Biol Reprod Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

BOR - Papers in Press, published online ahead of print December 19, 2007.
Biol Reprod 2007, 10.1095/biolreprod.107.063347
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
78/4/648    most recent
biolreprod.107.063347v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.
Agricola
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.
BIOLOGY OF REPRODUCTION 78, 648–660 (2008)
DOI: 10.1095/biolreprod.107.063347
© 2008 by the Society for the Study of Reproduction, Inc.

Neuroendocrinology; 

Developmental Programming: Impact of Prenatal Testosterone Excess on Pre- and Postnatal Gonadotropin Regulation in Sheep1

Mohan Manikkam 4 7, Robert C. Thompson 5 7, Carol Herkimer 4 7, Kathleen B. Welch 8, Jonathan Flak 3 4 7, Fred J. Karsch 6 7, and Vasantha Padmanabhan 2 4 6 7

Departments of Pediatrics,4 Psychiatry,5 and Molecular and Integrative Physiology,6 the Reproductive Sciences Program,7 and the Center for Statistical Consultation and Research,8 University of Michigan, Ann Arbor, Michigan 48109

ABSTRACT

The goal of this study was to explore mechanisms that mediate hypersecretion of LH and progressive loss of cyclicity in female sheep exposed during fetal life to excess testosterone. Our working hypothesis was that prenatal testosterone excess, by its androgenic action, amplifies GnRH-induced LH (but not FSH) secretion and, thus, hypersecretion of LH in adulthood, and that this results from altered developmental gene expression of GnRH and estradiol (E2) receptors, gonadotropin subunits, and paracrine factors that differentially regulate LH and FSH synthesis. We observed that, relative to controls, females exposed during fetal life to excess testosterone, as well as the nor-aromatizable androgen dihydrotestosterone, exhibited enhanced LH but not FSH responses to intermittent delivery of GnRH boluses under conditions in which endogenous LH (GnRH) pulses were suppressed. Luteinizing hormone hypersecretion was more evident in adults than in prepubertal females, and it was associated with development of acyclicity. Measurement of pituitary mRNA concentrations revealed that prenatal testosterone excess induced developmental changes in gene expression of pituitary GnRH and E2 receptors and paracrine modulators of LH and FSH synthesis in a manner consistent with subsequent amplification of LH release. Together, this series of studies suggests that prenatal testosterone excess, by its androgenic action, amplifies GnRH-induced LH response, leading to LH hypersecretion and acyclicity in adulthood, and that this programming involves developmental changes in expression of pituitary genes involved in LH and FSH release.

activin, developmental biology, estradiol receptor, follistatin, FSH, gonadotropin-releasing hormone receptor, inhibin, LH, neuroendocrinology, pituitary, pituitary responsiveness

INTRODUCTION

Treatment of pregnant sheep with testosterone from Days 30 to 90 of the 150-day gestation period leads to reproductive dysfunction in adult female offspring [1–8]. Disruptions include a progressive deterioration of estrous cycles [1–3], development of multifollicular ovaries [4, 5] and, ultimately, infertility [1, 3]. Associated with these abnormalities is an alteration in gonadotropin secretion manifested as hypersecretion of LH but not FSH [7, 8]. Hypersecretion of LH associated with reproductive abnormalities is also evident following prenatal exposure to excess testosterone in female rhesus monkeys [9], mice [10], and rats [11], and it is also seen in women with polycystic ovary syndrome [12, 13].

Although reproductive consequences of prenatal exposure of female sheep to excess testosterone are well described, the mechanisms producing these detrimental effects are not understood. For example, it is unclear whether the selective increase in LH secretion results from androgenic action or aromatization of androgens to estrogens. In addition, prenatal testosterone treatment leads to reduced responsiveness to E2 negative feedback, characterized by increased frequency and a tendency toward an increase in amplitude of LH pulses [8]. It is not known whether the increased LH pulse amplitude is due to heightened pituitary responsiveness to GnRH, altered amounts of GnRH or E2 receptors, and/or changes in gonadotropin gene expression. Further, the reduced responsiveness to E2 negative feedback in sheep exposed to excess testosterone during fetal life is evident only for LH; feedback effects of E2 on FSH are not affected. This suggests involvement of pituitary paracrine regulators, such as activin or inhibin, and their receptors or binding proteins (e.g., follistatin).

The goal of this study was to explore mechanisms that mediate hypersecretion of LH in prenatal testosterone-treated female sheep. Our working hypothesis was that prenatal testosterone excess, by its androgenic action, amplifies GnRH-induced LH (but not FSH) secretion and, thus, hypersecretion of LH in adulthood. This increased pituitary responsiveness to GnRH results from developmental changes in GnRH and E2 receptors, gonadotropin subunits, and paracrine factors that differentially regulate LH and FSH secretion.

MATERIALS AND METHODS

Breeding, Maintenance, and Prenatal Treatment

All procedures were approved by the Institutional Animal Care and Use Committee of the University of Michigan and were in conformity with the National Institutes of Health Guide for Use and Care of Animals. Suffolk sheep ages 2–3 yr were purchased from local farmers and were maintained in a farm inspected by the US Department of Agriculture and approved by the University of Michigan Department of Laboratory Animal Medicine. Prior to breeding, sheep were group fed with 0.5 kg shelled corn and 1 kg alfalfa hay per female sheep per day. During October to November of three successive years (2001–2003), ewes were mated by raddled rams. Pregnant sheep were maintained on pasture under natural photoperiods with a supplement feed of 1.25 kg alfalfa/brome mix hay per female sheep. After birth, each mother with its lambs was individually housed for the first 3 days and then group housed in a barn under natural photoperiods except for a 60-watt bulb in the lamb feed area at night. All lambs were weaned at 8 wk and maintained in Sheep Research Facility (Ann Arbor, MI), where they were maintained outdoors and fed commercial feed pellets ad libitum.

Prenatal treatments consisted of intramuscular injections twice a week of testosterone (T1875; Sigma-Aldrich Corp., St. Louis, MO) or dihydrotestosterone (DHT) propionate (A2595–000; Steraloids Inc., Newport, RI), 100 mg in 2 ml corn oil, to pregnant sheep from Day 30 to Day 90 of gestation. Control pregnant sheep received no injections because we have found no difference in cycle dynamics of offspring born to vehicle-injected and noninjected mothers [14].

Study 1: Pituitary Responsiveness to GnRH

This study tested the hypothesis that prenatal testosterone, via its androgenic action, increases pituitary responsiveness to GnRH and that this action is specific for LH (i.e., the FSH response is not affected). The approach was to compare pituitary responsiveness to GnRH in control females and females exposed prenatally to exogenous testosterone or DHT as well as aromatizable and nonaromatizable androgens, respectively (Fig. 1). Only one offspring from each mother was used when twin female births were involved. Responsiveness to GnRH was tested twice, first at 5.5 wk (prepubertal), when the lambs were ovary intact, and second at 23 mo of age, after they had been ovariectomized at the end of breeding season and studied 2 mo later during the second anestrus.


Figure 01
View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 1. Schematic showing experimental design of study 1, pituitary responsiveness to GnRH; and study 2, adult hypergonadotropism. In study 1, one 30-mm E2 implant was inserted subcutaneously in each sheep and retained throughout the experiment. Shaded boxes indicate periods of blood sampling. Pre-GnRH samples were taken 5 days after implant insertion in prepubertal sheep and 11 days after implant insertion in adult sheep.

Prepubertal GnRH testing involved seven control, six prenatal testosterone-treated, and five prenatal DHT-treated females. Adult testing involved five control, four prenatal testosterone-treated, and three prenatal DHT-treated females. To block pulsatile LH/GnRH release, sheep were inserted subcutaneously with one 30-mm Silastic E2 implant [15]. Prepubertal and anestrous control females are extremely sensitive to E2 negative feedback [16, 17], and this E2 treatment is known to abolish endogenous LH/GnRH pulsatility [16, 18]. Jugular blood samples (3 ml) were taken at 24-min intervals for 6 h at 1–2 wk after insertion of E2 implants to confirm the absence of LH pulses (pre-GnRH bleed). GnRH (2 ng/kg; Luteinizing hormone-releasing hormone human, acetate salt L7134; Sigma, St. Louis, MO) was then administered intravenously every 90 min for 48 h via a jugular catheter. This dose of GnRH produces LH responses of magnitude similar to those seen in prepubertal females [16]. Blood samples (3 ml) were collected into heparinized tubes at 10-min intervals for 9 h (spanning six GnRH injections) beginning 42 h after initial injection for characterization of LH and FSH responses.

Study 2: Prenatal Testosterone Effects on Adult Gonadotropin Release

In view of the progressive deterioration of the reproductive axis seen in prenatal testosterone-treated females [1–3], study 2 was performed during the second breeding season to determine whether the selective increase in LH release seen in prepubertal females in the first year of life [7, 8] persists into adulthood (second year of life). Adult control (n = 6) and prenatal testosterone-treated (n = 8) females (22 mo old) were given two intramuscular injections of prostaglandin F2{alpha} (PGF2{alpha}, 20 mg; Lutalyse; Pfizer Inc., Kalamazoo, MI) 11 days apart to induce regression of the corpora lutea, and thus synchronize the follicular phase of the cycle within a group of females (Fig. 1). Jugular blood samples were collected at 10-min intervals from 23 to 28 h after the second PGF2{alpha} injection for measurement of LH (all samples) and FSH (hourly samples). Twice-weekly progesterone measures were taken from the start of the second breeding season to determine whether the prenatal testosterone-treated females were cycling at the time of the experiment.

Study 3: Prenatal Testosterone Effects on Gonadotropin Subunits, GnRH Receptor, and Estrogen Receptors

Study 3 was undertaken to examine whether prenatal testosterone excess caused developmental changes in pituitary gene expression that might account for the selective increase in LH secretion and pituitary responsiveness to GnRH. For this purpose, real-time PCR was used to quantify changes in mRNA expression patterns of gonadotropin subunits (glycoprotein hormones, alpha peptide [CGA], LHB, FSHB), GnRH receptor (GNRHR), and E2 receptors (ESR1, ESR2) at four developmental stages. Table 1 summarizes the treatments, developmental stages, and number of animals studied. Where possible, umbilical arterial levels of LH and FSH were measured to relate changes in fetal gonadotropin subunit mRNA expression with fetal LH and FSH secretion. When twin female fetuses were involved, both fetuses were studied. Values of these two female fetuses were averaged to ensure independence of individual measures in all analyses.


View this table:
[in this window]
[in a new window]
[Download PPT slide]
  		TABLE 1. Details of the number of female sheep from treatment groups at various ages used in the pituitary gene expression study.

Pregnant ewes were anesthetized with 1000–1500 mg pentobarbital intravenously (Nembutal Na; Abbott Laboratories, Chicago, IL); anesthesia was maintained by inhalation of 1%–2% halothane (Halocarbon Laboratories, Riveredge, NJ) in oxygen-nitric oxide mixture (2:1). A midventral laparatomy was performed to expose the uterus with fetuses. Blood samples (~5 to 10 ml) were collected from umbilical arteries using a Butterfly-21 infusion set (no. 4492; Abbott Laboratories, North Chicago, IL). The ewes then were killed by a barbiturate overdose (Fatal Plus; Vortech Pharmaceuticals, Dearborn, MI), and fetuses were removed. The entire procedure took less than 30 min. The adult female sheep used in this study were the ones used in study 2; they were also killed by a barbiturate overdose. Pituitaries collected from female fetuses and adult females were frozen in liquid nitrogen and stored at –80°C until real-time PCR analysis.

Study 4: Prenatal Testosterone Effects on Pituitary Paracrine Regulators of Gonadotropins

This study was undertaken to determine whether prenatal testosterone excess causes developmental changes in the mRNA expression pattern of pituitary paracrine modulators of gonadotropins that might contribute to differential changes in LH and FSH secretion and pituitary responsiveness to GnRH. Pituitary mRNA expression patterns of the following modulators were determined by real-time PCR on RNA obtained from pituitaries collected in Study 3: inhibin alpha (INHA), inhibin beta A (INHBA), inhibin beta B (INHBB), their binding proteins—follistatin (FST), immunoglobulin superfamily member I, also known as inhibin binding protein (IGSF1), transforming growth factor beta receptor III, also known as betaglycan (TGFBR3), and the receptors through which they signal—activin A receptor type IA (ACVR1), activin A receptor type IB (ACVR1B), and activin A receptor type IIA (ACVR2A).

Hormone Assays

Plasma concentrations of LH and FSH were measured by validated radioimmunoassays described elsewhere [19, 20]. The sensitivity (2 SDs from the buffer control) of the LH assay was 1.1 ± 0.2 ng/ml of NIH-LH-S12 (n = 19 assays). Mean intraassay coefficients of variation at 80% and 20% displacement points were 11.6% ± 1.4% and 5.9% ± 0.6%, respectively. The interassay coefficients of variation in three quality control pools averaging 14.4, 12.9, and 23.0 ng/ml were 4.9%, 6.2%, and 7.3%, respectively. The sensitivity of the FSH assay was 0.9 ± 0.1 ng/ml of NIDDK-oFSH-1 (n = 10 assays). Mean intraassay coefficients of variation at 80% and 20% displacement points were 7.5% ± 1.0% and 3.8% ± 0.7%, respectively. The interassay coefficients of variation based on two quality control pools from ovary-intact, ovariectomized ewes averaging 4.6 and 11.1 ng/ml were 10.7% and 7.9%, respectively.

Pituitary RNA Isolation and Real-Time PCR

Pituitary RNA was isolated using TRIzol reagent (Invitrogen), measured by spectrophotometer or Ribogreen assay (Molecular Probes, Eugene, OR), and evaluated in a denaturing formaldehyde gel. Purification of RNA was done by DNase treatment and column clean up (Qiagen, Valencia, CA). First-strand cDNA from RNA was prepared using iScript cDNA synthesis kit (Bio-Rad, Hercules, CA) and column purified (QIAquick PCR purification kit; Qiagen). Gene-specific primers for real-time PCR (see details of primers in Table 2) [21–34] were designed using Laser gene Primer select software (DNASTAR Inc., Madison, WI) and were synthesized commercially (Invitrogen, Grand Island, NY). All cDNA samples were processed in parallel, and cycle threshold (CT) values were obtained from three replicate runs. Real-time PCR reactions (50 µl) were run in DNase- and RNase-free 96-well plates with IQ SYBR Green Supermix (Bio-Rad). The amplification conditions were: cycle 1, 95°C for 30 sec; cycle 2 (35x), step 1 denaturation at 95°C for 30 sec, step 2 annealing at 61°C (glyceraldehyde-3-phosphate dehydrogenase [GAPDH], GNRHR, LHB, FSHB) or 55°C (CGA) or 59.7°C (ESR1) or 56.6°C (ESR2) for 30 sec or 57.5°C for 30 sec for activin series of genes (INHBB, INHA, FST, ACVR1, ACVR1B, ACVR2A, IGSF1, and TGFBR3), step 3 elongation at 72°C for 30 sec (data collection and real-time analysis enabled here); cycle 3, final elongation at 72°C for 5 min; and cycle 4, hold at 4°C. Polymerase chain reaction product specificity was checked by melt curves and agarose gel electrophoresis. Three housekeeping genes were evaluated in this study: GAPDH, actin beta (ACTB; annealing at 57.5°C), and 18S ribosomal RNA (annealing at 60°C). Ratios calculated using GAPDH and ACTB as the housekeeping genes were qualitatively similar, but subtle differences were evident with 18S ribosomal RNA. Gene expression data are reported as mean fold-change in the ratio of target gene to GAPDH.


View this table:
[in this window]
[in a new window]
[Download PPT slide]
  		TABLE 2. Oligonucleotide primers used for real-time polymerase chain reaction.

The CT values obtained for each sample were analyzed by 2–{Delta}{Delta}CT method [35]. This method enables the exponential CT values to be converted into linear values of fold-change in mRNA amounts. It also enables one to remove variations in the amount of RNA used in first-strand synthesis and the amount of cDNA used for PCR. This 2–{Delta}{Delta}CT method of quantification was validated by examining the efficiency of amplification of target gene and GAPDH using real-time PCR. Using reverse transcriptase, cDNA was synthesized from 1 µg total RNA isolated from a nonexperimental sheep pituitary and DNase treated before PCR. Concentrations of cDNA over a 500-fold range (1, 0.2, 0.02, and 0.002 of original cDNA) were amplified by real-time PCR using target gene and GAPDH-specific primers. The {Delta}CT (CT for target gene minus CT for GAPDH) was calculated for each cDNA dilution, and the data were fit using least-squares linear regression analysis. The slope of the regression line was 0.029, suggesting that the amplification efficiencies of target genes and GAPDH were approximately equal to two. In this experiment, the formula 2–{Delta}{Delta}CT for control group was treated as 2–({Delta}CT1–{Delta}CT2), such that {Delta}CT1 = individual CT of GNRHR – individual CT of GAPDH, and {Delta}CT2 = mean CT of GNRHR – mean CT of GAPDH of the same sample. For the treated groups, the formula was treated as 2–({Delta}CT1–{Delta}CT2), such that {Delta}CT1 = individual CT of GNRHR of treated sample – individual CT of GAPDH of treated sample, and {Delta}CT2 = mean CT of GNRHR of control sample – mean CT of GAPDH of the control sample.

Statistical Analyses

In study 1, differences in mean LH and FSH concentrations, LH pulse amplitudes, and LH:FSH ratios among control, prenatal testosterone-treated, and DHT-treated females were analyzed using ANOVA with Tukey adjustment for post-hoc tests. LH pulse amplitude was calculated as the mean amplitude (peak minus preceding nadir) arising from the six administered GnRH pulses from individual animals.

For study 2, LH pulses in control and prenatal testosterone-treated adults were identified using the Cluster algorithm [36]. The minimum number of points in a peak and a nadir were set at 1 and 1, respectively, and the t-statistic values used to identify a significant increase from the preceding nadir and a decrease to the following nadir were 1 and 1, respectively. Group differences in mean LH and FSH concentrations, LH pulse frequency, LH pulse amplitude, and LH:FSH ratios were analyzed using ANOVA with Tukey adjustment for post-hoc tests. LH pulse amplitude was calculated as the mean from the Cluster-identified pulses from individual animals.

For studies 3 and 4, comparison of pituitary gene expression for the different treatments was carried out on the ratio of each gene to GAPDH expression (2–{Delta}{Delta}CT data). To allow comparison of the ratios across developmental ages, the ratios of target genes to GAPDH at developmental ages were calculated using the value for control at the earliest measurable time point as a calibrator in the calculations of 2–{Delta}{Delta}CT. For adult samples, calculation of fold-change in gene expression was done with the control group as calibrator. The mean of three replicate ratios was first calculated for each lamb, then the lamb means were calculated, so the mother was the unit of analysis. Outliers (defined using Tukey criteria [37]: values greater than the 75th percentile + 3*IQR, or less than the 25th percentile – 3*IQR) were excluded (separate analyses conducted without excluding outliers also yielded qualitatively similar results). A two-way ANOVA (allowing for unequal variances for the control and testosterone-treated animals) was used to compare the effects of treatment, age, and the treatment by age interaction for the fetal ages (gestation 65, 90, and 140 days). Post-hoc comparisons of means were carried out for specific fetal-age comparisons within each treatment and for the effect of treatment at each age. The effect of treatment on dam means at age 2 yr, after excluding outliers, was analyzed separately using independent sample t-tests. An alpha level of 0.05 was used to determine significance for all significance tests. SAS release 9.1.3 for Windows (SAS Institute Inc., Cary, NC) was used for statistical analyses.

RESULTS

Study 1: Pituitary Responsiveness to GnRH

Circulating LH and FSH patterns of three control, three prenatal testosterone-treated, and three DHT-treated lambs during the prepubertal period (ovary intact) prior to and during exogenous GnRH and composite data for all animals are presented in Figure 2. Insertion of a 30-mm E2 implant blocked endogenously generated LH pulses in all control and prenatal testosterone-treated females but not all prenatal DHT-treated females (only two of five blocked) during the 6-h sampling window prior to GnRH treatment. Each GnRH injection induced a pulse of LH in all animals (Fig. 2). Compared with control females, the resulting mean LH concentration was higher in prenatal testosterone-treated and prenatal DHT-treated females (one prenatal testosterone-treated lamb did not respond and was not included in this analysis; P < 0.05) compared with that of the control group. LH pulse amplitudes (ng/ml) also were significantly higher in prenatal testosterone-treated (P < 0.05) and prenatal DHT-treated females (P < 0.01) compared with control females. Mean circulating FSH concentration did not differ among groups, and FSH amplitudes were not calculated due to ambiguity in defining FSH pulses. Mean LH:FSH ratios also were higher in prenatal testosterone-treated (P < 0.01) and DHT-treated (P < 0.01) groups compared with the control group.


Figure 02
View larger version (34K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 2. Top: Representative LH and FSH profiles in ng/ml of three each of control (432, 442, 441), prenatal testosterone-treated (414, 415, 436), and prenatal DHT-treated (402, 410, 422) prepubertal female lambs. Patterns of LH and FSH released after E2 implant insertion are shown in panels labeled "Pre." LH and FSH responses to GnRH injections given every 90 min are shown in panels labeled "Post." Arrows indicate time of GnRH injections. Bottom: Mean LH and FSH concentrations and amplitudes of GnRH-induced LH releases, and LH:FSH ratios in prepubertal control (C), prenatal testosterone-treated (T), and prenatal DHT-treated females. Data were analyzed by ANOVA. Asterisks indicate significant treatment differences (*P < 0.05; **P < 0.01).

Figure 3 depicts results prior to and during the GnRH challenge in these same animals during adulthood (ovariectomized). Treatment with E2 blocked LH pulses in all control, three of four prenatal testosterone-treated, and two of three prenatal DHT-treated females during the 6-h sampling window prior to GnRH challenge. Each GnRH injection induced an LH response in all females (Fig. 3, post). Mean LH concentration was higher in prenatal testosterone-treated females (P < 0.01) but not in prenatal DHT-treated females. LH pulse amplitudes were significantly higher (P < 0.01) in prenatal testosterone-treated females. In spite of the small number of animals studied, LH pulse amplitudes tended to be higher (P = 0.07) in prenatal DHT-treated females compared with control females. Mean circulating FSH concentration did not differ among the three groups. Mean LH:FSH ratio was higher in prenatal testosterone-treated group (P < 0.01), but not the prenatal DHT-treated group, compared with the control group.


Figure 03
View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 3. Top: Representative LH and FSH profiles of three control, prenatal testosterone-treated (T), and prenatal DHT-treated adult ovariectomized females (same animals as shown in Fig. 2). Patterns of LH and FSH released after E2 implant insertion are shown in panels labeled "Pre." LH and FSH responses to GnRH injections given every 90 min are shown in panels labeled "Post." Arrows indicate time of GnRH injections. Bottom: Mean LH and FSH concentrations and amplitudes of GnRH-induced LH releases and LH:FSH ratios in control (C), prenatal testosterone-treated (T), and prenatal DHT-treated females. Data were analyzed by ANOVA. Asterisks indicate significant treatment differences (*P < 0.05; **P < 0.01).

Study 2: Prenatal Testosterone Effects on Adult Gonadotropin Release

Twice-weekly progesterone measures preceding the study found that five of the eight prenatal testosterone-treated females were not cycling at the time of the study. LH and FSH profiles from four each of the adult ovary-intact control and prenatal testosterone-treated females, mean LH and FSH concentrations, LH pulse amplitude, and LH:FSH ratios of the two treatment groups are shown in Figure 4. When all animals were included in the analysis, the mean LH concentration was higher (P < 0.01), and the mean LH pulse amplitude tended to increase (P = 0.07) in prenatal testosterone-treated females, but there was no group difference in pulse frequency (P > 0.05). LH:FSH ratio also was significantly higher in prenatal testosterone-treated females compared with control females (P < 0.05).


Figure 04
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 4. Top: Representative LH and FSH profiles in ng/ml of four each of control and prenatal testosterone (T)-treated adult females. Closed circles depict peaks of pulses identified by the Cluster analysis [36]. Bottom: Mean LH concentration, LH pulse frequency, LH pulse amplitude, mean FSH concentration, and LH:FSH ratios in adult control (open bars, C) and prenatal testosterone (T)-treated females (closed bars summarize results for the entire group, shaded bars for the cycling group, and hatched bars for the acyclic group). Data were analyzed by ANOVA. Asterisks indicate significant treatment differences (*P < 0.05; **P < 0.01).

Because five of the eight prenatal testosterone-treated females were not cycling at the time of study, separate analyses also were carried out to compare values in cyclic vs. acyclic prenatal testosterone-treated females with those of controls. The mean LH concentration and LH pulse amplitude were significantly higher in the acyclic prenatal testosterone-treated females compared with controls (Fig. 4). These group differences were not observed for prenatal testosterone-treated females that expressed estrous cycles at the time of the study. There were no differences among pulse frequencies of control, cycling prenatal testosterone-treated, or acyclic prenatal testosterone-treated groups. Mean FSH concentrations showed a trend for an increase in both acyclic and cyclic prenatal testosterone-treated females. LH:FSH ratios tended to be higher in acyclic testosterone females (P = 0.06) but not in cyclic testosterone females compared with control females.

Study 3: Prenatal Testosterone Effects on Gonadotropin Subunit, GnRH Receptor, and E2 Receptors

Ratios of pituitary mRNA expression of CGA, LHB, FSHB, GNRHR, ESR1, and ESR2 with that of GAPDH from control and prenatal testosterone-treated female fetuses (Days 65, 90, and 140) are presented in Figure 5. All three gonadotropin subunit mRNAs were influenced by age and treatment (Fig. 5, top). There were significant age (P < 0.01), treatment (P < 0.05), and age by treatment (P < 0.05) effects with CGA mRNA expression. CGA mRNA expression did not differ between Fetal Days 65 and 90 but increased from Days 90–140 in both control and prenatal testosterone-treated fetuses (control: P < 0.01; testosterone: P < 0.05). Prenatal testosterone excess reduced CGA mRNA expression significantly at Fetal Day 140 (P < 0.05) compared with control fetuses, but not at other fetal ages. A significant age effect (P < 0.05) was evident with LHB mRNA. Prenatal testosterone excess reduced the expression of LHB mRNA at Fetal Days 65 (P < 0.005) and 90 (P < 0.05) but not Day 140. There was also a significant age (P < 0.005), treatment (P < 0.05), and age by treatment (P < 0.05) effect in the expression patterns of FSHB mRNA. FSHB mRNA expression decreased from Day 65 to Day 90 in both control and prenatal testosterone-treated fetuses (control: P < 0.005; prenatal testosterone-treated: P < 0.01), increasing thereafter from Day 90 to Day 140 in both groups (control: P < 0.05; prenatal testosterone-treated: P < 0.01). Prenatal testosterone treatment increased FSHB mRNA expression in Day 65 fetuses (P < 0.05) but decreased (P < 0.05) it in Day 90 fetuses.


Figure 05
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 5. Mean (±SEM) ratios of pituitary mRNA expression of gonadotropin subunits (CGA, LHB, and FSHB), GnRH/E2 receptors (GNRHR, ESR1, and ESR2) to that of GAPDH at Fetal Days 65, 90, and 140 in control (open bars) and prenatal testosterone-treated (filled bars) females. Asterisks indicate significant treatment differences. Single asterisk indicates P < 0.05, two asterisks indicate P < 0.01. Letters with different superscripts within a given treatment group (a: control; b: prenatal testosterone treated) indicate significant age affects. One outlier was excluded from CGA and ESR2 mRNA measures of the prenatal testosterone-treated Day 140 group. n.d., not detectable.

Gonadotropin-releasing hormone and E2 receptor mRNAs are presented in Figure 5 (bottom panels). GNRHR mRNA was not detectable on Fetal Days 65 and 90 (except in two prenatal testosterone-treated fetuses at Day 90). GNRHR mRNA tended to be higher in prenatal testosterone-treated females compared with controls (P = 0.095) at Fetal Day 140. There was a significant age (P < 0.005), treatment (P < 0.001), and age by treatment interaction (P < 0.01) with ESR1 but not ESR2 mRNA expression. ESR1 mRNA expression was the highest on Fetal Day 65 and declined from Fetal Day 65 to Day 90 (P < 0.001) and from Day 90 to Day 140 (P < 0.05) in control fetuses. There was no age effect on ESR1 mRNA expression in prenatal testosterone-treated females. Expression of ESR1 mRNA was lower in prenatal testosterone-treated vs. control fetuses on Fetal Day 65 (P < 0.05), with a tendency toward a decline also on Fetal Day 90 (P = 0.066).

Circulating Levels of LH and FSH

Presence of fibrin in samples prevented accurate assessment of LH and FSH from Fetal Day 65 and Fetal Day 90 samples, and hence is not reported. Both LH and FSH measures were available from only four control and five prenatal testosterone-treated Day 140 female fetuses. The mean LH concentrations in umbilical arterial blood of Day 140 female fetuses averaged 3.3 ± 1.2 and 5.9 ± 4.8 ng/ml, respectively, for control and prenatal testosterone-treated groups. Corresponding mean FSH values were 2.4 ± 0.6 and 3.1 ± 1.4 ng/ml, respectively. The ratio of LH:FSH for this subset of prenatal testosterone-treated females was numerically higher (5.5 ± 2.2) than controls (1.1 ± 0.4) but did not differ significantly (P = 0.12).

Study 4: Prenatal Testosterone Effects on Pituitary Paracrine Regulators of Gonadotropins

Developmental changes in expression of mRNA for LH and FSH modulators at Days 65, 90, and 140 are shown in Figure 6.


Figure 06
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 6. Mean (±SEM) ratios of pituitary mRNA expression of INHA, INHBB (paracrine mediators of FSH), inhibin/activin binding proteins (FST, IGSF1, TGFBR3), and activin receptors (ACVR1, ACVR1B, and ACVR2A) to that of GAPDH at Fetal Days 65, 90, and 140. Results from control and prenatal testosterone-treated ovine females are shown as open and filled bars, respectively. Asterisks indicate significant treatment differences (single asterisk indicates P < 0.05; two asterisks indicate P < 0.01). Letters with differing superscripts within a given treatment group (a: control; b: prenatal testosterone-treated) indicate significant age affects (P < 0.05). Note FST measures in the Day 65 groups are based n = 2 (reliable estimates were not obtained from the others). n.d., not detectable.

Inhibin and activin. A significant age effect (P < 0.01) and a tendency toward a treatment (P = 0.078) effect were evident with INHA mRNA expression (Fig. 6, top left). INHA mRNA decreased from Fetal Day 65 to Day 90 in both control (P < 0.01) and prenatal testosterone-treated (P < 0.01) fetuses, but did not decrease from Day 90 to Day 140. Prenatal testosterone treatment tended to decrease INHA mRNA expression in Fetal Day 65 (P = 0.058) but not in other fetal ages. Inhibin beta A (INHBA) mRNA was not detected in sheep pituitary at any age studied (not shown). In terms of INHBB, a significant age (P < 0.01) and age by treatment effect (P < 0.05) was evident. INHBB mRNA expression decreased from Fetal Day 65 to Day 90 in both control (P < 0.001) and prenatal testosterone-treated (P < 0.01) fetuses. This expression increased from Fetal Day 90 to Day 140 in prenatal testosterone-treated (P < 0.01) but not in control fetuses. Prenatal testosterone treatment reduced INHBB mRNA expression on Fetal Day 90 (P < 0.05) but increased (P < 0.05) this expression on Fetal Day 140 compared with controls.

Inhibin/activin binding proteins. Prenatal testosterone excess also altered the developmental expression of binding proteins of activin and inhibin. There was an age (P <0.001) and an age by treatment (P < 0.05) effect in FST mRNA expression. Expression was lower in Day 140 compared with Day 90 control fetuses (P < 0.001), and a tendency toward a similar decline was also evident in prenatal testosterone-treated (P = 0.061) fetuses. However, FST mRNA expression was higher in Day 140 prenatal testosterone-treated fetuses compared with age-matched control fetuses (P < 0.001). An age effect (P < 0.01) and a tendency for treatment effect (P = 0.097) was evident with mRNA expression of the inhibin binding protein IGSF1. IGSF1 mRNA expression increased from Fetal Day 65 to Day 90 in both control (P < 0.005) and prenatal testosterone-treated (P < 0.05) fetuses, declining to Day 65 levels in both control (P < 0.005) and prenatal testosterone-treated (P < 0.05) Day 140 fetuses. There was no treatment effect of prenatal testosterone excess on IGSF1 mRNA expression during any of the fetal ages studied. A tendency toward an age effect (P = 0.08) and a significant age by treatment effect (P < 0.005) was evident in the expression of TGFBR3 mRNA (also known as betaglycan). TGFBR3 mRNA expression declined from Day 65 to Day 90 in control (P < 0.005) but not in prenatal testosterone-treated fetuses. TGFBR3 mRNA expression was lower in prenatal testosterone-treated females at Fetal Day 65 compared with controls (P < 0.005) but did not differ across other fetal ages.

Activin receptors. A prenatal testosterone treatment effect was observed in two of the three activin receptor mRNAs measured. Overall analyses revealed no age, treatment, or age by treatment effect with ACVR1 mRNA expression. An age effect of ACVR1B mRNA was evident in that it was detectable on Fetal Days 90 and 140 but not Day 65. Prenatal testosterone excess increased (P < 0.05) ACVR1B mRNA expression on Fetal Day 140 compared with controls. A significant age (P < 0.05) and age by treatment (P < 0.05) effect was also evident with ACVR2A mRNA expression. ACVR2A mRNA was first detectable on Fetal Day 90. ACVR2A expression level was higher (P < 0.01) in Day 90 prenatal testosterone-treated fetuses compared with Day 140 fetuses. Prenatal testosterone treatment increased (P < 0.05) ACVR2A mRNA expression in Day 90 fetuses.

Prenatal Testosterone Effects on Adult Pituitary mRNA Expression

Pituitary expression of all of the above regulators in pituitaries collected from adults is shown in Figure 7. Prenatal testosterone treatment increased CGA and ACVR1 mRNA expression (P < 0.05) but had no effect on expression of all other mRNAs.


Figure 07
View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 7. Mean (±SEM) ratios of pituitary mRNA expression of gonadotropin subunits (CGA, LHB, FSHB) GnRH/E2 receptors (GNRHR, ESR1, and ESR2), paracrine mediators of FSH (INHA, INHBB), inhibin/activin binding proteins (FST, IGSF1, TGFBR3), and activin receptors (ACVR1, ACVR1B, and ACVR2A) to that of GAPDH in 22-mo-old control (open bars) and prenatal testosterone-treated (filled bars) sheep. Asterisks indicate significant treatment differences (*P < 0.05).

DISCUSSION

This study documents that prenatal testosterone treatment, by its androgenic action, increases pituitary responsiveness to GnRH, thus contributing at least in part to the LH hypersecretion in prepubertal and adult prenatal testosterone-treated females and, possibly, loss of cyclicity in adulthood. Furthermore, prenatal testosterone treatment induced developmental changes in mRNA expression of GnRHR and ESR1, a regulator of pituitary sensitivity to E2 negative feedback [38], as well as paracrine factors involved in the differential control of LH and FSH release in a manner consistent with predicted changes in LH/FSH production/release. These results are discussed below relative to the four study objectives.

Study 1: Pituitary Responsiveness to GnRH

We first determined whether increased pituitary responsiveness to GnRH contributes to the selective LH hypersecretion seen in prenatal testosterone-treated peripubertal females [7, 8] and if it is programmed via the androgenic actions of testosterone. Increased GnRH-induced LH but not FSH release in both ovary-intact prepubertal and ovariectomized adult prenatal testosterone-treated females after suppression of endogenous LH pulses by a constant-release E2 implant provides evidence that increased pituitary responsiveness to GnRH contributes to selective LH hypersecretion later in life. Our similar findings in females exposed prenatally to excess DHT suggests that this amplification of responsiveness to GnRH results from the androgenic actions of testosterone.

Although androgen receptors are present in pituitary [39] for testosterone to exert its effect, the extent to which the increased LH responsiveness to GnRH seen in prenatal DHT-treated females is the result of direct pituitary effect as opposed to the added contribution from endogenous GnRH is unclear. Unfortunately, the 30-mm E2 implant did not completely suppress LH pulses in many prenatal DHT-treated females. Nonetheless, the low frequency of LH pulses during the pre-GnRH treatment period in concert with the similar magnitude of LH responses to sequential administration of GnRH suggests that the contribution from endogenous GnRH, if any, is likely to be little. It is interesting to note that increased pituitary responsiveness to GnRH also appears to contribute to LH hypersecretion in prenatal testosterone-treated monkeys [9] and in women with PCOS [12, 13, 40–42]. A limitation of all of those studies, however, is that endogenous GnRH action was not blocked prior to testing with GnRH; thus, increased GnRH secretion prior to GnRH challenge as opposed to the intrinsic responsiveness of the pituitary itself could have contributed to the increased responsiveness to GnRH. The present work with prenatal testosterone-treated sheep overcomes this limitation.

Study 2: Prenatal Testosterone Effects on Adult Gonadotropin Release

The foregoing observations of increased responsiveness to GnRH specific to heightened LH release fit nicely with the second study, which determined that LH hypersecretion persisted into adulthood and was associated with a disruption of estrous cyclicity in prenatal testosterone-treated females. We previously found that LH secretion was heightened in prenatal testosterone-treated prepubertal females, and this effect was more pronounced at 24 wk than at 12 wk [8]. This evidence for a progressive deterioration is in accord with our present finding; the magnitude of LH increase seen in adult prenatal testosterone-treated females (in this study) was several-fold higher than that seen in prepubertal lambs [8]; animals used in this study are an independent sample set from the same testosterone-treated breeding cohort used in our earlier prepubertal studies [8].

Relative to this, the mixed physiologic status of the prenatal testosterone-treated females in this study is especially interesting in that it is consistent with a progressive development of reproductive abnormalities as a consequence of prenatal programming. Analyses of the cycling and acyclic prenatal testosterone-treated groups separately revealed significant LH hypersecretion only in the acyclic females; LH values in the cyclic prenatal testosterone-treated females were intermediate between control and acyclic prenatal testosterone-treated females. Collectively, these findings are consistent with progressive amplification of LH hypersecretion and progressive loss of cyclicity [2, 3]. At the time of testing in the second year of life, the group comprised both cycling (36%) and acyclic (64%) females. Although differences in cycle status of these animals may reflect differences in fetal susceptibility to testosterone exposure or absolute levels of testosterone reaching the fetus, an intriguing possibility is that all of the animals would be acyclic if studied at a later age.

While changes in mean LH and pulse amplitude were evident in prenatal testosterone-treated females who had reached adulthood, contrary to our expectation, no differences were seen in LH pulse frequency. One caveat in interpreting these findings is that the sampling frequency of 10 min may have been inadequate to assess frequency accurately. Five-minute sampling has been used to estimate high-frequency LH pulses characterizing the late follicular phase of the cycle [43]. A second caveat in interpreting the lack of an effect on frequency may relate to when LH pulses were examined relative to onset of LH surge. We have observed a delayed onset of LH surge in prenatal testosterone-treated females [7], and thus the LH pulse analysis in the present study would have been at a relatively earlier time point in follicular phase in prenatal testosterone-treated females. Despite these caveats, the collective results of studies 1 and 2 provide important insight into the reproductive consequences of prenatal testosterone treatment by showing a progressive loss of cycles associated with LH hypersecretion and enhanced pituitary responsiveness to GnRH.

Study 3: Prenatal Testosterone Effects on Gonadotropin Subunit, GnRH Receptor, and E2 Receptors

The third objective of this study was to determine whether developmental changes in expression patterns of gonadotropin subunits and GnRH/E2 receptors induced by fetal testosterone excess set the stage for subsequent LH amplification. During normal ontogeny, serum LH concentrations are higher in female than in male fetuses between Days 55 and 100 of gestation [44, 45]. Although LH concentrations in Day 70 to Day 90 ovine female fetuses are reported to be high, falling progressively as gestation advanced, LH concentrations in male fetuses follow the opposite trend: low before 90 days but gradually increasing from Day 90 to Day 130 of gestation [45]. This sex difference in fetal LH levels was attributed to the early increase in testosterone (negative feedback) in male fetuses [46–48]. Similar characterization has not been undertaken at the mRNA level. Assuming that changes in LHB mRNA reflect changes in circulating LH, the prediction would be that the early increase in testosterone in prenatal testosterone-treated females (administered) would result in lower LHB mRNA levels in Day 65 and Day 90 fetuses. Findings from this study of reduced levels of LHB mRNA in Day 65 and Day 90 prenatal testosterone-treated fetuses compared with control fetuses support this prediction. Increases in LHB mRNA in Day 140 relative to Day 90 prenatal testosterone-treated fetuses are consistent with the LH increase reported in males. However, LHB mRNA or LH concentrations in Day 140 prenatal testosterone fetuses did not differ significantly from control fetuses of similar age (although there was a trend toward numerical increase in LH). A previous study found that prenatal testosterone excess induced an increase in LH pulse frequency at this time point [48]. Since LH is released in a pulsatile manner, the single time point measure of fetal LH and the small sample size in the current study may have masked the LH increase. The high variability in LH levels seen at this time supports this premise. Analogous to this, the increased LH pulse frequency seen in prenatal testosterone-treated females at 12 wk was not reflected in the mean LH levels [8]. Variability in LHB mRNA levels may also be reflective of the time when pituitary was collected relative to an LH pulse, as rapid changes in pituitary mRNA have been reported [49, 50].

Increases in GNRHR mRNA expression seen in prenatal testosterone-treated Day 140 fetuses (not detectable at Days 65 and 90) are also supportive of an increase in LH release. Because E2 infusion during gestation suppresses LH in control female fetuses [51], and increased LH secretion in prenatal testosterone-treated sheep involves decreased pituitary responsiveness to E2 [8], we predicted E2 receptor mRNA levels would be lower in prenatal testosterone-treated females. Our findings of decline in ESR1 mRNA expression at Days 65 and 90 are consistent with this premise. The absence of changes in ESR2 mRNA expression is consistent with the premise that ESR1 but not ESR2 is involved in negative feedback regulation of LH [38, 52].

Study 4: Prenatal Testosterone Effects on Pituitary Paracrine Regulators of Gonadotropins

Like LH, there are also sex differences in plasma FSH concentrations in ovine fetuses, possibly stemming from the early testosterone feedback [46–48] in the males. In general, in contrast to LH levels being low, circulating concentrations of FSH are higher in females than males from Gestational Days 71–150 [45, 53]. Gonadectomy before Day 130 increased mean FSH concentrations and LH pulse frequency in males but not in females [54, 55]. Furthermore, maternal passive immunization against GnRH caused a suppression of both LH and FSH secretion in males, but only LH in the females [56]. Findings from this study also show that prenatal testosterone treatment, while having similar suppressive effects on Day 90 LHB and FSHB mRNA, had opposing effects on LHB and FSHB mRNA in Day 65 fetuses, which was reflected as a reduction in LHB mRNA and increase in FSHB mRNA. Reported differences in developmental regulation of LH and FSH in males and females following gonadectomy [54, 55] and the differential effects of prenatal testosterone treatment on LHB and FSHB mRNA observed in this study suggest involvement of other selective pituitary modulators of LH and FSH.

If paracrine changes account for differential release of LH and FSH in prenatal testosterone-treated females, higher expression of negative regulators of FSH (INHA, FST, TGFBR3, IGSF1) or lower expression of positive regulators of FSH (INHBB, ACVR1, ACVR1B, and ACVR2A) would accompany or precede reduced FSHB mRNA or protein expression, and the opposite would be true with increased FSHB mRNA or protein expression. This premise held true for the most part. A decrease in mRNA of INHA and TGFBR3, an inhibin co-receptor that increases affinity of inhibin to bind activin receptors and mediates functional antagonism of activin signaling [57], was evident at Day 65 prenatal testosterone-treated fetuses, a time point at which an increase in FSHB mRNA was evident. Furthermore, because in vitro studies in sheep find inhibin and activin to have opposite effects on LH from that of FSH, with inhibin increasing LH and activin suppressing LH [58, 59], the reduced expression of TGFBR3 in d65 prenatal testosterone-treated fetuses may lead to reduction in activin antagonism, thus contributing toward the differential effects of prenatal testosterone treatment on LHB (decrease) and FSHB (increase) mRNA. Similarly, because activins initiate downstream signaling events by sequentially interacting with ACVR2A and the type I receptor [60, 61], the increase in ACVR2A mRNA at Fetal Day 90 and ACVR1B mRNA at Fetal Day 140 should provide a stimulatory input to FSH and an inhibitory input to LH. However, the effects of increase in activin and activin receptor expression on LH and FSH in the Day 140 female fetuses appear to be counterbalanced by increased expression of FST, a neutralizer of activin action.

Our findings provide evidence that prenatal testosterone excess induces changes in the developmental progression of various regulatory signals, the sum effect of which directs differential expression of LH and FSH. Figure 8 summarizes changes in key integrators of hypothalamic and ovarian inputs (GNRHR and ESR1) and pituitary paracrine mediators and the predicted stimulatory or inhibitory input to LH and FSH at Fetal Days 65, 90, and 140. While the reduction in ESR1 on Fetal Day 65 is likely to reduce E2 negative feedback input to LH, increase in pituitary activin action stemming from a reduction in INHA and TGFBR3 may override the positive effects of reduced ESR1, resulting in reduced expression of LHB mRNA and increased expression of FSHB mRNA. Similarly, an increase in the ACVR2A mRNA expression and consequent increase in activin action A may override the effects of reduced ESR1 mRNA at Fetal Day 90, resulting in reduced LHB mRNA expression. The increase in pituitary GNRHR mRNA expression at Fetal Day 140 may contribute toward the postnatal amplification of LH release [7, 8]. An increase in FST, neutralizer of activin action, stemming from an increase in FST mRNA expression may negate the potential for increased activin action stemming from increased INHBB and ACVR1B mRNA expression. Changes in LHB and FSHB mRNA expression seen in this study parallel reported changes in circulating LH/FSH of male fetuses (Fig. 8D, open triangles) [44, 45, 53] and the prenatal testosterone-treated female fetuses studied at one time point [48]. They are also consistent with the observed changes in expression of the various modulators, suggesting that the developmental integration of these regulatory signals likely sets the stage for the subsequent increase in LH release seen during postnatal life [8].


Figure 08
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIG. 8. Integrative model summarizing changes in mRNA expression of regulators of LH and FSH induced by prenatal testosterone excess. A) Changes in mRNA expression of key mediators GNRHR and ESR1. B) Changes in mRNA expression of paracrine mediators of LH and FSH. C) The effects of prenatal testosterone excess on LHB and FSHB mRNA expression. D) Reported differences in circulating LH/FSH between male and female fetuses (male fetuses differ from females in that they, like the prenatal testosterone-treated females, are exposed to an early increase in testosterone from the developing testes). Known effects of each gene on LH and FSH and the direction of prenatal testosterone-induced changes in mRNA expression are indicated as –, +, or closed triangles (see key in shaded area). The predicted effect of these changes on LHB and FSHB mRNA expression in prenatal testosterone-treated females (T) relative to controls (C) is provided in the bottom of each panel.

Prenatal Testosterone Programming of Genes Involved in Gonadotropin Biosynthesis in Adult Females

Since the regulation of gonadotropin secretion in adult females can be influenced by the cycle stage, the relationships between gonadotropin changes and relevant subunit mRNAs are discussed here separately from developmental changes in fetuses discussed earlier. Clearly, the cycling prenatal testosterone-treated females during the presumptive follicular phase showed increased LH and not FSH secretion. No changes in mRNA expression patterns of GNRHR or ESR1 were evident, consistent with lack of changes in LHB and FSHB mRNA expression. The only increase seen—namely, that of ACVR1 mRNA—is inconsistent with observed changes in LHB and FSHB mRNA expression. As such, the increased LH release seen in the adult animals must involve posttranscriptional events either at the translational (change in mRNA translational efficiency) or posttranslational levels (i.e., GNRHR and ESR1, receptor recycling, or degradation).

In summary, prenatal testosterone treatment in sheep, by its androgenic action, programs increased pituitary responsiveness and produces pronounced increase in LH secretion in adults. Furthermore, prenatal testosterone treatment alters developmental trajectory of pituitary mediators of gonadotropins in a manner consistent with the role they play in differential regulation of LH and FSH, the observed changes in LHB and FSHB mRNA expression, and/or the expected pattern of gonadotropin release.

ACKNOWLEDGMENTS

We are grateful to Mr. Douglas Doop for his role in breeding/lambing, animal care, and facility management; Drs. Teresa Steckler, Leslie Jackson, Puliyur S. Mohankumar, and Hiren Sarma, Mr. James Lee, Mr. James Dell'Orco, Mr. David Han, Ms. Olga Astapova, and Ms. Danielle Djombi for their assistance with the prenatal testosterone treatment, collection of blood samples, tissue procurement, and/or hormonal assays; and Mrs. Rebecca Demo for assistance with RNA analysis.

FOOTNOTES

1Supported by USPHS Grants P01-HD44232 and R01 HD41098 to V.P. Back

Correspondence: 2Vasantha Padmanabhan, Department of Pediatrics and Reproductive Sciences Program, University of Michigan, 300 N. Ingalls Bldg., Rm. 1109 SW, Ann Arbor, MI 48109-0404. FAX: 734 936 8620; e-mail: vasantha{at}umich.edu

3Current address: Department of Psychiatry, University of Cincinnati, Cincinnati, OH 45267. Back

Received: 7 June 2007.

First decision: 6 July 2007.

Accepted: 9 December 2007.

REFERENCES

  1. ; : Clarke IJ, Scaramuzzi RJ, Short RV. Ovulation in prenatal androgenized ewes. J Endocrinol 1977 73385–389[Abstract/Free Full Text]
  2. ; : Birch RA, Padmanabhan V, Foster DL, Robinson JE. Prenatal programming of reproductive neuroendocrine function: fetal androgen exposure produces progressive disruption of reproductive cycles in sheep. Endocrinology 2003 1441426–1434[Abstract/Free Full Text]
  3. ; : Manikkam M, Steckler TL, Welch KB, Inskeep EK, Padmanabhan V. Fetal programming: prenatal testosterone treatment leads to follicular persistence/luteal defects; partial restoration of ovarian function by cyclic progesterone treatment. Endocrinology 2006 1471997–2007[CrossRef][Medline]
  4. ; : West C, Foster DL, Evans NP, Robinson J, Padmanabhan V. Intra-follicular activin availability is altered in prenatal-androgenized lambs. Mol Cell Endocrinol 2001 18551–59[CrossRef][Medline]
  5. ; : Steckler T, Wang J, Bartol FF, Roy SK, Padmanabhan V. Fetal programming: prenatal testosterone treatment causes intrauterine growth retardation, reduces ovarian reserve and increases ovarian follicular recruitment. Endocrinology 2005 1463185–3193[Abstract/Free Full Text]
  6. ; : Steckler TL, Roberts EK, Doop DD, Lee TM, Padmanabhan V. Developmental programming in sheep: administration of testosterone during 60–90 days of pregnancy reduces breeding success and pregnancy outcome. Theriogenology 2007 67459–467[CrossRef][Medline]
  7. ; : Sharma TP, Herkimer C, West C, Ye W, Birch R, Robinson JE, Foster DL, Padmanabhan V. Fetal programming: prenatal androgen disrupts positive feedback actions of estradiol but does not affect timing of puberty in female sheep. Biol Reprod 2002 66924–933[Abstract/Free Full Text]
  8. ; : Sarma HN, Manikkam M, Herkimer C, Dell'Orco J, Welch KB, Foster DL, Padmanabhan V. Fetal programming: excess prenatal testosterone reduces postnatal luteinizing hormone, but not follicle-stimulating hormone responsiveness, to estradiol negative feedback in the female. Endocrinology 2005 1464281–4291[CrossRef][Medline]
  9. ; : Abbott DH, Barnett DK, Bruns CM, Dumesic DA. Androgen excess fetal programming of female reproduction: a developmental aetiology for polycystic ovary syndrome? Hum Reprod Update 2005 11357–374[Abstract/Free Full Text]
  10. ; : Sullivan SD and Moenter SM. Prenatal androgens alter GABAergic drive to gonadotropin-releasing hormone neurons: implications for a common fertility disorder. Proc Nat Acad Sci U S A 2004 1017129–7134[Abstract/Free Full Text]
  11. ; : Foecking EM, Szabo M, Schwartz NB, Levine JE. Neuroendocrine consequences of prenatal androgen exposure in the female rat: absence of luteinizing hormone surges, suppression of progesterone receptor gene expression, and acceleration of the gonadotropin-releasing hormone pulse generator. Biol Reprod 2005 721475–1483[Abstract/Free Full Text]
  12. ; : Katz M and Carr PJ. Abnormal luteinizing hormone response patterns to synthetic gonadotrophin releasing hormone in patients with polycystic ovarian syndrome. J Endocrinol 1976 70163–171[Abstract/Free Full Text]
  13. ; : Rebar R, Judd HL, Yen SS, Rakoff J, Vandenberg G, Naftolin F. Characterization of the inappropriate gonadotropin secretion in polycystic ovary syndrome. J Clin Invest 1976 571320–1329[Medline]
  14. ; : Veiga-Lopez A, Ye W, Phillips DJ, Herkimer C, Knight PG, Padmanabhan V. Developmental programming: deficits in reproductive hormone dynamics and ovulatory outcomes in prenatal testosterone-treated sheep. Biol Reprod 2008 78636–647[Abstract/Free Full Text]
  15. ; : Karsch FJ, Cummins JT, Thomas GB, Clarke IJ. Steroid feedback inhibition of pulsatile secretion of gonadotropin-releasing hormone in the ewe. Biol Reprod 1987 361207–1218[Abstract]
  16. ; : Wood RI, Ebling FJ, I'Anson H, Bucholtz DC, Yellon SM, Foster DL. Prenatal androgens time neuroendocrine sexual maturation. Endocrinology 1991 1282457–2468[Abstract/Free Full Text]
  17. ; : Goodman RL, Bittman EL, Foster DL, Karsch FJ. Alterations in the control of luteinizing hormone pulse frequency underlie the seasonal variation in estradiol negative feedback in the ewe. Biol Reprod 1982 27580–589[CrossRef][Medline]
  18. ; : Karsch FJ, Dahl GE, Evans NP, Manning JM, Mayfield KP, Moenter SM, Foster DL. Seasonal changes in gonadotropin-releasing hormone secretion in the ewe: alteration in response to the negative feedback action of estradiol. Biol Reprod 1993 491377–1383[Abstract]
  19. ; : Niswender GD, Reichert LE Jr, Midgley AR Jr, Nalbandov AV. Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 1969 841166–1173[Abstract/Free Full Text]
  20. ; : Padmanabhan V, McFadden K, Mauger DT, Karsch FJ, Midgley AR Jr. Neuroendocrine control of follicle-stimulating hormone (FSH) secretion. I. Direct evidence for separate episodic and basal components of FSH secretion. Endocrinology 1997 138424–432[Abstract/Free Full Text]
  21. ; : . Bello PA, Mountford PS, Brandon MR, Adams TE. Cloning and DNA sequence analysis of the cDNA for the common alpha-subunit of the ovine pituitary glycoprotein hormones. Nucl Acids Res 1989 1710494[Free Full Text]
  22. ; : Brown P, McNeilly JR, Wallace RM, McNeilly AS, Clark AJ. Characterization of the ovine LH beta-subunit gene: the promoter directs gonadotrope-specific expression in transgenic mice. Mol Cell Endocrinol 1993 93157–165[CrossRef][Medline]
  23. ; : . Mountford PS, Bello PA, Brandon MR, Adams TE. Cloning and DNA sequence analysis of the cDNA for the precursor of ovine follicle stimulating hormone beta-subunit. Nucl Acids Res 1989 176391[Free Full Text]
  24. ; : Illing N, Jacobs GF, Becker II, Flanagan CA, Davidson JS, Eales A, Zhou W, Sealfon SC, Millar RP. Comparative sequence analysis and functional characterization of the cloned sheep gonadotropin-releasing hormone receptor reveal differences in primary structure and ligand specificity among mammalian receptors. Biochem Biophys Res Commun 1993 196745–751[CrossRef][Medline]
  25. ; : Mitchell DC and Ing NH. Estradiol stabilizes estrogen receptor messenger ribonucleic acid in sheep endometrium via discrete sequence elements in its 3'-untranslated region. Mol Endocrinol 2003 17562–574[Abstract/Free Full Text]
  26. ; : Cardenas H, Burke KA, Bigsby RM, Pope WF, Nephew KP. Estrogen receptor beta in the sheep ovary during the estrous cycle and early pregnancy. Biol Reprod 2001 65128–134[Abstract/Free Full Text]
  27. ; : Hayashi K, Carpenter KD, Gray CA, Spencer TE. The activin-follistatin system in the neonatal ovine uterus. Biol Reprod 2003 69843–850[Abstract/Free Full Text]
  28. ; : Fleming JS, Galloway SM, Crawford RJ, Tisdall DJ, Greenwood PJ. Tissue-specific variation in the length of the 5' untranslated region of the beta A-inhibin mRNA in sheep. Mol Reprod Dev 1995 401–8[CrossRef][Medline]
  29. ; : Thompson DA, Cronin CN, Martin F. Genomic cloning and sequence analyses of the bovine alpha-, beta A- and beta B-inhibin/activin genes. Identification of transcription factor AP-2-binding sites in the 5'-flanking regions by DNase I footprinting. European J Biochem 1994 226751–764[Medline]
  30. ; : Tisdall DJ, Hill DF, Petersen GB, Fleming JS. Ovine follistatin: characterization of cDNA and expression in sheep ovary during the luteal phase of the oestrous cycle. J Mol Endocrinol 1992 8259–264[Abstract/Free Full Text]
  31. ; : Chapman SC and Woodruff TK. Modulation of activin signal transduction by inhibin B and inhibin-binding protein (InhBP). Mol Endocrinol 2001 15668–679[Abstract/Free Full Text]
  32. ; : Lopez-Casillas F, Cheifetz S, Doody J, Andres JL, Lane WS, Massague J. Structure and expression of the membrane proteoglycan betaglycan, a component of the TGF-beta receptor system. Cell 1991 67785–995[CrossRef][Medline]
  33. ; : Dupont J, McNeilly J, Vaiman A, Canepa S, Combarnous Y, Taragnat C. Activin signaling pathways in ovine pituitary and LbetaT2 gonadotrope cells. Biol Reprod 2003 681877–1887[Abstract/Free Full Text]
  34. ; : . Lee GH, Thonney ML, Richards HM. Partial clone and sequence of an ovine glyceraldehyde-3-phosphate dehydrogenase cDNA. J Animal Sci 1998 76917[Free Full Text]
  35. ; : Livak KJ and Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 2001 25402–408[CrossRef][Medline]
  36. ; : Veldhuis JD and Johnson ML. Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 1986 250E486–E493[Medline]
  37. Exploratory Data Analysis . . Tukey JW. 1977Reading, MA: Addison-Wesley;
  38. ; : Dorling AA, Todman MG, Korach KS, Herbison AE. Critical role for estrogen receptor alpha in negative feedback regulation of gonadotropin-releasing hormone mRNA expression in the female mouse. Neuroendocrinology 2003 78204–209[CrossRef][Medline]
  39. ; : Yuan X, He Y, Liu J, Luo H, Zhang J, Cui S. Expression of androgen receptor and its co-localization with estrogen receptor-alpha in the developing pituitary gland of sheep fetus. Histochem Cell Biol 2007 127423–432[CrossRef][Medline]
  40. ; : Arroyo A, Laughlin GA, Morales AJ, Yen SSC. Inappropriate gonadotropin secretion in polycystic ovary syndrome: influence of adiposity. J Clin Endocrinol Metab 1997 823728–3733[Abstract/Free Full Text]
  41. ; : Patel K, Coffler MS, Dahan MH, Malcom PJ, Deutsch R, Chang RJ. Relationship of GnRH-stimulated LH release to episodic LH secretion and baseline endocrine-metabolic measures in women with polycystic ovary syndrome. Clin Endocrinol 2003 6067–74[CrossRef]
  42. ; : White D, Leigh A, Wilson C, Donaldson A, Franks S. Gonadotrophin and gonadal steroid response to a single dose of long-acting agonist of gonadotropin-releasing hormone in ovulatory and anovulatory women with polycystic ovary syndrome. Clin Endocrinol 1995 42475–481[Medline]
  43. ; : Breen KM, Billings HJ, Wagenmaker ER, Wessinger EW, Karsch FJ. Endocrine basis for disruptive effects of cortisol on preovulatory events. Endocrinology 2005 1462107–2115[Abstract/Free Full Text]
  44. ; : Foster DL, Roach JF, Karsch FJ, Norton HW, Cook B, Nalbandov AV. Regulation of luteinizing hormone in the fetal and neonatal lamb. I. LH concentrations in blood and pituitary. Endocrinology 1972 90102–111[Abstract/Free Full Text]
  45. ; : Sklar CA, Mueller PL, Gluckman PD, Kaplan SL, Rudolph AM, Grumbach MM. Hormone ontogeny in the ovine fetus. VII. Circulating luteinizing hormone and follicle-stimulating hormone in mid-and late gestation. Endocrinology 1981 108874–880[Abstract/Free Full Text]
  46. ; : Attal JA. Levels of testosterone, androstenedione, estrone and estradiol-17 beta in the testes of fetal sheep. Endocrinology 1969 85280–289[Abstract/Free Full Text]
  47. ; : Pomerantz DK and Nalbandov AV. Androgen level in the sheep fetus during gestation. Proc Soc Exp Biol Med 1975 149413–416[Medline]
  48. ; : Matwijiw I and Faiman C. Control of gonadotropin secretion in the ovine fetus. IV. Male-specific entrainment of the hypothalamic control of luteinizing hormone secretion by testosterone in the female ovine fetus. Endocrinology 1991 1291447–1451[Abstract/Free Full Text]
  49. ; : Shorte SL, Leclerc GM, Vazquez-Martinez R, Leaumont DC, Faught WJ, Frawley LS, Boockfor FR. PRL gene expression in individual living mammotropes displays distinct functional pulses that oscillate in a noncircadian temporal pattern. Endocrinology 2002 1431126–1133[Abstract/Free Full Text]
  50. ; : Norris AJ, Stirland JA, McFerran DW, Seymour ZC, Spiller DG, Loudon AS, White MR, Davis JR. Dynamic patterns of growth hormone gene transcription in individual living pituitary cells. Mol Endocrinol 2003 17193–202[Abstract/Free Full Text]
  51. ; : Gluckman PD, Marti-Henneberg C, Kaplan SL, Grumbach MM. Hormone ontogeny in the ovine fetus: XIV. The effect of 17 beta-estradiol infusion on fetal plasma gonadotropins and prolactin and the maturation of sex steroid-dependent negative feedback. Endocrinology 1983 1121618–1623[Abstract/Free Full Text]
  52. ; : Wersinger SR, Haisenleder DJ, Lubahn DB, Rissman EF. Steroid feedback on gonadotropin release and pituitary gonadotropin subunit mRNA in mice lacking a functional estrogen receptor alpha. Endocrine 1999 11137–143[CrossRef][Medline]
  53. ; : Phillips DJ, Smith PR, Heath DA, Condell LA, McNatty KP. Bioactive and immunoreactive FSH and immunoreactive inhibin concentrations in the ovine fetus. J Endocrinol 1992 134287–295[Abstract/Free Full Text]
  54. ; : Matwijiw I and Faiman C. Control of gonadotropin secretion in the ovine fetus. II. A sex difference in pulsatile luteinizing hormone secretion after castration. Endocrinology 1989 1241352–1358[Abstract/Free Full Text]
  55. ; : Mesiano S, Hart CS, Heyer BW, Kaplan SL, Grumbach MM. Hormone ontogeny in the ovine fetus. XXVI. A sex difference in the effect of castration on the hypothalamic-pituitary gonadotropin unit in the ovine fetus. Endocrinology 1991 1293073–3079[Abstract/Free Full Text]
  56. ; : Miller DW, Fraser HM, Brooks AN. Suppression of fetal gonadotrophin concentrations by maternal passive immunization to GnRH in sheep. J Reprod Fertil 1998 11369–73[Abstract/Free Full Text]
  57. ; : Lewis KA, Gray PC, Blount AL, MacConell LA, Wiater E, Bilezikjian LM, Vale W. Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature 2000 404411–414[CrossRef][Medline]
  58. ; : Muttukrishna S and Knight PG. Inverse effects of activin and inhibin on the synthesis and secretion of FSH and LH by ovine pituitary cells in vitro. J Mol Endocrinol 1991 6171–178[Abstract/Free Full Text]
  59. ; : Ghosh BR, Wu JC, Strahl BD, Childs GV, Miller WL. Inhibin and estradiol alter gonadotropes differentially in ovine pituitary cultures: changing gonadotrope numbers and calcium responses to gonadotropin-releasing hormone. Endocrinology 1996 1375144–5154[Abstract]
  60. Inhibin, Activin and Follistatin–Recent Advances and Future Views . : Lebrun JJ, Chen Y, Vale WW. Receptor serine kinases and signaling by activins and inhibins. 1997Tokushima, Japan: Springer-Verlag;1–20. In:
  61. ; : Massague J. How cells read TGF-beta signals? Nature Rev Mol Cell Biol 2000 1169–178[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Exp PhysiolHome page
A. L. Fowden and A. J. Forhead
Hormones as epigenetic signals in developmental programming
Exp Physiol, June 1, 2009; 94(6): 607 - 625.
[Abstract] [Full Text] [PDF]

Home page
 EndocrinologyHome page
L. M. Jackson, K. M. Timmer, and D. L. Foster
Organizational Actions of Postnatal Estradiol in Female Sheep Treated Prenatally with Testosterone: Programming of Prepubertal Neuroendocrine Function and the Onset of Puberty
Endocrinology, May 1, 2009; 150(5): 2317 - 2324.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
P. Smith, T. L. Steckler, A. Veiga-Lopez, and V. Padmanabhan
Developmental Programming: Differential Effects of Prenatal Testosterone and Dihydrotestosterone on Follicular Recruitment, Depletion of Follicular Reserve, and Ovarian Morphology in Sheep
Biol Reprod, April 1, 2009; 80(4): 726 - 736.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
A. Veiga-Lopez, O. I. Astapova, E. F. Aizenberg, J. S. Lee, and V. Padmanabhan
Developmental Programming: Contribution of Prenatal Androgen and Estrogen to Estradiol Feedback Systems and Periovulatory Hormonal Dynamics in Sheep
Biol Reprod, April 1, 2009; 80(4): 718 - 725.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
T. L. Steckler, J. S. Lee, W. Ye, E. K. Inskeep, and V. Padmanabhan
Developmental Programming: Exogenous Gonadotropin Treatment Rescues Ovulatory Function But Does Not Completely Normalize Ovarian Function in Sheep Treated Prenatally with Testosterone
Biol Reprod, October 1, 2008; 79(4): 686 - 695.
[Abstract] [Full Text] [PDF]

Home page
 Biol. Reprod.Home page
A. Veiga-Lopez, W. Ye, D.J. Phillips, C. Herkimer, P.G. Knight, and V. Padmanabhan
Developmental Programming: Deficits in Reproductive Hormone Dynamics and Ovulatory Outcomes in Prenatal, Testosterone-Treated Sheep
Biol Reprod, April 1, 2008; 78(4): 636 - 647.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
78/4/648    most recent
biolreprod.107.063347v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.
Agricola
Right arrow Articles by Manikkam, M.
Right arrow Articles by Padmanabhan, V.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS