Identification Of Biomarkers And Their Functions Of Patients With Systemic Lupus Erythematosus Based On Bioinformatics

Background:Systemic lupus erythematosus(SLE)is a common autoimmune disease characterized with autoantibody production and multiple organ damage.Its pathogenesis remains unclear,and has been generally recognized to correlate to the overactivation of T and B cells and the deposition of antigen-antibody complexes.In recent years,transcriptome sequencing,methylation sequencing,single cell sequencing and other high-throughput sequencing technologies have emerged,which provide researchers with more macroscopical vision of the disease and the possibility of further analyzing the underlying mechanism of the disease occurrence and development.In the era of big data,upon analyzing the sequencing data uploaded by other researchers,we can find the general laws and key genes of the disease more quickly and accurately,so as to establish the research direction more efficiently,reduce the experimental error,obtain the research results with more scientific research value in less time,and promote the research on the pathogenesis of the disease and targeted drugs more vigorously.Objective:To screen the biomarkers of SLE through bioinformatics research on the samples of SLE patients in the public database,and further verify and explore the role of key genes in CD4~+T cell function and their specific mechanism in the disease occurrence through experiments.Methods:1.The biomarkers were screened out by bioinformatics methods in the SLE patient data set of the public database:(1)The differentially expressed genes were screened from blood samples of SLE patients and healthy control(HC)in the public database.GSE20864,GSE112087,GSE122459 and GSE144390 containing blood samples of SLE patients were retrieved from the Gene Expression Omnibus(GEO)database.The differentially expressed genes of SLE patients and HC in the above datasets were found using bioinformatics methods,and Gene Ontology(GO)and pathway enrichment analyses were performed on these differentially expressed genes using Metascape database.(2)Common differentially expressed genes were found in various target organs of SLE patients.Using the Robust Rank Aggreg package in R language to take the intersection of differentially expressed genes in(1),we could find the common differentially expressed genes.Taking the intersection of the differentially expressed genes found in the datasets of skin and kidney samples from SLE patients,we could find the key differentially expressed genes.(3)Immune and stromal cell analysis:the"ESTIMATE"package was used to score immune cells and stromal cells in kidney and skin samples of SLE patients.The blood samples of SLE patients and HC were scored by CIBERSORT,and the gene expression of 22 kinds of immune cells between SLE patients and HC were compared.GSE135779,a single cell sequencing dataset of SLE patients,was analyzed to obtain the expression level of key genes in various peripheral blood cell subsets of SLE patients and HC.(4)Epigenetic analysis:GSE96879,a methylation dataset,was analyzed to compare the DNA methylation levels between SLE patients and HC.By analyzing GSE37426,GSE102547 and GSE84655 datasets,we searched for the differentially expressed mi RNA,lnc RNA and circ RNA between SLE patients and HC,and mapped the ce RNA regulatory network according to the ce RNA hypothesis.(5)Identification of biomarkers:GSE17755 dataset includes sample data of HC,SLE patients and Rheumatoid arthritis(RA)patients.Receiver Operating Characteristic(ROC)was used to judge whether the key genes could be used as biomarkers to distinguish SLE patients from HC and RA patients.The basis of judgment was Area Under Curve(AUC).GSE100150 dataset includes sample data of SLE patients and dermatomyositis(DM)patients.Similarly,ROC was used to judge whether SLE and DM patients could be differentiated by the key genes.2.Experimental verification of biomarkers:(1)Real Time Quantitative PCR(RT-q PCR)was used to detect the expression level of biomarkers in peripheral blood mononuclear cells of HC,SLE patients and patients with other autoimmune diseases.(2)The expression level of biomarkers in the skin of SLE patients and HC was analyzed by immunohistochemistry.(3)Methylation sequencing technology was used to verify the DNA methylation level of biomarkers in B cells of SLE patients.(4)The expression level of key gene RSAD2 in T cell subsets of human tonsil was analyzed by flow cytometry.The peripheral blood samples of SLE patients and HC were collected,and the expression level of RSAD2 in peripheral blood T cell subsets was analyzed by flow cytometry.3.Study on the molecular mechanism of key genes in the pathogenesis of SLE:(1)The upstream regulatory molecules of RSAD2 were explored by interferon stimulation experiment.(2)The expression level of RSAD2 during the differentiation of Na(?)ve CD4~+T cells into Th17and Tfh cells was used to judge the effect of RSAD2 on the differentiation of Na(?)ve CD4~+T cells.(3)The effect of RSAD2 on the induction of differentiation of Na(?)ve CD4~+T cells into Th17 cells was explored in Na(?)ve CD4~+T cells with the knockdown of RSAD2.(4)The effect of RSAD2 on the induction of differentiation of Na(?)ve CD4~+T cells into Tfh cells was explored in Na(?)ve CD4~+T cells with the knockdown of RSAD2.Results:1.Screening of biomarkers:(1)432 high-expressed genes and 25 low-expressed genes were screened in GSE20864 dataset,918 high-expressed genes and 427low-expressed genes were found in GSE112087 dataset,571high-expressed genes and 670 low-expressed genes were screened in GSE122459 dataset,and 386 high-expressed genes and 60 low-expressed genes were searched in GSE144390 dataset.These differentially expressed genes were related to the regulation of innate immune response,cytokine signals in the immune system and the regulation of cytokine production according to the bioinformatics analysis based on Metascape database.(2)It was found that 26 genes appeared at least three times in the four differentially expressed gene sets using the"Robust Rank Aggreg"package,of which 19 upregulated genes were obversed in the blood samples of SLE patients using the mcode plugin in string database and Cytoscape software.1000 high-expressed genes were screened from the skin sample dataset GSE109248 of SLE patients and 196 high-expressed genes were screened from the kidney sample dataset GSE32591 of SLE patients.After taking the intersection,13 genes were obtained.These genes were highly expressed in blood,skin and kidney samples of SLE patients,and associated with interferon signaling pathway based on enrichment analysis.(3)According to the ESTIMATE score,it was found that the immune cell scores of both skin and kidney samples of SLE patients were higher than those of normal samples,and only kidney samples of SLE patients showed higher stromal cell scores.Meantime,the analysis found that there were differences in interferon receptor gene expression between SLE and HC kidney samples compared with skin samples.The results of CIBERSORT showed that compared with HC,the proportion of Na(?)ve B cells in SLE patients decreased,but the proportion of dendritic cells(DCs)increased.Single cell sequencing analysis revealed that compared with HC,SLE patients had high expression of key genes in cell subsets such as proliferative T cells(pro T),plasma B cells,plasmacytoid dendritic cells(p DCs)and monocytes.(4)41 hypomethylated genes and 1 hypermethylated gene were screened in GSE96879 dataset.Enrichment analysis showed that they were closely related to type I interferon signal pathway.According to the ce RNA hypothesis,8 low-expressed mi RNAs,25 high-expressed lnc RNAs and 2 high-expressed circ RNAs were found in SLE patients,and the ce RNA network map was constructed.(5)The results of ROC and AUC have proved that RSAD2,OAS2,IFIT1,IFIT3,PLSCR1,IFI44 and IFI44L could well distinguish SLE from HC,RA and DM.2.Verification of biomarkers:(1)The results of RT-q PCR verified that RSAD2,OAS2,IFIT1,IFIT3 and PLSCR1 were significantly overexpressed in SLE patients(P<0.05).(2)Immunohistochemistry also proved that RSAD2 and IFIT1 were highly expressed in the skin of SLE patients(P<0.0001).(3)Methylation sequencing analysis showed that DNA hypomethylation occurred in RSAD2,IFIT1,IFIT3,PLSCR1,IFI44 and IFI44L in B cells of SLE patients,and some genes even appeared multiple hypomethylation sites(P<0.05).3.Effects of key genes on T cell differentiation:(1)In tonsil and peripheral blood,flow cytometry showed that RSAD2 was highly expressed in activated CD4~+T cells and CD8~+T cells.Compared with HC,the expression of RSAD2 in peripheral blood T cell subsets of SLE patients was significantly higher.Among the subsets of CD4~+T cells in peripheral blood of SLE patients,the increase of RSAD2in Tfh-like cells,Th2 cells and Th17 cells was the most remarkable(P<0.0001).(2)The results of in vitro stimulation showed that,IFN-γhad the strongest effect on promoting RSAD2 expression at 48 h(P<0.05),while IFN-αhad the strongest effect on promoting RSAD2 expression at 72 h(P<0.05).(3)During the differentiation of Na(?)ve CD4~+T cells into Th17 and Tfh cells,the expression of RSAD2 increased continuously over time.(4)With the knockdown of RSAD2 using Smart Silencer,the proportion of Na(?)ve CD4~+T cells differentiating into Th17 and Tfh cells decreased significantly(P<0.05).Conclusion:13 high-expressed key genes in skin,kidney and blood of SLE patients were screened by joint analysis of the results of multiple datasets.RSAD2,OAS2,IFIT1,IFIT3,PLSCR1,IFI44 and IFI44L may be regulated by epigenetics,and the 7 key genes had good potential as biomarkers of SLE.The expression levels of 7 biomarkers in SLE patients were analyzed by RT-q PCR,immunohistochemistry and methylation sequencing.The expression of RSAD2 was regulated by IFN-γand IFN-α.RSAD2 could affect the differentiation of Na(?)ve CD4~+T cells into Th17 and Tfh cells and provide a potential target for the diagnosis and treatment of SLE.