Abstract
Exposure to excess androgens in utero induces irreversible changes in gonadotrophin secretion and results in disrupted reproductive endocrine and ovarian function in adulthood, in a manner reminiscent of the common clinical endocrinopathy of polycystic ovary syndrome (PCOS). We have recently identified an abnormality in early follicle development in PCOS which we suggested might be an androgenic effect. We propose that altered ovarian function in androgenized ewes is due to prenatal androgens not only causing an abnormality of gonadotrophin secretion, but also exerting a direct effect on the early stages of folliculogenesis. Therefore, in this study, we explored the possible differences between small preantral follicles in the ovarian cortex of androgenized female lambs with those of normal lambs. At 8 months of age, small ovarian cortical biopsies (approximately 5 mm3) were obtained at laparotomy from nine female lambs that had been exposed to androgens in utero from embryonic days 30 to 90 of a 147-day pregnancy, and 11 control female lambs. Further, ovarian tissue was obtained at 20 months of age from ten androgenized and nine control animals. Tissue was either fixed immediately for histology or cultured for up to 15 days prior to fixing. The number of follicles in haematoxylin and eosin-stained sections was counted and recorded along with the stage of development. Before culture, the total follicle density (follicles/mm3 tissue) was not statistically significantly different between the two types of ovary at either 8 or 20 months of age. Furthermore, there were no statistically significant differences in the density of follicles at each stage of development. However, there was a lower percentage of primordial follicles, but a higher percentage of primary follicles, in biopsies taken at 8 months from androgenized lambs when compared with controls. At 20 months, the proportions of follicles at the primordial and primary stages were not significantly different between the two groups, but this was mainly attributable to an increase in the proportion of growing follicles in biopsies from control animals. Culture of ovarian cortex from 8-month-old lambs resulted in a progressive increase in the proportion of growing follicles when compared with tissue fixed on the day of surgery. However, there was no difference between androgenized and control tissue in the percentage of growing follicles. The increase in the proportion of growing follicles in the cortex of androgenized animals is reminiscent of similar observations in human polycystic ovaries and suggests that excess exposure to androgen in early life plays a part in the accelerated progression of follicle development from the primordial to the primary stage in polycystic ovaries.
Introduction
Exposure of female mammals to male gonadal steroids during a species-specific window of early development induces irreversible morphological and functional changes in the reproductive axis in adulthood. In the sheep, this ‘critical period’ occurs during prenatal life when androgens from either an endogenous or an exogenous source can promote masculinization of the external genitalia (Short 1974, Wood & Foster 1998), the reproductive neuroendocrine system (Wood & Foster 1998, Sharma et al. 2002) and abnormal development of the ovary (West et al. 2001, Steckler et al. 2005). The focus of this report is on the effects of in utero exposure to testosterone on ovarian function in postnatal life. The observation that 3- and 5-week-old prenatally androgenized female lambs possess enlarged ovaries with an increased number of follicles (West et al. 2001), suggests that androgens may disrupt folliculogenesis in the androgenized sheep. There is additional evidence from other species (rats (Hillier & Ross 1979, Billig et al. 1993), mice (Murray et al. 1998), primates (Vendola et al. 1998, 1999) and humans (Hughesdon 1982) that the exposure to androgens disrupts the morphology and function of the ovary in post-pubertal mammals.
In humans, the common endocrine disorder of polycystic ovary syndrome (PCOS) is characterized by anovulation (and infertility) and ovarian androgen excess (Franks 1995, Rotterdam 2004). Anovulation reflects the arrest of follicle development in the late antral stages but recent work from our group (Webber et al. 2003), and that of Erickson and colleagues (Maciel et al. 2004), has highlighted a disorder of early, preantral follicle development. The mechanism of disordered early folliculogenesis in polycystic ovaries remains uncertain but exposure to excess androgen must be considered as a possible aetiological factor.
In this study, we made a detailed morphometric analysis of follicle populations in ovarian cortical tissue from androgenized and control lambs. We examined tissue that had been fixed immediately after surgery and also used a tissue culture system, well established in this laboratory, to assess the effects of culture on follicle development for up to 15 days (Hovatta et al. 1997, Wright et al. 1999). The aim of this study was to determine, in this putative model of human PCOS, the effects of prenatal androgenization on follicle density, growth and development in the ewe.
Materials and Methods
Animals and in utero steroid treatments
Studies were performed on a total of 25 female sheep of the Poll Dorset breed born at The Babraham Institute (52°12′N) between 25th March and 14th April 2000. Twelve of the female sheep had been exposed to exogenous androgen prenatally; their mothers had been given i.m. injections of testosterone for 60 days between days 30 and 90 of the 147-day pregnancy, as described in previous studies (Robinson et al. 1999, Birch et al. 2003, Unsworth et al. 2005). Specifically, pregnant ewes were treated with testosterone propionate (100 mg twice weekly in 1 ml vegetable oil; Sigma-Aldrich). This includes the ‘critical period’ for gonadal differentiation. Ovarian tissue was obtained at 8 months of age from 11 female lambs who had received no prenatal treatment (controls; i.e. nine of the previously reported animals and two additional controls) and nine androgenized animals. Samples were also obtained at 20 months of age from nine controls and ten androgenized ewes. Animals studied at 20 months included seven controls and seven androgenized animals who had been studied at 8 months, with an additional two controls and three androgenized ewes of the same age who had not previously been sampled. Lambs remained with their mothers until weaned at 10 weeks. The lambs were maintained outdoors at the Babraham Institute during the summer months and under shelter during the winter and were given food and water available ad libitum. At no time were any of the animals in these studies in contact with adult rams. Procedures were carried out under Home Office (UK) Project Licences PPL 80/ 1037 and PPL 80/1506.
Blood sampling
Blood samples were obtained at twice weekly intervals from about 15 weeks of age and continued until the animals were 20 months of age (Unsworth et al. 2005). Plasma was stored at −20 °C until preparation for progesterone immunoassay.
Tissue collection
Ovarian cortical tissue pieces <5 mm3 were collected by biopsy, using a small scalpel blade, during laparotomy under general anaesthesia (sodium pentobarbitone 20 mg/kg i.v.; Sagatal; RMB Animal Health Ltd, Dagenham, Essex, UK). The biopsy was taken from an area of cortex that was devoid of any large visible follicles or corpora lutea and any bleeding which resulted from taking the biopsy was stemmed with a single stitch (Proline). In each case, only one of the ovaries was biopsied. Following biopsy, the ovarian tissue was collected into HEPES-buffered minimum essential medium (MEM; Gibco, Life Technologies) for immediate transport to the laboratory, where it was divided into pieces of about 1 mm in diameter with a scalpel under sterile conditions. One control piece was immediately fixed in Bouin’s Solution (Sigma) for 4 h and stored in 70% alcohol.
Culture of ovarian cortex tissue
In the case of the biopsies obtained at 8 months, tissue that was not immediately fixed was divided into small pieces (approximately 1–2 mm in size) and the pieces were randomly distributed to one of the three Millicell culture plate well inserts (Millipore, Bedford, MA, USA). Up to three tissue pieces were transferred into each insert, which had been pre-coated with 100 μl artificial extracellular matrix (ECM; Matrigel; Becton Dickinson, Bedford, MA, USA). This was produced by diluting ECM in the ratio of 1:3 with serum-free Earle’s balanced salt solution (EBSS; Gibco). The inserts rested in 0.4 ml pre-equilibrated culture medium in the wells of 24-well plates (Nunclon, Rosklide, Denmark), with 0.1 ml within the insert and 0.3 ml in the well surrounding the insert. The medium consisted of Minimum Essential Medium alpha (α-MEM; Gibco) supplemented with 2.5% human serum albumin (HSA, Zenalb 20, Bio Products Laboratory, Elstree, Herts, UK; effective concentration 10%), 1% ITS (insulin/transferrin/selenium; Gibco; effective concentration: 10 μg/ml insulin, 5.5 μg/ml transferrin and 6.7 ng/ml sodium selenite) and antibiotics (50 U/ml penicillin G, 50 mg/ml streptomycin sulphate and 0.125 mg/ml amphotericin B; antibiotic anti-mycotic solution; Gibco). Every second day, the medium was replaced with fresh medium, and the spent medium was kept for analysis of steroid production. The plates were incubated at 37°C under 5% CO2 for 5, 10 or 15 days.
Tissue processing and follicle assessment
The tissue was dehydrated through a series of alcohols, embedded in paraffin wax and the blocks serially sectioned at 5 μm thickness on a Leica RM2135 microtome. The sections were stained with haematoxylin and eosin, and analysed microscopically using a Nikon Eclipse E600 microscope (×60 objective; Nikon, Kingston-upon-Thames, UK). Photographs of each section were captured using a DXM 1200 digital camera (Nikon) and analysed using a digital image analysis system (LUCIA, Nikon). Every serial section was examined and each follicle was identified and counted. We examined each follicle in every section in which it appeared to avoid double counting (Webber et al. 2003). The stage of development of each follicle was assessed. The stage of follicle development was defined (Webber et al. 2003) as follows: a primordial follicle comprised an oocyte surrounded by a single layer of granulosa cells, of which more than 50% had a flattened, squamous appearance and a primary follicle comprised a single layer of granulosa cells, of which more than 50% were cuboidal; secondary follicles were those with more than one layer of granulosa cells, and follicles with more than five layers of granulosa cells were designated multilayered preantral. The proportion of primordial follicles and the proportion of follicles that had initiated growth (i.e. was at the primary stage or beyond) were calculated as percentages of the total number of follicles counted for each biopsy. The volume of the piece of tissue analysed was calculated by measuring the cross-sectional area of every fifth section using the LUCIA software. Each area was then multiplied by 5 to determine the volume for five sections, and then multiplied by 5 μm (the thickness of each section) to give an approximate tissue volume. This allowed calculation of follicle density to be expressed in terms of number of follicles per cubic millimetre of tissue.
Hormone concentrations in culture medium
Progesterone and androstenedione concentrations were measured in samples of medium collected from ovarian cortex cultures using commercially available kits (Coat-a-Count; Diagnostic Products Corp., Los Angeles, CA, USA) by methods previously described (Rice et al. 2005). Samples of medium were pooled from days 0–5, 6–10 and 11–15 for each animal and steroid production during each time period analysed. Total steroid production over the entire culture period was also determined. All samples for each steroid were measured in the same assay. The intra-assay coefficient of variations were 6.4% (Rice et al. 2005) and 8.7% for progesterone and androstenedione respectively.
Statistical analysis
Follicle densities in control and androgenized animals were compared using the Mann–Whitney test. The relative proportions of primordial and growing follicles were compared with control and androgenized animals using binomial/logistic regression (blogit command; Stata8 for the Macintosh; Stata Corporation, College Station, TX, USA; Webber et al. 2003). This was also used to calculate the coefficients for the linear trends in such proportions with days in culture, and to compare the proportions between control and androgenized ovaries. Confidence intervals and P values for comparisons between proportions in control and androgenized groups were computed using robust standard errors with clustering by individual tissue samples. To analyse steroid production, group means were determined and analysed statistically using the Mann–Whitney U-test.
Results
Effect of prenatal androgen exposure on external appearance of female lambs
The administration of testosterone propionate had a virilizing effect on the external genitalia of female lambs as described previously (Barnes et al. 1995, Unsworth et al. 2005). Testosterone-treated lambs had genitalia that were very similar to males, with a penis and scrotal tissue, but with no testicular tissue. However, all had ovaries in the correct anatomical position. At the time of laparotomy, it was immediately clear that prenatal androgens also had a marked effect on the internal reproductive organs. Specifically, the uterus was dilated with secretions, probably the result of a blind-ended uterus. Furthermore, there were laterally placed tubular structures within the broad ligament not usually seen in female lambs, which were probably the remnants of the Wolffian ducts.
Reproductive cycles
Details of reproductive cycles during the first breeding season of androgenized and control sheep have been reported elsewhere (Birch et al. 2003, Unsworth et al. 2005). Whereas all control ewes showed cyclical progesterone secretion, none of the androgenized sheep had regular cycles; eight out of nine had erratic cyclical activity with occasional peaks of progesterone production but only one had no evidence of progesterone production. In all but three animals (two controls and one androgenized), the first biopsy (i.e. at 8 months) was taken before the onset of reproductive cycles. In two of the controls, there had been evidence ofcyclical activity(as judged by peaks of progesterone production exceeding 2 ng/ml) at 18 and 30 weeks respectively. Biopsies in these cases were obtained, in one animal, just after the second progesterone peak and, in the other, during the first progesterone peak. However, in one of these two animals, the biopsy contained no follicles and the data were therefore not included in the calculation of follicle proportions. In one of the androgenized animals, the first evidence of cyclicity was at 19 weeks of age and the 8-month biopsy was taken during the third progesterone peak.
Ovarian morphology
The gross morphology of the ovaries in androgenized animals was markedly different from that in control animals. Ovaries from androgenized animals were larger and contained either an increased number of large antral follicles, follicular cysts (with a thin or absent membrana granulosa) or an increased stroma (or both; Sharma et al. 2002). In one case, the ovary of an androgenized animal was almost completely occupied by a large follicular cyst. Microscopic analysis of cortical tissue revealed follicles at all stages of development in both normal and androgenized animals.
Follicle density
In tissue biopsies that were obtained at 8 months of age (and fixed immediately: ‘day 0’ samples), the density of follicles in the cortex of ovaries from androgenized ewes was not statistically significantly different from that in tissue from control animals. The total follicle density in biopsies from androgenized females was (median) 132/mm3 (range 15–588) when compared with 105 (0–204) (P=0.84; Mann–Whitney) in biopsies from control females (i.e. in one control piece, no follicles were seen). Focusing on follicles at specific stages of development, we found that the median density of primordial follicles was 39 (0–372) in androgenized ewes and 68 (0–137) in controls (P=0.67) and there was no statistically significant difference in density of primary follicles between the groups: 59 (10–212) when compared with 37 (0–68), P=0.25. There were too few follicles at the secondary stage or beyond to allow statistical comparison of follicle densities between androgenized and control sheep. A similar pattern was observed in ovarian cortical samples analysed in 20-month-old ewes. The overall follicle density was 21.2/mm3 (0–93) in androgenized animals and 30.4 (7–155) in control ewes (P=0.36). There were no differences between the groups when the follicles were further categorized as primordial, primary or later, more advanced stages.
Proportions of primordial and growing follicles
In biopsies obtained at 8 months of age, androgenized females contained a significantly lower proportion of primordial follicles when compared with controls (Fig. 1). In androgenized lambs, 46.9% (29–66) were at the primordial stage when compared with 71.1% (62–78) in control animals (mean (95% confidence intervals); P=0.02, binomial/logistic regression). Conversely, and reciprocally, biopsies in androgenized ewes contained a significantly (P=0.02) higher proportion of growing follicles in tissue from androgenized ewes than tissue from control animals (Fig. 1). In ovarian cortex obtained from 20-month-old ewes, the difference between androgenized and control animals in the proportions of primordial and growing follicles was not statistically significant. The proportion of primordial follicles in androgenized ewes was 36.9% (22–54) and in control animals, it was 43.8% (34–54; P=0.49; Fig. 1). The narrowing of the difference between the groups was largely attributable to a significant reduction between 8 and 20 months in the proportion of primordial follicles in cortex of control animals (P=0.002), whereas there was no significant change in the proportions of primordial or growing follicles over the same time interval in androgenized animals.
Although it is unlikely that circulating gonadotrophin concentrations have a significant impact on the progression from primordial to growing follicles, we considered the possible effect of cyclicity of the animals on follicle proportions. Only three animals biopsied at 8 months had started cycling and in only two of these biopsies were follicles detected. Exclusion of these two animals (one in each group) did not affect the overall proportions of resting follicles and there was no change in the level of statistical significance. Of the animals sampled at 20 months, only one ewe (in the androgenized group) was acyclical and its exclusion did not significantly influence the results.
Follicle proportions during culture
Tissue pieces from cortical biopsies taken at 8 months of age were cultured for 15 days as described previously. The difference between groups in the proportions of follicles at the primordial and primary stage that we observed in the tissue that was fixed immediately after removal (day 0) was not sustained during the culture of cortical tissue for 15 days (Fig. 2). The proportion of growing follicles was not statistically significantly different in tissue taken from androgenized ewes when compared with that in samples from control ewes after 5, 10 or 15 days in culture (P=0.33, 0.51, 0.19 respectively; logistic regression). There was a significant difference between the curves depicting the relationship between follicle proportions and days in culture (P=0.006), but this was primarily due to the difference invalues at day 0. There was no significant difference in the slopes of the curves (Fig. 2). There was no difference in the proportion of atretic follicles between the two groups during the culture period. The proportion (mean%±s.e.m.) of atretic follicles on day 5 of culture was 51.7±9.5 in control tissue and 45.9±4.0 in androgenized cortex; day 10, 57.5±5.3 vs 51.7±10.7; day 15, 64.0±11.7 vs 63.9±14.0.
Steroid production during culture
Conditioned medium from cultured cortical tissue (again from 8 month animals) contained measurable but variable concentrations of both progesterone and androstenedione (Fig. 3). The variability in hormone production was not related to the occurrence of ovarian cyclicity of the animals prior to collection of the biopsies. There was no statistically significant difference in steroid production between androgenized and control tissue at any time point during the culture period. The concentration (median (range)) of progesterone in pooled media (5+10+15 days) was 23.0 (6–258) nmol/l in the androgenized group and 15.0 (10.4–45) nmol/l in the control group (P=0.33, Mann–Whitney). Androstenedione concentration was 14.9 (9.3–27) nmol/l in the androgenized samples and 11.3 (7.5–16) nmol/l in control samples (P=0.29).
Discussion
Previously published studies of the effects of prenatal androgenization on the hypothalamic–pituitary–ovarian axis of the ewe have shown disruption of reproductive cycles associated with disordered neuroendocrine function (Sharma et al. 2002, Robinson et al. 2003, Unsworth et al. 2005). In this study, we show that prenatal androgenization with testosterone has a marked effect on ovarian morphology. There were significant differences between androgenized and control ewes in development of early preantral follicles. Androgenized female lambs, biopsied at 8 months, demonstrated a reduced proportion of primordial follicles, and a reciprocal increase in the proportion of follicles that had started to grow. These findings are strikingly similar to our observations in human polycystic ovaries (Webber et al. 2003), a condition in which exposure to excess androgen during development has been proposed as an aetiological factor (Abbott et al. 2002). In ovarian tissue obtained from animals at 20 months of age, the difference between androgenized and control ewes was less obvious. This was largely due to an age-dependent increase in the proportion of early growing follicles in the control group but not in the androgenized animals. A similar decrease in the proportion of primordial follicles as a function of increasing age has been observed in human ovaries (Gougeon 1998). Since both the 8 and 20 month animals were sampled at the same time of year, the findings could not be attributed to seasonality. There was no statistically significant difference between the two groups in the density of follicles in ovarian cortex. A previous study did report an increased number of follicles in androgenized ewes when compared with controls, but in this case, follicle counts were restricted to antral and multilayered preantral follicles (West et al. 2001).
We found no major differences between androgenized and control ovaries when cortical tissue was cultured for up to 15 days. The proportion of growing follicles increased with time in culture in both types of ovary, a finding similar to our observations of cultured cortical biopsies from normal (Wright et al. 1999) and polycystic human ovaries (Webber et al. 2003) as well as reported studies in baboon (Wandji et al. 1997) and cow (Wandji et al. 1996). It was of interest that both progesterone and androstenedione were detected in conditioned medium from cultured biopsies of both types of ovary in these sheep studies (although there were no significant differences between the groups). Large preantral and antral follicles were rarely present in these tissue pieces, suggesting that, at least under these culture conditions, either small preantral follicles or (more plausibly) ovarian stromal cells have the capacity to synthesize and secrete steroids. It is particularly noteworthy that androstenedione was produced by cultured tissue, since this is a theca cell product and a distinct theca cell layer is not found in the sheep ovary until the secondary stage of follicle development. This points to the presence of steroidogenic cells capable of synthesizing androgens before fully differentiating and organizing as a discrete theca layer around the follicle. There was no correlation between the hormone concentrations and the number of growing follicles. The presence of putative theca cells in the stroma surrounding medium-sized (and some small) preantral follicles has been suggested in a study of rat ovary (Hirshfield 1991) although, in rodents, the theca layer tends to form at an earlier stage of follicular development than in the ovine ovary (Fortune & Eppig 1979, Lundy et al. 1999). Nevertheless, the presence of P450c17 has been demonstrated in both follicular and stromal tissues of the fetal sheep ovary, suggesting that the ovine ovary has the capacity to produce androgen in the absence of mature follicles (Quirke et al. 2001).
Whilst the dysfunctional antral follicle development could, at least in part, be attributable to disordered gonadotrophin secretion, it is very unlikely that the abnormality in early follicle development is secondary to endocrine disruption. Initiation of follicle growth is essentially gonadotrophin-independent and local (paracrine and autocrine) factors are thought to play an essential role in this process (Gougeon 1996). These data therefore suggest a direct effect of testosterone on the earliest stages of ovarian follicle development. It is conceivable that the observed effects of testosterone are mediated by aromatization to oestradiol. Previous studies indicate that the effects of dihydro-testosterone on the H–P–O axis are of similar, though not identical to those of testosterone itself: estradiol-mediated positive feedback is disrupted in testosterone-treated (Unsworth et al. 2005) but not dihydrotestosterone (DHT)-treated (Masek et al. 1999) animals. Thus, whilst an androgenic rather than an oestrogenic action seems likely, it is not proven. Disrupted reproductive function seen in adult ewes (exposed to just androgen treatment in utero) may, therefore, result not only from abnormal prenatal programming of neuroendocrine function (and hence abnormal gonadotrophin secretion) but also from a direct effect of androgen on the developing ovary.
The hypothesis that prenatal exposure of the ovary to androgens accelerates the growth of the early stages of follicle development in sheep is supported by recent data in fetal sheep and from studies in other species. Examination of fetal sheep ovaries from animals exposed to high concentrations of androgen revealed a reduced proportion of primordial and an increased proportion of growing follicles. Although these researchers, in contrast to our observations in pubertal and cycling animals, observed a reduction in follicle density in androgen-treated animals, their findings are very much in keeping with those presented here (Steckler et al. 2005). An in vitro study in mice has shown that androgens promote growth of preantral follicles (Murray et al. 1998). In that study, isolated follicles cultured in the presence of androgen-free serum grew more slowly than control follicles, but this effect was reversed by the addition of androstenedione to the medium. Furthermore, follicles developed at a faster rate when cultured in the presence of the non-aromatizable androgen, DHT. Preantral follicles from a number of species possess androgen receptors (Tetsuka et al. 1995, Tetsuka & Hillier 1996, Hillier et al. 1997), although the evidence for their presence in primordial follicles is less clear in the non-human primate (Weil et al. 1998). Nevertheless, short-term androgen treatment in vivo stimulates the early stages of follicle growth in primates, resulting in a significant increase in the number of small growing preantral and antral follicles (Vendola et al. 1998, 1999).
It is possible that androgens may also stimulate follicle development indirectly. There are a number of paracrine and autocrine factors that are potential candidates for playing a role in controlling early folliculogenesis (McNatty et al. 2000), including growth differentiation factor-9 (GDF-9), c-kit, inhibins, activins, anti-Müllerian hormone (AMH; otherwise known as Müllerian-inhibiting substance or MIS; Durlinger et al. 1999) and insulin-like growth factors (IGFs). IGF-I is known to have mitogenic actions on, and to promote differentiation of, granulosa cells of antral and large preantral follicles in a number of mammalian species (Adashi et al. 1985, Bergh et al. 1991, Willis et al. 1998) and data in primates suggest that IGF-I and its receptor mediate androgen-induced follicular growth (Vendola et al. 1999).
The observation that exposure to androgens increases the proportion of follicles that have initiated growth is of considerable interest because it is reminiscent of observations of early follicle growth in PCOS (Webber et al. 2003, Maciel et al. 2004). PCOS is the commonest cause of anovulation in women of reproductive age and is associated with hypersecretion of androgens and luteinizing hormone (LH), and with multiple antral follicles in the ovaries (Franks 1995), characteristics that are also prominent in androgenized ewes (West et al. 2001, Richards et al. 2002). Polycystic ovaries not only have an increased number of antral follicles, but also recent data from our laboratory (Webber et al. 2003) and from that of Dr Gregory Erickson (Maciel et al. 2004) have demonstrated that the earliest stages of follicle development are also abnormal. We found that polycystic ovaries had a reduced proportion of primordial follicles, but an increased proportion of primary follicles, an identical pattern to that seen in androgenized ewes. In fact, the only difference between women with PCO and androgenized sheep was in the density of follicles in the cortical biopsies. In women, biopsies from anovulatory PCO had an increased density of small preantral follicles, although the density in ovulatory PCO (as seen in the present studies in the ewe) was the same as in normal ovarian cortex. The aetiology of polycystic ovaries and PCOS remain uncertain but, on the basis of similar findings in the Rhesus monkeys exposed to androgens in utero, we have proposed that the origins of PCOS occur during ovarian development and that exposure to excess androgen plays a major part in determining both the reproductive and the metabolic phenotypes (Abbott et al. 2002). Further support for the central role of androgens comes from observations of the ovarian morphology in women exposed to an extra-ovarian source of androgen excess during fetal life, i.e. those with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. The majority of these women have ovaries with a polycystic morphology similar to women with PCOS. The results of our studies of the prenatally androgenized sheep are consistent with the hypothesis that androgens play a major part in the aetiology of dysfunctional ovarian folliculogenesis in PCOS and also lend support to the proposal that the prenatally androgenized ewe is a good model for exploring the aetiology of this common reproductive disorder.
The proportion (percentage) of follicles in the primordial and growing (primary stage onwards) phase in cortical tissue taken (and immediately fixed) from control and androgenized ewes taken at 8 and 20 months. The proportion of primordial follicles was significantly lower (and the proportion of growing follicles higher) in tissue from androgenized animals at 8 months (P=0.02) but not at 20 months (P=0.49), mainly due to a significant decrease between 8 and 20 months in the proportion of primordial follicles in tissue from normal ovaries (P=0.002). Data are shown as means and 95% confidence intervals.
Citation: Journal of Endocrinology 192, 2; 10.1677/joe.1.07097
The proportions (percentage) of follicles at the primary stage or beyond in cortical tissue pieces taken from animals at 8 months and cultured for up to 15 days. Before culture, the proportion of growing follicles was higher in tissue from androgenized animals (P=0.023) but there was no significant difference between tissue from androgenized and control animals in the response to culture conditions (see results). Data shown are means with 95% confidence intervals (for clarity, only upper CI is shown for androgenized and lower CI only for controls).
Citation: Journal of Endocrinology 192, 2; 10.1677/joe.1.07097
Concentrations of androstenedione and progesterone in media (median and individual values) obtained after culture of cortical tissue pieces taken from animals at 8 months and cultured for up to 15 days. Both steroids were detected in spent medium over each time period but there were no differences between the two groups. In one sample in the androgen-treated group, the levels of progesterone were several times higher than in the remaining tissue incubations throughout the culture period and the results have been omitted from this figure.
Citation: Journal of Endocrinology 192, 2; 10.1677/joe.1.07097
We are extremely grateful to Andrew Dady, Jo Grindrod, Tony Jones and Martin White for assistance with the collection and processing of blood samples, surgery, the administration of steroids, to James Taylor and William Unsworth for carrying out the serum progesterone assays. Supported by Wellbeing and BBSRC and a Research Fellowship from the Medical Research Council (UK; R A F). L B was in receipt of a Clinical Endocrinology Trust medical undergraduate grant from the Society for Endocrinology (UK). The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
Footnotes
(Present address of JE Robinson is Division of Cell Sciences, Institute of Comparative Medicine, University of Glasgow, Bearsden Road, Glasgow G61 1QH, UK)
References
Abbott DH, Dumesic DA & Franks S 2002 Developmental origin of polycystic ovary syndrome – a hypothesis. Journal of Endocrinology 174 1–5.
Adashi EY, Resnick CE, D’Ercole AJ, Svoboda ME & Van Wyk JJ 1985 Insulin-like growth factors as intraovarian regulators of granulosa cell growth and function. Endocrine Reviews 6 400–420.
Barnes FL, Crombie A, Gardner DK, Kausche A, Lacham-Kaplan O, Suikkari AM, Tiglias J, Wood C & Trounson AO 1995 Blastocyst development and birth after in vitro maturation of human primary oocytes, intracytoplasmic sperm injection and assisted hatching. Human Reproduction 10 3243–3247.
Bergh C, Olsson JH & Hillensjo T 1991 Effect of insulin-like growth factor I on steroidogenesis in cultured human granulosa cells. Acta Endocrinologica 125 177–185.
Billig H, Furuta I & Hsueh AJ 1993 Estrogens inhibit and androgens enhance ovarian granulosa cell apoptosis. Endocrinology 133 2204–2212.
Birch RA, Padmanabhan V, Foster DL, Unsworth WP & Robinson JE 2003 Prenatal programming of reproductive neuroendocrine function: fetal androgen exposure produces progressive disruption of reproductive cycles in sheep. Endocrinology 144 1426–1434.
Durlinger AL, Kramer P, Karels B, de Jong FH, Uilenbroek JT, Grootegoed JA & Themmen AP 1999 Control of primordial follicle recruitment by anti-Mullerian hormone in the mouse ovary. Endocrinology 140 5789–5796.
Fortune JE & Eppig JJ 1979 Effects of gonadotropins on steroid secretion by infantile and juvenile mouse ovaries in vitro. Endocrinology 105 760–768.
Franks S 1995 Polycystic ovary syndrome. New England Journal of Medicine 333 853–861.
Gougeon A 1996 Regulation of ovarian follicular development in primates: facts and hypotheses. Endocrine Reviews 17 121–154.
Gougeon A 1998 Ovarian follicular growth in humans: ovarian ageing and population of growing follicles. Maturitas 30 137–142.
Hillier SG & Ross GT 1979 Effects of exogenous testosterone on ovarian weight, follicular morphology and intraovarian progesterone concentration in estrogen-primed hypophysectomized immature female rats. Biology of Reproduction 20 261–268.
Hillier SG, Tetsuka M & Fraser HM 1997 Location and developmental regulation of androgen receptor in primate ovary. Human Reproduction 12 107–111.
Hirshfield AN 1991 Theca cells may be present at the outset of follicular growth. Biology of Reproduction 44 1157–1162.
Hovatta O, Silye R, Abir R, Krausz T & Winston RM 1997 Extracellular matrix improves survival of both stored and fresh human primordial and primary ovarian follicles in long-term culture. Human Reproduction 12 1032–1036.
Hughesdon PE 1982 Morphology and morphogenesis of the Stein-Leventhal ovary and of so-called ‘hyperthecosis’. Obstetric and Gynecologic Survey 37 59–77.
Lundy T, Smith P, O’Connell A, Hudson NL & McNatty KP 1999 Populations of granulosa cells in small follicles of the sheep ovary. Journal of Reproduction and Fertility 115 251–262.
Maciel GA, Baracat EC, Benda JA, Markham SM, Hensinger K, Chang RJ & Erickson GF 2004 Stockpiling of transitional and classic primary follicles in ovaries of women with polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism 89 5321–5327.
Masek KS, Wood RI & Foster DL 1999 Prenatal dihydrotestosterone differentially masculinizes tonic and surge modes of luteinizing hormone secretion in sheep. Endocrinology 140 3459–3466.
McNatty KP, Fidler AE, Juengel JL, Quirke LD, Smith PR, Heath DA, Lundy T, O’Connell A & Tisdall DJ 2000 Growth and paracrine factors regulating follicular formation and cellular function. Molecular and Cellular Endocrinology 163 11–20.
Murray AA, Gosden RG, Allison V & Spears N 1998 Effect of androgens on the development of mouse follicles growing in vitro. Journal of Reproduction and Fertility 113 27–33.
Quirke LD, Juengel JL, Tisdall DJ, Lun S, Heath DA & McNatty KP 2001 Ontogeny of steroidogenesis in the fetal sheep gonad. Biology of Reproduction 65 216–228.
Rice S, Christoforidis N, Gadd C, Nikolaou D, Seyani L, Donaldson A, Margara R, Hardy K & Franks S 2005 Impaired insulin-dependent glucose metabolism in granulosa-lutein cells from anovulatory women with polycystic ovaries. Human Reproduction 20 373–381.
Richards JS, Sharma SC, Falender AE & Lo YH 2002 Expression of FKHR, FKHRL1, and AFX genes in the rodent ovary: evidence for regulation by IGF-I, estrogen, and the gonadotropins. Molecular Endocrinology 16 580–599.
Robinson JE, Forsdike RA & Taylor JA 1999 In utero exposure of female lambs to testosterone reduces the sensitivity of the gonadotropin-releasing hormone neuronal network to inhibition by progesterone. Endocrinology 140 5797–5805.
Robinson JE, Birch RA, Grindrod JA, Taylor JA & Unsworth WP 2003 Sexually differentiated regulation of GnRH release by gonadal steroid hormones in sheep. Reproduction Supplement 61 299–310.
Rotterdam 2004 Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Human Reproduction 19 41–47.
Sharma TP, Herkimer C, West C, Ye W, Birch R, Robinson JE, Foster DL & Padmanabhan V 2002 Fetal programming: prenatal androgen disrupts positive feedback actions of estradiol but does not affect timing of puberty in female sheep. Biology of Reproduction 66 924–933.
Short R 1974 Sexual differentiation of the brain of sheep. In The Sexual Endocrinology of the Perinatal Period, pp 121–142. Ed. GM Forest. Lyon: Colloque International INSERM.
Steckler T, Wang J, Bartol FF, Roy SK & Padmanabhan V 2005 Fetal programming: prenatal testosterone treatment causes intrauterine growth retardation, reduces ovarian reserve and increases ovarian follicular recruitment. Endocrinology 146 3185–3193.
Tetsuka M & Hillier SG 1996 Androgen receptor gene expression in rat granulosa cells: the role of follicle-stimulating hormone and steroid hormones. Endocrinology 137 4392–4397.
Tetsuka M, Whitelaw PF, Bremner WJ, Millar MR, Smyth CD & Hillier SG 1995 Developmental regulation of androgen receptor in rat ovary. Journal of Endocrinology 145 535–543.
Unsworth WP, Taylor JA & Robinson JE 2005 Prenatal programming of reproductive neuroendocrine function: the effect of prenatal androgens on the development of estrogen positive feedback and ovarian cycles in the ewe. Biology of Reproduction 72 619–627.
Vendola KA, Zhou J, Adesanya OO, Weil SJ & Bondy CA 1998 Androgens stimulate early stages of follicular growth in the primate ovary. Journal of Clinical Investigation 101 2622–2629.
Vendola K, Zhou J, Wang J & Bondy CA 1999 Androgens promote insulin-like growth factor-I and insulin-like growth factor-I receptor gene expression in the primate ovary. Human Reproduction 14 2328–2332.
Wandji SA, Srsen V, Voss AK, Eppig JJ & Fortune JE 1996 Initiation in vitro of growth of bovine primordial follicles. Biology of Reproduction 55 942–948.
Wandji SA, Srsen V, Nathanielsz PW, Eppig JJ & Fortune JE 1997 Initiation of growth of baboon primordial follicles in vitro. Human Reproduction 12 1993–2001.
Webber LJ, Stubbs S, Stark J, Trew GH, Margara R, Hardy K & Franks S 2003 Formation and early development of follicles in the polycystic ovary. Lancet 362 1017–1021.
Weil SJ, Vendola K, Zhou J, Adesanya OO, Wang J, Okafor J & Bondy CA 1998 Androgen receptor gene expression in the primate ovary: cellular localization, regulation, and functional correlations. Journal of Clinical Endocrinology and Metabolism 83 2479–2485.
West C, Foster DL, Evans NP, Robinson J & Padmanabhan V 2001 Intra-follicular activin availability is altered in prenatally-androgenized lambs. Molecular and Cellular Endocrinology 185 51–59.
Willis DS, Mason HD, Watson H & Franks S 1998 Developmentally regulated responses of human granulosa cells to insulin-like growth factors (IGFs): IGF-I and IGF-II action mediated via the type-I IGF receptor. Journal of Clinical Endocrinology and Metabolism 83 1256–1259.
Wood RI & Foster DL 1998 Sexual differentiation of reproductive neuroendocrine function in sheep. Reviews of Reproduction 3 130–140.
Wright CS, Hovatta O, Margara R, Trew G, Winston RM, Franks S & Hardy K 1999 Effects of follicle stimulating hormone and serum substitution on the in vitro growth and development of early human ovarian follicles. Human Reproduction 14 1555–1562.
