Integrated analysis of m6A regulator-mediated RNA methylation modification patterns and immune characteristics in Sjögren’s Syndrome

Integrated analysis of m6A regulator-mediated RNA methylation modification patterns and immune characteristics in Sjögren’s Syndrome | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Integrated analysis of m6A regulator-mediated RNA methylation modification patterns and immune characteristics in Sjögren’s Syndrome Junhao Yin, Jiayao Fu, Jiabao Xu, Changyu Chen, Hanyi Zhu, Yijie Zhao, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-2173202/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 01 Mar, 2024 Read the published version in Heliyon → Version 1 posted You are reading this latest preprint version Abstract Growing evidence suggests that N6-methyladenosine (m6A), the most abundant RNA internal modification, plays a critical role in immune regulation and thereby potentially contributes to the pathogenesis of autoimmune disorders. However, the role of m6A modification of the immune microenvironment of Sjögren’s syndrome (SS) remains unknown. In this study, we used data from public databases and our sequencing efforts to evaluate the expression levels of m6A regulators by profiling the data of whole peripheral blood of 220 SS patients and 62 healthy controls. We found that SS was associated with the expression of several m6A regulators, and this difference was correlated with activated CD4 + T cells. We screened key genes with a random forest (RF) machine learning algorithm and constructed a diagnostic model of SS using multivariate logistic regression analysis. Two distinct m6A modification patterns were determined by unsupervised clustering, with significant differences in immunocyte infiltration, immune reactivity, and enriched biological functions. Key m6A regulators, gene modules, and co-expression networks of m6A-related genes were identified by conventional bioinformatics methods. This identified three key m6A regulators ( METTL3 , ALKBH5 , and YTHDF1 ) and two m6A-related hub genes ( COMMD8 and SRP9 ) which may play an essential role in the diagnosis and treatment of SS. This study demonstrates the close relationship between m6A modification and the immune microenvironment in SS and provides a basis for an improved understanding of m6A modification patterns and the exploration of new therapeutic options for SS. Sjögren’s Syndrome epigenetics immune characteristics machine learning random forest Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Sjögren’s syndrome (SS) is a classical autoimmune disease with relatively high morbidity and a prevalence that is estimated at 0.5%-1% ( 1 ). SS is characterized by abnormal local inflammation resulting in salivary and lachrymal gland dysfunctions ( 2 )( 3 ). The clinical symptoms of SS are not limited to the secretory organs but also manifest in parenchymal organs, including kidney, lung, and liver, which may be related to periepithelial infiltrates in these organs ( 4 ). SS manifests three main stages. The first is characterized by the death of acinar epithelial cells. In the second, immune cell accumulation occurs, first involving macrophages and dendritic cells, then CD4 + T and B220 + B lymphocytes and autoantibodies. In the third stage, patients exhibit obvious clinical symptoms. In the middle or late stages of SS, there are abnormal humoral immune responses mediated by autoreactive B cells with positive expression of anti-ds-DNA antibodies, anti-nuclear antibodies, rheumatoid factor, anti-Ro/SSA antibodies, and anti-La/SSB antibodies. Such humoral immune responses are mainly regulated by CD27 + memory B cells and terminally differentiated plasma cells; this is characterized by the presence of chemokines such as CXCR3, CXCR4, CXCL12, and CXCL13, extensive infiltration of autoreactive B cells in salivary gland lesions, and ectopic germinal center generation. SS patients with germinal center-like structures found in labial gland biopsies have been shown to be more likely to develop non-Hodgkin’s lymphoma and have higher mortality ( 5 )( 6 ). Thus, understanding immune regulation in SS might help us uncover pathologic mechanisms that may prove to be amenable to targeted immuno-based therapies. The m6A modification was first discovered in 1974 and is the most common RNA epigenetic modification in eukaryotic cells ( 7 )( 8 ). Studies have shown that this kind of RNA methylation is a dynamic and reversible process. It can regulate RNA transcription, splicing, degradation and translation without changing the base sequence, thereby mediating the development of various diseases, such as systemic lupus erythematosus ( 9 ), rheumatoid arthritis ( 10 ), and autoimmune thyroid disease ( 11 ). Regulators of this methylation modification include methyltransferases, demethylases, and methylation readers ( 12 ).m6A methyltransferase is a protein complex composed of multiple components (such as METTL3, METTL14, and WTAP), and mainly catalyzes the m6A modification of adenylate on RNA. The function of demethylase is to demethylate bases that have been modified by m6A. FTO protein and ALKBH5 are the only two identified demethylases for m6A modification. M6A methylation reader protein is a type of RNA-binding protein that can specifically bind to m6A methylated regions and alter RNA secondary structure to affect protein-RNA interactions ( 13 ). Numerous studies have demonstrated that modulators of m6A participate in the progression of autoimmune disorders by regulating the function of immune cells. For example, a recent study reported that ALKBH5 promotes the expression of Interferon-γ in CD4 + T cells, thereby promoting the development of colitis in the murine adoptive transfer colitis model ( 14 ). Dysregulation of m6A reader YTHDF2 can lead to the activation of the NF-κB/TNFAIP3 signaling pathway in peripheral blood mononuclear cells (PBMCs), thereby promoting systemic lupus erythematosus ( 15 ). METTL3 activates the NF-κB signaling pathway and promotes the secretion of inflammatory factors by fibroblast-like synovial cells, which ultimately exacerbates the symptoms of rheumatoid arthritis ( 10 ). However, how m6A contributes to the abnormal immune response in SS remains unresolved. Our previous work confirmed that the hyperactivation and abnormal differentiation of CD4 + T cells play a significant role in SS ( 16 )( 17 )( 18 ). Interestingly, a recent study indicated that METTL3 -deficient murine T cells are unable to expand and differentiate in response to IL-7 ( 19 ). This study also reported that METTL3 can regulate the function and stability of regulatory T cells (Tregs) ( 20 ). These results indicated that the m6A modification may also play an immunomodulatory role in SS. In the present study, we applied a machine learning approach to screening for key m6A modulators in SS. Machine learning is one of the fastest-growing fields in artificial intelligence and relies on pattern recognition ( 21 ). This technology has been applied to all aspects of medical research, including predicting patient diagnosis, formulating the best course of treatment, assessing risk levels, pathological analysis, personalized medicine, precision medicine, and the development of predictive models ( 22 )( 23 ). We first utilized database gene expression profiles to identify differentially expressed m6A regulators between samples taken from SS patients and controls. Machine learning approaches were then used to screen key biomarkers to establish a diagnostic model. The resulting predictive model, based on m6A regulators, was able to accurately distinguish between control and SS patient-derived samples. In addition, we clustered the SS samples according to the m6A regulators, grouping them into two distinct m6A modification patterns. There were significant differences in immunocyte infiltration, immune response, and predicted biological functions in different categories of samples. Our results provide support for the role of m6A in the abnormal immune response of SS. 2. Materials And Methods 2.1. Raw data collection and preprocessing The workflow is presented in Fig. 1 . Human expression profiles (accession number: GSE51092) ( 24 ) were downloaded from the Gene Expression Omnibus public database ( 25 ), including the expression array of whole peripheral blood from 190 SS patients and 32 healthy controls. The platform for GSE51092 was GPL6884 (Illumina HumanWG-6 v3.0 expression BeadChip). Another dataset (GSE84844) contained 60 whole blood samples, including 30 SS and 30 controls ( 26 ). The platform for GSE84844 was GPL570 (Affymetrix Human Genome U133 Plus 2.0 Array). The clinical characteristic data for GSE84844 were also downloaded. After merging, all probes were annotated with corresponding gene names; those lacking annotations were eliminated. We performed quantile normalization on the expression data using the limma package ( 27 ). Next, we used the Surrogate Variable Analysis (sva) package to eliminate batch differences and divided the samples into training (47 control and 165 SS) and validation groups (15 control and 55 SS) by random sampling ( 28 ). The associated web link is shown in Additional file 1 . 2.2 Screening for SS-associated m6A regulators We sorted out most of the m6A regulators from previous studies ( 29 )( 30 )( 31 )( Additional file 2 ). We used the limma package to identify m6A regulators that were differentially expressed between the SS and control groups according to the cut-off criteria of an adjusted p -value 1 ( 27 ). The included genes were used in subsequent analyses. Since m6A regulators often function synergistically, we used Pearson correlation analysis to test and quantify the strength of relationships. 2.3 Diagnostic model We screened for predictors of the best mathematical classification model for SS diagnosis using the random forest (RF) model and the support vector machine (SVM) model. RF is an ensemble learning method that integrates many decision trees into a forest to predict the outcome. In the RF method, the importance of variables is indicated by a low Gini coefficient. SVM is a machine learning algorithm that can convert originally inseparable data into linearly separable data through a kernel function. The importance of SVM variables is determined by the discriminant function coefficient value w 2 . The two aforementioned methods were implemented using the randomForest and kernlab packages, respectively. With the help of the pROC and DALEX packages, we compared the AUC [area under the receiver operating characteristic (ROC) curve] values and residuals between the two models to identify the best model. After dimension reduction and feature selection, the selected m6A regulators were prepared to construct a predictive model via logistic regression analysis. ROC curves were used to evaluate the diagnostic efficiency of the predictive model. Calibration plots were used to calibrating agreement between the model predictions and observed values. 2.3 Relationship between m6A regulators and immune characteristics Single-sample gene-set enrichment analysis (ssGSEA) was utilized to estimate the abundance of 23 infiltrating immune cells in different groups using the GSVA package ( 32 )( 33 ) ( Additional file 3 ). The enrichment score representing the relative abundance of each immunocyte was compared between SS and controls using t-tests. In addition, we also assessed the activity of the immune response based on the gene sets of specific immune responses downloaded from the ImmPort database ( 34 ) ( Additional file 4 ). The gene list of major histocompatibility complex (MHC)-related genes was obtained from the HGNC database ( 35 ) ( Additional file 5 ). 2.5 Unsupervised cluster analysis of m6A modification patterns in SS We classified SS samples into distinct m6A modification patterns by unsupervised pattern clustering based on the expression of 16 m6A regulators. The ConsensusClusterPlus package was utilized to construct the cumulative distribution function curve corresponding to k = 2–9 and the delta area score, through which the appropriate number of clusters was determined ( 31 )( 36 ). Subsequently, we performed principal components analysis (PCA) to confirm the clustering effect of two m6A modification subgroups. Additionally, we assessed the differentially expressed genes (DEGs) of SS samples in different m6A clusters to identify m6A regulator mediated genes ( p < 0.05). Significant DEGs were used in subsequent analyses. 2.6 Biological enrichment analysis for different m6A clusters To further explore the biological functions of DEGs, we performed functional enrichment analysis, including Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, as previously described ( 37 )( 38 ). In the GO enrichment analysis, we annotated genes using the org.Hs.eg.db R package and performed enrichment analysis using clusterProfiler ( P < 0.05). As for KEGG pathway analysis, gene annotations were obtained from the KEGG rest API and the enrichment analysis was also implemented using the clusterProfiler package. Statistical thresholds were set at α = 0.05. 2.7 Identification of m6A mediated genes After m6A modification clustering of SS samples, differentially expressed genes in the two distinct clusters were defined as ‘m6A regulator mediated genes’. To find hub genes, we applied the weighted gene co-expression network analysis (WGCNA), which is an analytical method designed to identify cooperatively expressed gene modules and explore the association between gene networks and phenotypes of interest. WGCNA was conducted by using the WGCNA R package based on gene expression profiles of SS samples (n = 189). Gene and sample outliers were removed by the goodSamplesGenes function of the WGCNA package. Correlations between different modules and subgroups were measured by using Pearson’s correlation analyses. Furthermore, we calculated the correlation of m6A modification pattern and gene expression to obtain gene significance (GS), and the correlation of the module eigengene and the gene expression profile to obtain module membership (MM). Based on a previous study, we set the threshold for the hub genes at |MM|>0.8 and |GS|>0.1 ( 39 ). 2.8 Acquisition of the study samples This study was approved by the Ethics Committee of the Shanghai Ninth People’s Hospital affiliated to Shanghai Jiao Tong University School of Medicine (Approval IDs: SH9H-2019-T159-2 and SH9H-2021-TK69-1), and all subjects provided a written informed consent. Selected SS patients fulfilled the criteria of the American-European Consensus Group for SS ( 40 ). None of the patients received immunosuppressive or immunomodulatory drugs before sample harvest. All donors were middle-aged females (between 30 and 60 years old). Ten ml of peripheral blood was collected from each donor and PBMCs were isolated as described ( 16 ). Two subsets of human T cells were isolated from PBMC by positive selection microbeads (CD4 + T cells (130-045-101, Miltenyi Biotec) and CD8 + T cells (130-045-101, Miltenyi Biotec)) according to the manufacturer’s instructions. Besides, labial gland biopsies were collected for hematoxylin & eosin (H&E) staining. 2.9 Cell culture Female C57BL/6 mice were purchased from the Model Animal Research Center of Nanjing University (China). Animal use was in compliance with the criteria outlined in the Guide for the Care and Use of Medical Laboratory Animals ( 41 ). Mouse spleens were carefully ground in a cell strainer (Falcon) and rinsed with PBS containing 2% fetal bovine serum (FBS, Gibco) to obtain a single cell suspension. To isolate total CD4 + T cells, the suspension of splenic cells was purified using a Mouse CD4 + T Cell Isolation Kit (19852A, Stemcell Technologies). Cells were grown in culture medium containing 10% FBS and 1% penicillin/streptomycin (HyClone) and stimulated with 5 µg/ml plate-bound anti-CD3ε and anti-CD28 antibodies (BD Biosciences) for 48 h. 2.10 Gene expression analysis Total RNA was extracted from T cells with TRIzol Reagent (TaKaRa) according to the manufacturer’s protocol. 1000 ng of total RNA was reverse-transcribed to complementary DNA using Takara PrimeScript RT reagent kits (TaKaRa) and was subsequently used for real-time PCR on a LightCycler96 Instrument (Roche). The primer sequences are listed in Table 1 . ACTB/Actb was used as an internal mRNA control. All experiments were repeated in triplicate, and the relative RNA expression rates were calculated using the 2-△△Ct method. 2.8 Statistical analysis Statistical analyses were carried out by using R v4.1.1 and Bioconductor ( Additional file 1 ). All statistical tests were two-tailed and a statistical threshold of α = 0.05 was used throughout. Significant differences are annotated as: * p < 0.05; ** p < 0.01; *** p < 0.001, or **** p < 0.0001. Table 1 A list of Primers Gene and primer type Primer sequences (5' to 3') Actb Forward primer GATCAAGATCATTGCTCCTCCTG Reverse primer AGGGTGTAAAACGCAGCTCA ACTB Forward primer AACGACCCCTTCATTGAC Reverse primer TCCACGACATACTCAGCAC Mettl3 Forward primer CTGGGCACTTGGATTTAAGGAA Reverse primer TGAGAGGTGGTGTAGCAACTT METTL3 Forward primer GAGATATGCTCTTAACCACCCG Reverse primer GCTGCCCAATCCATCCAA Ythdf1 Forward primer ACAGTTACCCCTCGATGAGTG Reverse primer GGTAGTGAGATACGGGATGGGA YTHDF1 Forward primer ACCTGTCCAGCTATTACCCG Reverse primer TGGTGAGGTATGGAATCGGAG Alkbh5 Forward primer CGCGGTCATCAACGACTACC Reverse primer ATGGGCTTGAACTGGAACTTG ALKBH5 Forward primer CCAGCTATGCTTCAGATCGCCT Reverse primer GGTTCTCTTCCTTGTCCATCTCC 3. Results 3.1 Genetic variation of m6A regulators and immune activation SS First, we extracted the clinical data of GSE84844 (GSE51092 had no clinical data). The 30 SS samples were all from female patients, and 24 were older than 50 years old ( Additional file 6 ). In the data of SS patients (n = 270) in our biological sample bank, female patients accounted for about 88%, and 122 patients were older than 50 years old ( Additional file 7 ). These results suggest that middle-aged and old female patients are more likely to develop SS, which is consistent with previous reports ( 1 ). We utilized the inSilicoMerging package for data merging and the sva package for batch effect elimination. Following processing, the data distribution tended to be consistent among the datasets (Fig. 2 A-B, Additional file 8 ), indicating that batch effects were removed. Based on stratified random sampling, the dataset was divided into training and validation samples. The training set contained 165 SS and 47 controls, while the validation set contained 55 SS and 15 controls ( Additional file 9 ). We used the training set to establish a predictive model that was then tested for predictive accuracy and reliability against the validation set. We sorted and classified the accumulative 26 m6A regulators and used a schematic diagram to show how the dynamic process of m6A modification is involved in the immune microenvironment (Fig. 2 C, Additional file 2 ). First, we extracted the expression profiles of m6A regulators in the training cohort. In total, 16 m6A regulators were identified (Fig. 2 D). The basal expressions of YTHDC1 and ALKBH5 were higher than other m6A regulators. Significant expression differences in the 7 regulators were observed between different groups, including YTHDF3 , RBM15 , YTHDC2 , YTHDF1 , RBM15B , ELAVL1 , and ALKBH5 . The absolute fold change in YTHDC2 was the largest, followed by YTHDF3 . In contrast, RBM15B , YTHDF1 , ELAVL1 , and ALKBH5 expression was significantly decreased in SS. The expression profiles of the 7 differentially expressed genes are shown in a heatmap plot (Fig. 2 E). We then performed a correlation analysis to explore the association between different m6A modulators. We found that METTL3 was positively correlated with YTHDC1 , YTHDC2 , YTHDF1 , and YTHDF2 (Fig. 2 F), which may be related to the recruitment of METTL3 to these proteins ( 42 )( 43 ). Additionally, the correlation coefficient between YTHDC1 and YTHDF2 was the highest (r = 0.59). 3.2 Construction and validation of an m6A-based diagnostic model To further restrict the range of m6A regulators, we used the RF and SVM algorithms to construct two different models based on the 7 differentially expressed m6A regulators. As shown in Fig. 3 A, the median residuals obtained by the RF algorithm were lower, implying greater accuracy in the RF model. The reverse cumulative distribution plot also showed that the residuals of most samples of the RF model are relatively lower (Fig. 3 B). The RF model AUC value was also greater than that of the SVM, further supporting the former model (Fig. 3 C). Figure 3 D shows the relationship between the RF iteration times and the classification error; when the iteration times reached 300, the error became small and stable. The importance of the seven m6A regulators were ranked based on factors of the random forest model (Fig. 3 E). All seven feature genes had a Gini index greater than 2 and were selected as predictors ( 44 ). A nomogram model based on these seven predictors was constructed by using the rms package (Fig. 3 F). The discrimination power and calibration capability of the nomogram were evaluated using ROC and calibration curves, respectively. AUC was 0.770 for the derivation set and 0.989 for the validation set, indicating that the model had a good ability to discriminate between control and SS samples (Fig. 3 G-H). The calibration curve showed that the error between the observed and predicted values was small, suggesting that the nomogram model had a strong predictive value (Fig. 3 I-J). These results show that YTHDF3 , RBM15 , YTHDC2 , YTHDF1 , RBM15B , ELAVL1 , and ALKBH5 may play an essential role in the progression of SS. 3.3 Relationship between m6A regulators and the immune microenvironment Differences in the abundance of 23 immunocytes in the immune microenvironment of the control and SS groups are shown in Fig. 4 A. SS patients had higher infiltration of activated B and CD4 + T cells, and Th17 cells. This is consistent with our previous findings that hyperactivation of CD4 + T ( 18 )( 17 )( 45 ) and B cells ( 46 )( 47 ) are important biological features of SS. Notably, these gene expression profiles were derived from peripheral blood samples instead of salivary gland biopsy samples. Studies have shown that in the early stage of SS, peripheral blood CD4 + :CD8 + and Th17:Treg ratios are significantly elevated ( 48 ). We also performed immune infiltration analysis using ssGSEA on our previous transcriptional data of labial gland tissues derived from SS patients and healthy donors ( 49 ). We found a significant increase of activated CD4 + T and B cells in the labial glands (Fig. 4 B). In addition, we performed H&E staining on gland biopsy samples. Labial gland samples from patients with SS contained abundant lymphocytic infiltrates with disrupted acinar structures (Fig. 4 C-D)。Therefore, SS presents with an increase in activated CD4 + T and B cells in the blood and labial glands, suggesting that they are an important feature of the SS immune microenvironment. Recent studies have shown that m6A modification is involved in the regulation of the immune microenvironment and immune responses. To clarify the role of m6A regulators in the immune microenvironment, we performed a correlation analysis (|R|>0.2, p-value < 0.05). The abundance of activated CD4 + T cells were positively correlated with METTL3 , WTAP , RBM15 , YTHDC2 , YTHDF3 , LRPPRC , IGFBP3 , and ALKBH5 . There was also a strong correlation between activated CD8 + T cells and five m6A regulators ( METTL3 , RBM15 , YTHDF2 , LRPPRC , IGFBP3 ) (Fig. 4 E). Treg cells are important inhibitory regulators of the immune response, and were found to be positively correlated with YTHDC1 , YTHDC2 and YTHDF3 ; they were negatively correlated with RBM15B , YTHDF1 , IGFBP3 , ALKBH5 , and ELAVL1 . There was no obvious correlation between activated B cells and m6A-regulated genes. 3.4 The expression of m6A-regulators in activated CD4 + T cells To validate the results of correlation analysis, we examined the gene expression of CD4 + and CD8 + T cells from SS patients and normal donors, finding that the mRNA levels of METTL3 , ALKBH5 and YTHDF1 were significantly increased in SS patients (Fig. 5 A-B). Based on previous sequencing data, the expression of m6A-regulated genes in rest and activated CD4 + T cells is shown in the form of heatmaps (Fig. 5 C-E, Additional file 10 ). Wtap , Virma , Zc3h13 , Rbm15 , Rbm15b , Hnrnpa2b1 , and Alkbh5 were expressed at higher levels in splenic CD4 + T cells. Further, we carried out in vitro experiments to verify. It was found that the expressions of Mettl3, Alkbh5, and Ythdf1 were significantly increased in activated CD4 + T cells, which was consistent with the predicted results (Fig. 5 F-H). Mettl3 regulates the proliferation and differentiation of CD4 + T cells by targeting the IL-7/STAT5/SOCS signaling pathway. Alkbh5 reduced the methylation of CXCL2 and IFN-γ mRNA, resulting in enhanced CD4 + T cell-mediated inflammatory response and recruitment of neutrophils. However, there is currently no studies on the function of Ythdf1 in CD4 + T cells. 3.5 Expression pattern based on 16 m6A methylation modification regulators According to the expression profiles of 16 m6A modulators, we used an unsupervised clustering method to divide the dataset into different categories, thus obtaining distinct m6A modification patterns in SS (Fig. 6 A-C). Given that k = 2 was the best choice, we obtained two distinct patterns, where subgroup A contained 109 samples and subgroup B had 56 samples. Using a PCA analysis, we were able to visually demonstrate the similarity between samples from different clusters. Figure 6 D shows two distinct m6A modification groups. The expression of 16 m6A regulators between two clusters was shown in a heatmap (Fig. 6 E). To investigate the two patterns in-depth, we applied differential analysis to compare the gene expression levels of the two subgroups and found a total of 11 m6A regulators with changes in expression (Fig. 6 F). Compared with subgroup B, WTAP , RBM15 , YTHDF3 , YTHDC2 , and IGF2BP1 were more highly expressed in subgroup A, while RBM15B , CBLL1 , YTHDF1 , YTHDF2 , IGFBP2 , and IGFBP3 were more highly expressed in group B. There is cross interaction or competition between the different m6A reader proteins which constitute an interactive network. The highly expressed reader proteins in different groups may play a leading role to exert their intracellular functions, including in mRNA decay, mRNA stabilization, and enhanced translation. These results suggest that SS is associated with distinct m6A modification patterns. To clarify whether the m6A modification patterns are correlated with the immune microenvironment characteristics, we assessed the level of immune cell infiltration and found two patterns of cell composition (Fig. 7 A). Pattern A had a higher level of infiltrating activated CD4 + T cells, immature dendritic cells, and Tregs; pattern B had more activated CD8 + T cells, monocytes, and T follicular helper cells. This suggests that distinct m6A modification patterns may function in distinct immunocytes. Pattern B also showed more immunoreactivity, including antigen processing and presentation, chemokine receptors, cytokine receptors, natural killer cell cytotoxicity, TCR signaling pathway and TGF-beta family members (Fig. 7 B). In m6A cluster A, activation of the TCR signaling pathway was up-regulated, suggesting that CD4 + T cell activation may be up-regulated, which is consistent with the results of the cell infiltration analysis. The TGF-β signaling pathway, which acts as a suppressor of autoimmune T cell activation, had a low enrichment fraction in both clusters, also likely contributing to the autoimmune responses. Using data from the HGNC database, we also analyzed MHC-related genes for differential expression in two m6A modification patterns (Fig. 7 C). Compared with m6A modification pattern B, the expressions of four MHC class Ⅱ molecules ( HLA-DRA , HLA-DRB4 , HLA-DMB , and HLA-DRB6 ) and one MHC class Ⅰ molecule ( HLA-E) were higher in m6A modification pattern A. The enhanced expression of MHC class Ⅱ molecules is a key component of many autoimmune mouse models, including experimental allergic encephalomyelitis, and is one of the necessary conditions for the increase of autoimmune CD4 + T cells ( 50 ). These findings lead us to posit that patients with m6A modification pattern A are in the early stage of SS, and the number of CD4 + T cells in the peripheral blood is higher and the TCR signaling pathway is enhanced. In turn, patients with m6A modification pattern B are likely to be in the middle or late stages of SS, during which CD4 + T cells and B cells in peripheral blood are somewhat depleted. To test this, we extracted the clinical data of GSE84844 and performed differential analyses of the clinical characteristics of patients in the different m6A clusters. Serological indicators IgA, IgG, IgM, ANA, RF, anti-Ro/SSA, and anti-La/SSB were all related to over-activated humoral immunity (Fig. 7 D-F). Stronger SS humoral immune activation was related to more overt disease activity ( 26 ). Our results showed that the serum IgG and anti-La/SSB levels of m6A cluster B patients were significantly higher than those in cluster A. The SS activity index ESSDAI (European League Against Rheumatism Sjögren’s Syndrome Disease Activity Index) involves 12 domains such as systemic symptoms, lymph nodes, glands, blood system, and serological changes ( 51 ). According to the ESSDAI classification, most of our samples were from patients with mild symptoms; the ESSDAI scores of patients in the two m6A clusters were statistically different. The mean scores of cluster B were greater than those of group A, suggesting that patients in group B may be in a more advanced stage of SS (Fig. 7 G). Taken together, we speculated that compared with pattern B, the immunophenotype of m6A modification pattern A is more similar to the early stage of SS, which is more in line with our research field ( 16 )( 17 )( 18 ). 3.6 Biological characteristics of distinct m6A modification patterns To further explore the role of m6A modification patterns in SS, we conducted a functional enrichment analysis based on differentially expressed genes (DEGs) between the two subgroups to explore potential biological functions. There were a total of 1172 DEGs between the two m6A modification subgroups. The GO enrichment analysis revealed that DEGs mainly participated in the regulation of protein modification processes, response to cytokines and immune responses (Fig. 8 A, Additional file 11 ). These findings provided evidence for a correlation between m6A methylation modification and the immune microenvironment in SS. The MAPK signaling pathway was the most enriched pathway based on the DEGs between two m6A subgroups (Fig. 8 B, Additional file 12 ). The p38/MAPK signaling pathway is known to be closely related to SS. For example, the migratory ability of B cells is mediated by MAPK signaling; blockade of this pathway is shown to effectively alleviate symptoms in experimental SS mice ( 52 ). The MAPK pathway is also involved in the regulation of autophagy and apoptosis of salivary gland cells, which is of great significance in the early onset of SS ( 53 ). We also found that lncRNA NEAT1 can promote the activation of the MAPK signaling pathway in human CD4 + T cells and Jurkat T cells thereby participating in the progression of SS ( 17 ). Subsequently, we performed WGCNA analysis based on the above DEGs (Fig. 8 C). Figures 8 D and 8 E showed that the minimum soft threshold for building a scale-free co-expression network is 14. According to the optimal soft threshold, we constructed a co-expression network and divided genes into four different network modules as shown in Fig. 8 F. Among these modules, the grey invalid module is a gene set consisting of genes that do not belong to any other module. After a correlation analysis between modules and m6A modification patterns, we found that the turquoise-colored module was positively correlated with m6A modification pattern A and had the highest correlation coefficient (Fig. 8 G). Since the immune signature in m6A modification pattern A is closer to that of the early SS lesions that we are most interested in, we chose this module as the key module. To further explore the correlation between this turquoise module and m6A modification pattern A, we calculated their correlation to obtain gene significance (GS), and the correlation of module feature vector and gene expression to obtain module membership (MM). We found a positive correlation between MM and GS (Fig. 8 H), suggesting that these genes, which are highly correlated with m6A modification patterns, also play pivotal roles in the turquoise module. Based on the similarity among genes in the key modules, we used Cytoscape to construct a protein-protein interaction (PPI) network to visualize the interaction between genes (Fig. 8 I)( 54 ). Furthermore, we intersected the important genes in the turquoise module (|MM|>0.8, |GS|>0.1) and the important genes in the PPI network (node degree > 5) to screen reliable hub genes ( Additional file 13 ). As a result, COMMD8 and SRP9 were identified as hub genes. The protein complex composed of COMMD8 and COMMD3 can recruit specific G protein-coupled receptor kinases to catalyze the phosphorylation of these activated chemokine receptors, terminating the signal transduction pathway that is mediated by them ( 55 ). This mechanism makes COMMD8 important for lymphocyte migration and immune regulation. SRP9 is one of the components of the signal recognition particle (SRP) complex. SRP can bind to the nascent signal peptide chain on the ribosome, and then interact with the SRP receptor to deliver the protein to the correct organelle, such as the endoplasmic reticulum membrane ( 56 ). In addition, anti-SRP antibodies have been detected in the serum of patients with various autoimmune diseases including rheumatoid arthritis, multiple sclerosis, and Sjögren’s syndrome, and may be associated with disease progression ( 57 ). Discussion SS is a chronic autoimmune disease characterized by abnormal activation of T cells, production of autoantibodies, and formation of ectopic germinal centers. Despite advances in the understanding of SS pathogenesis, the precision of targeted therapy is limited by disease heterogeneity. Recently, the role of epigenetic factors such as m6A RNA methylation modifications has attracted considerable attention in the field of autoimmunity research. Patients with autoimmune disorders such as Rheumatoid Arthritis show higher PBMC RNA N6-methyladenine that is accompanied by a down-regulation of the m6A erasers ALKBH5 and FTO ( 58 ). In systemic lupus erythematosus, PBMC ALKBH5 and YTHDF2 are downregulated, a presentation that is correlated with several clinical characteristics ( 59 ). Mechanistically, m6A modification can affect the biological functions of various immune cells, thereby regulating the immune microenvironment. For example, conditional knockout of Mettl3 and Mettl14 significantly delays the onset of colitis in the murine T cell transfer model. The m6A modification of suppressor of cytokine signaling (Socs) family genes, an immediate early gene, by Mettl3 and Mettl14 determines the rate of RNA degradation in CD4 + T cells. High levels of SOCS proteins can inhibit the activation of the JAK-STAT5 signaling pathway, resulting in the inhibition of T cell expansion and differentiation after IL-7 stimulation ( 19 ). Knockout of Alkbh5 also protects against colitis and experimental autoimmune encephalomyelitis models, an effect that is due to the upregulation of m6A modification of Ifng and Cxcl2 in CD4 + T cells, resulting in decreased RNA stability and protein levels ( 14 ). Mettl3 can also cooperate with IGF2BP3 and YTHDF2 to regulate the proliferation of germinal center B cells ( 60 ). Apart from immune cells, stem cells and epithelial cells also play an important role in the immune microenvironment. In human kidney epithelial cells, IGF2BP2 recognizes the m6A-modified region of Cebpd and enhances its RNA stability, thereby promoting the translation of CCAAT/enhancer binding protein β/δ and lipocalin-2, both of which are critical for autoantibody-induced glomerulonephritis ( 61 ). Therefore, we believe that the regulatory role played by m6A modification in the immune microenvironment may also mediate the pathogenesis of SS. To explore the association between m6A modification regulators and the immune microenvironment in SS, we performed a systematic bioinformatics analysis based on expression profiles derived from public databases and our sequencing data. First, in peripheral blood samples obtained from SS patients, activated CD4 + T cells and activated B cells were significantly higher than that in healthy controls. This parallelled the immune cell infiltration seen in our labial gland samples, suggesting that SS patients have an active immune response. Subsequently, we found that there were differences in the expression of several m6A regulators between SS and control samples, which covered the types of m6A writers, readers and erasers. The key m6A regulators ( YTHDF3 , RBM15 , YTHDC2 , YTHDF1 , RBM15B , ELAVL1 , and ALKBH5 ) associated with SS were filtered by the random forest algorithm and the multivariate logistic regression analysis. We further analyzed the correlation between m6A regulators and immune cell infiltration levels and surprisingly found that activated CD4 + T cells were positively correlated with most m6A modulators; especially the m6A writers and readers. We validated this prediction in murine CD4 + T cells; after 24 hours of anti-CD3/CD28 antibody stimulation, the expression of multiple m6A regulators in CD4 + T cells was significantly upregulated. Among these up-regulated genes, Alkbh5 and Ythdf1 were activated CD4 + T cell-related genes that were related to SS diagnosis. These findings suggest directions for subsequent research into how m6A modifications regulate the SS immune microenvironment. Unsupervised clustering method based on m6A modulators can help elucidate the regulatory role of m6A modification patterns in relevant genes on the SS immune microenvironment. In turn, these data can provide potential research directions for subsequent biological experiments. This method was first used by Zhang et al. to explore the role of m6A modifications in the tumor immune microenvironment of gastric cancer ( 31 ). Since then, this approach has been used for various diseases, including non-small cell lung cancer ( 62 ), lupus nephritis ( 63 ), periodontitis ( 64 ), etc. Using these approaches, we found two distinct m6A methylation modification patterns based on 16 m6A modulators. The expression of m6A regulators differed among the two patterns which also presented with different immune microenvironment characteristics. Pattern A was characterized by greater activated CD4 + T cells, an enhanced TCR signaling pathway, and an up-regulation of multiple MHC class II molecules, which is similar to the early stages of SS. Correlation analysis of clinical characteristics showed that the serum IgG, anti-La/SSB, and anti-Ro/SSA antibody levels of patients in pattern A were lower than those in pattern B, and that the ESSDAI scores also showed the same trend, thus supporting the above hypothesis. Finally, the DEGs between the two m6A modification patterns were defined as m6A-related genes and were used for biological function prediction as well as the screening of m6A-related hub genes. As a result, these m6A-related genes appear to be related to the immune response and the MAPK signaling pathway. MAPK is one of the main signaling pathways of T cell activation signal transduction. Our previous studies showed that the TLR9-dependent p38/MAPK signaling pathway is abnormally activated in salivary gland cells and peripheral blood mononuclear cells of NOD/Ltj mice in the early stages of SS; inhibition of this pathway helps alleviate symptoms ( 65 )( 66 ). On this basis, we further found that activation of the TLR9/p38/MAPK signaling pathway led to increased apoptosis in human salivary gland cells, which is an important feature of SS ( 53 ). Additionally, two hub genes COMMD8 and SRP9 were screened by the WGCNA analysis and the topological analysis. COMMD8 is a member of the Copper metabolism Murr1 domain-containing protein family. The COMMD8/CCDC22 complex promotes activation of the proinflammatory signaling pathway NF-κB pathway by interacting with IκB-targeting ubiquitin ligase. Dysfunction of COMMOD8 results in the failure of IκB degradation and a failure of NF-κB pathway activation ( 67 ). The COMMD8/COMMD3 complex can recruit GRK6 to the chemokine CXCR4 and induce further phosphorylation of CXCR4, thereby initiating the MAPK signaling pathway and promoting the migration of B cells and humoral immune responses ( 55 ). COMMD8 has also been demonstrated to be regulated by several epigenetic mechanisms. For example, the long non-coding RNA MALAT1 upregulates the COMMD8 abundance by competitively binding to miR-613 , thus promoting the survival and migration of non-small cell lung cancer cells ( 68 ). Another long non-coding RNA LINC00657 , can act as an RNA decoy of miR-26b-5p to promote the expression of COMMD8, which ultimately promotes the proliferation of non-small cell lung cancer cells ( 69 ). Similarly, the LncRNA MNX1-AS1 / miR-218-5p / COMMD8 pathway mediates the migration ability of hepatocellular carcinoma cells ( 70 ). SRP9 is one of the components of the signal recognition particle (SRP) complex, whose main function is to transport proteins to the corresponding organelles ( 56 ). Multiple case reports including polymyositis ( 71 ), systemic lupus erythematosus ( 72 ), and Sjögren’s syndrome ( 73 ) have shown that autoantibodies targeting SRP are associated with disease progression. Ye et al. constructed a gene signature based on five genes including SRP9 to calculate risk scores and predict relapse-free survival in patients with multiple sclerosis, suggesting that SRP9 may play a regulatory function in this autoimmune disease ( 74 ). These predicted hub genes remain to be validated by MeRIP-seq and cytology experiments. At present, there is no m6A modification-related research in the field of SS. We present the first analysis of the relationship between m6A modifications and the immune microenvironment in SS. We achieved this by combining data from public databases and our sequencing data. These promising results help us to understand the complex biological mechanisms of hyperactivation and abnormal differentiation of pro-inflammatory immune cells in SS and help lay the groundwork for the development of precision therapy for SS. Conclusions Overall, this work revealed the underlying regulation mechanisms of m6A methylation modification in the immune microenvironment of SS. We discovered that m6A regulators are correlated with specific immune cell infiltration levels. Comprehensively exploring m6A modification patterns provided a new direction for the understanding of SS, guiding more effective immunotherapy strategies. Abbreviations m6A, N6-methyladenosine; SS, Sjögren’s syndrome; RF, random forest; Treg, regulatory T cell; SVM, support vector machine; ROC, receiver operating characteristic; AUC, area under the ROC curve; ssGSEA, single-sample gene-set enrichment analysis; PCA, principal component analysis; DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; WGCNA, weighted gene co-expression network analysis; MM, module membership; GS, gene significance; MHC, major histocompatibility complex; ESSDAI, European League Against Rheumatism Sjögren’s Syndrome Disease Activity Index; DEGs, differentially expressed genes. Declarations Ethic s approval and consent to participate The study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the Shanghai Ninth People’s Hospital affiliated to Shanghai Jiao Tong University School of Medicine (Approval sequence: SH9H-2019-T159-2 and SH9H-2021-TK69-1).”, and all subjects provided a written informed consent. Consent for publication Not applicable. Availability of data and materials Data banks/repositories corresponding to all datasets analyzed in this study were listed in Additional file 1 . Competing interests The authors declare no potential conflicts of interest. Funding This research was funded by the National Natural Science Foundation of China (Grants No. 82001064, 82170976, 81970951, 81771089), Fundamental research program funding of Ninth People’s Hospital affiliated to Shanghai Jiao Tong university School of Medicine (JYZZ132), Biological sample bank project of Ninth People’s Hospital Affiliated to Shanghai Jiao Tong university School of Medicine (YBKB201907, YBKB202107), The 15th undergraduate training program for innovation of Shanghai Jiaotong University School of medicine (1521Y591), and the Shanghai Summit & Plateau Disciplines. Shanghai Science and Technology Commission Venus Cultivation-Yang Fan Special Project (22YF1422100), Cross funding of Ninth People’s Hospital Affiliated to Shanghai Jiao Tong university School of Medicine (JYJC202126). Author’s contributions Conceptualization, LY. Z, JY. F. and JH. Y.; methology, JY. F.; software, JH.Y. and ZL. Z.; validation, ZL. Z. and JB. X.; formal analysis, JH. Y. and CY. C.; investigation, YJ.Z. and HY. Z.; resources, BL.W.; data curation, JB. X., CY. C.; writing original draft, JH. Y. Acknowledgments The functional enrichment analyses were performed using the Sangerbox tools, a free online platform for data analysis (http://vip.sangerbox.com/). The authors would like to express their gratitude to EditSprings (https://www.editsprings.cn) for the expert linguistic services provided. References Tang Y, Zhou T, Yu X, Xue Z and Shen N: The role of long non-coding RNAs in rheumatic diseases. Nat Rev Rheumatol 13: 657–669, 2017. Mavragani CP and Moutsopoulos HM: Sjögren’s syndrome. Annu Rev Pathol Mech Dis 9: 273–285, 2014. Fox RI: Sjögren’s syndrome. In: Lancet. vol. 366 Elsevier, pp321–331, 2005. Malladi AS, Sack KE, Shiboski SC, et al. : Primary Sjögren’s syndrome as a systemic disease: A study of participants enrolled in an international Sjögren’s syndrome registry. Arthritis Care Res 64: 911–918, 2012. Nocturne G and Mariette X: Advances in understanding the pathogenesis of primary Sjögren’s syndrome. Nat Rev Rheumatol 9: 544–556, 2013. Theander E, Vasaitis L, Baecklund E, et al. : Lymphoid organisation in labial salivary gland biopsies is a possible predictor for the development of malignant lymphoma in primary Sjögren’s syndrome. Ann Rheum Dis 70: 1363–1368, 2011. Desrosiers R, Friderici K and Rottman F: Identification of methylated nucleosides in messenger RNA from Novikoff hepatoma cells. Proc Natl Acad Sci U S A 71: 3971–3975, 1974. Wang Y, Li L, Li J, et al. : The Emerging Role of m6A Modification in Regulating the Immune System and Autoimmune Diseases. Front Cell Dev Biol 9: 1–14, 2021. Luo Q, Fu B, Zhang L, Guo Y, Huang Z and Li J: Decreased Peripheral Blood ALKBH5 Correlates with Markers of Autoimmune Response in Systemic Lupus Erythematosus. Dis Markers 2020, 2020. Shi W, Zheng Y, Luo S, et al. : METTL3 Promotes Activation and Inflammation of FLSs Through the NF-κB Signaling Pathway in Rheumatoid Arthritis. Front Med 8: 1–17, 2021. Song R hua, Zhao J, Gao C qun, Qin Q and Zhang J an: Inclusion of ALKBH5 as a candidate gene for the susceptibility of autoimmune thyroid disease. Adv Med Sci 66: 351–358, 2021. Yang Y, Hsu PJ, Chen YS and Yang YG: Dynamic transcriptomic m6A decoration: Writers, erasers, readers and functions in RNA metabolism. Cell Res 28: 616–624, 2018. Shulman Z and Stern-Ginossar N: The RNA modification N 6-methyladenosine as a novel regulator of the immune system. Nat Immunol 21: 501–512, 2020. Zhou J, Zhang X, Hu J, et al. : M6A demethylase ALKBH5 controls CD4 + T cell pathogenicity and promotes autoimmunity. Sci Adv 7: 1–14, 2021. Zhao X, Dong R, Zhang L, et al. : N6-methyladenosine-dependent modification of circGARS acts as a new player that promotes SLE progression through the NF-κB/A20 axis. Arthritis Res Ther 24: 37, 2022. Fu J, Shi H, Wang B, et al. : LncRNA PVT1 links Myc to glycolytic metabolism upon CD4 + T cell activation and Sjögren’s syndrome-like autoimmune response. J Autoimmun 107: 102358, 2020. Ye L, Shi H, Yu C, et al. : LncRNA Neat1 positively regulates MAPK signaling and is involved in the pathogenesis of Sjögren’s syndrome. Int Immunopharmacol 88: 106992, 2020. Fu J, Pu Y, Wang B, et al. : Pharmacological Inhibition of Glutaminase 1 Normalized the Metabolic State and CD4 + T Cell Response in Sjogren’s Syndrome. J Immunol Res 2022: 1–13, 2022. Li HB, Tong J, Zhu S, et al. : M 6 A mRNA methylation controls T cell homeostasis by targeting the IL-7/STAT5/SOCS pathways. Nature 548: 338–342, 2017. Tong J, Cao G, Zhang T, et al. : m(6)A mRNA methylation sustains Treg suppressive functions. Cell Res 28: 253–256, 2018. Baştanlar Y and Ozuysal M: Introduction to machine learning. Methods Mol Biol 1107: 105–128, 2014. Dai C, Sun B, Wang R and Kang J: The Application of Artificial Intelligence and Machine Learning in Pituitary Adenomas. Front Oncol 11: 784819, 2021. Kolluri S, Lin J, Liu R, Zhang Y and Zhang W: Machine Learning and Artificial Intelligence in Pharmaceutical Research and Development: a Review. AAPS J 24: 19, 2022. Lessard CJ, Li H, Adrianto I, et al. : Variants at multiple loci implicated in both innate and adaptive immune responses are associated with Sjögren’s syndrome. Nat Genet 45: 1284–1292, 2013. Edgar R: Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30: 207–210, 2002. Tasaki S, Suzuki K, Nishikawa A, et al. : Multiomic disease signatures converge to cytotoxic CD8 T cells in primary Sjögren’s syndrome. Ann Rheum Dis 76: 1458–1466, 2017. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W and Smyth GK: Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43: e47, 2015. Leek JT, Johnson WE, Parker HS, Jaffe AE and Storey JD: The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics 28: 882–883, 2012. Chen Y-T, Shen J-Y, Chen D-P, et al. : Identification of cross-talk between m(6)A and 5mC regulators associated with onco-immunogenic features and prognosis across 33 cancer types. J Hematol Oncol 13: 22, 2020. Li Y, Xiao J, Bai J, et al. : Molecular characterization and clinical relevance of m(6)A regulators across 33 cancer types. Mol Cancer 18: 137, 2019. Zhang B, Wu Q, Li B, Wang D, Wang L and Zhou YL: m(6)A regulator-mediated methylation modification patterns and tumor microenvironment infiltration characterization in gastric cancer. Mol Cancer 19: 53, 2020. Barbie DA, Tamayo P, Boehm JS, et al. : Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 462: 108–112, 2009. Hänzelmann S, Castelo R and Guinney J: GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14: 7, 2013. Bhattacharya S, Andorf S, Gomes L, et al. : ImmPort: disseminating data to the public for the future of immunology. Immunol Res 58: 234–239, 2014. Tweedie S, Braschi B, Gray K, Jones TEM, Seal RL, Yates B and Bruford EA: Genenames.org: the HGNC and VGNC resources in 2021. Nucleic Acids Res 49: D939–D946, 2021. Wilkerson MD and Hayes DN: ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics 26: 1572–1573, 2010. Yin J, Zeng X, Ai Z, Yu M, Wu Y and Li S: Construction and analysis of a lncRNA-miRNA-mRNA network based on competitive endogenous RNA reveal functional lncRNAs in oral cancer. BMC Med Genomics 13: 1–14, 2020. Yin J, Zheng Z, Zeng X, et al. : lncRNA MALAT1 mediates osteogenic differentiation of bone mesenchymal stem cells by sponging miR-129-5p. PeerJ 10: e13355, 2022. Tang J, Kong D, Cui Q, Wang K, Zhang D, Gong Y and Wu G: Prognostic Genes of Breast Cancer Identified by Gene Co-expression Network Analysis. Front Oncol 8: 374, 2018. Bruserud Ø: Effects of azoles on human acute myelogenous leukemia blasts and T lymphocytes derived from acute leukemia patients with chemotherapy-induced cytopenia. Int Immunopharmacol 1: 2183–2195, 2001. Guide for the Care and Use of Laboratory Animals. National Academies Press, Washington, D.C., Washington (DC), 2011. Yu D, Horton JR, Yang J, et al. : Human MettL3-MettL14 RNA adenine methyltransferase complex is active on double-stranded DNA containing lesions. Nucleic Acids Res 49: 11629–11642, 2021. Wang Q, Guo X, Li L, Gao Z, Su X, Ji M and Liu J: N6-methyladenosine METTL3 promotes cervical cancer tumorigenesis and Warburg effect through YTHDF1/HK2 modification. Cell Death Dis 11: 911, 2020. Li H, Wu H, Li W, Zhou J, Yang J and Peng W: Constructing a Multiple Sclerosis Diagnosis Model Based on Microarray. Front Neurol 12: 1–10, 2022. Fu J, Shi H, Wang B, et al. : LncRNA PVT1 links Myc to glycolytic metabolism upon CD4 + T cell activation and Sjögren’s syndrome-like autoimmune response. J Autoimmun 107: 102358, 2020. Shao Y, Fu J, Zhan T, Ye L and Yu C: Fangchinoline inhibited proliferation of neoplastic B-lymphoid cells and alleviated Sjögren’s syndrome-like responses in NOD/Ltj mice via the Akt/mTOR pathway. Curr Mol Pharmacol, 2022. Zhan T, Wang B, Fu J, Shao Y, Ye L, Shi H and Zheng L: Artesunate inhibits Sjögren’s syndrome-like autoimmune responses and BAFF-induced B cell hyperactivation via TRAF6-mediated NF-κB signaling. Phytomedicine 80: 153381, 2021. Van Woerkom JM, Kruize AA, Wenting-Van Wijk MJG, et al. : Salivary gland and peripheral blood T helper 1 and 2 cell activity in Sjögren’s syndrome compared with non-Sjögren’s sicca syndrome. Ann Rheum Dis 64: 1474–1479, 2005. Shi H, Cao N, Pu Y, Xie L, Zheng L and Yu C: Long non-coding RNA expression profile in minor salivary gland of primary Sjögren’s syndrome. Arthritis Res Ther 18: 109, 2016. Letterio JJ, Geiser AG, Kulkarni AB, et al. : Autoimmunity associated with TGF-β1-deficiency in mice is dependent on MHC class II antigen expression. J Clin Invest 98: 2109–2119, 1996. Seror R, Ravaud P, Bowman SJ, et al. : EULAR Sjögren’s syndrome disease activity index: Development of a consensus systemic disease activity index for primary Sjögren’s syndrome. Ann Rheum Dis 69: 1103–1109, 2010. Chen X, Zhang P, Liu Q, et al. : Alleviating effect of paeoniflorin-6’-O-benzene sulfonate in antigen-induced experimental Sjögren’s syndrome by modulating B lymphocyte migration via CXCR5-GRK2-ERK/p38 signaling pathway. Int Immunopharmacol 80: 106199, 2020. Fu J, Shi H, Cao N, et al. : Toll-like receptor 9 signaling promotes autophagy and apoptosis via divergent functions of the p38/JNK pathway in human salivary gland cells. Exp Cell Res 375: 51–59, 2019. Shannon P, Markiel A, Ozier O, et al. : Cytoscape: A software Environment for integrated models of biomolecular interaction networks. Genome Res 13: 2498–2504, 2003. Nakai A, Fujimoto J, Miyata H, et al. : The COMMD3/8 complex determines GRK6 specificity for chemoattractant receptors. J Exp Med 216: 1630–1647, 2019. Halic M and Beckmann R: The signal recognition particle and its interactions during protein targeting. Curr Opin Struct Biol 15: 116–125, 2005. Faoro C and Ataide SF: Noncanonical Functions and Cellular Dynamics of the Mammalian Signal Recognition Particle Components. Front Mol Biosci 8: 1–19, 2021. Luo Q, Gao Y, Zhang L, Rao J, Guo Y, Huang Z and Li J: Decreased ALKBH5, FTO, and YTHDF2 in Peripheral Blood Are as Risk Factors for Rheumatoid Arthritis. Biomed Res Int 2020, 2020. Luo Q, Rao J, Zhang L, Fu B, Guo Y, Huang Z and Li J: The study of METTL14, ALKBH5, and YTHDF2 in peripheral blood mononuclear cells from systemic lupus erythematosus. Mol Genet Genomic Med 8: 1–10, 2020. Grenov AC, Moss L, Edelheit S, et al. : The germinal center reaction depends on RNA methylation and divergent functions of specific methyl readers. J Exp Med 218, 2021. Bechara R, Amatya N, Bailey RD, et al. : The m6A reader IMP2 directs autoimmune inflammation through an IL-17 and TNFα-dependent C/EBP transcription factor axis. Sci Immunol 6, 2021. Fan Y, Zhou Y, Lou M, Li X, Zhu X and Yuan K: m6A Regulator-Mediated Methylation Modification Patterns and Characterisation of Tumour Microenvironment Infiltration in Non-Small Cell Lung Cancer. J Inflamm Res Volume 15: 1969–1989, 2022. Zhao H, Pan S, Duan J, Liu F, Li G, Liu D and Liu Z: Integrative Analysis of m6A Regulator-Mediated RNA Methylation Modification Patterns and Immune Characteristics in Lupus Nephritis. Front Cell Dev Biol 9: 1–16, 2021. Zhang X, Zhang S, Yan X, et al. : m6A regulator-mediated RNA methylation modification patterns are involved in immune microenvironment regulation of periodontitis. J Cell Mol Med 25: 3634–3645, 2021. Shi H, Yu C-Q, Xie L-S, Wang Z-J, Zhang P and Zheng L-Y: Activation of TLR9-dependent p38MAPK pathway in the pathogenesis of primary Sjögren’s syndrome in NOD/Ltj mouse. J oral Pathol Med Off Publ Int Assoc Oral Pathol Am Acad Oral Pathol 43: 785–791, 2014. Cao N, Shi H, Chen C, Zheng L and Yu C: Inhibition of the TLR9-dependent p38 MAPK signaling pathway improves the pathogenesis of primary Sjögren’s syndrome in the NOD/Ltj mouse. J Biol Regul Homeost Agents 35: 1103–1108, 2021. Starokadomskyy P, Gluck N, Li H, et al. : CCDC22 deficiency in humans blunts activation of proinfammatory NF-κB signaling. J Clin Invest 123: 2244–2256, 2013. Wang S, Wang T, Liu D and Kong H: LncRNA MALAT1 Aggravates the Progression of Non-Small Cell Lung Cancer by Stimulating the Expression of COMMD8 via Targeting miR-613. Cancer Manag Res 12: 10735–10747, 2020. Zhang R, Niu Z, Pei H and Peng Z: Long noncoding RNA LINC00657 induced by SP1 contributes to the non-small cell lung cancer progression through targeting miR-26b-5p/COMMD8 axis. J Cell Physiol 235: 3340–3349, 2020. Ji D, Wang Y, Sun B, Yang J and Luo X: Long non-coding RNA MNX1-AS1 promotes hepatocellular carcinoma proliferation and invasion through targeting miR-218-5p/COMMD8 axis. Biochem Biophys Res Commun 513: 669–674, 2019. Miller T, Al-Lozi MT, Lopate G and Pestronk A: Myopathy with antibodies to the signal recognition particle: clinical and pathological features. J Neurol Neurosurg Psychiatry 73: 420–428, 2002. Hengstman GJD, ter Laak HJ, Vree Egberts WTM, et al. : Anti-signal recognition particle autoantibodies: marker of a necrotising myopathy. Ann Rheum Dis 65: 1635–1638, 2006. Takahashi N, Nishida A, Tsugawa J, Okajima M, Fujioka S and Tsuboi Y: [A Case of Immune-Mediated Necrotizing Myopathy Associated with Primary Sjögren Syndrome]. Brain Nerve 73: 183–187, 2021. Ye F, Liang J, Li J, Li H and Sheng W: Development and Validation of a Five-Gene Signature to Predict Relapse-Free Survival in Multiple Sclerosis. Front Neurol 11: 1–12, 2020. Additional Declarations No competing interests reported. Supplementary Files Additionalfile1.xls Additionalfile2.xls Additionalfile3.xls Additionalfile4.xls Additionalfile5.xls Additionalfile6.xlsx Additionalfile7.xlsx Additionalfile8.tif Additionalfile9.xlsx Additionalfile10.xlsx Additionalfile11.xlsx Additionalfile12.xlsx Additionalfile13.xlsx Cite Share Download PDF Status: Published Journal Publication published 01 Mar, 2024 Read the published version in Heliyon → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-2173202","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":148724555,"identity":"ccc0420f-236f-4840-ade0-37df3a5f905f","order_by":0,"name":"Junhao Yin","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Junhao","middleName":"","lastName":"Yin","suffix":""},{"id":148724556,"identity":"09098a2c-2661-457f-a8d5-b955cce3d521","order_by":1,"name":"Jiayao Fu","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jiayao","middleName":"","lastName":"Fu","suffix":""},{"id":148724557,"identity":"264348c6-ceb5-4e4d-8d1f-a92b9b6b0f06","order_by":2,"name":"Jiabao Xu","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jiabao","middleName":"","lastName":"Xu","suffix":""},{"id":148724558,"identity":"c4779e16-f302-41c4-9876-cda5733821f6","order_by":3,"name":"Changyu Chen","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Changyu","middleName":"","lastName":"Chen","suffix":""},{"id":148724559,"identity":"0481ecf7-c757-4548-bdc3-ae7517c2979e","order_by":4,"name":"Hanyi Zhu","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hanyi","middleName":"","lastName":"Zhu","suffix":""},{"id":148724560,"identity":"6816a138-ac8f-4479-a629-f943d6db5176","order_by":5,"name":"Yijie Zhao","email":"","orcid":"","institution":"Department of Oral and Maxillofacial Surgery, Shanghai Stomatological Hospital, Fudan University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yijie","middleName":"","lastName":"Zhao","suffix":""},{"id":148724561,"identity":"58ff5e45-a6aa-4226-a4e2-d743f92d5404","order_by":6,"name":"Zhanglong Zheng","email":"","orcid":"","institution":"Department of Oral and Maxillofacial Surgery, School and Hospital of Stomatology, Tongji University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Zhanglong","middleName":"","lastName":"Zheng","suffix":""},{"id":148724562,"identity":"c6394936-1cb4-401a-8d2f-eb4dc34028ea","order_by":7,"name":"Baoli Wang","email":"","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Baoli","middleName":"","lastName":"Wang","suffix":""},{"id":148724563,"identity":"971bba1f-7e21-4641-b211-5fc86e424387","order_by":8,"name":"Lingyan Zheng","email":"data:image/png;base64,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","orcid":"","institution":"Department of Oral Surgery, Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Lingyan","middleName":"","lastName":"Zheng","suffix":""}],"badges":[],"createdAt":"2022-10-17 04:59:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-2173202/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-2173202/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1016/j.heliyon.2024.e28645","type":"published","date":"2024-03-01T16:10:47+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":28682085,"identity":"8396e5b2-292f-46ca-b482-7042749861ad","added_by":"auto","created_at":"2022-11-04 20:42:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":115289,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlow chart of this research programme\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"Figure1final.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/b44d9086bca8644485c3406e.png"},{"id":28682083,"identity":"94bcc51e-a538-4cf7-8254-d6504e3bbbdd","added_by":"auto","created_at":"2022-11-04 20:42:15","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":881726,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of differentially expressed m6A regulators.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A-B) The distribution of mRNA expression levels for two datasets before removing batch effects (A) or after removing batch effects (B).\u003c/p\u003e\n\u003cp\u003e(C) The overview of the dynamic reversible process of m6A RNA methylation modification, which regulated by ‘writers’, ‘erasers’ and ‘readers’ in SS and their potential biological functions for RNA.\u003c/p\u003e\n\u003cp\u003e(D) The violin plot showed the difference in 16 m6A-related genes between the SS and the normal group. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01, triple asterisks indicate P\u0026lt;0.001 vs. control.\u003c/p\u003e\n\u003cp\u003e(E) The heatmap of seven differentially expressed m6A methylation regulators between the SS and the normal group. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01, triple asterisks indicate P\u0026lt;0.001 vs. control.\u003c/p\u003e\n\u003cp\u003e(F) Correlation analysis based on the expression of 16 m6A regulators.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/45c0ff860c9e6fbcd627817e.png"},{"id":28682818,"identity":"2e1c6c73-fb8c-4daa-b168-d3b2622ad844","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":334913,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConstruction and validation of a nomogram model for SS diagnosis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) A residual boxplot based on random forest (RF) and support vector machine (SVM) algorithms.\u003c/p\u003e\n\u003cp\u003e(B) Cumulative residual distribution map of the samples based on RF and SVM. A small number of outliers will contribute many residuals (deviations from the true value), and the higher the position of the line, the larger the residuals for a large number of samples.\u003c/p\u003e\n\u003cp\u003e(C) ROC curves of RF and SVM models. The larger the area under the curve (AUC), the higher the prediction accuracy. ROC, receiver operating characteristic.\u003c/p\u003e\n\u003cp\u003e(D) Screening biomarkers based on RF. The relationship between the number of classification trees and the error in Random Forests. With the increase of the number classification trees, the classification error gradually decreases, and the model gradually tends to be stable.\u003c/p\u003e\n\u003cp\u003e(E) Gini index of predictors in the RF model, which reflects the importance of predictors.\u003c/p\u003e\n\u003cp\u003e(F) The Nomogram for predicting the risk of getting SS based on the expression of seven m6A regulators.\u003c/p\u003e\n\u003cp\u003e(G-H) The predictive value of the m6A regulator gene signature in the derivation (G) and validation (H) sets by calculating the pooled AUC. 0.7\u0026lt;AUC≤1 indicates that the gene signature has high accuracy. The ROC curve shows that the nomogram model constructed from the seven m6A regulators has optimal predictive power. The AUCs were 0.770, and 0.989 in the training and test cohorts, respectively.\u003c/p\u003e\n\u003cp\u003e(I-J) Calibration curve to assess the predictive power of the nomogram model.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/1e6158a73defa91c093cca19.png"},{"id":28682099,"identity":"8c534f09-4a14-41b6-a3e2-3d982cac4ada","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":887781,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe correlation between m6A regulators and immune microenvironment characteristics.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) The immunocyte infiltration in the peripheral blood samples from SS patients and normal donors.\u003c/p\u003e\n\u003cp\u003e(B) The immunocyte infiltration in the labial gland samples from SS patients and normal donors.\u003c/p\u003e\n\u003cp\u003e(C) H\u0026amp;E staining in the labial glands of SS patients and normal donors.\u003c/p\u003e\n\u003cp\u003e(D) The correlation between m6A regulators and infiltration of 23 immune cells. The abundance of activated CD4+T cells are positively correlated with METTL3, WTAP, RBM15, YTHDC2, YTHDF3, LRPPRC, IGFBP3, ALKBH5. |R|\u0026gt;0.2 and P \u0026lt;0.05 were considered significant.\u003c/p\u003e\n\u003cp\u003e(D) The expression of 26 m6A regulators in anti-CD3/CD28 stimulated CD4+T cells and un-activated CD4+T cells. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01 vs. control.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/2f6e86fe9bf428ff2ac40e8c.png"},{"id":28682815,"identity":"3e6f411a-fc81-494a-bd8a-f297cf1f35d1","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":288170,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eM6A regulators are related to CD4+T cell activation and the occurrence of SS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Total CD4+T cells were isolated from 9 healthy donors and 7 SS patients. Expression of each genes was normalized to β-actin expression.\u003c/p\u003e\n\u003cp\u003e(B) Total CD4+T cells were isolated from 9 healthy donors and 7 SS patients. Expression of each genes was normalized to β-actin expression.\u003c/p\u003e\n\u003cp\u003e(C-E) Heatmap of m6A-regulator expression between resting and activated murine CD4+T cells.\u003c/p\u003e\n\u003cp\u003e(F-H) Murine CD4+T cells were cultured in the presence of 5μg/ml plate-bound anti-CD3ε and 2μg/ml anti-CD28 for indicated time points. The expression of m6A regulators were evaluated by real-time qPCR. Data shown are means±SD. Three independent experiments were performed for qRT-PCR assays.\u003c/p\u003e","description":"","filename":"Figure5v2.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/8842478e3cb34de5e0bf41dc.png"},{"id":28683673,"identity":"2a60f126-2817-4e28-a1d8-fbc98ad43926","added_by":"auto","created_at":"2022-11-04 20:58:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":519668,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTwo distinct m6A modification pattern subtypes in SS.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) Cumulative distribution function (CDF) curve of each consensus matrix from k=2 to 9.\u003c/p\u003e\n\u003cp\u003e(B) Relative changes in the area under CDF curve based on different k values.\u003c/p\u003e\n\u003cp\u003e(C) Consensus clustering of the 165 SS samples for k=2.\u003c/p\u003e\n\u003cp\u003e(D) Principal component analysis for the transcriptome profiles of 2 m6A subtypes, showing a remarkable difference on transcriptome between different modification patterns.\u003c/p\u003e\n\u003cp\u003e(E) Heatmap of 16 m6A regulators in different m6A modification patterns.\u003c/p\u003e\n\u003cp\u003e(F) The expression status of 16 m6A regulators in the two m6A subtypes. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01, triple asterisks indicate P\u0026lt;0.001 vs. control.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/f72c94a37400c1cdbf0976b1.png"},{"id":28684607,"identity":"ae1dfbe0-c0c3-4ba2-815e-08f652fe0fc0","added_by":"auto","created_at":"2022-11-04 21:06:16","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":749903,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiversity of immune microenvironment characteristics between two distinct m6A modification patterns.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) The abundance differences of each immune microenvironment infiltrating immunocyte in two m6A modification patterns. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01, triple asterisks indicate P\u0026lt;0.001 vs. control.\u003c/p\u003e\n\u003cp\u003e(B) The activity differences of each immune reaction gene-set in two m6A modification patterns. Asterisk indicates P\u0026lt;0.05, double asterisks indicate P\u0026lt;0.01, quadruple asterisks indicate P\u0026lt;0.0001 vs. control.\u003c/p\u003e\n\u003cp\u003e(C) The expression differences of each MHC-related gene in two m6A modification patterns. Asterisk indicates P\u0026lt;0.05 vs. control.\u003c/p\u003e\n\u003cp\u003e(D-F) Levels of IgG (D), anti-La/SSB (E) and anti-Ro/SSA (F) antibodies in patient serum in different m6A clusters.\u003c/p\u003e\n\u003cp\u003e(G) ESSDAI (European League Against Rheumatism Sjogren’s Syndrome Disease Activity Index) scores of patients in different clusters.\u003c/p\u003e","description":"","filename":"Figure7.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/86fa952ac3b7367e88490609.png"},{"id":28683675,"identity":"1aac8026-c183-44cf-8021-4815efbd40fb","added_by":"auto","created_at":"2022-11-04 20:58:16","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":898348,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification and function analysis of m6A phenotype-related genes in SS.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A) The top 10 most significant Gene ontology terms. Different colors represent different P values.\u003c/p\u003e\n\u003cp\u003e(B) The top 10 most significant pathway terms. Different colors represent different P values.\u003c/p\u003e\n\u003cp\u003e(C) Clustering dendrogram of two m6A modification subtypes in SS.\u003c/p\u003e\n\u003cp\u003e(D-E) Analysis of the scale-free fitting index (D) and analysis of the mean connectivity (E) for soft-thresholding power from 1 to 30.\u003c/p\u003e\n\u003cp\u003e(F) Gene dendrogram obtained by average linkage hierarchical clustering. The color row underneath the dendrogram shows the module assignment determined by the dynamic tree cut, in which 4 modules were identified.\u003c/p\u003e\n\u003cp\u003e(G) Correlation heatmap between module eigen genes and the m6A modification patterns.\u003c/p\u003e\n\u003cp\u003e(H) A scatterplot of gene significance (GS) for m6A modification pattern A vs module membership (MM) in the turquoise module. There is a highly significant correlation between GS and MM in this module, implying that hub genes of the turquoise module also tend to be highly correlated with m6A modification pattern A.\u003c/p\u003e\n\u003cp\u003e(I) A protein-protein interaction (PPI) network based on the genes in the turquoise module. The red background represents a node degree greater than 9. The orange background represents a node degree greater than 5. The yellow background represents a degree less than or equal to 5.\u003c/p\u003e","description":"","filename":"Figure8.png","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/aac71f2a319ef091e96a3072.png"},{"id":53873429,"identity":"2c698370-19c8-4973-8db5-3fc9b40c2bdb","added_by":"auto","created_at":"2024-04-01 16:10:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4678160,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/92a803f9-6d90-42db-ba2b-9ff234b036dd.pdf"},{"id":28682082,"identity":"77e0adba-8570-4a89-918c-fa3f919ec152","added_by":"auto","created_at":"2022-11-04 20:42:15","extension":"xls","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":33280,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile1.xls","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/0475fb69cbeb92cabeb74ec6.xls"},{"id":28682084,"identity":"73b597b7-ed36-436d-afbc-29b032b730b3","added_by":"auto","created_at":"2022-11-04 20:42:15","extension":"xls","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":34816,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile2.xls","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/5f9a84dbedda8b207a131de6.xls"},{"id":28682087,"identity":"1e7bd721-e16d-4741-91bb-33355264f09e","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"xls","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":37888,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile3.xls","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/6faf71d3c2f7927262027c52.xls"},{"id":28682817,"identity":"0a4cdfb0-7fda-4b24-80e9-a9b4222a57e5","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"xls","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":49152,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile4.xls","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/007452f48c87849141e42c73.xls"},{"id":28684605,"identity":"9b70ec57-0b4b-4cb7-b593-285be3ab24bb","added_by":"auto","created_at":"2022-11-04 21:06:16","extension":"xls","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":37376,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile5.xls","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/d225bef89b11c6c064233928.xls"},{"id":28682091,"identity":"1cfabec3-6a6a-45ad-9ce1-b690c9003d59","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":15412,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/c370efb9ee9c1dec6731898c.xlsx"},{"id":28682819,"identity":"ac38471c-abc9-46f0-b064-4b605e70dd7c","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":49339,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile7.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/979bfe1b71a7228924239d16.xlsx"},{"id":28682101,"identity":"96d8700a-7feb-4665-a5ed-316a3c67f5d9","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"tif","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":6257068,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile8.tif","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/12c7d5907aad2a52509394df.tif"},{"id":28682093,"identity":"85b8640d-ce46-4b37-bd30-05ea0adfbc4a","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"xlsx","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":15487,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile9.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/c1c2b206a1cf1e36765ce409.xlsx"},{"id":28682825,"identity":"efdba0e2-d07a-4cb9-94ef-05f657371df0","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"xlsx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":6222566,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile10.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/ef2189395ad48f11708811d9.xlsx"},{"id":28684609,"identity":"bd483c5e-3798-43c6-a40a-f1b91b8cbf79","added_by":"auto","created_at":"2022-11-04 21:06:16","extension":"xlsx","order_by":11,"title":"","display":"","copyAsset":false,"role":"supplement","size":56367,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile11.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/284072b6b08cf618c26238b4.xlsx"},{"id":28682821,"identity":"362de4a6-702c-4af1-b8a4-c9b0d7c2941d","added_by":"auto","created_at":"2022-11-04 20:50:16","extension":"xlsx","order_by":12,"title":"","display":"","copyAsset":false,"role":"supplement","size":13317,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile12.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/ae5f043a793fab7234c1bd37.xlsx"},{"id":28682095,"identity":"f44f2e04-10a6-4b57-a444-37f62d18a25c","added_by":"auto","created_at":"2022-11-04 20:42:16","extension":"xlsx","order_by":13,"title":"","display":"","copyAsset":false,"role":"supplement","size":11565,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile13.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-2173202/v1/bd8e280379dd327ac3e08ee4.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Integrated analysis of m6A regulator-mediated RNA methylation modification patterns and immune characteristics in Sjögren’s Syndrome","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eSj\u0026ouml;gren\u0026rsquo;s syndrome (SS) is a classical autoimmune disease with relatively high morbidity and a prevalence that is estimated at 0.5%-1% (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). SS is characterized by abnormal local inflammation resulting in salivary and lachrymal gland dysfunctions (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e)(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). The clinical symptoms of SS are not limited to the secretory organs but also manifest in parenchymal organs, including kidney, lung, and liver, which may be related to periepithelial infiltrates in these organs (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). SS manifests three main stages. The first is characterized by the death of acinar epithelial cells. In the second, immune cell accumulation occurs, first involving macrophages and dendritic cells, then CD4\u003csup\u003e+\u003c/sup\u003eT and B220\u003csup\u003e+\u003c/sup\u003e B lymphocytes and autoantibodies. In the third stage, patients exhibit obvious clinical symptoms. In the middle or late stages of SS, there are abnormal humoral immune responses mediated by autoreactive B cells with positive expression of anti-ds-DNA antibodies, anti-nuclear antibodies, rheumatoid factor, anti-Ro/SSA antibodies, and anti-La/SSB antibodies. Such humoral immune responses are mainly regulated by CD27\u003csup\u003e+\u003c/sup\u003e memory B cells and terminally differentiated plasma cells; this is characterized by the presence of chemokines such as CXCR3, CXCR4, CXCL12, and CXCL13, extensive infiltration of autoreactive B cells in salivary gland lesions, and ectopic germinal center generation. SS patients with germinal center-like structures found in labial gland biopsies have been shown to be more likely to develop non-Hodgkin\u0026rsquo;s lymphoma and have higher mortality (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)(\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Thus, understanding immune regulation in SS might help us uncover pathologic mechanisms that may prove to be amenable to targeted immuno-based therapies.\u003c/p\u003e \u003cp\u003eThe m6A modification was first discovered in 1974 and is the most common RNA epigenetic modification in eukaryotic cells (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e)(\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Studies have shown that this kind of RNA methylation is a dynamic and reversible process. It can regulate RNA transcription, splicing, degradation and translation without changing the base sequence, thereby mediating the development of various diseases, such as systemic lupus erythematosus (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e), rheumatoid arthritis (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), and autoimmune thyroid disease (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Regulators of this methylation modification include methyltransferases, demethylases, and methylation readers (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e).m6A methyltransferase is a protein complex composed of multiple components (such as METTL3, METTL14, and WTAP), and mainly catalyzes the m6A modification of adenylate on RNA. The function of demethylase is to demethylate bases that have been modified by m6A. FTO protein and ALKBH5 are the only two identified demethylases for m6A modification. M6A methylation reader protein is a type of RNA-binding protein that can specifically bind to m6A methylated regions and alter RNA secondary structure to affect protein-RNA interactions (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNumerous studies have demonstrated that modulators of m6A participate in the progression of autoimmune disorders by regulating the function of immune cells. For example, a recent study reported that ALKBH5 promotes the expression of Interferon-γ in CD4\u003csup\u003e+\u003c/sup\u003eT cells, thereby promoting the development of colitis in the murine adoptive transfer colitis model (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Dysregulation of m6A reader YTHDF2 can lead to the activation of the NF-κB/TNFAIP3 signaling pathway in peripheral blood mononuclear cells (PBMCs), thereby promoting systemic lupus erythematosus (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). METTL3 activates the NF-κB signaling pathway and promotes the secretion of inflammatory factors by fibroblast-like synovial cells, which ultimately exacerbates the symptoms of rheumatoid arthritis (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). However, how m6A contributes to the abnormal immune response in SS remains unresolved. Our previous work confirmed that the hyperactivation and abnormal differentiation of CD4\u003csup\u003e+\u003c/sup\u003eT cells play a significant role in SS (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e)(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e)(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Interestingly, a recent study indicated that \u003cem\u003eMETTL3\u003c/em\u003e-deficient murine T cells are unable to expand and differentiate in response to IL-7 (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). This study also reported that \u003cem\u003eMETTL3\u003c/em\u003e can regulate the function and stability of regulatory T cells (Tregs) (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). These results indicated that the m6A modification may also play an immunomodulatory role in SS.\u003c/p\u003e \u003cp\u003eIn the present study, we applied a machine learning approach to screening for key m6A modulators in SS. Machine learning is one of the fastest-growing fields in artificial intelligence and relies on pattern recognition (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). This technology has been applied to all aspects of medical research, including predicting patient diagnosis, formulating the best course of treatment, assessing risk levels, pathological analysis, personalized medicine, precision medicine, and the development of predictive models (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e)(\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e). We first utilized database gene expression profiles to identify differentially expressed m6A regulators between samples taken from SS patients and controls. Machine learning approaches were then used to screen key biomarkers to establish a diagnostic model. The resulting predictive model, based on m6A regulators, was able to accurately distinguish between control and SS patient-derived samples. In addition, we clustered the SS samples according to the m6A regulators, grouping them into two distinct m6A modification patterns. There were significant differences in immunocyte infiltration, immune response, and predicted biological functions in different categories of samples. Our results provide support for the role of m6A in the abnormal immune response of SS.\u003c/p\u003e "},{"header":"2. Materials And Methods","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Raw data collection and preprocessing\u003c/h2\u003e \u003cp\u003eThe workflow is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Human expression profiles (accession number: GSE51092) (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) were downloaded from the Gene Expression Omnibus public database (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e), including the expression array of whole peripheral blood from 190 SS patients and 32 healthy controls. The platform for GSE51092 was GPL6884 (Illumina HumanWG-6 v3.0 expression BeadChip). Another dataset (GSE84844) contained 60 whole blood samples, including 30 SS and 30 controls (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). The platform for GSE84844 was GPL570 (Affymetrix Human Genome U133 Plus 2.0 Array). The clinical characteristic data for GSE84844 were also downloaded. After merging, all probes were annotated with corresponding gene names; those lacking annotations were eliminated. We performed quantile normalization on the expression data using the limma package (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). Next, we used the Surrogate Variable Analysis (sva) package to eliminate batch differences and divided the samples into training (47 control and 165 SS) and validation groups (15 control and 55 SS) by random sampling (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). The associated web link is shown in \u003cb\u003eAdditional file 1\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Screening for SS-associated m6A regulators\u003c/h2\u003e \u003cp\u003eWe sorted out most of the m6A regulators from previous studies (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e)(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e)(\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e)(\u003cb\u003eAdditional file 2\u003c/b\u003e). We used the limma package to identify m6A regulators that were differentially expressed between the SS and control groups according to the cut-off criteria of an adjusted \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and |log\u003csub\u003e2\u003c/sub\u003e (fold change) | \u0026gt; 1 (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). The included genes were used in subsequent analyses. Since m6A regulators often function synergistically, we used Pearson correlation analysis to test and quantify the strength of relationships.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Diagnostic model\u003c/h2\u003e \u003cp\u003eWe screened for predictors of the best mathematical classification model for SS diagnosis using the random forest (RF) model and the support vector machine (SVM) model. RF is an ensemble learning method that integrates many decision trees into a forest to predict the outcome. In the RF method, the importance of variables is indicated by a low Gini coefficient. SVM is a machine learning algorithm that can convert originally inseparable data into linearly separable data through a kernel function. The importance of SVM variables is determined by the discriminant function coefficient value w\u003csup\u003e2\u003c/sup\u003e. The two aforementioned methods were implemented using the randomForest and kernlab packages, respectively. With the help of the pROC and DALEX packages, we compared the AUC [area under the receiver operating characteristic (ROC) curve] values and residuals between the two models to identify the best model. After dimension reduction and feature selection, the selected m6A regulators were prepared to construct a predictive model via logistic regression analysis. ROC curves were used to evaluate the diagnostic efficiency of the predictive model. Calibration plots were used to calibrating agreement between the model predictions and observed values.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Relationship between m6A regulators and immune characteristics\u003c/h2\u003e \u003cp\u003eSingle-sample gene-set enrichment analysis (ssGSEA) was utilized to estimate the abundance of 23 infiltrating immune cells in different groups using the GSVA package (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e)(\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) (\u003cb\u003eAdditional file 3\u003c/b\u003e). The enrichment score representing the relative abundance of each immunocyte was compared between SS and controls using t-tests. In addition, we also assessed the activity of the immune response based on the gene sets of specific immune responses downloaded from the ImmPort database (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e) (\u003cb\u003eAdditional file 4\u003c/b\u003e). The gene list of major histocompatibility complex (MHC)-related genes was obtained from the HGNC database (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e) (\u003cb\u003eAdditional file 5\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e\u003cb\u003e2.5 Unsupervised cluster analysis of m6A modification patterns in SS\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eWe classified SS samples into distinct m6A modification patterns by unsupervised pattern clustering based on the expression of 16 m6A regulators. The ConsensusClusterPlus package was utilized to construct the cumulative distribution function curve corresponding to k\u0026thinsp;=\u0026thinsp;2\u0026ndash;9 and the delta area score, through which the appropriate number of clusters was determined (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e)(\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). Subsequently, we performed principal components analysis (PCA) to confirm the clustering effect of two m6A modification subgroups. Additionally, we assessed the differentially expressed genes (DEGs) of SS samples in different m6A clusters to identify m6A regulator mediated genes (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Significant DEGs were used in subsequent analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Biological enrichment analysis for different m6A clusters\u003c/h2\u003e \u003cp\u003eTo further explore the biological functions of DEGs, we performed functional enrichment analysis, including Gene Ontology (GO) enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, as previously described (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e)(\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). In the GO enrichment analysis, we annotated genes using the org.Hs.eg.db R package and performed enrichment analysis using clusterProfiler (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05). As for KEGG pathway analysis, gene annotations were obtained from the KEGG rest API and the enrichment analysis was also implemented using the clusterProfiler package. Statistical thresholds were set at α\u0026thinsp;=\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.7 Identification of m6A mediated genes\u003c/h2\u003e \u003cp\u003eAfter m6A modification clustering of SS samples, differentially expressed genes in the two distinct clusters were defined as \u0026lsquo;m6A regulator mediated genes\u0026rsquo;. To find hub genes, we applied the weighted gene co-expression network analysis (WGCNA), which is an analytical method designed to identify cooperatively expressed gene modules and explore the association between gene networks and phenotypes of interest. WGCNA was conducted by using the WGCNA R package based on gene expression profiles of SS samples (n\u0026thinsp;=\u0026thinsp;189). Gene and sample outliers were removed by the goodSamplesGenes function of the WGCNA package. Correlations between different modules and subgroups were measured by using Pearson\u0026rsquo;s correlation analyses. Furthermore, we calculated the correlation of m6A modification pattern and gene expression to obtain gene significance (GS), and the correlation of the module eigengene and the gene expression profile to obtain module membership (MM). Based on a previous study, we set the threshold for the hub genes at |MM|\u0026gt;0.8 and |GS|\u0026gt;0.1 (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Acquisition of the study samples\u003c/h2\u003e \u003cp\u003e This study was approved by the Ethics Committee of the Shanghai Ninth People\u0026rsquo;s Hospital affiliated to Shanghai Jiao Tong University School of Medicine (Approval IDs: SH9H-2019-T159-2 and SH9H-2021-TK69-1), and all subjects provided a written informed consent. Selected SS patients fulfilled the criteria of the American-European Consensus Group for SS (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). None of the patients received immunosuppressive or immunomodulatory drugs before sample harvest. All donors were middle-aged females (between 30 and 60 years old). Ten ml of peripheral blood was collected from each donor and PBMCs were isolated as described (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Two subsets of human T cells were isolated from PBMC by positive selection microbeads (CD4\u0026thinsp;+\u0026thinsp;T cells (130-045-101, Miltenyi Biotec) and CD8\u0026thinsp;+\u0026thinsp;T cells (130-045-101, Miltenyi Biotec)) according to the manufacturer\u0026rsquo;s instructions. Besides, labial gland biopsies were collected for hematoxylin \u0026amp; eosin (H\u0026amp;E) staining.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Cell culture\u003c/h2\u003e \u003cp\u003eFemale C57BL/6 mice were purchased from the Model Animal Research Center of Nanjing University (China). Animal use was in compliance with the criteria outlined in the Guide for the Care and Use of Medical Laboratory Animals (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e). Mouse spleens were carefully ground in a cell strainer (Falcon) and rinsed with PBS containing 2% fetal bovine serum (FBS, Gibco) to obtain a single cell suspension. To isolate total CD4\u0026thinsp;+\u0026thinsp;T cells, the suspension of splenic cells was purified using a Mouse CD4\u0026thinsp;+\u0026thinsp;T Cell Isolation Kit (19852A, Stemcell Technologies). Cells were grown in culture medium containing 10% FBS and 1% penicillin/streptomycin (HyClone) and stimulated with 5 \u0026micro;g/ml plate-bound anti-CD3ε and anti-CD28 antibodies (BD Biosciences) for 48 h.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Gene expression analysis\u003c/h2\u003e \u003cp\u003eTotal RNA was extracted from T cells with TRIzol Reagent (TaKaRa) according to the manufacturer\u0026rsquo;s protocol. 1000 ng of total RNA was reverse-transcribed to complementary DNA using Takara PrimeScript RT reagent kits (TaKaRa) and was subsequently used for real-time PCR on a LightCycler96 Instrument (Roche). The primer sequences are listed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. ACTB/Actb was used as an internal mRNA control. All experiments were repeated in triplicate, and the relative RNA expression rates were calculated using the 2-△△Ct method.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Statistical analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were carried out by using R v4.1.1 and Bioconductor (\u003cb\u003eAdditional file 1\u003c/b\u003e). All statistical tests were two-tailed and a statistical threshold of α\u0026thinsp;=\u0026thinsp;0.05 was used throughout. Significant differences are annotated as: *\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05; **\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.01; ***\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, or ****\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.0001.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eA list of Primers\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGene and primer type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrimer sequences (5' to 3')\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eActb\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGATCAAGATCATTGCTCCTCCTG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAGGGTGTAAAACGCAGCTCA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eACTB\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAACGACCCCTTCATTGAC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTCCACGACATACTCAGCAC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eMettl3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCTGGGCACTTGGATTTAAGGAA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTGAGAGGTGGTGTAGCAACTT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eMETTL3\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGAGATATGCTCTTAACCACCCG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCTGCCCAATCCATCCAA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eYthdf1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eACAGTTACCCCTCGATGAGTG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGGTAGTGAGATACGGGATGGGA\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eYTHDF1\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eACCTGTCCAGCTATTACCCG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTGGTGAGGTATGGAATCGGAG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eAlkbh5\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCGCGGTCATCAACGACTACC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eATGGGCTTGAACTGGAACTTG\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eALKBH5\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForward primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCCAGCTATGCTTCAGATCGCCT\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReverse primer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGGTTCTCTTCCTTGTCCATCTCC\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Genetic variation of m6A regulators and immune activation SS\u003c/h2\u003e \u003cp\u003eFirst, we extracted the clinical data of GSE84844 (GSE51092 had no clinical data). The 30 SS samples were all from female patients, and 24 were older than 50 years old (\u003cb\u003eAdditional file 6\u003c/b\u003e). In the data of SS patients (n\u0026thinsp;=\u0026thinsp;270) in our biological sample bank, female patients accounted for about 88%, and 122 patients were older than 50 years old (\u003cb\u003eAdditional file 7\u003c/b\u003e). These results suggest that middle-aged and old female patients are more likely to develop SS, which is consistent with previous reports (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e). We utilized the inSilicoMerging package for data merging and the sva package for batch effect elimination. Following processing, the data distribution tended to be consistent among the datasets (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-B, \u003cb\u003eAdditional file 8\u003c/b\u003e), indicating that batch effects were removed. Based on stratified random sampling, the dataset was divided into training and validation samples. The training set contained 165 SS and 47 controls, while the validation set contained 55 SS and 15 controls (\u003cb\u003eAdditional file 9\u003c/b\u003e). We used the training set to establish a predictive model that was then tested for predictive accuracy and reliability against the validation set. We sorted and classified the accumulative 26 m6A regulators and used a schematic diagram to show how the dynamic process of m6A modification is involved in the immune microenvironment (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, \u003cb\u003eAdditional file 2\u003c/b\u003e). First, we extracted the expression profiles of m6A regulators in the training cohort. In total, 16 m6A regulators were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). The basal expressions of \u003cem\u003eYTHDC1\u003c/em\u003e and \u003cem\u003eALKBH5\u003c/em\u003e were higher than other m6A regulators. Significant expression differences in the 7 regulators were observed between different groups, including \u003cem\u003eYTHDF3\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eELAVL1\u003c/em\u003e, and \u003cem\u003eALKBH5\u003c/em\u003e. The absolute fold change in \u003cem\u003eYTHDC2\u003c/em\u003e was the largest, followed by \u003cem\u003eYTHDF3\u003c/em\u003e. In contrast, \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eELAVL1\u003c/em\u003e, and \u003cem\u003eALKBH5\u003c/em\u003e expression was significantly decreased in SS. The expression profiles of the 7 differentially expressed genes are shown in a heatmap plot (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). We then performed a correlation analysis to explore the association between different m6A modulators. We found that \u003cem\u003eMETTL3\u003c/em\u003e was positively correlated with \u003cem\u003eYTHDC1\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, and \u003cem\u003eYTHDF2\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF), which may be related to the recruitment of \u003cem\u003eMETTL3\u003c/em\u003e to these proteins (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e)(\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). Additionally, the correlation coefficient between \u003cem\u003eYTHDC1\u003c/em\u003e and \u003cem\u003eYTHDF2\u003c/em\u003e was the highest (r\u0026thinsp;=\u0026thinsp;0.59).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Construction and validation of an m6A-based diagnostic model\u003c/h2\u003e \u003cp\u003eTo further restrict the range of m6A regulators, we used the RF and SVM algorithms to construct two different models based on the 7 differentially expressed m6A regulators. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, the median residuals obtained by the RF algorithm were lower, implying greater accuracy in the RF model. The reverse cumulative distribution plot also showed that the residuals of most samples of the RF model are relatively lower (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). The RF model AUC value was also greater than that of the SVM, further supporting the former model (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD shows the relationship between the RF iteration times and the classification error; when the iteration times reached 300, the error became small and stable. The importance of the seven m6A regulators were ranked based on factors of the random forest model (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). All seven feature genes had a Gini index greater than 2 and were selected as predictors (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e). A nomogram model based on these seven predictors was constructed by using the rms package (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). The discrimination power and calibration capability of the nomogram were evaluated using ROC and calibration curves, respectively. AUC was 0.770 for the derivation set and 0.989 for the validation set, indicating that the model had a good ability to discriminate between control and SS samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG-H). The calibration curve showed that the error between the observed and predicted values was small, suggesting that the nomogram model had a strong predictive value (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI-J). These results show that \u003cem\u003eYTHDF3\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eELAVL1\u003c/em\u003e, and \u003cem\u003eALKBH5\u003c/em\u003e may play an essential role in the progression of SS.\u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Relationship between m6A regulators and the immune microenvironment\u003c/h2\u003e \u003cp\u003eDifferences in the abundance of 23 immunocytes in the immune microenvironment of the control and SS groups are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA. SS patients had higher infiltration of activated B and CD4\u003csup\u003e+\u003c/sup\u003eT cells, and Th17 cells. This is consistent with our previous findings that hyperactivation of CD4\u003csup\u003e+\u003c/sup\u003eT (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e)(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e)(\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e) and B cells (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e)(\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e) are important biological features of SS. Notably, these gene expression profiles were derived from peripheral blood samples instead of salivary gland biopsy samples. Studies have shown that in the early stage of SS, peripheral blood CD4\u003csup\u003e+\u003c/sup\u003e:CD8\u003csup\u003e+\u003c/sup\u003e and Th17:Treg ratios are significantly elevated (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). We also performed immune infiltration analysis using ssGSEA on our previous transcriptional data of labial gland tissues derived from SS patients and healthy donors (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e). We found a significant increase of activated CD4\u003csup\u003e+\u003c/sup\u003eT and B cells in the labial glands (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). In addition, we performed H\u0026amp;E staining on gland biopsy samples. Labial gland samples from patients with SS contained abundant lymphocytic infiltrates with disrupted acinar structures (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC-D)。Therefore, SS presents with an increase in activated CD4\u003csup\u003e+\u003c/sup\u003eT and B cells in the blood and labial glands, suggesting that they are an important feature of the SS immune microenvironment.\u003c/p\u003e \u003cp\u003eRecent studies have shown that m6A modification is involved in the regulation of the immune microenvironment and immune responses. To clarify the role of m6A regulators in the immune microenvironment, we performed a correlation analysis (|R|\u0026gt;0.2, p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The abundance of activated CD4\u003csup\u003e+\u003c/sup\u003eT cells were positively correlated with \u003cem\u003eMETTL3\u003c/em\u003e, \u003cem\u003eWTAP\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, \u003cem\u003eYTHDF3\u003c/em\u003e, \u003cem\u003eLRPPRC\u003c/em\u003e, \u003cem\u003eIGFBP3\u003c/em\u003e, and \u003cem\u003eALKBH5\u003c/em\u003e. There was also a strong correlation between activated CD8\u0026thinsp;+\u0026thinsp;T cells and five m6A regulators (\u003cem\u003eMETTL3\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDF2\u003c/em\u003e, \u003cem\u003eLRPPRC\u003c/em\u003e, \u003cem\u003eIGFBP3\u003c/em\u003e) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). Treg cells are important inhibitory regulators of the immune response, and were found to be positively correlated with \u003cem\u003eYTHDC1\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e and \u003cem\u003eYTHDF3\u003c/em\u003e; they were negatively correlated with \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eIGFBP3\u003c/em\u003e, \u003cem\u003eALKBH5\u003c/em\u003e, and \u003cem\u003eELAVL1\u003c/em\u003e. There was no obvious correlation between activated B cells and m6A-regulated genes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.4 The expression of m6A-regulators in activated CD4\u003csup\u003e+\u003c/sup\u003eT cells\u003c/h2\u003e \u003cp\u003eTo validate the results of correlation analysis, we examined the gene expression of CD4\u003csup\u003e+\u003c/sup\u003e and CD8\u003csup\u003e+\u003c/sup\u003e T cells from SS patients and normal donors, finding that the mRNA levels of \u003cem\u003eMETTL3\u003c/em\u003e, \u003cem\u003eALKBH5\u003c/em\u003e and \u003cem\u003eYTHDF1\u003c/em\u003e were significantly increased in SS patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-B). Based on previous sequencing data, the expression of m6A-regulated genes in rest and activated CD4\u003csup\u003e+\u003c/sup\u003eT cells is shown in the form of heatmaps (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC-E, \u003cb\u003eAdditional file 10\u003c/b\u003e). \u003cem\u003eWtap\u003c/em\u003e, \u003cem\u003eVirma\u003c/em\u003e, \u003cem\u003eZc3h13\u003c/em\u003e, \u003cem\u003eRbm15\u003c/em\u003e, \u003cem\u003eRbm15b\u003c/em\u003e, \u003cem\u003eHnrnpa2b1\u003c/em\u003e, and \u003cem\u003eAlkbh5\u003c/em\u003e were expressed at higher levels in splenic CD4\u003csup\u003e+\u003c/sup\u003eT cells. Further, we carried out in vitro experiments to verify. It was found that the expressions of Mettl3, Alkbh5, and Ythdf1 were significantly increased in activated CD4\u003csup\u003e+\u003c/sup\u003eT cells, which was consistent with the predicted results (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF-H). Mettl3 regulates the proliferation and differentiation of CD4\u003csup\u003e+\u003c/sup\u003eT cells by targeting the IL-7/STAT5/SOCS signaling pathway. Alkbh5 reduced the methylation of CXCL2 and IFN-γ mRNA, resulting in enhanced CD4\u003csup\u003e+\u003c/sup\u003eT cell-mediated inflammatory response and recruitment of neutrophils. However, there is currently no studies on the function of \u003cem\u003eYthdf1\u003c/em\u003e in CD4\u003csup\u003e+\u003c/sup\u003eT cells.\u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Expression pattern based on 16 m6A methylation modification regulators\u003c/b\u003e\u003c/h2\u003e \u003cp\u003eAccording to the expression profiles of 16 m6A modulators, we used an unsupervised clustering method to divide the dataset into different categories, thus obtaining distinct m6A modification patterns in SS (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-C). Given that k\u0026thinsp;=\u0026thinsp;2 was the best choice, we obtained two distinct patterns, where subgroup A contained 109 samples and subgroup B had 56 samples. Using a PCA analysis, we were able to visually demonstrate the similarity between samples from different clusters. Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD shows two distinct m6A modification groups. The expression of 16 m6A regulators between two clusters was shown in a heatmap (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE). To investigate the two patterns in-depth, we applied differential analysis to compare the gene expression levels of the two subgroups and found a total of 11 m6A regulators with changes in expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). Compared with subgroup B, \u003cem\u003eWTAP\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDF3\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, and \u003cem\u003eIGF2BP1\u003c/em\u003e were more highly expressed in subgroup A, while \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eCBLL1\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eYTHDF2\u003c/em\u003e, \u003cem\u003eIGFBP2\u003c/em\u003e, and \u003cem\u003eIGFBP3\u003c/em\u003e were more highly expressed in group B. There is cross interaction or competition between the different m6A reader proteins which constitute an interactive network. The highly expressed reader proteins in different groups may play a leading role to exert their intracellular functions, including in mRNA decay, mRNA stabilization, and enhanced translation. These results suggest that SS is associated with distinct m6A modification patterns.\u003c/p\u003e \u003cp\u003eTo clarify whether the m6A modification patterns are correlated with the immune microenvironment characteristics, we assessed the level of immune cell infiltration and found two patterns of cell composition (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Pattern A had a higher level of infiltrating activated CD4\u003csup\u003e+\u003c/sup\u003eT cells, immature dendritic cells, and Tregs; pattern B had more activated CD8\u003csup\u003e+\u003c/sup\u003eT cells, monocytes, and T follicular helper cells. This suggests that distinct m6A modification patterns may function in distinct immunocytes. Pattern B also showed more immunoreactivity, including antigen processing and presentation, chemokine receptors, cytokine receptors, natural killer cell cytotoxicity, TCR signaling pathway and TGF-beta family members (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB). In m6A cluster A, activation of the TCR signaling pathway was up-regulated, suggesting that CD4\u003csup\u003e+\u003c/sup\u003eT cell activation may be up-regulated, which is consistent with the results of the cell infiltration analysis. The TGF-β signaling pathway, which acts as a suppressor of autoimmune T cell activation, had a low enrichment fraction in both clusters, also likely contributing to the autoimmune responses. Using data from the HGNC database, we also analyzed MHC-related genes for differential expression in two m6A modification patterns (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC). Compared with m6A modification pattern B, the expressions of four MHC class Ⅱ molecules (\u003cem\u003eHLA-DRA\u003c/em\u003e, \u003cem\u003eHLA-DRB4\u003c/em\u003e, \u003cem\u003eHLA-DMB\u003c/em\u003e, and \u003cem\u003eHLA-DRB6\u003c/em\u003e) and one MHC class Ⅰ molecule (\u003cem\u003eHLA-E)\u003c/em\u003e were higher in m6A modification pattern A. The enhanced expression of MHC class Ⅱ molecules is a key component of many autoimmune mouse models, including experimental allergic encephalomyelitis, and is one of the necessary conditions for the increase of autoimmune CD4\u003csup\u003e+\u003c/sup\u003eT cells (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). These findings lead us to posit that patients with m6A modification pattern A are in the early stage of SS, and the number of CD4\u003csup\u003e+\u003c/sup\u003eT cells in the peripheral blood is higher and the TCR signaling pathway is enhanced. In turn, patients with m6A modification pattern B are likely to be in the middle or late stages of SS, during which CD4\u003csup\u003e+\u003c/sup\u003eT cells and B cells in peripheral blood are somewhat depleted. To test this, we extracted the clinical data of GSE84844 and performed differential analyses of the clinical characteristics of patients in the different m6A clusters. Serological indicators IgA, IgG, IgM, ANA, RF, anti-Ro/SSA, and anti-La/SSB were all related to over-activated humoral immunity (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD-F). Stronger SS humoral immune activation was related to more overt disease activity (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Our results showed that the serum IgG and anti-La/SSB levels of m6A cluster B patients were significantly higher than those in cluster A. The SS activity index ESSDAI (European League Against Rheumatism Sj\u0026ouml;gren\u0026rsquo;s Syndrome Disease Activity Index) involves 12 domains such as systemic symptoms, lymph nodes, glands, blood system, and serological changes (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e). According to the ESSDAI classification, most of our samples were from patients with mild symptoms; the ESSDAI scores of patients in the two m6A clusters were statistically different. The mean scores of cluster B were greater than those of group A, suggesting that patients in group B may be in a more advanced stage of SS (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eG). Taken together, we speculated that compared with pattern B, the immunophenotype of m6A modification pattern A is more similar to the early stage of SS, which is more in line with our research field (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e)(\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e)(\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.6 Biological characteristics of distinct m6A modification patterns\u003c/h2\u003e \u003cp\u003eTo further explore the role of m6A modification patterns in SS, we conducted a functional enrichment analysis based on differentially expressed genes (DEGs) between the two subgroups to explore potential biological functions. There were a total of 1172 DEGs between the two m6A modification subgroups. The GO enrichment analysis revealed that DEGs mainly participated in the regulation of protein modification processes, response to cytokines and immune responses (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eA, \u003cb\u003eAdditional file 11\u003c/b\u003e). These findings provided evidence for a correlation between m6A methylation modification and the immune microenvironment in SS. The MAPK signaling pathway was the most enriched pathway based on the DEGs between two m6A subgroups (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eB, \u003cb\u003eAdditional file 12\u003c/b\u003e). The p38/MAPK signaling pathway is known to be closely related to SS. For example, the migratory ability of B cells is mediated by MAPK signaling; blockade of this pathway is shown to effectively alleviate symptoms in experimental SS mice (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). The MAPK pathway is also involved in the regulation of autophagy and apoptosis of salivary gland cells, which is of great significance in the early onset of SS (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). We also found that lncRNA \u003cem\u003eNEAT1\u003c/em\u003e can promote the activation of the MAPK signaling pathway in human CD4\u003csup\u003e+\u003c/sup\u003eT cells and Jurkat T cells thereby participating in the progression of SS (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Subsequently, we performed WGCNA analysis based on the above DEGs (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eC). Figures\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eD and \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eE showed that the minimum soft threshold for building a scale-free co-expression network is 14. According to the optimal soft threshold, we constructed a co-expression network and divided genes into four different network modules as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eF. Among these modules, the grey invalid module is a gene set consisting of genes that do not belong to any other module. After a correlation analysis between modules and m6A modification patterns, we found that the turquoise-colored module was positively correlated with m6A modification pattern A and had the highest correlation coefficient (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eG). Since the immune signature in m6A modification pattern A is closer to that of the early SS lesions that we are most interested in, we chose this module as the key module. To further explore the correlation between this turquoise module and m6A modification pattern A, we calculated their correlation to obtain gene significance (GS), and the correlation of module feature vector and gene expression to obtain module membership (MM). We found a positive correlation between MM and GS (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eH), suggesting that these genes, which are highly correlated with m6A modification patterns, also play pivotal roles in the turquoise module. Based on the similarity among genes in the key modules, we used Cytoscape to construct a protein-protein interaction (PPI) network to visualize the interaction between genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003eI)(\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e). Furthermore, we intersected the important genes in the turquoise module (|MM|\u0026gt;0.8, |GS|\u0026gt;0.1) and the important genes in the PPI network (node degree\u0026thinsp;\u0026gt;\u0026thinsp;5) to screen reliable hub genes (\u003cb\u003eAdditional file 13\u003c/b\u003e). As a result, \u003cem\u003eCOMMD8\u003c/em\u003e and \u003cem\u003eSRP9\u003c/em\u003e were identified as hub genes. The protein complex composed of COMMD8 and COMMD3 can recruit specific G protein-coupled receptor kinases to catalyze the phosphorylation of these activated chemokine receptors, terminating the signal transduction pathway that is mediated by them (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). This mechanism makes COMMD8 important for lymphocyte migration and immune regulation. SRP9 is one of the components of the signal recognition particle (SRP) complex. SRP can bind to the nascent signal peptide chain on the ribosome, and then interact with the SRP receptor to deliver the protein to the correct organelle, such as the endoplasmic reticulum membrane (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). In addition, anti-SRP antibodies have been detected in the serum of patients with various autoimmune diseases including rheumatoid arthritis, multiple sclerosis, and Sj\u0026ouml;gren\u0026rsquo;s syndrome, and may be associated with disease progression (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e).\u003c/p\u003e "},{"header":"Discussion","content":"\u003cp\u003eSS is a chronic autoimmune disease characterized by abnormal activation of T cells, production of autoantibodies, and formation of ectopic germinal centers. Despite advances in the understanding of SS pathogenesis, the precision of targeted therapy is limited by disease heterogeneity. Recently, the role of epigenetic factors such as m6A RNA methylation modifications has attracted considerable attention in the field of autoimmunity research. Patients with autoimmune disorders such as Rheumatoid Arthritis show higher PBMC RNA N6-methyladenine that is accompanied by a down-regulation of the m6A erasers \u003cem\u003eALKBH5\u003c/em\u003e and \u003cem\u003eFTO\u003c/em\u003e (\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e). In systemic lupus erythematosus, PBMC \u003cem\u003eALKBH5\u003c/em\u003e and \u003cem\u003eYTHDF2\u003c/em\u003e are downregulated, a presentation that is correlated with several clinical characteristics (\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e). Mechanistically, m6A modification can affect the biological functions of various immune cells, thereby regulating the immune microenvironment. For example, conditional knockout of \u003cem\u003eMettl3\u003c/em\u003e and \u003cem\u003eMettl14\u003c/em\u003e significantly delays the onset of colitis in the murine T cell transfer model. The m6A modification of suppressor of cytokine signaling (Socs) family genes, an immediate early gene, by \u003cem\u003eMettl3\u003c/em\u003e and \u003cem\u003eMettl14\u003c/em\u003e determines the rate of RNA degradation in CD4\u003csup\u003e+\u003c/sup\u003eT cells. High levels of SOCS proteins can inhibit the activation of the JAK-STAT5 signaling pathway, resulting in the inhibition of T cell expansion and differentiation after IL-7 stimulation (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Knockout of \u003cem\u003eAlkbh5\u003c/em\u003e also protects against colitis and experimental autoimmune encephalomyelitis models, an effect that is due to the upregulation of m6A modification of \u003cem\u003eIfng\u003c/em\u003e and \u003cem\u003eCxcl2\u003c/em\u003e in CD4\u003csup\u003e+\u003c/sup\u003eT cells, resulting in decreased RNA stability and protein levels (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e). Mettl3 can also cooperate with IGF2BP3 and YTHDF2 to regulate the proliferation of germinal center B cells (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e). Apart from immune cells, stem cells and epithelial cells also play an important role in the immune microenvironment. In human kidney epithelial cells, IGF2BP2 recognizes the m6A-modified region of \u003cem\u003eCebpd\u003c/em\u003e and enhances its RNA stability, thereby promoting the translation of CCAAT/enhancer binding protein β/δ and lipocalin-2, both of which are critical for autoantibody-induced glomerulonephritis (\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e). Therefore, we believe that the regulatory role played by m6A modification in the immune microenvironment may also mediate the pathogenesis of SS.\u003c/p\u003e \u003cp\u003eTo explore the association between m6A modification regulators and the immune microenvironment in SS, we performed a systematic bioinformatics analysis based on expression profiles derived from public databases and our sequencing data. First, in peripheral blood samples obtained from SS patients, activated CD4\u003csup\u003e+\u003c/sup\u003eT cells and activated B cells were significantly higher than that in healthy controls. This parallelled the immune cell infiltration seen in our labial gland samples, suggesting that SS patients have an active immune response. Subsequently, we found that there were differences in the expression of several m6A regulators between SS and control samples, which covered the types of m6A writers, readers and erasers. The key m6A regulators (\u003cem\u003eYTHDF3\u003c/em\u003e, \u003cem\u003eRBM15\u003c/em\u003e, \u003cem\u003eYTHDC2\u003c/em\u003e, \u003cem\u003eYTHDF1\u003c/em\u003e, \u003cem\u003eRBM15B\u003c/em\u003e, \u003cem\u003eELAVL1\u003c/em\u003e, and \u003cem\u003eALKBH5\u003c/em\u003e) associated with SS were filtered by the random forest algorithm and the multivariate logistic regression analysis. We further analyzed the correlation between m6A regulators and immune cell infiltration levels and surprisingly found that activated CD4\u003csup\u003e+\u003c/sup\u003eT cells were positively correlated with most m6A modulators; especially the m6A writers and readers. We validated this prediction in murine CD4\u003csup\u003e+\u003c/sup\u003eT cells; after 24 hours of anti-CD3/CD28 antibody stimulation, the expression of multiple m6A regulators in CD4\u003csup\u003e+\u003c/sup\u003eT cells was significantly upregulated. Among these up-regulated genes, \u003cem\u003eAlkbh5\u003c/em\u003e and \u003cem\u003eYthdf1\u003c/em\u003e were activated CD4\u003csup\u003e+\u003c/sup\u003eT cell-related genes that were related to SS diagnosis. These findings suggest directions for subsequent research into how m6A modifications regulate the SS immune microenvironment.\u003c/p\u003e \u003cp\u003eUnsupervised clustering method based on m6A modulators can help elucidate the regulatory role of m6A modification patterns in relevant genes on the SS immune microenvironment. In turn, these data can provide potential research directions for subsequent biological experiments. This method was first used by Zhang et al. to explore the role of m6A modifications in the tumor immune microenvironment of gastric cancer (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Since then, this approach has been used for various diseases, including non-small cell lung cancer (\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e), lupus nephritis (\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e), periodontitis (\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e), etc. Using these approaches, we found two distinct m6A methylation modification patterns based on 16 m6A modulators. The expression of m6A regulators differed among the two patterns which also presented with different immune microenvironment characteristics. Pattern A was characterized by greater activated CD4\u003csup\u003e+\u003c/sup\u003eT cells, an enhanced TCR signaling pathway, and an up-regulation of multiple MHC class II molecules, which is similar to the early stages of SS. Correlation analysis of clinical characteristics showed that the serum IgG, anti-La/SSB, and anti-Ro/SSA antibody levels of patients in pattern A were lower than those in pattern B, and that the ESSDAI scores also showed the same trend, thus supporting the above hypothesis. Finally, the DEGs between the two m6A modification patterns were defined as m6A-related genes and were used for biological function prediction as well as the screening of m6A-related hub genes. As a result, these m6A-related genes appear to be related to the immune response and the MAPK signaling pathway. MAPK is one of the main signaling pathways of T cell activation signal transduction. Our previous studies showed that the TLR9-dependent p38/MAPK signaling pathway is abnormally activated in salivary gland cells and peripheral blood mononuclear cells of NOD/Ltj mice in the early stages of SS; inhibition of this pathway helps alleviate symptoms (\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e)(\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e). On this basis, we further found that activation of the TLR9/p38/MAPK signaling pathway led to increased apoptosis in human salivary gland cells, which is an important feature of SS (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). Additionally, two hub genes \u003cem\u003eCOMMD8\u003c/em\u003e and \u003cem\u003eSRP9\u003c/em\u003e were screened by the WGCNA analysis and the topological analysis.\u003c/p\u003e \u003cp\u003eCOMMD8 is a member of the Copper metabolism Murr1 domain-containing protein family. The COMMD8/CCDC22 complex promotes activation of the proinflammatory signaling pathway NF-κB pathway by interacting with IκB-targeting ubiquitin ligase. Dysfunction of COMMOD8 results in the failure of IκB degradation and a failure of NF-κB pathway activation (\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e). The COMMD8/COMMD3 complex can recruit GRK6 to the chemokine CXCR4 and induce further phosphorylation of CXCR4, thereby initiating the MAPK signaling pathway and promoting the migration of B cells and humoral immune responses (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). COMMD8 has also been demonstrated to be regulated by several epigenetic mechanisms. For example, the long non-coding RNA \u003cem\u003eMALAT1\u003c/em\u003e upregulates the COMMD8 abundance by competitively binding to \u003cem\u003emiR-613\u003c/em\u003e, thus promoting the survival and migration of non-small cell lung cancer cells (\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e). Another long non-coding RNA \u003cem\u003eLINC00657\u003c/em\u003e, can act as an RNA decoy of \u003cem\u003emiR-26b-5p\u003c/em\u003e to promote the expression of COMMD8, which ultimately promotes the proliferation of non-small cell lung cancer cells (\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e). Similarly, the LncRNA \u003cem\u003eMNX1-AS1\u003c/em\u003e/\u003cem\u003emiR-218-5p\u003c/em\u003e/\u003cem\u003eCOMMD8\u003c/em\u003e pathway mediates the migration ability of hepatocellular carcinoma cells (\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e). SRP9 is one of the components of the signal recognition particle (SRP) complex, whose main function is to transport proteins to the corresponding organelles (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). Multiple case reports including polymyositis (\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e), systemic lupus erythematosus (\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e), and Sj\u0026ouml;gren\u0026rsquo;s syndrome (\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e) have shown that autoantibodies targeting SRP are associated with disease progression. Ye et al. constructed a gene signature based on five genes including \u003cem\u003eSRP9\u003c/em\u003e to calculate risk scores and predict relapse-free survival in patients with multiple sclerosis, suggesting that \u003cem\u003eSRP9\u003c/em\u003e may play a regulatory function in this autoimmune disease (\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e). These predicted hub genes remain to be validated by MeRIP-seq and cytology experiments.\u003c/p\u003e \u003cp\u003eAt present, there is no m6A modification-related research in the field of SS. We present the first analysis of the relationship between m6A modifications and the immune microenvironment in SS. We achieved this by combining data from public databases and our sequencing data. These promising results help us to understand the complex biological mechanisms of hyperactivation and abnormal differentiation of pro-inflammatory immune cells in SS and help lay the groundwork for the development of precision therapy for SS.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusions","content":" \u003cp\u003eOverall, this work revealed the underlying regulation mechanisms of m6A methylation modification in the immune microenvironment of SS. We discovered that m6A regulators are correlated with specific immune cell infiltration levels. Comprehensively exploring m6A modification patterns provided a new direction for the understanding of SS, guiding more effective immunotherapy strategies.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003em6A, N6-methyladenosine; SS, Sj\u0026ouml;gren\u0026rsquo;s syndrome; RF, random forest; Treg, regulatory T cell; SVM, support vector machine; ROC, receiver operating characteristic; AUC, area under the ROC curve; ssGSEA, single-sample gene-set enrichment analysis; PCA, principal component analysis; DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; WGCNA, weighted gene co-expression network analysis; MM, module membership; GS, gene significance; MHC, major histocompatibility complex; ESSDAI, European League Against Rheumatism Sj\u0026ouml;gren\u0026rsquo;s Syndrome Disease Activity Index; DEGs, differentially expressed genes.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthic\u003c/strong\u003e\u003cstrong\u003es\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of the Shanghai Ninth People\u0026rsquo;s Hospital affiliated to Shanghai Jiao Tong University School of Medicine (Approval sequence: SH9H-2019-T159-2 and SH9H-2021-TK69-1).\u0026rdquo;,\u0026nbsp;and all subjects provided a written informed consent.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData banks/repositories corresponding to all datasets analyzed in this study were listed in \u003cstrong\u003eAdditional file 1\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no potential conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was funded by the National Natural Science Foundation of China (Grants No. 82001064, 82170976, 81970951, 81771089), Fundamental research program funding of Ninth People\u0026rsquo;s Hospital affiliated to Shanghai Jiao Tong university School of Medicine (JYZZ132), Biological sample bank project of Ninth People\u0026rsquo;s Hospital Affiliated to Shanghai Jiao Tong university School of Medicine (YBKB201907, YBKB202107), The 15th undergraduate training program for innovation of Shanghai Jiaotong University School of medicine (1521Y591), and the Shanghai Summit \u0026amp; Plateau Disciplines. Shanghai Science and Technology Commission Venus Cultivation-Yang Fan Special Project (22YF1422100), Cross funding of Ninth People\u0026rsquo;s Hospital Affiliated to Shanghai Jiao Tong university School of Medicine (JYJC202126).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor\u0026rsquo;s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, LY. Z, JY. F. and JH. Y.; methology, JY. F.; software, JH.Y. and ZL. Z.; validation, ZL. Z. and JB. X.; formal analysis, JH. Y. and CY. C.; investigation, YJ.Z. and HY. Z.; resources, BL.W.; data curation, JB. X., CY. C.; writing original draft, JH. Y.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe functional enrichment analyses were performed using the Sangerbox tools, a free online platform for data analysis (http://vip.sangerbox.com/). The authors would like to express their gratitude to EditSprings (https://www.editsprings.cn) for the expert linguistic services provided.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eTang Y, Zhou T, Yu X, Xue Z and Shen N: The role of long non-coding RNAs in rheumatic diseases. Nat Rev Rheumatol 13: 657\u0026ndash;669, 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMavragani CP and Moutsopoulos HM: Sj\u0026ouml;gren\u0026rsquo;s syndrome. Annu Rev Pathol Mech Dis 9: 273\u0026ndash;285, 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFox RI: Sj\u0026ouml;gren\u0026rsquo;s syndrome. In: Lancet. vol.\u0026nbsp;366 Elsevier, pp321\u0026ndash;331, 2005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalladi AS, Sack KE, Shiboski SC, \u003cem\u003eet al.\u003c/em\u003e: Primary Sj\u0026ouml;gren\u0026rsquo;s syndrome as a systemic disease: A study of participants enrolled in an international Sj\u0026ouml;gren\u0026rsquo;s syndrome registry. Arthritis Care Res 64: 911\u0026ndash;918, 2012.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNocturne G and Mariette X: Advances in understanding the pathogenesis of primary Sj\u0026ouml;gren\u0026rsquo;s syndrome. Nat Rev Rheumatol 9: 544\u0026ndash;556, 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTheander E, Vasaitis L, Baecklund E, \u003cem\u003eet al.\u003c/em\u003e: Lymphoid organisation in labial salivary gland biopsies is a possible predictor for the development of malignant lymphoma in primary Sj\u0026ouml;gren\u0026rsquo;s syndrome. Ann Rheum Dis 70: 1363\u0026ndash;1368, 2011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDesrosiers R, Friderici K and Rottman F: Identification of methylated nucleosides in messenger RNA from Novikoff hepatoma cells. Proc Natl Acad Sci U S A 71: 3971\u0026ndash;3975, 1974.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y, Li L, Li J, \u003cem\u003eet al.\u003c/em\u003e: The Emerging Role of m6A Modification in Regulating the Immune System and Autoimmune Diseases. Front Cell Dev Biol 9: 1\u0026ndash;14, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo Q, Fu B, Zhang L, Guo Y, Huang Z and Li J: Decreased Peripheral Blood ALKBH5 Correlates with Markers of Autoimmune Response in Systemic Lupus Erythematosus. Dis Markers 2020, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi W, Zheng Y, Luo S, \u003cem\u003eet al.\u003c/em\u003e: METTL3 Promotes Activation and Inflammation of FLSs Through the NF-κB Signaling Pathway in Rheumatoid Arthritis. Front Med 8: 1\u0026ndash;17, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong R hua, Zhao J, Gao C qun, Qin Q and Zhang J an: Inclusion of ALKBH5 as a candidate gene for the susceptibility of autoimmune thyroid disease. Adv Med Sci 66: 351\u0026ndash;358, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang Y, Hsu PJ, Chen YS and Yang YG: Dynamic transcriptomic m6A decoration: Writers, erasers, readers and functions in RNA metabolism. Cell Res 28: 616\u0026ndash;624, 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShulman Z and Stern-Ginossar N: The RNA modification N 6-methyladenosine as a novel regulator of the immune system. Nat Immunol 21: 501\u0026ndash;512, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou J, Zhang X, Hu J, \u003cem\u003eet al.\u003c/em\u003e: M6A demethylase ALKBH5 controls CD4 + T cell pathogenicity and promotes autoimmunity. Sci Adv 7: 1\u0026ndash;14, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao X, Dong R, Zhang L, \u003cem\u003eet al.\u003c/em\u003e: N6-methyladenosine-dependent modification of circGARS acts as a new player that promotes SLE progression through the NF-κB/A20 axis. Arthritis Res Ther 24: 37, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu J, Shi H, Wang B, \u003cem\u003eet al.\u003c/em\u003e: LncRNA PVT1 links Myc to glycolytic metabolism upon CD4 + T cell activation and Sj\u0026ouml;gren\u0026rsquo;s syndrome-like autoimmune response. J Autoimmun 107: 102358, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe L, Shi H, Yu C, \u003cem\u003eet al.\u003c/em\u003e: LncRNA Neat1 positively regulates MAPK signaling and is involved in the pathogenesis of Sj\u0026ouml;gren\u0026rsquo;s syndrome. Int Immunopharmacol 88: 106992, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu J, Pu Y, Wang B, \u003cem\u003eet al.\u003c/em\u003e: Pharmacological Inhibition of Glutaminase 1 Normalized the Metabolic State and CD4 + T Cell Response in Sjogren\u0026rsquo;s Syndrome. J Immunol Res 2022: 1\u0026ndash;13, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi HB, Tong J, Zhu S, \u003cem\u003eet al.\u003c/em\u003e: M 6 A mRNA methylation controls T cell homeostasis by targeting the IL-7/STAT5/SOCS pathways. Nature 548: 338\u0026ndash;342, 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTong J, Cao G, Zhang T, \u003cem\u003eet al.\u003c/em\u003e: m(6)A mRNA methylation sustains Treg suppressive functions. Cell Res 28: 253\u0026ndash;256, 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaştanlar Y and Ozuysal M: Introduction to machine learning. Methods Mol Biol 1107: 105\u0026ndash;128, 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDai C, Sun B, Wang R and Kang J: The Application of Artificial Intelligence and Machine Learning in Pituitary Adenomas. Front Oncol 11: 784819, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKolluri S, Lin J, Liu R, Zhang Y and Zhang W: Machine Learning and Artificial Intelligence in Pharmaceutical Research and Development: a Review. AAPS J 24: 19, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLessard CJ, Li H, Adrianto I, \u003cem\u003eet al.\u003c/em\u003e: Variants at multiple loci implicated in both innate and adaptive immune responses are associated with Sj\u0026ouml;gren\u0026rsquo;s syndrome. Nat Genet 45: 1284\u0026ndash;1292, 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdgar R: Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30: 207\u0026ndash;210, 2002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTasaki S, Suzuki K, Nishikawa A, \u003cem\u003eet al.\u003c/em\u003e: Multiomic disease signatures converge to cytotoxic CD8 T cells in primary Sj\u0026ouml;gren\u0026rsquo;s syndrome. Ann Rheum Dis 76: 1458\u0026ndash;1466, 2017.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W and Smyth GK: Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 43: e47, 2015.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeek JT, Johnson WE, Parker HS, Jaffe AE and Storey JD: The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics 28: 882\u0026ndash;883, 2012.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Y-T, Shen J-Y, Chen D-P, \u003cem\u003eet al.\u003c/em\u003e: Identification of cross-talk between m(6)A and 5mC regulators associated with onco-immunogenic features and prognosis across 33 cancer types. J Hematol Oncol 13: 22, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Xiao J, Bai J, \u003cem\u003eet al.\u003c/em\u003e: Molecular characterization and clinical relevance of m(6)A regulators across 33 cancer types. Mol Cancer 18: 137, 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang B, Wu Q, Li B, Wang D, Wang L and Zhou YL: m(6)A regulator-mediated methylation modification patterns and tumor microenvironment infiltration characterization in gastric cancer. Mol Cancer 19: 53, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbie DA, Tamayo P, Boehm JS, \u003cem\u003eet al.\u003c/em\u003e: Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 462: 108\u0026ndash;112, 2009.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026auml;nzelmann S, Castelo R and Guinney J: GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 14: 7, 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBhattacharya S, Andorf S, Gomes L, \u003cem\u003eet al.\u003c/em\u003e: ImmPort: disseminating data to the public for the future of immunology. Immunol Res 58: 234\u0026ndash;239, 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTweedie S, Braschi B, Gray K, Jones TEM, Seal RL, Yates B and Bruford EA: Genenames.org: the HGNC and VGNC resources in 2021. Nucleic Acids Res 49: D939\u0026ndash;D946, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilkerson MD and Hayes DN: ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics 26: 1572\u0026ndash;1573, 2010.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin J, Zeng X, Ai Z, Yu M, Wu Y and Li S: Construction and analysis of a lncRNA-miRNA-mRNA network based on competitive endogenous RNA reveal functional lncRNAs in oral cancer. BMC Med Genomics 13: 1\u0026ndash;14, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin J, Zheng Z, Zeng X, \u003cem\u003eet al.\u003c/em\u003e: lncRNA MALAT1 mediates osteogenic differentiation of bone mesenchymal stem cells by sponging miR-129-5p. PeerJ 10: e13355, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang J, Kong D, Cui Q, Wang K, Zhang D, Gong Y and Wu G: Prognostic Genes of Breast Cancer Identified by Gene Co-expression Network Analysis. Front Oncol 8: 374, 2018.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBruserud \u0026Oslash;: Effects of azoles on human acute myelogenous leukemia blasts and T lymphocytes derived from acute leukemia patients with chemotherapy-induced cytopenia. Int Immunopharmacol 1: 2183\u0026ndash;2195, 2001.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuide for the Care and Use of Laboratory Animals. National Academies Press, Washington, D.C., Washington (DC), 2011.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu D, Horton JR, Yang J, \u003cem\u003eet al.\u003c/em\u003e: Human MettL3-MettL14 RNA adenine methyltransferase complex is active on double-stranded DNA containing lesions. Nucleic Acids Res 49: 11629\u0026ndash;11642, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Q, Guo X, Li L, Gao Z, Su X, Ji M and Liu J: N6-methyladenosine METTL3 promotes cervical cancer tumorigenesis and Warburg effect through YTHDF1/HK2 modification. Cell Death Dis 11: 911, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi H, Wu H, Li W, Zhou J, Yang J and Peng W: Constructing a Multiple Sclerosis Diagnosis Model Based on Microarray. Front Neurol 12: 1\u0026ndash;10, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu J, Shi H, Wang B, \u003cem\u003eet al.\u003c/em\u003e: LncRNA PVT1 links Myc to glycolytic metabolism upon CD4 + T cell activation and Sj\u0026ouml;gren\u0026rsquo;s syndrome-like autoimmune response. J Autoimmun 107: 102358, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShao Y, Fu J, Zhan T, Ye L and Yu C: Fangchinoline inhibited proliferation of neoplastic B-lymphoid cells and alleviated Sj\u0026ouml;gren\u0026rsquo;s syndrome-like responses in NOD/Ltj mice via the Akt/mTOR pathway. Curr Mol Pharmacol, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhan T, Wang B, Fu J, Shao Y, Ye L, Shi H and Zheng L: Artesunate inhibits Sj\u0026ouml;gren\u0026rsquo;s syndrome-like autoimmune responses and BAFF-induced B cell hyperactivation via TRAF6-mediated NF-κB signaling. Phytomedicine 80: 153381, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Woerkom JM, Kruize AA, Wenting-Van Wijk MJG, \u003cem\u003eet al.\u003c/em\u003e: Salivary gland and peripheral blood T helper 1 and 2 cell activity in Sj\u0026ouml;gren\u0026rsquo;s syndrome compared with non-Sj\u0026ouml;gren\u0026rsquo;s sicca syndrome. Ann Rheum Dis 64: 1474\u0026ndash;1479, 2005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi H, Cao N, Pu Y, Xie L, Zheng L and Yu C: Long non-coding RNA expression profile in minor salivary gland of primary Sj\u0026ouml;gren\u0026rsquo;s syndrome. Arthritis Res Ther 18: 109, 2016.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLetterio JJ, Geiser AG, Kulkarni AB, \u003cem\u003eet al.\u003c/em\u003e: Autoimmunity associated with TGF-β1-deficiency in mice is dependent on MHC class II antigen expression. J Clin Invest 98: 2109\u0026ndash;2119, 1996.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeror R, Ravaud P, Bowman SJ, \u003cem\u003eet al.\u003c/em\u003e: EULAR Sj\u0026ouml;gren\u0026rsquo;s syndrome disease activity index: Development of a consensus systemic disease activity index for primary Sj\u0026ouml;gren\u0026rsquo;s syndrome. Ann Rheum Dis 69: 1103\u0026ndash;1109, 2010.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen X, Zhang P, Liu Q, \u003cem\u003eet al.\u003c/em\u003e: Alleviating effect of paeoniflorin-6\u0026rsquo;-O-benzene sulfonate in antigen-induced experimental Sj\u0026ouml;gren\u0026rsquo;s syndrome by modulating B lymphocyte migration via CXCR5-GRK2-ERK/p38 signaling pathway. Int Immunopharmacol 80: 106199, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu J, Shi H, Cao N, \u003cem\u003eet al.\u003c/em\u003e: Toll-like receptor 9 signaling promotes autophagy and apoptosis via divergent functions of the p38/JNK pathway in human salivary gland cells. Exp Cell Res 375: 51\u0026ndash;59, 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon P, Markiel A, Ozier O, \u003cem\u003eet al.\u003c/em\u003e: Cytoscape: A software Environment for integrated models of biomolecular interaction networks. Genome Res 13: 2498\u0026ndash;2504, 2003.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakai A, Fujimoto J, Miyata H, \u003cem\u003eet al.\u003c/em\u003e: The COMMD3/8 complex determines GRK6 specificity for chemoattractant receptors. J Exp Med 216: 1630\u0026ndash;1647, 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHalic M and Beckmann R: The signal recognition particle and its interactions during protein targeting. Curr Opin Struct Biol 15: 116\u0026ndash;125, 2005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFaoro C and Ataide SF: Noncanonical Functions and Cellular Dynamics of the Mammalian Signal Recognition Particle Components. Front Mol Biosci 8: 1\u0026ndash;19, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo Q, Gao Y, Zhang L, Rao J, Guo Y, Huang Z and Li J: Decreased ALKBH5, FTO, and YTHDF2 in Peripheral Blood Are as Risk Factors for Rheumatoid Arthritis. Biomed Res Int 2020, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo Q, Rao J, Zhang L, Fu B, Guo Y, Huang Z and Li J: The study of METTL14, ALKBH5, and YTHDF2 in peripheral blood mononuclear cells from systemic lupus erythematosus. Mol Genet Genomic Med 8: 1\u0026ndash;10, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrenov AC, Moss L, Edelheit S, \u003cem\u003eet al.\u003c/em\u003e: The germinal center reaction depends on RNA methylation and divergent functions of specific methyl readers. J Exp Med 218, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBechara R, Amatya N, Bailey RD, \u003cem\u003eet al.\u003c/em\u003e: The m6A reader IMP2 directs autoimmune inflammation through an IL-17 and TNFα-dependent C/EBP transcription factor axis. Sci Immunol 6, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFan Y, Zhou Y, Lou M, Li X, Zhu X and Yuan K: m6A Regulator-Mediated Methylation Modification Patterns and Characterisation of Tumour Microenvironment Infiltration in Non-Small Cell Lung Cancer. J Inflamm Res Volume 15: 1969\u0026ndash;1989, 2022.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao H, Pan S, Duan J, Liu F, Li G, Liu D and Liu Z: Integrative Analysis of m6A Regulator-Mediated RNA Methylation Modification Patterns and Immune Characteristics in Lupus Nephritis. Front Cell Dev Biol 9: 1\u0026ndash;16, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Zhang S, Yan X, \u003cem\u003eet al.\u003c/em\u003e: m6A regulator-mediated RNA methylation modification patterns are involved in immune microenvironment regulation of periodontitis. J Cell Mol Med 25: 3634\u0026ndash;3645, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi H, Yu C-Q, Xie L-S, Wang Z-J, Zhang P and Zheng L-Y: Activation of TLR9-dependent p38MAPK pathway in the pathogenesis of primary Sj\u0026ouml;gren\u0026rsquo;s syndrome in NOD/Ltj mouse. J oral Pathol Med Off Publ Int Assoc Oral Pathol Am Acad Oral Pathol 43: 785\u0026ndash;791, 2014.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao N, Shi H, Chen C, Zheng L and Yu C: Inhibition of the TLR9-dependent p38 MAPK signaling pathway improves the pathogenesis of primary Sj\u0026ouml;gren\u0026rsquo;s syndrome in the NOD/Ltj mouse. J Biol Regul Homeost Agents 35: 1103\u0026ndash;1108, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStarokadomskyy P, Gluck N, Li H, \u003cem\u003eet al.\u003c/em\u003e: CCDC22 deficiency in humans blunts activation of proinfammatory NF-κB signaling. J Clin Invest 123: 2244\u0026ndash;2256, 2013.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang S, Wang T, Liu D and Kong H: LncRNA MALAT1 Aggravates the Progression of Non-Small Cell Lung Cancer by Stimulating the Expression of COMMD8 via Targeting miR-613. Cancer Manag Res 12: 10735\u0026ndash;10747, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang R, Niu Z, Pei H and Peng Z: Long noncoding RNA LINC00657 induced by SP1 contributes to the non-small cell lung cancer progression through targeting miR-26b-5p/COMMD8 axis. J Cell Physiol 235: 3340\u0026ndash;3349, 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJi D, Wang Y, Sun B, Yang J and Luo X: Long non-coding RNA MNX1-AS1 promotes hepatocellular carcinoma proliferation and invasion through targeting miR-218-5p/COMMD8 axis. Biochem Biophys Res Commun 513: 669\u0026ndash;674, 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller T, Al-Lozi MT, Lopate G and Pestronk A: Myopathy with antibodies to the signal recognition particle: clinical and pathological features. J Neurol Neurosurg Psychiatry 73: 420\u0026ndash;428, 2002.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHengstman GJD, ter Laak HJ, Vree Egberts WTM, \u003cem\u003eet al.\u003c/em\u003e: Anti-signal recognition particle autoantibodies: marker of a necrotising myopathy. Ann Rheum Dis 65: 1635\u0026ndash;1638, 2006.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakahashi N, Nishida A, Tsugawa J, Okajima M, Fujioka S and Tsuboi Y: [A Case of Immune-Mediated Necrotizing Myopathy Associated with Primary Sj\u0026ouml;gren Syndrome]. Brain Nerve 73: 183\u0026ndash;187, 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYe F, Liang J, Li J, Li H and Sheng W: Development and Validation of a Five-Gene Signature to Predict Relapse-Free Survival in Multiple Sclerosis. Front Neurol 11: 1\u0026ndash;12, 2020.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Sjögren’s Syndrome, epigenetics, immune characteristics, machine learning, random forest","lastPublishedDoi":"10.21203/rs.3.rs-2173202/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-2173202/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGrowing evidence suggests that N6-methyladenosine (m6A), the most abundant RNA internal modification, plays a critical role in immune regulation and thereby potentially contributes to the pathogenesis of autoimmune disorders. However, the role of m6A modification of the immune microenvironment of Sj\u0026ouml;gren\u0026rsquo;s syndrome (SS) remains unknown. In this study, we used data from public databases and our sequencing efforts to evaluate the expression levels of m6A regulators by profiling the data of whole peripheral blood of 220 SS patients and 62 healthy controls. We found that SS was associated with the expression of several m6A regulators, and this difference was correlated with activated CD4\u003csup\u003e+\u003c/sup\u003eT cells. We screened key genes with a random forest (RF) machine learning algorithm and constructed a diagnostic model of SS using multivariate logistic regression analysis. Two distinct m6A modification patterns were determined by unsupervised clustering, with significant differences in immunocyte infiltration, immune reactivity, and enriched biological functions. Key m6A regulators, gene modules, and co-expression networks of m6A-related genes were identified by conventional bioinformatics methods. This identified three key m6A regulators (\u003cem\u003eMETTL3\u003c/em\u003e, \u003cem\u003eALKBH5\u003c/em\u003e, and \u003cem\u003eYTHDF1\u003c/em\u003e) and two m6A-related hub genes (\u003cem\u003eCOMMD8\u003c/em\u003e and \u003cem\u003eSRP9\u003c/em\u003e) which may play an essential role in the diagnosis and treatment of SS. This study demonstrates the close relationship between m6A modification and the immune microenvironment in SS and provides a basis for an improved understanding of m6A modification patterns and the exploration of new therapeutic options for SS.\u003c/p\u003e","manuscriptTitle":"Integrated analysis of m6A regulator-mediated RNA methylation modification patterns and immune characteristics in Sjögren’s Syndrome","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-11-04 20:42:10","doi":"10.21203/rs.3.rs-2173202/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e53359fe-5401-43e6-9b04-017b58f1c877","owner":[],"postedDate":"November 4th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-04-01T16:10:47+00:00","versionOfRecord":{"articleIdentity":"rs-2173202","link":"https://doi.org/10.1016/j.heliyon.2024.e28645","journal":{"identity":"heliyon","isVorOnly":true,"title":"Heliyon"},"publishedOn":"2024-03-01 16:10:47","publishedOnDateReadable":"March 1st, 2024"},"versionCreatedAt":"2022-11-04 20:42:10","video":"","vorDoi":"10.1016/j.heliyon.2024.e28645","vorDoiUrl":"https://doi.org/10.1016/j.heliyon.2024.e28645","workflowStages":[]},"version":"v1","identity":"rs-2173202","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-2173202","identity":"rs-2173202","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}