Prenatal Testosterone and Dihydrotestosterone Exposure Disrupts Ovine Testes Development

INTRODUCTION

Estradiol and testosterone (T), two important gonadal steroid hormones mediate growth, maintenance and function of many reproductive tissues and regulate ovarian follicular development and spermatogenesis. The importance of steroid hormones for normal reproductive function is convincingly shown in knockout rodent models (Eddy et al. 1996, Couse & Korach 1999, Robertson et al. 1999, Curtis & Korach 2000, Hewitt & Korach 2003, Matsumoto et al. 2005). For instance, estrogen receptor knockout females and males are both infertile with females demonstrating hypoplastic uteri and hyperemic ovaries and males showing atrophy of the testes and seminiferous tubule dysmorphogenesis (Eddy et al. 1996).

Inappropriate steroid receptor signaling relative to timing, duration, and level of exposure to androgens culminate in dysfunctions at various levels. For instance, T emanating from male littermates sharing the same uterine horn leads to masculinization of female rats (Meisel & Ward 1981). Several studies have also documented that inappropriate exposure to T during development culminates in phenotypic, reproductive and behavioral anomalies in female offspring of many species (rat:(Wolf et al. 2002); mouse:(Ryan & Vandenbergh 2002); guinea pig:(Resko & Roselli 1997); sheep:(Padmanabhan et al. 2006, Steckler et al. 2007, Recabarren et al. 2008); and monkey:(Abbott et al. 1998, Dumesic et al. 2005). Congenital adrenal hyperplasia in humans, an autosomal recessive disease resulting from 21-hydroxylase deficiency, leads to excess androgen exposure during fetal development with females born with virilized external genitalia (Forest et al. 2004), similar to the effects caused by experimental treatments in animal models.

While extensive characterization of females has been unde rtaken following prenatal exposure to excess T, such in-depth studies in male offspring are limited. Testosterone and dihydrotestosterone (DHT), one of the metabolites of T, have specific roles during male sexual differentiation. Testosterone is directly involved in development and differentiation of Wolffian duct-derived structures (epididymides, vas deferens, seminal vesicles, and ejaculatory ducts), and DHT is the active ligand in a number of androgen target tissues, including the urogenital sinus and tubercle (Wilson et al. 1993, Randall 1994). Recent studies in sheep have found that exposure to excess T from days 30–90 of gestation leads to increased ano-genital ratio (Manikkam et al. 2004) and reduced sperm count and motility in the male offspring (Recabarren et al. 2008). The present studies expand upon these initial observations to determine if exposure to excess androgens during ram sexual differentiation disrupted testicular formation and sperm characteristics. Since T can be aromatized to estrogen, these studies compared prenatal exposure to excess T with DHT (non-aromatizable androgen), to distinguish androgenic contribution from estrogenic contribution.

MATERIALS AND METHODS

Luminal measurements

Five micrometer thick sections from paraffin embedded testes were prepared and stained with Hematoxylin and Eosin (H&E). Embedded tissues were cut at 3 different depths on 3 separate occasions to ensure differences were not due to a fixation artifact. Approximately 20 opened stages I to VI seminiferous tubule lumens from each ram were measured at 400× magnification using SPOT™ Imaging Solutions (Diagnostic Instruments Inc., Sterling Heights, MI). Luminal measurements were averaged between two separate measurements of each lumen. Measurements were averaged between a 180° horizontal measurement of the lumen and a 180° vertical measurement of the lumen. Measurements were made at a magnification of 400×. In addition, the percentage of opened lumens, partially closed lumens and fully closed lumens were calculated for each ram. Open lumens had a clear visible lumen with a well defined perimeter and no debris. Partially closed lumens had a distinct perimeter and contained debris filling up to50% of the lumen. Fully closed lumens had distinct perimeters and greater than 50% of the luminal space was filled with debris (Figure 2).

Figure 2

Representative images of (a) open, (b) partially closed, and (c) fully closed seminiferous tubule lumina from 9 month old Suffolk rams prenatally treated with T and DHT at 200× magnification. Image (d) represents an open seminiferous tubule lumen at 400× magnification.

Prenatal androgen treatment

Suffolk ewes were brought to the Sheep Research Facility at the University of Michigan for timed artificial insemination with Suffolk semen and for subsequent androgen treatment. All animal procedures were approved by University of Michigan Committee for Use and Care of Animals and were consistent with National Institutes of Health Guide for Use and Care of Animals. Pregnant ewes were administered twice weekly i.m. injections of 100 mg T-propionate (Sigma-Aldrich Corp., St. Louis, MO) or 100 mg DHT propionate in 2.4 ml cottonseed oil (Sigma-Aldrich Corp.) from 30–90 d gestation (term = 147 d). Days 30–90 were selected for treatment as this is the period of sexual differentiation in sheep. Control (C) ewes received an equal volume of vehicle. Prenatal T treatment elevates T levels in female fetuses from near detection level in controls to levels seen in male fetuses at 65 days of gestation (Veiga-Lopez et al. 2010). Lambs were all born in mid-March. The concentrations of T and DHT chosen resulted in similar ano-genital ratios in the female offspring. At birth, males were distinguished from prenatal T and DHT treated females by palpating the scrotal sac to determine presence of testes.

Seven C, six T and five DHT-treated rams were weaned at 8 weeks of age and vasectomized at 14 weeks to prevent unwanted pregnancies in the flock. The 18 rams were from 18 different mothers, some were singletons, while the remainder had either a male or female sibling. The males kept from male twin pairs were chosen at random. All rams had ad libitum access to water, alfalfa pellets and hay for the first 4 months of life, after which they had access only to hay and water. Males were reared in a mixed-sex, same-age group until 6 months of age; when they were moved to pasture with a group of older, intact ewes (N=50). When the breeding season began for pasture ewes in late September, these young vasectomized males had access to them, and all engaged in copulatory behavior when they had access to an estrous female.

Vasectomy

Males were anesthetized initially with ketamine hydrochloride (4 – 6 mg/kg, iv) and diazepam (0.2– 0.3 mg/kg, iv). Once this anesthetic had taken effect, the animal was intubated and transported to a surgical table where it was secured in a supine position. Anesthesia was maintained with a mixture of halothane gas (approximately 1.5%) in nitrous oxide (1/3) and oxygen (2/3). After clipping the wool on the dorsal aspect of the scrotum and adjacent abdomen, the skin was prepared for surgery by standard surgical scrub and draped with sterile surgical drape. An incision was made on one side of the scrotum and the subcutaneous tissue dissected to expose the blood vessels of the pampiniform plexus, which contains the testicular artery and vein. The vas deferens was palpated as a muscular tube, elevated and dissected free from the surrounding vasculature. After ligating both ends of the section of vas deferens to be removed, to ensure that the tube is tied shut, a 4–5 cm segment of vas deferens was cut out. Subcutaneous tissue was then closed with an interrupted cross stitch and the skin closed separately with an interrupted cross stitch. The suture material used for both closures was #1 chromic gut. The procedure was then repeated on the other side such that the vas deferens was bilaterally interrupted.

Testosterone measurement

Due to limitations in resources, rams were euthanized in mid-December, during the breeding season. Prior to sacrifice, rams were weighed, blood was collected via jugular venipuncture into heparinized tubes, and total T (Modified version of Siemens Coat-a-Count T assay; intra-assay coefficient of variation was 7.4%) and free-T (MP Biomedicals, Orangeburg, NY; intra-assay coefficient of variation was 5.2%) were measured. The minimum detectability of the free-T assay was 0.18 pg/ml and for the total T assay was 2 ng/dL.

Testicular biometry

Testes were collected immediately following euthanization (overdose by I.V. injection of FatalPlus which contains Sodium Pentobarbitol) of the 8.5–9 month old rams. Testicles were weighed in pairs. Penis length was measured from the base to the tip of the glans penis.

Histological evaluation

Approximately 0.5 mm sections of tissue from testis were fixed in 10% neutral buffered formalin (Protocol, Middletown, VA) for 24 h, and then washed in 70% ethanol. Tissue was dehydrated in alcohol and embedded in paraffin wax. Two 5-micrometer thick sections were cut at varying depths to avoid duplication of cell counts. Slides were stained with Hematoxylin and Eosin (H&E). Five round seminiferous tubules from Stages I to VI of spermatogenic wave per cross-section (10 round seminiferous tubules per ram) were selected for analysis. The number of cell layers of basal and adluminal compartment (lumen to lamina propria) were measured at 400× magnification. All histological evaluations were done by a single individual, in a random order without knowledge of treatment.

Luminal measurements

Five micrometer thick sections from paraffin embedded testes were prepared and stained with Hematoxylin and Eosin (H&E). Embedded tissues were cut at 3 different depths on 3 separate occasions to ensure differences were not due to a fixation artifact. Approximately 20 opened stages I to VI seminiferous tubule lumens from each ram were measured at 400× magnification using SPOT™ Imaging Solutions (Diagnostic Instruments Inc., Sterling Heights, MI). Luminal measurements were averaged between two separate measurements of each lumen. Measurements were averaged between a 180° horizontal measurement of the lumen and a 180° vertical measurement of the lumen. Measurements were made at a magnification of 400×. In addition, the percentage of opened lumens, partially closed lumens and fully closed lumens were calculated for each ram. Open lumens had a clear visible lumen with a well defined perimeter and no debris. Partially closed lumens had a distinct perimeter and contained debris filling up to50% of the lumen. Fully closed lumens had distinct perimeters and greater than 50% of the luminal space was filled with debris (Figure 2).

Figure 2

Representative images of (a) open, (b) partially closed, and (c) fully closed seminiferous tubule lumina from 9 month old Suffolk rams prenatally treated with T and DHT at 200× magnification. Image (d) represents an open seminiferous tubule lumen at 400× magnification.

Sperm assessment

Sperm were collected from caudal epididymides by cutting one end of the epididymus (between caudal and corpus) and repeatedly milking the epididymis into an empty dish. The volume collected from each male was not recorded. Sperm assessments were performed in a treatment-blinded fashion where motility, morphology and live-dead sperm stains were counted twice and averaged for each ram. Sperm motility was assessed at the light microscope level at 100× magnification. Motility was recorded as a percentage of forward-progressing sperm cells. Sperm were diluted 20-fold in HEPES buffered medium (Gibco, Grand Island, NY) prior to analysis to allow for clear motility visualization. A nigrosin-eosin “live-dead” stain was used to assess sperm morphology and to evaluate the integrity of the acrosome and sperm membrane.

Statistical Analyses

Differences between treatment groups were tested with ANOVA. If there were statistically significant differences then posthoc Tukey-Kramer multiple comparison method was utilized to look for specific differences between treatments. Differences were considered statistically significant at a level of P<0.05. Results are shown as mean ± standard error of mean (S.E.M.).

RESULTS

At the time of study, body weights of C, prenatal T and DHT-treated males did not differ from each other (Table 1). Free T concentration and testes mass were greater (P<0.05) in T-treated rams than DHT-treated rams, but neither group differed from controls (Table 1). Total T concentrations were higher than free-T and consistent with the free-T concentrations (Table 1), but did not differ significantly between groups, in part because of the greater variation between individuals. However, a non-parametric test (because of group differences in variation) detected a trend towards a difference between the DHT-treated and T-treated rams (Dwass-Steel Chrichlow-Fligner Test = 3.098, p = 0.07). Additionally, there were no differences in penis length between treatments (Table 1).

Epididymal sperm motility was significantly reduced in prenatal DHT-treated rams (P<0.05) but not T-treated males compared to control males (Table 2). There were no differences in percentage of sperm exhibiting normal morphology or in the percentage of live/dead sperm between the 3 treatment groups (Table 2).

Seminiferous tubule diameters of prenatal T-treated rams tended to be larger than the controls but did not achieve statistical significance. Prenatal DHT-treated ram seminiferous tubule diameters were significantly smaller than prenatal T-treated rams (P<0.05) but similar to that of controls (Table 3). The seminiferous tubule lumen diameter of prenatal T-treated males were significantly larger than control and prenatal DHT-treated males (P<0.05). There were no differences in the number of cell layers of the basal and adluminal compartment between C and prenatal T-treated males, however there was a reduction in cell layers in DHT treated males compared to the other 2 groups (P<0.05; Table 3). Significantly more prenatal DHT- and T-treated rams had closed seminiferous tubule lumina than C males (P<0.05; Figure 1).

Figure 1

Percentage of opened, partially closed and fully closed stages I to VI seminiferous tubule lumina from 9 month old Suffolk rams prenatally treated with T and DHT. Different superscripts ^{between treatments indicate statistical significances within state of luminal opening, P<0.05.}

DISCUSSION

Androgens play an important role in organization, development and function of many reproductive tissues and other biological processes. Results from this study demonstrate that increased androgen exposure during male fetal sexual differentiation (gestational days 30–90) altered testicular size, free-T concentrations in male lambs, seminiferous tubule size and function, germ cell number, and sperm motility. The impact of T and DHT differed for some measures. These differences may relate to degree of androgen exposure (DHT is a more potent androgen) or contribution from estradiol derived from aromatization of T. Prenatal T treatment elevates fetal T levels similar to those seen in intact males (Veiga-Lopez et al. 2010). Together these findings suggest that this increase in androgens during a critical period of development has a significant impact on seminiferous tubule formation and germ cell production, which could ultimately have negative effects on subsequent fertility.

The average number of cell layers from seminiferous tubule lumen to lamina propria was reduced in DHT-treated rams (Table 3). This decrease in cell layers likely results from a reduction in spermatocytes observed in this treatment group. Prenatal DHT-treated ram testes were significantly smaller than T-treated which did not differ from control males. In most species, testicular hypertrophy is associated with increased diameter and length of the seminiferous tubules (Voglmayr &amp; Mattner 1968, Boockfor et al. 1983, Kosco et al. 1989), increased numbers of germ and Sertoli cells (Waites et al. 1983, Hochereau-de Reviers et al. 1984, Orth 1984, Waites et al. 1985), and increased sperm production per testis at maturity (Voglmayr &amp; Mattner 1968, Barnes et al. 1980). The finding that DHT-treated ram testes were smaller is suggestive of reduced differentiation and proliferation.

One surprising finding in this study was a significant increase in percentage of seminiferous tubule lumens that were partially or completely occluded in DHT- and T-treated rams. Although it is unclear why these lumens are blocked, it is likely that occlusion would negatively affect fertility. Following spermiogenesis, each sperm cell is released by Sertoli cells into the seminiferous tubule lumen by the process of spermiation. Sperm migrate along the tubule to the rete testis and into the caput epididymis. Spermatozoa released into the seminiferous tubule lumen are suspended in fluid produced primarily by Sertoli cells (Zaneveld 1978). Occlusion of seminiferous tubules blocks transport of sperm, thus decreasing the concentration of sperm in the ejaculate. This could in part explain the greatly reduced concentrations of sperm in ejaculates from adult rams whose mothers were prenatally treated with testosterone (Recabarren et al. 2008). Further studies are required to determine if lumen blockage is due to increased apoptosis in seminiferous tubules or an inhibition of Sertoli cell function preventing fluid secretion and normal sperm transport within the seminiferous tubule lumen.

There were no morphological differences in mature sperm between the treatment groups indicating that prenatal androgen exposure does not affect the process of spermiogenesis. However, prenatal DHT-treated rams had significantly reduced percentage of forward progressing motile sperm compared to T-treated and C rams. The lack of difference between control and T-treated rams differ from the findings of Recabarren and colleagues (Recabarren et al. 2008). In their study there was a significant decrease in the straight line velocity of sperm from rams whose mothers were treated with T during gestation. The differences may relate to the duration of T treatment. In Recabarren’s study pregnant mothers were treated with T propionate from days 30 and 120 days of pregnancy, 30 days longer than in the current study.

To a certain degree, some effects (i.e. sperm motility and seminiferous tubule germ cell layers) were more pronounced in DHT-treated males compared to T-treated males. It is conceivable that such differences may be related to T having a lower affinity than DHT for the androgen receptor and its dissociation from the receptor is five times faster than DHT or that the more pronounced differences observed between the two androgen treatments are a result of the higher binding affinity of DHT to the androgen receptor. These differences may also relate to how DHT and T act locally and distally on androgen target tissues. DHT acts on distally located receptors of the urogenital sinus and tubercle. Testosterone plays a more significant role on locally located Wolffian duct-derived structures. In this study, rams were exposed to excess androgens, allowing both local and distal androgen receptors to be activated. To what extent aromatization of T to estrogen plays a role in the differential effects exerted by prenatal treatment with T and DHT is unclear. The mechanisms by which androgen exposure during male fetal development negatively affects reproductive function of mature rams remains to be elucidated.

Luteinizing hormone and FSH were not measured in this experiment. However, Recabarren and colleagues have demonstrated that rams born to mothers exposed to excess T produced less LH in response to GnRH agonist stimulation at 20 and 30 weeks of age (Recabarren et al. 2007). Additionally, rams exposed to T during development had significantly higher plasma T levels than controls at 20 weeks of age but were at lower levels than controls at 30 weeks (Recabarren et al. 2007). This advancement in the developmental trajectory of gonadal T responses to LH stimulation may account for the differences in free-T in male fetuses exposed to excess DHT and T.

Many differences observed in androgen-treated rams were similar to pathologies of human males with GnRH/gonadotropin-based disorders (Davidson et al. 1983, Booth et al. 1987, Nachtigall et al. 1997, Cabrera et al. 2001, Stikkelbroeck et al. 2001). Whether the disruptive effects of prenatal T treatment in testicular dynamics is the result of disrupted LH pulse characteristics as is the case with the female offspring (Masek et al. 1999, Sarma et al. 2005) remains to be ascertained.

In summary findings from this study clearly demonstrate that prenatal T and DHT treatment during male fetal sexual differentiation have differential effects on post-pubertal testicular size, seminiferous tubule size and function, and progressive sperm motility. The occlusion of the seminiferous tubule lumens and reduction is progressive sperm movement may greatly compromise the fertility of male offspring exposed to high androgens during critical periods of gestation.

ACKNOWLEDGEMENTS