Effect of Fetal Pituitary-Testes Suppression on Brain Sexual Differentiation and Reproductive Function in Male Sheep

Abstract

We previously demonstrated that treating fetal lambs on gestational day 62 with the long-acting gonadotrophin-releasing hormone (GnRH) antagonist degarelix (DG) suppresses pituitary-testicular function during midgestation. The objective of this study was to investigate whether impaired gonadotrophic drive during this fetal period has enduring effects on sexual differentiation and reproductive function in adult male sheep. We assessed the effects of prenatal administration of DG, with or without testosterone (T) replacement, on various sexually dimorphic behavioral traits in adult rams, including sexual partner preferences, as well as neuroendocrine responsiveness and testicular function. Our findings revealed that DG treatment had no effect on genital differentiation or somatic growth. There were some indications that DG treatment suppressed juvenile play behavior and adult sexual motivation; however, male-typical sexual differentiation of reproductive behavior, sexual partner preference, and gonadotropin feedback remained unaffected and appeared to be fully masculinized and defeminized. DG-treated rams showed an increased LH response to GnRH stimulation and a decreased T response to human chorionic gonadotropin stimulation, suggesting impaired Leydig cell function and reduced T feedback. Both effects were reversed by cotreatment with T propionate. DG treatment also suppressed the expression of CYP17 messenger RNA, a key enzyme for T biosynthesis. Despite the mild hypogonadism induced by DG treatment, ejaculate volume, sperm motility, and sperm morphology were not affected. In summary, these results suggest that blocking GnRH during midgestation does not have enduring effects on brain sexual differentiation but does negatively affect the testes' capacity to synthesize T.

The critical period for sexual differentiation in sheep occurs between gestational days (GD) 30 to 90 of a 146-day pregnancy (1, 2). Research on ewes exposed to testosterone (T) during pregnancy has shown that the period for genital masculinization (GD 30-60) occurs before the period for masculinization of the central nervous system and neuroendocrine function (GD 60-90) (3). However, administering the antiandrogen drug flutamide during the latter period does not completely prevent brain masculinization in male sheep (4). This is because the treatment leads to a compensatory increase in the secretion of luteinizing hormone (LH) and T, which reduces the effectiveness of flutamide by competing with it at the androgen receptor level. This compensatory response, along with the discovery that kisspeptin (the principal neuropeptide regulator of gonadotrophin-releasing hormone [GnRH] neuron activity) is expressed in the hypothalamus under the control of T during the critical period, suggests that the gonadotropic axis in male fetuses actively maintains an androgen environment sufficient for brain masculinization and defeminization during midgestation (5). Treatment with a GnRH antagonist is an effective way to inhibit gonadotrophic function and reduce plasma T levels in fetuses without eliciting a compensatory response (6). Degarelix (DG) is a quick, long-acting GnRH antagonist that acts to block GnRH receptors in the pituitary gland (7). We found that subcutaneous injection of DG into fetal sheep on GD 62 significantly suppressed basal plasma LH and T concentrations when measured on GD 76 and 85. In addition, the LH and T responses to a bolus injection of GnRH were inhibited when compared with controls at these ages. Finally, treatment with DG also caused highly significant reductions in the expression of genes for pituitary gonadotrophins and key testicular steroidogenic enzymes. In the present study, we used prenatal DG treatment to investigate whether impairment of gonadotrophic drive during fetal development has lasting effects on sexual differentiation of reproductive function and behavior in the adult sheep. To do this, we evaluated the effects of prenatal administration of DG with or without T replacement on juvenile play behavior, adult sexual behaviors, neuroendocrine responsiveness, testicular function, and various sexually dimorphic traits in adult rams.

Materials and Methods

Animals

Thirty-three time-bred adult pregnant Polypay ewes (Ovis aries) and their fetuses were used for this study. The ewes underwent estrous cycle synchronization with intravaginal progesterone pessaries and prostaglandin F2α injections, as previously described (8), to allow for accurate calculation of gestational age. All animal procedures used in the study were conducted in accordance with Public Health Services Policy on Humane Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committees of the Oregon Health & Science University (OHSU) and Oregon State University (OSU). All breeding, husbandry, and behavior testing was conducted at the OSU sheep research facilities using sheep from the resident flock.

Experimental Treatments

Ewes were assigned randomly to 3 groups according to the treatments their fetuses would receive: 1) control mannitol vehicle (C, n = 12); 2) DG (n = 12; Firmagon, Ferring Pharmaceuticals); or 3) DG + T propionate (DG + TP, n = 9; Sigma-Aldrich). Fetal injections were performed as described previously (6). Briefly, on GD 62, the pregnant ewes were sedated with ketamine and diazepam intravenously and then anesthetized with inhaled isoflurane and oxygen. The ewe's abdomen was opened through a left flank approach using a sterile surgical technique. The uterus was located and opened in layers, maintaining the fetal membranes in close approximation to the endometrium. Each lamb fetus was exposed, sexed, and given bilateral subcutaneous injections along the spine of either DG (2 mg/500 μL or 5% mannitol vehicle [500 μL]). The tail was docked to visually confirm treatment after parturition. The lamb was returned into the amnion sac along with 1 million U penicillin G (Pfizer) and 2 mg of ciprofloxacin (Fisher Scientific). When ewes carried multiple fetuses, the procedure was repeated for each fetus. When injections were completed, the uterus was closed with suture, returned into the abdomen, and the individual muscle layers of the flank closed with continuous absorbable suture. The ewe's skin was then stapled and sprayed with Alushield (Neogen) and a transdermal 100 μg/hour fentanyl patch (Duragesic, Janssen Labs) applied. To complete the treatments, the ewes received twice-weekly intramuscular injections of either corn oil (2 mL) or TP (100 mg in 2 mL corn oil) from GD 62 to 90, constituting the critical period for masculinization of the ovine sexually dimorphic nucleus (oSDN) (3). This dose and regimen of TP has been found previously to masculinize the volume of the oSDN in the preoptic area (9), advance the timing of puberty in female sheep (10) and produce concentrations of T in the fetus approximately equal to twice that of control male fetuses (9, 11). Pregnant dams were housed indoors at the OSU Sheep Center with ad libitum water and fed orchard grass hay and minerals with selenium. Their ration was supplemented with alfalfa hay and a 14% protein grain during late gestation and lactation formulated to meet National Research Council requirements.

Postnatal Treatment

Lambs were born during the first 2 weeks of March 2020. There were 16 C lambs born from 10 pregnancies (male, n = 9; female, n = 8), 18 DG lambs from 12 pregnancies (male, n = 12; female, n = 7), and 15 DG + TP lambs from 9 pregnancies (male, n = 11; female, n = 4). Two C pregnancies were lost. Seven lambs were born with birth defects and died or were euthanized after delivery (1 C male, 1 C female, 2 DG male, 1 DG female, and 2 DG + TP females). Two other DG males and one DG + TP female died from accidents in the first 3 months of life. No effects of treatments were noted for length of pregnancy (145 ± 0.3 days), lambing difficulty, or dam attentiveness to her lambs. C males (n = 8) were derived from 5 pregnancies (2 twin, and 3 triplet litters), DG males (n = 8) came from 5 DG pregnancies (3 twin and 2 triplet litters), and DG + TP males (n = 11) were from 8 pregnancies (3 single, 3 twin, and 2 triplets). Within the first 24 hours after birth, the lambs were weighed and several physical measurements were taken, including crown-rump length, anogenital distance, anoumbilical distance, and scrotal circumference. These measurements were repeated at various times during the first year of life. C females were studied for comparison in several tests. DG and DG + TP females were not studied. The experimental lambs were raised with their mothers until they were weaned at age 90 days. They were separated by sex at age 5 months into nonadjoining pastures and barns (2 miles apart) and given ad libitum access to grass, water, and mineral supplements. During the winter, they were fed orchard grass supplemented with 18% protein pellets. Gastrointestinal parasite infestation was routinely monitored and treated with antihelmintics as appropriate.

Sexually Dimorphic Behaviors and Luteinzing Hormone Responsiveness

Play behavior

To determine whether prenatal DG treatment altered play behavior, spontaneous interactions between lambs were recorded by observing lambs at age 5 and 6 months when the lambs were housed in single-sex groups. Two observers blinded to the lambs' treatments recorded behaviors while a third videotaped. Observation occurred in the morning for 30 minutes once a week for 8 weeks. Video recordings were reviewed to verify scoring. The following components of play behavior were counted: anogenital sniffs, mounts, foreleg kicks, and head butts.

Male-typical sexual behavior and sexual partner preference

To evaluate whether prenatal DG treatment demasculinized ram behavior, experimental rams were given sexual behavior tests between August and September of the second breeding season when they were aged 17 to 18 months. Estrus was induced in ovariectomized ewes by treating intravaginally with controlled internal drug release (CIDR) devices containing 0.3 g progesterone (Eazi-Breed CIDR Sheep Inserts; Pfizer). The CIDR devices were removed after 5 days, and 18 hours later the sheep were given an intramuscular injection of 50 μg 17β-estradiol (Sigma-Aldrich Corp) in 2 mL of corn oil. Sexual behavior tests began 24 hours after the estradiol injection when rams were paired individually with 2 estrous ewes in a 10 × 10 m pen. Rams were observed for 20 minutes, during which time the latency and frequency of precopulatory courtship behaviors (genital sniffs, foreleg kicks, flehmen, and vocalizations) and consummatory behaviors (mounts and ejaculations) were recorded by 2 observers. Each ram was tested once on any day and 3 times during a week. This was repeated for 3 weeks for a total of 9 separate tests.

Two weeks after sexual behavior tests were completed, rams were tested for sexual partner preferences. Briefly, rams were isolated for 5 days to prevent behavioral interactions between male pen mates prior to the preference tests. Then each ram was observed for 10 minutes in a testing pen that contained 2 restrained rams and 2 restrained estrus ewes (12). The sex of the preferred stimulus animal and the frequency of courtship and consummatory behaviors were recorded. The test was repeated three times at 5-day intervals.

Female-typical sexual behavior and luteinizing hormone surge (positive feedback)

To assess whether prenatal DG treatment interfered with defeminization in males, we evaluated the ability of experimental rams to show female-typical sexual behaviors and an LH surge. Control females were used as a comparison group. When sheep were aged 21 months, they received 2 progesterone-containing CIDR implants (described earlier) placed subcutaneously over the lateral thorax. The CIDRs were removed after 5 days. Then 18 hours later, all sheep received 50 µg 17β-estradiol intramuscularly in 2 mL corn oil. In females, this protocol reliably elicits an LH surge at 12 hours and sexual receptivity at 24 hours after the estradiol injection (13, 14). To capture the LH surge, LH was measured in serum samples collected at 30 and 0 minutes before and at 12, 16, and 18 hours after the injection of estradiol. The quality of female sexual behavior was evaluated 24 hours after estradiol stimulation when each study sheep was paired with a sexually experienced ram for 10 minutes. Proceptive (head turns, fanning, nudging) and receptive (standing) were recorded by 2 observers.

Hypothalamus-Pituitary-Testicular Axis Responsiveness

Gonadotrophin-releasing hormone challenge

To determine whether prenatal DG treatment altered anterior pituitary function/sensitivity, the experimental rams were administered exogenous GnRH (15). At age 6 to 7 months (first breeding season), the rams were given a 2-mL bolus intravenous injection of 100 µg GnRH (Cystorelin, Abbott Laboratories). Blood samples were taken at −30 minutes, 0 minutes (just prior to drug administration), and every 30 minutes thereafter for 4 hours. Plasma was harvested and analyzed for LH concentrations.

Kisspeptin challenge

To determine whether prenatal DG treatment altered hypothalamic responsiveness, experimental rams were administered the kisspeptin analogue murine kisspeptin-10 (KP-10). At age 15 months (second breeding season), the rams were given a 2-mL bolus intravenous injection of 50 μg KP-10 (AnaSpec) and blood samples were collected at 30-minute intervals before and for 2.5 hours after drug administration. Blood samples were taken at −30 minutes, 0 minutes (just prior to drug administration), and every 30 minutes thereafter for 2.5 hours. Plasma was harvested and analyzed for LH concentrations.

Human chorionic gonadotropin challenge

To determine whether prenatal DG treatment altered testicular Leydig cell responsiveness, experimental rams were administered human chorionic gonadotropin (hCG) (16). One week after the kisspeptin challenge test, the rams were given a 1-mL bolus intravenous injection of 1000 IU hCG (CHORULON [chorionic gonadotropin] Merck Animal Health). Blood samples were collected at 30 minutes prior to the hCG administration, and then at 1 and 2.5 hours after injection. Plasma was harvested and analyzed for T concentrations.

Quantitative Real-Time Polymerase Chain Reaction

Total RNA was extracted from testis of 5 randomly selected adult rams from each treatment group using the Zymo Quick-DNA/RNA Miniprep Plus Kit (Zymo Research) following the manufacturer's instructions. The RNA was then converted to complementary DNA (cDNA) using the SuperScript IV First-Strand Synthesis System (Invitrogen). Quantitative real-time polymerase chain reactions were run in triplicate using PowerSYBR Green Master Mix (Invitrogen) and previously validated sheep-specific primer sets (17). All reactions were run on a Quant Studio 7 Flex Thermal Cycler (Applied Biosystems). The polymerase chain reaction amplification efficiencies were 97% or greater, which allowed for quantification of gene expression to be performed by the ΔΔCt method. Each run per gene contained a pooled adult testis cDNA calibrator, a cDNA reaction without reverse transcriptase (negative control), and a reaction replacing cDNA with nuclease-free water (template negative control). Genes were normalized against the reference gene GAPDH.

Breeding Soundness Examinations

At approximately age 20 months rams were given breeding soundness examinations (BSEs) conducted by a veterinary theriogenologist (C.T.E.). The BSE was conducted according to the guidelines of the Society for Theriogenology as previously described (18). The examination consisted of (1) a physical examination, (2) inspection of the external genitalia and measurement of scrotal circumference, and (3) semen collection and evaluation. Scoring was performed by an examiner blinded to prenatal treatment of the ram. Rams were classified as unsatisfactory, questionable, satisfactory, or exceptional based on the results of their BSE according to Society for Theriogenology guidelines.

Euthanasia and Tissue Collection

Rams were euthanized at age 24 months with a lethal intravenous dose (15 mg/kg) of sodium pentobarbital (Euthasol; Delmarva Laboratories Inc). The head was removed and perfused through the carotid arteries with 500 mL physiological saline containing heparin (20 000 U/L), then with 3.5 L of ice-cold 4% paraformaldehyde (PFA) in Sorensen's buffer, pH 7.4. Following perfusion, the brain was removed from the skull and a preoptic area-hypothalamic block of tissue was dissected, immersion fixed in 4% PFA overnight, and then cryoprotected in 20% sucrose solution, frozen, and stored at −80 °C until sectioning. Frozen coronal sections were cut at 30-µm intervals with a freezing microtome into 4 parallel series and stored in cryopreservative solution at −20 °C. Each brain was assigned a nonidentifying number prior to sectioning so that all processing and subsequent analysis was performed without knowledge of the sex, treatment, or behavioral classification of the donor.

Ovine Sexually Dimorphic Nucleus Identification and Volume Measurements

Histology and RNAscope in situ hybridization

One series of brain sections, consisting of every fourth tissue section, was mounted onto Superfrost Plus glass slides (Fisher Scientific) and stained with thionin (0.1%). Brain sections containing the oSDN were identified by examining the thionin-stained series, and then matching adjacent sections were used for RNAscope. Fluorescent in situ hybridization was performed based on instructions from Advanced Cell Diagnostics and technical recommendations with minor modifications using the RNAscope Multiplex Fluorescent Reagent Kit v2 (Advanced Cell Diagnostics; catalog No. 323100). All incubations were conducted at 40 °C using an ACD HybEZ II Hybridization System with an EZ-Batch Slide System (Advanced Cell Diagnostics; catalog No. 321710). On day 1, brain sections were washed overnight in 0.5× Tris-buffered saline + 0.1% Tween 20 at 4 °C on a rocking shaker to remove cryoprotectant. On day 2, sections were mounted onto Superfrost Plus slides (Fisher Scientific), allowed to air dry for 2 hours, and then heated on a slide warmer at 60 °C for 1 hour. Slides were next submerged in chilled 4% PFA (1 hour at 4 °C), rinsed in 0.1 M phosphate-buffered saline (5 minutes/rinse), followed by incubation in increasing concentrations of ethanol (50%, 70%, 100%, and 100%; 5 minutes each), and dried at 50 °C on a slide warmer for 15 minutes. Slides were then equilibrated at room temperature (RT) and incubated with RNAscope Hydrogen Peroxide solution (Advanced Cell Diagnostics; catalog No. 322335) for 10 minutes. The slides were briefly rinsed with deionized water then dried at 60 °C for 15 minutes, washed again 3 to 5 times in the water and incubated with RNAscope Target Retrieval Reagent (95 °C for 10 minutes; Advanced Cell Diagnostics; catalog No. 322000). The slides were rinsed 4 times in deionized water, 1 time in 100% ethanol, and dried at 50 °C for 10 minutes. Next, a hydrophobic barrier was created around the tissue sections using an ImmEdge Pen (Advanced Cell Diagnostics; catalog No. 310018), and slides were stored overnight at RT. On day 3, sections were treated with RNAscope Protease III (Advanced Cell Diagnostics; catalog No.322337) for 30 minutes at 40 °C. After a brief wash in deionized water, each tissue section was covered with 3 to 5 drops of RNAscope aromatase messenger RNA (mRNA) target (Oa-cyp19-04; catalog No. 589681) and control probes (positive control, Oa-POLR2A; catalog No. 516171; 3-plex negative control; catalog No. 320871) and incubated for 2 hours at 40 °C. Next, the slides were washed twice with 1× Wash Buffer (Advanced Cell Diagnostics; catalog No. 310091) for 2 minutes per wash at RT. Sequential tissue application was then performed by adding 3 or 4 drops of the following reagents, followed by incubation at 40 °C with 2-minute washes using 1× Wash Buffer between each application: RNAscope Multiplex FL v2 Amp 1 (Advanced Cell Diagnostics; catalog No. 323101) for 30 minutes, RNAscope Multiplex FL v2 Amp 2 (Advanced Cell Diagnostics; catalog No. 323102) for 30 minutes, and RNAscope Multiplex FL v2 Amp 3 (Advanced Cell Diagnostics; catalog No. 323103) for 15 minutes. Following the final incubation with Amp 3, slides were rinsed twice with 1× Wash Buffer (2 minutes/wash at RT) followed by application of RNAscope Multiplex FL v2 HRP C1 (15 minutes at 40 °C; Advanced Cell Diagnostics; catalog No. 323104). Slides were then washed twice with 1× Wash Buffer (2 minutes/wash at RT), and incubated with 150 µL per slide of Opal 570 (Akoya Biosciences) diluted in RNAscope TSA buffer (Advanced Cell Diagnostics; catalog No. 322809) at a final concentration of 1:750 for 30 minutes at 40 °C. Following a rinse with 1× Wash Buffer twice (2 minutes/wash at RT), RNAscope Multiplex FL v2 HRP Blocker (Advanced Cell Diagnostics; catalog No. 323107) was applied to tissue (15 minutes at 40 °C). Slides were then rinsed twice with 1× Wash Buffer (2 minutes/wash at RT). Sections were covered with 3 to 5 drops of DAPI (4′,6-diamidino-2-phenylindole) solution, incubated for 1 minute at RT, then removed and replaced with Prolong Gold Antifade Mountant (Fisher Scientific; catalog No. 36930) before the coverslip was applied to the slide. The slides were stored in the dark at 4 °C until imaged.

Image analysis

Sample images were acquired on a Zeiss Axioscan 7 slide-scanner. Fluorescence images were illuminated with a Colibri7 LED light source (385 and 555 nm) and captured on a Hamamatsu Orca Flash 4.0 camera (0.324um/pixel) using a 20× 0.8NA objective. Only the central portion of the brain section (covering the oSDN) was imaged. Brightfield images were also acquired on the Axioscan 7 using a 10× 0.45NA objective on a Zeiss Axiocam 705 color camera (0.346 μm/pixel). Images were imported into Zen 3.6 software (Blue edition; Carl Zeiss Microscopy GmBH), where the outline of the oSDN was drawn by hand and the cross-sectional area was measured in each image in which it appeared. The volume of the oSDN was calculated by multiplying the average cross-sectional area by the number of tissue sections positive for the autoradiographic signal and the distance between sections (12).

Hormone Assays

LH concentrations in plasma were measured by Dr Robert Goodman at West Virginia University using radioimmunoassay reagents provided by the National Hormone and Peptide Program as previously described (19). LH assay sensitivity averaged 0.020 ng/mL (NIH S19) with intra-assay and interassay coefficients of variations equal to 4.1% and 7.4%, respectively. Total T levels in plasma were measured by the Oregon Health & Science University Endocrine Technologies Core using ultra-high-performance liquid chromatography–heated electrospray ionization-tandem triple quadrupole mass spectrometry as previously described (6). T assay sensitivity averaged 10 pg/mL and the intra-assay and interassay variations were 8.1% and 10%, respectively.

Statistics

Data were analyzed using GraphPad Prism version 10.0 (GraphPad Software). Response parameters to the GnRH, KP-10, and hCG challenges were calculated using the area under the curve (AUC) analysis with the baseline calculated as the average of the −30- and 0-minute values. Ordinary one-way analysis of variance (ANOVA) was employed for statistical comparisons of normally distributed data involving 3 or more treatment groups, followed by the Tukey test when statistical significance was observed. This was the case for mean data related to play behavior, ejaculation latency in test 1 and 2, oSDN volume, hormone response parameters, and gene expression. Ejaculation frequencies contained zero values, and thus were analyzed using a nonparametric Kruskal-Wallis ANOVA followed by Dunn's multiple comparison test. For testing behaviors among treatments over time, we used a 2-way repeated-measures ANOVA. All data are presented as mean ± SE of the mean, with P less than .05 considered statistically significant.

Results

Effect of Prenatal Degarelix Treatment on Genital Differentiation and Physical Measurements

The anatomy of the external genitalia conformed to genetic sex and was not affected by DG treatment. As expected, there was a statistically significant (P < .05) sex difference in the ratio of anogenital distance to anoumbilical distance that was not altered by DG treatment. Body weights, crown-rump lengths, and scrotal circumferences all increased progressively during the first year and were unaffected by DG treatment (Supplementary Fig. S1) (20).

Effect of Prenatal Degarelix Treatment on Play Behaviors

Play behavior observed at age 5 to 6 months consisted mainly of male sexual patterns (ie, anogenital sniffs, mounts, foreleg kicks) and head butts. For statistical analysis, all behaviors were combined and expressed as the average frequency of play behaviors per animal over the entire 4 hours of testing (Fig. 1). The average frequency of play was statistically significantly (P < .05) lower in C ewe lambs compared to C ram lambs. The average frequency of play behavior in DG males fell between that observed in C females and in C and DG + TP males, although it did not exhibit a statistically significant difference from any of the other 3 groups. The average level of play in DG + TP rams was significantly (P < .05) greater than in C females, while not differing significantly from the levels observed in either C males or DG males.

Figure 1.

Effect of prenatal degarelix (DG) treatment on play behaviors. Play behaviors were recorded for 30 minutes a week for 8 weeks when the lambs were aged 5 and 6 months. Play behaviors include the sum of anogenital sniffs, foreleg kicks, mounts and head butts. Data (mean frequency ± SEM) were analyzed by 1-way analysis of variance followed by a Tukey test. Bars with dissimilar letters are significantly different (P < .05). See “Materials and Methods” for description of prenatal treatments.

Effect of Prenatal Degarelix Treatment on Male-Typical Sexual Behavior and Sexual Partner Preference

Although there was no main effect of treatment, sexual performance improved across all treatment groups with repeated testing as evidenced by a significant main effect of time (Fig. 2). This was apparent as the mean frequencies of mounts and ejaculations increased and their latencies decreased. Closer inspection of the results showed that on the first test DG rams had significantly fewer ejaculations than C rams and greater latencies (Fig. 3A and 3B). By the second test, the rams became more active and the treatment difference was lost (Fig. 3C and 3D). There were no differences among treatment groups in sexual activity displayed in sexual partner preference tests (Fig. 4). Only 2 rams showed an exclusive preference for females (1 C; 1 DG + TP). No rams showed an exclusive preference for males. However, 3 DG and 3 DG + TP males ejaculated on males, while no control males exhibited this behavior.

Figure 2.

Effect of prenatal degarelix (DG) treatment on male sexual behavior. Data (mean ± SEM) were analyzed by 2-way repeated-measures analysis of variance. There was a significant effect of time, but no effect of treatment. Ejac, ejaculation; PreCop, precopulatory behaviors ie, anogenital sniffs, foreleg kicks, flehmen, and vocalizations. See “Materials and Methods” for description of prenatal treatments.

Figure 3.

Effect of prenatal degarelix (DG) treatment on ejaculatory behavior during the first 2 behavioral tests. A and B, DG rams exhibited significantly fewer ejaculations and a greater latency than control and DG + TP rams in test 1. C and D, No differences were observed between treatment groups in test 2. Data (mean ± SEM) were analyzed by 1-way analysis of variance followed by a Tukey test. *P less than .05. See “Materials and Methods” for description of prenatal treatments.

Figure 4.

Effect of prenatal degarelix (DG) treatment on sexual partner preferences. Each male was given 3 preference tests (see “Materials and Methods” for details) at age 18 months. The frequencies of precopulatory (PreCop) behaviors (ie, anogenital sniffs, foreleg kicks, flehmen, and vocalizations combined), mounts and ejaculations (Ejac) directed at either estrous female stimulus animals (female-directed) or male stimulus animals (male-directed) were recorded. Data (mean ± SEM) were analyzed by Kruskal-Wallis analysis of variance followed by Dunn's multiple comparison test. There were no differences among treatment groups in sexual activity displayed in sexual partner preference tests. See “Materials and Methods” for description of prenatal treatments.

Effect of Prenatal Degarelix Treatment on Female-Typical Sexual Behavior and Luteinizing Hormone Surge

All C ewes displayed receptive behavior (ie, turns, tail fans, and stands) and all received mounts from the test ram. In contrast, while a few rams (≤30%) in each treatment group performed turns, no experimental rams displayed fans and only one experimental ram (DG + TP) stood and was mounted (data not shown). A positive LH surge response to estradiol was elicited in all control ewes. By contrast, none of the treated or C rams exhibited an LH surge (Fig. 5).

Figure 5.

Effect of prenatal degarelix (DG) treatment on luteinizing hormone (LH) surge. LH secretion measured in response to a surge-inducing dose of estradiol (50 µg 17β-estradiol) when sheep were aged 21 months. Data (mean ± SEM) were analyzed by 2-way repeated-measures analysis of variance followed by a Tukey test. *P less than .05; **P less than .01, CTL ♀ vs all other groups. See “Materials and Methods” for description of prenatal treatments.

Effect of Prenatal Degarelix Treatment on Ovine Sexually Dimorphic Nucleus Volume

Fig. 6A shows representative photomicrographs of thionin-stained and fluorescent RNAscope images of oSDN in a C ram and C ewe. One-way ANOVA revealed that the oSDN volume, measured either by thionin stain (Fig. 6B) or by aromatase mRNA expression (Fig. 6C), was statistically significantly (P < .05) smaller for C ewes than for C, DG and DG + TP rams. There was no difference in volume among the C and treated ram groups.

Figure 6.

Effect of prenatal degarelix (DG) on ovine sexually dimorphic nucleus (oSDN) volume. A, Representative photomicrographs of oSDN in a control ram and ewe: a and b, thionin-stained sections; c and d, fluorescent RNAscope images of aromatase messenger RNA (mRNA) expression. Scale bar = 1000 µm. Differences in oSDN volume in brain sections B, stained with thionin or C, processed for aromatase in situ hybridization. Data (mean ± SEM) were analyzed with 1-way analysis of variance followed by Tukey test. **P less than .01. See “Materials and Methods” for description of prenatal treatments. ac, anterior commissure; oc, optic chiasm; 3 V, third ventricle.

Effect of Prenatal Degarelix Treatment on Hypothalamic-Pituitary-Testicular Responsiveness

Bolus intravenous injection of GnRH at approximately age 7 months induced a significantly greater (P < .05) discharge of LH in DG rams compared to C and DG + TP rams (Fig. 7A). Baseline concentrations of LH were unaffected by DG and DG + TP treatment (Fig. 7B). The overall secretory response to GnRH, estimated as the AUC, and the peak response was significantly greater (P < .01) in DG than in C and DG + TP rams (Fig. 7C and 7D). KP-10 injection at age 15 months elicited an unequivocal LH response that was smaller and more rapid than the GnRH response (Fig. 8A). At this age, baseline concentrations of LH were significantly lower in DG rams than in C and DG + TP rams (Fig. 8B). The LH secretory response to KP-10, estimated by the AUC, was not affected by treatment (Fig. 8C), but peak response was significantly higher in DG vs DG + TP rams (Fig. 8D). Finally, testicular responsiveness was evaluated with an intravenous injection of 1000 IU of hCG in 15-month-old rams. hCG-stimulated T secretion was significantly greater (P < .05) in C and DG + TP rams than in DG rams (Fig. 9A). This difference was reflected in the AUC and peak T responses (Fig. 9C and 9D). Baseline T concentrations were also significantly greater in C rams than in DG rams, which were not different from levels in DG + TP rams (Fig. 9B).

Figure 7.

Effect of prenatal degarelix (DG) treatment on anterior pituitary responsiveness. A, Luteinizing hormone (LH) secretion measured in response to an intravenous injection of gonadotropin-releasing hormone (GnRH; 100 µg/animal) when rams were aged 7 months. GnRH was administered at time 0. B, LH baseline; C, area under the curve (AUC); and D, peak response measures to bolus intravenous injection of GnRH. Data (mean ± SEM) were analyzed with A, 2-way repeated-measures analysis of variance or B to D, 1-way analysis of variance followed by Tukey tests. *P less than .05. See “Materials and Methods” for description of prenatal treatments.

Figure 8.

Effect of prenatal degarelix (DG) treatment on kisspeptin responsiveness. A, Luteinizing hormone (LH) secretion measured in response to an intravenous injection of murine kisspeptin-10 (KP-10; 50 µg/animal) when rams were aged 15 months. Murine KP-10 was administered at time 0. B, LH baseline; C, area under the curve (AUC); and D, peak response measures to bolus intravenous injection of KP-10. Data (mean ± SEM) were analyzed with A, 2-way repeated-measures analysis of variance or B to D, 1-way analysis of variance followed by Tukey test. *P less than .05; **P less than .01. See “Materials and Methods” for description of prenatal treatments.

Figure 9.

Effect of prenatal degarelix (DG) on testis response to bolus human chorionic gonadotropin (hCG) injection. Testosterone secretion measured in response to an intravenous injection of hCG (1000 IU/animal) when rams were aged 15 months. hCG was administered at time 0. B, Luteinizing hormone (LH) baseline; C, area under the curve (AUC); and D, peak response measures to bolus intravenous injection of hCG. Data (mean ± SEM) were analyzed by A, 2-way repeated-measures analysis of variance or B to D, 1-way analysis of variance followed by Tukey test. *P less than .05; **P less than .01. See “Materials and Methods” for description of prenatal treatments.

Effect of Prenatal Degarelix Treatment on Testicular Gene Expression

Fig. 10 shows that at the time of euthanasia (age 24 months), testicular expression of Cyp 17A, a key enzyme in the synthesis of T, was significantly lower in DG and DG + TP rams than in C rams. Two other components of the steroidogenic pathway ie, StAR and Cyp 11, were unaltered. No difference was found in the expression of the LH receptor, nor of inhibin or CYP19 (data not shown).

Figure 10.

Effect of prenatal degarelix (DG) treatment on testicular expression of A to C, steroidogenic enzyme messenger RNA and D, LHR in adult male rams. Data (mean ± SEM) were analyzed with 1-way analysis of variance followed by Tukey test. *P less than .05; **P less than .01. Cyp 11, cytochrome P450 family 11 (cholesterol side chain cleavage enzyme); Cyp17, cytochrome P450 family 17A1 (17α-hydroxylase/17,20-lyase); LHR, luteinizing hormone receptor; StAR, steroid acute regulatory protein. See “Materials and Methods” for description of prenatal treatments.

Breeding Soundness Examinations

No statistically significant group differences were observed in scrotal circumference or sperm motility and morphology at age 20 months (Supplementary Fig. S2) (20). Approximately 60% of rams met the criteria of satisfactory classification in all groups.

Discussion

We found that impairment of gonadotrophic drive to the testis during midgestation had little effect on various measures of masculinization and defeminization in male rams. Although there were effects of DG treatment on juvenile play behavior and on ejaculation frequency and latency in adults, no alterations were observed in other measures of male-typical sexual behavior or sexual partner preferences. Furthermore, there were no indications that DG treatment disrupted defeminization because estradiol failed to elicit LH surges or receptive behaviors in treated rams, similar to C rams.

However, prenatal DG treatment did alter pituitary and testicular function. DG-treated rams showed an increased LH response to GnRH stimulation and a decreased T response to hCG stimulation, suggesting enhanced pituitary sensitivity and impaired Leydig cell function. Both effects were reversed by cotreatment with TP. DG treatment also suppressed the expression of CYP17 mRNA, a key enzyme for T biosynthesis. Despite the mild hypogonadism displayed by DG rams, there was no indication that hypothalamic responsiveness to kisspeptin was affected or that breeding potential was altered as assessed by BSEs. Taken together, these results suggest that disrupting the gonadotropic axis during this early period of development in sheep can affect testicular development, yet does not interfere significantly with brain sexual differentiation nor with the development of a male-typical sexual partner preference.

Our results replicated previously reported sex differences in adolescent play behavior of lambs (21, 22). C male lambs performed more behaviors than C female lambs. DG males demonstrated levels of behavior that fell between the levels observed in C females and in C and DG + TP males, whereas DG + TP–treated males showed significantly more play than females. The observation that DG treatment eliminated the sex difference in play behavior, which was reestablished after combined treatment with DG + TP, implies that DG had an effect on the gonadotrophic axis that was reversed by T. These findings align with prior evidence (23) indicating that exposure to T during a critical developmental period for brain sexual differentiation influences play behavior and contributes to the behavioral sex differences in lambs, similar to what others observed in rats (24-26).

We observed a pattern of impaired ejaculation in DG rams during the first sexual behavior test. Although the amounts of precopulatory behaviors and mounts were not affected, the number of ejaculations was significantly lower and the latency longer compared to C and DG + TP rams. By the second test, there were no differences among the treatment groups in any measure of male sexual behavior. Sexual activity continued to change over time in all experimental groups as rams gained experience. Mount and ejaculatory latencies decreased as their frequencies increased. Thus, the ejaculatory defect may reflect a transient deficit of sexual motivation that resolves rapidly after exposure to estrous ewes. The finding that ejaculatory behavior was not affected in rams cotreated with TP suggests that the effect of DG was mediated by reduced prenatal T exposure. Studies in rats have demonstrated that suppression of pituitary gonadal function with a GnRH antagonist during the first week after birth, a time developmentally equivalent to the second trimester in sheep, also leads to reduced ejaculation (27). Similar to our results in sheep, the number of mounts and intromissions, and their latencies, were unaffected by GnRH antagonist treatment in rats. Additionally, neonatal blockage of GnRH transiently reduced fertility in male rats. Although we did not directly measure fertility in the present experiment, we evaluated the potential breeding ability of the rams with a BSE. We found no effect of treatment on the percentages of rams classified as satisfactory and no significant differences in scrotal circumference, sperm motility, or morphology.

The administration of prenatal DG did not have an effect on the proportion of rams exhibiting same-sex sexual partner preferences, nor did it significantly change the volume of the oSDN. In a previous study we reported that male-oriented rams, those that prefer other rams as sexual partners, possess a smaller oSDN volume compared to female-oriented rams, which prefer ewes (12). This difference is thought to be caused by variations in fetal exposure to T since the oSDN volume is enlarged in females treated prenatally with exogenous T and reduced in males treated prenatally with the androgen antagonist flutamide (4, 9). Therefore, if fetal GnRH neurons are essential for fetal T secretion, administering DG during the critical period was expected to result in a smaller oSDN and an increased proportion of male sheep exhibiting same-sex partner preference. It should also have disrupted the defeminization of estrogen-stimulated LH surge and receptive behavior. However, none of these effects were observed.

The unexpected lack of an antagonist effect on sexually dimorphic features in adults was surprising, given that our previous study clearly demonstrated a reduction of plasma T in treated fetuses (6). There could be several reasons why the DG treatment used in the present study did not inhibit masculinization and defeminization. First, although DG is a potent GnRH antagonist, it may not have completely blocked the action of GnRH. We found previously that basal T concentrations in DG-treated male fetuses were reduced by more than 90%, but were significantly increased after a bolus injection of GnRH (6). The low level of T and residual testicular responsiveness may be sufficient to masculinize partner preference and the oSDN. Past studies have shown that low levels of androgen can influence the developmental mechanisms organizing neuroendocrine systems of sheep, particularly when administered over an extended time period (28). Second, it is also possible that the treatment period was not long enough. We previously found that fetal lambs given a single subcutaneous injection of DG on GD 62 showed decreased LH and T at GD 76 and 85 (6). This duration should adequately cover the critical period for the masculinization of sexual behavior and oSDN volume, which occurs between GD 60 and GD 90 (1, 3). However, we do not know how quickly T secretion resumes after DG injection, nor how long the developing sheep brain remains responsive to androgen. We found previously that plasma T concentrations remain elevated in male fetuses through GD 100 and for 24 hours after birth (8, 14), and there is some evidence that sexual differentiation in sheep includes a postnatal period of steroid sensitivity that contributes to masculinization and defeminization (1, 29). Finally, T may not be the only factor involved in behavioral differentiation and oSDN development. While T is known to play a critical role, other factors may also be involved. For example, other hormones or environmental factors could override the effects of DG. Overall, the precise reasons for the lack of effect of DG on sexual differentiation of the sheep brain and behavior are not fully understood. Further experimentation will be necessary to test these possibilities and to more fully determine the relationship between hormones, brain development, and behavior in the sheep.

Prenatal DG treatment had a long-term detrimental effect on testicular function in adult males, which was characterized by significantly reduced basal and hCG-stimulated T secretion in the absence of any difference in testis size. Along with this effect, we observed a significant decrease in Cyp17 mRNA expression, which encodes the enzyme required for the final step of T biosynthesis. Prenatal gonadotropins are important in governing testicular development (30). Inhibition of GnRH activity during fetal development leads to impaired gonadotropin secretion and disrupts normal development and differentiation of the fetal testes, resulting in a reduction in the number of Leydig cells and decreased T synthesis (6, 31). This disruption in testes function has been shown to persist into neonatal life (32) and, based on the current results, appears to be permanent. The downregulation of CYP17 expression could be a consequence of the altered Leydig cell programming that contributes to reduced T production.

We found that cotreatment with TP mitigated the effect of DG on hCG-stimulated T secretion, but paradoxically did not alter the deficit in basal T levels and CYP17 mRNA expression. This observation supports the idea that prenatal androgens are involved in differentiation of Leydig cells (33, 34) and suggests they influence normal development of Leydig cell components involved in hCG sensitivity. However, the reason why cotreatment with TP did not change basal T levels and CYP17 mRNA expression remains unclear. Since gonadotropin secretion would be suppressed by DG regardless of T cotreatment, it is possible that the mechanism responsible for basal T secretion and Cyp17 expression is more sensitive to the effects of prenatal DG treatment on gonadotropin suppression and cannot be compensated for by TP cotreatment. Additionally, the suppression of fetal GnRH activity might have affected other aspects of the complex endocrine or paracrine control of Leydig cell function (35). Future studies will be necessary to assess the cause and extent of Leydig cell dysfunction resulting from in utero inhibition of gonadotropin secretion by DG treatment.

Prenatal DG treatment had complex and age-dependent effects on LH secretion in rams. At age 7 months, the treatment enhanced the LH response to GnRH stimulation, likely due to increased sensitivity of GnRH receptors and/or altered feedback mechanisms. By age 15 months, the basal LH levels were lower in the treated rams, indicating sustained effects on the hypothalamic-pituitary-gonadal axis, but the responsiveness to kisspeptin-induced GnRH release was unaffected, indicating that GnRH neuron sensitivity was unaffected. Cotreatment with TP reversed the enhanced LH response to GnRH in prenatal DG-treated 7-month-old rams, but did not restore resting LH levels in 15-month-old rams. The results suggest the effect on basal LH levels observed in 15-month-old rams are persistent and not amenable to reversal with prenatal T administration. The underlying mechanisms are likely complicated and may involve developmental reprogramming of the hypothalamic-pituitary-gonadal axis. Further research will be needed to gain a deeper understanding of these complex interactions and their implications for reproductive development and function.

To summarize, our findings indicate that inhibition of the gonadotropic axis during midgestation in lamb fetuses can affect testicular development, while having minimal effects on brain sexual differentiation and the formation of a male-typical sexual partner preference. Further studies are warranted to determine the extent to which impaired gonadotropic drive during fetal life may influence reproductive potential in adult life.

Glossary

Abbreviations

ANOVA

analysis of variance

AUC

area under the curve

BSE

breeding soundness examination

cDNA

complementary DNA

CIDR

controlled internal drug release

DG

degarelix

GD

gestational day

GnRH

gonadotrophin-releasing hormone

hCG

human chorionic gonadotropin

KP-10

kisspeptin-10

LH

luteinizing hormone

mRNA

messenger RNA

oSDN

ovine sexually dimorphic nucleus

OSU

Oregon State University

PFA

paraformaldehyde

RT

room temperature

T

testosterone

TP

T propionate

Funding

This work was supported by the National Institutes of Health (grants OD011047 to C.E.R. and P51 OD011092 to the Oregon National Primate Research Center).

Disclosures

The authors have nothing to disclose.

Data Availability

Some or all data sets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.