骨关节炎滑膜组织衰老相关生物标志物的筛选及免疫细胞浸润分析

Screening of senescence-related biomarkers and analysis of immune cell infiltration in synovial tissue of osteoarthritis

目的利用生物信息学方法筛选和验证骨关节炎(OA)滑膜组织衰老相关诊断标志物。方法从GEO数据库获取GSE206848、GSE55235和GSE554573个骨关节滑膜组织数据集,筛选出GSE206848差异表达基因(DEGs),从CellAge数据库获取衰老相关基因(SRGs),将DEGs和SRGs取交集,获得衰老相关差异表达基因(DESRGs),进行基因本体(GO)、京都基因和基因组百科全书(KEGG)分析和使用STRING在线数据库及Cytoscape软件进行蛋白互作网络(PPI)分析,再利用CytoHubba插件结合随机森林算法筛选出SRGs,进行曲线下面积(AUC)预测值和基因表达水平的验证,并通过GSE55235和GSE55457数据集对关键基因进行验证,通过CIBERSORT进行免疫浸润分析。结果最终得到30个DESRGs,GO富集在RNA聚合酶Ⅱ的调控转录调控、核质、蛋白结合等过程,KEGG主要富集在信号通路“细胞衰老、内分泌抵抗、乳腺癌”等通路上,SNAI1、NOX4、DHX9为滑膜细胞衰老相关诊断标志物。OA滑膜组织含有较少的静息记忆CD4^(-)T细胞(P=0.007),且诸多免疫细胞间存在相关性(P<0.05)。诊断标志物NOX4与多种免疫细胞存在相关性(P<0.05)。结论利用生物信息学分析筛选出OA滑膜组织衰老相关诊断标志物,发掘免疫细胞浸润情况及与诊断标志物间的关系,有助于了解OA的发病机制和为未来的靶向治疗提供研究方向。

Objective To screen and validate the aging-related diagnostic markers of synovial tissue in osteoarthritis(OA)by bioinformatics methods.Methods GSE206848,GSE55235 and GSE554573 datasets of bone and joint synovial tissue were obtained from Gene Expression Omnibus(GEO)database and GSE206848 differentially expressed genes(DEGs)were screened.Senescence-associated genes(SRGs)were downloaded from the CellAge database.Differentially expressed senescence-related genes(DESRGs)were obtained by intersection of DEGs and SRGs.Gene Ontology(GO),Kyoto Encyclopedia of Genes and Genomes(KEGG)were analyzed,and protein-protein interaction(PPI)was analyzed by using STRING online database and Cytoscape software.The predicted area under curve(AUC)value and gene expression level were verified.The key genes were verified by GSE55235 and GSE55457 datasets,and the immune infiltration analysis was performed by CIBERSORT.Results Finally,30 DESRGs were obtained.GOs were enriched in the regulation of transcription regulation,nucleoplasm and protein binding of RNA polymeraseⅡ,while KEGGs were mainly enriched in the signaling pathways such as“cellular senescence,endocrine resistance,and breast cancer”.SNAI1,NOX4 and DHX9 were identified as synovial cell senescence-related diagnostic genes.OA synovial tissue contained fewer resting memory CD4^(-)T cells(P=0.007),and there was a correlation between immune cells(P<0.05).Diagnostic marker NOX4 was correlated with a variety of immune cells(P<0.05).Conclusion Bioinformatics analysis is used to screen out the diagnostic markers related to synovial aging in OA,which is helpful to understand the pathogenesis of OA and provide research directions for targeted therapy in the future.