Abstract：
BACKGROUND: Osteoarthritis is a chronic degenerative joint disease, and increasing evidences suggest that the pathogenic mechanism involves immune system and inflammation.
OBJECTIVE: The aim of current study was to uncover hub genes linked to immune infiltration in osteoarthritis synovial tissue using comprehensive bioinformatics analysis and experimental confirmation.
METHODS: Multiple microarray datasets (GSE55457, GSE55235, GSE12021 and GSE1919) for osteoarthritis in Gene Expression Omnibus database were downloaded for analysis. Differentially expressed genes (DEGs) were identified using Limma package in R software, and immune infiltration was evaluated by CIBERSORT algorithm. Then weighted gene co-expression network analysis (WGCNA) was performed to uncover immune infiltration-associated gene modules. Protein-protein interaction (PPI) network was constructed to select the hub genes, and the tissue distribution of these genes was analyzed using BioGPS database. Finally, the expression pattern of these genes was confirmed by RT-qPCR using clinical samples.
RESULTS: Totally 181 DEGs between osteoarthritis and normal control were screened. Macrophages, mast cells, memory CD4 T cells and B cells accounted for the majority of immune cell composition in synovial tissue. Osteoarthritis synovial showed high abundance of infiltrating resting mast cells, B cells memory and plasma cells. WGCNA screened 93 DEGs related to osteoarthritis immune infiltration. These genes were involved in TNF signaling pathway, IL-17 signaling pathway, response to steroid hormone, glucocorticoid and corticosteroid. Ten hub genes including MYC, JUN, DUSP1, NFKBIA, VEGFA, ATF3, IL-6, PTGS2, IL1B and SOCS3 were selected by using PPI network. Among them, four genes (MYC, JUN, DUSP1 and NFKBIA) specifically expressed in immune system were identified and clinical samples revealed consistent change of these four genes in synovial tissue retrieved from patients with osteoarthritis.
CONCLUSIONS: A 4-gene-based diagnostic model was developed, which had well predictive performance in osteoarthritis. MYC, JUN, DUSP1 and NFKBIA might be biomarkers and potential therapeutic targets in osteoarthritis.
OBJECTIVE: The aim of current study was to uncover hub genes linked to immune infiltration in osteoarthritis synovial tissue using comprehensive bioinformatics analysis and experimental confirmation.
METHODS: Multiple microarray datasets (GSE55457, GSE55235, GSE12021 and GSE1919) for osteoarthritis in Gene Expression Omnibus database were downloaded for analysis. Differentially expressed genes (DEGs) were identified using Limma package in R software, and immune infiltration was evaluated by CIBERSORT algorithm. Then weighted gene co-expression network analysis (WGCNA) was performed to uncover immune infiltration-associated gene modules. Protein-protein interaction (PPI) network was constructed to select the hub genes, and the tissue distribution of these genes was analyzed using BioGPS database. Finally, the expression pattern of these genes was confirmed by RT-qPCR using clinical samples.
RESULTS: Totally 181 DEGs between osteoarthritis and normal control were screened. Macrophages, mast cells, memory CD4 T cells and B cells accounted for the majority of immune cell composition in synovial tissue. Osteoarthritis synovial showed high abundance of infiltrating resting mast cells, B cells memory and plasma cells. WGCNA screened 93 DEGs related to osteoarthritis immune infiltration. These genes were involved in TNF signaling pathway, IL-17 signaling pathway, response to steroid hormone, glucocorticoid and corticosteroid. Ten hub genes including MYC, JUN, DUSP1, NFKBIA, VEGFA, ATF3, IL-6, PTGS2, IL1B and SOCS3 were selected by using PPI network. Among them, four genes (MYC, JUN, DUSP1 and NFKBIA) specifically expressed in immune system were identified and clinical samples revealed consistent change of these four genes in synovial tissue retrieved from patients with osteoarthritis.
CONCLUSIONS: A 4-gene-based diagnostic model was developed, which had well predictive performance in osteoarthritis. MYC, JUN, DUSP1 and NFKBIA might be biomarkers and potential therapeutic targets in osteoarthritis.
摘要：
背景：骨关节炎是一种慢性退行性关节病，越来越多的证据表明其致病机制涉及免疫系统和炎症。
目的：本研究的目的是通过综合生物信息学分析和实验证实，揭示与骨关节炎滑膜组织中免疫浸润相关的hub基因。
方法:下载基因表达综合数据库中骨关节炎的多个微阵列数据集(GSE55457、GSE55235、GSE12021和GSE1919)用于分析。在R软件中使用Limma软件包鉴定差异表达基因(DEGs),并通过CIBERSORT算法评估免疫浸润。然后进行加权基因共表达网络分析（WGCNA）以揭示与免疫浸润相关的基因模块。构建了蛋白质-蛋白质相互作用(PPI)网络来选择集线器基因，并使用BioGPS数据库分析了这些基因的组织分布。最后,这些基因的表达模式通过使用临床样品的RT-qPCR证实。
结果：共筛选出骨关节炎与正常对照的181个DEG。巨噬细胞,肥大细胞,记忆CD4T细胞和B细胞在滑膜组织中占免疫细胞构成的年夜多半。骨关节炎滑膜显示高度丰富的浸润性静息肥大细胞，B细胞记忆和浆细胞。WGCNA筛选93个与骨关节炎免疫浸润相关的DEGs。这些基因参与TNF信号通路，IL-17信号通路,对类固醇激素的反应，糖皮质激素和皮质类固醇。包括MYC在内的十个hub基因，JUN,DUSP1,NFKBIA,VEGFA,使用PPI网络选择ATF3，IL-6，PTGS2，IL1B和SOCS3。其中,四个基因(MYC，JUN,鉴定了在免疫系统中特异性表达的DUSP1和NFKBIA），临床样品显示从骨关节炎患者中提取的滑膜组织中这四个基因的一致变化。
结论：建立了基于4基因的诊断模型，在骨关节炎中具有良好的预测性能。MYC,JUN,DUSP1和NFKBIA可能是骨关节炎的生物标志物和潜在治疗靶点。
目的：本研究的目的是通过综合生物信息学分析和实验证实，揭示与骨关节炎滑膜组织中免疫浸润相关的hub基因。
方法:下载基因表达综合数据库中骨关节炎的多个微阵列数据集(GSE55457、GSE55235、GSE12021和GSE1919)用于分析。在R软件中使用Limma软件包鉴定差异表达基因(DEGs),并通过CIBERSORT算法评估免疫浸润。然后进行加权基因共表达网络分析（WGCNA）以揭示与免疫浸润相关的基因模块。构建了蛋白质-蛋白质相互作用(PPI)网络来选择集线器基因，并使用BioGPS数据库分析了这些基因的组织分布。最后,这些基因的表达模式通过使用临床样品的RT-qPCR证实。
结果：共筛选出骨关节炎与正常对照的181个DEG。巨噬细胞,肥大细胞,记忆CD4T细胞和B细胞在滑膜组织中占免疫细胞构成的年夜多半。骨关节炎滑膜显示高度丰富的浸润性静息肥大细胞，B细胞记忆和浆细胞。WGCNA筛选93个与骨关节炎免疫浸润相关的DEGs。这些基因参与TNF信号通路，IL-17信号通路,对类固醇激素的反应，糖皮质激素和皮质类固醇。包括MYC在内的十个hub基因，JUN,DUSP1,NFKBIA,VEGFA,使用PPI网络选择ATF3，IL-6，PTGS2，IL1B和SOCS3。其中,四个基因(MYC，JUN,鉴定了在免疫系统中特异性表达的DUSP1和NFKBIA），临床样品显示从骨关节炎患者中提取的滑膜组织中这四个基因的一致变化。
结论：建立了基于4基因的诊断模型，在骨关节炎中具有良好的预测性能。MYC,JUN,DUSP1和NFKBIA可能是骨关节炎的生物标志物和潜在治疗靶点。
