Bioavailable testosterone and androgen receptor activation, but not total testosterone, are associated with muscle mass and strength in females

Testosterone, the major androgen, influences the reproductive and non-reproductive systems in males and females via binding to the androgen receptor (AR). Both circulating endogenous testosterone and muscle AR protein content are positively associated with muscle mass and strength in males, but there is no such evidence in females. Here, we tested whether circulating testosterone levels were associated with muscle mass, function, or the muscle anabolic response to resistance training in pre-menopausal females. Twenty-seven pre-menopausal, untrained females (aged 23.5 ± 4.8 years) underwent a 12-week resistance training programme. Muscle strength, size, power, and plasma and urine androgen hormone levels were measured. Skeletal muscle biopsies were collected before and after the training programme to quantify the effect of resistance training on AR content and nuclear localisation. Primary muscle cell lines were cultured from a subset (n = 6) of the participants' biopsies and treated with testosterone to investigate its effect on myotube diameter, markers of muscle protein synthesis and AR cellular localisation. Physiological levels of total testosterone were not associated with muscle mass or strength at baseline or with the changes in muscle mass and strength that occurred in response to resistance training in our cohort of pre-menopausal females. In contrast, bioavailable testosterone and the proportion of nuclear-localised AR were positively associated with skeletal muscle mass and strength in pre-menopausal females. In vitro, supra-physiological doses of testosterone increased myocyte diameter, but this did not occur via the Akt/mTOR pathway as previously suggested. Instead, we show a marked increase in AR nuclear localisation with testosterone administration in vitro. KEY POINTS: Total circulating testosterone was not related to muscle mass or strength before or after resistance training in pre-menopausal females. Bioavailable testosterone was positively related to exercise-induced muscle hypertrophy in pre-menopausal females. In vivo nuclear localisation of the androgen receptor was positively related to muscle mass in pre-menopausal females at baseline, but not to resistance training-induced hypertrophy. Testosterone treatment induced androgen receptor nuclear translocation but did not induce mTOR signalling in primary skeletal myocytes cultured from pre-menopausal female muscle.