Opposing effects of androgen and estrogen on pituitary-adrenal function in nonpregnant primates

Maternal administration of androstenedione produces a sustained fall in maternal plasma adrenocorticotropic hormone (ACTH) concentrations in the pregnant nonhuman primate. We hypothesize a negative feedback influence on the maternal hypothalamo-pituitary-adrenal (HPA) axis by androgens in primates. This may reflect an important maternal adaptation during pregnancy in primates preventing premature induction of labor by maternal stress. However, androstenedione is precursor for placental estradiol-17beta synthesis, and infusion of androstenedione into pregnant primates elevates maternal plasma estradiol-17beta to term concentrations. Thus, it could be argued that 1) the effects attributed to androstenedione on the maternal HPA axis are mediated by estrogen rather than by androgen and 2) the negative influence of androgens may be on placental ACTH rather than, or in addition to, pituitary ACTH. To discriminate between androgenic and estrogenic effects of androstenedione on pituitary and/or placental ACTH function in primates we measured plasma ACTH, cortisol, and dehydroepiandrosterone sulfate (DHEAS) concentrations in nonpregnant baboons after treatment with either androstenedione or estradiol-17beta. Nine female baboons were studied between 14 and 22 days postpartum prior to estrous cycling. After 2 days of baseline, a continuous i.v. infusion of androstenedione (1.5 mg/kg per h in 10% intralipid, IL) was started at 0900 h and maintained for 9 days in 3 baboons. A similar protocol was carried out in another 3 baboons that received a continuous i.v. infusion of estradiol-17beta (10 microg/kg per h in 10% IL) instead of androstenedione. Three additional baboons received continuous i.v. IL vehicle alone and served as controls. Arterial blood samples (0.5 ml) for measurement of plasma hormones were taken during baseline and after 1, 3, 5, 7, and 9 days of infusion. Baseline plasma ACTH, DHEAS, and cortisol concentrations were similar among all groups. Plasma ACTH did not change during IL, increased following estradiol-17beta, and fell during androstenedione treatment. Accordingly, plasma cortisol and DHEAS concentrations were also unaltered by IL, and both steroids increased during estradiol-17beta treatment. In contrast, plasma cortisol and DHEAS remained unaltered from baseline during androstenedione treatment, despite the fall in plasma ACTH measured at this time. These data in the nonpregnant baboon 1) are consistent with negative feedback on pituitary ACTH by androgens and 2) demonstrate a positive influence on pituitary-adrenal function by estrogen in primates.