Role of lipid metabolites in mediating causal effects of gut microbiota on type 2 diabetes mellitus.

Using two-sample Mendelian randomization with genome-wide association study (GWAS) data from European-ancestry populations, we identified a significant mediation pathway linking genus Akkermansia to type 2 diabetes (T2D) via phosphatidylcholine (O-18:1_20:3). GWAS data were obtained from the Dutch Microbiome Project (gut microbiota), Finnish cohorts (lipid metabolites), and FinnGen (T2D). Lower levels of phosphatidylcholine (O-18:1_20:3), a protective lipid, mediated the association. The indirect effect was IE = 0.0206 (SE = 0.00187; 95% CI 0.01694-0.02428; P = 3.06 ×10⁻²⁸; OR = e^IE = 1.0208), with a total effect TE = 0.0795 (SE = 0.00767; OR = 1.0827), corresponding to a mediation proportion of 25.9%. In contrast, higher-level Verrucomicrobiaceae taxa (phylum, class, order, and family aggregates) showed weaker, non-significant mediation (95% confidence intervals including zero). We also observed that the microbial pathway PWY.7221(de novo guanosine nucleotide biosynthesis) was inversely associated with T2D (OR = 0.903; 95% CI 0.852-0.957). ediation analyses further supported phosphatidylcholine (O-18:1_20:3) as a key lipid linking genus Akkermansia to T2D. Collectively, these findings support a causal gut microbiota-lipid-T2D axis and highlight phosphatidylcholine (O-18:1_20:3) as a potential target for therapeutic strategies that modulate lipid metabolism through microbiome interventions.