Abstract
Gestational testosterone (TS) excess, acting via both the androgenic and estrogenic pathways, advances puberty and disrupts the neuroendocrine estradiol (E2) feedback and periovulatory hormonal dynamics in female sheep. These prenatally programmed defects may be subject to postnatal modifications by continued organizational and/or activational effects of steroids. This study investigated (1) the organizational contribution of prenatal estrogen excess and (2) the impact of postnatal exposure to E2 in modulating the effects of prenatal androgen excess (TS and dihydrotestosterone (DHT)) on puberty, neuroendocrine feedback mechanisms, and periovulatory hormonal dynamics in sheep. Pregnant Suffolk sheep were treated with TS, DHT, E2, or E2 plus DHT (ED) from days 30 to 90 of gestation. A subset of the control (C), TS, and DHT female offspring received a constant-release E2 implant postnatally. Findings revealed that (1) prenatal E2-treatment failed to reproduce the neuroendocrine disruptions predicted to be programmed by the estrogenic pathway and (2) prenatal E2D-treatment did not adequately replicate the reproductive neuroendocrine defects induced by prenatal TS excess. More importantly, continuous postnatal E2-treatment, while delaying the onset of puberty and reducing the inhibitory effects of E2 on tonic luteinizing hormone (LH) release, failed to amplify the E2-positive feedback and periovulatory defects induced by prenatal TS-treatment. Our results indicate that disruptions in E2-positive feedback mechanisms and periovulatory gonadotropin secretion induced by prenatal TS-treatment are programmed predominantly during the prenatal life with postnatal exposure to E2 excess not contributing further to these disruptions.
Introduction
Many diseases in adults, such as obesity, metabolic syndrome, and polycystic ovary syndrome (PCOS), may have their origins during fetal life. As proposed by Barker’s developmental origins of adult disease hypothesis (Barker 2004), adverse prenatal insults can cause permanent changes in the physiology of the developing fetus leading to pathology during adulthood. These insults include inadvertent exposure to hormonal, nutritional, and environmental agents (Rhind et al. 2001, Tang & Ho 2007, Dumesic et al. 2014). The effects of increased exposure to native steroids in disease states (New 2006, Rosenfield 2007) and to environmental endocrine-disrupting chemicals (EDC) that can bind to steroid receptors (Vaiserman 2014) have been the focus of intense research. Due to the high rate of tissue differentiation and metabolism, a developing fetus is extremely sensitive to EDC exposures even at concentrations far below those detrimental to adults (Diamanti-Kandarakis et al. 2009, Schug et al. 2011).
Congenital adrenal hyperplasia (CAH) (New 2006) and PCOS (Sir-Petermann et al. 2002) are disease states characterized by elevated testosterone levels, among other endocrine imbalances. Experimentally, prenatal TS-treatment in various species, including sheep, leads to dysfunctions during adult life that resemble those seen in women with PCOS (Padmanabhan & Veiga-Lopez 2013a, b, Dumesic et al. 2014). For instance, prenatal TS-treated sheep show progressive loss of cyclicity (Clarke et al. 1977, Birch et al. 2003, Manikkam et al. 2006) and disrupted periovulatory hormonal dynamics (Veiga-Lopez et al. 2008), with defects evident at both the neuroendocrine and ovarian levels (Padmanabhan & Veiga-Lopez 2013b). The neuroendocrine disruptions induced by prenatal TS excess include reduced sensitivity to estradiol (E2) and progesterone (P) negative feedback (Wood & Foster 1998, Robinson et al. 2002, Sarma et al. 2005), dampened or absent E2-positive feedback, and disrupted periovulatory hormonal dynamics (Sharma et al. 2002, Unsworth et al. 2005, Veiga-Lopez et al. 2009). Comparative studies with dihydrotestosterone (DHT), a nonaromatizable androgen (Veiga-Lopez et al. 2009), as well as TS+ androgen antagonist-treated sheep (Jackson et al. 2008) provided evidence that E2-negative, but not E2-positive, feedback perturbations are programmed by androgenic actions of TS. This led to the premise that E2-positive feedback and periovulatory hormonal dynamics are likely programmed by the estrogenic actions of TS (Foster et al. 2006).
Although the organizational period may extend beyond birth (Jackson et al. 2013), evidence suggests that activational effects of the postnatal environment can modify the impact of such organizational changes. For instance, the organizational effects of excess prenatal steroid hormone exposure in masculinizing the behavioral circuits in the males become only apparent during puberty when gonadal steroid output determines the expression of sex-typical behavior (Schulz et al. 2009). This can be explained by the two-hit hypothesis, which states that an early-life adverse event (“first-hit”) programs a pathological condition that may be revealed only later in life by a subsequent exposure to an adverse influence or the so-called ‘‘second hit’’ (Bayer et al. 1999, Tang et al. 2008, Puttabyatappa et al. 2015). From an adverse exposure standpoint, humans are exposed throughout their lifespan to steroids involuntarily and sometimes unknowingly through food consumption (phytoestrogens), industrial byproducts (bisphenol A), diseases (CAH, PCOS), and/or voluntarily (contraception and anabolic steroids) (Bahrke et al. 1998, Jefferson et al. 2012, Pignatelli 2013, Peretz et al. 2014, Jensen et al. 2015). Such exposure(s) can be deleterious to fertility and may serve to mask/unmask or reduce/amplify prenatally programmed functions. For example, postnatal overfeeding (second hit) of prenatally TS-treated (first hit) sheep exacerbated the reproductive cyclicity defects, with a majority of animals ending their breeding season early (Steckler et al. 2009). Similarly, postnatal exposure to E2 (second hit) amplified the ovarian defects induced by prenatal TS excess (first hit) (Veiga-Lopez et al. 2014).
The objective of this study was to investigate the contributions of prenatal and postnatal E2 in programming and amplifying reproductive neuroendocrine defects. Specifically, this study tested the following hypotheses: (1) prenatal E2-treatment alone induces neuroendocrine defects that have been postulated to be programmed by the estrogenic effects of prenatal TS-treatment, (2) prenatal ED-treatment replicates the effects of prenatal TS-treatment, and (3) continuous postnatal E2 exposure amplifies the effects of prenatal androgen excess.
Materials and methods
All procedures used in this study were approved by the University of Michigan Animal Care and Use Committee and are consistent with the National Research Council’s Guide for the Care and Use of Laboratory Animals.
Generation of experimental animals
The study was conducted at the University of Michigan Sheep Research Facility, Ann Arbor, MI, USA. Adult Suffolk ewes purchased from local farmers were mated with Suffolk rams of proven fertility. General husbandry and nutrition have been described previously (Manikkam et al. 2004). Starting at 2–3 weeks before breeding, ewes were group-fed daily with 0.5kg shelled corn and 1.0–1.5kg alfalfa hay/ewe. Once mated, females were assigned randomly to the different treatment groups and housed under a natural photoperiod in the pasture and group-fed with a daily maintenance diet of 1.25kg alfalfa/brome mix hay/ewe. Lambs born to these ewes were fed a pelleted diet (Shur-Gain, Strykersville, NY, USA) consisting of 3.6MCal/kg digestible energy and 18% crude protein. At ~8 weeks of age, lambs were weaned and maintained outdoors. They were fed ad libitum until they attained ~40kg of body weight, at which point they were switched to a diet consisting of 2.3MCal/kg digestible energy and 11.3% crude protein. The treatment groups used in the study and the various studies performed are summarized in Fig. 1.
The lambs used in the study received either no treatment prenatally (control; C) or were treated prenatally from gestational days 30 to 90 with TS (TS-treated), DHT (D-treated), estradiol (E2-treated), or ED (ED-treated) (Fig. 1A). The number of female offspring born in each treatment group was C=13, E2=11, ED=6 and 7 each for TS and D prenatal treatment groups. A subset of prenatal C, TS-, and D-treated females received E2 implants postnatally starting at ~12 weeks of age. Sample size of postnatal groups was 7 for C+E2 and TS+E2, and 6 for the D+E group (Fig. 1A). There were insufficient animals to generate postnatal treatment groups from prenatal E2- and ED-treated animals. Changes in ovarian follicular dynamics in this cohort of animals have been reported previously (Veiga-Lopez et al. 2014).
Prenatal groups
Prenatal TS- and D-treated female sheep were generated as reported previously (Wood et al. 1991). Briefly, pregnant Suffolk ewes were injected (intramuscular) twice weekly from 30 to 90 days of gestation with either 100mg TS propionate (1.2mg/kg; Sigma–Aldrich) or 100mg DHT propionate (Steraloids, Newport, RI, USA) suspended in 2mL cottonseed oil. Females in the C group did not receive vehicle because no differences in reproductive parameters were found between vehicle-treated and untreated controls in our previous study (Veiga-Lopez et al. 2008). Prenatal E2-treated animals were generated by placing a 30mm SILASTIC implant (Dow Corning, Midland, MI, USA) filled with crystalline E2 subcutaneously in the axillary region for the same duration as described before (Jackson et al. 2008). Control females did not receive sham implants. The efficacy of this approach in suppressing LH secretion in sheep has been well established (Karsch et al. 1993). The 30mm implant produces circulating E2 concentrations of ~3pg/mL, similar to those seen during the follicular phase in the female sheep (Goodman et al. 1980, 1981, Jackson et al. 2013). The prenatal ED-treated group received E2 implants along with twice weekly injections of DHT.
Postnatal treatment
Approximately 12 weeks after birth, half of the prenatal C-, TS-, and D-treated female lambs received 10mm E2 implants, which produces circulating concentrations of E2 at ~1pg/mL (Goodman et al. 1980, 1981, Jackson et al. 2013). Figure 1 summarizes the treatment groups and the type and sequence of studies conducted.
Assessment of puberty
Beginning at ~21 weeks of age, P concentrations in blood samples collected twice a week were assessed in all animals. Age at puberty was defined as the age at the first sample having circulating concentrations of P greater than 0.5ng/mL.
Testing E2-negative feedback responses
This test was conducted during the prepubertal period (~18 weeks of age; Fig. 1B), when female lambs are extremely sensitive to the E2-negative feedback (Foster & Jackson 2006). Details of the E2-negative feedback test are summarized in Fig. 1C. The gonadotropin-releasing hormone antagonist (GnRH-A), acyline (Contraception and Reproductive Health Branch, National Institutes of Health, Bethesda, MD, USA), was administered subcutaneously (10µg/kg) every 12h for 72h to abolish LH pulsatility and reduce endogenous E2 concentrations (Sarma et al. 2005, Steckler et al. 2008). Blood samples were collected at 20min intervals for 6h after the last GnRH-A injection to determine the GnRH-A ablation of LH pulsatility (Period 1; Fig. 1C). To establish the impact of reduced endogenous E2 on LH pulse frequency, blood samples were collected for 6h beginning 72h after cessation of the GnRH-A treatment (Period 2; Pre E2; Fig. 1C). The E2-negative feedback was then evaluated by inserting a single 3mm SILASTIC E2 implant (subcutaneously) to produce circulating concentrations of <1pg/mL (Goodman et al. 1980, 1981, Jackson et al. 2013). Blood samples were collected beginning 72h after insertion of the E2 implant for 6h at 20min intervals (Period 3; Post E2; Fig. 1C).
Testing E2-positive feedback responses
This test was carried out following the E2-negative feedback testing (Fig. 1C) in prepubertal females at ~19 weeks of age in the absence of P priming, as reported previously (Sharma et al. 2002). All animals received four 30mm SILASTIC E2 implants (subcutaneously) to provide late follicular phase concentrations of E2 (Goodman et al. 1980, 1981, Jackson et al. 2013). Blood samples were collected every 2h for 72h starting 1h before insertion of E2 implants.
Periovulatory hormonal dynamics
The impact of experimental treatments on periovulatory hormonal dynamics was studied during the first and second breeding season at ~8 and ~20 months of age respectively (Fig. 1B). Studies were carried out during two breeding seasons to determine whether these animals show a progressive loss in cyclicity as observed previously in prenatal TS-treated animals (Manikkam et al. 2006). Estrous was synchronized with two injections of prostaglandin F2α (PGF2α, 10mg, intramuscular; Lutalyse, Pfizer Animal Health, Florham Park, NJ, USA) administered 11 days apart (Fig. 1D). Following the second PGF2α injection, blood samples were collected every 2h for 192h during the first breeding season and for 120h during the second breeding season to assess the periovulatory hormonal changes. This was followed by daily blood sampling for 19 additional days to assess luteal P secretion. LH and FSH were assayed every 2h, E2 every 4h, and P in daily samples.
Hormone measurements
Plasma LH, FSH and P concentrations were measured in duplicate using validated radioimmunoassays (RIA) (Niswender et al. 1969, Padmanabhan et al. 1995, 1997). The sensitivity of LH RIA was 0.1±0.01ng/mL (n=46 assays; mean±s.e.m.). Mean intra-assay coefficient of variation (CV) based on four quality control pools measuring 3.1±0.1, 7.2±0.1, 13.4±0.1, and 23.8±0.01ng/mL were 10.0, 6.0, 6.8, and 6.0% respectively. The corresponding interassay CVs averaged 20.5, 7.2, 5.6, and 4.5%. The sensitivity of FSH RIA was 0.06±0.01ng/mL (n=31 assays). The intra-assay CVs based on two quality control pools measuring 5.0±0.1 and 11.2±0.2ng/mL were 9.9 and 4.8% respectively. The corresponding interassay CVs averaged 13.9 and 8.7%. Plasma E2 for the E2-negative feedback study was measured using the commercially available MAIA kit (Polymedco Inc., Courtlandt Manor, NY) and for the periovulatory hormonal dynamics assessment was measured using a validated RIA (Rozell & Keisler 1990). For the MAIA kit sensitivity of E2 assay was 0.1±0.02pg/mL (n=6 assays). Mean intra-assay CVs based on two quality control pools measuring 1.09±0.08 and 5.30±0.054pg/mL was 14.22 and 13.76% and the corresponding inter-assay CVs averaged 17.02 & 24.59%. For the periovulatory hormonal dynamics study sensitivity of E2 RIA was 0.3±0.1pg/mL (n=36 assays). Mean intra-assay CVs based on two quality control pools measuring 0.53±0.02 and 3.32±0.15pg/mL were 12.8 and 12.6%, respectively. The corresponding inter-assay CVs averaged 18.36 and 25.38%. The sensitivity of P RIA (Coat-a-Count, DPC/Siemens) was 0.001±0.0002ng/mL (n=8 assays). Mean intra-assay CVs based on three quality control pools measuring 0.2±0.006, 1.7±0.02, and 14.3±0.2ng/mL were 4.8, 5.1, and 2.4% respectively. The corresponding interassay CVs averaged 12.2, 3.6, and 3.8%. Sample sets from experimental groups were randomly distributed such that each assay included samples from all treatment groups.
Statistical analysis
Age at puberty was assessed in all animals. Results from twin females were averaged for assessment of puberty before analysis to ensure mother was the experimental unit. Only a subset of animals was used for E2-negative and E2-positive feedback studies (n=7 each for C, TS, D, E2, C+E2, and TS+E2, and n=6 each for ED and D+E) and for assessment of periovulatory hormonal dynamics (n=6 each for C, TS, D, ED, C+E2, and D+E groups, n=7 for E2 and n=5 for TS+E2). During the second breeding season, only animals from C, TS, and E2 for prenatal treatment groups and C+E2 and TS+E2 for postnatal groups were available for this study. The number of animals in each treatment group during their second year were C=9, TS=5, E2=9, C+E2=7, and TS+E2=4. For the feedback and periovulatory dynamic studies, when twin female births were involved, only one offspring (selected randomly) was used.
For E2-negative feedback, LH values from serially collected samples were subjected to pulse analysis using the cluster algorithm (Veldhuis & Johnson 1986). For cluster analysis, the minimum number of data points to identify either a peak (the highest concentration reached during a pulse) or a nadir (basal level) for the 20min sampling frequency was set at one. The Student’s t statistic values used to identify significant increases from preceding nadirs and decreases to following nadirs were set at 1.0. Increases in LH concentrations more than two times the assay sensitivity over the preceding nadir were considered as a pulse. The number of pulses was assessed over a 6h period. For both E2-positive feedback and periovulatory hormonal characterization, the LH and FSH surges were defined based on the circulating gonadotropin concentrations being above the baseline plus two times the assay sensitivity and remaining high for at least 8h (Padmanabhan et al. 2015). For synchronization of the estrous cycle following PGF2α injection, daily P concentrations had to fall below 0.5ng/mL. For all statistical analysis, P<0.05 was considered significantly different.
Age at puberty
The age at puberty was analyzed by one-way ANOVA for prenatal treatment group comparisons and a two-way ANOVA for testing postnatal E2 modulation. In view of the large number of comparisons that reduces power, the magnitude of treatment effects on age at puberty was also examined by effect size analysis (Cohen 1992, Nakagawa & Cuthill 2007, Padmanabhan et al. 2015). This analysis allows comparison of the means between two treatments with respect to the magnitude of difference between them. The computed statistic is Cohen’s d value, and values above 0.2, 0.5, and 0.8 were considered as small, medium, and large effect sizes, respectively (Cohen 1992, Nakagawa & Cuthill 2007).
E2-negative feedback
Variables compared were E2 concentrations and LH pulse frequency. Number of pulses during the 6h period was square-root-transformed before analysis. For comparison among prenatal groups, ANOVA was used followed by post hoc analyses adjusting for multiple comparisons. For postnatal comparisons, paired t-test was used to compare postnatal groups with corresponding prenatal-only treated groups. To test whether number of pulses changed significantly between pre-E2 (72h post-GnRH-A) and post-E2 period in each treatment groups, Wilcoxon signed-rank test was used. In addition, Cohen’s effect size analysis was used as a secondary analysis to relate the magnitude of differences.
E2-positive feedback and periovulatory dynamics
Comparisons involved peak time, peak hormone concentration, total hormone released during the surge, the duration of surge, and time from E2 peak to LH peak. All continuous variables were log-transformed before analysis. For comparing categorical variables (% synchronized and % showing LH surges in the periovulatory dynamics study and % responded in the positive feedback study) between prenatal groups and each of the postnatal pairs, Fisher’s exact test was used. For comparison of LH and FSH surge dynamics among only prenatal treatment groups, ANOVA was used followed by post hoc tests after adjusting for multiple comparisons. For variables with only one TS subject providing data in E2-positive feedback test or two TS subjects in periovulatory hormonal dynamics examination, given all other prenatal groups are not significantly different, the mean and standard deviation (SD) of the outcome using combined data from the C, D, E2, and ED groups were first calculated. Then, the probability of observing a value larger or equal to the value observed in the single TS subject assuming it is drawn from a normal distribution with the above derived mean and SD was calculated. A smaller P-value indicates that TS subject was less likely to have the same distribution as other groups. SFor prenatal vs postnatal treatment comparisons, unpaired t-test was used.
Results
Effects of prenatal treatments
E2-negative feedback responses
The results of the E2-negative feedback study in animals that received only prenatal treatments are summarized in Fig. 2A and B, with representative circulating LH profiles shown in Supplementary Fig. 1, see section on supplementary data given at the end of this article. Administration of GnRH-A reduced E2 concentrations to ~0.3pg/mL in all groups (data not shown). Approximately 72h after the last injection of GnRH-A, endogenous E2 concentrations had increased slightly but were not statistically different among prenatal treatment groups (Fig. 2A, Pre E). Insertion of a 3mm E2 implant caused the concentration of E2 to increase to ~0.6pg/mL across all prenatal treatment groups (Fig. 2A, Post E). Effect size analysis found a large size effect of E2 implants in C, D, and ED groups, but not in TS and E2 groups. Administration of GnRH-A reduced the LH pulsatility in all prenatal treatment groups (data not shown). Seventy-two hours after cessation of GnRH-A injection, LH pulsatility increased in all prenatal treatment groups, and LH pulse frequency did not differ among groups (Fig. 2B, Pre E2). E2-treatment ablated LH pulses in C animals (Fig. 2B, Post E2). Prenatal TS-treated animals had significantly more LH pulses during the E2-treatment period than those of C females (Fig. 2B, Post E2). A reduction in LH pulses was evident in TS-, D-, and E2-treated animals relative to pre E2 period, albeit to a lesser degree than in C animals (Fig. 2B, Post E2). While not reaching statistical significance, five of six animals showed suppression following E2-treatment in the ED group (Fig. 2B, Post E2).
Age at puberty
The large number of treatment comparisons relative to the sample size precluded identification of statistical differences among prenatal groups (Fig. 2C). However, Cohen’s analysis found a large size effect between C and prenatal TS, E2, and ED groups, but not D treatment.
E2-positive feedback responses
Results of the E2-positive feedback test in prenatal groups are summarized in Fig. 3 and representative hormonal profiles are depicted in Supplementary Fig. 2. While all females in the C and prenatal D-, E2- and ED-treated groups demonstrated gonadotropin surges, only one prenatal TS-treated female responded to the E2-positive feedback challenge (Fig. 3A). Mean total release (Fig. 3B and F), peak time (Fig. 3C and G), peak concentrations (Fig. 3D and H), and duration (Fig. 3E and I) of the LH/FSH surges did not differ among C, D, E2, and ED groups. The only prenatal TS-treated female that responded to the E2-positive feedback showed a delayed and reduced gonadotropin surge.
Periovulatory hormonal dynamics
Results of the periovulatory hormonal dynamics during the first and second breeding season in prenatal groups are summarized in Fig. 4 and representative hormonal profiles are depicted in Supplementary Figs 3 and 4 respectively. During the first breeding season, 100% of the animals were synchronized in the C and E2 groups, while 66.7, 83.3, and 83.3% of the animals showed estrous synchronization following PGF2α administration in TS, D, and ED groups respectively (Fig. 4A). Of the female sheep that synchronized, 100% of the animals showed LH surges in C, D, and E2 groups, while only two of four and four of five animals showed definable LH surges in the TS and ED groups respectively (Fig. 4F). No differences in mean total release (Fig. 4B and G), peak time (Fig. 4C and H), peak gonadotropin concentration (Fig. 4D and I), and duration (Fig. 4E and J) were observed between C, E2, D, and ED groups. Of the two prenatal TS-treated animals that met the criteria of an LH surge (sustained increase in LH for over 8h), total LH was lower than that of the control animals, but no difference was observed with other surge parameters.
During the second breeding season, 88.9, 60.0, and 66.7% of the animals from C-, TS-, and E2-treatment groups synchronized following PGF2α administration respectively (Fig. 4A). Of these, 100% of the animals in C and E2 groups while none of the animals in the TS group showed definable LH surges (Fig. 4F). No differences in the gonadotropin surge characteristics were observed between C and E2 groups (Fig. 4B, C, D, E, G, H, I and J).
Effects of postnatal E2 exposure
E2 negative feedback responses
The impact of postnatal E2-treatment on modulating the effects of prenatal steroid treatment on E2-negative feedback is summarized in Fig. 5A and B and representative circulating LH profiles in Supplementary Fig. 1. GnRH-A markedly reduced E2 concentrations in the postnatal groups (C+E2, TS+E2, and D+E) to ~0.3pg/mL (data not shown), and E2 implant insertion increased concentrations of E2 to ~0.6pg/mL (Fig. 5A, Pre E2). Effect size analysis comparing postnatal E2-treatment groups with corresponding prenatal-only treated groups showed large effects following E2 implants. Suppression of LH pulsatility following GnRH-A treatment and an increase in the number of LH pulses following cessation of GnRH-A treatment in the postnatal groups were similar to changes described for the prenatal treatment-only groups. There were no differences in LH pulse frequency among treatment groups during pre-E2 period (Fig. 5B, Pre E2). While E2-treatment completely ablated LH pulsatility in the C group, this complete suppression was not achieved in four out of seven animals in the C+E2 group, which had pulses during the pre-E2 period (Fig. 5B, Post E2). The remaining three animals had no LH pulses during the pre-E2 period, but showed low pulse frequency (1–2 pulses/6h) during E2-treatment period. Postnatal E2 failed to reduce LH pulsatility in TS+E2 and D+E groups (Fig. 5B, Post E2) as opposed to suppressive effects being evident in prenatal TS- and D-only females.
Age at puberty
ANOVA revealed a significant postnatal E2-treatment effect in delaying age at puberty (Fig. 5C). Similarly, effect size analysis showed robust effects of postnatal E2-treatment in delaying the age at puberty in all treated groups.
E2-positive feedback responses
Results of the E2-positive feedback test in the postnatal groups are summarized in Fig. 6 and Supplementary Fig. 2. While five of seven C+E2 and two of seven TS+E2 animals responded to the positive feedback challenge, all D+E females demonstrated a response (Fig. 6A). Except for a reduction in total LH and FSH released in response to postnatal E2-treatment in the C+E2 group (Fig. 6B and F), and surge duration in the D+E group relative to corresponding prenatal control groups (Fig. 6E), postnatal E2-treatment had no effect on other LH and FSH surge parameters.
Periovulatory hormonal dynamics
The impact of postnatal E2-treatment on periovulatory hormonal dynamics during the first and second breeding season is summarized in Fig. 7, and representative hormonal profiles are depicted in Supplementary Figs 3 and 4 respectively. Among the animals that received postnatal E2-treatment, 71.4% of C+E2, 60.0% of TS+E2, and 66.7% of the D+E animals showed estrous synchronization following PGF2α administration during the first breeding season (Fig. 7A). Of those that synchronized, 100% of the animals showed definable LH surges in C+E2 and D+E groups, while only 1 of the 3 animals that synchronized had an LH surge in the TS+E2 group (Fig. 7F). Postnatal E2-treatment had no effect on any of the attributes of the preovulatory surge dynamics (Fig. 7). Because only one TS+E2 animal had a definable LH surge, comparison between TS and TS+E2 animals was not possible.
During the second breeding season, 85.7 and 75.0% of the animals in C+E2 and TS+E2 treatment groups synchronized following PGF2α administration respectively (Fig. 7A). Of these, 100% of the C+E2 and none of the TS+E2 animals showed definable LH surges (Fig. 7F). Postnatal E2-treatment had no effect on any of the parameters of preovulatory surge dynamics in C animals, and since TS+E2 animals did not surge, such comparisons were not possible (Fig. 7).
Discussion
The results from this study confirm our previous findings of neuroendocrine feedback, pubertal timing, and periovulatory disruptions in prenatal TS-treated female sheep. However, the data fail to support our hypotheses that (1) prenatal E2-treatment replicates the neuroendocrine disruptions attributed to estrogenic programming of E2-positive feedback and periovulatory hormonal dynamics in prenatal TS-treated females and (2) concomitant prenatal treatment with D and E2 reproduces the E2 neuroendocrine feedback and periovulatory defects induced by prenatal TS excess. Furthermore, our results indicate that postnatal E2-treatment (1) reduces the E2 inhibitory effects on tonic LH release in prenatal TS and D groups, (2) delays the onset of puberty in all groups, and (3) fails to amplify the dampening of LH/FSH surges during the E2-positive feedback challenge in prenatal TS- and D-treated animals. The significance of these findings and their relevance to the understanding of the steroid component in programming these neuroendocrine feedback disruptions are discussed below.
Estrogenic regulation of prenatal TS-induced disruptions
Our previous findings showing lack of disruptions in E2-positive feedback and periovulatory hormonal dynamics in prenatal D-treated sheep (Veiga-Lopez et al. 2009) and failure of androgen antagonist cotreatment to prevent E2-positive feedback disruptions induced by prenatal TS-treatment (Abi Salloum et al. 2012) are consistent with the premise that these defects are mediated by estrogenic programming stemming from aromatization of TS to E2. Paradoxically, the present observations that prenatal E2-treatment did not result in E2-positive feedback and periovulatory disruptions fail to support this premise. One possibility for this discrepancy is that not sufficient fetal E2 concentrations were achieved. Alternatively, these findings, in concert with previous findings of partial restoration of preovulatory LH surges in animals cotreated with TS and androgen antagonist (Padmanabhan et al. 2015), raise the possibility that both androgens and estrogens synergize in programming the prenatal TS-induced disruptions in E2 feedback and periovulatory hormonal dynamics.
Similarly, the failure of concomitant treatment with E2 and D to replicate the reproductive neuroendocrine defects might also be a function of inadequate estrogen levels achieved and consequent imbalance in androgen to estrogen ratio at the fetal level relative to what is attained with TS-treatment alone. An altered androgen to estrogen ratio could arise from insufficient estrogen levels being achieved due to placental metabolism of the peripherally administered E2 to lesser active metabolites as seen in monkeys (Slikker et al. 1982) or from increased androgen actions, since D has higher potency than TS (Grino et al. 1990). The ratio could also be altered, as D can be metabolized to 3β-diol, an estrogen receptorβagonist (Handa et al. 2008). Therefore, additional studies involving the use of estrogen antagonist alone or in combination with an androgen antagonist are required to clearly address the relative role of androgens and estrogens in mediating these effects.
Effects of postnatal E2 exposure
The lack of response to the E2-negative feedback challenge in control animals postnatally treated with E2 may be attributed to the downregulation of the estrogen receptor in the hypothalamus and pituitary gland due to continuous exposure to E2. In fact, E2-induced downregulation of estrogen receptor expression has been observed in other tissues, such as the uterus (Medlock et al. 1991, Farnell & Ing 2003). Therefore, the interval between the removal of the postnatal E2 implants and the E2-negative feedback challenge (7 days) may have been insufficient to allow recovery of estrogen receptor expression in the hypothalamus and pituitary to facilitate responsiveness to the E2-negative feedback challenge. Considering that puberty was delayed in postnatal E2-treated animals, this lack of E2-negative feedback response was surprising. To what extent this relates to the short interval between the removal of postnatal E2 implant and the E2-negative feedback testing remains to be determined. One possibility is that the delay in puberty could be a function of reduced gonadotropic support to the ovary (Foster et al. 1986).
Diminished gonadotropin release in response to E2-positive feedback was observed in prepubertal female sheep chronically treated with E2 in this study and in previous studies (Malcolm et al. 2006). However, a reduction in the periovulatory release of gonadotropins was not observed when examined during the first and second breeding season. The reason for this difference might relate to the heightened sensitivity of prepubertal females to E2-negative feedback compared with postpubertal females, resulting in reduced storage of LH in the pituitary that is available for release during the E2-positive feedback challenge. In contrast to normal periovulatory LH release in our study, postnatal exposure to E2 was found to block endogenous LH surge in female sheep (Ozturk et al. 1998). The differences in LH surge responses between this study and our study may be a function of the circulating E2 concentrations achieved by the treatment and/or the timing of treatment.
Postnatal E2 modulation of prenatal steroid effects
Comparison of the effects of prenatal TS plus postnatal E2-treatment with prenatal TS-treatment alone found that postnatal E2-treatment does not amplify the effects of prenatal TS-treatment in reducing responsiveness to E2-negative feedback and advancing puberty, a finding consistent with our previous studies (Sarma et al. 2005, Veiga-Lopez et al. 2009, Padmanabhan et al. 2015). Instead, postnatal E2-treatment delayed puberty relative to controls, while failing to reduce gonadotropin levels in response to the E2-negative feedback challenge. Considering that an escape from E2-negative feedback is an important event controlling the onset of puberty, this dissociation between delayed puberty and reduced sensitivity to E2-negative feedback is paradoxical. As discussed above, this dissociation may be a function of the short interval between the removal of the E2 implant and the testing for the E2-negative feedback.
Furthermore, the finding that E2-positive feedback and periovulatory hormonal dynamics were impaired in prenatal TS-treated sheep, which is in agreement with our previous findings (Sharma et al. 2002, Veiga-Lopez et al. 2009, Padmanabhan et al. 2015), coupled with failure of postnatal E2 exposure to amplify these disruptions, suggests these defects are organized primarily during the prenatal life, with little or no impact of continuous postnatal exposure to excess E2, at least at the dose tested. Alternatively, low concentrations of E2 produced by the ovary during postnatal life may be sufficient to further differentiate (masculinize) the GnRH/gonadotropin surge mechanism in ovary-intact, prenatal TS-treated female sheep, with exogenous postnatal E2-treatment having no additional impact on the gonadotropin surges. This possibility is supported by the observation that neonatal ovariectomy restores the E2-induced LH surge in prenatal TS-treated sheep (Jackson et al. 2013).
Contrary to our expectation and in disagreement with our previous studies (Veiga-Lopez et al. 2009), prenatal D-treatment failed to disrupt E2-negative feedback. The reason for this discrepancy is unclear, but it may involve the different ages at which the E2-negative feedback tests were carried out, suggesting that the desensitization to the E2-negative feedback may occur only at a later developmental stage in D-treated compared with TS-treated females. As prenatal D-treated animals do not develop a multifollicular ovarian morphology (Smith et al. 2009), the delay in the desensitization of the neuroendocrine axis to the E2-negative feedback could be due to lower secretion of E2 and/or other ovarian-derived factors unlike prenatal TS-treated females. Failure of postnatal E2 to alter the E2-positive feedback or periovulatory hormonal dynamics in prenatal D-treated animals is again consistent with such defects being programmed primarily during the prenatal life.
Conclusions
In conclusion, the results from this study combined with our earlier findings indicate that disruptions in E2 feedback mechanisms, timing of puberty, and periovulatory hormonal dynamics seen in prenatal TS-treated sheep require activation of both androgenic and estrogenic pathways. Additionally, the present findings indicate that these neuroendocrine disruptions are programmed primarily during the prenatal life and are not amplified or modified by postnatal exposure to E2 excess, at least at the dose tested. Unfortunately, likely due to a putative metabolism of D into 3β-diol or insufficient fetal E2 concentrations achieved, the relative role of androgens or estrogens in programming some of these reproductive neuroendocrine defects could not be completely discerned by this study. In the future, comparative experiments utilizing antagonists of androgen and estrogen receptors separately and in combination need to be undertaken in both the ovariectomized and ovary-intact models. Although postnatal E2 exposure did not amplify the adverse effects of prenatal TS exposure, it did negatively affect the timing of puberty and preovulatory gonadotropin release in control animals. These data indicate that postnatal exposure to EDC with estrogenic properties, through activational effects, may have the potential to adversely affect reproductive neuroendocrine function in females. Therefore, although extrapolation of findings in this sheep model to human pathology should be done cautiously, these observations highlight the potential for gestational and postnatal endocrine imbalances to negatively impact the offspring’s reproductive health.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/REP-16-0065.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by National Institutes of Health grant P01 HD44232.
Acknowledgments
We are grateful to Mr Douglas Doop for his expert animal care, facility management, and help with generation of the experimental lambs; Drs Mohan Manikkam and Teresa Steckler, Ms Jasmine Tavadia, Mr James Lee, Mr Rohit Sreedharan, staff, and undergraduate students from Dr Theresa Lee’s laboratory and students from University of Michigan Undergraduate Research Opportunity program (UROP) for assistance with prenatal steroid treatment, blood sampling, hormonal assays, and data analyses. We are also thankful to Dr Aimee K Wurst from the laboratory of Dr Keith Inskeep for performing the estradiol measures at West Virginia University, Dr Wen Ye for the help with statistical analysis, and Mr Jacob Moeller for proof-reading and editing the language.
References
Abi Salloum B, Herkimer C, Lee JS, Veiga-Lopez A & Padmanabhan V 2012 Developmental programming: prenatal and postnatal contribution of androgens and insulin in the reprogramming of estradiol positive feedback disruptions in prenatal testosterone-treated sheep. Endocrinology 153 2813–2822. (doi:10.1210/en.2011-2074)
Bahrke MS, Yesalis CE & Brower KJ 1998 Anabolic-androgenic steroid abuse and performance-enhancing drugs among adolescents. Child and Adolescent Psychiatric Clinics of North America 7 821–838.
Barker DJ 2004 The developmental origins of adult disease. Journal of the American College of Nutrition 23 588S–595S. (doi:10.1080/07315724.2004.10719428)
Bayer TA, Falkai P & Maier W 1999 Genetic and non-genetic vulnerability factors in schizophrenia: the basis of the “two hit hypothesis.” Journal of Psychiatric Research 33 543–548. (doi:10.1016/S0022-3956(99)00039-4)
Birch RA, Padmanabhan V, Foster DL, Unsworth WP & Robinson JE 2003 Prenatal programming of reproductive neuroendocrine function: fetal androgen exposure produces progressive disruption of reproductive cycles in sheep. Endocrinology 144 1426–1434. (doi:10.1210/en.2002-220965)
Clarke I, Scaramuzzi R & Short R 1977 Ovulation in prenatally androgenized ewes. Journal of Endocrinology 73 385–389. (doi:10.1677/joe.0.0730385)
Cohen J 1992 A power primer. Psychological Bulletin 112 155–159. (doi:10.1037/0033-2909.112.1.155)
Diamanti-Kandarakis E, Bourguignon JP, Giudice LC, Hauser R, Prins GS, Soto AM, Zoeller RT & Gore AC 2009 Endocrine-disrupting chemicals: an Endocrine Society scientific statement. Endocrine Reviews 30 293–342. (doi:10.1210/er.2009-0002)
Dumesic DA, Goodarzi MO, Chazenbalk GD & Abbott DH 2014 Intrauterine environment and polycystic ovary syndrome. Seminars in Reproductive Medicine 32 159–165. (doi:10.1055/s-0034-1371087)
Farnell YZ & Ing NH 2003 The effects of estradiol and selective estrogen receptor modulators on gene expression and messenger RNA stability in immortalized sheep endometrial stromal cells and human endometrial adenocarcinoma cells. Journal of Steroid Biochemistry and Molecular Biology 84 453–461. (doi:10.1016/S0960-0760(03)00066-9)
Foster DL & Jackson LM 2006 Puberty in the sheep. In Knobil and Neill’s Physiology of Reproduction, pp 2127–2176. Boston, MA, USA: Elsevier.
Foster DL, Ryan KD, Goodman RL, Legan SJ, Karsch FJ & Yellon SM 1986 Delayed puberty in lambs chronically treated with oestradiol. Journal of Reproduction and Fertility 78 111–117. (doi:10.1530/jrf.0.0780111)
Foster DL, Jackson LM & Padmanabhan V 2006 Programming of GnRH feedback controls timing puberty and adult reproductive activity. Molecular and Cellular Endocrinology 254–255 109–119. (doi:10.1016/j.mce.2006.04.004)
Goodman RL, Legan SJ, Ryan KD, Foster DL & Karsch FJ 1980 Two effects of estradiol that normally contribute to the control of tonic LH secretion in the ewe. Biology of Reproduction 23 415–422. (doi:10.1095/biolreprod23.2.415)
Goodman RL, Legan SJ, Ryan KD, Foster DL & Karsch FJ 1981 Importance of variations in behavioural and feedback actions of oestradiol to the control of seasonal breeding in the ewe. Journal of Endocrinology 89 229–240. (doi:10.1677/joe.0.0890229)
Grino PB, Griffin JE & Wilson JD 1990 Testosterone at high concentrations interacts with the human androgen receptor similarly to dihydrotestosterone. Endocrinology 126 1165–1172. (doi:10.1210/endo-126-2-1165)
Handa RJ, Pak TR, Kudwa AE, Lund TD & Hinds L 2008 An alternate pathway for androgen regulation of brain function: activation of estrogen receptor beta by the metabolite of dihydrotestosterone, 5alpha-androstane-3beta,17beta-diol. Hormones and Behavior 53 741–752. (doi:10.1016/j.yhbeh.2007.09.012)
Jackson LM, Timmer KM & Foster DL 2008 Sexual differentiation of the external genitalia and the timing of puberty in the presence of an antiandrogen in sheep. Endocrinology 149 4200–4208. (doi:10.1210/en.2007-1382)
Jackson LM, Mytinger A, Roberts EK, Lee TM, Foster DL, Padmanabhan V & Jansen HT 2013 Developmental programming: postnatal steroids complete prenatal steroid actions to differentially organize the GnRH surge mechanism and reproductive behavior in female sheep. Endocrinology 154 1612–1623. (doi:10.1210/en.2012-1613)
Jefferson WN, Patisaul HB & Williams CJ 2012 Reproductive consequences of developmental phytoestrogen exposure. Reproduction 143 247–260. (doi:10.1530/REP-11-0369)
Jensen ET, Daniels JL, Sturmer T, Robinson WR, Williams CJ, Vejrup K, Magnus P & Longnecker MP 2015 Hormonal contraceptive use before and after conception in relation to preterm birth and small for gestational age: an observational cohort study. BJOG 122 1349–1361. (doi:10.1111/bjo.2015.122.issue-10)
Karsch FJ, Dahl GE, Evans NP, Manning JM, Mayfield KP, Moenter SM & Foster DL 1993 Seasonal changes in gonadotropin-releasing hormone secretion in the ewe: alteration in response to the negative feedback action of estradiol. Biology of Reproduction 49 1377–1383. (doi:10.1095/biolreprod49.6.1377)
Malcolm KD, Jackson LM, Bergeon C, Lee TM, Padmanabhan V & Foster DL 2006 Long-term exposure of female sheep to physiologic concentrations of estradiol: effects on the onset and maintenance of reproductive function, pregnancy, and social development in female offspring. Biology of Reproduction 75 844–852. (doi:10.1095/biolreprod.106.053264)
Manikkam M, Crespi EJ, Doop DD, Herkimer C, Lee JS, Yu S, Brown MB, Foster DL & Padmanabhan V 2004 Fetal programming: prenatal testosterone excess leads to fetal growth retardation and postnatal catch-up growth in sheep. Endocrinology 145 790–798. (doi:10.1210/en.2003-0478)
Manikkam M, Steckler TL, Welch KB, Inskeep EK & Padmanabhan V 2006 Fetal programming: prenatal testosterone treatment leads to follicular persistence/luteal defects; partial restoration of ovarian function by cyclic progesterone treatment. Endocrinology 147 1997–2007. (doi:10.1210/en.2005-1338)
Medlock KL, Lyttle CR, Kelepouris N, Newman ED & Sheehan DM 1991 Estradiol down-regulation of the rat uterine estrogen receptor. Proceedings of The Society for Experimental Biology and Medicine 196 293–300. (doi:10.3181/00379727-196-43191)
Nakagawa S & Cuthill IC 2007 Effect size, confidence interval and statistical significance: a practical guide for biologists. Biological Reviews of the Cambridge Philosophical Society 82 591–605. (doi:10.1111/brv.2007.82.issue-4)
New MI 2006 Extensive clinical experience: nonclassical 21-hydroxylase deficiency. Journal of Clinical Endocrinology & Metabolism 91 4205–4214. (doi:10.1210/jc.2006-1645)
Niswender GD, Reichert LE Jr, Midgley AR Jr & Nalbandov AV 1969 Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 84 1166–1173. (doi:10.1210/endo-84-5-1166)
Ozturk M, Smith RF & Dobson H 1998 Effect of prolonged exposure to oestradiol on subsequent LH secretion in ewes. Journal of Reproduction and Fertility 114 1–9. (doi:10.1530/jrf.0.1140001)
Padmanabhan V & Veiga-Lopez A 2013a Animal models of the polycystic ovary syndrome phenotype. Steroids 78 734–740. (doi:10.1016/j.steroids.2013.05.004)
Padmanabhan V & Veiga-Lopez A 2013b Sheep models of polycystic ovary syndrome phenotype. Molecular and Cellular Endocrinology 373 8–20. (doi:10.1016/j.mce.2012.10.005)
Padmanabhan V, Evans NP, Dahl GE, McFadden KL, Mauger DT & Karsch FJ 1995 Evidence for short or ultrashort loop negative feedback of gonadotropin-releasing hormone secretion. Neuroendocrinology 62 248–258. (doi:10.1159/000127011)
Padmanabhan V, McFadden K, Mauger DT, Karsch FJ & Midgley AR Jr 1997 Neuroendocrine control of follicle-stimulating hormone (FSH) secretion. I. Direct evidence for separate episodic and basal components of FSH secretion. Endocrinology 138 424–432. (doi:10.1210/endo.138.1.4892)
Padmanabhan V, Veiga-Lopez A, Herkimer C, Abi Salloum B, Moeller J, Beckett E & Sreedharan R 2015 Developmental programming: prenatal and postnatal androgen antagonist and insulin sensitizer interventions prevent advancement of puberty and improve LH surge dynamics in prenatal testosterone-treated sheep. Endocrinology 156 2678–2692. (doi:10.1210/en.2015-1235)
Padmanabhan V, Veiga-Lopez A, Herkimer C, Abi Salloum B, Moeller J, Beckett E & Sreedharan R 2015 Developmental programming: prenatal and postnatal androgen antagonist and insulin sensitizer interventions prevent advancement of puberty and improve LH surge dynamics in prenatal testosterone-treated sheep. Endocrinology 156 2678–2692. (doi:10.1210/en.2015-1235)
Peretz J, Vrooman L, Ricke WA, Hunt PA, Ehrlich S, Hauser R, Padmanabhan V, Taylor HS, Swan SH & VandeVoort CA et al. 2014 Bisphenol a and reproductive health: update of experimental and human evidence, 2007–2013. Environmental Health Perspectives 122 775–786. (doi:10.1289/ehp.1307728)
Pignatelli D 2013 Non-classic adrenal hyperplasia due to the deficiency of 21-hydroxylase and its relation to polycystic ovarian syndrome. Frontiers of Hormone Research 40 158–170. (doi:10.1159/000342179)
Puttabyatappa M, Cardoso RC & Padmanabhan V 2015 Effect of maternal PCOS and PCOS-like phenotype on the offspring’s health. Molecular and Cellular Endocrinology In press. (doi:10.1016/j.mce.2015.11.030)
Rhind SM, Rae MT & Brooks AN 2001 Effects of nutrition and environmental factors on the fetal programming of the reproductive axis. Reproduction 122 205–214. (doi:10.1530/rep.0.1220205)
Robinson JE, Birch RA, Foster DL & Padmanabhan V 2002 Prenatal exposure of the ovine fetus to androgens sexually differentiates the steroid feedback mechanisms that control gonadotropin releasing hormone secretion and disrupts ovarian cycles. Archives of Sexual Behavior 31 35–41. (doi:10.1023/A:1014075016956)
Rosenfield RL 2007 Clinical review: identifying children at risk for polycystic ovary syndrome. Journal of Clinical Endocrinology & Metabolism 92 787–796. (doi:10.1210/jc.2006-2012)
Rozell TG & Keisler DH 1990 Effects of oestradiol on LH, FSH and prolactin in ovariectomized ewes. Journal of Reproduction and Fertility 88 645–653. (doi:10.1530/jrf.0.0880645)
Sarma HN, Manikkam M, Herkimer C, Dell’Orco J, Welch KB, Foster DL & Padmanabhan V 2005 Fetal programming: excess prenatal testosterone reduces postnatal luteinizing hormone, but not follicle-stimulating hormone responsiveness, to estradiol negative feedback in the female. Endocrinology 146 4281–4291. (doi:10.1210/en.2005-0322)
Schug TT, Janesick A, Blumberg B & Heindel JJ 2011 Endocrine disrupting chemicals and disease susceptibility. Journal of Steroid Biochemistry and Molecular Biology 127 204–215. (doi:10.1016/j.jsbmb.2011.08.007)
Schulz KM, Molenda-Figueira HA & Sisk CL 2009 Back to the future: the organizational-activational hypothesis adapted to puberty and adolescence. Hormones and Behavior 55 597–604. (doi:10.1016/j.yhbeh.2009.03.010)
Sharma TP, Herkimer C, West C, Ye W, Birch R, Robinson JE, Foster DL & Padmanabhan V 2002 Fetal programming: prenatal androgen disrupts positive feedback actions of estradiol but does not affect timing of puberty in female sheep. Biology of Reproduction 66 924–933. (doi:10.1095/biolreprod66.4.924)
Sir-Petermann T, Maliqueo M, Angel B, Lara HE, Perez-Bravo F & Recabarren SE 2002 Maternal serum androgens in pregnant women with polycystic ovarian syndrome: possible implications in prenatal androgenization. Human Reproduction 17 2573–2579. (doi:10.1093/humrep/17.10.2573)
Slikker W Jr, Hill DE & Young JF 1982 Comparison of the transplacental pharmacokinetics of 17 beta-estradiol and diethylstilbestrol in the subhuman primate. Journal of Pharmacology and Experimental Therapeutics 221 173–182.
Smith P, Steckler TL, Veiga-Lopez A & Padmanabhan V 2009 Developmental programming: differential effects of prenatal testosterone and dihydrotestosterone on follicular recruitment, depletion of follicular reserve, and ovarian morphology in sheep. Biology of Reproduction 80 726–736. (doi:10.1095/biolreprod.108.072801)
Steckler TL, Lee JS, Ye W, Inskeep EK & Padmanabhan V 2008 Developmental programming: exogenous gonadotropin treatment rescues ovulatory function but does not completely normalize ovarian function in sheep treated prenatally with testosterone. Biology of Reproduction 79 686–695. (doi:10.1095/biolreprod.108.068643)
Steckler TL, Herkimer C, Dumesic DA & Padmanabhan V 2009 Developmental programming: excess weight gain amplifies the effects of prenatal testosterone excess on reproductive cyclicity – implication for polycystic ovary syndrome. Endocrinology 150 1456–1465. (doi:10.1210/en.2008-1256)
Tang WY & Ho SM 2007 Epigenetic reprogramming and imprinting in origins of disease. Reviews in Endocrine and Metabolic Disorders 8 173–182. (doi:10.1007/s11154-007-9042-4)
Tang WY, Newbold R, Mardilovich K, Jefferson W, Cheng RY, Medvedovic M & Ho SM 2008 Persistent hypomethylation in the promoter of nucleosomal binding protein 1 (Nsbp1) correlates with overexpression of Nsbp1 in mouse uteri neonatally exposed to diethylstilbestrol or genistein. Endocrinology 149 5922–5931. (doi:10.1210/en.2008-0682)
Unsworth WP, Taylor JA & Robinson JE 2005 Prenatal programming of reproductive neuroendocrine function: the effect of prenatal androgens on the development of estrogen positive feedback and ovarian cycles in the ewe. Biology of Reproduction 72 619–627. (doi:10.1095/biolreprod.104.035691)
Vaiserman A 2014 Early-life exposure to endocrine disrupting chemicals and later-life health outcomes: an epigenetic bridge? Aging and Disease 5 419–429. (doi:10.14336/AD.2014.0500419)
Veiga-Lopez A, Ye W, Phillips DJ, Herkimer C, Knight PG & Padmanabhan V 2008 Developmental programming: deficits in reproductive hormone dynamics and ovulatory outcomes in prenatal, testosterone-treated sheep. Biology of Reproduction 78 636–647. (doi:10.1095/biolreprod.107.065904)
Veiga-Lopez A, Astapova OI, Aizenberg EF, Lee JS & Padmanabhan V 2009 Developmental programming: contribution of prenatal androgen and estrogen to estradiol feedback systems and periovulatory hormonal dynamics in sheep. Biology of Reproduction 80 718–725. (doi:10.1095/biolreprod.108.074781)
Veiga-Lopez A, Wurst AK, Steckler TL, Ye W & Padmanabhan V 2014 Developmental programming: postnatal estradiol amplifies ovarian follicular defects induced by fetal exposure to excess testosterone and dihydrotestosterone in sheep. Reproductive Sciences 21 444–455. (doi:10.1177/1933719113503412)
Veldhuis JD & Johnson ML 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. American Journal of Physiology 250 E486–E493.
Wood RI & Foster DL 1998 Sexual differentiation of reproductive neuroendocrine function in sheep. Reviews of Reproduction 3 130–140. (doi:10.1530/ror.0.0030130)
Wood RI, Ebling FJ, I’Anson H, Bucholtz DC, Yellon SM & Foster DL 1991 Prenatal androgens time neuroendocrine sexual maturation. Endocrinology 128 2457–2468. (doi:10.1210/endo-128-5-2457)