Elucidating causal relationships between circulating inflammatory factors and aging traits: Transcriptomics and Mendelian randomization analyses.

Abstract

Chronic inflammation is epidemiologically linked to aging, but the causal roles of specific inflammatory factors remain unclear. We performed bidirectional Mendelian randomization (Bi-MR) to assess causal relationships between 91 circulating inflammatory protein genes and aging traits, including telomere length, epigenetic clocks, frailty, cognitive function, and health span. Causal estimates were derived using inverse-variance weighted, with MR-Egger, weighted median, Bayesian weighted Mendelian randomization, and robust adjusted profile score (MR-RAPS) for sensitivity analysis. MRlap assessed sample overlap, and MR-PRESSO detected pleiotropy. Transcriptomic validation was performed using GTEx and GSE236927 datasets with CIBERSORT adjustment, followed by weighted gene coexpression network analysis and independent validation in GSE123697 and GSE237029. Twelve inflammatory factor genes showed significant forward associations, and 10 showed reverse associations after false discovery rate correction. Notable forward links included CASP8 with health span and IL1A and CXCL10 with telomere length. Reverse MR identified telomere length as a causal driver of IL10 mRNA levels. Transcriptomic analysis confirmed IL10 as a differentially expressed, aging-related inflammatory gene. This study provides genetic and transcriptomic evidence supporting bidirectional causal links between inflammation and aging. IL10 emerged as a robust candidate linking immune regulation to aging biology.