Multi-omics data reveal causal associations of cellular senescence-related genes in rheumatoid arthritis: A summary-data-based Mendelian randomization and co-localization analysis.

Rheumatoid arthritis (RA) is a complex autoimmune disease. Recently, cell senescence has been identified as a key factor in its pathogenesis. This study integrated multi-omics summary data and applied Mendelian randomization (MR) and co-localization analysis to systematically evaluate the causal relationships between cell senescence-related genes and RA. We collected summary data on blood methylation quantitative trait loci (mQTL), expression quantitative trait loci, and protein quantitative trait loci. The FinnGen database was the primary discovery dataset, validated by the UK Biobank and GWAS Catalog. We used the summary-data-based MR method to assess causal associations between the molecular traits of cell senescence-related genes and RA. Co-localization analysis was then performed to confirm shared genetic variants. After integrating multi-omics data on cell senescence-related mQTL and expression quantitative trait loci, we identified 5 key cell senescence-related genes potentially associated with RA: BCL2L1, DNMT3B, ERRFI1, NEK4, and RAF1. These genes demonstrated significant causal associations across multiple analyses. The mQTL signals based on summary-data-based MR analysis show that the genetically regulated methylation variations at the cg12873919 (odds ratio [OR] = 0.91, 95% CI [0.84-0.99]) and cg13989999 (OR = 0.90, 95% CI [0.82-1.00]) sites of the BCL2L1 gene are negatively associated with RA risk and may mediate disease risk by upregulating gene expression (OR = 0.82, 95% CI [0.76-0.88] and OR = 0.78, 95% CI [0.71-0.87]). Conversely, the mQTL effect size at the cg26432171 site of the RAF1 (OR = 1.17, 95% CI [1.02-1.33]) is positively associated with RA risk and is consistent with the upregulation of gene expression (OR = 1.83, 95% CI [1.49-2.25]), thereby enhancing RA susceptibility. Moreover, several sites in the DNMT3B gene (e.g., cg09149842) exhibited negative correlations with RA risk, suggesting that DNMT3B may play a critical role in RA pathogenesis by affecting gene expression. Methylation sites in ERRFI1 (cg13808198, cg22678073) and NEK4 (cg09524078) were also associated with RA risk, supporting their potential regulatory roles in RA. Co-localization analysis further validated the association between methylation sites and RA, particularly for BCL2L1, RAF1, DNMT3B, ERRFI1, and NEK4, where we identified shared causal signals with RA (posterior probability of H4 > 0.5). This study systematically evaluated the causal relationships between cell senescence-related genes and RA risk. These findings provide new insights into RA pathogenesis and reinforce the clinical value of these genes as potential therapeutic targets.