reveals the steps in filovirus budding. PLoS Pathog 6: 1000875 Stahelin RV (2014) Membrane binding and bending in Ebola VP40 assembly Winter SL, Chlanda P (2021) Dual-axis Volta phase plate cryo-electron and egress. Front Microbiol 5: 300 Stauffer CE (1971) The effect of pH on thermolysin activity. Arch Biochem Biophys 147: 568 – 570 Szule JA, Fuller NL, Peter Rand R (2002) The effects of acyl chain length and saturation of diacylglycerols and phosphatidylcholines on membrane monolayer curvature. Biophys J 83: 977 – 984 Takamatsu Y, Kolesnikova L, Becker S (2018) Ebola virus proteins NP, VP35, and VP24 are essential and sufficient to mediate nucleocapsid transport. Proc Natl Acad Sci U S A 115: 1075 – 1080 Templer RH, Khoo BJ, Seddon JM (1998) Gaussian curvature modulus of an amphiphilic monolayer. Langmuir 14: 7427 – 7434 Terzi MM, Deserno M (2017) Novel tilt-curvature coupling in lipid membranes. J Chem Phys 147: 084702 tomography of Ebola virus-like particles reveals Actin-VP40 interactions. J Struct Biol 213: 107742 Yamauchi Y, Greber UF (2016) Principles of virus Uncoating: cues and the snooker ball. Traffic 17: 569 – 592 Zucker B, Golani G, Kozlov M (2021) Model for ring closure in ER tubular network dynamics. bioRxiv https://doi.org/10.1101/2021.11.18.469198 [PREPRINT] License: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 20 of 20 The EMBO Journal 42: e113578 \| 2023 (cid:1) 2023 The Authors
10.3390_nu12040966	Article Salivary Microbiota Shifts under Sustained Consumption of Oolong Tea in Healthy Adults Zhibin Liu, Hongwen Guo, Wen Zhang and Li Ni * Institute of Food Science & Technology, Fuzhou University, Fuzhou 350108, China; [email protected] (Z.L.); [email protected] (H.G.); [email protected] (W.Z.) * Correspondence: [email protected]; Tel.: +86-591-2286-6378 Received: 16 February 2020; Accepted: 25 March 2020; Published: 31 March 2020 Abstract: Tea is the most widely consumed beverages next to water, however little is known about the inﬂuence of sustained tea consumption on the oral bacteria of healthy adults. In this study, three oral healthy adults were recruited and instructed to consume 1.0 L of oolong tea infusions (total polyphenol content, 2.83 g/L) daily, for eight weeks. Salivary microbiota pre-, peri-, and post-treatment were fully compared by high-throughput 16S rRNA sequencing and multivariate statistical analysis. It was revealed that oolong tea consumption reduced salivary bacterial diversity and the population of some oral disease related bacteria, such as Streptococcus sp., Prevotella nanceiensis, Fusobacterium periodonticum, Alloprevotella rava, and Prevotella elaninogenica. Moreover, via correlation network and Venn diagram analyses, seven bacterial taxa, including Streptococcus sp. (OTU_1), Ruminococcaceae sp. (OTU_33), Haemophilus sp. (OTU_696), Veillonella spp. (OTU_133 and OTU_23), Actinomyces odontolyticus (OTU_42), and Gemella haemolysans (OTU_6), were signiﬁcantly altered after oolong tea consumption, and presented robust strong connections (\|r\| > 0.9 and p < 0.05) with other oral microbiota. These results suggest sustained oolong tea consumption would modulate salivary microbiota and generate potential oral pathogen preventative beneﬁts. Additionally, diverse responses to oolong tea consumption among subjects were also noticed. Keywords: oolong tea; phenolic proﬁle; salivary microbiota; 16S rRNA sequencing; bacterial diversities; correlation network 1. Introduction An estimation of 700 diverse bacterial species have been identiﬁed in human oral cavities, which constitute complex microbial communities [1]. These bacteria generally inhabit at diﬀerent oral niches, including saliva, supragingival plaque, subgingival plaque, and mucosa. Of these niches, saliva harbors as much as 108 bacteria/mL and constitutes a reservoir of microorganisms regularly derived from dental plaque bioﬁlms adhering to gingival crevices, periodontal pockets, the dorsum of the tongue, and other oral mucosal surfaces [2]. As an integral part of oral microbiota, salivary microbiota has been found to be diﬀerentiated between patients with a healthy oral cavity and those with dental caries and periodontitis [3]. Additionally, several studies discovered marked clinical importance of salivary microbiota on the general health of the host, such as by either preventing or causing infections [4]. Thus, salivary microbiota may provide further insight into the integral microbiota structure within the human oral cavity, and even the oral and general health status of individuals. Since the oral cavity is exposed to the external environment, the salivary microbiota may be inﬂuenced by various factors, including oral hygiene, smoking, nutrients, mechanical stress, and the overall health condition of the host [5]. The impact of nutritional factors in shaping the oral microbial ecosystem cannot be ignored. Food residuals in the mouth can be utilized as substrates for oral bacteria; moreover, some food components have a selective eﬀect on microbial growth, by Nutrients 2020, 12, 966; doi:10.3390/nu12040966 www.mdpi.com/journal/nutrients nutrients(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Nutrients 2020, 12, 966 2 of 15 either stimulating or suppressing some speciﬁc bacteria. For example, a regular consumption of polyphenol-rich beverages and foods, such as tea, cranberry, coﬀee, grape, almond, and alcohol-free red wine, have been reported to inhibit oral pathogenic bacteria [6–8]. The suppression of oral, especially periodontal pathogenic, bacteria may ameliorate the control of plaque bioﬁlms, and thus reduce the inﬂammatory and immunological processes of oral and periodontal diseases [9]. Recently, the impact of nutraceutical dietary aliments, such as antioxidants, probiotics, natural agents, and vitamins, on oral health is gaining more and more attention [10]. Tea (Camellia sinensis), second only to water, is the most widely consumed beverage in the world. The major constituents of tea leaves are the ﬂavonoids, including ﬂavonols, ﬂavones, and ﬂavan-3-ols, of which over 60% are the ﬂavan-3-ols, commonly referred to as catechins. Based on the United States Department of Agriculture (USDA) Flavonoid Database, it has been estimated that the daily total ﬂavonoid intake is mainly from ﬂavan-3-ols (83.5%); while, the major source of ﬂavonoids is tea (157 mg), and citrus fruit juices come second (8 mg) [11]. There is a large population of heavy tea consumers all over the world, especially in the southern part of China, where people consume a substantial amount of tea infusions on a daily basis. A number of health-promoting eﬀects have been associated with tea consumption; these eﬀects are generally attributed to the phenolic compounds in tea. Tea polyphenols are well known for their antimicrobial properties, including on Streptococcus mutans and lactobacilli [12], and thus, they are believed to possess anti-cariogenic eﬀects [13,14]. Moreover, regular consumption of tea has proved to exert gut microbiota regulation eﬀect [15,16]. However, with regard to the normal balanced oral microbiota, little is known about the inﬂuence of tea drinking. Considering the wide range of biological properties, including anti-microbial, anti-oxidant, anti-inﬂammatory, anti-cariogenic, and gut microbiota regulation eﬀects of tea polyphenols, it is reasonable to assume that sustained tea drinking will result in certain oral ecological shifts. A better understanding of the oral ecological shifts under sustained and signiﬁcant tea consumption may contribute to oral health management for tea consumers. It is also worth noting that, due to the variability in genes, social habits, hormonal ﬂuctuation, diet, quality and quantity of saliva, etc., the oral environment diﬀers between subjects and represents huge inter-individual variations [17]. Moreover, the responses of oral microbiota of diﬀerent individuals to certain nutritional factors maybe also be diverse. To understand the inﬂuence of tea consumption on oral microbiota, tracking the temporal dynamic of salivary microbiota of subjects separately may provide useful information free from interference of inter-individual variations. In the current study, it is hypothesized that sustained tea consumption will alter the composition of salivary microbiota and exert oral health beneﬁts to the host. To test this hypothesis, three orally healthy subjects were recruited and instructed to consume a substantial amount of tea infusions on a daily basis and their salivary bacterial communities pre-, peri-, and post-treatment were quantiﬁed by utilizing a high-throughput HiSeq sequencing technique. Then, via several multivariate statistical analyses, the temporal dynamics of salivary microbiota of each individual were analyzed. Based on these, the impact of sustained consumption of tea on the normal balanced oral microbiota was discussed. 2. Materials and Methods 2.1. Oolong Tea Infusion Preparation and Phenolic Proﬁle Analysis The tea used in this study was an oolong tea variety, purchased from a local market in Fujian Province, China. The oolong tea was prepared in accordance with the tea consumption method of local residents. A certain amount of dry oolong tea (whole leaves) was immersed in 20 times the volume of ◦ distilled boiling water (temperature around 90–95 C) for 1 min, then the tea leaves were ﬁltered, and the liquor was retained as an oolong tea infusion. The phenolic proﬁle of the tea infusion was then analyzed by utilizing ultra-high performance liquid chromatography (UHPLC) coupled to quadrupole time-of-ﬂight mass spectrometer (Q-TOF MS/MS) approach, as previously described [15]. Brieﬂy, chromatography separation was performed Nutrients 2020, 12, 966 3 of 15 ◦ ◦ on an Acquity UHPLC system (Waters, Milford, MA, USA) with HSS T3 column (100 mm × 2.1 mm, 1.7 µm). A sample of 1 µL was injected and eluted with the mobile phase at 0.3 mL/min at 40 C; detection was at 280 nm. The mobile phase consisted of (A) 0.1% formic acid solution (v/v) and (B) acetonitrile with 0.1% formic acid (v/v), while the gradient program was as follows: 99%–93% (A) in 0–2 min; 93%–60% (A) in 2–13 min; 60%–1% (A) in 13–14 min. The eluent was then introduced to a SYNAPT G2-Si high-deﬁnition mass spectrometer (Waters, Milford, MA, USA) equipped with an electrospray ionization (ESI) source. The analyses were performed in negative-ion mode and positive-ion mode, with a sampling cone voltage of 40.0 V, and a capillary voltage of 2500 V. The source temperature was 120 C. The time-of-ﬂight (TOF) acquisition rate was 0.2 s/scan with 0.01 s inter-scan delay. Data were collected in centroid mode from 100 to 1200 Da in full scan during 0–14 min. The mass data were corrected during acquisition using a lock-mass calibrant of leucine enkephalin (200 ng/mL), via a lock spray interface at + = 556.2771) and a ﬂow-rate of 50 µL/min, generating a reference ion for positive ion mode ([M+H] − = 554.2615) to ensure accuracy during the MS analysis. All data analyses negative ion mode ([M–H] were conducted using the MarkerLynx application manager software (version 4.1, Waters, Milford, MA, USA). The total polyphenols content in the tea infusions was then measured by utilizing the Folin–Ciocalteu method [18]. Brieﬂy, 1 mL sample, 5 mL Folin–Ciocalteu’s reagent (diluted 10 times), and 4 mL sodium carbonate (7.5%, w/v) were mixed. After 60 min, the absorbance at 765 nm was measured. Total phenolic content was expressed as a mass percentage on dry matter basis. Gallic acid was used as an external standard. C, with a desolvation gas ﬂow of 800 L/h at a temperature of 450 ◦ 2.2. Subject Enrollment, Study Design, and Salivary Sample Collection The inclusion criteria for this study included: healthy adult individuals sharing a relatively similar living environment; no tea and antibiotics taken in the previous 3 months; and no smoking. After the screening process, three healthy adult Chinese individuals (2 females and 1 male), 23 years of age, were enrolled from the campus of Fuzhou University, Fuzhou, China. The plaque and gingival status was examined before and after tea intervention. No obvious change was observed either before or after tea usage. In addition, no adverse reaction was reported throughout the experimental period by participants. Written informed consent was obtained from each participant. This study was approved by the ethical committee of the Institute of Food Science and Technology of Fuzhou University (approval number: IFSTFZU20180301). This study consisted of a 3-day baseline period, an 8-week oolong tea infusion intervention period, and a 4-week follow-up period. During the intervention period, the three subjects (subject 1, subject 2, and subject 3) were required to consume 1.0 L of oolong tea infusion per day (0.5 L in the morning and 0.5 L in the afternoon). Moreover, they were also instructed to circulate or swish the infusion around in their mouths prior to swallowing the tea infusion. During the follow-up period, the subjects were asked not to consume any tea drinks. In addition to this, the subjects were asked to maintain their regular diet and oral hygiene habits, with the exception of the sampling occasions. Salivary samples were collected at 4 diﬀerent stages, each stage included 3 sequential days: (A) 3 sequential days of the baseline period, which was prior to the intervention period; (B) 3 sequential days after 4 weeks of the tea intervention; (C) 3 sequential days after 8 weeks of the tea intervention; and (D) 3 sequential days at the end of the follow-up period, which accounted for 4 weeks post-intervention. All salivary sample collections were conducted in the morning. Each subject was asked not to eat, drink, or brush their teeth before the sample collection. Then, 2 mL of unstimulated saliva were collected from the subjects by expectoration into a tube. In total, 36 salivary samples from the 3 subjects were sampled. 2.3. Salivary Bacterial DNA Extraction Salivary bacterial DNA was extracted from the 36 salivary samples by utilizing a rapid DNA extraction kit (BioTeke Corporation, Beijing, China), following the manufacturer’s instructions. The extracted bacterial DNA was then checked by agarose gel electrophoresis. Nutrients 2020, 12, 966 4 of 15 2.4. Illumina Sequencing of Salivary Bacteria (cid:48) (cid:48) (cid:48) Bacterial primers 341-F (5 -CCT AYG GGR BGC ASC AG-3 -GGA CTA CNN (cid:48) ) with speciﬁc barcodes were used to amplify the V3–V4 region of bacterial 16S GGG TAT CTA AT-3 rRNA genes. The sequencing library of bacterial 16S rRNA genes was generated for high-throughput sequencing, employing the TruSeq DNA PCR-Free Sample Preparation Kit (Illumina, San Diego, CA, USA). Next, the library was sequenced on an Illumina HiSeq2500 platform by Novogene Bioinformatics Technology Co., Ltd. (Beijing, China). ) and 806-R (5 ® 2.5. Bioinformatic Analysis Raw sequencing reads, obtained from the Illumina platform, were then merged by using FLASH software (Version 1.2.7) [19] and ﬁltered using QIIME software (Version 1.7), with the default parameter setting of ‘split_libraries_fastq.py’ script [20,21]. All quality ﬁltered sequencing reads were then clustered into operational taxonomic units (OTUs) with a threshold of 97% sequence similarity, by utilizing UPARSE software (Version 7.0) [22]. The representative sequence (most abundant) for each bacterial OTU was then annotated by utilizing the GreenGene Database [23] and Human Oral Microbiome Database (HOMD) [24]. The least total sequences number was 30,070 in this study. The total reads of each sample was normalized to 30,070 sequences/sample, and the OTUs abundance information was normalized correspondingly for further analysis. Based on these annotated and normalized output data, diﬀerent statistical methods were used to interpret the similarities of diverse data sets, or to plot the correlation network among the salivary microbiota. First, community diversity estimators including Shannon and Simpson indexes were calculated by R software (Version 3.2.5) with vegan package. Second, the multiple response permutation procedure (MRPP) and analysis of similarity (Anosim) were employed to compare the statistical diﬀerences within and between subjects in salivary microbiota proﬁles, by using R software with vegan package [23]. Third, principal component analysis (PCA) was applied to evaluate and visualize the diﬀerences of samples in OTU-level complexity, by using R software with mixOmics package. Next, the correlations among the OTUs with relative abundance over 0.1% of each subject were calculated, based upon Pearson’s correlation coeﬃcients, by using R software with Hmisc package. The strong connections (\|r\| > 0.9, p < 0.05) were further imported into Gephi software (Version 0.8.2), so as to generate correlation networks of these predominant microbiota [25]. The nodes (OTUs) with high strong connection numbers were deﬁned as the “hub microbiota”, which were likely to be more connected to other nodes when compared to non-hub nodes [26,27]. Moreover, the relative abundance of the hub microbiota was further visualized into heatmaps, by utilizing R software with pheatmap package. Hierarchical clustering of the columns (samples) was further calculated based on Euclidean distance and ward.D method, and indicated on the heatmaps. Lastly, in order to identify the shared and unique hub salivary microbiome of these three subjects, a Venn diagram was built according to the method as descripted by Heberle et al. [28]. Other data are expressed as mean ± SD. Furthermore, the statistical signiﬁcance among diﬀerent data sets was analyzed by Student’s t-test or Duncan’s multiple range test using SPSS software (Version 19.0.0), while the signiﬁcance threshold was established at 0.05. 3. Results 3.1. Phenolic Proﬁle of Oolong Tea Infusion The total polyphenols content and phenolic proﬁle of the oolong tea infusion used in this study were determined, and the results indicated that the total polyphenol content of the tea infusion was 2.83 ± 0.02 g/L. Following untargeted UHPLC Q-TOF-MS approach, the phenolic constituents present in the tea infusion were further analyzed. Table 1 gives the MS characteristics and tentative identiﬁcation of each chromatographic peak. These chromatographic peaks, along with their proposed chemical structure, are depicted in Figure S1. In summary, 33 constituents were tentatively identiﬁed from the Nutrients 2020, 12, 966 5 of 15 tea infusion, including 2 alkaloids, 7 ﬂavan-3-ols, 7 organic acids and esters, 4 proanthocyanidins, 11 ﬂavonoid glycosides, 1 theaﬂavin, and 1 amino acid. Of the total chromatographic peak areas, caﬀeine (peak 16), epigallocatechin (peak 13), epicatechin (peak 20), gallic acid (peak 5), and caﬀeoyl-hexoside (peak 1) were the most abundant constituents. Table 1. The phenolic proﬁles of the oolong tea infusion. Peak No. a tR (Min) Tentative Identiﬁcation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 1.05 1.40 1.97 2.49 2.74 2.92 3.80 3.84 4.37 4.41 4.54 4.81 4.91 5.16 5.36 5.60 5.68 5.79 6.14 6.23 6.34 6.41 6.68 6.92 7.01 7.11 7.21 7.36 7.62 7.72 8.00 8.43 8.78 Caﬀeoyl-hexoside L-Theanine Epigallocatechin-glucuronide Theasinensin C Gallic acid Theogallin Theobromine b Gallocatechin Theasinensin B Digalloyl-hexoside O-Methylgallic acid Theacitrin A Epigallocatechin p-Coumaroylquinic acid Catechin Caﬀeine b Procyanidin Epicatechin-epicatechin p-Coumaroylquinic acid Epicatechin Epigallocatechin gallate p-Coumaroylquinic acid Gallocatechin gallate Theaﬂavin Myricetin-hexoside Myricetin-hexoside Quercetin-hexosyl-hexosyl-deoxyhexoside Quercetin-hexosyl-hexosyl-deoxyhexoside Kaempferol-deoxyhexosyl-deoxyhexoside Kaempferol-hexosyl-hexosyl-deoxyhexoside Kaempferol-hexosyl-hexosyl-deoxyhexoside Kaempferol-hexosyl-hexoside Kaempferol-hexoside Chemical Formula C15H18O9 C7H14N2O3 C21H22O13 C30H26O14 C7H6O5 C14H16O10 C7H8N4O2 C15H14O7 C37H30O18 C20H20O14 C8H8O5 C37H28O18 C15H14O7 C16H18O8 C15H14O6 C8H10N4O2 C30H26O12 C30H26O12 C16H18O8 C15H14O6 C22H18O11 C16H18O8 C22H18O11 C29H24O12 C21H20O13 C21H20O13 C33H40O21 C33H40O21 C27H30O14 C33H40O20 C33H40O20 C27H30O15 C21H20O11 [M-H]− (m/z) Measured Mass (Da) Theoretical Exact Mass (Da) Mass Accuracy (ppm) 341.0875 173.0931 481.0991 609.1235 169.0140 343.0665 181.0736 305.0662 761.1348 483.0758 183.0295 759.1196 305.0689 337.0923 289.0718 195.0888 577.1356 577.1356 337.0923 289.0734 457.0777 337.0918 457.0773 563.1199 479.0827 479.0825 771.1986 771.1982 577.1555 755.2029 755.2048 593.1508 447.0927 341.0873 173.0927 481.0983 609.1245 169.0137 343.0666 181.0725 305.0662 761.1354 483.0775 183.0294 759.1198 305.0662 337.0924 289.0713 195.0882 577.1346 577.1346 337.0924 289.0713 457.0771 337.0924 457.0771 563.1190 479.0826 479.0826 771.1984 771.1984 577.1558 755.2035 755.2035 593.1507 447.0928 0.58 2.53 1.74 −1.60 1.52 −0.19 6.04 0.09 −0.83 −3.57 0.58 −0.24 8.94 −0.26 1.87 3.30 1.65 1.65 −0.26 7.41 1.24 −1.74 0.37 1.60 0.19 −0.23 0.22 −0.30 −0.48 −0.81 1.70 0.18 −0.18 + a Peaks were assigned from the chromatograms in Figure S1; b [M+H] mode. 3.2. Overall Salivary Bacterial Structure The salivary bacterial components of the three subjects during the 12-week experimental period were investigated and evaluated using the Illumina HiSeq sequencing analysis. A total of 1,983,489 (average length = 425 bp) quality ﬁltered sequencing reads corresponding to the V3–V4 region of bacterial 16S rRNA genes were obtained. Good’s coverage estimation values were within the range of 99.8%–100%, which indicated adequate sequence coverage to reliably describe the full bacterial communities present in all the samples. All sequences were clustered into 189 to 458 OTUs with a 97% similarity level for each sample. The summary of the sequencing results is listed in Table S1. After the taxonomic assignment, these sequences were then annotated into 25 phyla and 260 genera. At the phylum level, Firmicutes (41.04%), Bacteroidetes (24.23%), and Proteobacteria (23.31%) comprised the majority of OTUs (88.59%). While at the genus level, Streptococcus (28.24%), Haemophilus (15.97%), Prevotella (14.64%), Alloprevotella (5.27%), and Neisseria (4.21%) were the most prevalent bacterial taxa throughout the three subjects, which in totality accounted for 69.05% of all salivary bacteria. The relative abundance of these bacterial taxa at the phylum level and genus level are presented in Figure 1. These ﬁndings were generally in line with the ﬁndings of Belstrøm et al., which indicated the ﬁve most predominant genera identiﬁed were Streptococcus, Haemophilus, Prevotella, Rothia, and Neisseria, accounting for around 50% of the identiﬁed OTUs [29]. Nutrients 2020, 12, 966 6 of 15 Figure 1. Relative abundances of the most abundant phyla and genera in each salivary sample in the (A) phylum level and (B) genus level. 3.3. Comparisons of Salivary Bacterial Communities Based on the relative abundance of all the OTUs, salivary bacterial community diversity (expressed by the Shannon and Simpson indexes) was investigated ﬁrst, and the results are shown in Table 2. Compared with baseline (week 0), after eight weeks of tea consumption, a remarkable reduction in the community diversity was noticed across the three subjects, with the exception of the Shannon index of subject 3. Nutrients 2020, 12, x FOR PEER REVIEW 6 of 16 which indicated the five most predominant genera identified were Streptococcus, Haemophilus, Prevotella, Rothia, and Neisseria, accounting for around 50% of the identified OTUs [29]. Figure 1. Relative abundances of the most abundant phyla and genera in each salivary sample in the (A) phylum level and (B) genus level. 3.3. Comparisons of Salivary Bacterial Communities Based on the relative abundance of all the OTUs, salivary bacterial community diversity (expressed by the Shannon and Simpson indexes) was investigated first, and the results are shown in Table 2. Compared with baseline (week 0), after eight weeks of tea consumption, a remarkable Nutrients 2020, 12, 966 7 of 15 Table 2. The temporal changes of the salivary microbial community diversity in each subject. Baseline Week 0 Tea Intervention Follow-Up Week 4 Week 8 Week 12 5.28 ± 0.41 a 0.94 ± 0.02 a 4.68 ± 0.27 ab 0.89 ± 0.03 ab 4.00 ± 0.39 b 0.81 ± 0.04 b 4.17 ± 0.40 b 0.84 ± 0.06 b 4.79 ± 0.58 a 0.91 ± 0.06 a 4.63 ± 0.22 ab 0.88 ± 0.06 a 4.02 ± 0.27 b 0.83 ± 0.05 b 4.37 ± 0.13 ab 0.90 ± 0.02 a Subject 1 Shannon Simpson Subject 2 Shannon Simpson Subject 3 Shannon Simpson 3.93 ± 0.27 a 0.83 ± 0.03 a Values are expressed as the mean ± SD (n = 3). Means with diﬀerent superscript letters (a, b) within a row suggest signiﬁcant diﬀerences (p < 0.05); means with the same superscript letters (a, b) within a row suggest the diﬀerences are not signiﬁcant (p ≥ 0.05), as determined by Duncan’s multiple range test. 3.90 ± 0.17 a 0.80 ± 0.03 b 3.99 ± 0.57 a 0.85 ± 0.10 a 4.07 ± 0.65 a 0.83 ± 0.09 a In order to adequately compare the homogeneity of salivary bacterial communities among the three subjects, MRPP and Anosim tests were then performed. In the pairwise comparisons, positive delta values from MRPP tests and R values from Anosim tests were observed, which indicated a higher similarity within the groups (Table 3). Thus, diversities of salivary microbiota among individuals were much larger than the variation within individuals over the course of tea consumption. Table 3. Summary of multiple response permutation procedure (MRPP) and analysis of similarity (Anosim) tests between each subject. Compared Data Sets MRPP Anosim Delta p-Value R p-Value Subject 1 vs. Subject 2 Subject 1 vs. Subject 3 Subject 2 vs. Subject 3 Subject 1 vs. Subject 2 vs. Subject 3 0.1969 0.1479 0.1919 0.2227 0.001 0.001 0.001 0.001 0.7105 0.4886 0.7562 0.6482 0.001 0.001 0.001 0.001 The general proﬁles of salivary microbiota of each individual subject at diﬀerent sampling times were further compared with PCA (Figure 2). For subject 1, the salivary bacterial communities in the baseline period were separated from the tea intervention and follow-up period, while in the follow-up period bacterial communities gathered with those in the tea consumption period. For subject 2, clear distinctions in the bacterial communities were discovered between week 0 and the other experimental periods, while the bacterial communities in week 12 and week 4 were overlapped. In the case of subject 3, relatively higher similarities were found among the diﬀerent treatment periods, which might suggest a slighter or lower impact of tea consumption on the salivary microbiota. 3.4. Correlation Networks of Salivary Microbiota Based on the Illumina sequencing results, 67 OTUs were deﬁned as the predominant salivary microbiota of the three subjects, with relative abundance over 0.1%. Pearson’s correlations were calculated among the predominant salivary microbiota of each subject, and the strong connections (\|r\| > 0.9 and p < 0.05) were further visualized as networks (Figure 3A,C,E). When comparing the networks of the three subjects, subject 1 had the most complicated co-occurrence patterns of salivary bacteria, with a total strong connection number of 128. For subjects 2 and 3, the strong connection numbers were 49 and 41, respectively. Nutrients 2020, 12, 966 8 of 15 Figure 2. Principal component analysis (PCA) score plots based on the relative abundance of all operational taxonomic units (OTUs) of each subject. 3.5. Hub Salivary Microbiota Identiﬁcation The size of each node in the network represents the number of strong connections with other nodes. Thus, the OTUs with a larger node size were identiﬁed as the hub salivary microbiota of each subject, which had more connections with other bacteria. In this study, approximately 20 hub OTUs from each subject were intentionally selected. In particular, for subject 1, 21 OTUs were deﬁned as the hub microbiota (Figure 3A). The relative abundance changes of these bacteria were further visualized as a heatmap plot (Figure 3B). Of these, 8 OTUs (OTU 133, 23, 42, 5, 6, 7, 8, and 9) increased after tea intervention, while the remaining 13 OTUs decreased; moreover, OTU 1, 42, and 5 increased during the follow-up period (week 12). For subjects 2 and 3, 20 and 25 OTUs were identiﬁed as hub microbiota (Figure 3C,E). The successions of these hub salivary microbiota during the 12-week experimental period are illustrated in Figure 3D,F. Through a Venn diagram, seven OTUs, including OTU_1 (Streptococcus sp.), OTU_133 (Veillonella sp.), OTU_23 (Veillonella sp.), OTU_33 (Ruminococcaceae sp.), OTU_42 (Actinomyces odontolyticus), OTU_6 (Gemella haemolysans), and OTU_696 (Haemophilus sp.), were identiﬁed as the shared hub microbiota of the three subjects (Figure 4A). The unique hub salivary microbiome is also shown in Figure 4A. Based on the relative abundance of these shared hub bacteria during the entire experimental period for the three subjects, a PCA plot was further depicted (Figure 4B). A clear separation of the baseline period (week 0) from other score points was observed, which revealed a signiﬁcant change with regard to these seven OTUs which occurred after tea infusion drinking. The PCA score plots of week 4 and week 8 were gathered into two discrete clusters, which indicated a time-dependent response of these bacteria to tea drinking. For week 12, this cluster was in-between those of week 4 and week 8, indicating a relatively similar bacterial proﬁle pattern in the follow-up period with tea treatment. The temporal shifts of these seven shared hub salivary microbiota during the 12-week experimental period are reﬂected in Figure 5. In general, compared with the baseline period, in week 4, Ruminococcaceae sp. (OTU_33) and Haemophilus sp. (OTU_696) were suppressed signiﬁcantly (p < 0.05), while Veillonella sp. (OTU_133), Actinomyces odontolyticus (OTU_42), and Gemella haemolysans (OTU_6) were promoted signiﬁcantly (p < 0.05). After eight weeks of tea consumption, Streptococcus sp. (OTU_1), Ruminococcaceae sp. (OTU_33), and Haemophilus sp. (OTU_696) were suppressed signiﬁcantly (p < 0.05), while Veillonella spp. (OTU_133 and OTU_23), Actinomyces odontolyticus (OTU_42), and Gemella haemolysans (OTU_6) were promoted signiﬁcantly (p < 0.05). In the follow-up period, only Streptococcus sp. (OTU_1) was return to its initial level (p > 0.05). Nutrients 2020, 12, x FOR PEER REVIEW 8 of 16 Figure 2. Principal component analysis (PCA) score plots based on the relative abundance of all operational taxonomic units (OTUs) of each subject. 3.4. Correlation Networks of Salivary Microbiota Based on the Illumina sequencing results, 67 OTUs were defined as the predominant salivary microbiota of the three subjects, with relative abundance over 0.1%. Pearson’s correlations were calculated among the predominant salivary microbiota of each subject, and the strong connections (\|r\| > 0.9 and p < 0.05) were further visualized as networks (Figure 3A,C,E). When comparing the networks of the three subjects, subject 1 had the most complicated co-occurrence patterns of salivary bacteria, with a total strong connection number of 128. For subjects 2 and 3, the strong connection numbers were 49 and 41, respectively. Nutrients 2020, 12, 966 9 of 15 Figure 3. Correlation networks of the predominant salivary microbiota (A,C,E) and heatmaps of the hub salivary microbiota (B,D,F) in each subject. In correlation networks, each node represents an OTU; the color of nodes indicates the phylum information; the size of nodes represents the number of linkages; lines between nodes represent a strong correlation between these two OTUs (\|r\| > 0.9 and p < 0.05, Pearson’s correlation); red line represents a positive correlation and blue line represents a negative correlation. The nodes with high strong connection numbers were selected as the “hub microbiota” and their dynamic shifts of relative abundance were further depicted on heatmaps. The color of the data matrix in heatmaps corresponds to the normalized relative abundance of the OTUs; the color bar on the top right indicates the scale. Nutrients 2020, 12, x FOR PEER REVIEW 9 of 16 Figure 3. Correlation networks of the predominant salivary microbiota (A, C, and E) and heatmaps of the hub salivary microbiota (B, D, and F) in each subject. In correlation networks, each node represents an OTU; the color of nodes indicates the phylum information; the size of nodes represents the number of linkages; lines between nodes represent a strong correlation between these two OTUs (\|r\| > 0.9 and p < 0.05, Pearson’s correlation); red line represents a positive correlation and blue line represents a negative correlation. The nodes with high strong connection numbers were selected as the “hub microbiota” and their dynamic shifts of relative abundance were further depicted on heatmaps. The color of the data matrix in heatmaps corresponds to the normalized relative abundance of the OTUs; the color bar on the top right indicates the scale. Nutrients 2020, 12, 966 10 of 15 Figure 4. (A) The Venn diagram of the hub salivary microbiota in each subject. (B) PCA score plots based on the relative abundance of the shared hub microbiota across the three subjects. Figure 5. experimental period. The temporal shifts of the shared hub salivary microbiota during the 12-week Nutrients 2020, 12, x FOR PEER REVIEW 10 of 16 3.5. Hub Salivary Microbiota Identification The size of each node in the network represents the number of strong connections with other nodes. Thus, the OTUs with a larger node size were identified as the hub salivary microbiota of each subject, which had more connections with other bacteria. In this study, approximately 20 hub OTUs from each subject were intentionally selected. In particular, for subject 1, 21 OTUs were defined as the hub microbiota (Figure 3A). The relative abundance changes of these bacteria were further visualized as a heatmap plot (Figure 3B). Of these, 8 OTUs (OTU 133, 23, 42, 5, 6, 7, 8, and 9) increased after tea intervention, while the remaining 13 OTUs decreased; moreover, OTU 1, 42, and 5 increased during the follow-up period (week 12). For subjects 2 and 3, 20 and 25 OTUs were identified as hub microbiota (Figure 3C,E). The successions of these hub salivary microbiota during the 12-week experimental period are illustrated in Figure 3D,F. Through a Venn diagram, seven OTUs, including OTU_1 (Streptococcus sp.), OTU_133 (Veillonella sp.), OTU_23 (Veillonella sp.), OTU_33 (Ruminococcaceae sp.), OTU_42 (Actinomyces odontolyticus), OTU_6 (Gemella haemolysans), and OTU_696 (Haemophilus sp.), were identified as the shared hub microbiota of the three subjects (Figure 4A). The unique hub salivary microbiome is also shown in Figure 4A. Based on the relative abundance of these shared hub bacteria during the entire experimental period for the three subjects, a PCA plot was further depicted (Figure 4B). A clear separation of the baseline period (week 0) from other score points was observed, which revealed a significant change with regard to these seven OTUs which occurred after tea infusion drinking. The PCA score plots of week 4 and week 8 were gathered into two discrete clusters, which indicated a time-dependent response of these bacteria to tea drinking. For week 12, this cluster was in-between those of week 4 and week 8, indicating a relatively similar bacterial profile pattern in the follow-up period with tea treatment. The temporal shifts of these seven shared hub salivary microbiota during the 12-week experimental period are reflected in Figure 5. In general, compared with the baseline period, in week 4, Ruminococcaceae sp. (OTU_33) and Haemophilus sp. (OTU_696) were suppressed significantly (p < 0.05), while Veillonella sp. (OTU_133), Actinomyces odontolyticus (OTU_42), and Gemella haemolysans (OTU_6) were promoted significantly (p < 0.05). After eight weeks of tea consumption, Streptococcus sp. (OTU_1), Ruminococcaceae sp. (OTU_33), and Haemophilus sp. (OTU_696) were suppressed significantly (p < 0.05), while Veillonella spp. (OTU_133 and OTU_23), Actinomyces odontolyticus (OTU_42), and Gemella haemolysans (OTU_6) were promoted significantly (p < 0.05). In the follow-up period, only Streptococcus sp. (OTU_1) was return to its initial level (p > 0.05). Figure 4. (A) The Venn diagram of the hub salivary microbiota in each subject. (B) PCA score plots based on the relative abundance of the shared hub microbiota across the three subjects. Nutrients 2020, 12, x FOR PEER REVIEW 11 of 16 Figure 5. The temporal shifts of the shared hub salivary microbiota during the 12-week experimental period. 4. Discussion In the present study, an oolong tea infusion containing a total of 2.83 ± 0.02 g/L polyphenols, including catechin, epicatechin, epigallocatechin gallate, and at least 30 other components, was used to evaluate its salivary microbiota modification effect. Three subjects were required to consume 1.0 L of tea infusions daily, which equaled approximately 52 mg/kg body weight of tea polyphenols. The preparation method and intake amount of tea infusions followed the general tea drinking habits of individuals in the southern part of China, which would provide a more pragmatic and appropriate insight. Under this consumption amount, diverse responds of salivary microbiota were observed among the three subjects. Following the Illumina high-throughput sequencing, a highly diverse salivary bacterial community was observed. A total of 8801 OTUs with a 97% similarity level was identified from the 36 saliva samples and annotated into 25 phyla and 260 genera. In addition to the complexity, a high inter-individual variation in salivary microbiota was also discovered. It was revealed that the salivary microbial communities within the three subjects were significantly distinct from each other, exhibiting host-specific microbiota profiles; their overall collective responses to tea consumption also varied among each participant. The positive delta values from MRPP tests and R values from Anosim tests indicated the differences of salivary microbiota profiles among subjects were far more significant than that among different time points within one subject (Table 3). When all 36 salivary microbiota Nutrients 2020, 12, 966 4. Discussion 11 of 15 In the present study, an oolong tea infusion containing a total of 2.83 ± 0.02 g/L polyphenols, including catechin, epicatechin, epigallocatechin gallate, and at least 30 other components, was used to evaluate its salivary microbiota modiﬁcation eﬀect. Three subjects were required to consume 1.0 L of tea infusions daily, which equaled approximately 52 mg/kg body weight of tea polyphenols. The preparation method and intake amount of tea infusions followed the general tea drinking habits of individuals in the southern part of China, which would provide a more pragmatic and appropriate insight. Under this consumption amount, diverse responds of salivary microbiota were observed among the three subjects. Following the Illumina high-throughput sequencing, a highly diverse salivary bacterial community was observed. A total of 8801 OTUs with a 97% similarity level was identiﬁed from the 36 saliva samples and annotated into 25 phyla and 260 genera. In addition to the complexity, a high inter-individual variation in salivary microbiota was also discovered. It was revealed that the salivary microbial communities within the three subjects were signiﬁcantly distinct from each other, exhibiting host-speciﬁc microbiota proﬁles; their overall collective responses to tea consumption also varied among each participant. The positive delta values from MRPP tests and R values from Anosim tests indicated the diﬀerences of salivary microbiota proﬁles among subjects were far more signiﬁcant than that among diﬀerent time points within one subject (Table 3). When all 36 salivary microbiota data sets from the three subjects were depicted into one PCA plot, no clear cluster was observed (Figure S2). The host-speciﬁc salivary microbiota were also conﬁrmed by the distinct correlation networks of each participant. These results were consistent with the ﬁndings of Belstrøm et al., since the authors conﬁrmed that the ﬁve individuals in their study had a personalized salivary bacterial ﬁngerprint [29]. Hall et al. also stated that the oral bacterial community ﬁngerprint varied from person to person in their study [17]. Thus, in order to minimize the inter-individual variations, the data sets from diﬀerent subjects were analyzed separately, otherwise the eﬀect of tea may be obscured by the inter-individual variations, as shown in Figure S2. In general, it was revealed that oolong tea consumption led to a profound reduction in diversity of the salivary bacterial communities of subject 1 and subject 2. Takeshita et al. stated in their population-based study that good oral health was associated with a lower phylogenetic diversity of the salivary microbiome [5]. Moreover, Vestman et al. reported that the diversity of the tooth bioﬁlm samples was reduced after probiotics supplementation [30]. The increase of diversity of gut microbiota is normally associated with better gut health conditions, such as through the extension of the functional genes for facilitation of the absorption of nutrients and energy, or for appropriate development of immunity. In contrast to the commensal microbiota residents in the intestinal tract, which typically live in harmony with the host, the oral microbiota is responsible for the two most common diseases, including dental caries and periodontal diseases [31]. The increase of diversity in salivary microbiota may be associated with the ﬂourish of dental plaque which resulted from the accumulation of attached bacteria, and thus increase the risk of dental caries and periodontal diseases; while, the decrease in taxonomic diversity in saliva may indicate the shrinking of bacterial communities in dental plaque bioﬁlms, and thus lead to healthier oral ecological conditions. However, for subject 3, the decrease of the salivary microbial community diversity was not signiﬁcant, except for Simpson index of week 8, which might indicate a lower modulation eﬀect of tea on subject 3. Furthermore, according to PCA, signiﬁcant overall shifts of salivary microbiota composition were noted in subjects 1 and 2. However, in the case of subject 3, a higher variation was discovered amongst diﬀerent sampling time points, which may also suggest a lower eﬀectiveness of tea consumption upon the salivary microbiota of subject 3. The oral cavity, as the portal of entry to the gastrointestinal tract, is one of the most complex microbial colony sites within the human body [32]. In order to better understand the complex ecologic system, a correlation network was employed in this study to simplify and visualize the co-occurrence patterns of salivary bacteria. The bacteria taxa with robust connections with other salivary bacteria were deﬁned as “hub salivary bacteria”. Subsequently, via a heatmap plot, the temporal dynamic of Nutrients 2020, 12, 966 12 of 15 each individual hub salivary bacteria was clearly presented. Afterwards, through a Venn diagram, the shared hub microbiota across the three subjects were further identiﬁed. Separated correlation network analysis revealed the detailed inﬂuence of tea consumption on the salivary microbiota composition within the same contactable environment. Thus, it minimized the inter-individual variations between subjects. While, an additional Venn diagram further helped in seeking the common inﬂuences of tea consumption. In particular, seven shared hub OTUs across the three subjects were identiﬁed from the highly complex and personalized oral ecosystem. Notably, OTU_1 (Streptococcus sp.), as the most predominant taxon, also acted as a shared hub microbiota and favorably interacted with other oral bacteria. Due to the bioﬁlm formation and acid production ability of Streptococcus, multiple members of this genus, including Streptococcus mutans, Streptococcus sobrinus, Streptococcus salivarius, Streptococcus constellatus, and Streptococcus parasanguinis, were considered as opportunistic pathogens [33]. With regard to the shifts of Streptococcus sp., a signiﬁcant decrease (−16.94%, p = 0.035) was found after eight weeks of tea consumption. Therefore, a Streptococcus inhibitory eﬀect of tea was observed in this study, and the eﬀect may assist in the prevention of dental caries. There is a preponderance of evidence to support the beneﬁcial role of tea in protecting against this oral pathogen. Narotzki et al. reviewed the clinical and biological studies regarding the correlation between green tea and oral health and concluded that green tea may reduce dental caries through bacterial growth repression and enzyme activity inhibition [14]. With the exception of green tea, it has also been reported that black tea extracts could inhibit S. mutans adhesion in vitro [34]. Kawarai et al. compared the S. mutans bioﬁlm formation inhibitory eﬀect of Assam tea (a black tea variety) and green tea and ascertained that Assam tea exhibited a stronger bioﬁlm inhibition activity than green tea [35]. The inhibitory activity of speciﬁc teas against oral pathogens are commonly attributed to the phenolic components within the tea [14]. Similar inhibitory eﬀects were also observed on OTU_33 (Ruminococcaceae sp.) and OTU_696 (Haemophilus sp.), both of which were also hub microbiota across the three subjects. Haemophilus are a common bacteria which inhabit the mouth, vagina, and intestinal tract. The genus includes commensal organisms, along with some pathogenic species such as H. inﬂuenzae and H. ducreyi. The inhibitory eﬀect of tea on Haemophilus may also reduce the risk of infection. Ruminococcaceae, one of the most typical gut microbiotas, can be found in signiﬁcant numbers in the intestines of humans. However, the biological meaning regarding the depletion of this bacterium induced by tea drinking was not clear. Along with oolong tea consumption, a signiﬁcant elevation of OTU_133 (Veillonella sp.), OTU_23 (Veillonella sp.), OTU_42 (Actinomyces odontolyticus), and OTU_6 (Gemella haemolysans), which were all demonstrated as robust network nodes across the three subjects, was observed in this study. Lim et al. illustrated a signiﬁcant negative association between Haemophilus and Veillonella [36], which was consistent with our ﬁndings. Furthermore, it was reported that the establishment of some certain oral commensals was linked to oral health, such as the bacterial species belonging to Neisseria, Veillonella, and Actinomyces [37], although details regarding the exact mechanisms are not yet available. Moreover, the elevated eﬀect on these four hub bacteria continued throughout the follow-up period, which demonstrated the sustained eﬀect of tea drinking. With regard to the mechanisms behind the modiﬁcation eﬀect of tea on salivary microbiota, several hypotheses have been invoked to account for this particular eﬀect: (i) tea polyphenols possess antimicrobial properties, which are believed to aid in the inhibition of certain bacteria, including some pathogens [13,14]; (ii) tea polyphenols as antioxidants may alleviate oral oxidative stress and inﬂammation, which may further impact the oral immune system and induce a drift of the bacterial community [14]; (iii) tea polyphenols can precipitate salivary proteins and inhibit the activity of salivary alpha-amylase, and thus, induce the decrease of fermentation of carbohydrates involved in caries formation [38]. However, the precise mechanism is still ambiguous, resulting in the necessity for further studies. Numerous epidemiologic studies and clinical trials have validated that regular tea consumption could reduce the risk of cardiovascular disease, including coronary heart disease, stroke, and peripheral arterial disease [39]. Recent studies show a correlation between periodontal disease and Nutrients 2020, 12, 966 13 of 15 cardiovascular disease [40,41]. Thus, from the perspective of alleviating systematic inﬂammatory and immunological processes, explicating the underlying mechanisms (e.g., to link the levels of endogenous mediators, such as endothelin [42] and vitamins [9] of tea consumption may open an innovative avenue toward the development of new antibiotics with good safety and tolerability margin. It was also acknowledged that the inadequate number of subjects in this study might limit the statistical analysis. As explained previously, using limited subjects and following the time course of each individual may help to minimize the inter-individual variations. However, further studies with larger sample sizes are warranted to validate these ﬁndings. 5. Conclusions In summary, using three healthy adult volunteers as our subjects, our study demonstrated that a daily consumption of 1.0 L oolong tea for eight weeks caused a reduction in bacterial community diversity, as well as the disturbance of hub salivary bacterium with strong connections to other salivary microbiota. Additionally, it was also noticed that large inter-individual variations were found, implying diverse responses to oolong tea consumption may exist among subjects. Larger sample sizes and more in-depth mechanism studies are necessary to further clarify and elucidate the physiological relevance of the shifts of salivary microbiota to the oral health of the host. Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6643/12/4/966/s1, Figure S1: Chromatograms obtained from oolong tea infusion, using UHPLC Q-TOF-MS/MS in negative and positive ion modes, Figure S2: PCA score plot based on the relative abundance of all OTUs of the three subjects, Table S1: Summary of the sequencing results of all salivary samples. Author Contributions: Conceptualization, Z.L., W.Z., and L.N.; formal analysis, Z.L. and H.G.; main funding acquisition, L.N.; investigation, Z.L., H.G., and W.Z.; project administration, L.N.; supervision, L.N.; visualization, Z.L.; writing—original draft, Z.L.; writing—review and editing, Z.L., W.Z., and L.N. All authors read and approved the ﬁnal manuscript. Funding: This study was supported by the National Key Research and Development Plan of China (no. 2016YFD0400801). Acknowledgments: The authors thank Jun Lin, Fujian Agriculture and Forestry University, for his excellent technical assistance. Conﬂicts of Interest: The authors declare no conﬂicts of interest. References 1. 2. 3. 4. 5. 6. Jenkinson, H.F.; Lamont, R.J. Oral microbial communities in sickness and in health. Trends Microbiol. 2005, 13, 589–595. [CrossRef] Velden, U.V.D.; Winkelhoﬀ, A.J.V.; Abbas, F.; Graaﬀ, J.D. The habitat of periodontopathic micro-organisms. J. Clin. Periodontol. 1986, 13, 243–248. [CrossRef] Yang, F.; Zeng, X.; Ning, K.; Liu, K.-L.; Lo, C.-C.; Wang, W.; Chen, J.; Wang, D.; Huang, R.; Chang, X. Saliva microbiomes distinguish caries-active from healthy human populations. ISME J. 2012, 6, 1–10. [CrossRef] [PubMed] Lazarevic, V.; Whiteson, K.; Hernandez, D.; François, P.; Schrenzel, J. Study of inter-and intra-individual variations in the salivary microbiota. BMC Genom. 2010, 11, 523. [CrossRef] [PubMed] Takeshita, T.; Kageyama, S.; Furuta, M.; Tsuboi, H.; Takeuchi, K.; Shibata, Y.; Shimazaki, Y.; Akifusa, S.; Ninomiya, T.; Kiyohara, Y. Bacterial diversity in saliva and oral health-related conditions: The Hisayama Study. Sci. Rep. 2016, 6, 22164. [CrossRef] [PubMed] Chinsembu, K.C. Plants and other natural products used in the management of oral infections and improvement of oral health. Acta Trop. 2016, 154, 6–18. [CrossRef] [PubMed] 7. Musarra-Pizzo, M.; Ginestra, G.; Smeriglio, A.; Pennisi, R.; Sciortino, M.T.; Mandalari, G. The antimicrobial and antiviral activity of polyphenols from almond (Prunus dulcis L.) skin. Nutrients 2019, 11, 2355. [CrossRef] [PubMed] Tsou, S.-H.; Hu, S.-W.; Yang, J.-J.; Yan, M.; Lin, Y.-Y. Potential Oral Health Care Agent from Coﬀee against Virulence Factor of Periodontitis. Nutrients 2019, 11, 2235. [CrossRef] 8. Nutrients 2020, 12, 966 14 of 15 9. 10. Isola, G.; Polizzi, A.; Muraglie, S.; Leonardi, R.; Lo Giudice, A. Assessment of Vitamin C and Antioxidant Proﬁles in Saliva and Serum in Patients with Periodontitis and Ischemic Heart Disease. Nutrients 2019, 11, 2956. [CrossRef] Isola, G. Current Evidence of Natural Agents in Oral and Periodontal Health. Nutrients 2020, 12, 585. [CrossRef] 11. Chun, O.K.; Chung, S.J.; Song, W.O. Estimated dietary ﬂavonoid intake and major food sources of US adults. 12. 13. J. Nutr. 2007, 137, 1244–1252. [CrossRef] [PubMed] Ferrazzano, G.F.; Roberto, L.; Amato, I.; Cantile, T.; Sangianantoni, G.; Ingenito, A. Antimicrobial properties of green tea extract against cariogenic microﬂora: An in vivo study. J. Med. Food 2011, 14, 907–911. [CrossRef] [PubMed] Ferrazzano, G.F.; Amato, I.; Ingenito, A.; De Natale, A.; Pollio, A. Anti-cariogenic eﬀects of polyphenols from plant stimulant beverages (cocoa, coﬀee, tea). Fitoterapia 2009, 80, 255–262. [CrossRef] [PubMed] 14. Narotzki, B.; Reznick, A.Z.; Aizenbud, D.; Levy, Y. Green tea: A promising natural product in oral health. Arch. Oral Biol. 2012, 57, 429–435. [CrossRef] [PubMed] 15. Liu, Z.; Chen, Z.; Guo, H.; He, D.; Zhao, H.; Wang, Z.; Zhang, W.; Liao, L.; Zhang, C.; Ni, L. The modulatory eﬀect of infusions of green tea, oolong tea, and black tea on gut microbiota in high-fat-induced obese mice. Food Funct. 2016, 7, 4869–4879. [CrossRef] 16. Liu, Z.; Bruins, M.E.; Ni, L.; Vincken, J.-P. Green and black tea phenolics: Bioavailability, transformation by colonic microbiota, and modulation of colonic microbiota. J. Agric. Food Chem. 2018, 66, 8469–8477. [CrossRef] 17. Hall, M.W.; Singh, N.; Ng, K.F.; Lam, D.K.; Goldberg, M.B.; Tenenbaum, H.C.; Neufeld, J.D.; Beiko, R.; Senadheera, D.B. Inter-personal diversity and temporal dynamics of dental, tongue, and salivary microbiota in the healthy oral cavity. NPJ Bioﬁlm. Microbiomes 2017, 3, 2. [CrossRef] 18. Obanda, M.; Owuor, P.O.; Taylor, S.J. Flavanol composition and caﬀeine content of green leaf as quality potential indicators of Kenyan black teas. J. Sci. Food Agric. 1997, 74, 209–215. [CrossRef] 19. Magoˇc, T.; Salzberg, S.L. FLASH: Fast length adjustment of short reads to improve genome assemblies. Bioinformatics 2011, 27, 2957–2963. [CrossRef] 20. Caporaso, J.G.; Kuczynski, J.; Stombaugh, J.; Bittinger, K.; Bushman, F.D.; Costello, E.K.; Fierer, N.; Pena, A.G.; Goodrich, J.K.; Gordon, J.I. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 2010, 7, 335. [CrossRef] 21. Bokulich, N.A.; Subramanian, S.; Faith, J.J.; Gevers, D.; Gordon, J.I.; Knight, R.; Mills, D.A.; Caporaso, J.G. Quality-ﬁltering vastly improves diversity estimates from Illumina amplicon sequencing. Nat. Methods 2013, 10, 57. [CrossRef] [PubMed] 22. Edgar, R.C. UPARSE: Highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 2013, 10, 996. [CrossRef] [PubMed] 23. DeSantis, T.Z.; Hugenholtz, P.; Larsen, N.; Rojas, M.; Brodie, E.L.; Keller, K.; Huber, T.; Dalevi, D.; Hu, P.; Andersen, G.L. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 2006, 72, 5069–5072. [CrossRef] [PubMed] 24. Chen, T.; Yu, W.-H.; Izard, J.; Baranova, O.V.; Lakshmanan, A.; Dewhirst, F.E. The Human Oral Microbiome Database: A web accessible resource for investigating oral microbe taxonomic and genomic information. Database 2010, 2010. [CrossRef] [PubMed] 25. Bastian, M.; Heymann, S.; Jacomy, M. Gephi: An open source software for exploring and manipulating 26. networks. ICWSM 2009, 8, 361–362. Faust, K.; Lima-Mendez, G.; Lerat, J.-S.; Sathirapongsasuti, J.F.; Knight, R.; Huttenhower, C.; Lenaerts, T.; Raes, J. Cross-biome comparison of microbial association networks. Front. Microbiol. 2015, 6, 1200. [CrossRef] 27. Layeghifard, M.; Hwang, D.M.; Guttman, D.S. Disentangling interactions in the microbiome: A network perspective. Trends Microbiol. 2017, 25, 217–228. [CrossRef] 28. Heberle, H.; Meirelles, G.V.; da Silva, F.R.; Telles, G.P.; Minghim, R. InteractiVenn: A web-based tool for the analysis of sets through Venn diagrams. BMC Bioinform. 2015, 16, 169. [CrossRef] 29. Belstrøm, D.; Holmstrup, P.; Bardow, A.; Kokaras, A.; Fiehn, N.-E.; Paster, B.J. Temporal stability of the salivary microbiota in oral health. PLoS ONE 2016, 11, e0147472. [CrossRef] Nutrients 2020, 12, 966 15 of 15 30. Romani Vestman, N.; Chen, T.; Lif Holgerson, P.; Öhman, C.; Johansson, I. Oral microbiota shift after 12-week supplementation with Lactobacillus reuteri DSM 17938 and PTA 5289; a randomized control trial. PLoS ONE 2015, 10, e0125812. [CrossRef] 31. Wade, W.G. The oral microbiome in health and disease. Pharmacol. Res. 2013, 69, 137–143. [CrossRef] [PubMed] 32. Consortium, H.M.P. Structure, function and diversity of the healthy human microbiome. Nature 2012, 486, 33. 207–214. Jiang, W.; Ling, Z.; Lin, X.; Chen, Y.; Zhang, J.; Yu, J.; Xiang, C.; Chen, H. Pyrosequencing analysis of oral microbiota shifting in various caries states in childhood. Microb. Ecol. 2014, 67, 962–969. [CrossRef] [PubMed] 34. Limsong, J.; Benjavongkulchai, E.; Kuvatanasuchati, J. Inhibitory eﬀect of some herbal extracts on adherence of Streptococcus mutans. J. Ethnopharmacol. 2004, 92, 281–289. [CrossRef] 35. Kawarai, T.; Narisawa, N.; Yoneda, S.; Tsutsumi, Y.; Ishikawa, J.; Hoshino, Y.; Senpuku, H. Inhibition of Streptococcus mutans bioﬁlm formation using extracts from Assam tea compared to green tea. Arch. Oral Biol. 2016, 68, 73–82. [CrossRef] 36. Lim, M.Y.; Yoon, H.S.; Rho, M.; Sung, J.; Song, Y.-M.; Lee, K.; Ko, G. Analysis of the association between host genetics, smoking, and sputum microbiota in healthy humans. Sci. Rep. 2016, 6, 23745. [CrossRef] 37. Paropkari, A.D.; Leblebicioglu, B.; Christian, L.M.; Kumar, P.S. Smoking, pregnancy and the subgingival microbiome. Sci. Rep. 2016, 6, 30388. [CrossRef] 38. Hara, K.; Ohara, M.; Hayashi, I.; Hino, T.; Nishimura, R.; Iwasaki, Y.; Ogawa, T.; Ohyama, Y.; Sugiyama, M.; Amano, H. The green tea polyphenol (−)-epigallocatechin gallate precipitates salivary proteins including alpha-amylase: Biochemical implications for oral health. Eur. J. Oral Sci. 2012, 120, 132–139. [CrossRef] 39. Khan, N.; Mukhtar, H. Tea polyphenols in promotion of human health. Nutrients 2019, 11, 39. [CrossRef] 40. Isola, G.; Alibrandi, A.; Currò, M.; Matarese, M.; Ricca, S.; Matarese, G.; Ientile, R.; Kocher, T. Evaluation of salivary and serum ADMA levels in patients with periodontal and cardiovascular disease as subclinical marker of cardiovascular risk. J. Periodontol. 2019. [CrossRef] Isola, G.; Giudice, A.L.; Polizzi, A.; Alibrandi, A.; Patini, R.; Ferlito, S. Periodontitis and Tooth Loss Have Negative Systemic Impact on Circulating Progenitor Cell Levels: A Clinical Study. Genes 2019, 10, 1022. [CrossRef] [PubMed] Isola, G.; Polizzi, A.; Alibrandi, A.; Indelicato, F.; Ferlito, S. Analysis of Endothelin-1 concentrations in individuals with periodontitis. Sci. Rep. 2020, 10, 1652. [CrossRef] [PubMed] 41. 42. © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_s21248442	Article Manual Annotation of Time in Bed Using Free-Living Recordings of Accelerometry Data Esben Lykke Skovgaard , Jesper Pedersen , Niels Christian Møller, Anders Grøntved and Jan Christian Brønd Centre of Research in Childhood Health, Research Unit for Exercise Epidemiology, Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, 5230 Odense, Denmark; [email protected] (J.P.); [email protected] (N.C.M.); [email protected] (A.G.); [email protected] (J.C.B.) Correspondence: [email protected] Abstract: With the emergence of machine learning for the classiﬁcation of sleep and other human behaviors from accelerometer data, the need for correctly annotated data is higher than ever. We present and evaluate a novel method for the manual annotation of in-bed periods in accelerometer data using the open-source software Audacity®, and we compare the method to the EEG-based sleep monitoring device Zmachine® Insight+ and self-reported sleep diaries. For evaluating the manual annotation method, we calculated the inter- and intra-rater agreement and agreement with Zmachine and sleep diaries using interclass correlation coefﬁcients and Bland–Altman analysis. Our results showed excellent inter- and intra-rater agreement and excellent agreement with Zmachine and sleep diaries. The Bland–Altman limits of agreement were generally around ±30 min for the comparison between the manual annotation and the Zmachine timestamps for the in-bed period. Moreover, the mean bias was minuscule. We conclude that the manual annotation method presented is a viable option for annotating in-bed periods in accelerometer data, which will further qualify datasets without labeling or sleep records. Keywords: sleep; accelerometry; labeling; machine learning; physical activity; human behavior; circadian rhythms; classiﬁcation; sleep/wake cycles; annotation; wearable sensors 1. Introduction Utilizing machine learning with the identiﬁcation of sleep, physical activity behavior, or non-wear from accelerometry data provides the ability to model very complex and non-linear relationships, which is not possible with more simple statistical methods, like multiple linear or logistic regression [1]. However, the use of supervised machine learn- ing algorithms demands large amounts of accurate annotated data to provide sufﬁcient accuracy and to ensure generalizability [2]. Sleep is increasingly recognized as a critical component of the healthy development of children and overall health [3–5], and healthy sleep is generally deﬁned by adequate duration, appropriate timing, good quality, and the absence of sleep disturbances or disorders [6]. Nevertheless, only scarce efforts to assess sleep measures from accelerometer data have been successfully approached with advanced machine learning techniques [7] providing researchers with inexpensive and minimally invasive methods. Valid objective measures using accurate automated scoring of sleep and wake time are important to provide valuable insights into the circadian rhythms on an individual and population-wide basis. The gold standard for objective sleep assessment is polysomnography (PSG), which is based on the continuous recording of electroencephalographic (EEG), electromyographic (EMG), and electrooculographic (EOG) activity via surface electrodes. The assessment of sleep from PSG provides a detailed description of individuals’ sleep architecture through the identiﬁcation of various sleep stages in addition to the more general sleep outcomes, like sleep duration and timing [8]. The PSG method is costly and Citation: Skovgaard, E.L.; Pedersen, J.; Møller, N.C.; Grøntved, A.; Brønd, J.C. Manual Annotation of Time in Bed Using Free-Living Recordings of Accelerometry Data. Sensors 2021, 21, 8442. https://doi.org/10.3390/ s21248442 Academic Editor: Steven Vos Received: 12 November 2021 Accepted: 14 December 2021 Published: 17 December 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Sensors 2021, 21, 8442. https://doi.org/10.3390/s21248442 https://www.mdpi.com/journal/sensors sensors(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Sensors 2021, 21, 8442 2 of 14 burdensome in terms of technician support for sensor application/removal, overnight monitoring (for in-lab PSG), and manual record scoring, in addition to being intrusive for the patient due to the necessity of wearing multiple sensors on the scalp and face throughout the night. Recent studies have attempted to identify PSG-assessed sleep-wake classiﬁcation with wrist acceleration using machine-learning techniques [7,9,10]. One study by Sundararajan et al. [9] attempted this using a random forest machine learning algorithm. The results from the study revealed an F1 score of 73.9% as an estimate of overall accuracy for the identiﬁcation of sleep-wake stages. Moreover, the study showed a large false discovery rate (i.e., the true wake time was predicted as sleep by the algorithm) for the identiﬁcation of sleep. This may primarily be a result of the intrinsic limitations of a wrist-worn accelerometer incorrectly classifying quiet wakefulness as sleep and, secondly, may be explained by the number of participants and the use of single-night PSG-recordings restricting information on inter- and intraindividual variation. Thus, the poor false discovery rate could potentially be balanced out by increasing the number of subjects in the study and/or by increasing the number of consecutive days of recordings for more information on the variation in the movement behavior of the subjects during sleep hours. Accelerometry provides researchers with an inexpensive and minimally invasive method, which has the potential to play an important role in the assessment of sleep duration and timing characterization since it is more practical and suitable than PSG for prolonged recordings (i.e., multiple consecutive days) in non-laboratory settings [11]. How- ever, the limitations of accelerometry must be acknowledged, which is also emphasized by the poor results for the identiﬁcation of sleep vs. wakefulness with wrist acceleration presented with the study by Sundararajan et al. Thus, accelerometry does impose certain limitations as a methodology to identifying subject’s sleep behavior and developing new algorithms should not focus on sleep staging but rather the sleep timing (bedtime/wake-up time) and speciﬁcally the sleep/wake state of the subject. Moreover, the accurate identiﬁcation of the sleep-wake state from accelerometry is most optimally approached with accelerometry recordings covering at least 7–10 days of measurement for each subject to ensure the appropriate day-to-day variation and movement behavior during sleep hours. Developing supervised machine learning algorithms to identify the sleep/wake cycle from multiple days of accelerometry recordings of individuals requires the annotation of the data to identify time in bed and speciﬁcally when the participants go to bed and when they get out of bed. Although there is an obvious distinction between time in bed and actual sleep time, accelerometry as a surrogate measure of sleep is widely used in the literature [12–15] due to the many practical advantages of using accelerometry compared to more intricate methodologies for the detection of sleep. The time in bed annotation could be established from individual sleep diaries, EEG- based recordings [16], systems for the recording of tracheal sounds [17,18], etc.; however, such additional data are not recorded in conjunction with accelerometry within many studies, which leaves a substantial data resource ready for enrichment. If individual time in bed periods can be accurately annotated without additional data for correct labeling, it would provide the option to use the accelerometer data to facilitate the improvement of existing algorithms or the development of new supervised machine learning algorithms. Currently, there are no accurate or easy-to-use methods for the manual annotation of time in bed from accelerometry. The aims of the present study were to (1) describe a method for the manual annotation of subjects’ individual bedtime and time out of bed with raw unprocessed accelerometry, (2) evaluate the accuracy of the manual annotation to predict time in bed/out of bed obtained using a single channel EEG-based sleep staging system and a sleep diary, and (3) to evaluate the inter- and intra-rater reliability of annotations. Sensors 2021, 21, 8442 3 of 14 2. Materials and Methods 2.1. Study Population Data for the current study originates from the SCREENS pilot trial (www.clinicaltrials. gov (accessed on 15 May 2020), NCT03788525), which is a two-arm parallel-group cluster- randomized trial with two intervention groups and no control group [19,20]. Data were collected between October 2018 and March 2019. The collection of data was reported to the local data protection department SDU RIO (ID: 10.391) in agreement with the rules of the Danish Data Protection Agency. Families in the municipality of Middelfart in Denmark were invited to participate if they had at least one child aged 6–10 years residing in the household (n = 1686). Based on survey responses, families were eligible to participate if the contacted parent’s total screen media use was above the median amount (2.7 h/day) based on all respondents (n = 394) and if all children in the household were older than 3.9 years. The latter was to avoid potential disturbances of sleep measurement due to an infant or toddler’s pattern of nocturnal awakening. For further details on inclusion and exclusion criteria see Pedersen et al. [21] In total, data from 14 children and 19 adults were included in the present study. The included participants were not instructed to change their sleep and bedtime behavior as a part of the interventions. The napping behavior, if any, of the participants was irrelevant to the current study as we focused on their nightly sleep time monitored by the EEG-based sleep staging system. 2.2. Actigraphy Both adults and children underwent 24-h accelerometry recording using two Axivity AX3 (Axivity Ltd., Newcastle upon Tyne, UK) triaxial accelerometers. The Axivity AX3 is a small (dimensions: 23 mm × 32.5 mm × 7.6 mm) weighing only 11 g. Sensitivity was set to ±8 g and the sampling frequency to 50 Hz. The accelerometers were worn at two anatomical locations; one ﬁxated to the body in a pocket attached to a belt worn around the waist, where the sensor was placed on the right hip with the USB connector facing away from the right side of the body. A second belt was worn around the right thigh midway between the hip and the knee, where the accelerometer was placed in a pocket with the USB connector facing away from the body. The devices were worn for 1 week (seven consecutive days) at baseline and at follow-up, which corresponds to the recommended number of days required to reliably estimate habitual physical activity [22]. 2.3. Zmachine® Insight+ Concurrent with the accelerometer recordings, both adults and children underwent sleep assessments for 3–4 nights at baseline and 3 nights at follow-up using the Zmachine® (ZM) Insight+ model DT-200 (General Sleep Corporation, Cleveland, OH, USA), Firmware version 5.1.0) concurrently with the actigraphy recording. The device measures sleep by single-channel EEG from the differential mastoid (A1–A2) EEG location on a 30-s epoch basis. The sleep apparatus is developed for use in a free-living setting for objective mea- surement of sleep, including measurement of sleep duration and sleep stage classiﬁcation, as well as computation of sleep-speciﬁc quantities, e.g., latency to the respective sleep stages. The algorithm in ZM has been compared to polysomnography (PSG) in adults with and without chronic sleep issues within a laboratory setting [23,24], and we found that ZM can be feasibly applied to children and adults for multiple days of measurements in free-living [12]. The ZM device demonstrated a high degree of validity for detecting sleep versus awake with a sensitivity, speciﬁcity, positive predictive value, and negative predictive values of 95.5%, 92.5%, 98%, and 84.2%, respectively [24]. Three electrodes (Ambu A/S, Ballerup, Denmark, type: N-00-S/25) are mounted on the mastoids (signal) and the back of the neck (ground). Thirty minutes before the participants plan to go to bed to sleep, the skin areas are cleansed with an alcohol swab, Sensors 2021, 21, 8442 4 of 14 and then electrodes are attached to the skin. An EEG cable connects the three electrodes to the ZM device, where a sensor check is performed to detect whether one or more electrodes are not mounted correctly. If there are sensor problems, these are solved swiftly by a simple change of said electrodes. Participants also reported their bedtimes and time of awakening each day using a prospective daily diary. Parents reported the bedtimes and times of awakening on behalf of their child. 2.4. Audacity Audacity is a free software tool that was developed for audio editing. The software project was originally started by Dominic Mazzoni and Roger Dannenberg in the fall of 1999 as part of a research project at Carnegie Mellon University. Audacity was ini- tially released as an open-source audio editor in May 2000. Since then, the software has been community-developed adding hundreds of features, providing complete support for professional-quality 24-bit and 32-bit audio, a complete Manual and support for many different languages, and millions of copies have been distributed. Today, Audacity is being maintained by a team of volunteers located in many different countries. Audacity is distributed under the terms of the GNU General Public License. Everyone is free to use this application for any personal, educational, or commercial purpose. In the context of analyzing accelerometer data, Audacity is especially useful as it enables the researchers to effortlessly inspect high-resolution raw accelerometer data with a high degree of precision. It is possible to rapidly zoom in to inspect portions of the accelerometer recording in detail (e.g., to inspect certain behavior around bedtime) or zoom out to get an overview (e.g., a whole week). Moreover, Audacity provides a high-resolution labeling function that can be used for the annotation of the accelerometry data. All labels created can be stored in a separate ﬁle and subsequently used in the ML algorithm. The ability to manually inspect high-resolution raw accelerometer data at the level of detail provided by Audacity is, to the knowledge of the authors of the current study, unprecedented in other software. In the Audacity software, it is possible to combine more than 100 channels of data, which provides the ability to combine different signal features derived from the acceleration. Adding multiple signal features together provides an interesting option that might facilitate the visual interpretation and classiﬁcation of the underlying behavior. However, adding too many signal features could have negative consequences for the accurate identiﬁcation of the behavior of interest. We combined a total of seven independent signal features. The classiﬁcation of “lying” within the ﬁrst feature is derived as follows: the inclination of the hip accelerometer exceeds 65 degrees, and the thigh accelerometer is concurrently classiﬁed as “sitting” by the activity type classiﬁcation algorithm by Skotte et al. [25]. The remaining signal features, excluding “time”, were selected from algorithm by Skotte et al. directly. All included features are summarized and described in Table 1. The accelerometry signal features are described with respect to the longitudinal axis of the body. The features generated from the accelerometry data is processed using a window length of two seconds (60 samples) and a 50% overlap (30 samples), providing a resolution of one second. The algorithm by Skotte et al. and the algorithm producing the ﬁrst feature are solely based on the inclination(s) of the accelerometer(s) and, as such, can be used not only to assess the time in bed but also the posture of the participants. Therefore, this is not a precise indicator of the to-bed/out-of-bed timestamps. Figures 1 and 2 shows examples of the visual Audacity interface with all seven signal features as listed in Table 1. Figure 1 is a seven-day overview, and Figure 2 presents a zoomed view of approximately 24 h and with a single annotated night. Sensors 2021, 21, 8442 5 of 14 Table 1. Signal features for the detection of in-bed periods in Audacity. Description Values Visual Interpretation Name Lying The classiﬁcation of lying based on the thigh and back 1: lying −1: not lying Activity The classiﬁcation of activity type Time Time categorized into four-hour windows Thigh-SDacc Standard deviation of the acceleration on the longitudinal axis of the thigh Lying position This feature will guide the rater to identify periods of activity prior to correct bedtime Time of day 1: Standing, moving, or walking 0: Sitting −1: Other activity −1: 00:00–04:00 and so on throughout the 24 h cycle −1: No movement Proportion of leg movement Thigh-Inclination Inclination angle of the thigh device in relation to the longitudinal axis of the thigh The −1 to 1 range represents the −180 to 180 degrees inclination angle Inclination angle of the thigh Hip-SDacc Standard deviation of the hip acceleration on the longitudinal axis of the torso −1: No movement Proportion of whole-body movement Hip-Inclination Inclination of the hip device in relation to the longitudinal axis of the torso The −1 to 1 range represents the −180 to 180 degrees inclination angle Inclination angle of the body/hip Figure 1. Screenshot of the Audacity interface showing the seven horizontal panels representing the included signal features. See Table 1 for a detailed description of the features. Sensors 2021, 21, 8442 6 of 14 Figure 2. Screenshot of the Audacity interface when zoomed in on a single night for the labeling of the in-bed period. The seven horizontal panels represent the included signal features. See Table 1 for a detailed description of features. 2.5. Manual Annotation by the Raters The three raters were all researchers who had prior experience working with ac- celerometer data, and thus, had some understanding and knowledge on how to interpret the different data channels available. The raters labeled each wav-ﬁle independently of each other with in-bed and out-of-bed timestamps and exported the corresponding labels as text ﬁles. Each ﬁle was labeled twice (round 1 and round 2) for test–retest purposes. The raters were at no time aware of previous annotations made by themselves or by the other raters. 2.6. Ground Truth The ZM ground truth labels of the time in-bed and out-of-bed events were derived from the sleep staging data of the ZM as the ﬁrst and last non-sensor-problem event for the night. If the ZM reported the beginning or the end of the recording as having sensor problems, the corresponding night was discarded from further analysis. Sensor problems most commonly occur due to poor attachment of the electrodes. All subjects were instructed to attach the ZM and turn it on at the exact time when the participants went to bed and to remove upon awakening. The timestamps of these events were used as the ground truth values. 2.7. Statistical Analysis All statistical analyses were performed using R statistical (R Core Team, Vienna, Austria) software version 4.0.2 (22 June 2020), RStudio (RStudio Inc., Boston, MA, USA) version 1.1.456. Descriptive characteristics were computed using medians and interquartile ranges for continuous variables and proportions for categorical variables. Characteristics are presented separately for children and adults. Agreement analyses were performed using intraclass correlation coefﬁcient (ICC) and Bland–Altman analysis. Furthermore, to illustrate the overall agreement and symmetry of methods, probability density distribution plots are shown. The ICC is an index that, con- Sensors 2021, 21, 8442 7 of 14 trary to Pearson correlation, assesses not only how well correlated the two techniques are but also if they are equal. An ICC < 0.5 indicates poor agreement, 0.5 < ICC > 0.75 indicates moderate agreement, 0.75 < ICC > 0.9 indicates good agreement, and ICC > 0.90 indicates excellent agreement [26]. In the current study, interpretations of the ICCs are based on the corresponding 95% conﬁdence intervals in accordance with guidelines [26]. Bland–Altman analyses allow examining the degree of agreement between two measurement techniques [27]. The mean of the differences between two techniques (representing the mean bias) and limits of agreement (which are deﬁned as a deviation from the mean superior to two standard deviations) are calculated. A positive bias/mean difference indicates an underestimation (earlier) of the to-bed or out-of-bed timestamp, while a negative difference indicates an overestimation (later) compared to ZM. 3. Results Descriptive characteristics of the included subjects of the current study are reported in Table 2. Table 2. Descriptive characteristics of the study participants. Population (N = 33) Children N Gender (% female) Age (years) Adults N Gender (% female) Age (years) ISCED 0–3 (%) 4–6 (%) 7–8 (%) 14 28.6 9 (7–10) 19 57.9 42 (39–46) 36.8 47.4 15.8 ISCED, International Standard Classiﬁcation of Education 3.1. Intraclass Correlation Coefﬁcient Analyses The ICC analyses highlighted an excellent agreement between ZM and manual in-bed annotation for time to bed and time out of bed across both rounds 1 and 2 and at the baseline and follow-up with the lower limits of the conﬁdence intervals all above 0.9 (see Table 3). Table 3. Intraclass correlation coefﬁcients between ZM and the average of the manual annotations between the three raters. Baseline (n = 94 Nights) Follow-Up (n = 54 Nights) Round 1 ICC (95% CI) Round 2 ICC (95% CI) Round 1 ICC (95% CI) Round 2 ICC (95% CI) To bed Out of bed 0.98 (0.98; 0.99) 0.98 (0.97; 0.99) 0.98 (0.96; 0.98) 0.98 (0.96; 0.98) 0.96 (0.94; 0.98) 0.98 (0.97; 0.99) 0.95 (0.92; 0.97) 0.97 (0.95; 0.98) Round 1 and round 2 refers to the ﬁrst and second round of annotation. Excellent agreement was also observed between self-report and ZM for both baseline data and follow-up data, which is indicated by the lower limit of the 95% conﬁdence interval has values no less than 0.94 (see Table 4). Sensors 2021, 21, 8442 8 of 14 Table 4. Intraclass correlation coefﬁcients between self-report and ZM. To bed Out of bed Baseline (n = 94 Nights) Follow-Up (n = 54 Nights) ICC (95% CI) 0.98 (0.98; 0.99) 0.98 (0.97; 0.99) ICC (95% CI) 0.96 (0.94; 0.98) 0.98 (0.96; 0.99) The ICCs of the agreement between the three manual raters’ ability to annotate the to bed and out of bed timestamps showed good to excellent agreement as seen by the lower limits of the 95% confidence intervals no less than 0.88. A slight tendency of difference of the ICCs can be seen when comparing the to-bed to the out-of-bed timestamps (see Table 5). Table 5. Intraclass correlation coefﬁcients between manual raters. Baseline (n = 110 Nights) Follow-Up (n = 62 Nights) Round 1 ICC (95% CI) Round 2 ICC (95% CI) Round 1 ICC (95% CI) Round 2 ICC (95% CI) To bed Out of bed 0.91 (0.88; 0.94) 0.93 (0.9; 0.95) 0.92 (0.89; 0.94) 0.97 (0.96; 0.98) 0.94 (0.9; 0.96) 0.97 (0.96; 0.98) 0.97 (0.95; 0.98) 0.98 (0.98; 0.99) Round 1 and round 2 refers to the ﬁrst and second round of annotation. The ICCs for the test–retest reliability showed good to excellent agreement for each rater between rounds 1 and 2 for both baseline data and follow-up data (see Table 6). This is seen by the lower limits of the 95% conﬁdence intervals values of no less than 0.86. Although the ICCs are similar, it seems that raters 1 and 3 showed lower agreement when annotating the baseline to-bed timestamp compared to the later annotations, whereas the ICC scores of rater 2 did not elicit this behavior. Table 6. Test–retest intraclass correlation coefﬁcients between the ﬁrst and second round of manual annotations. Baseline (n = 110 Nights) Follow-Up (n = 62 Nights) To Bed ICC (95% CI) Out of Bed ICC (95% CI) To Bed ICC (95% CI) Out of Bed ICC (95% CI) Rater 1 Rater 2 Rater 3 0.91 (0.87; 0.94) 0.97 (0.96; 0.98) 0.91 (0.87; 0.94) 0.98 (0.98; 0.99) 0.91 (0.87; 0.94) 0.96 (0.94; 0.97) 0.96 (0.94; 0.98) 0.91 (0.86; 0.95) 0.98 (0.97; 0.99) 1.00 (0.99; 1.00) 0.99 (0.98; 0.99) 0.98 (0.97; 0.99) 3.2. Bland–Altman Analyses The bias and upper and lower limits of agreement with corresponding conﬁdence intervals for the comparison of the manual annotation and self-report in relation to ZM are presented in Table 7. Biases for the manual annotation compared to ZM is in the range of -6 min to 5 min, while self-report produced slightly lower biases in comparison to ZM. Generally, the limits of agreement seem to be of the same magnitude regardless of the method comparison. Sensors 2021, 21, 8442 9 of 14 Table 7. Bland–Altman analysis of inter-method agreement between manual annotation and ZM as well as self-report and ZM. All estimates are in minutes. Method Bias (95% CI) Upper LOA (95% CI) Lower LOA (95% CI) Baseline, to bed (n = 94) Manual, round 1 Manual, round 2 Self-report Baseline, out of bed (n = 94) Manual, round 1 Manual, round 2 Self-report Follow-up, to bed (n = 54) Manual, round 1 Manual, round 2 Self-report Follow-up, out of bed (n = 54) Manual, round 1 Manual, round 2 Self-report 3.02 (−0.44; 6.47) 0.48 (−2.42; 3.39) 1.23 (−1.57; 4.03) 0.53 (−2.34; 3.4) 0.98 (−1.47; 3.43) −2.79 (−5.26; −0.32) −6.08 (−11.34; −0.83) −0.4 (−5.3; 4.51) 0.77 (−4.08; 5.62) 4.95 (0.65; 9.25) 2.57 (−0.76; 5.89) 0.56 (−3.62; 4.74) −30.04 (−35.96; −24.12) −27.3 (−32.28; −22.32) −25.56 (−30.37; −20.76) −26.9 (−31.82; −21.99) −22.49 (−26.7; −18.28) −26.45 (−30.69; −22.21) −43.81 (−52.84; −34.77) −35.6 (−44.03; −27.17) −34.06 (−42.4; −25.72) −25.95 (−33.35; −18.55) −21.3 (−27.02; −15.59) −29.45 (−36.64; −22.26) 36.07 (30.15; 42) 28.27 (23.29; 33.24) 28.02 (23.21; 32.82) 27.96 (23.05; 32.88) 24.45 (20.24; 28.66) 20.87 (16.63; 25.11) 31.64 (22.61; 40.67) 34.8 (26.37; 43.23) 35.59 (27.25; 43.93) 35.85 (28.45; 43.25) 26.44 (20.72; 32.15) 30.57 (23.39; 37.76) 3.3. Density Plots The probability density distribution for the difference between the to-bed and out- of-bed scoring for the manual annotation and self-report compared to ZM is shown in Figure 3. These plots function as a visual representation of the bias and spread around zero of the manual annotations and self-report in comparison to ZM as seen previously [10]. Figure 3. Probability density distributions for differences between manual in-bed annotations and self-report compared to ZM. 4. Discussion This is the ﬁrst study to describe a method for the manual annotation of in-bed periods with accelerometry data and to evaluate the accuracy of the method with multiple raters and compared to sleep assessed with EEG -based methodology Zmachine Insight+® and self-reported sleep as reference methodologies. When all interpretations of the ICC analyses were based on the lower limit of the 95% conﬁdence interval, our results showed (1) good- to-excellent interrater reliability, (2) the test–retest reliability (or intra-rater reliability) Sensors 2021, 21, 8442 10 of 14 showed good to excellent agreement for all three raters between their ﬁrst and second round of in-bed annotations, (3) compared to ZM, the average of the manual in-bed annotation method for all three raters showed agreements ranging from good to excellent, and (4) the agreement between the self-reported in-bed timestamps and ZM were good to excellent. Furthermore, the Bland–Altman analysis revealed that the mean bias of the manual annotation and self-report compared to ZM was within ±6 min with LOA no larger than a span of ±45 min. Finally, the probability density distribution plots of the differences between the in-bed estimates of the manual raters and self-report compared to ZM were comparable in terms of the symmetry, spread around zero, and positioning of outliers. The excellent performance of the prospective sleep diaries in the current study may be explained by the synchronized use of ZM and the sleep diaries. Thus, having the subject manually initiate and end the ZM recording every morning and night will make it easier for the subject to recall the time going to bed and time out of bed, thus, avoiding much of the usual discrepancy between objectively and subjectively measured sleep duration [28]. We would not expect to see such good agreement between the manual annotation and sleep diaries as well as ZM and sleep diaries if the participants wre instructed to exclusively log sleep using subjective measurements without protocol disturbances as anchor time points. When compared to ZM, we found that the manual annotation of the in-bed period deviated more when estimating the going to bed timestamp. This could be caused by the difﬁculty the raters had in discriminating between inactive behaviors before bedtime and actual time in bed. However, these discrepancies may be minimized by further training of the rater’s ability to distinguish between inactive behaviors; however, this poses the most important limitation to the manual annotation method. Nevertheless, the accuracy obtained in the present study is reassuring as it is achieved based on little preliminary formal training or brieﬁng of the raters involved. In that sense, most of the work when determining the in-bed period is self-explanatory when provided with the information, given by the signal features selected in the present study in Audacity. This is further supported by the excellent ICC between the three manual raters. However, there appears to be a slight learning curve as the LOAs are consistently narrower during round 2 of the manual scoring compared to round 1. This is also evident in the density plots, which display a greater spread around zero during round 1 compared to round 2. This suggests that more than two rounds of manual scoring may homogenize the results further or that the raters may beneﬁt from revisiting the annotations from the beginning of the ﬁrst round of annotations. Alternatively, a form of training may be proﬁtable before the actual annotation takes place. Further research is warranted to investigate methods to optimize the homogenization of the manual annotations. Nevertheless, the evidence suggests that supervised machine learning, given a large amount of labeled data, is resistant to label noise [29], which means that the tradeoff for accuracy in favor of the sheer volume of labeled data may be preferable. This further advocates for the use of the manual annotation methodology in data sets with no self-reported sleep or other measures of interest without labels. There are other labeling tools for annotating time series data (e.g., Label Studio (Heartex Organization, San Francisco, CA, USA) [30] or Visplore (Visplore GmbH, Vienna, Austria) [31]); however, we found that Audacity was well suited for this speciﬁc task. Label Studio, for instance, may have difﬁculties handling week-long accelerometer data with 100 M+ entries, and Audacity is perfectly suited with its ability to seamlessly handle and navigate very large data structures. Furthermore, our feature selection was based on domain knowledge with the purpose of providing the right combination of features in limited number to avoid overﬂowing the raters with redundant information. This methodology can be extended to other behaviors, e.g., walking, which would likely require a different set of features. Although we do not provide clear-cut guidelines for the process of annotating the data, the raters in the present study were able to gain the right insights. Furthermore, the labeling of data is a step that inherently requires common knowledge in human behavior, and if the labels can sufﬁciently be described based on formal rules, Sensors 2021, 21, 8442 11 of 14 one can question whether the training of an AI model is necessary at all. Nevertheless, we suggest that further research investigating which features are the most important for successful annotations and, likewise, examine the effect of other sets of features that might provide important knowledge that could further facilitate the use of manual annotation of accelerometry time series data. To date, most studies that have investigated the validity of actigraphy and self-report compared to PSG or EEG-based methodologies, have routinely evaluated sleep parameters, such as the total sleep time, wake after sleep onset, sleep latency, and sleep efﬁciency. These measures are often an aggregate measure that includes sleep onset and wake onset, which would be comparable to the to-bed and out-of-bed timestamps in the current study. Moreover, the precision of these time points is not evaluated and, thus, is difﬁcult to compare to the measurements of the current study. The novel methodology for annotating the time in bed in the present study, however, provides ICC values on par with or better than previous studies comparing actigraphy sleep parameters to PSG [7,32]. Furthermore, one study presented mean absolute errors of 39.9 min and 29.9 min for the sleep onset time and waking up time, respectively, and 95% limits of agreement above ±3 h for sleep duration when comparing an algorithm to PSG [10]. Furthermore, the ability to precisely estimate the timestamp of certain events compared to durations of speciﬁc behaviors leaves less room for error in the effort to obtain good agreements. Additionally, our manual methodology performs strongly across age and gender as the included subjects in the present study consist of both children and adults of both genders. This suggests that our manual annotation method is accurate irrespective of the inclusion of different developmental age groups and genders and their speciﬁc behavior and that it may be a more precise tool for estimating exact time points compared to present state automated methodologies. Traditionally, the accuracy of the assessment of sleep parameters is highly dependent on the target population, and thus we view the current results with plausible high generalizability to populations of normal sleepers. Identifying periods of sleep (rather than simply lying) is an important component of a 24 h behavior proﬁle, and many studies examining sleep detection based on conventional accelerometers involve asking the participant to record their time in bed, sleep onset, and waking up time [33–35]. The use of self-reported measures of sleep may be replaced by annotating in-bed manually, thereby, lowering the participant burden and avoiding the inevitable recall bias associated with self-reported measures. Therefore, an important application of the manual annotation methodology using Audacity is that it can, with no difﬁculty, be employed on free-living data. Moreover, the application of our proposed methodology is manifold. Other suitable use cases could be the annotation of non-wear time, manual clock synchronization of several different devices, examining the validity of raw data, and more. Additionally, it is not limited to actigraphy data but can be utilized on a wide variety of multi-channel data for an increased overview, including orientation (gyroscopic data), temperature, battery voltage, and light as examples. Finally, Audacity provides a ﬂuid workﬂow even with very large multi-channel data ﬁles with the ability to swiftly zoom to every resolution needed and scroll through time with no lag and add labels that advocate the use of Audacity as a standard tool for researchers working with raw data and machine learning. For these purposes, the implementation of the Audacity- methodology on raw accelerometer data may help drive the development of future human behavior research. Although the work of changing from raw sensor data to operational predictive models using labelled data has been the standard method for years, no previous studies have proposed a methodology that enables researchers to make optimal use of their available accelerometry data. We show that with a few well-selected features, the annotation of sleep is comparable to EEG-based sleep classiﬁcation hardware. However, it is important to note that we do not currently recommend that our proposed methodology of manually annotating in-bed time is to be used as a replacement of other more well-established techniques for estimating sleep, e.g., EEG- or tracheal-sound-based options in ongoing Sensors 2021, 21, 8442 12 of 14 studies. Though, it may serve as a post-hoc procedure to enrich already collected data with a measure of sleep. The strengths of this study include the continuous data collection of both accelerome- try, sleep diary, and ZM in the home of the participants during multiple consecutive days of recordings making it high-quality free-living data. The limitations include the limited rater generalizability as the three manual raters were ﬁxed and not randomly chosen from a larger population of eligible raters to accommodate different characteristics. However, due to the scarce pre-brieﬁng instructions on how to label the raw data, we suggest that this methodology is generalizable to at least other researchers working with accelerometer data. A natural next step would be to develop and validate a procedure similar to what is available with sleep annotation with EEG (AASM Scoring Manual [36]) and propose guidelines to make the manual annotation methodology accessible to persons with limited experience within the ﬁeld of accelerometry. To record true free-living behavior using participant-mounted devices is difﬁcult and wearing the ZM during sleep may affect the behavior of the participants and, thus, poses as a limitation of the study. Additionally, although the criterion measure in the current study is validated against PSG, it would have been more optimal to use PSG as a criterion measure. Finally, we did not incorporate napping behavior in the current study as we focused on sleep in relation to circadian rhythms. Further research is needed to validate the manual methodology for use in detecting napping behavior. 5. Conclusions In conclusion, our results show that the manual annotation of the in-bed period from thigh- and hip-worn accelerometer data using Audacity demonstrated good agreement with a minimal mean bias and acceptable limits of agreement for the time to bed and out of bed when compared to the same estimates assessed with the use of an objective EEG-based sleep device and prospective sleep dairies. Furthermore, the manual annotation was highly reliable with excellent inter- and intra-rater agreement and has an accuracy with the EEG-based assessment similar to the sleep diaries. The study shows that the manual annotation can be used on already collected raw data when not accompanied by sleep records. This will facilitate the additional use of free-living data resources and, thus, could increase the amount of available training data when employing data-demanding machine learning algorithms. This has the potential to improve the generalizability of these algorithms in assessing human behavior from objective recordings. Author Contributions: Conceptualization, N.C.M., E.L.S. and J.C.B.; methodology, N.C.M., E.L.S. and J.C.B.; validation, J.P., J.C.B. and E.L.S.; formal analysis, E.L.S.; investigation, E.L.S., J.P. and J.C.B.; data curation, A.G. and J.P.; writing—original draft preparation, E.L.S.; writing—review and editing, J.C.B., J.P., N.C.M. and A.G.; visualization, E.L.S.; supervision, J.C.B.; funding acquisition, A.G. All authors have read and agreed to the published version of the manuscript. Funding: TrygFonden (grant number 130081) and European Research Council (grant number 716657). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The datasets generated and analyzed during the current study are not publicly available due to the general data protection regulations but will be shared on reasonable request using a safe platform by the corresponding author. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Fiorillo, L.; Puiatti, A.; Papandrea, M.; Ratti, P.-L.; Favaro, P.; Roth, C.; Bargiotas, P.; Bassetti, C.L.; Faraci, F.D. Automated sleep scoring: A review of the latest approaches. Sleep Med. Rev. 2019, 48, 101204. [CrossRef] [PubMed] Sensors 2021, 21, 8442 13 of 14 2. 3. 4. 5. 6. Van der Ploeg, T.; Austin, P.C.; Steyerberg, E.W. Modern modelling techniques are data hungry: A simulation study for predicting dichotomous endpoints. BMC Med. Res. Methodol. 2014, 14, 137. [CrossRef] [PubMed] Chaput, J.-P.; Gray, C.E.; Poitras, V.J.; Carson, V.; Gruber, R.; Birken, C.S.; MacLean, J.E.; Aubert, S.; Sampson, M.; Tremblay, M.S. Systematic review of the relationships between sleep duration and health indicators in the early years (0–4 years). BMC Public Health 2017, 17, 855. [CrossRef] [PubMed] Chaput, J.-P.; Gray, C.E.; Poitras, V.J.; Carson, V.; Gruber, R.; Olds, T.; Weiss, S.K.; Connor Gorber, S.; Kho, M.E.; Sampson, M.; et al. Systematic review of the relationships between sleep duration and health indicators in school-aged children and youth. Appl. Physiol. Nutr. Metab. 2016, 41, S266–S282. [CrossRef] [PubMed] St-Onge, M.-P.; Grandner, M.A.; Brown, D.; Conroy, M.B.; Jean-Louis, G.; Coons, M.; Bhatt, D.L. Impact on Lifestyle Behaviors and Cardiometabolic Health: A Scientiﬁc Statement From the American Heart Association. Circulation 2016, 134, e367–e386. [CrossRef] [PubMed] Gruber, R.; Carrey, N.; Weiss, S.K.; Frappier, J.Y.; Rourke, L.; Brouillette, R.T.; Wise, M.S. Position Statement on Pediatric Sleep for Psychiatrists. J. Can. Acad. Child Adolesc. Psychiatry 2014, 23, 174–195. 7. Haghayegh, S.; Khoshnevis, S.; Smolensky, M.H.; Diller, K.R. Application of deep learning to improve sleep scoring of wrist 8. 9. actigraphy. Sleep Med. 2020, 74, 235–241. [CrossRef] [PubMed] Vaughn, B.V.; Giallanza, P. Technical review of polysomnography. Chest 2008, 134, 1310–1319. [CrossRef] [PubMed] Sundararajan, K.; Georgievska, S.; te Lindert, B.H.W.; Gehrman, P.R.; Ramautar, J.; Mazzotti, D.R.; Sabia, S.; Weedon, M.N.; van Someren, E.J.W.; Ridder, L.; et al. Sleep classiﬁcation from wrist-worn accelerometer data using random forests. Sci. Rep. 2021, 11, 24. [CrossRef] [PubMed] 10. Van Hees, V.T.; Sabia, S.; Jones, S.E.; Wood, A.R.; Anderson, K.N.; Kivimäki, M.; Frayling, T.M.; Pack, A.I.; Bucan, M.; Trenell, M.I.; et al. Estimating sleep parameters using an accelerometer without sleep diary. Sci. Rep. 2018, 8, 12975. [CrossRef] [PubMed] 11. Van de Water, A.T.M.; Holmes, A.; Hurley, D.A. Objective measurements of sleep for non-laboratory settings as alternatives to polysomnography—A systematic review. J. Sleep Res. 2011, 20, 183–200. [CrossRef] [PubMed] 12. Van Hees, V.T.; Sabia, S.; Anderson, K.N.; Denton, S.J.; Oliver, J.; Catt, M.; Abell, J.G.; Kivimäki, M.; Trenell, M.I.; Singh-Manoux, A. A Novel, Open Access Method to Assess Sleep Duration Using a Wrist-Worn Accelerometer. PLoS ONE 2015, 10, e0142533. [CrossRef] 13. Madsen, M.T.; Rosenberg, J.; Gögenur, I. Actigraphy for measurement of sleep and sleep-wake rhythms in relation to surgery. J. 14. Clin. Sleep Med. 2013, 9, 387–394. [CrossRef] [PubMed] Schwab, K.E.; Ronish, B.; Needham, D.M.; To, A.Q.; Martin, J.L.; Kamdar, B.B. Actigraphy to Evaluate Sleep in the Intensive Care Unit. A Systematic Review. Ann. Am. Thorac. Soc. 2018, 15, 1075–1082. [CrossRef] [PubMed] 15. Barouni, A.; Ottenbacher, J.; Schneider, J.; Feige, B.; Riemann, D.; Herlan, A.; Hardouz, D.E.; McLennan, D. Ambulatory sleep scoring using accelerometers—distinguishing between nonwear and sleep/wake states. PeerJ 2020, 8, e8284. [CrossRef] [PubMed] 16. Younes, M.; Raneri, J.; Hanly, P. Staging Sleep in Polysomnograms: Analysis of Inter-Scorer Variability. J. Clin. Sleep Med. 2016, 12, 885–894. [CrossRef] [PubMed] 17. Dafna, E.; Tarasiuk, A.; Zigel, Y. Sleep-Wake Evaluation from Whole-Night Non-Contact Audio Recordings of Breathing Sounds. PLoS ONE 2015, 10, e0117382. [CrossRef] 18. Montazeri Ghahjaverestan, N.; Akbarian, S.; Hafezi, M.; Saha, S.; Zhu, K.; Gavrilovic, B.; Taati, B.; Yadollahi, A. Sleep/Wakefulness Detection Using Tracheal Sounds and Movements. Nat. Sci. Sleep 2020, 12, 1009–1021. [CrossRef] [PubMed] 19. Rasmusen, M.; Pedersen, J.; Olesen, L.; Kristensen, P.; Brønd, J.; Grøntved, A. Feasibility of two screen media reduction interventions: Results from the SCREENS pilot trial. PLoS ONE 2021, 16, e0259657. 20. Rasmussen, M.G.B.; Pedersen, J.; Olesen, L.G.; Brage, S.; Klakk, H.; Kristensen, P.L.; Brønd, J.C.; Grøntved, A. Short-term efﬁcacy of reducing screen media use on physical activity, sleep, and physiological stress in families with children aged 4–14: Study protocol for the SCREENS randomized controlled trial. BMC Public Health 2020, 20, 380. [CrossRef] [PubMed] 21. Pedersen, J.; Rasmussen, M.G.B.; Olesen, L.G.; Kristensen, P.L.; Grøntved, A. Self-administered electroencephalography-based 22. sleep assessment: Compliance and perceived feasibility in children and adults. Sleep Sci. Pract. 2021, 5, 8. [CrossRef] Jaeschke, L.; Steinbrecher, A.; Jeran, S.; Konigorski, S.; Pischon, T. Variability and reliability study of overall physical activity and activity intensity levels using 24 h-accelerometry-assessed data. BMC Public Health 2018, 18, 530. [CrossRef] 23. Wang, Y.; Loparo, K.A.; Kelly, M.R.; Kaplan, R.F. Evaluation of an automated single-channel sleep staging algorithm. Nat. Sci. Sleep 2015, 7, 101–111. [CrossRef] 24. Kaplan, R.F.; Wang, Y.; Loparo, K.A.; Kelly, M.R.; Bootzin, R.R. Performance evaluation of an automated single-channel sleep-wake 25. detection algorithm. Nat. Sci. Sleep 2014, 6, 113–122. [CrossRef] [PubMed] Skotte, J.; Korshøj, M.; Kristiansen, J.; Hanisch, C.; Holtermann, A. Detection of Physical Activity Types Using Triaxial Accelerom- eters. J. Phys. Act. Health 2014, 11, 76–84. [CrossRef] 26. Koo, T.K.; Li, M.Y. A Guideline of Selecting and Reporting Intraclass Correlation Coefﬁcients for Reliability Research. J. Chiropr. Med. 2016, 15, 155–163. [CrossRef] [PubMed] 27. Bland, J.M.; Altman, D.G. Measuring agreement in method comparison studies. Stat. Methods Med. Res. 1999, 8, 135–160. [CrossRef] 28. Aili, K.; Åström-Paulsson, S.; Stoetzer, U.; Svartengren, M.; Hillert, L. Reliability of Actigraphy and Subjective Sleep Measurements in Adults: The Design of Sleep Assessments. J. Clin. Sleep Med. 2017, 13, 39–47. [CrossRef] [PubMed] Sensors 2021, 21, 8442 14 of 14 29. Rolnick, D.; Veit, A.; Belongie, S.; Shavit, N. Deep Learning is Robust to Massive Label Noise. arXiv 2018, arXiv:170510694. 30. Label Studio–Open Source Data Labeling. Available online: https://labelstud.io (accessed on 6 December 2021). 31. Visplore–Software for Visual Time Series Analysis. Available online: https://visplore.com/home (accessed on 6 December 2021). 32. Yavuz-Kodat, E.; Reynaud, E.; Geoffray, M.-M.; Limousin, N.; Franco, P.; Bourgin, P.; Schroder, C.M. Validity of Actigraphy Compared to Polysomnography for Sleep Assessment in Children With Autism Spectrum Disorder. Front. Psychiatry 2019, 10, 551. [CrossRef] [PubMed] 33. Littner, M.; Kushida, C.A.; Anderson, W.M.; Bailey, D.; Berry, R.B.; Davila, D.G.; Hirshkowitz, M.; Kapen, S.; Kramer, M.; Loube, D.; et al. Practice parameters for the role of actigraphy in the study of sleep and circadian rhythms: An update for 2002. Sleep 2003, 26, 337–341. [CrossRef] [PubMed] 34. Lockley, S.W.; Skene, D.J.; Arendt, J. Comparison between subjective and actigraphic measurement of sleep and sleep rhythms. J. Sleep Res. 1999, 8, 175–183. [CrossRef] 35. Girschik, J.; Fritschi, L.; Heyworth, J.; Waters, F. Validation of self-reported sleep against actigraphy. J. Epidemiol. 2012, 22, 462–468. [CrossRef] [PubMed] 36. AASM Scoring Manual-American Academy of Sleep Medicine. Available online: https://aasm.org/clinical-resources/scoring- manual (accessed on 8 November 2021).
10.3390_nu13103343	Article Vitamin D and the Risks of Depression and Anxiety: An Observational Analysis and Genome-Wide Environment Interaction Study Zhen Zhang † Chuyu Pan, Jingxi Zhang , Xuena Yang †, Yumeng Jia, Yan Wen, Shiqiang Cheng, Peilin Meng, Chun’e Li, Huijie Zhang, , Yujing Chen and Feng Zhang * Key Laboratory of Trace Elements and Endemic Diseases of National Health and Family Planning Commission, School of Public Health, Health Science Center, Xi’an Jiaotong University, Xi’an 710000, China; [email protected] (Z.Z.); [email protected] (X.Y.); [email protected] (Y.J.); [email protected] (Y.W.); [email protected] (S.C.); [email protected] (P.M.); [email protected] (C.L.); [email protected] (H.Z.); [email protected] (C.P.); [email protected] (J.Z.); [email protected] (Y.C.) * Correspondence: [email protected] † The two authors contributed equally to this work. Abstract: Previous studies have suggested that vitamin D (VD) was associated with psychiatric diseases, but efforts to elucidate the functional relevance of VD with depression and anxiety from genetic perspective have been limited. Based on the UK Biobank cohort, we ﬁrst calculated polygenic risk score (PRS) for VD from genome-wide association study (GWAS) data of VD. Linear and logistic regression analysis were conducted to evaluate the associations of VD traits with depression and anxiety traits, respectively. Then, using individual genotype and phenotype data from the UK Biobank, genome-wide environment interaction studies (GWEIS) were performed to identify the potential effects of gene × VD interactions on the risks of depression and anxiety traits. In the UK Biobank cohort, we observed signiﬁcant associations of blood VD level with depression and anxiety traits, as well as signiﬁcant associations of VD PRS and depression and anxiety traits. GWEIS identiﬁed multiple candidate loci, such as rs114086183 (p = 4.11 × 10−8, LRRTM4) for self-reported depression status and rs149760119 (p = 3.88 × 10−8, GNB5) for self-reported anxiety status. Our study results suggested that VD was negatively associated with depression and anxiety. GWEIS identiﬁed multiple candidate genes interacting with VD, providing novel clues for understanding the biological mechanism potential associations between VD and psychiatric disorders. Keywords: vitamin D; depression; anxiety; genome-wide association study; polygenic risk score; genome-wide environment interaction study 1. Introduction Psychiatric disorders are a group of complex diseases, which are mainly characterized by varying degrees of obstacles in mental activities such as cognition, emotion, willpower, and behavior [1]. A meta-analysis showed that overall global prevalence of psychiatric disorders was increased with odds ratio of 1.179 (95% CI: 1.065–1.305) [2]. Additionally, another study has found that the prevalence of the common mental illnesses is continuously rising, particularly in low- and middle-income countries, with many people suffering from depression and anxiety disorders simultaneously [3]. Depression and anxiety disorders are the most common mental disorders in the general population [4]. According to the latest report of World Health Organization, it is estimated that there are 322 million people (4.4% of the world’s population) living with depression and more than 260 million people (3.6% of the global population) affected by anxiety disorders [3]. Furthermore, Wang et al. observed that depression and anxiety were signiﬁcantly associated with higher cancer incidence, cancer-speciﬁc mortality and all-cause mortality [5]. Traditional Citation: Zhang, Z.; Yang, X.; Jia, Y.; Wen, Y.; Cheng, S.; Meng, P.; Li, C.; Zhang, H.; Pan, C.; Zhang, J.; et al. Vitamin D and the Risks of Depression and Anxiety: An Observational Analysis and Genome-Wide Environment Interaction Study. Nutrients 2021, 13, 3343. https://doi.org/10.3390/ nu13103343 Academic Editors: Daniel-Antonio de Luis Roman and Ana B. Crujeiras Received: 25 August 2021 Accepted: 21 September 2021 Published: 24 September 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Nutrients 2021, 13, 3343. https://doi.org/10.3390/nu13103343 https://www.mdpi.com/journal/nutrients nutrients(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Nutrients 2021, 13, 3343 2 of 15 treatment of depression and anxiety can only relieve rather than cure the condition, causing a tremendous economic burden on individual families and high suicide rate. Therefore, it is urgent to ﬁnd new ways to treat and prevent depression and anxiety disorders. Vitamin D (VD) is a member of the steroid hormone family. As a necessary vitamin to human body, it is not only cheap, but also widely available. Besides being ingested from daily diet and VD supplements, it can also be synthesized from 7-dehydrocholesterol in the skin through ultraviolet radiation-b (UVB) [6]. Previous studies on VD have found that genetic and environmental factors can affect the status and metabolism of VD. The data from twin and family studies have suggested that circulating VD concentrations are partly determined by genetics, with heritability ranging from 23% to 80% [7]. Furthermore, genetic studies of VD have found that genetic variations and alterations (e.g., deletions, ampliﬁca- tions, inversions) in genes involved in VD metabolism, catabolism, transport, or binding to its receptors may affect VD levels [7]. Moreover, a study of gene-environment interactions with VD has showed that speciﬁc genetic variants associated with VD metabolism may be correlated with prostate cancer risk in VD-deﬁcient patients [8]. As for the biological function of VD, previous studies have demonstrated that VD plays an important role in bone health, reproduction and fertility, and immune function [9]. Growing evidence shows that VD also exerts a great inﬂuence on the development and adult brain, such as maintaining calcium balance and signaling, regulating neurotrophic factors, providing neuroprotection, modulating neurotransmission, and promoting synaptic plasticity [6]. In addition, a recent systematic review has shown that VD deﬁciency in adulthood may also be associated with adverse brain-related outcomes [6]. The discovery of the functions of VD in the brain and the continuous conﬁrmation of the effects of VD on disease provides a new research direction for the study of psychiatric disorders in recent years. Although a lot of research has been devoted to exploring the relationship between VD and depression and anxiety, it is still controversial whether VD was associated with the two mental disorders [10]. Some studies have observed an association between VD and depression symptoms [11–13], while other studies have not found this association [14,15]. Additionally, previous studies on the association between VD and anxiety symptoms observed inconsistent association [13,16]. The discrepant results may be not only related to potential confounding factors, but also to complex etiology of psychiatric disorders. Previous studies on the pathogenesis of psychiatric disorders have found that they are associated with a variety of factors. For example, numerous studies have proved that inﬂammation is related to the pathogenesis of psychiatric disorders, and that the increase in inﬂammation will affect the occurrence and development of some psychiatric disor- ders [17,18]. Other studies have found that both diet and the gut microbiome have a strong inﬂuence on emotional behavior and neural processes [19]. Dietary patterns and changes in the microbiome can affect symptoms of depression and anxiety disorder and increase the risk of both disease through the microbiome gut–brain axis (MGBA) [20]. Moreover, some studies have shown that psychiatric disorders are caused by the interplay between genetic susceptibility and environmental risk factors [6,21]. It has been reported that the heritability of major depression [22] and current anxiety symptoms [23] was estimated to be 38% and 31%, respectively. However, previous studies on exploring the associations between VD and depression and anxiety only focused on the inﬂuence of environmental or genetic factors, usually not considering the interactions between them, which may underestimate the effects of VD on the risks of depression and anxiety. Polygenic risk score (PRS) is proposed by running a GWAS on a study sample, se- lecting SNPs according to the relevant phenotypes, and creating the sum of phenotype related alleles (usually weighted by the SNP-speciﬁc coefﬁcients from the GWAS) that can be evaluated in other replication sample [24]. PRS analysis can not only explore the genetic associations between various complex diseases and traits, but also assess the inﬂuence of susceptible loci on disease risks [25]. For example, PRS-related studies conducted by Psychiatric Genomics Consortium have found signiﬁcant associations between PRS of symptom scale score and the risk of schizophrenia [26] as well as the efﬁcacy of antipsy- Nutrients 2021, 13, 3343 3 of 15 chotic medication [27]. Recently, Revez et al. [28] conducted a GWAS of 417,580 UK Biobank study participants and identiﬁed 143 genetic loci associated with VD. Using the PRS of VD traits as instrumental variables, we can further explore the associations between the PRS of VD traits and psychiatric disorders. GWAS study has strong ability in identifying susceptibility genetic loci associated with psychiatric disorders [29]. However, previous studies have shown that most of the phenotypic variation in complex diseases and traits cannot be explained solely by genetic factors, because phenotypic variation can also occur through genetic environment (G × E) interactions, in which the genotypes of different individuals vary in response to environmental stimuli [30]. Therefore, we further adopted the genome-wide environment interaction study (GWEIS), which can not only assess the effects of G × E interactions, but also evaluate the effects of genetic interactions on a genome-wide scale, helping to identify new genetic risk variants and understand the potential biological mechanisms [31]. For example, Rivera et al. found 53 and 34 single nucleotide polymorphisms (SNPS) in additive interactions with smoking in Lofgren’s syndrome (LS) and non-LS, respectively, but the association did not persist when assessing the effect of smoking on sarcoidosis without genetic information [32]. Utilizing the individual blood VD levels and calculated VD PRS data in UK Biobank cohort, we conducted regression analysis to evaluate the associations between VD (in- cluding blood VD levels and calculated VD PRS data) and depression status and anxiety status in the UK Biobank. Then, based on the results of regression analysis, genome-wide environment interaction studies (GWEIS) were performed to clarify the potential effects of gene × Vitamin D interactions on depression and anxiety. 2. Materials and Methods 2.1. UK Biobank Cohort The UK Biobank (UKB) is a large population-based prospective cohort study, with health-linked information, both regarding phenotype and genotype, on approximately 500,000 participants aged between 40 and 69 years from all over the United Kingdom in 2006 and 2010 [33]. All participants were asked to report a series of health status and demographic information through questionnaires and interviews, and approved to use their anonymous data for any health-related research. Informed consent was provided by UKB from all participants. This study was approved by UKB (Application 46478) and obtained participants’ health-related records. 2.2. UK Biobank Phenotypes of Vitamin D A total of 376,803 UKB participants’ blood samples were collected for quantita- tive measurement of 25(OH)D levels via chemiluminescent immunoassay (CLIA), and 343,334 (91.12%) individuals had their vitamin D 25(OH)D levels (UK Biobank data ﬁeld: 30,890) measured. The analysis was limited to the population of white British individuals (UK Biobank data ﬁeld: 21,000). 2.3. UK Biobank Phenotypes of Depression and Anxiety The phenotypes of depression and anxiety were deﬁned according to the previ- ous study [34]. The selection criteria of case group were deﬁned based on self-reports (UK Biobank data ﬁelds: 20,002; 20,126; and 20,544). In order to classify participants as much as possible, patient health questionnaire-9 (PHQ-9) [35], general anxiety disorder-7 (GAD-7) [36], and composite international diagnostic interview short-form (CIDI-SF) [37] were used as strict inclusion and exclusion criterion. PHQ-9 and GAD-7 are score scales for depression and anxiety, used to screen and measure the severity of depression and anxiety, respectively. PHQ-9 is a classiﬁcation scale focusing on nine depression symptoms and signs, with a total score (0–27) [38], and GAD-7 is a classiﬁcation scale focusing on seven anxious symptoms and signs, with a total score (0–21) [39]. Detailed classiﬁcation of depression and anxiety are presented in the Supplementary Information. The selection Nutrients 2021, 13, 3343 4 of 15 criteria of the control group were as follows: without symptoms of depression and anxiety deﬁned by CIDI and self -reported, depression PHQ scores and anxiety GAD scores ≤ 5, and without core symptoms. 2.4. UK Biobank Genotyping, Imputation and Quality Control A total of 488,377 participants of UKB cohort were genotyped by either the Affymetrix UK BiLEVE Axiom Array or the Affymetrix UKB Axiom arrays [40]. The imputation and the quality control of these genotype results were carried out based on UK10K project reference panel [41] and Haplotype Reference Consortium (HRC) [42] reference panel. Then, we removed the participants who reported inconsistencies between self-reported gender and genetic gender, without ethic consents and imputation data. Additionally, we excluded variants with the Hardy–Weinberg equilibrium test p > 1.0 × 10−5, a minor allele frequency (MAF) of < 0.01, and a genotype missing rate of > 0.05. Ultimately, we used KING software to exclude the genetically related individuals. 2.5. GWAS Data of Vitamin D The latest large-scale GWAS summary statistics of VD were used here. Brieﬂy, this GWAS dataset detected 18,864 independent SNPs that were statistically associated with VD [28]. The genotype data were quality-controlled and imputed against the HRC and UK10K by the UKB group. Then, a linear mixed model GWAS was implemented in fastGWA to identify the genetic loci associated with 25OHD concentrations. Additionally, a rank-based reverse normal transformation (RINT) was applied to the phenotype, age, and gender. The genotyping batch and the ﬁrst 40 ancestry PCs were used as covariates in the mixed model. In order to determine the independent associations, a conditional and joint (COJO) analysis [43] was employed on the GWAS results to explain the correlation structure between SNPs in the 10-Mb window (COJO default parameters). Detailed information of the GWAS can be obtained in the published study [28]. 2.6. PRS Analysis of Vitamin D Using the VD associated SNPs from the GWAS (p < 5 × 10−8) [28], the PRS of VD of each individual was calculated as the sum of the risk allele they carried, weighted by the effect size of the risk allele [28]. The PRS of VD was computed by PLINK2.0 [44], according to the formula: PRS = ∑n βiSNPim i=1 PRS denotes the PRS value of VD for UKB subject; n and i, respectively, denote the total number of sample size and genetic markers; βi is the effect parameter of risk allele of the signiﬁcant SNP associated with VD, which was obtained from the GWAS of VD [28]; and SNPim is the dosage (0, 1, 2) of the risk allele of the SNP associated with VD [28]. 2.7. Statistical Analysis In the UK Biobank cohort, we evaluated associations between vitamin D and de- pression and anxiety through regression analysis. Speciﬁcally, logistic regression analysis was employed to evaluate the associations of self-reported depression and anxiety status with blood VD, VD PRS before COJO adjustment, and VD PRS after COJO adjustment. Linear regression analysis was conducted to test the associations of the PHQ-9 score and the GAD-7 score with blood VD, VD PRS before COJO adjustment, and VD PRS after COJO adjustment. In this regression analysis, blood VD, VD PRS before COJO adjustment, and VD PRS after COJO adjustment were used as independent variables. Self-reported depression, self-reported anxiety, PHQ-9 score, and GAD-7 score were used as outcome variables. Sex, age, and 10 principle components (PCs) of population structure were used as covariates in the regression analysis. A p < 0.05 indicated an association. All analyses were conducted by R. Nutrients 2021, 13, 3343 5 of 15 2.8. Genome-Wide Environment Interaction Studies (GWEIS) The generalized linear regression model of PLINK2.0 [45] was used to estimate the gene × VD interaction effects on the risk of depression and anxiety, using age, gender, and the ﬁrst 10 PCs as covariates. According to previous research [46], we used PLINK2.0 genetic additive (ADD) models and selected high-quality SNPs through a quality control ﬁlters: SNPs with a low call rates (<0.90), low minor allele frequencies (<0.01), or low Hardy–Weinberg equilibrium exact test p-values (<0.01) were excluded. p < 5.0 × 10−8 and p < 5.0 × 10−7 were deﬁned as signiﬁcant and suggestive interactions, respectively. GWEIS results were visualized with the circular Manhattan plots generated by the “CMplot” R script. (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). 3. Results 3.1. General Population Characteristics 3.1.1. Characteristics of UK Biobank Subjects with Blood Vitamin D Data For depression traits, with self-reported depression status as the outcome variable, a total of 110,744 participants answered the depression-related questions, and 52,766 were classified into depression group. With the depression PHQ score as the outcome variable, a total of 109,543 participants completed the questionnaire. For anxiety traits, with self-reported anxiety status as the outcome variable, a total of 98,784 participants answered the anxiety- related questions, and 19,759 were classified into anxiety group. With the anxiety GAD score as the outcome variable, a total of 110,023 participants completed the questionnaire. 3.1.2. Characteristics of UK Biobank Subjects with Vitamin D PRS Data In the self-reported depression status, a total of 121,685 participants answered depression- related questions, of which 58,349 were included in depression group; in the depression PHQ scores, a total of 120,033 participants completed the questionnaire. In the self-reported anxiety status, a total of 108,309 participants answered anxiety-related questions, of which 21,807 were classiﬁed into anxiety group. In addition, in the anxiety GAD scores, a total of 120,590 participants completed the questionnaire. Detailed information is shown in Table 1. Table 1. General population characteristics of this study participants from UK Biobank. Outcome Variable Independent Variable Depression status Anxiety status Depression (PHQ score) Anxiety (GAD score) Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Number/ (Case/Control) 52,766/57,978 58,349/63,336 58,349/63,336 19,759/79,025 21,807/86,502 21,807/86,502 109,543 120,033 120,033 110,023 120,590 120,590 Sex (Female) 61,458 (55.50%) 68,365 (56.18%) 68,365 (56.18%) 53,541 (54.20%) 59,453 (54.89%) 59,453 (54.89%) 60,377 (55.12%) 66,934 (55.76%) 66,934 (55.76%) 60,629 (55.11%) 67,235 (55.76%) 67,235 (55.76%) Age ± SD 56.40 ±7.68 56.47 ± 7.65 56.47 ± 7.65 56.42 ± 7.60 56.50 ± 7.57 56.50 ± 7.57 56.16 ± 7.65 56.24 ± 7.62 56.24 ± 7.62 56.15 ± 7.65 56.23 ± 7.61 56.23 ± 7.61 Abbreviations: VD, Vitamin D; VDPRS, polygenic risk score of vitamin D; COJO, conditional and joint analysis; SD, age was described as mean ± standard deviation (SD); PHQ score, patient health questionnaire (PHQ) is used to describe the depression; GAD score, general anxiety disorder (GAD) is used to describe the anxiety of the participants. 3.2. Regression Analysis Result 3.2.1. Associations between Blood Vitamin D and Depression, Anxiety Traits in UK Biobank Cohort Signiﬁcant associations of blood VD level with self-reported depression status (odds ratio (OR) = 0.89, p = 5.92 × 10−77) and self-reported anxiety status (OR = 0.92, Nutrients 2021, 13, 3343 6 of 15 p = 1.46 × 10−22) were observed. Associations were also observed between the blood VD level, the depression PHQ score (Beta = −0.062, standard error (SE) = 0.003, p = 5.95 × 10−96), and the anxiety GAD score (Beta = −0.030, SE = 0.00, p = 1.21 × 10−21). 3.2.2. Associations between Vitamin D PRS and Depression, Anxiety Traits in UK Biobank Cohort We observed significant associations of VD PRS before COJO adjustment with self- reported depression status (OR = 0.99, p = 3.82 × 10−2), depression PHQ score (Beta = −0.0060, SE = 0.003, p = 3.25 × 10−2), and anxiety GAD score (Beta = −0.010, SE = 0.00, p = 4.36 × 10−2). In addition, we also observed signiﬁcant associations of VD PRS after COJO adjustment with a self-reported depression status (OR = 0.99, p = 1.84 × 10−2), a depression PHQ score (Beta = −0.0070, SE = 0.0030, p = 9.15 × 10−3), and an anxiety GAD score (Beta = −0.010, SE = 0.00, p = 1.02 × 10−2). Detailed information is shown in Table 2. Table 2. The associations between Vitamin D traits and traits of depression and anxiety. Outcome Variable Independent Variable Depression status Anxiety status Depression (PHQ score) Anxiety (GAD score) Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Blood VD VDPRS After COJO VDPRS Before COJO Beta −0.12 −0.014 −0.012 −0.080 0.00 0.00 −0.062 −0.007 −0.006 −0.030 −0.010 −0.010 SE 0.01 0.006 0.006 0.01 0.01 0.01 0.003 0.003 0.003 0.00 0.00 0.00 T −18.57 −2.36 −2.07 −9.77 −0.29 −0.32 −20.81 −2.61 −2.14 −9.56 −2.57 −2.02 p−Value 5.92 × 10−77 1.84 × 10−2 3.82 × 10−2 1.46 × 10−22 7.71 × 10−1 7.47 × 10−1 5.95 × 10−96 9.15 × 10−3 3.25 × 10−2 1.21 × 10−21 1.02 × 10−2 4.36 × 10−2 OR 0.89 0.99 0.99 0.92 1.00 1.00 – – – – – – Abbreviations: SE, standard error; T, t−test; OR, odd ratios; VD, vitamin D; VDPRS, polygenic risk score of vitamin D; COJO, conditional and joint analysis; PHQ score, patient health questionnaire (PHQ) is used to describe the depression; GAD score, general anxiety disorder (GAD) is used to describe the anxiety of the participants. Note. Logistic regression was used to evaluate the association between blood VD, VD PRS before COJO adjustment, VD PRS after COJO adjustment and self-reported depression and anxiety. Linear regression was used to evaluate the association between blood VD, VD PRS before COJO adjustment, VD PRS after COJO adjustment and PHQ score, GAD score. 3.3. GWEIS Analysis Results For self-reported depression status, GWEIS identiﬁed a signiﬁcant gene × VD PRS interaction (p < 5.0 × 10−8) at the LRRTM4 gene (rs114086183, p = 4.11 × 10−8). For self-reported anxiety status, signiﬁcant gene × VD PRS interaction was detected at the GNB5 gene (rs149760119, p = 3.88 × 10−8). For the depression PHQ score, two signiﬁcant gene × blood VD interactions were identiﬁed at SLC11A2 and HIGD1C (rs117102029, p = 4.02 × 10−8). For the anxiety GAD score, we detected multiple signiﬁcant gene × VD trait interactions, such as SMYD3 (rs142593645, p = 2.51 × 10−8), SEMA3E (rs76440131, p = 2.80 × 10−10), and VTI1A (rs17266687, p = 3.09 × 10−8). Among them, three genes (SEMA3E, DOCK8, TMCO3) were identiﬁed by VD PRS before and after COJO adjustment. The visualization of the results is shown in Figures 1–3. Additional detailed results are shown in Table 3. Nutrients 2021, 13, 3343 7 of 15 Figure 1. Genomic regions interacting with blood VD for the PHQ score and the GAD score. (a) Depression (PHQ score), A SNP allele was found to significantly interact with blood VD in depression (PHQ score); (b) Anxiety (GAD score), seven independent SNP alleles were found to significantly interact with blood VD in anxiety disorder (GAD score). From the center, the first circos depicts the −log10 p-values of each variant due to double exposure, i.e., the effect of both SNP allele and blood VD. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The figure was generated using the “CMplot” R script (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). Figure 2. Genomic regions interacting with VD PRS after COJO adjustment for depression status, anxiety status, and GAD score. (a) Depression status, a SNP allele was found to signiﬁcantly interact with blood VD in depression status; (b) Anxiety status, 2 independent SNP alleles were found to signiﬁcantly interact with blood VD in anxiety status; (c) Anxiety (GAD score), 8 independent SNP alleles interacted signiﬁcantly with blood VD in anxiety disorders (GAD score). From the center, the ﬁrst circos depicts the −log10 p-values of each variant due to double exposure, i.e., the effect of both SNP allele and VD PRS after COJO adjustment. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The ﬁgure was generated using the “CMplot” R script (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). Nutrients 2021, 13, x FOR PEER REVIEW 7 of 15 For self-reported depression status, GWEIS identified a significant gene × VD PRS interaction (p < 5.0 × 10–8) at the LRRTM4 gene (rs114086183, p = 4.11 × 10−8). For self-re-ported anxiety status, significant gene × VD PRS interaction was detected at the GNB5 gene (rs149760119, p = 3.88 × 10−8). For the depression PHQ score, two significant gene × blood VD interactions were identified at SLC11A2 and HIGD1C (rs117102029, p = 4.02 × 10−8). For the anxiety GAD score, we detected multiple significant gene × VD trait interac-tions, such as SMYD3 (rs142593645, p = 2.51 × 10−8), SEMA3E (rs76440131, p = 2.80 × 10−10), and VTI1A (rs17266687, p = 3.09 × 10−8). Among them, three genes (SEMA3E, DOCK8, TMCO3) were identified by VD PRS before and after COJO adjustment. The visualization of the results is shown in Figures 1–3. Additional detailed results are shown in Table 3. Table 3. Summary of gene−environment interaction analysis among SNP and VD traits for depression and anxiety traits. CHR SNP Model Beta SE Gene p−Value Depression status 2 rs114086183 ADD ×VD PRS afterCOJO 0.16 0.029 LRRTM4 4.11 × 10−8 Anxiety status 15 rs149760119 ADD × VD PRS afterCOJO −0.22 0.04 GNB5 3.88 × 10−8 Depression (PHQ score) 12 rs117102029 ADD × VD blood 0.01 0.002 SLC11A2, HIGD1C 4.02 × 10−8 Anxiety (GAD score) 1 rs142593645 ADD × VD blood 1.52 0.27 SMYD3 2.51 × 10−8 7 rs13228257 ADD × VD blood −1.25 0.22 DPP6 1.45 × 10−8 7 rs76440131 ADD × VD PRS afterCOJO −2.64 0.42 SEMA3E 2.80 × 10−10 9 rs78029983 ADD × VD PRS afterCOJO −1.08 0.19 DOCK8 2.43 × 10−10 13 rs76004204 ADD × VD PRS afterCOJO 2.17 0.37 TMCO3 6.38 × 10−9 7 rs76440131 ADD×VD PRS beforeCOJO −2.16 0.38 SEMA3E 1.76 × 10−8 9 rs78029983 ADD×VD PRS beforeCOJO −1.10 0.20 DOCK8 2.10 × 10−8 10 rs17266687 ADD×VD PRS beforeCOJO −1.20 0.21 VTI1A 2.48 × 10−8 13 rs76004204 ADD×VD PRS beforeCOJO 1.97 0.34 TMCO3 8.89 × 10−9 Abbreviations: CHR, chromosome; SNP, single nucleotide polymorphism; SE, standard error; ADD, additive effect; VD, vitamin D; VDPRS, polygenic risk score of vitamin D; COJO, conditional and joint analysis; PHQ score, patient health questionnaire (PHQ) is used to describe the depression; GAD score, general anxiety disorder (GAD) is used to describe the anxiety of the participants; p−value, estimates of the eﬀect of interaction on depression and anxiety traits by using ADD × VD traits. (a) (b) Nutrients 2021, 13, x FOR PEER REVIEW 8 of 15 Figure 1. Genomic regions interacting with blood VD for the PHQ score and the GAD score. (a) Depression (PHQ score), A SNP allele was found to significantly interact with blood VD in depression (PHQ score); (b) Anxiety (GAD score), seven independent SNP alleles were found to significantly interact with blood VD in anxiety disorder (GAD score). From the center, the first circos depicts the −log10 p-values of each variant due to double exposure, i.e., the effect of both SNP allele and blood VD. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The figure was generated using the “CMplot” R script (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). (a) (b) (c) Figure 2. Genomic regions interacting with VD PRS after COJO adjustment for depression status, anxiety status, and GAD score. (a) Depression status, a SNP allele was found to significantly interact with blood VD in depression status; (b) Anx-iety status, 2 independent SNP alleles were found to significantly interact with blood VD in anxiety status; (c) Anxiety (GAD score), 8 independent SNP alleles interacted significantly with blood VD in anxiety disorders (GAD score). From the center, the first circos depicts the −log10 p-values of each variant due to double exposure, i.e., the effect of both SNP allele and VD PRS after COJO adjustment. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The figure was generated using the “CMplot” R script (https://github.com/Yin-LiLin/R-CMplot) (accessed on 15 February 2021). Nutrients 2021, 13, 3343 8 of 15 Figure 3. Genomic regions interacting with VD PRS before COJO adjustment for GAD score. This graph shows 7 independent SNP alleles interacting signiﬁcantly with blood VD in anxiety disorders (GAD score). From the center, the ﬁrst circos depicts the −log10 p-values of each variant due to double exposure, i.e., the effect of both SNP allele and VD PRS before COJO adjustment. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The ﬁgure was generated using the “CMplot” R script (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). Table 3. Summary of gene−environment interaction analysis among SNP and VD traits for depression and anxiety traits. CHR SNP Model rs114086183 ADD ×VD PRS afterCOJO Beta 0.16 SE Gene p−Value 0.029 LRRTM4 4.11 × 10−8 Depression status Anxiety status Depression (PHQ score) Anxiety (GAD score) 2 15 12 1 7 7 9 13 7 9 10 13 rs149760119 ADD × VD PRS afterCOJO −0.22 0.04 GNB5 rs117102029 ADD × VD blood 0.01 0.002 rs142593645 rs13228257 rs76440131 rs78029983 rs76004204 rs76440131 rs78029983 rs17266687 rs76004204 ADD × VD blood ADD × VD blood ADD × VD PRS afterCOJO ADD × VD PRS afterCOJO ADD × VD PRS afterCOJO ADD × VD PRS beforeCOJO ADD × VD PRS beforeCOJO ADD × VD PRS beforeCOJO ADD × VD PRS beforeCOJO 1.52 −1.25 −2.64 −1.08 2.17 −2.16 −1.10 −1.20 1.97 0.27 0.22 0.42 0.19 0.37 0.38 0.20 0.21 0.34 SLC11A2, HIGD1C SMYD3 DPP6 SEMA3E DOCK8 TMCO3 SEMA3E DOCK8 VTI1A TMCO3 3.88 × 10−8 4.02 × 10−8 2.51 × 10−8 1.45 × 10−8 2.80 × 10−10 2.43 × 10−10 6.38 × 10−9 1.76 × 10−8 2.10 × 10−8 2.48 × 10−8 8.89 × 10−9 Abbreviations: CHR, chromosome; SNP, single nucleotide polymorphism; SE, standard error; ADD, additive effect; VD, vitamin D; VDPRS, polygenic risk score of vitamin D; COJO, conditional and joint analysis; PHQ score, patient health questionnaire (PHQ) is used to describe the depression; GAD score, general anxiety disorder (GAD) is used to describe the anxiety of the participants; p−value, estimates of the effect of interaction on depression and anxiety traits by using ADD × VD traits. 4. Discussion Since many complex diseases are associated with thousands of genetic variations, GWAS study merely calculates the association between a single SNP and a phenotype, which could easily lead to a decline in the interpretation of phenotypes inﬂuenced by multiple genetic variations. In this study, using individual blood VD level and calculated VD PRS data, we systematically evaluated the associations of VD traits with depression and anxiety traits. Then, we conducted GWEIS to clarify the potential effects of gene × VD interaction on depression and anxiety traits. We observed signiﬁcant associations between VD and depression and anxiety traits in the UK Biobank cohort, and GWEIS analysis identiﬁed the effects of multiple signiﬁcant gene × VD interactions on depression and anxiety traits. As mentioned before, previous studies on VD and psychiatric disorders merely fo- cused on the effects of environmental or genetic factors on the risks of depression and anxiety, usually without considering the interaction between them. For anxiety, multiple Nutrients 2021, 13, x FOR PEER REVIEW 9 of 15 Figure 3. Genomic regions interacting with VD PRS before COJO adjustment for GAD score. This graph shows 7 independent SNP alleles interacting significantly with blood VD in anxiety disorders (GAD score). From the center, the first circos depicts the −log10 p-values of each variant due to dou-ble exposure, i.e., the effect of both SNP allele and VD PRS before COJO adjustment. The second circos shows chromosome density. Red dots represent the p < 5 × 10−8 and green dots represent p < 1 × 10−7. The figure was generated using the “CMplot” R script (https://github.com/YinLiLin/R-CMplot) (accessed on 15 February 2021). 4. Discussion Since many complex diseases are associated with thousands of genetic variations, GWAS study merely calculates the association between a single SNP and a phenotype, which could easily lead to a decline in the interpretation of phenotypes influenced by multiple genetic variations. In this study, using individual blood VD level and calculated VD PRS data, we systematically evaluated the associations of VD traits with depression and anxiety traits. Then, we conducted GWEIS to clarify the potential effects of gene × VD interaction on depression and anxiety traits. We observed significant associations between VD and depression and anxiety traits in the UK Biobank cohort, and GWEIS analysis iden-tified the effects of multiple significant gene × VD interactions on depression and anxiety traits. As mentioned before, previous studies on VD and psychiatric disorders merely fo-cused on the effects of environmental or genetic factors on the risks of depression and anxiety, usually without considering the interaction between them. For anxiety, multiple studies have observed that VD level was associated with anxiety [47], and VD supplemen-tation can significantly improve patients’ anxiety symptoms after adjusting for covariates known to affect VD level [48]. However, there is also controversy about the association between VD and depression. For instance, Zhao et al. found that VD was not associated with an increased risk of depression after adjusting for potential confounders, such as age, gender, race, physical activity, alcohol use, and chronic diseases [49]. Some studies con-ducted in different regions failed to observe the association between VD and depression when controlled for potential confounding factors [50,51]. In contrast, Milaneschi et al. observed that low VD levels were associated with the presence and severity of depression, suggesting that VD represented a potential biological vulnerability for depression [52]. Additionally, a prospective association study of VD and depression using UKB cohort data found that both vitamin D deficiency and insufficiency may be risk factors for new onset depression in middle-aged adults [53]. Our results support an association between VD traits and depression and anxiety traits in the UK Biobank cohort, particularly from a genetic perspective. It is important to note that our study found an association between Nutrients 2021, 13, 3343 9 of 15 studies have observed that VD level was associated with anxiety [47], and VD supplemen- tation can signiﬁcantly improve patients’ anxiety symptoms after adjusting for covariates known to affect VD level [48]. However, there is also controversy about the association between VD and depression. For instance, Zhao et al. found that VD was not associated with an increased risk of depression after adjusting for potential confounders, such as age, gender, race, physical activity, alcohol use, and chronic diseases [49]. Some studies conducted in different regions failed to observe the association between VD and depression when controlled for potential confounding factors [50,51]. In contrast, Milaneschi et al. observed that low VD levels were associated with the presence and severity of depression, suggesting that VD represented a potential biological vulnerability for depression [52]. Additionally, a prospective association study of VD and depression using UKB cohort data found that both vitamin D deﬁciency and insufﬁciency may be risk factors for new onset depression in middle-aged adults [53]. Our results support an association between VD traits and depression and anxiety traits in the UK Biobank cohort, particularly from a genetic perspective. It is important to note that our study found an association between VD and depression and anxiety; however, further research is needed to determine whether there was a causal association and in what direction. Currently, to the best of our knowledge, there are limited researches to explore the genetic mechanism affecting the link between VD and depression and anxiety. Therefore, we performed GWEIS and identiﬁed multiple candidate genes interacting with VD, which are implicated in the brain or neural regulation and pathology, such as LRRTM4 for depression status and GNB5 for anxiety status. The LRRTM4 is a new four-membered family of genes from human and mice. Its main function is to encode a putative leucine-rich repeat transmembrane protein, which can not only facilitate the development of glutamate synapses, but also regulate many cellular events during nervous system development and disease [54,55]. In animal experiments, it has been found that LRRTM4 is expressed in many brain regions and nervous system neurons, suggesting that LRRTM4 plays a vital role in the development and maintenance of the vertebrate nervous system [54]. In addition, the role of VD in regulating brain axon growth has been observed in previous studies [56], and prenatal VD deﬁciency has been shown to alter many genes involved in synaptic plasticity [57]. Whether VD deﬁciency alters the LRRTM4 gene remains elusive and need further studies. For anxiety status, the identiﬁed GNB5 gene is the G protein subunit beta 5 (Gβ5) that encodes a heterotrimeric GTP binding protein. Gβ5 is enriched in the central nervous system. Its main function is to form a complex with regulatory factors of the G protein signal transduction protein family, thereby regulating and affecting the neurotransmitter signal transduction of many neurobehavioral results [58]. A previous study found that VD can affect adult brain development and function through signal transduction [59]. Furthermore, a study of the expression of genes associated with Alzheimer’s disease in the presence of VD deﬁciency found that GNB5 expression was signiﬁcantly reduced [60]. We may infer that VD can regulate G-protein-mediated signaling in the brain by inﬂuencing the GNB5 gene [60]. In addition, previous studies have indicated that GNB5 gene mutations can lead to severe speech disorders, motor delays, and attention deﬁcit hyperactivity disorder (ADHD) as the main manifestations of recessive neurodevelopmental disorders [58,61]. For the anxiety GAD score, our GWEIS results showed that multiple genes have signif- icant interactions with vitamin D. DPP6 is a single-channel type II transmembrane protein expressed in the brain, which mainly regulates the dendritic excitability of hippocampal neurons [62]. Cacace et al. found that DPP6 is a new genetic factor in dementia. DPP6 is involved in a variety of cellular pathways, including neurogenesis and neuronal excitabil- ity, and its deletion has been associated with low intelligence and neurodevelopmental disorders [63]. Another study in an animal model also found that DPP6 deletion affects hippocampal synaptic development and leads to behavioral impairments in recognition, learning, and memory [64]. Similar to the DPP6 gene, Tang’s study found that VTI1A is mainly involved in neuronal development and neurotransmission, and mutations are Nutrients 2021, 13, 3343 10 of 15 likely to lead to neurological dysfunction and neurological diseases [65]. In common genes identiﬁed by VD PRS before and after COJO adjustment, SEMA3E is a member of the signaling family that binds directly to the receptor Plexin-D1 to secrete brain signals. A study found that SEMA3E-Plexin-D1 signaling is not only involved in axon growth and guidance, but also determines synaptic recognition and speciﬁcity in multiple parts of the nervous system [66]. Although these genes have been found to play a certain role in the development and conduction of the nervous system, the potential biological mechanism of the interaction between these genes and VD to affect the nervous system function and disease has not been found, which needs further research and conﬁrmation. In addition, we identiﬁed multiple candidate genes which interacted with vitamin D for the depression PHQ score, such as SLC11A2 and HIGD1C. The SLC11A2 gene, also called DMT1, is an iron-responsive gene mainly involved in iron absorption [67]. Mutations in this gene will affect the changes in the body’s iron content. Bastian et al. [68] found that iron deﬁciency in the early life can damage the expression of hippocampal neuron genes, leading to long-term neurological dysfunction. Saadat et al. [69] observed that the TT genotype and the T allele of the 1254T > C polymorphism in the DMT1 gene may be a risk factor for Parkinson’s disease. At present, few researches were conducted to explain the function and role of the HIGD1C gene. However, the HIGD1A gene, which came from the same family as the HIGD1C gene (HIG1 hypoxia-inducible domain family), has been found to be related to the nervous system in previous studies. Research conducted by López’s et al. found that the HIGD1A gene is not only widely expressed in the rat brain, but also may play a protective role in certain areas of the central nervous system [70]. Nevertheless, no studies have shown a direct link between VD and the effects of SLC11A2 and HIGD1C genes on the nervous system and psychiatric disorders. It is worth mentioning that previous studies have found the important role of VD in the brain and nervous system, such as VD differentiates brain cells [71], which regulates axon growth [56] and can regulate calcium signaling [72]. VD can not only affect adult brain development and function through signal transduction, but also affect the nutritional support factors of developing and mature neurons and prevent the production of reactive oxygen species. These all support the importance of VD for development and function of human brain. Molecular genetic studies have conﬁrmed the presence of widespread pleiotropy across psychiatric disorders [73]. Previous studies have found that depression and anxiety are highly comorbid and share a common underlying basis, including symptom overlap, potential negative affectivity, shared familial risk, stress, negative cognitions, and similar neural-circuitry dysfunction related to emotion regulation [74–76]. According to Gray and McNaughton’s theory, this comorbidity is caused by the recursive interconnection of brain regions that connect fear, anxiety, and panic, as well as hereditary personality traits such as neuroticism [77]. Twin and familial studies have shown that comorbidity of depression and anxiety disorders is largely explained by shared genetic risk factors [77]. However, a recent factor analysis and genomic structural equation modelling study on depression and anxiety found that depressive and anxiety symptoms could be affected by different factors, although the genetic correlation between the factors was high [78]. In this study, we further compared and identiﬁed genetic loci between depression and anxiety; no overlapping loci were found, suggesting that VD may have different biological mechanisms in depression and anxiety. It is considered that environmental exposure can contribute to the development of depression and anxiety through different molecular mechanisms [79]. Furthermore, based on the results of genomic structural equation modeling [78] and the differences in etiology and pathogenesis of depression and anxiety [80], it is reasonable to infer that few genetic loci interacting with VD promote the occurrence and development of depression and anxiety at the same time. Genetic research that assesses the link between VD and psychiatric disorders are limited, and further exploration is needed to conﬁrm our ﬁndings. It is worth noting that our study has some limitations. First, all research data in our study were derived from the UK Biobank, and the research participants were limited to Nutrients 2021, 13, 3343 11 of 15 people of European descent. Due to different genetic backgrounds, the results of this study should be interpreted with caution when applying the results to other populations. Secondly, in our research, we mainly used self-reports and related questionnaire scores to characterize depression and anxiety states. Since there is no systematic method to classify all the symptoms, the self-reported analysis results may not be completely consistent with the analysis results of questionnaire scores; in addition, self-reported results may increase the possibility of measurement error and recall bias. Due to the lack of temporal sequence between variables and the absence of Mendelian Randomization study, it is not possible to draw evidence for causality directly, resulting in the lack of demonstration strength of the study. Finally, there is a lack of relevant researches to investigate the inﬂuence of the identiﬁed SNPs on the biological mechanisms of depression and anxiety. More large sample prospective studies and biological studies are needed to conﬁrm our results and elucidate the potential role of new genetic variants in the pathogenesis of psychiatric disorders. 5. Conclusions In summary, through regression analysis and GWEIS analysis, we observed that the VD was negatively associated with depression and anxiety, and further GWEIS analysis identiﬁed multiple candidate genes related to depression and anxiety. The interaction effects observed from the results provide new direction for understanding the genetic research of psychiatric disorders. Further research is needed to clarify the underlying mechanism of gene × VD interaction effects for psychiatric disorders. Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/nu13103343/s1, Table S1: Present and past depression and/or bipolar affective disorder. Table S2: Generalized anxiety disorder. Author Contributions: Z.Z. had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. X.Y. contributed equally to the work as co–senior authors. Y.W., Y.J. and F.Z. conceptualized and designed the study. Z.Z., X.Y., Y.J., Y.W., S.C., P.M., C.L., H.Z., C.P., J.Z., Y.C. and F.Z. contributed in acquisition, analysis, and interpretation of the data. Z.Z. drafted the manuscript. F.Z., X.Y. helped with critical revision of the manuscript for important intellectual content. S.C. performed statistical analysis. Y.W., Y.J. and F.Z. provided administrative, technical, or material support. F.Z. supervised the study. All authors have read and agreed to the published version of the manuscript. Funding: This study was supported by the National Natural Scientiﬁc Foundation of China (81922059). Institutional Review Board Statement: This study has been approved by UKB (Application 46478) and obtained participants’ health-related records. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The UK Biobank data are available through the UK Biobank Access Management System https://www.ukbiobank.ac.uk/ (accessed on 20 December 2020). We will return the derived data ﬁelds following UK Biobank policy; in due course, they will be available through the UK Biobank Access Management System. Acknowledgments: This study was conducted using the UK Biobank Resource (Application 46478). Conﬂicts of Interest: All authors report no biomedical financial interests or potential conflicts of interest. References 1. 2. 3. 4. Battle, D.E. Diagnostic and Statistical Manual of Mental Disorders (DSM). Codas 2013, 25, 191–192. [CrossRef] [PubMed] Richter, D.; Wall, A.; Bruen, A.; Whittington, R. Is the global prevalence rate of adult mental illness increasing? Systematic review and meta-analysis. Acta Psychiatr. Scand. 2019, 140, 393–407. [CrossRef] [PubMed] Friedrich, M.J. Depression Is the Leading Cause of Disability Around the World. JAMA 2017, 317, 1517. [CrossRef] [PubMed] Bekhuis, E.; Boschloo, L.; Rosmalen, J.G.; Schoevers, R.A. Differential associations of speciﬁc depressive and anxiety disorders with somatic symptoms. J. Psychosom. Res. 2015, 78, 116–122. [CrossRef] Nutrients 2021, 13, 3343 12 of 15 6. 7. 5. Wang, Y.H.; Li, J.Q.; Shi, J.F.; Que, J.Y.; Liu, J.J.; Lappin, J.M.; Leung, J.; Ravindran, A.V.; Chen, W.Q.; Qiao, Y.L.; et al. Depression and anxiety in relation to cancer incidence and mortality: A systematic review and meta-analysis of cohort studies. Mol. Psychiatry 2020, 25, 1487–1499. [CrossRef] Groves, N.J.; McGrath, J.J.; Burne, T.H. Vitamin D as a neurosteroid affecting the developing and adult brain. Annu. Rev. Nutr. 2014, 34, 117–141. [CrossRef] Bahrami, A.; Sadeghnia, H.R.; Tabatabaeizadeh, S.A.; Bahrami-Taghanaki, H.; Behboodi, N.; Esmaeili, H.; Ferns, G.A.; Mobarhan, M.G.; Avan, A. Genetic and epigenetic factors inﬂuencing vitamin D status. J. Cell Physiol. 2018, 233, 4033–4043. [CrossRef] [PubMed] Ahn, J.; Albanes, D.; Berndt, S.I.; Peters, U.; Chatterjee, N.; Freedman, N.D.; Abnet, C.C.; Huang, W.Y.; Kibel, A.S.; Crawford, E.D.; et al. Vitamin D-related genes, serum vitamin D concentrations and prostate cancer risk. Carcinogenesis 2009, 30, 769–776. [CrossRef] Szpunar, M.J. Association of antepartum vitamin D deﬁciency with postpartum depression: A clinical perspective. Public Health Nutr. 2020, 23, 1173–1178. [CrossRef] 9. 8. 10. Bersani, F.S.; Ghezzi, F.; Maraone, A.; Vicinanza, R.; Cavaggioni, G.; Biondi, M.; Pasquini, M. The relationship between Vitamin D and depressive disorders. Riv. Psichiatr. 2019, 54, 229–234. [CrossRef] [PubMed] 11. Anglin, R.E.; Samaan, Z.; Walter, S.D.; McDonald, S.D. Vitamin D deﬁciency and depression in adults: Systematic review and meta-analysis. Br. J. Psychiatry 2013, 202, 100–107. [CrossRef] 13. 12. Hoang, M.T.; Deﬁna, L.F.; Willis, B.L.; Leonard, D.S.; Weiner, M.F.; Brown, E.S. Association between low serum 25-hydroxyvitamin D and depression in a large sample of healthy adults: The Cooper Center longitudinal study. Mayo Clin. Proc. 2011, 86, 1050–1055. [CrossRef] Fond, G.; Godin, O.; Schürhoff, F.; Berna, F.; Bulzacka, E.; Andrianarisoa, M.; Brunel, L.; Aouizerate, B.; Capdevielle, D.; Chereau, I.; et al. Hypovitaminosis D is associated with depression and anxiety in schizophrenia: Results from the national FACE-SZ cohort. Psychiatry Res. 2018, 270, 104–110. [CrossRef] Ikonen, H.; Palaniswamy, S.; Nordstrom, T.; Jarvelin, M.R.; Herzig, K.H.; Jaaskelainen, E.; Seppala, J.; Miettunen, J.; Sebert, S. Vitamin D status and correlates of low vitamin D in schizophrenia, other psychoses and non-psychotic depression—The Northern Finland Birth Cohort 1966 study. Psychiatry Res. 2019, 279, 186–194. [CrossRef] 14. 15. Libuda, L.; Laabs, B.H.; Ludwig, C.; Bühlmeier, J.; Antel, J.; Hinney, A.; Naaresh, R.; Föcker, M.; Hebebrand, J.; König, I.R.; et al. Vitamin D and the Risk of Depression: A Causal Relationship? Findings from a Mendelian Randomization Study. Nutrients 2019, 11, 1085. [CrossRef] 16. Kim, S.Y.; Jeon, S.W.; Lim, W.J.; Oh, K.S.; Shin, D.W.; Cho, S.J.; Park, J.H.; Shin, Y.C. The Relationship between Serum Vitamin D Levels, C-Reactive Protein, and Anxiety Symptoms. Psychiatry Investig. 2020, 17, 312–319. [CrossRef] 17. Niles, A.N.; Smirnova, M.; Lin, J.; O’Donovan, A. Gender differences in longitudinal relationships between depression and anxiety symptoms and inﬂammation in the health and retirement study. Psychoneuroendocrinology 2018, 95, 149–157. [CrossRef] 18. Tanaka, M.; Tóth, F.; Polyák, H.; Szabó, Á.; Mándi, Y.; Vécsei, L. Immune Inﬂuencers in Action: Metabolites and Enzymes of the Tryptophan-Kynurenine Metabolic Pathway. Biomedicines 2021, 9, 734. [CrossRef] [PubMed] 19. Luna, R.A.; Foster, J.A. Gut brain axis: Diet microbiota interactions and implications for modulation of anxiety and depression. Curr. Opin. Biotechnol. 2015, 32, 35–41. [CrossRef] [PubMed] 20. Bear, T.L.K.; Dalziel, J.E.; Coad, J.; Roy, N.C.; Butts, C.A.; Gopal, P.K. The Role of the Gut Microbiota in Dietary Interventions for 21. Depression and Anxiety. Adv. Nutr. 2020, 11, 890–907. [CrossRef] [PubMed] Sullivan, P.F.; Neale, M.C.; Kendler, K.S. Genetic epidemiology of major depression: Review and meta-analysis. Am. J. Psychiatry 2000, 157, 1552–1562. [CrossRef] 22. Kendler, K.S.; Gatz, M.; Gardner, C.O.; Pedersen, N.L. A Swedish national twin study of lifetime major depression. Am. J. Psychiatry 2006, 163, 109–114. [CrossRef] 23. Purves, K.L.; Coleman, J.R.I.; Meier, S.M.; Rayner, C.; Davis, K.A.S.; Cheesman, R.; Bækvad-Hansen, M.; Børglum, A.D.; Wan Cho, S.; Jürgen Deckert, J.; et al. A major role for common genetic variation in anxiety disorders. Mol. Psychiatry 2020, 25, 3292–3303. [CrossRef] 24. Dudbridge, F. Power and predictive accuracy of polygenic risk scores. PLoS Genet. 2013, 9, e1003348. [CrossRef] 25. Euesden, J.; Lewis, C.M.; O’Reilly, P.F. PRSice: Polygenic Risk Score software. Bioinformatics 2015, 31, 1466–1468. [CrossRef] 26. Schizophrenia Working Group of the Psychiatric Genomics Consortium. Biological insights from 108 schizophrenia-associated genetic loci. Nature 2014, 511, 421–427. [CrossRef] [PubMed] 27. Zhang, J.P.; Robinson, D.; Yu, J.; Gallego, J.; Fleischhacker, W.W.; Kahn, R.S.; Crespo-Facorro, B.; Vazquez-Bourgon, J.; Kane, J.M.; Malhotra, A.K.; et al. Schizophrenia Polygenic Risk Score as a Predictor of Antipsychotic Efﬁcacy in First-Episode Psychosis. Am. J. Psychiatry 2019, 176, 21–28. [CrossRef] 28. Revez, J.A.; Lin, T.; Qiao, Z.; Xue, A.; Holtz, Y.; Zhu, Z.; Zeng, J.; Wang, H.; Sidorenko, J.; Kemper, K.E.; et al. Genome-wide association study identiﬁes 143 loci associated with 25 hydroxyvitamin D concentration. Nat. Commun. 2020, 11, 1647. [CrossRef] 29. Hou, L.; Bergen, S.E.; Akula, N.; Song, J.; Hultman, C.M.; Landén, M.; Adli, M.; Alda, M.; Ardau, R.; Arias, B.; et al. Genome-wide association study of 40,000 individuals identiﬁes two novel loci associated with bipolar disorder. Hum. Mol. Genet. 2016, 25, 3383–3394. [CrossRef] Nutrients 2021, 13, 3343 13 of 15 30. Rask-Andersen, M.; Karlsson, T.; Ek, W.E.; Johansson, Å. Gene-environment interaction study for BMI reveals interactions between genetic factors and physical activity, alcohol consumption and socioeconomic status. PLoS Genet. 2017, 13, e1006977. [CrossRef] van Os, J.; Rutten, B.P. Gene-environment-wide interaction studies in psychiatry. Am. J. Psychiatry 2009, 166, 964–966. [CrossRef] 31. 32. Rivera, N.V.; Patasova, K.; Kullberg, S.; Diaz-Gallo, L.M.; Iseda, T.; Bengtsson, C.; Alfredsson, L.; Eklund, A.; Kockum, I.; Grunewald, J.; et al. A Gene-Environment Interaction Between Smoking and Gene polymorphisms Provides a High Risk of Two Subgroups of Sarcoidosis. Sci. Rep. 2019, 9, 18633. [CrossRef] Sudlow, C.; Gallacher, J.; Allen, N.; Beral, V.; Burton, P.; Danesh, J.; Downey, P.; Elliott, P.; Green, J.; Landray, M.; et al. UK biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015, 12, e1001779. [CrossRef] 33. 34. Davis, K.A.S.; Cullen, B.; Adams, M.; Brailean, A.; Breen, G.; Coleman, J.R.I.; Dregan, A.; Gaspar, H.A.; Hübel, C.; Lee, W.; et al. Indicators of mental disorders in UK Biobank-A comparison of approaches. Int. J. Methods Psychiatr. Res. 2019, 28, e1796. [CrossRef] [PubMed] 35. Kroenke, K.; Spitzer, R.L.; Williams, J.B. The PHQ-9: Validity of a brief depression severity measure. J. Gen. Intern. Med. 2001, 16, 36. 606–613. [CrossRef] Spitzer, R.L.; Kroenke, K.; Williams, J.B.; Löwe, B. A brief measure for assessing generalized anxiety disorder: The GAD-7. Arch. Intern. Med. 2006, 166, 1092–1097. [CrossRef] 37. Gigantesco, A.; Morosini, P. Development, reliability and factor analysis of a self-administered questionnaire which originates from the World Health Organization’s Composite International Diagnostic Interview—Short Form (CIDI-SF) for assessing mental disorders. Clin. Pract. Epidemiol. Ment. Health 2008, 4, 8. [CrossRef] 38. Manea, L.; Gilbody, S.; McMillan, D. Optimal cut-off score for diagnosing depression with the Patient Health Questionnaire (PHQ-9): A meta-analysis. CMAJ 2012, 184, E191–E196. [CrossRef] 39. Kroenke, K.; Spitzer, R.L.; Williams, J.B.; Löwe, B. The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: A systematic review. Gen. Hosp. Psychiatry 2010, 32, 345–359. [CrossRef] 40. Bycroft, C.; Freeman, C.; Petkova, D.; Band, G.; Elliott, L.T.; Sharp, K.; Motyer, A.; Vukcevic, D.; Delaneau, O.; O’Connell, J.; et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 2018, 562, 203–209. [CrossRef] 41. Walter, K.; Min, J.L.; Huang, J.; Crooks, L.; Memari, Y.; McCarthy, S.; Perry, J.R.; Xu, C.; Futema, M.; Lawson, D.; et al. The UK10K project identiﬁes rare variants in health and disease. Nature 2015, 526, 82–90. [CrossRef] 42. McCarthy, S.; Das, S.; Kretzschmar, W.; Delaneau, O.; Wood, A.R.; Teumer, A.; Kang, H.M.; Fuchsberger, C.; Danecek, P.; Sharp, K.; et al. A reference panel of 64,976 haplotypes for genotype imputation. Nat. Genet. 2016, 48, 1279–1283. [CrossRef] [PubMed] 43. Yang, J.; Ferreira, T.; Morris, A.P.; Medland, S.E.; Madden, P.A.; Heath, A.C.; Martin, N.G.; Montgomery, G.W.; Weedon, M.N.; Loos, R.J.; et al. Conditional and joint multiple-SNP analysis of GWAS summary statistics identiﬁes additional variants inﬂuencing complex traits. Nat. Genet. 2012, 44, s361–s363. [CrossRef] [PubMed] 44. Purcell, S.; Neale, B.; Todd-Brown, K.; Thomas, L.; Ferreira, M.A.; Bender, D.; Maller, J.; Sklar, P.; de Bakker, P.I.; Daly, M.J.; et al. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am. J. Hum. Genet. 2007, 81, 559–575. [CrossRef] 45. Chang, C.C.; Chow, C.C.; Tellier, L.C.; Vattikuti, S.; Purcell, S.M.; Lee, J.J. Second-generation PLINK: Rising to the challenge of larger and richer datasets. Gigascience 2015, 4, 7. [CrossRef] 46. Clarke, G.M.; Morris, A.P. A comparison of sample size and power in case-only association studies of gene-environment interaction. Am. J. Epidemiol. 2010, 171, 498–505. [CrossRef] 47. Kimball, S.M.; Mirhosseini, N.; Rucklidge, J. Database Analysis of Depression and Anxiety in a Community Sample-Response to a Micronutrient Intervention. Nutrients 2018, 10, 152. [CrossRef] 48. Zhu, C.; Zhang, Y.; Wang, T.; Lin, Y.; Yu, J.; Xia, Q.; Zhu, P.; Zhu, D.M. Vitamin D supplementation improves anxiety but not depression symptoms in patients with vitamin D deﬁciency. Brain Behav. 2020, 10, e01760. [CrossRef] 49. Zhao, G.; Ford, E.S.; Li, C.; Balluz, L.S. No associations between serum concentrations of 25-hydroxyvitamin D and parathyroid hormone and depression among US adults. Br. J. Nutr. 2010, 104, 1696–1702. [CrossRef] 50. Nanri, A.; Mizoue, T.; Matsushita, Y.; Poudel-Tandukar, K.; Sato, M.; Ohta, M.; Mishima, N. Association between serum 25-hydroxyvitamin D and depressive symptoms in Japanese: Analysis by survey season. Eur. J. Clin. Nutr. 2009, 63, 1444–1447. [CrossRef] 51. Pan, A.; Lu, L.; Franco, O.H.; Yu, Z.; Li, H.; Lin, X. Association between depressive symptoms and 25-hydroxyvitamin D in middle-aged and elderly Chinese. J. Affect. Disord. 2009, 118, 240–243. [CrossRef] [PubMed] 52. Milaneschi, Y.; Hoogendijk, W.; Lips, P.; Heijboer, A.C.; Schoevers, R.; van Hemert, A.M.; Beekman, A.T.; Smit, J.H.; Penninx, B.W. The association between low vitamin D and depressive disorders. Mol. Psychiatry 2014, 19, 444–451. [CrossRef] [PubMed] 53. Ronaldson, A.; Arias de la Torre, J.; Gaughran, F.; Bakolis, I.; Hatch, S.L.; Hotopf, M.; Dregan, A. Prospective associations between vitamin D and depression in middle-aged adults: Findings from the UK Biobank cohort. Psychol. Med. 2020, 10, 1–9. [CrossRef] 54. Laurén, J.; Airaksinen, M.S.; Saarma, M.; Timmusk, T. A novel gene family encoding leucine-rich repeat transmembrane proteins differentially expressed in the nervous system. Genomics 2003, 81, 411–421. [CrossRef] Nutrients 2021, 13, 3343 14 of 15 55. Reichman, R.D.; Gaynor, S.C.; Monson, E.T.; Gaine, M.E.; Parsons, M.G.; Zandi, P.P.; Potash, J.B.; Willour, V.L. Targeted sequencing of the LRRTM gene family in suicide attempters with bipolar disorder. Am. J. Med. Genet. B Neuropsychiatr. Genet. 2020, 183, 128–139. [CrossRef] 56. Marini, F.; Bartoccini, E.; Cascianelli, G.; Voccoli, V.; Baviglia, M.G.; Magni, M.V.; Garcia-Gil, M.; Albi, E. Effect of 1alpha,25- dihydroxyvitamin D3 in embryonic hippocampal cells. Hippocampus 2010, 20, 696–705. [CrossRef] 57. Eyles, D.; Almeras, L.; Benech, P.; Patatian, A.; Mackay-Sim, A.; McGrath, J.; Féron, F. Developmental vitamin D deﬁciency alters the expression of genes encoding mitochondrial, cytoskeletal and synaptic proteins in the adult rat brain. J. Steroid Biochem. Mol. Biol. 2007, 103, 538–545. [CrossRef] Shamseldin, H.E.; Masuho, I.; Alenizi, A.; Alyamani, S.; Patil, D.N.; Ibrahim, N.; Martemyanov, K.A.; Alkuraya, F.S. GNB5 mutation causes a novel neuropsychiatric disorder featuring attention deﬁcit hyperactivity disorder, severely impaired language development and normal cognition. Genome Biol. 2016, 17, 195. [CrossRef] 58. 59. Gezen-Ak, D.; Dursun, E.; Yilmazer, S. The effects of vitamin D receptor silencing on the expression of LVSCC-A1C and LVSCC-A1D and the release of NGF in cortical neurons. PLoS ONE 2011, 6, e17553. [CrossRef] 60. Grimm, M.O.W.; Lauer, A.A.; Grösgen, S.; Thiel, A.; Lehmann, J.; Winkler, J.; Janitschke, D.; Herr, C.; Beisswenger, C.; Bals, R.; et al. Proﬁling of Alzheimer’s disease related genes in mild to moderate vitamin D hypovitaminosis. J. Nutr. Biochem. 2019, 67, 123–137. [CrossRef] [PubMed] 61. Mai, J.H.; Ou, Z.H.; Chen, L.; Duan, J.; Liao, J.X.; Han, C.X. Intellectual developmental disorder with cardiac arrhythmia syndrome 62. in a family caused by GNB5 variation and literature review. Chin. J. Pediatrics 2020, 58, 833–837. [CrossRef] Sun, W.; Mafﬁe, J.K.; Lin, L.; Petralia, R.S.; Rudy, B.; Hoffman, D.A. DPP6 establishes the A-type K(+) current gradient critical for the regulation of dendritic excitability in CA1 hippocampal neurons. Neuron 2011, 71, 1102–1115. [CrossRef] 63. Cacace, R.; Heeman, B.; Van Mossevelde, S.; De Roeck, A.; Hoogmartens, J.; De Rijk, P.; Gossye, H.; De Vos, K.; De Coster, W.; Strazisar, M.; et al. Loss of DPP6 in neurodegenerative dementia: A genetic player in the dysfunction of neuronal excitability. Acta Neuropathol. 2019, 137, 901–918. [CrossRef] [PubMed] 64. Lin, L.; Murphy, J.G.; Karlsson, R.M.; Petralia, R.S.; Gutzmann, J.J.; Abebe, D.; Wang, Y.X.; Cameron, H.A.; Hoffman, D.A. DPP6 Loss Impacts Hippocampal Synaptic Development and Induces Behavioral Impairments in Recognition, Learning and Memory. Front. Cell Neurosci. 2018, 12, 84. [CrossRef] 65. Tang, B.L. Vesicle transport through interaction with t-SNAREs 1a (Vti1a)’s roles in neurons. Heliyon 2020, 6, e04600. [CrossRef] 66. Oh, W.J.; Gu, C. The role and mechanism-of-action of Sema3E and Plexin-D1 in vascular and neural development. Semin Cell Dev. Biol. 2013, 24, 156–162. [CrossRef] [PubMed] 67. Casale, M.; Borriello, A.; Scianguetta, S.; Roberti, D.; Caiazza, M.; Bencivenga, D.; Tartaglione, I.; Ladogana, S.; Maruzzi, M.; Della Ragione, F.; et al. Hereditary hypochromic microcytic anemia associated with loss-of-function DMT1 gene mutations and absence of liver iron overload. Am. J. Hematol. 2018, 93, E58–E60. [CrossRef] 68. Bastian, T.W.; von Hohenberg, W.C.; Mickelson, D.J.; Lanier, L.M.; Georgieff, M.K. Iron Deﬁciency Impairs Developing Hip- pocampal Neuron Gene Expression, Energy Metabolism, and Dendrite Complexity. Dev. Neurosci. 2016, 38, 264–276. [CrossRef] [PubMed] Saadat, S.M.; De ˘girmenci, ˙I.; Özkan, S.; Saydam, F.; Özdemir Köro ˘glu, Z.; Çolak, E.; Güne¸s, H.V. Is the 1254T > C polymorphism in the DMT1 gene associated with Parkinson’s disease? Neurosci. Lett. 2015, 594, 51–54. [CrossRef] [PubMed] 69. 70. López, L.; Zuluaga, M.J.; Lagos, P.; Agrati, D.; Bedó, G. The Expression of Hypoxia-Induced Gene 1 (Higd1a) in the Central Nervous System of Male and Female Rats Differs According to Age. J. Mol. Neurosci. 2018, 66, 462–473. [CrossRef] 71. Cui, X.; McGrath, J.J.; Burne, T.H.; Mackay-Sim, A.; Eyles, D.W. Maternal vitamin D depletion alters neurogenesis in the developing rat brain. Int. J. Dev. Neurosci. 2007, 25, 227–232. [CrossRef] [PubMed] 72. Brewer, L.D.; Thibault, V.; Chen, K.C.; Langub, M.C.; Landﬁeld, P.W.; Porter, N.M. Vitamin D hormone confers neuroprotection in parallel with downregulation of L-type calcium channel expression in hippocampal neurons. J. Neurosci. 2001, 21, 98–108. [CrossRef] [PubMed] 73. Cross-Disorder Group of the Psychiatric Genomics Consortium. Genomic Relationships, Novel Loci, and Pleiotropic Mechanisms across Eight Psychiatric Disorders. Cell 2019, 179, 1469–1482.e1411. [CrossRef] 74. Garber, J.; Brunwasser, S.M.; Zerr, A.A.; Schwartz, K.T.; Sova, K.; Weersing, V.R. Treatment and Prevention of Depression and Anxiety in Youth: Test of Cross-Over Effects. Depress. Anxiety 2016, 33, 939–959. [CrossRef] 75. Balogh, L.; Tanaka, M.; Török, N.; Vécsei, L.; Taguchi, S. Crosstalk between Existential Phenomenological Psychotherapy and Neurological Sciences in Mood and Anxiety Disorders. Biomedicines 2021, 9, 340. [CrossRef] 76. Tanaka, M.; Vécsei, L. Editorial of Special Issue “Crosstalk between Depression, Anxiety, and Dementia: Comorbidity in Behavioral Neurology and Neuropsychiatry”. Biomedicines 2021, 9, 517. [CrossRef] [PubMed] 77. Middeldorp, C.M.; Cath, D.C.; Van Dyck, R.; Boomsma, D.I. The co-morbidity of anxiety and depression in the perspective of genetic epidemiology. A review of twin and family studies. Psychol. Med. 2005, 35, 611–624. [CrossRef] 78. Thorp, J.G.; Campos, A.I.; Grotzinger, A.D.; Gerring, Z.F.; An, J.; Ong, J.S.; Wang, W.; Shringarpure, S.; Byrne, E.M.; MacGregor, S.; et al. Symptom-level modelling unravels the shared genetic architecture of anxiety and depression. Nat. Hum. Behav. 2021, 1–11. [CrossRef] Nutrients 2021, 13, 3343 15 of 15 79. Gonda, X.; Petschner, P.; Eszlari, N.; Sutori, S.; Gal, Z.; Koncz, S.; Anderson, I.M.; Deakin, B.; Juhasz, G.; Bagdy, G. Effects of Different Stressors Are Modulated by Different Neurobiological Systems: The Role of GABA-A Versus CB1 Receptor Gene Variants in Anxiety and Depression. Front. Cell. Neurosci. 2019, 13, 138. [CrossRef] [PubMed] 80. Clark, L.A.; Cuthbert, B.; Lewis-Fernández, R.; Narrow, W.E.; Reed, G.M. Three Approaches to Understanding and Classifying Mental Disorder: ICD-11, DSM-5, and the National Institute of Mental Health’s Research Domain Criteria (RDoC). Psychol. Sci. Public Interes. 2017, 18, 72–145. [CrossRef]
10.3390_md19090504	Article Phlorotannin and Pigment Content of Native Canopy-Forming Sargassaceae Species Living in Intertidal Rockpools in Brittany (France): Any Relationship with Their Vertical Distribution and Phenology? Camille Jégou 1 and Valérie Stiger-Pouvreau 2,* , Solène Connan 2 , Isabelle Bihannic 2, Stéphane Cérantola 3, Fabienne Guérard 2 1 2 3 Laboratoire de Biotechnologie et Chimie Marine (LBCM) EA 3884, Université de Brest, 6 Rue de l’université, F-29334 Quimper, France; [email protected] Laboratoire des Sciences de l’Environnement (LEMAR) UMR 6539, Université de Brest, CNRS, IRD, Ifremer, F-29280 Plouzane, France; [email protected] (S.C.); [email protected] (I.B.); [email protected] (F.G.) Service Commun de RMN-RPE, Université de Brest, F-29200 Brest, France; [email protected] * Correspondence: [email protected]; Tel.: +33-2-9849-8806 Abstract: Five native Sargassaceae species from Brittany (France) living in rockpools were surveyed over time to investigate photoprotective strategies according to their tidal position. We gave ev- idences for the existence of a species distribution between pools along the shore, with the most dense and smallest individuals in the highest pools. Pigment contents were higher in lower pools, suggesting a photo-adaptive process by which the decreasing light irradiance toward the low shore was compensated by a high production of pigments to ensure efﬁcient photosynthesis. Conversely, no xanthophyll cycle-related photoprotective mechanism was highlighted because high levels of zeaxanthin rarely occurred in the upper shore. Phlorotannins were not involved in photoprotection either; only some lower-shore species exhibited a seasonal trend in phlorotannin levels. The structural complexity of phlorotannins appears more to be a taxonomic than an ecological feature: Ericaria produced simple phloroglucinol while Cystoseira and Gongolaria species exhibited polymers. Conse- quently, tide pools could be considered as light-protected areas on the intertidal zone, in comparison with the exposed emerged substrata where photoprotective mechanisms are essential. Keywords: phenolic compounds; phlorotannins; photoprotective pigments; phenolic compounds structure; bathymetry; tide pool; intertidal algal distribution 1. Introduction In recent years, more attention has been paid to speciﬁc habitats on rocky shores known as “tide pools”. These biotopes represent isolated puddles found all over the rocky substrate on the intertidal zone, naturally retaining water at low tide. Although pools are sometimes not considered as intertidal habitats [1] because of the absence of emersion periods, they cannot be classiﬁed as subtidal areas either, but rather as a refuge area for both intertidal and subtidal organisms which will remain immersed [2]. Nevertheless, as for the emerged substrata, physical and chemical variations of the water in pools have been analyzed according to their tidal level [3–7]. Actually, the composition in macroalgae found in tide pools can vary according to the pool position on the shore [5,8], such as on emerged substrata, or not [9,10]. Tide pools were once considered unstable systems under the dependence of sudden disturbance [11], but a contradictory conclusion was brought out by Astles [12]. In a whole, little is known about ecology in such pools, compared to the extensive literature concerning the emerged intertidal zone, and only fragmental or sometimes contradictory information has been evidenced. For the coast of Brittany (France), Citation: Jégou, C.; Connan, S.; Bihannic, I.; Cérantola, S.; Guérard, F.; Stiger-Pouvreau, V. Phlorotannin and Pigment Content of Native Canopy-Forming Sargassaceae Species Living in Intertidal Rockpools in Brittany (France): Any Relationship with Their Vertical Distribution and Phenology? Mar. Drugs 2021, 19, 504. https://doi.org/ 10.3390/md19090504 Academic Editor: Hitoshi Sashiwa Received: 30 July 2021 Accepted: 2 September 2021 Published: 4 September 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Mar. Drugs 2021, 19, 504. https://doi.org/10.3390/md19090504 https://www.mdpi.com/journal/marinedrugs marine drugs (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Mar. Drugs 2021, 19, 504 2 of 20 a pattern of seaweed distribution depending on the location of the pool along the sea shore has been introduced by Cabioc’h et al. [13], and a bathymetric distribution inside some pools, analog to the algal belts observed along the sea shore, has been evidenced [8]. A particularly important missing piece of data is, in our opinion, a monitoring about typical or even exclusive tide pool species, concerning both distribution patterns and spatial and temporal variations of their populations. Sargassaceae species, altogether with Fucaceae ones, are among the most abundant brown macroalgae occurring along the coasts of Brittany [14,15]. In Europe, main native species from the family Sargassaceae are typically located in intertidal rockpools and/or in the subtidal zone, especially those from the genera Cystoseira, Ericaria and Gongolaria. These genera are quite common in Brittany, and they occupy pools located at nearly all tide levels. Actually, the genera Cystoseira, Ericaria and Gongolaria, settling in the subtidal zone and/or in rockpools of the intertidal zone, have been subject of a few studies on their ecology in Brittany, in comparison with the numerous data available for the surrounding coralline algae [7,16] and Bifurcaria/Fucus/Sargassum species [14,17–19]. Within this Sargassaceae family, different native genera are present in intertidal rockpools from Brittany and constitute large biomass and canopies [13,20]: Cystoseira fœniculacea, C. humilis, Ericaria selaginoides (previously known as Cystoseira tamariscifolia), Gongolaria baccata and G. nodicaulis (previously C. baccata and C. nodicaulis). According to Cabioc’h et al. [13], Cystoseira humilis would be strictly located on the intertidal zone, in exposed pools. In Brittany (Atlantic Ocean), this species exclusively occurs in tide pools, contrary to the Mediterranean Sea, where it can extend down to the subtidal zone [21,22] and references therein. These species can represent an important biomass, making it easily suitable for such a study. Each of the ﬁve species has its own pattern of distribution on the shore [20]; consequently, the populations have to face different environmental conditions, in particular tidal variations followed by emersion of rockpools, and then variability in the quantity and quality of light, and seawater salinity and temperature. As they colonize tide pools, these native Sargassaceae species do not face desiccation trouble. However, there are several abiotic factors varying in pools along the shore [4,5] and especially in Brittany [16,20]. For example, macroalgae in tide pools are submitted to variations in temperature, even if the range of variation is narrower than on the emerged substrata due to the persistence of water in pools. In addition, the light irradiance can dramatically ﬂuctuate in pools depending on their location on the intertidal zone. In the uppermost pools, the intense photosynthetically active and ultraviolet radiations can disturb the development of macroalgae. In this way, different photo-adaptive strategies could be expected for the species, according to their distribution on the shore. Among strategies developed by brown macroalgae face to light, two kinds of compounds are implicated in the tolerance to excessive light irradiance: carotenoids and phlorotannins [23]. Carotenoids are the main class of photoprotective compounds, consisting in xantho- phylls and carotenes. Under high light irradiance, the thermal dissipation of excessive energy is ensured in brown macroalgae by the xanthophyll cycle [24,25]. It involves two de-epoxidations of the photosynthetic violaxanthin which is then transformed in the pho- toprotective zeaxanthin which can dissipate excessive energy [24]. These two pigments can also be involved in the detoxiﬁcation of reactive oxygen species formed during photosyn- thesis, with other carotenoids such as fucoxanthin and β-carotene [26,27]. The xanthophyll cycle has been unequivocally evidenced as a major photoprotective mechanism of the brown macroalgae on the intertidal zone. Gévaert et al. [28,29] showed that the kelp Sac- charina latissima (located on the low intertidal zone) used the xanthophyll cycle to dissipate energy at low tide. Furthermore, Harker et al. [30] proved that Pelvetia canaliculata had a greater xanthophyll pool than S. latissima. This could partially explain the aptitude of the alga to settle very high on the shore. Uhrmacher et al. [31] revealed that photoinhibition was realized by the xanthophyll cycle in the brown macroalga Dictyota dichotoma. On the other side, phlorotannins belongs to a family of biomolecules typical of brown macroalgae, derived from the polymerization of phloroglucinol (1,3,5-trihydroxybenzene, Mar. Drugs 2021, 19, 504 3 of 20 126 Da). They are thus part of the group of polyphenols. These phlorotannins are involved in both the primary and secondary metabolism of Phaeophyceae [32–34]. Phlorotannins are produced by condensation of malonate and acetate under the action of a polyketide synthase (PKS) enzyme [35,36]. The ﬁnal structure of each phlorotannin depends on three parameters: (1) The degree of polymerization, which indicates the number of phlorogluci- nol units assembled to form the molecule in question. This degree is extremely variable, from oligomers up to 650 kDa polymers [37]. (2) The type of polymerization, which reﬂects the way the phloroglucinol units are linked. Linkages occur in three different ways, accord- ing to [38–40]: Aryl-Aryl (C-C) in the case of Fucols, Diaryl-Ether (C-O-C) in the case of Phlorethols and (Iso) Fuhalols, Dibenzo-dioxin (C-O-C) in the case of Eckols and Carmalols, reviewed by Stiger-Pouvreau et al. [41] for Sargassaceae species from Brittany. (3) The presence of substitutions on the rings, including halogen [42] and sulphate [43]. Numerous studies have investigated their ecological and cellular roles. Phlorotannins are notably thought to have a sunscreen action, as demonstrated in Halidrys siliquosa [44]. Lalegerie et al. [45] reviewed the ecological and ecophysiological roles of phenolic compounds from red and brown seaweeds, and Gager et al. [46] their biological activities. Our study was designed to evidence the distinctive distribution of the main native canopy-forming genera of Sargassaceae (Phaeophyceae, Fucales), i.e., Cystoseira, Ericaria and Gongolaria, occurring in tide pools from Brittany, according to their position on the shore. For this purpose, we have compiled three different studies (from winter 2009 to spring 2011) in order to investigate the involvement of pigments and phlorotannins (composition and type) on the phenology and distribution of the considered species. We have also analyzed the distribution of the ﬁve species inhabiting tide pools in Brittany, at three tidal heights, and surveyed the spatial and temporal variations of some biological and chemical (phlorotannins and pigments) characteristics of their populations. Our purpose was to verify whether the photosynthetic capacity and the photoprotective mechanisms are speciﬁc characters, hence explaining the distribution of the species, or if they can be modulated within populations according to the tidal height where individuals settle. Finally, we present the type of phlorotannins produced by each species, as protective compounds for their persistence in intertidal rockpools. 2. Results 2.1. Distribution of the Five Species along the Intertidal Rockyshore The transect showed a typical distribution pattern of the ﬁve species. On Figure 1 are plotted the position of the quadrats where individuals of each species were found. The species richness of each pool is also presented, as well as the correspondence of the level on the shore with the classical algal belts present on intertidal temperate rocky shores. In summary, the species succession from the upper to the lower shore was as follows: Cystoseira humilis, Gongolaria nodicaulis, C. fœniculacea, G. baccata, and Ericaria selaginoides, with a decrease in rockpool species richness in the upper shore (Figure 1). Cystoseira humilis and E. selaginoides had a narrow distribution on the shore. Cystoseira humilis was strictly settled in the upper pools, just underneath the Fucus spiralis belt. The macroalgal species richness associated with the rockpools dominated by C. humilis was low (seven species) (Figure 1). Ericaria selaginoides was located roughly from the F. serratus belt to the Bifurcaria bifurcata belt, in the very lower shore. In such pools, the high species richness (from 13 to 35 different taxa) is mainly due to the occurrence of red macroalgae. Gongolaria baccata, C. fœniculacea and G. nodicaulis had a wider distribution and colonized several pools along the rocky shore. The ﬁrst had a scattered distribution under the Ascophyllum nodosum/F. vesiculosus belt (species richness from 10 to 35), and the two others were located on the mid shore between F. spiralis and F. serratus belts (species richness from 7 to 13), where brown, green and red macroalgae coexist. The salinity of each pool was quite stable and around 35.3 over the 6-h period of measurement during the one-year survey (from spring 2 to spring 3). Conversely, temperature varied with seasons, and interestingly, during the day, in ranges depending on the considered pool. A monitoring of Mar. Drugs 2021, 19, 504 4 of 20 the variability of the seawater temperature was made during a winter day as presented in Figure 1: a range of nearly 7 ◦C was noted for the upper rockpools, while on the lower shore the range was of nearly 5 ◦C. The middle part of the shore represented an area where only a 3 ◦C difference was observed between the coldest and the warmest temperature Figure 1. Compiled distribution of individuals from the ﬁve species of Cystoseira, Gongolaria and Ericaria along the intertidal zone at Penmarc’h (Brittany) over a one-year survey, from spring 2 to spring 3. The presence of individuals in one quadrat on a precise position along the transect is indicated by one semi-transparent grey circle. Correspondence between levels and Fucales belts is given (right side), as well as the observed macroalgal species richness, temperature amplitude of rockpools for 6 h on a winter day (left side). Proﬁles of young apical branches of the ﬁve studied species from seaweed specimens are also drawn; scale bar: 1 cm. 2.2. Phenological Variables On the Table 1, the density varied signiﬁcantly (p < 0.001), with maximal values for the populations of C. humilis especially in summer; then the density decreased for C. fœniculacea, G. nodicaulis and G. baccata, and was minimal for E. selaginoides. The percentage cover is maximal in lower shore species, reaching 90% and 91.67% for G. baccata and E. selaginoides, respectively. Regarding the maturity index observed in the quadrats, it appeared maximal for G. nodicaulis and G. baccata (with no statistical difference between them), and the individuals of the other species less frequently bore receptacles (Table 1). The G. baccata population turned out to be only made of mature individuals, no matter the season. The individual length signiﬁcantly differed between species (p < 0.001), and from the smallest to the tallest, the order was Cystoseira humilis, G. nodicaulis/C. fœniculacea, E. selaginoides and the largest G. baccata. The maximal lengths of individuals were found in populations where the density remained low (i.e., G. baccata and E. selaginoides), and at the opposite, C. humilis which could present the densest population was characterized by the smallest individuals (Table 1). Cystoseira fœniculacea and G. nodicaulis were intermediate; individuals of these species were neither small nor tall, and a medium density range was observed in the quadrats. Site, season and their interaction affected the length in the C. humilis populations, and fertile individuals could be observed anytime of the year. Cystoseira fœniculacea and G. nodicaulis thalli appeared taller than C. humilis (Table 1). Gongolaria baccata and E. selaginoides were characterized by the tallest thalli and the lowest density. The length of both species was affected by season, but only the latter has a marked Mar. Drugs 2021, 19, x FOR PEER REVIEW 4 of 20 Figure 1. Compiled distribution of individuals from the five species of Cystoseira, Gongolaria and Ericaria along the intertidal zone at Penmarc’h (Brittany) over a one-year survey, from spring 2 to spring 3. The presence of individuals in one quadrat on a precise position along the transect is indi-cated by one semi-transparent grey circle. Correspondence between levels and Fucales belts is given (right side), as well as the observed macroalgal species richness, temperature amplitude of rock-pools for 6 h on a winter day (left side). Profiles of young apical branches of the five studied species from seaweed specimens are also drawn; scale bar: 1 cm. Cystoseira humilis and E. selaginoides had a narrow distribution on the shore. Cystoseira humilis was strictly settled in the upper pools, just underneath the Fucus spiralis belt. The macroalgal species richness associated with the rockpools dominated by C. humilis was low (seven species) (Figure 1). Ericaria selaginoides was located roughly from the F. serratus belt to the Bifurcaria bifurcata belt, in the very lower shore. In such pools, the high species richness (from 13 to 35 different taxa) is mainly due to the occurrence of red macroalgae. Gongolaria baccata, C. fœniculacea and G. nodicaulis had a wider distribution and colonized several pools along the rocky shore. The first had a scattered distribution under the Ascophyllum nodosum / F. vesiculosus belt (species richness from 10 to 35), and the two oth-ers were located on the mid shore between F. spiralis and F. serratus belts (species richness from 7 to 13), where brown, green and red macroalgae coexist. The salinity of each pool was quite stable and around 35.3 over the 6-h period of measurement during the one-year survey (from spring 2 to spring 3). Conversely, temperature varied with seasons, and in-terestingly, during the day, in ranges depending on the considered pool. A monitoring of the variability of the seawater temperature was made during a winter day as presented in Figure 1: a range of nearly 7 °C was noted for the upper rockpools, while on the lower shore the range was of nearly 5 °C. The middle part of the shore represented an area where only a 3 °C difference was observed between the coldest and the warmest temperature 2.2. Phenological Variables On the Table 1, the density varied significantly (p < 0.001), with maximal values for the populations of C. humilis especially in summer; then the density decreased for C. fœnic-ulacea, G. nodicaulis and G. baccata, and was minimal for E. selaginoides. The percentage cover is maximal in lower shore species, reaching 90% and 91.67% for G. baccata and Mar. Drugs 2021, 19, 504 5 of 20 seasonal growth, beginning in late winter, and ending around the summer when receptacles are fully developed, in accordance with the observations on the maturity index (Table 1). Table 1. Phenological variables from the ﬁve species of concern: the upper species Cystoseira humilis, the two median species C. fœniculacea and Gongolaria nodicaulis and the two lower species G. baccata and Ericaria selaginoides. Density represents the number of individuals counted within a surface of 0.25 m2. Variables were monitored from winter 2009 (=winter 1) to summer 2010 (=summer 2). W: winter, Sp: spring, Su: summer, A: autumn. Density (ind/0.25 m2) Percentage cover Maturity Mean length of individuals Min. Max. Min. Max. Min. Max. Min. Max. Upper Position C. humilis 2 (W1) 8.25 (Su2) 12.5% (W1) 60% (A1) 16% (W1) 100% (Sp/Su2) 1.33 (W1) 5.33 (Su2) 17.5% (Sp1) 56.67% (A1/W1) 0% (A) 100% (Sp) Position of the Rockpool along the Intertidal Zone Median Position G. nodicaulis C. fœniculacea G. baccata 1 (W1) 4.33 (Su1) 1 2 (A1) 13.33% (W1) 26.67% (W1) 10% (Sp2) 50% (Su2) 90% (Sp1) 91.67% (Su1) 22.36% (Su1) 100% (W/Sp1) 100% all year round Lower Position E. selaginoides 1 (Sp1/A1/W1) 1.33 (Su1) 0% (W/Sp) 100% (Su/A) 25.66 cm (W2) 56.67 cm (Sp1/A1) 21.63 cm (Su2) 15.68 cm (Su2) 22.36 cm (Su1) 53.67 cm (Su2) 33.05 cm (W2) 48.40 cm (W1) 58.33 cm (Sp1) 102.33 cm (Sp1) Number of macroalgal taxa in the rockpool 7 7–13 7–13 10–35 13–35 2.3. Pigment Composition and Inter-Species Variability In all samples, the major pigments were chlorophyll a, fucoxanthin, and chlorophyll c2 (Figure 2). Their maximal mean contents reached 2.8, 1.0 and 0.3 mg·g−1 algal DW, respectively. Maximal values were always observed in E. selaginoides. Violaxanthin, β- carotene and zeaxanthin contents were characterized by lower values, under 0.2, 0.14 and 0.08 mg·g−1 DW, respectively. All pigment contents depended on the considered species (Table S1, p < 0.05). The “tidal height” effect could be evidenced only for chlorophyll a, fucoxanthin and violaxanthin. There was no signiﬁcant interaction between the two factors, “species” and “tidal heights”, which means that the effect of tidal height on chlorophyll a, fucoxanthin and violaxanthin content did not depend on the species, and vice versa. Hence, for these three pigments, the content globally increased from the upper to the lower intertidal pools (HSD, p < 0.05; Figure 2). The low shore species E. selaginoides and the low/mid shore G. baccata were characterized by high content for all pigments, except zeaxanthin for which they had minimal values (Figure 2, HSD test: p < 0.05). Conversely, the upper alga C. humilis had the maximal zeaxanthin level and low contents for the other pigments. Even if this was not always supported by statistical analyses, the levels of chlorophyll a, fucoxanthin, chlorophyll c2, violaxanthin and β- carotene decreased for each species, from one location on the shore to a lower one (Figure 2). We evidenced a similar evolution of several pigments within species and tidal height using Pearson’s correlation test (Table S2): except zeaxanthin, all pigments were highly correlated (0.71 < ρ <0.90; p < 0.001) and thus evolved in the same manner through species and tidal height. The results for zeaxanthin are less clear; its highest levels occurred in the uppermost pools with C. humilis (around 0.06 mg.g−1 DW), but no statistical difference was observed in the rest of the dataset (Tukey’s HSD test, Figure 2). Mar. Drugs 2021, 19, 504 6 of 20 Figure 2. Pigment contents (mg·g−1 algal DW) in spring 2011 (=spring 3) in the ﬁve native Sargassaceae species from Brittany (France) according to their position on the shore. Ericaria selaginoides “S”, Gongolaria baccata “B”, Cystoseira fœniculacea “F”, G. nodicaulis “N” and C. humilis “H”. a, b: for each pigment analysis, values sharing a common letter were not statistically different according to Tukey’s HSD test. The scale for each graph is not similar. Mean + sd; n = 6 for each species and position along the shore. 2.4. Inter-Species and Inter-Seasonal Variabilities of Phenolic Contents The average phenolic compound (PC) content varied according to the ﬁve species and the sampling season. The ﬁve species showed signiﬁcantly different PC contents (Figure 3, Kruskal–Wallis, p < 0.001): Ericaria selaginoides had the highest mean PC content (0.42 ± 0.21% DW), followed by G. nodicaulis and G. baccata with, in average, 0.25 ± 0.16% DW and 0.21 ± 0.10% DW, respectively. The PC content of C. humilis (0.19 ± 0.09% DW) was not signiﬁcantly different than G. baccata one. Finally, C. fœniculacea (0.13 ± 0.06% DW) contained the lowest PC level. Among the ﬁve species studied, two groups can be distinguished (Figure 3): a ﬁrst group of two species with no signiﬁcant seasonal trend in phenolic content (C. fœniculacea and C. humilis), with PC contents ranging from 0.08 ± 0.03% DW (spring 1) to 0.15 ± 0.06% DW (summer 2) and 0.10 ± 0.04% DW (spring 1) and 0.21 ± 0.01% DW (winter 2), respectively. Mar. Drugs 2021, 19, x FOR PEER REVIEW 6 of 20 Figure 2. Pigment contents (mg·g−1 algal DW) in spring 2011 (=spring 3) in the five native Sargassaceae species from Brit-tany (France) according to their position on the shore. Ericaria selaginoides “S”, Gongolaria baccata “B”, Cystoseira fœniculacea “F”, G. nodicaulis “N” and C. humilis “H”. a, b: for each pigment analysis, values sharing a common letter were not statis-tically different according to Tukey’s HSD test. The scale for each graph is not similar. Mean + sd; n = 6 for each species and position along the shore. Conversely, the upper alga C. humilis had the maximal zeaxanthin level and low con-tents for the other pigments. Even if this was not always supported by statistical analyses, the levels of chlorophyll a, fucoxanthin, chlorophyll c2, violaxanthin and β-carotene de-creased for each species, from one location on the shore to a lower one (Figure 2). We evidenced a similar evolution of several pigments within species and tidal height using Pearson’s correlation test (Table S2): except zeaxanthin, all pigments were highly corre-lated (0.71 < ρ <0.90; p < 0.001) and thus evolved in the same manner through species and tidal height. The results for zeaxanthin are less clear; its highest levels occurred in the uppermost pools with C. humilis (around 0.06 mg.g−1 DW), but no statistical difference was observed in the rest of the dataset (Tukey’s HSD test, Figure 2). Mar. Drugs 2021, 19, 504 7 of 20 Figure 3. Evolution of the phenolic contents (% DW), in phloroglucinol equivalent, of semi-puriﬁed extracts from the ﬁve species of Cystoseira, Gongolaria and Ericaria. a, b, c: two same letters found in a species during two seasons indicate that their contents are not statistically different, after ANOVA. N.D.: no data, as E. selaginoides was represented only by basal parts with no lateral. Mean ± standard deviation; n = 18 for each studied species and each season. In contrast, the three other species showed marked seasonal variation: Gongolaria baccata showed a decrease in PC content from 0.27 ± 0.14% DW in summer 1 to 0.12 ± 0.03% DW in winter 2 (Figure 3). Conversely, the PC levels of G. nodicaulis oscillated between a maximum (0.35 ± 0.19% DW) from autumn 1 to winter 2, and a minimum observed before and after this period (0.04 ± 0.01% DW in summer 1). Ericaria selaginoides showed averaged phenolic content from spring to summer 1 (0.38 ± 0.15% DW). Then, its PC content decreased to reach its minimum in autumn 1 (0.11 ± 0.05% DW). In winter 2, due to the senescence of the species (only the basal parts remained), no sample has been collected. In spring 2, high levels can be observed (Figure 3), which increased again to a maximum in the following summer (0.60 ± 0.13% DW). Considering the variability of phenolic content within a year, the three species from the upper shore, Cystoseira humilis, C. fœniculacea and Gongolaria nodicaulis, have higher levels in winter and lower levels in spring/summer, while the two lower species, G. baccata and Ericaria selaginoides, have higher levels in spring/summer and lower levels in winter. Mar. Drugs 2021, 19, x FOR PEER REVIEW 7 of 20 2.4. Inter-Species and Inter-Seasonal Variabilities of Phenolic Contents The average phenolic compound (PC) content varied according to the five species and the sampling season. The five species showed significantly different PC contents (Fig-ure 3, Kruskal–Wallis, p < 0.001): Ericaria selaginoides had the highest mean PC content (0.42 ± 0.21% DW), followed by G. nodicaulis and G. baccata with, in average, 0.25 ± 0.16% DW and 0.21 ± 0.10% DW, respectively. The PC content of C. humilis (0.19 ± 0.09% DW) was not significantly different than G. baccata one. Finally, C. fœniculacea (0.13 ± 0.06% DW) contained the lowest PC level. Among the five species studied, two groups can be distin-guished (Figure 3): a first group of two species with no significant seasonal trend in phe-nolic content (C. fœniculacea and C. humilis), with PC contents ranging from 0.08 ± 0.03% DW (spring 1) to 0.15 ± 0.06% DW (summer 2) and 0.10 ± 0.04% DW (spring 1) and 0.21 ± 0.01% DW (winter 2), respectively. Figure 3. Evolution of the phenolic contents (% DW), in phloroglucinol equivalent, of semi-purified extracts from the five species of Cystoseira, Gongolaria and Ericaria. a, b, c: two same letters found in a species during two seasons indicate that their contents are not statistically different, after ANOVA. N.D.: no data, as E. selaginoides was represented only by basal parts with no lateral. Mean ± standard deviation; n = 18 for each studied species and each season. Mar. Drugs 2021, 19, 504 8 of 20 2.5. Puriﬁcation of Phlorotannins and NMR Analyses Figure 4 shows 2D 1H-13C HMBC and 1D 1H NMR spectra of the puriﬁed fractions for the ﬁve Cystoseira, Ericaria and Gongolaria species. All puriﬁed fractions presented different aromatic signals in the chemical shift zone between 5.7 and 6.5 ppm, attesting to the potential presence of phlorotannins (Figure 4). Moreover, these signals were different depending on the species. We can also highlight the presence of aliphatic signals of non- negligible intensity in the fractions of G. baccata, C. fœniculacea, C. humilis and G. nodicaulis, which attests the presence of compounds other than phlorotannins. On the other hand, the puriﬁed fraction of E. selaginoides indicated the overwhelming presence of an aromatic compound. It appears as a singlet at 5.78 ppm in deuterated methanol (Figure 4), indicating a single type of proton in the molecule, all chemically equivalent. This signal is due to phloroglucinol, the monomer at the origin of phlorotannins. By observing the diversity of phlorotannins between each species (form and number of peaks around 6 ppm, Figure 4), one should note that these molecules, as polymers of phloroglucinol, share a structural characteristic: among all these molecules, the protons visible in 1H NMR in MeOD are those of the methine groups (C-H of the aromatic rings). Following the results obtained on 1H NMR spectra, a 2D 1H-13C NMR analysis of the puriﬁed fractions was carried out for four of the ﬁve species studied. Indeed, the low phenolic content in C. fœniculacea, associated with the chemical instability of the puriﬁed fraction, did not allow us to present a Heteronuclear Multiple Bond Correlation (HMBC) spectrum for this species. The HMBC spectrum (Figure 4) of Gongolaria nodicaulis showed a multitude of signals on the 1H dimension, which correlated with methine-like carbons (C-H of the aromatic rings), phenol-like carbons (C-OH), rare carbons involved in aryl-aryl ring bonds (C-C) and a majority of carbons involved in diaryl-ether ring bonds (C-O-C). In G. baccata, a smaller number of correlation spots were observed (Figure 4). We can thus note, opposite the signal at 5.98 ppm in 1H dimension, a large correlation spot at 101 ppm in 13C dimension. However, there is no correlation spot between 120 and 150 ppm, suggesting the presence of a large linear polymer of the fucol type. Indeed, lateral branching would generate small variations in chemical shifts in the 1H dimension, and a less well-deﬁned correlation spot would then be observed. By similar reasoning, we can detect a phlorotannin of the phlorethol type (δ = 5.92 ppm in 1H dimension), and a fucophlorethol (δ = 6.10 ppm). There was also a signal at δ = 5.78 ppm (isolated signal in 1H dimension), which correlated only with methines and phenolic carbons (Figure 4). This compound, which translated to a singlet in 1H dimension, present in the pure state in the puriﬁed fraction of E. selaginoides (Figure 4), was also the major compound in the puriﬁed fraction of C. humilis (Figure 4) and identiﬁed as phloroglucinol. In C. humilis, we also found a linear or slightly branched phlorethol-type compound (δ = 5.92 ppm in 1H dimension), as well as another compound at δ = 5.88 ppm (1H dimension), which showed neither an aryl-aryl bond nor a diaryl-ether bond (Figure 4). This compound, which “looked like” phloroglucinol (1H δ = 5.78 ppm), differed by a doubling of the C-OH correlation spot around 160–163 ppm in 13C dimension. In E. selaginoides, the puriﬁed fraction showed only one major compound identiﬁed as phloroglucinol. Mar. Drugs 2021, 19, 504 9 of 20 Figure 4. 1H-13C HMBC NMR spectra of the puriﬁed fractions (solvent: MeOD). Horizontal axis: 1H dimension; vertical axis: 13C dimension. Only a 1H spectrum is given for C. fœniculacea. Grey box: phlorotannins area (circa 5.7–6.5 ppm in 1H dimension); green (90–100 ppm in 13C dimension): methine zone; red (100–105 ppm): carbons involved in an aryl-aryl bond between two phloroglucinol units; yellow (120–150 ppm): diaryl-ether bond; blue (150–165 ppm): carbons with phenol function. Mar. Drugs 2021, 19, x FOR PEER REVIEW 9 of 20 Figure 4. 1H-13C HMBC NMR spectra of the purified fractions (solvent: MeOD). Horizontal axis: 1H dimension; vertical axis: 13C dimension. Only a 1H spectrum is given for C. fœniculacea. Grey box: phlorotannins area (circa 5.7–6.5 ppm in 1H dimension); green (90–100 ppm in 13C dimension): methine zone; red (100–105 ppm): carbons involved in an aryl-aryl bond between two phloroglucinol units; yellow (120–150 ppm): diaryl-ether bond; blue (150–165 ppm): carbons with phenol function. Mar. Drugs 2021, 19, 504 10 of 20 In summary, Table 2 presents the type of phlorotannins that could be identiﬁed in the ﬁve species of concern. The upper species C. humilis produced phloroglucinol and phlorethols. The median species G. nodicaulis did not produce phloroglucinol but three types of phlorotannins: fucols, phlorethols and fucophlorethols. Table 2. Types of phlorotannins observed in our study in the ﬁve species of Cystoseira, Gongolaria and Ericaria. Examples of phlorotannin structures are also given. Mean total phenolic content (TPC) was calculated from spring 1 to summer 2. Mean TPC (% DW) Type of Phlorotannins (from 2D NMR Results) Phloroglucinol Fucol Phlorethol Fucophlorethol C. humilis C. fœniculacea G. nodicaulis G. baccata E. selaginoides 0.19 ± 0.09 0.13 ± 0.06 0.25 ± 0.16 0.21 ± 0.10 0.42 ± 0.21 Yes / No Traces Yes No / Yes Yes No Yes / Yes Yes No No / Yes Yes No Examples of chemical structures of phlorotannins Both lower species, G. baccata and E. selaginoides produced different types of phlorotan- nins: E. selaginoides synthetized only the monomer phloroglucinol, conversely to G. baccata which was able to produce phloroglucinol (traces), and the three types of polymers, fucols, phlorethols and fucophlorethols (Table 2). 3. Discussion This study focuses on the zonation pattern of ﬁve native canopy-forming Sargassaceae species occupying intertidal rockpools in Brittany (France) and investigates a potential relationship between tidal height, chemical content (pigments, phlorotannins) and phe- nological variables. Using ﬁeld data, we specify the distribution of each species on the foreshore of Penmarc’h, which is generalizable to the Brittany region. Thus, Gongolaria baccata and Ericaria selaginoides live at the limit of low tide levels; G. nodicaulis and Cystoseira fœniculacea are at the mid-tide level, and C. humilis at the high tide level. 3.1. Photo-Adaptation along the Shore but No Photoprotective Pigments in the Upper Shore Chlorophyll a, fucoxanthin and chlorophyll c2 are naturally the main pigments for all species at all tidal levels, as they are involved in photosynthesis [47]. Similar results were obtained from two brown macroalgae, Pelvetia canaliculata [48] and Saccharina latissima [49]. We also observed minimal photosynthetic pigments contents for macroalgae settled in the upper shore, and maximum in the lower shore (Figure 2). For example, E. selaginoides produced more pigments than C. humilis (except zeaxanthin), and G. baccata produced more chlorophyll when living in the lower intertidal zone than in the mid-zone. This increase in pigment production in species living in lower levels counterbalanced the reduced level of irradiance in this lower intertidal zone, as already suggested [50–52]. Furthermore, on the higher shore, high levels of irradiance or UV radiations could reduce pigment levels, as experimentally evidenced for Fucus vesiculosus [53]. On the other hand, no clear pattern could be observed looking at the evolution of the photoprotective zeaxanthin levels on different tidal heights (Figure 2). Only differences between species were observed, due to high levels occurring in Cystoseira humilis and in the mid C. fœniculacea populations. However, low zeaxanthin levels were determined in the upper individuals of G. nodicaulis and C. fœniculacea (Figure 2). The literature reports indicated that zeaxanthin production was deeply implicated in photoprotective mechanisms through the xanthophyll cycle via the double de-epoxidations of violaxanthin as it was demonstrated in the lower shore Saccharina latissima [28,29] and in the upper shore Pelvetia canaliculata [30]. Moreover, very Mar. Drugs 2021, 19, x FOR PEER REVIEW 10 of 20 Following the results obtained on 1H NMR spectra, a 2D 1H-13C NMR analysis of the purified fractions was carried out for four of the five species studied. Indeed, the low phe-nolic content in C. fœniculacea, associated with the chemical instability of the purified frac-tion, did not allow us to present a Heteronuclear Multiple Bond Correlation (HMBC) spec-trum for this species. The HMBC spectrum (Figure 4) of Gongolaria nodicaulis showed a multitude of signals on the 1H dimension, which correlated with methine-like carbons (C-H of the aromatic rings), phenol-like carbons (C-OH), rare carbons involved in aryl-aryl ring bonds (C-C) and a majority of carbons involved in diaryl-ether ring bonds (C-O-C). In G. baccata, a smaller number of correlation spots were observed (Figure 4). We can thus note, opposite the signal at 5.98 ppm in 1H dimension, a large correlation spot at 101 ppm in 13C dimen-sion. However, there is no correlation spot between 120 and 150 ppm, suggesting the pres-ence of a large linear polymer of the fucol type. Indeed, lateral branching would generate small variations in chemical shifts in the 1H dimension, and a less well-defined correlation spot would then be observed. By similar reasoning, we can detect a phlorotannin of the phlorethol type (δ = 5.92 ppm in 1H dimension), and a fucophlorethol (δ = 6.10 ppm). There was also a signal at δ = 5.78 ppm (isolated signal in 1H dimension), which correlated only with methines and phenolic carbons (Figure 4). This compound, which translated to a singlet in 1H dimension, present in the pure state in the purified fraction of E. selaginoides (Figure 4), was also the major compound in the purified fraction of C. humilis (Figure 4) and identified as phloroglucinol. In C. humilis, we also found a linear or slightly branched phlorethol-type compound (δ = 5.92 ppm in 1H dimension), as well as another compound at δ = 5.88 ppm (1H dimension), which showed neither an aryl-aryl bond nor a diaryl-ether bond (Figure 4). This compound, which “looked like” phloroglucinol (1H δ = 5.78 ppm), differed by a doubling of the C-OH correlation spot around 160–163 ppm in 13C dimen-sion. In E. selaginoides, the purified fraction showed only one major compound identified as phloroglucinol. In summary, Table 2 presents the type of phlorotannins that could be identified in the five species of concern. The upper species C. humilis produced phloroglucinol and phlorethols. The median species G. nodicaulis did not produce phloroglucinol but three types of phlorotannins: fucols, phlorethols and fucophlorethols. Table 2. Types of phlorotannins observed in our study in the five species of Cystoseira, Gongolaria and Ericaria. Examples of phlorotannin structures are also given. Mean total phenolic content (TPC) was calculated from spring 1 to summer 2. Mean TPC (% DW) Type of Phlorotannins (from 2D NMR Results) Phloroglucinol Fucol Phlorethol Fucophlorethol C. humilis 0.19 ± 0.09 Yes No Yes No C. fœniculacea 0.13 ± 0.06 / / / / G. nodicaulis 0.25 ± 0.16 No Yes Yes Yes G. baccata 0.21 ± 0.10 Traces Yes Yes Yes E. selaginoides 0.42 ± 0.21 Yes No No No Examples of chemical structures of phlorotannins Both lower species, G. baccata and E. selaginoides produced different types of phloro-tannins: E. selaginoides synthetized only the monomer phloroglucinol, conversely to G. baccata which was able to produce phloroglucinol (traces), and the three types of poly-mers, fucols, phlorethols and fucophlorethols (Table 2). 3. Discussion Mar. Drugs 2021, 19, x FOR PEER REVIEW 10 of 20 Following the results obtained on 1H NMR spectra, a 2D 1H-13C NMR analysis of the purified fractions was carried out for four of the five species studied. Indeed, the low phe-nolic content in C. fœniculacea, associated with the chemical instability of the purified frac-tion, did not allow us to present a Heteronuclear Multiple Bond Correlation (HMBC) spec-trum for this species. The HMBC spectrum (Figure 4) of Gongolaria nodicaulis showed a multitude of signals on the 1H dimension, which correlated with methine-like carbons (C-H of the aromatic rings), phenol-like carbons (C-OH), rare carbons involved in aryl-aryl ring bonds (C-C) and a majority of carbons involved in diaryl-ether ring bonds (C-O-C). In G. baccata, a smaller number of correlation spots were observed (Figure 4). We can thus note, opposite the signal at 5.98 ppm in 1H dimension, a large correlation spot at 101 ppm in 13C dimen-sion. However, there is no correlation spot between 120 and 150 ppm, suggesting the pres-ence of a large linear polymer of the fucol type. Indeed, lateral branching would generate small variations in chemical shifts in the 1H dimension, and a less well-defined correlation spot would then be observed. By similar reasoning, we can detect a phlorotannin of the phlorethol type (δ = 5.92 ppm in 1H dimension), and a fucophlorethol (δ = 6.10 ppm). There was also a signal at δ = 5.78 ppm (isolated signal in 1H dimension), which correlated only with methines and phenolic carbons (Figure 4). This compound, which translated to a singlet in 1H dimension, present in the pure state in the purified fraction of E. selaginoides (Figure 4), was also the major compound in the purified fraction of C. humilis (Figure 4) and identified as phloroglucinol. In C. humilis, we also found a linear or slightly branched phlorethol-type compound (δ = 5.92 ppm in 1H dimension), as well as another compound at δ = 5.88 ppm (1H dimension), which showed neither an aryl-aryl bond nor a diaryl-ether bond (Figure 4). This compound, which “looked like” phloroglucinol (1H δ = 5.78 ppm), differed by a doubling of the C-OH correlation spot around 160–163 ppm in 13C dimen-sion. In E. selaginoides, the purified fraction showed only one major compound identified as phloroglucinol. In summary, Table 2 presents the type of phlorotannins that could be identified in the five species of concern. The upper species C. humilis produced phloroglucinol and phlorethols. The median species G. nodicaulis did not produce phloroglucinol but three types of phlorotannins: fucols, phlorethols and fucophlorethols. Table 2. Types of phlorotannins observed in our study in the five species of Cystoseira, Gongolaria and Ericaria. Examples of phlorotannin structures are also given. Mean total phenolic content (TPC) was calculated from spring 1 to summer 2. Mean TPC (% DW) Type of Phlorotannins (from 2D NMR Results) Phloroglucinol Fucol Phlorethol Fucophlorethol C. humilis 0.19 ± 0.09 Yes No Yes No C. fœniculacea 0.13 ± 0.06 / / / / G. nodicaulis 0.25 ± 0.16 No Yes Yes Yes G. baccata 0.21 ± 0.10 Traces Yes Yes Yes E. selaginoides 0.42 ± 0.21 Yes No No No Examples of chemical structures of phlorotannins Both lower species, G. baccata and E. selaginoides produced different types of phloro-tannins: E. selaginoides synthetized only the monomer phloroglucinol, conversely to G. baccata which was able to produce phloroglucinol (traces), and the three types of poly-mers, fucols, phlorethols and fucophlorethols (Table 2). 3. Discussion Mar. Drugs 2021, 19, x FOR PEER REVIEW 10 of 20 Following the results obtained on 1H NMR spectra, a 2D 1H-13C NMR analysis of the purified fractions was carried out for four of the five species studied. Indeed, the low phe-nolic content in C. fœniculacea, associated with the chemical instability of the purified frac-tion, did not allow us to present a Heteronuclear Multiple Bond Correlation (HMBC) spec-trum for this species. The HMBC spectrum (Figure 4) of Gongolaria nodicaulis showed a multitude of signals on the 1H dimension, which correlated with methine-like carbons (C-H of the aromatic rings), phenol-like carbons (C-OH), rare carbons involved in aryl-aryl ring bonds (C-C) and a majority of carbons involved in diaryl-ether ring bonds (C-O-C). In G. baccata, a smaller number of correlation spots were observed (Figure 4). We can thus note, opposite the signal at 5.98 ppm in 1H dimension, a large correlation spot at 101 ppm in 13C dimen-sion. However, there is no correlation spot between 120 and 150 ppm, suggesting the pres-ence of a large linear polymer of the fucol type. Indeed, lateral branching would generate small variations in chemical shifts in the 1H dimension, and a less well-defined correlation spot would then be observed. By similar reasoning, we can detect a phlorotannin of the phlorethol type (δ = 5.92 ppm in 1H dimension), and a fucophlorethol (δ = 6.10 ppm). There was also a signal at δ = 5.78 ppm (isolated signal in 1H dimension), which correlated only with methines and phenolic carbons (Figure 4). This compound, which translated to a singlet in 1H dimension, present in the pure state in the purified fraction of E. selaginoides (Figure 4), was also the major compound in the purified fraction of C. humilis (Figure 4) and identified as phloroglucinol. In C. humilis, we also found a linear or slightly branched phlorethol-type compound (δ = 5.92 ppm in 1H dimension), as well as another compound at δ = 5.88 ppm (1H dimension), which showed neither an aryl-aryl bond nor a diaryl-ether bond (Figure 4). This compound, which “looked like” phloroglucinol (1H δ = 5.78 ppm), differed by a doubling of the C-OH correlation spot around 160–163 ppm in 13C dimen-sion. In E. selaginoides, the purified fraction showed only one major compound identified as phloroglucinol. In summary, Table 2 presents the type of phlorotannins that could be identified in the five species of concern. The upper species C. humilis produced phloroglucinol and phlorethols. The median species G. nodicaulis did not produce phloroglucinol but three types of phlorotannins: fucols, phlorethols and fucophlorethols. Table 2. Types of phlorotannins observed in our study in the five species of Cystoseira, Gongolaria and Ericaria. Examples of phlorotannin structures are also given. Mean total phenolic content (TPC) was calculated from spring 1 to summer 2. Mean TPC (% DW) Type of Phlorotannins (from 2D NMR Results) Phloroglucinol Fucol Phlorethol Fucophlorethol C. humilis 0.19 ± 0.09 Yes No Yes No C. fœniculacea 0.13 ± 0.06 / / / / G. nodicaulis 0.25 ± 0.16 No Yes Yes Yes G. baccata 0.21 ± 0.10 Traces Yes Yes Yes E. selaginoides 0.42 ± 0.21 Yes No No No Examples of chemical structures of phlorotannins Both lower species, G. baccata and E. selaginoides produced different types of phloro-tannins: E. selaginoides synthetized only the monomer phloroglucinol, conversely to G. baccata which was able to produce phloroglucinol (traces), and the three types of poly-mers, fucols, phlorethols and fucophlorethols (Table 2). 3. Discussion Mar. Drugs 2021, 19, x FOR PEER REVIEW 10 of 20 Following the results obtained on 1H NMR spectra, a 2D 1H-13C NMR analysis of the purified fractions was carried out for four of the five species studied. Indeed, the low phe-nolic content in C. fœniculacea, associated with the chemical instability of the purified frac-tion, did not allow us to present a Heteronuclear Multiple Bond Correlation (HMBC) spec-trum for this species. The HMBC spectrum (Figure 4) of Gongolaria nodicaulis showed a multitude of signals on the 1H dimension, which correlated with methine-like carbons (C-H of the aromatic rings), phenol-like carbons (C-OH), rare carbons involved in aryl-aryl ring bonds (C-C) and a majority of carbons involved in diaryl-ether ring bonds (C-O-C). In G. baccata, a smaller number of correlation spots were observed (Figure 4). We can thus note, opposite the signal at 5.98 ppm in 1H dimension, a large correlation spot at 101 ppm in 13C dimen-sion. However, there is no correlation spot between 120 and 150 ppm, suggesting the pres-ence of a large linear polymer of the fucol type. Indeed, lateral branching would generate small variations in chemical shifts in the 1H dimension, and a less well-defined correlation spot would then be observed. By similar reasoning, we can detect a phlorotannin of the phlorethol type (δ = 5.92 ppm in 1H dimension), and a fucophlorethol (δ = 6.10 ppm). There was also a signal at δ = 5.78 ppm (isolated signal in 1H dimension), which correlated only with methines and phenolic carbons (Figure 4). This compound, which translated to a singlet in 1H dimension, present in the pure state in the purified fraction of E. selaginoides (Figure 4), was also the major compound in the purified fraction of C. humilis (Figure 4) and identified as phloroglucinol. In C. humilis, we also found a linear or slightly branched phlorethol-type compound (δ = 5.92 ppm in 1H dimension), as well as another compound at δ = 5.88 ppm (1H dimension), which showed neither an aryl-aryl bond nor a diaryl-ether bond (Figure 4). This compound, which “looked like” phloroglucinol (1H δ = 5.78 ppm), differed by a doubling of the C-OH correlation spot around 160–163 ppm in 13C dimen-sion. In E. selaginoides, the purified fraction showed only one major compound identified as phloroglucinol. In summary, Table 2 presents the type of phlorotannins that could be identified in the five species of concern. The upper species C. humilis produced phloroglucinol and phlorethols. The median species G. nodicaulis did not produce phloroglucinol but three types of phlorotannins: fucols, phlorethols and fucophlorethols. Table 2. Types of phlorotannins observed in our study in the five species of Cystoseira, Gongolaria and Ericaria. Examples of phlorotannin structures are also given. Mean total phenolic content (TPC) was calculated from spring 1 to summer 2. Mean TPC (% DW) Type of Phlorotannins (from 2D NMR Results) Phloroglucinol Fucol Phlorethol Fucophlorethol C. humilis 0.19 ± 0.09 Yes No Yes No C. fœniculacea 0.13 ± 0.06 / / / / G. nodicaulis 0.25 ± 0.16 No Yes Yes Yes G. baccata 0.21 ± 0.10 Traces Yes Yes Yes E. selaginoides 0.42 ± 0.21 Yes No No No Examples of chemical structures of phlorotannins Both lower species, G. baccata and E. selaginoides produced different types of phloro-tannins: E. selaginoides synthetized only the monomer phloroglucinol, conversely to G. baccata which was able to produce phloroglucinol (traces), and the three types of poly-mers, fucols, phlorethols and fucophlorethols (Table 2). 3. Discussion Mar. Drugs 2021, 19, 504 11 of 20 high quantities of violaxanthin occurring in the last species were considered as a potential zeaxanthin pool that would be used as a way to dissipate energy in case of extreme light conditions. In this study, neither high content of violaxanthin nor high quantity of zeaxanthin could characterize the macroalgae settled in the upper shore. Apparently, there is no need for particular photoprotective mechanisms in these species living in the upper pools. Consequently, our study suggests that photo-adaptation exists in rockpools. Depend- ing on the tidal height, the species receive different amounts of light and regulate their pigment synthesis in order to maintain a good photosynthetic activity. In opposition, the xanthophyll cycle does not seem to be particularly active in the upper pools, indicating that light irradiance is not high enough to over-activate photosynthesis. The thin layer of water that persists in rockpools at low tide may represent an efﬁcient screen against excessive light irradiance. Further investigations should be undertaken particularly during the sunniest days of summer to state whether photoprotective mechanisms could be employed when the algae are submitted to exceptional light irradiance. Measuring the variations of the xanthophyll pigments in situ could also be a successful approach, as experienced by Gévaert et al. [29]. 3.2. What Are the Drivers of the Variability of Phenolic Content in the Five Species? To our knowledge, this study is the ﬁrst to characterize the seasonal variability of phlorotannins within the genera Cystoseira, Ericaria and Gongolaria from Brittany. In this study, these phenolic compounds were ﬁrst puriﬁed by a liquid–liquid ethyl acetate (EA) extraction before being quantiﬁed by the Folin-Ciocalteu method; this puriﬁcation step is commonly used to isolate phlorotannins [41,44,54,55] and more generally phenolic compounds from marine plants [56,57]. Interspeciﬁc variability. Our study demonstrates an interspeciﬁc variability in phe- nolic contents between the ﬁve species. Over the monitoring at Penmarc’h, we were able to determine that the levels of phenolic compounds were higher in E. selaginoides, G. nodi- caulis and G. baccata, while C. fœniculacea and C. humilis had lower levels (below 0.2% DW) (Figure 3). Among the different roles attributed to phlorotannins, a photoprotective role was considered by Pavia et al. [58,59]. These authors showed an induction of phlorotannin production in Ascophyllum nodosum following the exposure of thalli to high doses of UV-B light. Due to the numerous aromatic rings constituting these molecules, they strongly absorb certain UV radiation. Related to the distribution of each species along the intertidal rocky shore, C. hu- milis received the greatest amount of UV radiation because the shallow pools in which it settled are only submerged for a short time during a tidal cycle. However, its phenolic content was much lower than E. selaginoides, which was less exposed to strong radiation due to its low position on the foreshore. It is therefore impossible to establish a direct relationship between species distribution on the foreshore and phenolic content. However, phlorotannins are also known to be exuded into the surrounding environment, as already demonstrated [60,61]. Connan et al. [14] hypothesized this exudation to explain the lower content in the upper Pelvetia canaliculata compared to the high phenolic content determined in the median Ascophyllum nodosum. If the phlorotannins of C. humilis have an anti-UV action, perhaps they were rather exuded into the rockpool. The levels observed in vivo are probably dependent on other factors, like the diversity associated with each rockpool. The levels of phenolic compounds could also reﬂect the taxonomic group of the algal species. Indeed, the genus Ericaria (0.42% DW) produced more phenolic compounds than Gongo- laria (0.21–0.25% DW) followed by Cystoseira (0.13–0.19% DW). Our quantitative study of phlorotannins supports the idea that C. fœniculacea and C. humilis adopt a fundamentally different chemical defense strategy than G. baccata, G. nodicaulis and E. selaginoides. Seasonal variability. The phenolic contents in Cystoseira humilis did not show a seasonal pattern (Figure 3). This result is not surprising; indeed, the population of this species was in dynamic equilibrium, and perpetually renewed (many recruits constituted Mar. Drugs 2021, 19, 504 12 of 20 the density in summertime). No seasonality was observed, not in the size of the individuals, the density or the presence of mature individuals. The same result was observed for C. fœniculacea, whose phenolic content did not vary signiﬁcantly during the six studied seasons (Figure 3). On the contrary, this species developed seasonally, with individuals shedding their shoots in late spring, and developing progressively until late winter. Thus, phlorotannin levels in this species appeared to be independent of the phenological state of the individuals. Although we did not determine a phenological cycle in Gongolaria baccata, whose individuals were mature throughout the year in Penmarc’h (South Finistère), we noted lower levels of phenolic compounds in winter, and higher levels in summer (Figure 3). Our results are contradictory to those established by Le Lann et al. [19], who observed higher levels during winter and summer from samples taken in Porsmeur (North Finistère). The origin of this difference may be due to the puriﬁcation step that we included in our study, whereas Le Lann et al. [19] determined the polyphenols by the Folin–Ciocalteu method using crude extracts. Further studies are needed to determine the origin of the observed seasonal variability because here the relative stability of the population parameters con- trasts with the seasonality of the phenolic contents. An environmental factor that does not inﬂuence the size, density or maturity of the individuals could explain these seasonal variations in phlorotannin concentrations, such as epiphytism or herbivore pressure. Regarding G. nodicaulis, its phenolic content evolved in parallel with the phenology of the species: contents were minimal in spring/beginning of summer, when the branches started to fall naturally. It can thus be considered that a form of progressive senescence from spring onwards led to a decrease in concentrations, after which the branches were regenerated from the autumn onwards, with an increase in phenolic contents. Finally, in winter, some twigs began to detach, and levels decreased, heralding the resting period and lower levels observed in early summer (see results in Figure 3). Ericaria selaginoides is characterized by even clearer seasonal variations (Figure 3). Indeed, the end of summer can be contrasted with the other times of the year: the phenolic levels were particularly low (0.1% DW) in autumn, in contrast to spring or summer when they exceeded 0.4% DW. Such as G. nodicaulis, twig loss in E. selaginoides was initiated during the summer, and was maximal at the end of summer (Table 1). Thus, the summer senescence of shoots is a plausible explanation for these particularly low phenolic levels. Our results are not inconsistent with the observations of Abdala-Díaz et al. [62] who showed a progressive increase in levels between spring/summer, followed by a decrease from summer to winter in Spain. The authors also showed that the levels of phenolic compounds are correlated with the level of light radiation, and they deduced a photoprotective role of these compounds. In our study, we observed a more drastic decrease in levels in summer, which was probably related to the senescence that occurred in this period. Unfortunately, Abdala-Díaz et al. [62] did not specify the phenology of E. selaginoides in Spain. This data would be particularly interesting to be able to compare the strategy of this species between French and Spanish populations. Each species was characterized by its own seasonal dynamics of phlorotannin levels. In the case of E. selaginoides and G. nodicaulis, phlorotannin levels seemed to evolve in parallel with the phenology of the alga. These compounds were accumulated in the tissues before the maturity period, probably to protect the receptacles from herbivore grazing. Further investigations, extending the monitoring over several years and also measuring the exuded phenolic compounds in the rockpools, could shed some light on the ecological role of phlorotannins. 3.3. Do the Five Species Produced the Same Phlorotannins? Our study demonstrates that the ﬁve species did not produce similar phlorotan- nins. Ericaria selaginoides produced the monomer phloroglucinol and not more complex phlorotannins, Cystoseira humilis produced phloroglucinol and phlorethol, Gongolaria nod- icaulis produced three types of phlorotannins, fucol, phlorethol and fucophlorethol and Mar. Drugs 2021, 19, 504 13 of 20 ﬁnally, G. baccata was able to produce the three types of phlorotannins and the monomer (Table 2). The types of phlorotannins identiﬁed in both species of Gongolaria were essentially the same: fucols as well as phloroethols. The absence of signals at 145–150 ppm indicates that the extracts did not contain a fuhalol-type polymer [63]. This seems surprising because fuhalols have been characterized by the past in G. baccata [64] and in G. nodicaulis [65]. At the same time, small amounts of phloroglucinol were found in G. baccata and C. humilis. In E. selaginoides, large amounts of phloroglucinol were observed, but no traces of bifuhalol and diphlorethol, already identiﬁed in this species by Glombitza et al. [66]. According to the NMR data of diphlorethol and bifuhalol (also described in Bifurcaria bifurcata by Glombitza and Rösener [67]), the NMR signals of the aromatic protons of these compounds would be one doublet and one triplet, which are not visible on the 1H NMR spectrum of E. selaginoides (only a singulet; Figure 4). According to the results of our study, phlorotannin types are not a relevant criterion from a chemotaxonomic point of view. Indeed, fucols and phloroethols are very common molecules among the species considered. However, phloroglucinol can be used as a chemotaxonomic marker for E. selaginoides because it is the only species that exclusively produces the monomer phloroglucinol. By HR-MAS NMR analyses, Jégou et al. [68] highlighted that E. selaginoides produced phloroglucinol in spring and summer until the beginning of autumn and then the species entered a dormancy-like phase during autumn and winter in Brittany. Using an innovative qNMR technique, authors were able to speciﬁcally quantify the intra-individual and seasonal variations of phloroglucinol [68]. Further experiments on E. selaginoides under controlled conditions should be carried out in order to determine which parameters (temperature, salinity, light radiation, nutrients, and/or herbivore pressure) induce the synthesis of phenolic compounds. The diversity of phlorotannins within Sargassaceae species was already demonstrated by Stiger-Pouvreau et al. [41] on the eight species encountered in Brittany, including the ﬁve species studied here, together with Bifurcaria bifurcata, Halidrys siliquosa and Sargassum muticum. Additionally, Ferreres et al. [69], reviewed by de Sousa et al. [70], highlighted that G. nodicaulis and E. selaginoides, collected in West Portugal, produced also differ- ent phlorotannins belonging to eckol and fucophloroethol types, not observed in this present study. 3.4. Do Chemical Characteristics Could Help to Understand the Distribution of the Five Species along Brittany Rockyshores? The distribution of Sargassaceae populations in rockpools results from a balance between (1) the capacity of the species to survive to a variable environment—the most changing zones being in the upper shore [71]—and the species needing to be tolerant regarding temperature and salinity and (2) their capacity to settle on a substrate yet colonized by competitors. Cystoseira humilis must be particularly resistant to changing environments so it can colonize the upper shore. In addition, only few species have such an ability, as suggested by the low species richness of its pools (equal to 7, see Figure 1), and by the low proportion of substrate covered by living organisms (Jégou, pers. obs.). Thus, the competition is low, and the recruits of C. humilis have much space to settle, which permits high densities within populations. Small length of individuals could be regarded as an adaptation of this species to shallow pools, or to pools in which depth can become very low due to evaporation in summer [13]. At the very opposite, E. selaginoides cannot face such intense environmental changes, and thus it cannot develop on mid and upper pools. However, it can grow on lower tide pools, but because the competition for the substrate is strong (high species richness up to 35, see Figure 1, Table 1), only low densities occur on the ﬁeld. The high depth occurring more frequently in lower rock pools (in our survey, from 30 cm to more than 1 m) let the individuals grow taller than anywhere else. Gongolaria baccata turns out to follow this tendency; however, it has greater ecological amplitude, allowing its presence in mid pools. Gongolaria nodicaulis and C. fœniculacea can be regarded as intermediate species. Living in mid tidepools implicates a slightly higher variability of the water parameters than in lower pools, and a lower intensity of competition. Because Mar. Drugs 2021, 19, 504 14 of 20 these mid pools are frequently less deep than in the lower intertidal, this would lead to the occurrence of medium-sized individuals with an intermediate population density. Throughout the survey, seasonality was, at ﬁrst sight, not a typical feature within the ﬁve species. Only three of them, C. fœniculacea, G. nodicaulis and E. selaginoides showed unambiguous seasonal patterns, each species being characterized by a cycle in three phases: (1) active growth via the production of primary, secondary, tertiary ( . . . ) axes leading in (2) the development of mature receptacles, and after the release of the gametes, a loss of secondary axes that precedes (3) a dormancy period. This typical growth process has been known since Sauvageau [72] and Roberts [73–75] described it fairly. However, a time shift in these phases was observed between species. For example, C. fœniculacea and C. nodicaulis began dormancy in the middle of summer, while E. selaginoides was in its full maturity period (Table 1). Such an evident phenology could not be retrieved for G. baccata and C. humilis (Table 1). Cystoseira humilis was mature between spring and summer, but maturity can be observed all year long, depending on the dynamism of the population. Moreover, in Portugal, the reproduction period seems to be summer [76]. Reproduction period(s) must be under the inﬂuence of environmental parameters. Regarding G. baccata, it was not possible to determine seasonal patterns. Contrary to Le Lann et al. [19,77], we cannot deduce a unique reproductive period (Table 1) because all observed thalli were mature, all along the survey. However, the state of the algae differed throughout the seasons, even if it cannot be deduced from the results. It was obvious that during some periods, notably around summer, the number of receptacles was drastically reduced (Jégou, pers. obs.). A supplementary study on the evolution of the number of receptacles would probably highlight a seasonal pattern in the reproduction of G. baccata. 4. Materials and Methods 4.1. Native Canopy-Forming Sargassaceae Species and Sampled Rockpools The correct identiﬁcation of the specimens was here ensured using characteristic morphological and chemical features (see [78] for details about the criteria). In Brittany, the sites where the ﬁve species can be observed altogether are very rare because they have to ﬁt some conditions, such as a sheltered site and an extended rocky intertidal zone. We chose to set up our experiment at Saint-Pierre (Penmarc’h, Brittany, France; 47◦48(cid:48)03(cid:48)(cid:48) N; 4◦22(cid:48)39.0(cid:48)(cid:48) W), where the slope of the shore is so low that the intertidal zone goes beyond 400 m in direction to the sea. The location of the ﬁve species was checked on the ﬁeld using a Magellan™ Triton 200 GPS system, with an approximate 1 m precision. These tracks were deployed with a 60 m Stanley™ long tape rule. The presence of the Sargassaceae species was investigated using a 0.5 m × 0.5 m quadrat. For example, a 40 m track consisted of the observation of 80 aligned quadrats. For each species, the presence of individuals and their precise location on the intertidal zone were noticed. For statistical analyses purposes, we converted the location of all the individuals (latitude and longitude) into a one-dimensional position on the sea shore. We also evaluated the macroalgal species richness along the transect. This operation was repeated ﬁve times, from spring 2010 to spring 2011 (regarded as spring 2 to spring 3, respectively), to ensure the validity of the results, as suggested by Underwood [79]. Seasonal data were compiled to give the distribution of each species on Figure 1. 4.2. Extraction and Analysis of Pigments Sampling. The ﬁve species were sampled on May 2011 and regarded as spring 3. We sampled three individuals without epiphytes per species and along the distribution of each species along the shore in order to take in consideration the large or the narrow distribution of species along the shore. The different pools were selected on the shore on the basis of similar characteristics, especially regarding depth (approximately 20 cm). Just after sampling, the thalli were thoroughly rinsed with deionized water, freeze-dried and stored in the dark at −20 ◦C until extraction. Mar. Drugs 2021, 19, 504 15 of 20 Extraction. Algal samples (apical parts) were ground using liquid nitrogen with a mortar and a pestle. Then, 50 mg of the powdered alga was subsequently extracted twice with 1 mL of a mixture of acetone/water (90:10, v/v), at 4 ◦C and under agitation. The ﬁrst extraction lasted 30 min, the second 12 h. The two extracts were then pooled, centrifuged at 5000 rpm during 5 min. The resulting supernatants were combined and then ﬁltered using a 0.45 µm Nylon membrane (Millipore, Guyancourt, France). Prior to injection in the HPLC system, 150 µL of the ﬁltered extract were mixed with 50 µL of buffer (ammonium acetate aqueous solution at 0.5 M, pH 7.2). HPLC analysis. These samples were analyzed using an HPLC system using UV- Visible detection. For this purpose, a Waters HPLC system (Waters, Guyancourt, France) equipped with a Waters 717 Plus autosampler, a Water 600 Controller pump and a Photodi- ode Array Detector was used. The wavelength range of the detector was set at 210–700 nm. The pigments were separated following the method described by Wright et al. [80] modiﬁed by Bidigare et al. [81]. Each analysis included the injection of 10 µL of a sample. Pigments were separated using a Zorbax Eclipse XDB-C18 column (4.6×150 mm; 5 µm; Agilent Technologies, Les Ulis, France) equipped with a C18 guard cartridge (SecurityGuard, Phe- nomenex, Le Pecq, France) and maintained at 40 ◦C. A ternary mobile phase was used: eluent A was constituted by 80% methanol, 20% ammonium acetate buffer at pH 7.2, and BHT at 0.1 g/L; eluent B was constituted by 87.5% acetonitrile, 12.5% water, and BHT at 0.1 g/L; eluent C was pure ethyl acetate. The gradient of elution is indicated in Table S3 (Supplementary Material). Chlorophyll a, chlorophyll c2, fucoxanthin, violaxanthin, zeax- anthin and β-carotene (DHI, Hørsholm, Denmark) were used as standards. 4.3. Extraction of Phlorotannins and Determination of Phenolic Content Samples were collected from spring 2009 (=spring 1) to summer 2010 (=summer 2) in order to follow the variability of phenolic content during six seasons. Extraction. For each sample (3 individuals per species), phenolic compounds were extracted twice, successively from 200 mg of powdered algae with a methanol/water (1:1) mixture during 2 h, at 40 ◦C in the dark. The two extracts were pooled and semi-puriﬁed; methanol was evaporated and the volume of the resulting crude extract was set to 10 mL of aqueous solution. Then 5 mL of ethyl acetate was added and the resulting 15 mL was mixed and centrifuged (5000 rpm, 4 ◦C). The organic phase was isolated while the remaining aqueous phase was re-extracted once with 5 mL ethyl acetate. The two organic phases were combined and the solvent was removed using a rotary evaporator. The organic molecules were dissolved in 10 mL of water (containing less than 1% ethanol for a better solubility). Folin–Ciocalteu Assay. The phenolic content of the semi-puriﬁed fraction was evalu- ated using the Folin–Ciocalteu assay slightly adapted from previous studies [17,82] and modiﬁed by Le Lann et al. [83]. Brieﬂy, 100 mL of semi-puriﬁed fraction was mixed with 50 mL of Folin–Ciocalteu reagent, 200 mL of Na2CO3 (15%) and 650 mL of distilled water. This mixture was heated during 20 min at 70 ◦C and put on ice for 10 min to stop the reaction. The absorbance was measured at 750 nm. Standard phloroglucinol solutions were also submitted to this assay to obtain a calibration curve. The results were expressed as eq. phloroglucinol content in % DW. 4.4. NMR Analyses of Phlorotannins Puriﬁcation steps. Different puriﬁcation steps of the crude extract were followed: three washings using dichloromethane (DCM), a precipitation step using ethanol following by another precipitation step using acetone, and a liquid–liquid water:ethyl acetate (EA) extraction was performed with the phenolic compounds concentrated in the EA phase, as described by several authors [41,44,55,56]. The last EA phase, containing the phenolic compounds, was recovered for further analyses. Fractionation on a silica column. A 50 g silica column (Normal Phase Silica, 0.63– 0.200 mm, Merck, Fontenay sous Bois, France) was packed in ethyl acetate. The sample was deposited at the top of the column, and three fractions were recovered by elution with (1) Mar. Drugs 2021, 19, 504 16 of 20 ethyl acetate, (2) ethanol and (3) methanol. The ethyl acetate fraction was further analyzed using 2D NMR analysis in the aim to determine the structure of phlorotannins. NMR analysis. The various crude extracts and fractions were monitored by 1H 1D NMR analysis on a BRUKER Avance 400 MHz spectrometer (Bruker, Wissembourg, France) with a tunable triple resonance multi-nucleus probe (acquisition performed at 25 ◦C). For each species, the most interesting EA fraction was characterized in 2D 1H-13C NMR by heteronuclear correlation sequence on multiple bonds (HMBC) on a BRUKER Avance 500 Mhz spectrometer equipped with a cryoprobe. The HMBC sequence allows the determination of correlations between 1H and 13C at more than 2 or 3 bonds. The interest here was to show the proximity between aromatic protons compatible with phlorotannins (5.7 ppm ≤ δ 1H ≤ 6.5 ppm) with the carbons of the phenolic rings, and especially those involved in the linkage between phloroglucinol units, which allowed the determination of the major types of phlorotannins in each species [63]. 4.5. Statistical Analyses All analyses were performed using the R statistical software [84], with a type I error level α = 0.05. Considering the study of the distribution across the shore, we evaluated the differences between the positions of the species. When homogeneity of variances was rejected (Fligner-Killeen’s test), even after the use of common transformations of data, as proposed by Underwood [79], we used a Kruskal–Wallis test, and an associated multiple comparison test using the package “pgirmess” [85]. We investigated the effects of the factors “species” and “tidal height” on pigment and phenolic contents, using two-way ANOVA, and differences were highlighted by Tukey’s HSD test. For this purpose, we checked the required assumptions of normality (using Shapiro–Wilk test) and homoscedasticity (Levene’s test). Correlations between variables were assessed using Pearson’s method. 5. Conclusions On the intertidal zone of the rocky shores of Brittany, Cystoseira, Ericaria and Gongolaria species can be under the control of analogue parameters to the ones determining the macroalgal assemblages on the emerged substrata. Considering phlorotannins, our study highlights differences in content between species, without being possible to directly link phlorotannin content and distribution on the foreshore. Moreover, each species was characterized by its own seasonal dynamics of phlorotannin levels. Further investigations, extending the monitoring over several years and also measuring the exuded phenolic compounds in the rockpools could shed some light on the ecological role of phlorotannins. From a qualitative point of view, numerous polymers make up the mixture of phlorotannins in all species, except E. selaginoides which produces the monomer phloroglucinol. Thus, this species is a very good candidate for ecophysiological or functional genomics approaches. Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/md19090504/s1, Table S1: Results of the two-way ANOVA (pigment levels as a function of the species and tidal height of settlement), Df: degree of freedom, Signiﬁcant results are highlighted., Table S2: Correlation between pigment levels according to Pearson’s test; all indicated values are signiﬁcant (p < 0.001), n.s.: no signiﬁcant correlation (β-car: β-carotene; Chl: chlorophyll; Fuco: fucoxanthin; Viola: violaxanthin and Zea: zeaxanthin)., Table S3: Solvent gradient used during HPLC analysis of pigments in Cystoseira, Ericaria and Gongolaria species. Author Contributions: C.J. and I.B. performed the extraction of phlorotannins. S.C. (Solène Connan) supervised the extraction and analysis of pigments. S.C. (Stéphane Cérantola) supervised the NMR analysis and did the 2D NMR analysis. C.J. did the analysis of the samples, drafted the manuscript and conducted the statistical analyses. F.G. supervised the HPLC analysis. V.S.-P. designed and supervised the project. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Ministère de l’Education Nationale, de l’Enseignement Supérieur et de la Recherche and supported by the Ecole Doctorale des Sciences de la Mer (EDSM) from the Université de Bretagne Occidentale (UBO), and by the Interreg IVB project Biotecmar. Mar. Drugs 2021, 19, 504 17 of 20 Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Acknowledgments: This study is part of the Ph.D. thesis work carried out by the ﬁrst author within the Laboratory of Ecophysiology and Biotechnologies of Halophytes and Marine Algae (LEBHAM EA 3877) set at the IUEM (UBO-UEB). Gratitude is due to Jean-Jacques Jégou and Ludovic Jean for assistance during ﬁeld measurements. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 2. 1. Underwood, A.J. Structure of a rocky intertidal community in New South Wales: Patterns of vertical distribution and seasonal changes. J. Exp. Mar. Biol. Ecol. 1981, 51, 57–85. [CrossRef] Underwood, A.J.; Skilleter, G.A. Effects of patch-size on the structure of assemblages in rock pools. J. Exp. Mar. Biol. Ecol. 1996, 197, 63–90. [CrossRef] Daniel, M.J.; Boyden, C.R. Diurnal variations in physico-chemical conditions within intertidal rockpools. Field Stud. 1975, 4, 161–176. 3. 4. Morris, S.; Taylor, A.C. Diurnal and seasonal variation in physico-chemical conditions within intertidal rock pools. Estuar. Coast. Shelf Sci. 1983, 17, 339–355. [CrossRef] 5. Huggett, J.; Grifﬁths, C.L. Some relationships between elevation, physico-chemical variables and biota of intertidal rock pools. Mar. Ecol. Prog. Ser. 1986, 29, 189–197. [CrossRef] 7. 6. Noël, L.M.-L.J.; Grifﬁn, J.N.; Thompson, R.C.; Hawkins, S.J.; Burrows, M.T.; Crowe, T.P.; Jenkins, S.R. Assessment of a ﬁeld incubation method estimating primary productivity in rockpool communities. Estuar. Coast. Shelf Sci. 2010, 88, 153–159. [CrossRef] Legrand, E.; Riera, P.; Pouliquen, L.; Bohner, O.; Cariou, T.; Martin, S. Ecological characterization of intertidal rockpools: Seasonal and diurnal monitoring of physico-chemical parameters. Reg. Stud. Mar. Sci. 2018, 17, 1–10. [CrossRef] Kooistra, W.H.C.F.; Joosten, A.M.T.; van den Hoek, C. Zonation patterns in intertidal pools and their possible causes: A multivariate approach. Bot. Mar. 1989, 32, 9–26. [CrossRef] Araújo, R.; Sousa-Pinto, I.; Bárbara, I.; Quintino, V. Macroalgal communities of intertidal rock pools in the northwest coast of Portugal. Acta Oecol. 2006, 30, 192–202. [CrossRef] 9. 8. 10. Wallenstein, F.M.; Peres, S.D.; Xavier, E.D.; Neto, A.I. Phytobenthic communities of intertidal rock pools in the eastern islands of Azores and their relation to position on shore and pool morphology. Arquipélago—Life Mar. Sci. 2010, 27, 9–20. 11. Dethier, M.N. Disturbance and recovery in intertidal pools: Maintenance of mosaic patterns. Ecol. Monogr. 1984, 54, 99–118. [CrossRef] 12. Astles, K.L. Patterns of abundance and distribution of species in intertidal rock pools. J. Mar. Biol. Assoc. U. K. 1993, 73, 555–569. [CrossRef] 13. Cabioc’h, J.; Floc’h, J.Y.; Le Toquin, A.; Boudouresque, C.F.; Meinesz, A.; Verlaque, M. Guide des Algues des Mers d’Europe; Delachaux et Niestlé: Paris, France, 2006. 14. Connan, S.; Goulard, F.; Stiger, V.; Deslandes, E.; Ar Gall, E. Interspeciﬁc and temporal variation in phlorotannin levels in an assemblage of brown algae. Bot. Mar. 2004, 47, 410–416. [CrossRef] 15. Dizerbo, A.; Herpe, E. Liste et Répartition des Algues Marines des Côtes Françaises de la Manche et de L’atlantique, Iles Anglo-Normandes Incluses; Éditions Scientiﬁques Anaximandre: Landerneau, France, 2007. 16. Legrand, E.; Riera, P.; Bohner, O.; Coudret, J.; Schlicklin, F.; Derrien, M.; Martin, S. Impact of ocean acidiﬁcation and warming on the productivity of a rock pool community. Mar. Environ. Res. 2018, 136, 78–88. [CrossRef] [PubMed] 17. Plouguerné, E.; Le Lann, K.; Connan, S.; Jechoux, G.; Deslandes, E.; Stiger-Pouvreau, V. Spatial and seasonal variation in density, reproductive status, length and phenolic content of the invasive brown macroalga Sargassum muticum (Yendo) Fensholt along the coast of Western Brittany (France). Aquat. Bot. 2006, 85, 337–344. [CrossRef] 18. Billard, E.; Serrão, E.; Pearson, G.; Destombe, C.; Valero, M. Fucus vesiculosus and spiralis species complex: A nested model of local adaptation at the shore level. Mar. Ecol. Prog. Ser. 2010, 405, 163–174. [CrossRef] 19. Le Lann, K.; Connan, S.; Stiger-Pouvreau, V. Phenology, TPC and size-fractioning phenolics variability in temperate Sargassaceae (Phaeophyceae, Fucales) from Western Brittany: Native versus introduced species. Mar. Environ. Res. 2012, 80, 1–11. [CrossRef] [PubMed] Jégou, C. Étude du Genre Cystoseira des Côtes Bretonnes: Taxinomie, Ecologie et Caractérisation de Substances Naturelles. Doctoral Dissertation, Université de Bretagne Occidentale, Brest, France, 2011. 20. 21. Pardi, G.; Piazzi, L.; Cinelli, F. Demographic study of a Cystoseira humilis Kützing (Fucales: Cystoseiraceae) population in the 22. Western Mediterranean. Bot. Mar. 2000, 43, 81–86. [CrossRef] Jódar-Pérez, A.B.; Terradas-Fernández, M.; López-Moya, F.; Asensio-Berbegal, L.; López-Llorca, L.V. Multidisciplinary Analysis of Cystoseira sensu lato (SE Spain) Suggest a Complex Colonization of the Mediterranean. J. Mar. Sci. Eng. 2020, 8, 961. [CrossRef] Mar. Drugs 2021, 19, 504 18 of 20 23. Karsten, U. Defense Strategies of Algae and Cyanobacteria against Solar UVR. In Algal Chemical Ecology; Amsler, C.D., Ed.; Springer: Berlin/Heidelberg, Germany, 2008; pp. 273–296. 24. Goss, R.; Jakob, T. Regulation and function of xanthophyll cycle-dependent photoprotection in algae. Photosynth. Res. 2010, 106, 103–122. [CrossRef] 25. Demmig-Adams, B.; Adams, W.W., III. The role of xanthophyll cycle carotenoids in the protection of photosynthesis. Trends Plant Sci. 1996, 1, 21–26. [CrossRef] 26. Heo, S.-J.; Jeon, Y.-J. Protective effect of fucoxanthin isolated from Sargassum siliquastrum on UV-B induced cell damage. J. Photochem. Photobiol. B 2009, 95, 101–107. [CrossRef] 27. Burton, G.; Ingold, K. beta-Carotene: An unusual type of lipid antioxidant. Science 1984, 224, 569–573. [CrossRef] [PubMed] 28. Gévaert, F.; Créach, A.; Davoult, D.; Holl, A.-C.; Seuront, L.; Lemoine, Y. Photo-inhibition and seasonal photosynthetic perfor- mance of the seaweed Laminaria saccharina during a simulated tidal cycle: Chlorophyll ﬂuorescence measurements and pigment analysis. Plant Cell Environ. 2002, 25, 859–872. [CrossRef] 29. Gévaert, F.; Créach, A.; Davoult, D.; Migné, A.; Levavasseur, G.; Arzel, P.; Holl, A.-C.; Lemoine, Y. Laminaria saccharina photosynthesis measured in situ: Photoinhibition and xanthophyll cycle during a tidal cycle. Mar. Ecol. Prog. Ser. 2003, 247, 43–50. [CrossRef] 30. Harker, M.; Berkaloff, C.; Lemoine, Y.; Britton, G.; Young, A.J.; Duval, J.-C.; Rmiki, N.-E.; Rousseau, B. Effects of high light and desiccation on the operation of the xanthophyll cycle in two marine brown algae. Eur. J. Phycol. 1999, 34, 35–42. [CrossRef] 31. Uhrmacher, S.; Hanelt, D.; Nultsch, W. Zeaxanthin content and the degree of photoinhibition are linearly correlated in the brown alga Dictyota dichotoma. Mar. Biol. 1995, 123, 159–165. [CrossRef] 32. Amsler, C.D.; Fairhead, V.A. Defensive and Sensory Chemical Ecology of Brown Algae. Adv. Bot. Res. 2006, 43, 1–91. 33. 34. 35. Arnold, T.M.; Targett, N.M. Marine tannins: The importance of a mechanistic framework for predicting ecological roles. J. Chem. Schoenwaelder, M.E. The occurrence and cellular signiﬁcance of physodes in brown algae. Phycologia 2002, 41, 125–139. [CrossRef] Schoenwaelder, M.E. The biology of phenolic containing vesicles. Algae 2008, 23, 163–175. [CrossRef] Ecol. 2002, 28, 1919–1934. [CrossRef] [PubMed] 36. Meslet-Cladière, L.; Delage, L.; Leroux, C.J.J.; Goulitquer, S.; Leblanc, C.; Creis, E.; Ar Gall, E.; Stiger-Pouvreau, V.; Czjzek, M.; Potin, P. Structure/function analysis of a type III polyketide synthase in the brown alga Ectocarpus siliculosus reveals a biochemical pathway in phlorotannin monomer biosynthesis. Plant Cell 2013, 25, 3089–3103. [CrossRef] 37. Arnold, T.M.; Targett, N.M. Quantifying in situ rates of phlorotannin synthesis and polymerization in marine brown algae. J. Chem. Ecol. 1998, 24, 577–595. [CrossRef] 38. Ragan, M.; Glombitza, K.-W. Phlorotannins, brown algal polyphenols. Prog. Phycol. Res. 1986, 4, 129–241. 39. Li, S.M.; Glombitza, K.W. Carmalols and phlorethofuhalols from the brown alga Carpophyllum maschalocarpum. Phytochemistry 1991, 30, 3417–3421. [CrossRef] 40. Li, Y.; Qian, Z.-J.; Ryu, B.; Lee, S.-H.; Kim, M.-M.; Kim, S.-K. Chemical components and its antioxidant properties in vitro: An 41. edible marine brown alga, Ecklonia cava. Bioorgan. Med. Chem. 2009, 17, 1963–1973. [CrossRef] Stiger-Pouvreau, V.; Jegou, C.; Cerantola, S.; Guérard, F.; Le Lann, K. Phlorotannins in Sargassaceae species from Brittany (France): Interesting molecules for ecophysiological and valorisation purposes. Adv. Bot. Res. 2014, 71, 379–411. 42. Koch, M.; Gregson, R.P. Brominated phlorethols and nonhalogenated phlorotannins from the brown alga Cystophora congesta. Phytochemistry 1984, 23, 2633–2637. [CrossRef] 43. Glombitza, K.W.; Knöss, W. Sulphated phlorotannins from the brown alga Pleurophycus gardneri. Phytochemistry 1992, 31, 279–281. [CrossRef] 44. Le Lann, K.; Surget, G.; Couteau, C.; Coiffard, L.; Cérantola, S.; Gaillard, F.; Larnicol, M.; Zubia, M.; Guérard, F.; Poupart, N.; et al. Sunscreen, antioxidant, and bactericide capacities of phlorotannins from the brown macroalga Halidrys siliquosa. J. Appl. Phycol. 2016, 28, 3547–3559. [CrossRef] 45. Lalegerie, F.; Gager, L.; Stiger-Pouvreau, V.; Connan, S. The stressful life of red and brown seaweeds on the temperate intertidal zone: Effect of abiotic and biotic parameters on the physiology of macroalgae and content variability of particular metabolites. Adv. Bot. Res. 2020, 95, 247–287. 46. Gager, L.; Lalegerie, F.; Connan, S.; Stiger-Pouvreau, V. Marine Algal Derived Phenolic Compounds and their Biological Activities for Medicinal and Cosmetic Applications. In Recent Advances in Micro and Macroalgal Processing: Food and Health Perspectives; Rajauria, G., Yuan, Y.V., Eds.; Wiley-Blackwell: Hoboken, NJ, USA, 2021; pp. 278–334. 47. Alberte, R.S.; Friedman, A.L.; Gustafson, D.L.; Rudnick, M.S.; Lyman, H. Light-harvesting systems of brown algae and diatoms. Isolation and characterization of chlorophyll a/c and chlorophyll a/fucoxanthin pigment-protein complexes. BBA-Bioenergetics 1981, 635, 304–316. [CrossRef] 48. De Martino, A.; Douady, D.; Rousseau, B.; Duval, J.-C.; Caron, L. Characterization of two light-harvesting subunits isolated from the brown alga Pelvetia canaliculata: Heterogeneity of xanthophyll distribution. Photochem. Photobiol. 1997, 66, 190–197. [CrossRef] 49. De Martino, A.; Douady, D.; Quinet-Szely, M.; Rousseau, B.; Crépineau, F.; Apt, K.; Caron, L. The light-harvesting antenna of 50. brown algae. Eur. J. Biochem. 2000, 267, 5540–5549. [CrossRef] Stengel, D.; Dring, M. Seasonal variation in the pigment content and photosynthesis of different thallus regions of Ascophyllum nodosum (Fucales, Phaeophyta) in relation to position in the canopy. Phycologia 1998, 37, 259–268. [CrossRef] Mar. Drugs 2021, 19, 504 19 of 20 51. Aguilera, J.; Bischof, K.; Karsten, U.; Hanelt, D.; Wiencke, C. Seasonal variation in ecophysiological patterns in macroalgae from an Arctic fjord. II. Pigment accumulation and biochemical defence systems against high light stress. Mar. Biol. 2002, 140, 1087–1095. 52. Ramus, J.; Lemons, F.; Zimmerman, C. Adaptation of light-harvesting pigments to downwelling light and the consequent photosynthetic performance of the eulittoral rockweeds Ascophyllum nodosum and Fucus vesiculosus. Mar. Biol. 1977, 42, 293–303. [CrossRef] 53. Döhler, G.; Hagmeier, E.; David, C. Effects of solar and artiﬁcial UV irradiation on pigments and assimilation of 15N ammonium and 15N nitrate by macroalgae. J. Photochem. Photobiol. B 1995, 30, 179–187. [CrossRef] 54. Gager, L.; Connan, S.; Molla, M.; Couteau, C.; Arbona, J.-F.; Coiffard, L.; Cérantola, S.; Stiger-Pouvreau, V. Temporal variation of active phlorotannins determined by 1H NMR and in vitro assays. J. Appl. Phycol. 2020, 32, 2375–2386. [CrossRef] 55. Ar Gall, E.; Lelchat, F.; Hupel, M.; Jégou, C.; Stiger-Pouvreau, V. Extraction and puriﬁcation of phlorotannins from brown algae. 56. 57. In Natural Products from Marine Algae; Stengel, D.B., Connan, S., Eds.; Humana Press: New York, NY, USA, 2015; pp. 131–143. Surget, G.; Stiger-Pouvreau, V.; Le Lann, K.; Kervarec, N.; Couteau, C.; Coiffard, L.J.; Gaillard, F.; Cahier, K.; Guérard, F.; Poupart, N. Structural elucidation, in vitro antioxidant and photoprotective capacities of a puriﬁed polyphenolic-enriched fraction from a saltmarsh plant. J. Photochem. Photobiol. B 2015, 143, 52–60. [CrossRef] Surget, G.; Roberto, V.P.; Le Lann, K.; Mira, S.; Guérard, F.; Laizé, V.; Poupart, N.; Cancela, M.L.; Stiger-Pouvreau, V. Marine green macroalgae: A source of natural compounds with mineralogenic and antioxidant activities. J. Appl. Phycol. 2017, 29, 575–584. [CrossRef] 58. Pavia, H.; Brock, E. Extrinsic factors inﬂuencing phlorotannin production in the brown alga Ascophyllum nodosum. Mar. Ecol. Prog. Ser. 2000, 193, 285–294. [CrossRef] 59. Pavia, H.; Cervin, G.; Lindgren, A.; Åberg, P. Effects of UV-B radiation and simulated herbivory on phlorotannins in the brown 60. 61. alga Ascophyllum nodosum. Mar. Ecol. Prog. Ser. 1997, 157, 139–146. [CrossRef] Swanson, A.K.; Druehl, L.D. Induction, exudation and the UV protective role of kelp phlorotannins. Aquat. Bot. 2002, 73, 241–253. [CrossRef] Jennings, J.G.; Steinberg, P.D. In situ exudation of phlorotannins by the sublittoral kelp Ecklonia radiata. Mar. Biol. 1994, 121, 349–354. [CrossRef] 62. Abdala-Díaz, R.T.; Cabello-Pasini, A.; Pérez-Rodríguez, E.; Álvarez, R.C.; Figueroa, F.L. Daily and seasonal variations of optimum quantum yield and phenolic compounds in Cystoseira tamariscifolia (Phaeophyta). Mar. Biol. 2006, 148, 459–465. [CrossRef] 63. Cerantola, S.; Breton, F.; Ar Gall, E.; Deslandes, E. Co-occurrence and antioxidant activities of fucol and fucophlorethol classes of polymeric phenols in Fucus spiralis. Bot. Mar. 2006, 49, 347–351. [CrossRef] 64. Glombitza, K.W.; Schnabel, C.; Koch, M. Antibiotica aus Algen, 27. Mitt. Niedermolekulare Phlorotannine der Braunalge Cystoseira baccata (Gmelin) Silva, Teil II. Arch. Pharm. 1981, 314, 602–608. [CrossRef] 65. Glombitza, K.W.; Wegner-Hambloch, S.; Schulten, H.R. Antibiotics from algae, XXXVI. 1, 2 Phlorotannins from the brown alga Cystoseira granulata. Planta Med. 1985, 51, 116–120. [CrossRef] 66. Glombitza, K.W.; Rösener, H.U.; Müller, D. Bifuhalol und diphlorethol aus Cystoseira tamariscifolia. Phytochemistry 1975, 14, 1115–1116. [CrossRef] 67. Glombitza, K.W.; Rösener, H.U. Bifuhalol: Ein Diphenyläther aus Bifurcaria bifurcata. Phytochemistry 1974, 13, 1245–1247. 68. 69. [CrossRef] Jégou, C.; Kervarec, N.; Cérantola, S.; Bihannic, I.; Stiger-Pouvreau, V. NMR use to quantify phlorotannins: The case of Cystoseira tamariscifolia, a phloroglucinol-producing brown macroalga in Brittany (France). Talanta 2015, 135, 1–6. [CrossRef] Ferreres, F.; Lopes, G.; Gil-Izquierdo, A.; Andrade, P.B.; Sousa, C.; Mouga, T.; Valentão, P. Phlorotannin extracts from Fucales characterized by HPLC-DAD-ESI-MSn: Approaches to hyaluronidase inhibitory capacity and antioxidant properties. Mar. Drugs 2012, 10, 2766–2781. [CrossRef] 70. de Sousa, C.B.; Gangadhar, K.N.; Macridachis, J.; Pavao, M.; Morais, T.R.; Campino, L.; Varela, J.; Lago, J.H.G. Cystoseira algae (Fucaceae): Update on their chemical entities and biological activities. Tetrahedron-Asymmetry 2017, 28, 1486–1505. [CrossRef] 71. Kemp, J. Effects of temperature and salinity on resting metabolism in two South African rock pool Fish: The resident gobiid Caffrogobius caffer and the transient sparid Diplodus Sargus Capensis. Afr. Zool. 2009, 44, 151–158. [CrossRef] Sauvageau, C. A propos des Cystoseira de Banyuls et de Guéthary. Bull. Stn. Biol. Arcachon 1912, 14, 133–556. 72. 73. Roberts, M. Studies on marine algae of the British Isles. 6. Cystoseira foeniculacea (Linnaeus) Greville. Eur. J. Phycol. 1968, 3, 547–564. 74. Roberts, M. Studies on marine algae of the British Isles. 8. Cystoseira tamariscifolia (Hudson) Papenfuss. Eur. J. Phycol. 1970, 5, 201–210. [CrossRef] 75. Roberts, M. Studies on marine algae of the British Isles. 9. Cystoseira nodicaulis (Withering) M. Roberts. Eur. J. Phycol. 1977, 12, 175–199. 76. Engelen, A.W.; Espirito-Santo, C.; Simões, T.; Monteiro, C.; Serrão, E.A.; Pearson, G.A.; Santos, R.O.P. Periodicity of propagule expulsion and settlement in the competing native and invasive brown seaweeds, Cystoseira humilis and Sargassum muticum (Phaeophyta). Eur. J. Phycol. 2008, 43, 275–282. [CrossRef] 77. Le Lann, K.; Stiger-Pouvreau, V. Spatio-temporal phenologies of temperate Sargassaceae: Coexistence of invasive and native species. Phycologia 2009, 48, 74. Mar. Drugs 2021, 19, 504 20 of 20 78. Jégou, C.; Culioli, G.; Kervarec, N.; Simon, G.; Stiger-Pouvreau, V. LC/ESI-MSn and 1H HR-MAS NMR analytical methods as useful taxonomical tools within the genus Cystoseira C. Agardh (Fucales; Phaeophyceae). Talanta 2010, 83, 613–622. [CrossRef] [PubMed] 79. Underwood, A.J. Experimental ecology of rocky intertidal habitats: What are we learning? J. Exp. Mar. Biol. Ecol. 2000, 250, 51–76. [CrossRef] 80. Wright, S.W.; Jeffrey, S.W.; Mantoura, R.F.C.; Llewellyn, C.A.; Bjornland, T.; Repeta, D.; Welschmeyer, N. Improved HPLC method for the analysis of chlorophylls and carotenoids in marine phytoplankton. Mar. Ecol. Prog. Ser. 1991, 77, 183–196. [CrossRef] 81. Bidigare, R.; Van Heukelem, L.; Trees, C. Analysis of algal pigments by High-Performance Liquid Chromatography. In Algal Culturing Techniques; Andersen, R.A., Ed.; Academic Press: London, UK, 2005; pp. 327–345. 82. Connan, S.; Delisle, F.; Deslandes, E.; Ar Gall, E. Intra-thallus phlorotannin content and antioxidant activity in Phaeophyceae of temperate waters. Bot. Mar. 2006, 49, 39–46. [CrossRef] 83. Le Lann, K.; Jegou, C.; Stiger-Pouvreau, V. Effect of different conditioning treatments on total phenolic content and antioxidant activities in two Sargassacean species: Comparison of the frondose Sargassum muticum (Yendo) Fensholt and the cylindrical Bifurcaria bifurcata R. Ross. Phycol. Res. 2008, 56, 238–245. [CrossRef] 84. R Development Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2021; Available online: http://www.r-project.org (accessed on 25 June 2021). Siegel, S.; Castellan, N.J. Nonparametric Statistics for the Behavioral Sciences; McGraw-Hill: New York, NY, USA, 1988. 85.
10.3390_ijms241210404	Article Genome-Wide Association Study of Lint Percentage in Gossypium hirsutum L. Races Yuanyuan Wang 1,†, Xinlei Guo 1,†, Xiaoyan Cai 2,3, Yanchao Xu 2, Runrun Sun 1, Muhammad Jawad Umer 2 Kunbo Wang 2, Tengfei Qin 4, Yuqing Hou 2, Yuhong Wang 2, Pan Zhang 1, Zihan Wang 1, Fang Liu 2,3,5 Qinglian Wang 1,* and Zhongli Zhou 2,* , , 1 Collaborative Innovation Center of Modern Biological Breeding of Henan Province, Henan Institute of Science 2 and Technology, Xinxiang 453003, China State Key Laboratory of Cotton Biology, Institute of Cotton Research, Chinese Academy of Agricultural Sciences, Anyang 455000, China; [email protected] (F.L.) 3 Hainan Yazhou Bay Seed Laboratory, National Nanfan Research Institute of Chinese Academy of Agriculture Sciences, Sanya 572025, China Institute of Crop Science, Chinese Academy of Agricultural Sciences, Beijing 100081, China School of Agricultural Sciences, Zhengzhou University, Zhengzhou 450001, China 4 5 * Correspondence: [email protected] (Q.W.); [email protected] (Z.Z.) † These authors contributed equally to this work. Abstract: Lint percentage is one of the most essential yield components and an important economic index for cotton planting. Improving lint percentage is an effective way to achieve high-yield in cotton breeding worldwide, especially upland cotton (Gossypium hirsutum L.). However, the genetic basis controlling lint percentage has not yet been systematically understood. Here, we performed a genome-wide association mapping for lint percentage using a natural population consisting of 189 G. hirsutum accessions (188 accessions of G. hirsutum races and one cultivar TM-1). The results showed that 274 single-nucleotide polymorphisms (SNPs) signiﬁcantly associated with lint percentage were detected, and they were distributed on 24 chromosomes. Forty-ﬁve SNPs were detected at least by two models or at least in two environments, and their 5 Mb up- and downstream regions included 584 makers related to lint percentage identiﬁed in previous studies. In total, 11 out of 45 SNPs were detected at least in two environments, and their 550 Kb up- and downstream region contained 335 genes. Through RNA sequencing, gene annotation, qRT-PCR, protein–protein interaction anal- ysis, the cis-elements of the promotor region, and related miRNA prediction, Gh_D12G0934 and Gh_A08G0526 were selected as key candidate genes for ﬁber initiation and elongation, respectively. These excavated SNPs and candidate genes could supplement marker and gene information for deciphering the genetic basis of lint percentage and facilitate high-yield breeding programs of G. hirsutum ultimately. Keywords: G. hirsutum race; semi-wild cotton; lint percentage; genome-wide association study 1. Introduction Cotton is one of the most important cash crops, which supplies raw material to more than 50 industries globally and is extensively planted in more than 70 countries throughout the world [1]. Of the four cultivated species, upland cotton (Gossypium hirsutum L.), a main supplier for the textile industry, contributes approximately 95% of cotton production, owing to its high yield, attractive ﬁber quality, and wide adaptability [2,3]. Developing high ﬁber yield has always been one of the essential and constant targets in cotton breeding and cultivation, for yield is the foundation for the beneﬁt of cotton planting [4]. Lint percentage (LP) is the fraction of ﬁber weight in seed–cotton weight, which is one of the most essential yield components and an important economic index for cotton planting [5–7]. It is close to cotton ﬁber yield and is signiﬁcantly positively related to cotton yield but negatively Citation: Wang, Y.; Guo, X.; Cai, X.; Xu, Y.; Sun, R.; Umer, M.J.; Wang, K.; Qin, T.; Hou, Y.; Wang, Y.; et al. Genome-Wide Association Study of Lint Percentage in Gossypium hirsutum L. Races. Int. J. Mol. Sci. 2023, 24, 10404. https://doi.org/ 10.3390/ijms241210404 Academic Editors: Qian-Hao Zhu and Jie Sun Received: 5 May 2023 Revised: 8 June 2023 Accepted: 15 June 2023 Published: 20 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Mol. Sci. 2023, 24, 10404. https://doi.org/10.3390/ijms241210404 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10404 2 of 17 correlated with the seed index. However, its genetic architecture is not fully understood; discovering the genetic variation and the candidate genes is still greatly meaningful for underlying lint percentage. Lint percentage is a complex quantitative trait controlled by quantitative trait loci (QTL), which is also signiﬁcantly inﬂuenced by different natural environments [8]. More- over, the complex correlation between lint percentage and other ﬁber traits makes it more difﬁcult to improve yield and quality traits simultaneously [3], only depending on tradi- tional breeding techniques. Detecting QTL-linked or QTL-associated markers and applying for marker-assisted selection (MAS) breeding are effective for precluding the disturbance of environments, effectively breaking the negative correlation among traits, improving the ef- ﬁciency of breeding. Linkage mapping is a traditional method for QTL/gene detection. The ﬁrst cotton linkage map reported 11 linked groups located on chromosomes and mapped some QTL, including two for LP [9]. Since then, linkage mapping has been widely utilized in cotton with the construction of high-density molecular genetic maps using the popula- tions from limited parents. Ample QTL for lint percentage were yielded using interspeciﬁc populations [10] and intraspeciﬁc populations [7,10–13]. These QTL provided meaningful insight for deciphering the genetic basis for lint percentage; however, the time consumption of constructing a mapping population and the low mapping accuracy of linkage analysis limited the exploitation of QTL through MAS breeding [14,15]. On the contrary, association mapping, as an alternative strategy, has lots of advantages, such as high resolution, no need for population construction, great efﬁciency, and low cost [5]. In recent years, association mapping also has been widely used for excavating marker–trait associations, which helps dissect the genetic basis of various complex traits in cotton [16–19]. Through genome-wide association mapping, a large number of SNPs associated with lint percentage have been discovered [20]. These associated markers supplied abundant marker resources and will be helpful for lint percentage improvement of modern cotton cultivars. However, upland cotton has been experiencing domestication and breeding for thou- sands of years. The repeated selection of closely related genotypes has dramatically re- duced the genetic basis of modern cotton cultivars, which has greatly hindered the breeding progress of yield breeding in G. hirsutum [21,22]. G. hirsutum races, also named semiwild cotton, are the progenitors of cultivated G. hirsutum, which have much more abundant phenotypic and genetic variations compared to the cultivars. Furthermore, no reproductive isolation between wild and cultivated types promoted the genetic mechanism analysis of complex traits. In the past years, with G. hirsutum races, several researchers executed linkage mapping, committing to broadening the genetic base for improving yield [23–26], ﬁber quality [22,27], and ﬁber color [25,28]. In our previous reports, SSR and SNP markers were selected to implement genome-wide association mapping for abiotic stress resistance, including glyphosate resistance [29] and drought resistance [30]. Furthermore, we per- formed a transcriptome analysis of G. hirsutum race accessions under alkali-salt stress and excavated the related candidate genes and signaling pathways [31]. The results above provided proliﬁc information on markers and gene resources for the genetic improvement of cultivated upland cotton. To detect the genetic architecture associated with lint percentage and the related can- didate genes, we used the Cotton80KSNP chip to genotype the association mapping panel, comprising 188 accessions of G. hirsutum races and one cultivar (TM-1); we also investi- gated the lint percentage from ﬁve different environments and conducted the genome- wide association study (GWAS) to detect the genetic loci associated with lint percentage. Finally, we predicted the candidate genes that might affect lint percentage based on gene annotation and the gene expression pattern during ﬁber development. The results not only provided molecular markers and candidate genes for further understanding the genetic architecture of lint percentage but also facilitated designing high-yielding cotton through MAS breeding. Int. J. Mol. Sci. 2023, 24, 10404 3 of 17 2. Results and Discussion 2.1. Large Variation in Lint Percentage in G. hirsutum Races We evaluated the lint percentage of 189 accessions (188 accessions of G. hirsutum races and one cultivar (TM-1)) across ﬁve environments during 2014–2015 and 2015–2016. The phenotype of lint percentage showed continuous and extensive variations among the accessions under every individual environment (Table 1). The values ranged from 8.29 to 54.77%, with a mean value of 27.87% across the ﬁve environments. Moreover, the coefﬁcients of variation (CV) exhibited relatively large values among the different environments, which ranged from 19.34 to 23.39%. However, the broad-sense heritability (H2) of the lint percentage was 77.20%. The correlation analysis of the lint percentage across the ﬁve environments ranged from 0.79 to 0.92, which uncovered signiﬁcant and highly positive correlations among different environments (Table S1). The results suggested that lint percentage may have been controlled by multiple genes. Table 1. Descriptive statistical analysis of lint percentage of G. hirsutum races across ﬁve environments. Env. E1 E2 E3 E4 E5 Max 47.55 54.77 43.8 43.86 42.43 Min 12.03 8.29 14.47 12.35 14.87 Mean 30.31 28 28.04 26.61 26.37 Std. 5.86 6.55 6.21 6.21 5.76 CV 19.34 23.39 22.14 23.34 21.83 Skewness Kurtosis 0.14 0.43 0.48 0.39 0.51 −0.15 0.85 −0.24 −0.39 −0.03 E1, E2, E3, E4, and E5 indicate the ﬁve environments: 2014–2015 Yacheng, 2014–2015 Damao, 2015–2016 Yacheng, 2015–2016 Damao in the greenhouse, and 2015–2016 Damao, respectively; CV, coefﬁcient of variation. Several previous publications reported the evaluation of LP using the cultivated G. hirsutum. Song et al. [5], Su et al. [4], Sun et al. [8], Huang et al. [17], and Xing et al. [32] measured LP values of 276, 290, 719, 503, and 196 upland cotton accessions, respectively. The LP values ranged from 10.49 to 49.62%, from 28.49 to 47.01%, from 23.68 to 56.92%, from 27.07 to 47.38%, and from 22.1 to 49.9%, with the mean values of 37.60%, 40.89%, 39.85%, 38.29%, and 37.54%, respectively. The CV of LP were 9.60%, 7.08%, 7.62%, 7.93%, and 8.47%, respectively. In comparison with these studies, the LP value of G. hirsutum races is much lower than that of cultivated G. hirsutum, whereas the CV value of G. hirsutum races was higher, which implied that the LP of the cultivars was greatly improved but the genetic diversity rapidly reduced during the years of repeated selection in cotton yield breeding. 2.2. Genome-Wide Association Study for Lint Percentage We carried out Genome-wide association study (GWAS) to discover loci underlying lint percentage across ﬁve environments using both the GAPIT 3.6.0 and Tassel 5 software. SNPs with −log10p values greater than 4.71 (−log10(1/51,268)) [18] were considered as signiﬁcantly associated with lint percentage. Based on FaST-LMM, FarmCPU, BLINK, MLMM, and MLM of GAPIT 3.6.0 software and MLM of Tassel 5 software, we identiﬁed a total of 274 signiﬁcantly associated SNPs randomly distributed on 24 chromosomes (Table S2) (these 274 SNPs were also identiﬁed signiﬁcantly associated with lint percentage by GLM and super models of GAPIT 3.6.0 software, and GLM model of Tassel 5 soft- ware). Similarly, we summarized QTL-linkage and QTL-associated with lint percentage based on 54 previous reports and found the QTL were distributed unevenly across all the 26 chromosomes (Table S3; Figure 1), which meant that lint percentage was controlled by multiple genes. Int. J. Mol. Sci. 2023, 24, 10404 4 of 17 Figure 1. The distribution of markers related to lint percentage in G. hirsutum. Among the 274 SNPs, 45 were simultaneously identiﬁed at least in two environments or/and at least by two models (Table 2). These SNPs were randomly distributed on 18 chromosomes: A3–A5, A7, A8, A11–A13, D2–D5, and D8–D13. Speciﬁcally, four SNPs on chromosome A3 (TM6245, TM6246, TM6247, and TM6248) were located close together, and they were distributed in a 20 Kb region; ten SNPs on chromosome A11 were distributed at a distance of 1.86 Mb, of which nine SNPs (TM39097, TM39105, TM39111, TM39118, TM39119, TM39120, TM39121, TM39122, and TM39136) positioned in a 23 Kb region; two SNPs (TM55817 and TM55819) on chromosome D4; and two SNPs (TM57478 and TM57486) on chromosome D5 were distributed in 8.4 Kb and 248 Kb regions, respectively. Table 2. SNPs detected signiﬁcantly associated with lint percentage at least in two environments or at least by two models. FaST-LMM FarmCPU BLINK MLMM MLM GAPIT Tassel TMLM SNP Chr TM6245 TM6246 TM6247 TM6248 TM8562 TM8787 TM10387 TM10953 TM13159 TM21439 TM22226 TM38921 TM39097 TM39105 A3 A3 A3 A3 A4 A4 A5 A5 A5 A7 A8 A11 A11 A11 Pos. (bp) 6,000,457 6,008,064 6,015,363 6,020,227 1,704,160 5,125,340 7,465,603 21,122,062 83,646,730 BLUP (4.90) BLUP (5.10) BLUP (5.10) BLUP (4.90) E5 (4.71) 73,820,001 E2 (5.42) 8,180,911 65,498,934 77,128,215 77,162,385 BLUP (4.90) BLUP (5.10) BLUP (5.10) BLUP (4.90) E5 (4.71) E2 (5.42) BLUP (6.99) E5 (6.23) E1 (10.43) E3 (9.45) E3 (17.39); E4 (6.90); E5 (9.84) E4 (11.19); E5 (6.70) E2 (7.92) E5 (6.37) E1 (7.94) E3 (4.97) E3 (6.75); E4 (6.30); E5 (6.66) E4 (15.51) E2 (6.45) E2 (5.41) E2 (4.94) E2 (5.01) E2 (5.23) Int. J. Mol. Sci. 2023, 24, 10404 4 of 17 chromosomes (Table S3; Figure 1), which meant that lint percentage was controlled by multiple genes. Figure 1. The distribution of markers related to lint percentage in G. hirsutum. Among the 274 SNPs, 45 were simultaneously identified at least in two environments or/and at least by two models (Table 2). These SNPs were randomly distributed on 18 chromosomes: A3–A5, A7, A8, A11–A13, D2–D5, and D8–D13. Specifically, four SNPs on chromosome A3 (TM6245, TM6246, TM6247, and TM6248) were located close together, and they were distributed in a 20 Kb region; ten SNPs on chromosome A11 were distrib-uted at a distance of 1.86 Mb, of which nine SNPs (TM39097, TM39105, TM39111, TM39118, TM39119, TM39120, TM39121, TM39122, and TM39136) positioned in a 23 Kb re-gion; two SNPs (TM55817 and TM55819) on chromosome D4; and two SNPs (TM57478 and TM57486) on chromosome D5 were distributed in 8.4 Kb and 248 Kb regions, respectively. Table 2. SNPs detected significantly associated with lint percentage at least in two environments or at least by two models. SNP Chr Pos. (bp) GAPIT Tassel FaST-LMM FarmCPU BLINK MLMM MLM TMLM TM6245 A3 6,000,457 BLUP (4.90) BLUP (4.90) TM6246 A3 6,008,064 BLUP (5.10) BLUP (6.99) BLUP (5.10) TM6247 A3 6,015,363 BLUP (5.10) BLUP (5.10) TM6248 A3 6,020,227 BLUP (4.90) BLUP (4.90) TM8562 A4 1,704,160 E5 (6.37) E5 (6.23) TM8787 A4 5,125,340 E1 (7.94) E1 (10.43) TM10387 A5 7,465,603 E5 (4.71) E5 (4.71) TM10953 A5 21,122,062 E3 (4.97) E3 (9.45) TM13159 A5 83,646,730 E3 (6.75); E4 (6.30); E5 (6.66) E3 (17.39); E4 (6.90); E5 (9.84) TM21439 A7 73,820,001 E2 (5.42) E2 (5.42) TM22226 A8 8,180,911 E4 (15.51) E4 (11.19); E5 (6.70) TM38921 A11 65,498,934 E2 (6.45) E2 (7.92) TM39097 A11 77,128,215 E2 (5.41) E2 (5.01) Int. J. Mol. Sci. 2023, 24, 10404 5 of 17 Table 2. Cont. SNP TM39111 TM39118 TM39119 TM39120 TM39121 TM39122 Chr A11 A11 A11 A11 A11 A11 Pos. (bp) 77,201,721 77,245,152 77,250,629 77,254,990 77,259,287 77,267,198 TM39136 A11 77,356,582 TM42527 TM42563 TM46588 TM47373 TM52184 TM53197 TM53588 TM55461 TM55484 TM55582 TM55817 TM55819 TM57002 TM57478 TM57486 TM58314 TM69172 TM72587 TM73642 TM75017 TM76204 TM78526 TM80159 TM81325 A12 A12 A13 A13 D2 D2 D3 D4 D4 D4 D4 D4 D5 D5 D5 D5 D8 D9 D10 D10 D11 D12 D13 D13 73,222,779 73,856,986 60,134,206 72,835,782 51,543,776 64,206,361 4,016,585 2,087,134 2,902,882 3,824,554 7,622,861 7,631,258 6,810,289 19,820,229 20,068,395 49,232,321 53,049,470 43,752,600 8,030,303 58,300,092 24,111,532 34,893,826 2,590,468 42,144,468 FaST-LMM FarmCPU BLINK MLMM MLM GAPIT E2 (6.57) E2 (7.05) E2 (4.99) E2 (6.57) E2 (4.85) E2 (5.41) E2 (5.48) E2 (5.35) E2 (5.39) E2 (5.33) E3 (9.26); E5 (12.70) E1 (5.91) E2 (5.34) E1 (6.86) E3 (8.73) E1 (4.89); E4 (10.43) E1 (4.77) E4 (15.33) E5 (9.59) E2 (6.80) E4 (22.33); E5 (4.79) E2 (5.45) E3 (13.74) E3 (7.45) E3 (5.42) E4 (4.72) E3 (8.00) E3 (5.16) E5 (5.39) E1 (7.99); E2 (9.96) E2 (6.19) E2 (5.35) E2 (5.35) E4 (4.95) E2 (4.84) E1 (5.59); E2 (4.84); E3 (14.49); E4 (5.33); E5 (13.96) E1 (13.91) E2 (7.14) E1 (14.22) E3 (7.06) E4 (6.60) E3 (6.30) E1 (13.39); E4 (14.39); E5 (10.63) E2 (9.29) E5 (13.00) E2 (6.19) E2 (5.23) E2 (5.35) E2 (5.35) E4 (12.47) E4 (4.95) E4 (4.76) E3 (8.17) E3 (7.83) E3 (6.93) E4 (13.12) E3 (14.52) E5 (9.39) E1 (6.54) E2 (4.84) Tassel TMLM E2 (5.06) E2 (4.73) E2 (4.80) E2 (5.42) E2 (4.72) E2 (5.42) E2 (6.86) E1 (4.85); E3 (5.82) E3 (4.87) E1 (4.90); E3 (4.77) E1, E2, E3, E4, and E5 indicate the ﬁve environments 2014–2015 Yacheng, 2014–2015 Damao, 2015–2016 Yacheng, 2015–2016 Damao in the greenhouse, and 2015–2016 Damao, respectively; Fast-LMM, FarmCPU, BLINK, MLMM, and MLM were models used in GAPIT 3.6.0 software; TMLM 5 was the model used in Tassel 5 software. All the SNPs were also detected by GLM and SUPER in GAPIT 3.6.0 software and were also identiﬁed by GLM of Tassel 5 software. Of the 274 SNPs, 151 and 123 were positioned in the A and D sub-genomes, respec- tively; of the 45 SNPs simultaneously identiﬁed at least in two environments or/and at least by two models, 25 and 20 were distributed in the A and D sub-genomes, respectively. The results showed that the number of associated markers in the A sub-genome was not signiﬁcantly different from that in the D sub-genome. Consistent with our opinion, in the previous reports, one GWAS identiﬁed eleven and eight associated SNPs in the A and D sub-genome [16], based on the genome resequencing of 318 accessions of core G. hirsutum germplasm; one GWAS detected eleven and ten lint-percentage-associated SNPs in the Int. J. Mol. Sci. 2023, 24, 10404 6 of 17 A and D sub-genomes, based on a CottonSNP63K array and a worldwide population consisting of 503 G. hirsutum accessions [17]; another GWAS identiﬁed nine and seven SNPs associated with lint percentage in the A and D sub-genomes, using SSR markers and 241 Upland cotton collections [33]. In addition, we surveyed 54 studies about linkage and association mapping on lint percentage (Table S3; Figure 1) and found that lint percentage QTL were distributed evenly between the A and D sub-genomes (1392 and 1285). Niu et al. [34] summarized QTL for lint percentage and found that the A sub-genome contained similar unique, tightly linked, and major QTL to that of the D sub-genome. Similarly, Said et al. [35] also reported that yield-related QTL were almost evenly distributed between the A and D sub-genomes [35]. The results suggested that the A and D sub-genomes contributed equally for lint percentage. We also found that 220 (80.29%) and 54 (19.71%) SNPs were distributed as intergenic and intragenic; of the 45 SNPs detected at least in two environments or/and at least by two models, 36 (80%) and 9 (20%) were located as intergenic and intragenic (Table S4). The results suggested that most of the signiﬁcant markers were positioned as intergenic. To conﬁrm this, we analyzed the physical positions of the markers associated with lint percentage and the nearest makers of the QTL related, based on previous reports on QTL linkage and QTL associated with lint percentage, and found that most (1811, 87.57%) of the associated loci (totally 2068 SNPs investigated) were located as intergenic based on the genome sequence of the TM-1 (NAU v1.1) [36] (Table S3). The above results supported that most of the associated loci were distributed on non-protein-coding regions of the genome [37], which suggested that many loci implicated by GWAS might work through altering the genetic regulation of one or more target genes by regulating changes in target gene expression. Fortunately, transcriptome sequencing of the natural population could provide both the expression proﬁle of each gene and the genetic variations in the gene coding regions. 2.3. Validation of the Stability of 45 SNPs Associated with Lint Percentage To validate the stability of these 45 signiﬁcantly associated SNPs detected at least in two environments or by two models, we investigated 2677 markers/QTL for lint percentage mapped previously from 54 publications in upland cotton (Table S3). Then, the physical locations of 2831 markers were obtained through BLASTN [38], and 584 makers were shown being located within 5 Mb up- and downstream of the 45 SNPs (Figure 2). In addition, ﬁfteen SNPs identiﬁed in this study (TM6245, TM6246, TM6247, TM6248, TM21439, TM42563, TM53197, TM53588, TM55461, TM555582, TM57478, TM57486, TM73642, TM75017, and TM80159), were localized within or near previously reported markers/QTLs. Speciﬁcally, four SNPs (TM6245, TM6246, TM6247, and TM6248) on A3 were located on the overlap of qLP-C-3 (CGR6528–BNL2443) detected by Wang et al. [39] and were 100 Kb away from TM6282 detected by Zhu et al. [13]. Eight SNPs (TM21439 on A7, TM42563 on A12, TM53197 on D2, TM53588 on D3, TM55461 on D4, TM55582 on D4, TM57478 and TM57486 on D5) were within 200 Kb away from TM21493 detected by Zhu et al. [13], LDB_12_93252980 detected by Su et al. [4], TM53255 detected by Zhu et al. [13], TM53212 detected by Xing et al. [32], DPL0281 detected by Zhang et al. [26], SWU20808 detected by Yang et al. [40], TM55475 detected by Zhu et al. [13], Marker26949 detected by Wang et al. [41], TM57480 and TM57483 detected by Zhu et al. [13], or D05_20162316 detected by Sarfraz et al. [20]. TM73642 on D10 was 70 Kb away from TM73693 detected by Zhu et al. [13]. TM75017 on D10 also was approximately 70 Kb away from i12188Gh, i12189Gh, and i12190Gh detected by Song et al. [5] and was 70 Kb and 150 Kb away from qLP-Dt10- 3 (Marker36441—Marker36437) and qLP-Dt10-2 (Marker36463—Marker36442) detected by Wang et al. [41], respectively. TM80159 on D13 was 34–170 Kb away from three SNPs (TM80151, TM80163, and TM80171) detected by Xing et al. [32]. The results above suggested that the 45 SNPs corresponded to markers/QTL reported previously. These markers were repeatedly detected in different populations with different genetic backgrounds and could potentially be considered in the MAS of target traits. Int. J. Mol. Sci. 2023, 24, 10404 7 of 17 Figure 2. Comparison of the location of markers identiﬁed in previous and present studies. The markers marked red meant the markers associated with lint percentage identiﬁed in the present study; the markers marked blue meant there were at least two markers between the two markers marked up and down. Int. J. Mol. Sci. 2023, 24, 10404 8 of 17 2.4. Discovery of Candidate Genes for Lint Percentage In total, 11 of the 45 SNPs were identiﬁed at least in two environments. We compared the lint percentage between alleles of the eleven associated SNPs in each environment and found that signiﬁcant differences existed between the superior and inferior alleles (Table S5; Figure 3). The superior alleles might be integrated properly by marker-assisted selection in order to improve lint percentage. We then discovered the candidate genes near these eleven associated SNPs, based on available genes’ annotation information of the G. hirsutum TM-1 genome [36]. A total of 335 candidate genes was obtained from the 550 Kb up- and downstream regions of 11 SNPs. Based on RNA-seq data of the ovule (including −3, −1, 0, 1, and 3 dpa) and ﬁber (including 5, 10, and 20 dpa) tissues in TM-1 [36], we predicted the candidate genes involved in ﬁber initiation and ﬁber elongation, respectively. After removing the genes with FPKM < 2, the gene expression changed within twofold, and the homolog without annotation, 125 genes (including 80 for the ovule and 86 for the ﬁber, 41 genes were overlapped in both ovule and ﬁber) was retained for further analysis (Figure S1; Tables S6 and S7). Based on the gene annotation information of these 125 genes in Arabidopsis, eleven genes were found to be related to root development; eight genes were related to plant-type cell wall biogenesis; six genes were related to auxin polar transport or its signaling pathway; ﬁve genes were related to ﬂower development and response to abscisic acid, respectively; four genes were related to cell differentiation and cell division, respectively; three genes were related to plant epidermis development, pollen maturation, anther dehiscence, and cell tip growth, respectively; and two genes were related to anther development, pollen development, pollen tube, and unidimensional cell growth, respectively (Tables S6 and S7). The gene expression patterns of these 125 genes were reﬂected (Figure S1) using RNA-seq data of TM-1 [36]. For 80 genes related to the ovule tissues, twenty-one and eight genes were expressed highly at −3 dpa and −2 dpa, which may have played roles in the differentiation of ﬁber cells; nine genes had high gene expressions at 0 dpa, which may been involved in producing ﬁber cell initials; 13 genes were expressed highly at 1 dpa, which may participated in the transformation from non-polar expansion to polar elongation; 14 genes expressed strongly at 3 dpa, which may participated in cotton ﬁber elongation. For 86 genes related to the ﬁber tissues, fourteen, eight, and ten genes, were expressed highly at 5 dpa, 10 dpa, and both stages, respectively, which may been involved in cell elongation; 41 genes were expressed highly at 20 dpa, suggesting that these genes may have functioned in cell elongation and/or primary wall biosynthesis. Referring to gene annotation and the RNA-seq analysis, we selected several genes to perform qRT-PCR using ovule (−2, −1, 0, 1, and 3 dpa) and ﬁber (5, 10, and 20 dpa) tissues. The results showed that the expression patterns of most genes determined by RNA-seq and qRT-PCR were similar (Figure 3). In the ovule tissues, Gh_D05G3191 and Gh_D12G0955 were highly expressed at 3 dpa; Gh_D03G0337 and Gh_D12G0934 were highly expressed at 1 dpa. Of these four genes mentioned above, Gh_D12G0934 was found to be involved in the progress of cell differentiation, cell division, circadian rhythm, regulation of transcription, DNA-templating, and root hair cell tip growth. The occurrence of root hair was similar to ﬁber initiation, suggesting that Gh_D12G0934 might have been involved in ﬁber initiation. In the ﬁber tissues, Gh_D03G0345 was strongly expressed at 10 dpa, Gh_A08G0526 was highly expressed from 10 dpa to 20 dpa, whereas Gh_A08G0520 and Gh_D12G0955 were preferentially expressed at 20 dpa. Among these four genes, Gh_A08G0526 was related to the progress of plant-type cell wall biogenesis, regulation of transcription, DNA-templating, and shoot system development. We predicted that Gh_A08G0526 might function in secondary wall deposition of cotton ﬁber. Int. J. Mol. Sci. 2023, 24, 10404 9 of 17 Figure 3. Gene expression pattern of the candidate genes of lint percentage. (A) The expression pattern of candidate genes detected by RNA-seq in the ovule tissue. (B) The expression pattern of candidate genes detected by RNA-seq in the ﬁber tissue. (C) The expression pattern of candidate genes detected by qRT-PCR in the ovule tissue. (D) The expression pattern of candidate genes detected by qRT-PCR in the ﬁber tissue. (E) Box plots for lint percentage of TM78526. (F) Box plots for lint percentage of TM22226. “”, signiﬁcant at a = 0.01 level. “.”, the outliers. Genes control phenotypes, and the identiﬁcations and isolations of the genes reg- ulating LP are essential for MAS and gene engineering breeding. Through GWAS for lint percentage, fruitful key candidate genes have been yielded directly based on linkage disequilibrium (Table S8), such as Gh_A02G1268 (MIPS, a member of the myo-inositol-1- phosphate synthase gene family) [42]; WD40 [15]; Gh_D08G2376 (a homolog of AT3G07020 and GhSGT1) [17]; Gh_A02G1392 (AIL6, a homolog of the AP2/ETHYLENE RESPONSE FACTOR (ERF)-type transcription-factor-encoding gene AINTEGUMENTA-like 6 in Ara- bidopsis) and Gh_D08G0312 (EIL, encoding ethylene insensitive 3-like family protein) [16]; Gh_D03G1064 (FRI, encoding a FRIGIDA-like protein) and Gh_D12G2354 (encoding a GPR107-like protein) [8]; Gh_D02G0025 (encoding tetratricopeptide repeat (TPR)-like su- perfamily protein) [19]; Gh_D05G1124 (encoding Protein phosphatase 2C family protein) and Gh_D05G0313 (the ortholog of AtLUT2) [5]; Gh_A02G1269 (encoding chaperone protein dnaJ 20), Gh_A02G1278 (encoding E3 ubiquitin-protein ligase RHA2A), Gh_A02G1280, and Gh_A02G1295 (AtAMP1, ALTERED MERISTEM PROGRAM1 in Arabidopsis) [43]; Gh_A05G2488 (a member of auxin transport facilitator family named PIN-FORMED LIKES proteins (PILS)), Gh_D13G0342 (RAB GTPase homolog G3F (RABG3F)), and Gh_A10G2138 (PRA1, encoding prenylated RAB acceptor protein 1) [13]; and GH_A07G1389 (encoding a tetratricopeptide repeat (TPR)-like superfamily protein) [44]. According to the pre- diction of gene function, transcription; signaling pathways (such as hormone, calcium Int. J. Mol. Sci. 2023, 24, 10404 9 of 17 Figure 3. Gene expression pattern of the candidate genes of lint percentage. (A) The expression pat-tern of candidate genes detected by RNA-seq in the ovule tissue. (B) The expression pattern of can-didate genes detected by RNA-seq in the fiber tissue. (C) The expression pattern of candidate genes detected by qRT-PCR in the ovule tissue. (D) The expression pattern of candidate genes detected by qRT-PCR in the fiber tissue. (E) Box plots for lint percentage of TM78526. (F) Box plots for lint per-centage of TM22226. “”, significant at 𝑎 = 0.01 level. “.”, the outliers. Genes control phenotypes, and the identifications and isolations of the genes regu-lating LP are essential for MAS and gene engineering breeding. Through GWAS for lint percentage, fruitful key candidate genes have been yielded directly based on linkage dis-equilibrium (Table S8), such as Gh_A02G1268 (MIPS, a member of the myo-inositol-1-phosphate synthase gene family) [42]; WD40 [15]; Gh_D08G2376 (a homolog of AT3G07020 and GhSGT1) [17]; Gh_A02G1392 (AIL6, a homolog of the AP2/ETHYLENE RESPONSE FACTOR (ERF)-type transcription-factor-encoding gene AINTEGUMENTA-like 6 in Ara-bidopsis) and Gh_D08G0312 (EIL, encoding ethylene insensitive 3-like family protein) [16]; Gh_D03G1064 (FRI, encoding a FRIGIDA-like protein) and Gh_D12G2354 (encoding a GPR107-like protein) [8]; Gh_D02G0025 (encoding tetratricopeptide repeat (TPR)-like su-perfamily protein) [19]; Gh_D05G1124 (encoding Protein phosphatase 2C family protein) and Gh_D05G0313 (the ortholog of AtLUT2) [5]; Gh_A02G1269 (encoding chaperone pro-tein dnaJ 20), Gh_A02G1278 (encoding E3 ubiquitin-protein ligase RHA2A), Gh_A02G1280, and Gh_A02G1295 (AtAMP1, ALTERED MERISTEM PROGRAM1 in Ara-bidopsis) [43]; Gh_A05G2488 (a member of auxin transport facilitator family named PIN-FORMED LIKES proteins (PILS)), Gh_D13G0342 (RAB GTPase homolog G3F (RABG3F)), Int. J. Mol. Sci. 2023, 24, 10404 10 of 17 +), and MAPK); energy metabolism (such as starch and sucrose metabolism); substance (Ca2 transport; cell division; and cell wall development are important biological processes for LP formation. 2.5. Gh_A08G0526 and Gh_D12G0934 Were Candidate Genes for Lint Percentage We separately investigated the protein interaction of Gh_A08G0526 and Gh_D12G0934. Until now, there has been no database available to predict protein interactions of genes in cotton directly. Here, we predicted the protein interactions of the homologous genes in Arabidopsis (Figure 4). For Gh_A08G0526, its homolog in Arabidopsis is AT2G44745.1 (WRKY12), which is involved in plant-type cell wall biogenesis, regulation of transcription, DNA-templateing, and/or shoot system development. Ten interactions protein of WRKY12 (AT2G44745.1) were identified, and a total of twenty-five protein interactive relationships was identified between these eleven genes (Table S9), of which four (AT1G79180.1, AT5G12870.1, AT4G22680.1, and AT3G08500.1, homologous with Gh_D08G2456, Gh_D09G1082, Gh_ D09G1690, and Gh_A08G1308, respectively) and two (AT3G61910.1 and AT1G32770.1 ho- mologous with Gh_D11G1062, Gh_A11G0915, and Gh_D12G2359, respectively) genes were members of the MYB and NAC family, respectively. The interacting genes included MYB63, MYB83, MYB85, MYB46, NAC012, NAC066, NST1, and VND7. The related genes in cotton were reported mainly to be involved in ﬁber initiation, ﬁber elongation, and/or secondary wall deposition, then contributing to the formation of lint. For instance, STV106, an NAC- domain-containing protein, could regulate secondary wall biogenesis [45]. GhMYB212 regulated the gene expression pattern of GhSWEET12, and further played an important role during the ﬁber elongation; plants with silenced GhMYB212 could accumulate less sucrose and glucose during the development of ﬁber and produce shorter ﬁbers and lower lint yields [46]. R2R3 MYB (GhMYB25-like) RNAi led to ﬁberless seeds, whereas trichomes elsewhere were normal [47]; GhMYB25-silenced cotton altered the timing of ﬁber elongating rapidly, resulting in the short ﬁbers and trichomes of other tissues reducing dramatically and, ﬁnally, reducing the seed production [48]. GhMYB7-overexpressed cotton plants accel- erated the cellulose biosynthesis of the secondary cell walls and resulted in shorter ﬁbers with thicker walls, but GhMYB7 RNAi plants delayed cellulose synthesis and produced longer ﬁbers with thinner walls [49]. In Arabidopsis, XND1 could regulate the activity of NST1 during the formation of the secondary cell walls of ﬁber cells [50]. GhFSN1, as a positive regulator, activated its downstream genes involved in the secondary cell wall and further controlled the secondary cell wall formation of cotton ﬁbers [51]. The critical genes related to the formation of the secondary cell wall, such as the transcription factor gene MYB46 and its homolog MYB83, could directly be regulated by VND7 [52–54]. The overexpression of MYB46 or MYB83 led to the secondary cell walls thickening, whereas the myb46 and myb83 loss-of-function double mutant did not yield any observable sec- ondary cell walls [55]. GhWRKY16 functioned as a positive regulator during ﬁber initiation and elongation, silencing GhWRKY16 in cotton, produced shorter ﬁbers and reducing the number of ﬁber protrusions on the ovule [56]. Taken together, NAC domain transcription factors function as master switches during the thickening of the secondary cell wall; MYB transcription factors, downstream from the NST genes, are also important in secondary wall biosynthesis [26,45]. The above results suggested that Gh_A08G0526 might be involved in ﬁber development through regulating MYB and NAC. Int. J. Mol. Sci. 2023, 24, 10404 11 of 17 Figure 4. Interaction network of these two key candidate genes in Arabidopsis. (A) Interaction network of Gh_A08G0526. (B) Interaction network of Gh_D12G0934. For Gh_D12G0934, its homolog in Arabidopsis AT4G00150.1 (HAM3), is involved in cell differentiation, cell division, circadian rhythm, regulation of transcription, DNA-templating, and/or root hair cell tip growth. Ten pairs of HAM3 (AT4G00150.1) protein interactions were identiﬁed, and a total of twenty-one protein interactive relationships was identiﬁed between these eleven genes (Table S10), of which three (AT3G11260.1, AT1G46480.1, and AT5G05770.1, homologous with Gh_A10G2087, Gh_D05G1962, and Gh_A10G2087, respec- tively) and two genes (AT4G17340.1 and AT5G47450.1, homologous with Gh_Sca059366G01 and Gh_D03G1568, respectively) were also family members of the WOX and TIP. These related genes functioned as being related to root development. For instance, ACT7 partici- pated in root growth, epidermal cell speciﬁcation, cell division, and root architecture [57]; the act7-4 mutant reduced the root growth, twisted root apical cells, and obliqued junc- tions between cells dramatically [58]. WOX4 functioned to facilitate the differentiation and/or maintenance of the initial cells of the developing vasculature; silencing WOX4 in Arabidopsis-generated small plants, with differentiated xylem and phloem reducing severely [59]. The expressions of WOX11/12 and WOX5/7 were vital for the initiation of the root primordium during the formation of root tissue in Arabidopsis [60]. The lateral roots of the 35S: AtASL5 transgenic cockscombs had a central–peripheral defect [61]. Atham1,2,3 mutant plants generated signiﬁcantly smaller root meristems in both the longitudinal and radial axes than those of a wild type by reducing the division rates of root meristem cells [62]. Furthermore, we analyzed the cis-elements of the promoter region (2000 bp sequence upstream of the coding sequence) for the two candidate genes (Gh_A08G0526 and Gh_ D12G0934) and found that seven elements were related to light in the promoter regions of Gh_A08G0526 and Gh_D12G0934, respectively (Table S11). An auxin-responsive element (TGA-element) was also identiﬁed in the promoter region of Gh_D12G0934. Moreover, we identiﬁed miRNAs targeting Gh_D12G0934 in G. hirsutum and found that ghr-miR171 targeted Gh_D12G0934. In cotton, miR171 showed signiﬁcantly higher expression in the fuzzless–lintless mutant Xu-142-ﬂ than in the wildtype [63]. Meanwhile, miR171 was expressed much more abundantly in the ovule during the initiation than in ﬁber during elongation and secondary wall thickening in 3–79 [64]. Generally speaking, the root mor- phogenesis had a similar mechanism to the cotton ﬁber initiation; therefore, Gh_D12G0934 might have affected lint percentage by controlling ﬁber initiation. However, its function needs to be veriﬁed by transgenic technologies, such as gene overexpression and RNA silencing. The above results suggested that Gh_A08G0526 and Gh_D12G0934 might be crucial candidate genes involved in ﬁber elongation or initiation. Int. J. Mol. Sci. 2023, 24, 10404 11 of 17 Figure 4. Interaction network of these two key candidate genes in Arabidopsis. (A) Interaction net-work of Gh_A08G0526. (B) Interaction network of Gh_D12G0934. For Gh_D12G0934, its homolog in Arabidopsis AT4G00150.1 (HAM3), is involved in cell differentiation, cell division, circadian rhythm, regulation of transcription, DNA-tem-plating, and/or root hair cell tip growth. Ten pairs of HAM3 (AT4G00150.1) protein inter-actions were identified, and a total of twenty-one protein interactive relationships was identified between these eleven genes (Table S10), of which three (AT3G11260.1, AT1G46480.1, and AT5G05770.1, homologous with Gh_A10G2087, Gh_D05G1962, and Gh_A10G2087, respectively) and two genes (AT4G17340.1 and AT5G47450.1, homologous with Gh_Sca059366G01 and Gh_D03G1568, respectively) were also family members of the WOX and TIP. These related genes functioned as being related to root development. For instance, ACT7 participated in root growth, epidermal cell specification, cell division, and root architecture [57]; the act7-4 mutant reduced the root growth, twisted root apical cells, and obliqued junctions between cells dramatically [58]. WOX4 functioned to facilitate the differentiation and/or maintenance of the initial cells of the developing vasculature; si-lencing WOX4 in Arabidopsis-generated small plants, with differentiated xylem and phloem reducing severely [59]. The expressions of WOX11/12 and WOX5/7 were vital for the initiation of the root primordium during the formation of root tissue in Arabidopsis [60]. The lateral roots of the 35S: AtASL5 transgenic cockscombs had a central–peripheral defect [61]. Atham1,2,3 mutant plants generated significantly smaller root meristems in both the longitudinal and radial axes than those of a wild type by reducing the division rates of root meristem cells [62]. Furthermore, we analyzed the cis-elements of the promoter region (2000 bp sequence upstream of the coding sequence) for the two candidate genes (Gh_A08G0526 and Gh_D12G0934) and found that seven elements were related to light in the promoter re-gions of Gh_A08G0526 and Gh_D12G0934, respectively (Table S11). An auxin-responsive element (TGA-element) was also identified in the promoter region of Gh_D12G0934. Moreover, we identified miRNAs targeting Gh_D12G0934 in G. hirsutum and found that ghr-miR171 targeted Gh_D12G0934. In cotton, miR171 showed significantly higher expres-sion in the fuzzless–lintless mutant Xu-142-fl than in the wildtype [63]. Meanwhile, miR171 was expressed much more abundantly in the ovule during the initiation than in fiber during elongation and secondary wall thickening in 3–79 [64]. Generally speaking, the root morphogenesis had a similar mechanism to the cotton fiber initiation; therefore, Gh_D12G0934 might have affected lint percentage by controlling fiber initiation. However, its function needs to be verified by transgenic technologies, such as gene overexpression and RNA silencing. The above results suggested that Gh_A08G0526 and Gh_D12G0934 might be crucial candidate genes involved in fiber elongation or initiation. Int. J. Mol. Sci. 2023, 24, 10404 12 of 17 3. Materials and Methods 3.1. Natural Population Germplasm and Lint Percentage Evaluation A natural population, including 189 accessions of G. hirsutum, was used for association mapping in the present study. One-hundred-and eighty-eight accessions consisted of all the seven G. hirsutum races, and the other was a representative of the cultivars, the upland cotton genetic standard G. hirsutum L. acc. Texas Marker-1 (TM-1) [30]. All the G. hirsutum races were originally introduced from the USDA-ARS Southern Agricultural Research Center in College Station, Texas, USA, were perennially puriﬁed and preserved in the National Wild Cotton Nursery, Sanya, Hainan, China, and were supervised by the Institute of Cotton Research of Chinese Academy of Agricultural Sciences (ICR-CAAS), Anyang, Henan, China. All the accessions were legally planted in ﬁve environments (E1: 2014–2015 Yacheng; E2: 2014–2015 Damao; E3: 2015–2016 Yacheng; E4: 2015–2016 Damao in the greenhouse; E5: 2015–2016 Damao). In each experimental environment, 20–25 plants were planted in a single row plot (the row length and the row interval were 5.0 m and 1.0 m, respectively). Standard local ﬁeld management practices were performed throughout the whole planting season. At the open-boll bloom stage, we randomly collected 30 naturally full open bolls from the middle of each row and weighted the seed–cotton yield of each sample. After ginning, the ﬁber yield was weighted, and the lint percentage was calculated based on the ratio of the lint-to-seed–cotton weight. Then, we used R 4.1.2 software to carry out the descriptive statistical analysis, the correlation analysis, the best linear unbiased prediction (BLUP) of each accession, and H2 calculation of lint percentage [65,66]. 3.2. SNP Genotyping and Population Structure Assessment Detailed descriptions of the SNP genotyping processes were published previously [30]. In brief, high-quality genomic DNA of each accession was extracted from the young leaf tissue and was diluted to 50 ng/µL. All the 189 accessions were genotyped by the CottonSNP80K chip. The SNP data were analyzed according to GenomeStudio Geno- typing software (v1.9.4, Illumina, San Diego, CA, USA) and then ﬁltered with a minor allele frequency (MAF) < 0.05, integrity < 50%, or call rate < 90%. Finally, a total of 51,268 high-quality SNPs were obtained and were used for further analysis, including pop- ulation structure, the kinship coefﬁcient of every pair of accessions, linkage disequilibrium (LD) analysis, and genome-wide association mapping. 3.3. Genome-Wide Association Study of Lint Percentage Genome-wide association study for lint percentage was conducted using GAPIT 3.6.0 [67] and Tassel 5.0 [68] software. For GAPIT 3.6.0 software, we selected seven models, including GLM, SUPER, MLM, Fast-LMM, FarmCPU, BLINK, and MLMM; the PCA.total was set as two. For Tassel 5.0 software, we used GLM and MLM models. The signiﬁcant threshold for the marker–trait associations was set as p = 1/n (where n = 51,268, the number of SNPs totally used for GWAS, −log10 p = 4.71) [18]. To validate the stability of the associated SNPs, we investigated markers/QTL for lint percentage from previously publications in upland cotton. Based on the TM-1 reference genome [36], the physical locations of markers were obtained by BLASTN [38] and then were compared with the physical location of the SNPs detected in this study. For the SNPs detected in two or more environments, the phenotypic values of every haplotype/allele were analyzed by the two-tailed t-tests using SAS 9.4 software (SAS Institute Inc., Cary, NC, USA). 3.4. Prediction and Identiﬁcation of Related Candidate Gene Based on gene annotation and gene expression pattern of the genes in the G. hirsutum acc. TM-1 genome [36], we predicted the candidate genes within an LD (550 Kb) up- and downstream of signiﬁcantly associated SNPs detected in more than one environment. Firstly, we obtained the raw RNA-seq data of TM-1 tissues (including ovule of −3, −1, 0, 1, and 3 days post-anthesis (dpa); and ﬁber of 5, 10, and 20 dpa) from NCBI Int. J. Mol. Sci. 2023, 24, 10404 13 of 17 Sequence Read Archive (accession no. PRJNA248163). Then, we carried out the gene expression through TopHat 2.1.1 and Cufﬂinks 2.2.1 software [69] and extracted the nor- malized FPKM values as the gene expression levels. Secondly, we ﬁltered the genes with FPKM < 2 in ovule or in ﬁber, and deleted the genes with gene expression changed within twofold. Thirdly, we picked up the homologous genes in Arabidopsis, searched their anno- tation from the database of the Arabidopsis information resource, and removed the genes without annotation. According to the annotation, the candidate genes of lint percentage were explored. The expression of the candidate genes was further analyzed by qRT-PCR. Using EASYspin plus plant RNA mini kit (RN38, Aidlab, Beijing, China),we extracted high- quality total RNA from TM-1 tissues, including ovules at −2, −1, 0, 1, and 3 dpa and ﬁbers at 5, 10, and 20 dpa, and then reverse-transcribed into cDNA following the protocol guidelines of GoScript™ Reverse Transcription System (A5001, Promega, Madison, WI, USA) with a 20 µL RT reaction: 5 µL GoScriptTM 5* Reaction buffer, 5 µL 25 mM MgCl2, 2 µL PCR Nucleotide Mix, 1.0 µL GoScriptTM Reverse Transcriptase, 0.5 µL Recombinant RNasin Ribonuclease inhibitor, 0.5 µL 500 µg/mg Oligo(dT)15Primer, 0.5 µL Random Primers, and 1 µg total RNAs and nuclease-free water. After pre-denaturation at 70 ◦C for 10 min, RNAs turned to the ice, then the RT reaction was performed following the temperature program: 5 min at 25 ◦C, 60 min at 42 ◦C, 15 min at 70 ◦C, and was kept at −20 ◦C. qRT-PCR ampliﬁcation for these candidate genes was implemented using Power SYBR Green PCR Master Mix (4309155, Applied Biosystems, Foster City, CA, USA) on a QuantStudioTM 6 Flex 384-well system (Applied Biosystems™, Foster City, CA, USA). Speciﬁcally, the qRT- PCR was performed in a 20 µL reaction, including 1 µL diluted cDNA, 10 µL SyberGreen PCR Master Mix, 2 µL primers, and nuclease-free water. The temperature program was 10 min at 95 ◦C, 45 cycles of 15 s at 95 ◦C, and 60 s at 60 ◦C. The G. hirsutum actin gene Ghactin7 (LOC107959437) was the internal reference [70], its primer and the speciﬁc primers of target genes were listed in Table S12. The expression pattern of candidate genes were analyzed using the comparative 2−∆∆Ct method [71] as follows: −∆∆Ct = [(CT gene of interest − CT internal control) sample A − (CT gene of interest − CT internal control) sample B]. For the ovule tissues, one biological replicate of ovules at −2 dpa was sample B; while, for the ﬁber tissues, one biological replicate of ﬁbers at 5 dpa was sample B. Finally, the heatmaps of the candidate gene expression patterns for lint percentage were produced with Mev 4.9 software [72]. 3.5. The Protein–Protein Interaction Analysis, the Cis-Elements, and the Predicted miRNA Targeting the Candidate Genes We imported the gene IDs of homologous candidate genes in Arabidopsis to the STRING database [73] and obtained the interaction network information of the candidate genes in Arabidopsis. Then, the interaction network was visualized using Cytoscape 3.7.2 [74]. Finally, the Arabidopsis interaction protein IDs were changed to G. hirsutum IDs for the prediction of the interaction proteins in G. hirsutum. To predict the possible cis-acting elements of the crucial candidate genes, the 2000 bp upstream region of the candidate genes, considered as the promoter region, was extracted and delivered to PlantCARE [75]. The statistics of predicted cis-acting elements were compiled by category. Furthermore, the miRNA database of the G. hirsutum (v1.1) was downloaded from the plant microRNA database [76] and was delivered to the psRNATarget to investigate the potential miRNAs of the candidate genes [77]. 4. Conclusions A genome-wide association study was conducted, using 188 accessions of G. hirsutum races and TM-1. Forty-ﬁve SNPs were identiﬁed at least in two environments or/and at least with two models, which were located near 584 markers/QTL related to lint percentage identiﬁed in previous publications. Eleven SNPs were detected as signiﬁcantly associated Int. J. Mol. Sci. 2023, 24, 10404 14 of 17 with lint percentage at least in two environments, and their 550 Kb (a linkage disequi- librium) up- and downstream region contained 335 genes. Based on RNA sequencing, qRT-PCR, gene annotations, protein–protein interaction networks, the cis-elements of the promotor region, and the related miRNA predictions, Gh_D12G0934 and Gh_A08G0526 were determined as crucial candidate genes for ﬁber initiation and elongation, respec- tively. These SNPs and candidate genes identiﬁed loci/genes for analyzing the genetic architectures of lint percentages in G. hirsutum. Supplementary Materials: The supporting information can be downloaded at: https://www.mdpi. com/article/10.3390/ijms241210404/s1. Author Contributions: Conceptualization, K.W., Q.W., F.L. and Z.Z.; software, Y.W. (Yuanyuan Wang), X.G., X.C., Y.X., R.S. and T.Q.; validation, Y.W. (Yuanyuan Wang), X.G., R.S., T.Q., P.Z. and Z.W.; investigation, X.C., Y.X., Y.H., Y.W. (Yuhong Wang), F.L. and Z.Z.; resources, K.W., F.L., Z.Z. and X.C.; writing—original draft preparation, Y.W. (Yuanyuan Wang) and X.G.; writing—review and editing, Y.W. (Yuanyuan Wang), X.G., M.J.U., K.W., Q.W., F.L. and Z.Z.; visualization, X.G. and Y.W. (Yuanyuan Wang). All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the National Natural Science Foundation of China (32201763, 32171994), the Science and Technology Development Project of Henan Province (212102110054, 222102110146), the Key Scientific Research Projects of Higher Education of Henan Province (22A210014), the Project of Sanya Yazhou Bay Science and Technology City (SCKJ-JYRC-2022-88), and the State Key Laboratory of Cotton Biology Open Fund (CB2021A02). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors have no conﬂict of interest to declare. References 1. Baytar, A.A.; Peynircio ˘glu, C.; Sezener, V.; Frary, A.; Do ˘ganlar, S. Molecular mapping of QTLs for ﬁber quality traits in Gossypium hirsutum multi-parent recombinant inbred lines. Euphytica 2021, 217, 181. [CrossRef] 2. Mehboob-ur-Rahman; Shaheen, T.; Tabbasam, N.; Iqbal, M.A.; Ashraf, M.; Zafar, Y.; Paterson, A.H. Cotton genetic resources. A 7. 5. 3. 4. 6. review. Agron. Sustain. Dev. 2012, 32, 419–432. Imran, M.; Shakeel, A.; Azhar, F.M.; Farooq, J.; Saleem, M.F.; Saeed, A.; Nazeer, W.; Riaz, M.; Naeem, M.; Javaid, A. Combining ability analysis for within-boll yield components in upland cotton (Gossypium hirsutum L.). Genet. Mol. Res. 2012, 11, 2790–2800. [CrossRef] Su, J.; Wang, C.; Ma, Q.; Zhang, A.; Shi, C.; Liu, J.; Zhang, X.; Yang, D.; Ma, X. An RTM-GWAS procedure reveals the QTL alleles and candidate genes for three yield-related traits in upland cotton. BMC Plant Biol. 2020, 20, 416. [CrossRef] Song, C.; Li, W.; Pei, X.; Liu, Y.; Ren, Z.; He, K.; Zhang, F.; Sun, K.; Zhou, X.; Ma, X.; et al. Dissection of the genetic variation and candidate genes of lint percentage by a genome-wide association study in upland cotton. Theor. Appl. Genet. 2019, 132, 1991–2002. [CrossRef] Allen, E.; Xie, Z.; Gustafson, A.M.; Carrington, J.C. microRNA-directed phasing during trans-acting siRNA biogenesis in plants. Cell 2005, 121, 207–221. [CrossRef] [PubMed] Abdurakhmonov, I.Y.; Buriev, Z.T.; Saha, S.; Pepper, A.E.; Musaev, J.A.; Almatov, A.; Shermatov, S.E.; Kushanov, F.N.; Mavlonov, G.T.; Reddy, U.K.; et al. Microsatellite markers associated with lint percentage trait in cotton, Gossypium hirsutum. Euphytica 2007, 156, 141–156. [CrossRef] Sun, Z.; Wang, X.; Liu, Z.; Gu, Q.; Zhang, Y.; Li, Z.; Ke, H.; Yang, J.; Wu, J.; Wu, L.; et al. A genome-wide association study uncovers novel genomic regions and candidate genes of yield-related traits in upland cotton. Theor. Appl. Genet. 2018, 131, 2413–2425. [CrossRef] Endrizzi, J.E.; Turcotte, E.L.; Kohel, R.J. Genetics, Cytology, and evolution of Gossypium. In Advances in Genetics; Caspari, E.W., Scandalios, J.G., Eds.; Academic Press: Cambridge, MA, USA, 1985; Volume 23, pp. 271–375. Said, J.I.; Song, M.; Wang, H.; Lin, Z.; Zhang, X.; Fang, D.D.; Zhang, J. A comparative meta-analysis of QTL between intraspeciﬁc Gossypium hirsutum and interspeciﬁc G. hirsutum × G. barbadense populations. Mol. Genet. Genom. 2015, 290, 1003–1025. [CrossRef] 11. Diouf, L.; Magwanga, R.O.; Gong, W.; He, S.; Pan, Z.; Jia, Y.H.; Kirungu, J.N.; Du, X. QTL Mapping of ﬁber quality and yield-related traits in an intra-speciﬁc Upland cotton using genotype by sequencing (GBS). Int. J. Mol. Sci. 2018, 19, 441. [CrossRef] 10. 8. 9. Int. J. Mol. Sci. 2023, 24, 10404 15 of 17 12. Abdelraheem, A.; Fang, D.D.; Zhang, J.F. Quantitative trait locus mapping of drought and salt tolerance in an introgressed recombinant inbred line population of Upland cotton under the greenhouse and ﬁeld conditions. Euphytica 2018, 214, 8. [CrossRef] 13. Zhu, G.; Gao, W.; Song, X.; Sun, F.; Hou, S.; Liu, N.; Huang, Y.; Zhang, D.; Ni, Z.; Chen, Q.; et al. Genome-wide association reveals genetic variation of lint yield components under salty ﬁeld conditions in cotton (Gossypium hirsutum L.). BMC Plant Biol. 2020, 20, 23. [CrossRef] 14. Cavanagh, C.; Morell, M.; Mackay, I.; Powell, W. From mutations to MAGIC: Resources for gene discovery, validation and delivery in crop plants. Curr. Opin. Plant Biol. 2008, 11, 215–221. [CrossRef] [PubMed] 16. 15. Nie, X.; Huang, C.; You, C.; Li, W.; Zhao, W.; Shen, C.; Zhang, B.; Wang, H.; Yan, Z.; Dai, B.; et al. Genome-wide SSR-based association mapping for ﬁber quality in nation-wide upland cotton inbreed cultivars in China. BMC Genom. 2016, 17, 352. [CrossRef] [PubMed] Fang, L.; Wang, Q.; Hu, Y.; Jia, Y.; Chen, J.; Liu, B.; Zhang, Z.; Guan, X.; Chen, S.; Zhou, B.; et al. Genomic analyses in cotton identify signatures of selection and loci associated with ﬁber quality and yield traits. Nat. Genet. 2017, 49, 1089–1098. [CrossRef] 17. Huang, C.; Nie, X.; Shen, C.; You, C.; Li, W.; Zhao, W.; Zhang, X.; Lin, Z. Population structure and genetic basis of the agronomic traits of upland cotton in China revealed by a genome-wide association study using high-density SNPs. Plant Biotechnol. J. 2017, 15, 1374–1386. [CrossRef] Sun, Z.; Wang, X.; Liu, Z.; Gu, Q.; Zhang, Y.; Li, Z.; Ke, H.; Yang, J.; Wu, J.; Wu, L.; et al. Genome-wide association study J. 2017, discovered genetic variation and candidate genes of ﬁbre quality traits in Gossypium hirsutum L. Plant Biotechnol. 15, 982–996. [CrossRef] 18. 19. Ma, Z.; He, S.; Wang, X.; Sun, J.; Zhang, Y.; Zhang, G.; Wu, L.; Li, Z.; Liu, Z.; Sun, G.; et al. Resequencing a core collection of upland cotton identiﬁes genomic variation and loci inﬂuencing ﬁber quality and yield. Nat. Genet. 2018, 50, 803–813. [CrossRef] Sarfraz, Z.; Iqbal, M.S.; Geng, X.; Iqbal, M.S.; Nazir, M.F.; Ahmed, H.; He, S.; Jia, Y.; Pan, Z.; Sun, G.; et al. GWAS mediated elucidation of heterosis for metric traits in cotton (Gossypium hirsutum L.) across multiple environments. Front. Plant Sci. 2021, 12, 565552. [CrossRef] 20. 21. Tyagi, P.; Gore, M.A.; Bowman, D.T.; Campbell, B.T.; Udall, J.A.; Kuraparthy, V. Genetic diversity and population structure in the 22. US Upland cotton (Gossypium hirsutum L.). Theor. Appl. Genet. 2014, 127, 283–295. [CrossRef] Feng, L.; Zhang, S.; Xing, L.; Yang, B.; Gao, X.; Xie, X.; Zhou, B. QTL analysis for yield and ﬁbre quality traits using three sets of introgression lines developed from three Gossypium hirsutum race stocks. Mol. Genet. Genom. 2019, 294, 789–810. [CrossRef] 23. Zhang, S.; Wang, T.; Liu, Q.; Gao, X.; Zhu, X.; Zhang, T.; Zhou, B. Quantitative trait locus analysis of boll-related traits in an intraspeciﬁc population of Gossypium hirsutum. Euphytica 2015, 203, 121–144. [CrossRef] 24. Liu, D.; Liu, F.; Shan, X.; Zhang, J.; Tang, S.; Fang, X.; Liu, X.; Wang, W.; Tan, Z.; Teng, Z.; et al. Construction of a high-density genetic map and lint percentage and cottonseed nutrient trait QTL identiﬁcation in upland cotton (Gossypium hirsutum L.). Mol. Genet. Genom. 2015, 290, 1683–1700. [CrossRef] [PubMed] 25. Liu, D.; Liu, X.; Su, Y.; Zhang, X.; Guo, K.; Teng, Z.; Zhang, J.; Liu, D.; Zhang, Z. Genetic mapping and identiﬁcation of Lgf loci controlling green fuzz in Upland cotton (Gossypium hirsutum L.). Crop J. 2021, 9, 777–784. [CrossRef] 26. Zhang, S.; Feng, L.; Xing, L.; Yang, B.; Gao, X.; Zhu, X.; Zhang, T.; Zhou, B. New QTLs for lint percentage and boll weight mined in introgression lines from two feral landraces into Gossypium hirsutum acc TM-1. Plant Breed. 2016, 135, 90–101. [CrossRef] Feng, L.; Zhou, C.; Su, Q.; Xu, M.; Yue, H.; Zhang, S.; Zhou, B. Fine-mapping and candidate gene analysis of qFS-Chr. D02, a QTL for ﬁbre strength introgressed from a semi-wild cotton into Gossypium hirsutum. Plant Sci. 2020, 297, 110524. [CrossRef] 27. 28. Liu, X.; Yang, L.; Wang, J.; Wang, Y.; Guo, Z.; Li, Q.; Yang, J.; Wu, Y.; Chen, L.; Teng, Z.; et al. Analyzing quantitative trait Loci for ﬁber quality and yield-related traits from a recombinant inbred line population with Gossypium hirsutum race palmeri as one parent. Front. Plant Sci. 2021, 12, 817748. [CrossRef] 29. Wang, Y.Y.; Zhou, Z.L.; Wang, X.X.; Cai, X.Y.; Li, X.N.; Wang, C.Y.; Wang, Y.H.; Fang, L.; Wang, K.B. Genome-wide association mapping of glyphosate-resistance in Gossypium hirsutum races. Euphytica 2016, 209, 209–221. [CrossRef] 30. Guo, X.; Wang, Y.; Hou, Y.; Zhou, Z.; Sun, R.; Qin, T.; Wang, K.; Liu, F.; Wang, Y.; Huang, Z.; et al. Genome-wide dissection of the genetic basis for drought tolerance in Gossypium hirsutum L. races. Front. Plant Sci. 2022, 13, 876095. [CrossRef] 31. Xu, Y.; Magwanga, R.O.; Yang, X.; Jin, D.; Cai, X.; Hou, Y.; Wei, Y.; Zhou, Z.; Wang, K.; Liu, F. Genetic regulatory networks for salt-alkali stress in Gossypium hirsutum with differing morphological characteristics. BMC Genom. 2020, 21, 15. [CrossRef] 32. Xing, H.; Yuan, Y.; Zhang, H.; Wang, L.; Mao, L.; Tao, J.; Wang, X.; Feng, W.; Wang, H.; Wang, Q.; et al. Multi-environments and multi-models association mapping identiﬁed candidate genes of lint percentage and seed index in Gossypium hirsutum L. Mol. Breed. 2019, 39, 149. [CrossRef] 33. Qin, H.; Chen, M.; Yi, X.; Bie, S.; Zhang, C.; Zhang, Y.; Lan, J.; Meng, Y.; Yuan, Y.; Jiao, C. Identiﬁcation of associated SSR markers for yield component and ﬁber quality traits based on frame map and Upland cotton collections. PLoS ONE 2015, 10, e0118073. [CrossRef] [PubMed] 34. Niu, H.; Ge, Q.; Shang, H.; Yuan, Y. Inheritance, QTLs, and candidate genes of lint percentage in Upland cotton. Front. Genet. 35. 2022, 13, 855574. [CrossRef] [PubMed] Said, J.I.; Lin, Z.; Zhang, X.; Song, M.; Zhang, J. A comprehensive meta QTL analysis for ﬁber quality, yield, yield related and morphological traits, drought tolerance, and disease resistance in tetraploid cotton. BMC Genom. 2013, 14, 776. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10404 16 of 17 36. Zhang, T.; Hu, Y.; Jiang, W.; Fang, L.; Guan, X.; Chen, J.; Zhang, J.; Saski, C.A.; Schefﬂer, B.E.; Stelly, D.M.; et al. Sequencing of allotetraploid cotton (Gossypium hirsutum L. acc. TM-1) provides a resource for ﬁber improvement. Nat. Biotechnol. 2015, 33, 531–537. [CrossRef] 37. Gallagher, M.D.; Chen-Plotkin, A.S. The post-GWAS era: From association to function. Am. J. Hum. Genet. 2018, 102, 717–730. [CrossRef] 38. Altschul, S.F.; Gish, W.; Miller, W.; Myers, E.W.; Lipman, D.J. Basic local alignment search tool. J. Mol. Biol. 1990, 215, 403–410. [CrossRef] 39. Wang, C.; Zhang, T.; Guo, W. The im mutant gene negatively affects many aspects of ﬁber quality traits and lint percentage in cotton. Crop Sci. 2013, 53, 27–37. [CrossRef] 40. Yang, P.; Sun, X.; Liu, X.; Wang, W.; Hao, Y.; Chen, L.; Liu, J.; He, H.; Zhang, T.; Bao, W.; et al. Identiﬁcation of candidate genes for lint percentage and ﬁber quality through QTL mapping and transcriptome analysis in an allotetraploid interspeciﬁc cotton CSSLs population. Front. Plant Sci. 2022, 13, 882051. [CrossRef] 41. Wang, H.; Jia, X.; Kang, M.; Li, W.; Fu, X.; Ma, L.; Lu, J.; Wei, H.; Yu, S. QTL mapping and candidate gene identiﬁcation of lint 42. 43. percentage based on a recombinant inbred line population of upland cotton. Euphytica 2021, 217, 102. [CrossRef] Su, J.; Fan, S.; Li, L.; Wei, H.; Wang, C.; Wang, H.; Song, M.; Zhang, C.; Gu, L.; Zhao, S.; et al. Detection of favorable QTL alleles and candidate genes for lint percentage by GWAS in Chinese Upland cotton. Front. Plant Sci. 2016, 7, 1576. [CrossRef] Su, J.; Wang, C.; Hao, F.; Ma, Q.; Wang, J.; Li, J.; Ning, X. Genetic detection of lint percentage applying single-locus and multi-locus genome-wide association studies in Chinese early-maturity Upland cotton. Front. Plant Sci. 2019, 10, 964. [CrossRef] 44. Zhu, L.; Andres, R.J.; Zhang, K.; Kuraparthy, V. High-density linkage map construction and QTL analysis of ﬁber quality and lint percentage in tetraploid cotton. Crop Sci. 2021, 61, 3340–3360. [CrossRef] 45. Wang, H.; Zhao, Q.; Chen, F.; Wang, M.; Dixon, R.A. NAC domain function and transcriptional control of a secondary cell wall 46. master switch. Plant J. 2011, 68, 1104–1114. [CrossRef] Sun, W.; Gao, Z.; Wang, J.; Huang, Y.; Chen, Y.; Li, J.; Lv, M.; Wang, J.; Luo, M.; Zuo, K. Cotton ﬁber elongation requires the transcription factor GhMYB212 to regulate sucrose transportation into expanding ﬁbers. New Phytol. 2019, 222, 864–881. [CrossRef] 47. Walford, S.A.; Wu, Y.; Llewellyn, D.J.; Dennis, E.S. GhMYB25-like: A key factor in early cotton ﬁbre development. Plant J. 2011, 65, 785–797. [CrossRef] 48. Machado, A.; Wu, Y.; Yang, Y.; Llewellyn, D.J.; Dennis, E.S. The MYB transcription factor GhMYB25 regulates early ﬁbre and trichome development. Plant J. 2009, 59, 52–62. [CrossRef] [PubMed] 49. Huang, J.; Chen, F.; Guo, Y.; Gan, X.; Yang, M.; Zeng, W.; Persson, S.; Li, J.; Xu, W. GhMYB7 promotes secondary wall cellulose deposition in cotton ﬁbres by regulating GhCesA gene expression through three distinct cis-elements. New Phytol. 2021, 232, 1718–1737. [CrossRef] [PubMed] 50. Zhang, Q.; Luo, F.; Zhong, Y.; He, J.; Li, L. Modulation of NAC transcription factor NST1 activity by XYLEM NAC DOMAIN1 regulates secondary cell wall formation in Arabidopsis. J. Exp. Bot. 2020, 71, 1449–1458. [CrossRef] [PubMed] 51. Zhang, J.; Huang, G.Q.; Zou, D.; Yan, J.Q.; Li, Y.; Hu, S.; Li, X.B. The cotton (Gossypium hirsutum) NAC transcription factor (FSN1) as a positive regulator participates in controlling secondary cell wall biosynthesis and modiﬁcation of ﬁbers. New Phytol. 2018, 217, 625–640. [CrossRef] 52. Zhong, R.; Lee, C.; Ye, Z.H. Global analysis of direct targets of secondary wall NAC master switches in Arabidopsis. Mol. Plant 2010, 3, 1087–1103. [CrossRef] [PubMed] 53. Yamaguchi, M.; Mitsuda, N.; Ohtani, M.; Ohme-Takagi, M.; Kato, K.; Demura, T. VASCULAR-RELATED NAC-DOMAIN7 directly regulates the expression of a broad range of genes for xylem vessel formation. Plant J. 2011, 66, 579–590. [CrossRef] 54. Ohashi-Ito, K.; Iwamoto, K.; Fukuda, H. LOB DOMAIN-CONTAINING PROTEIN 15 positively regulates expression of VND7, a master regulator of tracheary elements. Plant Cell Physiol. 2018, 59, 989–996. [CrossRef] [PubMed] 55. McCarthy, R.L.; Zhong, R.; Ye, Z.H. MYB83 is a direct target of SND1 and acts redundantly with MYB46 in the regulation of secondary cell wall biosynthesis in Arabidopsis. Plant Cell Physiol. 2009, 50, 1950–1964. [CrossRef] [PubMed] 56. Wang, N.N.; Li, Y.; Chen, Y.H.; Lu, R.; Zhou, L.; Wang, Y.; Zheng, Y.; Li, X.B. Phosphorylation of WRKY16 by MPK3-1 is essential for its transcriptional activity during ﬁber initiation and elongation in cotton (Gossypium hirsutum). Plant Cell 2021, 33, 2736–2752. [CrossRef] [PubMed] 57. Kandasamy, M.K.; McKinney, E.C.; Meagher, R.B. A single vegetative actin isovariant overexpressed under the control of multiple regulatory sequences is sufﬁcient for normal Arabidopsis development. Plant Cell 2009, 21, 701–718. [CrossRef] 58. Gilliland, L.U.; Pawloski, L.C.; Kandasamy, M.K.; Meagher, R.B. Arabidopsis actin gene ACT7 plays an essential role in 59. germination and root growth. Plant J. 2003, 33, 319–328. [CrossRef] Ji, J.; Strable, J.; Shimizu, R.; Koenig, D.; Sinha, N.; Scanlon, M.J. WOX4 promotes procambial development. Plant Physiol. 2010, 152, 1346–1356. [CrossRef] 60. Hu, X.; Xu, L. Transcription factors WOX11/12 directly activate WOX5/7 to promote root primordia initiation and organogenesis. 61. Plant Physiol. 2016, 172, 2363–2373. [CrossRef] Sun, X.; Feng, Z.; Meng, L. Ectopic expression of the Arabidopsis ASYMMETRIC LEAVES2-LIKE5 (ASL5) gene in cockscomb (Celosia cristata) generates vascular-pattern modiﬁcations in lateral organs. Plant Cell Tissue Organ Cult. 2012, 110, 163–169. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10404 17 of 17 62. Engstrom, E.M.; Andersen, C.M.; Gumulak-Smith, J.; Hu, J.; Orlova, E.; Sozzani, R.; Bowman, J.L. Arabidopsis homologs of the petunia hairy meristem gene are required for maintenance of shoot and root indeterminacy. Plant Physiol. 2011, 155, 735–750. [CrossRef] [PubMed] Sun, R.; Li, C.; Zhang, J.; Li, F.; Ma, L.; Tan, Y.; Wang, Q.; Zhang, B. Differential expression of microRNAs during ﬁber development between fuzzless-lintless mutant and its wild-type allotetraploid cotton. Sci. Rep. 2017, 7, 3. [CrossRef] [PubMed] 63. 64. Liu, N.; Tu, L.; Tang, W.; Gao, W.; Lindsey, K.; Zhang, X. Small RNA and degradome proﬁling reveals a role for miRNAs and their targets in the developing ﬁbers of Gossypium barbadense. Plant J. 2014, 80, 331–344. [CrossRef] [PubMed] 65. Bates, D.; Mächler, M.; Bolker, B.; Walker, S. Fitting linear mixed-effects models using lme4. J. Stat. Softw. 2015, 67, 1–48. [CrossRef] 66. Wickham, H. Ggplot2: Elegant Graphics for Data Analysis, 2nd ed.; Spring: New York, NY, USA, 2018. 67. Lipka, A.E.; Tian, F.; Wang, Q.; Peiffer, J.; Li, M.; Bradbury, P.J.; Gore, M.A.; Buckler, E.S.; Zhang, Z. GAPIT: Genome association and prediction integrated tool. Bioinformatics 2012, 28, 2397–2399. [CrossRef] 68. Bradbury, P.J.; Zhang, Z.; Kroon, D.E.; Casstevens, T.M.; Ramdoss, Y.; Buckler, E.S. TASSEL: Software for association mapping of complex traits in diverse samples. Bioinformatics 2007, 23, 2633–2635. [CrossRef] 69. Trapnell, C.; Roberts, A.; Goff, L.; Pertea, G.; Kim, D.; Kelley, D.R.; Pimentel, H.; Salzberg, S.L.; Rinn, J.L.; Pachter, L. Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufﬂinks. Nat. Protoc. 2012, 7, 562–578. [CrossRef] 72. 71. 70. Artico, S.; Nardeli, S.M.; Brilhante, O.; Grossi-de-Sa, M.F.; Alves-Ferreira, M. Identiﬁcation and evaluation of new reference genes in Gossypium hirsutum for accurate normalization of real-time quantitative RT-PCR data. BMC Plant Biol. 2010, 10, 49. [CrossRef] Schmittgen, T.D.; Livak, K.J. Analyzing real-time PCR data by the comparative CT method. Nat. Protoc. 2008, 3, 1101–1108. [CrossRef] Saeed, A.I.; Bhagabati, N.K.; Braisted, J.C.; Liang, W.; Sharov, V.; Howe, E.A.; Li, J.; Thiagarajan, M.; White, J.A.; Quackenbush, J. TM4 microarray software suite. In Methods in Enzymology; Academic Press: Cambridge, MA, USA, 2006; Volume 411, pp. 134–193. Szklarczyk, D.; Gable, A.L.; Lyon, D.; Junge, A.; Wyder, S.; Huerta-Cepas, J.; Simonovic, M.; Doncheva, N.T.; Morris, J.H.; Bork, P.; et al. STRING v11: Protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019, 47, D607–D613. [CrossRef] Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N.S.; Wang, J.T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504. [CrossRef] [PubMed] 74. 73. 75. Lescot, M.; Dehais, P.; Thijs, G.; Marchal, K.; Moreau, Y.; Van de Peer, Y.; Rouze, P.; Rombauts, S. PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Res. 2002, 30, 325–327. [CrossRef] [PubMed] 76. Zhang, Z.; Yu, J.; Li, D.; Zhang, Z.; Liu, F.; Zhou, X.; Wang, T.; Ling, Y.; Su, Z. PMRD: Plant microRNA database. Nucleic Acids Res. 2010, 38, D806–D813. [CrossRef] [PubMed] 77. Dai, X.; Zhuang, Z.; Zhao, P.X. psRNATarget: A plant small RNA target analysis server (2017 release). Nucleic Acids Res. 2018, 46, W49–W54. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_molecules24122333	Article Development of a Method for the Quantiﬁcation of Clotrimazole and Itraconazole and Study of Their Stability in a New Microemulsion for the Treatment of Sporotrichosis Patricia Garcia Ferreira 1, Carolina Guimarães de Souza Lima 2, Letícia Lorena Noronha 1, Marcela Cristina de Moraes 2, Fernando de Carvalho da Silva 2 Débora Omena Futuro 1 and Vitor Francisco Ferreira 1,* , Alessandra Lifsitch Viçosa 3 , 1 Departamento de Tecnologia Farmacêutica, Faculdade de Farmácia, Universidade Federal Fluminense, Niterói-RJ 24241-000, Brazil; [email protected] (P.G.F.); [email protected] (L.L.N.); [email protected]ﬀ.br (D.O.F.) 2 Departamento de Química Orgânica, Instituto de Química, Universidade Federal Fluminense, Niterói-RJ 24210-141, Brazil; [email protected] (C.G.d.S.L.); [email protected]ﬀ.br (M.C.d.M.); [email protected]ﬀ.br (F.d.C.d.S.) Fundação Oswaldo Cruz (FIOCRUZ), Farmanguinhos-Manguinhos, Avenida Sinzenando Nabuco 100, Rio de Janeiro-RJ 21045-900, Brazil; alessandra.vicosa@far.ﬁocruz.br 3 * Correspondence: [email protected]ﬀ.br; Tel.: +55-21-998578148 Academic Editors: Clinio Locatelli, Marcello Locatelli and Dora Melucci Received: 6 June 2019; Accepted: 20 June 2019; Published: 25 June 2019 ® Abstract: Sporotrichosis occurs worldwide and is caused by the fungus Sporothrix brasiliensis. This agent has a high zoonotic potential and is transmitted mainly by bites and scratches from infected felines. A new association between the drugs clotrimazole and itraconazole is shown to be eﬀective against S. brasiliensis yeasts. This association was formulated as a microemulsion containing benzyl alcohol as oil, Tween 60 and propylene glycol as surfactant and cosurfactant, respectively, and water. Initially, the compatibility between clotrimazole and itraconazole was studied using diﬀerential scanning calorimetry (DSC), thermogravimetric analysis (TG), Fourier transform infrared spectroscopy (FTIR), and X-ray powder diﬀraction (PXRD). Additionally, a simple and eﬃcient analytical HPLC method was developed to simultaneously determine the concentration of clotrimazole and itraconazole in the novel microemulsion. The developed method proved to be eﬃcient, robust, and reproducible for both components of the microemulsion. We also performed an accelerated stability study of this formulation, and the developed analytical method was applied to monitor the content of active ingredients. Interestingly, these investigations led to the detection of a known clotrimazole degradation product whose structure was conﬁrmed using NMR and HRMS, as well as a possible interaction between itraconazole and benzyl alcohol. Keywords: validation; sporotrichosis pre-development process; clotrimazole; itraconazole; stability; method 1. Introduction Sporotrichosis is a subcutaneous infectious disease with subacute to chronic evolution and with a worldwide distribution. The etiologic agent of sporotrichosis is Sporothrix schenckii, which is a thermo-dimorphic fungus that lives saprophytically in nature and is pathogenic to humans and animals [1,2]. The occurrence of sporotrichosis in animals, especially cats, as well as its transmission to humans has been reported in several countries [3]. In this context, the Brazilian state of Rio de Janeiro Molecules 2019, 24, 2333; doi:10.3390/molecules24122333 www.mdpi.com/journal/molecules molecules(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Molecules 2019, 24, 2333 2 of 15 is an epidemic area for this disease and the ﬁrst one associated with zoonotic transmission related to sick felines by Sporothrix brasiliensis, the most virulent species from the S. schenckii complex [4]. The treatment of both feline and human sporotrichosis is based on the use of itraconazole 1, which contains the 1,2,4-triazole scaﬀold in its structure and inhibits the synthesis of sterol, a vital component of the fungus cell membrane [5,6]. Clotrimazole 2, on the other hand, is an imidazole derivative with antifungal activity that is only indicated for topical use due to its toxicity (Figure 1). Similarly to itraconazole, clotrimazole is a synthetic antifungal and its mechanism of action involves the inhibition of sterol biosynthesis [7]. In this sense, Gagini et al. [8] reported the eﬀectiveness of the combination of itraconazole with clotrimazole against S. brasiliensis yeasts (the infective form) from feline and human sporotrichosis isolates, suggesting that clotrimazole by itself or in combination with itraconazole is potentially a new option for the treatment of sporotrichosis. Figure 1. Chemical structures of clotrimazole and itraconazole. Accordingly, the development of new pharmaceutical technologies for the use of clotrimazole and itraconazole associations is highly desirable in order to increase their eﬃciency in therapy, decrease adverse eﬀects and provide, especially for felines, alternative treatments. Moreover, the use of a combination antifungal therapy is a promising approach to avoid resistance [9]. Allied to all the mentioned features, the association of known drugs is highly advantageous for the pharmaceutical industry to ﬁnd innovations for the market, since they can reformulate their products in a more economically advantageous way when compared to the development of new drugs. In addition, the association of drugs already in use in the pharmaceutical market may increase their eﬃciency with known safety and eﬀectiveness, reintroducing forgotten and/or discarded ones. Considering the development of new formulations, microemulsions (MEs) have attracted great interest as potential drug delivery systems, mainly due to their unique physicochemical properties such as drug solubilization and enhanced absorption properties [10,11]. MEs are a thermodynamically stable, isotropic, transparent liquid system consisting of two immiscible liquids (usually water and oil) stabilized by a ﬁlm of surfactant compounds, suitably combined with a cosurfactant [12,13]. The presence of the surfactant helps to reduce the interfacial tension, making it possible to join the oil and aqueous phases [14,15]. MEs have been proposed as an innovative formulation approach to improve solubility and eﬃcacy and reduce of the toxicity of various drugs. Therefore, when the known hydrophobicity of clotrimazole and itraconazole are taken into account, such systems could be particularly advantageous for their delivery. In light of the aforementioned concepts, this paper reports the initial research phase for the pre-development of a clotrimazole–itraconazole formulation, the ﬁrst step towards a new antifungal combination. In this sense, the development and characterization of this new pharmaceutical formulation requires the evaluation of parameters such as drug release and stability. Therefore, as a further extension of our work in the ﬁeld, we have developed a simple, sensitive, and speciﬁc HPLC method for the simultaneous quantiﬁcation of clotrimazole and itraconazole in microemulsion. Although many researchers have investigated clotrimazole and itraconazole singly or in combination with other compounds, to the best of our knowledge, no HPLC method has been developed for the simultaneous determination of both drugs simultaneously, especially in microemulsion systems [16,17]. Finally, we performed an accelerated stability study of this formulation and the developed analytical Molecules 2019, 24, x FOR PEER REVIEW 2 of 15 to humans has been reported in several countries [3]. In this context, the Brazilian state of Rio de Janeiro is an epidemic area for this disease and the first one associated with zoonotic transmission related to sick felines by Sporothrix brasiliensis, the most virulent species from the S. schenckii complex [4]. The treatment of both feline and human sporotrichosis is based on the use of itraconazole 1, which contains the 1,2,4-triazole scaffold in its structure and inhibits the synthesis of sterol, a vital component of the fungus cell membrane [5,6]. Clotrimazole 2, on the other hand, is an imidazole derivative with antifungal activity that is only indicated for topical use due to its toxicity (Figure 1). Similarly to itraconazole, clotrimazole is a synthetic antifungal and its mechanism of action involves the inhibition of sterol biosynthesis [7]. In this sense, Gagini et al. [8] reported the effectiveness of the combination of itraconazole with clotrimazole against S. brasiliensis yeasts (the infective form) from feline and human sporotrichosis isolates, suggesting that clotrimazole by itself or in combination with itraconazole is potentially a new option for the treatment of sporotrichosis. Accordingly, the development of new pharmaceutical technologies for the use of clotrimazole and itraconazole associations is highly desirable in order to increase their efficiency in therapy, decrease adverse effects and provide, especially for felines, alternative treatments. Moreover, the use of a combination antifungal therapy is a promising approach to avoid resistance [9]. Allied to all the mentioned features, the association of known drugs is highly advantageous for the pharmaceutical industry to find innovations for the market, since they can reformulate their products in a more economically advantageous way when compared to the development of new drugs. In addition, the association of drugs already in use in the pharmaceutical market may increase their efficiency with known safety and effectiveness, reintroducing forgotten and/or discarded ones. Figure 1. Chemical structures of clotrimazole and itraconazole. Considering the development of new formulations, microemulsions (MEs) have attracted great interest as potential drug delivery systems, mainly due to their unique physicochemical properties such as drug solubilization and enhanced absorption properties [10,11]. MEs are a thermodynamically stable, isotropic, transparent liquid system consisting of two immiscible liquids (usually water and oil) stabilized by a film of surfactant compounds, suitably combined with a cosurfactant [12,13]. The presence of the surfactant helps to reduce the interfacial tension, making it possible to join the oil and aqueous phases [14,15]. MEs have been proposed as an innovative formulation approach to improve solubility and efficacy and reduce of the toxicity of various drugs. Therefore, when the known hydrophobicity of clotrimazole and itraconazole are taken into account, such systems could be particularly advantageous for their delivery. In light of the aforementioned concepts, this paper reports the initial research phase for the pre-development of a clotrimazole–itraconazole formulation, the first step towards a new antifungal combination. In this sense, the development and characterization of this new pharmaceutical formulation requires the evaluation of parameters such as drug release and stability. Therefore, as a further extension of our work in the field, we have developed a simple, sensitive, and specific HPLC method for the simultaneous quantification of clotrimazole and itraconazole in microemulsion. Although many researchers have investigated clotrimazole and itraconazole singly or in combination with other compounds, to the best of our knowledge, no HPLC method has been developed for the simultaneous determination of both drugs simultaneously, especially in microemulsion systems [16,17]. Finally, we performed an accelerated stability study of this formulation and the developed Molecules 2019, 24, 2333 3 of 15 method was applied to monitor the content of active ingredients. Interestingly, these investigations led to the detection of a known clotrimazole degradation product whose structure was conﬁrmed using NMR and HRMS, as well as a possible interaction between itraconazole and benzyl alcohol. 2. Results and Discussion 2.1. Study of the Compatibility between Clotrimazole and Itraconazole We initiated our studies by analyzing the physicochemical properties of both active ingredients as well as their compatibility using diﬀerent techniques such as thermal analyses (diﬀerential scanning calorimetry (DSC) and thermogravimetric/derivative thermogravimetry (TG)/DTG) analysis), powder X-ray diﬀraction (XRD) and FTIR. Initially, we proceeded to characterize the active ingredients and their combination using thermal analyses, which oﬀer the ability to quickly screen for potential drug–drug incompatibilities. Such interactions can be of a physical or chemical nature and may aﬀect the stability and bioavailability of the ﬁnal product, compromising the therapeutic eﬃcacy and safety [18]. The TG and DTG curves of clotrimazole (Figure 2a) showed that it is thermally stable up to 340 C, C, as showed in when its thermal decomposition starts; the highest rate of weight loss occurs at 388.6 C, where a loss of 60% of the total weight is observed. As for the DTG curve, and is ﬁnished at 421.1 C, with a maximum itraconazole, its thermal decomposition starts at 200 rate at 295.3 C and a total weight loss of 87%. The TG proﬁle of the binary mixture of clotrimazole and itraconazole (1:1 ratio) showed two decomposition steps, indicating that the compounds undergo thermal degradation independently, although a small shift in the initial temperature of decomposition was observed, as expected. C and is ﬁnished at 348.4 ◦ ◦ ◦ ◦ ◦ ◦ Figure 2. Thermogravimetric (TG) and derivative thermogravimetry (DTG) curves for (a) clotrimazole, (b) itraconazole and (c) the binary mixture of clotrimazole and itraconazole (1:1). Next, the DSC technique was employed to further analyze the occurrence of events related to possible interactions between the drugs [19]. It is noteworthy that although such analyses are conducted upon heating the sample to high temperatures, which is not consistent with the process of drug production nor its administration to patients, they aﬀord important information regarding the physical properties of the sample [18]. ◦ The DSC curves of the drugs showed endothermic peaks attributed to the melting of the drugs −1) between 158.5 and 175.0 for clotrimazole. On the other hand, a single endothermic event was observed in the DSC curve of the −1), which suggests a strong binary mixture, starting at 127.7 and ﬁnishing at 137.1 (∆H = −25.35 J g interaction between clotrimazole and itraconazole (Figure 3). −1) for itraconazole and 136.8 and 153.1 C (∆H = 41.6 J g C (∆H = 31.5 J g ◦ Molecules 2019, 24, x FOR PEER REVIEW 3 of 15 analytical method was applied to monitor the content of active ingredients. Interestingly, these investigations led to the detection of a known clotrimazole degradation product whose structure was confirmed using NMR and HRMS, as well as a possible interaction between itraconazole and benzyl alcohol. 2. Results and Discussion 2.1. Study of the Compatibility between Clotrimazole and Itraconazole We initiated our studies by analyzing the physicochemical properties of both active ingredients as well as their compatibility using different techniques such as thermal analyses (differential scanning calorimetry (DSC) and thermogravimetric/derivative thermogravimetry (TG)/DTG) analysis), powder X-ray diffraction (XRD) and FTIR. Initially, we proceeded to characterize the active ingredients and their combination using thermal analyses, which offer the ability to quickly screen for potential drug–drug incompatibilities. Such interactions can be of a physical or chemical nature and may affect the stability and bioavailability of the final product, compromising the therapeutic efficacy and safety [18]. The TG and DTG curves of clotrimazole (Figure 2a) showed that it is thermally stable up to 340 °C, when its thermal decomposition starts; the highest rate of weight loss occurs at 388.6 °C, as showed in the DTG curve, and is finished at 421.1 °C, where a loss of 60% of the total weight is observed. As for itraconazole, its thermal decomposition starts at 200 °C and is finished at 348.4 °C, with a maximum rate at 295.3 °C and a total weight loss of 87%. The TG profile of the binary mixture of clotrimazole and itraconazole (1:1 ratio) showed two decomposition steps, indicating that the compounds undergo thermal degradation independently, although a small shift in the initial temperature of decomposition was observed, as expected. Figure 2. Thermogravimetric (TG) and derivative thermogravimetry (DTG) curves for (a) clotrimazole, (b) itraconazole and (c) the binary mixture of clotrimazole and itraconazole (1:1). Next, the DSC technique was employed to further analyze the occurrence of events related to possible interactions between the drugs [19]. It is noteworthy that although such analyses are conducted upon heating the sample to high temperatures, which is not consistent with the process of drug production nor its administration to patients, they afford important information regarding the physical properties of the sample [18]. The DSC curves of the drugs showed endothermic peaks attributed to the melting of the drugs between 158.5 and 175.0 °C (ΔH = 31.5 J g−1) for itraconazole and 136.8 and 153.1 °C (ΔH = 41.6 J g−1) for clotrimazole. On the other hand, a single endothermic event was observed in the DSC curve of the binary mixture, starting at 127.7 and finishing at 137.1 (ΔH = −25.35 J g−1), which suggests a strong interaction between clotrimazole and itraconazole (Figure 3). Molecules 2019, 24, 2333 4 of 15 Figure 3. Diﬀerential scanning calorimetry (DSC) proﬁle of itraconazole, clotrimazole, and the clotrimazole/itraconazole binary mixture (1:1). In order to further explore the possibility of interactions between the active ingredients, powder X-ray diﬀraction (PXRD) analyses were conducted. Interestingly, the diﬀractogram of the binary mixture (Figure 4) contained virtually all the peaks of clotrimazole and itraconazole, with no marked displacement of the peaks being observed. Furthermore, it is important to highlight that it was not possible to notice the appearance of any new peaks, which means that if there is any interaction between the drugs, it probably is not strong enough to take place in the solid state. The same observations were made in the FTIR spectra of the binary mixture, which showed the characteristic bands observed for the isolated active ingredients (For more details, see the Supplementary Materials). Figure 4. X-ray diﬀractograms of clotrimazole, itraconazole, and the binary mixture (1:1). With the characterization of the active ingredients and the binary mixture in hand, we proceeded to develop an HPLC method for their quantiﬁcation in a newly developed microemulsion for the treatment of sporotrichosis. Molecules 2019, 24, x FOR PEER REVIEW 4 of 15 80100120140160180200220Heat flow (mW/mg)Temperature (°C) Mixture (1:1) Clotrimazole Itraconazole Figure 3. Differential scanning calorimetry (DSC) profile of itraconazole, clotrimazole, and the clotrimazole/itraconazole binary mixture (1:1). In order to further explore the possibility of interactions between the active ingredients, powder X-ray diffraction (PXRD) analyses were conducted. Interestingly, the diffractogram of the binary mixture (Figure 4) contained virtually all the peaks of clotrimazole and itraconazole, with no marked displacement of the peaks being observed. Furthermore, it is important to highlight that it was not possible to notice the appearance of any new peaks, which means that if there is any interaction between the drugs, it probably is not strong enough to take place in the solid state. The same observations were made in the FTIR spectra of the binary mixture, which showed the characteristic bands observed for the isolated active ingredients (For more details, see the Supplementary Materials). With the characterization of the active ingredients and the binary mixture in hand, we proceeded to develop an HPLC method for their quantification in a newly developed microemulsion for the treatment of sporotrichosis. 10203040ICCCCCCCCCCCCCCCCCCCCIIIIIIIICCCCCIIICCCRelative intensity (a.u.)2θ (°) Binary mixture (1:1) Itraconazole (I) Clotrimazole (C)I Figure 4. X-ray diffractograms of clotrimazole, itraconazole, and the binary mixture (1:1). Molecules 2019, 24, x FOR PEER REVIEW 4 of 15 80100120140160180200220Heat flow (mW/mg)Temperature (°C) Mixture (1:1) Clotrimazole Itraconazole Figure 3. Differential scanning calorimetry (DSC) profile of itraconazole, clotrimazole, and the clotrimazole/itraconazole binary mixture (1:1). In order to further explore the possibility of interactions between the active ingredients, powder X-ray diffraction (PXRD) analyses were conducted. Interestingly, the diffractogram of the binary mixture (Figure 4) contained virtually all the peaks of clotrimazole and itraconazole, with no marked displacement of the peaks being observed. Furthermore, it is important to highlight that it was not possible to notice the appearance of any new peaks, which means that if there is any interaction between the drugs, it probably is not strong enough to take place in the solid state. The same observations were made in the FTIR spectra of the binary mixture, which showed the characteristic bands observed for the isolated active ingredients (For more details, see the Supplementary Materials). With the characterization of the active ingredients and the binary mixture in hand, we proceeded to develop an HPLC method for their quantification in a newly developed microemulsion for the treatment of sporotrichosis. 10203040ICCCCCCCCCCCCCCCCCCCCIIIIIIIICCCCCIIICCCRelative intensity (a.u.)2θ (°) Binary mixture (1:1) Itraconazole (I) Clotrimazole (C)I Figure 4. X-ray diffractograms of clotrimazole, itraconazole, and the binary mixture (1:1). Molecules 2019, 24, 2333 5 of 15 2.2. Determination of the Concentration of Clotrimazole and Itraconazole in Microemulsions Using HPLC Analyses Considering the unique properties presented by microemulsions, in the present work, benzyl alcohol was used as an oil phase, Tween 60 as a surfactant, and propylene glycol as a cosolvent in the presence of water. These components were chosen on the basis in their previously reported applications in other pharmaceutical forms available on the international market. ® In this context, HPLC-DAD (diode array detector) was selected as an analytical tool for the simultaneous quantiﬁcation of clotrimazole and itraconazole in the developed microemulsion through a rapid, simple, and isocratic method [20]. In our study, the best separation condition was achieved using a C18 analytical column with a mobile phase composed of acetonitrile and a phosphate buﬀered saline 0.05 M (pH 8.0 with ammonium hydroxide 1 M) in the ratio (v/v) 60:40, respectively, with a −1 ﬂow rate and UV detection at 190 nm. A typical chromatogram is presented in Figure 5, 1 mL min with a retention time of 9.1 min being observed for clotrimazole and 10.9 min for itraconazole. Figure 5. Chromatograms of the (a) mobile phase and (b) standard solution containing a binary mixture of itraconazole and clotrimazole. Molecules 2019, 24, x FOR PEER REVIEW 5 of 15 2.2. Determination of the Concentration of Clotrimazole and Itraconazole in Microemulsions Using HPLC Analyses Considering the unique properties presented by microemulsions, in the present work, benzyl alcohol was used as an oil phase, Tween® 60 as a surfactant, and propylene glycol as a cosolvent in the presence of water. These components were chosen on the basis in their previously reported applications in other pharmaceutical forms available on the international market. In this context, HPLC-DAD (diode array detector) was selected as an analytical tool for the simultaneous quantification of clotrimazole and itraconazole in the developed microemulsion through a rapid, simple, and isocratic method [20]. In our study, the best separation condition was achieved using a C18 analytical column with a mobile phase composed of acetonitrile and a phosphate buffered saline 0.05 M (pH 8.0 with ammonium hydroxide 1 M) in the ratio (v/v) 60:40, respectively, with a 1 mL min−1 flow rate and UV detection at 190 nm. A typical chromatogram is presented in Figure 5, with a retention time of 9.1 min being observed for clotrimazole and 10.9 min for itraconazole. (a) (b) Figure 5. Chromatograms of the (a) mobile phase and (b) standard solution containing a binary mixture of itraconazole and clotrimazole. Molecules 2019, 24, 2333 6 of 15 To evaluate the linearity of the method, calibration standards of clotrimazole (5–200 µg mL −1) −1) were analyzed. A linear relationship was established for the injected and itraconazole (5–160 µg mL concentration ranges versus the peak area for both analytes, with determination coeﬃcients greater than 0.9988 (see the calibration curves in the Supplementary Materials). The calibration curve parameters are reported in Table 1, with the linearity parameters of the method shown in Table 2. Table 1. Summary of the validation data obtained for the proposed HPLC method developed for the quantiﬁcation of clotrimazole and itraconazole in microemulsions. LOD—limit of detection; LOQ—limit of quantiﬁcation. Standard Solutions Parameters of the Method Validation Results Clotrimazole Itraconazole Linearity LOD LOQ Slope Interception Linearity LOD LOQ Slope Interception Calibration range (µg mL −1): 5–200 y = 233647.7939x − 312039.9299 (R2 = 0.9988) 0.84 µg mL 2.54 µg mL 233647.7939 ± 976.8015153 −312039.9299 ± 59416.57811 −1 −1 Calibration range (µg mL −1): 5–160 y = 89946.6896x − 79996.5373 (R2 = 0.9999) 0.86 µg mL 2.60 µg mL 89946.6896 ± 780.1420761 −79996.53731 ± 23351.48986 −1 −1 Table 2. Data related to the linearity of the developed HPLC method with its respective average, precision, and accuracy. Concentration (µg/mL) Clotrimazole Itraconazole Average (µg/mL) Accuracy (%) Precision (%) Average (µg/mL) Accuracy (%) Precision (%) 5 10 20 40 80 160 200 4.883 9.292 19.233 38.927 77.731 151.888 204.631 97.7 92.9 96.2 97.3 97.2 94.9 102.3 0.20 0.57 0.40 0.01 1.08 0.71 0.65 5.593 10.115 20.029 39.621 79.154 160.488 - 111.9 101.2 100.1 99.1 98.9 100.3 - 0.86 0.91 0.58 1.12 1.81 0.82 - The method’s selectivity was conﬁrmed by the absence of interferences at the retention times of itraconazole and clotrimazole in the microemulsion prepared without the drugs (Figure 6). The purity of the compounds was checked using PDA (photodiode array) detection. The within-assay precision (repeatability) was carried out by performing six consecutive analyses of standard solution at three diﬀerent concentrations for each drug on the same day. The samples were also analyzed on diﬀerent days to evaluate the between-assay precision (intermediate precision). The obtained values were evaluated through the dispersion of the results by calculating the standard deviation of the measurement series. The intra- and inter-day precision relative standard deviation (RSD %) was between 1.18 and 0.8 for clotrimazole and 1.48 and 0.84 for itraconazole. The recovery of the drugs was in the range of 93.8–100.9% with RSDs below 2.35% for clotrimazole and in the range of 100.5–104.3% with RSDs below 2.40% for itraconazole. The results are given in Table 3. Molecules 2019, 24, 2333 7 of 15 Figure 6. Chromatogram obtained from the injection of the microemulsion using the developed HPLC method. Table 3. Data related to the repeatability and intermediate precision of the developed HPLC method. Samples (µg mL−1) Intra-Day Precision (Repeatability) Inter-Day Precision (Intermediate Precision) Clotrimazole Concentration −1) Found (µg mL 7 15 120 6.818 14.510 116.679 Itraconazole Concentration −1) Found (µg mL 7 70 150 7.206 70.809 152.745 Accuracy (%) 97.4 ±2.25 96.7 ±1.13 97.2 ±0.27 Accuracy (%) 102.9 ± 1.33 101.2 ± 1.15 101.8 ± 0.85 Precision (%) Concentration −1) Found (µg mL 0.47 1.18 0.28 6.865 14.075 121.108 Precision (%) Concentration −1) Found (µg mL 1.48 1.16 0.84 7.305 70.374 160.98 Accuracy (%) 98.07 ± 1.17 93.83 ± 3.17 100.92 ± 4.28 Accuracy (%) 104.35 ± 1.25 100.53 ± 2.41 100.61 ± 4.9 Precision (%) 2.35 0.95 0.28 Precision (%) 1.20 2.40 1.59 No changes were observed in the drug concentrations of the stock solutions under storage conditions. Indeed, further analyses showed that the percent recovery of clotrimazole and itraconazole were, respectively, 97.3% ± 3.15 and 91.3% ± 2.71 at room temperature (25 C) and 94.2 ± 0.34 and 88.7 ± 1.63 under refrigeration (−5 C, Table 4). Moreover, the drugs were stable for at least 30 days under storage conditions, with RSDs below 8%. ◦ ◦ Table 4. Data related to the stability of the assay of the developed HPLC method. N = 2 for each day and condition. Days Accuracy (%) Precision (%) Accuracy (%) Precision (%) Clotrimazole Itraconazole 0 7 15 30 ◦ 97.3 ± 0.94 (25 105.7 ± 0.89 (25 104.9 ± 0.07 (−5 105.3 ± 1.51 (25 105.5 ± 0.39 (−5 ◦ 97.3 ± 3.15 (25 94.2 ± 0.34 (−5 C) ◦ C) C) C) C) ◦ ◦ ◦ C) ◦ C) 1.18 0.85 0.07 1.45 0.38 0.62 0.73 101.4 ± 0.62 98.6 ± 4.48 98.4 ± 7.79 101.3 ± 0.51 100.1 ± 3.71 91.3 ± 2.71 88.7 ± 1.63 0.84 4.57 7.96 0.50 3.74 3.21 3.04 In order to evaluate the robustness of the chromatographic method, assays were carried out by changing both the column brand and ratio of the mobile phase for acetonitrile 70:30 (v/v) and a phosphate buﬀered saline 0.05 M (pH 8.0 with ammonium hydroxide 1 M). The alteration of the Molecules 2019, 24, x FOR PEER REVIEW 7 of 15 Figure 6. Chromatogram obtained from the injection of the microemulsion using the developed HPLC method. Table 3. Data related to the repeatability and intermediate precision of the developed HPLC method. Samples (µg mL−1) Intra-Day Precision (Repeatability) Inter-Day Precision (Intermediate Precision) Clotrimazole Concentration Found (µg mL−1) Accuracy (%) Precision (%) Concentration Found (µg mL−1) Accuracy (%) Precision (%) 7 6.818 97.4 ±2.25 0.47 6.865 98.07 ± 1.17 2.35 15 14.510 96.7 ±1.13 1.18 14.075 93.83 ± 3.17 0.95 120 116.679 97.2 ±0.27 0.28 121.108 100.92 ± 4.28 0.28 Itraconazole Concentration Found (µg mL−1) Accuracy (%) Precision (%) Concentration Found (µg mL−1) Accuracy (%) Precision (%) 7 7.206 102.9 ± 1.33 1.48 7.305 104.35 ± 1.25 1.20 70 70.809 101.2 ± 1.15 1.16 70.374 100.53 ± 2.41 2.40 150 152.745 101.8 ± 0.85 0.84 160.98 100.61 ± 4.9 1.59 No changes were observed in the drug concentrations of the stock solutions under storage conditions. Indeed, further analyses showed that the percent recovery of clotrimazole and itraconazole were, respectively, 97.3% ± 3.15 and 91.3% ± 2.71 at room temperature (25 °C) and 94.2 ± 0.34 and 88.7 ± 1.63 under refrigeration (−5 °C, Table 4). Moreover, the drugs were stable for at least 30 days under storage conditions, with RSDs below 8%. Table 4. Data related to the stability of the assay of the developed HPLC method. N = 2 for each day and condition. Days Accuracy (%) Precision (%) Accuracy (%) Precision (%) Clotrimazole Itraconazole 0 97.3 ± 0.94 (25 °C) 1.18 101.4 ± 0.62 0.84 7 105.7 ± 0.89 (25 °C) 0.85 98.6 ± 4.48 4.57 104.9 ± 0.07 (−5 °C) 0.07 98.4 ± 7.79 7.96 15 105.3 ± 1.51 (25 °C) 1.45 101.3 ± 0.51 0.50 105.5 ± 0.39 (−5 °C) 0.38 100.1 ± 3.71 3.74 30 97.3 ± 3.15 (25 °C) 0.62 91.3 ± 2.71 3.21 94.2 ± 0.34 (−5 °C) 0.73 88.7 ± 1.63 3.04 In order to evaluate the robustness of the chromatographic method, assays were carried out by changing both the column brand and ratio of the mobile phase for acetonitrile 70:30 (v/v) and a Molecules 2019, 24, 2333 8 of 15 column brand and the mobile phase did not promote any signiﬁcant variations in the retention time of clotrimazole and itraconazole peaks; a good resolution was observed with retention times of 8 min for clotrimazole and 10.7 min for itraconazole (Figure 7). Figure 7. Chromatogram of clotrimazole and itraconazole obtained in the robustness studies. 2.3. Study of the Stability of a Novel Microemulsion Containing Clotrimazole and Itraconazole Subsequently, the developed method was used in the determination of clotrimazole and itraconazole in the newly developed microemulsion with the purpose of quantifying the drugs in the formulation, as well as in the accelerated stability study. Based on the assumption that possible interactions and incompatibilities may arise from the contact between the drugs over time, they were left to stand for three months, both under refrigeration and heating conditions, and further analyzed. The initial drug content of the microemulsion was taken as 100%, and the drug content over time was plotted (Figure 8), with all data being represented as mean ± SD (n = 3). For the samples stored at 5 C, no signiﬁcant changes were observed for both drugs when compared to the ﬁrst day. Furthermore, it is noteworthy that there was no evident interaction between clotrimazole and itraconazole at this temperature, since the peaks of both drugs were detected independently without the appearance of any additional peaks. On the other hand, when the samples that were stored at 40 C were analyzed, it was possible to notice a signiﬁcant decrease in the concentration of the drugs over time, especially for clotrimazole. Additionally, a new peak could also be observed in the chromatogram of such samples (Figure 9). ◦ ◦ Molecules 2019, 24, x FOR PEER REVIEW 8 of 15 phosphate buffered saline 0.05 M (pH 8.0 with ammonium hydroxide 1 M). The alteration of the column brand and the mobile phase did not promote any significant variations in the retention time of clotrimazole and itraconazole peaks; a good resolution was observed with retention times of 8 min for clotrimazole and 10.7 min for itraconazole (Figure 7). Figure 7. Chromatogram of clotrimazole and itraconazole obtained in the robustness studies. 2.3. Study of the Stability of a Novel Microemulsion Containing Clotrimazole and Itraconazole Subsequently, the developed method was used in the determination of clotrimazole and itraconazole in the newly developed microemulsion with the purpose of quantifying the drugs in the formulation, as well as in the accelerated stability study. Based on the assumption that possible interactions and incompatibilities may arise from the contact between the drugs over time, they were left to stand for three months, both under refrigeration and heating conditions, and further analyzed. The initial drug content of the microemulsion was taken as 100%, and the drug content over time was plotted (Figure 8), with all data being represented as mean ± SD (n = 3). For the samples stored at 5 °C, no significant changes were observed for both drugs when compared to the first day. Furthermore, it is noteworthy that there was no evident interaction between clotrimazole and itraconazole at this temperature, since the peaks of both drugs were detected independently without the appearance of any additional peaks. On the other hand, when the samples that were stored at 40 °C were analyzed, it was possible to notice a significant decrease in the concentration of the drugs over time, especially for clotrimazole. Additionally, a new peak could also be observed in the chromatogram of such samples (Figure 9). 020406080020406080100Concentration (%)Time (days) Clotrimazole 40°C Clotrimazole 5°C Itraconazole 40°C Itraconazole 5°C Figure 8. Graph showing the concentration of clotrimazole and itraconazole over time in different conditions. All data is represented as mean ± SD (n = 3). Molecules 2019, 24, 2333 9 of 15 Figure 8. Graph showing the concentration of clotrimazole and itraconazole over time in diﬀerent conditions. All data is represented as mean ± SD (n = 3). Figure 9. Cont. Molecules 2019, 24, x FOR PEER REVIEW 8 of 15 phosphate buffered saline 0.05 M (pH 8.0 with ammonium hydroxide 1 M). The alteration of the column brand and the mobile phase did not promote any significant variations in the retention time of clotrimazole and itraconazole peaks; a good resolution was observed with retention times of 8 min for clotrimazole and 10.7 min for itraconazole (Figure 7). Figure 7. Chromatogram of clotrimazole and itraconazole obtained in the robustness studies. 2.3. Study of the Stability of a Novel Microemulsion Containing Clotrimazole and Itraconazole Subsequently, the developed method was used in the determination of clotrimazole and itraconazole in the newly developed microemulsion with the purpose of quantifying the drugs in the formulation, as well as in the accelerated stability study. Based on the assumption that possible interactions and incompatibilities may arise from the contact between the drugs over time, they were left to stand for three months, both under refrigeration and heating conditions, and further analyzed. The initial drug content of the microemulsion was taken as 100%, and the drug content over time was plotted (Figure 8), with all data being represented as mean ± SD (n = 3). For the samples stored at 5 °C, no significant changes were observed for both drugs when compared to the first day. Furthermore, it is noteworthy that there was no evident interaction between clotrimazole and itraconazole at this temperature, since the peaks of both drugs were detected independently without the appearance of any additional peaks. On the other hand, when the samples that were stored at 40 °C were analyzed, it was possible to notice a significant decrease in the concentration of the drugs over time, especially for clotrimazole. Additionally, a new peak could also be observed in the chromatogram of such samples (Figure 9). 020406080020406080100Concentration (%)Time (days) Clotrimazole 40°C Clotrimazole 5°C Itraconazole 40°C Itraconazole 5°C Figure 8. Graph showing the concentration of clotrimazole and itraconazole over time in different conditions. All data is represented as mean ± SD (n = 3). Molecules 2019, 24, x FOR PEER REVIEW 9 of 15 020406080020406080100 Concentration (%)Time (days) Clotrimazole 40°C Clotrimazole 5°C Itraconazole 40°C Itraconazole 5°C Figure 8. Graph showing the concentration of clotrimazole and itraconazole over time in different conditions. All data is represented as mean ± SD (n = 3). (A) 30 days (40 °C) (B) 60 days (40 °C) Molecules 2019, 24, 2333 10 of 15 Figure 9. HPLC chromatograms for the samples in the stability study after (A) 30 days, (B) 60 days, and (C) 90 days. In order to investigate the formation of this compound, which might be a result of the interaction between clotrimazole and itraconazole, we conducted further studies. Initially, we sought to investigate which degradation products could be formed from the degradation of both drugs and found that the degradation of clotrimazole is well-reported under acidic conditions, giving product 3 (Figure 10). Figure 10. Reaction scheme showing the degradation of clotrimazole in acid medium. ◦ With these concepts in mind, we conducted the synthesis of compound 3 from clotrimazole by heating it at 80 C in the presence of acetonitrile and concentrated hydrochloric acid for 2 h; the product identity was conﬁrmed using NMR and HRMS by comparing the obtained data with previous reports (for details, see the Supplementary Materials) [21]. Next, we conducted the forced degradation of a mixture of itraconazole and clotrimazole by heating both at 50 C for 24 h in a solution of acetonitrile, water, and benzyl alcohol-mimicking the microemulsion composition—and isolated the formed product using column chromatography. ◦ With both compounds in hand, we analyzed product 3 and the degradation product by HPLC using the developed method, and the comparison of the retention times of both compounds proved that, indeed, product 3 is formed from the degradation of clotrimazole under acidic conditions. Furthermore, the retention time was also a match for the product previously detected in the stability studies conducted at 40 C, which proves that under speciﬁc conditions, clotrimazole may undergo degradation in the presence of traces of acid, forming 3. However, the formation of 3 was not observed in the stability studies conducted at 5 C, which shows the viability of this novel microemulsion and encourages carrying out further studies for its development. ◦ ◦ The content of itraconazole (% w/w) in the microemulsions stored in climatic chambers also underwent a slight decrease, which was less signiﬁcant when compared to clotrimazole. In order to exclude the possibility of interaction between the drugs, the decrease in itraconazole content was also investigated. However, unlike clotrimazole, degradation studies of itraconazole are not found in the literature. Thus, an aliquot was collected directly from the chromatographic system at the same retention time as the degradation product formed during the stability study; at a low intensity, it Molecules 2019, 24, x FOR PEER REVIEW 10 of 15 (C) 90 days (40 °C) Figure 9. HPLC chromatograms for the samples in the stability study after (A) 30 days, (B) 60 days, and (C) 90 days. In order to investigate the formation of this compound, which might be a result of the interaction between clotrimazole and itraconazole, we conducted further studies. Initially, we sought to investigate which degradation products could be formed from the degradation of both drugs and found that the degradation of clotrimazole is well-reported under acidic conditions, giving product 3 (Figure 10). Figure 10. Reaction scheme showing the degradation of clotrimazole in acid medium. With these concepts in mind, we conducted the synthesis of compound 3 from clotrimazole by heating it at 80 °C in the presence of acetonitrile and concentrated hydrochloric acid for 2 h; the product identity was confirmed using NMR and HRMS by comparing the obtained data with previous reports (for details, see the Supplementary Materials) [21]. Next, we conducted the forced degradation of a mixture of itraconazole and clotrimazole by heating both at 50 °C for 24 h in a solution of acetonitrile, water, and benzyl alcohol-mimicking the microemulsion composition—and isolated the formed product using column chromatography. With both compounds in hand, we analyzed product 3 and the degradation product by HPLC using the developed method, and the comparison of the retention times of both compounds proved that, indeed, product 3 is formed from the degradation of clotrimazole under acidic conditions. Furthermore, the retention time was also a match for the product previously detected in the stability studies conducted at 40 °C, which proves that under specific conditions, clotrimazole may undergo degradation in the presence of traces of acid, forming 3. However, the formation of 3 was not observed in the stability studies conducted at 5 °C, which shows the viability of this novel microemulsion and encourages carrying out further studies for its development. The content of itraconazole (% w/w) in the microemulsions stored in climatic chambers also underwent a slight decrease, which was less significant when compared to clotrimazole. In order to exclude the possibility of interaction between the drugs, the decrease in itraconazole content was also investigated. However, unlike clotrimazole, degradation studies of itraconazole are not found in the literature. Thus, an aliquot was collected directly from the chromatographic system at the same retention time as the degradation product formed during the stability study; at a low intensity, it was Molecules 2019, 24, x FOR PEER REVIEW 10 of 15 NClN2H2OH+ (cat)OHCl3+NHN4+ Figure 10. Reaction scheme showing the degradation of clotrimazole in acid medium. With these concepts in mind, we conducted the synthesis of compound 3 from clotrimazole by heating it at 80 °C in the presence of acetonitrile and concentrated hydrochloric acid for 2 h; the product identity was confirmed using NMR and HRMS by comparing the obtained data with previous reports (for details, see the Supplementary Materials) [21]. Next, we conducted the forced degradation of a mixture of itraconazole and clotrimazole by heating both at 50 °C for 24 h in a solution of acetonitrile, water, and benzyl alcohol-mimicking the microemulsion composition—and isolated the formed product using column chromatography. With both compounds in hand, we analyzed product 3 and the degradation product by HPLC using the developed method, and the comparison of the retention times of both compounds proved that, indeed, product 3 is formed from the degradation of clotrimazole under acidic conditions. Furthermore, the retention time was also a match for the product previously detected in the stability studies conducted at 40 °C, which proves that under specific conditions, clotrimazole may undergo degradation in the presence of traces of acid, forming 3. However, the formation of 3 was not observed in the stability studies conducted at 5 °C, which shows the viability of this novel microemulsion and encourages carrying out further studies for its development. The content of itraconazole (% w/w) in the microemulsions stored in climatic chambers also underwent a slight decrease, which was less significant when compared to clotrimazole. In order to exclude the possibility of interaction between the drugs, the decrease in itraconazole content was also investigated. However, unlike clotrimazole, degradation studies of itraconazole are not found in the literature. Thus, an aliquot was collected directly from the chromatographic system at the same retention time as the degradation product formed during the stability study; at a low intensity, it was possible to observe a product with a mass-to-charge ratio (m/z) of 437.1931. Considering that the itraconazole concentration change was lower than for clotrimazole, we hypothesized that an interaction of itraconazole with some other excipient of the microemulsion may be taking place. In that sense, we propose that the degradation product may be formed by the nucleophilic addition of benzyl alcohol to the methylene group linking the phenolic aromatic part with the 1,3-dioxolane ring (Figure 11); indeed, the mass-to-charge ratio was a match for the proposed product. It is worth mentioning that although we have observed a good match in a mass-to-charge ratio of 437.1931, further studies are necessary to confirm whether the proposed structure is indeed the correct one, such as the isolation and complete spectroscopic characterization of this compound, which was not possible at the scale we were working. Figure 11. Scheme showing the reaction between itraconazole and benzyl alcohol. Molecules 2019, 24, 2333 11 of 15 was possible to observe a product with a mass-to-charge ratio (m/z) of 437.1931. Considering that the itraconazole concentration change was lower than for clotrimazole, we hypothesized that an interaction of itraconazole with some other excipient of the microemulsion may be taking place. In that sense, we propose that the degradation product may be formed by the nucleophilic addition of benzyl alcohol to the methylene group linking the phenolic aromatic part with the 1,3-dioxolane ring (Figure 11); indeed, the mass-to-charge ratio was a match for the proposed product. It is worth mentioning that although we have observed a good match in a mass-to-charge ratio of 437.1931, further studies are necessary to conﬁrm whether the proposed structure is indeed the correct one, such as the isolation and complete spectroscopic characterization of this compound, which was not possible at the scale we were working. Figure 11. Scheme showing the reaction between itraconazole and benzyl alcohol. 3. Experimental Methods 3.1. Materials for Analytical Method Development Clotrimazole and itraconazole (as a mixture of stereoisomers) standards were purchased from Merck, São Paulo, SP, Brazil. Microemulsions were prepared using Tween 60, propylene glycol, and benzyl alcohol, all purchased from Merck, São Paulo, SP, Brazil. HPLC-grade acetonitrile was acquired from J.T. Baker Inc., Phillipsburg, NJ, USA. Clotrimazole (Jintan Zhongxing Pharmaceutical Chemical Co., Ltd., Mainland, China) and itraconazole (Metrochem API, Telangana, India) were donated by Valdequimica Produtos Quimicos Ltd., São Paulo, Brazil. All solutions were prepared with ultra-pure Milli-Q water obtained from a Milli-Q Water Millipore puriﬁcation system (Burlington, MA, USA). ® 3.2. Compatibility Study of Clotrimazole and Itraconazole 3.2.1. Preparation of Clotrimazole/Itraconazole Binary Mixtures The binary mixtures were prepared and homogenized by taking clotrimazole and itraconazole in a 1:1 proportion (w:w). These mixtures were further used for X-ray powder diﬀraction, Fourier transform infrared spectroscopy (FTIR), and thermal analyses. 3.2.2. X-ray Powder Diﬀraction (PXRD) PXRD patterns were collected on a Bruker D8 Venture diﬀractometer system (Bruker, Billerica, MA, USA) operating at 1.5406 Å, 40 kV voltage, and a current of 40 mA using a Cu Kα radiation source. The samples were contained in a ﬂat poly(methyl methacrylate) sample holder and the data acquisition ◦ was done in a range of 5 to 70 ◦/0.1 s step size over a total period of 10 min. (2θ) at 0.019 3.2.3. Fourier Transform Infrared Spectroscopy (FTIR) The FTIR spectra of the solid samples were obtained using a Varian FT-IR 660 equipment (Varian Inc., Walnut Creek, CA, USA). A hydraulic press was used to prepare pellets for analysis. The KBr pellets contained 3 mg of a sample and 100 mg of KBr. Spectra were collected with a resolution of 4 cm −1 on the spectral domain of 3800–600 cm −1. Molecules 2019, 24, x FOR PEER REVIEW 10 of 15 NClN2H2OH+ (cat)OHCl3+NHN4+ Figure 10. Reaction scheme showing the degradation of clotrimazole in acid medium. With these concepts in mind, we conducted the synthesis of compound 3 from clotrimazole by heating it at 80 °C in the presence of acetonitrile and concentrated hydrochloric acid for 2 h; the product identity was confirmed using NMR and HRMS by comparing the obtained data with previous reports (for details, see the Supplementary Materials) [21]. Next, we conducted the forced degradation of a mixture of itraconazole and clotrimazole by heating both at 50 °C for 24 h in a solution of acetonitrile, water, and benzyl alcohol-mimicking the microemulsion composition—and isolated the formed product using column chromatography. With both compounds in hand, we analyzed product 3 and the degradation product by HPLC using the developed method, and the comparison of the retention times of both compounds proved that, indeed, product 3 is formed from the degradation of clotrimazole under acidic conditions. Furthermore, the retention time was also a match for the product previously detected in the stability studies conducted at 40 °C, which proves that under specific conditions, clotrimazole may undergo degradation in the presence of traces of acid, forming 3. However, the formation of 3 was not observed in the stability studies conducted at 5 °C, which shows the viability of this novel microemulsion and encourages carrying out further studies for its development. The content of itraconazole (% w/w) in the microemulsions stored in climatic chambers also underwent a slight decrease, which was less significant when compared to clotrimazole. In order to exclude the possibility of interaction between the drugs, the decrease in itraconazole content was also investigated. However, unlike clotrimazole, degradation studies of itraconazole are not found in the literature. Thus, an aliquot was collected directly from the chromatographic system at the same retention time as the degradation product formed during the stability study; at a low intensity, it was possible to observe a product with a mass-to-charge ratio (m/z) of 437.1931. Considering that the itraconazole concentration change was lower than for clotrimazole, we hypothesized that an interaction of itraconazole with some other excipient of the microemulsion may be taking place. In that sense, we propose that the degradation product may be formed by the nucleophilic addition of benzyl alcohol to the methylene group linking the phenolic aromatic part with the 1,3-dioxolane ring (Figure 11); indeed, the mass-to-charge ratio was a match for the proposed product. It is worth mentioning that although we have observed a good match in a mass-to-charge ratio of 437.1931, further studies are necessary to confirm whether the proposed structure is indeed the correct one, such as the isolation and complete spectroscopic characterization of this compound, which was not possible at the scale we were working. Figure 11. Scheme showing the reaction between itraconazole and benzyl alcohol. Molecules 2019, 24, 2333 12 of 15 3.2.4. Thermal Analyses DSC data were collected on a Shimadzu Diﬀerential Scanning Calorimeter DSC-60A (Shimadzu, Quioto, Japan). Approximately 4 mg samples were placed in aluminum pans, and the temperature −1 under nitrogen ﬂow program was set to increase from 30 to 250 (50 mL min C with a heating rate of 10 C min −1). ◦ ◦ Thermogravimetric (TG) analyses were performed using a Netzsch STA 409 PC/PG (Netzsch, −1 at a heating rate of Selb, Germany) under a nitrogen atmosphere with a ﬂow rate of 60 mL min ◦ 10 −1 over the range of 30 to 300 C and using 6 mg of sample in an aluminum cell. C min ◦ 3.3. Instruments and Chromatographic Conditions Chromatographic experiments were performed on a Shimadzu SPD-M20A system (Shimadzu, Quioto, Japan). The chromatographic separations were performed using a 150 mm × 4.6 mm i.d. (5 µm particle size) Fortis C18 column in isocratic elution mode with acetonitrile and phosphate buﬀered −1. saline 0.05 M pH 8.0 adjusted with ammonium hydroxide 1 M (60:40, v/v) at a ﬂow rate of 1.0 mL min The detection wavelength was set at 190 nm, and the injection volume was 20 µL. 3.4. Standard Stock Solutions and Calibration Standards Standard stock solutions of clotrimazole and itraconazole were freshly prepared by dissolving the −1) Calibration standards in the concentration range of 5, 10, 20, 40, 80, −1 were prepared in the appropriate volumetric ﬂasks by diluting the stock solution drugs in methanol (0.2 mg mL 160, and 200 µg mL in the mobile phase. An aliquot (20 µL) of the solution was then directly injected into the HPLC. 3.5. Sample Preparation An amount of microemulsion was accurately weighted to contain 25 mg clotrimazole and C. The sample was itraconazole in a 50 mL centrifuge tube and heated for 5 min in a water bath at 50 then removed from the bath, shaken until cooled to room temperature, and placed in an ice-methanol bath. Next, the sample was centrifuged for 5 min and extracted with chloroform (5 mL). Finally, the solvent was removed under a stream of gaseous nitrogen, and the residue was diluted in the mobile phase. ◦ 3.6. Method Validation Protocol The proposed method was validated under the optimized conditions regarding its linearity range, selectivity, sensitivity, precision, accuracy and stability of the assay according to the regulatory guidelines requirements (FDA). 3.6.1. Linearity Range The linearity range was evaluated by measuring the chromatographic peak area responses of the drugs at seven concentration levels and in triplicate. Analytical curves were constructed by plotting the peak area against the concentration of itraconazole and clotrimazole (Figures 2 and 3), which gives the regression equation. The results are presented in Table 1. 3.6.2. Selectivity To ensure the selectivity of the proposed method, drug-free microemulsions were prepared and analyzed in the described chromatographic conditions. 3.6.3. Sensitivity The sensitivity was determined by means of the limit of detection (LOD) and limit of quantiﬁcation (LOQ). One of the ways to calculate the LOD (Equation (1)) and LOQ (Equation (2)) is based on the Molecules 2019, 24, 2333 13 of 15 standard deviation (σ) of the y-intercept from the regression of the calibration standard. The results are given in Table 1. LOD = 3, 3.σ s LOD (σ—standard deviation; s—slope of the calibration standard). LOQ = 10.σ s LOQ (σ—standard deviation; s—slope of the calibration standard). 3.6.4. Precision and Accuracy (1) (2) The accuracy and precision of the method were estimated by quintuplicate quality control −1 (medium QC), (QC) samples prepared using the mobile phase: 7 µg mL −1 (medium QC), −1 (high QC) for clotrimazole and 7 µg mL and 120 µg mL −1 (high QC) for itraconazole. Accuracy was established through back-calculation and 150 µg mL and expressed as the percent diﬀerence between the found and the nominal concentration for each compound, and the precision was calculated as the coeﬃcient of variation (CV) of the replicate measurements. Calibration standards and QC samples were analyzed in three diﬀerent batches in order to determine the intra and inter-batch variability. −1 (low QC), 15 µg mL −1 (low QC), 70 µg mL 3.6.5. Stability The stability of the standard solutions was investigated after storage for 7, 15, and 30 days at room temperature (25 ◦ C) and under refrigeration (−5 ◦ C) using the working solution. 3.6.6. Robustness The robustness of an analytical method is a measure of its capacity to resist changes due to small variations in parameter conditions, e.g., by using a diﬀerent column. In this way, the method robustness was assessed as a function of changing the column brand for a C18 Agilent column (Agilent Technologies Inc, Santa Clara, CA, USA), (150 × 4.6 mm × 5 µm) and the ratio of the mobile phase. 3.7. Application of the Method 3.7.1. Microemulsion Preparation ® With the developed method in hand, the next step was to develop a stable microemulsion using a combination of clotrimazole and itraconazole. MEs were composed of benzyl alcohol, the non-ionic surfactant Tween 60, propylene glycol, and water. The optimum weight ratios of the components and MEs’ areas were determined using a pseudo-ternary phase diagram (data not shown in this work). The systems were prepared as previously described [22]; the surfactant (Tween 60) and cosolvent (propylene glycol) were prepared separately, and clotrimazole and itraconazol were solubilized in benzyl alcohol and added to the mixture. The pseudo-ternary phase diagrams of oil, surfactant/cosolvent, and water were set up using the water titration method. ® 3.7.2. Stability Study The stability proﬁle of the prepared microemulsion at accelerated conditions was studied according to the ICH guidelines. The formulation was placed separately in an amber-colored screw-capped glass container and stored at 40 ± 2 C for 3 months, with sampling at 0, 30, 60, and 90 days. The samples were then evaluated for drug content using the developed HPLC method. C and 5–8 ± 3 ◦ ◦ Molecules 2019, 24, 2333 14 of 15 3.8. Characterization of the Synthetized Compounds NMR spectra were obtained using a Varian Unity Plus VXR (Varian Inc., Walnut Creek, CA, USA), 500 MHz instrument in CDCl3 solutions. The chemical shifts were reported in units of d (ppm) downﬁeld from tetramethylsilane, which was used as an internal standard; coupling constants (J) are reported in hertz and refer to apparent peak multiplicities. High-resolution mass spectra (HRMS) were recorded on a MICROMASS Q-TOF mass spectrometer (Waters, Milford, MA, USA). 4. Conclusions The combination of clotrimazole and itraconazole in a pharmaceutical formulation is of great importance owing to the potential of generating a new option for the treatment of sporotrichosis. In this sense, the preformulation investigation using diﬀerent techniques (DSC, TG, PXRD, FTIR) was essential to examine the existence of possible clotrimazole–itraconazole interactions. Furthermore, an HPLC method was developed and validated according to standard guidelines, and it is the ﬁrst reported method for the simultaneous determination of clotrimazole and itraconazole in nanotechnology-based products such as microemulsions. Based on our results, it was possible to conclude that there is no other co-eluting peak along with those of interest, the method being speciﬁc for the estimation of clotrimazole and itraconazole. Interestingly, accelerated stability studies showed that a product derived from clotrimazole was formed, as well as a possible interaction between itraconazole and benzyl alcohol, when the microemulsion was conditioned at elevated temperatures (40 C). On the other hand, the studies C showed that the microemulsion is stable for at least 3 months, as no degradation conducted at 5 peaks were observed in the HPLC analysis, which allows us to infer that it is possible to guarantee the stability of the formulation under refrigeration. ◦ ◦ Supplementary Materials: Supplementary materials are available online. Figure S1: Analytical calibration curve for clotrimazole, Figure S2: Analytical calibration curve for itraconazole, Figure S3: IR spectra of clotrimazole, itraconazole and their binary mixture (1:1), Figure S4: HPLC chromatogram of compound 3, Figure S5: HPLC chromatogram of the decomposition product formed via the forced degradation of clotrimazole in the presence of itraconazole. Author Contributions: All authors have read this manuscript and concur with its submission. The contributions of each author are listed as follows: P.G.F.—Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Roles/Writing—original draft; C.G.d.S.L.—Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—review & editing; L.L.N.—Data curation, Formal analysis, Investigation; Methodology; M.C.d.M.—Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Funding acquisition, Writing—review & editing; F.d.C.d.S.—Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Validation, Visualization; A.L.V.—Conceptualization; D.O.F.—Conceptualization, Funding acquisition, Supervision, Writing—review & editing; V.F.F.—Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing—review & editing. Funding: This study was ﬁnanced in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior-Brasil (CAPES)-Finance Code 001, CNPq (303713/2014-3) and FAPERJ (E-26/2002.800/2017, E-26/200.930/2017). Conﬂicts of Interest: All authors declare that there is no conﬂict of interest. References 1. 2. 3. 4. Chakrabarti, A.; Bonifaz, A.; Gutierrez-Galhardo, M.C.; Mochizuki, T.; Li, S. Global epidemiology of sporotrichosis. Med. Mycol. 2015, 53, 3–14. [CrossRef] [PubMed] Gremião, I.D.; Miranda, L.H.; Reis, E.G.; Rodrigues, A.M.; Pereira, S.A. Zoonotic epidemic of sporotrichosis: Cat to human transmission. PLoS Pathog. 2017, 13, e1006077. [CrossRef] [PubMed] Rodrigues, A.M.; de Hoog, G.S.; de Camargo, Z.P. Sporothrix species causing outbreaks in animals and humans driven by animal-animal transmission. PLoS Pathog. 2016, 12, e1005638. [CrossRef] [PubMed] Barros, M.B.L.; Schubach, T.P.; Coll, J.O.; Gremião, I.D.; Wanke, B.; Schubach, A. Esporotricose: A evolução e os desaﬁos de uma epidemia. Rev. Panam. Salud Publica 2010, 27, 455–460. [PubMed] Molecules 2019, 24, 2333 15 of 15 5. 6. 7. 8. 9. Gremião, I.D.; Menezes, R.C.; Schubach, T.M.; Figueiredo, A.B.; Cavalcanti, M.C.; Pereira, S.A. Feline sporotrichosis: Epidemiological and clinical aspects. Med. Mycol. 2015, 53, 15–21. [CrossRef] [PubMed] Bustamante, B.; Campos, P.E. Sporotrichosis: A forgotten disease in the drug research. Expert Rev. Anti-Infect. Ther. 2004, 2, 85–94. [CrossRef] [PubMed] Kadavakollu, S.; Stailey, C.; Kunapareddy, C.S.; White, S. Clotrimazole as a cancer drug: A short review. Med. Chem. 2014, 4, 722–724. [CrossRef] Gagini, T.; Borba-Santos, L.P.; Rodrigues, A.M.; Camargo, Z.P.; Rozental, S. Clotrimazole is highly eﬀective in vitro against feline Sporothrix brasiliensis isolates. J. Med. Microbiol. 2011, 66, 1573–1580. [CrossRef] Pai, V.; Ganavalli, A.; Kikkeri, N.N. Antifungal resistance in dermatology. Indian J. Dermatol. 2018, 63, 361–368. [CrossRef] 10. Carvalho, A.L.M.; da Silva, J.A.; Lira, A.A.M.; Conceição, T.M.F.; Nunes, R.S.; Junior, R.L.C.A.; Sarmento, V.H.V.; Leal, L.B.; Santana, D.P. Evaluation of microemulsion and lamellar liquid crystalline systems for transdermal zidovudine delivery. J. Pharm. Sci. 2016, 105, 1–6. [CrossRef] 11. Padula, C.; Telò, I.; Ianni, A.D.; Pescina, S.; Nicoli, S.; Santi, P. Microemulsion containing triamcinolone acetonide for buccal administration. Eur. J. Pharm. Sci. 2018, 115, 233–239. [CrossRef] 12. Rashida, M.A.; Naza, T.; Abbasa, M.; Nazirb, S.; Younasa, N.; Majeeda, S.; Qureshic, N.; Akhtard, M.N. Chloramphenicol loaded microemulsions: Development, characterization and stability. Colloid Interface Sci. Commun. 2019, 28, 41–48. [CrossRef] Seok, S.H.; Lee, S.-A.; Park, E.-S. Formulation of a microemulsion-based hydrogel containing celecoxib. J. Drug Deliv. Sci. Technol. 2018, 43, 409–414. [CrossRef] 13. 14. Kumar, S.K.; Dhancinamoorthi, D.; Sarvanan, R.; Gopal, U.K.; Shanmugam, V. Microemulsions as a carrier for novel drug delivery: A review. Int. J. Pharm. Sci. Rev. Res. 2011, 10, 37–45. 15. Hu, X.-B.; Kang, R.-R.; Tang, T.-T.; Li, Y.-J.; Wu, J.-Y.; Wang, J.-M.; Liu, X.-Y.; Xiang, D.-X. Topical delivery of -trimethoxy-trans-stilbene-loaded microemulsion-based hydrogel for the treatment of osteoarthritis in 3,5,4 a rabbit model. Drug Deliv. Transl. Res. 2019, 9, 357–365. [CrossRef] (cid:48) 16. Hájková, R.; Sklenárová, H.; Matysová, L.; Svecová, P.; Solich, P. Development and validation of HPLC method for determination of clotrimazole and its two degradation products in spray formulation. Talanta 2007, 73, 483–489. [CrossRef] 17. Abdel-Moety, E.M.; Khattab, F.I.; Kelani, K.M.; AbouAl-Alamein, A.M. Chromatographic determination of clotrimazole, ketoconazole and ﬂuconazole in pharmaceutical formulations. Farmaco 2002, 57, 931–938. [CrossRef] 18. Bharate, S.S.; Bharate, S.B.; Bajaj, A.N. Interactions and incompatibilities of pharmaceutical excipientes with active pharmaceutical ingredients: A comprehensive review. J. Excipients and Food Chem. 2010, 1, 3–26. [CrossRef] 19. Ceschel, G.C.; Badiello, R.; Ronchi, C.; Maﬀei, P. Degradation of components in drug formulations: A comparison between HPLC and DSC methods. J. Pharm. Biomed. Anal. 2003, 32, 1067–1072. [CrossRef] 20. Deshmukha, P.R.; Gaikwadb, V.L.; Tamanea, P.K.; Mahadikc, K.R.; Purohit, R.N. Development of stability-indicating HPLC method and accelerated stability studies for osmotic and pulsatile tablet formulations of Clopidogrel Bisulfate. J. Pharm. Biomed. Anal. 2019, 165, 346–356. [CrossRef] 21. Lee, T.-K.; Ryoo, S.-J.; Lee, Y.-S. A new method for the preparation of 2-chlorotrityl resin and its application to solid-phase peptide synthesis. Tetrahedron Lett. 2007, 48, 389–391. [CrossRef] 22. Nandi, I.; Bari, M.; Joshi, H. Study of isopropyl myristate micremulsion systems containing cyclodextrins to improve the solubility of two model hydrophobic drugs. AAPS PharmaSciTech. 2003, 4, 1–9. [CrossRef] Sample Availability: Samples of the compounds are available from the authors. © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_s22093318	Article Intra- and Inter-Subject Perspectives on the Detection of Focal Onset Motor Seizures in Epilepsy Patients Sebastian Böttcher 1,2,3,* Martin Glasstetter 1 Mark P. Richardson 3,4 , Elisa Bruno 3,4 , Valentina Ticcinelli 3,5 , Nino Epitashvili 1 , Matthias Dümpelmann 1,3 , Nicolas Zabler 1 , , Sarah Thorpe 3,6 , Simon Lees 3,6 , Kristof Van Laerhoven 2 , and Andreas Schulze-Bonhage 1,3 1 Epilepsy Center, Department of Neurosurgery, Medical Center—University of Freiburg, 79106 Freiburg im Breisgau, Germany; [email protected] (N.E.); [email protected] (M.D.); [email protected] (N.Z.); [email protected] (M.G.); [email protected] (A.S.-B.) 2 Ubiquitous Computing, Department of Electrical Engineering and Computer Science, University of Siegen, 3 57076 Siegen, Germany; [email protected] The RADAR-CNS Consortium, London WC2R 2LS, UK; [email protected] (E.B.); [email protected] (V.T.); [email protected] (S.T.); [email protected] (S.L.); [email protected] (M.P.R.) 4 Division of Neuroscience, Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London SE5 9RT, UK 5 UCB Pharma, 1070 Anderlecht, Belgium 6 The RADAR-CNS Patient Advisory Board, King’s College London, London WC2R 2LS, UK * Correspondence: [email protected] Abstract: Focal onset epileptic seizures are highly heterogeneous in their clinical manifestations, and a robust seizure detection across patient cohorts has to date not been achieved. Here, we assess and discuss the potential of supervised machine learning models for the detection of focal onset motor seizures by means of a wrist-worn wearable device, both in a personalized context as well as across patients. Wearable data were recorded in-hospital from patients with epilepsy at two epilepsy centers. Accelerometry, electrodermal activity, and blood volume pulse data were processed and features for each of the biosignal modalities were calculated. Following a leave-one-out approach, a gradient tree boosting machine learning model was optimized and tested in an intra-subject and inter-subject evaluation. In total, 20 seizures from 9 patients were included and we report sensitivities of 67% to 100% and false alarm rates of down to 0.85 per 24 h in the individualized assessment. Conversely, for an inter-subject seizure detection methodology tested on an out-of-sample data set, an optimized model could only achieve a sensitivity of 75% at a false alarm rate of 13.4 per 24 h. We demonstrate that robustly detecting focal onset motor seizures with tonic or clonic movements from wearable data may be possible for individuals, depending on speciﬁc seizure manifestations. Keywords: wearables; epilepsy; seizure detection; multimodal; mHealth; eHealth; mobile health; digital health 1. Introduction With a worldwide incidence of up to 100 per 100,000 per year, epilepsy is one of the most prevalent neurological disorders, affecting over 70 million people worldwide [1]. Epilepsy manifests in a multitude of different symptoms with varying severities generally denoted as epileptic seizures [2], and the current gold standard in diagnosis and seizure monitoring is in-hospital video-electroencephalography (vEEG). However, this diagnostic tool, while being accurate and widely used for diagnosis and determination of treatment, is only practicable in relatively short-term applications at a hospital or special epilepsy monitoring unit [3]. At home or in their daily life, patients with epilepsy cannot feasibly be Citation: Böttcher, S.; Bruno, E.; Epitashvili, N.; Dümpelmann, M.; Zabler, N.; Glasstetter, M.; Ticcinelli, V.; Thorpe, S.; Lees, S.; Van Laerhoven, K.; et al. Intra- and Inter-Subject Perspectives on the Detection of Focal Onset Motor Seizures in Epilepsy Patients. Sensors 2022, 22, 3318. https://doi.org/ 10.3390/s22093318 Academic Editor: Lorenzo Chiari Received: 22 March 2022 Accepted: 22 April 2022 Published: 26 April 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Sensors 2022, 22, 3318. https://doi.org/10.3390/s22093318 https://www.mdpi.com/journal/sensors (cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)sensorsSensors 2022, 22, 3318 2 of 20 monitored in such a way and with the same beneﬁt. Yet, seizure tracking in an ultra-long- term context at home is needed, both as an automatic alarm to potentially alert carers in the event of a seizure, and as a tool for monitoring and forecasting the disease trajectory [4,5]. Thus, different technologies and methodologies of seizure monitoring need to be employed, and wearables such as smartwatches and ﬁtness trackers in combination with machine learning may ﬁll this gap while being easily available for a wider audience. Epileptic seizures are deﬁned by a period of abnormal neuronal activity in the brain, and are generally divided into two main groups by their neurological onset [2]. Seizures with an early bilateral involvement are called generalized seizures, while seizures with just a single point of onset are denoted as focal onset seizures, but epileptic activity can propagate across the brain resulting in “focal to bilateral” seizures with characteristic motor manifestations. Bilateral tonic–clonic seizures have been assessed in numerous studies as to the viability of wearables for the detection during recent years, and detection has been demonstrated to be feasible in multiple retrospective studies [6–14]. Conversely, focal seizures in general are still a relatively unexplored ﬁeld with respect to wearable non-EEG detection [15–20]. Symptoms and manifestations of these seizures are much more heterogeneous as compared to those of bilateral tonic–clonic seizures, with some barely or not at all captured by typical wearable biosignal modalities, such as accelerometry (ACC), electrodermal activity (EDA), or photoplethysmography (PPG). Focal seizures can be roughly divided into two categories: those with motor and those without motor manifestations. Non-motor symptoms, that is, those without involuntary movement of the body, may include partial loss of awareness or consciousness, cognitive impairment, or emotional or sensory symptoms. Motor symptoms, on the other hand, can include tonic or clonic movements of the limbs or body in general, hyperkinetic movements, or automatisms. A single epileptic seizure can thereby be composed of multiple types of manifestations, in parallel or sequentially. This work highlights difﬁculties in the detection of focal onset epileptic seizures, specif- ically those with focal tonic or clonic motor symptoms but without bilateral propagation, from biosignal data captured by wearable devices. A data set consisting of multimodal data from a wrist-worn wearable was recorded from patients with epilepsy during their in- hospital stay at an epilepsy monitoring unit (EMU). Classical supervised machine learning is applied in order to assess the utility of this kind of data for the detection of seizures, in the context of an automated seizure diary. As written diaries created by the patients themselves have been demonstrated to be very inaccurate and often severely undercount seizures even for convulsive seizures [21–23], an automated diary tool implementing an objective seizure identiﬁcation and quantiﬁcation is needed, for example, as a basis for treatment decisions made by epileptologists. In the supervised methodology employed here, data are ﬁrst labeled, based on parallel vEEG monitoring, as “seizure” or not “seizure”, and then processed into meaningful features and given to the machine learning model for training. Thus, the trained model can then be used to automatically classify new data. Speciﬁcally, in this study we employ a gradient tree boosting machine as the seizure detection model. To evaluate such an approach, and subsequently also determine the application in a real- world system, two procedures can be applied: intra-subject or inter-subject evaluation. The intra-subject evaluation focuses on the performance of the methodology when applied to data from a single patient, while the inter-subject evaluation assesses the performance over multiple patients with potentially different types of epilepsy and seizure manifestations. The former requires multiple seizures recorded per subject and will produce individual- ized models tailored to a single patient, while the latter requires seizures recorded from multiple different participants and will give inter-subject models, to be used over wider populations. Here, we aim to determine which of these approaches may work best for focal motor seizures going forward, giving guidance for the design of future studies in the ﬁeld. Following the study classiﬁcation suggested by Beniczky and Ryvlin in 2018 [24], the work presented here could be classiﬁed as a phase 1 retrospective proof-of-principle study. The main contribution of this work is the evaluation of supervised machine learn- Sensors 2022, 22, 3318 3 of 20 ing methodologies on focal onset epileptic motor seizures in a data set recorded from a non-EEG wearable device. A comparison between two evaluation approaches, intra- and inter-subject, provides additional context and facilitates recommendations towards future studies in the ﬁeld. 2. Materials and Methods 2.1. Data Set Data from wearable devices was recorded from a total of 243 patients with epilepsy across two EMUs in the period between July 2017 and February 2020. Both at the neuro- physiological department of King’s College Hospital, London (KCL, 71/243), and at the Epilepsy Center, Medical Center—University of Freiburg (UKF, 172/243), patients in the age range of 7 to 80 with a diagnosis of epilepsy were recruited sequentially as part of their standard clinical epilepsy care, for example, in the course of standard presurgical evaluation. Patients with predominantly (suspected) psychogenic non-epileptic seizures or other involuntary movements were not included in the study. As part of their stay in the EMU study, participants may have had seizures provoked, for example, by temporary reduction of their anti-epileptic medication or through other means, such as sleep depri- vation or hyperventilation techniques. The vEEG data was retrospectively reviewed and labeled by clinical experts (E.B., N.E.). Primarily, they marked seizure type and semiologies, electrographic and clinical onset and offset, and other meta-data including state of vigilance and body position at seizure onset. While participants wore different kinds of wearable devices, and sometimes more than one in parallel, the retrospective study presented here only includes data from the wrist-worn Empatica E4 (Figure A1; Empatica Inc., Boston MA, USA). It is a research-grade device designed speciﬁcally for epilepsy seizure detection recording 3-axis ACC at a sample rate of 32 Hz, EDA at 4 Hz, skin temperature at 4 Hz, and PPG at 64 Hz, the latter of which was internally pre-processed into a blood volume pulse (BVP) signal. The wearable has a European CE class IIa certiﬁcation as a medical device. The data recording mode used in this study was the online Bluetooth streaming mode. Battery life could range between 12 h to 48 h depending on the condition of the battery. Par- ticipants were given two devices, such that one would always be recording while the other was charging. The study and recording procedures are further described in Bruno et al. 2021 [25] and Ranjan et al. 2019 [26]. As part of the study recruitment, all participants gave written informed consent, and the study protocols and consent forms were approved by the local ethics committees (London Fulham Research Ethics Committee—16/LO/2209; Ethics Committee at the University of Freiburg—538/16). 2.2. Feature Set To facilitate the detection of focal motor seizures in this data set of non-EEG wearable data, a selection of derived features are calculated from each of the three raw data modalities. These features are chosen to meaningfully represent the changes in the signal between ictal (seizure) and inter-ictal (non-seizure) phases. Each feature vector is calculated consecutively from the raw time series data at a constant interval of two seconds, regardless of the actual length of the feature window (see Figure 1). The choice of features in this study was informed primarily by previous research in the ﬁeld. The following details the feature calculations for the biosignal modalities, ACC, EDA, and BVP. For the ACC features, a number of parameters calculated from the recurrence plot are used as features. Recurrence plots are a statistical tool to analyze recurrence in time series data [27,28], and have been successfully used in the detection of motor movements from accelerometer data before [7,29,30]. Here, we speciﬁcally calculate, from the recurrence quantiﬁcation analysis, the determinism (percentage of points that form diagonal lines of a minimal length), the Shannon entropy (probability that a line has a certain length), the average diagonal line length, and the recurrence rate (density of recurrence points). All of these values are derived from overlapping data windows of a length of 10 s, centered at each two-second interval. Sensors 2022, 22, 3318 4 of 20 The EDA features used here are calculated from the skin conductance level (SCL) and the skin conductance response rate (SCRR) [31–34]. The former is essentially a low-pass ﬁltered version of the original raw EDA signal and thus represents the slower tonic changes in the EDA data. It is represented in the feature set by the difference of area under the curve and maximum between the ﬁve minutes before (feature window) and after (baseline) each two-second interval point. Additionally, the SCRR feature is calculated against the baseline in the same way, representing the higher-frequency phasic changes of the EDA signal. The SCRR is calculated as the number of threshold crossings of the ﬁrst derivative of the EDA signal in the window. Finally, the BVP raw data (derived device-internally from the PPG sensor) is processed to a heart rate (HR) estimation following the procedure described in Glasstetter et al. 2021 [35]: A peak tracking algorithm was applied to ﬁnd local minima in the raw time series [36], and the resulting inter-beat-intervals were processed to the HR estimation employing several ﬁlters to produce a smooth and meaningful output. This HR estimation as well as a spectral entropy score representing BVP signal quality [35,37] was used as feature values. As the BVP signal is highly sensitive to motion artifacts [38], using a signal quality index like this as a feature for classiﬁcation follows the principle of regarding artifacts as additional information, instead of discarding them outright. Furthermore, this feature can be observed as a sort of indication for the quality of the model; a model that is highly dependent on the data quality of a signal may not be regarded as a particularly stable model. Additionally, the mean and maximum of the calculated HR feature over a 60-s window are used as features as well, which are baseline-corrected by the difference of values between the feature and baseline window. An overview of the different feature and baseline windows can be found in Figure 1. A comprehensive listing of the individual features is shown below: 1. Four features calculated from the recurrence plot of the ACC signal in a ten-second window: (a) (b) (c) (d) Determinism, that is, the percentage of points that form diagonal lines of a minimal length. The Shannon entropy of the probability that a line has a certain length. The average diagonal line length. Recurrence rate, that is, the density of recurrence points. 2. EDA-based features over a ﬁve-minute window, minus the same value in the ﬁve minutes before the feature window: (a) (b) (c) The area under the curve of the skin conductance level calculated as the moving mean of the raw EDA signal over a one-minute window. The maximum value of the skin conductance level calculated as above. The skin conductance response rate calculated as the number of threshold crossings of the ﬁrst derivative of the smoothed EDA signal within the window. 3. Heart rate-based features calculated from the BVP signal: (a) (b) (c) (d) The local maximum of the heart rate estimation in a 60-s window, minus the baseline value from the prior 60-s window. The mean of the heart rate estimation in a 60-s window, minus the baseline value from the prior 60-s window. The spectral entropy data quality index of the raw BVP signal, sampled at two-second intervals. The heart rate estimation calculated from the raw BVP signal, sampled at two-second intervals. Sensors 2022, 22, 3318 5 of 20 Figure 1. Overview of how the feature and baseline windows were chosen, for the three different groups of features by modality. This calculation would result in one feature vector, for the next the windows would all be shifted by an interval of T = 2 s to the right. Abscissa not to scale. 2.3. Evaluation To assess the possibilities of detecting focal epileptic seizures by wearable biosig- nal data, we investigate two different approaches: intra-subject and inter-subject. The distinction is an important addition to this work, as focal motor seizures have not been investigated to a degree that allows making the choice outright. While an inter-subject approach, i.e., creating models that can detect seizures across a patient population without individual adjustments, is certainly the best possible outcome, the heterogeneity of focal seizures may dictate the use of individualized models. To examine the effect that this might have with the given data set, the evaluation is divided into two parts. First, a subset of participants with at least three seizures recorded is isolated, and the detection model is evaluated per participant in a parameter-optimized leave-one-seizure-out cross-validation. As the data set does not provide data with more than six seizures recorded for a single participant, or with multiple independent recordings of a participant, this is performed without a dedicated test set which is truly “out-of-sample”. Rather, the model is trained with the data of all but one seizure and the respective peri-ictal data of 10 min before and after each seizure. These data are standardized using the z-score method before training, and the normalization parameters (centering mean and scaling standard deviation) are stored. The resulting model is then tested on the remaining participant data and standard- ized using the previously stored normalization parameters from the training step. This test data includes the complete data set of the participant, including the left-out seizure, but not any of the data used for training the model. Nevertheless, due to the high imbalance between inter-ictal versus ictal phases, the proportion of data between the test and training set is usually far greater than 10:1. This process is repeated such that each seizure of the participant is left out once. Secondly, the seizure data from all the participants with three or more seizures recorded, selected in the ﬁrst step, is used to validate the performance on data from all the remaining participants with one or two focal motor seizures recorded. Thereby, the model is ﬁrst parameter-optimized in a leave-one-participant-out cross-validation on those training participants. Thus, each of the participants in this optimization set is omitted from the model training process once and used as a validation data set. The mean performance scores over the cross-validation runs are then used to determine the optimal parameter combination. In a second evaluation step, the optimized model, now trained with all the peri-ictal seizure data from the training subjects, is then applied to all the data from the test set participants. During the training of this model, the data are again ﬁrst standardized using the z-score method, and those normalization parameters are then applied to the incoming test data. Overall this results in a model trained and optimized on data from one set of participants, which is then tested on data from another separate set of participants. Sensors 2022, 22, 3318 6 of 20 2.4. Classiﬁcation Model The gradient tree boosting machine (GTBM) [39,40] methodology was chosen as the model used to detect ictal states, as it is relatively straightforward in its application and already validated on the same cohort, albeit on data from patients with convulsive seizures [7]. Due to this methodology’s requirement for parameter tuning to achieve good performance, hyperparameter optimization was conducted in both the intra- and inter-subject evaluations, as described above. For the optimization of the intra-subject model, an optimal parameter combination was found for each of the three included participants by performing a leave-one-seizure-out cross-validation. Thereby, the model was trained on the data of all but one seizure and the respective peri-ictal data, and tested on all remaining data including the left-out seizure for that participant, minus the training data. This was repeated for all seizures, and the performance scores were averaged. This procedure was then repeated for each combination of parameters. For the inter-subject evaluation, a similar procedure was implemented for the three selected participants, but in a leave-one-participant-out manner. The model with the best parameter combination was then tested on out-of-sample data from previously unseen participants. Four different model parameters were optimized: the learning rate, the maximum number of weak learners, the maximum tree depth per weak learner, and the misclassiﬁca- tion cost for false positives. Conversely, the misclassiﬁcation cost for false negatives was not tuned and kept unweighted, and only one type of boosting was used, namely adaptive boosting for binary classiﬁcation (“AdaBoost”) [41]. In total, the number of different pa- rameter combinations over which the grid search optimization was performed added up to 600. The best parameter combination was chosen as the one with the highest sensitivity and lowest number of false positives, in that order. In the case of a tie, the parameter combination with a higher learning rate or lower number of trees was chosen as the best one, as it would be computationally more efﬁcient. To gauge the signiﬁcance of the various features for the creation of the model, the predictor importance of each of the optimized models was analyzed. Therefore, the im- portance scores for each of the models resulting from the single cross-validation runs was averaged in the intra-subject leave-one-seizure-out evaluation, resulting in one set of scores per participant included there. Moreover, for the inter-subject leave-one-participant-out evaluation, only the importance scores of the optimal model, trained on all seizures from those intra-subject evaluation participants, were noted. The feature importance was based on decision tree node impurity, using the Gini diversity index and calculated such that the smallest possible value was 0 [42,43]. Thereby, the importance scores for each of the predictors are the averages over all the trained trees in the boosting ensemble for the GTBM model. 2.5. Performance Measurement The main indicators of performance used in this evaluation are the mean sensitivity, false alarm rate (FAR) per 24 h (FAR24) and positive predictive value (PPV). These scores are calculated from the number of overlaps of seizure events in the ground truth and predicted labels. The label data, analogous to the feature data, are stored at two-second intervals, and seizure events here are deﬁned as consecutive intervals of labels classiﬁed as a seizure of at least 6 s and at most 10 min. The prediction output of the classiﬁcation model is furthermore smoothed before this scoring computation, by ﬁlling out gaps between seizure labels of at most 30 s, and removing any orphan seizure labels. After this processing of the model output, it is compared to the ground truth and any overlaps of seizure events are counted as true positives, seizure events in the ground truth but not in the predictions are counted as false negatives, and vice versa; seizure events in the predictions that are not present in the ground truth are counted as false positives. Note that the comparisons described above are given a 2-min margin before and after seizure events in the ground truth, wherein overlaps with prediction events still count as true positives. This is performed to account Sensors 2022, 22, 3318 7 of 20 for some of the uncertainty related to seizure manifestations, as well as the in-hospital setting providing a certain degree of nurse intervention after a seizure. True negatives are not counted in this evaluation, as they do not give any more worthwhile information for performance measurement. We also calculated the false alarm rate per night (FARn), that is, during a standard eight-hour night between 23:00 and 07:00. Thereby, we counted how many of the false alarms produced by the model occur during that time period, and divide by the number of hours that were recorded during nights, taking into account any data loss that may have occurred during these hours. The FAR per night is therefore calculated as FARn = number of FP during night · hours per night nightly hours recorded . All data analysis, feature extraction, and performance evaluation was implemented using MATLAB R2021b (MathWorks, Natick, MA, USA). 2.6. Data Set Selection The complete data set of wearable data from 243 patients with epilepsy was ﬁltered to include only data relevant to the premise of this study. First, only data from those participants who had at least one focal seizure recorded that involved tonic or clonic motor manifestations were included. Thereby, these manifestations could co-occur with other seizure manifestations; however, focal to bilateral tonic–clonic seizures were excluded. Moreover, the data set was not ﬁltered further by overlap of symptom location versus device location. Therefore, the data set can, for example, include instances of motor seizures that manifest primarily on the right hand side, but where the wearable device was attached to the left wrist. Excluding these seizure instances would signiﬁcantly reduce the number of seizures and included participants for the analysis presented here, especially for the inter- subject evaluation, to the point of impracticality. During the study recordings, the Empatica E4 device was used in a Bluetooth streaming mode [25], which unfortunately led to a signiﬁcant loss of data due to regular problems with connectivity of the wearable device to a base device that stores the data. Thereby, more than 50% of the potential data to be recorded, and correspondingly as many potential seizures, were lost, leading to a signiﬁcantly reduced number of relevant focal motor seizures recorded for this study. Furthermore, the data set was ﬁltered for total length of recording per participant, where only those recordings with at least 24 h of data were included, and for length of seizures, where only those seizures with a duration between 10 s and 10 min are included. Limiting the duration of seizures excludes very short seizures of just a few seconds, such as myoclonic seizures, and very long seizures, such as status epilepticus. This is performed to exclude outliers and to have deﬁned limits of duration within which to detect potential seizure events. Lastly, the data quality during all remaining seizures was visually checked, and speciﬁcally those with bad EDA signal quality were excluded. A poor EDA signal can either be a ﬂat zero-line, indicating loss of contact of the electrodes with the skin, or multiple periods of high rates of amplitude change, indicating a loosely ﬁtting device. The BVP raw signal was speciﬁcally not ﬁltered for signal quality, ﬁrstly because it would ﬁlter out nearly every remaining seizure due to its high susceptibility to motion artifacts, and secondly because the feature set for this evaluation indeed includes a data quality index as a feature itself. Figure 2 visualizes the data set selection process, and Table A2 lists clinical and demographic information of the ﬁnally selected participants. Sensors 2022, 22, 3318 8 of 20 Figure 2. Data set ﬂowchart of the participant selection process. KCL: King’s College London; UKF: University Medical Center Freiburg; E4: Empatica E4 wrist-worn wearable device. 3. Results 3.1. Data Set and Examples The resulting data set used for this evaluation thus included 20 relevant seizures from a total of nine study participants. The participants were 44% female (4 of 9) and had a mean age of 45 years (range 9 to 69 years) at study enrollment. Three of these participants had more than two seizures recorded for a total of twelve seizures (Table A1), and the data from these was thus used for the intra-subject evaluation, as well as the training set for the leave-one-participant-out cross-validation in the inter-subject evaluation. The remaining six participants had either one or two seizures recorded, for a total of eight seizures included in the inter-subject evaluation test set. The mean recording length per participant in this data set was 83.2 h (range 35.8–127.6 h). To give a better overview of the three participants with multiple seizures recorded and used in the intra-subject evaluation, Figure 3 presents one example seizure for each of these participants. Participant UKF1 had six seizures recorded with the wearable device, all of them focal onset motor seizures with tonic and clonic manifestations, ictal tachycardia, and impaired awareness. Furthermore, all of these seizures occurred while the patient was sleeping in his hospital bed, and all have a characteristic progression. Overall, these seizure symptoms the closest to focal to bilateral or generalized tonic–clonic seizures in the data set, yet noticeably lack the severity of the larger seizures, both regarding the vEEG and also the movements captured with the ACC signal. UKF2, on the other hand, had three focal onset motor seizures recorded with only tonic manifestations, ictal tachycardia, and miscellaneous awareness during the seizure. Notably though, all seizures occurred while the patient was awake. Another important distinguishing factor for this participant is that he was only nine years old at the time of enrolment, and as such the only pediatric patient in the relevant data sets regarded here. Epilepsy in pediatric patients generally manifests in different ways than for adults. The sole data set from the KCL site, KCL1, had three seizures recorded that match the criteria for the seizure type. The motor manifestations for them were more heterogeneous than for the other two participants. All had tonic components, but there were also some oral automatisms, and one seizure also had clonic components. Furthermore, one seizure did not prompt ictal tachycardia, and there was a high variance between the seizure durations, with one being over two minutes and occurring while awake, and the other two only 22 s, occurring from sleep. Aside from the movements during the seizures, Figure 3 also gives a good overview of the typical EDA and BVP responses in the data, which can be observed most clearly in the ﬁrst presented seizure UKF1-4. The EDA Sensors 2022, 22, 3318 9 of 20 signal shows a clear response to the seizure, and the feature, the difference in the maximum of the skin conductance level, accordingly, is at its highest during the seizure. The heart rate estimated from the BVP sensor signal also clearly demonstrates some response after the seizure onset for all three examples; however, at the same time, the signal quality also drops signiﬁcantly, and as such the estimated heart rate should not be regarded as representative for these periods. (a) (b) (c) Figure 3. Selection of examples of true positive detections for each of the three participants in the intra-subject evaluation. Seizures shown are: (a) UKF1-4; (b) UKF2-3; (c) and KCL1-3 (see Table A1). Due to the grace period of 2 minutes around a seizure event, the detection for KCL1-3 counts as a true positive. Each plot of a seizure shows the raw ACC signal (top), the raw EDA signal and feature 2b (middle), and the estimated heart rate and signal quality index of the BVP signal (bottom). The regions highlighted in red mark the ground truth as labeled by experts, and those highlighted in green mark the seizure intervals, as predicted by the respective model, trained on the data of all the other seizures of the participant. The seizure onset and offset are additionally marked by the black vertical bars. All signals shown are normalized between −1 to 1 only for these plots. The original value ranges before normalization can be found in Table A3. Sensors 2022, 22, 3318 10 of 20 3.2. Intra-Subject Evaluation Data from three participants were selected for the intra-subject evaluation. One patient was selected from the London cohort with three seizures recorded (KCL1), and two from the Freiburg cohort with six (UKF1) and three (UKF2) seizures recorded, respectively. Out of these twelve seizures, only one seizure, for participant UKF2, could not be identiﬁed in the individual optimized leave-one-seizure-out evaluation. All other seizures were consistently detected in the complete participant data by the optimized models, when trained on the other seizures for the respective participant. In terms of false alarm rate however, the methodology exhibited vastly different performances over the three participants. The cross-validation runs for participant UKF1 showed the lowest number of false positives at just three on average, ranging from 1 to 5 depending on which of the seizures was left out for testing. Overall, this results in a low false alarm rate of less than one per 24 h (0.85/24 h). For the other two cases, the false alarm rate was considerably higher, at almost two per hour for UKF2 (41.5/24 h) and somewhat less than one per hour for KCL1 (17.7/24 h), on average. An overview of the per-participant results of this evaluation can be found in Table 1 under “Intra-Subject Evaluation”. Table 1. Evaluation results for the intra-subject leave-one-seizure-out evaluation, and the inter-subject leave-one-participant-out evaluation, respectively. Means and ranges are always across the single folds of the validations, that is, across the held-back data for the ﬁrst part and across the held-back data, and test set participants, in the second part. FP: false positive; FAR24: false alarm rate per 24 h; PPV: positive predictive value; FAR: false alarm rate; LOPO: leave-one-patient-out cross-validation. Patient ID Sensitivity Mean FP [Range] Mean FAR24 [Range] Mean PPV [Range] Mean FAR per Night [Range] Recording Duration Device on Same Hand as Seizure UKF1 100% (6/6) 3 [1–5] UKF2 67% (2/3) 79 [18–126] KCL1 100% (3/3) 37 [0–58] LOPO UKF1 LOPO UKF2 LOPO KCL1 LOPO test (N = 6) 50% (3/6) 100% (3/3) 67% (2/3) 75% (6/8) 28 124 1 55 [16–87] Intra-Subject Evaluation 0.85 [0.28–1.42] 41.52 [9.42–65.94] 17.69 [0–27.72] 28.3% [16.7–50%] 0.6% [0–1.1%] 34.5% [1.7–100%] Inter-Subject Evaluation 7.96 64.9 0.48 13.4 [4.4–22.7] 9.7% 2.4% 67% 2.1% [0–5.9%] 0 84.4 h 100% (6/6) 6.3 [1–11.5] 45.9 h 100% (3/3) 1.9 [0–3.0] 50.2 h 0% (0/3) 3.1 9.5 0 2.0 [0.7–3.2] 84.4 h 45.9 h 50.2 h 100% (6/6) 100% (3/3) 0% (0/3) 568.6 h 38% (3/8) 3.3. Inter-Subject Evaluation To assess the performance of seizure detection across multiple patients, the GTBM model is ﬁrst trained using the peri-ictal seizure data of the 12 seizures from the three participants mentioned above. The model is thereby parameter-optimized in a leave-one- participant-out manner, as explained in the Materials and Methods. In this cross-validation, the model with the best-performing parameter combination was able to recognize a total of eight of the twelve seizures (overall sensitivity 67%, mean 72%, and range 50–100%) in the validation set, with a mean false alarm rate of approximately one per hour, averaged over the three participants (mean 24.4/24 h; range 0.5/24 h–64.9/24 h). The resulting model is then applied to the complete data sets of six other participants, including a total of eight epileptic focal motor seizures. In this out-of-sample test set, the model was overall able to detect six of the eight seizures (overall sensitivity 75%, mean 75%, and range 0–100%) with a mean false alarm rate of 13.4 per 24 h (range 4.4/24 h–22.7/24 h). Table 1 shows a summary of these across-participant results under “Inter-Subject Evaluation”. Sensors 2022, 22, 3318 11 of 20 3.4. Feature Importance The feature importance for each of the three optimal models trained on data from the three per-subject evaluation participants was calculated as outlined in the Materials and Methods. Figure 4 shows these importance scores per participant and feature, and the mean scores of each feature group by modality. These feature scores are unitless and can be interpreted qualitatively to determine whether some speciﬁc feature or general modality is contributing more than the others. Here, the EDA features were more inﬂuential than the others in both the participants, UKF1 and KCL1. Conversely, the BVP features were unexpectedly more meaningful for the model of participant UKF2 than the other two modalities. The same kind of feature importance scores for the inter-subject model trained on all three of these participants for the inter-subject evaluation can be found in Figure 4d. Here, the EDA features are demonstrated to be more important than the others. (a) (c) (b) (d) Figure 4. Feature importance scores per intra-subject evaluation for the seizure detection models of the three selected participants: (a) UKF1; (b) UKF2; (c) KCL1 (see Table A2); (d) Feature importance scores of the model resulting from training the GTBM model on the seizure data of all three inter- subject training participants. Blue, red, and yellow bars show the importance scores for the features grouped by biosignal modality ACC, EDA, and BVP, respectively. Horizontal lines mark the mean scores of the groups. The ordinate is unitless; the scores can be interpreted qualitatively. The feature labels correspond to the listing of features in the Materials and Methods. 4. Discussion 4.1. Principal Findings The main ambition of the evaluation presented here was to qualitatively assess the utility of multimodal biosignal data from wearables in creating worthwhile and robust seizure detection systems. Thereby, two principal avenues of potential study design were investigated: intra-subject and inter-subject schemes. Speciﬁcally, we focused our Sensors 2022, 22, 3318 12 of 20 evaluation on focal motor seizures with tonic or clonic components, as opposed to bilateral tonic–clonic seizures. These focal seizures have a multitude of possible physical and psychological manifestations that can occur in sequence or in parallel and be repeated or not occur at all, in a single seizure. Furthermore, while there may oftentimes be little change in the semiology of seizures for a single patient with epilepsy, they can be very heterogeneous across populations [2,44]. These circumstances are also reﬂected in our results. Among the three participants with at least three seizures recorded, the individually optimized model could robustly recover the left-out seizures in the leave-one-seizure-out cross-validation for two participants. In one other participant, however, out of three seizures, one could not be restored by the model when trained on the other two (Table A1 and Figure 5). These three seizures had roughly the same semiology with tonic manifestations and ictal tachycardia. In the wearable data, however, one clear difference can be found between this seizure and the other two; that is, it had no discernable EDA response before, during, or after the seizure. Additionally, this participant UKF2 had an important demographic difference to all other included participants in this data set, in that they were the only pediatric patient at nine years old. Age has been linked, for example, to signiﬁcant changes in seizure semiologies [45]. These circumstances likely led to this speciﬁc seizure falling out of the scope of this methodology. Figure 5. Seizure UKF2-2, a false negative. Compare also to Figure 3. Data shown from top to bottom: raw ACC, raw EDA and feature 2b, heart rate and BVP signal quality index. The red overlay is the seizure ground truth. The seizure onset and offset are additionally marked by the black vertical bars. All signals shown are normalized between −1 to 1 only for these plots. The original value ranges before normalization can be found in Table A3. These results suggest that a methodology such as the one presented here, optimized on individual participants, can robustly detect seizures for some patients with epilepsy, but it may fail, especially when the seizures have differing semiologies that are not represented in the training data for the model. Furthermore, when looking at the false alarm rate per 24 h (FAR24) and positive predictive values (PPV), the heterogeneity of focal seizure detection is especially highlighted. The FAR24 performance of the seizure detection, ranging from less than one false positive (FP) per day to almost two FP per hour, is an important factor when it comes to actually applying the methodology to a real-world setting. Similarly, this false alarm rate also carries over to the nighttime, with multiple false positives per night for some participants. Thus, a high sensitivity in detecting seizures is in vain if an automated seizure diary is ﬁlled with dozens of false seizure events per day. Yet, further data recordings and model optimization may produce robust seizure detection systems for individual patients. With respect to the inter-subject evaluation across multiple study participants, the results for the methodology applied here further demonstrate the heterogeneity of the focal motor seizures in this data set, and clearly demonstrate the resulting difﬁculties. Inter-subject models applied in a leave-one-participant-out manner to data of the three Sensors 2022, 22, 3318 13 of 20 selected participants from the intra-subject evaluation perform worse than if trained in an individualized manner, at least either in terms of sensitivity or false alarm rates. Likewise, testing the model on out-of-sample data of six other participants resulted in a tolerable sen- sitivity but a high false alarm rate and low PPV. A model such as this would be ineffective in real-world settings, be it as an automated seizure diary or an alarm system, and patients’ and caregivers’ needs, in particular, would not be fulﬁlled [46–48]. Overall, the results so far suggest that not only are focal onset motor seizures varied in their clinical manifestations, they are also sensitive to changes in common wearable biosignal modalities, when investigated across patients. However, in some individual patients, seizure semiologies are similar enough across seizures to enable robust seizure detection models for less-severe focal onset motor seizures, if the models are optimized in a personalized manner. 4.2. Related Work To compare our results to some of the current and past state-of-the-art seizure detection studies, we compiled a list of twelve works featuring focal motor seizures in some form in their set of analyzed seizure types (Table 2). The main source of this list was the extensive literature review by Beniczky et al., 2021 [4], ﬁltered for relevant seizure types and relevant biosignal modalities that are most closely related to the modalities ACC, EDA, and PPG. Furthermore, to add some variety, we also included two recent studies employing wearable EMG- and EEG-based focal seizure detection. We aimed to compare all three of our performance measures, namely sensitivity, FAR24, and PPV; however, as most of these works include focal motor seizures as part of a larger group of seizure types, almost always dominated by generalized and focal to bilateral seizures, we were only able to ﬁnd three studies where this was possible for all three measures [9,11,20]. All three of these studies classify the focal onset seizures in their respective data sets as complex partial seizures (CPS), which is an older type of epileptic seizure classiﬁcation meaning an interval with ictal impaired awareness without giving information about motor manifestations during the seizures. Therefore it is unclear if, with respect to movements during the seizures, the seizure types investigated in these works are comparable to those in the study presented here. Moreover, all three studies evaluated their seizure detection in an inter-subject manner across a population of patients with epilepsy. Summarizing the relevant results from these three works, Cogan et al. 2017 [9] use a combination of two different wearable devices to record the biosignal modalities EDA, ECG, and SpO2, and report an algorithm sensitivity of 50% with a false alarm rate of 0.28 per day, in CPS only. Notably, during their analysis they also look into personalization of their algorithm, and conclude that a minimum of 6 to 8 seizures per patient would be required to sufﬁciently train and optimize their algorithm parameters, based on a worst-case scenario. Kusmakar et al. 2019 [11] employ ACC sensors in a leave-one-participant-out inter-subject evaluation and report a sensitivity of 67% and a FAR24 of 4, regarding the participants with CPS. They duly conclude that these seizures are much more similar to inter-ictal data than they are to ictal GTCS data, and as their data set includes only a very small number of CPS, a good performance on this seizure type is not to be expected. Lastly, Vandecasteele et al. 2017 [20] compared two wearable devices recording ECG and PPG, respectively, and reported an overall sensitivity of 32% with a FAR24 of 43.2 for the PPG-based wearable device, which was the same as the one used in this study. They conclude that the PPG- based detection was signiﬁcantly hindered by motion artifacts even for these possibly non-generalized seizures. Comparing these performances, and further results from other related work, which did not report their outcomes per seizure type or per participant (Table 2), in our results (Table 1), it becomes clear that a sensitivity of 75% in an independent test set of focal motor seizures is in fact among the top performing methodologies regarding this seizure type only. Furthermore, works that reported lower numbers of false positive rates also consistently had lower sensitivities, and some studies that reported similar numbers nevertheless still Sensors 2022, 22, 3318 14 of 20 had lower sensitivities. Overall, a comparison of our results to any of these works should be taken with caution, as it is to the best of our knowledge the ﬁrst work analyzing speciﬁcally focal motor seizures with multi-modal non-EEG wearable data. Table 2. Related work compiled from Beniczky et al., 2021 [4], and this study as comparison. Only those works are included that involve seizure types relevant to this study, that is, any of focal motor seizures, SPS, CPS, or other non-generalized seizures. FAR24: false alarm rate per 24 h; PPV: positive predictive value; FS: focal seizures; SPS: simple partial seizures; CPS: complex partial seizures; hyper: hypermotoric seizures; myo: myoclonic seizures; FS min mot: focal seizures with minimal motor component. Study Modalities Seizure Types # Pat. w/Seizures # Seizures Sensitivity FAR24 PPV this study [6] + [8] + [9] +,,§ [10] + [18] [11] +, [49] + [12] + [14] + [20] [50] [51] ACC, EDA, PPG ACC, PPG FS tonic or clonic 9 20 67–100%/75% 0.85–41.5/13.4 0.6– 34.5%/2.1% FS hyper/other major (28 total) 5/14 73%/84% Not reported per seizure type ECG FS/SPS/CPS/other (31 total) 8/26/31/5 Not reported per seizure type EDA, ECG, SpO2 ECG ACC ACC CPS 8 23 16.7%/50% 0.7/0.28 6.25%/50% SPS/CPS (16 total) 37/38 Tonic/tonic-clonic CPS 15 3 2 22 5 6 19%/71% 67%/100% 67% 50% Not reported per seizure type Not reported per seizure type 4.19 22.5 Not reported per seizure type ACC, EDA FS tonic-clonic ACC ACC, ECG Myo, tonic/FS hyper/FS min mot FS hyper/myo, tonic cluster ECG/PPG CPS EMG GTCS/tonic/clonic/ other motor EEG, ECG, ACC Focal tonic/focal nonmotor (41 total) 140 6%/24%/2% Not reported per seizure type 5/5 11 20 3 18/9 47 Not reported per seizure type 70%/32% 50.6/43.2 2.15%/1.12% 18/9/3/17 83%/56%/33%/76% - 83%/50% (t+c)/76% 47/9 + 9 84%/100% 8/13 + 5 - + The study also contained other seizure types, most notably generalized seizures, however the presented data only relates to those seizure types speciﬁcally mentioned. * Performance scores only include CPS, calculated by authors from original reported numbers. § Performance scores represent a non-optimized detection, and a reﬁned analysis, respectively. 4.3. Modality Importance Overall, the distributions of feature importance scores, as a proxy for the importance of biosignal modalities, seem to be heterogeneous and no clear winner can be found amongst the three examples. The features calculated from the EDA signal, however, seem to be the most informative with respect to epileptic seizure phases, within this data set of focal onset motor seizures. This is concurrent with some prior research on generalized seizures as well [52–55], and suggests that electrodermal activity could be an important clinical marker of epileptic seizures beyond highly convulsive episodes, which usually induce heavy sweating. However, it is not a universally applicable biomarker, as these results also suggest, with at least one participant and several seizures exhibiting no signiﬁcant (post-)ictal EDA response, as is also concluded in further literature on the topic [34,56,57]. In Glasstetter et al., 2021 [35], the authors explore the utility of wearable PPG signals for the detection of focal onset seizures with ictal tachycardia, and conclude that in some patients the tachycardia thresholds can be found regardless of seizure-related movements. It seems to be the case that in some seizures the ictal tachycardia presents itself some few seconds before the electrographic seizure onset, and could therefore be used as indicators for seizures, albeit not effectively in a monomodal system. The authors, however, also Sensors 2022, 22, 3318 15 of 20 note that arbitrary non-seizure-related movements may hinder this detection from PPG signals due to motion artifacts, and generally, pre-ictal tachycardia seems to be an isolated phenomenon not generalizable over patient cohorts. Our results, demonstrating that BVP features are important only for one of three patients with at least three seizures recorded, coincide with these conclusions. 4.4. Limitations The main limitation of the study and analysis presented here is the limited size of the data set, with just 20 seizures recorded from nine patients with epilepsy. Unfortunately, publicly available data sets focusing on classical wrist-worn wearable data which combine movement and autonomous nervous system biosignals are scarce as it is, and practically nonexistent in the ﬁeld of epilepsy research. To be considered a useful supplement to the data collected here, a public data set would need to provide at least ACC, EDA, and BVP data recorded from a wrist-worn wearable in a cohort of patients with epilepsy, and at least EEG seizure onset and offset would need to be labeled by experts. To the best of our knowledge, such a data set does not currently exist in an open source form. The relatively small size of the data set used here is also the main reason we used “classical” supervised machine learning with a set of speciﬁcally tailored features as opposed to some deep learning methodology. Deep learning may, however, be an avenue to pursue in a later study, with a more extensive data set. As the seizure types regarded here were carefully chosen, a large percentage of the overall recorded seizures were not eligible for inclusion. Furthermore, during their usually week-long stay at the epilepsy monitoring units, patients rarely have more than a few seizures in the ﬁrst place, as seizure provocation is often conducted only for a limited time until enough seizures have been observed to serve some clinical purpose. This leads to a generally low number of study participants with more than two seizures recorded, a necessary minimum requirement for any kind of analysis of personalized, intra-subject evaluations. Future research may include more seizures and seizure types from our already recorded data set, but new studies need to be conceptualized to aim for higher numbers of seizures recorded per participant, not just over the whole cohort, in order to push focal seizure detection research to produce better, more realistically applicable results [46]. Another limitation of this study, as well as many other related works, is the conﬁne- ment of participants to a hospital room throughout the data recording procedures. The data collected and analyzed here cannot be regarded as representative of real-world situations, and new ambulatory studies with a prospective goal of recording focal onset seizures with wearables from patients in their daily living environment are needed. However, even with generalized tonic–clonic seizures as a target, these phase 3 and 4 studies are still rare, and potential methodologies and pitfalls need ﬁrst be explored in these in-hospital studies before worthwhile out-of-hospital studies can be designed. 5. Conclusions Seizure detection for focal onset seizures without generalization by means of wear- able non-EEG devices is a so far little-researched problem. We demonstrate that for those seizures in this category with tonic or clonic movements, that is, those closest in semiol- ogy to generalized tonic–clonic seizures, robustly detecting seizures from wearable data may be possible for individual patients with epilepsy, depending on their speciﬁc seizure manifestation. Overall, electrodermal activity signals seem to provide the most informative features for seizure detection, suggesting it to be an essential part of any future seizure detection research. Detection across patients with purely inter-subject models without personalization is, however, not possible to a worthwhile degree, at least with the method- ology and data set presented here. Both a low sensitivity that misses a quarter or more of the seizures and high false alarm rates in the order of one per hour make current results of these inter-subject models ineffective for clinical applications. Our results thus demonstrate that individualized models are needed to robustly detect focal onset seizures, and future Sensors 2022, 22, 3318 16 of 20 research in this domain should include at least some degree of personalization in the modeling process. Author Contributions: Conceptualization, S.B., E.B., M.D., S.T., S.L., M.P.R. and A.S.-B.; methodology, S.B., M.D. and A.S.-B.; software, S.B. and M.G.; validation, S.B.; formal analysis, S.B.; investigation, S.B., E.B., N.E. and M.D.; resources, A.S.-B.; data curation, S.B., E.B. and N.E.; writing—original draft preparation, S.B.; writing—review and editing, S.B., M.D., N.Z., M.G., V.T., S.T., S.L., K.V.L., M.P.R. and A.S.-B.; visualization, S.B.; supervision, K.V.L., M.D. and A.S.-B.; project administration, M.P.R. and A.S.-B.; funding acquisition, M.P.R. and A.S.-B. All authors have read and agreed to the published version of the manuscript. Funding: This study received funding from the Innovative Medicines Initiative 2 Joint Undertaking under grant agreement No. 115902. This joint undertaking received support from the European Union’s Horizon 2020 Research and Innovation program and the European Federation of Pharmaceu- tical Industries and Associations. This communication reﬂects the views of the Remote Assessment of Disease and Relapse-Central Nervous System Consortium and neither Innovative Medicines Initiative nor the European Union and the European Federation of Pharmaceutical Industries and Associations are liable for any use that may be made of the information contained herein. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of the University of Freiburg (538/16, July 2017) and the London Fulham Research Ethics Committee (16/LO/2209, December 2016). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: Not applicable. Conﬂicts of Interest: A.S. receives research funding from the German Ministry of Science, European Union, National Institute of Health, and from the companies BIAL, Precisis, and UNEEG; is an advisory board member of SEER Medical; and has received honoraria for lectures or advice from Arvelle, BIAL, EISAI, GW, Precisis, and UCB. M.P.R. holds or co-holds research funding from the UK Medical Research Council, UK National Institute for Health Research, Wellcome Trust, UK Engineering and Physical Sciences Research Council, Epilepsy Research UK, Epilepsy Foundation of America, European Commission, Canadian Institutes of Health Research, Xenon Pharma, and GW Pharma; has research collaborations with UNEEG Medical, Seer Medical, UCB Pharma, ANT Neuro, and IMEC; is a Trustee of Epilepsy Research UK and an advisory board member of SUDEP Action; M.P.R. does not receive personal remuneration from any of these sources. M.P.R. holds a patent WO2013182848A1. V.T. is an employee of UCB Pharma. Appendix A Figure A1. The Empatica E4 wrist-worn wearable device used in this study (left), and the Android phone application that connects to the wearable via Bluetooth and records the data stream (right). Sensors 2022, 22, 3318 17 of 20 Table A1. Seizures recorded for the three participants used in the intra-subject evaluation. iTC: ictal tachycardia; UI: urinary incontinence. Seizure ID Seizure Duration Motor Symptoms Autonomic Symptoms Awareness Vigilance/Body Position UKF1-1 UKF1-2 UKF1-3 UKF1-4 UKF1-5 UKF1-6 UKF2-1 UKF2-2 * UKF2-3 KCL1-1 KCL1-2 KCL1-3 ID UKF1 UKF2 UKF3 UKF4 UKF5 UKF6 UKF7 UKF8 KCL1 82 s 86 s 55 s 73 s 43 s 47 s 23 s 39 s 107 s 128 s 22 s 22 s Tonic, clonic Tonic, clonic, myoclonic, automatisms (arms, legs) Tonic, clonic, myoclonic Tonic, clonic, myoclonic, automatisms (legs) Tonic, clonic Tonic, clonic, myoclonic Tonic Tonic Tonic Tonic, clonic, automatisms (arms, face) Tonic, automatisms (face) Tonic, automatisms (face) iTC iTC, UI iTC iTC iTC iTC iTC iTC, ﬂushing iTC, ﬂushing iTC iTC - Impaired Impaired Impaired Impaired Impaired Impaired Aware Impaired Impaired Impaired Impaired Impaired Asleep/lying Asleep/lying Asleep/lying Asleep/lying Asleep/lying Asleep/lying Awake/sitting Awake/lying Awake/sitting Awake/sitting Asleep/lying Asleep/lying * Seizure was not recognized by the model during evaluation. Table A2. Demographic and clinical information for the nine selected participants. TLE: temporal lobe epilepsy; FLE: frontal lobe epilepsy; xTLE: extratemporal lobe epilepsy. Gender Age Total Recording Duration # Seizures Recorded Epilepsy Origin Epilepsy Type m m f f m f m f m 55 9 27 69 50 34 48 46 65 84.4 h 45.9 h 92.0 h 120.8 h 127.6 h 35.8 h 105.1 h 87.2 h 50.2 h 6 3 2 1 2 1 1 1 3 Structural Structural Structural Structural Structural Unknown Structural Structural Structural Focal (TLE) Focal (xTLE) Focal (TLE) Focal (TLE) Focal (FLE) Focal (FLE) Focal (TLE) Focal (TLE) Focal (TLE) Table A3. The original value ranges for the plotted data in Figures 3 and 5, before normalization. ACC: accelerometry; EDA: electrodermal activity; BVP: blood volume pulse; HR: heart rate. Seizure ID Data Range ACC x [g] ACC y [g] ACC z [g] EDA Raw [µs] EDA Feature [a.u.] BVP HR [bpm] BVP Quality [a.u.] UKF1-4 UKF2-2 UKF2-3 KCL1-3 ictal non-ictal ictal non-ictal ictal non-ictal ictal non-ictal min max min max min max min max min max min max min max min max −1.7031 0.6094 −1.1563 0.9219 −1.5000 1.8906 −2.0000 1.9844 −2.0000 1.4219 −2.0000 1.9844 −0.9219 −0.7188 −1.4688 0.9063 −0.9688 0.2344 −1.3750 0.0625 −1.1406 1.9844 −2.0000 1.9844 −1.2500 1.7031 −2.0000 1.9844 0.2969 0.4219 0.0469 0.6875 −0.6406 1.9844 −0.5313 1.1875 −2.0000 1.0938 −2.0000 1.9844 −2.0000 1.9844 −2.0000 1.9844 0.3281 0.6875 −1.4688 1.9063 0.2602 5.7441 2.0375 11.1237 0.2770 0.6421 0.2975 0.8484 0.0051 0.3472 0.0000 0.6789 5.9464 6.3181 0.0000 51.4717 5.0397 7.8344 −4.4675 8.0093 −0.0152 −0.0152 −0.0385 0.0117 0.1558 0.2559 −0.2570 0.2959 33.3491 33.5858 −40.6902 33.7534 50.7644 121.0513 48.9167 137.2188 70.0432 102.9488 57.6646 119.5292 51.2767 124.4696 43.3582 147.5059 66.5748 136.6889 54.5837 118.9363 0.0917 0.2325 0.1241 0.3376 0.1664 0.2530 0.1367 0.32395 0.1242 0.2306 0.1185 0.312448 0.1346 0.1641 0.1157 0.3457851 References 1. 2. Thijs, R.D.; Surges, R.; O’Brien, T.J.; Sander, J.W. Epilepsy in adults. Lancet 2019, 393, 689–701. [CrossRef] Fisher, R.S.; Cross, J.H.; French, J.A.; Higurashi, N.; Hirsch, E.; Jansen, F.E.; Lagae, L.; Moshé, S.L.; Peltola, J.; Roulet Perez, E.; et al. Operational classiﬁcation of seizure types by the International League Against Epilepsy: Position Paper of the ILAE Commission for Classiﬁcation and Terminology. Epilepsia 2017, 58, 522–530. [CrossRef] [PubMed] Sensors 2022, 22, 3318 18 of 20 3. 4. 5. 6. 7. 8. 9. 10. Schulze-Bonhage, A.; Bruno, E.; Brandt, A.; Shek, A.; Heers, M.; Martinez-Lizana, E.; Altenmüller, D.M.; Richardson, M.P.; San Antonio, V. Diagnostic yield and limitations of in-hospital documentation in patients with epilepsy. Epilepsia 2022, in press. Beniczky, S.; Wiebe, S.; Jeppesen, J.; Tatum, W.O.; Brazdil, M.; Wang, Y.; Herman, S.T.; Ryvlin, P. Automated seizure detection using wearable devices: A clinical practice guideline of the International League Against Epilepsy and the International Federation of Clinical Neurophysiology. Epilepsia 2021, 62, 632–646. [CrossRef] [PubMed] Brinkmann, B.H.; Karoly, P.J.; Nurse, E.S.; Dumanis, S.B.; Nasseri, M.; Viana, P.F.; Schulze-Bonhage, A.; Freestone, D.R.; Worrell, G.; Richardson, M.P.; et al. Seizure Diaries and Forecasting With Wearables: Epilepsy Monitoring Outside the Clinic. Front. Neurol. 2021, 12, 1–14. [CrossRef] [PubMed] Arends, J.; Thijs, R.D.; Gutter, T.; Ungureanu, C.; Cluitmans, P.; Van Dijk, J.; Van Andel, J.; Tan, F.; De Weerd, A.; Vledder, B.; et al. Multimodal nocturnal seizure detection in a residential care setting A long-term prospective trial. Neurology 2018, 91, E2010–E2019. [CrossRef] Böttcher, S.; Bruno, E.; Manyakov, N.V.; Epitashvili, N.; Claes, K.; Glasstetter, M.; Thorpe, S.; Lees, S.; Dümpelmann, M.; Van Laerhoven, K.; et al. Detecting Tonic-Clonic Seizures in Multimodal Biosignal Data From Wearables: Methodology Design and Validation. JMIR mHealth uHealth 2021, 9, e27674. [CrossRef] Boon, P.; Vonck, K.; van Rijckevorsel, K.; Tahry, R.E.; Elger, C.E.; Mullatti, N.; Schulze-Bonhage, A.; Wagner, L.; Diehl, B.; Hamer, H.; et al. A prospective, multicenter study of cardiac-based seizure detection to activate vagus nerve stimulation. Seizure 2015, 32, 52–61. [CrossRef] Cogan, D.; Birjandtalab, J.; Nourani, M.; Harvey, J.; Nagaraddi, V. Multi-Biosignal Analysis for Epileptic Seizure Monitoring. Int. J. Neural Syst. 2017, 27, 1650031. [CrossRef] Fisher, R.S.; Afra, P.; Macken, M.; Minecan, D.N.; Bagi´c, A.; Benbadis, S.R.; Helmers, S.L.; Sinha, S.R.; Slater, J.; Treiman, D.; et al. Automatic Vagus Nerve Stimulation Triggered by Ictal Tachycardia: Clinical Outcomes and Device Performance—The U.S. E-37 Trial. Neuromodul. Technol. Neural Interface 2016, 19, 188–195. [CrossRef] 11. Kusmakar, S.; Karmakar, C.K.; Yan, B.; O’Brien, T.J.; Muthuganapathy, R.; Palaniswami, M. Automated Detection of Convulsive Seizures Using a Wearable Accelerometer Device. IEEE Trans. Biomed. Eng. 2019, 66, 421–432. [CrossRef] [PubMed] 12. Patterson, A.L.; Mudigoudar, B.; Fulton, S.; McGregor, A.; Poppel, K.V.; Wheless, M.C.; Brooks, L.; Wheless, J.W. SmartWatch by SmartMonitor: Assessment of Seizure Detection Efﬁcacy for Various Seizure Types in Children, a Large Prospective Single-Center Study. Pediatr. Neurol. 2015, 53, 309–311. [CrossRef] 13. Ulate-Campos, A.; Coughlin, F.; Gaínza-Lein, M.; Fernández, I.S.; Pearl, P.L.; Loddenkemper, T. Automated seizure detection 14. systems and their effectiveness for each type of seizure. Seizure 2016, 40, 88–101. [CrossRef] [PubMed] van Andel, J.; Ungureanu, C.; Arends, J.; Tan, F.; Van Dijk, J.; Petkov, G.; Kalitzin, S.; Gutter, T.; de Weerd, A.; Vledder, B.; et al. Multimodal, automated detection of nocturnal motor seizures at home: Is a reliable seizure detector feasible? Epilepsia Open 2017, 2, 424–431. [CrossRef] [PubMed] 15. Böttcher, S.; Manyakov, N.V.; Epitashvili, N.; Folarin, A.; Richardson, M.P.; Dümpelmann, M.; Schulze-Bonhage, A.; Van Laerhoven, K. Using multimodal biosignal data from wearables to detect focal motor seizures in individual epilepsy patients. In Proceedings of the 6th international Workshop on Sensor-based Activity Recognition and Interaction, Rostock, Germany, 16–17 September 2019; pp. 1–6. [CrossRef] 16. De Cooman, T.; Vandecasteele, K.; Varon, C.; Hunyadi, B.; Cleeren, E.; Van Paesschen, W.; Van Huffel, S. Personalizing Heart 17. Rate-Based Seizure Detection Using Supervised SVM Transfer Learning. Front. Neurol. 2020, 11, 1–13. [CrossRef] Jeppesen, J.; Fuglsang-Frederiksen, A.; Johansen, P.; Christensen, J.; Wüstenhagen, S.; Tankisi, H.; Qerama, E.; Hess, A.; Beniczky, S. Seizure detection based on heart rate variability using a wearable electrocardiography device. Epilepsia 2019, 60, 2105–2113. [CrossRef] 18. Kramer, U.; Kipervasser, S.; Shlitner, A.; Kuzniecky, R. A Novel Portable Seizure Detection Alarm System: Preliminary Results. J. Clin. Neurophysiol. 2011, 28, 36–38. [CrossRef] 19. Ryvlin, P.; Cammoun, L.; Hubbard, I.; Ravey, F.; Beniczky, S.; Atienza, D. Noninvasive detection of focal seizures in ambulatory patients. Epilepsia 2020, 61, 1–8. [CrossRef] 20. Vandecasteele, K.; De Cooman, T.; Gu, Y.; Cleeren, E.; Claes, K.; Van Paesschen, W.; Van Huffel, S.; Hunyadi, B. Automated epileptic seizure detection based on wearable ECG and PPG in a hospital environment. Sensors 2017, 17, 2338. [CrossRef] 21. Cook, M.J.; O’Brien, T.J.; Berkovic, S.F.; Murphy, M.; Morokoff, A.; Fabinyi, G.; D’Souza, W.; Yerra, R.; Archer, J.; Litewka, L.; et al. Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: A ﬁrst-in-man study. Lancet Neurol. 2013, 12, 563–571. [CrossRef] 22. Elger, C.E.; Hoppe, C. Diagnostic challenges in epilepsy: Seizure under-reporting and seizure detection. Lancet Neurol. 2018, 17, 279–288 . [CrossRef] 23. Hoppe, C.; Poepel, A.; Elger, C.E. Epilepsy: Accuracy of patient seizure counts. Arch. Neurol. 2007, 64, 1595–1599. [CrossRef] [PubMed] 24. Beniczky, S.; Ryvlin, P. Standards for testing and clinical validation of seizure detection devices. Epilepsia 2018, 59, 9–13. [CrossRef] 25. Bruno, E.; Böttcher, S.; Viana, P.F.; Amengual-Gual, M.; Joseph, B.; Epitashvili, N.; Dümpelmann, M.; Glasstetter, M.; Biondi, A.; Laerhoven, K.; et al. Wearable devices for seizure detection: Practical experiences and recommendations from the Wearables for Epilepsy And Research (WEAR) International Study Group. Epilepsia 2021, 62, 2307–2321. [CrossRef] Sensors 2022, 22, 3318 19 of 20 26. Ranjan, Y.; Rashid, Z.; Stewart, C.; Conde, P.; Begale, M.; Verbeeck, D.; Boettcher, S.; Hyve, T.; Dobson, R.; Folarin, A. RADAR-base: Open source mobile health platform for collecting, monitoring, and analyzing data using sensors, wearables, and mobile devices. J. Med. Internet Res. 2019, 21, e11734. [CrossRef] [PubMed] 27. Eckmann, J.P.; Kamphorst, S.O.; Ruelle, D. Recurrence Plots of Dynamical Systems. Europhys. Lett. 1987, 4, 973–977. [CrossRef] 28. Webber, C.L.; Zbilut, J.P. Dynamical assessment of physiological systems and states using recurrence plot strategies. J. Appl. Physiol. 1994, 76, 965–973. [CrossRef] 29. Großekathöfer, U.; Manyakov, N.V.; Mihajlovi´c, V.; Pandina, G.; Skalkin, A.; Ness, S.; Bangerter, A.; Goodwin, M.S. Automated detection of stereotypical motor movements in autism spectrum disorder using recurrence quantiﬁcation analysis. Front. Neuroinform. 2017, 11, 9. [CrossRef] 30. Lu, J.; Tong, K.Y. Robust Single Accelerometer-Based Activity Recognition Using Modiﬁed Recurrence Plot. IEEE Sens. J. 2019, 19, 6317–6324. [CrossRef] 31. Boucsein, W. Electrodermal Activity; Springer: New York, NY, USA, 2012; ISBN 978-1-4614-1125-3. 32. Posada-Quintero, H.F.; Chon, K.H. Innovations in Electrodermal Activity Data Collection and Signal Processing: A Systematic Review. Sensors 2020, 20, 479. [CrossRef] 33. Vieluf, S.; Amengual-Gual, M.; Zhang, B.; El Atrache, R.; Ufongene, C.; Jackson, M.C.; Branch, S.; Reinsberger, C.; Loddenkemper, T. Twenty-four-hour patterns in electrodermal activity recordings of patients with and without epileptic seizures. Epilepsia 2021, 62, 960–972. [CrossRef] [PubMed] 34. Vieluf, S.; Reinsberger, C.; El Atrache, R.; Jackson, M.; Schubach, S.; Ufongene, C.; Loddenkemper, T.; Meisel, C. Autonomic nervous system changes detected with peripheral sensors in the setting of epileptic seizures. Sci. Rep. 2020, 10, 1–8. [CrossRef] [PubMed] 35. Glasstetter, M.; Böttcher, S.; Zabler, N.; Epitashvili, N.; Dümpelmann, M.; Richardson, M.P.; Schulze-Bonhage, A. Identiﬁcation of Ictal Tachycardia in Focal Motor- and Non-Motor Seizures by Means of a Wearable PPG Sensor. Sensors 2021, 21, 6017. [CrossRef] [PubMed] Scholkmann, F.; Boss, J.; Wolf, M. An efﬁcient algorithm for automatic peak detection in noisy periodic and quasi-periodic signals. Algorithms 2012, 5, 588–603. [CrossRef] 36. 37. Nasseri, M.; Nurse, E.; Glasstetter, M.; Böttcher, S.; Gregg, N.M.; Laks Nandakumar, A.; Joseph, B.; Pal Attia, T.; Viana, P.F.; Bruno, E.; et al. Signal quality and patient experience with wearable devices for epilepsy management. Epilepsia 2020, 61, S25–S35. [CrossRef] 38. Bent, B.; Goldstein, B.A.; Kibbe, W.A.; Dunn, J.P. Investigating sources of inaccuracy in wearable optical heart rate sensors. NPJ Digit. Med. 2020, 3, 18. [CrossRef] Friedman, J.H. Greedy function approximation: A gradient boosting machine. Ann. Stat. 2001, 29, 1189–1232. [CrossRef] 39. 40. Natekin, A.; Knoll, A. Gradient boosting machines, a tutorial. Front. Neurorobot. 2013, 7, 21. [CrossRef] 41. Freund, Y.; Schapire, R.E. A Decision-Theoretic Generalization of On-Line Learning and an Application to Boosting. J. Comput. Syst. Sci. 1997, 55, 119–139. [CrossRef] 42. Lemon, S.C.; Roy, J.; Clark, M.A.; Friedmann, P.D.; Rakowski, W. Classiﬁcation and regression tree analysis in public health: Methodological review and comparison with logistic regression. Ann. Behav. Med. 2003, 26, 172–181. [CrossRef] 43. MathWorks. Estimates of Predictor Importance for Classiﬁcation Ensemble of Decision Trees—MATLAB. 2022. Available online: https://mathworks.com/help/stats/compactclassiﬁcationensemble.predictorimportance.html (accessed on 21 March 2022). 44. Cragar, D.E.; Berry, D.T.; Fakhoury, T.A.; Cibula, J.E.; Schmitt, F.A. A review of diagnostic techniques in the differential diagnosis 45. of epileptic and nonepileptic seizures. Neuropsychol. Rev. 2002, 12, 31–64. [CrossRef] [PubMed] Fernandez-Baca Vaca, G.; Mayor, C.L.; Losarcos, N.G.; Park, J.T.; Lüders, H.O. Epileptic seizure semiology in different age groups. Epileptic Disord. 2018, 20, 179–188. [CrossRef] [PubMed] 46. Bruno, E.; Viana, P.F.; Sperling, M.R.; Richardson, M.P. Seizure detection at home: Do devices on the market match the needs of 47. 48. people living with epilepsy and their caregivers? Epilepsia 2020, 61, S11–S24. [CrossRef] [PubMed] Simblett, S.K.; Biondi, A.; Bruno, E.; Ballard, D.; Stoneman, A.; Lees, S.; Richardson, M.P.; Wykes, T. Patients’ experience of wearing multimodal sensor devices intended to detect epileptic seizures: A qualitative analysis. Epilepsy Behav. 2020, 102, 106717. [CrossRef] [PubMed] van Westrhenen, A.; Souhoka, T.; Ballieux, M.E.; Thijs, R.D. Seizure detection devices: Exploring caregivers’ needs and wishes. Epilepsy Behav. 2021, 116, 107723. [CrossRef] 49. Onorati, F.; Regalia, G.; Caborni, C.; Migliorini, M.; Bender, D.; Poh, M.Z.Z.; Frazier, C.; Kovitch Thropp, E.; Mynatt, E.D.; Bidwell, J.; et al. Multicenter clinical assessment of improved wearable multimodal convulsive seizure detectors. Epilepsia 2017, 58, 1870–1879. [CrossRef] 50. Baumgartner, C.; Whitmire, L.E.; Voyles, S.R.; Cardenas, D.P. Using sEMG to identify seizure semiology of motor seizures. Seizure 2021, 86, 52–59. [CrossRef] 51. Munch Nielsen, J.; Zibrandtsen, I.C.; Masulli, P.; Lykke Sørensen, T.; Andersen, T.S.; Wesenberg Kjær, T. Towards a wearable multi-modal seizure detection system in epilepsy: A pilot study. Clin. Neurophysiol. 2022, 136, 40–48. [CrossRef] Sensors 2022, 22, 3318 20 of 20 52. Heldberg, B.E.; Kautz, T.; Leutheuser, H.; Hopfengartner, R.; Kasper, B.S.; Eskoﬁer, B.M. Using wearable sensors for semiology- independent seizure detection-Towards ambulatory monitoring of epilepsy. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS, Milan, Italy, 25–29 August 2015; pp. 5593–5596. [CrossRef] 53. Onorati, F.; Regalia, G.; Caborni, C.; LaFrance, W.C.; Blum, A.S.; Bidwell, J.; De Liso, P.; El Atrache, R.; Loddenkemper, T.; Mohammadpour-Touserkani, F.; et al. Prospective Study of a Multimodal Convulsive Seizure Detection Wearable System on Pediatric and Adult Patients in the Epilepsy Monitoring Unit. Front. Neurol. 2021, 12, 1–15. [CrossRef] 54. Poh, M.Z.; Loddenkemper, T.; Reinsberger, C.; Swenson, N.C.; Goyal, S.; Sabtala, M.C.; Madsen, J.R.; Picard, R.W. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor. Epilepsia 2012, 53, e93–e97. [CrossRef] 55. Regalia, G.; Onorati, F.; Lai, M.; Caborni, C.; Picard, R.W. Multimodal wrist-worn devices for seizure detection and advancing research: Focus on the Empatica wristbands. Epilepsy Res. 2019, 153, 79–82. [CrossRef] [PubMed] 56. Bruno, E.; Böttcher, S.; Biondi, A.; Epitashvili, N.; Manyakov, N.V.; Lees, S.; Schulze-Bonhage, A.; Richardson, M.P. Post-ictal 57. accelerometer silence as a marker of post-ictal immobility. Epilepsia 2020, 61, 1397–1405. [CrossRef] [PubMed] Sarkis, R.A.; Thome-Souza, S.; Poh, M.Z.; Llewellyn, N.; Klehm, J.; Madsen, J.R.; Picard, R.; Pennell, P.B.; Dworetzky, B.A.; Loddenkemper, T.; et al. Autonomic changes following generalized tonic clonic seizures: An analysis of adult and pediatric patients with epilepsy. Epilepsy Res. 2015, 115, 113–118. [CrossRef] [PubMed]
10.5152_tjg.2023.22590	Identification of Hub Genes of NAFLD-Related HCC Liu et al. 34 4 ORIGINAL ARTICLE LIVER Identification of Hub Genes and Immune Infiltration in Non-alcoholic Fatty Liver Disease -Related Hepatocellular Carcinoma by Bioinformatics Analysis Xu Liu , Yan Wang , Tao Li , Yundong Qu Department of Infectious Diseases and Hepatology, The Second Hospital of Shandong University, Jinan, Shandong, China Cite this article as: Liu X, Wang Y, Li T, Qu Y. Identification of hub genes and immune infiltration in non-alcoholic fatty liver disease -related hepatocellular carcinoma by bioinformatics analysis. Turk J Gastroenterol. 2023;34(4):383-393. ABSTRACT Background: Non-alcoholic fatty liver disease has been a significant risk factor for hepatocellular carcinoma. In the study, we aimed to identify the key genes associated with the transition from non-alcoholic fatty liver disease to hepatocellular carcinoma through bioin- formatics analysis. Methods: The GSE164760 dataset was used for identifying differentially expressed genes. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analysis were performed to explore the potential function of the differentially expressed genes. Subsequently, the protein–protein interaction network was constructed to select hub genes, and the immune cell infiltration was ana- lyzed. Finally, the receiver operating characteristic analysis was performed to assess the diagnostic ability of the crucial genes. Results: A total of 156 differentially expressed genes were identified. Gene Ontology enrichment analysis indicated that differentially expressed genes were strongly associated with cellular hormone metabolic process, response to xenobiotic stimulus, collagen-contain- ing extracellular matrix, detoxification, and regulation of growth. In the protein–protein interaction network, ESR1, CAT, CXCL8, CD4, SPP1, CYP2E1, CYP3A4, UGT2B7, GSTA1 and THBS1 were selected as the hub genes. Immune infiltration analysis demonstrated that M0 macrophages, plasma cells, CD8+T cell and M2 macrophages were significantly changed in tumor tissues. Finally, we verified the hub gene expression and selected CD4, UGT2B7, and CYP3A4 as the potential diagnostic biomarkers. Conclusion: CD4, UGT2B7, and CYP3A4 were selected as the potential diagnostic biomarkers of non-alcoholic fatty liver disease–hepa- tocellular carcinoma. Keywords: Non-alcoholic fatty liver disease, hepatocellular carcinoma, bioinformatics analysis, hub genes INTRODUCTION Non-alcoholic fatty liver disease (NAFLD) comprises a spectrum of diseases varying from liver steatosis to non- alcoholic steatohepatitis (NASH); the latter can progress to cirrhosis and become an emerging risk factor for hepa- tocellular carcinoma (HCC).1-3 Although the current inci- dence of NAFLD-associated HCC is still lower than other etiologies such as hepatitis B, its prevalence is expected to increase with the obesity and metabolic syndrome epi- demic.4 Therefore, NAFLD-induced HCC is entirely wor- thy of clinical attention. Non-alcoholic fatty liver disease-associated HCC, which often occurs in the absence of cirrhosis, is less likely to be detected than HCC arising from other etiologies.5,6 The majority of patients with NAFLD-HCC are diagnosed at advanced unresectable stage requiring systemic treat- ment. Currently, immunotherapy has been approved for NAFLD-HCC. However, emerging evidence indicates that HCC arising from NAFLD might be less responsive to immunotherapy.7,8 Both tumor and tumor immune microenvironment (TME) factors determine the poor response.9 Therefore, more studies are needed to explore the mechanism and identify more effective diagnostic biomarkers for NAFLD-HCC. In this study, we downloaded the NASH-HCC microar- ray gene expression profile (GSE164760) from the Gene Expression Omnibus (GEO) database for bioinformat- ics analysis. By comparing the gene expression between NASH tissues and NASH-HCC tissues, we screened out differentially expressed genes (DEGs) and constructed protein–protein interaction (PPI) network for selecting hub genes. Furthermore, we performed the immune cell infiltration analysis and verified the hub genes in another dataset (GSE164441). Finally, we discovered that CD4, Corresponding author: Yundong Qu and Tao Li, e-mail: [email protected] and [email protected] Received: August 19, 2022 Accepted: February 13, 2023 Publication Date: April 17, 2023 DOI: 10.5152/tjg.2023.22590 383 Copyright @ Author(s) – Available online at https://www.turkjgastroenterol.org.Content of this journal is licensed under a Creative Commons Attribution (CC BY) 4.0 International LicenseTurk J Gastroenterol 2023; 34(4): 383-393 Liu et al. Identification of Hub Genes of NAFLD-Related HCC UGT2B7, and CYP3A4 may have the potential to serve as diagnostic biomarkers in the progression of NASH to HCC. The results may provide novel biomarkers for NAFLD-related HCC diagnosis and facilitate the develop- ment of targeted therapeutics. Immune Cell Infiltration The CIBERSORT algorithm was used to estimate the immune cell composition fractions.15 The algorithm was running with LM22 at 1000 permutations. Boxplots and stacked histograms were used to visualize the results of immune cell infiltration. MATERIALS AND METHODS Data Collection and Preprocessing (GSE16476010 and The datasets for NAFLD-HCC GSE16444111) were downloaded from the GEO (https:// www.n cbi.n lm.ni h.gov /geo/ ) database by R version 4.2.0 using the “GEOquery” R package. The dataset GSE164760 includes 53 NASH-HCC samples and 74 NASH samples. The dataset GSE164441 contains 10 paired NAFLD-HCC tissues and adjacent non-tumor liver tissues. We filtered out probes without corresponding gene symbols. For multiple probes annotated to the same gene, we retained the gene randomly. Differentially Expressed Gene Screening The “limma” R package was used to select DEGs between NASH-HCC samples and NASH samples, as previously described. Threshold values were set as P <.01 and \|log2FoldChange\|>1. The “ggplot2” and “pheatmap” pack- ages were applied to visualize the significant DEGs. Functional and Pathway Enrichment Analysis Metascape (http://metascape.org) was used for Gene Ontology (GO) enrichment analysis.12 The “clusterProfiles” R package was used for Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. Protein–Protein Interaction Network The STRING database (https://string-db.org/) is used to predict and construct PPI networks between the candi- date genes.13 The DEGs were uploaded to the STRING website to obtain the PPIs. The PPI network information was imported into Cytoscape software.14 CytoHubba was used to identify the top 10 hub genes with high degree. Main Points • Non-alcoholic fatty liver disease (NAFLD) has become an important risk factor for hepatocellular carcinoma (HCC). • M0 macrophages, plasma cells, CD8+T cells, and M2 macro- phages were significantly changed in NAFLD-HCC tissues. CD4, UGT2B7, and CYP3A4 were selected as the potential diagnostic biomarkers of NAFLD-HCC. • Receiver Operating Characteristic Analysis The receiver operating characteristic (ROC) analysis was performed, and area under the curve (AUC) was calcu- lated to assess the diagnostic ability of the key genes using the “pROC” package. Statistical Analysis In this study, statistical analysis was analyzed with the R version 4.2.0. Student’s t-tests or Wilcoxon rank-sum tests were used for analyzing the differences between the 2 groups. Correlations were performed by Pearson or Spearman’s analysis. P < .05 was defined as statistically significant. RESULTS Gene Expression and Enrichment Analysis Revealed Differentially Expressed Genes in Non-alcoholic Steat ohepa titis –Hepa tocel lular Carcinoma Tissues and Their Association with the Process of Material Metabolism and Regulation of Growth Based on 53 NASH-HCC samples and 74 NASH sam- ples, a total of 156 DEGs from the GSE164760 data- set were acquired, including 43 upregulated and 113 downregulated DEGs in NASH-HCC tissues (Table 1). Heatmap clearly showed the landscape of genomic dif- ferences between NASH tissues and NASH-HCC tissues (Figure 1A). The volcano plot was drawn to display the dis- tribution of DEGs between NASH-HCC tissues and NASH tissues (Figure 1B). The results of GO enrichment analysis indicated that DEGs were strongly associated with cel- lular hormone metabolic process, response to xenobiotic stimulus, collagen-containing extracellular matrix, detox- ification, and regulation of growth (Figure 1C and 1D). The KEGG pathway demonstrated that those genes were cor- related with drug metabolism–cytochrome P450, retinol metabolism, and chemical carcinogenesis–DNA adducts. (Figure 1E). Protein–Protein Interaction Network Established for Hub Gene Screening The PPI network was constructed to study the relation- ship among those DEGs at the protein level (Figure 2A). 384 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Turk J Gastroenterol 2023; 34(4): 383-393 Table 1. DEGs in NASH-HCC Samples Compared with NASH Samples DEGs Upregulated Gene Name IGFBP1, SPP1, CXCR4, CXCL8, CDH13, SPINK1, CDKN2A, AKR1B10, LCN2, CHI3L1, EGR1, EFEMP1, APOD, ANXA2P2, C1orf198, GEM, KDM5D, VCAN, TXN, PLVAP, SULF1, ZIC2, SIK1, CAP2, FAT1, SNORA71D, FOSB, MUC13, ANXA2, THBS1, COL4A1, ITGA6, YWHAZ, CCN2, MAP2, TXNRD1, SPRY1, COL15A1, OLFML2B, STC1, MAP1B, GPC3, NSMCE4A Downregulated CLEC4G, FCN2, VIPR1, CLEC4M, STAB2, FCN3, CRHBP, CXCL14, SLC19A3, CD5L, GLYAT, ACACB, GNAO1, IGFBP3, DNASE1L3, ACAA1, IL13RA2, FAM13A, GHR, CD4, NR1I2, DGAT2, C1RL-AS1, COLEC10, CYP39A1, OBSL1, CLTRN, ADRA1A, CNDP1, NAT2, SERPINA4, GSTZ1, BCO2, SKAP1, SULT1E1, LIFR, F11, RAPH1, MT1F, FTCD, KCNN2, LYVE1, GLUD1, RABEP1, AFM, FXYD1, ACSM3, CDHR2, THRSP, ZG16, GASK1A, CXCL12, CYP2C9, CYP2C8, CYP2C19, PPP1R1A, HSD17B13, GSTA1, AGBL2, PLAC8, ADH4, APOA1, MT1H, SRD5A2, ADH1A, GBA3, RDH16, CYP1A2, MT1G, PHGDH, PZP, PBLD, BCHE, BHMT, ESR1, HPX, PPP1R3C, HAMP, GNMT, ATF5, CAT, SLC22A7, FYB2, MT1JP, INHBE, GSK3B, SERPINA5, PLIN1, TRO, HAO2, PPP1R3B, SLC38A4, RHEB, IL16, ANXA10, ALDOB, MT1M, SLC22A1, CYP2E1, UGT2B7, GLS2, HPGD, FBP1, CYP3A4, ADH1C, C6, PXMP2, MT1X, NCAM2, ADH1B, C9, CFHR3, XIST A total of 43 upregulated DEGs and 113 downregulated DEGs were identified in NASH-HCC tissues, compared with NASH tissues. The hub genes are shown in boldface. DEGs, differentially expressed genes; HCC, hepatocellular carcinoma; NASH, non-alcoholic steatohepatitis. The top 10 genes (ESR1, CAT, CXCL8, CD4, SPP1, CYP2E1, CYP3A4, UGT2B7, GSTA1, and THBS1) with the highest degree in the above network were selected as the hub genes (Figure 2B). We then investigated the correlation between these hub genes (Figure 2C). The relationship among hub genes showed that SPP1, CXCL8, and THBS1 had a negative co-relationship with the other genes. In addition, the strongest correlation was found between CXCL8 with THBS1 (correlation coefficient = 0.73) and between UGT2B7 with ESR1 (correlation coefficient = 0.73). Then, the expression pattern of these hub genes was analyzed (Figure 2D). The expression levels of SPP1, CXCL8, and THBS1 were higher in NASH-HCC tissues than those in NASH tissues, while the expression lev- els of ESR1, CAT, CD4, CYP2E1, CYP3A4, UGT2B7, and GSTA1 were lower in NASH-HCC tissues than those in NASH tissues. Immune Cell Infiltration Analysis Showed the Different Proportion of M0 Macrophages, M2 Macrophages, Plasma Cells, and CD8+T Cells Between Non-alcoholic Steat ohepa titis –Hepa tocel lular Carcinoma Tissues and Non-alcoholic Steatohepatitis Tissues In order to reveal the immune microenvironment of NASH and NASH-HCC tissues, the CIBERSORT algorithm was used to analyze specific immune cell types that infiltrated into NASH and NASH-HCC tissues. The proportion of 22 immune cells in NASH and NASH-HCC tissues is shown in Figure 3A and 3B. In NASH tissues, the top 3 categories were M2 macrophages, regulatory T cells (Tregs), and rest- ing mast cells. In NASH-HCC tissues, Tregs, resting mast cells, and M2 macrophages ranked in the top 3. Figure 3C and 3D show that the proportion of immune cells varied among samples in the NASH group and the NASH-HCC group. M2 macrophages and Tregs comprised the largest 385 Turk J Gastroenterol 2023; 34(4): 383-393 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Figure 1. Gene expression and enrichment analysis revealed DEGs in NASH-HCC tissues and their association with the process of material metabolism and regulation of growth. (A) Heatmap of all DEGs in NASH tissues and NASH-HCC tissues. (B) Volcano map of all DEGs when comparing NASH-HCC tissues to NASH tissues. Red indicates upregulated genes, blue represents downregulated genes, and grey means genes that were not differentially expressed. (C) Heatmap of GO enrichment terms of DEGs in Metascape, colored by P-value. (D) Network of GO enrichment terms, colored by cluster ID. (E) The dot plot of the top 10 terms of KEGG enrichment analysis. DEG, differentially expressed genes; GO, Gene Ontology; HCC, hepatocellular carcinoma; KEGG, Kyoto Encyclopedia of Genes and Genomes; NASH, non- alcoholic steatohepatitis. 386 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Turk J Gastroenterol 2023; 34(4): 383-393 Figure 2. Identification of hub genes. (A) Construction of the PPI network of DEGs. The color depth and shape size of the nodes are positively correlated with degree. (B) The top 10 hub genes at protein level. (C) The correlation between the 10 hub genes. Positive correlation was marked with red and negative with blue. (D) The expression of 10 hub genes in NASH tissues and NASH-HCC tissues in GSE164760. DEG, differentially expressed genes; HCC, hepatocellular carcinoma; NASH, non-alcoholic steatohepatitis; PPI, protein–protein interaction. 387 Turk J Gastroenterol 2023; 34(4): 383-393 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Figure 3. Immune cell infiltration analysis showed the different proportion of M0 macrophages, M2 macrophages, plasma cells, and CD8+T cells between NASH-HCC tissues and NASH tissues. (A) Composition of 22 immune cells in 74 NASH tissues. (B) Composition of 22 immune cells in 53 NASH-HCC tissues. (C) Fractions of 22 immune cells in 74 NASH tissues. (D) Fractions of 22 immune cells in 53 NASH-HCC tissues. (E) Comparisons of 22 immune cells between NASH tissues and NASH-HCC tissues. HCC, hepatocellular carcinoma; NASH, non-alcoholic steatohepatitis. 388 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Turk J Gastroenterol 2023; 34(4): 383-393 proportion of immune cell subtypes in NASH samples, and Tregs and resting mast cells occupied the highest proportion in NASH-HCC tissues. Figure 3E shows the different immune cell occupancies in NASH samples and NASH-HCC samples. M0 macrophages, plasma cells, and CD8+T cells were higher in NASH-HCC tissues, while M2 macrophages were higher in NASH tissues. Then, the correlation matrix was drawn to visualize the interaction of 22 immune cells. As shown in Figure 4A, by further analyzing the CIBERSORT scores, M2 mac- rophages had the highest negative relationship with M0 macrophages, while resting mast cells had the highest positive correlation with M0 macrophages. Furthermore, we explored the relationship between the 10 hub genes and the significantly changed immune cells. As shown in Figure 4B, M2 macrophages showed a consistent corre- lation among the 10 hub genes. Moreover, we analyzed the relationship between the 10 hub genes and 8 immune checkpoint genes, including “CD274,” “CTLA4,” “PDCD1,” “PDCD1LG2,” “CD86,” “CD80,” “CD276,” and “VTCN1” (Figure 4C). UGT2B7 showed the highest negative corre- lation with CTLA4 (R = –0.47). Hub Gene Verification and Diagnostic Ability Assessment Indicated That CD4, UGT2B7, and CYP3A4 may be the Potential Diagnostic Biomarkers of Non-alcoholic Fatty Liver Disease–Hepatocellular Carcinoma To verify the expression of the 10 hub genes in NAFLD- HCC, we downloaded the dataset GSE164441 (RNA- Seq). As it is shown in Figure 5A, of the 10 hub genes, only the expression levels of CD4, UGT2B7, and CYP3A4 were significantly different in the dataset GSE164441, and all the 3 genes were downregulated in tumor tissues. Then, the ROC analysis was performed to evaluate the sensitivity and specificity of CD4, UGT2B7, and CYP3A4 for the diagnosis of NAFLD-related HCC. In the dataset GSE164760, the AUC of CD4, UGT2B7, and CYP3A4 was 0.913, 0.838, and 0.757, respectively (Figure 5B). In the verification dataset GSE164441, the AUC values of CD4, UGT2B7, and CYP3A4 were 0.805, 0.816, and 0.666, respectively (Figure 5C). The results indicated that CD4, UGT2B7, and CYP3A4 may be potential diagnostic bio- markers of NAFLD-HCC. DISCUSSION In the study, we obtained 156 DEGs between NASH tis- sues and NASH-HCC tissues and discovered that those genes were related to the process of material metabolism and regulation of growth. Subsequently, 10 hub genes were selected via the PPI network. The results of immune infiltration analysis showed that M2 macrophages, Tregs, and resting mast cells comprised the largest proportion of immune cell subtypes both in NASH samples and in NASH-HCC samples. The proportion of M0 macrophages, plasma cells, and CD8+T cells in NASH-HCC tissues is sig- nificantly increased compared with NASH tissues; in con- trast, the proportion of M2 macrophages in NASH-HCC tissues is significantly decreased. Then, we verified hub genes in another dataset, and only the expression of CD4, UGT2B7, and CYP3A4 was significantly different, which may be due to the differences in detection methods and samples. Finally, we found the 3 genes could serve as the prospective markers for NAFLD-to-HCC transition via ROC analysis. Tumor immune microenvironment plays a significant role in the transition from NAFLD to HCC. Immunotherapy such as immune checkpoint inhibitors holds great prom- ise for advanced HCC; however, recent data suggested that patients with NAFLD-associated HCC are less sen- sitive to conventional immune checkpoint inhibition due to the altered immune components.16-18 Studies have reported that the activation of intrahepatic CD8+T cells promotes the NASH-to-HCC transition.19,20 Pfister et al7 revealed that aberrantly activated CD8+PD1+ T cells in the TME involve in NASH-induced liver injury and pro- gression of NASH to HCC, which may limit the response to immunotherapy. Macrophages include 3 subtypes (M0, M1, and M2), and each subtype has different functions in tumorigenesis and development. Previous studies have reported that M0 macrophage enrichment is associated with poor prognosis and sorafenib response in HCC.21,22 M2 macrophages serve a significant role in facilitat- ing cancer initiation, angiogenesis of tumor stroma, and immunosuppression.23-25 In our study, although the pro- portion of M2 macrophages in NASH-HCC tissues was lower than that in NASH tissues, its proportion in the NASH-HCC tissues was still significantly higher than that of other immune cell types. CD4 molecule plays an important role in the develop- ment, differentiation, activation, and antigen recognition of T cells. CD4+T cells exert an important role in tumor immune surveillance. The absolute number of CD4+T cells is decreased in the inflamed liver and NAFLD-promoted HCC.20,26 A study suggested that selective loss of CD4+T lymphocytes in NAFLD promotes hepatocarcinogenesis.27 It is also reported that CD4+T cells inhibit the initiation of HCC and contribute to tumor regression.28,29 389 Turk J Gastroenterol 2023; 34(4): 383-393 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Figure 4. Correlation of hub genes and immune cells. (A) The correlation of 22 immune cells. Red represents positive correlation and blue represents negative correlation. Lower panel was Pearson’s correlation coefficient. In the upper panel, the size of the circle is positively correlated with the correlation coefficient. (B) The correlation between each hub gene and changed immune cells. Red, positive; blue, negative. (C) Correlation between the 10 hub genes and 8 immune checkpoint genes. 390 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Turk J Gastroenterol 2023; 34(4): 383-393 Figure 5. Hub gene verification and ROC analysis indicated that CD4, UGT2B7, and CYP3A4 may be the potential diagnostic biomarkers of NAFLD-HCC. (A) The differential expression of CD4, UGT2B7, and CYP3A4 in paired tumor tissues and adjacent tissues of GSE164441. (B) ROC analysis of CD4, UGT2B7, and CYP3A4 in GSE164760. (C) ROC analysis of CD4, UGT2B7, and CYP3A4 in GSE164441. HCC, hepatocellular carcinoma; NAFLD, non-alcoholic fatty liver disease; ROC, receiver operating characteristic. 391 Turk J Gastroenterol 2023; 34(4): 383-393 Liu et al. Identification of Hub Genes of NAFLD-Related HCC UGT2B7 (UDP glucu ronos yltra nsfer ase family 2 mem- ber B7) is a member of the UGT2B family, is mainly expressed in the liver, and is involved in biotransfor- mation and detoxification. The altered expression of UGT2B7 may affect the toxicity and efficacy of drugs.30 It was reported that the expression of UGT2B7 was downregulated in HCC, and the impaired function was closely related to the hepatocarcinogenesis.31 In addi- tion, UGT2B7 was strongly associated with microvascu- lar invasion in HCC.32 CYP3A4 (cytochrome P450 family 3 subfamily A member 4), mainly expressed in the liver and intestine, is involved in the majority of drug metabolism, participating in the metabolism of some pre-carcinogens.33,34 The down- regulation of CYP3A4 might influence the metabolism of some pre-carcinogens, which might be associated with carcinogens. CYP3A4 was reported as a potential diagno- sis and prognosis biomarker for HCC.35-37 However, this study has some limitations that need to be focused in future work. First, we only used 1 gene expres- sion profile for differential gene analysis, which may result in selection bias. Second, the study was based on the bio- informatics analysis of the GEO dataset and experiments should be conducted next to validate the results. Besides, we did not analyze the clinicopathological parameters due to the lack of clinical information. In summary, by bioinformatics analysis, we identified and verified the hub genes associated with NAFLD-to-HCC transition and found that CD4, UGT2B7, and CYP3A4 may serve as the potential biomarkers of NAFLD-HCC. This recognition may provide new into the understanding of the pathogenesis of NAFLD-to-HCC transition. insight Ethics Committee Approval: GEO belong to public databases. The patients involved in the database have obtained ethical approval. Users can download relevant data for free for research and publish relevant articles. As our study is based on open source data, ethical approval is not necessary. Informed Consent: N/A. Peer-review: Externally peer-reviewed. Author Contributions: Concept – X.L., T.L.; Design – Y.Q., X.L.; Supervision – Y.W.; Materials – X.L.; Data Collection and/or Processing – X.L.; Analysis and/or Interpretation – Y.Q., T.L.; Literature Review – T.L., Y.Q., Y.W.; Writing – X.L.; Critical Review – Y.Q., T.L., Y.W. Declaration of Interests: The authors have no conflict of interest to declare. Funding: This study received no funding. REFERENCES 1. Powell EE, Wong VW-S, Rinella M. Non-alcoholic fatty liver disease. Lancet. 2021;397(10290):2212-2224. [CrossRef] 2. Eslam M, Valenti L, Romeo S. Genetics and epigenetics of NAFLD and NASH: clinical impact. J Hepatol. 2018;68(2):268-279. [CrossRef] 3. Friedman SL, Neuschwander-Tetri BA, Rinella M, Sanyal AJ. Mech- anisms of NAFLD development and therapeutic strategies. Nat Med. 2018;24(7):908-922. [CrossRef] 4. Huang DQ, El-Serag HB, Loomba R. Global epidemiology of NAFLD-related HCC: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol. 2021;18(4):223-238. [CrossRef] 5. Pais R, Fartoux L, Goumard C, et al. Temporal trends, clinical pat- terns and outcomes of NAFLD-related HCC in patients undergoing liver resection over a 20-year period. Aliment Pharmacol Ther. 2017; 46(9):856-863. [CrossRef] 6. Piscaglia F, Svegliati-Baroni G, Barchetti A, et al. Clinical patterns of hepatocellular carcinoma in nonalcoholic fatty liver disease: a multicenter prospective study. Hepatology. 2016;63(3):827-838. [CrossRef] 7. Pfister D, Núñez NG, Pinyol R, et al. NASH limits anti-tumour sur- veillance in immunotherapy-treated HCC. Nature. 2021;592(7854): 450-456. [CrossRef] 8. Chan LL, Chan SL. Novel perspectives in immune checkpoint inhibitors and the management of non-alcoholic steat ohepa titis -rela ted hepatocellular carcinoma. Cancers (Basel). 2022;14(6):1526. [CrossRef] 9. Anstee QM, Reeves HL, Kotsiliti E, Govaere O, Heikenwalder M. From NASH to HCC: current concepts and future challenges. Nat Rev Gastroenterol Hepatol. 2019;16(7):411-428. [CrossRef] 10. Pinyol R, Torrecilla S, Wang H, et al. Molecular characterisation of hepatocellular carcinoma in patients with non-alcoholic steatohep- atitis. J Hepatol. 2021;75(4):865-878. [CrossRef] 11. Yang W, Feng Y, Zhou J, et al. A selective HDAC8 inhibitor poten- tiates antitumor immunity and efficacy of immune checkpoint blockade in hepatocellular carcinoma. Sci Transl Med. 2021;13(588). [CrossRef] 12. Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist- oriented resource for the analysis of systems-level datasets. Nat Commun. 2019;10(1):1523. [CrossRef] 13. Szklarczyk D, Gable AL, Lyon D, et al. STRING v11: protein-protein association networks with increased coverage, supporting func- tional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019;47(D1):D607-D613. [CrossRef] 14. Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software envi- ronment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498-2504. [CrossRef] 15. Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453- 457. [CrossRef] 16. Hindson J. T cells in NASH and liver cancer: pathology and immunotherapy. Nat Rev Gastroenterol Hepatol. 2021;18(6):367. [CrossRef] 17. O’Leary K. T cell drivers in NASH-HCC. Nat Rev Cancer. 2021; 21(6):341. [CrossRef] 392 Liu et al. Identification of Hub Genes of NAFLD-Related HCC Turk J Gastroenterol 2023; 34(4): 383-393 18. Albano E, Sutti S. The paradox role of cytotoxic T-lymphocytes in NAFLD-associated hepatocellular carcinoma. Hepatobiliary Surg Nutr. 2021;10(5):705-707. [CrossRef] 19. Wolf MJ, Adili A, Piotrowitz K, et al. Metabolic activation of intra- hepatic CD8+ T cells and NKT cells causes nonalcoholic steatohepa- titis and liver cancer via cross-talk with hepatocytes. Cancer Cell. 2014;26(4):549-564. [CrossRef] 20. Yang YM, Kim SY, Seki E. Inflammation and liver cancer: molecu- lar mechanisms and therapeutic targets. Semin Liver Dis. 2019; 39(1):26-42. [CrossRef] 21. Hsiao YW, Chiu LT, Chen CH, Shih WL, Lu TP. Tumor-infiltrating leukocyte composition and prognostic power in hepatitis B- and hepatitis C-related hepatocellular carcinomas. Genes (Basel). 2019;10(8). [CrossRef] 22. Farha M, Jairath NK, Lawrence TS, El Naqa I. Characterization of the tumor immune microenvironment identifies M0 macrophage- enriched cluster as a poor prognostic factor in hepatocellular carci- noma. JCO Clin Cancer Inform. 2020;4:1002-1013. [CrossRef] 23. Hume DA. The many alternative faces of macrophage activation. Front Immunol. 2015;6:370. [CrossRef] 24. Pollard JW. Tumour-educated macrophages promote tumour progression and metastasis. Nat Rev Cancer. 2004;4(1):71-78. [CrossRef] 25. Bingle L, Brown NJ, Lewis CE. The role of tumour-associated macrophages in tumour progression: implications for new antican- cer therapies. J Pathol. 2002;196(3):254-265. [CrossRef] 26. Levite M, Safadi R, Milgrom Y, Massarwa M, Galun E. Neurotrans- mitters and Neuropeptides decrease PD-1 in T cells of healthy sub- jects and patients with hepatocellular carcinoma (HCC), and increase their proliferation and eradication of HCC cells. Neuropeptides. 2021;89:102159. [CrossRef] 27. Ma C, Kesarwala AH, Eggert T, et al. NAFLD causes selective CD4(+) T lymphocyte loss and promotes hepatocarcinogenesis. Nature. 2016;531(7593):253-257. [CrossRef] 28. Corthay A, Skovseth DK, Lundin KU, et al. Primary antitumor immune response mediated by CD4+ T cells. Immunity. 2005;22(3): 371-383. [CrossRef] 29. Rakhra K, Bachireddy P, Zabuawala T, et al. CD4(+) T cells con- tribute to the remodeling of the microenvironment required for sus- tained tumor regression upon oncogene inactivation. Cancer Cell. 2010;18(5):485-498. [CrossRef] 30. Shen ML, Xiao A, Yin SJ, et al. Associations between UGT2B7 polymorphisms and cancer susceptibility: a meta-analysis. Gene. 2019;706:115-123. [CrossRef] 31. Lu L, Zhou J, Shi J, et al. Drug-metabolizing activity, protein and gene expression of UDP-g lucur onosy ltran sfera ses are significantly altered in hepatocellular carcinoma patients. PLoS One. 2015; 10(5):e0127524. [CrossRef] 32. Beaufrère A, Caruso S, Calderaro J, et al. Gene expression signa- ture as a surrogate marker of microvascular invasion on routine hepatocellular carcinoma biopsies. J Hepatol. 2022;76(2):343-352. [CrossRef] 33. Ingelman-Sundberg M. Human drug metabolising cytochrome P450 enzymes: properties and polymorphisms. Naunyn Schmiede- bergs Arch Pharmacol. 2004;369(1):89-104. [CrossRef] 34. Kamdem LK, Meineke I, Gödtel-Armbrust U, Brockmöller J, Wojnowski L. Dominant contribution of P450 3A4 to the hepatic carcinogenic activation of aflatoxin b1. Chem Res Toxicol. 2006; 19(4):577-586. [CrossRef] 35. Ashida R, Okamura Y, Ohshima K, et al. CYP3A4 gene is a novel biomarker for predicting a poor prognosis in hepatocellular carci- noma. Cancer Genomics Proteomics. 2017;14(6):445-453. [CrossRef] 36. Liu J, Han F, Ding J, et al. Identification of multiple hub genes and pathways in hepatocellular carcinoma: a bioinformatics analysis. BioMed Res Int. 2021;2021:8849415-. [CrossRef] 37. Mao J-X, Zhao Y-Y, Dong J-Y, et al. UBE2T and CYP3A4: hub genes regulating the transformation of cirrhosis into hepatocellular carcinoma. All Life. 2021;14(1):509-521. [CrossRef] 393
10.1242_jcs.261100	© 2023. Published by The Company of Biologists Ltd \| Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 RESEARCH ARTICLE GIPC3 couples to MYO6 and PDZ domain proteins, and shapes the hair cell apical region Paroma Chatterjee1, Clive P. Morgan1,, Jocelyn F. Krey1, Connor Benson1,‡, Jennifer Goldsmith1, Michael Bateschell1, Anthony J. Ricci2 and Peter G. Barr-Gillespie1,§ ABSTRACT GIPC3 has been implicated in auditory function. Here, we establish that GIPC3 is initially localized to the cytoplasm of inner and outer hair cells of the cochlea and then is increasingly concentrated in cuticular plates and at cell junctions during postnatal development. Early postnatal Gipc3KO/KO mice had mostly normal mechanotransduction currents, but had no auditory brainstem response at 1 month of age. Cuticular plates of Gipc3KO/KO hair cells did not flatten during development as did those of controls; moreover, hair bundles were squeezed along the cochlear axis in mutant hair cells. Junctions between inner hair cells and adjacent inner phalangeal cells were also severely disrupted in Gipc3KO/KO cochleas. GIPC3 bound directly to MYO6, and the loss of MYO6 led to altered distribution of GIPC3. Immunoaffinity purification of GIPC3 from chicken inner ear extracts identified co-precipitating proteins associated with adherens junctions, intermediate filament networks and the cuticular plate. Several of immunoprecipitated proteins contained GIPC family consensus PDZ-binding motifs (PBMs), including MYO18A, which bound directly to the PDZ domain of GIPC3. We propose that GIPC3 and MYO6 couple to PBMs of cytoskeletal and cell junction proteins to shape the cuticular plate. KEY WORDS: Hair cell, Cuticular plate, Actin, Myosin, Cytoskeleton INTRODUCTION The PDZ (PSD95, DLG1, ZO1) protein GIPC3 is located in sensory hair cells and spiral ganglion neurons, and is necessary for auditory function in mice (Charizopoulou et al., 2011) and humans (Charizopoulou et al., 2011; Rehman et al., 2011). GIPC3 belongs to the GIPC family, which also includes GIPC1 and GIPC2; GIPC proteins interact with key transmembrane proteins, such as receptor tyrosine kinases, G-protein-coupled receptors, and integrins, and play roles in their trafficking, signaling and internalization (Katoh, 2013). GIPC proteins contain a single PDZ domain that is flanked by an N-terminal GIPC-homology 1 1Oregon Hearing Research Center & Vollum Institute, Oregon Health & Science University, Portland, OR 97239, USA. 2Department of Otolaryngology—Head & Neck Surgery, Stanford University, Stanford, CA 94305, USA. Present address: Thermo Fisher Scientific, Eugene, OR 97402, USA. ‡Present address: Stanford Medicine Children’s Health, Stanford University, Stanford, CA 94305, USA. §Author for correspondence ([email protected]) C.P.M., 0000-0001-8451-6837; C.B., 0009-0008-6592-7241; P.G.B.-G., 0000- 0002-9787-5860 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. Handling Editor: Giampietro Schiavo Received 1 March 2023; Accepted 13 April 2023 (GH1) domain and a C-terminal GH2 domain (Katoh, 2013). In the absence of ligand, GIPC1 resides in an autoinhibited configuration, where sites that interact with other proteins are masked (Shang et al., 2017). The GIPC1 PDZ domain binds to many PDZ-binding motifs (PBMs), which are typically C-terminal tails of interacting proteins (Katoh, 2013); binding to PBMs then releases autoinhibition, allowing GH2 to bind to other proteins (Shang et al., 2017). MYO6 is a well-characterized interacting partner for GIPC1 (Naccache et al., 2006; Reed et al., 2005; Shang et al., 2017). MYO6 is an unconventional myosin motor that is highly expressed in hair cells and is essential for their function (Avraham et al., 1995, 1997; Hasson et al., 1997); preliminary evidence suggests that MYO6 also interacts with GIPC3 (Shang et al., 2017). MYO6 interacts with GH2 of GIPC1 (Naccache et al., 2006), so when proteins with PBMs bind to the GIPC1 PDZ domain, the GIPC1 GH2 domain is released and can bind to MYO6. The original ahl5 allele identified for Gipc3, which contains a missense mutation in the PDZ domain, caused elevated auditory brainstem response (ABR) thresholds but not profound deafness (Charizopoulou et al., 2011). By contrast, the International Mouse Phenotyping Consortium (IMPC) (Brown et al., 2018) reported that a full Gipc3-knockout (KO) mouse was profoundly deaf (www. mousephenotype.org), prompting us to examine the location and role of GIPC3 in hair cell function in more detail. We developed monoclonal antibodies against GIPC3; we used them to reveal that GIPC3 was initially located in the cytoplasm of cochlear inner hair cells (IHCs), then increasingly concentrated at cell–cell junctions (Corwin and Warchol, 1991) and the cuticular plate (Pollock and McDermott, 2015) as development proceeded. Using the IMPC Gipc3-KO mouse line, we found that cuticular plates of Gipc3KO/KO hair cells were significantly disrupted, appearing rounder than those of controls, and cell–cell junctions were significantly disordered; Gipc3KO/KO mice also had stereocilia organization defects. Later in development, stereocilia begin to fuse together in Gipc3KO/KO hair cells, eventually producing giant stereocilia. This phenotype resembled the phenotype of Myo6KO/KO hair cells (Self et al., 1999), suggesting that GIPC3 and MYO6 might be in the same pathway; indeed, GIPC3 and MYO6 interacted directly. Although MYO6 was not mislocalized in Gipc3KO/KO hair cells, GIPC3 no longer localized to cuticular plates and junctional regions of Myo6KO/KO hair cells. One of the anti-GIPC3 monoclonal antibodies was used to immunoaffinity purify GIPC3 and its complexes, and the results suggested that GIPC3 associated with actin-rich networks associated with cell junctions, including those of MYO6, MYH9, MYH10 and MYO18A. Several co-immunoprecipitated proteins were predicted to have high-affinity PBM interactions with the GIPC family, including APPL2, MYO18A, ACTN1 and ACTN4. We demonstrated that the GH2 domain of GIPC3 interacts with MYO6, and that the PDZ domain of GIPC3 interacts with MYO18A through its C-terminal PBMs. We propose that GIPC3 interacts with MYO6 to couple apical junctions to 1 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 the cuticular plate, which is necessary for proper anchoring of stereocilia. RESULTS GIPC3 is enriched in hair cells and localizes to apical regions To determine the distribution of GIPC3 in the inner ear, we examined published mass spectrometry datasets that used data- dependent acquisition (DDA), where relatively abundant tryptic peptides detected in the first stage of analysis (MS1) are selected for fragmentation and sequencing in the second stage (MS2). Using a mass spectrometry dataset that included mouse utricle and utricle hair bundle results (Krey et al., 2015), we noted that GIPC3 was enriched in bundles as compared to the whole epithelium. We also analyzed a dataset that reported mass spectrometry of isolated cochlea and utricle cells from Pou4f3-Gfp mice (Krey et al., 2018); these mice express GFP exclusively in hair cells (Scheffer et al., 2015). Following FACS sorting of dissociated cochlea and utricle cells, GIPC3 was readily detected in GFP-positive cells; GIPC3 was present both in cochlear (Fig. 1A) and utricle (Fig. 1B) hair cells, but was at low levels in GFP-negative cells. These results show that GIPC3 was highly enriched in hair cells. We antibodies developed monoclonal against GIPC3, immunizing mice with a mixture of recombinant mouse and chicken GIPC3 proteins; we chose antibodies designated as 6B4, 3A7 and 10G5 for further characterization of GIPC3. Using these antibodies, which were validated in Gipc3-null cochleas (see below), GIPC3 was detected in the cytoplasm of IHCs and outer hair cells (OHCs) and was modestly enriched in the cuticular plate at early developmental ages; it was particularly concentrated near apical junctions and the circumferential actin belt (arrows in Fig. 1C), in a region called the pericuticular necklace (Hasson et al., 1997). When introduced into hair cells using adeno-associated virus (AAV) transduction via in utero electroporation, GFP-GIPC3 localized to similar locations (Fig. 1D); junctions was notable its concentration at apical (arrows in Fig. 1D,E). GIPC3 detected by any of the three antibodies was found in cuticular plates and at apical junctions throughout development (Fig. 1F–I; Fig. S2) but was increasingly concentrated apically by post-natal day (P)15.5 (Fig. 1J–L). Lattice structured illumination microscopy (SIM) showed a punctate pattern in the cuticular plate and strong labeling at apical junctions (Fig. 1M, arrows). Fig. 1. GIPC3 in the cochlea and utricle. (A) Mass spectrometry quantification from cochlea using DDA. For A and B, green corresponds to Pou4f3-GFP- positive cells (hair cells), whereas black corresponds to GFP-negative cells; riBAQ measures relative molar abundance, and mean±range are plotted (n=2 each; biological replicates). (B) Mass spectrometry quantification from utricle using DDA. (C) Localization of GIPC3 in P3.5 cochlear hair cells. OHC, outer hair cell. IHC, inner hair cell. Arrows point to concentration of GIPC3 at the periphery of the cuticular plate, at or adjacent to the plasma membrane in the pericuticular necklace region. (D,E) Localization of GFP–GIPC3 introduced into inner (D) and outer (E) hair cells using AAV. Reslice images below are the X-Z images from the transects shown in yellow in the upper panels. Arrows point to concentration of GFP–GIPC3 at the pericuticular necklace region. (F–M) Developmental progression of GIPC3 labeling in IHCs (F–H,J–M) and OHCs (I). M is a lattice SIM image; arrows indicate GIPC3 located at circumferential actin ring of the displayed IHC. AAV experiments were performed twice; immunolocalization experiments were performed more than three times. Panel widths: C, 17.5 µm; D,E, 10 µm; F–I, 15 µm; J–-M, 10 µm. Superresolution modality: C–L, Airyscan; M, lattice SIM. 2 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Gipc3KO/KO mice have reduced auditory function We permanently deleted Gipc3 exons 2 and 3 and the neomycin cassette from Gipc3tm1a(KOMP)Wtsi mice using the Cre deleter strain (Schwenk et al., 1995); here we refer to the resulting tm1b allele, a global Gipc3 knockout, as Gipc3KO. The signal for monoclonal antibodies 6B4 and 3A7 was lost completely in Gipc3KO/KO cochlea and utricle hair cells (Fig. S1A–F). Most 10G5 signal was lost in Gipc3KO/KO mice; a very low level of immunoreactivity remained in the junctional region (Fig. S1G,H) and resembled that of GIPC1 immunostaining (Giese et al., 2012). The IMPC has previously reported elevated auditory brainstem response (ABR) measurements (www.mousephenotype.org); we replicated those results, showing not only that Gipc3KO/KO mice were profoundly deaf at 5–6 weeks of age, but that there was a modest threshold elevation in heterozygotes (Fig. 2A). We found that mechanoelectrical transduction (MET) currents at P10 were of similar maximum amplitudes in Gipc3KO/+ and Gipc3KO/KO hair cells (Fig. 2B,C). Given the increased number of stereocilia rows, we expected larger currents from Gipc3KO/KO, but noticed that the fluid jet was less effective at moving their hair bundles. Larger stimulations led to separation of rows and loss of MET currents. By using an offset stimulus, we were able to evoke larger currents from the Gipc3KO/KO bundles without causing damage. These data were quantified as the ratio of maximum currents after the offset stimulus to that before the stimulus (Fig. 2B,D). The larger currents suggest that the additional rows of to stimulate. Application of mechanical stimuli while the cell was subjected to voltage steps allowed us to construct current-voltage (I-V ) relationships for Gipc3KO/+ and Gipc3KO/KO MET (Fig. 2E); I-V curves for the two genotypes did not differ (Fig. 2F), suggesting that the MET channel properties were unaffected by the lack of GIPC3. stereocilia were functional, albeit harder FM1-43 dye a proxy for labeling is often used as mechanotransduction (Meyers et al., 2003). Dye loading was reduced by ∼65% in wild-type littermates using the transduction channel blocker tubocurarine, consistent with most dye entering hair cells via the transduction channels; the reduction was less (∼50%) in heterozygotes, and dye labeling in Gipc3KO/KO hair cells was only slightly reduced (Fig. S1I–O). These results are consistent with an increased number of transduction channels and an increased open probability at rest. Distorted hair bundles in Gipc3KO/KO IHCs and OHCs We next examined the morphology of apical IHCs in Gipc3KO/KO mice. Although hair bundles of P2.5 Gipc3KO/KO mutant hair cells resembled those of heterozygous controls (Fig. 2G,K), by P6.5 bundles appeared to be squeezed along the cochlear longitudinal axis (Fig. 2H,L). This compression was also apparent by scanning electron microscopy (SEM) at P8.5 (Fig. 2Q,R). As development progressed, bundles became more disorganized and abnormal (Fig. 2I,J,M,N); after P14.5, fused and elongated stereocilia were often seen (Fig. 2P), resembling those seen in Myo6-null mutants (Self et al., 1999). IHC stereocilia for Gipc3KO/KO mice were gathered into a wedge shape, rather than a curled block, and diameters of row 3 stereocilia were greater than those in control bundles (Fig. 2S,U). OHC stereocilia shapes were similar in Gipc3KO/KO and controls, but the wings of the OHC bundles were considerably closer together in the mutants (Fig. 2T,V). TRIOBP labeling (Fig. 2W) also demonstrated that the squeezing distortion seen in bundles was due to altered positions of the stereocilia rootlets; rootlets of row 1 stereocilia in both IHCs and OHCs were reproducibly closer together (Fig. 2X). GIPC3 interacts directly with MYO6 Because the paralog GIPC1 directly binds to MYO6, we examined whether GIPC3 also interacts with MYO6. MYO6 was present in cochlear and vestibular hair cells at levels well above those of GIPC3 (Fig. 1A,B). In immunolocalization experiments with IHCs from P6.5 and P15.5 C57BL/6 mice, MYO6 and GIPC3 had overlapping distributions, although MYO6 was relatively more concentrated at apical junctions and the pericuticular region, and GIPC3 was more concentrated in the cuticular plate, especially at later ages (Fig. 3A,B). We used glutathione-S-transferase (GST) pulldown experiments to confirm that GIPC3 and MYO6 interact directly. GIPC1 interacts with the helical cargo-binding domain (HCBD) of MYO6 (Shang et al., 2017), so we used either HCBD–GST or GST alone to precipitate HA-tagged GIPC3 or GIPC3 fragments containing the GIPC3 PDZ and GH2 domains or the PDZ domain alone (Fig. 3C,D). HCBD-GST precipitated full-length GIPC3, albeit only a small amount (red arrow in Fig. 3D); because both GH1 and GH2 domains are present, likely in the autoinhibited state and incapable of binding efficiently to MYO6 (Shang et al., 2017). Deletion of the GH1 domain allowed much more strong binding to HCBD–GST, but eliminating the GH2 domain abolished the MYO6 interaction (Fig. 3D). We conclude that the GH2 domain of GIPC3 interacts with HCBD of MYO6, especially when autoinhibition is relieved. this molecule is We exploited the NanoSPD method (Bird et al., 2017) to provide additional evidence for the GIPC3–MYO6 interaction (Fig. 3E–G). In these experiments, we expressed MYO10NANOTRAP, a construct that targets filopodia tips of HeLa cells because of its MYO10 motor and binds to GFP fusion proteins through its nanotrap-anti-GFP single- chain antibody (Bird et al., 2017). Constructs used in NanoSPD experiments are listed in Table S1. Co-expressed GFP-tagged constructs, like the MYO6 tails, bound to MYO10NANOTRAP and were also transported to filopodial tips. Proteins tagged with a fluorescent protein that does not bind to MYO10NANOTRAP, for example mCherry, can then be co-transported to filopodial tips if the GFP-tagged and mCherry-tagged proteins interact. As a positive control, we confirmed that, when MYO10NANOTRAP was co-expressed, GFP–MYO7ATAIL enabled transport of mCherry–PDZD7 to filopodial tips (Fig. S2A,B) as previously reported (Morgan et al., 2016). We compared the tail domains from three splice forms of MYO6 that were predicted in Ensembl (http://useast.ensembl.org/Mus_ musculus/Gene/Summary?db=core;g=ENSMUSG00000033577; r=9:80072313-80219011): Myo6-212 (its protein product is referred to here as MYO6-A), Myo6-201 (MYO6-B) and Myo6-203 (MYO6-C); these three splice forms respectively correspond to the large-insert, small-insert and no-insert splice forms previously reported (Buss et al., 2001). The two splice sites used for these isoforms flank the HCBD domain, raising the possibility that splicing regulates GIPC3–MYO6 interaction. Nevertheless, when co-expressed with MYO10NANOTRAP, each of the three splice forms of GFP-MYO6TAIL transported mCherry–GIPC3 to filopodial tips of HeLa cells (Fig. 3E–G; Fig. S2C–E). mCherry–GIPC3 was located in the cytoplasm of HeLa cells when expressed alone (Fig. S2I). When expressed with MYO10NANOTRAP, neither GFP–MYO7ATAIL (Fig. S2F) nor GFP alone (Fig. S2G) facilitated targeting of mCherry–GIPC3 to filopodial tips. The elevation of GFP–MYO6TAIL signal at filopodial tips was significant for each of the three MYO6 constructs compared to the 3 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 2. Characterization of Gipc3-knockout mice. (A) ABR thresholds for Gipc3 genotypes at 5–6 weeks (mean±s.e.m.). Gipc3KO/KO completely lacked responses, preventing statistical testing. Unpaired two-tailed t-test P-values for Gipc3KO/+ comparison to Gipc3+/+ were: 4 kHz, 0.0045; 8 kHz, 0.0061; 12 kHz, 0.0207; 16 kHz, 0.0103; 24 kHz, 0.1097; 32 kHz, 0.0751. Sample sizes (n; individual ears) were (respectively Gipc3+/+, Gipc3KO/+ and Gipc3KO/KO): 4 kHz, 8 kHz, 16 kHz, 24 kHz, and 32 kHz (8, 28, 18); 12 kHz and 24 kHz (6, 26, 8). (B–F) Mechanoelectrical transduction from P10 IHCs; holding potential of −84 mV. (B) MET in response to sinusoidal displacement bursts, interceded with a positive adapting step. (C) Maximum MET current. Sample sizes were n=7 (Gipc3KO/+) or 9 (Gipc3KO/KO); mean±s.e.m. plotted. Unpaired two-tailed t-tests with P-values as indicated in figure (applies to D as well). (D) Ratio of average responses from second sinusoidal burst divided by responses from first burst (n=7 each). (E) MET currents at different voltages; voltage steps ranged from −120 mV to 100 mV in steps of 20 mV. (F) MET current–voltage relationships. Current at each voltage was divided by the absolute value of the current at −84 mV (mean ±s.d.). Sample sizes were n=7 (Gipc3KO/+) or n=4 (Gipc3KO/KO); mean±s.e.m. plotted. (G–N) Bundle morphology visualized by phalloidin staining in cochlear hair cells of Gipc3KO/+ (G–J) and Gipc3KO/KO (K–M) mice during development. (O,P) High magnification view of IHC bundles. Note elongated and thickened Gipc3KO/KO stereocilia in P (arrows), presumably arising from fusion of several stereocilia. (Q,R) SEM of Gipc3KO/+ (Q) and Gipc3KO/KO (R) cochleas. (S–V) Magnified views of IHC (S,U) and OHC (T) bundles. Gipc3KO/KO IHC bundles had increased numbers of rows with thick stereocilia (U); Gipc3KO/KO OHC bundles were often squeezed inwards (V). (W) Examples of TRIOBP labeling to detect rootlets of Gipc3KO heterozygote and knockout IHCs and OHCs. (X) Overlays of row 1 stereocilia for Gipc3KO heterozygote and knockout IHCs and OHCs. Both IHCs and OHC row 1 patterns show inward squeezing. Images in G–X representative of at least two repeats. Scale bar: 2 µm (X). Panel widths: G–N, 40 µm; O,P, 20 µm; Q,R, 17.3 µm; S–V, 5 µm; W, 10 µm. controls (Fig. 3F). For GFP–MYO6TAIL (bait) plus mCherry– GIPC3 ( prey) experiments, increased prey fluorescence at filopodial tips correlated with higher levels of bait fluorescence (Fig. 3G); there was no such correlation for GFP–MYO7ATAIL (Fig. 3G). GIPC3 is mislocalized in Myo6-null mice In Gipc3KO/KO mice, the distribution of MYO6 was not different from that in heterozygous controls (Fig. 3H,I); MYO6 thus did not localization. To test whether GIPC3 depend on GIPC3 for e c n e i c S l l e C f o l a n r u o J 4 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 3. See next page for legend. e c n e i c S l l e C f o l a n r u o J 5 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 3. GIPC3 interacts with MYO6. (A,B) Colocalization of GIPC3 and MYO6 in IHCs at P6.5 and P15.5 in C57BL/6 cochleas. Localization was performed more than three times. (C) Domain and construct structure of GIPC3 and MYO6. (D) Coomassie-stained gel showing interaction of GIPC3 and MYO6 using GST pulldowns. The GIPC3 full-length construct (red arrow) and the GIPC3 construct containing the PDZ and GH2 domains both interacted with the MYO6 HCBD construct; the GIPC3 construct containing only the PDZ domain did not. Experiment was performed twice. (E–G) NanoSPD analysis of interactions with GIPC3. All transfected cells had mCherry–GIPC3 and MYO10NANOTRAP; GFP–MYO6TAIL constructs were also included, which interacted with MYO10NANOTRAP and were targeted to filopodial tips (arrows). (E) Example of interaction of mCherry–GIPC3 with GFP–MYO6TAIL-A. (F) Prey fluorescence with various constructs. Mean±s.d. plotted. P-values are one-way ANOVA comparisons with Dunnett correction for multiple comparisons to no-bait condition with Dunnett correction for multiple comparisons. au, arbitrary units. (G) Intensity correlation analysis, using scatter plot of bait (X-axis) and prey (Y-axis) fluorescence at individual filopodia tips (from three independent determinations). We used linear fits through the origin; values for slope and R2 were: MYO6A-A (0.172, 0.19), MYO6A-B (0.156, 0.21), MYO6A-C (0.207, 0.39), MYO7A (0.015, −0.32), no bait (0.032, −0.85). Dashed lines are 95% confidence intervals. Sample sizes (n) were MYO6A-A (79 filopodia), MYO6A-B (160), MYO6A-C (52), MYO7A (101), no bait (116). (H,I) MYO6 localization did not change in Gipc3KO/KO IHCs. (J,K) GIPC3 was mislocalized in Myo6 CRISPR knockout G0 IHCs. Panel widths: A,B, 25 µm; E, 30 µm; H–K, 35 µm. localization depends on MYO6, we used the ‘improved genome editing via oviductal nucleic acids delivery’ technique (i-GONAD) (Ohtsuka et al., 2018), a CRISPR-Cas9 strategy, to create G0 animals that had null mutations in both Myo6 alleles. We found a high correlation between biallelic targeting in genotyping assays of mouse tails and the loss of MYO6 immunoreactivity in hair cells (Table S2), suggesting that CRISPR-Cas9 gene modification occurred early during development. Animals with disrupted Myo6 alleles had severely disrupted hair bundles, including fused stereocilia, as has been reported for the Myo6sv allele (Self et al., 1999). Hair cells from i-GONAD-generated G0 mouse pups that had two wild-type Myo6 alleles had normal distribution of GIPC3 (Fig. 3J). By contrast, in G0 pups with presumptive null Myo6 mutations in both alleles, GIPC3 distribution was significantly perturbed the hair cell (Fig. 3K); GIPC3 no longer concentrated at periphery, and instead was found in an apparently cytoplasmic pattern. Localization of GIPC3 near cellular junctions thus depends on MYO6. Misshapen cuticular plates in Gipc3KO/KO inner hair cells As IHCs of C57BL/6 mice developed, their cuticular plates elongated along the cochlear longitudinal axis (Fig. 4A). This shape transition matched that of the apical circumference, which has been described as shifting from a near-circular shape around P1 to a rounded rectangular one from P5 on (Etournay et al., 2010). While the cross-sectional area of the cuticular plate did not change between P0 and P20 (Fig. 4A and P, left), the vertical depth increased (Fig. 4G,Q). Using the simplifying geometric assumption that the cuticular plate was a hemisphere, we estimated that the cuticular plate volume increased ∼30% over this period (Fig. 4R, left). An alternative simplifying assumption, that the cuticular plate was a flat cylinder, gave similar results for volume. In Gipc3KO/KO IHCs, the shift of the cuticular plate circumference from circular to rounded rectangular did not occur (Fig. 4B,C). The cross-sectional area of the Gipc3KO/KO cuticular plate was reduced compared to that in Gipc3KO/+ at all time points during development (Fig. 4P, right). The decrease in cuticular plate area was, however, initially offset by a corresponding increase in cuticular plate depth (Fig. 4H,I,Q), such that the estimated cuticular-plate volume in Gipc3KO/+ and Gipc3KO/KO IHCs was nearly identical (Fig. 4R, the cuticular plate thinned considerably in right). After P21, Gipc3KO/KO hair cells (Fig. 4Q, right), which led to a decrease in cuticular plate volume (Fig. 4R, right). GIPC3 thus plays a role in the developmental elongation and flattening of IHC cuticular plates. Labeling for components of the apical region highlighted these changes. We used antibodies against TJP1 (ZO1) to label apical junctions and antibodies against LMO7 to label the cuticular plate (Du et al., 2019). In Gipc3KO/KO mutants, LMO7 labeling remained throughout the apical area even though the cross-sectional area of the cuticular plate, as determined by phalloidin labeling, decreased (Fig. 4D,E). We used image rendering to reconstruct cuticular plates marked by LMO7 labeling at P16.5 (Fig. 4L–O). The distinction between the thin, flattened cuticular plates of Gipc3KO/+ controls (Fig. 4L,M) and the more rounded cuticular plates of Gipc3KO/KO mutants (Fig. 4N,O) was apparent. Volumes were directly estimated from the rendered LMO7 images, not using simplifying geometric assumptions; Gipc3KO/+ control LMO7 volumes were 63±18 µm3 (mean±s.d.; n=13), whereas Gipc3KO/KO mutants were 71±18 µm3 (P=0.165; unpaired two-tailed Student’s t-test). Altered apical junctions of Gipc3KO/KO hair cells IHCs are packed closely together along the cochlear axis, albeit separated by luminal processes of inner phalangeal cells (IPhCs) (Driver and Kelley, 2009). In control IHCs, antibodies against TJP1 labeled the junctional regions of IHCs, but not the IPhCs, revealing two parallel lines of immunostaining separated by a very small gap (Fig. 5A, arrows). Examination of a serial block-face SEM dataset from mouse cochlea (Hua et al., 2021) showed that two adjacent IHCs sandwiched microvilli projecting from the IPhCs; the microvilli and IHC plasma membranes appeared to be close enough that the interaction was mediated by cell–cell contacts (Fig. S3). In addition, some direct IHC–IHC contacts also appeared to be present in areas with fewer microvilli (Fig. S3). In Gipc3KO/KO cochleas, however, the gap between IHCs widened noticeably (Fig. 5B, arrows); this space was presumably filled in by expansion of the apical surfaces of IPhCs. The consistent difference between the association of adjacent IHCs is illustrated in Fig. 5C,D; whereas Gipc3KO/+ IHC outlines were relatively rectangular, Gipc3KO/KO IHC outlines were considerably more rounded and did not appear to be tightly coupled together. In addition, the apical circumference of Gipc3KO/KO IHCs was reduced significantly when compared to wild-type or heterozygote IHCs (Fig. 5G). Immunoreactivity for the MYH9 (the heavy chain of nonmuscle myosin IIa) in Gipc3KO/+ and Gipc3KO/KO IHCs (Fig. 5E,F) highlighted the relatively square appearance of the apical junction region in Gipc3KO/+ compared to the more rounded appearance in Gipc3KO/KO. In Gipc3KO/+, MYH9 was concentrated at the junction between one IHC and two pillar cells, a tripartite junction that forms an apex pointing laterally in the cochlea (Fig. 5E, arrows). By contrast, MYH9 was relatively uniformly distributed around the apical junctional region in Gipc3KO/KO IHCs (Fig. 5F). The loss of close opposition of IHCs in Gipc3KO/KO cochleas was apparent by SEM (Fig. 5H,I). In heterozygotes, a few IPhC microvilli projected between the apical surfaces of two adjacent IHCs, which appeared to have some direct contacts (Fig. 5H, inset). By contrast, the microvilli-endowed apical surfaces of IPhCs expanded substantially in Gipc3KO/KO cochleas, and apical surfaces of IHCs flanking each IPhC were relatively far apart (Fig. 5I, inset). e c n e i c S l l e C f o l a n r u o J 6 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 4. See next page for legend. Immunoaffinity purification of GIPC3 protein complexes To understand how GIPC3 exerts its effects on dimensions of cuticular plates and apical surfaces, we examined the GIPC3 protein interaction network in hair cells. Because of its superior recognition of chicken GIPC3, we exploited the 10G5 anti-GIPC3 monoclonal antibody to immunoaffinity purify GIPC3 from crosslinked chicken inner ear extracts. We used a fraction that was enriched for stereocilia, but still contained large amounts of hair cell cytoplasmic proteins (Morgan et al., 2016). We carried out two separate experiments, each with ∼1000 chicken ears, where we stabilized protein complexes using primary amine-reactive homo-bifunctional N-hydroxysuccimide ester crosslinkers, which are thiol-cleavable and hence reversible (Mattson 7 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 4. Cuticular plate defects in Gipc3-knockout mice. (A) Examples of phalloidin labeling at the level of the cuticular plate for C57BL/6 IHCs (X-Y slices). Red dashed line in P2.5 example outlines cuticular plate. Lateral and medial sides of the hair cell indicated in the P4.5 panel. CP, cuticular plate; AB, circumferential actin belt. (B,C) Examples of phalloidin labeling at the level of the cuticular plate for Gipc3KO heterozygote (B) and homozygote (C) IHCs. (D,E) Triple labeling for TJP1 (ZO1; showing apical cell junctions), F-actin and LMO7 (showing cuticular plate). (F) Diagrams illustrating X-Y slice used for panels A–E and X-Z reslice used for panels G–K. Lateral and medial sides of the hair cell are indicated. (G) Examples of phalloidin labeling through the cuticular plate for C57BL/6 IHCs (X-Z reslices). SC, stereocilia; M, medial edge of hair cell; L, lateral edge. (H,I) Examples of phalloidin labeling through the cuticular plate for Gipc3KO heterozygote (H) and homozygote (I) IHCs. (J,K) Triple labeling for TJP1, actin and LMO7. (L–O) Imaris reconstruction (rendering) of LMO7 labeling in Gipc3KO heterozygote (L,M) and homozygote (N,O) IHCs at P16.5. The lateral (kinocilium) edge of the hair cell is at top in L and N; the medial edge is at the bottom. M and O are the same cuticular plates as in L and N but are rotated in two axes. (P–R) Quantification of cuticular plate dimensions (mean ±s.d.). (P) Quantification of IHC cuticular plate area from P2.5 to P19.5 in C57BL/6 IHCs (left) and from P2.5 to P21.5 Gipc3KO heterozygote and homozygote IHCs (right). For Gipc3KO, unpaired two-tailed t-tests were used for statistical comparisons. Mean±s.e.m. are plotted in right panels (also for M and N). P-values for cuticular plate area were: P2.5, 0.0073 (n=24 and 13); P6.5, P<0.00001 (n=28 and 27); P8.5, P<0.00001 (n=32 and 41); P15.5, P<0.00001 (n=18 and 25); P21.5, P<0.00001 (n=17 and 14). (Q) Quantification of cuticular plate depth from P2.5 to P19.5 in C57BL/6 IHCs (left) and from P2.5 to P25.5 in Gipc3KO heterozygote and homozygote IHCs (right). P-values for cuticular plate depth were: P2.5, 0.011 (n=33 and 19 for heterozygote and knockout); P8.5, 0.0003 (n=30 and 35); P15.5, P<0.0001 (n=60 and 53); P21.5, P<0.0001 (n=27 and 31). P<0.05; P<0.001; *P<0.0001. (R) Quantification of cuticular plate volume from P2.5 to P19.5 in C57BL/6 IHCs (left) and from P2.5 to ≥P21.5 in Gipc3KO heterozygote and homozygote IHCs (right). Data were determined from area and depth averages; mean±s.d. plotted. Panel widths: A, 17.5 µm; B–K, 12 µm; L–O, 50 µm. et al., 1993). In one experiment, we used dithiobis(succinimidyl propionate) (DSP), a membrane-permeable crosslinker that crosslinks extracellular and intracellular complexes; in the other experiment, we used 3,3′-dithiobis(sulfosuccinimidyl propionate) (DTSSP), which is membrane impermeant and thus only stabilizes extracellular and transmembrane complexes. We prepared soluble extracts of crude, crosslinked stereocilia (S7 in Fig. 6A) and used these fractions (S7 from DSP or DTSSP) for identifying GIPC3-interacting proteins. When proteins were crosslinked with DSP, very few proteins were purified from S7 by control mouse IgG (Fig. 6B). By contrast, 429 proteins were precipitated from S7 (out of 1061 total) with 10G5 (Fig. 6D). Many proteins were enriched substantially by precipitation – their relative abundance in the immunoaffinity precipitate was greater than that in the starting material – including GIPC3 itself, which was detected in the 10G5 immunoaffinity precipitate but not in the S7 starting extract. The immunoaffinity purification was repeated using the flow-through from this DSP extract experiment as the starting material, which gave very similar results (Fig. S4); this suggests that the antibody-coupled beads were saturated with GIPC3 and its partners in the first immunoprecipitation. Immunoaffinity purification results with DTSSP were broadly similar, although GIPC3 made up a larger fraction of the precipitate and other proteins were at lower levels in the precipitate than in the DSP experiment (Fig. 6D,E), consistent with the lack of intracellular crosslinking. Gene ontology (GO) analysis of the top 35 proteins most enriched in the GIPC immunoprecipitations (Table S3) indicated that many were associated with actin components (red in Fig. 6F). Key cellular component terms included focal adhesion, myosin complex, actin cytoskeleton, stress fiber, cell cortex, and the myosin II complex. This analysis suggests that the GIPC3 complexes purified included co-precipitating fragments of adherens junctions, circumferential actin belts and cuticular plates. These results are consistent with localization of GIPC3 in these regions and morphological disruptions to the same areas in the Gipc3KO/KO mice. To determine how the protein networks of key immunoprecipitated proteins overlapped, we examined pairwise binary interactions listed in the BioGRID protein-interaction database (Oughtred et al., 2021) for APPL2, MYO6, MYO18A, MYH9 and MYH10. We chose these proteins based on their known interaction with actin structures at cell junctions (MYO18A, MYH9 and MYH10) or their known interaction with the paralog GIPC1 and association with early endosomes (APPL2 and MYO6). We compared the overlap of these six networks (including GIPC3), identifying interacting proteins shared by two or more networks. Many interactions were shared by these six proteins including many proteins that were identified in the (Fig. 6G), immunoprecipitation experiments (bold in Fig. 6G). Color coding was the same as in Fig. 6B–E. The APPL2 network overlap was sparser than those of the other proteins, but still many proteins were shared with the other target proteins. Binding partners for GIPC3 are located in hair cells As GIPC3 is likely to interact with target proteins using its PDZ domain, we inspected the top 35 enriched proteins for C-terminal PBMs (Table S3). Four proteins had PBMs that met the consensus for high-affinity binding to the PDZ domains of the GIPC family (Table S4) – APPL2, MYO18A, ACTN1 and ACTN4 (Fig. 7A). A paralog of APPL2, namely APPL1, is a well-characterized partner of GIPC1 (Lin et al., 2006; Varsano et al., 2006), suggesting that a GIPC3–APPL2 interaction is plausible. Indeed, we found using NanoSPD that mCherry–GIPC3 interacts with GFP–APPL2 (Fig. S2M). Immunoreactivity for APPL2 was detected in inner pillar cells (Fig. S5); although APPL2 was not definitively located in IHCs, Appl2 transcripts were enriched in cochlear hair cells (https://umgear.org/index.html?multigene_plots=0&gene_symbol_ exact_match=1&gene_symbol=appl2). These results suggest that our antibody is insufficiently sensitive or that its epitope is masked in hair cells. Myo18a transcripts are highly enriched in hair cells (https:// umgear.org/index.html?multigene_plots=0&gene_symbol_exact_ match=1&gene_symbol=myo18a), and MYO18A protein was enriched in cochlear and vestibular hair cells (Fig. 1A,B), supporting MYO18A as a candidate for interaction with GIPC3. Using the Atlas antibody specific for MYO18A (Fig. S6A,B), we found in C57BL/6 and Gipc3KO/+ heterozygote IHCs that MYO18A immunoreactivity surrounded the cuticular plate, especially underneath it (Fig. 7B). Similar results were seen with a second antibody (Fig. S5D). Lattice SIM revealed significant MYO18A immunoreactivity at the apical periphery of IHCs (Fig. 7C,D), similar to the location of GIPC3. MYO18A immunoreactivity was similarly distributed in Gipc3KO/KO mutants and heterozygote controls (Fig. 7E,F). α-Actinins have been localized to hair cell cuticular plates (Slepecky and Chamberlain, 1985). Actn1 and Actn4 transcripts were present in cochlear hair cells, although not enriched (https:// umgear.org/index.html?multigene_plots=0&gene_symbol_exact_ match=0&gene_symbol=actn). We detected a modest increase of ACTN4 labeling in IHC cuticular plates compared to that in cell bodies; expression levels were much higher in pillar cells (Fig. S5C). We were unable to detect ACTN1. Other proteins might bind to GIPC3 in hair cells. Although MYO6 is responsible for strong adhesion of cell–cell contacts mediated by 8 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 5. Apical junction defects in Gipc3-knockout mice. (A,B) TJP1 immunoreactivity highlights apical junctions in Gipc3KO IHCs. X-Y (top) and X-Z (middle) slices. Yellow lines in upper panels indicate transects used for X-Z reslices in middle panels; magnified boxed regions of reslices are below. immunolocalization experiments were performed more than three times. IPhC, inner phalangeal cell. (C,D) Tracings of TJP1 labeling from four Gipc3KO/+ (C) and six Gipc3KO/KO (D) cochleas. (E,F) MYH9 immunoreactivity highlights apical junctions in Gipc3KO cochleas. (G) Apical circumference area for the indicated genotypes. Number of cochleas and cells per cochlea analyzed: Gipc3+/+, 2 and 4; Gipc3KO/+, 4 and 4; Gipc3KO/KO, 6 and 2–4. P-values determined from nested one-way ANOVA with Tukey correction. (H,I) SEM of P21.5 Gipc3KO/+ and Gipc3KO/KO IHC region. Insets show the junction region between an IHC, an IPhC, and the next IHC. The yellow dashed lines outline the region in between the two IHCs that is occupied by IPhC microvilli. P21.5 SEM was carried out once, with similar results seen in more than 25 IHCs in each genotype. Panel widths: A,B upper and middle panels, 80 µm; A,B lower panels, 20 µm; E,F, 35 µm; G,H, 30 µm (insets, 2 µm). CDH1 (E-cadherin) (Maddugoda et al., 2007), CDH1 expression was low in IHCs (Etournay et al., 2010). CDH2 has reciprocal expression in the cochlea, however, and was present at high levels in IHCs (Etournay et al., 2010; Simonneau et al., 2003), raising the possibility that MYO6 plays a role in CDH2-mediated cell–cell contacts. CTNNB1 (β-catenin) binds directly to CDH1 and CDH2 (Valenta et al., 2012); CTNNB1 and its partner CTNNA1 were co-purified with GIPC3 in immunoaffinity purification experiments (Fig. 6D, E), albeit not enriched relative to the starting extract. CTNNB1 has a C-terminal PBM (NQLAWFDTDL) that is similar to the GIPC3 consensus sequence; in NanoSPD experiments, we demonstrated that mEMERALD–CTNNB1 interacted with mCherry–GIPC3 (Fig. S2O,P). CTNNB1 could thus mediate interaction of GIPC3 with adhesion complexes, presumably through the PBM in CTNNB1. MYO18A forms aggregates and interacts with GIPC3 Experiments using NanoSPD demonstrated that full-length MYO18A tagged at the N-terminus with GFP interacted with GIPC3 (Fig. 7G). A comparison of MYO18A deletion constructs (Fig. 7I) showed that constructs that sequentially lacked the IQ calmodulin-binding N-terminal extension, motor domain, domain and coiled-coil domain still interacted with GIPC3; only when the C-terminal PBM was deleted was the interaction of MYO18A and GIPC3 significantly reduced (Fig. 7K; Figs S6, S7). Unsurprisingly, the interaction of MYO18A with GIPC3 required the presence of the PDZ domain in GIPC3 (Fig. 7J,L; Figs S6, S7). MYO6 did not interact with MYO18A (Fig. 7L). a as When expressed full-length co-expression, MYO18A formed protein without MYO10NANOTRAP large aggregates within the cytoplasm of HeLa cells (Fig. 7H; Fig. S6D), resembling the biomolecular condensates arising from self- association (Banani et al., 2017). These aggregates were still seen with constructs that lacked the N-terminal extension, motor domain, IQ calmodulin-binding domain or C-terminal PDB (Fig. S6C–G). A construct with the C-terminus alone, including the PBM, was cytoplasmic (Fig. S6H), however, the coiled-coil domain has a role in mediating MYO18A self-association. Formation two MYO18A–MYO18A of aggregates suggested that at interaction domains exist within the coiled-coil domain. implying that least When co-expressed with GFP alone, mCherry–GIPC3 was largely cytoplasmic, although some mCherry–GIPC3 did localize to an unknown punctate target in the HeLa cell cytoplasm (Fig. S6I). By contrast, most mCherry–GIPC3 was recruited to aggregates when e c n e i c S l l e C f o l a n r u o J 9 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 6. See next page for legend. GFP–FL-MYO18A was co-expressed (Fig. S6J), which indicates that GIPC3 does not obscure the domains responsible for aggregation. Removal of the N-terminal extension, motor domain, IQ calmodulin- binding domain, or coiled-coil domain had no effect on recruitment of mCherry–GIPC3 to GFP–MYO18A aggregates (Fig. S6K,L). Although aggregates were absent when C-PBM-MYO18A was localization of co-expressed mCherry–GIPC3 still expressed, matched that of C-PBM-MYO18A (Fig. S6M). By contrast, when the four C-terminal amino acids of FL-MYO18A were deleted, mCherry–GIPC3 no longer interacted with MYO18A (Fig. S6N). Taken together, the experiments of Fig. 7 and Fig. S6 demonstrate that the PBM of MYO18A interacts with the PDZ domain of GIPC3. 10 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 6. GIPC3 interaction networks identified through immunoaffinity purification and protein mass spectrometry. (A) Flow chart for anti- GIPC3 immunoaffinity purification from crude chick stereocilia extracts. F/T, flow through. (B–E) Comparison of abundance (riBAQ) of proteins detected in DSP1 total or DTSSP total (starting S7 extract; plotted on X-axis) compared to the immunoprecipitates (Y-axis) for mouse IgG control (B,C) and 10G5 anti-GIPC3 (D,E) experiments. Panels B and D show results with the DSP-crosslinked starting extract, whereas panels C and E show results with DTSSP crosslinking. Each point represents the average abundance of that protein in two experiments (biological replicates); symbol colors were arbitrarily chosen. Red dashed line is the unity line (equal riBAQ in total and IP). Key proteins are called out. Mouse IgG protein from immunoprecipitation is highlighted in gray. (F) Gene ontology analysis (cellular component) with DAVID of the top 50 proteins from the DTSSP 10G5 eluate. Red, actin- associated components; orange, intermediate filament-associated components; blue, microtubule-associated components; gray, other components. (G) Overlap of protein interaction networks of key proteins from the DTSSP 10G5 eluate (GIPC3, APPL2, MYO6, MYO18A, MYH9 and MYH10). BioGRID-defined protein networks for APPL2, MYO6, MYO18A, MYH9 and MYH10 were compared with the top 100 proteins from the DTSSP 10G5 eluate. Only proteins identified as interactors of two or more of the key proteins were included. Proteins in bold were present in the top 100 proteins from the DTSSP 10G5 eluate. DISCUSSION Our data show that GIPC3 is essential for hair cell function, and roles for this protein include shaping the cuticular plate and contributing to normal cell–cell junctions. Our results confirm that the GH2 domain of GIPC3 interacts with the molecular motor MYO6; moreover, we show that the PDZ domain interacts with a number of proteins situated at locations that mean they are likely to be involved in cuticular plate and junction function, including MYO18A and the α-actinins ACTN1 and ACTN4. Moreover, the GIPC3 network also includes the non-muscle myosin II proteins MYH9 and MYH10, perhaps coupled through their binding partner MYO18A (Billington et al., 2015). Our working hypothesis is that GIPC3 establishes multiple complexes that include MYO6 and other key molecules, and that these complexes operate at the apex of the hair cell to connect the cuticular plate to apical junctions, directly or indirectly. Cuticular plate and hair-bundle shape changes in Gipc3KO/KO Cochlear hair cells undergo a notable change in shape of their apical region in the early postnatal period (Fig. 8A), which has been ascribed to internal tension generated by MYO7A (Etournay et al., 2010). Disruption of hair bundle integrity interferes with the apical shape change, which suggests a reciprocal interaction between stereocilia and their rootlets with the cuticular plate and apical junction complex (Etournay et al., 2010). GIPC3-engaged MYO6 might also contribute to this internal tension, albeit with a less dramatic impact on bundle integrity and without the loss of MET current. A prominent consequence of the loss of GIPC3 in hair cells was that the cuticular plate remained more round, both for the apical-to- basal axis (Z axis) and the perpendicular axis (X-Y axis) of the cell. In Gipc3KO/KO IHCs, the cross-sectional area of the cuticular plate never reached the size seen in Gipc3KO/+ controls, whereas the depth of the cuticular plate did not reduce as it does in controls. Accordingly, the volume of the cuticular plate did not change appreciably until after P15.5 in Gipc3KO/KO IHCs, suggesting that the cuticular plate was formed normally early during development but was then subjected to different internal forces in the mutant. In particular, the cuticular plate did not flatten and extend normally during development in Gipc3KO/KO IHCs, that is, it did not proceed from hemisphere-like to disk-like (Fig. 8). Gipc3KO/KO hair bundles differ from those of Gipc3KO/+ controls in several ways. Stereocilia in IHC bundles appeared to be pushed together, and the shorter stereocilia were generally thicker than the shorter stereocilia of controls. Maximum transduction currents were higher in Gipc3KO/KO IHCs, suggesting that there were more channels, but offset stimuli were required to fully elicit the maximum current. Both IHC and OHC bundles of Gipc3KO/KO mice had a squeezed appearance, where the flanking wings of the bundle (stereocilia most distal from the fonticulus and basal body) were closer together. This phenotype correlates well with the lack of flattening and extension of the cuticular plate, as if the stereocilia were coupled within the cuticular plate and were, similar to the cuticular plate, not subjected to the internal forces that spreads them out along the cochlear axis. and junctional cytoskeletal experiments with The GIPC3 protein network anti-GIPC3 Immunoaffinity purification precipitated many proteins, presumably in large complexes that were crosslinked either both extracellularly and intracellularly (with DSP experiments) or extracellularly alone (DTSSP). The molar abundance in the immunoprecipitates (estimated from the Y-axis in Fig. 6D,E) of most cytoskeletal proteins was much greater than that of GIPC3, which suggests that a small number of GIPC3 molecules engaged large cytoskeletal networks. These networks likely included the cuticular plate itself, as well as the apical cell junctions connecting hair cells to surrounding supporting cells. In addition, cytoskeletal proteins concentrated in the pericuticular necklace were likely also present, given GIPC3 is localized there. GIPC3 includes a PDZ domain, and several proteins enriched in the GIPC3 immunoaffinity purification experiments contained sequences that matched the consensus C-terminal PBM for the GIPC family. Although these immunoaffinity purification experiments were carried out with chick inner ear, we confirmed interaction of mouse MYO18A, APPL2 and CTNNB1 with mouse GIPC3 using NanoSPD assays. The interaction of GIPC1 and MYO6 is well established, and preliminary evidence suggested that GIPC3 and MYO6 interact (Shang et al., 2017). We confirmed this GIPC3–MYO6 interaction using in vitro GST pulldown experiments, which showed that the GH2 domain of GIPC3 interacted with the helical cargo-binding domain (HCBD) of MYO6. Moreover, immunoaffinity purification and NanoSPD experiments showed that MYO6 associates with complexes containing GIPC3, presumably because of the direct GIPC3–MYO6 interaction. Additional evidence for this interaction includes similar phenotypes of the Myo6- and Gipc3-null mouse lines, as well as the mislocalization of GIPC3 in Myo6-knockout model that was generated using i-GONAD and CRISPR/Cas9. In particular, the latter results showed that MYO6 is responsible for localization of GIPC3 at the apical periphery of hair cells, near cell– cell junctions. Full-length GIPC3 forms an autoinhibited dimer, and so the interaction with MYO6 likely occurs only after GIPC3 has been activated by a ligand with a PBM that binds the GIPC3 PDZ domain (Shang et al., 2017), which in turn suggests that GIPC3 not only interacts with MYO6 at the apical periphery, but also with an activating ligand. These experiments highlighted MYO18A, another unconventional myosin that is expressed in hair cells. Hair cells are notable for their reliance on myosins, presumably because their actin-rich cytoskeletal structures, especially the stereocilia and cuticular plate, are substrates for myosins (Friedman et al., 2020; Hasson et al., 1997). The location of MYO18A below the cuticular e c n e i c S l l e C f o l a n r u o J 11 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 7. See next page for legend. e c n e i c S l l e C f o l a n r u o J 12 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 7. MYO18A is located in the hair cell apical domain. (A) Top, sequence logo for binding of ligands to GIPC1 PDZ domain; bottom, C-terminal ten amino acids of APPL2, MYO18A, ACTN1 and ACTN4. (B) Immunolocalization of MYO18A in P14.5 mouse cochlea; slices from a three-dimensional image stack. Transects for other image axes are shown in yellow; the X and Y transects in the main X-Y image show the locations for the Y-Z and X-Z images. Arrow indicates concentration of MYO18A immunoreactivity below the IHC cuticular plate. IHC, inner hair cell; IPC, inner pillar cell; OPC, outer pillar cell; OHC, outer hair cell. Immunolocalization experiments for MYO18A were performed more than five times. (C,D) MYO18A immunoreactivity in P15.5 IHCs using lattice SIM imaging. (C) Image showing four IHCs at the stereocilia/cuticular plate level. (D) Image showing a single IHC (labeled with asterisk in C) at the cuticular plate level (different plane than in C). Arrows delineate the gap between cuticular plate actin and the circumferential actin belt. (E,F) MYO18A immunoreactivity in P14.5 IHCs from folded cochleas using Airyscan imaging. E is from a Gipc3KO/+ mouse and F is from a Gipc3KO/KO mouse. (G) NanoSPD of MYO18A–GIPC3. Example of filopodial targeting of mCherry–GIPC3 by GFP–FL-MYO18A, mediated by MYO10NANOTRAP. (H) Expression of GFP–FL-MYO18A and mCherry–FL-GIPC3 constructs in HeLa cells (no MYO10NANOTRAP expressed). Arrows indicate large cytoplasmic aggregates containing GFP and mCherry. squeezing. Images in C–H representative of at least three repeats. (I) GFP–MYO18A constructs. ‘Motor’, actin- and ATP-binding domains are homologous to myosin motor domains in active myosins; IQ, isoleucine/glutamine calmodulin-binding; PBM, PDZ-binding motif. (J) mCherry–GIPC3 constructs. mCh, mCherry; GH1, GIPC-homology 1; GH2, GIPC-homology 2. (K) Prey (mCh–GIPC3) fluorescence with GFP–MYO18A constructs or GFP control. Mean±s.d. plotted in K and L. One-way ANOVA comparisons to no-bait condition with Dunnett correction for multiple comparisons. Sample sizes (n) were GFP– FL-MYO18A (105 filopodia), GFP–IQ-PBM-MYO18A (109), GFP–CC-PBM- MYO18A (116), GFP–C-PBM-MYO18A (129), GFP–ΔPBM-MYO18A (127), GFP (208). (L) Prey (mCh–GIPC3 constructs or mCh–MYO6) fluorescence with GFP-MYO18A constructs or GFP control. One-way ANOVA comparisons to no-bait condition with Dunnett correction for multiple comparisons. Sample sizes (n) were mCh–D1-GIPC3 (128 filopodia for GFP–FL-MYO18A and 88 for GFP alone), mCh–D3-GIPC3 (180 and 47) and mCh–MYO6 (107 and 93). Panel widths: B, 37.5 µm for X-Y plot (same scale applies to Y-Z and X-Z panels); C, 50 µm; D, 12 µm; E-F, 45 µm; G,H, 15 µm. plate suggests that it is involved in constraining the cuticular plate, a role also proposed for the striated organelle, an enigmatic structure also found in hair cells below the cuticular plate (Slepecky et al., 1981; Vranceanu et al., 2012). MYO18A is plausibly a component of the striated organelle. In addition, MYO18A localizes near apical junctions connecting IHCs to surrounding supporting cells, like the apical distribution of GIPC3. Mice homozygous for a global Myo18a null allele show preweaning lethality with complete penetrance (https://www.mousephenotype.org/data/genes/ MGI:2667185), thus preventing simple genetic analysis. Model for cuticular plate shaping Etournay et al. (2010) showed that apical surfaces of IHCs shift from near circular to rounded rectangular by P7.5 (Fig. 8A); the distance along the cochlear axis remained approximately the same, however, whereas the distance along the lateral-medial axis shortened (Etournay et al., 2010). Our results suggest that force is generated during this shape change that is also coupled to lengthening of the cuticular plate along the same axis, and that transmission of this force requires GIPC3. As development proceeds, the apical circumference remodels, the cuticular plate flattens, and anchoring at the sides of the plate and base ensures that it forms an immovable platform for insertion of stereocilia. Loss of GIPC3 does not affect the total amount of cuticular plate material, however, because the volume of the cuticular plate was not altered in Gipc3KO/KO hair cells. As internal forces are generated, the cuticular plate extends; because a constant volume is maintained, the plate thins (Fig. 8). The hair bundle phenotype suggests that the stereocilia are moved into place because of their anchoring within the cuticular plate; bundles start as a tight cluster of stereocilia (Kaltenbach et al., 1994), but as development proceeds, they are separated into rows and columns by the forces extending the cuticular plate. When GIPC3 is absent, this separation does not occur properly. Based on localization of MYH9 and MYO7A, apical circumference remodeling might result from forces applied within hair cells (Etournay et al., 2010). MYO18A itself likely cannot generate this tension; it has very low myosin ATPase activity (Guzik-Lendrum et al., 2013), although it can form mixed filaments with MYH9 (Billington et al., 2015). We suggest that GIPC3– MYO6 complexes assist in anchoring the cuticular plate to apical cell–cell junctions and coupling to internal forces (Fig. 8B–E); whether MYO6 contributes to force elongating the cuticular plate or simply acts as an anchor is unknown. Several mechanisms for the formation of the rounded rectangular apical circumference in IHCs are plausible (Fig. 8). For example, internal contraction force could be generated along the lateral- medial axis, squeezing these two sides of the hair cell closer together (Fig. 8F). Given that the apical surface area of the cell does not change, tension could then develop within the cell along the cochlear axis; this tension could be used to reshape the cuticular plate if it were coupled to the apical junctions. Alternatively, forces might be applied to hair cells externally by the surrounding supporting cells; supporting cells could squeeze IHCs along their lateral-medial axis or elongate them along the cochlear axis. The rounded shape of cuticular plates in Gipc3KO/KO IHCs argues against this possibility, however, and suggests that the forces are reshaping forces are generated generated internally. Whether internally or externally, GIPC3 and MYO6 might be located in the ideal place to couple tension generated at the apical circumference to the cuticular plate, shaping it during development. MATERIALS AND METHODS Reagents Mouse monoclonal antibodies against GIPC3 were produced by the Monoclonal Antibody Core of the OHSU Vaccine and Gene Therapy Institute using standard hybridoma techniques. We used HEK293 cells to express mouse or chicken GIPC3 tagged with TwinStrepII, then purified them with affinity chromatography. A 1:1 mix of tagged mouse and chicken GIPC3 was used for immunization, and initial culture supernatants were screened by enzyme-linked immunoassay (ELISA) and immunoblotting. Antibodies that recognized GIPC3 were subsequently screened using immunocytochemistry with Gipc3KO/+ control and Gipc3KO/KO mutant cochleas. Hybridomas were cloned; antibodies were expressed in serum-free medium using a bioreactor and were purified using Protein A chromatography. We utilized the 6B4 (P.B.-G.; Oregon Health and Science University, Cat# PGBG-mAb002, RRID: AB_2895259; IgG2a-κ isotype), 3A7 (P.B.-G.; Oregon Health and Science University, Cat# PGBG-mAb003, RRID:AB_2895260; IgG2b-κ isotype) and 10G5 (P.B.-G.; Oregon Health and Science University, Cat# PGBG-mAb001, RRID:AB_2895258; IgG2a-κ isotype) antibodies here. Dilutions of 1:50 were used for immunostaining. Other primary antibodies used were: Atlas Antibody anti-MYO18A (Cat# HPA021121, RRID:AB_1854250; dilution of 1:200 for immunostaining) from Sigma-Aldrich (St Louis, MO, USA); Proteintech (Rosemont, IL, USA) anti- MYO18A (Cat# 14611-1-AP, RRID:AB_2201447; dilution of 1:100); Proteintech anti-APPL2 (Cat# 14294-1-AP, RRID:AB_2878041; dilution of 1:100); Thermo Fisher Scientific (Waltham, MA, USA) anti-TJP1 (also known as ZO1; Cat# 33-9100, RRID:AB_2533147; dilution of 1:200); Santa Cruz Biotechnology (Dallas, TX, USA) anti-LMO7 (Cat# sc-376807, RRID: AB_2892126; dilution of 1:100); Proteintech anti-TARA (TRIOBP; Cat# 16124-1-AP, RRID:AB_2209237; dilution of 1:200); Proteintech anti-ACTN4 (Cat# 19096-1-AP, RRID:AB_10642150; dilution of 1:100); and Biolegend e c n e i c S l l e C f o l a n r u o J 13 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Fig. 8. Model for GIPC3 coupling of apical cell junctions to the cuticular plate. (A) Tracings of averaged TJP1-labeled apical cell borders at indicated ages (adapted from fig. 2 of Etournay et al., 2010 with permission). (B,C) Key complexes and processes in P1 IHC. The cuticular plate is rounded up early in development. Apical junctions in blue, cuticular plate in dark orange, stereocilia in gray. The pericuticular necklace is the gap between the apical junctions and the cuticular plate. MYO18A is not shown at this age. (D,E) P7.5 IHC. The cuticular plate is flattened later in development. GIPC3– MYO6 complexes are in red; the MYO18A structure is in green; MYO18A at the apical junction region is not shown. (F) As the apical circumference is remodeled between P1 and P7.5, the IHC narrows along the lateral-medial axis. Inside the cell, active force generated along the cochlear axis stretches the cuticular plate; maintaining constant volume, the cuticular plate passively shrinks along the lateral-medial axis. GIPC3–MYO6 complexes either generate or are simply coupled to the force production that stretches the cuticular plate. Loss of GIPC3 prevents the elongation of the cuticular plate. The MYO18A structure underneath the cuticular plate might assist in its flattening; GIPC3 could couple MYO18A there with MYO6, ACTN1 or ACTN4 in the cuticular plate. B and D show medial-lateral sections through IHC centers; C, E and F show cross-section of IHCs at level of dashed line in B and D. The pericuticular necklace was removed for clarity in C–F, and the apical junctions were removed in F. Fisher Scientific donkey anti-rabbit-IgG conjugated to Alexa Fluor 488 (2 mg l−1; Cat# A21206, RRID:AB_2535792) or Thermo Fisher Scientific donkey anti-mouse-IgG conjugated to Alexa Fluor 568 (2 mg l−1; Cat# A10037, RRID:AB_2534013). Labeled phalloidins were: Biotium (Fremont, CA, USA) CF405 phalloidin (1 U per ml; Cat# 00034) or Biotium CF568 phalloidin (Cat# 00044). Plasmids were constructed using standard cloning techniques and are available upon request. Animal models All animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at Oregon Health & Science University ( protocol IP00000714) or Stanford University. Mouse pups were assumed to be born at midnight, so the animal age on the first day is referred to as P0.5. Both female and male pups were used for all experiments. C57BL/6J mice (RRID:IMSR_JAX:000664, Jackson Laboratories, Bar Harbor, ME) were used as wild-type mice. The C57BL/6N- Gipc3tm1a(KOMP)Wtsi line was obtained as resuscitated mice from the Knockout Mouse Project (KOMP) at the University of California Davis, USA. We used the Cre deleter strain (Schwenk et al., 1995) to remove the neomycin cassette and Gipc3 exons 2-3, generating a tm1b LacZ-tagged null allele. This line was backcrossed to C57BL/6J for more than six generations and propagated for the experiments described in this study. and The Myo6 locus was targeted for CRISPR-mediated knockout using guide RNAs (gRNAs) designed to exons 2 and 4 (http://crispor.tefor.net/). gRNAs were delivered via in situ electroporation using the i-GONAD procedure (Gurumurthy et al., 2019; Ohtsuka and Sato, 2019). Necessary components, including guides (Alt-R CRISPR-Cas9 crRNA, exon 2 guides – 5′- GTGGGGTGGGGTGCCCAAAC-3′ 5′-GGTTCAATTGTTAA- exon 4 guide – 5′-TGTACCGAACTTTGACATTG-3′, GCTGTC-3′, tracrRNA (Cat# 1072532), and Cas9 protein (Cat# 1081060) were obtained from Integrated DNA Technologies (IDT). Timed crosses with two female and one male C57BL/6J mice were set for an embryonic day (E)0.7 pregnancy. Two guides targeting exon 2 were used to increase the likelihood of a disruption to exon 2, the first coding exon, reducing experimental time and number of pregnant females. At this point in pregnancy, the zygote is at the single-cell stage and has lost its cumulus cells, allowing higher efficiency electroporation of the zygotes (Gurumurthy et al., 2019). Pregnancy was confirmed by checking plugs. Ribonucleoprotein (RNP) was prepared. To anneal tracrRNA and crRNA, tracrRNA and crRNA were mixed to a final concentration of 90 and 30 µM, respectively, in duplex buffer (IDT Cat# 1072570), then heated to 95°C and allowed to cool slowly back to room temperature. Cas9 protein was added at a final concentration of 1.5 mg l−1 to the annealed tracrRNA/crRNA mix, and the sample heated to 37°C and allowed to cool slowly back to room temperature. Fast Green (Thermo Fisher Scientific Cat# 2353-45-9) was prepared in duplex e c n e i c S l l e C f o l a n r u o J 14 anti-MYH9 (Cat# 909802, RRID:AB_2734686; dilution of 1:200). Two MYO6 antibodies were used; one (dilution of 1:250) was a gift from the laboratory of John Kendrick-Jones (MRC Laboratory of Molecular Biology, Cambridge, UK), and the other was Proteus anti-MYO6 (Cat# 25-6791, RRID: AB_10013626; dilution of 1:250). Secondary antibodies used were: Thermo RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 buffer, and then sterile filtered through a 0.22 µm filter (MilliporeSigma, Burlington, MA, USA; Cat# SLGP033RS). Filtered Fast Green was added at 3 g l−1 to the RNP sample so that the mixture could be visualized, once injected within the oviduct. To prepare for electroporation of RNPs, pregnant dams (E0.7) were given an intraperitoneal injection of anesthetic (working stock: 9 g l−1 Nembutal, Sigma-Aldrich Cat# P37610; 20.8 g l−1 MgSO4, Sigma- Aldrich Cat# 63138; 10% ethanol, Sigma-Aldrich Cat# 459836; 40% propylene glycol, Thermo Fisher Scientific Cat# P335-1) at 7.8 µl per gram body weight; anesthesia was confirmed by the lack of toe pinch reflex. Surgery was performed to expose the ovary and oviduct, and an estimated 0.5 to 1 µl of the RNP mixture injected into the lumen of the oviductal ampulla. The paddles of the electrode were placed around the portion of the oviduct where the Fast Green was visible, and electroporation performed, using three pulses of 5 ms on and 50 ms off at 30 V. The range of currents achieved under this protocol was 100–500 mA; optimal results were obtained when currents measured 150–250 mA. After electroporation, the ovary and oviduct were gently returned to the abdominal cavity, and the incision closed with two stitches and a wound clip. Throughout the surgery, tissue was kept hydrated with prewarmed lactated Ringers solution (Baxter Cat# 2B2323). For the first 3 days after surgery, dams were treated with a dose of meloxicam (MWI Animal Health Cat# 501080) at 1 µl per gram body weight for pain management; the first dose was administered soon after surgery was complete. G0 pups were screened for mutations by PCR and sequencing. Genomic DNA was extracted from G0 tail tissue samples using the DNeasy Blood and Tissue kit (Qiagen, Hilden, Germany; Cat# 69506). Primers sets were designed to amplify across either exon 2 or 4. Additionally, the forward primer for exon 2 and the reverse from exon 4 were paired to screen for large indels between exon 2 and 4. All amplicons for the exons 2 and 4, plus any amplicon smaller than the 9246 bp wild-type (WT) band for the large indel screen were gel purified with the NucleoSpin Gel clean up kit (Takara Bio, Kusatsu, Shiga, Japan; Cat# 740609) and sequenced. From the sequence, pups were placed in two categories; those with clearly definable indels, with quality sequence on either side of the change, or as mosaic, where the at the start of an indel there was an abrupt change in the chromatogram from clean single peaks to sequence with multiple peaks, indicating mixed template in the extracted gel band. Immunocytochemistry (see below) was carried out with G0 pups without knowledge of genotype, although the behavioral Myo6-null phenotype (circling) and morphological defects in hair cells were obvious in mice that were P15.5 or older. Data-independent mass spectrometry DDA mass spectrometry data were obtained from a dataset that is described in detail elsewhere (Krey et al., 2018) and located at https://www.ebi.ac.uk/ pride/archive/projects/PXD006240. For each developmental time point, the relative molar intensities for each protein were determined using the relative intensity-based absolute-quantification method (riBAQ method) (Krey et al., 2014); the mean±range was plotted (n=2 for each). Immunoaffinity purification mass spectrometry Immunoaffinity purification experiments using the 10G5 anti-GIPC3 monoclonal antibody used soluble extracts of partially purified chicken inner ear stereocilia prepared with methods described previously (Morgan et al., 2016). A simplified flow chart of the purification scheme is provided in Fig. 6A. Fertilized chicken eggs were obtained from the Department of Poultry Sciences, Texas A&M University (College Station, TX). Temporal bones were removed from E19–E21 chicks and were placed in ice-cold oxygenated chicken saline (155 mM NaCl, 6 mM KCl, 2 mM MgCl2, 4 mM CaCl2, 3 mM D-glucose, 10 mM HEPES, pH 7.25) for no more than 2 h, with an exchange of saline after 1 h. Sensory inner ear organs were removed using micro-dissection and were stored in ice-cold oxygenated saline for up to 4 h during dissection. Organs were rinsed with 4–5 changes of chicken saline (minimum 10-fold dilution per rinse) to remove excess soluble protein. Inner ears were treated with 1 mM dithiobis(succinimidyl propionate) (DSP; Thermo Fisher Scientific Cat# 22585), a membrane-permeable protein crosslinking reagent, or 0.1 mM 3,3′-dithiobis(sulfosuccinimidyl propionate) (DTSSP; Thermo Fisher Scientific Cat# 21578), a membrane-impermeant crosslinker, for 1 h at 4°C. The crosslinker solution was replaced with 100 mM Tris in saline to quench the reaction; the solution was reduced to 3 ml for each 100 ear lot, which was then snap-frozen in the presence of 1:100 Protease Inhibitor Cocktail (Sigma-Aldrich Cat# P8340) and stored at −80°C. Organs were thawed with chicken saline with 1:100 Protease Inhibitor Cocktail and 2% normal donkey serum (NDS; Jackson ImmunoResearch, West Grove, PA, USA; Cat# 017-000-121) at ∼5 ml per 100 ears. A glass/Teflon homogenizer was used to homogenize tissues (20 strokes at 2400 rpm). After centrifuging the homogenate at 120 g for 5 min at 4°C, the supernatant was collected; homogenization was carried out two more times. Chicken saline containing NDS and protease inhibitors was used to wash the pellet two or three more times. All supernatants (typically 50–60 ml per 1000 ears) were combined as the post-nuclear supernatant (S1); the nuclear pellet (P1) was discarded. S1 (11 ml each centrifuge tube) was layered on 2.2 M sucrose cushions (1 ml cushion) and was spun in a Beckman SW41 rotor at 8400 g for 30 min at 4°C. The supernatant was removed (S2); to collect the dense-membrane pellet, the cushion was removed and the tubes were washed out with chicken saline with protease inhibitors and serum. Dense membranes (P2) were homogenized using five strokes in a glass/Teflon homogenizer to remove lumps. The volume yield was usually ∼20–25 ml for 500 ears. D10 or 10G5 monoclonal antibodies were coupled to 1 µm MyOne Tosylactivated Dynabeads (Life Technologies, Grand Island, NY; Cat# 65502) at 40 µg antibody per mg of beads in 0.1 M sodium borate pH 9.5, 1 M ammonium sulfate overnight at 37°C with shaking. Unreacted groups were blocked overnight at 37°C with shaking in PBS containing 0.05% Tween 20, and 0.5% BSA. Antibody-coupled beads were stored at 4°C in the same buffer with 0.02% NaN3. The bead stock concentration was 50 g l−1, with the coupled antibody at 2 g l−1. D10 beads were washed with chicken saline with serum and were added to the P2 homogenate at 1 µl per ear; the mixture was then rotated overnight at 4°C. After collecting beads with a magnet, they were washed five times with chicken saline containing serum and three times with chicken saline. Pooled D10 beads were sonicated (Sonics & Materials sonicator, Newtown, CT; Cat# VCX 130) with a 2 mm probe in saline with protease inhibitors in 2–3 ml batches (in ice water). Sonication was for 5–10 s at 25–50% power, followed by cooling in ice water for 1–2 min. A magnet was used to concentrate the beads and the solution was removed. The sonication was repeated for a total of 20 ml of eluate; this solution was spun at 112,500 g (rmax; 35,000 rpm in a Beckman 70Ti rotor); the pellet was retained. Sonication was repeated on the D10 beads with six additional 3 ml aliquots; these aliquots were pooled and centrifuged. The supernatants from the two centrifugation steps were pooled (cytosolic fraction). Membrane pellets were resuspended using sonication with saline plus protease inhibitors and were combined; the pool was diluted to ∼500 ear- equivalents per tube. The solution was spun at 125,000 g (rmax; 45,000 rpm in Beckman TLA55 rotor) for 30 min at 4°C. The supernatant (S7) was removed and the pellet (crude stereocilia membranes) was frozen at −80°C. S7 was sonicated with 500 µl RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 0.1% SDS, 1% NP-40, 0.5% deoxycholate, 1:100 protease inhibitors) as above for each 500 ears; extracts were spun at 125,000 g (rmax) for 15 min at 4°C. The extraction was repeated twice on the pellet and the three supernatants were combined and diluted to 1.5 ml total volume (10 ears/30 µl). Immunoaffinity purification was carried out serially; the RIPA extract was first incubated with beads with control mouse IgG, then the unbound material was then incubated with beads coupled with 10G5 anti-GIPC3 antibody. The RIPA extract (1.5 ml; 500 ear equivalents) or flow-through material was added to 50 µl antibody-coupled beads; the beads and extract were rotated for 1 h at room temperature. Beads were collected with a magnet, washed at least five times with RIPA buffer, and eluted five times with 20 µl 2% SDS. eFASP was used to digest proteins to peptides and prepare samples for mass spectrometry (Erde et al., 2014). Reduction and alkylation were performed prior to filter aided exchange. Ammonium bicarbonate was added to 50 mM along with 10 mM dithiothreitol (DTT) and the samples heated at 95°C for 10 min. Iodoacetamide was then added to 20 mM and samples were incubated in the dark at 37°C for 1 h. Finally, DTT was added 15 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 to 10 mM to neutralize remaining iodoacetamide. 30 K Amicon Ultra centrifuge filters (MilliporeSigma Cat# UFC903024) were passivated with 5% Tween 20. The samples were exchanged four times using 0.1 M ammonium bicarbonate, 8 M urea, and 0.2% deoxycholic acid. They were then equilibrated three times in digestion buffer (50 mM ammonium bicarbonate and 0.2% deoxycholic acid). Finally, 200 ng trypsin was added in 100 µl digestion buffer and incubated at 37°C overnight. Deoxycholic acid was removed using ethyl acetate as described. Protein digests were separated using liquid chromatography with a NanoAcquity UPLC system (Waters, Milford, MA, USA); analytes were ionized using electrospray with a Nano Flex Ion Spray Source (Thermo Fisher Scientific) fitted with a 20 μm stainless steel nano-bore emitter spray tip and 2.6 kV source voltage, and were delivered to a QExactive HF (Thermo Fisher Scientific). Xcalibur version 4.1 was used to control the system. Samples were first bound to a trap cartridge (Symmetry C18 trap cartridge; Waters) at 10 μl min−1 for 10 min; the system then switched to a 75 μm×250 mm NanoAcquity BEH 130 C18 column with 1.7 μm particles (Waters) using mobile phases of water and acetonitrile containing 0.1% formic acid. A 7.5–30% acetonitrile gradient was delivered over 90 min at a flow rate of 300 nl min−1. Survey mass spectra were acquired in m/z 375−1400 at 120,000 resolution (at 200 m/z); data-dependent acquisition selected the top ten most abundant ions precursor ions for tandem mass spectrometry using an isolation width of 1.2 m/z. HCD fragmentation used normalized collision energy of 30 and a resolution of 30,000. Dynamic exclusion was set to auto, charge state for MS/MS +2 to +7, maximum ion time 100 ms, minimum AGC target of 3×106 in MS1 mode and 5×103 in MS2 mode. MaxQuant (Cox and Mann, 2008) and the search engine Andromeda (Cox et al., 2011) were used to identify peptides and assemble proteins from the mass spectrometry raw files. MaxQuant was used to calculate iBAQ (Schwanhäusser et al., 2011) for each protein, and we used an Excel spreadsheet to calculate riBAQ (Krey et al., 2014; Shin et al., 2013) and enrichment values. In vitro GST pulldown experiments Full-length GIPC3 (residues 1–297), GIPC3 PDZ-GH2 domains (93–297) and GIPC3 PDZ (93–181) were cloned into pMal-T-Avi-His/BirA (Addgene #102962) with N-terminal maltose-binding protein (MBP) and C-terminal biotin-acceptor peptide and His tags. Proteins were purified on amylose resin (New England Biolabs, Ipswich, MA, USA; Cat# E8021) and eluted with maltose using the manufacturer’s protocol. The HBCD domain of mouse Myo6 (residues 1052–1096) was cloned into a C-terminal GST expression vector pGEX-43T from the laboratory of James Bartles (formerly at Northwestern, Chicago, IL). The expressed protein was purified using glutathione–Sepharose 4B (Sigma-Aldrich Cat# GE17-0756- 01) using the manufacturer’s protocol. MBP- and GST-tagged proteins were purified by gel filtration on a Superdex 200 column (Sigma-Aldrich Cat# GE17-5175-01) and concentrated using a 30 K Amicon filter. Proteins were mixed together at 1 µM each final concentration in 40 µl of pulldown solution (150 mM NaCl, 1 mM EDTA, 10 mM Tris-HCl pH 7.5). After incubating for 1 h at room temperature, 40 µl of glutathione– Sepharose 4B was added to each tube and was incubated with rotation for an additional 1 h. The beads were washed three times with pulldown solution, transferred to mini-columns, then spun for 3 min to remove excess buffer. SDS at 95°C (25 µl) was added to each mini-column, incubated for 10 min, and spun to elute. Samples were separated by SDS-PAGE using a 4–12% NuPAGE Bis-Tris (Thermo Fisher Scientific Cat# NP0321BOX) in NuPAGE MOPS SDS running buffer (Thermo Fisher Scientific Cat# NP0001). After washing gels 4× with water, they were stained with Imperial Protein Stain (Thermo Fisher Scientific Cat# 24615) for 2 h and destained overnight. gel HeLa cell expression and NanoSPD experiments We maintained HeLa cells (ATCC Cat# CCL-2, RRID:CVCL_0030) in a humidified 5% (v/v) CO2 incubator at 37°C, using Eagle’s minimal essential medium (EMEM; ATCC Cat# 30-2003) that was supplemented with 10% serum (Serum Plus II, Sigma-Aldrich #14009C) and 10 ml l−1 penicillin- line was streptomycin (Sigma-Aldrich Cat# P4333). The HeLa cell authenticated by ATCC, and was free of mycoplasma contamination (mycoplasma detection kit, ATCC Cat# 30-1012 K). Cells were grown on acid-washed #1.5 thickness 22×22 mm cover glasses (Corning Cat# 2850-22) placed in six-well plates and coated with 0.025 poly-L-lysine (Sigma-Aldrich Cat# P1274). Cells were transfected at ∼60–70% confluency with Lipofectamine 3000 (Thermo Fisher Scientific Cat# following the manufacturer’s protocol and using 3.75 µl L3000015) Lipofectamine and 2.5 µg total plasmid DNA per well. At 24–36 h post transfection, cells were fixed in 4% formaldehyde (Electron Microscopy Sciences, Hatfield, PA, USA; Cat# 1570) in PBS for 15 min at room temperature and rinsed twice in PBS prior to staining. HeLa cells were double transfected at 60–70% confluency using Lipofectamine 3000. Cells were incubated at 37°C for ∼24 h post transfection, then fixed for 15 min in 4% formaldehyde at room temperature. Fixed cells were washed three times in 1× PBS before permeabilizing in 0.1% Triton X-100 in 1× PBS for 10 min, and then incubated with 1 U per ml CF405 phalloidin in 1× PBS for 2–3 h at room temperature. Cells were washed three times in 1× PBS, and coverslips mounted in Vectashield mounting medium (Thermo Fisher Scientific Cat # H-1000). Imaging was performed with the same setup described above. Cells were imaged such that the field of view contained one or two cells with clearly extended filopodia that did not contact other cells, and so the Z-stack encompassed the filopodia (i.e. the entire cell body was not imaged). If the confluency exceeded 80%, cells were re-seeded for the NanoSPD experiments to avoid overlapping filopodia; after re-seeding, transfection was performed as described here. Immunofluorescence Most IHCs imaged were from the higher frequency half of the apical region (from 1/6th to 2/6th of the distance from apex to base); we refer to these cells as apical IHCs. Some Airyscan images were acquired using a 63×, 1.4 NA Plan-Apochromat objective on a Zeiss 32-channel LSM 880 laser-scanning confocal microscope equipped with an Airyscan detector and run under software; Zeiss, Oberkochen, Germany) ZEISS ZEN (v2.6, 64-bit acquisition software. Other Airyscan images were acquired using a 63×, 1.4 NA Plan-Apochromat objective on a Zeiss 3-channel LSM 980 laser- scanning confocal microscope equipped with an Airyscan 2 detector and run under ZEISS ZEN (v3.1, 64-bit software; Zeiss) acquisition software. Settings for X-Y pixel resolution and Z-spacing, as well as pinhole diameter and grid selection, were set according to software-suggested settings for optimal Nyquist-based resolution. Processing of raw data for Airyscan- acquired images was performed using manufacturer-implemented automated settings. Display adjustments in brightness and contrast and reslices and/or average Z-projections were made in Fiji/ImageJ software. Surface rendering of LMO7 images was performed using Imaris version 9.9.0 (Oxford Instruments, Abingdon, UK), creation parameters. For cochlea imaging, for each antibody, two to four images were acquired from one or two cochlea per genotype per age for each experiment, and experiments were repeated at least twice. Ears from control and mutant littermates or from different ages of C57BL/6J mice of both sexes were stained and imaged on the same days for each experiment to limit variability. Genotyping was performed either prior to dissection or performed on tails collected during dissection for younger animals (<P8). Genotypes were known by the experimenter during staining and image acquisition. During image acquisition, the gain and laser settings for the antibody and phalloidin signals were adjusted to reveal the staining pattern in control samples, and the corresponding KO samples used the same settings. Image acquisition parameters and display adjustments were kept constant across ages and genotypes for every antibody/fluorophore combination. software-guided following Structured illumination (SIM) images were acquired with a 63×1.4 NA oil immersion lens on a Zeiss Elyra 7 microscope with dual PCO.edge 4.2 sCMOS cameras for detection. Grid selection and Z-spacing was guided by the software and kept consistent across images. Grid spacing was relaxed when imaging CF405 phalloidin as the illumination pattern lacked modulation and was kept consistent across all images. Post-acquisition processing was performed with software-recommended standard filtering for 488 and 568 nm channels. Processing was performed without baseline subtraction and with ‘scale to raw’ checked. Contrast was manually adjusted to retain both dim and bright structures in channels with high dynamic range. 16 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 the phalloidin channel Measurements of hair cell structures For measuring the apical dimensions during development, Z-stack images of apical IHCs from C57BL/6J, Gipc3KO/+, and Gipc3KO/KO mice at different developmental ages were collected using same image acquisition parameters. For cuticular plate area measurements, a Z-projection of a sub-stack of that showed the cuticular plate distinctively was created using Fiji/ImageJ from the original Z-stack images. Once the X-Y projections were generated, a drawing tablet with stylus was used to manually draw regions of interest (ROIs) of the cuticular plates using the freehand selection tool in Fiji/ImageJ. All ROIs were saved for each image. Next, using the ‘Analyze’ function in Fiji/ImageJ, the corresponding area under the bounding region was measured and tabulated for statistical analysis. To measure the cuticular plate depth at different time points during development, Z-stack images of the apical IHCs were collected from different developmental timepoints. For depth measurement using the phalloidin channel, X-Z reslices were generated using Fiji/ImageJ for each cell in the image field. Reslices were created by drawing a line passing through the fonticulus, through the middle of the IHC, for consistency across all the groups. Cuticular plate depths were measured individually in Fiji/ ImageJ by manually drawing a line from the top of the cuticular plate to the bottom of the plate in each of the re-slices. Next using the Analyze function in Fiji/ImageJ, the length of the line (indicating the depth of the cuticular plate) was measured and tabulated for statistical analysis. To measure the circumference of the cortical actin belt, P25.5–P27.5 cochleas were labeled with the anti-TJP1 antibody. Perimeters of the cortical actin belt were measured from Z-stacks of maximum projection images. Using a drawing tablet with stylus, ROIs of the cortical actin belt were manually drawn using the freehand selection tool in Fiji. All ROIs were saved for each image. Next using the ‘Analyze’ function in Fiji/ImageJ, the perimeter of the cortical actin belt was measured and tabulated for statistical analysis. For rootlet tracing, IHCs and OHCs labeled with anti-TRIOBP antibody were examined at P25.5. The row 1 rootlets of hair bundles were traced using the Multipoint tool function in Fiji/ImageJ. The X and Y positions were collected, then plotted to get the respective traces of the row 1 positions in the cuticular plate. Scanning electron microscopy For SEM, periotic bones with cochleas were dissected in Leibovitz’s L-15 medium (Thermo Fisher Scientific Cat# 21083-027) from P8.5 control and mutant littermates from Gipc3KO/+×Gipc3KO/KO crosses. An age-matched C57BL/6J control group was also included. Several small holes were made in periotic bones to provide access for fixative solutions; encapsulated cochleas were fixed for an hour in 2.5% glutaraldehyde in 0.1 M cacodylate buffer (Electron Microscopy Sciences, Cat# 15960) supplemented with 2 mM CaCl2. After washing with distilled water, the cochlear sensory epithelium was dissected out and the tectorial membrane was lifted off manually. Cochlear tissues were dehydrated in an ethanol series and critical- point dried using liquid CO2 (Leica EM CPD300). After immobilization on aluminum specimen holders using carbon tape, specimens were sputter coated with 3–4 nm of platinum (Leica EM ACE600). Samples were imaged using a Helios Nanolab 660 DualBeam Microscope (FEI). Auditory brainstem response measurements ABR experiments were carried out as described previously (Krey et al., 2016), using 8 Gipc3+/+, 28 Gipc3KO/+, and 18 Gipc3KO/KO animals. Animals were anesthetized with xylazine [10 mg/kg body weight, intramuscularly (i.m.), IVX; Animal Health Inc., Greeley, CO, USA] and ketamine [40 mg/kg, i.m.; Hospira, Inc., Lake Forest, IL, USA], and placed on a heating pad in a sound-isolated chamber. Needle electrodes were placed subcutaneously near the test ear, both at the vertex and at the shoulder of the test ear side. A closed-tube sound-delivery system, sealed into the ear canal, was used to stimulate each ear. ABR measurements used tone bursts with a 1 ms rise time, applied at 4, 8, 12, 16, 24 and 32 kHz. Responses were obtained for each ear, and the tone-burst stimulus intensity was increased in steps of 5 dB. The threshold was defined as an evoked response of 0.2 µV from the electrodes. Mechanotransduction measurements MET measurements were similar to those described previously (Krey et al., 2022). Mice were killed by decapitation using methods approved by the Stanford University Administrative Panel on Laboratory Animal Care, and inner ear tissue dissected from postnatal day 8–12 mice of either sex; genotype was typically unknown but usually could be determined by inspection of hair bundle morphology. Organ of Corti tissues from the 5– 12 kHz region were placed into a recording chamber as previously described (Peng et al., 2013). Hair cells were imaged on a BX51 upright fixed-stage microscope (Olympus, Pittsburgh, PA, USA) using a 100×1.0 NA dipping lens. The dissection and extracellular solution contained (in mM) 140 NaCl, 2 CaCl2, 0.5 MgCl2, 10 HEPES, 2 Na-ascorbate, 2 Na-pyruvate and 6 dextrose. The osmolality was 300–310 mOsm and pH was 7.4. The tectorial membrane was peeled off prior to mounting in the dish. After the recording chamber was placed onto the stage and apical and bath perfusion were started with the same solution. After a whole-cell recording was obtained, the apical perfusion was turned off to limit additional mechanical stimulation of the hair bundle or disruption of fluid jet flow. Whole-cell patch recordings were obtained using thick-walled borosilicate pipettes with electrode resistances of 3–5 MΩ, tip size of 1.5–2.2 µm inner diameter, pulled on a P95 micropipette puller from Sutter (Novato, CA, USA). The internal solution contained (in mM) 100 CsCl, 30 ascorbate, 3 Na2ATP, 5 phosphocreatine, 10 HEPES, 1 Cs4BAPTA [1,2-bis(o-aminophenoxy)ethane- N,N,N′,N′-tetraacetic acid]; osmolality was 290 mOsm and pH was 7.2. An Axopatch 200b amplifier (Molecular Devices, San Jose, CA, USA), coupled to a data acquisition board from IOtech (Measurement Computing Corporation, Norton, MA, USA; Cat# 3000) were used for all measurements. Data were sampled at 100 kHz and filtered with an 8-pole Bessel filter at 10 kHz. Junction potentials were estimated at 4 mV and corrected off-line. Uncompensated series resistance was 9±4 MΩ (n=19) and whole-cell capacitance was 10±3 pF (n=19). Cells were held at −80 mV for all experiments and Ca2+ currents were used as a quality control test for the recording. No difference was found in Ca2+ current properties between genotypes. Cells were included only if recordings remained stable throughout the timeframe of data capture. Hair bundles were stimulated with a fluid jet driven by a piezo electric disc bender 592 (27 mm 4.6 kHz; Murata Electronics, Nagaokakyo, Japan; Cat# 7BB-27-4L0). Discs were mounted in a 3D printed housing to minimize fluid volume being moved by the disc. Thin-walled borosilicate glass was used to deliver fluid to the bundle. Tip sizes of 10–15 µm diameter were selected as they uniformly stimulate inner hair cell bundles when placed 1–3 µm from the bundle face (Peng et al., 2021). Three cycles of a 40 Hz sine wave were used to activate MET channels. Voltage was varied to the disc bender to maximize current amplitudes. Maximal current was identified by the flattening of the peak response when channels were opened. littermates; cochleae were dissected out FM1-43 labeling To minimize entry via endocytosis, all solutions were prechilled to 4°C. Inner ears were isolated from P7.5–P8.5 C57BL/6 mice or from heterozygote and mutant in Hank’s balanced salt solution (HBSS; Thermo Fisher Scientific Cat# 14025092), supplemented with 5 mM HEPES, working as quickly as possible and treating the tissue as gently as possible to avoid link breakage. Cochleas were also left attached to the modiolus to avoid disruption. To control for non-specific dye uptake, one cochlea per animal was treated with 100 µM tubocurarine (Sigma-Aldrich Cat# T2379) and the other one without a transduction inhibitor. The dissected cochleas were transferred into wells with ice-cold HBSS, with or without tubocurarine, and left on ice for 5 min. During this time, wells were prepared with ice-cold HBSS containing 6 µM FM1-43FX (Thermo Fisher Scientific Cat# F35355) with or without tubocurarine. Cochleas were incubated in FM-143FX solution for 30 s on ice, then immediately transferred to wells with ice cold HBSS with or without tubocurarine and no FM1-43FX. Cochleas were washed twice with cold HBSS, then fixed in 4% formaldehyde in HBSS for 15–20 min. Next cochleas were rinsed twice with cold HBSS, the tectorial membrane was removed, and the cochleas were mounted in Vectashield mounting medium. Samples were immediately imaged on the confocal e c n e i c S l l e C f o l a n r u o J 17 RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 microscope. For analysis, regions of interest (ROIs) were drawn in Fiji/ ImageJ and average intensity projections were made for 3-4 sections of the Z-stack from the center. Signal intensities were measured in Fiji from ROIs drawn on the slices. Statistical analysis Experimental groups were specified by genotype and thus investigators did not allocate animals or samples. No samples were excluded. The investigator was not aware of the genotype during analysis, but the genotype was usually obvious from cochlear morphology. Unless otherwise stated, statistical comparisons between two sets of data used the unpaired two-tailed Student’s t-test with unpaired data and the untested assumptions of normal distribution and equal variance. In Fig. 5, we used the nested one- way ANOVA test with Tukey’s correction for multiple comparisons in Prism (www.graphpad.com) from different genotypes; while comparing the results from multiple cochleas per condition, the nested one-way ANOVA approach takes into account the structure of the data, that is, the variance in individual cell measurements for each condition (Eisner, 2021). The following software packages were used for data analysis: Microsoft Excel (www.microsoft.com/en-us/microsoft- 365/excel), GraphPad Prism (www.graphpad.com), Fiji/ImageJ (https:// imagej.net/software/fiji/), MaxQuant In figures, asterisks indicate: P<0.05; P<0.01; P<0.001; ***P<0.0001. to compare the results (www.maxquant.org). Acknowledgements We carried out mass spectrometry in the OHSU Proteomics Shared Resource, (partially supported by NIH core grants P30EY010572 and P30CA069533, and S10OD012246 for the Orbitrap Fusion) confocal microscopy in the OHSU Advanced Light Microscopy Core at The Jungers Center (P30NS061800 provided support for imaging), electron microscopy from the OHSU Multiscale Microscopy Core, and monoclonal antibody production from the Monoclonal Antibody Core of the OHSU Vaccine and Gene Therapy Institute. Competing interests The authors declare no competing or financial interests. Author contributions Conceptualization: P.C., C.P.M., A.J.R., P.G.B.-G.; Methodology: C.P.M., P.C., A.J.R., P.G.B.-G.; Validation: C.P.M., J.F.K.; Formal analysis: P.C., C.P.M., A.J.R., P.G.B.-G.; Investigation: P.C., C.P.M., J.F.K., C.B., J.G., M.B., A.J.R.; Data curation: P.G.B.-G.; Writing - original draft: P.G.B.-G.; Writing - review & editing: P.C., C.P.M., J.F.K., A.J.R., P.G.B.-G.; Visualization: P.C., J.F.K., A.J.R., P.G.B.-G.; Supervision: A.J.R., P.G.B.-G.; Project administration: A.J.R., P.G.B.-G.; Funding acquisition: A.J.R., P.G.B.-G. Funding A.J.R. was supported by National Institute on Deafness and Other Communication Disorders grant R01DC0003896; P.G.B.-G. was supported by National Institute on Deafness and Other Communication Disorders grant R01DC002368. Open access funding provided by Oregon Health & Science University. Deposited in PMC for immediate release. Data availability Mass spectrometry data, as well as spreadsheets with all derived values, are available from ProteomeXchange (http://www.proteomexchange.org) using the accession number PXD038234; information conforming to Minimal Information About a Proteomics Experiment (MIAPE) standards (Taylor et al., 2007) was included in the submission. Peer review history The peer review history is available online at https://journals.biologists.com/jcs/ lookup/doi/10.1242/jcs.261100.reviewer-comments.pdf References Avraham, K. B., Hasson, T., Steel, K. P., Kingsley, D. M. and Russell, L. B. (1995). The mouse Snell’s waltzer deafness gene encodes an unconventional myosin. Nat. Genet. 11, 369-375. doi:10.1038/ng1295-369 Avraham, K. B., Hasson, T., Sobe, T., Balsara, B., Testa, J. R., Skvorak, A. B., Morton, C. C., Copeland, N. G. and Jenkins, N. A. (1997). Characterization of unconventional MYO6, the gene responsible for deafness in Snell’s waltzer mice. Hum. Mol. Genet. 6, 1225-1231. doi:10.1093/ hmg/6.8.1225 the human homologue of Banani, S. F., Lee, H. O., Hyman, A. A. and Rosen, M. K. (2017). Biomolecular condensates: organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 18, 285-298. doi:10.1038/nrm.2017.7 Billington, N., Beach, J. R., Heissler, S. M., Remmert, K., Guzik-Lendrum, S., Nagy, A., Takagi, Y., Shao, L., Li, D., Yang, Y. et al. (2015). Myosin 18A coassembles with nonmuscle myosin 2 to form mixed bipolar filaments. Curr. Biol. 25, 942-948. doi:10.1016/j.cub.2015.02.012 Bird, J. E., Barzik, M., Drummond, M. C., Sutton, D. C., Goodman, S. M., Morozko, E. L., Cole, S. M., Boukhvalova, A. K., Skidmore, J., Syam, D., et al. (2017). Harnessing molecular motors for nanoscale pulldown in live cells. Mol. Biol. Cell 28, 463-475. doi:10.1091/mbc.e16-08-0583 Brown, S. D. M., Holmes, C. C., Mallon, A. M., Meehan, T. F., Smedley, D. and Wells, S. (2018). High-throughput mouse phenomics for characterizing mammalian gene function. Nat. Rev. Genet. 19, 357-370. doi:10.1038/s41576- 018-0005-2 Buss, F., Arden, S. D., Lindsay, M., Luzio, J. P. and Kendrick-Jones, J. (2001). Myosin VI isoform localized to clathrin-coated vesicles with a role in clathrin- mediated endocytosis. EMBO J. 20, 3676-3684. doi:10.1093/emboj/20.14.3676 Charizopoulou, N., Lelli, A., Schraders, M., Ray, K., Hildebrand, M. S., Ramesh, A., Srisailapathy, C. R., Oostrik, J., Admiraal, R. J., Neely, H. R. et al. (2011). Gipc3 mutations associated with audiogenic seizures and sensorineural hearing loss in mouse and human. Nat. Commun. 2, 201. doi:10.1038/ncomms1200 Corwin, J. T. and Warchol, M. E. (1991). Auditory hair cells: structure, function, development, and regeneration. Annu. Rev. Neurosci. 14, 301-333. doi:10.1146/ annurev.ne.14.030191.001505 Cox, J. and Mann, M. (2008). MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 26, 1367-1372. doi:10.1038/nbt.1511 Cox, J., Neuhauser, N., Michalski, A., Scheltema, R. A., Olsen, J. V. and Mann, M. (2011). Andromeda: a peptide search engine integrated into the MaxQuant environment. J. Proteome Res. 10, 1794-1805. doi:10.1021/pr101065j Driver, E. C. and Kelley, M. W. (2009). Specification of cell fate in the mammalian cochlea. Birth Defects Res. C Embryo Today 87, 212-221. doi:10.1002/bdrc. 20154 Du, T. T., Dewey, J. B., Wagner, E. L., Cui, R., Heo, J., Park, J.-J., Francis, S. P., Perez-Reyes, E., Guillot, S. J., Sherman, N. E. et al. (2019). LMO7 deficiency reveals the significance of the cuticular plate for hearing function. Nat. Commun. 10, 1117. doi:10.1038/s41467-019-09074-4 Eisner, D. A. (2021). Pseudoreplication in physiology: more means less. J. Gen. Physiol. 153, e202012826. doi:10.1085/jgp.202012826 Erde, J., Loo, R. R. and Loo, J. A. (2014). Enhanced FASP (eFASP) to increase proteome coverage and sample recovery for quantitative proteomic experiments. J. Proteome Res. 13, 1885-1895. doi:10.1021/pr4010019 Etournay, R., Lepelletier, L., Boutet de Monvel, J., Michel, V., Cayet, N., Leibovici, M., Weil, D., Foucher, I., Hardelin, J. P. and Petit, C. (2010). Cochlear outer hair cells undergo an apical circumference remodeling constrained by the hair bundle shape. Development 137, 1373-1383. doi:10.1242/dev.045138 Friedman, T. B., Belyantseva, I. A. and Frolenkov, G. I. (2020). Myosins and hearing. Adv. Exp. Med. Biol. 1239, 317-330. doi:10.1007/978-3-030-38062-5_13 Giese, A. P., Ezan, J., Wang, L., Lasvaux, L., Lembo, F., Mazzocco, C., Richard, E., Reboul, J., Borg, J. P., Kelley, M. W. et al. (2012). Gipc1 has a dual role in Vangl2 trafficking and hair bundle integrity in the inner ear. Development 139, 3775-3785. doi:10.1242/dev.074229 Gurumurthy, C. B., Sato, M., Nakamura, A., Inui, M., Kawano, N., Islam, M. A., Ogiwara, S., Takabayashi, S., Matsuyama, M., Nakagawa, S. et al. (2019). Creation of CRISPR-based germline-genome-engineered mice without ex vivo handling of zygotes by i-GONAD. Nat. Protoc. 14, 2452-2482. doi:10.1038/ s41596-019-0187-x Guzik-Lendrum, S., Heissler, S. M., Billington, N., Takagi, Y., Yang, Y., Knight, P. J., Homsher, E. and Sellers, J. R. (2013). Mammalian myosin-18A, a highly divergent myosin. J. Biol. Chem. 288, 9532-9548. doi:10.1074/jbc.M112.441238 Hasson, T., Gillespie, P. G., Garcia, J. A., MacDonald, R. B., Zhao, Y., Yee, A. G., Mooseker, M. S. and Corey, D. P. (1997). Unconventional myosins in inner-ear sensory epithelia. J. Cell Biol. 137, 1287-1307. doi:10.1083/jcb.137.6.1287 Hua, Y., Ding, X., Wang, H., Wang, F., Lu, Y., Neef, J., Gao, Y., Moser, T. and Wu, H. (2021). Electron microscopic reconstruction of neural circuitry in the cochlea. Cell Rep. 34, 108551. doi:10.1016/j.celrep.2020.108551 Kaltenbach, J. A., Falzarano, P. R. and Simpson, T. H. (1994). Postnatal development of the hamster cochlea. II. Growth and differentiation of stereocilia bundles. J. Comp. Neurol. 350, 187-198. doi:10.1002/cne.903500204 Katoh, M. (2013). Functional proteomics, human genetics and cancer biology of GIPC family members. Exp. Mol. Med. 45, e26. doi:10.1038/emm.2013.49 Krey, J. F., Wilmarth, P. A., Shin, J. B., Klimek, J., Sherman, N. E., Jeffery, E. D., Choi, D., David, L. L. and Barr-Gillespie, P. G. (2014). Accurate label-free protein quantitation with high- and low-resolution mass spectrometers. J. Proteome Res. 13, 1034-1044. doi:10.1021/pr401017h Krey, J. F., Sherman, N. E., Jeffery, E. D., Choi, D. and Barr-Gillespie, P. G. (2015). The proteome of mouse vestibular hair bundles over development. Sci. Data 2, 150047. doi:10.1038/sdata.2015.47 18 e c n e i c S l l e C f o l a n r u o J RESEARCH ARTICLE Journal of Cell Science (2023) 136, jcs261100. doi:10.1242/jcs.261100 Krey, J. F., Drummond, M., Foster, S., Porsov, E., Vijayakumar, S., Choi, D., Friderici, K., Jones, S. M., Nuttall, A. L. and Barr-Gillespie, P. G. (2016). Annexin A5 is the most abundant membrane-associated protein in stereocilia but is dispensable for hair-bundle development and function. Sci. Rep. 6, 27221. doi:10.1038/srep27221 Krey, J. F., Scheffer, D. I., Choi, D., Reddy, A., David, L. L., Corey, D. P. and Barr- Gillespie, P. G. (2018). Mass spectrometry quantitation of proteins from small pools of developing auditory and vestibular cells. Sci. Data 5, 180128. doi:10. 1038/sdata.2018.128 Krey, J. F., Liu, C., Belyantseva, I. A., Bateschell, M., Dumont, R. A., Goldsmith, J., Chatterjee, P., Morrill, R. S., Fedorov, L. M., Foster, S. et al. (2022). ANKRD24 organizes TRIOBP to reinforce stereocilia insertion points. J. Cell Biol. 221, e202109134. doi:10.1083/jcb.202109134 Lin, D. C., Quevedo, C., Brewer, N. E., Bell, A., Testa, J. R., Grimes, M. L., Miller, F. D. and Kaplan, D. R. (2006). APPL1 associates with TrkA and GIPC1 and is required for nerve growth factor-mediated signal transduction. Mol. Cell. Biol. 26, 8928-8941. doi:10.1128/MCB.00228-06 Maddugoda, M. P., Crampton, M. S., Shewan, A. M. and Yap, A. S. (2007). Myosin VI and vinculin cooperate during the morphogenesis of cadherin cell cell contacts in mammalian epithelial cells. J. Cell Biol. 178, 529-540. doi:10.1083/jcb. 200612042 Mattson, G., Conklin, E., Desai, S., Nielander, G., Savage, M. D. and Morgensen, S. (1993). A practical approach to crosslinking. Mol. Biol. Rep. 17, 167-183. doi:10.1007/BF00986726 Meyers, J. R., MacDonald, R. B., Duggan, A., Lenzi, D., Standaert, D. G., Corwin, J. T. and Corey, D. P. (2003). Lighting up the senses: FM1-43 loading of sensory cells through nonselective ion channels. J. Neurosci. 23, 4054-4065. doi:10.1523/ JNEUROSCI.23-10-04054.2003 Morgan, C. P., Krey, J. F., Grati, M., Zhao, B., Fallen, S., Kannan-Sundhari, A., Liu, X. Z., Choi, D., Mü ller, U. and Barr-Gillespie, P. G. (2016). PDZD7-MYO7A complex identified in enriched stereocilia membranes. Elife 5, e18312. doi:10. 7554/eLife.18312 Naccache, S. N., Hasson, T. and Horowitz, A. (2006). Binding of internalized receptors to the PDZ domain of GIPC/synectin recruits myosin VI to endocytic vesicles. Proc. Natl. Acad. Sci. USA 103, 12735-12740. doi:10.1073/pnas. 0605317103 Ohtsuka, M. and Sato, M. (2019). i-GONAD: A method for generating genome- edited animals without ex vivo handling of embryos. Dev. Growth Differ. 61, 306-315. doi:10.1111/dgd.12620 Ohtsuka, M., Sato, M., Miura, H., Takabayashi, S., Matsuyama, M., Koyano, T., Arifin, N., Nakamura, S., Wada, K. and Gurumurthy, C. B. (2018). i-GONAD: a robust method for in situ germline genome engineering using CRISPR nucleases. Genome Biol. 19, 25. doi:10.1186/s13059-018-1400-x Oughtred, R., Rust, J., Chang, C., Breitkreutz, B. J., Stark, C., Willems, A., Boucher, L., Leung, G., Kolas, N., Zhang, F., et al. (2021). The BioGRID database: A comprehensive biomedical resource of curated protein, genetic, and chemical interactions. Protein Sci. 30, 187-200. doi:10.1002/pro.3978 Peng, A. W., Effertz, T. and Ricci, A. J. (2013). Adaptation of mammalian auditory hair cell mechanotransduction is independent of calcium entry. Neuron 80, 960-972. doi:10.1016/j.neuron.2013.08.025 Peng, A. W., Scharr, A. L., Caprara, G. A., Nettles, D., Steele, C. R. and Ricci, A. J. (2021). Fluid jet stimulation of auditory hair bundles reveal spatial non- uniformities and two viscoelastic-like mechanisms. Front. Cell Dev. Biol. 9, 725101. doi:10.3389/fcell.2021.725101 Pollock, L. M. and McDermott, B. M. (2015). The cuticular plate: a riddle, wrapped in a mystery, inside a hair cell. Birth Defects Res. C Embryo Today 105, 126-139. doi:10.1002/bdrc.21098 Reed, B. C., Cefalu, C., Bellaire, B. H., Cardelli, J. A., Louis, T., Salamon, J., Bloecher, M. A. and Bunn, R. C. (2005). GLUT1CBP(TIP2/GIPC1) interactions with GLUT1 and myosin VI: evidence supporting an adapter function for GLUT1CBP. Mol. Biol. Cell 16, 4183-4201. doi:10.1091/mbc.e04-11-0978 Rehman, A. U., Gul, K., Morell, R. J., Lee, K., Ahmed, Z. M., Riazuddin, S., Ali, R. A., Shahzad, M., Jaleel, A. U., Andrade, P. B. et al. (2011). Mutations of GIPC3 cause nonsyndromic hearing loss DFNB72 but not DFNB81 that also maps to chromosome 19p. Hum. Genet. 130, 759-765. doi:10.1007/s00439-011- 1018-5 Scheffer, D. I., Shen, J., Corey, D. P. and Chen, Z. Y. (2015). Gene expression by mouse inner ear hair cells during development. J. Neurosci. 35, 6366-6380. doi:10.1523/JNEUROSCI.5126-14.2015 Schwanhä usser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., Chen, W. and Selbach, M. (2011). Global quantification of mammalian gene expression control. Nature 473, 337-342. doi:10.1038/nature10098 Schwenk, F., Baron, U. and Rajewsky, K. (1995). A cre-transgenic mouse strain for the ubiquitous deletion of loxP-flanked gene segments including deletion in germ cells. Nucleic Acids Res. 23, 5080-5081. doi:10.1093/nar/23.24.5080 Self, T., Sobe, T., Copeland, N. G., Jenkins, N. A., Avraham, K. B. and Steel, K. P. (1999). Role of myosin VI in the differentiation of cochlear hair cells. Dev. Biol. 214, 331-341. doi:10.1006/dbio.1999.9424 Shang, G., Brautigam, C. A., Chen, R., Lu, D., Torres-Vázquez, J. and Zhang, X. (2017). Structure analyses reveal a regulated oligomerization mechanism of the PlexinD1/GIPC/myosin VI complex. Elife 6, e27322. doi:10.7554/eLife.27322 Shin, J.-B., Krey, J. F., Hassan, A., Metlagel, Z., Tauscher, A. N., Pagana, J. M., Sherman, N. E., Jeffery, E. D., Spinelli, K. J., Zhao, H. et al. (2013). Molecular architecture of the chick vestibular hair bundle. Nat. Neurosci. 16, 365-374. doi:10. 1038/nn.3312 Simonneau, L., Gallego, M. and Pujol, R. (2003). Comparative expression patterns of T-, N-, E-cadherins, beta-catenin, and polysialic acid neural cell adhesion molecule in rat cochlea during development: implications for the nature of Kö lliker’s organ. J. Comp. Neurol. 459, 113-126. doi:10.1002/cne.10604 Slepecky, N. and Chamberlain, S. C. (1985). Immunoelectron microscopic and immunofluorescent localization of cytoskeletal and muscle-like contractile proteins in inner ear sensory hair cells. Hear. Res. 20, 245-260. doi:10.1016/ 0378-5955(85)90029-2 Slepecky, N., Hamernik, R. and Henderson, D. (1981). The consistent occurrence of a striated organelle (Friedmann body) in the inner hair cells of the normal chinchilla. Acta Otolaryngol. 91, 189-198. doi:10.3109/00016488109138499 Taylor, C. F., Paton, N. W., Lilley, K. S., Binz, P.-A., Julian, R. K., Jones, A. R., Zhu, W., Apweiler, R., Aebersold, R., Deutsch, E. W. et al. (2007). The minimum information about a proteomics experiment (MIAPE). Nat. Biotechnol. 25, 887-893. doi:10.1038/nbt1329 Valenta, T., Hausmann, G. and Basler, K. (2012). The many faces and functions of β-catenin. EMBO J. 31, 2714-2736. doi:10.1038/emboj.2012.150 Varsano, T., Dong, M. Q., Niesman, I., Gacula, H., Lou, X., Ma, T., Testa, J. R., Yates, J. R. and Farquhar, M. G. (2006). GIPC is recruited by APPL to peripheral TrkA endosomes and regulates TrkA trafficking and signaling. Mol. Cell. Biol. 26, 8942-8952. doi:10.1128/MCB.00305-06 Vranceanu, F., Perkins, G. A., Terada, M., Chidavaenzi, R. L., Ellisman, M. H. and Lysakowski, A. (2012). Striated organelle, a cytoskeletal structure positioned to modulate hair-cell transduction. Proc. Natl. Acad. Sci. USA 109, 4473-4478. doi:10.1073/pnas.1101003109 e c n e i c S l l e C f o l a n r u o J 19
10.3390_ijms241210388	Article Quantitative Loop-Mediated Isothermal Ampliﬁcation Detection of Ustilaginoidea virens Causing Rice False Smut Yu Zhang, Xinyue Li †, Shuya Zhang, Tianling Ma , Chengxin Mao and Chuanqing Zhang Department of Plant Pathology, Zhejiang Agriculture and Forest University, Hangzhou 311300, China; [email protected] (Y.Z.); [email protected] (X.L.); [email protected] (S.Z.); [email protected] (C.M.) * Correspondence: [email protected] (T.M.); [email protected] (C.Z.) † Current address: Station of Agriculture Techniques of Zhenhai District, Ningbo 315200, China. Abstract: Rice false smut caused by Ustilaginoidea virens is one of the most devastating diseases in rice worldwide, which results in serious reductions in rice quality and yield. As an airborne fungal disease, early diagnosis of rice false smut and monitoring its epidemics and distribution of its pathogens is particularly important to manage the infection. In this study, a quantitative loop-mediated isothermal ampliﬁcation (q-LAMP) method for U. virens detection and quantiﬁcation was developed. This method has higher sensitivity and efﬁciency compared to the quantitative real-time PCR (q-PCR) method. The species-speciﬁc primer that the UV-2 set used was designed based on the unique sequence of the U. virens ustiloxins biosynthetic gene (NCBI accession number: BR001221.1). The q-LAMP assay was able to detect a concentration of 6.4 spores/mL at an optimal reaction temperature of 63.4 ◦C within 60 min. Moreover, the q-LAMP assay could even achieve accurate quantitative detection when there were only nine spores on the tape. A linearized equation for the standard curve, y = −0.2866x + 13.829 (x is the ampliﬁcation time, the spore number = 100.65y), was established for the detection and quantiﬁcation of U. virens. In ﬁeld detection applications, this q-LAMP method is more accurate and sensitive than traditional observation methods. Collectively, this study has established a powerful and simple monitoring tool for U. virens, which provides valuable technical support for the forecast and management of rice false smut, and a theoretical basis for precise fungicide application. Keywords: rice false smut; quantitative loop-mediated isothermal ampliﬁcation (q-LAMP); detection; ustiloxins biosynthetic gene 1. Introduction Rice false smut is a disease affecting rice spikes that occurs from the ﬂowering to the milking stage [1,2]. Its most typical and visible symptom is the replacement of rice grains with false smut balls [3,4]. It occurs mainly in Asian countries such as China, Japan, Korea, the Philippines, and India, and is one of the most devastating diseases in the world’s major rice producing regions [5–8]. In recent years, due to the promotion of short-stalked compact and high-yielding rice varieties, indica–japonica interspeciﬁc hybrid rice combinations, changes in cultivation patterns, and the excessive use of nitrogen fertilizer during the tillering and gestation periods, the occurrence of rice false smut has become increasingly serious and has gradually risen from a previously minor or sporadic disease to become one of the three new major diseases affecting rice in China [9,10]. The damage caused not only results in a decrease in rice quality and yield, but also the generation of mycotoxin ustiloxins on infected rice spikelets [11,12]. As antimitotic cyclopeptide mycotoxins, the ustiloxins produced within a false smut ball can inhibit microtubule assembly and cell skeleton formation, which poses a serious threat to farmland preservation and ecosystems, as well as the health of humans and animals [13]. Strategies to manage this devastating disease are therefore urgently needed. Citation: Zhang, Y.; Li, X.; Zhang, S.; Ma, T.; Mao, C.; Zhang, C. Quantitative Loop-Mediated Isothermal Ampliﬁcation Detection of Ustilaginoidea virens Causing Rice False Smut. Int. J. Mol. Sci. 2023, 24, 10388. https://doi.org/10.3390/ ijms241210388 Academic Editor: Fucheng Lin Received: 28 April 2023 Revised: 14 June 2023 Accepted: 19 June 2023 Published: 20 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Int. J. Mol. Sci. 2023, 24, 10388. https://doi.org/10.3390/ijms241210388 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10388 2 of 13 It is widely accepted that Ustilaginoidea virens (teleomorph Villosiclava virens) is the causal agent of rice false smut [14,15]. As a typical airborne disease, virulent pathogen spores land on the surface of a rice spikelet and germinate hyphae as well as false smut balls on the spikelet [3,12,16–18]. Thus, the epidemic of rice false smut is closely related to the amount of U. virens spores in the ﬁeld, and the diagnosis of rice false smut, combined with accurate detection and spore quantiﬁcation, is of great importance for its prevention and management [5,19]. Traditionally, the microscopic counting of spores after capture is widely used in rice false smut diagnosis; however, this method requires specialist taxonomic technicians [20]. Given the complexity of environmental samples and human subjectivity, it is difﬁcult to obtain reliable data with high efﬁciency via microscopic analysis. A quantitative real-time PCR (q-PCR) technique has been applied for the early identiﬁcation and quantiﬁcation of pathogens in airborne diseases [21]. However, this technique is susceptible to interference from environmental dust and other pathogens, making it difﬁcult to quantify the low concentrations of spores captured [20]. in Loop-mediated isothermal ampliﬁcation (LAMP), developed by Notomi et al. 2000, is a non-PCR-based nucleic acid ampliﬁcation technique that can be used for the molecular detection of various bacteria, viruses, fungi in disease diagnosis [22–24]. The LAMP reaction is carried out at a constant temperature (between 60 and 65 ◦C) in less than an hour through the use of two pairs of primers—an inner primer (FIP/BIP) and an outer primer (F3/B3). These two pairs of primers constitute the basic LAMP primer set for the LAMP reaction, in order to recognize speciﬁc nucleic acid sequences of monitored targets [25–27]. An additional pair of LAMP primers, loop primers, can also be used to signiﬁcantly improve LAMP efﬁciency. A simple and visual LAMP assay was developed by Yang et al. in 2018 for the rapid diagnosis of U. virens [28]. However, this assay cannot be used directly for quantitative detection of complex DNA samples. The quantitative- LAMP (q-LAMP) assay (DiaSorin S.p.A., Saluggia, Italy) is a technical improvement from the classical LAMP, which combines LAMP technology with the real-time ﬂuorescence quantitative PCR technique [29]. It is based on the addition of nucleic acid ﬂuorescent dyes, such as SYBR Green or SYTO, resulting in a more sophisticated method suitable for the needs of ﬁeld diagnosis [30,31]. In this study, we aimed to design and develop a speciﬁc and sensitive q-LAMP assay for detection and quantiﬁcation U. virens, which can be applied in the early diagnosis of rice false smut for preventing the spread of this devastating airborne disease. Additionally, this study is the ﬁrst report to describe a quantitative diagnostic test for the detection of U. virens using q-LAMP. 2. Results 2.1. Design of Primers for U. virens Detection The best LAMP UV-2 primers were designed based on the ustiloxins biosynthetic gene sequence of U. virens (NCBI accession number: BR001221.1) that did not show any simi- larities to other sequences available in the National Center for Biotechnology Information (NCBI) GenBank database, in order to allow speciﬁc ampliﬁcation of U. virens (Figure 1, Table 1) [32–34]. Additionally, the UV-2 primer sets met the requirement that ∆G values must be less than or equal to −4 Kcal/mol at the 3(cid:48)end of F3/B3 and F2/B2 and 5(cid:48)ends of F1c and B1c. 2.2. Optimization of the q-LAMP Assay To optimize the q-LAMP assay system, the q-LAMP assay was carried out using the UV-2 primer sets at temperatures ranging from 61.8 ◦C to 66 ◦C. As shown in Figure 2, the ﬂuorescence quantitative results show that the strongest ﬂuorescence intensity and the shortest reaction time were obtained when the reaction temperature was 63.4 ◦C (which reached the ampliﬁcation peak at 30 min). Thus, 63.4 ◦C was chosen as the reaction temperature at which to carry out the optimal q-LAMP assays. Int. J. Mol. Sci. 2023, 24, 10388 3 of 13 Figure 1. The species-speciﬁc primers for detecting Ustilaginoidea virens in the quantitative loop- mediated isothermal ampliﬁcation (q-LAMP) and quantitative real-time PCR (q-PCR). The species- speciﬁc primers designed based on the sequence of the ustiloxins biosynthetic gene segments for identiﬁcation and quantiﬁcation of U. virens in q-LAMP assay and q-PCR assay. The forward and reverse primer sequences were highlighted with shade and arrow for orientation. Table 1. The sequences of species-speciﬁc primers used in the quantitative loop-mediated isothermal ampliﬁcation (q-LAMP) assay and quantitative real-time PCR (q-PCR) assay. Serial Number UV-2 Sequence (5(cid:48)–3(cid:48)) F3 B3 FIP(F1c-F2) BIP(B1c-B2) GGCACAGCATGACAGGATG TGCTCCCACACTGGTAGT CCTGACATGGCCGGTTTCCCGACGCATGGCCAATAACTCC AGCGGGGCACTTAGGTTCTGCCAATCAAGGCAGCTGATCT Figure 2. Optimization of the q-LAMP assay via reaction temperature screening. The inﬂuence of temperature ranged from 61.8 ◦C to 66 ◦C in the q-LAMP detection system and showed that the strongest ﬂuorescence intensity and the shortest reaction time were obtained at 63.4 ◦C (black line). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 3 of 14 Table 1. The sequences of species-speciﬁc primers used in the quantitative loop-mediated isother-mal ampliﬁcation (q-LAMP) assay and quantitative real-time PCR (q-PCR) assay. Serial Number Sequence (5′–3′) UV-2 F3 GGCACAGCATGACAGGATG B3 TGCTCCCACACTGGTAGT FIP(F1c-F2) CCTGACATGGCCGGTTTCCCGACGCATGGCCAATAACTCC BIP(B1c-B2) AGCGGGGCACTTAGGTTCTGCCAATCAAGGCAGCTGATCT Figure 1. The species-speciﬁc primers for detecting Ustilaginoidea virens in the quantitative loop-mediated isothermal ampliﬁcation (q-LAMP) and quantitative real-time PCR (q-PCR). The species-speciﬁc primers designed based on the sequence of the ustiloxins biosynthetic gene segments for identiﬁcation and quantiﬁcation of U. virens in q-LAMP assay and q-PCR assay. The forward and reverse primer sequences were highlighted with shade and arrow for orientation. 2.2. Optimization of the q-LAMP Assay To optimize the q-LAMP assay system, the q-LAMP assay was carried out using the UV-2 primer sets at temperatures ranging from 61.8 °C to 66 °C. As shown in Figure 2, the ﬂuorescence quantitative results show that the strongest ﬂuorescence intensity and the shortest reaction time were obtained when the reaction temperature was 63.4 °C (which reached the ampliﬁcation peak at 30 min). Thus, 63.4 °C was chosen as the reaction tem-perature at which to carry out the optimal q-LAMP assays. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of 14 Figure 2. Optimization of the q-LAMP assay via reaction temperature screening. The inﬂuence of temperature ranged from 61.8 °C to 66 °C in the q-LAMP detection system and showed that the strongest ﬂuorescence intensity and the shortest reaction time were obtained at 63.4 °C (black line). 2.3. Speciﬁcity Validation of the q-LAMP Assay System The speciﬁcity validation of design in the q-LAMP assay system (using UV-2 primer sets and 63.4 °C as reaction temperature) was tested using the U. virens stain and the other nine fungi. The results show that ﬂuorescence signals were detected in the samples with the DNA template of U. virens, while the samples with the DNA template of the other nine fungi or ddH2O (negative control) did not show any ﬂuorescence signal (Figure 3), indi-cating that the design of the q-LAMP assay system was highly speciﬁc to the detection of U. virens. Figure 3. Speciﬁcity validation of the q-LAMP assay system. The q-LAMP assay system (using UV-2 primer sets and 63.4 °C as reaction temperature) was highly speciﬁc for the detection of U. virens. The q-LAMP assay system showed that ﬂuorescence signals were only detected in the samples with DNA template of U. virens (black line), while the samples with DNA template of the other 9 fungi (including Fusarium fujikuroi, F. oxysporum, F. proliferatum, F. solani, F. graminearum, Penicillium sp., Int. J. Mol. Sci. 2023, 24, 10388 4 of 13 2.3. Speciﬁcity Validation of the q-LAMP Assay System The speciﬁcity validation of design in the q-LAMP assay system (using UV-2 primer sets and 63.4 ◦C as reaction temperature) was tested using the U. virens stain and the other nine fungi. The results show that ﬂuorescence signals were detected in the samples with the DNA template of U. virens, while the samples with the DNA template of the other nine fungi or ddH2O (negative control) did not show any ﬂuorescence signal (Figure 3), indicating that the design of the q-LAMP assay system was highly speciﬁc to the detection of U. virens. Figure 3. Speciﬁcity validation of the q-LAMP assay system. The q-LAMP assay system (using UV-2 primer sets and 63.4 ◦C as reaction temperature) was highly speciﬁc for the detection of U. virens. The q-LAMP assay system showed that ﬂuorescence signals were only detected in the samples with DNA template of U. virens (black line), while the samples with DNA template of the other 9 fungi (including Fusarium fujikuroi, F. oxysporum, F. proliferatum, F. solani, F. graminearum, Penicillium sp., Pyricularia oryzae, Alternaria alternata and Rhizoctonia solani) or negative control (nucleic acid-free water) did not show any ﬂuorescence signal. 2.4. Sensitivity Validation of the q-LAMP Assay System The sensitivity validation of the q-LAMP assay was determined using the genomic DNA of gradient dilution of U. virens spores as templates under optimal conditions (using primer sets UV-2 and 63.4 ◦C as reaction temperature). As shown in Table 2 and Figure 4, the fluo- rescence signals were detected in the samples with 2 × 104 spores/mL, 4 × 103 spores/mL, 8 × 102 spores/mL, 1.6 × 102 spores/mL, 32 spores/mL, and 6.4 spores/mL within 60 min (the spore extracts were used as DNA template in q-LAMP assay system), while no signals were detected in the sample with the DNA template of 1.28 spores/mL. Thus, theoretically, the q-LAMP assay was able to detect the sample with a concentration of 6.4 spores/mL. We also compared the sensitivity of the q-LAMP assay system with quantitative real-time PCR (q-PCR) for U. virens detection. The q-PCR assay was carried out using the FB/B3 primer set and the effective ampliﬁcation reactions were detected in samples with spore concentrations of 2 × 104, 4 × 103, 8 × 102, 1.6 × 102 spores/mL, but not 32 spores/mL (Supplementary Figure S1). Thus, the q-LAMP assay system is more sensitive and efﬁcient compared to the q-PCR system used in this study. 2.5. Establishment of a Standard Curve for q-LAMP Detection of U. virens A standard curve between the ampliﬁcation time (x) and the Log10 value of spore number (y) was constructed based on the q-LAMP assay: y = −0.2866x + 13.829 (Figure 5, Supplementary Table S1), the formula used for calculating spore number is 100.65y, and the correlation coefﬁcient R2 = 0.9942, showing a good linear relationship. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of 14 Figure 2. Optimization of the q-LAMP assay via reaction temperature screening. The inﬂuence of temperature ranged from 61.8 °C to 66 °C in the q-LAMP detection system and showed that the strongest ﬂuorescence intensity and the shortest reaction time were obtained at 63.4 °C (black line). 2.3. Speciﬁcity Validation of the q-LAMP Assay System The speciﬁcity validation of design in the q-LAMP assay system (using UV-2 primer sets and 63.4 °C as reaction temperature) was tested using the U. virens stain and the other nine fungi. The results show that ﬂuorescence signals were detected in the samples with the DNA template of U. virens, while the samples with the DNA template of the other nine fungi or ddH2O (negative control) did not show any ﬂuorescence signal (Figure 3), indi-cating that the design of the q-LAMP assay system was highly speciﬁc to the detection of U. virens. Figure 3. Speciﬁcity validation of the q-LAMP assay system. The q-LAMP assay system (using UV-2 primer sets and 63.4 °C as reaction temperature) was highly speciﬁc for the detection of U. virens. The q-LAMP assay system showed that ﬂuorescence signals were only detected in the samples with DNA template of U. virens (black line), while the samples with DNA template of the other 9 fungi (including Fusarium fujikuroi, F. oxysporum, F. proliferatum, F. solani, F. graminearum, Penicillium sp., Int. J. Mol. Sci. 2023, 24, 10388 5 of 13 Table 2. The time for ﬂuorescence signal to reach the ﬂuorescence threshold and ﬂuorescence signal records in the q-LAMP assays for testing samples with known spore concentration. Spore Concentration (Spores/mL) Time a (min) (Mean ± Standard Deviation) Fluorescence Signals b 2 × 104 4 × 103 8 × 102 1.6 × 102 3.2 × 101 6.4 1.28 CK c 17.90 ± 1.64 31.44 ± 0.71 34.68 ± 1.26 37.92 ± 1.53 41.16 ± 0.98 44.40 ± 2.42 + + + + + + − − a Time for ﬂuorescence signal reaching the ﬂuorescence threshold in the q-LAMP assays. b “+” indicates a successful ﬂuorescence signal detection, “−” indicates no ﬂuorescence signal detected in the q-LAMP assays. c The nucleic acid-free water was used negative control (CK) in the q-LAMP assay. Figure 4. Sensitivity validation of q-LAMP system. The ﬂuorescence signals in q-LAMP assays were detected in the samples with DNA template of 2 × 104 spores/mL, 4 × 103 spores/mL, 8 × 102 spores/mL, 1.6 × 102 spores/mL, 32 spores/mL, and 6.4 spores/mL within 60 min, while no signals were detected in sample with DNA template of 1.28 spores/mL and CK. The bolded dark green line (horizontal) indicates ﬂuorescence threshold. Fluorescence signals above this threshold marked as a successful detection of U. virens in q-LAMP assays. 2.6. Application of q-LAMP Assay for U. virens Spore Calculation The standard curve of q-LAMP was applied to calculate U. virens spore number on tapes, and each tape sample contained 450, 116, 29, and 9 manually added spores, respectively. As shown in Table 3 and Figure 6, the ampliﬁcation times quantitated using the cycle threshold (Ct) values for the tested samples were 34.03, 37.12, 40.46, and 43.17, corresponding to 446.07, 118.51, 28.29, and 8.85 predicted spores per tape, respectively, which is very close to the actual spore number on each Melinex tape. Thus, this q-LAMP system can efﬁciently quantitate U. virens spore number with high accuracy. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 6 of 14 Figure 4. Sensitivity validation of q-LAMP system. The ﬂuorescence signals in q-LAMP assays were detected in the samples with DNA template of 2 × 104 spores/mL, 4 × 103 spores/mL, 8 × 102 spores/mL, 1.6 × 102 spores/mL, 32 spores/mL, and 6.4 spores/mL within 60 min, while no signals were detected in sample with DNA template of 1.28 spores/mL and CK. The bolded dark green line (horizontal) indicates ﬂuorescence threshold. Fluorescence signals above this threshold marked as a successful detection of U. virens in q-LAMP assays. 2.5. Establishment of a Standard Curve for q-LAMP Detection of U. virens A standard curve between the ampliﬁcation time (x) and the Log10 value of spore number (y) was constructed based on the q-LAMP assay: y = −0.2866x + 13.829 (Figure 5, Supplementary Table S1), the formula used for calculating spore number is 100.65y, and the correlation coeﬃcient R2 = 0.9942, showing a good linear relationship. Figure 5. Standard curve of q-LAMP detection system. A standard curve between logarithmic values of the spore number (y) and the ampliﬁcation time quantitated using the cycle threshold (Ct) values (x): y = −0.2866x + 13.829. The correlation coeﬃcient (R2) is 0.9942, showing a good linear relation-ship. Int. J. Mol. Sci. 2023, 24, 10388 6 of 13 Figure 5. Standard curve of q-LAMP detection system. A standard curve between logarithmic values of the spore number (y) and the ampliﬁcation time quantitated using the cycle threshold (Ct) values (x): y = −0.2866x + 13.829. The correlation coefﬁcient (R2) is 0.9942, showing a good linear relationship. Table 3. Quantitative detection of U. virens spores using q-LAMP system. Ct a 34.03 37.12 40.46 43.17 Manually Added Spores (Spores/mL) Predictive Spores (Spores/mL) R2 p Value 450 116 29 9 446.07 118.51 28.29 8.85 0.999 0.639 a the ampliﬁcation times (x) quantitated using the cycle threshold (Ct) values. Figure 6. Quantitative detection of U. virens spores on Melinex tape using q-LAMP system. Serial numbers 1, 2, 3, and 4 represent 450, 116, 29, and 9 spores, respectively. The green line (horizontal) indicates ﬂuorescence threshold. Fluorescence signals above this threshold marked as a successful detection of U. virens in q-LAMP assays. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 6 of 14 Figure 4. Sensitivity validation of q-LAMP system. The ﬂuorescence signals in q-LAMP assays were detected in the samples with DNA template of 2 × 104 spores/mL, 4 × 103 spores/mL, 8 × 102 spores/mL, 1.6 × 102 spores/mL, 32 spores/mL, and 6.4 spores/mL within 60 min, while no signals were detected in sample with DNA template of 1.28 spores/mL and CK. The bolded dark green line (horizontal) indicates ﬂuorescence threshold. Fluorescence signals above this threshold marked as a successful detection of U. virens in q-LAMP assays. 2.5. Establishment of a Standard Curve for q-LAMP Detection of U. virens A standard curve between the ampliﬁcation time (x) and the Log10 value of spore number (y) was constructed based on the q-LAMP assay: y = −0.2866x + 13.829 (Figure 5, Supplementary Table S1), the formula used for calculating spore number is 100.65y, and the correlation coeﬃcient R2 = 0.9942, showing a good linear relationship. Figure 5. Standard curve of q-LAMP detection system. A standard curve between logarithmic values of the spore number (y) and the ampliﬁcation time quantitated using the cycle threshold (Ct) values (x): y = −0.2866x + 13.829. The correlation coeﬃcient (R2) is 0.9942, showing a good linear relation-ship. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 7 of 14 2.6. Application of q-LAMP Assay for U. virens Spore Calculation The standard curve of q-LAMP was applied to calculate U. virens spore number on tapes, and each tape sample contained 450, 116, 29, and 9 manually added spores, respec-tively. As shown in Table 3 and Figure 6, the ampliﬁcation times quantitated using the cycle threshold (Ct) values for the tested samples were 34.03, 37.12, 40.46, and 43.17, cor-responding to 446.07, 118.51, 28.29, and 8.85 predicted spores per tape, respectively, which is very close to the actual spore number on each Melinex tape. Thus, this q-LAMP system can eﬃciently quantitate U. virens spore number with high accuracy. Table 3. Quantitative detection of U. virens spores using q-LAMP system. Ct a Manually Added Spores (Spores/mL) Predictive Spores (Spores/mL) R2 p Value 34.03 450 446.07 0.999 0.639 37.12 116 118.51 40.46 29 28.29 43.17 9 8.85 a the ampliﬁcation times (x) quantitated using the cycle threshold (Ct) values. Figure 6. Quantitative detection of U. virens spores on Melinex tape using q-LAMP system. Serial numbers 1, 2, 3, and 4 represent 450, 116, 29, and 9 spores, respectively. The green line (horizontal) indicates ﬂuorescence threshold. Fluorescence signals above this threshold marked as a successful detection of U. virens in q-LAMP assays. 2.7. Field Application of q-LAMP Assay System The q-LAMP system results show that spores of U. virens were ﬁrst observed on the 27 August 2018, while the results obtained using the microscope show that spores were observed for the ﬁrst time on the 2nd of September. Then, the number of spores began to rise rapidly and reached its peak on the 20 September and obvious symptoms of rice false trot were found in the ﬁeld on the 25th of September. In the following year (2019), the q-LAMP system and microscope manual observation were used to monitor the spores of U. virens in the ﬁeld again. The results showed that the q-LAMP system detected the spores for the ﬁrst time on the 31st of August, while microscopic observation led to the detection of only a handful of spores on the 6th of September, and the concentration of spores reached its peak on the 30th of September. The symptoms of rice false smut were found Int. J. Mol. Sci. 2023, 24, 10388 7 of 13 2.7. Field Application of q-LAMP Assay System The q-LAMP system results show that spores of U. virens were ﬁrst observed on the 27 August 2018, while the results obtained using the microscope show that spores were observed for the ﬁrst time on the 2nd of September. Then, the number of spores began to rise rapidly and reached its peak on the 20 September and obvious symptoms of rice false trot were found in the ﬁeld on the 25th of September. In the following year (2019), the q-LAMP system and microscope manual observation were used to monitor the spores of U. virens in the ﬁeld again. The results showed that the q-LAMP system detected the spores for the ﬁrst time on the 31st of August, while microscopic observation led to the detection of only a handful of spores on the 6th of September, and the concentration of spores reached its peak on the 30th of September. The symptoms of rice false smut were found in the ﬁeld on the 5th of October. Through monitoring the dynamic changes in the spore number of U. virens in the ﬁeld for two consecutive years (Figure 7), it was clearly seen that the q-LAMP system was faster and more efﬁcient than the traditional microscopic observation method. Figure 7. Field application of U. virens spores using q-LAMP system. (A) Flow chart of ﬁeld U. virens spore sample detection, q-LAMP assay system and microscope observation were used for the collected samples, respectively. (B) The results of U. virens spore concentration measured by different methods in rice ﬁelds in 2018 (q-LAMP assay system is the gray line; microscope observation method is the blue line). (C) The results of U. virens spore concentration measured by different methods in rice ﬁelds in 2019 (q-LAMP assay system is the gray line; microscope observation method is the blue line). The green arrow is the ﬁrst detection of spores by q-LAMP assay system, the orange arrow is the ﬁrst observation of spores by microscope observation, and the red arrow is the occurrence time of rice false smut in the ﬁeld. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 8 of 14 in the ﬁeld on the 5th of October. Through monitoring the dynamic changes in the spore number of U. virens in the ﬁeld for two consecutive years (Figure 7), it was clearly seen that the q-LAMP system was faster and more eﬃcient than the traditional microscopic observation method. Figure 7. Field application of U. virens spores using q-LAMP system. (A) Flow chart of ﬁeld U. virens spore sample detection, q-LAMP assay system and microscope observation were used for the col-lected samples, respectively. (B) The results of U. virens spore concentration measured by diﬀerent methods in rice ﬁelds in 2018 (q-LAMP assay system is the gray line; microscope observation method is the blue line). (C) The results of U. virens spore concentration measured by diﬀerent meth-ods in rice ﬁelds in 2019 (q-LAMP assay system is the gray line; microscope observation method is the blue line). The green arrow is the ﬁrst detection of spores by q-LAMP assay system, the orange arrow is the ﬁrst observation of spores by microscope observation, and the red arrow is the occur-rence time of rice false smut in the ﬁeld. 3. Discussion Currently, rice false smut disease caused by U. virens is one of the most devastating rice diseases in China, as well as many other countries [35]. The occurrence of rice false smut disease not only results in the decrease in rice quality and the serious loss of rice yield, but also threatens food safety due to its production of toxic mycotoxins within the false smut balls [10,11]. However, it has been found that rice false smut disease is diﬃcult to control. As a typical airborne disease, the epidemic of rice false smut is closely related to the number of U. virens spores in the ﬁeld; thus, early detection and warning are critical for preventing and mitigating rice false smut. In this study, a q-LAMP assay system was developed. The results show that the species-speciﬁc UV-2 primer sets in the q-LAMP assay system could correctly distinguish U. virens from the other nine air-dispersed fungi, including Fusarium fujikuroi, F. oxysporum, F. proliferatum, F. solani, F. graminearum, Penicil-lium sp, Pyricularia oryzae, Alternaria alternata, and Rhizoctonia solani (Figure 3). Addition-ally, sensitivity validation found that the q-LAMP assay was able to detect a concentration Int. J. Mol. Sci. 2023, 24, 10388 8 of 13 3. Discussion Currently, rice false smut disease caused by U. virens is one of the most devastating rice diseases in China, as well as many other countries [35]. The occurrence of rice false smut disease not only results in the decrease in rice quality and the serious loss of rice yield, but also threatens food safety due to its production of toxic mycotoxins within the false smut balls [10,11]. However, it has been found that rice false smut disease is difﬁcult to control. As a typical airborne disease, the epidemic of rice false smut is closely related to the number of U. virens spores in the ﬁeld; thus, early detection and warning are critical for preventing and mitigating rice false smut. In this study, a q-LAMP assay system was developed. The results show that the species-speciﬁc UV-2 primer sets in the q-LAMP assay system could correctly distinguish U. virens from the other nine air-dispersed fungi, including Fusarium fujikuroi, F. oxysporum, F. proliferatum, F. solani, F. graminearum, Penicillium sp., Pyricularia oryzae, Alternaria alternata, and Rhizoctonia solani (Figure 3). Additionally, sensitivity validation found that the q-LAMP assay was able to detect a concentration of 6.4 U. virens spores/mL at an optimal reaction temperature of 63.4 ◦C within 60 min (Figure 4), and the q-LAMP assay could even achieve accurate quantitative detection when there were only nine U. virens spores on the Melinex tape (Figure 6). Moreover, there was a good linear relationship between the spore amount (y) and the ampliﬁcation time (x) (Figure 5), which enables accurate quantiﬁcation of U. virens and early diagnosis of U. virens infection via q-LAMP assay. The LAMP primer set consisted of two outer primers (forward primer F3 and backward primer B3), two inner primers (forward inner primer FIP and backward inner primer BIP), and two loop primers (forward loop F and backward loop B) (Supplementary Figure S2). The outer primers (F3 and B3) were used in the initial steps of the LAMP reactions but later, during the isothermal cycling, only the inner primers were used for strand-displacement DNA synthesis. Outer and inner primers are necessary for LAMP primer design, while the loop primers can be used to accelerate ampliﬁcation reactions and improve the LAMP assay efﬁciency [36]. In this study, the q-LAMP primer set was designed according to the work of Wang et al. [20] and Li et al. [37], containing a forward inner primer (FIP), a backward inner primer (BIP), and two outer (F3 and B3) primers. The ustiloxins biosynthetic gene sequence was used to design primers to ensure their speciﬁcity. Meanwhile, we sequenced the targeted region of ustiloxins biosynthetic gene in 15 U. virens stains and designed the primer sets elaborately to eliminate the interference from nucleotide polymorphisms, ensuring the ampliﬁcation efﬁciency in U. virens detection (Figure 3). For U. virens diagnosis, besides traditional disease diagnosis that includes the iden- tiﬁcation of symptoms, isolation of pathogens, and microscopic techniques, a conven- tional nested-PCR assay has been developed for the detection U. virens in rice [6]. How- ever, the nested-PCR has less sensitivity and cannot be used in accurate quantiﬁcation of U. virens [38]. Recently, the q-PCR technique and q-LAMP assay have been applied for the identiﬁcation and quantiﬁcation of pathogens in disease diagnosis. In this study, we have established these two systems for U. virens quantiﬁcation. The q-PCR assay was carried out using the F3/FB primer set and the effective ampliﬁcation reactions were detected in samples with spore concentrations of 2 × 104, 4 × 103, 8 × 102, 1.6 × 102 spores/mL, but not 32 spores/mL (Supplementary Figure S1), indicating a lower sensitivity of q-PCR for U. virens detection compared to the q-LAMP assay system. Rice false smut has no symptoms in the early stage and can only be identiﬁed in the late stage when the smut balls appear. Chemical control is the main means of rice false smut prevention and control [39]. The previous study showed that the ﬁrst 4~15 d of ear bud breakage was the main period of control, and the ﬁrst 4~7 d of control was the best [40]. If the key window in the infection of U. virens in rice is not grasped, the efﬁcacy of management will be inadequate [16,40]. Therefore, for rice false smut that relies on airborne transmission, early detection and early warning can aid in disease prevention and control. In this study, we collected spore samples of U. virens from the ﬁeld for two consecutive years using the q-LAMP assay system and microscopic observation. Compared with manual Int. J. Mol. Sci. 2023, 24, 10388 9 of 13 observation, the q-LAMP assay system could detect spores in the air more accurately and quickly, providing a theoretical basis for precise fungicide application (Figure 7). Therefore, the q-LAMP assay, with higher efﬁciency and sensitivity, is a better choice for the early diagnosis of rice false smut. In conclusion, this is the ﬁrst assay developed for the detection of U. virens using q-LAMP assays. Compared with other U. virens detection methods, the newly developed LAMP assay has superior operability, speciﬁcity, and sensitivity, and is more suitable for the quantitative detection of U. virens and early diagnosis. 4. Materials and Methods 4.1. Fungal Isolates Isolates of Ustilaginoidea virens and the nine other fungal pathogens used in this study were isolated and identiﬁed in our lab, and detailed information on each fungus is listed in Table 4. Isolates were maintained on potato dextrose agar (PDA, prepared by 200 g potato, 20 g glucose, and 20 g agar per 1 L pure water) slants at 4 ◦C. Table 4. The information of strains used in the speciﬁcity validation of the q-LAMP assay system. Species Fusarium fujikuroi F. oxysporum F. proliferatum F. solani F. graminearum Penicillium sp. Ustilaginoidea virens Pyricularia oryzae Alternaria alternata Rhizoctonia solani Isolate NO. / ACCC30927 a CICC2489 b ACCC37119 ACCC37680 ACCC31507 ACCC2711 ACCC37631 ACCC36843 ACCC36246 Host Rice Rice Rice Rice Wheat Soil Rice Rice Rice Rice Origin Zhejiang, China Hainan, China Anhui, China Hebei, China Jiangxi, China Shandong, China Hunan, China Fujian, China Hainan, China Beijing, China a ACCC (Agricultural Culture Collection of China). b CICC (China Center of Industrial Culture Collection). 4.2. DNA Template Preparation from Mycelium and Spores for q-PCR and q-LAMP Analysis Preparation of mycelial DNA template for optimum conditions and speciﬁcity of the q- LAMP assay, after mycelia grew covering two-thirds of the PDA plate surfaces, the hyphae were then transferred to a mortar and ground with liquid nitrogen. The resultant powder was then placed into a 2-mL centrifuge tube and the mycelial DNA of each fungus was extracted using a Genomic DNA Kit (Sangon Biotech Co., Ltd., Shanghai, China) according to the manufacturer’s instructions. The extracted DNA was used as DNA template in q- LAMP analyses and stored at −20 ◦C. For spore DNA template preparation, after growing on PDA medium at 25 ◦C in darkness for 20 days, 5 mm diameter mycelial plugs taken from colony margin were placed into the potato sucrose (PS, prepared by 200 g potato and 20 g sucrose per 1 L pure water) medium at 25 ◦C 150 rpm, in darkness for 7 days. Spores were separated from medium with ﬁltration through four layers of lens tissue and washed twice with distilled water. Then, spores were diluted with 10% sodium dodecylsulfate (SDS) solution into a series of concentration gradients. An amount of 1-mL spore suspension sample of known concentration mixed with 200-µL 10% Chelex-100 solution [20], 50-µL 10% SDS solution and 0.4 g acid-washed glass beads was placed into a 2.0-mL centrifuge tube. The sample was lysed by Fast Prep Apparatus (JXFSTPRP-24L, Jingxin Technology, Shanghai, China) for 40 s at speed of 6 m/s and placed in boiling water bath for 5 min. The grinding and heating steps were repeated three times, after which the sample was placed on ice for 2 min. The cooled lysate was used directly as DNA template in q-PCR and q-LAMP analyses and stored at −20 ◦C. Int. J. Mol. Sci. 2023, 24, 10388 10 of 13 4.3. Design of q-LAMP Primers for U. virens Detection Ustiloxin A and Ustiloxin B of U. virens are synthesized by ustiloxins biosynthetic gene that was found to be species-speciﬁc to U. virens [13,41]. Thus, the sequence of ustiloxins biosynthetic gene (NCBI accession number: BR001221.1) was chosen for q-LAMP primer design using Primer Explore V5 (online web service, http://primerexplorer.jp/e/) ensuring the speciﬁcity and accuracy of q-LAMP assay system for U. virens detection. The q-LAMP primers contain forward inner primer (FIP), backward inner primer (BIP), and two outer (F3 and B3) primers (Supplementary Figure S2). The primers were designed according to the following rules: ∆G values of less than or equal to −4 Kcal/mol at the 3(cid:48)end of F3/B3 and F2/B2 and 5(cid:48)ends of F1c and B1c. 4.4. Determination of Optimum Condition of the q-LAMP Assay To better facilitate the efﬁciency of q-LAMP reaction, the LAMP reaction system was improved via screening for the optimal reaction temperature based on a reference from Notomi [42]. The LAMP reaction was carried out in the following reaction mixtures containing 0.25 µM·L−1 of the primers, FIP and BIP; 0.2 µM·L−1 of the primers, F3 and B3; 1.0 mM·L−1 betaine; 2.0 mM·L−1 dNTPs (Takara Bio Inc., 108, San Jose, CA, USA); 25 mM·L−1 Tris-HCl (pH 8.8); 12.5 mM·L−1 KCl, 12.5 mM·L−1 (NH4)2SO4; 10 mM·L−1 MgCl2; 0.125% (v/v) Triton X-100; 0.2 U·L−1 of Bst DNA polymerase (New England Biolabs, 110, Beijing, China); 0.5 µL 1 × SYBR Green I; and 1 µL of DNA template extracted as described above, and the volume was adjusted to 25 µL with nucleic-acid-free water. The screened reaction temperature gradients were 61.8 ◦C, 62.1 ◦C, 62.6 ◦C, 63.4 ◦C, 64.4 ◦C, 65.2 ◦C, 65.6 ◦C, and 66 ◦C. LAMP reactions were performed using a Bio-Rad quantitative ﬂuorescent PCR instrument (Bio-Rad CFX96, Hercules, CA, USA) for 80 cycles each, each cycle for 60 s, and the reaction was terminated at 80 ◦C for 10 min. Optimal reaction temperature screening experiments were repeated three times. 4.5. Validation of the Speciﬁcity for q-LAMP Assay Systems The speciﬁcity of the reaction system was tested by performing q-LAMP reactions at the optimal reaction temperature with UV-2 primers in above 25-µL reaction mixtures for 70 min. The assay results were compared with the DNA of U. virens and the 9 other fungi listed in Table 4. The nucleic acid-free water was set as negative control. Additionally, the DNA template of U. virens and the 9 other fungi were prepared as descripted in 4.2 mycelial DNA template preparation. The extracted DNA of U. virens and the 9 other fungi were stored at −20 ◦C and their concentration were more than 150 µg·mL−1. The results were rigorously validated with the assessment that the detectable peak of ﬂuorescence signals detected by Bio-Rad CFX96 as positive; no ﬂuorescence signal as negative. The speciﬁcity testing experiment was repeated three times. 4.6. Sensitivity Validation of q-LAMP and q-PCR Assay Systems The sensitivity validation of q-LAMP reactions was performed at the optimal reaction temperature with UV-2 primers in reaction mixtures above 25-µL for 60 min. An amount of 1 µL of DNA lysate from U. virens spores of known concentration was used as a DNA template in the LAMP reaction system. The nucleic-acid-free water was used as a DNA template in the negative control (CK). The detectable peak of ﬂuorescence signals detected by Bio-Rad CFX96 was regarded as positive, while no ﬂuorescence signal was regarded as negative. Sensitivity assay experiments were repeated three times. The sensitivity of the q-PCR reaction system was assayed via performing q-PCR ampliﬁcation using primers, UV-2 F3/B3. The q-PCR reaction system was 12.5 µL SYBR® Premix Ex Taq II (Tli RNaseH Plus, 2×), 1.0 µL of forward primer F3 (10 µM), 1.0 µL of reverse primer B3 (10.0 µM), 1.0 µL DNA template (in CK, nucleic-acid-free water was used as DNA template), and the volume was adjusted to 25 µL with nucleic-acid-free water. The reaction conditions were: pre-denaturation at 95 ◦C for 2 min, denaturation at 95 ◦C for 5 s, annealing at 60 ◦C for 30 s, extension at 72 ◦C for 6 s. The ﬂuorescence signal was collected during the extension Int. J. Mol. Sci. 2023, 24, 10388 11 of 13 for a total of 40 cycles, and ﬁnally the ampliﬁcation curve was plotted. The detectable peak of ﬂuorescence signals detected by Bio-Rad CFX96 was regarded as positive, while no ﬂuorescence signal was regarded as negative. Sensitivity assay experiments were repeated three times. 4.7. Establishment of Standard Curves for q-LAMP Assay Systems A standard curve was constructed using software SPSS 13.0 by analyzing the associa- tion of logarithmic values of the spore number (y) and the ampliﬁcation time quantitated using the cycle threshold (Ct) values (x). The correlation coefﬁcient R2 was used for assess- ing the linear relationship between the spore number in sample (y) and ampliﬁcation time (x). The experiments were repeated three times. 4.8. Calculating of U. virens Spore Using q-LAMP System Spores of U. virens were artificially added to each of the four Melinex tape (1 cm × 2 cm) in the ultra-clean bench, with 450, 116, 29 and 9 spores in each tape. The collected spore- adsorbed Melinex tape was cut and placed in 2-mL centrifuge tubes, and the genomic DNA of the spores on the Melinex tape was then extracted according to the method mentioned above. An amount of 1 µL of the cooled lysate was used directly as DNA template. q-LAMP assay was performed with the optimal reaction conditions in reaction mixtures above 25-µL for 60 min, and the time quantitated using the cycle threshold (Ct) values detected by Bio-Rad CFX96 was recorded as the ampliﬁcation time (x). The linearized equation for the standard curve was used for converting the ampliﬁcation time to the corresponding spore number. Then, the calculated spore number was compared to the amount of actual added (listed above) to test the accuracy efﬁciency of this q-LAMP system. 4.9. Field Application of q-LAMP Assay by U. virens An air borne spore catcher (DIANJIANG, DJ-0723) with Melinex tape was established in Yongyou 1540 cultivation area, Jiangtang Village, Jinhua City, Zhejiang Province for the collection of spores of U. virens, and samples were collected at six-day intervals for 11 consecutive times, starting on the 9th of August 2018. Similarly, starting on the 13th of August of the following year (2019), 11 consecutive samples were collected every six days. The spores of U. virens were adsorbed on the Melinex tape and the tapes (1 cm × 1 cm) with spores were cut and placed in a 2-mL centrifuge tube, and the conidial DNA was extracted according to the methods mentioned above. An amount of 1 µL of the cooled lysate was used directly as DNA template. q-LAMP assay was performed with the optimal reaction conditions in the reaction mixtures above 25-µL for 60 min, and the ampliﬁcation time quantitated using the cycle threshold (Ct) values was recorded. According to the established standard curve, the number of spores was calculated. The spore population of U. virens in the Melinex tape was recorded via q-LAMP assay at six-day intervals. Meanwhile, the spores of U. virens (1 cm × 1 cm) adsorbed on the slide were suspended in 1 mL of ddH2O, and the spore suspension was counted using a hemocytometer under the microscope to determine the spore concentration. There were three spore catchers placed at the collection site, and the data collected by each instrument were used as a repetition. Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/ijms241210388/s1. Author Contributions: Conceptualization, T.M. and C.Z.; methodology, C.Z., Y.Z. and X.L.; software, T.M. and Y.Z.; validation, T.M. and C.Z.; formal analysis, Y.Z. and S.Z.; investigation, S.Z, C.M. and X.L.; writing—original draft preparation, Y.Z. and T.M.; writing—review and editing, T.M. and C.Z.; visualization, C.M. and X.L.; supervision, T.M. and C.Z. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Key Research and Development Project of Zhejiang Province, China (2015C02019), Science &Technology Program of Agriculture and Country in Zhenhai District. Institutional Review Board Statement: Not applicable. Int. J. Mol. Sci. 2023, 24, 10388 12 of 13 Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 2. 1. Wei, S.; Wang, Y.; Zhou, J.; Xiang, S.; Sun, W.; Peng, X.; Li, J.; Hai, Y.; Wang, Y.; Li, S. The conserved effector UvHrip1 interacts with OsHGW and infection of Ustilaginoidea virens regulates defense- and heading date-related signaling pathway. Int. J. Mol. Sci. 2020, 21, 3376. [CrossRef] Andargie, M.; Li, J. Arabidopsis thaliana: A model host plant to study plant–pathogen interaction using rice false smut isolates of Ustilaginoidea virens. Front. Plant Sci. 2016, 7, 192. [CrossRef] Song, J.-H.; Wei, W.; Lv, B.; Lin, Y.; Yin, W.-X.; Peng, Y.-L.; Schnabel, G.; Huang, J.-B.; Jiang, D.-H.; Luo, C.-X. Rice False smut fungus hijacks the rice nutrients supply by blocking and mimicking the fertilization of rice ovary. Environ. Microbiol. 2016, 18, 3840–3849. [CrossRef] 3. 4. Meng, S.; Xiong, M.; Jagernath, J.S.; Wang, C.; Qiu, J.; Shi, H.; Kou, Y. UvAtg8-mediated autophagy regulates fungal growth, 5. 6. 7. 8. 9. stress responses, conidiation, and pathogenesis in Ustilaginoidea virens. Rice 2020, 13, 56. [CrossRef] [PubMed] Devi, T.K.; Singh, N.I. Aerobiology and epidemiology of false smut disease of rice by Ustilagnoidea virens (Syn. Claviceps oryzae sativae) in Thoubal District. J. Mycopatholog. Res. 2007, 45, 107–108. Zhou, Y.-L.; Izumitsu, K.; Sonoda, R.; Nakazaki, T.; Tanaka, E.; Tsuda, M.; Tanaka, C. PCR-based speciﬁc detection of Ustilaginoidea virens and Ephelis japonica. J. Phytopathol. 2003, 151, 513–518. [CrossRef] Ashizawa, T.; Takahashi, M.; Moriwaki, J.; Hirayae, K. Quantiﬁcation of the rice false smut pathogen Ustilaginoidea virens from soil in Japan using real-time PCR. Eur. J. Plant Pathol. 2010, 128, 221–232. [CrossRef] Tanaka, E.; Ashizawa, T.; Sonoda, R.; Tanaka, C. Villosiclava virens gen. nov., comb. nov., teleomorph of Ustilaginoidea virens, the causal agent of rice false smut. Mycotaxon 2008, 106, 491–501. [CrossRef] Zhang, Y.; Zhang, K.; Fang, A.; Han, Y.; Yang, J.; Xue, M.; Bao, J.; Hu, D.; Zhou, B.; Sun, X.; et al. Speciﬁc adaptation of Ustilaginoidea virens in occupying host ﬂorets revealed by comparative and functional genomics. Nat. Commun. 2014, 5, 3849. [CrossRef] [PubMed] 10. Zhou, Y.-L.; Xie, X.-W.; Zhang, F.; Wang, S.; Liu, X.-Z.; Zhu, L.-H.; Xu, J.-L.; Gao, Y.-M.; Li, Z.-K. Detection of quantitative resistance loci associated with resistance to rice false smut (Ustilaginoidea virens) using introgression lines. Plant Pathol. 2013, 63, 365–372. [CrossRef] Sun, X.; Kang, S.; Zhang, Y.; Tan, X.; Yu, Y.; He, H.; Zhang, X.; Liu, Y.; Wang, S.; Sun, W.; et al. Genetic diversity and population structure of rice pathogen Ustilaginoidea virens in China. PLoS ONE 2013, 8, e76879. [CrossRef] 11. 12. Zhou, L.; Lu, S.; Shan, T.; Wang, P.; Wang, S. Chemistry and biology of mycotoxins from rice false smut pathogen. In Mycotoxins: Properties, Applications, and Hazards; Melbor, B., Greene, J., Eds.; Nova Science Publishers, Inc.: New York, NY, USA, 2012. Fu, X.; Wang, A.; Wang, X.; Lin, F.; He, L.; Lai, D.; Liu, Y.; Li, Q.X.; Zhou, L.; Wang, B. Development of a monoclonal antibody-based icELISA for the detection of Ustiloxin B in rice false smut balls and rice grains. Toxins 2015, 7, 3481–3496. [CrossRef] 13. 14. Yu, M.; Yu, J.; Cao, H.; Song, T.; Pan, X.; Qi, Z.; Du, Y.; Zhang, R.; Huang, S.; Liu, W.; et al. SUN-family protein UvSUN1 regulates the development and virulence of Ustilaginoidea virens. Front. Microbiol. 2021, 12, 739453. [CrossRef] 15. Yu, J.; Yu, M.; Song, T.; Cao, H.; Pan, X.; Yong, M.; Qi, Z.; Du, Y.; Zhang, R.; Yin, X.; et al. A homeobox transcription factor UvHOX2 regulates chlamydospore formation, conidiogenesis, and pathogenicity in Ustilaginoidea virens. Front. Microbiol. 2019, 10, 1071. [CrossRef] [PubMed] 16. Tang, Y.-X.; Jin, J.; Hu, D.-W.; Yong, M.-L.; Xu, Y.; He, L.-P. Elucidation of the infection process of Ustilaginoidea virens (teleomorph: Villosiclava virens) in rice spikelets. Plant Pathol. 2012, 62, 1–8. [CrossRef] 17. Hu, Y. Infection processes of Ustilaginoidea virens during artiﬁcial inoculation of rice panicles. Eur. J. Plant Pathol. 2014, 139, 67–77. [CrossRef] 18. Yong, M.; Deng, Q.; Fan, L.; Miao, J.; Lai, C.; Chen, H.; Yang, X.; Wang, S.; Chen, F.; Jin, L.; et al. The role of Ustilaginoidea virens sclerotia in increasing incidence of rice false smut disease in the subtropical zone in China. Eur. J. Plant Pathol. 2017, 150, 669–677. [CrossRef] 19. Tsukui, T.; Nagano, N.; Umemura, M.; Kumagai, T.; Terai, G.; Machida, M.; Asai, K. Ustiloxins, fungal cyclic peptides, are ribosomally synthesized in Ustilaginoidea virens. Bioinformatics 2014, 31, 981–985. [CrossRef] [PubMed] 20. Wang, Q.W.; Zhang, C.-Q. Q-LAMP assays for the detection of Botryosphaeria dothidea causing Chinese hickory canker in trunk, water, and air samples. Plant Dis. 2019, 103, 3142–3149. [CrossRef] [PubMed] 21. Harrison, N.A.; Womack, M.; Carpio, M.L. Detection and characterization of a lethal yellowing (16SrIV) group phytoplasma in Canary Island date palms affected by lethal decline in Texas. Plant Dis. 2002, 86, 676–681. [CrossRef] 22. Notomi, T.; Okayama, H.; Masubuchi, H.; Yonekawa, T.; Watanabe, K.; Amino, N.; Hase, T. Loop-mediated isothermal ampliﬁca- tion of DNA. Nucleic Acids Res. 2000, 28, E63. [CrossRef] [PubMed] 23. Aryan, E.; Makvandi, M.; Farajzadeh, A.; Huygen, K.; Bifani, P.; Mousavi, S.-L.; Fateh, A.; Jelodar, A.; Gouya, M.-M.; Romano, M. A novel and more sensitive loop-mediated isothermal ampliﬁcation assay targeting IS6110 for detection of Mycobacterium tuberculosis complex. Microbiol. Res. 2010, 165, 211–220. [CrossRef] [PubMed] Int. J. Mol. Sci. 2023, 24, 10388 13 of 13 24. McKenna, J.P.; Fairley, D.J.; Shields, M.D.; Cosby, S.L.; Wyatt, D.E.; McCaughey, C.; Coyle, P.V. Development and clinical validation of a loop-mediated isothermal ampliﬁcation method for the rapid detection of Neisseria meningitidis. Diagn. Microbiol. Infect. Dis. 2011, 69, 137–144. [CrossRef] [PubMed] 25. Xie, L.; Xie, Z.; Zhao, G.; Liu, J.; Pang, Y.; Deng, X.; Xie, Z.; Fan, Q.; Luo, S. A loop-mediated isothermal ampliﬁcation assay for the 26. visual detection of duck circovirus. Virol. J. 2014, 11, 76. [CrossRef] Soleimani, M.; Shams, S.; Majidzadeh, A.K. Developing a real-time quantitative loop-mediated isothermal ampliﬁcation assay as a rapid and accurate method for detection of Brucellosis. J. Appl. Microbiol. 2013, 115, 828–834. [CrossRef] 27. Tomita, N.; Mori, Y.; Kanda, H.; Notomi, T. Loop-mediated isothermal ampliﬁcation (LAMP) of gene sequences and simple visual detection of products. Nat. Protoc. 2008, 3, 877–882. [CrossRef] 28. Yang, X.; Al-Attala, M.N.; Zhang, Y.; Zhang, A.-F.; Zang, H.-Y.; Gu, C.-Y.; Gao, T.-C.; Chen, Y.; Ali, F.; Li, Y.-F.; et al. Rapid detection of Ustilaginoidea virens from rice using Loop-Mediated Isothermal Ampliﬁcation Assay. Plant Dis. 2018, 102, 1741–1747. [CrossRef] Stella, S.; Gottardi, E.M.; Favout, V.; Gonzalez, E.B.; Errichiello, S.; Vitale, S.R.; Fava, C.; Luciano, L.; Stagno, F.; Grimaldi, F.; et al. The q-LAMP method represents a valid and rapid alternative for the detection of the BCR-ABL1 rearrangement in Philadelphia- positive leukemias. Int. J. Mol. Sci. 2019, 20, 6106. [CrossRef] 29. 30. Wang, Y.; Li, K.; Xu, G.; Chen, C.; Song, G.; Dong, Z.; Lin, L.; Wang, Y.; Xu, Z.; Yu, M.; et al. Low-cost and Scalable platform with multiplexed microwell array biochip for rapid diagnosis of COVID-19. Research 2021, 2021, 2813643. [CrossRef] 31. Ku, J.; Chauhan, K.; Hwang, S.-H.; Jeong, Y.-J.; Kim, D.-E. Enhanced speciﬁcity in loop-mediated isothermal ampliﬁcation with poly(ethylene glycol)-engrafted graphene oxide for detection of viral genes. Biosensors 2022, 12, 661. [CrossRef] 32. Huang, Y.; Tang, X.; Zheng, L.; Huang, J.; Zhang, Q.; Liu, H. Development of generic immuno-magnetic bead-based enzyme-linked immunoassay for Ustiloxins in rice coupled with enrichment. Toxins 2021, 13, 907. [CrossRef] 33. Wang, X.; Fu, X.; Lin, F.; Sun, W.; Meng, J.; Wang, A.; Lai, D.; Zhou, L.; Liu, Y. The contents of Ustiloxins A and B along with their 34. distribution in rice false smut balls. Toxins 2016, 8, 262. [CrossRef] Fu, R.; Chen, C.; Wang, J.; Liu, Y.; Zhao, L.; Lu, D. Transcription proﬁling of rice panicle in response to crude toxin extract of Ustilaginoidea virens. Front. Microbiol. 2022, 13, 701489. [CrossRef] 35. Abbas, H.K.; Shier, W.T.; Cartwright, R.D.; Sciumbato, G.L. Ustilaginoidea virens infection of rice in Arkansas: Toxicity of false smut galls, their extracts and the Ustiloxin fraction. Am. J. Plant Sci. 2014, 05, 3166–3176. [CrossRef] 36. Rovira, A.; Abrahante, J.; Murtaugh, M.; Claudia, M.-Z. Reverse transcription loop-mediated isothermal ampliﬁcation for the detection of porcine reproductive and respiratory syndrome virus. J. Veter-Diagn. Investig. 2009, 21, 350–354. [CrossRef] 37. Li, L.; Zhang, S.Y.; Zhang, C.-Q. Establishment of a rapid detection method for rice blast fungus based on one-step loop-mediated isothermal ampliﬁcation (LAMP). Plant Dis. 2019, 103, 1967–1973. [CrossRef] 38. Li, H.; Ni, D.; Duan, Y.; Chen, Y.; Li, J.; Song, F.; Li, L.; Wei, P.; Yang, J. Quantitative detection of the rice false smut pathogen Ustilaginoidea virens by real-time PCR. Genet. Mol. Res. 2013, 12, 6433–6441. [CrossRef] [PubMed] 39. Wang, Z.; Yang, X.; Lyu, L.; Yuan, B.; Chang, X.; Zhang, S. Progress and prospective of Villosiclava virens infection mechanism. Hubei Agric. Sci. 2019, 58, 5. [CrossRef] 40. Zhang, S.; Zhang, Q.; Luo, H. Test pesticides against rice false smut and choose optimum application period. J. Huazhong Agric. 41. Univ. 2007, 26, 178. Fu, X.; Xie, R.; Wang, J.; Chen, X.; Wang, X.; Sun, W.; Meng, J.; Lai, D.; Zhou, L.; Wang, B. Development of colloidal gold-based lateral ﬂow immunoassay for rapid qualitative and semi-quantitative analysis of Ustiloxins A and B in rice samples. Toxins 2017, 9, 79. [CrossRef] 42. Tomlinson, J.; Boonham, N. Real-time LAMP for Chalara fraxinea diagnosis. Methods Mol. Biol. 2015, 1302, 75–83. [CrossRef] [PubMed] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_ijms241210237	Article Role of the IL-33/ST2 Activation Pathway in the Development of the Hepatic Fibrosis Induced by Schistosoma mansoni Granulomas in Mice Laura Maggi 1, Genil Mororó Araújo Camelo 1 João Marcelo Peixoto Moreira 1 Ary Correa, Jr. 6, Roselene Ecco 7 , Izabella Chrystina Rocha 1,2 , William Pereira Alves 3, , Thiago Almeida Pereira 4 , Wagner Luiz Tafuri 5, Élida Mara Leite Rabelo 3, and Deborah Aparecida Negrão-Corrêa 1,* 1 Laboratório de Esquistossomose e Imunohelmintologia, Departamento de Parasitologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, MG, Brazil; [email protected] (L.M.); [email protected] (G.M.A.C.); [email protected] (I.C.R.); [email protected] (J.M.P.M.) 2 Curso de Enfermagem, Instituto de Ciências Biológicas e Saúde, Universidade Federal de Mato Grosso, 3 5 4 Barra do Garça 78698-000, MG, Brazil Laboratório de Parasitologia Molecular, Departamento de Parasitologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, MG, Brazil; [email protected] (W.P.A.); [email protected] (É.M.L.R.) Institute for Stem Cell Biology and Regenerative Medicine, Stanford University School of Medicine, Stanford, CA 94305, USA; [email protected] Laboratório de Patologia das Leishmanioses, Departamento de Patologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, MG, Brazil; [email protected] Laboratório de Micologia, Departamento de Microbiologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, MG, Brazil; [email protected] Setor de Patologia, Escola Veterinária, Universidade Federal de Minas Gerais, Belo Horizonte 31270-901, MG, Brazil; [email protected] * Correspondence: [email protected]; Tel.: +55-31-3409-2855 6 7 Citation: Maggi, L.; Camelo, G.M.A.; Rocha, I.C.; Pereira Alves, W.; Moreira, J.M.P.; Almeida Pereira, T.; Tafuri, W.L.; Rabelo, É.M.L.; Correa, A., Jr.; Ecco, R.; et al. Role of the IL-33/ST2 Activation Pathway in the Development of the Hepatic Fibrosis Induced by Schistosoma mansoni Granulomas in Mice. Int. J. Mol. Sci. 2023, 24, 10237. https://doi.org/ 10.3390/ijms241210237 Academic Editor: J. B. Helms Received: 26 May 2023 Revised: 6 June 2023 Accepted: 13 June 2023 Published: 16 June 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Abstract: Schistosoma mansoni eggs retained in host tissues induce innate cytokine release, contributing to the induction of Type-2 immune responses and granuloma formation, important to restrain cytotoxic antigens, but leading to ﬁbrosis. Interleukin(IL)-33 participates in experimental models of inﬂammation and chemically induced ﬁbrosis, but its role in S. mansoni-induced ﬁbrosis is still unknown. To explore the role of the IL-33/suppressor of the tumorigenicity 2 (ST2) pathway, serum and liver cytokine levels, liver histopathology, and collagen deposition were comparatively evaluated in S. mansoni-infected wild-type (WT) and IL-33-receptor knockout (ST2−/−) BALB/c mice. Our data show similar egg counts and hydroxyproline in the livers of infected WT and ST2−/− mice; however, the extracellular matrix in ST2−/− granulomas was loose and disorganised. Pro-ﬁbrotic cytokines, such as IL-13 and IL-17, and the tissue-repairing IL-22 were signiﬁcantly lower in ST2−/− mice, especially in chronic schistosomiasis. ST2−/− mice also showed decreased α-smooth muscle actin (α-SMA) expression in granuloma cells, in addition to reduced Col III and Col VI mRNA levels and reticular ﬁbres. Therefore, IL-33/ST2 signalling is essential for tissue repairing and myoﬁbroblast activation during S. mansoni infection. Its disruption results in inappropriate granuloma organisation, partly due to the reduced type III and VI collagen and reticular ﬁbre formation. Keywords: Schistosoma mansoni infection; liver ﬁbrosis; granuloma; collagen deposition; reticular ﬁbres; IL-33/ST2 activation pathway 1. Introduction Fibrotic diseases, such as liver cirrhosis, pulmonary ﬁbrosis, chronic kidney disease, and cardiovascular disease, are responsible for approximately 45% of deaths in indus- trialised nations, showing that ﬁbrosis is a major global health problem and should be Int. J. Mol. Sci. 2023, 24, 10237. https://doi.org/10.3390/ijms241210237 https://www.mdpi.com/journal/ijms International Journal of Molecular SciencesInt. J. Mol. Sci. 2023, 24, 10237 2 of 20 better understood [1,2]. Fibrosis has been associated with excessive tissue deposition of the extracellular matrix (ECM) due to the proliferation and activation of ﬁbroblasts and myoﬁbroblasts that result in exacerbated growth of the injured tissue, forming hard scars that destroy the normal architecture of the affected organ [3–5]. A wide variety of stimuli from different etiological origins, such as autoimmune re- actions, allergic responses, tissue damage, and persistent infections, including parasitic diseases, can trigger ﬁbrotic processes [3–5]. One of the main causes of disease in schisto- somiasis, caused by infections with Schistosoma spp., is ﬁbrosis, and the liver is the main affected organ in this case [6–9]. It is estimated that more than 250 million individuals are infected by Schistosoma spp., and approximately 780 million people live in areas at risk of infection [10]. Among all the known species causing human schistosomiasis, Schistosoma mansoni is the most prevalent in the human population and is the only species transmitted in the Americas [11,12]. The morbidity of schistosomiasis is related to parasite load, time elapsed since infection, rate of reinfection, nutritional status, and comorbidities [13–15]. These different factors will determine the type and intensity of the host’s immune response elicited during infection, and a chronic, exacerbated type-2 response has been associated with the degree of ﬁbrosis in this disease [13,15]. The initial immune response to S. mansoni infection is predominantly a type-1 response, induced by larval and adult worm antigens [16,17]. When oviposition begins, egg antigens are responsible for a shift towards a type-2 immune response, thereby triggering a heteroge- neous extracellular matrix deposition around the parasite egg, which conﬁnes the cytotoxic effects of secreted antigens and restrains tissue damage. However, the type-2 response also leads to tissue ﬁbrosis and portal hypertension, which is associated with the severity of chronic schistosomiasis [18–23]. In parallel, an immunoregulatory response is induced and becomes predominant around the 11th week onwards in most infected individuals [24–26]. The induction of an immunomodulatory response reduces T-helper (Th)2 inﬂammation, affecting granuloma size and formation, and controls morbidity [25,27,28]. As many parasite eggs are swept through the hepatic portal system and become trapped in the liver parenchyma, the induction and modulation of the immune response in this organ are essential for the outcome of schistosomiasis mansoni [29,30]. In this disease, resident liver cells play a key role in granuloma formation, tissue remodelling, and ﬁbrosis. Activation of hepatic stellate cells (HSCs) and myoﬁbroblast differentiation are necessary for collagen deposition and ﬁbrosis in liver tissues, participating in granuloma formation during schistosomiasis [7,31]. HSC activation can be inﬂuenced by a multitude of signals from other resident and inﬁltrating cells, as well as autocrine and paracrine signalling from HSCs [7]. Proﬁbrotic cytokines, including transforming growth factor (TGF)-β1, connective tissue growth factor (CTGF), IL-13, and IL-33, which are stimulated by S. mansoni egg anti- gens during granuloma formation, can also activate HSCs, triggering their differentiation into myoﬁbroblasts, thereby contributing to Schistosoma-induced ﬁbrosis [32]. Therefore, ﬁbrosis is preceded by inﬂammation, and elements of the innate and adaptive immune responses are fundamental for the induction and regulation of ﬁbrotic processes [2]. IL-33 is an alarmin released by resident liver cells, such as hepatocytes, endothelial cells, and HSCs, when Schistosoma eggs injure the host’s tissues [33,34]. The biological functions of IL-33 depend on the activation of its only known receptor, the suppressor of tumorigenicity 2 receptor (ST2), which is expressed in several liver-resident and -inﬁltrating cells [35–37]. IL33/ST2 binding on type-2 innate lymphoid cells (ILC2s) and eosinophils activates IL-13 production [36,38,39]. This cytokine, in turn, binds to the IL-13Rα2 receptor in HSCs and activates extracellular signal regulated kinases 1 and 2 (ERK1-ERK2), subsequently triggering activin receptor-like kinases (ALKs) to activate Smad 1 and Smad 2, resulting in CTGF expression in HSCs and inducing the expression of type I and III collagen, and matrix metalloproteinases [7,38,40]. Moreover, recombinant IL-33 was capable of activating HSCs in vitro, leading to the expression of IL-6, TGF-β, α-SMA, and collagen [32]. The IL33/ST2 activation pathway also stimulates the early production of Int. J. Mol. Sci. 2023, 24, 10237 3 of 20 the Th2 cytokines IL-13 and IL-5, which increase the expression of cell adhesion molecules (CAMs) in endothelial cells and leukocytes, thereby promoting inﬁltration and activation of CD4+ T cells, macrophages, and eosinophils, besides activating tissue-resident cells, such as ﬁbroblasts [10,23,41–44]. The production of IL-33 has been associated with cellular recruitment, activation, and ECM deposition in several experimental models of ﬁbrosis. Bleomycin-induced lung ﬁbrosis has demonstrated that IL-33 promotes macrophage differentiation and activation towards an M2 phenotype, collaborating with ﬁbrosis [45,46]. Liver ﬁbrosis induced by thioacetamide and CCl4 administration or Helicobacter hepaticus infection also showed the essential role of the IL-33/ST2 signalling pathway in the pathological alterations [47]. However, the role of IL-33/ST2 signalling in ﬁbrosis caused by schistosomiasis remains poorly understood. In experimental infections with S. japonicum, exogenous administration of IL-33 leads to larger liver granulomas and a stronger Th2 response, whereas inhibition of the IL- 33/ST2 pathway hinders ﬁbrosis, collagen deposition, and differentiation of HSCs into myoﬁbroblasts [36,43,48,49]. In S. mansoni infections, Vanella et al. [34] reported that the disruption of IL-33/ST2 signalling does not alter granuloma size or ﬁbrosis and only hampers the expression of Th2 cytokines when blocked, along with IL-25 and thymic stromal lymphopoietin (TSLP) signalling. On the other hand, results from a study by Maggi et al. [50] indicate that ST2 knockout mice have impaired modulation of granulomas in the chronic phase of the schistosomiasis mansoni, leading to larger granulomas with an increased cellular inﬁltrate, resulting in higher mortality. Therefore, the effects of the IL-33/ST2 pathway on hepatic ﬁbrosis and HSC activation and differentiation during S. mansoni infection require further elucidation. In the current study, we provide data demonstrating that IL-33/ST2 signalling is an essential step for HSC activation during experimental schistosomiasis, and disruptions in this pathway result in inappropriate granuloma organisation with altered extracellular matrix composition. 2. Results 2.1. Systemic Immune Response To assess the systemic immune response during S. mansoni infection, we quantiﬁed serum levels of IL-12p70, IL-13, IL-33, TGF-β, CCL24, and IL-22 (Figure 1). The majority of serum samples tested from both mouse strains had no detectable levels of IL-12p70 and IL-13. We also observed that uninfected ST2−/− mice have higher serum levels of IL-33, CCL24, and TGF-β when compared with WT mice. When infected, IL-33 levels remain similar between mouse strains throughout the course of infection (Figure 1A). Moreover, as expected, the serum concentrations of the regulatory cytokine TGF-β increased progressively, peaking in the 12th week post infection in both WT and ST2−/− mice (Figure 1B). In contrast, the serum level of CCL24, a chemokine that attracts eosinophils, and IL-22, a cytokine associated with tissue repair mechanisms, showed signiﬁcantly different levels in WT and ST2−/− infected mice. Serum levels of CCL24 were signiﬁcantly higher in ST2−/− mice compared with WT mice between the 8th and 10th weeks of S. mansoni infection; however, in the chronic phase of schistosomiasis, the serum levels of CCL24 were similar in both mouse strains (Figure 1C). IL-22 levels remained similar between the 2 strains until the 10th week of infection, while in the 12th week, the levels of this cytokine were signiﬁcantly reduced in infected ST2−/− mice (Figure 1D). 2.2. Liver Immune Response Consistent with our previous data [50], the number of S. mansoni eggs retained in the liver of both WT and ST2−/− mouse strains was statistically similar throughout the infection (Figure 2A). Unlike that observed in serum, IL-33 levels in the liver are signiﬁcantly lower in ST2−/− mice compared with WT mice in both stages of infection (Figure 2B). Furthermore, we observed that egg retention induces the activation of a Th2 response even in the absence of the ST2 receptor, as seen by the concentrations of IL-4 and IL-13 in the liver; however, this Int. J. Mol. Sci. 2023, 24, 10237 4 of 20 response is not maintained in the chronic phase of infection, when ST2−/− mice showed signiﬁcantly lower levels of these cytokines than the infected WT mice (Figure 2C,D). Figure 1. Serum cytokines levels. Evaluation of IL-33 (A), TGF-β (B), CCL24 (C), and IL-22 (D) levels estimated by enzyme-linked immunosorbent assay (ELISA) in serum samples from WT (blue squares) and ST2−/− (black triangles) mice. Mice were subcutaneously infected with 50 S. mansoni cercariae and the serum was collected at weeks 4, 8, 10, and 12 post infection. Data are shown as kinetics and are presented as mean ± standard error of the mean (SEM). Normality was determined by the Kolmogorov–Smirnov test. Comparisons between normally distributed groups were carried out by Student’s t-test, and p-values were assigned. Differences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. The concentrations of IL-17 in the liver homogenate were signiﬁcantly higher in knockout mice compared with WT mice in the acute phase of infection, but the levels of this cytokine decreased signiﬁcantly in ST2−/− mice compared with WT mice in the chronic phase (Figure 2E). Just as in serum, the levels of IL-22 in ST2−/− mice are signiﬁcantly lower compared with WT mice during chronic schistosomiasis (Figure 2F). 2.3. Histopathology In order to analyse how the IL-33/ST2 pathway affects the formation of hepatic gran- ulomas, we evaluated haematoxylin-eosin (HE) liver slides (Figure 3). Histopathological analysis revealed the presence of adult parasites in some liver sections within the portal vessels of both infected mouse strains. Transverse sections of the parasite revealed a thin tegument and digestive tube ﬁlled with a brown granular pigment (Figure 3A). Many vessels were hyperaemic and showed marked perivascular inﬂammation with moderate inﬁltration of plasma cells, lymphocytes, neutrophils, and eosinophils, associated with ductal proliferation (Figure 3A). In acute schistosomiasis, livers from both infected groups (Figure 3B) were characterised by multiple regions showing loss of parenchyma, which had approximately 40% of its area replaced by inﬂammatory cells associated with schistosome eggs and disorganised ﬁbroblasts. Most of the inﬂammatory areas contained, centrally, a schistosome egg surrounded by numerous neutrophils, eosinophils, lymphocytes, and macrophages (Figure 3C,D). Fibroblasts intermixed with these inﬂammatory cells with minimal to moderate quantities of collagen ﬁbres formed additional layers in the outer Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 4 of 21 Figure 1. Serum cytokines levels. Evaluation of IL-33 (A), TGF-β (B), CCL24 (C), and IL-22 (D) levels estimated by enzyme-linked immunosorbent assay (ELISA) in serum samples from WT (blue squares) and ST2−/− (black triangles) mice. Mice were subcutaneously infected with 50 S. mansoni cercariae and the serum was collected at weeks 4, 8, 10, and 12 post infection. Data are shown as kinetics and are presented as mean ± standard error of the mean (SEM). Normality was determined by the Kolmogorov–Smirnov test. Comparisons between normally distributed groups were carried out by Student’s t‐test, and p‐values were assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. 2.2. Liver Immune Response Consistent with our previous data [50], the number of S. mansoni eggs retained in the liver of both WT and ST2−/− mouse strains was statistically similar throughout the infection (Figure 2A). Unlike that observed in serum, IL-33 levels in the liver are significantly lower in ST2−/− mice compared with WT mice in both stages of infection (Figure 2B). Furthermore, we observed that egg retention induces the activation of a Th2 response even in the absence of the ST2 receptor, as seen by the concentrations of IL-4 and IL-13 in the liver; however, this response is not maintained in the chronic phase of infection, when ST2−/− mice showed sig-nificantly lower levels of these cytokines than the infected WT mice (Figure 2C,D). The concentrations of IL-17 in the liver homogenate were signiﬁcantly higher in knockout mice compared with WT mice in the acute phase of infection, but the levels of this cytokine decreased signiﬁcantly in ST2−/− mice compared with WT mice in the chronic phase (Figure 2E). Just as in serum, the levels of IL-22 in ST2−/− mice are signiﬁcantly lower compared with WT mice during chronic schistosomiasis (Figure 2F). Int. J. Mol. Sci. 2023, 24, 10237 5 of 20 zone of these areas and were generally more intense in liver granulomas from WT mice (Figure 3C) than in ST2−/− (Figure 3D). Inﬁltration of multinucleated giant cells was less frequent, and there were some pre-granulomatous areas. The remaining parenchyma among the well-circumscribed lesion areas had a preserved tissue architecture; however, scattered neutrophils and eosinophils were also seen between the cords of hepatocytes or some foci of lesion. These focal lesions were characterised by necrotic foci associated with ﬁbrin, haemorrhage, moderate eosinophil inﬁltration, and macrophages containing cytoplasmic hemosiderin. Figure 2. Parasite load and immunological parameters in the liver. Quantiﬁcation of eggs retained in the liver of WT (blue squares) and ST2−/− (black triangles) mice at 8 (acute) or 12 (chronic) weeks after infection with 50 cercariae (A). Quantiﬁcation of Th2, Th17, and regulatory cytokines in liver homogenates from WT and ST2−/− BALB/c mice during acute and chronic infections. Mice were subcutaneously infected with 50 S. mansoni cercariae and euthanised at weeks 8 (acute phase) or 12 post infection (chronic phase). Data compile results from two independent experiments and are presented as dot plots and mean ± SEM, each point representing an animal, of the IL-33 (B), IL-4 (C), IL-13 (D), IL-17 (E), and IL-22 (F) concentrations estimated by sandwich ELISA. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between normally distributed groups were carried out by two-way ANOVA followed by Bonferroni’s post-test (A–C), while the Kruskal– Wallis test (D–F) was performed for non-parametric distributions, and the p-value were assigned. Differences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. * represents signiﬁcant differences between chronically infected mice in comparison with mice of the same strain infected in the acute phase (* p < 0.05; p < 0.01; ** p < 0.001). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 5 of 21 Figure 2. Parasite load and immunological parameters in the liver. Quantiﬁcation of eggs retained in the liver of WT (blue squares) and ST2−/− (black triangles) mice at 8 (acute) or 12 (chronic) weeks after infection with 50 cercariae (A). Quantiﬁcation of Th2, Th17, and regulatory cytokines in liver homogenates from WT and ST2−/− BALB/c mice during acute and chronic infections. Mice were sub-cutaneously infected with 50 S. mansoni cercariae and euthanised at weeks 8 (acute phase) or 12 post infection (chronic phase). Data compile results from two independent experiments and are pre-sented as dot plots and mean ± SEM, each point representing an animal, of the IL-33 (B), IL-4 (C), IL-13 (D), IL-17 (E), and IL-22 (F) concentrations estimated by sandwich ELISA. Normality was de-termined by the Kolmogorov–Smirnov test. Comparisons between normally distributed groups were carried out by two-way ANOVA followed by Bonferroni’s post-test (A–C), while the Kruskal–Wallis test (D–F) was performed for non-parametric distributions, and the p‐value were assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. * represents signiﬁcant diﬀerences between chronically infected mice in comparison with mice of the same strain infected in the acute phase (* p < 0.05; p < 0.01; p < 0.001). 2.3. Histopathology In order to analyse how the IL-33/ST2 pathway affects the formation of hepatic granu-lomas, we evaluated haematoxylin-eosin (HE) liver slides (Figure 3). Histopathological anal-ysis revealed the presence of adult parasites in some liver sections within the portal vessels of both infected mouse strains. Transverse sections of the parasite revealed a thin tegument Int. J. Mol. Sci. 2023, 24, 10237 6 of 20 Figure 3. Histopathology of hepatic granulomas of WT and ST2−/− mice infected with S. mansoni. Photomicrograph of a HE-stained liver section with sections of adult worms within the portal vessel of an infected ST2−/− mouse (A). Photomicrograph of a HE-stained liver section slide of granulomas during acute phase from WT (C) and ST2−/− (B,D) mice and chronic schistosomiasis from WT (E) and ST2−/− (F). Scale bars = 100 µm (A,C–F) or 200 µm (B). In chronic schistosomiasis, the liver sections of infected mice from the WT and ST2−/− BALB/c groups were characterised by multifocal to coalescing areas with loss of parenchyma and replacement by well-formed granulomas associated or not with schisto- some eggs. These areas replaced approximately 50–60% of the parenchyma. Inﬂammatory cell inﬁltrate was comprised of macrophages, multinucleated giant cells, lymphocytes, neu- trophils, and eosinophils in different proportions in the different stages of the granulomas (Figure 3E,F). Some recent necrotic and acute inﬂammatory areas with viable eggs were scattered throughout the parenchyma. The periportal areas had numerous plasma cells, lymphocytes, and macrophages, and a mild increase in ﬁbroplasia. Most of the liver granu- lomas from infected WT mice showed lower cellular inﬁltration and had organised layers of ﬁbroblasts and eosinophilic ﬁbres, concentrically arranged and surrounding the central area (Figure 3E), while liver granulomas from infected ST2−/− mice still presented intense cellular inﬁltration and more disorganised ﬁbre deposition (Figure 3F). In some granulo- mas, the eggs were disrupted and partially or completely mineralised, with less frequent Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 7 of 21 Figure 3. Histopathology of hepatic granulomas of WT and ST2−/− mice infected with S. mansoni. Photomicrograph of a HE-stained liver section with sections of adult worms within the portal vessel of an infected ST2−/− mouse (A). Photomicrograph of a HE-stained liver section slide of granulomas during acute phase from WT (C) and ST2−/− (B,D) mice and chronic schistosomiasis from WT (E) and ST2−/− (F). Scale bars = 100 µm (A,C–F) or 200 µm (B). 2.4. Hepatic Stellate Cell Diﬀerentiation As the differentiation of HSCs into myofibroblasts is characterised by the expression of α-SMA [7,51,52], the expression of this myofibroblast marker was evaluated in the livers of the experimental groups. Immunohistochemical (IHC) analysis revealed that many cells that composed Schistosoma-induced granulomas in the livers of infected WT mice expressed α-SMA, especially in the chronic phase (Figure 4A). In contrast, liver granulomas from in-fected ST2−/− mice showed very few α-SMA-expressing cells (Figure 4B). Morphometric anal-ysis of the brown-stained areas in hepatic granulomas of chronically infected mice from both mouse strains confirmed a significantly lower expression of α-SMA+ cells in granulomas formed in the absence of the activation of the IL-33/ST2 pathway (Figure 4C). Int. J. Mol. Sci. 2023, 24, 10237 7 of 20 eosinophils. Pigmented (hemosiderin-containing) macrophages were comparatively higher in these granulomas, although some were observed between the cords of hepatocytes. In summary, the histopathological evaluation revealed differences in inﬂammation and ﬁbre deposition in liver granulomas between infected WT and ST2−/− BALB/c mice. Thus, the livers from these mouse strains were further evaluated to reveal the pattern of collagen deposition. 2.4. Hepatic Stellate Cell Differentiation As the differentiation of HSCs into myoﬁbroblasts is characterised by the expression of α-SMA [7,51,52], the expression of this myoﬁbroblast marker was evaluated in the livers of the experimental groups. Immunohistochemical (IHC) analysis revealed that many cells that composed Schistosoma-induced granulomas in the livers of infected WT mice expressed α-SMA, especially in the chronic phase (Figure 4A). In contrast, liver granulomas from infected ST2−/− mice showed very few α-SMA-expressing cells (Figure 4B). Morphometric analysis of the brown-stained areas in hepatic granulomas of chronically infected mice from both mouse strains conﬁrmed a signiﬁcantly lower expression of α-SMA+ cells in granulomas formed in the absence of the activation of the IL-33/ST2 pathway (Figure 4C). Figure 4. Analysis of myoﬁbroblast activation in S. mansoni-induced hepatic granulomas. Photomi- crographs of IHC-stained liver sections slides of granulomas from WT (A) and ST2−/− (B) mice after incubation with anti-α-SMA antibody in chronic schistosomiasis. Scale bars = 50 µm. Morphometric analysis showing the percentage of the area marked with α-SMA per ﬁeld in hepatic granulomas during the chronic phase of infection (C). The chart bars show the mean ± SEM of three mice per group. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between the groups were carried out by the Mann–Whitney U test, and the p-value was assigned. Differences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. 2.5. Collagen Deposition and Expression of Pro-Fibrotic Genes in Livers of S. mansoni-Infected Mice To evaluate collagen production and deposition in the hepatic granulomas of WT and ST2−/− mice, hydroxyproline levels were measured. The hydroxyproline concentration in the liver increased signiﬁcantly during the chronic phase of infection in both WT and ST2−/− mice; however, there was no difference between the strains (Figure 5A). Histopatho- logical analysis of the liver parenchyma stained with Masson’s trichrome, a stain speciﬁc for collagen deposition, revealed that, although the hepatic hydroxyproline content in infected mice of both strains was similar, the pattern of collagen deposition in the granulomas showed marked differences between groups, especially in the chronic phase of S. mansoni infection (Figure 5B,C). In the chronic phase of schistosomiasis, the collagen deposition in the liver granulomas of infected WT mice was compact and organised in well-delimited ﬁbres around the eggs (Figure 5B), while in infected ST2−/− mice, the collagen was loosely and widely deposited around the parasite eggs with no well-deﬁned ﬁbre organisation (Figure 5C). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 8 of 21 Figure 4. Analysis of myoﬁbroblast activation in S. mansoni-induced hepatic granulomas. Photomi-crographs of IHC-stained liver sections slides of granulomas from WT (A) and ST2−/− (B) mice after incubation with anti-α-SMA antibody in chronic schistosomiasis. Scale bars = 50 µm. Morphometric analysis showing the percentage of the area marked with α-SMA per ﬁeld in hepatic granulomas during the chronic phase of infection (C). The chart bars show the mean ± SEM of three mice per group. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between the groups were carried out by the Mann–Whitney U test, and the p‐value was assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. 2.5. Collagen Deposition and Expression of Pro‐Fibrotic Genes in Livers of S. mansoni‐Infected Mice To evaluate collagen production and deposition in the hepatic granulomas of WT and ST2−/− mice, hydroxyproline levels were measured. The hydroxyproline concentration in the liver increased signiﬁcantly during the chronic phase of infection in both WT and ST2−/− mice; however, there was no diﬀerence between the strains (Figure 5A). Histopatho-logical analysis of the liver parenchyma stained with Masson’s trichrome, a stain speciﬁc for collagen deposition, revealed that, although the hepatic hydroxyproline content in in-fected mice of both strains was similar, the pattern of collagen deposition in the granulo-mas showed marked diﬀerences between groups, especially in the chronic phase of S. mansoni infection (Figure 5B,C). In the chronic phase of schistosomiasis, the collagen dep-osition in the liver granulomas of infected WT mice was compact and organised in well-delimited ﬁbres around the eggs (Figure 5B), while in infected ST2−/− mice, the collagen was loosely and widely deposited around the parasite eggs with no well-deﬁned ﬁbre organisation (Figure 5C). To assess the role of the activation of the IL-33/ST2 pathway on the expression of genes associated with liver ﬁbrosis in infected ST2−/− mice, qPCR analysis of collagen genes was performed (Figure 5). Regarding mRNA expression of Col I (Figure 5D), the S. mansoni infection induced a signiﬁcant increase in Col I expression in WT and ST2−/− mice, with no signiﬁcant diﬀerences between mouse strains. As for Col III and Col VI, the infection in-duced a signiﬁcant increase in the mRNA expression of both genes in WT mice, while in infected ST2−/− mice, the expressions of Col III (Figure 5D) and Col VI (Figure 5E) were statistically similar to that of uninfected animals, both in the acute and chronic phases of schistosomiasis. Thus, the mRNA expression of Col III and Col VI in the livers of infected ST2−/− mice was signiﬁcantly lower than in infected WT mice (Figure 5E,F). The reduction of type III collagen ﬁbres in liver granulomas of infected ST2−/− mice was conﬁrmed by Picrosirius staining. Although red-stained collagen ﬁbres were ob-served in the granulomatous reactions around parasite eggs trapped in the livers of both mouse strains in chronic schistosomiasis (Figure 6A,C), the analysis of stained liver sec-tions under polarised light conﬁrmed that granulomas from WT mice (Figure 6B) pre-sented red and green collagen ﬁbres, which represent type I and III collagens, respectively, with overlapping type I and III ﬁbres shown in yellow. In contrast, collagen ﬁbres in liver granulomas from infected ST2−/− mice were almost all red under polarised light, indicating a clear predominance of type I collagen (Figure 6D). As indicated by Gomori’s ammoniacal silver staining of livers from S. mansoni-in-fected mice, the reduction of collagen III and VI deposition in hepatic granulomas formed Int. J. Mol. Sci. 2023, 24, 10237 8 of 20 Figure 5. Evaluation of collagen induction and deposition in the liver of WT and ST2−/− mice infected with S. mansoni. Quantiﬁcation of hydroxyproline concentrations in the livers of WT and ST2−/− mice after experimental infection with S. mansoni (A). Photomicrographs of Masson’s Trichrome-stained hepatic granulomas of WT (B) or ST2−/− (C) mice chronically infected (14 weeks) with S. mansoni. Scale bars = 50 µm. Relative quantiﬁcation of mRNA expression of Col I (D), Col III (E), and Col VI (F) in the livers of WT and ST2−/− mice. RT-PCR was performed by extracting mRNA from the liver homogenate from uninfected WT and ST2−/− mice or mice at weeks 8 (acute infection) and 12 (chronic infection) of S. mansoni infection. Chart bars show mean ± SEM of 12 mice per group (A) and 5 mice per group (D–F). Normality was determined by the Kolmogorov–Smirnov test. Comparisons between groups were performed by two-way ANOVA followed by Bonferroni’s post-test, and p-values were assigned. Differences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. # represents signiﬁcant differences between infected mice and their uninfected controls; asterisk represents signiﬁcant differences between chronically infected mice compared with mice of the same strain in the acute phase of infection (# p < 0.05; or ## p < 0.01; #### p < 0.0001). To assess the role of the activation of the IL-33/ST2 pathway on the expression of genes associated with liver ﬁbrosis in infected ST2−/− mice, qPCR analysis of collagen genes was performed (Figure 5). Regarding mRNA expression of Col I (Figure 5D), the S. mansoni infection induced a signiﬁcant increase in Col I expression in WT and ST2−/− mice, with no signiﬁcant differences between mouse strains. As for Col III and Col VI, the infection induced a signiﬁcant increase in the mRNA expression of both genes in WT mice, while in infected ST2−/− mice, the expressions of Col III (Figure 5D) and Col VI (Figure 5E) were statistically similar to that of uninfected animals, both in the acute and chronic phases of schistosomiasis. Thus, the mRNA expression of Col III and Col VI in the livers of infected ST2−/− mice was signiﬁcantly lower than in infected WT mice (Figure 5E,F). The reduction of type III collagen ﬁbres in liver granulomas of infected ST2−/− mice was conﬁrmed by Picrosirius staining. Although red-stained collagen ﬁbres were observed in the granulomatous reactions around parasite eggs trapped in the livers of both mouse strains in chronic schistosomiasis (Figure 6A,C), the analysis of stained liver sections under polarised light conﬁrmed that granulomas from WT mice (Figure 6B) presented Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 9 of 21 in the absence of the activation of the IL-33/ST2 pathway was accompanied by modiﬁca-tions in the arrangement of ﬁbres (Figure 7). As shown in Figure 7B and conﬁrmed by morphometric analysis (Figure 7C), liver granulomas from chronically infected ST2−/− mice showed a signiﬁcant decrease in the formation of reticular ﬁbres (Figure 7B,C). In contrast, WT mice developed granulomas with well-organised reticular ﬁbres around the eggs, which can be seen as the black-stained regions on the periphery of the egg (Figure 7A). Figure 5. Evaluation of collagen induction and deposition in the liver of WT and ST2−/− mice infected with S. mansoni. Quantiﬁcation of hydroxyproline concentrations in the livers of WT and ST2−/− mice after experimental infection with S. mansoni (A). Photomicrographs of Masson’s Trichrome-stained hepatic granulomas of WT (B) or ST2−/− (C) mice chronically infected (14 weeks) with S. mansoni. Scale bars = 50 µm. Relative quantiﬁcation of mRNA expression of Col I (D), Col III (E), and Col VI (F) in the livers of WT and ST2−/− mice. RT-PCR was performed by extracting mRNA from the liver homogenate from uninfected WT and ST2−/− mice or mice at weeks 8 (acute infection) and 12 (chronic infection) of S. mansoni infection. Chart bars show mean ± SEM of 12 mice per group (A) and 5 mice per group (D–F). Normality was determined by the Kolmogorov–Smirnov test. Comparisons be-tween groups were performed by two-way ANOVA followed by Bonferroni’s post-test, and p-values were assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are repre-sented by their p-value. # represents signiﬁcant diﬀerences between infected mice and their unin-fected controls; asterisk represents signiﬁcant diﬀerences between chronically infected mice com-pared with mice of the same strain in the acute phase of infection (# p < 0.05; ** or ## p < 0.01; #### p < 0.0001). Int. J. Mol. Sci. 2023, 24, 10237 9 of 20 red and green collagen ﬁbres, which represent type I and III collagens, respectively, with overlapping type I and III ﬁbres shown in yellow. In contrast, collagen ﬁbres in liver granulomas from infected ST2−/− mice were almost all red under polarised light, indicating a clear predominance of type I collagen (Figure 6D). Figure 6. Collagen ﬁbre deposition on Picrosirius Red staining in hepatic granulomas of WT and ST2−/− mice chronically infected with S. mansoni. Photomicrographs of liver sections from infected WT and ST2−/− mice stained with Picrosirius Red. Scale bar = 100 µm. Hepatic granuloma of WT mouse in chronic phase under optical microscopy (A) and under polarised light (B). Hepatic granuloma of ST2−/− mouse in chronic phase under optical microscopy (C) and under polarised light (D). The green arrow indicates the presence of collagen III (greenish colour), and the yellow arrow indicates the presence of collagens I and III (yellowish colour) (B), while the presence of collagen I (reddish colour) is indicated by a red arrow (D). As indicated by Gomori’s ammoniacal silver staining of livers from S. mansoni-infected mice, the reduction of collagen III and VI deposition in hepatic granulomas formed in the absence of the activation of the IL-33/ST2 pathway was accompanied by modiﬁcations in the arrangement of ﬁbres (Figure 7). As shown in Figure 7B and conﬁrmed by morphometric analysis (Figure 7C), liver granulomas from chronically infected ST2−/− mice showed a signiﬁcant decrease in the formation of reticular ﬁbres (Figure 7B,C). In contrast, WT mice developed granulomas with well-organised reticular ﬁbres around the eggs, which can be seen as the black-stained regions on the periphery of the egg (Figure 7A). Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 10 of 21 Figure 6. Collagen ﬁbre deposition on Picrosirius Red staining in hepatic granulomas of WT and ST2−/− mice chronically infected with S. mansoni. Photomicrographs of liver sections from infected WT and ST2−/− mice stained with Picrosirius Red. Scale bar = 100 µm. Hepatic granuloma of WT mouse in chronic phase under optical microscopy (A) and under polarised light (B). Hepatic gran-uloma of ST2−/− mouse in chronic phase under optical microscopy (C) and under polarised light (D). The green arrow indicates the presence of collagen III (greenish colour), and the yellow arrow indi-cates the presence of collagens I and III (yellowish colour) (B), while the presence of collagen I (red-dish colour) is indicated by a red arrow (D). Figure 7. Reticular ﬁbres of collagen in hepatic granulomas from S. mansoni-infected WT and ST2−/− mice. Representative photomicrographs of liver sections from WT (A) and ST2−/− (B) mice chroni-cally infected with S. mansoni stained with Gomori’s Ammoniacal Silver. Scale bar = 50 µm. Mor-phometric analysis of reticular ﬁbres in liver sections from WT (blue squares) and ST2−/− (black tri-angles) mice during the acute and chronic phases of infection (C). Data are represented as a dot plot and mean ± SEM, each point representing the average of 20 granulomas per mice, with 4–8 mice per group. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between groups were performed by Student’s t‐test, and p-values were assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. Int. J. Mol. Sci. 2023, 24, 10237 10 of 20 Figure 7. Reticular ﬁbres of collagen in hepatic granulomas from S. mansoni-infected WT and ST2−/− mice. Representative photomicrographs of liver sections from WT (A) and ST2−/− (B) mice chronically infected with S. mansoni stained with Gomori’s Ammoniacal Silver. Scale bar = 50 µm. Morphometric analysis of reticular ﬁbres in liver sections from WT (blue squares) and ST2−/− (black triangles) mice during the acute and chronic phases of infection (C). Data are represented as a dot plot and mean ± SEM, each point representing the average of 20 granulomas per mice, with 4–8 mice per group. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between groups were performed by Student’s t-test, and p-values were assigned. Differences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. 3. Discussion Parasitic diseases such as schistosomiasis are an important cause of liver ﬁbrosis. Liver lesions caused by the S. mansoni infection are initiated by the retention of eggs in the hepatic capillaries, which activate elements of the innate and adaptive immune responses and lead to the formation of granulomas, a process that often results in liver ﬁbrosis, associated with the severity of chronic schistosomiasis [7,24,53–55]. In the current study, we evaluated the role of the IL-33/ST2 activation pathway in the development of liver ﬁbrosis caused by the experimental S. mansoni infection. Our data revealed a signiﬁcant reduction of pro-ﬁbrotic cytokines, such as IL-13 and IL-17, and the tissue-repairing IL-22 in the livers of chronically infected ST2−/− mice. Moreover, infected ST2−/− mice showed lower activation of HSCs and, consequently, lower differentiation of these cells into myoﬁbroblasts. The decreased myoﬁbroblast differentiation was accompanied by a reduced expression of type III and type VI collagen and lower formation of reticular collagen ﬁbres in liver granulomas of infected ST2−/− mice, indicating that the IL-33/ST2 activation pathway participates in liver- repairing mechanisms and the appropriate organisation of collagen around parasite eggs. Interventions in particular immune pathways, such as IL-33/ST2 signalling, may impact pathology as long as they can interfere with broader immune-mediated mechanisms. Vannella et al. [34] have shown that the ablation of the IL-33/ST2 pathway during S. mansoni- induced inﬂammation may only signiﬁcantly alter type-2 responses in the lung and liver in conjunction with the blockage of TSLP and IL-25. In addition, we observed in a previous work [50] that abrogation of the IL-33/ST2 pathway in experimental S. mansoni infections does not affect type-2 response, but is sufﬁcient for inducing changes in global inﬂammation on its own, compromising immunoregulatory processes in chronic schistosomiasis. Conﬁrming our published results [50], the current data also demonstrated that, during experimental infection with S. mansoni, there is an activation of Th2 and Th17 responses, regardless of the absence of the IL-33 receptor; however, the IL-22 levels decreased sys- temically and locally. IL-22 is a cytokine homologous to IL-10, as their receptors share the IL-10Rb chain. Yet, unlike IL-10, which acts on immune cells, such as macrophages and T cells, IL-22 performs its function in tissue cells, such as hepatocytes [56,57]. During in- ﬂammatory processes, IL-22 directly protects the host tissue from destructive damage from the exacerbated immune response; in the liver, this cytokine promotes repair and tissue regeneration by stimulating the proliferation and survival of hepatocytes [57–59]. Aside from the possible role in tissue repairing, IL-22 and IL-10 may also play an important role in the modulation of liver ﬁbrosis, decreasing HSC activation and tissue inﬂammation [59–61]. Int. J. Mol. Sci. 2023, 24, x FOR PEER REVIEW 10 of 21 Figure 6. Collagen ﬁbre deposition on Picrosirius Red staining in hepatic granulomas of WT and ST2−/− mice chronically infected with S. mansoni. Photomicrographs of liver sections from infected WT and ST2−/− mice stained with Picrosirius Red. Scale bar = 100 µm. Hepatic granuloma of WT mouse in chronic phase under optical microscopy (A) and under polarised light (B). Hepatic gran-uloma of ST2−/− mouse in chronic phase under optical microscopy (C) and under polarised light (D). The green arrow indicates the presence of collagen III (greenish colour), and the yellow arrow indi-cates the presence of collagens I and III (yellowish colour) (B), while the presence of collagen I (red-dish colour) is indicated by a red arrow (D). Figure 7. Reticular ﬁbres of collagen in hepatic granulomas from S. mansoni-infected WT and ST2−/− mice. Representative photomicrographs of liver sections from WT (A) and ST2−/− (B) mice chroni-cally infected with S. mansoni stained with Gomori’s Ammoniacal Silver. Scale bar = 50 µm. Mor-phometric analysis of reticular ﬁbres in liver sections from WT (blue squares) and ST2−/− (black tri-angles) mice during the acute and chronic phases of infection (C). Data are represented as a dot plot and mean ± SEM, each point representing the average of 20 granulomas per mice, with 4–8 mice per group. Normality was determined by the Kolmogorov–Smirnov test. Comparisons between groups were performed by Student’s t‐test, and p-values were assigned. Diﬀerences between WT and ST2−/− mice in the same phase of infection are represented by their p-value. Int. J. Mol. Sci. 2023, 24, 10237 11 of 20 Interestingly, as demonstrated in our previous articles [50], ST2−/− mice also showed a reduction in IL-10 levels from the acute to the chronic phase of the infection, which may also be associated with the exacerbated inﬂammatory response developed by this strain. HSCs are resident cells in the hepatic parenchyma, and the transdifferentiation of these cells from their quiescent state to myoﬁbroblasts results in pro-ﬁbrotic and proliferative activity, which plays a central role in the pathogenesis of liver ﬁbrosis induced by different harmful or inﬂammatory agents [62–64]. Experimental evidence indicates that activated HSCs are also a major source of collagen deposition and are essential for liver ﬁbrosis and extracellular matrix remodelling in schistosomiasis [7,62]. However, the role of IL-33 in the activation of HSCs and liver ﬁbrosis induced by the deposition of Schistosoma spp. eggs remain poorly understood [7,38,65]. Our data revealed that S. mansoni infection in ST2−/− mice resulted in lower α-SMA expression in liver granuloma cells compared with infected WT mice. Since α-SMA expression is a marker of myoﬁbroblast differentiation [7,51], this indicates that the IL-33/ST2 activation pathway is essential for the differentiation and activation of HSC-derived myoﬁbroblasts in the livers of S. mansoni-infected mice. The weak activation of HSCs in ST2−/− mice, shown in this study, also corroborated the ﬁndings of Tan et al. [32] in an experimental model of bile duct ligation (BDL) ﬁbro- sis. HSCs isolated from ST2−/− mice, even when stimulated with high concentrations (100 ng/mL) of recombinant IL-33, could not be activated, nor could they activate the c-Jun N-terminal kinase (JNK)/ERK/p38 transcription factors of the mitogen-activated protein kinase (MAPK) pathway, which are associated with the functionality of the IL-33/ST2 pathway, thereby indicating that the biological effect of IL-33 is dependent on the activation of the MAPK pathway and ST2 signalling in HSCs. By inhibiting the JNK/ERK/p38-MAPK pathway in HSCs isolated from C57BL/6 mice, a drastic decrease in collagen production was observed, indicating that IL-33/ST2 signalling through this pathway is essential for the activation and functionality of HSCs. Furthermore, a previous study from our research group using the same experimental model showed that the concentration of IL-13, which is one of the major factors involved in the activation of HSCs and ﬁbrosis, was also lower in the liver homogenate of ST2−/− mice, while systemic and local TGF-β levels remain similar between both strains infected with S. mansoni [38,50]. Kaviratne et al. [66] demonstrated that IL-13−/− mice infected with S. mansoni presented an almost complete abrogation of the ﬁbrosis process, despite the continuous and unaltered production of TGF-β. The same study also demonstrated that, by inhibiting a part of the TGF-β signalling cascade during experimental schistosomiasis mansoni, there was no change in the development of ﬁbrosis or the production of IL- 13, suggesting that the process of liver ﬁbrosis induced by S. mansoni infections, while dependent on IL-13, is TGF-β-independent. As demonstrated in other models of liver ﬁbrosis, IL-13 can be activated via the IL-33 pathway, implying that the liver ﬁbrosis mechanisms during schistosomiasis are mainly driven by a balance between IL-33 and IL-13 instead of TGF-β [7]. Our data further support that the sustained TGF-β levels are not sufﬁcient for the induction of proper ﬁbrosis in S. mansoni infections when the IL-33/ST2 pathway is not active. However, the lower levels of IL-13 during the chronic phase may have prevented appropriate extracellular matrix deposition, thus affecting granuloma formation in the absence of IL-33/ST2 signalling. Other than that, a Th17 response has also been associated with granuloma formation and ﬁbrosis [67,68], whose activation is related to TGF-β and IL-6 [69,70]. Our data show that the levels of IL-17 and IL-22, cytokines produced by Th17 cells, were affected by the lack of IL-33/ST2 signalling. The increased IL-17 concentrations during acute infection may have contributed to the observed unchecked periovular inﬂammation since this cytokine has been associated with increased granuloma size and pathogenesis in schistosomia- sis [68,71]. Despite the lower IL-17 levels in chronic infection, they are accompanied by reduced IL-22 levels, both systemically and locally. The concomitant reduced IL-22 levels could have not been sufﬁcient to compensate for the expectedly decreased IL-17-driven liver damage. Therefore, Th17-mediated pathogenesis could still be responsible for the main- Int. J. Mol. Sci. 2023, 24, 10237 12 of 20 tained granuloma size during the chronic phase. Supporting our ﬁndings, Nady et al. [72] observed that the in vitro formation of granulomas induced by S. mansoni soluble egg antigen (SEA) signiﬁcantly decreased in the presence of IL-22 and IL-17 together, while in the presence of IL-17 alone, there is granuloma growth due to the inﬂammatory role of this cytokine. Despite the compromised differentiation of HSC-derived myoﬁbroblasts during S. man- soni infection in ST2−/− mice, the hydroxyproline content in the livers of these animals, an indirect measure of tissue collagen deposition, was similar to that of infected WT mice. Corroborating our data, Vannella et al. [34] also observed high and similar levels of hydrox- yproline in the livers of IL-33−/− mice infected with S. mansoni. Although our data showed that the tissue content of hydroxyproline was similar between the two strains of mice analysed, the differences in the deposition of collagen in the granulomas between WT and ST2−/− animals, as portrayed by Masson’s trichrome staining, were markedly signiﬁcant. ST2-deﬁcient mice deposited collagen in a completely disorganised and exudative manner. Consistent with the increase in hydroxyproline in infected mice, there was an increase in Col I mRNA expression in the livers of infected WT and ST2−/− BALB/c mice, whereas collagens III and VI increased only in the WT mice. Type I and III collagen production and deposition are commonly found in schistosomiasis granulomas, with collagen I being prominent during the chronic phase of schistosomiasis [6,73,74]. Our data showed that Col I expression was maintained in the absence of the IL- 33/ST2 pathway, but the lack of the ST2 receptor signiﬁcantly reduced Col III and Col VI expression in the acute phase of schistosomiasis, which was conﬁrmed by histological analysis with Picrosirius staining under polarised light. The induction of Col I expression in infected ST2−/− animals may be related to the initial increase in the liver levels of IL-17, as this cytokine is described as a potent inducer of type I collagen production [64,75,76]. Meanwhile, the reduction in type III collagen levels is in accordance with data found in the literature for the human and murine tendinopathy models, in which it was possible to relate IL-33 with type III collagen synthesis, showing that this cytokine plays a key role in the transition of type III collagen synthesis through the regulation of miRNA29a [77]. The lower production of type III and VI collagens in the absence of ST2 receptor activation resulted in a reduced deposition and greater disorganisation of reticular ﬁbres in these animals. Reticular ﬁbres are continuous collagen ﬁbres composed of type I and III collagens in association with other collagens, such as V [78], and this network of ﬁbres helps to protect the injured tissue, thereby contributing mainly to the containment of egg antigens [23,53]. As a result of this disorganisation in the deposition of the extracellular matrix and the signiﬁcant reduction in the formation of reticular ﬁbres in the Schistosoma-induced granulomas formed in ST2−/− mice, these structures may not correctly conﬁne the antigens released by S. mansoni eggs, developing greater microscopic liver lesions and, consequently, higher mortality [50]. This difference in the production of collagen types was observed in another experimen- tal model, as demonstrated by Li et al. [45] in a model of pulmonary ﬁbrosis induction in ST2−/− and WT C57BL/6 mice subjected to inoculation with bleomycin. In these animals, although bleomycin stimulated type I collagen expression in both mouse strains, there was a signiﬁcant reduction in Col III expression in ST2−/− mice, which was associated with early ﬁbrosis repair. When WT mice received mIL-33 together with bleomycin, an increase in inﬂammatory cells, such as alternatively activated macrophages (M2) and ILC2, was observed, which increased the production of IL-13 and TGF-β, soluble collagen, and the expression of Col III, consequently increasing the ﬁbrosis score. In addition to HSCs, alternatively activated macrophages play a special role in the ﬁbrosis process. Alternatively activated macrophages typically express low levels of inﬂammatory cytokines and high levels of TGF-β, which promote collagen expression and ﬁbrosis [79]. Experimental evidence indicates that M2 macrophages play a key role in the development of ﬁbrosis and tissue repair in schistosomiasis [6,80,81]. Considering its role as an alarmin with a type-2 immune proﬁle, IL-33 seems to be a potential activator of M2 macrophages during innate and adaptive immune responses, as they strongly amplify Int. J. Mol. Sci. 2023, 24, 10237 13 of 20 the expression of arginase I and cytokines that contribute to ﬁbrotic processes [82]. Some authors have already reported this linkage of IL-33 with macrophages and the development of pulmonary [45], renal [83], and liver ﬁbrosis due to S. japonicum infection [48]. However, there is still little clariﬁcation, and further studies are needed on the role of IL-33 in the activation and differentiation of macrophages involved in liver ﬁbrosis during S. mansoni infection, which may be the key point for understanding the development of this process. 4. Methods 4.1. Mice Six-to-eight-week-old female WT and ST2−/− BALB/c mice were used in the experi- ments. WT BALB/c mice were acquired from an established colony at the university mouse facility. ST2−/− mice, deﬁcient in the receptor that is activated by IL-33 binding [84], were originally purchased from Jackson Immuno Research Laboratories and kindly provided by Dr. João Santana (Faculdade de Medicina de Ribeirão Preto, USP, Brazil). Both strains were maintained at the animal facility for infected animals of the Department of Parasitology (ICB, UFMG, Brazil). Experimental mice were fed standard chow (Presence, Primor, Brazil) and provided tap water ad libitum. The experimental procedures were approved by the Ethics Committee on Animal Use (protocols 104/2007, 159/2012, and 368/2018) of the Universidade Federal de Minas Gerais (UFMG, Belo Horizonte, Brazil). 4.2. Parasite and Infection S. mansoni from the LE strain, originally isolated from a human patient in Belo Hori- zonte, Brazil, has been maintained by successive passages through Biomphalaria glabrata snails and Mesocricetus auratus hamsters at the Laboratory of Schistosomiasis and Helminth Immunology of the Department of Parasitology (ICB, UFMG, Brazil), and it was used for the experimental infections performed in this study. S. mansoni cercariae were harvested from infected B. glabrata snails, washed, counted, and subcutaneously injected into each mouse (25 or 50 cercariae per mouse, as deﬁned for each experiment), as previously de- scribed by Pellegrino and Macedo [85]. To conﬁrm parasite infection and burden, the left liver lobe of each infected mouse was digested in a 5% KOH solution, and the number of recovered parasite eggs was estimated, as described previously [86]. 4.3. Experimental Design To evaluate the role of IL-33/ST2 activation on the liver pathology and ﬁbrosis as- sociated with S. mansoni infection, WT and ST2−/− BALB/c mice were infected with 25 or 50 cercariae and followed for up to 14 weeks. During this period, animals showing severe clinical signs of schistosomiasis, such as accentuated weight loss and apathy, were euthanised. At 0, 4, 8, 10, and 12 weeks after S. mansoni infection, 6 mice from each experi- mental group were immobilised, and blood was collected from the tail of each individual to quantify cytokines and chemokines. At weeks 8 and 12 or 8 and 14 post infection (acute and chronic phases of schistosomiasis, respectively), 6–10 mice from each experimental group (WT and ST2−/−) were euthanised by exsanguination via the brachial plexus under deep anaesthesia by intraperitoneal injection of ketamine (80 mg/kg—Dopalen—Sespo Industry and Commerce Ltda., Jacareí, Brazil) and xylazine (10 mg/kg—Kensol—Köing Laboratories S/A, Avellaneda, Argentine). The left liver lobes of mice infected with 50 cercariae were immediately removed and processed to quantify the gene expression of collagen types I, III, and VI using real-time quantitative polymerase chain reaction (qPCR). Pathological evalua- tion was carried out in experiments with 25 cercariae, and the chronic phase was analysed at 14 weeks. Due to the higher mortality rate of mice infected with a 50 cercariae inoculum, they were analysed at the early chronic phase, i.e., 12 weeks post infection. The left liver lobe was processed to estimate the number of parasite eggs retained in the parenchyma, the median liver lobe was harvested to estimate the hydroxyproline content, and the right liver lobe was promptly ﬁxed and processed for histopathological analysis. Liver sec- Int. J. Mol. Sci. 2023, 24, 10237 14 of 20 tions were subjected to immunohistochemical procedures or were stained using either haematoxylin-eosin, Masson’s trichrome, Picrosirius red, or Gomori’s ammoniacal silver. 4.4. Cytokine and Chemokine Assay The concentrations of cytokines and chemokines were quantiﬁed in serum and liver homogenate of WT and ST2−/− BALB/c mice by ELISA using commercially available kits (Duoset, R&D Systems, Minneapolis, MN, USA), following the manufacturer’s instructions. In serum samples collected at weeks 0, 4, 8, 10, and 12 post infection, the concentrations of IL-13, IL-12p70, IL-33, TGF-β, CCL24, and IL-22 were measured. To measure TGF-β, each serum sample was activated with a HCl solution and diluted, as described by the manufacturer. IL-4, IL-13, IL-33, IL-17, and IL-22 cytokine levels were quantiﬁed in liver homogenate samples of all experimental groups, as previously described [50]. 4.5. Quantiﬁcation of Hydroxyproline As an indirect determination of collagen deposition, the hydroxyproline content was measured, as described previously [87]. A sample of the median liver lobe from each experimental mouse was homogenised in 0.9% saline solution, frozen at −80 ◦C, and lyophilised (Liotop K105—Liobras, São Carlos, Brazil). The assay was carried out with 20 mg of the lyophilised product that was subjected to alkaline hydrolysis and treated with chloramine T oxidising reagent, and the colorimetric reaction was assessed by the addition of Ehrlich’s reagent [88]. The standard curve for the assay was prepared using serial dilutions of known concentrations of hydroxyproline, and the absorbance was measured at 550 nm using a microplate reader (VersaMax, Molecular Devices, San Jose, CA, USA). 4.6. Histopathological, Immunohistochemical, and Morphometric Analyses of Liver Histopathological evaluation of hepatic granuloma and collagen deposition was per- formed in liver sections from infected mice. Liver samples from each mouse were collected, ﬁxed in 10% buffered formalin, embedded in parafﬁn, and cut into 5 µm thick sections. The liver sections were stained with haematoxylin-eosin to evaluate the overall inﬂammation and tissue damage. Masson’s trichrome and Picrosirius red staining were performed to illustrate extracellular collagen deposition around parasite eggs, and Gomori’s ammoniacal silver staining to characterise the reticular ﬁbre organisation [89–92]. Moreover, picrosirius- stained slides were examined under polarised light microscopy (Opticam, O500R, São Paulo, Brazil) to characterise collagen composition; type I collagen appears as red ﬁbres, and type III collagen appears as green ﬁbres under polarised light [89]. The HE and picrosirius-stained images were captured with an Opticam LOPT14003 camera and anal- ysed using OPTHD software (version x64, 4.10.17015.20200426). Liver sections stained with Gomori’s silver from infected mice in both the acute and chronic phases of schistosomiasis (6 mice/group) were used for morphometric analysis of reticular ﬁbres. From each mouse, 20 microscopic ﬁelds containing granulomas were randomly selected, and the images were taken with an Olympus Bx40 microscope with an attached Qcolor 3 camera (Olympus, Tokyo, Japan). The area of the reticular ﬁbres was quantiﬁed using the software package Ks-400 (version V.3.0) [93]. Immunohistochemistry (IHC) was performed on slides with 5 µm liver sections of mice from the different experimental groups to evaluate the expression of α-SMA. Each section was deparafﬁnised, rehydrated, and subjected to antigen retrieval using a 0.01 M sodium citrate solution (pH 6.0) by microwave heating at 80% power for 10 min. The slides were then incubated with anti-α-SMA antibody (Abcam 32575, Cambridge, MA, USA) at a dilution of 1:800 for 24 h at 4 ◦C. After incubation with the primary antibody, the slides were washed, and the presence of α-SMA in the liver parenchyma was revealed using the EnVision+ Dual Link System-horseradish peroxidase (HRP) kit (Agilent Dako, Santa Clara, CA, USA), in which the secondary antibody is associated with HRP-labelled polymers, and peroxidase activity was revealed by the chromogen 3,3(cid:48)-diaminobenzidine (DAB; Agilent Dako, Santa Clara, CA, USA). All slides were counterstained with Harris Int. J. Mol. Sci. 2023, 24, 10237 15 of 20 modiﬁed haematoxylin (Sigma-Aldrich, St. Louis, MO, USA) for 2 min and with bluing reagent (Thermo Fisher Scientiﬁc, Waltham, MA, USA) for 1 min. Subsequently, the samples were analysed under an optical microscope, and α-SMA was quantitatively estimated in the livers of infected ST2−/− and WT mice using the ImageJ software (version 1.43). The amount of α-SMA in granuloma cells was expressed as a percentage of the α-SMA-stained area per ﬁeld, analysing exudative granulomas containing viable miracidia from chronically infected mice of both strains (3 mice/group), using an Olympus Bx41TF microscope coupled to a DP12-camera 2 (Olympus Optical, Tokyo, Japan). 4.7. Quantiﬁcation of mRNA Expression by RT-qPCR The quantiﬁcation of gene expression of collagen types I, III, and VI was performed in liver samples using quantitative reverse transcription polymerase chain reaction (qPCR). Liver fragments of mice from different experimental groups were collected, stored in RNalater™ Stabilisation Solution (Invitrogen™, Waltham, MA, USA), and processed for RNA extraction using a tissue homogeniser (Omni TH) and TRIzol kit (Invitrogen), accord- ing to the manufacturer’s instructions. After the quantiﬁcation of RNA in a spectropho- tometer (Epoch™ Microplate, Biotek, VT, USA), the samples were subjected to complete digestion using a DNAse Turbo DNA-free kit (Ambion, Life Technologies, Thermo Fisher Scientiﬁc, Waltham, MA, USA), followed by cDNA synthesis using a commercially available High-Capacity cDNA reverse transcription kit (Ambion, Life Technologies). For quantiﬁca- tion of gene expression, each PCR experiment was performed in triplicate in a total volume of 10 µL, combining 5 µL of SYBR Green Master Mix (Ambion, Life Technologies), 0.3 µL of each primer (forward and reverse) for the targeted gene, 2.4 µL of Milli-Q water, and 2.0 µL of cDNA. The speciﬁc primers for the collagen type I, collagen type III, and collagen type VI genes are listed in Table 1. The housekeeping gene GAPDH was used as an internal con- trol. qPCR was performed in a StepOnePlus Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) using the following parameters: Taq polymerase was activated at 95 ◦C for 10 min, followed by 40 cycles of ampliﬁcation (95 ◦C for 15 s for denaturation, and 60 ◦C for 60 s for primer annealing and extension). Relative quantiﬁcation of the mRNA ∆CT method [94]. Data were expression for each target gene was performed using the 2 analysed using StepOne 2.3 (Life Technologies). Table 1. Primer sequences for GAPDH, Collagen type I, Collagen type III, and Collagen type VI used in the RT-qPCR. Gene GAPDH [95] Col I [95] Col III [94] Col VI [96] Forward Reverse Forward Reverse Forward Reverse Forward Reverse Sequence (5(cid:48)–3(cid:48)) AGG TCG GTG TGA ACG GAT TTG TGT AGA CCA TGT AGT TGA GGT CA ACT GGA CTG TCC CAA CCC C CTT AGT TTG GAC AGG ATC TGG AAC CTG GTT TCT TCT CAC CCT TC ACT CAT AGG ACT GAC CAA GGT GG CGC CCT TCC CAC TGA CAA GCG TTC CCT TTA AGA CAG TTG AG 4.8. Statistical Analysis Statistical analysis was performed, and graphs were generated using GraphPad Prism 8.0.2. Data were subjected to ROUT tests for the identiﬁcation and removal of outliers, and their normal distribution was veriﬁed using the Kolmogorov–Smirnov test. Comparisons between the experimental groups at different time points of infection were carried out by two-way ANOVA, followed by Bonferroni’s post-test, and comparisons between only one phase of infection were carried out by Student’s t-test. For non-parametric datasets, Int. J. Mol. Sci. 2023, 24, 10237 16 of 20 comparisons between multiple groups were performed by the Kruskal–Wallis test, while comparisons between only two groups were carried out by the Mann–Whitney U test. After being assigned, p-values ≤ 0.05 were considered signiﬁcant. 5. Conclusions The severity of schistosomiasis has been associated with the intensity and extent of tissue damage and ﬁbrosis induced by parasite eggs trapped in the host’s tissues. This study shows that, in S. mansoni-infected mice, the IL-33/ST2 activation pathway did not signiﬁcantly change the quantity of collagen deposited around the parasite eggs retained in the host’s liver. However, activation of this pathway participates in the induction of some pro-ﬁbrotic cytokines, such as IL-13 and IL-17, and the tissue-repairing IL-22 in the livers of chronically infected ST2−/− mice. Moreover, IL-33/ST2 signalling is essential for the differentiation of HSCs into myoﬁbroblasts and, consequently, for the production of type III and VI collagen. This results in the proper organisation of collagen reticular ﬁbres within the granulomatous reaction, enabling the adequate development of hepatic granulomas and antigen retention during schistosomiasis, protecting the liver parenchyma. Thus, the current study provides evidence that supports the role of the IL-33/ST2 pathway in liver-repairing mechanisms and the appropriate organisation of collagen around parasite eggs, participating in S. mansoni-induced liver ﬁbrosis and controlling disease severity. Author Contributions: L.M. and D.A.N.-C. conceptualised and designed the study; L.M., G.M.A.C. and I.C.R. performed animal experiments and data analysis; L.M. and W.P.A. conducted the q-PCR procedure and data analysis; É.M.L.R. supervised and provided resources to carry out the q-PCR experiments; T.A.P. performed the immunohistochemical analysis; A.C.J. and J.M.P.M. executed the morphometric analysis; W.L.T. and R.E. supervised the histopathological analysis; L.M., G.M.A.C., R.E. and D.A.N.-C. wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: Financial support for this work was provided by the Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG-Brazil, grant #PPM-00500-15 and grant #APQ 01637-17). D.A.N.- C., L.M. and G.M.A.C. were supported by a fellowship from CNPq/Brazil. J.M.P.M. was supported by CAPES/Brazil. Institutional Review Board Statement: The animal study protocol was approved by the Institutional Ethics Committee on Animal Use of Universidade Federal de Minas Gerais (protocols code 104/2007, 159/2012, and 368/2018). Informed Consent Statement: Not applicable. Data Availability Statement: All data supporting our results is present in the paper. Acknowledgments: Acknowledgement is due to José Carlos dos Reis for the technical support provided. Conﬂicts of Interest: The authors declare no conﬂict of interest. The funders had no role in the study design, data collection, and analysis, or decision to publish the manuscript. References 1. Wick, G.; Grundtman, C.; Mayerl, C.; Wimpissinger, T.-F.; Feichtinger, J.; Zelger, B.; Sgonc, R.; Wolfram, D. The Immunology of 2. Fibrosis. Annu. Rev. Immunol. 2013, 31, 107–135. [CrossRef] [PubMed] Pellicoro, A.; Ramachandran, P.; Iredale, J.P.; Fallowﬁeld, J.A. Liver Fibrosis and Repair: Immune Regulation of Wound Healing in a Solid Organ. Nat. Rev. Immunol. 2014, 14, 181–194. [CrossRef] [PubMed] 3. Wynn, T. Cellular and Molecular Mechanisms of Fibrosis. J. Pathol. 2008, 214, 199–210. [CrossRef] [PubMed] 4. Wick, G.; Backovic, A.; Rabensteiner, E.; Plank, N.; Schwentner, C.; Sgonc, R. The Immunology of Fibrosis: Innate and Adaptive Responses. Trends Immunol. 2010, 31, 110–119. [CrossRef] 5. Weiskirchen, R.; Weiskirchen, S.; Tacke, F. Organ and Tissue Fibrosis: Molecular Signals, Cellular Mechanisms and Translational 6. 7. Implications. Mol. Asp. Med. 2019, 65, 2–15. [CrossRef] Anthony, B.; Allen, J.T.; Li, Y.S.; McManus, D.P. Hepatic Stellate Cells and Parasite-Induced Liver Fibrosis. Parasites Vectors 2010, 3, 60. [CrossRef] Carson, J.P.; Ramm, G.A.; Robinson, M.W.; McManus, D.P.; Gobert, G.N. Schistosome-Induced Fibrotic Disease: The Role of Hepatic Stellate Cells. Trends Parasitol. 2018, 34, 524–540. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10237 17 of 20 8. 9. Kamdem, S.D.; Moyou-Somo, R.; Brombacher, F.; Nono, J.K. Host Regulators of Liver Fibrosis during Human Schistosomiasis. Front. Immunol. 2018, 9, 2781. [CrossRef] Chen, T.T.; Peng, S.; Wang, Y.; Hu, Y.; Shen, Y.; Xu, Y.; Yin, J.; Liu, C.; Cao, J. Improvement of Mitochondrial Activity and Fibrosis by Resveratrol Treatment in Mice with Schistosoma Japonicum Infection. Biomolecules 2019, 9, 658. [CrossRef] 10. McManus, D.P.; Dunne, D.W.; Sacko, M.; Utzinger, J.; Vennervald, B.J.; Zhou, X.-N. Schistosomiasis. Nat. Rev. Dis. Primers 2018, 4, 13. [CrossRef] 11. Chitsulo, L.; Engels, D.; Montresor, A.; Savioli, L. The Global Status of Schistosomiasis and Its Control. Acta Trop. 2000, 77, 41–51. [CrossRef] 12. Masamba, P.; Adenowo, A.; Oyinloye, B.; Kappo, A. Universal Stress Proteins as New Targets for Environmental and Therapeutic Interventions of Schistosomiasis. Int. J. Environ. Res. Public Health 2016, 13, 972. [CrossRef] 13. De Jesus, A.R.; Silva, A.; Santana, L.B.; Magalhães, A.; de Jesus, A.A.; de Almeida, R.P.; Rêgo, M.A.V.; Burattini, M.N.; Pearce, E.J.; Carvalho, E.M. Clinical and Immunologic Evaluation of 31 Patients with Acute Schistosomiasis Mansoni. J. Infect. Dis. 2002, 185, 98–105. [CrossRef] 14. Hiatt, R.A.; Ottesen, E.A.; Sotomayor, Z.R.; Lawley, T.J. Serial Observations of Circulating Immune Complexes in Patients with Acute Schistosomiasis. J. Infect. Dis. 1980, 142, 665–670. [CrossRef] 15. Abath, F.G.C.; Morais, C.N.L.; Montenegro, C.E.L.; Wynn, T.A.; Montenegro, S.M.L. Immunopathogenic Mechanisms in Schisto- somiasis: What Can Be Learnt from Human Studies? Trends Parasitol. 2006, 22, 85–91. [CrossRef] 16. El Ridi, R.; Tallima, H.; Mahana, N.; Dalton, J.P. Innate Immunogenicity and in vitro Protective Potential of Schistosoma mansoni Lung Schistosomula Excretory-Secretory Candidate Vaccine Antigens. Microbes Infect. 2010, 12, 700–709. [CrossRef] 17. Egesa, M.; Lubyayi, L.; Tukahebwa, E.M.; Bagaya, B.S.; Chalmers, I.W.; Wilson, S.; Hokke, C.H.; Hoffmann, K.F.; Dunne, D.W.; Yazdanbakhsh, M.; et al. Schistosoma Mansoni Schistosomula Antigens Induce Th1/Pro-Inﬂammatory Cytokine Responses. Parasite Immunol. 2018, 40, e12592. [CrossRef] 18. Okano, M.; Satoskar, A.R.; Nishizaki, K.; Abe, M.; Harn, D.A. Induction of Th2 Responses and IgE Is Largely Due to Carbohydrates Functioning as Adjuvants on Schistosoma mansoni Egg Antigens. J. Immunol. 1999, 163, 6712–6717. [CrossRef] 19. Okano, M.; Satoskar, A.R.; Nishizaki, K.; Harn, D.A. Lacto-N-Fucopentaose III Found on Schistosoma mansoni Egg Antigens Functions as Adjuvant for Proteins by Inducing Th2-Type Response. J. Immunol. 2001, 167, 442–450. [CrossRef] 20. Thomas, P.G.; Carter, M.R.; Atochina, O.; Da’Dara, A.A.; Piskorska, D.; McGuire, E.; Harn, D.A. Maturation of Dendritic Cell 2 Phenotype by a Helminth Glycan Uses a Toll-Like Receptor 4-Dependent Mechanism. J. Immunol. 2003, 171, 5837–5841. [CrossRef] 21. Herbert, D.R.; Hölscher, C.; Mohrs, M.; Arendse, B.; Schwegmann, A.; Radwanska, M.; Leeto, M.; Kirsch, R.; Hall, P.; Mossmann, H.; et al. Alternative Macrophage Activation Is Essential for Survival during Schistosomiasis and Downmodulates T Helper 1 Responses and Immunopathology. Immunity 2004, 20, 623–635. [CrossRef] [PubMed] 22. Everts, B.; Perona-Wright, G.; Smits, H.H.; Hokke, C.H.; van der Ham, A.J.; Fitzsimmons, C.M.; Doenhoff, M.J.; van der Bosch, J.; Mohrs, K.; Haas, H.; et al. Omega-1, a Glycoprotein Secreted by Schistosoma mansoni Eggs, Drives Th2 Responses. J. Exp. Med. 2009, 206, 1673–1680. [CrossRef] [PubMed] 23. Costain, A.H.; MacDonald, A.S.; Smits, H.H. Schistosome Egg Migration: Mechanisms, Pathogenesis and Host Immune Responses. Front. Immunol. 2018, 9, 3042. [CrossRef] [PubMed] 24. Taylor, J.J.; Mohrs, M.; Pearce, E.J. Regulatory T Cell Responses Develop in Parallel to Th Responses and Control the Magnitude and Phenotype of the Th Effector Populatio. J. Immunol. 2006, 176, 5839–5847. [CrossRef] 25. Rutitzky, L.I.; Smith, P.M.; Stadecker, M.J. T-Bet Protects against Exacerbation of Schistosome Egg-Induced Immunopathology by Regulating Th17-Mediated Inﬂammation. Eur. J. Immunol. 2009, 39, 2470–2481. [CrossRef] 26. Haeberlein, S.; Obieglo, K.; Ozir-Fazalalikhan, A.; Chayé, M.A.M.; Veninga, H.; van der Vlugt, L.E.P.M.; Voskamp, A.; Boon, L.; den Haan, J.M.M.; Westerhof, L.B.; et al. Schistosome Egg Antigens, Including the Glycoprotein IPSE/Alpha-1, Trigger the Development of Regulatory B Cells. PLoS Pathog. 2017, 13, e1006539. [CrossRef] 27. Hoffmann, K.F.; Cheever, A.W.; Wynn, T.A. IL-10 and the Dangers of Immune Polarization: Excessive Type 1 and Type 2 Cytokine Responses Induce Distinct Forms of Lethal Immunopathology in Murine Schistosomiasis. J. Immunol. 2000, 164, 6406–6416. [CrossRef] 28. Hesse, M.; Piccirillo, C.A.; Belkaid, Y.; Prufer, J.; Mentink-Kane, M.; Leusink, M.; Cheever, A.W.; Shevach, E.M.; Wynn, T.A. The Pathogenesis of Schistosomiasis Is Controlled by Cooperating IL-10-Producing Innate Effector and Regulatory T Cells. J. Immunol. 2004, 172, 3157–3166. [CrossRef] 29. Ross, A.G.P.; Bartley, P.B.; Sleigh, A.C.; Olds, G.R.; Li, Y.; Williams, G.M.; McManus, D.P. Schistosomiasis. N. Engl. J. Med. 2002, 346, 1212–1220. [CrossRef] 30. Chen, M.G. Schistosoma japonicum and S. japonicum-like Infections: Epidemiology, Clinical and Pathological Aspects. In Human Schistosomiasis; Jordan, P., Webbe, G., Sturrock, R.F., Eds.; CAB International: Wallingfordm, CT, USA, 1993; pp. 237–270. 31. Chang, D.; Ramalho, L.N.Z.; Ramalho, F.S.; Martinelli, A.L.C.; Zucoloto, S. Hepatic Stellate Cells in Human Schistosomiasis Mansoni: A Comparative Immunohistochemical Study with Liver Cirrhosis. Acta Trop. 2006, 97, 318–323. [CrossRef] 32. Tan, Z.; Liu, Q.; Jiang, R.; Lv, L.; Shoto, S.S.; Maillet, I.; Quesniaux, V.; Tang, J.; Zhang, W.; Sun, B.; et al. Interleukin-33 Drives Hepatic Fibrosis through Activation of Hepatic Stellate Cells. Cell Mol. Immunol. 2018, 15, 388–398. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10237 18 of 20 33. Hams, E.; Bermingham, R.; Wurlod, F.A.; Hogan, A.E.; O’Shea, D.; Preston, R.J.; Rodewald, H.; McKenzie, A.N.J.; Fallon, P.G. The Helminth T2 RNase Ω1 Promotes Metabolic Homeostasis in an IL-33- and Group 2 Innate Lymphoid Cell-dependent Mechanism. FASEB J. 2016, 30, 824–835. [CrossRef] 34. Vannella, K.M.; Ramalingam, T.R.; Borthwick, L.A.; Barron, L.; Hart, K.M.; Thompson, R.W.; Kindrachuk, K.N.; Cheever, A.W.; White, S.; Budelsky, A.L.; et al. Combinatorial Targeting of TSLP, IL-25, and IL-33 in Type 2 Cytokine–Driven Inﬂammation and Fibrosis. Sci. Transl. Med. 2016, 8, 337ra65. [CrossRef] Schmitz, J.; Owyang, A.; Oldham, E.; Song, Y.; Murphy, E.; McClanahan, T.K.; Zurawski, G.; Moshreﬁ, M.; Qin, J.; Li, X.; et al. IL-33, an Interleukin-1-like Cytokine That Signals via the IL-1 Receptor-Related Protein ST2 and Induces T Helper Type 2-Associated Cytokines. Immunity 2005, 23, 479–490. [CrossRef] 35. 36. He, X.; Xie, J.; Wang, Y.; Fan, X.; Su, Q.; Sun, Y.; Lei, N.; Zhang, D.; Gao, G.; Pan, W. Down-Regulation of MicroRNA-203-3p Initiates Type 2 Pathology during Schistosome Infection via Elevation of Interleukin-33. PLoS Pathog. 2018, 14, e1006957. [CrossRef] Schmitt, P.; Girard, J.-P.; Cayrol, C. Interleukin-33: From Biology to Potential Treatments. Med. Sci. 2019, 35, 440–451. [CrossRef] 37. 38. Mchedlidze, T.; Waldner, M.; Zopf, S.; Walker, J.; Rankin, A.L.; Schuchmann, M.; Voehringer, D.; McKenzie, A.N.J.; Neurath, M.F.; Pﬂanz, S.; et al. Interleukin-33-Dependent Innate Lymphoid Cells Mediate Hepatic Fibrosis. Immunity 2013, 39, 357–371. [CrossRef] 39. Higashi, T.; Friedman, S.L.; Hoshida, Y. Hepatic Stellate Cells as Key Target in Liver Fibrosis. Adv. Drug Deliv. Rev. 2017, 121, 27–42. [CrossRef] 40. Liu, Y.; Meyer, C.; Müller, A.; Herweck, F.; Li, Q.; Müllenbach, R.; Mertens, P.R.; Dooley, S.; Weng, H.-L. IL-13 Induces Connective Tissue Growth Factor in Rat Hepatic Stellate Cells via TGF-β–Independent Smad Signaling. J. Immunol. 2011, 187, 2814–2823. [CrossRef] 41. Teixeira, M.M.; Talvani, A.; Tafuri, W.L.; Lukacs, N.W.; Hellewell, P.G. Eosinophil Recruitment into Sites of Delayed-Type Hypersensitivity Reactions in Mice. J. Leukoc. Biol. 2001, 69, 353–360. [CrossRef] 42. Hams, E.; Aviello, G.; Fallon, P.G. The Schistosoma Granuloma: Friend or Foe? Front. Immunol. 2013, 4, 89. [CrossRef] [PubMed] 43. Yu, Y.; Deng, W.; Lei, J. Interleukin-33 Promotes Th2 Immune Responses in Infected Mice with Schistosoma Japonicum. Parasitol. Res. 2015, 114, 2911–2918. [CrossRef] [PubMed] 44. da Paz, V.R.F.; Figueiredo-Vanzan, D.; dos Santos Pyrrho, A. Interaction and Involvement of Cellular Adhesion Molecules in the Pathogenesis of Schistosomiasis Mansoni. Immunol. Lett. 2019, 206, 11–18. [CrossRef] [PubMed] 46. 45. Li, D.; Guabiraba, R.; Besnard, A.-G.; Komai-Koma, M.; Jabir, M.S.; Zhang, L.; Graham, G.J.; Kurowska-Stolarska, M.; Liew, F.Y.; McSharry, C.; et al. IL-33 Promotes ST2-Dependent Lung Fibrosis by the Induction of Alternatively Activated Macrophages and Innate Lymphoid Cells in Mice. J. Allergy Clin. Immunol. 2014, 134, 1422–1432.e11. [CrossRef] [PubMed] Fanny, M.; Nascimento, M.; Baron, L.; Schricke, C.; Maillet, I.; Akbal, M.; Riteau, N.; le Bert, M.; Quesniaux, V.; Ryffel, B.; et al. The IL-33 Receptor ST2 Regulates Pulmonary Inﬂammation and Fibrosis to Bleomycin. Front. Immunol. 2018, 9, 1476. [CrossRef] 47. Cao, S.; Zhu, L.; Zhu, C.; Feng, J.; Yin, J.; Lu, J.; Xu, Y.; Yang, H.; Huang, Y.; Zhang, Q. Helicobacter Hepaticus Infection-Induced IL-33 Promotes Hepatic Inﬂammation and Fibrosis through ST2 Signaling Pathways in BALB/c Mice. Biochem. Biophys. Res. Commun. 2020, 525, 654–661. [CrossRef] 48. Peng, H.; Zhang, Q.; Li, X.; Liu, Z.; Shen, J.; Sun, R.; Wei, J.; Zhao, J.; Wu, X.; Feng, F.; et al. IL-33 Contributes to Schistosoma Japonicum-Induced Hepatic Pathology through Induction of M2 Macrophages. Sci. Rep. 2016, 6, 29844. [CrossRef] 49. Li, Z.-Y.; Xiao, L.; Lin, G.; Tang, J.; Chen, Y.; Chen, L.; Li, B.; Wu, M.; Liu, S.; Huang, C.; et al. Contribution of Tissue Transglutaminase to the Severity of Hepatic Fibrosis Resulting from Schistosoma Japonicum Infection through the Regulation of IL-33/ST2 Expression. Parasites Vectors 2019, 12, 302. [CrossRef] 50. Maggi, L.; Rocha, I.C.; Camelo, G.M.A.; Fernandes, V.R.; Negrão-Corrêa, D. The IL-33/ST2 Pathway Is Not Essential to Th2 Stimulation but Is Key for Modulation and Survival during Chronic Infection with Schistosoma mansoni in Mice. Cytokine 2021, 138, 155390. [CrossRef] 51. Bachem, M.G.; Meyer, D.; Melchior, R.; Sell, K.M.; Gressner, A.M. Activation of Rat Liver Perisinusoidal Lipocytes by Transforming Growth Factors Derived from Myoﬁbroblastlike Cells. A Potential Mechanism of Self Perpetuation in Liver Fibrogenesis. J. Clin. Investig. 1992, 89, 19–27. [CrossRef] 52. Geerts, A. Formation of Normal Desmin Intermediate Filaments in Mouse Hepatic Stellate Cells Requires Vimentin. Hepatology 2001, 33, 177–188. [CrossRef] 53. Pearce, E.J.; MacDonald, A.S. The Immunobiology of Schistosomiasis. Nat. Rev. Immunol. 2002, 2, 499–511. [CrossRef] 54. Smith, P.M.; Shainheit, M.G.; Bazzone, L.E.; Rutitzky, L.I.; Poltorak, A.; Stadecker, M.J. Genetic Control of Severe Egg-Induced Immunopathology and IL-17 Production in Murine Schistosomiasis. J. Immunol. 2009, 183, 3317–3323. [CrossRef] 55. Yu, Y.; Wang, J.; Wang, X.; Gu, P.; Lei, Z.; Tang, R.; Wei, C.; Xu, L.; Wang, C.; Chen, Y.; et al. Schistosome Eggs Stimulate Reactive Oxygen Species Production to Enhance M2 Macrophage Differentiation and Promote Hepatic Pathology in Schistosomiasis. PLoS Negl. Trop. Dis. 2021, 15, e0009696. [CrossRef] 56. Wolk, K.; Kunz, S.; Witte, E.; Friedrich, M.; Asadullah, K.; Sabat, R. IL-22 Increases the Innate Immunity of Tissues. Immunity 2004, 21, 241–254. [CrossRef] 57. Zenewicz, L.A.; Yancopoulos, G.D.; Valenzuela, D.M.; Murphy, A.J.; Karow, M.; Flavell, R.A. Interleukin-22 but Not Interleukin-17 Provides Protection to Hepatocytes during Acute Liver Inﬂammation. Immunity 2007, 27, 647–659. [CrossRef] 58. Zenewicz, L.A.; Flavell, R.A. Recent Advances in IL-22 Biology. Int. Immunol. 2011, 23, 159–163. [CrossRef] Int. J. Mol. Sci. 2023, 24, 10237 19 of 20 59. Xing, Z.; Wu, Y.; Liu, N. IL-22 Alleviates the Fibrosis of Hepatic Stellate Cells via the Inactivation of NLRP3 Inﬂammasome Signaling. Exp. Ther. Med. 2021, 22, 1088. [CrossRef] 60. Lu, D.-H.; Guo, X.; Qin, S.; Luo, W.; Huang, X.; Chen, M.; Wang, J.; Ma, S.; Yang, X.; Jiang, H. Interleukin-22 Ameliorates Liver Fibrogenesis by Attenuating Hepatic Stellate Cell Activation and Downregulating the Levels of Inﬂammatory Cytokines. World J. Gastroenterol. 2015, 21, 1531–1545. [CrossRef] 61. Zhang, L.-J.; Zheng, W.-D.; Chen, Y.-X.; Huang, Y.-H.; Chen, Z.-X.; Zhang, S.-J.; Shi, M.-N.; Wang, X.-Z. Antiﬁbrotic Effects of Interleukin-10 on Experimental Hepatic Fibrosis. Hepatogastroenterology 2007, 54, 2092–2098. 62. Benyon, R.C.; Arthur, M.J.P. Extracellular Matrix Degradation and the Role of Hepatic Stellate Cells. Semin. Liver Dis. 2001, 21, 63. 373–384. [CrossRef] Seki, E.; Brenner, D.A. Recent Advancement of Molecular Mechanisms of Liver Fibrosis. J. Hepatobiliary Pancreat Sci. 2015, 22, 512–518. [CrossRef] [PubMed] 64. Tsuchida, T.; Friedman, S.L. Mechanisms of Hepatic Stellate Cell Activation. Nat. Rev. Gastroenterol. Hepatol. 2017, 14, 397–411. [CrossRef] [PubMed] 65. Osman, A.; Niles, E.G.; Verjovski-Almeida, S.; LoVerde, P.T. Schistosoma Mansoni TGF-β Receptor II: Role in Host Ligand-Induced Regulation of a Schistosome Target Gene. PLoS Pathog. 2006, 2, e54. [CrossRef] [PubMed] 66. Kaviratne, M.; Hesse, M.; Leusink, M.; Cheever, A.W.; Davies, S.J.; McKerrow, J.H.; Wakeﬁeld, L.M.; Letterio, J.J.; Wynn, T.A. IL-13 Activates a Mechanism of Tissue Fibrosis That Is Completely TGF-β Independent. J. Immunol. 2004, 173, 4020–4029. [CrossRef] 67. Chuah, C.; Jones, M.K.; Burke, M.L.; McManus, D.P.; Gobert, G.N. Cellular and Chemokine-Mediated Regulation in Schistosome- Induced Hepatic Pathology. Trends Parasitol. 2014, 30, 141–150. [CrossRef] 68. Angeles, J.M.M.; Mercado, V.J.P.; Rivera, P.T. Behind Enemy Lines: Immunomodulatory Armamentarium of the Schistosome Parasite. Front. Immunol. 2020, 11, 1018. [CrossRef] 69. Veldhoen, M.; Hocking, R.J.; Atkins, C.J.; Locksley, R.M.; Stockinger, B. TGF-β in the Context of an Inﬂammatory Cytokine Milieu Supports De Novo Differentiation of IL-17-Producing T Cells. Immunity 2006, 24, 179–189. [CrossRef] 70. Mangan, P.R.; Harrington, L.E.; O’Quinn, D.B.; Helms, W.S.; Bullard, D.C.; Elson, C.O.; Hatton, R.D.; Wahl, S.M.; Schoeb, T.R.; Weaver, C.T. Transforming Growth Factor-β Induces Development of the TH17 Lineage. Nature 2006, 441, 231–234. [CrossRef] 71. Bai, Y.; Guan, F.; Zhu, F.; Jiang, C.; Xu, X.; Zheng, F.; Liu, W.; Lei, J. IL-33/ST2 Axis Deﬁciency Exacerbates Hepatic Pathology by Regulating Treg and Th17 Cells in Murine Schistosomiasis Japonica. J. Inﬂamm. Res. 2021, 14, 5981–5998. [CrossRef] 72. Nady, S.; Shata, M.T.M.; Mohey, M.A.; El-Shorbagy, A. Protective Role of IL-22 against Schistosoma mansoni Soluble Egg Antigen- Induced Granuloma In Vitro. Parasite Immunol. 2017, 39, e12392. [CrossRef] 73. Olds, G.R.; Grifﬁn, A.; Kresina, T.F. Dynamics of Collagen Accumulation and Polymorphism in Murine Schistosoma Japonicum. Gastroenterology 1985, 89, 617–624. [CrossRef] 74. Al Adnani, M.S. Concomitant Immunohistochemical Localization of Fibronectin and Collagen in Schistosome Granulomata. J. Pathol. 1985, 147, 77–85. [CrossRef] 75. Hellerbrand, C.; Stefanovic, B.; Giordano, F.; Burchardt, E.R.; Brenner, D.A. The Role of TGFβ1 in Initiating Hepatic Stellate Cell Activation In Vivo. J. Hepatol. 1999, 30, 77–87. [CrossRef] 76. Meng, F.; Wang, K.; Aoyama, T.; Grivennikov, S.I.; Paik, Y.; Scholten, D.; Cong, M.; Iwaisako, K.; Liu, X.; Zhang, M.; et al. Interleukin-17 Signaling in Inﬂammatory, Kupffer Cells, and Hepatic Stellate Cells Exacerbates Liver Fibrosis in Mice. Gastroen- terology 2012, 143, 765–776.e3. [CrossRef] 77. Millar, N.L.; Gilchrist, D.S.; Akbar, M.; Reilly, J.H.; Kerr, S.C.; Campbell, A.L.; Murrell, G.A.C.; Liew, F.Y.; Kurowska-Stolarska, M.; McInnes, I.B. MicroRNA29a Regulates IL-33-Mediated Tissue Remodelling in Tendon Disease. Nat. Commun. 2015, 6, 6774. [CrossRef] 78. Ushiki, T. Collagen Fibers, Reticular Fibers and Elastic Fibers. A Comprehensive Understanding from a Morphological Viewpoint. 79. Arch. Histol. Cytol. 2002, 65, 109–126. [CrossRef] Spencer, M.; Yao-Borengasser, A.; Unal, R.; Rasouli, N.; Gurley, C.M.; Zhu, B.; Peterson, C.A.; Kern, P.A. Adipose Tissue Macrophages in Insulin-Resistant Subjects Are Associated with Collagen VI and Fibrosis and Demonstrate Alternative Activation. Am. J. Physiol.-Endocrinol. Metab. 2010, 299, E1016–E1027. [CrossRef] 80. Gordon, S. Alternative Activation of Macrophages. Nat. Rev. Immunol. 2003, 3, 23–35. [CrossRef] 81. Biswas, S.K.; Mantovani, A. Macrophage Plasticity and Interaction with Lymphocyte Subsets: Cancer as a Paradigm. Nat. Immunol. 2010, 11, 889–896. [CrossRef] 82. Kurowska-Stolarska, M.; Stolarski, B.; Kewin, P.; Murphy, G.; Corrigan, C.J.; Ying, S.; Pitman, N.; Mirchandani, A.; Rana, B.; van Rooijen, N.; et al. IL-33 Ampliﬁes the Polarization of Alternatively Activated Macrophages That Contribute to Airway Inﬂammation. J. Immunol. 2009, 183, 6469–6477. [CrossRef] [PubMed] 83. Li, Y.; Liu, J.; Yu, T.; Yan, B.; Li, H. Interleukin-33 Promotes Obstructive Renal Injury via Macrophages. Mol. Med. Rep. 2019, 20, 1353–1362. [CrossRef] [PubMed] 84. Townsend, M.J.; Fallon, P.G.; Matthews, D.J.; Jolin, H.E.; McKenzie, A.N.J. T1/ST2-Deﬁcient Mice Demonstrate the Importance of T1/ST2 in Developing Primary T Helper Cell Type 2 Responses. J. Exp. Med. 2000, 191, 1069–1075. [CrossRef] [PubMed] 85. Pellegrino, J.; Macedo, D.G. A Simpliﬁed Method for the Concentration of Cercariae. J. Parasitol. 1955, 41, 329. [CrossRef] 86. Cheever, A.W. Conditions Affecting the Accuracy of Potassium Hydroxide Digestion Techniques for Counting Schistosoma mansoni Eggs in Tissues. Bull. World Health Organ. 1968, 39, 328–331. Int. J. Mol. Sci. 2023, 24, 10237 20 of 20 87. Reddy, G.K.; Enwemeka, C.S. A Simpliﬁed Method for the Analysis of Hydroxyproline in Biological Tissues. Clin. Biochem. 1996, 29, 225–229. [CrossRef] 88. De Rezende, M.C.; Moreira, J.M.P.; Fernandes, L.L.M.; Rodrigues, V.F.; Negrão-Corrêa, D. Strongyloides Venezuelensis-Infection Alters the Proﬁle of Cytokines and Liver Inﬂammation in Mice Co-Infected with Schistosoma Mansoni. Cytokine 2020, 127, 154931. [CrossRef] Junqueira, L.C.U.; Bignolas, G.; Brentani, R.R. Picrosirius Staining plus Polarization Microscopy, a Speciﬁc Method for Collagen Detection in Tissue Sections. Histochem. J. 1979, 11, 447–455. [CrossRef] Junqueira, L.C.; Junqueira, L.M.M. Técnicas Básicas de Citologia e Histologia; Santos Editora: São Paulo, Brazil, 1983. 90. 91. Melo, F.; Amaral, M.; Oliveira, P.; Lima, W.; Andrade, M.; Michalick, M.; Raso, P.; Tafuri, W.; Tafuri, W. Diffuse Intralobular Liver 89. Fibrosis in Dogs Naturally Infected with Leishmania (Leishmania) chagasi. Am. J. Trop. Med. Hyg. 2008, 79, 198–204. [CrossRef] 92. Alves, A.F.; Pereira, R.A.; Andrade, H.M.; Mosser, D.M.; Tafuri, W.L. Immunohistochemical Study of Renal Fibropoiesis Associated with Dogs Naturally and Experimentally Infected with Two Different Strains of Leishmania (L.) Infantum. Int. J. Exp. Pathol. 2019, 100, 222–233. [CrossRef] 93. Caliari, M. V Princípios de Morfometria Digital: KS300 Para Iniciantes; Editora da UFMG: Belo Horizonte, Brazil, 1997. 94. Schwartz, C.; Oeser, K.; da Costa, C.P.; Layland, L.E.; Voehringer, D. T Cell–Derived IL-4/IL-13 Protects Mice against Fatal Schistosoma mansoni Infection Independently of Basophils. J. Immunol. 2014, 193, 3590–3599. [CrossRef] 95. Liu, K.; Zhao, E.; Ilyas, G.; Lalazar, G.; Lin, Y.; Haseeb, M.; Tanaka, K.E.; Czaja, M.J. Impaired Macrophage Autophagy Increases the Immune Response in Obese Mice by Promoting Proinﬂammatory Macrophage Polarization. Autophagy 2015, 11, 271–284. [CrossRef] 96. Wilson, M.S.; Elnekave, E.; Mentink-Kane, M.M.; Hodges, M.G.; Pesce, J.T.; Ramalingam, T.R.; Thompson, R.W.; Kamanaka, M.; Flavell, R.A.; Keane-Myers, A.; et al. IL-13Rα2 and IL-10 Coordinately Suppress Airway Inﬂammation, Airway-Hyperreactivity, and Fibrosis in Mice. J. Clin. Investig. 2007, 117, 2941–2951. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
10.3390_molecules25040938	Article Large-Scale Virtual Screening Against the MET Kinase Domain Identiﬁes a New Putative Inhibitor Type Emmanuel Bresso 1,† Flavio Maina 2 , Rosanna Dono 2 and Bernard Maigret 1,* , Alessandro Furlan 2,3,† , Philippe Noel 1, Vincent Leroux 1 , 1 Université de Lorraine, CNRS, Inria, LORIA, F-54000 Nancy, France; [email protected] (E.B.); [email protected] (P.N.); [email protected] (V.L.) 2 Aix Marseille Univ, CNRS, Developmental Biology Institute of Marseille (IBDM), UMR7288, Parc Scientiﬁque de Luminy, 13009 Marseille, France; [email protected] (A.F.); ﬂ[email protected] (F.M.); [email protected] (R.D.) 3 University of Lille, CNRS, UMR 8523, PhLAM-Physique des Lasers Atomes et Molécules, F-59000 Lille, France * Correspondence: [email protected] † These authors contributed equally to this work. Received: 27 January 2020; Accepted: 14 February 2020; Published: 19 February 2020 Abstract: By using an ensemble-docking strategy, we undertook a large-scale virtual screening campaign in order to identify new putative hits against the MET kinase target. Following a large molecular dynamics sampling of its conformational space, a set of 45 conformers of the kinase was retained as docking targets to take into account the ﬂexibility of the binding site moieties. Our screening funnel started from about 80,000 chemical compounds to be tested in silico for their potential afﬁnities towards the kinase binding site. The top 100 molecules selected—thanks to the molecular docking results—were further analyzed for their interactions, and 25 of the most promising ligands were tested for their ability to inhibit MET activity in cells. F0514-4011 compound was the most efﬁcient and impaired this scattering response to HGF (Hepatocyte Growth Factor) with an IC50 of 7.2 µM. Interestingly, careful docking analysis of this molecule with MET suggests a possible conformation halfway between classical type-I and type-II MET inhibitors, with an additional region of interaction. This compound could therefore be an innovative seed to be repositioned from its initial antiviral purpose towards the ﬁeld of MET inhibitors. Altogether, these results validate our ensemble docking strategy as a cost-effective functional method for drug development. Keywords: virtual screening; ensemble-docking; structure-based drug design; cross-docking validation; induced ﬁt; conformational selection; MET kinase 1. Introduction MET receptor is a multifunctional receptor tyrosine kinase that plays a pivotal role in human development and tumorigenesis. Upon binding of its ligand HGF (Hepatocyte Growth Factor), MET triggers several intracellular signaling cascades, including MAPK and PI3K pathways, leading to various cellular responses, such as survival, proliferation, and migration, among others. MET activation can be driven in cancer by several mechanisms: HGF or MET overexpression, and also mutations [1]. Aberrant activation of MET signaling does not only affect cancer development and progression, but it also contributes to resistance against other cancer drugs [2–11]. Consequently, MET represents a pharmaceutically relevant target in anticancer drug design and has been the focus of several anticancer strategies [12–18]. Pioneer MET inhibitors such as SU11274, PHA665752, or AM7 were helpful for Molecules 2020, 25, 938; doi:10.3390/molecules25040938 www.mdpi.com/journal/molecules molecules(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Molecules 2020, 25, 938 2 of 19 demonstrating the efﬁcacy of MET inhibition to impair tumor growth in preclinical models. Then, further developments in the ﬁeld led to the approval by the FDA of crizotinib and cabozantinib in the 2010s for treating non-small cell lung cancers and medullary thyroid cancers, respectively. Even though promising results have been reported, the therapeutic activity of these drugs is still relative and efforts are required to identify new MET inhibitors with physicochemical properties optimized for clinical efﬁciency [19,20]. Moreover, new alterations in MET sequence have been recently identiﬁed, such as MET exon 14 skipping in lung cancers and the emergence of MET mutations in the kinase domain following treatment with MET inhibitors [21]. Novel inhibitor structures may possibly target these mutations with increased efﬁciency. Designing new putative hits against MET therefore remains a valuable challenge to be tackled. In the present work, we carried out a virtual screening campaign in order to identify innovative compounds able to become new hits for further lead development. As MET plasticity upon ligand binding had been previously highlighted [22,23], we took into account this aspect for the molecular docking simulations. Indeed, MET can accommodate several distinct ligand binding modes and associated receptor conformations, a feature that is particular to the kinase family [24]. We reasoned that it should be taken into account for designing drugs with improved efﬁciency and selectivity proﬁles [25,26]. To be efﬁcient, the molecular docking engines embedded within the virtual screening approaches must be adapted to handle such ﬂexibility [27,28]. In the present work, we used the ensemble-docking strategy—previously recognized for its efﬁciency in drug design [29]—and show the beneﬁt of an investigation using a large ensemble-docking on MET. In previous medicinal chemistry works, we already identiﬁed novel MET inhibitors [30,31] and characterized the different MET ligand binding modes as shown by the stream of released X-ray data [32]. Here, we provide the results of a large-scale ensemble-docking investigation on MET, in which MET conformations are extended from available X-ray data to molecular dynamics and normal mode analysis. A limited number of candidate compounds were selected from the ensemble-docking results and one of them was subsequently validated experimentally as a possible new MET inhibitor, providing a valuable seed for further investigations. 2. Materials and Methods 2.1. Screened Chemical Libraries The choice of an appropriated set of compounds to explore the virtual screening space is critical for assuming a good rate of success [33]. Today, millions of compounds can be selected for high-throughput screening, and a suitable selection strategy must be designed. In our case and according to previous success stories [34–36], we chose to use a set of libraries selected in order: 1. 2. 3. 4. To use the highest possible chemical diversity, in order to cover a large spectrum of chemical structures [37–40]; To include kinase-targeted compounds, as such a choice is already proven to be successful [41,42]; To explore natural products, which are a promising source of anticancer drugs [43–45]; To take into account the repositioning of approved compounds, as drug repurposing presents an increasing interest in cancer research, by removing many costs associated with several steps of drug development [46–49]. Several proofs of concept are now available and a typical case of viral-to-cancer drug repositioning is gemcitabine with US patent No 4,808,614, aimed at treating viral infections, and the later-issued US No 5,464,826, which claims of its use to treat cancer. Therefore, we also considered chemical libraries dedicated to antiviral compounds. According to the criteria listed above concerning the choice of the chemical libraries, we ﬁrstly downloaded around 200,000 compounds from the chemical providers listed in Table 1. After eliminating redundancies in compounds and in scaffolds to assume a large chemical diversity and in respect of general druglike deﬁnitions [50], we ﬁnally retained about 80,000 compounds for our screening campaign. Molecules 2020, 25, 938 3 of 19 Table 1. List of the selected chemical libraries used in the present virtual screening campaign, providing a total of 76,251 compounds. Supplier Library Names French laboratories chimiotheque-nationale.enscm.fr Collaboration medchem.u-strasbg.fr ChemBridge www.chembridge.com Life Chemicals www.lifechemicals.com TimTec www.timtec.net Otava otavachemicals.com TargetMol www.targetmol.com Selected subsets (kinase, essential, etc.) laboratory collection kinase library diversity library kinase general library antiviral library 15K diversity library NDL + NPL natural derivatives library all kinase library 10K diversity library natural productlike library antivirus library natural compounds library 2.2. Selected MET Conformational Ensemble In ensemble-based docking calculations, a well-suited choice of the protein target conformational sample is required to reproduce the protein plasticity and the possible induced-ﬁt phenomena [51,52]. Concerning MET conformational ﬂexibility, our previous exploration by molecular dynamics and normal mode simulations [22] was limited to the 26 PDB structures available at that time (Table 2). We have now extended this analysis by considering all the X-ray structures available for the wild-type MET in the PDB [53] deposited after 2012. From the additional structures found, only 19 were considered in this work (Table 3) as we discarded those where three regions were missing in the X-ray structure and those where the number of missing residues in a single region was larger than 20. This selection aimed to improve our protein ensemble sample by covering most of the kinase structural characteristics, such as the position of the c-helix (in or out) or of the aspartate-phenylalanine-glycine (DFG) motif (in or out) as deﬁned in Kinametrix [54], thus covering most of the inhibitor type binding modes. 3D structures considered in this ensemble of 45 conformers looked representative of the diversity of MET structures, as shown from the dendrogram, the heat maps, and correspondence maps in Figure 1. These results obtained thanks to the Dali server [55] clearly illustrate how MET 3D structures used here are organized into several families covering most of the protein conformational space presently known. Molecules 2020, 25, 938 4 of 19 Table 2. List of the 26 PDB MET kinase domain structures selected in the previous work [22] and reused in the present one. The kinase conformation associated to each structure is annotated according to the KinaMetrix web resource [56]: DO means DFG-out, DI means DFG-in, CO means α-cHelix-out, CI means α-cHelix-in, and ωCD indicates DFG intermediate. No PDB ID Ligand PDB ID Deposition Date Annotation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 1R0P 2RFN 2RFS 3C1X 3CCN 3CD8 3CE3 3CTH 3CTJ 3F66 3F82 3EFJ 3EFK 2WD1 3DKC 3DKF 2WGJ 3A4P 3L8V 3I5N 3LQ8 2WKM 3Q6W 3QTI 3RHK 3ZXZ KSA AM7 AM8 CKK LKG L5G 1FN 319 320 IHX 353 MT3 MT4 ZZY ATP SX8 VGH DFQ L8V B2D 88Z PFY Q6W 3QT M97 KRW 2003 2007 2007 2008 2008 2008 2008 2008 2008 2008 2008 2008 2008 2009 2009 2009 2009 2010 2010 2010 2010 2010 2011 2011 2011 2011 Inactive CODI ? Inactive CODI Inactive CIDO Inactive CODI Inactive CODI Inactive CODO Inactive CIDO Inactive CIDO Inactive CODI Inactive CIDO Inactive CODO Inactive CODI Inactive CODI Inactive CODI Inactive CODI Active CIDI Inactive CODI Inactive CIDO Inactive CODI Inactive CODO Inactive CODI Active CIDI Inactive CODI Inactive ωCD Inactive CODI Table 3. List of the PDB MET kinase domain structures added to the ones coming from our previous work [22]. The kinase conformation associated to each structure is annotated according to the KinaMetrix web resource [56]: DO means DFG-out, DI means DFG-in, CO means α-cHelix-out, CI means α-cHelix-in, and ωCD indicates DFG intermediate. PDB ID Ligand ID Ligand IC50 (nM) Date Missing Sequence # Missing Residues Annotation 4DEI 4GG5 4EEV 3VW8 4IWD 3ZCL 3ZC5 3ZBX 4KNB 4MXC 4XYF 4R1V 4XMO 5DG5 5EYD 5EOB 5EYC 5UAF 5HTI 0JL 0J3 L1X DF6 1JC 5TF W9Z 6XE 1RU DWF 44X 3E8 46G 5B4 5T1 5QQ 5SZ 84P 66L 0.6–2 0.9 4.7/42 2 1 19 6 5 47/410 6.7 1/5 400 2 ? 1 0.24 3 ? ? 2012 2012 2012 2013 2013 2013 2013 2013 2013 2014 2015 2015 2015 2015 2016 2016 2016 2017 2017 1100–1103 1115–1117 1146–1151 1225–1243 1237–1242 1286–1290 1234–1235 1240–1243 1100–1102 1099–1102 1089–1102 1099–1103 1113–1115 1238–1242 1099–1103 1150–1151 1098–1103 - 1098–1103 1151–1152 1238–1240 1099–1103 1098–1105 1145–1152 1238–1242 7 6 19 11 6 3 4 14 8 5 5 2 6 - 8 3 5 16 5 Inactive CODI Inactive CODI Inactive CIDO ? Active CIDI Inactive CODI Inactive CODI Inactive CODI Inactive CODI Inactive CODO Inactive CODI Inactive CODI Inactive CODO Inactive CODI Inactive CODI Inactive CODI Active CIDI Inactive CODO No 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 To be in accordance with the 26 conformers coming from our previous work [22], the 19 added crystal structures were prepared and cleaned following the same protocol: missing residues, side Molecules 2020, 25, 938 5 of 19 chains, and hydrogens were added when necessary; unnecessary water molecules, ions, and additives were removed; basic and acidic side chains were ionized according to a pH set to 7. To consider possible binding sites ﬂuctuations, short molecular dynamics (MD) simulations were undertaken for each of these 19 structures. For that purpose, these structures were placed in a solvent box of 80 Å and counter ions were added for electrostatic neutrality. NAMD [57] molecular dynamics program was used with the same CHARMM36 force ﬁeld and same protocol as previously described [22]. After minimization and equilibrium steps (64,000 conjugate gradients and 1 ns MD, respectively), 10 ns of MD were recorded, with a frame length of 1 ps. These 19 MD trajectories were analyzed, and the most stable representative conformer was retained for each of them and added in the ensemble-docking set. Figure 1. (a) Dendrogram showing the relationships between the 45 PDB conformers listed in Tables 2 and 3 and used to sample MET structure plasticity. (b) Similarity heat-map showing the relationships between the 45 PDB conformers and used to sample MET structure plasticity. The color scale corresponds to the Dali Z-score values. (c) Correspondence analysis of the 45 ensemble PDB-related conformers. This plot positions data points with the most similar structural neighborhoods near each other according to a multidimensional scaling method. 2.3. Description of the Ensemble-Docking Protocol The ensemble-docking facility proposed in the GOLD docking program was used [58]. This GOLD feature evaluates different receptor conformations concurrently during the docking exploration. The protein ensemble used in this work thus contained 45 MET conformers (26 from our previous work and 19 added in this one). As these conformers must be superimposed before being used in GOLD ensemble-docking program, they were structurally aligned according to their conserved and most rigid secondary structure patterns, as previously described [22] and summarized in Table 4. Molecules 2020, 25, 938 6 of 19 Table 4. List of the secondary structure elements used for aligning all the conformers. Domain Secondary Structure Name Residues N-terminal C-terminal β1 β2 β2 β2 β2 αE αF αH αI 1076 to 1081 1092 to 1098 1104 to 1110 1144 to 1146 1154 to 1158 1178 to 1198 1263 to 1278 1310 to 1320 1330 to 1343 When docking an ensemble of conformations for a given protein, their binding sites must be deﬁned using a method that is not conformer speciﬁc. In the present ensemble-GOLD version, as it was not possible to deﬁne the active site by a list of atoms or residues, the only way was to use the centroid of the binding cavity and a sphere radius around this point. Therefore, for each of the 45 aligned protein conformers used here, protein cavities and their center of mass were detected by the LIGSITE program [59]. From these data, we obtained an average center point as the ensemble binding site deﬁnition for GOLD. Figure 2 presents the position of this average center point within the 45 protein conformers. A radius of 20 Å was associated to this average point to deﬁne the binding cavity of each conformer in order to correctly encompass the receptor for all the conformations in the ensemble, including conformational variations around the center. We also veriﬁed that the resulting sphere was encompassing all groups of residues previously identiﬁed as potential interaction areas for MET ligands [32]. For each docking run, we used 50 starting poses/molecule for the GOLD generic algorithm. Tested compounds were ranked by the standard goldscore scoring function. Figure 2. Position of the average center-point (as a green sphere) found from the 45 used conformers and used for the ensemble-dockings. 2.4. Computer Grid Facilities Due to the massive calculations needed ( 80,000 molecules × 48 protein ensemble conformers × 50 poses/molecule), and considering the computing time to achieve only one run, we used the Grid5000 facility [60] providing the required computer power in order to distribute the calculations using the PVM framework embedded in GOLD. A total of 1300 cpus (mostly Xeons) with 4 GB RAM/core and inﬁniband connections were used for each run. The docking performances run around 300-docked ligands/ensemble/hour. The calculations were spread on the clusters using the same strategy as previously described [61]. Molecules 2020, 25, 938 2.5. Scattering Assays 7 of 19 The experimental protocols for measuring the potency of MET inhibitors are detailed in previous publications [30,62]. MDCK cells were preincubated with compounds overnight at 0.1–100 µM concentrations at 37 ◦C in a humidiﬁed atmosphere of 5% CO2, followed by a 24 h stimulation with 20 ng/mL HGF (R&D Systems). Cells were further incubated at 37 ◦C in an atmosphere of 5% CO2 for 24–48 h, washed with phosphate buffered saline (PBS; Gibco BRL), and ﬁxed with 4% PFA (paraformaldehyde, Sigma). The quantiﬁcation of scattering response was performed by counting the number of cells with scattered morphology in 30 independent colonies. The IC50 corresponds to the concentration of compounds leading to a 50% inhibition of MET-triggered cell scattering. 3. Results 3.1. Preliminary Validation Concerning the GOLD Ensemble-Docking Protocol Used The coordinates of the 45 aligned conformers and of the sphere representing their common binding sites constituted our ensemble-docking protein reference. The ﬁrst question here concerned the accuracy of this binding site deﬁnition compared to ones that are more classical. For that, we compared the docking results for some of the selected 45 MET conformers using three binding site deﬁnitions; namely, a residue list, an existing ligand, and the center point of the binding cavity, respectively. For each individual docking target, the three deﬁnitions provided almost the same rank and docking score for the associated PDB ligand (Table 5). Moreover, the poses of this ligand found using the three binding site deﬁnitions were similar to the pose found in the crystal structures, as illustrated with the example of the AM7 ligand on Figure 3. Table 5. Comparison of the docking results using the 3 binding site deﬁnitions. Deﬁnition of the Binding Site Target PDB Name Ligand PDB Name Rank Number Score Value Center + radius 20 Å Residues list From its PDB ligand 3DKC 2RFN 3DKC 2RFN 3DKC 2RFN ATP AM7 ATP AM7 ATP AM7 1 1 1 1 1 1 105.5 100.8 102.8 98.0 107.1 106.6 Figure 3. Poses of the AM7 ligand in the X-ray 2RFN structure compared to the docking results. In black, the original pose of the ligand in its PDB protein conformation; in colors, the best docking poses obtained by GOLD on the 2RFN target using a deﬁnition of the binding site from a list of residues (orange), from the original ligand (green), and from a center-point (purple). The second question was related to the ability of the ensemble-docking process to retrieve a given PDB ligand to its PDB structure among the 45 ones. To evaluate that point, an ensemble-docking calculation was carried out on the 45 protein target conformers using a short chemical library built Molecules 2020, 25, 938 8 of 19 from their own 45 associated ligands (the list is given in Tables 2 and 3). We checked whether we could associate the right PDB target for a given PDB ligand (with possibly similar rank, score and pose compared to the ones found for the individual target dockings) in the protein ensemble. This was achieved for almost 80% of the compounds (Table S1). For example, the KSA ligand was able to preferentially retrieve its original 1R0P partner among the ensemble of the 45 PDB conformers of the protein target. From these results, it appeared that the ensemble-docking procedure we used was a satisfactory method to tackle multiple conformers docking and to achieve a valuable virtual screening. 3.2. Selection of Candidate Hits from the Virtual Screening Campaign Once the screening campaign was achieved for the 80,000 compounds ﬁltered from the chosen libraries, we kept the top-100 ranked compounds according to their GOLD scores (ranging from 100 to 114) for further analysis. We started the docking analysis with the Life Chemicals compound F0725-0356 giving the best docking score of 114. A comparison between the X-ray complex 3EFK/MT4 structure and the MD_3EFK/F025-0356 one presented quite similar poses and protein/ligand interactions. Indeed, the most important residues known in MET interactions (namely, Met1160, Asp1222, Tyr1159) were found in both complexes. We next analyzed the protein-ligand interactions for the other top-100 compounds in order to compare them to the ones found in the 45 original PDB structures (Tables 2 and 3). For that, we used the PLIP program [63] by focusing on two important interaction types: hydrogen bonds and π-stacking. Protein residues Met1160 (45/45), Asp1222 (34/45), and Lys1110 (6/45) concentrated the vast majority of hydrogen bonds with ligands; while Tyr1230 (25/45) and Phe1223 (7/45) dealt with most of the π-stacking. In order to limit our biological tests on possible promising compounds, we eliminated from the top-100 list the molecules not presenting at least one hydrogen bond and one π-stacking from the ones described above in the PDB complexes. After this ﬁlter, we retained only 41 compounds as satisfying these criteria. As most of these compounds came from the Life Chemical antiviral library and given the simplicity of comparing molecules from the same supplier, we decided to only test compounds from Life Chemicals. As some of these molecules were not available in stock from this provider, only the 25 compounds listed in Table 6 were ﬁnally kept to proceed further. Molecules 2020, 25, 938 9 of 19 Table 6. Ligands selected from the Life Chemical (LC) antiviral library and experimentally tested. ”-“: the compound (assessed at a concentration up to 100µM) did not affect MDCK cell scattering in response to HGF/SF. ”+“: the compound impaired MDCK cell scattering in response to HGF/SF with an IC50 > 10 µM. ”+++“: the compound impaired MDCK cell scattering in response to HGF/SF with an IC50 < 10 µM. Mol ID Life Chemicals Name GOLD Score Best Protein Conformer Biological Activity 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 F0725-0356 F0772-0607 F0816-0342 F0737-0405 F0737-0393 F0301-0263 F0721-0868 F0715-0299 F0539-1482 F0385-0029 F0385-0334 F0514-4011 F0174-0048 F1620-0074 F0011-0324 F0721-0906 F0012-0227 F0721-0911 F0715-0300 F2252-0240 F0772-2099 F0473-0261 F0721-0900 F0772-2147 F0526-0094 120.7 111.8 111.2 110.6 110.1 105.5 105.1 105.0 104.0 103.8 103.4 103.3 102.4 102.1 102.0 102.0 101.9 101.9 101.8 101.1 100.9 100.9 100.5 100.5 100.3 MD_3EFK MD_3EFK MD_3F82 MD_3EFJ MD_3EFK MD_3EFK MD_3EFK MD_3EFK MD_3EFJ MD_3EFK MD_3EFJ MD_3EFJ MD_3EFK MD_3EFJ MD_3EFK MD_3EFJ MD_3EFJ MD_3EFJ MD_2RFN MD_3EFK MD_3EFJ MD_3CE3 MD_3EFK MD_3EFJ MD_3EFJ - - - - - - - - + - - +++ - - - - - - - - - - - - - 3.3. F0539-1482 and F0514-4011 Inhibit MET-Induced Cell Scattering These 25 compounds were then experimentally tested for their ability to restrain MET-triggered biological activities. We previously efﬁciently screened compounds for their inhibitory properties towards MET-triggered biological responses by using cell scattering assays [31,64]. In particular, MDCK epithelial cells acquire a “scattered phenotype” after stimulation with MET ligand HGF. Among the 25 tested compounds, two compounds were found active, namely, F0539-1482 and F0514-4011. F0514-4011 was the most efﬁcient and impaired this scattering response to HGF with an IC50 of 7.2 µM (Figure 4). No toxic effects were observed at biologically active concentrations. This study thus demonstrates that our strategy actually allows the identiﬁcation of compounds able to inhibit MET-driven biological activities. Molecules 2020, 25, 938 10 of 19 Figure 4. (a) F0514-4011 impairs cell scattering in response to MET ligand HGF: MDCK epithelial cells were treated with 20 ng/mL HGF, with or without preincubation with F0514-4011 for 2 h. F0514-4011 IC50 is 7.2 µM. (b) Dose-response curve for F0514-4011. 3.4. Compared Docking of F0514-4011 Compound Versus Known Inhibitors In order to understand why the compound F0514-4011 (Figure 5) was the most potent compound among the 25 experimentally tested ones while not presenting the highest GOLD score, we compared its docking data with those of potent existing inhibitors. For that, we collected the structures of ligands found in the PDB related to MET kinase domain in complex with already marketed inhibitors with binding IC50 found in the nM range (Table 7). All these compounds were submitted to the ensemble-docking GOLD protocol already used for our virtual screening campaign. From these calculations, it appeared that the best docking scores ranged from 111 for merestinib (L1X ID in PDB 4EEV) to 83 for AMG337 (5T1 ID in PDB 5EYD), so that the score of 103 obtained for our active F0514-4011 compound was in this range of active compounds. Considering now IC50, one possibility to explain the higher IC50 of 7.2 µM obtained for F0514-4011 (compared to 0.4–14 µM range found for compounds listed in Table 7) could be its weaker solubility (cLogP of 5.7, greater than that of all compounds listed in Table 7). Table 7. Data used for some known marketed inhibitors with nano-molar range IC50 found in the PDB. PDB ID Ligand ID Name IC50 Solubility cLogP Docking Score 2RFS 2WGJ 2WKM 3DKF 3RHK 3LQ8 3Q6W 3QTI 3ZXZ 4EEV 5EYD AM8 VGH PFY SX8 M97 88Z Q6W 3QT KRW L1X 5T1 SU11274 Criotinib PHA-665752 SGX-523 Tivantinib Fortinib MK-2461 NVP-BVU972 PF-04217903 Merestinib AMG337 10 nM 11 nM 9 nM 4 nM 4 nM 0.4 nM 0.4 nM 14 nM 5 nM 5 nM 1 nM 2.9 3.5 4.4 1.4 3.1 4.3 3.3 1.6 0.2 3.4 0.3 86 82 88 84 83 99 93 84 87 111 83 Molecules 2020, 25, 938 11 of 19 Figure 5. F0514-4011: N-[[4-(4-Ethoxyphenyl)-5-[2-[3-(4-methylphenyl)-5-thiophen-2-yl-3,4-dihydropyrazol -2-yl]-2-oxoethyl]sulfanyl-1,2,4-triazol-3-yl]methyl]-2-phenylacetamide. Another point concerned the interaction of F0514-4011 with amino acid residues within the protein-binding region. In Table 8, we have listed the protein residues/ligand interactions found from Table 7 PDB complexes, as calculated by the PLIP program. These interactions were compared to the ones obtained for F0514-4011 from its best pose MD_3EFJ in the ensemble MET conformations. From this comparison, it appears that several of the most important amino acid residues found from the PDB protein/ligand analysis were also found for F0514-4011, at the exception of Met1160, common to all PDB structures of Table 8, replaced possibly by Met1131 and Met1229 in our case. This situation is mostly due to the conformation of the large DFG loop acting as a highly ﬂexible lid protecting the binding sites which was quite different in the MD_3EFJ conformation, found as the most suitable one to bind F0514-4011 when compared to the PDB ones (see Figure 6 for an example with the 5DG5 and 4DEI structures). Therefore, our docking results concerning the best pose proposed by GOLD for F0514-4011 appear quite in agreement with most of data obtained from all the PDB concerning MET kinase domain complexed with inhibitors. Molecules 2020, 25, 938 12 of 19 Table 8. List of the protein residues interacting with a nM. inhibitor from the PDB complexes of Table 7 ranked by their number of occurrence. In bold, the residues also found in the interactions with F0514-4011 with the MD-3EFJ MET conformation. According to the PLIP results, a residue was marked ”+“ when at least one protein-ligand interaction was found, whatever its quality (hydrophobic, H-bond, π-stacking, ionic, etc.) and marked by ”-“ when no protein-ligand interaction was found. Residue 4EEV 2WGJ 5EYD 3ZXZ 2RFS PDB IDs 3RHK 3QTI 3Q6W 2WKM 3DKF 3LQ8 MET1160 LEU1157 ASP1222 ALA1108 TYR1230 VAL1092 ILE1084 TYR1159 PRO1158 LEU1140 ALA1221 PHE1223 ASP1164 LYS1110 ASN1209 GLU1127 PHE1134 VAL1139 PHE1200 ARG1086 ARG1208 THR1343 GLU1347 PHE1089 ASP1231 ARG1166 ASN1167 ILE1130 ASN1171 + + + + - - + - - - - + - + - - + + + - - - - - - - - + - + + - - + + + - + - - - - - - - - - - - - - - - - - - - - + + + + + - + - - - - - - - - - - - - - - - - - - - - - - + - + + + + + - - - - - - - - - - - - - - - - - + + + - - + + + - + - - + + - - - - - - - - - - - - - - - - - - - + + - - - + + - + - - + - + - - - - - - - - - + - - - - - + + + + + + + - - - + - - - - - - - - - - + + - - - - - - + + - + - + + + - + + - - - - - - - - + + - - - - - - - - + + - - + - - + + + - - + - + - - - - - - - - - - - - - - + + + + + + + - - - + - + - - - - - - - - - - - - - - - - + + + + - - + + - + - + - - - + + + + - - - - - - - - - - Figure 6. Differences for the lid DFG loop between selected PDB structures and our MD-reﬁned MD_3EFJ conformations. The proteins are depicted from their Cα ribbon-like traces. To further characterize the F0514-4011 inhibitor type, we have considered the general 3D shape of known kinase inhibitors as analyzed in several papers [7,65–67]. Concerning MET, such compounds Molecules 2020, 25, 938 13 of 19 are generally classiﬁed as type-I or -II. Type-I ligands essentially bind at the ATP binding site and present a U-shaped conformation, with the protein in the DFG-in structure; while type II are found in an extended shape and correspond to the DFG-out protein form. We illustrate this in Figure 7, showing the conformations of two typical ligands, namely, type-I AMG337 (from PDB 5EYD) and the type-II altiratinin analog DP-4157 (from PDB 5DG5). From this picture, it appears that F0514-4011 presents both the U and linear shapes while also showing another region of interaction, including three of the protein residues already found in MET complexes—namely, Asp1222, Tyr1230, and Arg1208 (found only 2 times for 3C1X and 3YW8 among our 45 ensemble conformations). Asp1204 and Asn1209 residues, still not involved in MET complex PDB structures, complement this supplementary binding pocket. The thiophene moiety of F0514-4001 was placed central within this pocket by the thiophene-pyrazole group which also oriented the associated toluene ring to close the U-shape part. Therefore, one could postulate that F0514-4011 molecule describes a possibly novel type of inhibitor. Figure 7. Comparison of the conformations between F0514-4011, the U-shape inhibitor 5T1 (AMG337), and the linear-shape 5B4 (altiratinib), as observed in their respective binding sites. Nevertheless, considering the limitations of any docking program, the stability of F0514-4011’s best docking pose could be questioned. In order to validate it, we have performed a molecular dynamics simulation using the same conditions as those used for the PDB complexes (cubic water 3 ). The results show that the GOLD docking pose is very stable and still conserved after box of 80 Å 10 ns of MD (Figure 8). The protein/ligand interactions found for F0514-4011 after the MD simulation were similar to those discussed above, thus giving conﬁdence to the robustness of the docking results. Our ﬁnal question concerns the originality of F0514-4011 compared to the known MET ligands. The Tanimoto similarity index calculated between F0514-4011 and most of the published MET ligands shows that the molecule identiﬁed by our virtual screening campaign seems to be an innovative hit as all the Tanimoto values are low, ranging from 0.39 (with the pioneer inhibitor PHA-665752) to 0.12 (for norcantharidin) (Table S2). We have completed this quite elementary similarity search by using the ChemDes web server [68], which allows a large panel of similarity ﬁngerprint types as well as ﬁngerprints descriptors and similarity measures. Using this web server, we mined several databases collecting MET known inhibitors (such as the PDB or PubChem [69]), already in clinical trials (such as MDDR [70]), or described as putative inhibitors (such as in Life Chemical or sellekchem [71] providers). The results obtained with this method conﬁrmed the lack of similarity suggested with the Tanimoto distance. With the Sokal similarity method and DTRF ﬁngerprint types, the similarities ranged from 0.46 to 0.19 (in the PDB list, a maximum of 0.40 was obtained for compound ID 75H found in PDB ID 5T3Q (data not shown)).This could be due to the thiophene moiety of F0514-4011, as we have found only two papers and one patent in the literature referring to thiophene-related MET inhibitors [72–74] and only one reference to the role of thiophene-pyrazole moiety in kinase inhibition [75]. Molecules 2020, 25, 938 14 of 19 Figure 8. (a) Comparison between the initial docking pose (in orange) of F0514-4011 and the ﬁnal one after the 10-ns MD simulation (in cyan). (b) Evolution of F0514-4011 root mean square deviation (RMSD) during 11 ns (1 ns equilibrium and 10 ns production) of MD simulation. Poses were aligned on the initial one and the curve was smoothed. 4. Discussion Molecular docking, molecular dynamics, and virtual screening approaches can now be efﬁciently used for the design of new inhibitors of the MET kinase domain [27,56,76–80]. From all these approaches, new potent compounds were obtained and more highlights revealed about MET kinase domain conformational behavior. In this vein, our study merges both simulations and experiments and highlights a novel scaffold for MET inhibition. Using an ensemble-docking approach associated to short molecular dynamics runs in order to take into account the ﬂexibility of the used X-ray structures in the protein conformational ensemble, we were faced with the fundamental question of the relevance of this strategy for handling the difﬁcult problem of predicting ligand-binding modes on a ﬂexible target. This is especially true for MET kinase, the active site of which exhibits important structural variations, as observed in their available crystal structures [81,82]. We believe that this work brings a positive answer to this question and can constitute a working line for other simulations in the future. Ensemble-docking is now widely used, and incorporating this approach to short molecular dynamics simulations looks promising. Still, a couple of simple questions have to be answered prior to initiating the docking calculation: how do we generate a relevant ensemble for a given receptor [51], and how can we be sure that the possible energy differences obtained between conformations in the ensemble are properly accounted for? Interestingly, F0514-4011 compound (also referenced in PubChem with ID 5237313) is not a newcomer in drug design as it has been already tested as a possible activator of E3 ligase (FBW7) and inhibitor of microphthalmia-associated transcription factor (MITF), but was found inactive in both assays . Our study suggests that it could be repositioned for MET inhibition, as evidenced by its biological activity against MET-driven cell scattering. Some drug properties such as solubility and lack of toxicity were already known. With regard to its molecular weight of 650Da, which could be Molecules 2020, 25, 938 15 of 19 considered as limiting its possible therapeutic action, it should be noted that other inhibitors currently on the market have similar characteristics such as tarloxatinib (679Da), foretinib (632Da), or golvatinib (633Da). Therefore, it should not be a major hurdle if lead optimization provides us with a promising drug in terms of activity and/or selectivity. This will be the topic of future investigations. This virtual screening work presents F0414-4011 as a valuable compound that could be a seed for developing new and innovative leads against MET kinase. Its novelty and originality might be used to overcome the resistance problem found presently for several existing inhibitors. Supplementary Materials: The following are available online. Table S1: Comparison of the ensemble-docking results to the individual ones (a ligand against its own PDB-related structure), Table S2: most used c-Met inhibitors as pointed by SelleckChem and AdooQ Biosciences. Author Contributions: Conceptualization, A.F., V.L., and B.M.; software, E.B. and P.N.; validation, A.F., F.M., and R.D.; formal analysis, E.B., A.F., P.N., and B.M.; investigation, E.B, P.N., V.L., and B.M.; writing–original draft preparation, E.B., A.F., P.N., V.L., and B.M.; writing–review and editing, E.B., A.F., P.N., and B.M.; supervision, F.M., R.D., and B.M. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Acknowledgments: Experiments presented in this paper were carried out using the Grid’5000 testbed, supported by a scientiﬁc interest group hosted by Inria and including CNRS, RENATER, and several Universities as well as other organizations (see https://www.grid5000.fr). We are much grateful to all members of our labs for helpful discussion and advice. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations The following abbreviations are used in this manuscript: aspartate-phenylalanine-glycine DFG MD Molecular dynamics RMSD Root mean square deviation References 1. 2. 3. 4. 5. 6. Furlan, A.; Kherrouche, Z.; Montagne, R.; Copin, M.C.; Tulasne, D. Thirty Years of Research on Met Receptor to Move a Biomarker from Bench to Bedside. Cancer Res. 2014, 74, 6737–6744. [CrossRef] [PubMed] Zhang, H.; Feng, Q.; Chen, W.D.; Wang, Y.D. HGF/c-MET: A Promising Therapeutic Target in the Digestive System Cancers. Int. J. Mol. Sci. 2018, 19, 3295. [CrossRef] [PubMed] Rashed, W.M. C-MET as a Potential Target Therapy toward Personalized Therapy in Some Pediatric Tumors: An Overview. Crit. Rev. Oncol. 2018, 131, 7–15. [CrossRef] [PubMed] Bahrami, A.; Shahidsales, S.; Khazaei, M.; Ghayour-Mobarhan, M.; Maftouh, M.; Hassanian, S.M.; Avan, A. C-Met as a Potential Target for the Treatment of Gastrointestinal Cancer: Current Status and Future Perspectives. J. Cell. Physiol. 2017, 232, 2657–2673. [CrossRef] [PubMed] Refaat, T.; Donnelly, E.D.; Sachdev, S.; Parimi, V.; El Achy, S.; Dalal, P.; Farouk, M.; Berg, K.N.; Helenowksi, I.; Gross, J.P.; et al. C-Met Overexpression in Cervical Cancer, a Prognostic Factor and a Potential Molecular Therapeutic Target. Am. J. Clin. Oncol. 2017, 40, 590. [CrossRef] [PubMed] Trovato, M.; Campennì, A.; Giovinazzo, S.; Siracusa, M.; Ruggeri, R.M. Hepatocyte Growth Factor/C-Met Insights 2017, Axis in Thyroid Cancer: From Diagnostic Biomarker to Therapeutic Target. 12, 1177271917701126. [CrossRef] Biomark. 7. Wu, P.; Clausen, M.H.; Nielsen, T.E. Allosteric small-molecule kinase inhibitors. Pharmacol. Ther. 2015, 156, 59–68. [CrossRef] 8. Mo, H.N.; Liu, P. Targeting MET in Cancer Therapy. Chronic Dis. Transl. Med. 2017, 3, 148–153. [CrossRef] Li, C.; Wu, J.; Hynes, M.; Dosch, J.; Sarkar, B.; Welling, T.H.; di Magliano, M.P.; Simeone, D.M. C-Met Is 9. a Marker of Pancreatic Cancer Stem Cells and Therapeutic Target. Gastroenterology 2011, 141, 2218–2227. [CrossRef] Molecules 2020, 25, 938 16 of 19 10. 11. Sawada, K.; Radjabi, A.R.; Shinomiya, N.; Kistner, E.; Kenny, H.; Becker, A.R.; Turkyilmaz, M.A.; Salgia, R.; Yamada, S.D.; Woude, G.F.V.; et al. C-Met Overexpression Is a Prognostic Factor in Ovarian Cancer and an Effective Target for Inhibition of Peritoneal Dissemination and Invasion. Cancer Res. 2007, 67, 1670–1679. [CrossRef] Sierra, J.R.; Tsao, M.S. C-MET as a Potential Therapeutic Target and Biomarker in Cancer. Ther. Adv. Med. Oncol. 2011, 3, S21–S35. [CrossRef] [PubMed] 12. Cui, J.J. Targeting Receptor Tyrosine Kinase MET in Cancer: Small Molecule Inhibitors and Clinical Progress. J. Med. Chem. 2014, 57, 4427–4453. [CrossRef] [PubMed] 13. Parikh, P.K.; Ghate, M.D. Recent Advances in the Discovery of Small Molecule C-Met Kinase Inhibitors. Eur. J. Med. Chem. 2018, 143, 1103–1138. [CrossRef] [PubMed] 14. Pasquini, G.; Giaccone, G. C-MET Inhibitors for Advanced Non-Small Cell Lung Cancer. Expert Opin. Investig. Drugs 2018, 27, 363–375. [CrossRef] [PubMed] 15. Yuan, H.; Liu, Q.; Zhang, L.; Hu, S.; Chen, T.; Li, H.; Chen, Y.; Xu, Y.; Lu, T. Discovery, Optimization and Biological Evaluation for Novel c-Met Kinase Inhibitors. Eur. J. Med. Chem. 2018, 143, 491–502. [CrossRef] [PubMed] 16. Zhu, K.; Kong, X.; Zhao, D.; Liang, Z.; Luo, C. C-MET Kinase Inhibitors: A Patent Review (2011–2013). Expert Opin. Ther. Patents 2014, 24, 217–230. [CrossRef] [PubMed] 17. Lv, P.C.; Wang, Z.C.; Zhu, H.L. Recent Advances in the Design and Synthesis of C-Met Inhibitors as 18. Anticancer Agents (2014-Present). Curr. Med. Chem. 2017, 24, 57–64. [CrossRef] Sun, Z.G.; Yang, Y.A.; Zhang, Z.G.; Zhu, H.L. Optimization techniques for novel c-Met kinase inhibitors. Expert Opin. Drug Discov. 2019, 14, 59–69. [CrossRef] 19. Hughes, V.S.; Siemann, D.W. Have Clinical Trials Properly Assessed C-Met Inhibitors? Trends Cancer 2018, 4, 94–97. [CrossRef] 20. Miranda, O.; Farooqui, M.; Siegfried, J. Status of Agents Targeting the HGF/c-Met Axis in Lung Cancer. Cancers 2018, 10, 280. [CrossRef] 21. Cortot, A.B.; Kherrouche, Z.; Descarpentries, C.; Wislez, M.; Baldacci, S.; Furlan, A.; Tulasne, D. Exon 14 deleted MET receptor as a new biomarker and target in cancers. JNCI J. Natl. Cancer Inst. 2017, 109, djw262. [CrossRef] [PubMed] 22. Asses, Y.; Venkatraman, V.; Leroux, V.; Ritchie, D.W.; Maigret, B. Exploring C-Met Kinase Flexibility by Sampling and Clustering Its Conformational Space. Proteins Struct. Funct. Bioinform. 2012, 80, 1227–1238. [CrossRef] [PubMed] 23. Dussault, I.; Bellon, S.F. C-Met Inhibitors with Different Binding Modes: Two Is Better than One. Cell Cycle 24. 2008, 7, 1157–1160. [CrossRef] [PubMed] Jacobs, M.D.; Caron, P.R.; Hare, B.J. Classifying Protein Kinase Structures Guides Use of Ligand-Selectivity Proteins Struct. Proﬁles to Predict Inactive Conformations: Structure of Lck/Imatinib Complex. Funct. Bioinform. 2008, 70, 1451–1460. [CrossRef] 25. Druker, B.J.; Talpaz, M.; Resta, D.J.; Peng, B.; Buchdunger, E.; Ford, J.M.; Lydon, N.B.; Kantarjian, H.; Capdeville, R.; Ohno-Jones, S.; et al. Efﬁcacy and Safety of a Speciﬁc Inhibitor of the BCR-ABL Tyrosine Kinase in Chronic Myeloid Leukemia. N. Engl. J. Med. 2001, 344, 1031–1037. [CrossRef] 26. Capdeville, R.; Buchdunger, E.; Zimmermann, J.; Matter, A. Glivec (STI571, Imatinib), a Rationally Developed, Targeted Anticancer Drug. Nat. Rev. Drug Discov. 2002, 1, 493. [CrossRef] 27. Aliebrahimi, S.; Kouhsari, S.M.; Ostad, S.N.; Arab, S.S.; Karami, L. Identiﬁcation of Phytochemicals Targeting C-Met Kinase Domain Using Consensus Docking and Molecular Dynamics Simulation Studies. Cell Biochem. Biophys. 2018, 76, 135–145. [CrossRef] 28. Li, M.J.; Wu, G.Z.; Kaas, Q.; Jiang, T.; Yu, R.L. Development of Efﬁcient Docking Strategies and Structure-Activity Relationship Study of the c-Met Type II Inhibitors. J. Mol. Graph. Model. 2017, 75, 241–249. [CrossRef] 29. Amaro, R.E.; Baudry, J.; Chodera, J.; Demir, O.; McCammon, J.A.; Miao, Y.; Smith, J.C. Ensemble Docking in Drug Discovery. Biophys. J. 2018, 114, 2271–2278. [CrossRef] 30. Patané, S.; Pietrancosta, N.; Hassani, H.; Leroux, V.; Maigret, B.; Kraus, J.L.; Dono, R.; Maina, F. A New Met Inhibitory-Scaffold Identiﬁed by a Focused Forward Chemical Biological Screen. Biochem. Biophys. Res. Commun. 2008, 375, 184–189. [CrossRef] Molecules 2020, 25, 938 17 of 19 31. Furlan, A.; Colombo, F.; Kover, A.; Amat, M.; Asses, Y.; et al. Benzothiazol-2-Ylphenyl Moiety as Inhibitors of Tumorigenesis by Oncogenic Met Signaling. Eur. Med. Chem. 2012, 47, 239–254. [CrossRef] [PubMed] Issaly, N.; Tintori, C.; Angeli, L.; Leroux, V.; Letard, S.; Identiﬁcation of New Aminoacid Amides Containing the Imidazo [2, 1-b] J. 32. Asses, Y.; Leroux, V.; Tairi-Kellou, S.; Dono, R.; Maina, F.; Maigret, B. Analysis of C-Met Kinase Domain Complexes: A New Speciﬁc Catalytic Site Receptor Model for Deﬁning Binding Modes of ATP-Competitive Ligands. Chem. Biol. Drug Des. 2009, 74, 560–570. [CrossRef] [PubMed] 33. Gimeno, A.; Ojeda-Montes, M.; Tomás-Hernández, S.; Cereto-Massagué, A.; Beltrán-Debón, R.; Mulero, M.; Pujadas, G.; Garcia-Vallvé, S. The Light and Dark Sides of Virtual Screening: What Is There to Know? Int. J. Mol. Sci. 2019, 20, 1375. [CrossRef] [PubMed] 34. Kioshima, E.S.; Shinobu-Mesquita, C.S.; Abadio, A.K.R.; Felipe, M.S.S.; Svidzinski, T.I.E.; Maigret, B. Selection of potential anti-adhesion drugs by in silico approaches targeted to ALS3 from Candida albicans. Biotechnol. Lett. 2019, 41, 1391–1401. [CrossRef] 35. Bresso, E.; Fernandez, D.; Amora, D.X.; Noel, P.; Petitot, A.S.; de Sa, M.E.L.; Albuquerque, E.V.S.; Danchin, E.G.J.; Maigret, B.; Martins, N.F. A Chemosensory GPCR as a Potential Target to Control the Root-Knot Nematode Meloidogyne incognita Parasitism in Plants. Molecules 2019, 24, 3798. [CrossRef] 36. Rodrigues-Vendramini, F.A.V.; Faria, D.R.; Arita, G.S.; Capoci, I.R.G.; Sakita, K.M.; Caparroz-Assef, S.M.; Becker, T.C.A.; de Souza Bonﬁm-Mendonça, P.; Felipe, M.S.; Svidzinski, T.I.E.; et al. Antifungal activity of two oxadiazole compounds for the paracoccidioidomycosis treatment. PLoS Negl. Trop. Dis. 2019, 13, e0007441. [CrossRef] 37. Gilad, Y.; Nadassy, K.; Senderowitz, H. A Reliable Computational Workﬂow for the Selection of Optimal Screening Libraries. J. Cheminform. 2015, 7, 61. [CrossRef] 38. Petrone, P.M.; Wassermann, A.M.; Lounkine, E.; Kutchukian, P.; Simms, B.; Jenkins, J.; Selzer, P.; Glick, M. Biodiversity of Small Molecules–a New Perspective in Screening Set Selection. Drug Discov. Today 2013, 18, 674–680. [CrossRef] 39. Huggins, D.J.; Venkitaraman, A.R.; Spring, D.R. Rational Methods for the Selection of Diverse Screening Compounds. ACS Chem. Biol. 2011, 6, 208–217. [CrossRef] 40. Ma, C.; Lazo, J.S.; Xie, X.Q. Compound Acquisition and Prioritization Algorithm for Constructing Structurally Diverse Compound Libraries. ACS Comb. Sci. 2011, 13, 223–231. [CrossRef] 41. Xi, H.; Lunney, E.A. The Design, Annotation, and Application of a Kinase-Targeted Library. In Chemical Library Design; Humana Press: Totowa, NJ, USA, 2011; pp. 279–291. 42. Deanda, F.; Stewart, E.L.; Reno, M.J.; Drewry, D.H. Kinase-Targeted Library Design through the Application of the PharmPrint Methodology. J. Chem. Inf. Model. 2008, 48, 2395–2403. [CrossRef] [PubMed] 43. Dutta, S.; Mahalanobish, S.; Saha, S.; Ghosh, S.; Sil, P.C. Natural Products: An Upcoming Therapeutic Approach to Cancer. Food Chem. Toxicol. 2019, 128, 240–255. [CrossRef] [PubMed] 44. Henkin, J.M.; Ren, Y.; Soejarto, D.D.; Kinghorn, A.D. The Search for Anticancer Agents from Tropical Plants. In Progress in the Chemistry of Organic Natural Products 107; Springer: New York, NY, USA, 2018; pp. 1–94. 45. Mondal, S.; Bandyopadhyay, S.; K Ghosh, M.; Mukhopadhyay, S.; Roy, S.; Mandal, C. Natural Products: (Former Curr. Anti-Cancer Agents Med. Chem. Promising Resources for Cancer Drug Discovery. Med.-Chem.-Anti-Cancer Agents) 2012, 12, 49–75. [CrossRef] 46. Nowak-Sliwinska, P.; Scapozza, L.; i Altaba, A.R. Drug Repurposing in Oncology: Compounds, Pathways, Phenotypes and Computational Approaches for Colorectal Cancer. Biochim. Biophys. Acta (BBA) Rev. Cancer 2019, 1871, 434–454. [CrossRef] [PubMed] 47. Cheng, F. In Silico Oncology Drug Repositioning and Polypharmacology. In Cancer Bioinformatics; Springer: New York, NY, USA, 2019; pp. 243–261. 48. Abdelaleem, M.; Ezzat, H.; Osama, M.; Megahed, A.; Alaa, W.; Gaber, A.; Shafei, A.; Refaat, A. Prospects for Repurposing CNS Drugs for Cancer Treatment. Oncol. Rev. 2019, 13, 411. [CrossRef] [PubMed] 49. Yadav, V.; Talwar, P. Repositioning of Fluoroquinolones from Antibiotic to Anti-Cancer Agents: An Underestimated Truth. Biomed. Pharmacother. 2019, 111, 934–946. [CrossRef] 50. Lipinski, C.A. Lead-and Drug-like Compounds: The Rule-of-Five Revolution. Drug Discov. Today Technol. 2004, 1, 337–341. [CrossRef] Molecules 2020, 25, 938 18 of 19 51. Evangelista Falcon, W.; Ellingson, S.R.; Smith, J.C.; Baudry, J. Ensemble Docking in Drug Discovery: How Many Protein Conﬁgurations from Molecular Dynamics Simulations are Needed To Reproduce Known Ligand Binding? J. Phys. Chem. B 2019, 123, 5189–5195. [CrossRef] 52. Motta, S.; Bonati, L. Modeling Binding with Large Conformational Changes: Key Points in Ensemble-Docking Approaches. J. Chem. Inf. Model. 2017, 57, 1563–1578. [CrossRef] 53. Berman, H.M. The Protein Data Bank. Nucleic Acids Res. 2000, 28, 235–242. [CrossRef] 54. Rahman, R.; Ung, P.M.U.; Schlessinger, A. KinaMetrix: A web resource to investigate kinase conformations and inhibitor space. Nucleic Acids Res. 2019, 47, D361–D366. [CrossRef] [PubMed] 55. Holm, L.; Laakso, L.M. Dali server update. Nucleic Acids Res. 2016, 44, W351–W355. [CrossRef] [PubMed] Ibrahim, H.S.; Albakri, M.E.; Mahmoud, W.R.; Allam, H.A.; Reda, A.M.; Abdel-Aziz, H.A. Synthesis and 56. Biological Evaluation of Some Novel Thiobenzimidazole Derivatives as Anti-Renal Cancer Agents through Inhibition of c-MET Kinase. Bioorg. Chem. 2019, 85, 337–348. [CrossRef] [PubMed] 57. Phillips, J.C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, E.; Villa, E.; Chipot, C.; Skeel, R.D.; Kalé, L.; 58. Schulten, K. Scalable Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781–1802. [CrossRef] Jones, G.; Willett, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Development and validation of a genetic algorithm for ﬂexible docking 1 1Edited by F. E. Cohen. J. Mol. Biol. 1997, 267, 727–748. [CrossRef] 59. Hendlich, M.; Rippmann, F.; Barnickel, G. LIGSITE: Automatic and Efﬁcient Detection of Potential Small Molecule-Binding Sites in Proteins. J. Mol. Graph. Model. 1997, 15, 359–363. [CrossRef] 60. Bolze, R.; Cappello, F.; Caron, E.; Daydé, M.; Desprez, F.; Jeannot, E.; Jégou, Y.; Lanteri, S.; Leduc, J.; Melab, N.; et al. Grid’5000: A Large Scale and Highly Reconﬁgurable Experimental Grid Testbed. Int. J. High Perform. Comput. Appl. 2006, 20, 481–494. [CrossRef] 62. 61. Ghemtio, L.; Jeannot, E.; Maigret, B. Efﬁciency of a Hierarchical Protocol for High Throughput Structure-Based Virtual Screening on GRID5000 Cluster Grid. Open Access Bioinform. 2010, 2, 41–53. Furlan, A.; Roux, B.; Lamballe, F.; Conti, F.; Issaly, N.; Daian, F.; Guillemot, J.F.; Richelme, S.; Contensin, M.; Bosch, J.; et al. Combined Drug Action of 2-Phenylimidazo [2, 1-b] Benzothiazole Derivatives on Cancer Cells According to Their Oncogenic Molecular Signatures. PLoS ONE 2012, 7, e46738. [CrossRef] Salentin, S.; Schreiber, S.; Haupt, V.J.; Adasme, M.F.; Schroeder, M. PLIP: Fully Automated Protein–Ligand Interaction Proﬁler. Nucleic Acids Res. 2015, 43, W443–W447. [CrossRef] 63. 64. Colombo, F.; Tintori, C.; Furlan, A.; Borrelli, S.; Christodoulou, M.S.; Dono, R.; Maina, F.; Botta, M.; Amat, M.; Bosch, J.; et al. ‘Click’ synthesis of a triazole-based inhibitor of Met functions in cancer cells. Bioorg. Med. Chem. Lett. 2012, 22, 4693–4696. [CrossRef] [PubMed] 65. Roskoski, R. Classiﬁcation of small molecule protein kinase inhibitors based upon the structures of their drug-enzyme complexes. Pharmacol. Res. 2016, 103, 26–48. [CrossRef] [PubMed] 66. Zhao, Z.; Xie, L.; Bourne, P.E. Structural Insights into Characterizing Binding Sites in Epidermal Growth Factor Receptor Kinase Mutants. J. Chem. Inf. Model. 2019, 59, 453–462. [CrossRef] [PubMed] 67. Panicker, R.C.; Chattopadhaya, S.; Coyne, A.G.; Srinivasan, R. Allosteric Small-Molecule Serine/Threonine Kinase Inhibitors. In Protein Allostery in Drug Discovery; Springer: Singapore, 2019; Volume 1163, pp. 253–278. 68. Dong, J.; Cao, D.S.; Miao, H.Y.; Liu, S.; Deng, B.C.; Yun, Y.H.; Wang, N.N.; Lu, A.P.; Zeng, W.B.; Chen, A.F. ChemDes: An integrated web-based platform for molecular descriptor and ﬁngerprint computation. J. Cheminform. 2015, 7, 60. [CrossRef] [PubMed] 69. Kim, S.; Thiessen, P.A.; Bolton, E.E.; Chen, J.; Fu, G.; Gindulyte, A.; Han, L.; He, J.; He, S.; 2016, Shoemaker, B.A.; et al. PubChem Substance and Compound databases. 44, D1202–D1213. [CrossRef] [PubMed] Nucleic Acids Res. 70. MDDR. Available online: www.3dsbiovia.com/products/collaborative-science/databases/bioactivity- databases/mddr.html (accessed on 18 February 2020). Selleckchem.Com-Inhibitor Expert. Available online: www.selleckchem.com (accessed on 18 February 2020). 71. 72. Mohareb, R.M.; Hilmy, K.M.; Elshehawy, Y.A. Discovery of new thiophene, pyrazole, isoxazole derivatives as antitumor, c-Met, tyrosine kinase and Pim-1 kinase inhibitors. Bull. Chem. Soc. Ethiop. 2018, 32, 285. [CrossRef] Shi, L.; Wu, T.T.; Wang, Z.; Xue, J.Y.; Xu, Y.G. Discovery of quinazolin-4-amines bearing benzimidazole fragments as dual inhibitors of c-Met and VEGFR-2. Bioorg. Med. Chem. 2014, 22, 4735–4744. [CrossRef] Fancelli, D.; Pevarello, P.; Varasi, M. Thiophene Derivatives Active as Kinase Inhibitors, Process for Their Preparation and Pharmaceutical Compositions Comprising Them. WO2001EP06763, 14 June 2001. 73. 74. Molecules 2020, 25, 938 19 of 19 75. Zhan, W.; Che, J.; Xu, L.; Wu, Y.; Hu, X.; Zhou, Y.; Cheng, G.; Hu, Y.; Dong, X.; Li, J. Discovery of pyrazole-thiophene derivatives as highly Potent, orally active Akt inhibitors. Eur. J. Med. Chem. 2019, 180, 72–85. [CrossRef] 76. Zhang, S.; Song, Q.; Wang, X.; Wei, Z.; Yu, R.; Wang, X.; Jiang, T. Virtual Screening Guided Design, Synthesis and Bioactivity Study of Benzisoselenazolones (BISAs) on Inhibition of c-Met and Its Downstream Signalling Pathways. Int. J. Mol. Sci. 2019, 20, 2489. [CrossRef] 77. Balasubramanian, P.K.; Balupuri, A.; Bhujbal, S.P.; Cho, S.J. 3D-QSAR-Aided Design of Potent c-Met Inhibitors Using Molecular Dynamics Simulation and Binding Free Energy Calculation. J. Biomol. Struct. Dyn. 2019, 37, 2165–2178. [CrossRef] 78. Zhang, Q.W.; Ye, Z.D.; Shen, C.; Tie, H.X.; Wang, L.; Shi, L. Synthesis of Novel 6, 7-Dimethoxy-4-Anilinoquinolines as Potent c-Met Inhibitors. J. Enzym. Inhib. Med. Chem. 2019, 34, 124–133. [CrossRef] [PubMed] Singh, P.K.; Silakari, O. Molecular Dynamics Guided Development of Indole Based Dual Inhibitors of EGFR (T790M) and c-MET. Bioorg. Chem. 2018, 79, 163–170. [CrossRef] [PubMed] 79. 80. Yan, M.; Wang, H.; Wang, Q.; Zhang, Z.; Zhang, C. Allosteric inhibition of c-Met kinase in sub-microsecond molecular dynamics simulations induced by its inhibitor, tivantinib. Phys. Chem. Chem. Phys. 2016, 18, 10367–10374. [CrossRef] [PubMed] 81. Modi, V.; Dunbrack, R.L. Deﬁning a new nomenclature for the structures of active and inactive kinases. Proc. Natl. Acad. Sci. USA 2019, 116, 6818–6827. [CrossRef] [PubMed] 82. Mashayekh, K.; Shariﬁ, S.; Damghani, T.; Elyasi, M.; Avestan, M.S.; Pirhadi, S. Clustering and Sampling of the c-Met Conformational Space: A Computational Drug Discovery Study. Comb. Chem. High Throughput Screen. 2020, 22, 635–648. [CrossRef] c(cid:13) 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
10.3390_biom11050662	Article Investigation of Mitochondrial Adaptations to Modulation of Carbohydrate Supply during Adipogenesis of 3T3-L1 Cells by Targeted 1H-NMR Spectroscopy Manon Delcourt 1,2,, Virginie Delsinne 2, Jean-Marie Colet 2 , Anne-Emilie Declèves 1 and Vanessa Tagliatti 2 1 Metabolic and Molecular Biochemistry Unit, Faculty of Medicine and Pharmacy, Research Institute for Health Sciences and Technology, UMONS, 20 Place du Parc, 7000 Mons, Belgium; [email protected] 2 Human Biology and Toxicology Unit, Faculty of Medicine and Pharmacy, Research Institute for Health Sciences and Technology, UMONS, 20 Place du Parc, 7000 Mons, Belgium; [email protected] (V.D.); [email protected] (J.-M.C.); [email protected] (V.T.) Correspondence: [email protected]; Tel.: +32-(0)65-373506 Abstract: (1) Background: White adipose tissue (WAT) is a dynamic and plastic tissue showing high sensitivity to carbohydrate supply. In such a context, the WAT may accordingly modulate its mitochondrial metabolic activity. We previously demonstrated that a partial replacement of glucose by galactose in a culture medium of 3T3-L1 cells leads to a poorer adipogenic yield and improved global mitochondrial health. In the present study, we investigate key mitochondrial metabolic actors reﬂecting mitochondrial adaptation in response to different carbohydrate supplies. (2) Methods: The metabolome of 3T3-L1 cells was investigated during the differentiation process using different glucose/galactose ratios and by a targeted approach using 1H-NMR (Proton nuclear magnetic resonance) spectroscopy; (3) Results: Our ﬁndings indicate a reduction of adipogenic and metabolic overload markers under the low glucose/galactose condition. In addition, a remodeling of the mitochondrial function triggers the secretion of metabolites with signaling and systemic energetical homeostasis functions. Finally, this study also sheds light on a new way to consider the mitochondrial metabolic function by considering noncarbohydrates related pathways reﬂecting both healthier cellular and mitochondrial adaptation mechanisms; (4) Conclusions: Different carbohydrates supplies induce deep mitochondrial metabolic and function adaptations leading to overall adipocytes function and proﬁle remodeling during the adipogenesis. Keywords: carbohydrates; mitochondria; adaptation; metabolism; metabolomics; adipogenesis 1. Introduction Today, white adipose tissue (WAT) is no longer considered a simple storage container for excessive nutrients. As other organs showing extraordinary plastic features, WAT exhibits cellular mechanisms of adaptation essential to maintain the overall systemic energy and metabolic homeostasis. Through a mechanistic approach, it is accepted now that WAT dynamics are challenged by several environmental features in both physiological and pathological contexts, such as obesity and metabolic disorders. Those adaptations lead to cellular changes known as WAT remodeling, including processes leading to the expansion of the tissue storage capacity by inducing an increase of adipocytes number. In this context, adipogenesis, the mechanism through which precursors cells differentiate into mature and functional adipocytes, has been extensively studied [1,2]. Adipogenesis is highly sensitive to the cell metabolic status and ﬁnely regulated by master gene regulators [3–5]. The 3T3-L1 cell line has been extensively used to study adipogenesis and is now well characterized in terms of transcriptional and proteomic events occurring through the differentiation process [6,7]. The same is not true for metabolic changes in differentiating adipocytes. To Citation: Delcourt, M.; Delsinne, V.; Colet, J.-M.; Declèves, A.-E.; Tagliatti, V. Investigation of Mitochondrial Adaptations to Modulation of Carbohydrate Supply during Adipogenesis of 3T3-L1 Cells by Targeted 1H-NMR Spectroscopy. Biomolecules 2021, 11, 662. https:// doi.org/10.3390/biom11050662 Academic Editor: Ladislav Andˇera Received: 22 March 2021 Accepted: 27 April 2021 Published: 29 April 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Biomolecules 2021, 11, 662. https://doi.org/10.3390/biom11050662 https://www.mdpi.com/journal/biomolecules biomolecules(cid:1)(cid:2)(cid:3)(cid:1)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)Biomolecules 2021, 11, 662 2 of 18 orchestrate this complex metabolic landscape remodeling, mitochondria play a central role [8,9]. Indeed, as crucial metabolic sensors, mitochondria are found at the crossroad of several biosynthetic pathways, especially focused on ATP production [10,11]. In addition, mitochondria are increasingly considered signaling hubs [10,12–14]. To better understand the inﬂuence of carbohydrate supply, both in terms of quantity and quality, on WAT plasticity, previous studies focused on studying how nutrients could modulate adipogenesis associated with WAT expansion [15–17]. In this context, we previ- ously reported that the partial replacement of glucose by galactose in the culture medium of 3T3-L1 cells had an anti-adipogenic effect and globally improved their mitochondrial health [18]. This study also underlined important changes regarding mitochondrial and metabolic enzyme expression, suggesting a deep impact on the mitochondrial landscape in differentiating cells. Based on these ﬁrst ﬁndings, we intend now to investigate the mito- chondrial metabolic function changes occurring in differentiating 3T3-L1 exposed to either LG-GAL ((Low glucose (5 mM glucose) + galactose (20 mM galactose)) or HG (25 mM glucose) by using targeted 1H-NMR (proton nuclear magnetic resonance) spectroscopy of cellular ﬂuids (intracellular extract and extracellular media). Hence, selected mitochondrial metabolic markers were relatively scored from their spectral intensities to perform the statistical comparison of their abundances in both carbohydrate conditions all along the adipogenesis process. 2. Materials and Methods 2.1. Cell Culture and Differentiation The murine 3T3-L1 cell-line was purchased from the American Type Culture Collec- tion (ATCC) and subcultured in growth media consisting of Dulbecco’s modiﬁed Eagle’s medium (DMEM) HG (25 mM glucose, Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS) premium (Pan-Biotech, Aidenbach, Germany). At passage 7 (P7), cells were cultured until reaching 100% conﬂuence and then let for two additional days before starting the differentiation (D0). At D0, cells were exposed for 3 days to a differentiation medium (DM) consisting of culture medium + 10% FBS + an adi- pogenic cocktail (insulin 1 mg/mL (Sigma-Aldrich, St. Louis, MO, USA), IBMX 12 mg/mL (Sigma-Aldrich, St. Louis, MO, USA) and DEX 0.4 mg/mL (Sigma-Aldrich, St. Louis, MO, USA)). The culture medium was different depending on the condition: DMEM HG (25 mM glucose, Sigma-Aldrich St. Louis, MO, USA) was used for the HG condition and a DMEM LG (5 mM glucose, Sigma-Aldrich, St. Louis, MO, USA) + GAL (20 mM galactose, Sigma-Aldrich, St. Louis, MO, USA) for the LG-GAL condition. At D3, the DM was replaced by a maintenance medium (MM) containing the same components as DM except for IBMX and DEX. MM was replaced every two days until D7 that corresponds to mature adipocytes. Cells were harvested on D0, D3, D5 and D7 to perform intracellular metabolites extraction steps. Medium of cells was sampled for extracellular metabolites levels assessment before being changed or renewed. 2.2. Extracellular (EC) and Intracellular (IC) Metabolic Extracts Preparation and 1H-NMR Spectra Acquirement EC and IC metabolites levels were determined on D0, D3 and D7 of differentiating 3T3- L1 cells cultured either in LG-GAL or HG medium. EC metabolites levels were assessed on culture media, and IC metabolites levels were measured after chloroform–methanol– water extraction separating the cell content into two main phases (one hydrophilic and one hydrophobic). Brieﬂy, cells were washed twice with warmed DPBS, quenched with cold water:methanol solution (1:4) and successively vortexed and sonicated following the addition of the different solvents (methanol, water and chloroform; (ﬁnal ratio: 1:0.5:0.5)). Following this mechano-chemical extraction phase, the extract solution was centrifuged 10 min at 14,000× g to obtain the different phases. The upper phase (methanol–water phase) containing hydrophilic metabolites was collected, evaporated with a speed vacuum, and resuspended using 700 µL phosphate buffer (80:20 D2O), pH 7.4. Extracellular media Biomolecules 2021, 11, 662 3 of 18 samples (500 µL) were mixed with a phosphate-buffered (80:20 D2O) solution (250 µL). IC and EC samples were centrifuged at 10,000× g for 10 min. The supernatant was collected (650 µL), mixed with trimethylsilylpropionic acid (TSP) at a ﬁnal concentration of 1 mM and transferred into 5 mm diameter NMR tubes. 1H-NMR spectra of the intracellular aqueous phase and extracellular ﬂuids were acquired using a Bruker Avance 600 MHz spectrometer with a 5 mm PABBO BB-probe. A NOESYPRESAT-1D sequence was used with 256 scans. The acquired FIDs were Fourier-transformed to obtain spectra. The program Mestre Nova 11 (Mestrelab Research, Santiago de Compostela, Spain) was used for further spectral processing (baseline, reference and phase corrections). 2.3. Analytical Steps of the IC and EC Metabolic Contents through a Two-Step 1H-NMR-Based Metabonomic Study This study included two successive analytical steps: (1) preliminary exploratory untargeted metabonomic study and (2) selection and semiquantitative assessment of key metabolites. 3. Untargeted Exploratory Approach 3.1. Multivariate Date Analyses The ﬁrst step of this metabonomic study was based on multivariate data analyses of the NMR spectral data set in terms of observations (biological samples) and variables (metabolites) discrimination between the two carbohydrate conditions (HG vs. LG-GAL). First, NMR spectra were handled using Mestre Nova 11 software. The spectral area from 0.08 to 10 ppm was subdivided into subregions of 0.04 ppm width, and each spectral subregion (descriptor) was integrated. After removal of the water (from 4.2 to 5.1 ppm) and TSP resonances (from 0.08 to 0.8 ppm), each descriptor integral was normalized to the total spectral area. Finally, the numerical values were gathered in an Excel table and exported to SIMCA P + 12 (Umetrics®) software to perform multivariate data analysis (statistical projection of the data set). On those projections (centered method), the two main components (1 and 2) were used to evaluate the most discriminant descriptors and, after identiﬁcation based on their chemical shift and multiplicity, the corresponding metabolites (as shown on the loadings plot in Figure 2A mostly affected in 3T3-L1 cells exposed to LG-GAL medium (in red on the scores plot on Figure 2A) versus HG medium (in black on the scores plot on Figure 2A). Those changes were followed all along the adipogenesis (D0, D3 and D5). 3.2. Identiﬁcation of Discriminant Variables (Metabolites) Metabolites of interest (corresponding to the discriminant descriptors) were identiﬁed from several databases, including the Human Metabolome Database (HMDB), as well as by using the Chenomx Proﬁler software 8.3 (Chenomx, Edmonton, AB, Canada) (Figure 2B). This part was performed without selecting speciﬁc markers, and all the metabolites were considered by performing non-guided analyses of the overall data. Finally, key intracellular metabolic markers (Figure 2C) were listed and used in the second part of this study consisting of selecting the most relevant mitochondrial metabolic markers. 3.3. Relative Levels of the Mitochondrial Metabolic Markers’ Assessment The metabolites selected in the ﬁrst step were submitted to a relative quantitation based on the AUC (Area Under the Curve) measurement of their best-resolved resonance. For example, glucose was assessed from its doublet resonance at 5.25 ppm. Then, the data were expressed as a percentage of the corresponding value obtained on Day 0 under the HG condition. For instance, in the case of glucose, the mean AUC of the doublet resonance measured on D3 in LG-GAL condition was divided by its mean AUC measured on D0 in HG condition. Such a relative assessment of the metabolic levels allowed us to statistically express some metabolic differences between the two conditions. Biomolecules 2021, 11, 662 4 of 18 Figure 1. Cont. Biomolecules 2021, 11, 662 5 of 18 Figure 2. Preliminary and exploratory metabonomic study: mitochondrial markers selection. (A) Preliminary multivariate data analyses: (a) Scores scatterplot (PLS-DA model) of the intracellular metabolites of 3T3-L1 cells cultured either with LG-GAL or HG media and harvested at the different time-points (D0, D3 and D5). R2 cum = 0.525; Hotelling’s T2 = 0.95; p-value (CV-ANOVA) = 0.0223. (b) Loadings scatterplot (PLS-DA model) of the intracellular metabolites of 3T3-L1 cells cultured either with LG-GAL or HG media and harvested at the different time-points (D0, D3 and D5). (B) 1 H-NMR intracellular metabolome of 3T3-L1 cells (identiﬁed hydrophilic metabolites). (C) Identiﬁed polar intracellular metabolites based on the previous multivariate analyses of the dataset and on identifying the detected peaks on (A): Preliminary and exploratory metabonomic study—mitochondrial markers the 1H-NMR spectra of the polar intracellular extract through Chenomx tool®. Selected metabolites are in bold and underlined. Statistical (* p< 0.05; p < 0.01; * p < 0.001; ** p < 0.0001) comparison between LG-GAL vs. HG were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison. cum = 0.871; Q2 For a particular metabolite of interest, the relative expression of its levels was based on the following formulation (Equation (1)) Mean AUC o f best resolved resonance at (D0 or D3 or D7) in (LG − GAL or HG) Mean AUC o f best resolved resonance at D0 in HG condition (1) Relative metabolites of interest levels calculation The AUC integration was performed using a “peak picking tool” available in Mestre Nova 11 software (Mestrelab Research, Santiago de Compostela, Spain). This (Table 1) semiquantitative method was validated by assessing the relative glucose levels in our LG (5 mM glucose) vs. HG media (25 mM glucose). A ratio of ca 1:5 was calculated, as expected from the glucose levels in LG versus HG conditions. Biomolecules 2021, 11, 662 6 of 18 Table 1. Validation of the relative metabolite levels assessment. Integrated AUC values were determined using the peak-picking tool, and relative values were determined based on the previously explained relative levels formulation. Total AUC of Doublet at 5.25 ppm % (Relative to the Mean of the Glucose AUC in HG Media) Glucose EC levels in HG media Glucose EC levels LG media 4.87 0.97 100% 19.9% 3.4. Selection of Mitochondrial Metabolic Markers Once the relative levels of the metabolites were determined, metabolites (underlined and in bold in Figure 2C) were considered as statistically relevant and somehow related to the mitochondrial metabolism (catabolism vs. anabolism, anaplerotic reactions, oxidative vs. anaerobic metabolism) were selected. Those selected intracellular markers were also checked and semiquantitatively assessed in the extracellular media of differentiating 3T3-L1 cells exposed to either HG or LG-GAL. 4. Targeted Approach After selecting mitochondrial metabolites of interest, a targeted 1H-NMR-based metabonomic study of the mitochondrial metabolome was performed. Based on histograms and statistical analyses of the relative metabolite levels, a semiquantitative investigation of the different related mitochondrial metabolic pathways was performed. Different aspects of the mitochondrial metabolome were assessed through selecting those markers and are detailed in the discussion. Statistical Analyses Data from at least three independent experiments (n = 3, biological triplicate) were analyzed using two-way ANOVA and Holm–Sidak’s multiple-comparison test. Statistical analyses were performed using GraphPad Prism 6 software. Results are presented as mean values ± SEM. The level for statistical signiﬁcance was deﬁned as p < 0.05. 5. Results 5.1. Selection of Mitochondrial Metabolic Markers Carbohydrates Supply Intake and Consumption Exposing differentiating 3T3-L1 cells to either 25 mM glucose or to 5 mM glucose + 20 mM galactose was expected to give different EC (Extracellular) and IC (Intracellu- lar) metabolic proﬁles from cells harvested at the different time-points and from the two compared conditions [18]. As expected, EC and IC glucose levels (Figure 3A,B) were consid- erably lower in the LG-GAL condition, whose exclusive EC and IC galactose (Figure 3C,D) availability showed a different carbohydrate proﬁle. Regarding IC galactose in differentiat- ing 3T3-L1cells, part of it was converted into galactitol. Galactose conversion into glycolytic subproduct (Glucose-1-phosphate further isomerized into glucose-6-phosphate) occurring in the liver is represented using a dotted arrow in Figure 3. In the WAT, the other pathways using galactose as a substrate were not detected and are also represented by a dotted arrow ending by a question mark. 5.2. Carbohydrate Metabolism-Anaerobic Glycolysis Evaluating both glucose and lactate as indicators of anaerobic glycolysis was per- formed to better investigate the downstream catabolic changes associated with glucose and partial galactose replacement (Figure 4). In the LG-GAL condition, the decrease of glucose in both the EC and IC media (Figure 3) indicates the progressive restriction in glucose availability and increased glucose transport and consumption over adipogenesis (D0-D7). In addition, inducing adipogenesis in lower glucose conditions (LG-GAL), compared to HG conditions, led to signiﬁcantly lower IC and EC lactate levels (D7 *), indicating lower production and secretion of this anaerobic glycolysis rate marker. Biomolecules 2021, 11, 662 7 of 18 Figure 3. Evolution of IC (Intracellular) and EC (Extracellular) levels of key carbohydrates through adipogenesis of 3T3-L1 cells in HG vs. LG-Gal conditions. Dotted arrows mean uncer- tain/hypothetical galactose conversions. (A) Relative glucose EC levels. (B) Relative glucose IC levels. (C) Relative galactose EC levels. (D) Relative galactose and galactitol IC levels. Results are the means ± SEM (biological (n = 3) triplicates). Statistical (n.s, p < 0.05; p < 0.01; * p < 0.001; **** p < 0.0001) comparison of values obtained at each key time points and compared to D0 (in black) and between HG vs. LG-GAL (in gray) were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison test. Figure 4. Key anaerobic glycolysis-related metabolites IC (Intracellular) and EC (Extracellular) levels evolution through adipogenesis in both HG vs. LG-Gal conditions scheme. Each graph individually represents either the relative IC or the EC metabolite levels evolution in 3T3-L1 cells. (A) Relative lactate EC levels. (B) Relative lactate IC levels. Results are the means ± SEM (biological (n = 3) triplicates). Statistical (n.s, * p < 0.05; p < 0.01; * p < 0.001; ** p < 0.0001) comparison of each key time points to the D0 situation (in black) and between HG vs. LG-GAL (in gray) were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison test. 5.3. Carbohydrate and Related Mitochondrial Catabolism—TCA (TriCarboxylic Acid) Cycle To continue the carbohydrate metabolism assessment of 3T3-L1 cells differentiating in either HG or LG-GAL conditions, a closer look at the mitochondrial metabolism was performed (Figure 5). Closely linked to their function of the “powerhouse” of the cells, Biomolecules 2021, 11, 662 8 of 18 the mitochondrial TCA (TriCarboxylic Acid) cycle is associated with both carbohydrate catabolism and cellular energy production [19,20]. Several key markers of the TCA cycle were semiquantitatively assessed, giving an overview of the TCA cycle activity in both conditions. Three main TCA cycle markers were relatively quantified in the IC media of 3T3-L1, and two of them were also detected and evaluated in the EC media. Our results pinpoint lower levels of the selected TCA markers in the intracellular fraction with statistically significant differences reached at D7 for all of them (pyruvate (), succinate () and fumarate (*)) in LG-GAL condition (Figure 5). Those highly significant differences were correlated to two opposite trends observed in TCA cycle activity through adipogenesis depending on the carbohydrate supply. Whereas in HG condition, TCA markers levels become significantly higher at D7, levels of those metabolites constantly decrease over the adipogenesis in LG-GAL condition. Pyruvate, initially present at similar levels in both HG and LG DMEM culture media (Figure 5A), appears to be slightly lower in LG-GAL condition than in HG condition. In this last condition, EC pyruvate levels were kept constant over adipogenesis. On the other hand, in the LG-GAL condition, EC media displayed lower levels reflecting lower IC pyruvate levels (Figure 5C) due to lower consumption and/or secretion. As for succinate (Figure 5B), whereas its EC levels tended to increase in HG condition, those levels decreased in LG-GAL condition through adipogenesis. Absent from both HG and LG DMEM culture media, succinate levels modulation could only be attributed to altered secretion. Figure 5. Key oxidative carbohydrates catabolism and TCA cycle-related metabolites IC and EC levels evolution through adipogenesis in both HG vs. LG-Gal conditions scheme. Each graph individually represents either the relative IC or the EC metabolite levels evolution in 3T3-L1 cells. (A) Relative pyruvate EC levels. (B) Relative succinate EC levels. (C) Relative pyruvate IC levels. (D) Relative fumarate IC levels. (E) Relative succinate IC levels. Results are the means ± SEM (biological (n = 3) triplicates). Statistical (n.s, p < 0.05 ; p < 0.01 ; p < 0.0001) comparison of each key time points to the D0 situation (in black) and between HG vs. LG-GAL (in gray) were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison test. 5.4. Mitochondrial Metabolism—Non-Glycolytic “Glutamine–Glutamate–Pyroglutamate” Group Besides mitochondrial metabolism related to carbohydrate combustion, other metabo- lites could be assessed to evaluate the mitochondrial metabolic function status (Figure 6). The glutamine–glutamate–pyroglutamate group can be considered as non-glycolytic since there are not directly coming from a glycolytic source. In addition, the glutamine oxidation into glutamate further converted into alpha-ketoglutarate is a decisive reaction to sustain Biomolecules 2021, 11, 662 9 of 18 TCA cycle activity [21]. For this reason, glutamine and glutamate were considered, in our study, as two markers of mitochondrial metabolic function. Thanks to our previous 1H-NMR metabolomics study [18], pyroglutamate was retained as a third informative metabolite. In- deed, less is known about pyroglutamate in mammalian cells if it is only its production from glutamate cyclization [22]. Although different trends were observed inside this glutamine– glutamate–pyroglutamate group, the combined findings could be used to partially reflect the mitochondrial metabolic status by considering glutamate as a central hub. Indeed, IC glutamate levels (Figure 6C) significantly () and similarly decreased in both conditions through adipogenesis (D3 and D7). However, EC glutamate levels did not follow the same trend, with a significant increase in the HG condition and a late significant decrease at D7 in the LG-GAL condition (Figure 6D). This could indicate an increase in glutamate consumption and an associated decrease in its extracellular release in the LG-GAL condition. On the opposite, an increase in glutamate release was observed in HG-condition. The changes observed in IC and EC glutamine levels (Figure 6A,B) clearly indicate a significant increase occurring during the late maturation phase of the adipogenesis (D7 ) in HG condition, whereas those levels look significantly lower in LG-GAL condition (). Finally, higher IC and EC pyroglutamate levels (Figure 6E,F) were noticed in LG-GAL (***) condition, most likely due to greater production and release during the early adipogenic phase (D3). During the late adipogenic phase, IC pyroglutamate levels were clearly decreased in LG-GAL conditions and associated with signif- icantly high and constant EC pyroglutamate levels. This could be correlated to a reduction of glutamate into pyroglutamate conversion and/or an increase of its release. Figure 6. The glutamine, glutamate and pyroglutamate group IC and EC levels evolution through adipogenesis in both HG vs. LG-Gal conditions scheme. Each graph individually represents either the relative IC or the EC metabolite levels evolution in 3T3-L1 cells. (A) Relative glutamine IC levels. (B) Relative glutamine EC levels. (C) Relative glutamate IC levels. (D) Relative glutamate EC levels. (E) Relative pyroglutamate IC levels. (F) Relative pyroglutamate EC levels. Results are the means ± SEM (biological (n = 3) triplicates). Statistical (n.s, p < 0.05; p < 0.01; * p < 0.001; ** p < 0.0001) comparison of each key time points to the D0 situation (in black) and between HG vs. LG-GAL (in gray) were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison test. Biomolecules 2021, 11, 662 10 of 18 5.5. Mitochondrial Metabolism Markers—Acetate and β-hydroxybutyrate To go deeply into the different pathways that could inﬂuence mitochondrial metabolism, we decided to look closer at acetate and the β-hydroxybutyrate, two poorly assessed metabolites in the WAT (Figure 7). Both are known to be closely related to the Acetyl-CoA metabolism. As shown in Figure 7, the early adipogenic phase is associated with a high and signiﬁcant increase of IC and EC acetate (Figure 7A,B) in both conditions (*). However, the increase of IC acetate was signiﬁcantly greater in LG-GAL condition than in HG one. However, at the end of the late maturation phase, IC and EC acetate levels signiﬁcantly decreased in both conditions and particularly in the HG condition (). At D7, EC levels followed a similar decreasing trend in both conditions. IC β-hydroxybutyrate levels sig- niﬁcantly decreased in both conditions after the early adipogenic phase (D0 vs. D3; ). However, such an observation was not correlated to changes in its levels in the EC media at the same time. β-hydroxybutyrate, like acetate, was not initially present in our culture media, and changes in its EC levels (Figure 7C) could only be correlated to a different secretion. During the late maturation phase, two opposite trends were observed depending on the condition. Whereas β-hydroxybutyrate IC levels (Figure 7D) continued to decrease in HG condition, signiﬁcantly greater levels were observed in LG-GAL condition (). In the EC media, a similar signiﬁcant increase of β-hydroxybutyrate levels was observed in both HG and LG-GAL conditions (**). Figure 7. Acetate-β-hydroxybutyrate IC and EC levels evolution through adipogenesis in both HG vs. LG-Gal conditions scheme. Each graph individually represents either the relative IC or the EC metabolite levels evolution in 3T3-L1 cells. (A) Relative acetate EC levels. (B) Relative acetate IC levels. (C) Relative β-hydroxybutyrate EC levels. (D) Relative β-hydroxybutyrate IC levels. Results are the means ± SEM (biological (n = 3) triplicates). Statistical (n.s, p < 0.05; p < 0.01; * p < 0.001; **** p < 0.0001) comparison of each key time points to the D0 situation (in black) and between HG vs. LG-GAL (in gray) were performed by two-way ANOVA and Holm–Sidak’s multiple-comparison test. Biomolecules 2021, 11, 662 11 of 18 6. Discussion After evidencing mitochondrial dynamics and functional changes in our previous studies [18], we wanted to look deeper at the IC and EC mitochondrial metabolome changes linked to carbohydrate supply changes in an adipogenic context. Such an investigation aims to characterize the mitochondrial functional changes by evaluating different aspects of this organelle metabolic function (anabolism vs. catabolism, oxidative vs. anaerobic metabolism and anaplerotic pathways) and thanks to a targeted 1H-NMR-based metabonomic study of the mitochondrial metabolome. Previously to this targeted approach, an exploratory assessment of the 3T3-L1 intracellular metabolome evolution during adipogenesis was performed in both conditions and by using multivariate analytical tools (Figure 2). This preliminary step was essential to determine the most relevant and biologically signiﬁcant markers of the mitochondrial metabolic function to be addressed in this targeted metabo- nomic study (Figure 2C). Therefore, this targeted study was structured into different parts assessing metabolites’ relative levels compared to carbohydrate conditions and linking them to a marker function reﬂecting a speciﬁc part of the mitochondrial metabolic function. First, we evaluated the ﬂuctuations in IC and EC levels of carbohydrates during adipogenesis when 3T3-L1 cells were exposed to either 5 mM glucose + 20 mM galactose) (LG-GAL) or 25 mM glucose (HG). Unsurprisingly, differentiating 3T3-L1 cells in LG-GAL condition had lower EC and IC glucose levels associated with important galactose intake related to this carbohydrate EC availability (Figure 2). According to our results, in LG- GAL condition, galactose appears to be converted into galactitol, an end-product of a biochemical pathway not associated with energy production. In other words, as expected, adipocytes are not enzymatically equipped to convert galactose back to glucose. However, such an important galactose transport and metabolism indicate that 3T3-L1 cells are highly active in terms of “carbohydrates catabolism” when the carbohydrates supply is high. Next, ﬁguring out such greed for carbohydrates, we investigated different aspects of the metabolism directly linked to carbohydrates intake modulation. During the hyperglycemia period, WAT avidly consumes glucose [23,24] and, for stor- age purposes, converts it into a lipid form, triacylglycerols [25]. Also, WAT produces high amounts of lactate [26,27], indicating a pro-glycolytic proﬁle that known to be exacerbated in an obesity context [28]. Our ﬁndings conﬁrm that differentiating 3T3-L1 has a higher glycolytic rate in HG condition as displayed by higher glucose consumption and lactate production and secretion (Figure 3). Conversely, reducing glucose availability by a partial replacement with galactose led to lower lactate production, indicating a poorer anaerobic glycolysis stimulation to maintain cell metabolism. Prior studies focused on anaerobic glycolysis when studying 3T3-L1 differentiation [29,30]. A study evidenced increasing IC lactate levels in differentiating 3T3-L1 as a management strategy to deal with an excess of glucose [31]. In addition to this high glycolytic rate, it has been proven in vitro (in 3T3-L1) that lactate secretion is high in HG conditions. Transposing to the in vivo situation, this would mimic the liver conversion of lactate to promote neoglucogenesis, essential to face potential further fasting periods [32]. Such an LG-GAL condition, if strictly considered as a glucose restriction situation under the in vivo glycolytic point of view, does not stimulate excess glucose management strategy and slow down lactate release usually secreted by the WAT under high glycemia period. Interestingly, those observations highlight that although being an in vitro model, 3T3-L1 cells demonstrate endocrine-like features by keeping organ crosstalk strategies through metabolites secretion modulation in different metabolic con- texts. In addition, our ﬁndings emphasize previous observations of differentiating 3T3-L1 adipocytes as obligatory anaerobic glycolytic cells. Such an observation is comforted by recent reports suggesting a signaling role of lactate [33–36] and evidencing it as mandatory actors to promote 3T3-L1 differentiation [30,37]. In such a context, we propose lactate as a marker of high glucose-associated metabolic stress conditions promoting adipocytes differentiation in WAT expansion contexts. After evidencing such importance of the anaerobic glycolytic metabolism in differ- entiating and mature adipocytes, we focused on the mitochondrial metabolic status of Biomolecules 2021, 11, 662 12 of 18 differentiating 3T3-L1. Previously, we demonstrated an improvement in the mitochondrial network health status in 3T3-L1 differentiating in LG-GAL condition compared to HG one [18]. As an overall key feature, mitochondria were less stressed under this low glucose condition, and, according to the prior discussion point, this could be correlated to a change in the mitochondrial metabolism associated with less deleterious nutritional conditions. Here, we intended to speciﬁcally evaluate selected metabolic markers reﬂecting different aspects of the mitochondrial function using the 1H-NMR metabonomic approach. TCA cycle appeared as the ﬁrst key mitochondrial metabolic pathway we wanted to have a closer look at (Figure 5) and mainly due to its potential direct relation with the anaerobic glycolysis rate. In addition, assessing this oxidative catabolic pathway is a common feature of most mitochondrial function assessment methods, such as MTT and cellular oximetry tests. Pyruvate, by being the end-product of aerobic glycolysis and the key precursors of acetyl-CoA further feeding the TCA cycle, appears to be an excellent marker of oxidative carbohydrate catabolism. In turn, succinate and fumarate, by playing the reversible role of substrate and product of the reaction catalyzed by the succinate dehydrogenase (TCA cycle enzyme and complex II of the mitochondrial respiratory chain), were more considered as speciﬁc markers of the oxidative mitochondrial catabolism. It is well-known that oxidative mitochondrial metabolism gets stimulated when reaching an advanced adipogenic maturation phase in nearly mature adipocytes [38].. However, it is important to underline that those observations were done in HG conditions and could be a consequence of such “excess glucose” conditions. To this point, our LG-GAL condition, by reducing the glycolytic rate and potentially modulating the mitochondrial metabolism, could give rise to new metabolic considerations. It suggests that in LG-GAL condition, the oxidative catabolism is broadly reduced as evidenced by pyruvate, succinate and fumarate IC lower levels. It is also important to note that some TCA cycle intermediates are known to play signaling roles, which can be accordingly tuned. Recent evidence of succination, a post-translational modiﬁcation linked to fumarate levels, seems to link this mechanism stimulation to glucotoxicity in the WAT tissue, a phenomenon increasing associated with obesity development and linked mitochondrial dysfunction onset [39–42]. In addition, It is increasingly acknowledged that under stressful metabolic circumstances, mitochondria tend to release TCA intermediates whose levels can considerably increase in the extrami- tochondrial compartments [43]. For this purpose, pyruvate EC levels were interestingly lower in LG-GAL conditions than in HG ones. Such an observation could be correlated to higher consumption of this substrate and a lower leaking of it from the intracellular to the extracellular compartment. Remarkably, it appears that IC and EC succinate levels, a substrate not initially present in the DMEM media we used, were considerably higher in HG media after cell harvesting. For this reason, one can reasonably propose that in LG-GAL condition, mitochondrial metabolites tend to stay in the intramitochondrial compartment, reﬂecting a poorer release of TCA intermediates, markers of mitochondrial stress onset. In other words, the high stimulation of the oxidative catabolism seems to be linked to a glucose overload, consequently stimulating the mitochondrial catabolic function as a protective mechanism justifying our previous observation of more catabolically stressed mitochondria in HG condition. When considering the mitochondrial metabolism, the metabolism related to the glutamine–glutamate group represents another pathway to keep an eye on (Figure 6). Over the past decades, numerous studies evaluated that ﬂuctuations in glutamine–glutamate levels could reﬂect relevant changes in mitochondrial function [44]. Interestingly, the obser- vation of very similar glutamate levels, clearly decreasing over time in both LG-GAL and HG conditions, underlines a decisive management mechanism of this metabolite during the differentiation process. However, obvious discrepancies in the evolutions of glutamine and pyroglutamate levels clearly indicate different mitochondrial metabolic proﬁles in both conditions. In such a context, the signiﬁcant increase of both IC and EC glutamine levels in the HG conditions, associated with the parallel decrease of glutamate, suggests stimulation of the glutamate conversion to glutamine. This could reﬂect a mitochondrial overload due Biomolecules 2021, 11, 662 13 of 18 to the overstimulation by the carbohydrate catabolism, consequently stimulating glutamate reduction into glutamine. Preventively, such a reaction could prevent further mitochondrial feeding through alpha-ketoglutarate production and alleviate additional mitochondrial catabolic function requirements. The signiﬁcant increase of the glutamine EC levels sug- gests an extracellular release mechanism avoiding intracellular glutamine excesses. In agreement with this, a very recent publication highlighted that WAT glutamine levels could be inversely correlated to adiposity and could inhibit glycolysis [45]. Our observations of higher EC glutamine EC release during the late maturation phase of the adipogenesis pro- cess could be explained by such a signaling role of this metabolite. Finally, the signiﬁcant increase of glutamate release and the glutamate conversion into pyroglutamate during the late maturation phase strengthens the hypothesis of a progressive mitochondrial overload and glutamate disposal mechanisms setup. In the LG-GAL condition, the quite different glutamate levels management seems to be more correlated to glutamate into pyroglutamate conversion mechanisms. Less is known about pyroglutamate in mammalian cells, except that its production comes from glutamate cyclization [22]. However, our results show that this metabolite is an additional marker helping to better assess glutamate levels evolution, itself inﬂuenced by the mitochondrial metabolic function. We fascinatingly found that 3T3-L1 cells display signiﬁcantly higher IC pyroglutamate levels in the early adipogenic phase and higher EC pyroglutamate levels across adipogenesis in the LG-GAL condition. Recently, a correlation between poorer visceral adiposity and higher pyroglutamate lev- els in the WAT of physically trained mice was reported [46]. It clearly appears that the progressive glutamate decrease in LG-GAL condition is ﬁrst due to its conversion into pyroglutamate, an early anti-adipogenic marker. Over time, pyroglutamate is less pro- duced but is greatly released in the extracellular compartment of mature adipocytes. As for other mitochondrial metabolites, this could underline an extracellular signaling function of pyroglutamate to extend anti-adipogenic power, consequently affecting adipogenesis of still undifferentiated peripheric cells in LG-GAL condition. Through all those obser- vations, we clearly identify different “glutamine–pyroglutamate group” proﬁles in both conditions, mostly reﬂecting the prevention role of LG-GAL media against mitochondrial metabolic overload and the associated reduction of the adipogenic yield in the 3T3-L1 cell line. On the other side, the mitochondrial overload occurring in HG seems to prevent anaplerotic mitochondrial pathway triggering by avoiding glutamate levels rising through its reduction into glutamine. To expect reaching the most comprehensive overview of the mitochondrial metabolism evaluation, we also included some markers reﬂecting the acetyl-CoA metabolism. As previously noted, acetyl-CoA, a mitochondrial metabolite, is at the crossroad of several biochemical pathways, and its levels can deeply inﬂuence key metabolic and cellular pathways [47,48]. Commonly known to be produced from pyruvate glycolytic supply, acetyl-CoA can also be linked to acetate. Some authors already associated acetate levels with mitochondrial metabolism, but also the nutrient availability in several cell types [13,49]. Interestingly, we found that acetate levels are very sensitive to the differentiation induction as evidenced by the signiﬁcant IC and EC increase of its levels in 3T3-L1 in both conditions during the early adipogenic phase. Such an observation could be correlated to the deep signaling network remodeling associated with the adipogenesis setup and requiring, at some key point, acetylation phenomenon. The similar acetate EC increase in both con- ditions, reﬂecting an important metabolite release, could reﬂect this correlation between differentiation induction and an acetate peak. However, the sharper IC acetate increase in LG-GAL conditions during the early adipogenic phase gave rise to questions about acetate sources. Recently, some authors advanced the hypothesis of a de novo production of acetate in case of environmental metabolic changes. Under certain circumstances, such as mitochondrial poorer function, de novo acetate production seems to be stimulated [49]. Assuming that the mitochondrial network is still not fully functional during the early adipogenic phase and that mitochondrial catabolism stimulation can be increased through glucose overload, acetate production from pyruvate can be a possible explanation. At this Biomolecules 2021, 11, 662 14 of 18 stage of differentiation, 3T3-L1 cells in LG-GAL condition display a poorer stimulation of the mitochondrial network maturation reducing the requirement of acetate conversion into acetyl-CoA. Acetate, through a secretion mechanism, is also expected to play an ex- tracellular role, as observed by higher EC acetate levels in the late maturation phase of adipogenesis in LG-GAL conditions. In this condition, acetate production and secretion in- crease reﬂect lower acetyl-CoA-dependent mitochondrial function associated with a poorer acetyl-CoA-dependent lipogenic process, a phenomenon reﬂecting an anti-adipogenic effect. Moreover, to a greater extent, high acetate secretion could also reﬂect the deep cellular and metabolic changes associated with conditions that are not associated with a nu- tritional excess, informing by this way, the peripheral tissues of this environmental change. Acetate released by the WAT seems to play a signaling function and could also constitute a metabolizable fuel for peripheric tissue. Interestingly, β-hydroxybutyrate, a ketone body known to be produced by the liver to fuel peripheric tissues when blood glucose levels are reduced [50], also appears to be highly produced and secreted by mature adipocytes in LG-GAL condition. In other words, 3T3-L1 clearly seems to reduce mitochondrial acetyl-CoA metabolism in favor of β-hydroxybutyrate production, therefore, reducing the mitochondrial carbohydrates catabolism and the acetyl-CoA-dependent lipogenic process. In this way, β-hydroxybutyrate can be assessed as a metabolite playing an extra-WAT feeding function. Such an observation reinforces the secretory function of differentiating adipocytes modulating their metabolism to promote systemic metabolic homeostasis in conditions where the glucose is not in excess and probably deleterious. In addition to that, it has been recently proved that β-hydroxybutyrate secretion can be associated with a reduction of ﬁbrosis and can facilitate beige adipogenesis [51]. The observation of such phenomena, linked to a healthier adipose tissue function jeopardizing obesity development or progress, underlines the potential systemic beneﬁts of this LG-GAL condition. 7. Conclusions This study aiming at a deeper investigation of the anti-adipogenic and mitochondrial metabolic function modulation in LG-GAL condition gave rise to new considerations about the metabolism assessment thanks to a targeted 1H-NMR-based metabolomics approach used. From a biological point of view, this study conﬁrmed that such a glucose reduction and partial replacement by galactose is associated with anti-adipogenic effect and several mito- chondrial metabolic-related pathway changes (as summarized in Figure 8). HG condition appeared as an “excess glucose” condition highly stimulating carbohydrates catabolism (glycolysis and acetyl-CoA related mitochondrial carbohydrates catabolism), further stimu- lating the adipogenic and storage function of 3T3-L1 cells. The reduction of the glycolytic rate associated with the stimulation of mitochondrial metabolic pathways not associated with the carbohydrates catabolism (higher glutamine into glutamate and subsequent glu- tamate into pyroglutamate transformation, higher acetate and β-hydroxybutyrate levels, and lower lactate and TCA intermediates levels), in LG-GAL condition, evidence a deep mitochondrial metabolic function changes linked to a reduction of the adipogenic yield and stimulation of the secretion of signaling metabolites (lower lactate and TCA intermediates release associated with higher pyroglutamate, acetate and β-hydroxybutyrate secretion). Consequently, this study throws light on clear but still poorly known metabolite functions as markers of the improvement of the mitochondrial network health status associated with poorer deleterious and less adipogenic nutritional conditions in differentiating 3T3-L1 cells. Biomolecules 2021, 11, 662 15 of 18 Figure 8. Summary of the main metabolic markers’ changes in 3T3-L1 differentiating in LG-GAL condition compared to the HG condition. E and L surrounded letters stands, respectively, for early and late adipogenic phases. The ﬁlled arrow stands for a pathway that is higher in the LG-GAL condition compared to the HG condition. A discontinued arrow stands for a pathway that is lower in LG-GAL condition compared to the HG. Condition. IC and EC stand, respectively, for intracellular and extracellular media. Finally, by assessing different aspects of the mitochondrial metabolism (non-exclusively related to the carbohydrates metabolism), this targeted metabonomics study throws light on a new way to evaluate the mitochondrial function, not only dedicated to the mitochon- drial oxidative metabolism. Most of the current mitochondrial metabolism assays (MTT, TCA and mitochondrial respiratory assessments) are only focusing on one part of the mito- chondrial function and lose a plethora of information to evaluate a potential mitochondrial dysfunction onset. The present study highlights that claiming a mitochondrial dysfunction based on only an oxidative carbohydrates catabolism assessment is irrelevant, and our results should stimulate other mitochondrial markers investigation assessment (acetate, glutamine, glutamate, pyroglutamate, acetate, β-hydroxybutyrate levels) to better reﬂect the mitochondrial function and health status. Author Contributions: Conceptualization, M.D.; data curation, M.D. and V.T.; formal analysis, M.D.; investigation, M.D.; methodology, M.D., V.T. and V.D.; project administration, M.D.; software, M.D.; supervision, J.-M.C. and A.-E.D.; validation, M.D.; visualization, M.D.; writing—original draft, M.D.; writing—review and editing, V.T., J.-M.C. and A.-E.D. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Data Availability Statement: The authors conﬁrm that the data supporting the ﬁndings of this study are available within the present article. Biomolecules 2021, 11, 662 16 of 18 Acknowledgments: Manon Delcourt was the recipient of a grant supported by the University of Mons. We would like to thank all the members of our laboratories for their untiring support for this work. We are grateful to the Department of General, Organic and Biomedical Chemistry (S. Laurent), Faculty of Medicine and Pharmacy, University of Mons, Belgium, for 1H-NMR technical support. The bioproﬁling platform (Bruker Avance NMR spectrometer) was supported by the European Regional Development Fund and the Walloon Region, Belgium. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Wang, Q.A.; Tao, C.; Gupta, R.K.; Scherer, P.E. Tracking adipogenesis during white adipose tissue development, expansion and 2. 3. 4. 5. 6. regeneration. Nat. Med. 2013, 19, 1338–1344. [CrossRef] Vishvanath, L.; Gupta, R.K. Contribution of adipogenesis to healthy adipose tissue expansion in obesity. J. Clin. Investig. 2019, 129, 4022–4031. [CrossRef] Fève, B. Adipogenesis: Cellular and molecular aspects. Best Pract. Res. Clin. Endocrinol. Metab. 2005, 19, 483–499. [CrossRef] Lowe, C.E.; O’Rahilly, S.; Rochford, J.J. Adipogenesis at a glance. J. Cell Sci. 2011, 124, 2681–2686. [CrossRef] Sarjeant, K.; Stephens, J.M. Adipogenesis. Cold Spring Harb. Perspect. Biol. 2012, 4, a008417. Available online: https://www.ncbi. nlm.nih.gov/pmc/articles/PMC3428766/ (accessed on 18 November 2020). [CrossRef] Zebisch, K.; Voigt, V.; Wabitsch, M.; Brandsch, M. Protocol for effective differentiation of 3T3-L1 cells to adipocytes. Anal. Biochem. 2012, 425, 88–90. [CrossRef] 7. Morrison, S.; McGee, S.L. 3T3-L1 adipocytes display phenotypic characteristics of multiple adipocyte lineages. Adipocyte 2015, 4, 8. 9. 295–302. [CrossRef] De Pauw, A.; Tejerina, S.; Raes, M.; Keijer, J.; Arnould, T. Mitochondrial (Dys)function in Adipocyte (De)differentiation and Systemic Metabolic Alterations. Am. J. Pathol. 2009, 175, 927–939. [CrossRef] [PubMed] Li, Q.; Gao, Z.; Chen, Y.; Guan, M.-X. The role of mitochondria in osteogenic, adipogenic and chondrogenic differentiation of mesenchymal stem cells. Protein Cell 2017, 8, 439–445. [CrossRef] [PubMed] 10. Anderson, A.J.; Jackson, T.D.; Stroud, D.A.; Stojanovski, D. Mitochondria—Hubs for regulating cellular biochemistry: Emerging 11. concepts and networks. Open Biol. 2019, 9, 190126. [CrossRef] [PubMed] Spinelli, J.B.; Haigis, M.C. The Multifaceted Contributions of Mitochondria to Cellular Metabolism. Nat. Cell Biol. 2018, 20, 745–754. [CrossRef] [PubMed] 12. Tait, S.W.G.; Green, D.R. Mitochondria and cell signalling. J. Cell Sci. 2012, 125, 807–815. [CrossRef] [PubMed] 13. Chandel, N.S. Evolution of Mitochondria as Signaling Organelles. Cell Metab. 2015, 22, 204–206. [CrossRef] 14. Aon, M.A.; Camara, A.K.S. Mitochondria: Hubs of cellular signaling, energetics and redox balance. A rich, vibrant, and diverse landscape of mitochondrial research. Front. Physiol. 2015, 6, 94. Available online: https://www.ncbi.nlm.nih.gov/pmc/articles/ PMC4374463/ (accessed on 26 October 2020). [CrossRef] [PubMed] 15. Wang, B.; Yang, Q.; Harris, C.L.; Nelson, M.L.; Busboom, J.R.; Zhu, M.-J.; Du, M. Nutrigenomic regulation of adipose tissue development—Role of retinoic acid: A review. Meat Sci. 2016, 120, 100–106. [CrossRef] 16. Khalilpourfarshbaﬁ, M.; Gholami, K.; Murugan, D.D.; Abdul Sattar, M.Z.; Abdullah, N.A. Differential effects of dietary ﬂavonoids on adipogenesis. Eur. J. Nutr. 2019, 58, 5–25. [CrossRef] 17. Bouwman, L.M.S.; Fernández-Calleja, J.M.S.; van der Stelt, I.; Oosting, A.; Keijer, J.; van Schothorst, E.M. Replacing Part of Glucose with Galactose in the Postweaning Diet Protects Female but Not Male Mice from High-Fat Diet-Induced Adiposity in Later Life. J. Nutr. 2019, 149, 1140–1148. [CrossRef] 18. Delcourt, M.; Tagliatti, V.; Delsinne, V.; Colet, J.-M.; Declèves, A.-E. Inﬂuence of Nutritional Intake of Carbohydrates on Mitochondrial Structure, Dynamics, and Functions during Adipogenesis. Nutrients 2020, 12, 2984. [CrossRef] 19. Cowan, K.J. The Mitochondria: Powerhouse of the Cell. In Functional Metabolism: Regulation and Adaptation; Storey, K.B., Ed.; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2004; pp. 211–241. Available online: https://onlinelibrary.wiley.com/doi/abs/10.1002/ 047167558X.ch8 (accessed on 26 October 2020). Javadov, S.; Kuznetsov, A.V. Mitochondria: The cell powerhouse and nexus of stress. Front. Physiol. 2013, 4, 207. Available online: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3735979/ (accessed on 26 October 2020). [CrossRef] 20. 21. Yang, C.; Ko, B.; Hensley, C.T.; Jiang, L.; Wasti, A.T.; Kim, J.; Sudderth, J.; Calvaruso, M.A.; Lumata, L.; Mitsche, M.; et al. Glutamine Oxidation Maintains the TCA Cycle and Cell Survival during Impaired Mitochondrial Pyruvate Transport. Mol. Cell 2014, 56, 414–424. [CrossRef] 22. Liu, Y.D.; Goetze, A.M.; Bass, R.B.; Flynn, G.C. N-terminal Glutamate to Pyroglutamate Conversion In Vivo for Human IgG2 Antibodies. J. Biol. Chem. 2011, 286, 11211–11217. [CrossRef] 23. Weber, G.; Banerjee, G.; Ashmore, J. Activities of enzymes involved in glycolysis, glucogenesis and hexosemonophosphate shunt in rat adipose tissue. Biochem. Biophys. Res. Commun. 1960, 3, 182–186. [CrossRef] Biomolecules 2021, 11, 662 17 of 18 24. Atsumi, T.; Nishio, T.; Niwa, H.; Takeuchi, J.; Bando, H.; Shimizu, C.; Yoshioka, N.; Bucala, R.; Koike, T. Expression of Inducible 6-Phosphofructo-2-Kinase/Fructose-2,6-Bisphosphatase/PFKFB3 Isoforms in Adipocytes and Their Potential Role in Glycolytic Regulation. Diabetes 2005, 54, 3349–3357. [CrossRef] [PubMed] 25. Acheson, K.J.; Schutz, Y.; Bessard, T.; Anantharaman, K.; Flatt, J.P.; Jéquier, E. Glycogen storage capacity and de novo lipogenesis 26. during massive carbohydrate overfeeding in man. Am. J. Clin. Nutr. 1988, 48, 240–247. [CrossRef] [PubMed] Jansson, P.-A.; Smith, U.; Lönnroth, P. Evidence for lactate production by human adipose tissue in vivo. Diabetologia 1990, 33, 253–256. [CrossRef] 27. DiGirolamo, M.; Newby, F.D.; Lovejoy, J. Lactate production in adipose tissue: A regulated function with extra-adipose implications. FASEB J. 1992, 6, 2405–2412. [CrossRef] [PubMed] 28. King, J.L.; DiGirolamo, M. Lactate production from glucose and response to insulin in perifused adipocytes from mesenteric and 29. epididymal regions of lean and obese rats. Obes. Res. 1998, 6, 69–75. [CrossRef] [PubMed] Soares, A.F.; Guichardant, M.; Cozzone, D.; Bernoud-Hubac, N.; Bouzaïdi-Tiali, N.; Lagarde, M.; Géloën, A. Effects of oxidative stress on adiponectin secretion and lactate production in 3T3-L1 adipocytes. Free Radic. Biol. Med. 2005, 38, 882–889. [CrossRef] [PubMed] 30. Harada, N.; Hirano, I.; Inui, H.; Yamaji, R. Stereoselective effects of lactate enantiomers on the enhancement of 3T3-L1 adipocyte 31. differentiation. Biochem. Biophys. Res. Commun. 2018, 498, 105–110. [CrossRef] [PubMed] Sabater, D.; Arriarán, S.; del Mar Romero, M.; Agnelli, S.; Remesar, X.; Fernández-López, J.A.; Alemany, M. Cultured 3T3L1 adipocytes dispose of excess medium glucose as lactate under abundant oxygen availability. Sci. Rep. 2014, 4, 3663. [CrossRef] 32. Krycer, J.R.; Quek, L.-E.; Francis, D.; Fazakerley, D.J.; Elkington, S.D.; Diaz-Vegas, A.; Cooke, K.C.; Weiss, F.C.; Duan, X.; Kurdyukov, S.; et al. Lactate production is a prioritized feature of adipocyte metabolism. J. Biol. Chem. 2020, 295, 83–98. [CrossRef] [PubMed] 33. Philp, A.; Macdonald, A.L.; Watt, P.W. Lactate—A signal coordinating cell and systemic function. J. Exp. Biol. 2005, 208, 4561–4575. [CrossRef] 34. Carrière, A.; Lagarde, D.; Jeanson, Y.; Portais, J.-C.; Galinier, A.; Ader, I.; Casteilla, L. The emerging roles of lactate as a redox substrate and signaling molecule in adipose tissues. J. Physiol. Biochem. 2020, 76, 241–250. [CrossRef] 35. Brooks, G.A. Lactate as a fulcrum of metabolism. Redox Biol. 2020, 35, 101454. [CrossRef] 36. Baltazar, F.; Afonso, J.; Costa, M.; Granja, S. Lactate Beyond a Waste Metabolite: Metabolic Affairs and Signaling in Malignancy. Front. Oncol. 2020, 10, 231. Available online: https://www.frontiersin.org/articles/10.3389/fonc.2020.00231/full (accessed on 30 September 2020). [CrossRef] 37. Petersen, C.; Nielsen, M.D.; Andersen, E.S.; Basse, A.L.; Isidor, M.S.; Markussen, L.K.; Viuff, B.M.; Lambert, I.H.; Hansen, J.B.; Pendersen, S.F. MCT1 and MCT4 Expression and Lactate Flux Activity Increase during White and Brown Adipogenesis and Impact Adipocyte Metabolism. Sci. Rep. 2017, 7, 13101. [CrossRef] 38. Newton, B.W.; Cologna, S.M.; Moya, C.; Russell, D.H.; Russell, W.K.; Jayaraman, A. Proteomic analysis of 3T3-L1 adipocyte 39. mitochondria during differentiation and enlargement. J. Proteome Res. 2011, 10, 4692–4702. [CrossRef] Fedotcheva, N.I.; Sokolov, A.P.; Kondrashova, M.N. Nonezymatic formation of succinate in mitochondria under oxidative stress. Free Radic. Biol. Med. 2006, 41, 56–64. [CrossRef] 41. 40. Nagai, R.; Brock, J.W.; Blatnik, M.; Baatz, J.E.; Bethard, J.; Walla, M.D.; Thorpe, S.R.; Baynes, J.W.; Frizzell, N. Succination of protein thiols during adipocyte maturation: A biomarker of mitochondrial stress. J. Biol. Chem. 2007, 282, 34219–34228. [CrossRef] [PubMed] Frizzell, N.; Rajesh, M.; Jepson, M.J.; Nagai, R.; Carson, J.A.; Thorpe, S.R.; Baynes, J.W. Succination of thiol groups in adipose tissue proteins in diabetes: Succination inhibits polymerization and secretion of adiponectin. J. Biol. Chem. 2009, 284, 25772–25781. [CrossRef] [PubMed] Frizzell, N.; Thomas, S.A.; Carson, J.A.; Baynes, J.W. Mitochondrial stress causes increased succination of proteins in adipocytes in response to glucotoxicity. Biochem. J. 2012, 445, 247–254. [CrossRef] 42. 43. Martínez-Reyes, I.; Chandel, N.S. Mitochondrial TCA cycle metabolites control physiology and disease. Nat. Commun. 2020, 11, 102. [CrossRef] 44. Chen, Q.; Kirk, K.; Shurubor, Y.I.; Zhao, D.; Arreguin, A.J.; Shahi, I.; Valsecchi, F.; Primiano, G.; Calder, E.L.; Carelli, V.; et al. Rewiring of glutamine metabolism is a bioenergetic adaptation of human cells with mitochondrial DNA mutations. Cell Metab. 2018, 27, 1007–1025. [CrossRef] 45. Petrus, P.; Lecoutre, S.; Dollet, L.; Wiel, C.; Sulen, A.; Gao, H.; Tavira, B.; Laurencikiene, J.; Rooyackers, O.; Checa, A.; et al. Glutamine Links Obesity to Inﬂammation in Human White Adipose Tissue. Cell Metab. 2020, 31, 375–390. [CrossRef] 46. Brennan, A.M.; Tchernof, A.; Gerszten, R.E.; Cowan, T.E.; Ross, R. Depot-Speciﬁc Adipose Tissue Metabolite Proﬁles and Corresponding Changes Following Aerobic Exercise. Front. Endocrinol. 2018, 9, 759. Available online: https://www.ncbi.nlm.nih. gov/pmc/articles/PMC6297272/ (accessed on 20 November 2020). [CrossRef] [PubMed] Shi, L.; Tu, B.P. Acetyl-CoA and the Regulation of Metabolism: Mechanisms and Consequences. Curr. Opin. Cell Biol. 2015, 33, 125–131. [CrossRef] 47. Biomolecules 2021, 11, 662 18 of 18 48. Pietrocola, F.; Galluzzi, L.; Bravo-San Pedro, J.M.; Madeo, F.; Kroemer, G. Acetyl Coenzyme A: A Central Metabolite and Second Messenger. Cell Metab. 2015, 21, 805–821. [CrossRef] [PubMed] 49. Liu, X.; Cooper, D.E.; Cluntun, A.A.; Warmoes, M.O.; Zhao, S.; Reid, M.A.; Liu, J.; Lund, P.J.; Lopes, M.; Garcia, B.A.; et al. Acetate Production from Glucose and Coupling to Mitochondrial Metabolism in Mammals. Cell 2018, 175, 502–513. [CrossRef] [PubMed] 50. Alberti, K.G.; Johnston, D.G.; Gill, A.; Barnes, A.J.; Orskov, H. Hormonal regulation of ketone-body metabolism in man. Biochem. Soc. Symp. 1978, 43, 163–182. 51. Wang, W.; Ishibashi, J.; Trefely, S.; Shao, M.; Cowan, A.J.; Sakers, A.; Lim, H.-W.; O’Connor, S.; Doan, M.T.; Cohen, P.; et al. A PRDM16-driven metabolic signal from adipocytes regulates precursor cell fate. Cell Metab. 2019, 30, 174–189. [CrossRef]

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.