Spermatogenesis without gonadotropins: maintenance has a lower testosterone threshold than initiation.

We showed previously that testosterone (T) alone could induce spermatogenesis and produce normally fertile spermatozoa in the absence of circulating gonadotropins. These studies used the hpg mouse, which is characterized by a congenital gonadotrophin deficiency due to a major deletion in the GnRH gene. Administering T by a subdermal implant of a SILASTIC brand tube impregnated with crystalline T showed that the androgenic requirement for full induction of spermatogenesis was a 1-cm length implant. Using this unique model of spermatogenesis without gonadotropins, we have now investigated the quantitative requirement for androgens to maintain spermatogenesis by testing the hypothesis that the androgenic threshold required for induction and maintenance of spermatogenesis are the same. Spermatogenesis was induced in homozygous hpg mice by T administration for 6 weeks. The first experiment determined the time-course of the regression of spermatogenesis after removal of the T-impregnated SILASTIC brand implant. Elongated spermatids were absent by 3 weeks and testicular weight regression was maximal by 4 weeks after androgen withdrawal. The second experiment examined the effects on maintenance of spermatogenesis of reducing the T dose. After full induction of spermatogenesis in homozygous hpg mice, the T implants were replaced with a range of smaller size T-impregnated SILASTIC brand implants for a further 4 weeks. All androgen-sensitive end-points (testis weight, tubular, and luminal diameters, round spermatids) were fully maintained with T implants of 0.06 cm and elongated spermatids with T implants of 0.25 cm. A further experiment showed that at very low T doses (0.06, 0.125 cm) the T effects observed at 4 weeks were maintained at 6 and 11 weeks duration. We conclude that the androgenic threshold to maintain spermatogenesis in the mouse is an order of magnitude lower than the threshold required for inducing spermatogenesis. This distinction suggests that the mechanism of action of testosterone in inducing spermatogenesis may involve regulation of a genetic switch to complete meiosis, whereas the maintenance involves a different locus of action. These findings suggest that further studies of androgen-dependent meiotic genes may be central to understanding the regulation and molecular basis of androgen-driven induction and maintenance of spermatogenesis.