Androgen receptor gene polymorphisms and the fat-bone axis in young men and women

Androgen receptor (AR) CAG(n) (polyglutamine) and GGN(n) (polyglycine) repeat polymorphisms determine part of the androgenic effect and may influence adiposity. The association of fat mass, and its regional distribution, with the AR CAG(n) and GGN(n) polymorphisms was studied in 319 and 78 physically active nonsmoker men and women (mean ± SD: 28.3 ± 7.6 and 24.8 ± 6.2 years old, respectively). The length of CAG and GGN repeats was determined by polymerase chain reaction and fragment analysis, and confirmed by DNA sequencing of selected samples. Men were grouped as CAG short (CAG(S)) if harboring repeat lengths ≤ 21, the rest as CAG long (CAG(L)). The corresponding cutoff CAG number for women was 22. GGN was considered short (GGN(S)) if GGN ≤ 23, the rest as GGN long (GGN(L)). No association between AR polymorphisms and adiposity or the hormonal variables was observed in men. Neither was there a difference in the studied variables between men harboring CAG(L) + GGN(L),CAG(S) + GGN(S),CAG(S) + GGN(L), and CAG(L) + GGN(S) combinations. However, in women, GGN(n) was linearly related to the percentage of body fat (r = 0.30, P < .05), the percentage of fat in the trunk (r = 0.28, P < .05), serum leptin concentration (r = 0.40, P < .05), and serum osteocalcin concentration (r = 0.32, P < .05). In men, free testosterone was inversely associated with adiposity and serum leptin concentration, and positively with osteocalcin, even after accounting for differences in CAG(n), GGN(n), or both. In summary, this study shows that the AR repeat polymorphism has little influence on absolute and relative fat mass or its regional distribution in physically active men. In young women, GGN length is positively associated with adiposity, leptin, and osteocalcin.