help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2003-0414
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/9/2224    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tomic, D.
Right arrow Articles by Flaws, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tomic, D.
Right arrow Articles by Flaws, J. A.
Molecular Endocrinology 18 (9): 2224-2240
Copyright © 2004 by The Endocrine Society

Ovarian Follicle Development Requires Smad3

Dragana Tomic, Kimberly P. Miller, Hilary A. Kenny, Teresa K. Woodruff, Patricia Hoyer and Jodi A. Flaws

Department of Epidemiology and Preventive Medicine (D.T., K.P.M., J.A.F.), University of Maryland School of Medicine, Baltimore, Maryland 21201; Department of Neurobiology and Physiology (H.A.K., T.K.W.), Northwestern University, Evanston, Illinois 60208; Robert H. Lurie Comprehensive Cancer Center (T.K.W.), Northwestern University, Chicago, Illinois 60611; and Department of Physiology (P.H.), University of Arizona, Tucson, Arizona 85724

Address all correspondence and requests for reprints to: Jodi A. Flaws, Ph.D., Department of Epidemiology and Preventive Medicine, 660 West Redwood Street, Howard Hall 133B, Baltimore, Maryland 21201. E-mail: jflaws{at}epi.umaryland.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Smad3 is an important mediator of the TGFß signaling pathway. Interestingly, Smad3-deficient (Smad3–/–) mice have reduced fertility compared with wild-type (WT) mice. To better understand the molecular mechanisms underlying the reduced fertility in Smad3–/– animals, this work tested the hypothesis that Smad3 deficiency interferes with three critical aspects of folliculogenesis: growth, atresia, and differentiation. Growth was assessed by comparing the size of follicles, expression of proliferating cell nuclear antigen, and expression of cell cycle genes in Smad3–/– and WT mice. Atresia was assessed by comparing the incidence of atresia and expression of bcl-2 genes involved in cell death and cell survival in Smad3–/– and WT mice. Differentiation was assessed by comparing the expression of FSH receptor (FSHR), estrogen receptor (ER) {alpha}, ERß, and inhibin {alpha}-, ßA-, and ßB-subunits in Smad3–/– and WT mice. Because growth, atresia, and differentiation are regulated by hormones, estradiol, FSH, and LH levels were compared in Smad3–/– and WT mice. Moreover, because alterations in folliculogenesis can affect the ability of mice to ovulate, the number of corpora lutea and ovulated eggs in response to gonadotropin treatments were compared in Smad3–/– and WT animals. The results indicate that Smad3 deficiency slows follicle growth, which is characterized by small follicle diameters, low levels of proliferating cell nuclear antigen, and low expression of cell cycle genes (cyclin-dependent kinase 4 and cyclin D2). Smad3 deficiency also causes atretic follicles, degenerated oocytes, and low expression of bcl-2. Furthermore, Smad3 deficiency affects follicular differentiation as evidenced by decreased expression of ERß, increased expression of ER{alpha}, and decreased expression of inhibin {alpha}-subunits. Smad3 deficiency causes low estradiol and high FSH levels. Finally, Smad3–/– ovaries have no corpora lutea, and they do not ovulate after ovulatory induction with exogenous gonadotropins. Collectively, these data provide the first evidence that reduced fertility in Smad3–/– mice is due to impaired folliculogenesis, associated with altered expression of genes that control cell cycle progression, cell survival, and cell differentiation. The findings that Smad3–/– follicles have impaired growth, increased atresia, and altered differentiation in the presence of high FSH levels, normal expression of FSHR, and lower expression of cyclin D2, suggest a possible interaction between Smad3 and FSH signaling downstream of FSHR in the mouse ovary.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
FEMALE MAMMALS ARE born with a finite reserve of primordial follicles in the ovary (1). Throughout the fertile life span, primordial follicles continuously grow into primary follicles, preantral, and antral follicles. Only antral follicles are capable of releasing oocytes at a stage when the oocyte can be fertilized (1). Thus, a continuous growth of primordial follicles to the antral stage is essential for female fertility. The process of follicular growth is controlled by both extraovarian and intraovarian factors, and the importance of each of these factors depends on the developmental stage of the follicle (1). Extraovarian factors regulate antral and preovulatory follicle growth, whereas intraovarian factors regulate preantral and early antral growth (1, 2).

One group of growth factors that has been implicated in many different aspects of follicle development is the TGFß family (3). The TGFß family consists of more than 35 members in vertebrates, and several TGFß family members are expressed in the ovary (3, 4). Growth differentiation factor (GDF)-9, bone morphogenetic protein (BMP)-6, and BMP-15 are expressed in oocytes, whereas other TGFß superfamily members are expressed in granulosa cells (inhibins, activins, TGFß1, TGFß2, and TGFß3) and theca cells (BMP-4 and BMP-7) within the ovary (3, 4). Interestingly, these factors have very different roles in regulating female fertility (3). Mice lacking GDF-9 are infertile due to a block in folliculogenesis at the primary follicular stage (5). BMP-15 null females are subfertile and demonstrate decreased ovulation and fertilization rates, but have minimal ovarian histopathological defects (6). Inhibin {alpha}-deficient (inhibin {alpha}–/–) mice have few follicles beyond the preantral stage, and they develop ovarian sex-cord stromal tumors during postnatal life (7). Activin ßB-deficient females have reproductive abnormalities including large litters, delays in delivery, and failure to nurse their offspring (8, 9). TGFß1-deficient mice (10) and TGFß-receptor type 2-deficient mice (11) die during embryogenesis, whereas mice lacking TGFß2 (12), and TGFß3 (13) die perinatally.

Recently, the Smad proteins, discovered through genetic studies in Drosophila (14) and Caenorhabditis elegans (15), have been shown to mediate the TGFß signaling pathway through a cascade of ligand-induced phosphorylation (3, 16, 17, 18, 19, 20). Downstream Smad signaling proteins might be the essential determinants of the different actions of TGFß superfamily members, but little is known about their function in the mammalian ovary. Smad3 is a receptor regulated Smad, and it is highly expressed in the ovarian surface epithelium, granulosa cells, and oocytes of several animal models (21, 22, 23, 24). Smad3 and its closely related homolog Smad2, mediate intracellular signaling pathways from TGFß and activin, each of which has been implicated as an important factor in ovarian development and function (3, 4). Evidence is also growing to support the hypothesis that cross talk occurs between many different signaling pathways in the ovary. Thus, to identify selective targets of Smad3 signaling pathways in vivo, this work investigated the role of Smad3 in ovarian development and function using mice with a targeted deletion of Smad3 gene in exon 8.

Female mice with a homozygous deletion of Smad3 gene in exon 8 (Smad3–/–) have reduced fertility compared with wild-type (WT) mice (24). These data differ from the data obtained from mice with a targeted disruption of Smad3 in exon 2, because adult homozygous mutant mice with disruption of Smad3 in exon 2 were fertile and produced homozygous litters (25). The reason that Smad3 mutant mice harboring a targeted disruption in exon 2 have a distinct reproductive phenotype from those harboring a disruption in exon 8 is unknown. The difference may stem from the fact that deletion of Smad3 in exon 8 by homologous recombination truncates 89 amino acids from the C-terminal region, a portion that contains an essential site for interacting with TGFß receptors (26), whereas deletion of Smad3 in exon 2 prevents the production of the C-terminal active domain of Smad3 (25).

Reduced fertility observed in mice with a homozygous deletion of Smad3 gene in exon 8 is due to impaired follicular development during adult life because ovaries isolated from Smad3–/– mice on postnatal days (PDs) 7–90 have an increased number of primordial and a decreased number of antral follicles compared with WT animals (24). To better understand the molecular mechanisms involved in impaired follicle development, studies were conducted to test the hypothesis that Smad3 deficiency causes diminished follicular growth, increased atresia and altered follicular differentiation leading to infertility. First, the effect of Smad3 deletion on follicular growth was assessed by morphologically examining the size of follicles, investigating granulosa cell proliferation via immunostaining with proliferating cell nuclear antigen (PCNA) and investigating the expression of genes involved in cell cycle progression [cyclin-dependent kinase 4 (cdk4) and cyclin D2] in the ovary. Second, the effect of Smad3 deletion on follicular atresia was assessed by examining the incidence of follicular atresia, investigating granulosa cell apoptosis using TUNEL (terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling) assays, and investigating the expression of genes involved in cell death (bax, bad, bid, and bok) and cell survival (bcl-2, mcl-1, bcl-xl) in the ovary. Third, the effect of Smad3 deletion on follicular differentiation was assessed by examining the expression of FSH receptor (FSHR), estrogen receptor (ER) {alpha}, ERß, and inhibin {alpha}-, ßA-, and ßB-subunits in the mouse ovary. In addition, because hormones may regulate many of these factors, levels of estradiol, FSH, and LH were measured and compared in Smad3–/– and WT animals. Finally, because alterations in follicular growth, atresia and differentiation can affect the ability of Smad3–/– mice to ovulate, the number of corpora lutea and the number of ovulated eggs in response to gonadotropin treatments were compared in Smad3–/– and WT animals. This approach has led to more insight about how the Smad3 signaling pathway in the ovary normally interfaces with cell cycle machinery, the Bcl-2 family of protooncogenes, ERs, and inhibin {alpha}-, ßA-, and ßB-subunits to regulate cell proliferation, cell death, and differentiation.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Effect of Smad3 Deletion on Ovarian Follicle Size and Morphology
To further extend previous morphological descriptions of the Smad3–/– ovary and provide evidence about whether the decreased number of preantral and antral follicles observed in Smad3–/– ovaries vs. WT ovaries during postnatal life (24) was due to a decreased number of follicles that grow to preantral and antral stages, Smad3–/– and WT ovaries were subjected to morphological assessment of follicle size over time. Smad3–/– ovaries contained preantral and antral follicles that were 10% and 20% smaller in diameter, respectively, compared with WT follicles on PD 18 (Fig. 1Go; n = 6–8; P ≤ 0.01 for preantral; P ≤ 0.001 for antral), and 20% and 22% smaller in diameter, respectively, than WT follicles on PD 90 (Fig. 1Go; n = 6–8; P ≤ 0.004 for preantral; P ≤ 0.03 for antral). To determine whether differences in follicle size were due to differences in proliferation of granulosa cells in WT and Smad3–/– follicles, the effect of Smad3 deletion on granulosa cell proliferation was examined via immunostaining for PCNA. PCNA, a cofactor of DNA polymerase {delta}, is expressed during G1, increases during G1/S transition, is high in G2, declines in M phase, and is not expressed during the Go phase of the cell cycle (27). Similar levels of PCNA proteins were previously observed in the granulosa cells of ovaries isolated from WT and Smad3–/– mice on PD 18 (24). On PD 30, however, levels of PCNA differed between WT and Smad3–/– mice. Positive PCNA staining was observed in the granulosa cells of both Smad3–/– and WT ovaries (Fig. 2Go), but Smad3–/– ovaries had less staining for PCNA protein (Fig. 2Go, D–F) compared with WT ovaries (Fig. 2Go, A–C). The most highly proliferative granulosa cells were detected in the antral follicles of WT animals, where 50–80% of granulosa cells were PCNA positive (Fig. 2Go, A–C). In contrast, antral follicles in Smad3–/– ovaries demonstrated 10–20% positively stained granulosa cells per cross section (Fig. 2Go, D–F), suggesting that nearly all of the granulosa cells in Smad3–/– ovaries are blocked at Go without progression to G1 phase of the cell cycle on PD 30.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1. Effect of Smad3 Deletion on Ovarian Follicle Size

Ovaries were collected from WT and Smad3–/– mice on PD 18 and PD 90 and complete serial sections were prepared for histological evaluation of follicle diameter as described in Materials and Methods. Statistically significant differences were assessed by ANOVA followed by a Scheffe post hoc test. Each bar represents the mean ± SEM (n = 6–8 ovaries per genotype at each time point). Asterisks indicate statistically significant differences between genotypes within the same age group.

 


View larger version (154K):
[in this window]
[in a new window]
  		Fig. 2. Effect of Smad3 Deletion on Proliferation and Apoptosis in Ovarian Follicles

Ovaries were collected from WT and Smad3–/– mice and subjected to immunohistochemistry for PCNA or measurement of apoptosis using TUNEL assays as described in Materials and Methods. A–C, Representative micrographs of ovaries collected on PD 30 from WT mice and stained with anti-PCNA monoclonal antibody (A, original magnification x100; B and C, original magnification x400; red staining = PCNA staining; blue staining = counterstain). D–F, Representative micrographs of ovaries collected on PD 30 from Smad3–/– mice and stained with anti-PCNA monoclonal antibody (D, original magnification, x100; E and F, original magnification, x400; red staining = PCNA staining, blue staining = counterstain). G–I, Representative micrographs of ovaries collected on PD 30 from WT mice and subjected to apoptosis detection using TUNEL assays (G–I, original magnification, x400; dark staining = apoptotic cells; blue staining = counterstain). J–L, Representative micrographs of ovaries collected on PD 30 from Smad3–/– mice and subjected to apoptosis detection using TUNEL assays (J–L, original magnification, x400; dark staining = apoptotic cells, blue staining = counterstain).

 
Characterization of the Expression of Cell Cycle Regulators in Smad3–/– Follicles
To further investigate the mechanism involved in the diminished follicle growth observed in Smad3-deficient mice, the expression of genes that control cell cycle progression (cdk4 and cyclin D2) were examined and compared in WT and Smad3–/– follicles. Cyclin D2 acts as a positive regulator of cell cycle progression by its ability to bind cdk4 and cdk6, thereby activating a cascade of events, which permits progression through the G1 phase of the cell cycle (28). A significantly lower expression of cdk4 and cyclin D2 was detected in preantral and antral follicles isolated from Smad3–/– ovaries compared with WT follicles on PD 90 (Fig. 3AGo). Specifically, real-time PCR analysis indicated that Smad3–/– follicles demonstrated a 50% and 40% reduction in cdk4 and cyclin D2, respectively, compared with WT follicles (Fig. 3BGo).



View larger version (10K):
[in this window]
[in a new window]
  		Fig. 3. Effect of Smad3 Deletion on Cdk4 and Cyclin D2 Levels in Isolated Antral Ovarian Follicles

RNA was isolated from antral follicles collected from WT and Smad3–/– ovaries as described in Materials and Methods. Cdk4 and cyclin D2 mRNA expression were analyzed by RT-PCR and real-time PCR using specific primers and ß-actin was used as an internal control. Data were collected from three independent experiments. A, Expression of cdk4 and cyclin D2 mRNA in WT and Smad3–/– ovaries as detected by RT-PCR. B, Real-time PCR analysis of relative levels of cdk4 and cyclin D2 mRNA in Smad3–/– ovaries, expressed as percentage of control (WT ovaries) after normalization to ß-actin levels. Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 8–10 mice per genotype, 5–7 follicles/ovary). Asterisks indicate statistically significant differences between genotypes.

 
Effect of Smad3 Deletion on Ovarian Follicular Atresia and Apoptosis
To determine whether the decreased number of preantral and antral follicles previously detected in Smad3–/– ovaries on PDs 18 and 90 was due to an increased incidence of atresia in Smad3–/– ovaries vs. WT ovaries, Smad3–/– and WT ovaries were subjected to morphometric assessment of the percentage of atretic follicles over time. On PD 18, 5–7% of all the follicles were atretic in both WT and Smad3–/– ovaries (Fig. 4AGo; n = 4–6; P = 0.7). By PD 90, however, Smad3–/– ovaries contained a significantly higher percentage of atretic follicles compared with WT ovaries (Fig. 4AGo; n = 5–7; P ≤ 0.03). Approximately 24% of the preantral and antral follicles were atretic in Smad3–/– ovaries, whereas only 7% of the follicles were atretic in WT ovaries. In addition, Smad3–/– ovaries on PD 90 contained numerous degenerated oocytes (i.e. fragmented oocytes), which are the remnants of atretic follicles. The number of atretic oocytes in Smad3–/– mice was 2-fold higher than in WT mice on PD 90 (Fig. 4BGo). Because follicular atresia is mediated via apoptosis, the effect of Smad3 deletion on granulosa cell apoptosis was further examined using TUNEL assays. Granulosa cells of Smad3–/– antral follicles contained more than 60% positively labeled cells per ovarian section (Fig. 2Go, J–K), whereas WT ovaries contained less than 10% positively labeled cells per ovarian section (Fig. 2Go, G and H). Thus, in the absence of Smad3, the majority of granulosa cells appear to undergo apoptosis.



View larger version (9K):
[in this window]
[in a new window]
  		Fig. 4. Effect of Smad3 Deletion on Follicular Atresia

Ovaries were collected from WT and Smad3–/– mice on PD 18 and PD 90 and complete serial sections were prepared for histological examination of healthy and atretic follicles as described in Materials and Methods. A, Effect of Smad3 deletion on the percentage of atretic follicles in WT and Smad3–/– ovaries. B, Effect of Smad3 deletion on the number of atretic oocytes in WT and Smad3–/– ovaries. A, Statistically significant differences were assessed by ANOVA followed by a Scheffe post hoc test. B, Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 4–7 ovaries per genotype at each time point). Asterisk indicates a statistically significant difference between genotypes.

 
Effect of Smad3 Deletion on the Expression of Bcl-2 Family Members in the Ovary
To further investigate the mechanism involved in the increased incidence of follicle atresia observed in Smad3-deficient mice, the expression of Bcl-2 family members in WT and Smad3–/– ovaries was compared using RT-PCR and real-time PCR. A significantly lower expression of bcl-2 (an antiapoptotic factor) was detected in Smad3–/– ovaries compared with WT ovaries (Fig. 5AGo). Specifically, real-time PCR analysis indicated that Smad3–/– follicles demonstrated a 50% reduction in bcl-2 mRNA levels, compared with WT follicles (Fig. 5BGo). In contrast, similar levels of mcl-1 and bcl-xl (antiapoptotic factors) and bax, bad, bid, and bok (proapoptotic factors) were observed in ovaries from Smad3–/– and WT mice (Fig. 5AGo).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Effect of Smad3 Deletion on the Expression of Bcl-2 Family Members

RNA was isolated from ovaries collected from WT and Smad3–/– mice as described in Materials and Methods. Expression of bcl-2 family members was analyzed by RT-PCR and real-time PCR using specific primers and ß-actin was used as an internal control. Data were collected from three independent experiments. A, Expression of bcl-2, mcl-1, bcl-xl (antiapoptotic proteins), bax, bad, bid, and bok (proapoptotic proteins) mRNA in Smad3–/– and WT ovaries as detected by RT-PCR. B, Real-time PCR analysis of relative levels of bcl-2 mRNA in Smad3–/– ovaries, expressed as percentage of control (WT ovaries) after normalization to ß-actin levels. Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 9–12 mice per genotype). Asterisk indicates a statistically significant difference between genotypes.

 
Expression of FSHR, ER{alpha}, and ERß in the Smad3–/– Ovary
To investigate whether Smad3 deletion affects follicular differentiation in the ovary, the expression of markers associated with the functional differentiation of granulosa cells of antral follicles including FSHR, ERß, and ER{alpha} was examined. Smad3–/– ovaries had a significantly lower expression of ERß mRNA and ERß protein (Fig. 6Go), significantly higher expression of ER{alpha} mRNA and higher expression of ER{alpha} protein (Fig. 6Go) and similar expression of FSHR mRNA and FSHR protein (Fig. 6Go) compared with WT ovaries. To further investigate whether the altered expression of receptors in the whole ovary was due to altered expression of receptors in ovarian follicles, individual antral follicles were isolated from Smad3–/– and WT ovaries and subjected to RT-PCR and real-time PCR analysis for FSHR, ERß, and ER{alpha}. Isolated antral follicles from Smad3–/– ovaries had significantly higher expression of ER{alpha} mRNA and similar expression of ERß mRNA and FSHR mRNA compared with follicles isolated from WT ovaries (Fig. 7Go, A and B).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 6. Effect of Smad3 Deletion on FSHR, ER{alpha}, and ERß Levels in the Ovary

RNA and protein lysate were isolated from ovaries collected from WT and Smad3–/– mice as described in Materials and Methods. FSHR, ER{alpha}, and ERß mRNA expression were analyzed by RT-PCR and real-time PCR using specific primers for FSHR, ER{alpha}, and ERß. Protein levels were analyzed by Western blot analysis using specific antibodies. ß-Actin was used as an internal control. Data were collected from three independent experiments. A, Expression of FSHR, ER{alpha}, and ERß mRNA in WT and Smad3–/– ovaries as detected by RT-PCR. B, Real-time PCR analysis of relative levels of FSHR, ER{alpha}, and ERß mRNA in Smad3–/– ovaries, expressed as percentage of control (WT ovaries) after normalization to ß-actin. C, Protein levels of FSHR, ER{alpha}, and ERß in WT and Smad3–/– ovaries as detected by Western blot analysis. D, Quantification of the protein levels for each protein in Smad3–/– ovaries, expressed as percentage of control (WT ovaries) after normalization to ß-actin levels. Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 9–12 mice per genotype). Asterisks indicate statistically significant difference between genotypes.

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 7. Effect of Smad3 Deletion on FSHR, ER{alpha}, and ERß Levels in Isolated Antral Ovarian Follicles

RNA was collected from antral follicles isolated from WT and Smad3–/– ovaries as described in Materials and Methods. FSHR, ER{alpha}, and ERß mRNA expression were analyzed by RT-PCR and real-time PCR using specific primers and analyzed relative to a ß-actin internal control. Data were collected from three independent experiments. A, Expression of FSHR, ER{alpha}, and ERß mRNA in WT and Smad3–/– follicles as detected by RT-PCR. B, Real-time PCR analysis of relative levels of FSHR, ER{alpha}, and ERß mRNA in Smad3–/– follicles, expressed as percentage of control (WT follicles) after normalization to ß-actin. Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 8–10 mice per genotype, 5–7 follicles/ovary). The asterisk indicates a statistically significant difference between genotypes.

 
Serum Estradiol, FSH, and LH Concentrations in Smad3–/– Mice
Follicular growth, atresia, and differentiation are often regulated by hormones (29, 30, 31). Thus, the levels of estradiol, FSH and LH hormones were compared in Smad3–/– and WT mice. Estradiol levels were significantly lower in Smad3–/– mice compared with WT animals on PD 18 (WT = 245 ± 18 pg/ml, n = 7; Smad3–/– = 167 ± 18 pg/ml, n = 4; P ≤ 0.02), PD 30 (WT = 289.1 ± 16.8 pg/ml, n = 9; Smad3–/– = 174.5 ± 37.8 pg/ml, n = 7, P ≤ 0.009), and PD 90 (WT = 212.9 ± 27.4 pg/ml, n = 8; Smad3–/– = 111.2 ± 17.3 pg/ml, n = 4, P ≤ 0.01) (Fig. 8AGo). FSH levels were significantly higher in Smad3–/– mice compared with WT animals on PD 18 (WT = 3.73 ± 0.63 ng/ml, n = 15; Smad3–/– = 18 ± 5.36 ng/ml, n = 8, P ≤ 0.002), PD 30 (WT = 3.1 ± 0.77 ng/ml, n = 9; Smad3–/– = 6.7 ± 0.63 ng/ml, n = 5, P ≤ 0.009), and PD 90 (WT = 6.1 ± 2 ng/ml, n = 6; Smad3–/– = 14.1 ± 2 ng/ml, n = 5, P ≤ 0.03) (Fig. 8BGo). In contrast, there were no significant differences between the levels of LH in WT and Smad3–/– mice on PD 30 (WT = 1.09 ± 0.06 ng/ml, n = 8; Smad3–/– = 0.99 ± 0.11 ng/ml, n = 7, P = 0.408) or PD 90 (WT = 0.98 ± 0.14 ng/ml, n = 7; Smad3–/– = 1.21 ± 0.19 ng/ml, n = 5, P = 0.352) (Fig. 8CGo).



View larger version (14K):
[in this window]
[in a new window]
  		Fig. 8. Serum Estradiol, FSH, and LH Concentrations in WT and Smad3–/– Mice

Blood was collected from WT and Smad3–/– mice on PDs 18, 30, and 90 and subjected to assays for measurement of estradiol, FSH, and LH levels as described in Materials and Methods. Statistically significant differences were assessed by ANOVA followed by a Scheffe post hoc test. A, Serum estradiol concentrations in WT and Smad3–/– mice. Each bar represents the mean ± SEM (n = 7 for WT and n = 4 for Smad3–/– on PD 18; n = 9 for WT and n = 7 for Smad3–/– on PD 30; n = 8 for WT and n = 4 for Smad3–/– on PD 90). B, Represents serum FSH concentrations in WT and Smad3–/– mice. Each bar represents the mean ± SEM (n = 15 for WT and n = 8 for Smad3–/– on PD 18; n = 9 for WT and n = 5 for Smad3–/– on PD 30; n = 6 for WT and n = 5 for Smad3–/– on PD 90). C, Represents serum LH concentrations in WT and Smad3–/– mice. Each bar represents the mean ± SEM (n = 8 for WT and n = 7 for Smad3–/– on PD 30; n = 7 for WT and n = 5 for Smad3–/– on PD 90). Asterisks indicate statistically significant differences between genotypes within the same age group.

 
Expression of Inhibin {alpha}, ßA, and ßB-Subunits in Smad3–/– and WT Ovaries
Inhibins are key regulators of granulosa cell proliferation and follicle differentiation both in vivo and in vitro (32, 33). Thus, production of the inhibin subunits was investigated in Smad 3–/– ovaries by protein blot analyses (Fig. 9Go). Smad3–/– ovaries had significantly lower levels of mature inhibin {alpha}-subunit than WT ovaries (Fig. 9BGo). Smad3–/– ovaries had similar levels of mature ßA-subunit (Fig. 9DGo) and mature ßB-subunit (Fig. 9FGo) compared with WT ovaries.



View larger version (38K):
[in this window]
[in a new window]
  		Fig. 9. Western Blot Analysis of Inhibin {alpha}-, ßA-, and ßB-Subunits in Ovarian Lysates from WT and Smad3–/– Mice

Proteins were extracted from individual ovaries (n = 9–11) and 30 µg protein were loaded in each well, under reducing conditions as described in Materials and Methods. Data were collected from three independent experiments. A, Immunoblot analysis of ovarian extracts using the {alpha}-subunit polyclonal antibody. One mature form ({alpha}C) and precursor forms ({alpha}N{alpha}C and arrowhead) of the {alpha}-subunit were detected. The standard, rh inhibin A, was detected in its nongylcosylated and diglycosylated mature {alpha}-subunit form ({alpha}C) and precursor proteins ({alpha}N{alpha}C and arrowhead). B, Average protein expression of mature {alpha}-subunit form in WT and Smad3–/– ovaries. C, Immunoblot analysis of ovarian extracts using the ßA-subunit polyclonal antibody. The mature (ßA) and precursor ßA-subunit were detected in ovaries. The standard, rh activin A conditioned media, was detected as mature (ßA) and precursor ßA-subunit. D, Average protein expression of mature ßA-subunit form in WT and Smad3–/– ovaries. E, Immunoblot analysis of ovarian extracts using ßB-subunit polyclonal antibody. The mature (ßB) and precursor ßB-subunit were detected in ovaries. The standard, rh activin B conditioned media, was detected as mature (ßB) and precursor ßB-subunit. F, Average protein expression of mature ßB-subunit form in WT and Smad3–/– ovaries. Statistically significant differences were assessed by t tests. Each bar represents the mean ± SEM (n = 9–12). Arrows point to bands of interest. Asterisk indicates a statistically significant difference between genotypes.

 
Effect of Smad3 Deletion on Ovulation
To further analyze reproductive deficit of the Smad3–/– females, the ability of Smad3–/– mice to ovulate was compared with the ability of WT mice to ovulate. Ovaries were collected from WT and Smad3–/– mice on PD 90 and prepared for examination of the total number of corpora lutea as described in Materials and Methods. WT ovaries contained 6.8 ± 0.6 corpora lutea per ovary (n = 5). In contrast, no corpora lutea were observed in Smad3–/– ovaries (n = 5). To determine whether ovulation was possible in Smad3–/– females, the ability of Smad3–/– mice to ovulate in response to gonadotropin treatment was compared with that of WT animals. Ovarian stimulation with exogenous gonadotropins was successful in eliciting ovulation only in WT animals. After stimulation with pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hCG), each oviduct from WT mice had a localized swelling and the average number of oocytes collected from each WT animal was 28.8 ± 3.8 (n = 12). The Smad3–/– animals did not have an obvious localized swelling; therefore, the entire oviduct was carefully flushed under the direct microscopic visualization. No oocytes were obtained from Smad3–/– mice (n = 4). Histological evaluation of ovaries from the superovulated WT females indicated the presence of multiple corpora lutea. In contrast, the ovaries from stimulated Smad3–/– females contained no corpora lutea, few large antral follicles, and many atretic follicles.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Previous work demonstrated that deletion of Smad3 in exon 8 decreases female fertility (24). Moreover, previous work indicated that the decreased fertility in Smad3–/– mice was due in part to a decreased number of antral follicles in Smad3–/– mice compared with WT mice (24). Thus, to identify selective targets of Smad3 signaling pathways in vivo, the present study investigated the effect of Smad3 deletion on follicular growth, atresia, and differentiation in the mouse ovary (summarized in Fig. 10Go).



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 10. A Model for the Role of Smad3 in Ovarian Folliculogenesis

Based on the results of studies using Smad3–/– mice, Smad3 is required for three critical aspects of folliculogenesis: growth, atresia, and differentiation. Smad3 deficiency slows follicle growth, which is characterized by small follicle diameters, low expression of PCNA, and low expression of cell cycle genes (cdk4 and cyclin D2). Smad 3 deficiency increases follicular atresia, which is characterized by a high number of atretic follicles, degenerated oocyctes, and low expression of bcl-2. Furthermore, Smad3 deficiency affects follicular differentiation as evidenced by decreased expression of ERß in the whole ovary, increased expression of ER{alpha} in the whole ovary and antral follicles, and decreased expression of inhibin {alpha}-subunit in the whole ovary. Finally, Smad3 deficiency causes low estradiol and high FSH levels.

 
The results of this study indicate that the reduced number of antral follicles in Smad3–/– ovaries is likely due to diminished follicle growth and decreased granulosa cell proliferation. Decreased follicle growth and decreased granulosa cell proliferation may be due, in part, to altered expression of genes (cdk4 and cyclin D2) that regulate the G1 stage of cell cycle progression. This is in agreement with previous findings suggesting that TGFß may affect cellular proliferation by affecting the G1 stage of cell cycle progression (17, 34) and that decreased levels of cyclin D2 (35) and cdk4 (36, 37) are associated with impaired granulosa cell proliferation, differentiation and luteinization in the mouse ovary. The reason that similar levels of PCNA proteins in the granulosa cells of WT and Smad3–/– ovaries on PD 18 were observed in a previous study (24) and higher levels of PCNA staining in the granulosa cells of WT ovaries compared with granulosa cells of Smad3–/– ovaries were observed in this study is likely due to the differences in the ages of the animals in the two studies. It is likely that on PD 18, the phenotype in the Smad3–/– is less pronounced than on PD 30, and that this makes it difficult to observe differences in PCNA staining on PD 18 as compared with PD 30.

To further analyze molecular changes within the Smad3–/– ovary, the effects of Smad3 deletion on follicle atresia were examined. Follicular death in the ovary is tightly regulated by antiapoptotic (bcl-2, mcl-1, bcl-xl) and proapoptotic (bax, bad, bid, and bok) factors as well as cross talk among multiple intracellular pathways (30, 38, 39, 40, 41, 42). The results of this study suggest that Smad3 deletion increases the incidence of follicle atresia in the mouse ovary and that this increased atresia is associated with an altered expression of members of the bcl-2 family of protooncogenes. These data are consistent with a previous study, which demonstrated that Smad3–/– mice had higher levels of Bax protein and lower levels of Bcl-2 protein compared with WT mice (24). These findings are also in agreement with in vitro studies suggesting that TGFß regulates expression of bcl-2 in ovarian surface epithelial cells (43) and ovarian thecal/interstitial cells (44), and studies demonstrating that bcl-2 is an important cell survival factor in the ovary (45, 46). Because Smad3 deletion did not increase bax mRNA expression but did increase Bax protein levels (24), it is possible that Smad3 deletion interferes with posttranscriptional and/or posttranslational mechanisms related to bax levels.

To further characterize the follicles of Smad3deficient mice, the expression of markers (FSHR, ER{alpha}, and ERß) associated with functional differentiation of antral follicles was examined. These markers are critical for antral follicle formation and normal ovarian function because FSHR-deficient females are infertile due to a block in folliculogenesis before antral follicle formation (47), ERß-deficient mice have decreased fertility with a reduced number of both antral follicles and corpora lutea (48), and ER{alpha}-deficient mice are infertile due to arrested follicular development at the late antral stage (49). FSHR was equally expressed in Smad3–/– and WT ovaries, and there was no difference in the expression of FSHR between antral follicles collected from Smad3–/– and WT ovaries. Whereas there was no difference in total FSHR levels in follicles, it is still possible that the levels of FSHR differ in Smad3–/– and WT granulosa cells. Thus, further studies will address this issue. ERß was expressed at lower levels in Smad3–/– ovaries compared with WT ovaries, whereas there was no difference in the expression of ERß between antral follicles isolated from Smad3–/– and WT ovaries. ER{alpha} was highly expressed in whole Smad3–/– ovaries compared with WT ovaries, and there was higher expression of ER{alpha} in isolated Smad3–/– antral follicles compared with WT follicles. The reasons for up-regulation of ER{alpha} in ovaries and follicles isolated from Smad3–/– animals vs. ovaries and follicles isolated from WT animals are unclear. It is possible that alternative signaling routes exist via Smad3 in the ovarian follicles that ensure the function of key genes in the ER signaling pathway and in the absence of Smad3 these genes are inactive, altering expression of ER{alpha} in the ovary. This is in agreement with previous in vitro reports that Smad3 is an important regulator of ER signaling in human kidney carcinoma cells (50). Thus, further studies will include determining the relationship between ER signaling pathways and Smad3 protein in the mouse ovary.

Hormonal regulation of follicular growth, apoptosis, and differentiation is critical for normal ovarian development and function (29, 30, 31, 42). Smad3–/– mice have significantly lower levels of estradiol, significantly higher levels of FSH, and similar LH levels compared with WT animals. The finding of lower levels of estradiol observed in Smad3–/– mice is consistent with the observed presence of lower numbers of antral follicles (i.e. follicles responsible for estradiol production) in Smad3–/– ovaries compared with WT ovaries (24). Moreover, Smad3–/– antral follicles are smaller, due to a reduced number of granulosa cells, and thus may have a reduced capacity for aromatization of androgens to estrogens. The high FSH levels observed in Smad3–/– animals may be due to increased FSH production from the pituitary gland due to disruption of the Smad3 signaling pathway, increased FSH production from the pituitary gland due to negative feedback caused by low estradiol levels in Smad3–/– mice, and/or increased FSH production from the pituitary gland due to decreased levels of inhibin in Smad3–/– animals. Although there are currently no reports on the effect of Smad3 deletion on the production of FSH from the pituitary gland, the data presented here indicate that there are lower estradiol levels in Smad3–/– animals, and it is possible that the low estradiol levels increase FSH production from the pituitary gland through a negative feedback mechanism. Recent studies, however, suggest that FSH synthesis and secretion may be primarily regulated by the inhibin family of hormones and less dependent on estradiol feedback (51, 52). The inhibins are heterodimers of one activin ß-subunit disulfide-linked to a unique {alpha}-subunit to produce two isoforms of inhibin: inhibin A or inhibin B. Inhibins have been characterized as paracrine acting factors within the ovary, which modulate follicular growth and steroidogenesis (53, 54, 55). Inhibins are also endocrine hormones, acting on the pituitary gland to inhibit the secretion of FSH (56). Thus, it is possible that loss of inhibin {alpha}-subunits in Smad3–/– ovaries may contribute to increased FSH production and secretion from the pituitary.

Finally, because alterations in follicular growth, atresia, and differentiation can affect the ability of Smad3–/– mice to ovulate, the number of corpora lutea and ovulated eggs in response to gonadotropin treatment were compared in Smad3–/– and WT animals. Smad3–/– ovaries have no corpora lutea, and they do not ovulate after ovulatory induction with exogenous gonadotropins. A significant reduction of ovulatory capacity has been described in mice after targeted disruption of the genes for ER{alpha} (49), ERß (48), and cyclin D2 (35) suggesting that intraovarian action of ER or cyclin D2 may facilitate ovulation. Thus, it is possible that reduced ovulatory capacity observed in Smad3–/– animals may be partially due to the altered expression of ERs and cyclin D2 in the Smad3–/– mouse ovary.

The findings of impaired follicular growth, increased apoptosis, altered differentiation in the presence of high FSH levels, decreased estradiol levels, and failure of follicles to be ovulated in respond to exogenous gonadotropins suggest that a mutation compromising Smad3 in the mouse ovary significantly impairs female fertility. One model to explain this finding is that mouse Smad3 is required to facilitate TGFß and activin effects in the ovary. This model is based on evidence that Smad3 mediates TGFß and activin signaling pathways (57, 58). TGFß is expressed in follicular cells and has proliferative actions on granulosa cells (55). Moreover, TGFß increases the diameter of follicles in adult mice (59) and thus, the impaired follicular growth observed in the absence of Smad3 may be due to impaired TGFß signaling in the ovary. In addition, the gonadal defects observed in Smad3–/– mice are partially similar to those observed in mice lacking activin receptor type 2 (ACVR2) (60), suggesting that absence of activin signaling through ACVR2 partially mimics Smad3 loss of function. Female ACVR2 mice are infertile due to a block at the antral stage of ovarian folliculogenesis. Moreover, ovaries collected from ACVR2 mice were smaller compared with controls, rarely had corpora lutea, and follicle atresia was often seen (60). However, mice lacking ACVR2 had suppressed pituitary and serum FSH levels that resulted in gonadal growth defect (60), whereas Smad3–/– mice have impaired follicular growth and increased atresia in the presence of high FSH levels. This suggests that Smad3 deletion might cause phenotypic effects by interfering with signaling from different TGFß family members.

Evidence is also growing to support the hypothesis that different Smads may be required for different biological responses to TGFß members. Smad2 and Smad3 are structurally homologous and are assumed to mediate signals from an identical set of ligand and receptors (16, 17, 18). However, the phenotypes of Smad2 null and Smad3 null mice are quite distinct (61) and different genes are regulated by different Smads in vitro (62). Moreover, the finding that the Smad2/Smad3 null heterozygote is embryonically lethal, whereas the Smad3 null homozygote is not embryonically lethal, suggests that the two Smads are not functionally interchangeable. Although expression of Smad2 protein in Smad3–/– embryos and in the tissues of adult Smad3–/– and WT mice is identical (26), the issue of whether Smad3 and Smad2 show any functional redundancy in the ovary can be definitively addressed only when the Smad2 conditional null and Smad2/Smad3 conditional double null mice become available.

The present studies open up new avenues of research for understanding the intraovarian role of the TGFß signaling pathway. The results of this study provide further evidence that the deletion of Smad3 affects folliculogenesis by affecting the expression of genes that control cell cycle progression (cdk4 and cyclin D2), cell survival (bcl-2 family of protooncogenes), and cell differentiation (ER{alpha}, ERß, and inhibin {alpha}). The findings that Smad3–/– follicles have impaired follicular growth, increased apoptosis, and altered differentiation in the presence of high FSH levels and normal expression of FSHR in the ovary, suggest that Smad3 deletion affects FSH signaling pathways downstream of FSHR in the ovary. This hypothesis is supported by the observation that Smad3–/– follicles have lower expression of cyclin D2, an FSH-responsive gene (35), compared with WT follicles. Further studies will include determining the relationship between the FSH signaling pathway and Smad3 protein.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Animals
Smad3-deficient mice (Smad3–/–) were generated as described (26) and generously provided by Dr. Chuxia Deng (National Institute of Diabetes and Digestive and Kidney Disease, National Institutes of Health). Smad3–/–, Smad3+/–, and WT animals were housed in the University of Maryland School of Medicine Central Animal Facility and provided food and water ad libitum. The University of Maryland School of Medicine Institutional Animal Use and Care Committee approved all procedures involving animal care, euthanasia, and tissue collection.

Screening/Genotyping of Mice
Mice heterozygous for Smad3 (Smad3+/–) were mated and offspring were genotyped to determine whether they were Smad3–/–, Smad3+/–, or WT, as previously described (24). Briefly, DNA was extracted from 3-mm ear punches of tissue and subjected to PCR using the following primers: 1) 5'-CCACTTCATTGCCATATGCCCTG-3', 2) 5'-CCCGAACAGTTGGATTCACACA-3', and 3) 5'-CCAGACTGCCTTGGGAAAAGC-3'. The conditions for PCR were as follows: 94 C for 3 min followed by 30 cycles of 94 C for 30 sec, 60 C for 30 sec, 72 C for 90 sec, and final extension at 72 C for 10 min. PCR products were subjected to agarose gel electrophoresis and visualized under UV light. The presence of a 400-bp band indicated WT mice, the presence of a 250-bp band indicated Smad3–/– mice, whereas the presence of both bands indicated Smad3+/– mice.

Measurement of Follicular Growth
To determine whether follicular growth differed by genotype, ovaries were subjected to morphological measurements of follicular size over time. First, ovaries were collected from Smad3–/– and WT mice on PDs 18 and 90. PD 18 was selected because it is when follicular growth is beginning, but mice are still sexually immature, and PD 90 was selected because it is when mice are usually cycling adults (1). In addition, these days were chosen because a previous study showed that Smad3–/– mice had significantly higher numbers of primordial follicles and lower numbers of preantral and antral follicles compared with WT mice on PDs 18 and 90 (24).

Ovaries were fixed in Kahle’s solution (4% formalin, 28% ethanol, and 0.34 N glacial acetic acid), serially sectioned (8 µm), mounted on glass slides, and stained with Weigert’s hematoxylin-picric acid methyl blue. Every tenth section was marked for analysis. Follicles were classified as preantral if they contained an oocyte with a visible nucleolus, more than one layer and less than five layers of granulosa cells, and lacked an antral space. Follicles were classified as antral if they contained an oocyte with a visible nucleolus, more than five layers of granulosa cells, and/or an antral space (63). The diameters of all preantral and antral follicles in the marked sections were measured in two perpendicular axes using a calibrated micrometer on a light microscope at x25 as described (64). The mean of the two diameters was calculated to obtain the mean diameter per follicle, and then the mean of all diameters was calculated to estimate the total mean diameter of all preantral and antral follicles present in each ovary. To avoid double counting follicles and to help ensure that a consistent area of each follicle was measured, only follicles containing an oocyte with a visible nucleolus were measured. To avoid bias, all ovaries were analyzed without knowledge of genotype or age. The data are presented as mean diameter of preantral follicles by genotype and mean diameter of antral follicles by genotype.

Measurement of Proliferation in the Ovary
Ovaries were removed from WT and Smad3–/– mice, fixed in Kahle’s solution, dehydrated in gradient alcohols, cleared in xylene, and embedded in paraffin. Sections were cut at 5 µm and subjected to staining for PCNA using a commercially available monoclonal antibody (1:100 dilution; Oncogene Research Products, Boston, MA) as described previously (22, 24). The biotinylated streptavidin Histomouse-SP system (Zymed Laboratories, Inc., San Francisco, CA) with background blocking was used with 3-amino-9-ethyl carbazole chromogenic visualization and Mayer’s hematoxylin counterstaining. In all experiments, negative controls were run in parallel. To obtain an estimate of the percentage of proliferating cells, the percentage of nuclei positively stained for PCNA was estimated in at least three different sections per ovary using at least three different ovaries (one ovary per animal) per genotype as previously described (22, 65). Positive staining was defined as intense red staining of the nucleus and negative was defined as light or diffuse staining of the nucleus and cytoplasm (65). To avoid bias, all ovaries were analyzed without knowledge of genotype or age.

Measurements of Follicular Atresia
Atresia was compared in Smad3–/– and WT mice by morphometric assessment of the percentage of atretic follicles over time. Ovaries were harvested from Smad3–/– and WT mice on PD 18 and 90. The ovaries were fixed in Kahle’s fixative and processed for histological evaluation as described above. Healthy and atretic follicles were classified using strict morphological criteria and without knowledge of genotype or age. Follicles were classified as healthy if they contained an intact oocyte, organized granulosa cell layers, and few (<10%) pyknotic bodies. Atretic follicles were classified as those containing a degenerated oocyte, disorganized granulosa cell layers, and/or more than 10% of the granulosa cells were pyknotic in appearance. The percentage of atretic follicles was estimated by taking the total number of atretic preantral and antral follicles divided by the total number of preantral and antral follicles (both healthy and atretic) and multiplying by 100. Ovaries collected from WT and Smad3–/– mice contained degenerated oocytes, i.e. fragmented oocytes, which are the remnants of atretic follicles. No diameter could be estimated for these follicular remnants because the layer of granulosa cells could not be distinguished very well from the surrounding interstitial tissue. To prevent double counting of these follicle remnants (atretic oocytes), they were counted in every tenth section.

Detection of Apoptosis by TUNEL Assay
Histological tissue sections from both Smad3–/– and WT ovaries were stained for apoptotic DNA using an ApopTag Peroxidase kit (Intergen Co., Purchase, NY) as previously described (22). To obtain an estimate of the percentage of apoptotic cells, the percentage of nuclei positively stained for apoptotic DNA was estimated in at least three different sections per ovary using at least three different ovaries (one ovary per animal) per genotype. To avoid bias, all ovaries were analyzed without knowledge of genotype or age.

Hormone Assays
Blood samples were obtained by retro-orbital sinus sampling from Smad3–/– and WT mice on PDs 18, 30, and 90 and subjected to measurements of 17ß-estradiol (estradiol), FSH, and LH. Blood was collected on the same day of the estrous cycle (estrus) to minimize natural fluctuation in hormone levels. Estradiol assays were performed using an ELISA kit obtained from Diagnostic System Laboratories, Inc. (Webster, TX) as previously described (22). The supplied protocol was followed without modifications and all samples were run in duplicate in a single assay. The minimum detection limit, as stated in the instructions of the kit, was 7 pg/ml. The average intraassay coefficient of variation was 4.2% and the average interassay coefficient of variation was 8.2%. Plasma FSH and LH were measured by RIA using reagents from the National Hormone and Pituitary Distribution Program. Iodination reagents (IODO-BEADS 28665, 28666) were purchased from Pierce (Rockford, IL). For both FSH and LH, a standard curve was prepared and cold standards and samples (100 µl) were added to labeled tubes along with primary antibody (FSH at 1:1400 dilution and LH at 1:500) and iodinated FSH or LH. Samples were stored at 4 C overnight. On d 2, secondary antibody was added (1:10 dilution) along with 2% normal rabbit serum (Sigma Aldrich, St. Louis, MO) and incubated at room temperature for 5 min. The tubes were centrifuged for 15 min at 3000 rpm, supernatant was decanted and pellets were counted in a {gamma} counter for 1 min each. All samples were run in duplicate. Sensitivity for the FSH assay was 200 pg/ml with inter- and intraassay coefficients of variation of 2.7% and 6.7%, respectively. Sensitivity for the LH assay was 86 pg/ml with an inter- and intraassay coefficient of variation of 5.3% and 2.5%, respectively. Safety precautions established by the University of Arizona Department of Environmental Health Safety were used in handling radioactive iodine.

Measurement of ER{alpha}, ERß, and FSHR Protein Levels
Ovaries were collected from Smad3–/– and WT animals immediately after they were killed, snap-frozen, and stored at –70 C until further processing. The protein levels of ER{alpha}, ERß, and FSHR were estimated in Smad3–/– and WT mice using Western blot analysis (39). Commercially available antibodies directed against ER{alpha} (1:200 dilution; Santa Cruz, Biotechnology, Inc., Santa Cruz, CA), ERß (1:200 dilution; Zymed Laboratories), and FSHR (1:500 dilution; Santa Cruz, Biotechnology, Inc.) were used. To ensure that the proteins were loaded equally, the blots were stripped and incubated with ß-actin (1:1000 dilution; Santa Cruz Biotechnology, Inc.). Data were collected from at least three independent experiments, signal intensity was analyzed by computer densitometry using Molecular Analyst Software (Bio-Rad Laboratories, Hercules, CA), and the values were normalized using ß-actin as an internal control.

Measurement of Inhibin {alpha}-, ßA-, and ßB-Subunits
Protein was extracted from whole PD 30 ovaries on dry ice with a mortar and pestle. The tissue was collected in protein extraction buffer (10 mM Tris, 0.5 M NaCl, 1 mM MgCl, 0.1% Triton X-100) containing protease inhibitors (Roche Laboratories, Indianapolis, IN). Levels of inhibin {alpha}-, ßA-, and ßB-subunits were measured and compared in Smad3–/– and WT ovaries using Western blot analysis as previously described (66). Data were collected from three independent experiments, results were analyzed using National Institutes of Health Image comparing the levels of subunits to total protein present on blot and results were expressed as the ratio of subunit pixel to total protein pixel.

RNA Isolation and Semiquantitative RT-PCR Analysis
Ovaries were collected from Smad3–/– and WT animals immediately after they were killed, snap-frozen, and stored at –70 C until further processing. Total RNA was extracted from ovarian and follicular tissue using the RNeasy Mini Kit (QIAGEN, Inc., Valencia, CA) according to the manufacturer’s protocol. Reverse-transcriptase generation of cDNA was performed with 0.5–1 µg of total RNA using an Omniscript reverse transcriptase kit (QIAGEN) with random primers according to the manufacturer’s protocols. Subsequent PCR analysis was carried out on 3 µl of the cDNA, and the products were analyzed by electrophoresis on a 1.5% agarose gel. Primer sequences for each product were as follows: FSHR (forward) 5'-AGCAAGTTTGGCTGTTATGAGG, (reverse) 5'-GTTCTGGACTGAATGATTTAGAGG-3' (67); ER{alpha} (forward) 5'-AATTCTGACAATCGACGCCAG-3', (reverse) 5'-GTGCTTCAACATTCTCCCTCCTC-3' (68); ERß (forward) 5'-CTTGGTCACGTACCCCTTAC-3', (reverse) 5'-GTATCGCGTCACTTTCCTTT-3' (68); cdk4 (forward) 5'-TGGCTGCCACTCGATATGAAC-3', (reverse) 5'-CCTCAGGTCCTGGTCTATATG-3' (69); cyclin D2 (forward) 5'-AGCTGTCCCTGATCCGCAAG-3', (reverse) 5'-GTCAACATCCCGC ACGTCTG-3' (70); bid (forward) 5'-ACACAGCTTGTGCCATGGAC-3', (reverse) 5'-AGGCTGTCTTCACCTCATCA-3'; bad (forward) 5'-CAATAACCATCGCAACGACC-3', (reverse) 5'-CTGGAACATACTCTGGGCTG-3'; bcl-xl (forward) 5' GAGAGCGTTCAGTGATCTAAC-3', (reverse) 5'-TCAGTGTCTGGTCACTTCCGA-3'; bcl-2 (forward) 5'-ATGATAACCGGGAGATCGTG-3', (reverse) 5'-GTTCAGGTACTCAGTCATCC-3'; bax (forward) 5'-ACCAGCTCTGAACAGATCATG-3', (reverse) 5'-TGGTCTGGATCCAGACAAG-3'; bok (forward) 5'-CACATCTTCTCAGCAGGTATC-3', (reverse) 5'-AGGTGCTTTGTAGGTACTGGA-3'; mcl-1 (forward) 5'-TGGAGATCATCTCGCGCTAC-3', (reverse) 5'-CTTCTAGGTCCTGTACGTGG. To ensure that the PCR was in the exponential phase, different PCR cycles from 20–40 were tested. Primers specific for mouse ß-actin were used as an internal control as previously described (68). Data were collected from at least three independent experiments, signal intensity was analyzed by computer densitometry using Molecular Analyst Software (Bio-Rad Laboratories), and the values were normalized using ß-actin as an internal control.

Real-Time PCR Analysis
Real-time PCR analysis was performed as previously described (71). Briefly, reverse-transcriptase generation of cDNA was performed with 0.5–1 µg of total RNA using an Omniscript reverse transcriptase kit (QIAGEN) with random primers according to the manufacturer’s protocols. Real-time PCR was conducted using a MJ Research (OPTICON) real-time PCR machine and accompanying software according to the manufacturer’s instructions. The OPTICON quantifies the amount of PCR product generated by measuring the dye (SYBR green) that fluoresces when bound to double-stranded DNA. A standard curve was generated from five serial dilutions of purified PCR product. Primer sequences were described above, and for each primer a melting curve was performed. An initial incubation of 95 C for 10 min was followed by 40–50 cycles of 94 C for 10 sec, 55–60 C for 10–20 sec, and 72 C for 10–30 sec, with final extension at 72 C for 10 min. Arbitrary numbers were assigned for each standard. Values were calculated for the experimental samples from the standard curve. ß-Actin mRNA was measured in each sample as an internal control. Data were collected from three independent experiments.

Follicular Isolation
Antral follicles were isolated from ovaries of unprimed female Smad3–/– and WT mice (n = 8–10 mice per genotype, 5–7 follicles/ovary) on PD 90. Antral follicles were isolated mechanically, and cleaned of interstitial tissue. Follicles were snap-frozen and stored at –70 C until RNA isolation and RT-PCR analysis as described above.

Measurement of Ovulation
The ability of Smad3–/– and WT mice to ovulate was determined by morphological assessment of the number of corpora lutea in the ovary (64, 72, 73). Briefly, ovaries were collected from WT and Smad3–/– mice on PD 90, fixed in Kahle’s solution and processed for histological evaluation as previously described. Sections were used to count the number of corpora lutea without knowledge of genotype. To avoid double counting, each corpus luteum was followed through consecutive sections to ensure that it was only counted once. In addition, WT and Smad3–/– female mice (28–32 d old) were injected with a single sc injection of pregnant mare serum gonadotropin (5 IU/mouse) followed 48 h later by a single sc injection of hCG (5 IU/mouse). Exactly 18 h after hCG treatment, the oviduct and ovaries were removed. The oocyte/cumulus mass was surgically extracted from the oviduct and the oocytes were counted after enzymatic dissociation from the surrounding cumulus. The ovaries were fixed and embedded in paraffin for hematoxylin and eosin staining followed by morphological examination for the presence of antral follicles and corpora lutea.

Statistical Analysis
All data were analyzed using SPSS statistical software (SPSS, Inc., Chicago, IL). Both the mean diameter of preantral and antral follicles per ovary, as well as the mean percentage of atretic follicles per ovary were calculated using ovaries from at least three different animals. Similarly, means ± SEM for mRNA and protein levels, hormone levels, PCNA staining, and TUNEL staining were calculated using at least three mice per group. Differences between means were evaluated by one-way ANOVA if more than two groups were compared and t tests were used for single comparisons, with statistical significance assigned at P ≤ 0.05. When a significant P value was obtained by ANOVA, the Scheffe test was used in the post hoc analysis.


   ACKNOWLEDGMENTS
 
We thank Dr. C. Deng and Dr. S. G. Brodie for generously providing Smad3–/– mice, Sam Marion for his assistance with the FSH and LH assays, Chuck Greenfeld for his provision of bcl-2 family primers, Dr. J. Jones, Armina Kazi, and Dr. R. Koos for their assistance with real-time PCR experiments, and Dr. Wylie Vale for providing inhibin subunit polyclonal antibodies. The authors also acknowledge Janice Babus for her technical assistance and Lynn Lewis for her help with formatting this document.


   FOOTNOTES
 
We gratefully acknowledge the support of National Institutes of Health (NIH) HD 38955 (to J.A.F.), the Lalor Foundation Fellowship and University of Maryland WHRG Fellowship (to D.T.), and NIH HD 37096 (to T.K.W.), During part of this study, H.A.K. was a fellow of the Northwestern University Reproductive Biology Training Grant HD 007068, and K.P.M. was a fellow of the University of Maryland Toxicology Training Grant T32ES07263.

Abbreviations: ACVR2, Activin receptor type 2; BMP, bone morphogenetic protein; cdk, cyclin-dependent kinase; ER, estrogen receptor; FSHR, FSH receptor; hCG, human chorionic gonadotropin; PCNA, proliferating cell nuclear antigen; PD, postnatal day; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling; WT, wild-type.

Received for publication October 23, 2003. Accepted for publication June 1, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Hirshfield AN 1991 Development of follicles in the mammalian ovary. Int Rev Cytol 124:43–101[Medline]
  2. Roy SK 2000 Regulation of transforming growth factor-ß-receptor type I and type II messenger ribonucleic acid expression in the hamster ovary by gonadotropins and steroid hormones. Biol Reprod 62:1858–1865[Abstract/Free Full Text]
  3. Chang H, Brown CW, Matzuk MM 2002 Genetic analysis of the mammalian transforming growth factor-ß superfamily. Endocr Rev 23:787–823[Abstract/Free Full Text]
  4. Drummond AE, Dyson M, Tan Le M, Ethier JF, Findlay JK 2003 Ovarian follicle populations of the rat express TGF-ß signaling pathways. Mol Cell Endocrinol 202:53–57[Medline]
  5. Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N, Matzuk MM 1996 Growth differentiation factor-9 is required during early ovarian folliculogenesis. Nature 383:531–535[CrossRef][Medline]
  6. Yan C, Wang P, DeMayo J, De Mayo FJ, Elvin JA, Carino C, Prasad SV, Skinner SS, Dunbar BS, Dube JL, Celeste AJ, Matzuk MM 2001 Synergistic roles of bone morphogenetic protein 15 and growth differentiation factor 9 in ovarian function. Mol Endocrinol 15:854–866[Abstract/Free Full Text]
  7. Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A 1992 {alpha}-Inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360:313–319[CrossRef][Medline]
  8. Vassalli A, Matzuk MM, Gardner HA, Lee KF, Jaenisch R 1994 Activin/inhibin ß B subunit gene disruption leads to defects in eyelid development and female reproduction. Genes Dev 8:414–427[Abstract/Free Full Text]
  9. Schrewe H, Gendron-Maguire M, Harbison ML, Gridley T 1994 Mice homozygous for a null mutation of activin ß B are viable and fertile. Mech Dev 47:43–51[CrossRef][Medline]
  10. Dickson MC, Martin JS, Cousins FM, Kulkarni AB, Karlsson S, Akhurst RJ 1995 Defective haematopoiesis and vasculogenesis in transforming growth factor-ß 1 knock out mice. Development 121:1845–1854[Abstract]
  11. Oshima M, Oshima H, Taketo MM 1996 TGF-ß receptor type II deficiency results in defects of yolk sac hematopoiesis and vasculogenesis. Dev Biol 179:297–302[CrossRef][Medline]
  12. Sanford LP, Ormsby I, Gittenberger-de Groot AC, Sariola H, Friedman R, Boivin GP, Cardell EL, Doetschman T 1997 TGFß2 knockout mice have multiple developmental defects that are non-overlapping with other TGFß knockout phenotypes. Development 124:2659–2670[Abstract]
  13. Kaartinen V, Voncken JW, Shuler C, Warburton D, Bu D, Heisterkamp N, Groffen J 1995 Abnormal lung development and cleft palate in mice lacking TGF-ß 3 indicates defects of epithelial-mesenchymal interaction. Nat Genet 11:415–421[CrossRef][Medline]
  14. Sekelsky JJ, Newfeld SJ, Raftery LA, Chartoff EH, Gelbart WM 1995 Genetic characterization and cloning of mothers agains dpp, a gene required for decapentaplegic function in Drosophila melanogaster. Genetics 139:1347–1358[Abstract]
  15. Savage C, Das P, Finelli AL, Townsend SR, Sun CY, Baird SE, Padgett RW 1996 Caenorhabditis elegans genes sma-2, sma-3, and sma-4 define a conserved family of transforming growth factor ß pathway components. Proc Natl Acad Sci USA 93:790–794[Abstract/Free Full Text]
  16. Zimmerman CM, Padgett RW 2000 Transforming growth factor-ß signaling mediators and modulators. Gene 249:17–30[CrossRef][Medline]
  17. Hu PP, Datto MB, Wang XF 1998 Molecular mechanisms of transforming growth factor-ß signaling. Endocr Rev 19:349–363[Abstract/Free Full Text]
  18. Souchelnytskyi S, Ronnstrand L, Heldin CH, ten Dijke P 2001 Phosphorylation of Smad signaling proteins by receptor serine/threonine kinases. Methods Mol Biol 124:107–120[Medline]
  19. Souchelnytskyi S, Tamaki K, Engstrom U, Wernstedt C, ten Dijke P, Heldin CH 1997 Phosphorylation of Ser465 and Ser467 in the C terminus of Smad2 mediates interaction with Smad4 and is required for transforming growth factor-ß signaling. J Biol Chem 272:28107–28115[Abstract/Free Full Text]
  20. Lagna G, Hata A, Hemmati-Brivanlou A, Massague J 1996 Partnership between DPC4 and SMAD proteins in TGF-ß signaling pathways. Nature 383:832–836[CrossRef][Medline]
  21. Kano K, Notani A, Nam SY, Fujisawa M, Kurohmaru M, Hayashi Y 1999 Cloning and studies of the mouse cDNA encoding Smad3. J Vet Med Sci 61:213–219[CrossRef][Medline]
  22. Symonds D, Tomic D, Borgeest C, McGee E, Flaws JA 2003 Smad3 regulates proliferation of the mouse ovarian surface epithelium. Anat Rec 273A:681–686
  23. Xu J, Oakley J, McGee EA 2002 Stage-specific expression of Smad2 and Smad3 during folliculogenesis. Biol Reprod 66:1571–1578[Abstract/Free Full Text]
  24. Tomic D, Brodie SG, Deng C, Hickey RJ, Babus JK, Malkas LH, Flaws JA 2002 Smad3 may regulate follicular growth in the mouse ovary. Biol Reprod 66:917–923[Abstract/Free Full Text]
  25. Zhu Y, Richardson JA, Parada LF, Graff JM 1998 Smad3 mutant mice develop metastatic colorectal cancer. Cell 94:703–714[CrossRef][Medline]
  26. Yang X, Letterio JJ, Lechleider RJ, Chen L, Hayman R, Gu H, Roberts AB, Deng C 1999 Targeted disruption of SMAD3 results in impaired mucosal immunity and diminished T cell responsiveness to TGF-ß. EMBO J 18:1280–1291[CrossRef][Medline]
  27. Oktay K, Schenken RS, Nelson JF 1995 Proliferating cell nuclear antigen marks the initiation of follicular growth in the rat. Biol Reprod 53:295–301[Abstract]
  28. Sherr CJ 2000 The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 60:3689–3695[Abstract/Free Full Text]
  29. Robker RL, Richards JS 1998 Hormonal control of the cell cycle in ovarian cells: proliferation versus differentiation. Biol Reprod 59:476–482[Free Full Text]
  30. Robker RL, Richards JS 1998 Hormone-induced proliferation and diffe