Endocrine Abstracts

Background: Healthy skeletal muscle and adipose tissue determine physical and cardiometabolic resilience. Age-related attrition of either produces adverse health outcomes, while sentinel disorders demonstrate that their disproportion can be a hallmark either of harm [e.g. sarcopenic obesity (high adiposity, reduced lean mass), or lipodystrophy (reduced adiposity, high muscle mass)] or supranormal function [e.g. myostatin deficiency (increased muscle, low adiposity)]. We hypothesised that the genetic determinants of their relative proportions may yield insights into mechanisms maintaining healthy body composition. We thus conducted bi-trait analysis of whole-body fat-free and fat mass normalized to height (FFMI and FMI respectively), using UK Biobank bioelectrical impedance data.

Methods and Results: Variant-level associations for each trait were first determined (P < 10^-3), with resulting variants next analysed by cross-phenotype association to identify those with diverging effects on FFMI and FMI (p_pairwise< 5x10^-8). The list was finally filtered for overlap at locus level for both traits. 6 variants were thus identified in men, 5 in women, and 14 in sex-combined analysis. These were assessed for association with a wider raft of cardiometabolic and body composition traits, and the pattern of association compared with that for rare variant burden scores for potentially mediating genes. PLCE1, SLC30A10, IRS1, and WNT2B showed concordant association patterns. In a complementary approach, variant-level association with (FFMI-FMI) was studied in exome sequences. Resulting variants (P < 1x10^-3) were clumped into 500kB linkage disequilibrium blocks, and lead variants within blocks with the highest probability of overlapping association for FFMI and FMI selected. For each variant, gene burden association testing for potential mediating genes was undertaken, using both prior traits and selected serum protein measures. 43 additional genes, and replication of signals for PLCE1 and ADAMTSL3 were identified.

Conclusion: These findings give novel insights into the genetic architecture of cardiometabolic resilience and suggest numerous avenues for downstream mechanistic study.