Associations between immune cells signatures and osteoarthritis: An integrated analysis of bidirectional Mendelian randomization and Bayesian colocalization

• Published in: Cytokine (Philadelphia, Pa.) Vol. 179; p. 156633
• Main Authors: Teng, Menghao, Wang, Jiachen, Su, Xiaochen, Hu, Jiale, Tian, Ye, Zhang, Yingang, Li, Meng
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.07.2024
• Subjects: Bayes Theorem; Bayesian colocalization; Genetic Predisposition to Disease; Genome-wide association study; Humans; Immune cells signatures; Mendelian randomization; Mendelian Randomization Analysis - methods; Monocytes - immunology; Monocytes - metabolism; Osteoarthritis; Osteoarthritis - genetics; Osteoarthritis - immunology; Polymorphism, Single Nucleotide - genetics; PP; MR-PRESSO; GWAS; OR; MR; IVs; NEK7; Treg; Mendelian randomization; IVW; IL-1β; LYZ; Immune cells signatures; NF-κB; SD; OA; Genome-wide association study; HVEM; TNF-α; Bayesian colocalization; SNPs; PDL-1; TLR-2; Osteoarthritis
• ISSN: 1043-4666; 1096-0023
• DOI: 10.1016/j.cyto.2024.156633

•MR analysis revealed 196 immune cell signatures were associated with types of OA.•Reverse MR showed the impact of OA on the levels of 140 immune cell signatures.•OA shares genetic variants with some immune cell characteristics. Previous investigations have explored the associations between immune cell signatures and osteoarthritis (OA); however, causality remains unclear. This study employs an integrated analysis, combining bidirectional Mendelian randomization (MR) and Bayesian colocalization (Coloc), to investigate causal relationships between 731 immune cells signatures and OA, identifying shared causal variants. Utilizing publicly available summary data, this study primarily employs inverse variance weighting (IVW). Supplementary methods include MR-Egger regression, weighted median, weight mode, and simple mode. Various sensitivity tests, including Cochran’s Q test, MR pleiotropy Residual Sum and Outlier, and leave-one-out tests, were conducted to assess the robustness of the analysis results. Coloc was employed to identify shared causal genetic variants among potential associations. IVW analysis revealed 196 immune cell signatures potentially linked to OA across diverse subtypes. Reverse MR analyses indicated the causal impact of OA on the levels of 140 immune cell signatures, with subtype-specific variations. Notably, several specific associations, including CD64 on CD14-CD16 + monocyte for Hip OA (OR = 1.0593, 95 % CI: 1.0260–1.0938, P = 0.0004), HLA-DR on CD14 + CD16- monocyte (OR = 0.9664, 95 % CI: 0.9497–0.9834, P = 0.0001), HLA-DR on CD14 + monocyte (OR = 0.9680, 95 % CI: 0.9509–0.9853, P = 0.0003) in the Knee or Hip OA, PDL-1 on CD14-CD16 + monocyte by All OA (OR = 1.7091, 95 %CI:1.2494–2.3378, P = 0.0008), and herpesvirus entry mediator on effector memory CD4 + T cell by Spine OA (OR = 0.5200, 95 %CI:0.3577–0.7561, P = 0.0006) remained significant post-Bonferroni correction. Sensitivity tests validated the credibility of the IVW analysis. Additionally, Coloc revealed several potential associations among shared genetic variants, including rs115328872, rs1800973, and rs317667. Our findings provide evidence for the potential involvement of immune cell signatures in OA development, revealing avenues for early prevention and innovative therapeutic strategies.