| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Instituto de Biología y Medicina Experimental—CONICET and Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,1428 Buenos Aires, Argentina
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
estradiol, follicle-stimulating hormone, granulosa cells, growth factors
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Transforming growth factor ß1 (TGFß1) is a member of a family of structurally related polypeptides that includes TGFßs, activins, inhibins, Müllerian inhibiting substance, and bone morphogenetic proteins [1]. TGFß is synthesized by cells in a latent form that must be activated to be recognized by cell surface receptors and to trigger biological responses [1].
Expression of different TGFß isoforms has been demonstrated in the ovary by a variety of methods [2]. However, TGFß1 has been reported to modulate granulosa cell function in vitro. In primary granulosa cell cultures, TGFß enhances FSH-stimulated production of progesterone and expression of aromatase activity and antagonizes the inhibitory effect of inhibin on this enzyme [2]. Most importantly, the autocrine TGFß production potentiates the FSH-dependent LH receptor induction in rat granulosa cells [3].
The actions of TGFß on granulosa cell growth are controversial. Although it is generally accepted that TGFß inhibits proliferation of epithelial and endothelial cells, stimulatory effects have been reported in certain mesenchyme-derived cell types [4]. The effects on granulosa cell growth also seem to range from being stimulatory to being inhibitory, depending on the species and cell culture conditions. TGFß inhibits DNA synthesis in porcine [5] and bovine [6] cells. In contrast, Dorrington and coworkers reported that TGFß acts synergistically with FSH to promote rat granulosa cell proliferation [7]. They also proposed that the growth promoting effect of estradiol is mediated by the local production of TGFß [8].
One of the most well-characterized biological actions of TGFß is the regulation of secretion of extracellular matrix proteins [1]. Fibronectin is a large glycoprotein present in plasma and extracellular matrices whose synthesis is stimulated by TGFß in many cell types [1]. Although fibronectin is known to be a major secreted protein of granulosa cells from several species [9–11], its role in follicular development has not been clearly established.
The present study was designed to examine the effects of TGFß on the hormonally stimulated DNA synthesis and to evaluate the possible autocrine role of this growth factor on rat granulosa cell proliferation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ovine FSH was obtained from the National Hormone and Pituitary Program (Torrance, CA), and [methyl-3H]thymidine (10 Ci/mmol) was obtained from American Radiolabeled Chemicals (St. Louis, MO). Diethylstilbestrol (DES), human fibronectin, and fibronectin-related peptides (RGDSPC and RGESPC) were purchased from Sigma (St. Louis, MO), and TGFß1 from porcine platelets was obtained from R&D Systems (Minneapolis, MN). Collagen was prepared from rat tails as previously described [12].
Granulosa Cell Preparation and Culture
Ovaries were obtained from 24- to 25-day-old female Sprague-Dawley rats after 3 days of DES treatment (s.c. Silastic implants containing 5 mg DES). Granulosa cells were prepared and cultured as a modification of the method of Campbell [13] as previously described [12]. The ovaries were punctured with a 30-gauge needle and incubated in Dulbecco modified Eagle medium (DMEM; 4.5 g glucose/L) with Ham F12 (1:1; Gibco, Gaithersburg, MD), EGTA (6.8 mM), and Hepes (10 mM; 15 min at 37°C) and were then washed twice and incubated in DMEM-F12 (1:1), sucrose (0.5 M), and Hepes (10 mM; 5 min at 37°C). After incubation of the ovaries, the medium was diluted with two volumes of DMEM-F12 and Hepes (10 mM), and ovaries were allowed to sediment in the same 50-ml tube where they were incubated. Granulosa cells were obtained by pressing ovaries within two pieces of nylon mesh (Nytex 50, Geneva, Switzerland). To eliminate contaminating theca/interstitial cells, the crude granulosa cell suspension was layered over a 40% Percoll solution in saline and centrifuged at 400 x g for 20 min. The purified granulosa cell layer was aspirated from the top of the Percoll solution and resuspended in DMEM-F12 (1:1) containing bicarbonate (2.2 g/L; pH 7.4).
Unless otherwise indicated, cells were seeded on plastic 96-well plates (Nunc, Copenhagen, Denmark) precoated with rat tail collagen. The initial plating density was 3 x 105 viable cells/cm2. Cells were maintained at 37°C with 5% CO2. After 2 h, the medium was changed to remove nonviable cells that did not attach to the culture dish and was replaced with various media containing the different factors to be tested.
DNA Synthesis Assay
DNA synthesis was determined by 3H-thymidine incorporation according to a method previously validated in these culture conditions [12]. Granulosa cells were cultured in 96-well plates in the presence of different hormones and peptides. Tritiated thymidine (4 µCi/ml) was added to the cultures 24 h after addition of stimuli. Unless otherwise indicated, cells were harvested 24 h later in hollow glass fibers with a multiwell cell harvester (Nunc) as previously described. Radioactivity was measured in a scintillation counter. The thymidine incorporation method used in the present study had been previously validated by the following criteria: suppression by a DNA synthesis inhibitor [14], correlation with cell numbers [15], and labeling index [12]. Because media were not changed before harvest, measured radioactivity corresponds to 3H-thymidine incorporated into intact DNA.
Assay of TGFß Activity
TGFß activity in conditioned culture media was assessed by 3H-thymidine incorporation into mink lung epithelial cells (Mv1Lu CCL-64) as described by Danielpour et al. [16]. Cultures were maintained in DMEM supplemented with 10% fetal bovine serum (FBS). For thymidine incorporation assays, cell were plated in 96-well plastic dishes at 7.5 x 103 cells/well in DMEM supplemented with 0.2% FBS.
To activate TGFß in the latent form, the media were acidified (25 µl HCl 5N/ml of conditioned medium, pH 3.0) and then neutralized with NaOH/Hepes (35 µl/ml of conditioned medium) and assayed immediately.
One hour after plating, aliquots of fresh or acid-activated conditioned media were added to the cell culture. A TGFß1 standard curve (0.025–2.5 ng/ml) was included in each assay. Twenty-two hours later, 3H-thymidine (final concentration, 10 µCi/ml) was added, and cells were harvested 2 h later. Radioactivity incorporated into the DNA was measured in a liquid scintillation counter. Concentration of TGFß activity in the medium was estimated by comparison of half-maximal inhibition (ID50) of DNA synthesis in CCL-64 cells with that produced by standard TGFß1.
Labeling Index
To determine the percentage of cells synthesizing DNA, granulosa cells were cultured in 35-mm plastic dishes coated with collagen. Twenty-four hours after plating, duplicate cultures were radioactively labeled by exposure to 3H-thymidine (3 µCi/ml) for 24 h. Incubation was stopped, and the cells were fixed by washing them with PBS (0–4°C) twice, with 5% trichloroacetic acid (0–4°C), and with ethanol twice. Dishes were then processed for autoradiography. The percentage of radioactively labeled cells was determined after counting at least 500 cells/dish [12].
Statistical Analysis
Treatments were applied to triplicate wells in each of three different experiments. The results shown are the mean and SD for triplicate wells from one experiment. Effects reported as significant in representative experiments were also significant at the same confidence interval in each replicate experiment. The pattern of the responses was the same in each repeated experiment; however, the magnitudes of the responses, measured as a proportion of the baseline, were not identical for each experiment. Therefore, we used a Friedman two-way variance test with replications, a nonparametric test to determine whether several (more than two) correlated samples are from different populations. We considered each granulosa cell preparation provided by a pool of the same number of DES-primed rats as one population and the replicate wells of treatments as replications for each population. When we applied this analysis and a Dunn multiple comparisons test, we found significant differences between treatments (see figure captions). Individual comparisons within a given group of treatments were made using an ANOVA and a Tukey-Kramer test for multiple comparisons after logarithmic transformation of data [17]. Analysis of the dose-response curves and calculation of the ID50 values were performed using a computer program (ALLFIT) based on a four-parameter logistic equation [18].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A series of experiments was carried out to characterize the effect of TGFß1 on granulosa cell proliferation. DNA synthesis was evaluated by measuring thymidine incorporation during consecutive 12-h periods, starting 24 h after plating. In the presence of 5 ng/ml TGFß1, a stimulatory effect (6.5 ± 0.3-fold increase in basal thymidine incorporation) was observed between 36 and 48 h of culture. Thus, in all subsequent experiments, thymidine incorporation was performed from 24 to 48 h after plating. Data shown in Figure 1A indicate that TGFß1 triggers a dose-dependent stimulation of thymidine incorporation, and the effects were significant at concentrations as low as 1 ng/ml.
|
Interaction with FSH
Figure 1B shows the thymidine incorporation on rat granulosa cells treated with different doses of FSH in the presence of TGFß1. Addition of increasing concentrations of FSH alone did not produce a significant increase in the rates of thymidine incorporation at any of the doses used. In the presence of TGFß1, however, FSH elicited biphasic stimulation in the incorporation rates, with maximal effects observed with doses between 5 and 20 ng/ml. At these concentrations, FSH did not alter the ID50 for TGFß1 (Fig. 1A).
To verify that the changes in thymidine incorporation were associated with changes in the number of cells entering DNA synthesis, labeling indexes were determined in cultures treated with TGFß1 and FSH for 48 h. Results indicated that TGFß alone increased the percentage of cells synthesizing DNA and that its effect was amplified by FSH (control: 0.55 ± 0.11a, FSH: 0.57 ± 0.21a, TGFß: 5.11 ± 0.41b, FSH + TGFß: 13.78 ± 0.20c; values with different superscripts are significantly different, P < 0.05, ANOVA and Tukey-Kramer multiple range test).
Interaction with Estradiol
In the rat granulosa cell culture system used in the present study, estradiol is not stimulatory by itself but rather interacts with a theca cell-derived factor [15] or with FSH [19] to promote DNA synthesis. Based on a previous study in which Bendell and Dorrington [20] suggested that the growth promoting actions of estradiol were mediated by TGFß, we sought to determine whether addition of exogenous TGFß1 was able to mimic estrogen effects. Data shown in Table 1 indicate the existence of synergistic effects for combinations of TGFß1 with either FSH or estradiol. Furthermore, the synergism between FSH and estradiol was markedly amplified in the presence of maximally effective doses of TGFß1.
|
Effect of Conditioned Medium on DNA Synthesis in Granulosa Cells Is Partially Due to TGFß
Kim and Schomberg have shown that TGFß produced by granulosa cells potentiates FSH-induced LH receptor formation in the same cells [3]. Therefore, we investigated the possibility of an autocrine role of TGFß on rat granulosa cell proliferation. As shown in Figure 2, the granulosa cell-conditioned medium stimulated thymidine incorporation in a concentration-dependent manner. Acid activation produced a marked increase in the potency of the conditioned media. However, the effects were not fully comparable to those elicited by a commercial preparation of TGFß1. The activated conditioned medium shows a biphasic effect, whereas the conditioned medium without treatment reached the maximum observed effect at the highest of the assayed doses. The mitogenic effect of conditioned medium on granulosa cells was neutralized by the addition of an anti-TGFß antibody (Fig. 2B).
|
Regulation of TGFß Production
We measured the levels of TGFß bioactivity in the medium with or without activation by acidic treatment, comparing the ID50 of DNA synthesis in CCL-64 cells with that produced by standard TGFß1 (ID50 = 0.26 ± 0.04 ng/ml). The conditioned media, with or without prior acid activation, inhibited thymidine incorporation with dose-response curves that were parallel to those obtained with standard TGFß1 (data not shown). Therefore, the TGFß content in the conditioned media was estimated using a four-parameter logistic program (ALLFIT) that compares the ID50 values. Most of the TGFß produced at 48 and 72 h of culture was present in the latent form (active/total = 11% ± 3% and 17.7% ± 5.3%, respectively).
To study the hormonal regulation of TGFß production we collected the conditioned media after different hormonal treatments. FSH (20 ng/ml) treatment elicited a significant suppression of both active and total TGFß activity (Table 2) but did not affect the proportion of active/total TGFß activity. As shown in Table 2, TGFß-like activity was produced by granulosa cells in relatively high amounts, regardless of the hormonal treatment. The proportion of active/total TGFß activity was highest in cultures treated with estradiol alone. Simultaneous addition of FSH and estradiol, however, elicited a significant suppression of active and latent TGFß levels. Therefore, the previously described direct stimulatory activity of FSH and estradiol on granulosa cells DNA synthesis [15, 19] (Table 1) could not be explained by an increased production of this growth factor.
|
We then tried to determine whether the hormonal treatments affected the responsiveness of granulosa cells to autocrine stimulation by TGFß. When conditioned medium from untreated cells was activated and added back to granulosa cell cultures, it elicited a 1.4-fold increase in thymidine incorporation (Fig. 3). This effect was markedly amplified by FSH (3-fold), estadiol (4.3-fold), and a combination of both (8.7-fold) (Fig. 3). Both the activity of the conditioned medium from untreated cells and that of an equipotent amount of TGFß1 showed a similar modulation by FSH and estradiol (Fig. 3). Treatment of the granulosa cells with FSH, estradiol, or a combination of both did not affect the response of the control activated and latent conditioned medium (data not shown).
|
Effect of a Fibronectin Antagonist
During the course of these experiments, granulosa cells cultured in the presence of TGFß1 showed extensive cell spreading. Furthermore, the cell rounding induced by FSH was partially suppressed in cultures treated with TGFß1. This observation could be indicative of changes in the production of extracellular matrix proteins and the fact that TGFß stimulates fibronectin production in a variety of cell types, including granulosa cells [10]. Therefore, we decided to test whether the effect of TGFß1 on DNA synthesis might be mediated by an increase in fibronectin production. Granulosa cells were cultured in the presence of a synthetic peptide containing the RGD sequence (RGDSPC), which is known to antagonize fibronectin binding to integrins [21]. The synergism between TGFß1 and FSH was partially blocked by RGDSPC, whereas the peptide containing the inactive sequence (RGESPC) had a slight stimulatory effect (Fig. 4A).
|
Effect of Exogenous Fibronectin
We next tried to determine whether the efect of TGFß1 could be reproduced by adding fibronectin to the culture medium. Exogenous fibronectin was able to stimulate thymidine incorporation at levels comparable to those observed in TGFß1-treated cells (Fig. 4B). Furthermore, in the presence of fibronectin this growth factor was unable to further stimulate DNA synthesis. However, in contrast to TGFß1, exogenous fibronectin did not interact with FSH (Fig. 4B).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The observation of a synergism between FSH and TGFß is in agreement with previous reports by Dorrington and coworkers [7] showing the stimulation of granulosa cell proliferation by a combination of TGFß and FSH. In the present study, the dose-response curve for FSH in the presence of TGFß1 was biphasic. Such bell-shaped curves have also been observed in the interaction between FSH or cAMP and insulin/insulin-like growth factor I in this system [12] and for other cAMP-mediated responses in human granulosa cell cultures [22]. Therefore, the biphasic nature of FSH responses in the presence of TGFß does not seem to be related to the previously described differential action of this growth factor in the presence of either low or high doses of the gonadotropin [23].
Bendell and Dorrington [20] proposed that the stimulatory actions of estradiol on rat granulosa cell DNA synthesis are mediated by the increased production of TGFß. In agreement with these results, we found that estradiol induced a small but significant increase in active TGFß secretion (1.6-fold, see Table 2). In addition, this steroid elicited a marked amplification of TGFß action (2.4-fold, see Table 1).
Several factors can account for the discrepancies in estrogen effects. First, in their study, Bendell and Dorrington [20] did not evaluate the presence of contaminating theca/interstitial cells, which are known to produce both active and latent TGFß [24]. Second, Bendell and Dorrington observed a stimulatory effect of estradiol on TGFß accumulation only after 60 h in culture, whereas in the present study estrogen effects on DNA synthesis were assessed between 24 and 48 h after plating.
The fresh and the acid-activated conditioned media stimulated thymidine incorporation in a concentration-dependent manner (8-fold for 10% of activated medium and 5-fold for 40% of fresh conditioned medium). However, the effects were not comparable to the pattern obtained with standard TGFß1. The activated conditioned medium showed a biphasic effect, whereas the conditioned medium without treatment reached a maximal effect at the highest of the assayed doses, reaching only 50% of the effect observed with saturation of TGFß1. These results suggest the presence of antagonists for the TGFß stimulatory effects in the conditioned medium. Tumor necrosis factor 
(TNF) is a possible antagonist; exogenous TNF inhibits the interaction between FSH and TGFß on granulosa cell DNA synthesis [25].
TGFß is known to be secreted in latent inactive forms that can be activated by proteolytic enzymes such as plasmin and urokinase [26]. In addition, mature TGFß can be inactivated by binding to 2-macroglobulin [27]. Both urokinase type plasminogen activator and 2-macroglobulin are produced by granulosa cells, and their secretion is hormonally regulated [28, 29]. However, the synthesis of thrombospondins, a multifunction glycoprotein that is also able to selectively activate one of the latent forms of TGFß in adrenocortical cells [30], is inhibited during FSH-induced differentiation of granulosa cells [31]. Thus, we decided to determine whether FSH and estradiol treatments had any effect on the ratio of active/latent TGFß produced in our system.
Treatment with estradiol increased the ratio of active/latent TGFß activity produced by granulosa cells, whereas neither FSH nor a combination of estradiol and FSH increased this ratio. Therefore, although these hormones amplify TGFß action once it is activated, activation itself may be controlled by a different factor or cell type. In this regard, the oocyte and theca cells are logical candidates to play a role in the modulation of the generation of the active mitogen.
Under our culture conditions, the combination of FSH, estradiol, and TGFß1 was the most powerful stimulus for DNA synthesis of all the combinations of growth factors tested so far, reaching stimulation of almost 200-fold over basal levels.
In view of the widely reported inhibitory effects on cell proliferation elicited by TGFß, the mitogenic action observed in some cell types has been ascribed to indirect effects mediated by changes either in the components of the extracellular matrix [1] or in other autocrine mitogens such as platelet-derived growth factors (PDGFs) [32].
The stimulatory effect of fibronectin on rat granulosa cell DNA synthesis is consistent with previous reports that addition of fibronectin promotes cell proliferation and cytokinesis, presumably by increasing cell adhesion to the substrate [33]. The fibronectin-mediated increase in cell adhesion seems to be responsible for the stimulation of DNA synthesis elicited by TGFß1 alone. Nevertheless, the exogenous protein was unable to synergize with FSH, and the interaction between FSH and TGFß1 was only partially suppressed in the presence of a fibronectin antagonist. This finding suggests the existence of additional mechanisms for the interaction between the signaling cascades triggered by FSH and TGFß.
The apparent failure of exogenous fibronectin to fully mimic TGFß1 effects should be interpreted with caution. Fibronectins are in fact a family of adhesive glycoproteins that have variable primary structures owing to cell type-specific splicing of a primary transcript [34]. Cellular fibronectins differ from plasma fibronectins in that they have so-called extra domains (EDs) in the molecule. The fibronectin used in the present study is a plasma isoform and therefore may not be equivalent to the form produced by granulosa cells under TGFß stimulation. In addition, fibronectin fragments produced by proteolytic cleavage (but not the intact fibronectin molecule) are mitogenic for cultured fibroblasts [35]. In this regard, we have provided evidence indicating that TGFß1 treatment modulates both the alternative splicing and the proteolytic processing of fibronectin in a bovine granulosa cell line [36, 37]. Furthermore, in bovine granulosa cells a synthetic peptide corresponding to the ED-I exon (not present in plasma fibronectin) was found to be mitogenic, whereas the plasmatic form was without effect [37].
The stimulatory actions of TGFß on cell proliferation in some cell types are mediated by an autocrine PDGF loop [38]. PDGF has been shown to stimulate proliferation of cultured porcine granulosa cells under some conditions [39]. In our experiment, neither PDGF isoform AA nor PDGF isoform BB had a significant effect on basal thymidine incorporation. Furthermore, these factors had an inhibitory action in the presence of FSH or FSH plus TGFß (data not shown). However, the possibility of autocrine loops mediated by other growth factors cannot be excluded. We have recently observed that the stimulatory effect of TGFß on rat granulosa cell DNA synthesis can be partially blocked by follistatin, suggesting the existence of an activin-mediated process [40].
Although the precise role(s) of ovarian TGFß has not been clearly established, the present data suggest that in addition to its previously demonstrated effect on cell differentiation, this factor may amplify the hormonal regulation of rat granulosa cell proliferation through a mechanism that partially involves formation of the extracellular matrix.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by grants from CONICET, the Buenos Aires University (TX218 to J.L.B.), and a Carrillo Oñativia Fellowship to J.L.B. 
2 Correspondence: Patricia Saragüeta, Obligado 2490, 1428 Buenos Aires, Argentina. FAX: 54 11 4786 2564; sarag{at}dna.uba.ar
3 These authors contributed equally to this work
Accepted: January 8, 2002.
Received: October 5, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) on rat granulosa cell DNA synthesis: interaction with FSH and transforming growth factor-ß1. In: Program of the 78th annual meeting of the Endocrine Society; 1996; San Francisco, CA. Abstract 145 2-macroglobulin. J Biol Chem 1990; 265:6973-6977 receptors. J Cell Physiol 1990; 145::181-186[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  P. Yang and S. K. Roy Transforming Growth Factor B1 Stimulated DNA Synthesis in the Granulosa Cells of Preantral Follicles: Negative Interaction with Epidermal Growth Factor Biol Reprod, July 1, 2006; 75(1): 140 - 148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.L. Juengel and K.P. McNatty The role of proteins of the transforming growth factor-{beta} superfamily in the intraovarian regulation of follicular development Hum. Reprod. Update, March 1, 2005; 11(2): 144 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Gueripel, M. Benahmed, and A. Gougeon Sequential Gonadotropin Treatment of Immature Mice Leads to Amplification of Transforming Growth Factor {beta} Action, Via Upregulation of Receptor-Type 1, Smad 2 and 4, and Downregulation of Smad 6 Biol Reprod, March 1, 2004; 70(3): 640 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Roy, J. Wang, and P. Yang Dexamethasone Inhibits Transforming Growth Factor-{beta} Receptor (T{beta}R) Messenger RNA Expression in Hamster Preantral Follicles: Possible Association with NF-YA Biol Reprod, June 1, 2003; 68(6): 2180 - 2188. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
