- Email: [email protected]
Intra-follicular activin availability is altered in prenatally-androgenized lambs
Molecular and Cellular Endocrinology 185 (2001) 51 – 59 www.elsevier.com/locate/mce
Intra-follicular activin availability is altered in prenatally-androgenized lambs Christine West a, Douglas L. Foster a,c, Neil P. Evans d, Jane Robinson e, Vasantha Padmanabhan a,b,* a
The Uni6ersity of Michigan, Reproducti6e Sciences Program, 300 N. Ingalls Building, Room. 1101 Ann Arbor, MI 48109 -0404, USA b Department of Pediatrics, The Uni6ersity of Michigan, Ann Arbor, MI 48109 -0404, USA c Department of Obstetrics and Gynecology, The Uni6ersity of Michigan, Ann Arbor, MI 48109 -0404, USA d Department of Veterinary Preclinical Studies, Uni6ersity of Glasgow, Glasgow, UK e Laboratory of Neuroendocrinology, The Babraham Institute, Cambridge, UK
Abstract Prenatal exposure of sheep to testosterone (T) disrupts ovarian cyclicity and leads to anovulation in adulthood. We propose that the disruption of ovarian function in prenatally-androgenized sheep is mediated via follicular defects stemming from reduced intrafollicular activin availability/action. The intra-follicular activin availability/action that facilitates follicular development is dictated by the relative proportions of activins, inhibins (antagonists of activin action) and follistatins (FS; binding proteins of activin and negator of activin action). Inhibin a, bA, bB, and FS mRNA expression were determined by in situ hybridization in 5 week-old ovaries from control (C) lambs or those exposed to testosterone (T) or DHT from 30 – 90 days of gestation. In utero exposure to T, but not DHT, increased total ovarian weight (0.4 9 0.1, 1.5 90.5 and 0.3 9 0.1 g, C, T and DHT, respectively) and total number of follicles (16.5 92.8, 37.8 97.9, and 18.8 9 3.0). With the exception of two follicles in T animals, all follicles were 52 mm in diameter. All follicles 5 2 mm in all groups expressed FSH receptor mRNA in the granulosa cells and LH receptor only in the thecal cells. The percentage of follicles expressing FS mRNA was increased (P B 0.05) in sheep prenatally-androgenized with either T (80.4 98) or DHT (80.3 9 5.5) as compared to C (50.8 98.2). In contrast, the percentage of follicles expressing activin bB mRNA tended to be lower (P= 0.06) in the T (30.9 9 7.1) and DHT (40.5 9 3.3) groups as compared to C (66.1 915.6). Increased expression of FS along with the reduced expression of activin bB mRNA provides evidence for compromised intra-follicular activin availability in the majority of follicles in the androgenized groups. The increase in ovarian weight and follicular number in the T, but not in the DHT group, suggests that the effects of T are mediated through the action of estrogen. We speculate that the decrease in relative abundance of activin may contribute to the selection defects in prenatally-androgenized sheep. If true, this may be a useful model to understand the etiology of polycystic ovarian syndrome. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Inhibin; Follistatin; Folliculogenesis; PCOS
1. Introduction Over a quarter century ago Gilbert Greenwald stated, ‘‘… one of the most intriguing mysteries in ovarian physiology is what factors determine whether Portions of this work were presented at the 31st Annual Meeting of the Society for the Study of Reproduction and the 81st Annual Meeting of the Endocrine Society. * Corresponding author. Tel.: + 1-734-647-0276; fax: + 1-734-9368620. E-mail address: [email protected] (V. Padmanabhan).
one follicle remains quiescent, another begins to develop but later becomes atretic, while still a third matures and ovulates’’ (Greenwald, 1972). Despite substantive research efforts, the mechanisms that underlie the regulation of folliculogenesis are still unclear. Folliculogenesis is a key reproductive event in the female that begins before birth and continues throughout reproductive life and involves progression of a follicle from the primordial to preovulatory state (Hirshfield, 1991; McNeilly, 1991; Findlay, 1991; Adashi, 1995; Gougeon, 1996; Roche, 1996). Failure of follicles to undergo developmental changes at the correct time and
0303-7207/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 1 ) 0 0 6 3 2 - 3
52
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
in an exact sequence leads to failure of folliculogenesis and subsequent deterioration of follicles through atresia. The mechanism by which some follicles are selected to develop to preovulatory status, in preference to other follicles that may have begun growth at the same time, is largely unknown. It has been proposed that selection and dominance, and subsequent escape from atresia, is associated with the ability of some follicles to continue growth and development despite the decline in FSH concentrations that occurs during the late follicular phase of the cycle (McNeilly, 1991; Marshall et al., 1991; Findlay and Clarke, 1997), possibly due to the action of local factors within the ovary such as members of the transforming growth factor family— activin, inhibin, and follistatin. These factors have all been shown to be able to act in a paracrine and autocrine manner, to regulate follicular proliferation and differentiation and as such could play an important role in folliculogenesis (Hillier, 1991; Rabinovici et al., 1992; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). Indeed, it has been shown experimentally that administration of activin leads to superovulation in rats (Erickson et al., 1995), and importantly this effect is mediated directly at the ovarian level, as it is maintained even in hypophysectomized animals (Doi et al., 1992). To properly assess activin availability it is essential to measure not only activin but follistatin and inhibin as well. Follistatins are activin binding proteins that negate activin’s action (Nakamura et al., 1990; Robertson, 1992; de Winter et al., 1996). Inhibins are functional opposites of activins, and they antagonize activin’s action by either binding with activin receptor type II (Xu et al., 1995a; Gray et al., 2000) or interfering with its signal transduction (Lewis et al., 2000; Chong et al., 2000). Adding to the complexity of these functional overlaps, these regulators also exist in multiple forms (Ying, 1988). The isomers identified in ovarian follicles are inhibin A, inhibin B, activin A, and activin B (Hillier, 1991; Rabinovici et al., 1992; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). These isomers appear to be similar functionally, but may differ in potency. Viewed in this context, the relative proportion of activins, follistatins and inhibins are important determinants of activin action. Impairment in expression of any of these regulators would alter intra-ovarian equilibrium, lead to disruption of folliculogenesis and subsequently anovulation. Animal models that exhibit either disrupted or arrested follicular development, provide valuable resources to investigate the intra-follicular environment that is required to facilitate folliculogenesis. One such model is the prenatally-androgenized female sheep. These are reproductively compromised, exhibit cyclic disruption and become anovulatory during adulthood (Clarke et al., 1977; DeHaan et al., 1987). Furthermore,
they manifest a multifolliculate condition (Clarke et al., 1977) such as that in women with PCOS (Dunaif, 1997; Ehrmann et al., 1995). The aim of this study was to investigate the cause of altered folliculogenesis in prenatally androgenized female sheep by determination of: (1) the effects of prenatal androgenization on follicular development; (2) the patterns of gonadotropin secretion; and (3) the intra-follicular activin environment in these animals relative to normal controls.
2. Methods
2.1. Animals and treatments 2.1.1. Pilot study To gain a preliminary understanding of the effects of prenatal androgenization on ovarian follicular development, a study was conducted on a small number of ovaries recovered from Poll Dorset sheep as part of a neuroendocrine study conducted at the Babraham Institute (Cambridge, UK) (Robinson et al., 1999). Ovaries were collected at 3 weeks of age from seven control lambs and seven lambs that had been prenatally androgenized by treatment of the dams with an aromatizable androgen, testosterone propionate (im; 100 mg twice weekly), for 60 days from fetal day 30 to 90 (term, 147 days). Following removal, all ovaries were weighed and processed for histology. 2.1.2. Main study Experiments were conducted using prenatally-androgenized Suffolk lambs maintained under normal husbandry conditions at the Sheep Research Facility in Ann Arbor, MI. For prenatal-androgenization, pregnant ewes were injected im twice weekly with 100 mg T propionate (T, aromatizable androgen) (n= 20 ewes), or 400 mg DHT propionate twice weekly (non-aromatizable androgen) (n= 12 ewes) from 30–90 days of pregnancy. Controls received no treatment. Fetal sex was unknown at the time of the treatments. Discounting the males born in this study, there were five females born in the DHT group and three females in the T-group. At 5 weeks of age, the ovaries were removed and weighed from C (n=5), T (n= 3) and DHT (n= 5) groups. Both ovaries from three animals in each group were used to determine follicular number and charcaterize by in situ hybridization the expression pattern of markers of follicular development (FSH receptor, LH receptor, aromatase), and factors involved in controlling local activin action (inhibin a, inhibin/activin bA, bB and follistatin). To determine the pattern and concentrations of LH and FSH experienced by the ovaries, blood samples were collected every 20-min for 4 h the day before ovariectomy.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
2.1.3. Assays Concentrations of FSH were measured using NIDDK-oFSH-1 (AFP 5679C) as the standard and NIDDK-anti-oFSH-1 (AFP-C5288113) antiserum at a working dilution of 1:12,000 (Padmanabhan et al., 1992). The sensitivity (2 S.D. of buffer control) and ED50 (50% displacement point) of the FSH assay averaged 0.06 and 0.37 ng respectively (n =1 assay). All plasma samples were measured in 100-ml volume in duplicate. The intra-assay coefficient of variation (C.V.) values at 80 and 20% displacement points averaged 7.3 and 3.6% respectively. Plasma LH was measured by a previously validated RIA (Niswender et al., 1969) using a serum standard calibrated against NIH-LH-S12 (biopotency 9.82×S1). The LH assay sensitivity and ED50 averaged 0.53 and 1.56 ng respectively (n= 2
53
assays). All plasma samples were measured initially in 200 ml in duplicate. Repeat assays were performed as needed in the 20–200 ml range. The intra-assay C.V. values at 80 and 20% displacement points averaged 4.9 and 2.4% respectively. The interassay C.V. values based on two quality control pools measuring 16.3 and 20.2 ng/ml, averaged 7.1 and 9.2%, respectively.
2.2. Riboprobe synthesis Both sense and antisense [35S]UTP-labeled cRNA probes for ovine follistatin (Tisdall et al., 1992), ovine inhibin a (Tisdall et al., 1994), ovine inhibin/activin bA (Tisdall et al., 1994), ovine inhibin/activin bB (Rodgers, 1991), bovine FSH-receptor (Xu et al., 1995b), ovine LH-receptor (Xu et al., 1995a) and bovine aromatase (Xu et al., 1995c), were transcribed from linearized cDNA templates using a transcription kit (Promega) according to the manufacturer’s recommendations and purified by centrifugation on a Sephadex G-50 column and diluted in hybridization buffer [50% formamide, 0.3 M NaCl, 10 mM Tris (pH 8.0), 1 mM EDTA (pH 8.0), 1× Denhardt’s solution [1× = 0.02% (wt./vol.) Ficoll, 0.02% polyvinylpyrrolidone, and 0.02% (wt./vol.) BSA], 10 mM dithiothreitol, 500 mg/ml salmon sperm DNA, and 10% dextran sulfate] to about 1×106 cpm/ml.
2.3. In situ hybridization
Fig. 1. Follicular morphology of an ovary from a control lamb (A) and lambs treated prenatally with testosterone (B, C). Ovaries were collected at 3 weeks of age. Note the disrupted nature of follicular development in prenatally-androgenized sheep. Scale bar bottom right is equal to 2.5 mm.
Sections (10 mm) of ovarian tissue were cut at − 20 °C using a cryostat (Reichert-Jung 2800 Frigocut), mounted onto microscope slides (Superfrost Plus; Fisher Scientific, Pittsburgh, PA), and stored at − 80 °C in boxes until analysis. Mounted slides (3 slides/ovary with two sections in each slide) were dried on a slide-warmer at 37 °C for 30 min and fixed in 4% paraformaldehyde in 0.1 M PBS for 60 min. Following rinsing with double strength standard sodium citrate solution (SSC; 1× = 150 mM NaCl and 15 mM Na3 citrate, pH 7.0) they were treated for 10 min in 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0), rinsed twice in 2× SSC, dehydrated in ethanol, and air dried. Hybridization was performed using 50 ml diluted probe in a humidified oven at 55 °C for 16 h. Sections were hybridized to both the antisense probe and the sense probe for each mRNA of interest. After hybridization, slides were soaked first in 2× SSC to remove the coverslips, twice in 2× SSC for 2 min each time at room temperature and then treated with ribonuclease-A (20 mg/ml in 0.5 M NaCl, 10 mM Tris, and 1 mM EDTA) for 60 min at 37 °C. Slides were washed at room temperature in gradient strengths of SCC for 2 min each at room temperature, dehydrated in increasing concentrations of ethanol and air-dried. Autoradiography was performed overnight to determine the success of the hybridization. Subsequently the
54
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
Fig. 2. Follicular morphology of C lambs and lambs androgenized prenatally with T or DHT at 5 weeks of age. Both ovaries from two animals in each group are shown. Note the mulitfollicular nature of the ovaries collected from lambs androgenized with T. The majority of ovaries in the T group were also larger in size. Table 1 Total ovarian weight and number of follicles per ovary in control and prenatally-androgenized lambs at 5 weeks of age Treatment
n
Total ovarian weight (g)
Number of follicles per ovary
Control Testosterone (200 mg) DHT (800 mg)
5 3 5
0.41 9 0.08 1.52 9 0.46* 0.29 9 0.08*
16.50 9 2.84 37.83 9 7.91* 18.83 9 2.95
Values are mean 9 SEM. * Significance at PB0.05 level.
sections were dipped in NTB-2 emulsion (Eastman Kodak, Rochester, NY), and exposed at 4 °C in a light-tight box (exposure varied with time). The slides were then developed, lightly counterstained with hematoxylin and eosin, and mounted for microscopic examination. Follicles were determined to exhibit specific hybridization to a given probe if the hybridization intensity for the antisense probe was greater than the background hybridization observed to the sense probe. All mRNA expression was quantified with NIH Image on digital images captured from the autoradiography films. The number of follicles that expressed the individual mRNAs of interest were determined and expressed as a percentage of the total number of follicles on that ovary. For each ewe, the percentage of follicles labeled on each ovary was av-
eraged to derive a value for each animal. All individual measurements of follicular diameter were made on the section that had the largest diameter for each follicle.
2.4. Statistical analysis All data were analyzed by ANOVA (General Linear Models procedure of SAS, Cary, NC) to evaluate the effects of treatment. Before ANOVA, each set of data was subjected to Levene’s test for heterogeneity of variance. When heterogeneity was present, data were transformed before ANOVA by a method that yielded homogeneity of variance. Post-ANOVA comparisons among means were made using Bonferroni t test (PB 0.05). All values given are the mean9SEM of non-transformed values.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
55
2.5. Results
2.5.2. Main study
2.5.1. Pilot study At 3 weeks of age, mean weights of the pairs of ovaries from prenatally testosterone-treated lambs (0.46 9 0.19 g, n= 7) were greater (P B 0.05) than C lambs (0.1490.04 g, n =7). Histological examination revealed that all C ovaries appeared normal and contained multiple primordial and early primary follicles (Fig. 1). In contrast, none of the ovaries from the androgenized animals appeared normal. Morphologically the ovaries could be divided into two distinct types. The first type of ovary was one with a single very large follicle that occupied 70– 90% of the ovarian volume but also contained primordial follicles (Fig. 1B). The large follicle had a thin, disorganized granulosa cell layer and red blood cells in its lumen. There was little sign of luteinization or atresia (pyknotic nuclei). The second type of ovary was one containing many smaller ‘cysts’, giving the ovary a honeycomb appearance (Fig. 1C). Primordial follicles were still present.
2.5.2.1. Total o6arian weight and follicle numbers. Ovarian weights of 5-week old lambs exposed prenatally to T were greater than those of controls (PB0.05) and lambs exposed to DHT in utero (PB 0.05) (see Fig. 2, Table 1). The total number of preantral and antral follicles per ovary in the T (37.89 7.9 follicles) group was increased (P B 0.04) compared to that for either the C (16.59 2.8 follicles) or the DHT group (18.892.9 follicles). In C lambs, there were no ovarian follicles \2 mm in diameter, and follicles appeared to be in various stages of follicular development that ranged from primordial to preantral. By contrast, in the T-treated lambs, a majority of follicles were in the preantral stage, and two of the three females had an enlarged follicle (\ 2 mm in diameter). One lamb had one follicle 7 mm in diameter whereas the other lamb had a 4-mm follicle. In the DHT treated lambs, there were no large follicles.
Fig. 3. Left. Circulating patterns of LH and FSH in individual control lambs and lambs treated prenatally with testosterone or DHT at 5 weeks of age (three /group). Right. Mean concentrations of LH, FSH and LH/FSH ratio for the entire group studied (Control, n =6; T, n = 3; and DHT n= 5; open bar) and for the subset (n = 3/group) on which ovarian studies were performed (closed bars). Note the testosterone group was comprised of only 3 animals and is shown as closed bar only.
56
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
2.5.2.4. Expression pattern of follistatin, inhibin alpha, inhibin/acti6in beta A and beta B. Fifty percent of the follicles in the C animals expressed follistatin mRNA Fig. 5, Table 2. Relative to controls, the percentage of follicles expressing follistatin mRNA was increased (PB 0.05) in the T and DHT groups. There were no statistical differences in the percentage of follicles expressing inhibin a, activin bA, activin bB mRNA, although follicles expressing bB mRNA tended to be higher in the control animals (P=0.06).
3. Discussion
Fig. 4. Expression pattern of FSH receptor and aromatase mRNA from a representative ovary from the C group and lambs androgenized prenatally with T or DHT at 5 weeks of age. FSH receptor expression was localized to granulosa cells. Aromatase expression was present in granulosa cells of large follicles.
2.5.2.2. Concentrations of FSH, LH, and the ratio of LH: FSH. Fig. 3 presents the individual (three lambs) and mean concentrations of circulating LH and FSH from C-, T- and DHT-treated lambs; mean LH: FSH ratios are also shown. There were no differences in mean concentrations of LH, FSH or in LH/FSH ratios although FSH tended to be lower in the androgenized groups. The LH pulse frequency ranged from 0 to 2 pulses in all groups of animals with the exception of one DHT-treated animal which showed 4 pulses during the collection period (not shown). 2.5.2.3. Expression pattern of FSH receptor, LH receptor and aromatase. All follicles in each treatment group expressed FSH-receptor mRNA in the granulosa cells (Fig. 4) and LH-receptor mRNA in the thecal cells (not shown). The 4- and 7-mm follicles in the T-treated animals expressed little FSH-receptor mRNA in the granulosa cells, but a considerable amount of LH-receptor mRNA (not shown) was expressed in both the granulosa cells and thecal cells. In addition, these two follicles heavily expressed aromatase mRNA (Fig. 4). None of the 5 2 mm follicles expressed aromatase mRNA.
The results of this study show that, in sheep, prenatal androgenization with testosterone increases ovarian volume, follicle number and increases the percentage of follicles expressing follistatin mRNA. Interestingly, an increase in follicular number following prenatal androgenization was evident only when androgenization occurred with the aromatizable androgen, testosterone and not the non-aromatizable androgen, DHT. It is unclear how much of the ovarian dysfunction results from aberrant gonadotropic drive stemming from masculanization of the brain, or from direct effects of prenatal androgen on ovarian development. Characterization of gonadotropin profiles at 5 weeks of age revealed no discernable differences between C, T and DHT groups despite an increase in ovarian weight in the T-treated lambs. Previous studies have shown that prenatal androgenization of female sheep with testosterone advances the onset of neuroendocrine sexual maturation (as determined by increased LH secretion). In the model used (ovariectomized females with estradiol maintained at invariant levels by Silastic implant), the pubertal rise in LH began at 10–15 weeks of age rather than the usual 30 weeks of age (Wood and Foster, 1998). Because the biologic basis supporting an intra-follicular role for activin as a facilitator of folliculogenesis is strong, our paracrine study focused on this member of the TGF-beta family. With the exception of one report (Woodruff et al., 1990), most studies support a role for activin in folliculogenesis (Hillier, 1991; Doi et al., 1992; Rabinovici et al., 1992; Erickson et al., 1995; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). Studies in vitro have revealed that activin, in synergy with FSH, leads to formation of large ‘Graafian-like’ follicles (Li et al., 1995). Furthermore, overexpression of follistatin (a negator of activin action) in mice results in arrested follicular development, even in transgenic lines that failed to show FSH suppression (Guo et al., 1998). Activins have been found to potentiate FSH receptor expression (Hasegawa et al., 1988; Xiao et al., 1992; Findlay, 1993), LH receptor induction (Nakamura et al., 1994), FSH-induced estra-
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
diol production (Hutchinson et al., 1987; Xiao et al., 1990; Miro et al., 1991), and suppress progesterone secretion (Li et al., 1992; Shukovski and Findlay, 1990). The dual ability of activin to potentiate estradiol production and simultaneously suppress progesterone production in fully differentiated granulosa cells complements the attributes of the preovulatory follicle. Furthermore, activins increase the sensitivity of follicles to respond to FSH (LaPolt et al., 1989). Consistent with local action, activin-binding sites have been identified in granulosa cells (Feng et al., 1993; Woodruff et al., 1993). Comparison of intrafollicular expression of inhibin a, bA, bB and follistatin mRNA in control and prenatally-androgenized lambs revealed that the majority of follicles (over 80%) in the prenatally androgenized lambs, whether they were androgenized with T or DHT, expressed follistatin mRNA. This suggests that this is an androgen-specific effect and not facilitated by
57
aromatization of testosterone to estradiol. In contrast, the number of follicles expressing activin bB mRNA tended to be lower in lambs androgenized in utero with the aromatizable T as compared to controls and DHT groups and may be related to the action of estrogen. In concert, these findings suggest that the intrafollicular availability of activin is reduced in a majority of the follicles of the prenatally-androgenized group. Previous studies have shown that neutralization of activin activity with follistatin, a binding protein of activin (Nakamura et al., 1990; Robertson, 1992; de Winter et al., 1996), suppresses FSH-induced estradiol production (Xiao et al., 1990; Xiao and Findlay, 1991; Xiao et al., 1992; Shukovski et al., 1991) and that the expression of follistatin mRNA is decreased in preovulatory follicles of sheep (Tisdall et al., 1994; Braw-Tal, 1994). The differences in intrafollicular expression of inhibin and activin subunits and follistatin in the follicles of lambs exposed to androgens before birth may not be a
Fig. 5. Expression pattern of follistatin, inhibin a, activin/inhibin bA, bB in a representative ovary from a C lamb and lambs androgenized prenatally with testosterone or DHT at 5 weeks of age. Table 2 Percent of follicles expressing follistatin, inhibin alpha, activin beta A, activin beta B in control and prenatally androgenized sheep at 5 weeks of age Treatment
n
Follistatin
Inhibin alpha
Beta A
Beta B
Control Testosterone (200 mg) DHT (800 mg)
3 3 3
50.89 11.4 80.49 6.5* 80.39 5.5*
46.1 9 11.8 48.8 9 11.8 42.4 9 15.5
8.7 93.6 23.6 9 8.5 13.7 9 4.3
66.1 9 15.6 30.9 97.1 40.5 93.3
Values are mean 9 SEM. * Significance at PB0.05 level.
58
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
function of reduced FSH drive as no differences were found in the FSH milieu of control and prenatally-androgenized lambs, at least shortly before the ovaries were examined. Furthermore, granulosa cells of all follicles examined in the control and prenatally-androgenized lambs also expressed FSH receptor mRNA. The two large follicles in the prenatally androgenized lambs with granulosa cell expression of LH-R and aromatase were also the ones that had strong expression of bA. The presence of large follicles with attributes of a preovulatory follicle in these animals are consistent with their ability to cycle during the first breeding season. Based on reports that these sheep eventually become anovulatory in adulthood (Clarke et al., 1977; DeHaan et al., 1987), it is likely such large follicles may not exist throughout adulthood. Determination that the intrafollicular activin milieu of all follicles in adult anovulatory prenatally-androgenized sheep is compromised would strengthen our hypothesis that activin availability is crucial for follicular development. The multifolliculate condition of the ovaries of the prenatally-androgenized lamb is reminiscent of the aberrant follicular development characteristic of polycystic ovarian syndrome (PCOS) of women (Ehrmann et al., 1995; Dunaif, 1997). In these women, selection of a dominant preovulatory follicle occurs infrequently, thereby leading to accumulation of many cystic follicles in the ovary and chronic anovulation. Whether the multifolliculate condition in the prenatally-androgenized sheep is the outcome of increased ovarian androgen production such as that which occurs in PCOS remains to be determined. Women with virilizing forms of congenital adrenal hyperplasia, who are exposed to high levels of androgen during fetal life, are highly likely to exhibit PCOS-like features, such as anovulation, ovarian hyperandrogenism and multifolliculate ovaries (Barnes et al., 1994). This raises the possibility that prenatal exposure to androgens may play an important role in the pathophysiology of PCOS and may predispose women to develop PCOS later in life. As such, the findings of this study may be of relevance to understanding ovarian cycle disruption in women with PCOS.
Acknowledgements We are grateful to Douglas D. Doop and Gary McCalla for help with the animal experimentation; Kaye Brabec for the assistance in analyzing the FSHreceptor data and the preparation of photomicrographs; Dr David Tisdall (Reproductive Biology Group at Wallaceville Animal Research Centre, Wallaceville, New Zealand) for providing the cDNAs for ovine follistatin, inhibin a, and inhibin/activin bA; Dr Roy Rodgers (Australia) for providing the inhibin/activin
bB cDNA;. Dr Alan Garverick (University of Columbia, MO) for providing the cDNAs for FSH-receptor, LH-receptor and aromatase; Drs Gordon D. Niswender and Leo E. Reichert, Jr. for the generous supply of LH RIA reagents; and Dr Albert F. Parlow and the National Hormone & pituitary prgram for the generous supply of FSH reagents. Supported by USPHS HD23812, an educational grant from Parke Davis and Cores of the Center for the Study of Reproduction, NIH P30 HD18258 (Assay and Reagents, Sheep Research, Biostatistics).
References Adashi, E.Y., 1995. The ovarian follicular apparatus. In: Adashi, E.Y., Rock, J.A., Rosenwaks, Z. (Eds.), Reproductive Endocrinology, Surgery and Technology. Lippincott-Raven publishers, New York, pp. 17 –40. Barnes, R.B., Rosenfield, R.L., Ehrmann, D.A., Cara, J.F., Cuttler, L., Levitsky, L.L., Rosenthal, I.M., 1994. Ovarian hyperandrogenism as a result of congenital adrenal virilizing disorders: evidence for prenatal masculanization of neuroendocrine function in women. J. Clin. Endocrinol. Metab. 79, 1328 – 1333. Braw-Tal, R., 1994. Expression of mRNA for follistatin and inhibin/ activin subunits during follicular growth and atresia. J. Mol. Endocrinol. 13, 253 – 264. Chong, H., Pangas, S.A., Bernard, D.J., Wang, E., Gitch, J., Chen, W., Draper, L.B., Cox, E.T., Woodruff, T.K., 2000. Structure and expression of a membrane component of the inhibin receptor system. Endocrinology 141, 2600 – 2607. Clarke, I.J., Scaramuzzi, R.J., Short, R.V., 1977. Ovulation in prenatally androgenized ewes. J. Endocrinol. 73, 385 – 389. DeHaan, K.C., Berger, L.L., Kesler, D.J., McKeith, F.K., Thomas, D.L., Nash, T.G., 1987. Effect of prenatal androgenization on lamb performance, carcass composition and reproductive function. J. Anim. Sci. 65, 1465 – 1470. de Winter, J.P., ten Dijke, P., de Vries, C.J., van Achterberg, T.A., Sugino, H., de Waele, P., Huylebroeck, D., Verschueren, K., va den Eijnden-van Raaij, A.J., 1996. Follistatins neutralize activin bioactivity by inhibition of activin binding to its type II receptors. Mol. Cell Endocrinol. 116, 105 – 114. Doi, M., Igarashi, M., Hasegawa, Y., Eto, Y., Shibai, H., Miura, T., Ibuki, Y., 1992. In vivo action of activin-A on pituitary-gonadal system. Endocrinology 130, 139 – 144. Dunaif, A., 1997. Insulin resistance and the polycystic ovarian syndrome: mechanism and implications for pathogenesis. Endocr. Rev. 18, 774 – 800. Ehrmann, D.A., Barnes, R.B., Rosenfield, R.L., 1995. Polycystic ovary syndrome as a form of functional ovarian hyperandrogenism due to dysregulation of androgen secretion. Endocr. Rev. 16, 322 – 353. Erickson, G.F., Kokka, S., Rivier, C., 1995. Activin causes premature superovulation. Endocrinology 136, 4804 – 4813. Feng, Z.M., Madigan, M.B., Chen, C.L., 1993. Expression of Type II activin receptor genes in the male and female reproductive tissues of the rat. Endocrinology 132, 2593 – 2600. Findlay, J.K., 1991. The ovary. Baillieres Clinical Endocrinology & Metabolism 5, 755 – 769. Findlay, J.K., 1993. An update on the roles of inhibin, activin and follistatin as local regulators of folliculogenesis. Biol. Reprod. 48, 15 – 23. Findlay, J.K., Clarke, I.J., 1997. Regulation of the secretion of FSH in domestic ruminants. J. Reprod. Fert (Suppl) 34, 27 – 37.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59 Gougeon, A., 1996. Regulation of ovarian follicular development in primates: facts and hypotheses. Endocr. Rev. 17, 121 –155. Gray, P.C., Greenwald, J., Blount, A.L., Kunitake, K.S., Donaldson, C.J., Choe, S., Vale, W., 2000. Identification of a binding site on the type II activin receptor for activin and inhibin. J. Biol. Chem. 275, 3206 – 3212. Greenwald, G.S., 1972. Editorial: of eggs and follicles. Am. J. Anat. 135, 1 – 3. Guo, Q., Kumar, T.R., Woodruff, T., Hadsell, L.A., DeMayo, F.J., Matzuk, M.M., 1998. Overexpression of mouse follistatin causes reproductive defects in transgenic mice. Mol. Endocrinol. 12, 96– 106. Hasegawa, Y., Miyamoto, K., Abe, Y., Nakamura, T., Sugino, H., Eto, Y., Shibai, H., Igarashi, M., 1988. Induction of follicle stimulating hormone receptor by erythroid differentiation factor on rat granulosa cell. Biochem. Biophys. Res. Commun. 156, 668 –674. Hillier, S.G., 1991. Regulatory functions for inhibin and activin in human ovaries. J. Endocrinol. 131, 171 –175. Hirshfield, A.N., 1991. Development of follicles in the mammalian ovary. Int. Rev. Cytol. 124, 43 –101. Hutchinson, L.A., Findlay, J.K., de Vos, F.L., Robertson, D.M., 1987. Effects of bovine inhibin, transforming growth factor beta and bovine activin-A on granulosa cell differentiation. Biochem. Biophys. Res. Commun. 146, 1405 –1412. Knight, P.G., 1996. Roles of inhibins, activins and follistatin in the female reproductive system. Frontiers in Neuroendocrinology 17, 476 – 509. LaPolt, P.S., Soto, D., Su, J.G., Campen, C.A., Vaughan, J., Vale, W., Hseuh, A.J.W., 1989. Activin stimulation of inhibin secretion and messenger RNA levels in cultured granulosa cells. Mol. Endocrinol. 3, 1666 – 1673. Lewis, K.A., Gray, P.C., Blount, A.L., MacConell, L.A., Wiater, E., Bilezikjian, L.M., Vale, W., 2000. Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature 404, 411 – 414. Li, W., Yuen, B.H., Leung, P.C.K., 1992. Inhibition of progesterone accumulation by activin-A in human granulosa cells. J. Clin. Endocrinol. Metab. 75, 285 –289. Li, R., Phillips, D.M., Mather, J.P., 1995. Activin promotes ovarian follicle development in vitro. Endocrinology 136, 849 –856. Marshall, J.C., Dalkin, A.C., Haisenleder, D.J., Paul, S.J., Ortolano, G.A., Kelch, R.P., 1991. Gonadotropin-releasing hormone pulses: regulators of gonadotropin synthesis and ovulatory cycles. Recent. Prog. Horm. Res. 47, 155 –187. McNeilly, A.S., 1991. The ovarian follicle and fertility. J. Steroid Biochem. Molec. Biol. 40, 29 – 33. Miro, F., Smyth, C.D., Hillier, S.G., 1991. Development-related effects of recombinant activin on steroid synthesis in rat granulosa cells. Endocrinology 129, 3388 –3394. Nakamura, K., Nakamura, M., Igarashi, S., Miyamoto, K., Eto, Y., Ibuki, Y., Minegishi, T., 1994. Effect of activin on luteinizing hormone-human chorionic gonadotropin receptor messenger ribonucleic acid in granulosa cells. Endocrinology 134, 2329 – 2335. Nakamura, T., Takio, K., Eto, Y., Shibai, H., Titani, K., Sugino, H., 1990. Activin-binding protein from rat ovary is follistatin. Science 247, 836 – 838. Niswender, G.D., Reichert, L.E. Jr, Midgley, A.R. Jr, Nalbandov, A.V., 1969. Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 84, 1166 –1173. Padmanabhan, V., Mieher, C.D., Borondy, M., I’Anson, H., Wood, R.I., Landefeld, T.D., Foster, D.L., Beitins, I.Z., 1992. Circulating bioactive follicle-stimulating hormone and less acidic follicle-stimulating hormone isoforms increase during experimental induction of puberty in female lambs. Endocrinology 131, 213 –220. Rabinovici, J., Spencer, S.J., Doldi, N., Goldsmith, P.C., Schwall, R., Jaffe, R.B., 1992. Activin-A as an intraovarian modulator: actions, localization, and regulation of the intact dimer in human ovarian
59
cells. J. Clin. Invest. 89, 1528 – 1536. Robertson, D.M., 1992. Follistatin/Activin-binding protein. Trends Endocrinol. Metab. 3, 65 – 68. Robinson, J.E., Forsdike, R.A., Taylor, J.A., 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. Roche, J.F., 1996. Control and regulation of folliculogenesis-a symposium in perspective. Rev. Reprod. 1, 19 – 27. Rodgers, R.J., 1991. Cloning of the inhibin/activin beta B subunit gene from the Booroola merino sheep. J. Mol. Endocrinol. 6, 87 –93. Shukovsk, I.L., Findlay, J.K., 1990. Activin-A inhibits oxytocin and progesterone production by preovulatory bovine granulosa cells in vitro. Endocrinology 126, 2222 – 2224. Shukovski, L., Findlay, J.K., Robertson, D.M., 1991. The effect of follicle-stimulating hormone-suppressing protein or follistatin on luteinizing bovine granulosa cells in vitro and its antagonistic effect on the action of activin. Endocrinology 129, 3395 – 3402. Tisdall, D.J., Hill, D.F., Petersen, G.B., Fleming, J.S., 1992. Ovine follistatin: characterization of cDNA and expression in sheep ovary during the luteal phase of the oestrous cycle. J. Mol. Endocrinol. 8, 259 – 264. Tisdall, D.J., Hudson, N., Smith, P., McNatty, K.P., 1994. Localization of ovine follistatin and and alpha and beta A inhibin mRNA in the sheep ovary during the oestrous cycle. J. Mol. Endocrinol. 12, 181 – 193. Wood, R.I., Foster, D.L., 1998. Sexual differentiation of reproductive neuroendocrine function in sheep. Rev. Reprod. 3, 130 –140. Woodruff, T.K., Lyon, R.J., Hansen, S.E., Rice, G.C., Mather, J.P., 1990. Inhibin and activin locally regulate rat ovarian folliculogenesis. Endocrinology 127, 3196 – 3205. Woodruff, T.K., Krummen, L., McCray, G., Mather, J.P., 1993. In situ ligand binding of recombinant [125I]activin-A and recombinant human [125I]inhibin-A to the adult rat ovary. Endocrinology 133, 2998 – 3006. Woodruff, T.K., Mather, J.P., 1995. Inhibin, activin and the female reproductive axis. Ann. Rev. Physiol. 57, 219 – 244. Xiao, S., Findlay, J.K., Robertson, D.M., 1990. The effect of bovine activin and follicle-stimulating hormone (FSH) suppressing protein/follistatin on FSH-induced differentiation of rat granulosa cells in vitro. Mol. Cell Endocrinol. 69, 1 – 8. Xiao, S., Findlay, J.K., 1991. Interactions between activin and folliclestimulating hormone-suppressing protein and their mechanisms of action on cultured rat granulosa cells. Mol. Cell Endocrinol. 79, 99 – 107. Xiao, S., Robertson, D.M., Findlay, J.K., 1992. Effects of activin and follicle-stimulating hormone (FSH)- suppressing protein/follistatin on FSH receptors and differentiation of cultured rat granulosa cells. Endocrinology 131, 1009 – 1016. Xu, J, McKeehan, K., Matsuzaki, K., McKeehan, W.L., 1995a. Inhibin antagonizes inhibition of liver cell growth by activin by a dominantnegative mechanism. J. Biol. Chem. 270, 6308 – 6313. Xu, Z., Garverick, H.A., Smith, G.W., Smith, M.F., Hamilton, S.A., Youngquist, R.S., 1995b. Expression of follicle-stimulating hormone and luteinizing hormone receptor messenger ribonucleic acids in bovine follicles during the first follicular wave. Biol. Reprod. 53, 951 – 957. Xu, Z., Garverick, H.A., Smith, G.W., Smith, M.F., Hamilton, S.A., Youngquist, R.S., 1995c. Expression of messenger ribonucleic acid encoding cytochrome P450 side-chain cleavage, cytochrome p450 17 alpha-hydroxylase, and cytochrome P450 aromatase in bovine follicles during the first follicular wave. Endocrinology 136, 981 – 989. Ying, S.Y., 1988. Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr. Rev. 9, 267 – 293.
Intra-follicular activin availability is altered in prenatally-androgenized lambs Christine West a, Douglas L. Foster a,c, Neil P. Evans d, Jane Robinson e, Vasantha Padmanabhan a,b,* a
The Uni6ersity of Michigan, Reproducti6e Sciences Program, 300 N. Ingalls Building, Room. 1101 Ann Arbor, MI 48109 -0404, USA b Department of Pediatrics, The Uni6ersity of Michigan, Ann Arbor, MI 48109 -0404, USA c Department of Obstetrics and Gynecology, The Uni6ersity of Michigan, Ann Arbor, MI 48109 -0404, USA d Department of Veterinary Preclinical Studies, Uni6ersity of Glasgow, Glasgow, UK e Laboratory of Neuroendocrinology, The Babraham Institute, Cambridge, UK
Abstract Prenatal exposure of sheep to testosterone (T) disrupts ovarian cyclicity and leads to anovulation in adulthood. We propose that the disruption of ovarian function in prenatally-androgenized sheep is mediated via follicular defects stemming from reduced intrafollicular activin availability/action. The intra-follicular activin availability/action that facilitates follicular development is dictated by the relative proportions of activins, inhibins (antagonists of activin action) and follistatins (FS; binding proteins of activin and negator of activin action). Inhibin a, bA, bB, and FS mRNA expression were determined by in situ hybridization in 5 week-old ovaries from control (C) lambs or those exposed to testosterone (T) or DHT from 30 – 90 days of gestation. In utero exposure to T, but not DHT, increased total ovarian weight (0.4 9 0.1, 1.5 90.5 and 0.3 9 0.1 g, C, T and DHT, respectively) and total number of follicles (16.5 92.8, 37.8 97.9, and 18.8 9 3.0). With the exception of two follicles in T animals, all follicles were 52 mm in diameter. All follicles 5 2 mm in all groups expressed FSH receptor mRNA in the granulosa cells and LH receptor only in the thecal cells. The percentage of follicles expressing FS mRNA was increased (P B 0.05) in sheep prenatally-androgenized with either T (80.4 98) or DHT (80.3 9 5.5) as compared to C (50.8 98.2). In contrast, the percentage of follicles expressing activin bB mRNA tended to be lower (P= 0.06) in the T (30.9 9 7.1) and DHT (40.5 9 3.3) groups as compared to C (66.1 915.6). Increased expression of FS along with the reduced expression of activin bB mRNA provides evidence for compromised intra-follicular activin availability in the majority of follicles in the androgenized groups. The increase in ovarian weight and follicular number in the T, but not in the DHT group, suggests that the effects of T are mediated through the action of estrogen. We speculate that the decrease in relative abundance of activin may contribute to the selection defects in prenatally-androgenized sheep. If true, this may be a useful model to understand the etiology of polycystic ovarian syndrome. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Inhibin; Follistatin; Folliculogenesis; PCOS
1. Introduction Over a quarter century ago Gilbert Greenwald stated, ‘‘… one of the most intriguing mysteries in ovarian physiology is what factors determine whether Portions of this work were presented at the 31st Annual Meeting of the Society for the Study of Reproduction and the 81st Annual Meeting of the Endocrine Society. * Corresponding author. Tel.: + 1-734-647-0276; fax: + 1-734-9368620. E-mail address: [email protected] (V. Padmanabhan).
one follicle remains quiescent, another begins to develop but later becomes atretic, while still a third matures and ovulates’’ (Greenwald, 1972). Despite substantive research efforts, the mechanisms that underlie the regulation of folliculogenesis are still unclear. Folliculogenesis is a key reproductive event in the female that begins before birth and continues throughout reproductive life and involves progression of a follicle from the primordial to preovulatory state (Hirshfield, 1991; McNeilly, 1991; Findlay, 1991; Adashi, 1995; Gougeon, 1996; Roche, 1996). Failure of follicles to undergo developmental changes at the correct time and
0303-7207/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 1 ) 0 0 6 3 2 - 3
52
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
in an exact sequence leads to failure of folliculogenesis and subsequent deterioration of follicles through atresia. The mechanism by which some follicles are selected to develop to preovulatory status, in preference to other follicles that may have begun growth at the same time, is largely unknown. It has been proposed that selection and dominance, and subsequent escape from atresia, is associated with the ability of some follicles to continue growth and development despite the decline in FSH concentrations that occurs during the late follicular phase of the cycle (McNeilly, 1991; Marshall et al., 1991; Findlay and Clarke, 1997), possibly due to the action of local factors within the ovary such as members of the transforming growth factor family— activin, inhibin, and follistatin. These factors have all been shown to be able to act in a paracrine and autocrine manner, to regulate follicular proliferation and differentiation and as such could play an important role in folliculogenesis (Hillier, 1991; Rabinovici et al., 1992; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). Indeed, it has been shown experimentally that administration of activin leads to superovulation in rats (Erickson et al., 1995), and importantly this effect is mediated directly at the ovarian level, as it is maintained even in hypophysectomized animals (Doi et al., 1992). To properly assess activin availability it is essential to measure not only activin but follistatin and inhibin as well. Follistatins are activin binding proteins that negate activin’s action (Nakamura et al., 1990; Robertson, 1992; de Winter et al., 1996). Inhibins are functional opposites of activins, and they antagonize activin’s action by either binding with activin receptor type II (Xu et al., 1995a; Gray et al., 2000) or interfering with its signal transduction (Lewis et al., 2000; Chong et al., 2000). Adding to the complexity of these functional overlaps, these regulators also exist in multiple forms (Ying, 1988). The isomers identified in ovarian follicles are inhibin A, inhibin B, activin A, and activin B (Hillier, 1991; Rabinovici et al., 1992; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). These isomers appear to be similar functionally, but may differ in potency. Viewed in this context, the relative proportion of activins, follistatins and inhibins are important determinants of activin action. Impairment in expression of any of these regulators would alter intra-ovarian equilibrium, lead to disruption of folliculogenesis and subsequently anovulation. Animal models that exhibit either disrupted or arrested follicular development, provide valuable resources to investigate the intra-follicular environment that is required to facilitate folliculogenesis. One such model is the prenatally-androgenized female sheep. These are reproductively compromised, exhibit cyclic disruption and become anovulatory during adulthood (Clarke et al., 1977; DeHaan et al., 1987). Furthermore,
they manifest a multifolliculate condition (Clarke et al., 1977) such as that in women with PCOS (Dunaif, 1997; Ehrmann et al., 1995). The aim of this study was to investigate the cause of altered folliculogenesis in prenatally androgenized female sheep by determination of: (1) the effects of prenatal androgenization on follicular development; (2) the patterns of gonadotropin secretion; and (3) the intra-follicular activin environment in these animals relative to normal controls.
2. Methods
2.1. Animals and treatments 2.1.1. Pilot study To gain a preliminary understanding of the effects of prenatal androgenization on ovarian follicular development, a study was conducted on a small number of ovaries recovered from Poll Dorset sheep as part of a neuroendocrine study conducted at the Babraham Institute (Cambridge, UK) (Robinson et al., 1999). Ovaries were collected at 3 weeks of age from seven control lambs and seven lambs that had been prenatally androgenized by treatment of the dams with an aromatizable androgen, testosterone propionate (im; 100 mg twice weekly), for 60 days from fetal day 30 to 90 (term, 147 days). Following removal, all ovaries were weighed and processed for histology. 2.1.2. Main study Experiments were conducted using prenatally-androgenized Suffolk lambs maintained under normal husbandry conditions at the Sheep Research Facility in Ann Arbor, MI. For prenatal-androgenization, pregnant ewes were injected im twice weekly with 100 mg T propionate (T, aromatizable androgen) (n= 20 ewes), or 400 mg DHT propionate twice weekly (non-aromatizable androgen) (n= 12 ewes) from 30–90 days of pregnancy. Controls received no treatment. Fetal sex was unknown at the time of the treatments. Discounting the males born in this study, there were five females born in the DHT group and three females in the T-group. At 5 weeks of age, the ovaries were removed and weighed from C (n=5), T (n= 3) and DHT (n= 5) groups. Both ovaries from three animals in each group were used to determine follicular number and charcaterize by in situ hybridization the expression pattern of markers of follicular development (FSH receptor, LH receptor, aromatase), and factors involved in controlling local activin action (inhibin a, inhibin/activin bA, bB and follistatin). To determine the pattern and concentrations of LH and FSH experienced by the ovaries, blood samples were collected every 20-min for 4 h the day before ovariectomy.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
2.1.3. Assays Concentrations of FSH were measured using NIDDK-oFSH-1 (AFP 5679C) as the standard and NIDDK-anti-oFSH-1 (AFP-C5288113) antiserum at a working dilution of 1:12,000 (Padmanabhan et al., 1992). The sensitivity (2 S.D. of buffer control) and ED50 (50% displacement point) of the FSH assay averaged 0.06 and 0.37 ng respectively (n =1 assay). All plasma samples were measured in 100-ml volume in duplicate. The intra-assay coefficient of variation (C.V.) values at 80 and 20% displacement points averaged 7.3 and 3.6% respectively. Plasma LH was measured by a previously validated RIA (Niswender et al., 1969) using a serum standard calibrated against NIH-LH-S12 (biopotency 9.82×S1). The LH assay sensitivity and ED50 averaged 0.53 and 1.56 ng respectively (n= 2
53
assays). All plasma samples were measured initially in 200 ml in duplicate. Repeat assays were performed as needed in the 20–200 ml range. The intra-assay C.V. values at 80 and 20% displacement points averaged 4.9 and 2.4% respectively. The interassay C.V. values based on two quality control pools measuring 16.3 and 20.2 ng/ml, averaged 7.1 and 9.2%, respectively.
2.2. Riboprobe synthesis Both sense and antisense [35S]UTP-labeled cRNA probes for ovine follistatin (Tisdall et al., 1992), ovine inhibin a (Tisdall et al., 1994), ovine inhibin/activin bA (Tisdall et al., 1994), ovine inhibin/activin bB (Rodgers, 1991), bovine FSH-receptor (Xu et al., 1995b), ovine LH-receptor (Xu et al., 1995a) and bovine aromatase (Xu et al., 1995c), were transcribed from linearized cDNA templates using a transcription kit (Promega) according to the manufacturer’s recommendations and purified by centrifugation on a Sephadex G-50 column and diluted in hybridization buffer [50% formamide, 0.3 M NaCl, 10 mM Tris (pH 8.0), 1 mM EDTA (pH 8.0), 1× Denhardt’s solution [1× = 0.02% (wt./vol.) Ficoll, 0.02% polyvinylpyrrolidone, and 0.02% (wt./vol.) BSA], 10 mM dithiothreitol, 500 mg/ml salmon sperm DNA, and 10% dextran sulfate] to about 1×106 cpm/ml.
2.3. In situ hybridization
Fig. 1. Follicular morphology of an ovary from a control lamb (A) and lambs treated prenatally with testosterone (B, C). Ovaries were collected at 3 weeks of age. Note the disrupted nature of follicular development in prenatally-androgenized sheep. Scale bar bottom right is equal to 2.5 mm.
Sections (10 mm) of ovarian tissue were cut at − 20 °C using a cryostat (Reichert-Jung 2800 Frigocut), mounted onto microscope slides (Superfrost Plus; Fisher Scientific, Pittsburgh, PA), and stored at − 80 °C in boxes until analysis. Mounted slides (3 slides/ovary with two sections in each slide) were dried on a slide-warmer at 37 °C for 30 min and fixed in 4% paraformaldehyde in 0.1 M PBS for 60 min. Following rinsing with double strength standard sodium citrate solution (SSC; 1× = 150 mM NaCl and 15 mM Na3 citrate, pH 7.0) they were treated for 10 min in 0.25% acetic anhydride in 0.1 M triethanolamine (pH 8.0), rinsed twice in 2× SSC, dehydrated in ethanol, and air dried. Hybridization was performed using 50 ml diluted probe in a humidified oven at 55 °C for 16 h. Sections were hybridized to both the antisense probe and the sense probe for each mRNA of interest. After hybridization, slides were soaked first in 2× SSC to remove the coverslips, twice in 2× SSC for 2 min each time at room temperature and then treated with ribonuclease-A (20 mg/ml in 0.5 M NaCl, 10 mM Tris, and 1 mM EDTA) for 60 min at 37 °C. Slides were washed at room temperature in gradient strengths of SCC for 2 min each at room temperature, dehydrated in increasing concentrations of ethanol and air-dried. Autoradiography was performed overnight to determine the success of the hybridization. Subsequently the
54
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
Fig. 2. Follicular morphology of C lambs and lambs androgenized prenatally with T or DHT at 5 weeks of age. Both ovaries from two animals in each group are shown. Note the mulitfollicular nature of the ovaries collected from lambs androgenized with T. The majority of ovaries in the T group were also larger in size. Table 1 Total ovarian weight and number of follicles per ovary in control and prenatally-androgenized lambs at 5 weeks of age Treatment
n
Total ovarian weight (g)
Number of follicles per ovary
Control Testosterone (200 mg) DHT (800 mg)
5 3 5
0.41 9 0.08 1.52 9 0.46* 0.29 9 0.08*
16.50 9 2.84 37.83 9 7.91* 18.83 9 2.95
Values are mean 9 SEM. * Significance at PB0.05 level.
sections were dipped in NTB-2 emulsion (Eastman Kodak, Rochester, NY), and exposed at 4 °C in a light-tight box (exposure varied with time). The slides were then developed, lightly counterstained with hematoxylin and eosin, and mounted for microscopic examination. Follicles were determined to exhibit specific hybridization to a given probe if the hybridization intensity for the antisense probe was greater than the background hybridization observed to the sense probe. All mRNA expression was quantified with NIH Image on digital images captured from the autoradiography films. The number of follicles that expressed the individual mRNAs of interest were determined and expressed as a percentage of the total number of follicles on that ovary. For each ewe, the percentage of follicles labeled on each ovary was av-
eraged to derive a value for each animal. All individual measurements of follicular diameter were made on the section that had the largest diameter for each follicle.
2.4. Statistical analysis All data were analyzed by ANOVA (General Linear Models procedure of SAS, Cary, NC) to evaluate the effects of treatment. Before ANOVA, each set of data was subjected to Levene’s test for heterogeneity of variance. When heterogeneity was present, data were transformed before ANOVA by a method that yielded homogeneity of variance. Post-ANOVA comparisons among means were made using Bonferroni t test (PB 0.05). All values given are the mean9SEM of non-transformed values.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
55
2.5. Results
2.5.2. Main study
2.5.1. Pilot study At 3 weeks of age, mean weights of the pairs of ovaries from prenatally testosterone-treated lambs (0.46 9 0.19 g, n= 7) were greater (P B 0.05) than C lambs (0.1490.04 g, n =7). Histological examination revealed that all C ovaries appeared normal and contained multiple primordial and early primary follicles (Fig. 1). In contrast, none of the ovaries from the androgenized animals appeared normal. Morphologically the ovaries could be divided into two distinct types. The first type of ovary was one with a single very large follicle that occupied 70– 90% of the ovarian volume but also contained primordial follicles (Fig. 1B). The large follicle had a thin, disorganized granulosa cell layer and red blood cells in its lumen. There was little sign of luteinization or atresia (pyknotic nuclei). The second type of ovary was one containing many smaller ‘cysts’, giving the ovary a honeycomb appearance (Fig. 1C). Primordial follicles were still present.
2.5.2.1. Total o6arian weight and follicle numbers. Ovarian weights of 5-week old lambs exposed prenatally to T were greater than those of controls (PB0.05) and lambs exposed to DHT in utero (PB 0.05) (see Fig. 2, Table 1). The total number of preantral and antral follicles per ovary in the T (37.89 7.9 follicles) group was increased (P B 0.04) compared to that for either the C (16.59 2.8 follicles) or the DHT group (18.892.9 follicles). In C lambs, there were no ovarian follicles \2 mm in diameter, and follicles appeared to be in various stages of follicular development that ranged from primordial to preantral. By contrast, in the T-treated lambs, a majority of follicles were in the preantral stage, and two of the three females had an enlarged follicle (\ 2 mm in diameter). One lamb had one follicle 7 mm in diameter whereas the other lamb had a 4-mm follicle. In the DHT treated lambs, there were no large follicles.
Fig. 3. Left. Circulating patterns of LH and FSH in individual control lambs and lambs treated prenatally with testosterone or DHT at 5 weeks of age (three /group). Right. Mean concentrations of LH, FSH and LH/FSH ratio for the entire group studied (Control, n =6; T, n = 3; and DHT n= 5; open bar) and for the subset (n = 3/group) on which ovarian studies were performed (closed bars). Note the testosterone group was comprised of only 3 animals and is shown as closed bar only.
56
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
2.5.2.4. Expression pattern of follistatin, inhibin alpha, inhibin/acti6in beta A and beta B. Fifty percent of the follicles in the C animals expressed follistatin mRNA Fig. 5, Table 2. Relative to controls, the percentage of follicles expressing follistatin mRNA was increased (PB 0.05) in the T and DHT groups. There were no statistical differences in the percentage of follicles expressing inhibin a, activin bA, activin bB mRNA, although follicles expressing bB mRNA tended to be higher in the control animals (P=0.06).
3. Discussion
Fig. 4. Expression pattern of FSH receptor and aromatase mRNA from a representative ovary from the C group and lambs androgenized prenatally with T or DHT at 5 weeks of age. FSH receptor expression was localized to granulosa cells. Aromatase expression was present in granulosa cells of large follicles.
2.5.2.2. Concentrations of FSH, LH, and the ratio of LH: FSH. Fig. 3 presents the individual (three lambs) and mean concentrations of circulating LH and FSH from C-, T- and DHT-treated lambs; mean LH: FSH ratios are also shown. There were no differences in mean concentrations of LH, FSH or in LH/FSH ratios although FSH tended to be lower in the androgenized groups. The LH pulse frequency ranged from 0 to 2 pulses in all groups of animals with the exception of one DHT-treated animal which showed 4 pulses during the collection period (not shown). 2.5.2.3. Expression pattern of FSH receptor, LH receptor and aromatase. All follicles in each treatment group expressed FSH-receptor mRNA in the granulosa cells (Fig. 4) and LH-receptor mRNA in the thecal cells (not shown). The 4- and 7-mm follicles in the T-treated animals expressed little FSH-receptor mRNA in the granulosa cells, but a considerable amount of LH-receptor mRNA (not shown) was expressed in both the granulosa cells and thecal cells. In addition, these two follicles heavily expressed aromatase mRNA (Fig. 4). None of the 5 2 mm follicles expressed aromatase mRNA.
The results of this study show that, in sheep, prenatal androgenization with testosterone increases ovarian volume, follicle number and increases the percentage of follicles expressing follistatin mRNA. Interestingly, an increase in follicular number following prenatal androgenization was evident only when androgenization occurred with the aromatizable androgen, testosterone and not the non-aromatizable androgen, DHT. It is unclear how much of the ovarian dysfunction results from aberrant gonadotropic drive stemming from masculanization of the brain, or from direct effects of prenatal androgen on ovarian development. Characterization of gonadotropin profiles at 5 weeks of age revealed no discernable differences between C, T and DHT groups despite an increase in ovarian weight in the T-treated lambs. Previous studies have shown that prenatal androgenization of female sheep with testosterone advances the onset of neuroendocrine sexual maturation (as determined by increased LH secretion). In the model used (ovariectomized females with estradiol maintained at invariant levels by Silastic implant), the pubertal rise in LH began at 10–15 weeks of age rather than the usual 30 weeks of age (Wood and Foster, 1998). Because the biologic basis supporting an intra-follicular role for activin as a facilitator of folliculogenesis is strong, our paracrine study focused on this member of the TGF-beta family. With the exception of one report (Woodruff et al., 1990), most studies support a role for activin in folliculogenesis (Hillier, 1991; Doi et al., 1992; Rabinovici et al., 1992; Erickson et al., 1995; Findlay, 1993; Woodruff and Mather, 1995; Knight, 1996). Studies in vitro have revealed that activin, in synergy with FSH, leads to formation of large ‘Graafian-like’ follicles (Li et al., 1995). Furthermore, overexpression of follistatin (a negator of activin action) in mice results in arrested follicular development, even in transgenic lines that failed to show FSH suppression (Guo et al., 1998). Activins have been found to potentiate FSH receptor expression (Hasegawa et al., 1988; Xiao et al., 1992; Findlay, 1993), LH receptor induction (Nakamura et al., 1994), FSH-induced estra-
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
diol production (Hutchinson et al., 1987; Xiao et al., 1990; Miro et al., 1991), and suppress progesterone secretion (Li et al., 1992; Shukovski and Findlay, 1990). The dual ability of activin to potentiate estradiol production and simultaneously suppress progesterone production in fully differentiated granulosa cells complements the attributes of the preovulatory follicle. Furthermore, activins increase the sensitivity of follicles to respond to FSH (LaPolt et al., 1989). Consistent with local action, activin-binding sites have been identified in granulosa cells (Feng et al., 1993; Woodruff et al., 1993). Comparison of intrafollicular expression of inhibin a, bA, bB and follistatin mRNA in control and prenatally-androgenized lambs revealed that the majority of follicles (over 80%) in the prenatally androgenized lambs, whether they were androgenized with T or DHT, expressed follistatin mRNA. This suggests that this is an androgen-specific effect and not facilitated by
57
aromatization of testosterone to estradiol. In contrast, the number of follicles expressing activin bB mRNA tended to be lower in lambs androgenized in utero with the aromatizable T as compared to controls and DHT groups and may be related to the action of estrogen. In concert, these findings suggest that the intrafollicular availability of activin is reduced in a majority of the follicles of the prenatally-androgenized group. Previous studies have shown that neutralization of activin activity with follistatin, a binding protein of activin (Nakamura et al., 1990; Robertson, 1992; de Winter et al., 1996), suppresses FSH-induced estradiol production (Xiao et al., 1990; Xiao and Findlay, 1991; Xiao et al., 1992; Shukovski et al., 1991) and that the expression of follistatin mRNA is decreased in preovulatory follicles of sheep (Tisdall et al., 1994; Braw-Tal, 1994). The differences in intrafollicular expression of inhibin and activin subunits and follistatin in the follicles of lambs exposed to androgens before birth may not be a
Fig. 5. Expression pattern of follistatin, inhibin a, activin/inhibin bA, bB in a representative ovary from a C lamb and lambs androgenized prenatally with testosterone or DHT at 5 weeks of age. Table 2 Percent of follicles expressing follistatin, inhibin alpha, activin beta A, activin beta B in control and prenatally androgenized sheep at 5 weeks of age Treatment
n
Follistatin
Inhibin alpha
Beta A
Beta B
Control Testosterone (200 mg) DHT (800 mg)
3 3 3
50.89 11.4 80.49 6.5* 80.39 5.5*
46.1 9 11.8 48.8 9 11.8 42.4 9 15.5
8.7 93.6 23.6 9 8.5 13.7 9 4.3
66.1 9 15.6 30.9 97.1 40.5 93.3
Values are mean 9 SEM. * Significance at PB0.05 level.
58
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59
function of reduced FSH drive as no differences were found in the FSH milieu of control and prenatally-androgenized lambs, at least shortly before the ovaries were examined. Furthermore, granulosa cells of all follicles examined in the control and prenatally-androgenized lambs also expressed FSH receptor mRNA. The two large follicles in the prenatally androgenized lambs with granulosa cell expression of LH-R and aromatase were also the ones that had strong expression of bA. The presence of large follicles with attributes of a preovulatory follicle in these animals are consistent with their ability to cycle during the first breeding season. Based on reports that these sheep eventually become anovulatory in adulthood (Clarke et al., 1977; DeHaan et al., 1987), it is likely such large follicles may not exist throughout adulthood. Determination that the intrafollicular activin milieu of all follicles in adult anovulatory prenatally-androgenized sheep is compromised would strengthen our hypothesis that activin availability is crucial for follicular development. The multifolliculate condition of the ovaries of the prenatally-androgenized lamb is reminiscent of the aberrant follicular development characteristic of polycystic ovarian syndrome (PCOS) of women (Ehrmann et al., 1995; Dunaif, 1997). In these women, selection of a dominant preovulatory follicle occurs infrequently, thereby leading to accumulation of many cystic follicles in the ovary and chronic anovulation. Whether the multifolliculate condition in the prenatally-androgenized sheep is the outcome of increased ovarian androgen production such as that which occurs in PCOS remains to be determined. Women with virilizing forms of congenital adrenal hyperplasia, who are exposed to high levels of androgen during fetal life, are highly likely to exhibit PCOS-like features, such as anovulation, ovarian hyperandrogenism and multifolliculate ovaries (Barnes et al., 1994). This raises the possibility that prenatal exposure to androgens may play an important role in the pathophysiology of PCOS and may predispose women to develop PCOS later in life. As such, the findings of this study may be of relevance to understanding ovarian cycle disruption in women with PCOS.
Acknowledgements We are grateful to Douglas D. Doop and Gary McCalla for help with the animal experimentation; Kaye Brabec for the assistance in analyzing the FSHreceptor data and the preparation of photomicrographs; Dr David Tisdall (Reproductive Biology Group at Wallaceville Animal Research Centre, Wallaceville, New Zealand) for providing the cDNAs for ovine follistatin, inhibin a, and inhibin/activin bA; Dr Roy Rodgers (Australia) for providing the inhibin/activin
bB cDNA;. Dr Alan Garverick (University of Columbia, MO) for providing the cDNAs for FSH-receptor, LH-receptor and aromatase; Drs Gordon D. Niswender and Leo E. Reichert, Jr. for the generous supply of LH RIA reagents; and Dr Albert F. Parlow and the National Hormone & pituitary prgram for the generous supply of FSH reagents. Supported by USPHS HD23812, an educational grant from Parke Davis and Cores of the Center for the Study of Reproduction, NIH P30 HD18258 (Assay and Reagents, Sheep Research, Biostatistics).
References Adashi, E.Y., 1995. The ovarian follicular apparatus. In: Adashi, E.Y., Rock, J.A., Rosenwaks, Z. (Eds.), Reproductive Endocrinology, Surgery and Technology. Lippincott-Raven publishers, New York, pp. 17 –40. Barnes, R.B., Rosenfield, R.L., Ehrmann, D.A., Cara, J.F., Cuttler, L., Levitsky, L.L., Rosenthal, I.M., 1994. Ovarian hyperandrogenism as a result of congenital adrenal virilizing disorders: evidence for prenatal masculanization of neuroendocrine function in women. J. Clin. Endocrinol. Metab. 79, 1328 – 1333. Braw-Tal, R., 1994. Expression of mRNA for follistatin and inhibin/ activin subunits during follicular growth and atresia. J. Mol. Endocrinol. 13, 253 – 264. Chong, H., Pangas, S.A., Bernard, D.J., Wang, E., Gitch, J., Chen, W., Draper, L.B., Cox, E.T., Woodruff, T.K., 2000. Structure and expression of a membrane component of the inhibin receptor system. Endocrinology 141, 2600 – 2607. Clarke, I.J., Scaramuzzi, R.J., Short, R.V., 1977. Ovulation in prenatally androgenized ewes. J. Endocrinol. 73, 385 – 389. DeHaan, K.C., Berger, L.L., Kesler, D.J., McKeith, F.K., Thomas, D.L., Nash, T.G., 1987. Effect of prenatal androgenization on lamb performance, carcass composition and reproductive function. J. Anim. Sci. 65, 1465 – 1470. de Winter, J.P., ten Dijke, P., de Vries, C.J., van Achterberg, T.A., Sugino, H., de Waele, P., Huylebroeck, D., Verschueren, K., va den Eijnden-van Raaij, A.J., 1996. Follistatins neutralize activin bioactivity by inhibition of activin binding to its type II receptors. Mol. Cell Endocrinol. 116, 105 – 114. Doi, M., Igarashi, M., Hasegawa, Y., Eto, Y., Shibai, H., Miura, T., Ibuki, Y., 1992. In vivo action of activin-A on pituitary-gonadal system. Endocrinology 130, 139 – 144. Dunaif, A., 1997. Insulin resistance and the polycystic ovarian syndrome: mechanism and implications for pathogenesis. Endocr. Rev. 18, 774 – 800. Ehrmann, D.A., Barnes, R.B., Rosenfield, R.L., 1995. Polycystic ovary syndrome as a form of functional ovarian hyperandrogenism due to dysregulation of androgen secretion. Endocr. Rev. 16, 322 – 353. Erickson, G.F., Kokka, S., Rivier, C., 1995. Activin causes premature superovulation. Endocrinology 136, 4804 – 4813. Feng, Z.M., Madigan, M.B., Chen, C.L., 1993. Expression of Type II activin receptor genes in the male and female reproductive tissues of the rat. Endocrinology 132, 2593 – 2600. Findlay, J.K., 1991. The ovary. Baillieres Clinical Endocrinology & Metabolism 5, 755 – 769. Findlay, J.K., 1993. An update on the roles of inhibin, activin and follistatin as local regulators of folliculogenesis. Biol. Reprod. 48, 15 – 23. Findlay, J.K., Clarke, I.J., 1997. Regulation of the secretion of FSH in domestic ruminants. J. Reprod. Fert (Suppl) 34, 27 – 37.
C. West et al. / Molecular and Cellular Endocrinology 185 (2001) 51–59 Gougeon, A., 1996. Regulation of ovarian follicular development in primates: facts and hypotheses. Endocr. Rev. 17, 121 –155. Gray, P.C., Greenwald, J., Blount, A.L., Kunitake, K.S., Donaldson, C.J., Choe, S., Vale, W., 2000. Identification of a binding site on the type II activin receptor for activin and inhibin. J. Biol. Chem. 275, 3206 – 3212. Greenwald, G.S., 1972. Editorial: of eggs and follicles. Am. J. Anat. 135, 1 – 3. Guo, Q., Kumar, T.R., Woodruff, T., Hadsell, L.A., DeMayo, F.J., Matzuk, M.M., 1998. Overexpression of mouse follistatin causes reproductive defects in transgenic mice. Mol. Endocrinol. 12, 96– 106. Hasegawa, Y., Miyamoto, K., Abe, Y., Nakamura, T., Sugino, H., Eto, Y., Shibai, H., Igarashi, M., 1988. Induction of follicle stimulating hormone receptor by erythroid differentiation factor on rat granulosa cell. Biochem. Biophys. Res. Commun. 156, 668 –674. Hillier, S.G., 1991. Regulatory functions for inhibin and activin in human ovaries. J. Endocrinol. 131, 171 –175. Hirshfield, A.N., 1991. Development of follicles in the mammalian ovary. Int. Rev. Cytol. 124, 43 –101. Hutchinson, L.A., Findlay, J.K., de Vos, F.L., Robertson, D.M., 1987. Effects of bovine inhibin, transforming growth factor beta and bovine activin-A on granulosa cell differentiation. Biochem. Biophys. Res. Commun. 146, 1405 –1412. Knight, P.G., 1996. Roles of inhibins, activins and follistatin in the female reproductive system. Frontiers in Neuroendocrinology 17, 476 – 509. LaPolt, P.S., Soto, D., Su, J.G., Campen, C.A., Vaughan, J., Vale, W., Hseuh, A.J.W., 1989. Activin stimulation of inhibin secretion and messenger RNA levels in cultured granulosa cells. Mol. Endocrinol. 3, 1666 – 1673. Lewis, K.A., Gray, P.C., Blount, A.L., MacConell, L.A., Wiater, E., Bilezikjian, L.M., Vale, W., 2000. Betaglycan binds inhibin and can mediate functional antagonism of activin signalling. Nature 404, 411 – 414. Li, W., Yuen, B.H., Leung, P.C.K., 1992. Inhibition of progesterone accumulation by activin-A in human granulosa cells. J. Clin. Endocrinol. Metab. 75, 285 –289. Li, R., Phillips, D.M., Mather, J.P., 1995. Activin promotes ovarian follicle development in vitro. Endocrinology 136, 849 –856. Marshall, J.C., Dalkin, A.C., Haisenleder, D.J., Paul, S.J., Ortolano, G.A., Kelch, R.P., 1991. Gonadotropin-releasing hormone pulses: regulators of gonadotropin synthesis and ovulatory cycles. Recent. Prog. Horm. Res. 47, 155 –187. McNeilly, A.S., 1991. The ovarian follicle and fertility. J. Steroid Biochem. Molec. Biol. 40, 29 – 33. Miro, F., Smyth, C.D., Hillier, S.G., 1991. Development-related effects of recombinant activin on steroid synthesis in rat granulosa cells. Endocrinology 129, 3388 –3394. Nakamura, K., Nakamura, M., Igarashi, S., Miyamoto, K., Eto, Y., Ibuki, Y., Minegishi, T., 1994. Effect of activin on luteinizing hormone-human chorionic gonadotropin receptor messenger ribonucleic acid in granulosa cells. Endocrinology 134, 2329 – 2335. Nakamura, T., Takio, K., Eto, Y., Shibai, H., Titani, K., Sugino, H., 1990. Activin-binding protein from rat ovary is follistatin. Science 247, 836 – 838. Niswender, G.D., Reichert, L.E. Jr, Midgley, A.R. Jr, Nalbandov, A.V., 1969. Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 84, 1166 –1173. Padmanabhan, V., Mieher, C.D., Borondy, M., I’Anson, H., Wood, R.I., Landefeld, T.D., Foster, D.L., Beitins, I.Z., 1992. Circulating bioactive follicle-stimulating hormone and less acidic follicle-stimulating hormone isoforms increase during experimental induction of puberty in female lambs. Endocrinology 131, 213 –220. Rabinovici, J., Spencer, S.J., Doldi, N., Goldsmith, P.C., Schwall, R., Jaffe, R.B., 1992. Activin-A as an intraovarian modulator: actions, localization, and regulation of the intact dimer in human ovarian
59
cells. J. Clin. Invest. 89, 1528 – 1536. Robertson, D.M., 1992. Follistatin/Activin-binding protein. Trends Endocrinol. Metab. 3, 65 – 68. Robinson, J.E., Forsdike, R.A., Taylor, J.A., 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. Roche, J.F., 1996. Control and regulation of folliculogenesis-a symposium in perspective. Rev. Reprod. 1, 19 – 27. Rodgers, R.J., 1991. Cloning of the inhibin/activin beta B subunit gene from the Booroola merino sheep. J. Mol. Endocrinol. 6, 87 –93. Shukovsk, I.L., Findlay, J.K., 1990. Activin-A inhibits oxytocin and progesterone production by preovulatory bovine granulosa cells in vitro. Endocrinology 126, 2222 – 2224. Shukovski, L., Findlay, J.K., Robertson, D.M., 1991. The effect of follicle-stimulating hormone-suppressing protein or follistatin on luteinizing bovine granulosa cells in vitro and its antagonistic effect on the action of activin. Endocrinology 129, 3395 – 3402. Tisdall, D.J., Hill, D.F., Petersen, G.B., Fleming, J.S., 1992. Ovine follistatin: characterization of cDNA and expression in sheep ovary during the luteal phase of the oestrous cycle. J. Mol. Endocrinol. 8, 259 – 264. Tisdall, D.J., Hudson, N., Smith, P., McNatty, K.P., 1994. Localization of ovine follistatin and and alpha and beta A inhibin mRNA in the sheep ovary during the oestrous cycle. J. Mol. Endocrinol. 12, 181 – 193. Wood, R.I., Foster, D.L., 1998. Sexual differentiation of reproductive neuroendocrine function in sheep. Rev. Reprod. 3, 130 –140. Woodruff, T.K., Lyon, R.J., Hansen, S.E., Rice, G.C., Mather, J.P., 1990. Inhibin and activin locally regulate rat ovarian folliculogenesis. Endocrinology 127, 3196 – 3205. Woodruff, T.K., Krummen, L., McCray, G., Mather, J.P., 1993. In situ ligand binding of recombinant [125I]activin-A and recombinant human [125I]inhibin-A to the adult rat ovary. Endocrinology 133, 2998 – 3006. Woodruff, T.K., Mather, J.P., 1995. Inhibin, activin and the female reproductive axis. Ann. Rev. Physiol. 57, 219 – 244. Xiao, S., Findlay, J.K., Robertson, D.M., 1990. The effect of bovine activin and follicle-stimulating hormone (FSH) suppressing protein/follistatin on FSH-induced differentiation of rat granulosa cells in vitro. Mol. Cell Endocrinol. 69, 1 – 8. Xiao, S., Findlay, J.K., 1991. Interactions between activin and folliclestimulating hormone-suppressing protein and their mechanisms of action on cultured rat granulosa cells. Mol. Cell Endocrinol. 79, 99 – 107. Xiao, S., Robertson, D.M., Findlay, J.K., 1992. Effects of activin and follicle-stimulating hormone (FSH)- suppressing protein/follistatin on FSH receptors and differentiation of cultured rat granulosa cells. Endocrinology 131, 1009 – 1016. Xu, J, McKeehan, K., Matsuzaki, K., McKeehan, W.L., 1995a. Inhibin antagonizes inhibition of liver cell growth by activin by a dominantnegative mechanism. J. Biol. Chem. 270, 6308 – 6313. Xu, Z., Garverick, H.A., Smith, G.W., Smith, M.F., Hamilton, S.A., Youngquist, R.S., 1995b. Expression of follicle-stimulating hormone and luteinizing hormone receptor messenger ribonucleic acids in bovine follicles during the first follicular wave. Biol. Reprod. 53, 951 – 957. Xu, Z., Garverick, H.A., Smith, G.W., Smith, M.F., Hamilton, S.A., Youngquist, R.S., 1995c. Expression of messenger ribonucleic acid encoding cytochrome P450 side-chain cleavage, cytochrome p450 17 alpha-hydroxylase, and cytochrome P450 aromatase in bovine follicles during the first follicular wave. Endocrinology 136, 981 – 989. Ying, S.Y., 1988. Inhibins, activins and follistatins: gonadal proteins modulating the secretion of follicle-stimulating hormone. Endocr. Rev. 9, 267 – 293.