Although androgen receptor (AR) within myocytes is thought to mediate many of the effects of testosterone and other androgens on skeletal muscle, little is known about the functions of AR within these cells. We, therefore, studied the ultrastructure of skeletal muscle of HSA-AR transgenic (Tg) mice that overexpress AR selectively in myocytes and exhibit neuromuscular atrophy. We examined male HSA-AR mice from two different founding lines: L78 (lower copy number and less severe phenotype) and L141 (higher copy number and more severe phenotype) and compared these to wild-type (Wt) brothers. We also examined testosterone-treated female mice from these two lines and compared them both to their Wt sisters and to vehicle-treated controls. Ultrastructural examination of extensor digitorum longus sections using transmission electron microscopy revealed remarkably disorganized myofibrils in male Tg and testosterone-treated female Tg mice. Quantification of ultrastructural pathology indicated reduced myofibril width, hypertrophic and hyperplastic intermyofibrillar mitochondria, and pronounced glycogen accumulation in HSA-AR males of both lines. Reduced myofibrillar width and increases in mitochondrial number, size, and volume density were also observed in testosterone-treated HSA-AR females, although glycogen accumulation was not observed. Structural abnormalities in mitochondria were also associated with increases in electron transport chain activity and systemic resting metabolic rate, indicative of hypermetabolism. We find that overexpression of AR in myocytes of HSA-AR mice results in alterations in myofibrils, mitochondria, and glycogen. Alterations in myofibrils and mitochondria appear to result from acute actions of testosterone, whereas those on glycogen do not. Pathology of myofibrils and/or mitochondria may, therefore, mediate in part the neuromuscular atrophy observed in HSA-AR mice.

Prairie voles (Microtus ochrogaster) are exceptional among rodents in that many aspects of their brain and behavior are not masculinized by exogenous aromatizable androgens. However, the sexually differentiated endpoints studied to date rely on estrogenic mechanisms in other mammals. We examined whether sexual differentiation of an androgen receptor-dependent sex difference would be similarly distinct in prairie voles. Male mammals have more and larger motoneurons projecting to perineal muscles than do females. This sex difference normally arises from males' perinatal androgen exposure and can be eliminated by treating developing females with androgens. Gross dissection revealed bulbospongiosus muscles in adult male, but not female, prairie voles. Retrograde tracing from males' bulbocavernosus muscles and the external anal sphincter from both sexes revealed sexually dimorphic populations of labeled motoneurons in the ventral horn of the lumbar spinal cord. Similar to other rodents, males had twice as many motoneurons as females, although no sex difference in motoneuron size was detected. Unexpectedly, prenatal or early postnatal exposure to testosterone propionate had no effect on adult females' motoneuron number or size. In adulthood, gonadectomy alone or followed by chronic testosterone treatment also had no effect on females' motoneuron size or number, although castration reduced motoneuron size in males. Comparing gonadally intact weanlings confirmed that the sex difference in motoneuron number exists before adulthood. As with some other sexually dimorphic traits, and perhaps related to their unique social organization, sexual differentiation of the prairie vole spinal cord differs from that found in most other laboratory rodents.