BIOLOGY OF REPRODUCTION 66, 1571–1578 (2002)
Stage-Specific Expression of Smad2 and Smad3 During Folliculogenesis1 Jian Xu, Jennifer Oakley, and Elizabeth A. McGee2 Magee-Womens Research Institute and the Department of Obstetrics and Gynecology and Reproductive Biology of the University of Pittsburgh, Pittsburgh, Pennsylvania 15213 ABSTRACT Paracrine and autocrine growth factors can affect many different aspects of ovarian follicle development. Many members of the transforming growth factor b (TGFb) family of growth factors and their receptors are expressed in developing follicles. However, the presence and function of the family of the TGFb signaling molecules known as Smads have not been evaluated during follicle development. We have demonstrated that two Smad family members that function as mediators for both activin and TGFb are expressed in granulosa cells of preantral follicles but not in large antral follicles. Smad2 expression, but not Smad3 expression, returns in luteal cells. Both Smad2 and Smad3 are translocated to the nucleus of granulosa cells in response to treatment with either TGFb or activin. However, Smad2 is more responsive to activin stimulation, and Smad3 is more responsive to TGFb stimulation. Stage-specific expression and differing ligand sensitivity of signaling molecules may work together to allow different effects of TGFb family ligands using the same signaling pathways over the course of follicular development.
activin, corpus luteum, follicular development, granulosa cells, growth factors
INTRODUCTION
The follicle is an essential functional unit of the ovary. Folliculogenesis is the development of the follicle from the primordial stage through a series of morphologically defined stages: primary, preantral, antral, culminating in the Graafian or preovulatory follicle stage. After ovulation, the follicle undergoes transformation into a corpus luteum [1]. The role of gonadotropins in regulating folliculogenesis is well established [2, 3], but over the last several years, it has become apparent that paracrine and autocrine growth factors also have a role in ovarian follicle development [4]. One group of growth factors that has been implicated in many critical aspects of follicle development is the transforming growth factor b (TGFb) superfamily [5–10]. The TGFb superfamily includes TGFb, activin, inhibin, mu¨llerian inhibiting substance, growth and development factor 9, and the bone morphogenetic proteins [11]. These factors play important roles in a wide range of biologic effects, such as cell growth, morphogenesis, cell differentiation, and apoptosis. The TGFb family members initiate their celThis work was supported by Magee-Womens Research Institute Postdoctoral Fellowship (J.X.), the American Society of Reproductive Medicine/ Ortho MacNeil Award in Reproductive Medicine (E.A.M.), and Women’s Reproductive Health Research Center Program sponsored by the National Institute of Child Health and Human Development (E.A.M.). 2 Correspondence: Elizabeth A. McGee, MWRI, 204 Craft Ave., Pittsburgh, PA 15213. FAX: 412 641 5290; e-mail:
[email protected] 1
Received: 30 July 2001. First decision: 29 August 2001. Accepted: 19 December 2001. Q 2002 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org
lular actions via heterodimeric serine-threonine kinase receptor complexes. The ligands bind to their specific type II receptor, then activate the type I receptor by phosphorylation of its glycine-serine-rich domain. The activated type I receptor can then phosphorylate downstream signaling molecules [12]. The Smad family of signal transduction molecules is a relatively new group of proteins that transmit TGFb signals from the cell surface into the nucleus. They were initially described in Caenorhabditis elegans as Sma (small) genes [13] and in Drosophila as Mad (mothers against Decapentaplegic) genes [14]. Currently, eight different Sma- and Mad-related genes have been identified in mammals and are termed Smads [15]. They can be subdivided into three distinct subclasses based on function: receptor-activated Smads (R-Smads: Smad1, Smad2, Smad3, Smad5, Smad8, and Smad9), common mediator Smads (Co-Smad: Smad4), and inhibitory or antagonistic Smads (anti-Smads: Smad6 and Smad7) [16]. Activated type I receptors associate with specific R-Smads, phosphorylating them at the COOH-terminus of the protein. The phosphorylated R-Smad dissociates from the receptor and forms a heteromeric complex with the Co-Smad Smad4, and together the heteromeric complex translocates to the nucleus, mediating the target gene responses. Anti-Smads function as antagonists, blocking both Smad phosphorylation and association with Smad4 [17]. Different TGFb family members can have very different effects on follicle development [18]. Downstream signaling proteins might be the determinants of these different actions, but very little is known about the expression and regulation of the Smad signaling proteins in the ovary. Smad2 and Smad3 are highly homologous R-Smads that have been associated with both TGFb and activin signaling [19]. In these studies, we have determined the expression and cellular localization of these two important signaling molecules over the course of follicular development and luteogenesis. We have also determined the ability of TGFb and activin to affect nuclear translocation of Smad2 and Smad3 in granulosa cells. MATERIALS AND METHODS
Hormones and Reagents Human FSH (ISIAFP-1; 8466 IU/mg) was obtained from National Hormone and Pituitary Distribution Program, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health (NIH) (Baltimore, MD). The hormones eCG (6000 IU/mg) and hCG (3000 IU/mg) were purchased from Sigma Chemical Company (St. Louis, MO). TGFb1 and recombinant human activin A were purchased from R&D Systems (Minneapolis, MN). All other reagents were purchased from Sigma.
Animals All animal experiments were performed in accordance with NIH guidelines with institutional approval. Sprague-Dawley rats were obtained from Harlan Company (Indianapolis, IN) and housed under standard conditions. Ovaries were obtained at 1300 h from 4-day cycling rats on the day of
1571
1572
XU ET AL. tions were performed under microscopic visualization using fine needles. All animals were anesthetized with CO2 and killed by cervical dislocation.
Western Blot Analysis Ovaries were homogenized in lysis buffer (NaCl 150 mM, Tris-HCl 50 mM, MnCl2 0.5 mM, MgCl2 0.5 mM, EGTA 5 mM, PMSF 0.2 mM, SDS 1%) and then centrifuged for 20 min, and the supernatant was collected. Protein content was quantitated with the BCA analysis kit (Pierce, Rockford, IL). Fifteen micrograms of protein lysate from each sample was subjected to denaturing PAGE and transferred to polyvinylidene fluoride membrane. The resultant membranes were blocked with Tris-buffered saline (TBS) containing 5% (w/w) fat-free dry milk at 48C overnight, then washed with TBST (a mixture of TBS and 0.05% Tween 20) three times for 10 min each, and then incubated for 2 h at room temperature with 3 mg/ml specific rabbit polyclonal antibody against Smad2 or Smad3 (Zymed Lab, San Francisco, CA). After three 10-min washes in TBST, the membranes were incubated with peroxidase-conjugated affinipure goat anti-rabbit immunoglobulin G (IgG) (Jackson ImmunoResearch Lab, West Grove, PA) diluted 1:125 000 in 1% BSA in TBST for 1 h at room temperature. They were then washed with TBST three times for 10 min each and imaged with the ECLPlus Western Blotting Detection System (Amersham, Buckinghamshire, England). ECL-incubated blots were exposed to Hyperfilm (Amersham). Groups of 20 preantral follicles from 23-day-old rats, 10 preovulatory follicles from eCG-treated rats, and 10 corpora lutea from 24-h and 10 corpora lutea from 48-h hCG-treated rats were homogenized. Ten micrograms of each protein was loaded per lane, and Western blot analysis was performed as previously described.
Granulosa Cell Culture and Treatment Ovaries from 25-day-old female rats were removed and cleaned of adipose and connective tissue. Ovaries were placed in McCoy medium containing 0.2% BSA, 26 mM sodium bicarbonate, and 6.8 mM EGTA and incubated for 10 min at 378C and then transferred to a solution containing McCoy medium with 0.5 M sucrose. Ovaries were then washed and suspended in fresh McCoy medium [21]. Follicle puncture was performed with fine needles under microscopic visualization. The cells were gently pelleted and rinsed and resuspended in McCoy medium. Cells were counted using a hemocytometer, and viability was determined by trypan blue staining. Cells were cultured on glass coverslips in McCoy medium containing 10% fetal bovine serum at 378C in a humid atmosphere containing 5% CO2. After 24 h, the medium was changed to serum-free McCoy medium containing 0.1% BSA and 10 ml/ml of ITS 1 Culture Supplement (contains 1.0 mg/ml insulin, 0.55 mg/ml human transferrin, 0.5 mg/ml sodium selenite, 50 mg/ml BSA, and 0.47 mg/ml linoleic acid; Collaborative Biomedical, Bedford, MA) for 24 h. The cells were then incubated for 24 h in the aforementioned serum-free medium containing FSH (10 ng/ml). After FSH treatment, groups of cells were incubated with TGFb (0.1–10 ng/ml) or with activin A (2–200 ng/ml) for 60 min. Cells were washed with PBS three times for 5 min each and fixed in 4% paraformaldehyde before immunofluorescence staining.
Biotin-Avidin D Cell Sorter System Indirect Immunofluorescence Techniques
FIG. 1. Western analysis of whole ovary lysates (15 mg) obtained from 4-day cycling rats. Relative expression of Smad2 (A) and Smad3 (B) proteins on each day of the estrous cycle. Signals were normalized to the day of estrus, which was arbitrarily designated as one. The bars represent the average of relative band intensity from six experiments 6 SEM. ✸ P , 0.05 (vs. estrus).
proestrus, estrus, metestrus, or diestrus. For the hormone-treated groups, 23-day-old female rats were injected with 10 IU eCG, and 48 h later were injected with 10 IU hCG. Ovaries were obtained at 0, 1, 3, 6, 12, and 24 h after hCG treatment. Ovaries were either frozen for later Western analysis or fixed in 4% paraformaldehyde and embedded in OCT (Sakura Finetek USA, Inc., Torrance, CA) [20] for immunohistochemistry. Each time point or cycle day contained at least six animals. Follicles and corpora lutea were dissected from fresh ovaries of some treatment groups. Dissec-
Frozen sections were cut, air-dried, and incubated with 10% goat serum in PBS for 1 h at 48C in a humidified chamber. Specific rabbit polyclonal antibodies against Smad2 or Smad3 (1:100 dilution in 10% goat serum in PBS, 5 mg/ml; Zymed Lab) were applied to the sections and slides were incubated overnight at 48C. For negative control, the primary antibody was preincubated with 50 mg/ml Smad2 or Smad3 peptide for 2 h. Slides were washed with cold PBS and biotinylated anti-rabbit IgG antibody (1: 200 in 10% goat serum in PBS; Vector Laboratories, Burlingame, CA) was applied to the tissue sections for 1 h. Slides were washed in PBS and fluorescein avidin D cell sorter (1:200 in sodium bicarbonate buffer; Vector Labs) was applied to the sections, and the slides were incubated for 1 h at 48C. After a wash with cold PBS, slides were mounted with Vectashield mounting medium (Vector) and visualized on a fluorescence microscope using a 510-nm filter.
Data Analysis All experiments were repeated at least three times. The intensities of Western blotting signals were quantitated using a scanning transmittance densitometer with computerized analysis (Hoeffer Science, San Francisco, CA). Signals obtained from Smad2 and Smad3 protein hybridizations were
SMAD2 AND SMAD3 EXPRESSION DURING FOLLICULOGENESIS
1573
then normalized against signal from the estrous group (Fig. 1) or the control group (Fig. 2). For the granulosa cell culture experiments, 100 cells from each group were counted based on a semirandom grid with at least one field from each of the four quadrants. Virtually all of the cultured cells exhibited some level of staining for both proteins. For each counted cell, the observer (who was blinded to the treatment group) determined visually whether the nuclear staining was greater than the cytoplasmic staining. This was defined as a nuclear-stained cell. The counts are reported as the percentage of cells with staining based on this definition. Statistical significance between mean values was determined by ANOVA followed by post hoc testing and was accepted at the 0.05 level.
RESULTS
Variation of Smad2 and Smad3 Protein Expression During the Estrous Cycle
Western analysis for Smad2 and Smad3 resulted in specific bands at 58 and 56 kDa, respectively. As shown in Figure 1, both Smad2 and Smad3 proteins were expressed on each day of the estrous cycle and had the lowest levels of expression on the day of estrus. Smad2 protein expression increased by a factor of 2.6 on the day of proestrus and 3.3 on the day of metestrus compared with estrous levels of expression (P , 0.05, Fig. 1A). The expression level of Smad3 almost doubled on the day of proestrus and diestrus (P , 0.05, Fig. 1B). Expression of Smad2 and Smad3 Proteins in Ovaries from eCG/hCG-Treated Rats
To more closely evaluate the expression pattern of Smad2 and Smad3 over the course of follicular development and luteogenesis, we determined the effect of sequential gonadotropin treatment on ovarian expression of the Smads (Fig. 2). As seen in Figure 2, Western analysis demonstrated a high level of expression for both Smad2 and Smad3 in control ovaries. After eCG treatment, which induces preovulatory follicle development, Smad2 expression was reduced to 53% and Smad3 expression to 68% of the control level (P , 0.05). After hCG treatment to trigger ovulation, expression for both Smad proteins was further suppressed in a time-dependant manner. Three hours after hCG treatment, Smad2 expression was 37% and Smad3 expression was 53% of the control level (P , 0.05). Expression of both proteins reached a nadir at 12 h, when Smad2 expression was only 29% and Smad3 was 37% of the control level (P , 0.05). However, 24 h after hCG treatment, the expression levels of the two Smads differed dramatically. Smad2 expression had increased to 71% of the control level (P , 0.05), whereas Smad3 expression remained near the nadir at 48% of the control level. Localization of Smad2 and Smad3 Proteins in the Ovary
To determine the cellular localization of Smad2 and Smad3 protein expression, we performed immunohistochemical analysis on ovarian sections from 23-day-old and cycling animals as described in Materials and Methods. As shown in Figure 3, A and B, the oocytes of primordial and primary follicles exhibited very strong positive staining (seen as green) for Smad2 protein; however, the oocytes of larger follicles did not have strong staining. This pattern was the same in immature and cycling ovaries. In the preantral and small antral follicles, granulosa cells had strong positive signals; however, staining was diminished markedly in granulosa cells of large antral follicles. Theca cells maintained low levels of expression for Smad2 in preantral and large antral follicles.
FIG. 2. Western analysis of whole ovary lysates (15 mg) obtained from eCG- and eCG-hCG-treated rats. Data points are control from ovaries from 25-day-old rats; PMSG from ovaries 48 h after eCG treatment; hCG3 from ovaries 51 h after eCG and 3 h after hCG; hCG12 from ovaries 60 h after eCG and 12 h after hCG; hCG24 from ovaries 72 h after eCG and 24 h after hCG. Signals were normalized to the control group, which was arbitrarily designated as 1. Bars represent the average of relative band intensity from six experiments 6 SEM. ✸ P , 0.05 (vs. control group of Smad2), ù P , 0.05 (vs. control group of Smad3), ❋ P , 0.05 (vs. hCG12 group of Smad2).
Smad3 protein was also localized in the oocytes of primordial and primary follicles. Strong positive signal was seen in granulosa cells in the preantral and small antral follicles. In large antral follicles, there was minimal granulosa cell staining. Light staining for Smad3 was seen in the theca cells (Fig. 3, D and E). Negative control sections with primary antibody preadsorbed with Smad2 or Smad3 peptide demonstrated negligible background staining (Fig. 3, C and F). In gonadotropin-treated animals, immunohistochemical analysis revealed minimal positive staining for both Smad2 and Smad3 in granulosa cells of antral follicles after eCG treatment (Fig. 4A). However, occasional granulosa cells did exhibit cytoplasmic staining for Smad3 (Fig. 4E), and a low level of theca staining was maintained for both proteins (Fig. 4, A and E). Twenty-four hours after hCG treatment, the newly formed corpora lutea exhibited minimal signal for Smad3 (Fig. 4F). However, Smad2 signal was readily apparent in the corpora lutea (Fig. 4B), although the intensity was less than that seen in preantral follicles. Higher magnification (4003) of the corpus luteum clearly revealed a cytoplasmic staining pattern for Smad2 (Fig. 4C) but not for Smad3 (Fig. 4G). Negative control sections with primary antibody preadsorbed with Smad2 or Smad3 peptide demonstrated negligible background staining (Fig. 4, D and H).
1574
XU ET AL.
FIG. 3. Immunohistochemical staining of rat ovary for Smad2 (top row) and Smad3 (bottom row). Slides were visualized at 2003 magnification using a fluorescence microscope with a 510-nm filter. The positive signal is seen as green. Negative control sections with primary antibody preadsorbed with Smad2 or Smad3 peptide are shown in C and F. P, Primordial follicle; 10, primary follicle; PA, preantral follicle; A, antral follicle; G, granulosa; T, theca; CL, corpus luteum.
Expression of Smad2 and Smad3 Proteins in Isolated Follicles and Corpora Lutea
To confirm a reduction in Smad3 protein expression in corpora lutea, we dissected individual follicles at different stages of development as well as corpora lutea at 24 and 48 h after hCG treatment and performed Western analysis on the resulting lysates. Figure 5 is a representative Western analysis of these proteins probed for Smad2 (upper panel) and Smad3 (lower panel). As expected, Smad2 and Smad3 were expressed at high levels in isolated preantral follicles. Expression of both proteins was reduced in preovulatory follicles. Smad2 protein expression was clearly maintained in corpora lutea at 24 and 48 h after hCG treatment. In contrast, Smad3 expression was very low in isolated corpora lutea. Jurkat cells and rat liver lysate were used for positive controls for Smad2 and Smad3, respectively. Differential Effects of TGFb and Activin A on Smad2 and Smad3 Protein Localization in Cultured Granulosa Cells
Cultured granulosa cells were used to determine whether TGFb or activin A could alter Smad2 or Smad3 protein localization. As shown in Figure 6, both Smad2 and Smad3 proteins were primarily expressed in the cytoplasm of cultured granulosa cells in the absence of the growth factors. Granulosa cells exposed to FSH exhibited a morphologic change described as rounding [22], but nuclear expression of both Smad proteins was fairly low, even in rounded cells. Smad2 was localized to the nucleus in 14.5% of cells, and Smad3 was localized to the nucleus in 22.5% of cells (Fig. 6A). When cells were treated with FSH for 24 h and then treated with either TGFb (1 ng/ml) or activin A (20 ng/ml), nuclear staining for both Smad2 and Smad3 was markedly increased. TGFb treatment for 60 min resulted in 30% and 50%
nuclear staining for Smad2 and Smad3, respectively (Fig. 6A; P , 0.05 compared with FSH-treated cells). Activin treatment for 60 min increased the level of nuclear staining for Smad2 to 53% and Smad3 to 36% (Fig. 6A, P , 0.05) compared with FSH-treated cells. Dose responses for TGFb and activin confirmed this relationship (Fig. 6, B and C). At higher doses of TGFb (10 ng/ml), the percentage of nuclear expression of Smad2 becomes similar to that for Smad3. DISCUSSION
TGFb family members are considered to be important regulators of follicle development; however, the specific mechanisms of action of these growth factors in the ovary remain undefined. Recently, we [18] and others [23] have reported distinct effects of TGFb and activin on preantral follicle development in vitro. Over the last several years, there have been many reports of effects of TGFb and activin in granulosa cells and larger follicles [5–9, 24–28]. However, these reports often differ on whether activin or TGFb have roles in follicle growth, atresia, or differentiation. Characterization of downstream signaling of the TGFb growth factors may help explain the seeming contradictions of various studies. TGFb and activin are both considered to work through both Smad2 and Smad3. We hypothesized that the differential effects of TGFb and activin in follicular development might result from differential expression or activation of the mediating Smad proteins during folliculogenesis. To address this hypothesis, we have characterized the expression of these two signaling proteins over the course of follicular development in cycling and gonadotropin-stimulated ovaries. We have also demonstrated the effects of TGFb and activin on the translocation of Smad2 and Smad3 to the nucleus in granulosa cells. Western analysis of whole ovarian lysates from each day
SMAD2 AND SMAD3 EXPRESSION DURING FOLLICULOGENESIS
1575
FIG. 4. Immunohistochemical staining of Smad2 and Smad3 in the ovary of eCGand hCG-treated rats. A, E) Serial sections from an ovary after eCG treatment (magnification 3200). B, F) Serial sections from ovaries obtained 24 h after hCG treatment (magnification 3200). C, G) Higher-power magnifications (magnification 3400) of the corpus luteum in B and G, respectively. D, H) Negative control sections with primary antibody preadsorbed with Smad2 or Smad3 peptide.
of the estrous cycle demonstrated distinct fluctuations of protein expression on different days of the cycle. In the rat, ovulation occurs early on the morning of estrus. On the following day (metestrus), progesterone levels rise, consistent with the formation of corpora lutea [29, 30]. Both Smad proteins are expressed at high levels on the day of proestrus and lower levels on the day of estrus. Notably, on metestrus, Smad2 protein expression is increased relative to estrous levels of expression, whereas Smad3 protein expression does not increase on metestrus. This pattern suggests that Smad2 may play a more significant role than Smad3 in corpus luteum function. To more closely evaluate the expression patterns of Smad2 and Smad3 during the later stages of folliculogenesis and luteogenesis, we induced follicular growth and ovulation with gonadotropin treatments. Before hormonal treatments, the ovaries had high basal levels of Smad pro-
tein expression. Treatment with eCG resulted in only a small suppression of expression. However, the induction of ovulation with hCG induced a marked suppression of the expression of both proteins, reaching a nadir at 12 h after hCG treatment. In this model, ovulation generally occurs 8–13 h after hCG administration. Consistent with the expression levels seen in the cycling animals, Smad2 and Smad3 proteins are expressed at low levels during ovulation. Of interest, 24 h after hCG treatment, Smad2 protein levels again become elevated. At this time, ovulation has occurred, and the early corpus luteum is present. Therefore, in both naturally cycling and hormonally induced early corpora lutea, Smad2 expression increases, whereas Smad3 expression does not. In addition to the quantity of regulatory protein present, the location of protein expression is also essential in determining its role in function. The cellular localization of
1576
XU ET AL.
FIG. 5. Western analysis of isolated follicles and corpora lutea. Bands are from representative paired blots subjected to immunodetection for Smad2 (top row) or Smad3 (bottom row). Jurkat cells and rat liver lysate were used as positive controls for Smad2 and Smad3, respectively. This experiment was repeated three times with tissue from different groups of animals.
Smad2 and Smad3 proteins during folliculogenesis was determined by immunohistochemistry. High expression was noted in the granulosa cells of preantral follicles for both Smad2 and Smad3. Both proteins were expressed at low levels in the granulosa cells of large antral and preovulatory follicles. As predicted by the Western analyses, Smad2 expression increased in the corpora lutea, yet minimal signal was seen for Smad3 in corpora lutea (Figs. 4 and 5). Oocyte expression of Smad proteins also changes during folliculogenesis. Both Smad2 and Smad3 are expressed at high levels in oocytes of primordial and primary follicles. As oocytes reach full size, expression of both proteins declines and is no longer apparent in oocytes of antral follicles (Figs. 3 and 4). However, after gonadotropin stimulation but before ovulation, some oocytes again express Smad2 and Smad3 (data not shown). This finding is consistent with a report demonstrating the presence of Smad2 and Smad3 expression in unfertilized, ovulated human oocytes and preimplantation embryos up to the eight-cell stage [31]. We have demonstrated that both activin and TGFb can
FIG. 6. Nuclear translocation of Smad2 and Smad3 after treatment with TGFb or activin. A) Percentage of nuclear stained cells for Smad2 or Smad3 after 60 min of treatment with TGFb (1 ng/ml) or activin (20 ng/ml). Inset: representative micrographs of control and treated cells stained for SMAD2 (upper panel) and Smad3 (lower panel). B) Dose response of TGFb effects on Smad2 (dashed line) and Smad 3 (solid line). Cells were pretreated with FSH (10 ng/ml) and then treated with 0.1, 1, or 10 ng/ml TGFb for 60 min as described in Materials and Methods. C) Dose response for activin. Cells were pretreated with FSH and then treated with 2, 20, or 200 ng/ml of activin.
SMAD2 AND SMAD3 EXPRESSION DURING FOLLICULOGENESIS
induce nuclear translocation of both Smad2 and Smad3 in granulosa cells. For receptor-activated Smads, nuclear translocation is a function of receptor-induced phosphorylation [12]. The Smad protein is activated by phosphorylation of a serine residue in the MH1 domain. This promotes a structural change that allows association with the Co-Smad Smad4, and then the complex undergoes nuclear translocation [17]. Thus our data demonstrating nuclear translocation of the proteins also signifies receptor-specific activation of the protein. Both activin and TGFb activate both Smad2 and Smad3 [16, 32]. However, the relative effectiveness of the growth factors in our study suggests a functional preference of Smad2 for activation by activin and Smad3 for activation by TGFb in granulosa cells. Both type I and type II receptors for activin and TGFb are present in the developing follicle. Although detailed analysis in the rat ovary has not been performed for all of the receptor types, studies in humans and hamsters [33–35] have shown that both pairs of activin receptors and the TGFb receptor pair are present in the granulosa, theca, oocytes, and interstitium of the ovary. Also, a recent study in the rat has localized mRNA for all four of the activin receptors in isolated oocytes and granulosa cells by real-time polymerase chain reaction [36]. Further studies are needed to correlate the stage-specific expression of all of the receptor subtypes that may function in activin and TGFb signaling during folliculogenesis. Functional studies are also necessary to associate specific Smads to specific receptor functioning. The overlap of growth factor effects in ovarian function and the lack of specificity in growth factor-receptor interactions have made it difficult to delineate the exact roles of specific growth factors in the ovary. In these experiments, we have begun to characterize the role of the TGFb family of signaling proteins known as Smads in ovarian function. We have found that the two Smads most closely associated with TGFb and activin signaling; Smad2 and Smad3, have stage-specific differential expression during folliculogenesis. We have also demonstrated that these two Smads also exhibit functionally different sensitivity to TGFb and activin in the granulosa cell. Further characterization of the functioning of these signaling molecules may reveal unique roles in regulating follicular development. ACKNOWLEDGMENTS The authors are grateful to Dr. Sandra Legan and Dr. Thomas Curry for providing tissues and also for many useful discussions of estrous cycle physiology.
REFERENCES 1. Hirshfield AN. Development of follicles in the mammalian ovary. Int Rev Cytol 1991; 124:43–101. 2. Zeleznik AJ, Benyo DF. Control of follicular development, corpus luteum function, and the recognition of pregnancy in higher primates. In: Knobil E, Neill J (eds.), The Physiology of Reproduction. New York: Raven Press; 1994: 751–782. 3. McGee EA, Hsueh AJ. Initial and cyclic recruitment of ovarian follicles. Endocr Rev 2000; 21:200–214. 4. Adashi EY. With a little help from my friends—the evolving story of intraovarian regulation. Endocrinology 1995; 136:4161–4162. 5. Findlay JK, Drummond AE, Britt KL, Dyson M, Wreford NG, Robertson DM, Groome NP, Jones ME, Simpson ER. The roles of activins, inhibins and estrogen in early committed follicles. Mol Cell Endocrinol 2000; 163:81–87. 6. Elvin JA, Yan C, Matzuk M. Oocyte-expressed TGF-beta superfamily members in female fertility. Mol Cell Endocrinol 2000; 159:1–5.
1577
7. Adashi EY, Resnick CE, Hernandez ER, May JV, Purchio AF, Twardzik DR. Ovarian transforming growth factor-beta (TGF beta): cellular site(s), and mechanism(s) of action. Mol Cell Endocrinol 1989; 61: 247–256. 8. Minegishi T, Kishi H, Tano M, Kameda T, Hiakawa T, Miyamoto K. Control of FSH receptor mRNA expression in rat granulosa cells by 39,59-cyclic adenosine monophosphate, activin, and follistatin. Mol Cell Endocrinol 1999; 149:71–77. 9. Dodson WC, Schomberg DW. The effect of transforming growth factor-beta on follicle stimulating hormone-induced differentiation of cultured rat granulosa cells. Endocrinology 1987; 120:512–516. 10. Roy SK, Kole AR. Ovarian transforming growth factor-beta (TGFbeta) receptors: in-vivo effects of follicle stimulating hormone, epidermal growth factor and TGF-beta on receptor expression in human preantral follicles. Mol Hum Reprod 1998; 4:207–214. 11. Attisano L, Wrana JL. Mads and Smads in TGF beta signaling. Curr Opin Cell Biol 1998; 10:188–194. 12. Feng XH, Derynck R. A kinase subdomain of transforming growth factor-beta (TGF-beta) type I receptor determines the TGF-beta intracellular signaling specificity. EMBO J 1997; 16:3912–3923. 13. Savage C, Das P, Finelli AL, Townsend SR, Sun C-Y, Baird SE, Padgett RW. Caenorhabditis elegans genes sma-2, sma-3, and sma4 define a conserved family of transforming growth factor beta pathway components. Proc Natl Acad Sci U S A 1996; 93:790– 794. 14. Sekelsky JJ, Newfeld SJ, Raftery LA, Chartoff EH, Gelbart WM. Genetic characterization and cloning of mothers against dpp, a gene required for decapentaplegic function in Drosophila melanogaster. Genetics 1995; 139:1347–1358. 15. Derynck R, Gelbart WM, Harland RM, Heldin CH, Kern SE, Massague J, Melton DA, Mlodzik M, Padgett RW, Roberts AB, Smith J, Thomsen GH, Vogelstein B, Wang XF. Nomenclature: vertebrate mediators of TGF beta family signals. Cell 1996; 87:173. 16. Piek E, Heldin C-H, Ten Dijke P. Specificity, diversity, and regulation in TGFb superfamily signaling. FASEB J 1999; 13:2105–2124. 17. Wrana JL. Crossing Smads. Science’s STKE. Review 2000; 1:1–6. 18. McGee EA, Spears N, Smith R, Nachtigal M, Ingraham H, Hsueh AW. Mullerian inhibitory substance induces growth of rat preantral ovarian follicles. Biol Reprod 2000; 64:293–298. 19. Derynck R, Zhang Y, Feng XH. Smads: transcriptional activators of TGF beta responses. Cell 1998; 95:737–740. 20. Hsu SY, Kaipia A, McGee E, Lomeli M, Hsueh AJ. Bok is a proapoptotic Bcl-2 protein with restricted expression in reproductive tissues and heterodimerizes with selective anti-apoptotic Bcl-2 family members. Proc Natl Acad Sci U S A 1997; 94:12401– 12406. 21. Campbell KL. Ovarian granulosa cells isolated with EGTA and hypertonic sucrose: cellular integrity and function. Biol Reprod 1979; 21:773–786. 22. Carnegie JA, Byard R, Dardick I, Tsang BK. Culture of granulosa cells in collagen gels: the influence of cell shape on steroidogenesis. Biol Reprod 1988; 38:881–890. 23. Liu XW, Andoh K, Abe Y, Kobayashi J, Yamaka K, Mizunuma H, Ibuki Y. A comparative study on TGFb and activin for preantral follicles from adult, immature, and DES-primed immature mice. Endocrinology 1999; 140:2480–2485. 24. Smitz J, Cortvrindt R, Hu Y, Vanderstichele H. Effects of recombinant activin A on in vitro culture of mouse preantral follicles. Mol Reprod Dev 1998; 50:294–304. 25. Mizunuma H, Liu X, Andoh K, Abe Y, Kobayashi J, Yamada K, Yokota H, Ibuki Y, Hasegawa Y. Activin from secondary follicles causes small preantral follicles to remain dormant at the resting stage. Endocrinology 1999; 140:37–42. 26. Li R, Phillips DM, Mather JP. Activin promotes ovarian follicle development in vitro. Endocrinology 1995; 136:849–856. 27. Carr BR, McGee EA, Sawetawan C, Clyne CD, Rainey WE. The effect of transforming growth factor-beta on steroidogenesis and expression of key steroidogenic enzymes with a human ovarian thecal-like tumor cell. Am J Obstet Gynecol 1996; 174:1109– 1116. 28. Sawetawan C, Carr BR, McGee E, Bird IM, Hong TL, Rainey WE. Inhibin and activin differentially regulate androgen production and 17 alpha-hydroxylase expression in human ovarian thecal-like tumor cells. J Endocrinol 1996; 148:312–321. 29. Butcher RL, Collins WE, Fugo NW. Plasma concentration of LH, FSH, prolactin, progesterone and estradiol-17b throughout the 4-day estrous cycle of the rat. Endocrinology 1974; 94:1704–1708.
1578
XU ET AL.
30. Smith MS, Freeman ME, Neill JD. The control of progesterone secretion during the estrous cycle and early pseudopregnancy in the rat: prolactin, gonadotropin and steroid levels associated with rescue of the corpus luteum of pseudopregnancy. Endocrinology 1975; 96:219– 226. 31. Osterlund C, Fried G. TGFb receptor types I and II and the substrate proteins Smad2 and Smad3 are present in human oocytes. Mol Hum Reprod 2000; 6:498–503. 32. Bernard DJ, Chapman SC, Woodruff TK. An emerging role for coreceptors in inhibin signal transduction. Mol Cell Endocrinol 2001; 180:55–62. 33. Schilling B, Yeh J. Expression of transforming growth factor (TGF)beta1, TGF-beta2, and TGF-beta3 and of type I and II TGF-beta re-
ceptors during the development of the human fetal ovary. Fertil Steril 1999; 72:147–153. 34. Roy SK, Kole AR. Ovarian transforming growth factor-beta (TGFbeta) receptors: in-vitro effects of follicle stimulating hormone, epidermal growth factor and TGF-beta on receptor expression in human preantral follicles. Mol Hum Reprod 1998; 4:207–214. 35. Roy SK, Kole AR. Transforming growth factor-beta receptor type II expression in the hamster ovary: cellular site(s), biochemical properties, and hormonal regulation. Endocrinology 1995; 136: 4610–4620. 36. Findlay JK, Drummond AE, Dyson M, Baillie AJ, Robertson DM, Ethier J. Production and actions of inhibin and activin during folliculogenesis in the rat. Mol Cell Endocrinol 2001; 180:139–144.
