Gut microbiota, lipid metabolism, and PCOS: A Mendelian randomization and mediation analysis

Polycystic ovary syndrome (PCOS) is a common reproductive and metabolic disorder. Gut microbiota (GM) and lipid metabolism are increasingly implicated in its pathogenesis. We aimed to evaluate their causal roles and the potential mediating effect of lipid metabolites on PCOS. We conducted a 2-sample Mendelian randomization using GWAS summary statistics for 211 GM taxa, lipid metabolites, and PCOS cases from the FinnGen consortium. Instrumental variables were selected at genome-wide significance (P < 5 × 10⁻⁸, linkage disequilibrium r² < 0.001). Causal effects were estimated with inverse-variance weighted as the primary method, complemented by MR-Egger and weighted median. Mediation analysis quantified the indirect effect of lipids. Twenty-nine GM taxa showed significant causal associations with PCOS, including protective taxa (Acetobacterales, Bifidobacterium longum) and risk-enhancing taxa (Proteus, Methanobacterium B). Seventeen lipid metabolites were linked to PCOS, with phosphatidylcholines and triacylglycerols increasing risk, while sterol ester (SE [27:1/14:0]) and phosphatidylethanolamine (PE [O-16:1_20:4]) were protective. Mediation analysis indicated that taxa such as Gordonibacter exerted indirect effects on PCOS through phospholipid pathways, with mediation proportions up to 30%. This study provides evidence for a GM-lipid metabolism-PCOS axis. Dysbiosis and lipid disturbances jointly contribute to PCOS risk. Gut microbial and lipidomic signatures may serve as targets for early diagnosis and intervention. Further validation in diverse populations and experimental models is required.