Developmental Programming: Postnatal Estradiol Amplifies Reproductive Defects Induced by Fetal Exposure to Excess Testosterone and Dihydrotestosterone in Sheep.

Abstract

Prenatal testosterone (T) excess causes defects in gonadotropin surge and follicular dynamics culminating in infertility. Prenatal dihydrotestosterone (DHT) treatment, while not affecting gonadotropin surge dynamics, leads to luteal defects. Considering that humans are exposed to estrogenic environmental steroids from variety of sources throughout their lifespan, we tested the hypothesis that constant exposure to estradiol-17β (E) postnatally would exaggerate the defects in periovulatory hormonal and follicular dynamics induced by prenatal T- and DHT excess. Pregnant sheep were treated with either T propionate or DHT propionate (100 mg im twice weekly) in cottonseed oil from 30 to 90 days of gestation. At nine weeks of age, a subset of control C (n=7), T (n=7), and DHT (n=6) females received a 10 mm Silastic implant containing E (produces circulating levels of <1 pg/mL E) to generate CE, TE, and DHTE groups. Six C, 7 T and 6 DHT females served as corresponding prenatal controls (no postnatal E). All animals were treated with 2 prostaglandin injections 11 days apart during their first breeding season to synchronize estrus. Blood samples were taken at 2h-intervals for 8 days to assess changes in periovulatory LH/FSH dynamics and daily for an additional 28 days to assess luteal progesterone dynamics during two cycles. Transrectal ultrasonography was performed daily to monitor ovarian follicular dynamics. Temporal changes in hormone dynamics and follicular size classes were analyzed by ANOVA, Wilcoxon test and linear mixed model. As expected, prenatal T but not DHT treatment severely dampened primary LH surges. Postnatal E treatment did not affect primary LH/FSH surge dynamics of C, T or DHT females, but obliterated secondary FSH surges in TE females. Postnatal E treatment increased (P<0.05) the incidence of subluteal cycles (≤2 ng/ml) in TE and tended to do so in DHTE females relative to their corresponding prenatal only counterparts (T: 0% vs TE: 80%, P=0.01; DHT: 16.7% vs DHTE: 50%). Postnatal E tended to increase incidence of short cycles (≤10 days) in the DHTE group (DHT: 16.7%, DHTE: 50%). Transrectal ultrasonography revealed fewer TE females developed corpora lutea relative to T (T: 85.7, TE: 14.3%, P=0.03). Postnatal E treatment tended to decrease the number of medium sized follicles (3 to ≤4mm) in all groups. Postnatal E treatment also reduced number of follicles >4 to ≤8 mm in the DHTE relative to DHT group (6.7±0.8 vs 3.7±0.9, P<0.05). All the T and 60% of the TE females had at least one luteinized follicle. Only few CE (16.6%), DHT (16.6%) and DHTE (25%) and none of C females had luteinized follicles. The luteinized follicle also tended to persist longer in the TE relative to T group (T: 31.5±3.5 vs TE: 22.1±2.4 days, P=0.07). These findings suggest that postnatal E treatment further deteriorates periovulatory hormonal/ovarian follicular dynamics and subsequent luteal outcome in prenatal steroid-treated females with the quality of prenatal steroid treatment determining the type and nature of disruption. Supported by NIH P01 HD44232.