The causal relationship between human blood metabolites and the risk of visceral obesity: a mendelian randomization analysis

Background We aimed to explore the causal relationship between blood metabolites and the risk of visceral obesity, as measured by visceral adipose tissue (VAT). Methods Summary statistics for 486 blood metabolites and total, as well as sex-stratified, MRI-derived VAT measurements, adjusted for body mass index (BMI) and height, were collected from previous genome-wide association studies (GWAS). A two-sample Mendelian Randomization (MR) design was used. Comprehensive evaluation was further conducted, including sensitivity analysis, linkage disequilibrium score (LDSC) regression, Steiger test, and metabolic pathway analysis. Results After multiple testing correction, arachidonate (20:4n6) has been implicated in VAT accumulation (beta = 0.35, 95%CI:0.18-0.52, P < 0.001; FDR = 0.025). Additionally, several blood metabolites were identified as potentially having causal relationship (FDR < 0.10). Among them, lysine (beta = 0.67, 95%CI: 0.28-1.06, P < 0.001; FDR = 0.074), proline (beta = 0.30, 95%CI:0.13-0.48, P < 0.001; FDR = 0.082), valerate (beta = 0.50, 95%CI:0.23-0.78, P < 0.001, FDR = 0.091) are associated with an increased risk of VAT accumulation. On the other hand, glycine (beta=-0.21, 95%CI: -0.33-0.09), P < 0.001, FDR = 0.076) have a protective effect against VAT accumulation. Most blood metabolites showed consistent trends between different sexes. Multivariable MR analysis demonstrated the effect of genetically predicted arachidonate (20:4n6) and proline on VAT remained after accounting for BMI and glycated hemoglobin (HbA1c). There is no evidence of heterogeneity, pleiotropy, and reverse causality. Conclusion Our MR findings suggest that these metabolites may serve as biomarkers, as well as for future mechanistic exploration and drug target selection of visceral obesity.