Skip to main content
NCBI home page
Endocrinology logoLink to Endocrinology
. 2009 Jan 8;150(5):2317–2324. doi: 10.1210/en.2008-1307

Organizational Actions of Postnatal Estradiol in Female Sheep Treated Prenatally with Testosterone: Programming of Prepubertal Neuroendocrine Function and the Onset of Puberty

Leslie M Jackson 1, Kathleen M Timmer 1, Douglas L Foster 1
PMCID: PMC2671902  PMID: 19131574

Abstract

Prenatal testosterone (T) exposure defeminizes reproductive neuroendocrine function in female sheep, although the LH surge dysfunctions are initially less severe in gonadally intact females than in females subject to neonatal ovariectomy and estradiol (E) replacement. Because prepubertal ovarian production of E differs quantitatively and qualitatively from chronic E replacement, we tested the hypothesis that postnatal E exacerbates the consequences of prenatal T on the positive, but not the negative, steroid feedback controls of GnRH secretion. Our approach was to characterize prepubertal sensitivity to E negative feedback, the onset and maintenance of progestagenic cycles, and the LH surge response in ovary intact, prenatally untreated (control), and T-treated (T) sheep that were exposed postnatally to only endogenous E, or exposed to excess E by sc implant. Sensitivity to E negative feedback was reduced in T females, but excess postnatal E did not further increase LH pulse frequency. Excess E prevented ovarian cycles in several control females, and increased cycle irregularity in T females. However, the LH surge mechanism was functional in all control females (regardless of postnatal E exposure) and in some T females without excess E, but nonfunctional in T females with excess E. These findings suggest that postnatal E does not program increased resistance to E negative feedback, but excess postnatal E does disrupt other mechanisms required for ovarian cyclicity. We conclude that in this precocial species, prenatal steroids are sufficient to program controls of tonic LH secretion, but the LH surge mechanism is susceptible to further programming by postnatal E.

The disruptive effects of prenatal testosterone exposure on the GnRH surge mechanism and female reproductive function are exacerbated by exposure to excess estradiol during postnatal life.

In precocial species like the sheep, sexual differentiation of the neural circuits that control reproductive neuroendocrine function and sexual behavior is considered to occur during a critical period of prenatal development due to the organizational actions of testicular steroids (1). Furthermore, exposing female fetuses to testosterone (T) during established critical periods defeminizes their brains, and this disruption is classically characterized by the absence of a GnRH/LH surge in response to follicular phase concentrations of estradiol (E). The consequences of prenatal T exposure on female reproductive neuroendocrine function have been studied extensively in the sheep, often using a neuroendocrine model in which animals are ovariectomized shortly after birth and treated chronically with an E implant to standardize the hormonal milieu, and facilitate evaluation of steroid feedback controls of GnRH secretion. Based on this approach, both the negative and positive feedback actions of E have been defeminized by prenatal T (2,3). The results of these initial studies led to the expectation that when T-treated females remained gonadally intact, they would be hypergonadotropic due to reduced sensitivity to E negative feedback but would not have a functional GnRH surge mechanism or exhibit estrous cycles at puberty (4). Surprisingly, this prediction was not entirely correct. Ovarian intact prenatally T-treated females are hypergonadotropic (5,6), but they have progesterone (P) cycles in their first year of life that begin at the same time as cycles in untreated females (5,7). In addition, T-treated females exhibit an LH surge in response to exogenous E (7) or during a spontaneous ovarian cycle (8) in their first year of life, although the surge is typically delayed and dampened. However, ovarian cycles eventually become irregular and cease during the second year of life (5,7,9).

One possible explanation for the incongruity between the expected absence of a functioning surge mechanism and the presence of ovarian cycles in the gonadally intact T-treated female is that prenatal exposure to T programs sex differences in neuroendocrine function, but complete disruption of the surge system also requires postnatal exposure to E. This requirement is met in the normal male exposed prenatally and postnatally to testicular T that can be aromatized peripherally to E. Similarly, prenatally T-treated female lambs that are neonatally ovariectomized and steroid replaced are exposed continuously to E after birth. Because the immature ovary is less steroidogenic than the developing testis (10,11), females exposed prenatally to excess T and postnatally to their own gonadal steroids do not experience continuous or significant elevations in circulating E until ovarian steroidogenesis increases at puberty. Consequently, these females have ovarian cycles (presumably including adequate GnRH surges) until endogenous E ultimately completes defeminization of the surge mechanism, and renders it inoperative in subsequent years.

This study was designed to test the hypothesis that the programming of responsiveness to the negative feedback action of E by T is completed during gestation, but differentiation of the positive feedback action of E requires the organizational actions of both prenatal T and postnatal E. Our approach was to provide ovary intact, untreated, and prenatally T-treated females with excess postnatal E using a chronic implant identical to that used in studies of the neuroendocrine model. Prepubertal patterns of pulsatile LH secretion, the onset of ovarian cycles at puberty, and the postpubertal LH surge response to exogenous E were compared with the same parameters in prenatally untreated and T-treated females without excess postnatal E. We predicted that untreated females with excess postnatal E would be hypogonadotropic before puberty due to their inherent sensitivity to the negative feedback effects of E, but the GnRH surge mechanism and ovarian cyclicity would not be affected by excess E. In T-treated females, excess postnatal E would not further increase hypergonadotropism, but it would ablate the surge response to E and disrupt ovarian cycles.

Our approach also included neonatally ovariectomized untreated and T-treated females without steroid replacement that lacked a significant source of postnatal E. Prepubertal LH pulsatility was studied in these females to examine the possibility of the programming of steroid-independent controls of GnRH by prenatal T. However, the ovariectomized females were not included in our initial testing of the LH surge response after puberty because the overall design of a larger study required that they remain steroid free until their second year of life, when they could be tested along with T-treated ovary intact females in which reproductive cycles had deteriorated over time.

Materials and Methods

Animals and treatments

Naturally cycling female Suffolk sheep (proven breeders at least 2 yr of age) were maintained outdoors at the Reproductive Sciences Program Sheep Research Facility at the University of Michigan as previously described (12). Mating was detected by the presence of paint marks transferred from breeding males to the females during copulation. Mated females were stratified by body weight (BW) and body condition score (BCS) (13), and then randomly assigned to an untreated (control) or prenatal T treatment group such that the distributions of BW and BCS did not differ between groups [control: BW (mean ± sem) = 89.6 ± 1.9 kg, BCS = 3.1 ± 0.1; T: BW = 89.9 ± 1.8 kg, BCS = 3.1 ± 0.1]. Prenatal T treatment (100 mg T propionate injected im twice a week; Sigma-Aldrich Corp., St. Louis, MO) was administered from d 30–90 of the 147 d gestation. The University Committee for the Use and Care of Animals at the University of Michigan approved the protocol for all animal care and experimental procedures.

Female lambs (23 control and 18 T) were born in the spring between March 25 and April 11, 2005. Lambs were weighed weekly, and, at 2 wk of age, were stratified by BW and randomly assigned to one of three postnatal treatment groups, except for the condition that twin females (control: three sets; T: two sets) received different postnatal treatments. The groups differed with respect to their source of E exposure and included: 1) no treatment in which females were exposed to endogenous, E; 2) a chronic E treatment (E) that exposed females to endogenous E plus E from a sc implant inserted at 2 wk; and 3) neonatal ovariectomy (OVX) at 2–3 wk in which females matured in the absence of gonadal steroids. The E implant was a small capsule made of SILASTIC brand tubing (outer diameter = 0.46 cm, inner diameter = 0.34 cm; Dow Corning Corp., Midland, MI) packed with a 3-cm column of crystalline 17β-E (Sigma-Aldrich), and sealed with SILASTIC brand adhesive Type A (Dow Corning). Implants of this size produce low-circulating concentrations of E (<3–5 pg/ml) (14). Ovaries were removed via a midline abdominal incision under ketamine (20 mg/kg, im) and xylazine (0.1–0.2 mg/kg, im) anesthesia. Lambs were weaned at 8 wk of age, and thereafter maintained outdoors with access to water ad libitum and a diet of alfalfa hay and commercial alfalfa pellets (Shur-Gain, Elma, NY) supplemented by vitamins and minerals. The six treatment groups are designated by paired terms (prenatal/postnatal) that represent the prenatal treatment and the principle source of E during postnatal life: control/O, n = 8; control/E, n = 8; control/OVX, n = 7; T/O, n = 6; T/E, n = 6; and T/OVX, n = 6.

Prepubertal sensitivity to E negative feedback

Sensitivity to E negative feedback control of tonic secretion of LH was assessed at two time points before the expected time of puberty at 25–30 wk (15). The prepubertal ovary undergoes irregular waves of follicular development (16), and there is considerable individual variation in LH pulse frequency throughout the prepubertal period (17,18), most likely because of variations in ovarian production of E. Thus, sampling on a single day could easily overestimate or underestimate the average frequency of LH pulses at a given stage of development. Therefore, frequent blood samples for characterization of pulsatile patterns of LH were collected by jugular venipuncture from ovary intact groups (control/O, control/E, T/O, and T/E) every 20 min for 6 h every other day for 5 d (i.e. on Monday, Wednesday, and Friday) at 14 and 22 wk. Because the frequency of LH pulses was expected to be higher in the ovariectomized animals, and not subject to modulation by endogenous steroids, frequent samples were collected from control/OVX and T/OVX females every 10 min for 4 h on 1 d at 14 and 22 wk. Due to the sampling intensity required for this portion of the study, we included only a subset of animals from each treatment group (n = 6 control/O, n = 6 control/E, n = 6 control/OVX, n = 5 T/O, n = 5 T/E, and n = 6 T/OVX), which facilitated simultaneous collection of samples from all groups. Plasma was separated by centrifugation, and 100-μl aliquots of the plasma collected on the hour from each female were pooled and stored separately for E assay to quantify the steroid feedback signal present during each sampling period. Plasma samples and pools were stored at −20 C until assayed for LH or E.

Onset of puberty and ovarian cyclicity

The developmental decrease in sensitivity to E negative feedback and the onset and maintenance of reproductive cycles at puberty were determined in gonadally intact females by measuring LH and P in blood samples collected twice weekly. Sampling began at 2 wk, and continued until 50 wk, or until P measures indicated the onset of seasonal anestrous.

Postpubertal LH surge response to E

The LH surge response to an E challenge was tested in gonadally intact females during the first seasonal anestrous period at 14 months of age. Chronic E implants were removed from control/E and T/E females, and 5 d later an artificial luteal phase was created by sc insertion (19) of two controlled-release P implants (CIDR; DEC International, Hamilton, New Zealand) and a SILASTIC brand capsule (outer diameter = 0.46 cm, inner diameter = 0.34 cm) containing a 1-cm column of E. The luteal phase implants were removed after 11 d, and 24 h later, four 3-cm E implants were inserted sc to produce late follicular phase concentrations (12–15 pg/ml) of E (20). Blood samples were collected every 2 h from 4 h before until 60 h after the E implants were inserted. After 60 h, E implants were removed from control/O and T/O females; one implant was left in place in control/E and T/E females.

Hormone assays

Concentrations of LH were measured in duplicate 20–200 μl aliquots of serum using modifications (21,22) of an RIA developed by Niswender et al. (23). Assay sensitivity, defined as 2 sd values from the buffer control, averaged 0.56 ± 0.03 ng/ml National Institutes of Health oLH-S12 (National Institutes of Health, Bethesda, MD) for 200 μl serum (n = 32 assays). Intraassay coefficients of variation (CVs) calculated from six replicates of three standard sera binding at 29, 50, and 82% of buffer controls averaged 3.6, 4.2, and 15.1%, respectively. Interassay CVs for the same standards were 3.0, 4.5, and 10.8%. Concentrations of E were measured in duplicate diethyl ether extracts of 200 μl plasma using a commercially available E MAIA assay RIA kit (Adaltis Italia, Bologna, Italy) modified for use in the sheep as previously described (24,25). Sensitivity of the assay averaged 0.43 ± 0.04 pg/ml for 200 μl plasma (n = 2 assays). Mean intraassay and interassay CVs were 11.8 and 9.3%. Concentrations of P were measured in duplicate 100-μl aliquots of unextracted plasma using a commercially available RIA kit (Coat-A-Count P4; Diagnostic Products Corp., Los Angeles, CA) previously validated for use in sheep (26). Sensitivity of this assay averaged 0.23 ± 0.04 ng/ml for 100 μl plasma (n = 9 assays). Mean intraassay and interassay CVs were 4.2 and 4.1%. The limit of assay sensitivity was assigned to those samples in which the concentration of LH, E, or P was below the sensitivity of the assay.

Data and statistical analyses

Pulses of LH during the frequent sampling periods at 14 and 22 wk were quantified by Cluster analysis (27). For samples collected every 20 min, the minimum number of data points in nadir and peak clusters was two and one, respectively, and the t statistics for significant increases and decreases were both 1.0. The t statistics were modified for analysis of samples collected every 10 min (from ovariectomized females) to 2.0 for both increases and decreases. For gonadally intact females, the age-specific pulse frequency for an individual was calculated by averaging pulse frequency during the three consecutive sampling periods. If no pulses were detected during a 6-h sampling period, pulse frequency was recorded as zero.

The onset of the developmental increase in LH measured in twice-weekly samples was evaluated using a criterion previously established in our laboratory (28), and was defined as the first of at least six consecutive samples in which the concentration of LH exceeded 1 ng/ml. The onset of puberty and maintenance of ovarian cycles in gonadally intact females were determined from P concentrations measured in samples collected twice weekly from 10–50 wk. Puberty was defined as the age when P concentrations first increased above 0.5 ng/ml for at least three consecutive samples (indicative of a 10- to 14 d luteal phase), and then decreased to less than 0.5 ng/ml for at least one sample (representing the 3 d follicular phase) (5,29). The length of a progestagenic cycle was calculated as the number of days from the first sample of a luteal phase to the first sample of the subsequent follicular phase. It should be noted that periods of elevated P in T/O and T/E females were irregular, and typically exceeded 14 d. However, the term “progesterone cycle” refers to any increase in circulating P followed by a return to basal levels (<0.5 ng/ml), while recognizing that such fluctuations may not represent a normal, ovulatory cycle.

The LH surge response to E in the hormone-induced ovarian cycle during anestrus was identified using criteria previously described (7,30). A surge was defined by concentrations of LH that exceeded baseline by 2 sd values for at least 8 h, with peak concentrations of LH exceeding twice the average concentrations measured during the 6 h before insertion of the E implants. The onset of the surge was the time of the first sample that exceeded baseline by 2 sd values, and peak amplitude was the maximum concentration of LH measured during the surge.

Differences in LH pulse frequency among groups at each age, differences in the age of the developmental increase in LH secretion, the age at the onset of P cycles, the number of cycles during the breeding season, and the mean length of cycles were compared using two-way ANOVA. Developmental increases in LH pulse frequency within a group were compared using a paired Student’s t test. The proportion of females exhibiting an LH surge response was analyzed using a χ2 statistic. For females exhibiting a surge, the time to surge onset and peak amplitude were compared among treatment groups using ANOVA. All data are expressed as mean ± sem; statistical significance for all analyses was defined as P < 0.05.

Results

E negative feedback control of tonic secretion of LH

Patterns of pulsatile secretion of LH from representative gonadally intact females at 14 and 22 wk are shown in Fig. 1A, and the LH and E data are summarized in Fig. 1B. At 14 wk, the frequency of LH pulses was similar among control/O, T/O, and T/E females, but no LH pulses were detected in any control/E females. At 22 wk, T/O females exhibited higher (P < 0.05) frequency LH pulses than control/O females, and the frequency of LH pulses in T/E females did not differ from either control/O (P = 0.057) or T/O females. Two control/E females had low- frequency (0.23 ± 0.02 pulses per hour) LH pulses, but the remaining females in this group were agonadotropic. The frequency of LH pulses was approximately hourly in all ovariectomized females at both 14 and 22 wk (insets in Fig. 1B, top panel), and higher than in any ovary intact group. Prenatal T treatment had no effect on steroid-independent secretion of LH, and there were no significant developmental increases in LH pulse frequency in any group.

Figure 1.

Figure 1

Open in a new tab

A, Pulsatile patterns of LH (solid circles) and plasma concentrations of E (shaded areas) in representative prepubertal control (C) and prenatally T-treated (T) female sheep exposed postnatally to endogenous, ovarian steroids (C/O, first panel; T/O, third panel) or excess E from a chronic sc implant (C/E, second panel; T/E, fourth panel) at 14 (left) and 22 wk (right) of age. Concentrations of LH were measured in frequent samples collected every 20 min for 6 h on d 1, 3, and 5 of a 5-d period. Concentrations of E were measured in pooled hourly samples from each sampling period. B, Mean (+sem) pulse frequency (top panel), LH concentration (middle panel), and E concentration (bottom panel) in C/O (white bars), C/E (gray bars), C/OVX (black bars), T/O (white-striped bars), T/E (gray-striped bars), and T/OVX (black-striped bars) sheep at 14 (left) and 22 wk (right). Concentrations of E in C/OVX and T/OVX females were undetectable and are plotted as the limit of assay sensitivity. Insets in top panels show pulsatile patterns of LH secretion in representative C/OVX and T/OVX females. Different letters denote significant (P < 0.05) differences among groups.

The mean concentration of LH in control/E females was lower than in the other ovary intact groups at 14 wk, but this difference was not significant at 22 wk (Fig. 1B, middle panel). At both ages, concentrations of LH were higher in ovariectomized females than in gonadally intact females, but there was no effect of prenatal treatment or postnatal age on steroid-independent secretion of LH.

Circulating concentrations of E were similar in control/O and T/O females at 14 wk (Fig. 1B, left-lower panel), but the presence of a chronic E implant in control/E and T/E females resulted in higher (P < 0.05) concentrations of E in these groups. At 22 wk, concentrations of E remained higher in control/E females than in control/O females (Fig. 1B, right-lower panel), but increased ovarian steroidogenesis was apparent in T-treated females because E concentrations in both T/O and T/E females were greater than in either control group.

Developmental resistance to E negative feedback

Profiles of LH in twice-weekly samples in two representative females from each group are shown in Fig. 2 (solid lines). In control/O and T/O females, a clear developmental increase in circulating concentrations of LH due to decreased resistance to E negative feedback was not evident due to the highly variable concentrations of LH in single, infrequent samples (Fig. 2, first and third panels). However, the continuous and uniform release of E from sc implants in control/E and T/E females suppressed LH until sensitivity to E negative feedback diminished during sexual maturation, and concentrations of LH increased (Fig. 2, second and fourth panels). The increase in LH occurred in control/E females at a mean age of 26.1 ± 0.7 wk, but in T/E females, developmental resistance to E negative feedback was advanced to 14.9 ± 2.9 wk.

Figure 2.

Figure 2

Open in a new tab

Concentrations of LH (solid lines) and P (shaded areas) in twice-weekly samples from representative control and T-treated females without (C/O, first panels; T/O third panels) and with (C/E, second panels; T/E, fourth panels) postnatal exposure to excess E. The onset of puberty is identified as the time of the first sustained increase in P. C, Control.

Pubertal onset of P cycles

As illustrated by the shaded areas in Fig. 2, P cycles occurred in all ovary intact groups. However, five of eight individuals in the control/E group had no cycles (e.g. Fig. 2, second-right panel, and Fig. 3A), despite the pubertal increase in LH. By contrast, all females in the three remaining groups had at least one P cycle before the end of the breeding season (Figs. 2 and 3A). When cycles did occur in control/E females, they began later than in control/O females (Fig. 3B). There was no effect of prenatal T alone on the age of puberty, although the onset of cycles in T/E females was delayed relative to control/O females. Regular P cycles were exhibited by control/O females and by the three control/E females that did cycle (Fig. 3, C and D). As illustrated in Fig. 2, third and fourth panels, T-treated females had fewer P cycles during the breeding season, and cycle length was highly variable. Due to the variability in cycle length, mean cycle length in T-treated females was not different than in control females. Mean BW was similar among all groups at the time of the first P cycles in control/O females (control/O = 50.2 ± 1.4 kg, control/E = 48.9 ± 2.1 kg, T/O = 42.2 ± 2.6 kg, T/E = 56.0 ± 2.8 kg), and was sufficient to promote the onset of puberty.

Figure 3.

Figure 3

Open in a new tab

Summary of prepubertal growth and ovarian cyclicity during the first breeding season in prenatally untreated and T-treated sheep with and without postnatal exposure to excess E. Panel A, Proportion of females exhibiting P cycles at puberty. Panel B, Mean (+sem) age of first P cycle. Panel C, Mean (+sem) number of P cycles during the first breeding season. Panel D, Mean (+sem) length of P cycles. Different letters denote significant differences (P < 0.05) among groups. C, Control.

LH surge response to E

Representative profiles of LH in response to exogenous E during anestrus are shown in Fig. 4, and a summary of the surge response is shown in Fig. 5. All control females had an LH surge (Figs. 4 and 5A), but the proportion of T/O females exhibiting a surge response (three of six) was significantly lower, and only one of six T/E females produced an LH surge. In control females, postnatal E had no effect on either the time to the surge onset (Fig. 5B) or the peak amplitude of the surge (Fig. 5C). For those T/O females exhibiting an LH surge, the onset of the surge was delayed, and the peak amplitude was lower than in control females.

Figure 4.

Figure 4

Open in a new tab

Profiles of LH secretion in representative seasonally anestrous C/O, C/E, T/O, and T/E females in response to exogenous E during an artificial ovarian cycle. The one T/E female shown with an LH surge was the only individual in that group that had a surge response. C, Control.

Figure 5.

Figure 5

Open in a new tab

Summary of LH surge responses in C/O, C/E, T/O, and T/E females. A, Percentage of females in each group that exhibited an LH surge response. B, Mean (+sem) number of hours between insertion of E implants at 0 h and the onset of the LH surge. C, Mean (+sem) amplitude of the peak of the LH surge. Different letters denote significant differences (P < 0.05) among groups. Data from the one T/E female that exhibited an LH surge are shown, but they were excluded from statistical analyses. C, Control.

Discussion

This study is the first to examine the combined consequences of excess prenatal T and excess postnatal E on reproductive neuroendocrine function, the onset of puberty, and ovarian cyclicity in female sheep. Our findings confirm the predictions that exposure to excess postnatal E does not further increase prepubertal resistance to E negative feedback and the ensuing hypergonadotropism that is programmed by prenatal T, but postnatal E does exacerbate the disruptive organizational actions of prenatal T on the positive feedback action of E and the GnRH surge mechanism. Furthermore, in contrast to our expectations, excess postnatal E dramatically disrupted ovarian cyclicity in prenatally untreated females, although the positive feedback action of E was functional when tested during anestrus. The relevance of these findings and potential mechanisms underlying the dysfunctions are discussed below.

Postnatal E and E negative feedback

In the prepubertal female sheep, the hypothalamus is extremely sensitive to E negative feedback (15), and very small amounts of E are sufficient to limit LH pulses to less than the hourly frequency characteristic of a sexually mature individual (14). In this study, consistent with the heightened sensitivity of the prepubertal lamb to E negative feedback, the addition of a physiologically relevant amount of E by sc implant in control/E females suppressed of LH pulsatility at 14 wk, with hypogonadotropism persisting in approximately 70% of the females in this group at 22 wk. Production of high-frequency pulses of LH (Fig. 1) and a precocial developmental increase in LH (Fig. 2) in T/E females, despite excess E, is consistent with prenatal exposure to excess T reducing sensitivity to E negative feedback resulting in hypersecretion of LH (6). The absence of differences in LH pulse frequency between T-treated females and control/O females at 14 wk may relate to the high variability seen among T females. Because this particular time point approximates the mean age of the developmental increase in LH observed in T/E females, the frequency of LH pulses among T females could differ dramatically depending on whether an individual had achieved developmental resistance to E negative feedback. Hypergonadotropism in T-treated females was evident at 22 wk when LH pulse frequency in T/O females was greater than in control/O females. Increased resistance to E negative feedback in both groups of T-treated females at this age is clearly supported by the fact that LH pulse frequency equals or exceeds that in control females, despite higher circulating concentrations of E in T-treated females.

Prenatal T treatment does not appear to program an inherent, steroid-independent increase in LH pulsatility based on our results from ovariectomized females. Because of an earlier study of prepubertal, neonatally gonadectomized sheep in which steroid-independent LH pulse frequency was higher in males than females (14), we expected that LH secretion in T/OVX females would be greater than in control/OVX females, and hypergonadotropism would not be attributable solely to resistance to E negative feedback. The lack of a difference in LH pulse frequency between control/OVX and T/OVX females could be due to the fact that the developing male is exposed to his own testicular steroids throughout most of gestation, and steroid exposure before d 30 or after d 90 could be responsible for sex differences in steroid-independent LH secretion. Nonetheless, exposure to excess T from d 30–90 gestation is sufficient to program hypergonadotropism by reducing hypothalamic sensitivity to E negative feedback. Chronic postnatal exposure to excess E did not further increase prepubertal gonadotropin secretion in T/E females, although it is important to note that peripheral concentrations of E differed among individuals. Additional experiments in which gonadotropin secretion in T/O and T/E females is studied in response to fixed doses of E would be necessary to exclude any organizational effect of postnatal E on E negative feedback and the control of tonic secretion of GnRH.

Postnatal E and E positive feedback

Our findings of an inoperative surge mechanism in half of the T/O females, and a delayed and dampened surge in the remaining half (Fig. 5), are consistent with previous reports (7,8), and suggest a partial disruption of the surge mechanism by prenatal T. This disruption could include programming at the hypothalamic level to reduce responsiveness to E, or changes in the pituitary that limit the release of LH in response to GnRH, or malfunctions at both levels. Prenatal T has increased pituitary sensitivity to GnRH (31), but it remains possible that pituitary responsiveness in the presence of follicular phase concentrations of E is impaired. Changes at the hypothalamic level most likely involve the E-sensitive neurons that provide afferent input to GnRH neurons. The number and distribution of GnRH neurons are similar in developing male and female lambs (32), but in studies using the neuroendocrine model (OVX plus E), prenatal T reduced the number of synaptic inputs to GnRH neurons in the adult brain (33), and also reduced the proportion of Fos-positive GnRH neurons within the preoptic area and hypothalamus during an E-induced LH surge (34). Thus, prenatal T programs the development of the neural circuits regulating the surge mechanism, and postnatal E may continue to organize E-sensitive stimulatory inputs to GnRH neurons to reduce further the neuronal population or modify synapse formation, resulting in ablation of the surge response as seen in all but one T/E female. An important test of this hypothesis that both prenatal and postnatal steroids are required to render the surge mechanism inoperative will examine the surge response in T/OVX females during the second breeding season.

Postnatal E and the onset and maintenance of ovarian cycles

Postnatal E treatment delayed or prevented the onset of P cycles in control females, which is a more severe outcome than in previous studies using a lower dose of E (1.3 cm implant). Foster et al. (35) reported that three of eight females exposed to excess E failed to exhibit ovarian cycles before 45 wk, and Malcolm et al. (30) noted a delay in the onset of puberty, although fertility was not impaired. This latter finding suggests that excess E has activational, but not organizational, effects on neuroendocrine function. Control/E females were not hypogonadotropic during the breeding season (Fig. 2), so acyclicity does not appear to involve suppression of LH secretion by excess E. However, in a study of ovariectomized ewes, the LH surge response to E was blocked in 95% of ewes pretreated for 10 d with 2 or 4 3-cm E implants (36), suggesting that chronically elevated E masks the incremental increases needed to induce a GnRH surge. Other possible neuroendocrine consequences of excess E may occur at the level of the anterior pituitary (37), and could include reduced responsiveness to GnRH, and limited stores of releasable LH in response to GnRH. Finally, excess E may impair folliculogenesis by down-regulating E receptors, reducing the ability of E to interact with FSH to stimulate differentiation of preovulatory follicles (38). This would limit ovarian production of E, a critical signal for the preovulatory GnRH surge, and prevent ovulation despite the surge mechanism being fully responsive to an E challenge.

In contrast to control/E females failing to exhibit P cycles, despite having a functional surge mechanism, T/E and T/O females exhibited P increases during the breeding season but did not have normal LH surge responses to E during anestrus. The first possible explanation for this conundrum is that all T females initially had an adequate surge mechanism that then deteriorated by the time it was tested with exogenous E during anestrus. Consequently, ovulation occurred at least once, but resistance to P negative feedback (19), or a defective response to a luteolytic signal (29), resulted in an extended luteal phase. Evidence of a functioning LH surge response in prepubertal T/E females would support this explanation, but this would not be expected based on studies of the OVX plus E model (2,3). A second possibility is that hypersecretion of LH results in abnormal follicular development, including increased androgen production and follicular persistence (29), or the development of luteinized follicles, and P increases occur in the absence of ovulation. Both mechanisms may contribute to abnormal cycles in T/O females until the surge response and ovarian cyclicity ultimately deteriorate.

We conclude that excess E during postnatal life has disruptive effects on female reproductive function and fertility, even when the amount of additional E is within normal physiological levels. Most importantly, postnatal E accelerates the ablation of the positive feedback response to E in prenatally T-treated females, which strongly suggests the existence of an expanded, steroid-sensitive, developmental period for organization of the surge mode GnRH secretion in precocial species. The detrimental effects of excess postnatal E on puberty and adult reproductive function are particularly important because of the increasing number of environmental substances that act as estrogen mimics or endocrine disruptors (39). Further delineation of the organizational roles of prenatal and postnatal steroids, and defining the critical periods during which they act, will expand our understanding of the development and differentiation of neuroendocrine function, and enhance our ability to predict or even prevent the consequences of exposure to excess steroids or endocrine-disrupting compounds.

Acknowledgments

We thank Mr. Doug Doop for his conscientious and expert animal care, technical assistance, and facility management. We also thank Mr. Karl Malcolm, Mr. Paul Slotten, Ms. Eila Roberts, Ms. Carol Herkimer, Dr. Mohan Manikkam, Mr. Michael Zakalik, and Ms. Erica LaVire for assistance with prenatal treatments and blood sampling, and Dr. Vasantha Padmanabhan for her insightful comments on an earlier draft of this manuscript.

Footnotes

This work was supported by National Institutes of Health Grant P01 HD-44232.

Disclosure Summary: The authors have nothing to disclose.

First Published Online January 8, 2009

Abbreviations: BCS, Body condition score; BW, body weight; CV, coefficient of variation; E, estradiol; O, ovary; OVX, ovariectomy; P, progesterone; T, testosterone.

References

  1. MacLusky NJ, Naftolin F 1981 Sexual Differentiation of the central nervous system. Science 211:1294–1302 [DOI] [PubMed] [Google Scholar]
  2. Wood RI, Mehta V, Herbosa CG, Foster DL 1995 Prenatal testosterone differentially masculinizes tonic and surge modes of luteinizing hormone secretion in the developing sheep. Neuroendocrinology 62:238–247 [DOI] [PubMed] [Google Scholar]
  3. Herbosa CG, Dahl GE, Evans NP, Pelt J, Wood RI, Foster DL 1996 Sexual differentiation of the surge mode of gonadotropin secretion: prenatal androgens abolish the gonadotropin-releasing hormone surge in the sheep. J Neuroendocrinol 8:627–633 [PubMed] [Google Scholar]
  4. Foster DL, Padmanabhan V, Wood RI, Robinson JE 2002 Sexual differentiation of the neuroendocrine control of gonadotrophin secretion: concepts derived from sheep models. Reprod Suppl 59:83–99 [PubMed] [Google Scholar]
  5. 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] [PubMed] [Google Scholar]
  6. 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] [PubMed] [Google Scholar]
  7. 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. Biol Reprod 66:924–933 [DOI] [PubMed] [Google Scholar]
  8. 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. Biol Reprod 78:636–647 [DOI] [PubMed] [Google Scholar]
  9. 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] [PubMed] [Google Scholar]
  10. Olster DH, Foster DL 1986 Control of gonadotropin secretion in the male during puberty: a decrease in response to steroid inhibitory feedback in the absence of an increase in steroid-independent drive in the sheep. Endocrinology 118:2225–2234 [DOI] [PubMed] [Google Scholar]
  11. Ryan KD, Goodman RL, Karsch FJ, Legan SJ, Foster DL 1991 Patterns of circulating gonadotropins and ovarian steroids during the first periovulatory period in the developing sheep. Biol Reprod 45:471–477 [DOI] [PubMed] [Google Scholar]
  12. 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] [PMC free article] [PubMed] [Google Scholar]
  13. Russel AJF, Donney JM, Gunn RG 1969 Subjective assessment of body fat in live sheep. J Agr Sci 72:451–454 [Google Scholar]
  14. Claypool LE, Foster DL 1990 Sexual differentiation of the mechanism controlling pulsatile secretion of luteinizing hormone contributes to sexual differences in the timing of puberty in sheep. Endocrinology 126:1206–1215 [DOI] [PubMed] [Google Scholar]
  15. Foster DL, Jackson LM 2006 Puberty in the sheep. In: Neill JD, ed. Knobil and Neill’s physiology of reproduction. 3rd ed. Amsterdam: Elsevier; 2127–2176 [Google Scholar]
  16. Rawlings NC, Evans AC, Honaramooz A, Bartlewski PM 2003 Antral follicle growth and endocrine changes in prepubertal cattle, sheep and goats. Anim Reprod Sci 78:259–270 [DOI] [PubMed] [Google Scholar]
  17. Foster DL, Lemons JA, Jaffe RB, Niswender GD 1975 Sequential patterns of circulating luteinizing hormone in female sheep from early postnatal life through the first estrous cycles. Endocrinology 97:975–994 [DOI] [PubMed] [Google Scholar]
  18. Huffman LJ, Inskeep EK, Goodman RL 1987 Changes in episodic luteinizing hormone secretion leading to puberty in the lamb. Biol Reprod 37:755–761 [DOI] [PubMed] [Google Scholar]
  19. Robinson JE, Forsdike RA, Taylor JA 1999 In utero exposure of female lambs to testosterone reduces the sensitivity of the gonadotropin-releasing hormone neuronal network to inhibition by progesterone. Endocrinology 140:5797–5805 [DOI] [PubMed] [Google Scholar]
  20. Foster DL 1984 Preovulatory gonadotropin surge system of prepubertal female sheep is exquisitely sensitive to the stimulatory feedback action of estradiol. Endocrinology 115:1186–1189 [DOI] [PubMed] [Google Scholar]
  21. Hauger RL, Karsch FJ, Foster DL 1977 A new concept for control of the estrous cycle of the ewe based upon temporal relationships between luteinizing hormone, estradiol, and progesterone in peripheral serum and evidence that progesterone inhibits tonic LH secretion. Endocrinology 101:807–817 [DOI] [PubMed] [Google Scholar]
  22. Ebling FJ, Wood RI, Karsch FJ, Vannerson LA, Suttie JM, Bucholtz DC, Schall RE, Foster DL 1990 Metabolic interfaces between growth and reproduction. III. Central mechanisms controlling pulsatile luteinizing hormone secretion in the nutritionally growth-limited female lamb. Endocrinology 126:2719–2727 [DOI] [PubMed] [Google Scholar]
  23. Niswender GD, Reichert Jr LE, Midgley Jr AR, Nalbandov AV 1969 Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 84:1166–1173 [DOI] [PubMed] [Google Scholar]
  24. Evans NP, Dahl GE, Glover BH, Karsch FJ 1994 Central regulation of pulsatile gonadotropin-releasing hormone (GnRH) secretion by estradiol during the period leading up to the preovulatory GnRH surge in the ewe. Endocrinology 134:1806–1811 [DOI] [PubMed] [Google Scholar]
  25. Breen KM, Billings HJ, Wagenmaker ER, Wessinger EW, Karsch FJ 2005 Endocrine basis for disruptive effects of cortisol on preovulatory events. Endocrinology 146:2107–2115 [DOI] [PubMed] [Google Scholar]
  26. 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] [PubMed] [Google Scholar]
  27. Veldhuis JD, Johnson ML 1986 Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 250(4 Pt 1):E486–E493 [DOI] [PubMed] [Google Scholar]
  28. Ebling FJ, Foster DL 1988 Photoperiod requirements for puberty differ from those for the onset of the adult breeding season in female sheep. J Reprod Fertil 84:283–293 [DOI] [PubMed] [Google Scholar]
  29. Steckler T, Manikkam M, Inskeep EK, Padmanabhan V 2007 Developmental programming: follicular persistence in prenatal testosterone-treated sheep is not programmed by androgenic actions of testosterone. Endocrinology 148:3532–3540 [DOI] [PubMed] [Google Scholar]
  30. 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. Biol Reprod 75:844–852 [DOI] [PubMed] [Google Scholar]
  31. Manikkam M, Thompson RC, Herkimer C, Welch KB, Flak J, Karsch FJ, Padmanabhan V 2008 Developmental programming: impact of prenatal testosterone excess on prenatal and postnatal gonadotropin regulation in sheep. Biol Reprod 78:648–660 [DOI] [PubMed] [Google Scholar]
  32. Wood RI, Newman SW, Lehman MN, Foster DL 1992 GnRH neurons in the fetal lamb hypothalamus are similar in males and females. Neuroendocrinology 55:427–433 [DOI] [PubMed] [Google Scholar]
  33. Kim SJ, Foster DL, Wood RI 1999 Prenatal testosterone masculinizes synaptic input to GnRH neurons in sheep. Biol Reprod 61:599–605 [DOI] [PubMed] [Google Scholar]
  34. Wood RI, Kim SJ, Foster DL 1996 Prenatal androgens defeminize activation of GnRH neurons in response to estradiol stimulation. J Neuroendocrinol 8:617–625 [PubMed] [Google Scholar]
  35. Foster DL, Ryan KD, Goodman RL, Legan SJ, Karsch FJ, Yellon SM 1986 Delayed puberty in lambs chronically treated with oestradiol. J Reprod Fertil 78:111–117 [DOI] [PubMed] [Google Scholar]
  36. Ozturk M, Smith RF, Dobson H 1998 Effect of prolonged exposure to oestradiol on subsequent LH secretion in ewes. J Reprod Fertil 114:1–9 [DOI] [PubMed] [Google Scholar]
  37. Arrequin-Arevalo JA, Nett TM 2006 A nongenomic action of estradiol as the mechanism underlying the acute suppression of secretion of luteinizing hormone in ovariectomized ewes. Biol Reprod 74:202–208 [DOI] [PubMed] [Google Scholar]
  38. Hulshof SC, Figueiredo JR, Beckers JF, Bevers MM, van der Donk JA, van den Hurk R 1995 Effects of fetal bovine serum, FSH and 17beta-estradiol on the culture of bovine preantral follicles. Theriogenology 44:217–226 [DOI] [PubMed] [Google Scholar]
  39. Colburn T, vom Saal FS, Soto AM 1993 Developmental effects of endocrine-disrupting chemicals in wildlife and humans. Environ Health Perspect 10:378–384 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES