Sex-stratified GWAS of Body Fat Percentage after Adjusting for Testosterone and SHBG in the UK Biobank

IntroductionAdiposity, particularly centripetal adiposity/reduced gluteofemoral adiposity, increases dyslipidemia, type 2 diabetes (T2D) and coronary artery disease (CAD). Genome-wide association studies (GWAS) to date have identified 12 loci associated with body fat percentage (BFP). Biological sex influences both overall adiposity and fat distribution. Further, testosterone and sex hormone binding globulin (SHBG) influence adiposity and metabolic function, with differential effects of testosterone in men and women. MethodsWe performed sex-stratified GWAS of BFP in white British individuals from the UK biobank adjusting for SHBG and testosterone. We further investigated association of the identified loci with high density cholesterol (HDL), triglyceride (TG), T2D, CAD, and MRI-derived abdominal subcutaneous adipose tissue (ASAT), visceral adipose tissue (VAT) and gluteofemoral adipose tissue (GFAT) using publicly available data from large GWAS. We also performed 2-sample Mendelian Randomization (MR) using identified BFP variants as instruments to investigate causal effect of BFP on HDL, TG, T2D and CAD in males and females separately. ResultsWe identified 193 and 174 autosomal loci explaining 3.35% and 2.60% of the variation in BFP in males and females, respectively. In addition, we identified 2 Chr X loci in men. Only 38 of these loci associated with BFP in both males and females. Seven loci in men including the 2 loci on Chr X and ten loci in females have not been associated with any adiposity or cardiometabolic traits previously. The majority of BFP loci did not associate with cardiometabolic traits. Of the BFP loci associated with cardiometabolic traits several had paradoxically beneficial cardiometabolic effects with favourable fat distribution. Consistent with that sex-stratified MR analyses using identified BFP variants as instruments did not find convincing supportive evidence that increased BFP has deleterious cardiometabolic effects in either sex with highly significant heterogeneity. ConclusionsAdjusting for testosterone and SHBG with sex-stratified analyses substantially increased the number of BFP associated loci. Despite this adjustment, there was limited genetic overlap in BFP in males and females. Further identified loci in general did not have adverse cardiometabolic effects which may reflect the protective effect of favourable fat distribution and the modulation of cardiometabolic risk by testosterone and SHBG.