Prenatal BPA Exposure Disrupts Ovarian Follicular Wave Dynamics in Adult Sheep.

Abstract

Concerns regarding the exposure to environmental endocrine disruptors especially those with steroidogenic potential, such as bisphenol A (BPA), are escalating. Developmental exposure to BPA has been found to adversely affect reproductive function in rodents. Translating findings from a multiparous species such as rodents to the human scenario may not be ideal, when the organ system being studied does not follow a similar developmental trajectory. The similarity of neuroendocrine and ovarian changes that occur in sheep relative to human allows a better translational perspective. Prenatal BPA treatment induces reproductive defects in sheep, manifested as LH excess and dampened LH surge. In this study we hypothesized that prenatal BPA treatment will also disrupt ovarian follicular dynamics. Pregnant sheep were treated from days 30 to 90 of gestation with 3 different BPA doses (B1: 0.05, B2: 0.5, or B3: 5 mg/kg BW/day, s.c., in corn oil) or corn oil (control group [C]). At ~18 months of age, all female offspring (C, B1, B2, B3; n=7, 8, 10, and 6, respectively) were treated with 2 PGF2a injections 11 days apart during their second breeding season to synchronize estrus. Starting on day 10, transrectal ultrasonography was performed daily for 22 days to monitor ovarian follicular dynamics (follicles and corpora lutea (CL)). Blood samples were taken daily for 28 days to assess luteal progesterone dynamics during two consecutive cycles. Mantel-Haenszel chi-square test was used to compare follicle size distribution between groups. Follicular size classes were analyzed by a linear mixed effect model. Follicular wave dynamics, CL, and P4 analysis included Fisher's exact test, ANOVA with repeated measures, Mann-Whitney test, and Pearson correlation. Transrectal ultrasonography revealed that none of the three BPA doses had an effect on the number, size, or length of the CLs at the first or second progestogenic cycle studied or the correlation between CL mass and progesterone release. None of the prenatal BPA treatment groups had an effect on the total number or duration of any follicular class (2-3mm, 4-5mm, and = 6mm) or the number of ovulatory follicles relative to the C group. Number of follicular waves (FW) per cycle was restricted to 3 or 4 in C females, while it tended to be more variable in the BPA groups ranging from 2 to 5 FW (P=0.076). However, FW analysis revealed no differences in the mean number, mean day of start, or duration of the FW. Number of small, medium, or large follicles within each FW was numerically higher in C vs. BPA groups in the 1st, 2nd, and 4th FW. Number of medium follicles was higher in C relative to all other BPA groups in the 2nd FW (P<0.05). These findings suggest that the effects of prenatal BPA at the current exposure levels on follicular dynamics are subtle and are evident as variability of FW occurrence and number of emerging follicles per FW. The subtle changes in follicular dynamics coupled with defects in LH release may result in subfertility and early ovarian aging in prenatal BPA-treated females. Supported by NIH ES016541.