Association between genetically determined leptin and blood lipids considering alcohol consumption: a Mendelian randomisation study.

Abstract

The objective of this study was to evaluate the association of genetically determined leptin with lipids. We conducted a Mendelian randomisation study to assess a potential causal relationship between serum leptin and lipid levels. We also evaluated whether alcohol drinking modified the associations of genetically determined leptin with blood lipids. 3860 participants of the Framingham Heart Study third generation cohort. Both genetic risk scores (GRSs), the GRS generated using leptin loci independent of body mass index (BMI) and GRS generated using leptin loci dependent of BMI, were positively associated with log-transformed leptin (log-leptin). The BMI-independent leptin GRS was associated with log-transformed triglycerides (log-TG, β=-0.66, p=0.01), but not low-density lipoprotein cholesterol (LDL-C, p=0.99), high-density lipoprotein cholesterol (HDL-C, p=0.44) or total cholesterol (TC, p=0.49). Instrumental variable estimation showed that per unit increase in genetically determined log-leptin was associated with 0.55 (95% CI: 0.05 to 1.00) units decrease in log-TG. Besides significant association with log-TG (β=-0.59, p=0.009), the BMI-dependent GRS was nominally associated with HDL-C (β=-10.67, p=0.09) and TC (β=-28.05, p=0.08). When stratified by drinking status, the BMI-dependent GRS was associated with reduced levels of LDL-C (p=0.03), log-TG (p=0.004) and TC (p=0.003) among non-current drinkers only. Significant interactions between the BMI-dependent GRS and alcohol drinking were identified for LDL-C (p=0.03), log-TG (p=0.03) and TC (p=0.02). These findings together indicated that genetically determined leptin was negatively associated with lipid levels and the association may be modified by alcohol consumption.