Identification of diagnostic genes and vital microRNAs involved in rheumatoid arthritis: based on data mining and experimental verification

Abstract

Background The pathogenesis of rheumatoid arthritis (RA) is complex. This study aimed to identify diagnostic biomarkers and transcriptional regulators that underlie RA based on bioinformatics analysis and experimental verification. Material and Methods We applied weighted gene co-expression network analysis (WGCNA) to analyze dataset GSE55457 and obtained the key module most relevant to the RA phenotype. We then conducted gene function annotation, gene set enrichment analysis (GSEA) and immunocytes quantitative analysis (CIBERSORT). Moreover, the intersection of differentially expressed genes (DEGs) and genes within the key module were entered into the STRING database to construct an interaction network and to mine hub genes. We predicted microRNA (miRNA) using a web-based tool (miRDB). Finally, hub genes and vital miRNAs were validated with independent GEO datasets, RT-qPCR and Western blot. Results A total of 367 DEGs were characterized by differential expression analysis. The WGCNA method divided genes into 14 modules, and we focused on the turquoise module containing 845 genes. Gene function annotation and GSEA suggested that immune response and inflammatory signaling pathways are the molecular mechanisms behind RA. Nine hub genes were screened from the network and seven vital regulators were obtained using miRNA prediction. CIBERSORT analysis identified five cell types enriched in RA samples, which were closely related to the expression of hub genes. Through ROC curve and RT-qPCR validation, we confirmed five genes that were specific for RA, including CCL25, CXCL9, CXCL10, CXCL11, and CXCL13. Moreover, we selected a representative gene (CXCL10) for Western blot validation. Vital miRNAs verification showed that only the differences in has-miR-573 and has-miR-34a were statistically significant. Conclusion Our study reveals diagnostic genes and vital microRNAs highly related to RA, which could help improve our understanding of the molecular mechanisms underlying the disorder and provide theoretical support for the future exploration of innovative therapeutic approaches.

Keywords

• Rheumatoid arthritis
• Weighted gene co-expression network analysis
• Diagnostic genes
• MicroRNAs