Effects of castration and chronic steroid treatments on hypothalamic gonadotropin-releasing hormone content and pituitary gonadotropins in male wild-type and estrogen receptor-alpha knockout mice.

Testicular androgens are integral components of the hormonal feedback loops that regulate circulating levels of LHbeta and FSH. The sites of feedback include hypothalamic areas regulating GnRH neurons and pituitary gonadotropes. To better define the roles of androgen receptor (AR), estrogen receptor-alpha (ERalpha), and estrogen receptor-beta (ERbeta) in mediating feedback effects of sex steroids on reproductive neuroendocrine function, we have determined the effects of castration and steroid replacement therapy on hypothalamic GnRH content, pituitary LHbeta and FSHbeta messenger RNA (mRNA) levels, and serum gonadotropins in male wild-type (WT) and estrogen receptor-alpha knockout (ERKO) mice. Hypothalami from intact WT and ERKO males contained similar amounts of GnRH, whereas castration significantly reduced GnRH contents in both genotypes. Replacement therapy with estradiol (E2), testosterone (T), or dihydrotestosterone (DHT) restored hypothalamic GnRH content in castrated (CAST) WT mice; only the androgens were effective in CAST ERKOs. Analyses of pituitary function revealed that LHbeta mRNA and serum LHbeta levels in intact ERKOs were 2-fold higher than those in intact WT males. Castration increased levels of LHbeta mRNA (1.5- to 2-fold) and serum LHbeta (4- to 5-fold) in both genotypes. Both E2 and T treatments significantly suppressed LHbeta mRNA and serum LH levels in CAST WT males. However, E2 was completely ineffective, and T was only partially effective in suppressing these two indexes in the CAST ERKO males. DHT treatments stimulated a 50% increase in LHbeta mRNA and serum LH levels in WT males, whereas serum LH was significantly suppressed in DHT-treated ERKO males. Although the pituitaries from intact ERKO males contained similar amounts of FSHbeta mRNA, serum FSH levels were 20% higher than those in the intact WT males. Castration increased FSHbeta mRNA levels only in WT males, but significantly increased serum FSH levels in both genotypes. Both E2 and T treatments significantly suppressed serum FSH in CAST WT males, whereas only E2 suppressed FSHbeta mRNA. DHT treatments of CAST WT mice stimulated a small increase in serum FSH, but failed to alter FSHbeta mRNA levels. None of the steroid treatments exerted any significant effect on FSHbeta mRNA or serum FSH levels in CAST ERKOs. These data suggest that hypothalamic GnRH contents can be maintained solely through AR signaling pathways. However, normal regulation of gonadotrope function requires aromatization of T and activation of ERalpha signaling pathways in the gonadotrope. In addition, serum FSH levels in male ERKOs appear to be regulated largely by nonsteroidal testicular factors such as inhibin. Finally, these data suggest that hypothalamic ERbeta may not be involved in mediating the negative feedback effects of T on serum LH and FSH in male mice.