help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (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 Pernasetti, F.
Right arrow Articles by Mellon, P. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Pernasetti, F.
Right arrow Articles by Mellon, P. L.
Endocrinology Vol. 142, No. 6 2284-2295
Copyright © 2001 by The Endocrine Society

ARTICLES

Cell-Specific Transcriptional Regulation of Follicle-Stimulating Hormone-ß by Activin and Gonadotropin-Releasing Hormone in the LßT2 Pituitary Gonadotrope Cell Model1

Flavia Pernasetti2, Vyacheslav V. Vasilyev3,4, Suzanne B. Rosenberg4, Janice S. Bailey5, Huey-Jing Huang, William L. Miller and Pamela L. Mellon

Departments of Reproductive Medicine (F.P., V.V.V., S.B.R., J.S.B., P.L.M.) and Neuroscience (V.V.V., P.L.M.), University of California-San Diego, La Jolla, California 92093-0674; and Department of Biochemistry, North Carolina State University (H.-J.H., W.L.M.), Raleigh, North Carolina 27695

Address all correspondence and requests for reprints to: Pamela L. Mellon, Ph.D., Department of Reproductive Medicine 0674, University of California-San Diego, 9500 Gilman Drive, La Jolla, California 92093-0674. E-mail: pmellon{at}ucsd.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
FSH is secreted by gonadotropes of the anterior pituitary and plays a crucial role in mammalian reproduction. However, little is known about FSH gene regulation due to the lack of a gonadotrope cell line that synthesizes FSH. The LßT2 mouse pituitary cell line, isolated by targeted tumorigenesis in transgenic mice, has the characteristics of a mature gonadotrope, including expression of GnRH receptor, steroidogenic factor 1, and both the {alpha}- and ß-subunits of LH, but was thought not to express FSH. Using RT-PCR, we show that these cells synthesize FSH ß- subunit messenger RNA, which is induced by activin and inhibited by follistatin. Furthermore, in transient transfections an ovine FSHß 5'-regulatory region (5.5 kb) confers LßT2 cell-specific expression to a reporter gene compared with other pituitary and nonpituitary cell lines. This FSHß regulatory region responds to activin specifically in LßT2 cells, an effect that is blocked by follistatin. The LHß, {alpha}-subunit, and GnRH receptor regulatory regions are induced by activin and blocked by follistatin. Furthermore, LßT2 cells express the components of the activin system, and addition of follistatin alone reduces FSHß gene expression, demonstrating that an endogenous activin autocrine loop regulates FSH in these cells. In addition, GnRH stimulates both the FSHß and LHß regulatory regions, specifically in LßT2 cells. Surprisingly, GnRH induction is reduced by follistatin, suggesting its dependence on endogenous activin. As the mouse GnRH receptor promoter is inhibited by follistatin, reduction of GnRH receptor levels might be one mechanism by which follistatin interferes with GnRH induction of gonadotropin genes. In summary, LßT2 cells exhibit the characteristics of fully differentiated gonadotropes, including the expression of LH, FSH, GnRH receptor, and components of the activin/follistatin system, as well as display the appropriate responses to activin and GnRH.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE MATURE ANTERIOR pituitary gonadotrope serves as a control center for integration of hormonal signals in the reproductive axis. The gonadotrope cells synthesize and secrete both pituitary gonadotropins, FSH and LH. FSH and LH are members of the family of glycoprotein hormones that also includes TSH and CG (1). Each of these heterodimeric hormones shares a common {alpha}-subunit, but also contains a unique ß-subunit that confers physiological specificity.

Pituitary gonadotropin messenger RNA (mRNA) expression and protein secretion are regulated by pulses of GnRH from the hypothalamus (2). Differential control of LH vs. FSH production is determined by the frequency and amplitude of the GnRH pulses (3). Another important point of control at the pituitary level is the synthesis and activity of the seven-transmembrane G protein-coupled GnRH receptor (GnRH-R), which, along with the gonadotropins, is regulated by both steroid and peptide hormones (4, 5).

Activin and inhibin, members of the transforming growth factor-ß family of growth factors, also play important roles in the modulation of hormone secretion from pituitary gonadotropes. Activin stimulates the synthesis of FSH (6), and inhibin represses the synthesis of FSH in the pituitary (7, 8). Activin is secreted by the gonads and within the pituitary gland, where it acts as an autocrine and/or paracrine regulator of the reproductive axis. Activin occurs as a homo- or heterodimer of two highly related ß-subunits (ßA and ßB) (9). Inhibin is an activin antagonist composed of an activin ß- subunit (ßA or ßB) and a unique {alpha}-subunit (8) secreted by the pituitary, gonads, adrenal gland, and brain (10). Follistatin, a potent activin-binding protein, is primarily secreted by gonadotrope and folliculostellate cells within the pituitary (11). It inhibits binding of activin to its receptors and thereby specifically inhibits the synthesis and secretion of FSH by the pituitary (12). This remarkable confluence of interacting hormones, growth factors, and receptors within a single cell type indicates a complex set of autocrine loops and endocrine responses that can be differentially modulated to regulate the transcription and secretion of the gonadotropins, LH and FSH.

Currently, very little is known about FSHß regulation at the transcriptional level, largely due to the lack of a differentiated gonadotrope cell line that produces endogenous FSHß in which to perform these studies. Primary cell culture, transfections into heterologous cells, and transgenic mice have been employed in the past to study transcriptional regulation of the FSHß gene. Each of these approaches has inherent disadvantages. A problem encountered when using primary pituitary cell culture preparations has been the multiplicity of endocrine cell types in this tissue. Gonadotropes comprise only 5–10% of pituitary cells, making it difficult to distinguish cell-specific responses over the background of activities in multiple other cell types. Moreover, the magnitude of a given response may be small due to the limited proportion of each cell type in the culture, making the study of modestly expressed genes difficult. The use of heterologous cells (13, 14) does not accurately reflect cell-specific responses, because distinct sets of transcription factors, kinases, G proteins, receptors, and other classes of molecules are expressed in different cell types. Finally, the use of transgenic mice has proven to be a very laborious and time-consuming approach (15) that does not allow analysis of signaling pathways or DNA-protein interactions.

Targeted expression of oncogenes in transgenic mice has been used to immortalize specific cell types that can then serve as valuable cultured model systems. Our laboratory has used 1.8 kb of the rat LHß promoter to direct expression of the oncogene simian virus 40 T antigen to the pituitary in mice (16). One of the cell lines isolated from the tumors induced by this transgene, LßT2, has many characteristics of a mature gonadotrope cell, including expression of GnRH-R, steroidogenic factor 1 (SF-1), estrogen, and progesterone receptors (17) and both the {alpha}- and ß-subunits of LH (16).

In the present study we show that the LßT2 cell line also expresses FSH ß-subunit mRNA and that it is induced by activin and repressed by follistatin. To investigate the potential of LßT2 cells as a model for the study of FSH gene regulation, we performed transient transfections with the 5.5-kb flanking region of the ovine FSHß (oFSHß) gene, a region that successfully targets expression to the pituitary of transgenic mice (18). We found that this 5.5-kb region is specifically expressed in the LßT2 gonadotrope cells compared with other pituitary and nonpituitary cell lines. It also responds to activin, an effect that is inhibited by follistatin. Surprisingly, activin regulation is LßT2 cell specific, failing to occur in {alpha}T3–1 cells despite the fact that they are known to be activin responsive (19, 20). Furthermore, LßT2 cells express all four known activin receptors as well as activin ßB-subunit, follistatin, and inhibin {alpha}-subunit. In addition, GnRH stimulates transcription from FSHß and LHß, and this response is inhibited by follistatin, suggesting that the GnRH response depends on endogenously produced activin. In concert with this observation, the GnRH-R promoter is repressed by follistatin in LßT2 cells, indicating that the activin dependence of the FSHß response to GnRH may be due at least partially to activin regulation of the GnRH-R gene. In summary, the 5'-flanking region of the ovine FSHß gene is specifically expressed in LßT2 cells and responds appropriately to its primary regulators, activin and GnRH. Thus, these cells constitute an excellent model system to study the mechanisms involved in FSHß gene regulation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
Recombinant human activin A and recombinant human follistatin-288 were obtained from Dr. A. F. Parlow, at the National Hormone and Pituitary Program (Torrance, CA) and Dr. Shunichi Shimasaki. GnRH agonist (des-Gly10, [D-Ala6]LHRH ethylamide) was purchased from Sigma (St. Louis, MO).

Construction of plasmids for transfection analysis
A 5.5-kb region of the ovine FSHß gene (oFSHß) encompassing 4741 bp of the promoter and 759 bp downstream from the +1 transcription start site, including the first intron of the oFSHß gene (the oFSHß translation start site was mutated to permit translation to initiate at the reporter gene AUG), was subcloned into the KpnI and XbaI restriction sites of the pGL3-Basic luciferase reporter plasmid (Promega Corp., Madison, WI). The resulting plasmid was named oFSHß-Luc.

The mouse GnRH-R promoter sequence was amplified by PCR. The 1227-bp band was verified by DNA sequencing and cloned in the BamHI/XhoI sites of the pUC18 backbone and named pGRBX. This plasmid was then digested with SphI (5'-end), filled with Klenow polymerase to generate a blunt end, and subsequently digested with XhoI. The isolated fragment was then cloned into the SmaI/XhoI sites of pGL3-Basic (Promega Corp.) and named GnRH-R-Luc.

The {alpha}GSU-Luc plasmid contains -507 to +46 bp of the mouse glycoprotein {alpha}-subunit promoter and was provided by Dr. Richard Maurer (21). The LHß-Luc plasmid contains the region between -1800 and +5 bp of the rat LHß promoter (provided by Dr. James Roberts) cloned upstream of a luciferase reporter gene in the pUC18 plasmid backbone.

The thymidine kinase (TK)-Luc plasmid contains the -81 bp region of the herpes simplex I thymidine kinase promoter cloned in pXP2, described by Dr. Steve Nordeen (22), provided by Dr. Sylvia Evans. This region of the TK promoter is missing both the CCAAT box and the distal GC box present in the 109-bp fragment of the promoter.

Plasmid DNA was prepared from overnight bacterial cultures using plasmid columns according to the supplier’s protocol (QIAGEN, Chatsworth, CA) or a cesium chloride protocol adapted from Sambrook et al. (23).

Cell culture and transient transfections
Cells were grown in 60-mm diameter dishes to 60–70% confluence in DMEM (Cellgro, Mediatech, Inc., Herndon, VA) supplemented with 10% FBS (Omega Scientific, Inc., Tarzana, CA) at 37 C with 5% CO2. Transient transfections were performed using FuGENE 6 transfection reagent (Roche Molecular Biochemicals, Indianapolis, IN), following the manufacturer’s protocol. For the activin/follistatin experiments, cells were pretreated with follistatin (250 ng/ml) at the time of transfection. After overnight incubation (16 h), cells were washed twice with 1 x PBS, and each independent treatment was then added in DMEM with 1% FBS: activin (100 ng/ml), follistatin (250 ng/ml), or activin plus follistatin. Cells were then incubated for an additional 24 h. GnRH treatments were performed at 1 nM for 6 h before harvest. The cells were then harvested, and luciferase and ß-galactosidase assays performed. All experiments were performed using 0.44 pmol of the reporter plasmids, unless noted differently. Rous sarcoma virus (RSV)-ß-galactosidase (0.5 µg) was used as an internal control unless noted differently. All transfection experiments were performed at least three times.

Luciferase and ß-galactosidase assays
Cells were washed twice in 1 x PBS and then 1 ml harvesting buffer (0.15 M NaCl, 1 mM EDTA, and 40 mM Tris-HCl, pH 7.4) was added to each dish. Cells were then scraped, transferred to microcentrifuge tubes, and collected by centrifugation at 3000 rpm for 5 min. The supernatant was discarded, and the cells were resuspended in 50–100 µl lysis solution (Galacto-light assay system, Tropix, Bedford, MA).

Luciferase activity was measured using an E.G.&G. Berthold Microplate Luminometer by injecting 100 µl of a buffer containing 100 mM Tris-HCl (pH 7.8), 15 mM MgSO4, 10 mM ATP, and 65 µM luciferin/well. ß-Galactosidase assays were performed using the Galacto-light assay system following the manufacturer’s protocol. Before each ß-galactosidase assay, cell extracts were heat-inactivated at 48 C for 50 min.

Reverse transcriptase, PCR, and Southern blotting
Total RNA was isolated by the method of Chomczynski and Sacchi (24). Polyadenylated [poly(A)+] mRNA was extracted using the PolyATract mRNA Isolation System IV (Promega Corp.) according to the manufacturer’s protocol. The concentration of each RNA sample was determined by UV absorbance measurement. Twenty nanograms of each RNA sample were annealed to 300 pmol of random hexamers (pdN6, Pharmacia Biotech, Piscataway, NJ) by heating at 70 C for 10 min and then chilling on ice. The RT reaction buffer, 7.5 mM dithiothreitol, and 0.37 mM deoxy-NTP mix were added to the annealed sample, and the mix was heated for 10 min at 25 C, and then incubated for 2 min at 42 C, after which 200 U Superscript TM II RT (Life Technologies, Inc., Gaithersburg, MD) were added. The reaction was then incubated for an additional 50 min at 42 C, after which the enzyme was inactivated by incubation at 70 C for 10 min. A 1-µl aliquot of the RT reaction was used as a template for the PCR reaction. Mouse FSHß primers (sense, 5'-GGACGTAGCTGTTTACTTCCCAG-3'; antisense, 5'-AGGGAGTCTGAGTGGCGGGC-3') were designed to span intron 2. The expected FSHß DNA products were 235 bp (spanning 97 bp of the 3'-end of exon 2 plus 138 bp of the 5'-end of exon 3) and 512 bp (235 bp of exons 2 and 3 sequences as described above plus 227 bp corresponding to intron 2) for complementary DNA (cDNA) and genomic products (25), respectively. The PCR reaction was performed in a final volume of 30 µl with 1.25 U Taq DNA polymerase (Amersham Pharmacia Biotech), 3 µl 10 x Taq buffer, 0.5 mM deoxy-NTP mix, and 10 pmol of each primer. We then performed 35 PCR cycles: 1 min at 94 C, 1 min at 63 C, and 1 min at 68 C, followed by an elongation step for 10 min at 68 C. PCR products were electrophoresed on 2% agarose and transferred to a nylon membrane (Hybond N+, Amersham Pharmacia Biotech, Arlington Heights, IL) following the Southern blot protocol described by Sambrook et al. (23). The rat FSHß cDNA (26) probe fragment was labeled by random priming (27).

Northern blotting
Total RNA was isolated using the TRIzol reagent, (Life Technologies, Inc.), following the manufacturer’s protocol and poly(A)+ RNA was extracted using PolyATract mRNA Isolation System IV (Promega Corp., Madison, WI) according to the manufacturer’s protocol. Two µg of each polyA+ RNA sample were electrophoresed in a denaturing 1% agarose gel containing formaldehyde and transferred to Hybond N+ nylon membrane (Amersham Pharmacia Biotech). The RNA was fixed by UV cross-linking using a UV chamber (Bio-Rad Laboratories, Inc., Richmond, CA). Membranes were hybridized overnight with 32P-labeled cDNA probe at 55 C in a 25% formamide solution. DNA probes were labeled by random priming (27). Excess probe was removed by washing at 65 C with 0.2 x SSPE (36 mM NaCl, 2 mM sodium phosphate, and 0.2 mM EDTA) and 0.1% SDS.

Statistics
In the transfection experiments, luciferase to ß-galactosidase ratios were compared for each independent condition by the ANOVA factorial test, followed by the Fisher’s protected least significant difference post-hoc test. In all analyses, P < 0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The endogenous mouse FSH ß-subunit mRNA is expressed in LßT2 cells and induced by activin
In our original characterization of the LßT2 cell line, Northern blots using LßT2 total RNA showed expression of {alpha}-subunit, LH ß-subunit, and other gonadotrope-specific markers, such as GnRH-R and SF-1; however, FSHß was not detected by this method (28). The FSHß gene is regulated differently from the LHß gene via mechanisms including activin stimulation and low pulsatile GnRH frequency. We hypothesized that FSHß might be expressed at low levels in LßT2 cells in the absence of these physiological stimulators. We then performed RT-PCR, a more sensitive method, to detect even low levels of FSHß RNA in LßT2 cells. LßT2 cell poly(A)+ mRNA was extracted from cells pretreated for 16 h with follistatin to block endogenously produced activin. The pretreatment was followed by a 48-h treatment with activin, follistatin, or a combination of both. Poly(A)+ mRNA from {alpha}T3–1 cells treated under the same conditions was used as a negative control, as these cells represent a precursor gonadotrope that expresses GnRH-R, SF-1, and {alpha}-subunit, but neither of the ß-subunits (29). Multiple RT-PCR products were observed on an agarose gel, including the expected size for the amplified mouse FSHß cDNA fragment (235 bp; data not shown). A Southern blot of the RT-PCR gel, probed with a rat FSHß cDNA fragment (26), detected the 235-bp FSHß-specific band (Fig. 1Go) in pretreated (-) LßT2 cell RNA. Treatment with activin increased the expression of FSHß in LßT2 cells, as indicated by the stronger intensity of the hybridization signal of the 235-bp PCR-amplified band. Follistatin blocks the synthesis of FSHß RNA in LßT2 cells alone or in combination with activin, as shown by the absence of detectable FSHß signal under these conditions. Furthermore, no FSHß signal was present in {alpha}T3–1 cell samples regardless of activin and/or follistatin treatments.



View larger version (63K):
[in this window]
[in a new window]
  		Figure 1. FSH ß-subunit mRNA is expressed in LßT2 cells, induced by activin, and blocked by follistatin. LßT2 and {alpha}T3–1 cells were pretreated with follistatin (250 ng/ml) for 16 h, followed by individual treatments with activin (Act; 100 ng/ml), follistatin (FS; 250 ng/ml), or a combination of both (Act+FS) for 48 h. Poly(A)+ RNA was extracted, and cDNA was synthesized by RT-PCR. The RT-PCR products were then run on a 2% agarose gel. To demonstrate the specificity of the bands, Southern blotting was performed, using rat FSHß cDNA (26 ) as a probe. The membrane was exposed to x-ray film overnight. The size of the expected cDNA RT-PCR product for the mouse FSHß is 235 bp.

 
A transfected 5.5-kb ovine FSH ß-subunit regulatory region targets reporter gene expression specifically to LßT2 cells
The expression of endogenous FSHß mRNA in LßT2 cells led us to pursue the hypothesis that LßT2 cells could be employed as a model to study FSHß gene regulation. To compare the level of expression from the ovine FSHß 5'-flanking region in LßT2 cells to a series of other pituitary- and nonpituitary-derived cell lines (Fig. 2Go), we performed transient transfection experiments using a plasmid encompassing 5.5 kb of the ovine FSHß 5'-regulatory region driving expression of the luciferase reporter gene (oFSHß-Luc). An RSV long-terminal repeat promoter driving expression of a ß-galactosidase gene (RSV-ßgal) was cotransfected as an internal control. A RSV luciferase plasmid (RSV-Luc) was transfected in parallel with the oFSHß-Luc reporter gene to control for differences in transfection efficiency, transcription rates, reporter mRNA half-life, luciferase and ß-galactosidase protein stability, and metabolic factors between distinct cell types. The ratio of the RSV-Luc value (parallel control) to the RSV-ßgal value (internal control) was set at 100 for each cell line, and the oFSHß-Luc values from each cell line were normalized to the RSV-Luc/RSV-ßgal value. To validate this approach, we tested the activities of the promoterless luciferase expression vector pGL3 and a TK-Luc plasmid and found each to be expressed at the same level in {alpha}T3–1 and LßT2 cells relative to RSV-Luc (data not shown). Basal expression of the oFSHß-Luc plasmid is approximately 4-fold higher in LßT2 cells than in the nongonadotrope cell lines tested (P < 0.001). The luciferase activity in the gonadotrope progenitor cells ({alpha}T3–1) is greater than in other cells tested, but is still 2.2-fold lower than in LßT2 cells. The corticotrope-derived (AtT20), hypothalamic neuroendocrine (GT1–7) (30), fibroblast-derived (NIH-3T3), and kidney-derived (CV1) cells show equivalently low levels of expression of the reporter gene, indicating that the 5.5-kb oFSHß regulatory region confers LßT2 cell specificity.



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. The regulatory region of the oFSH ß-subunit gene is specifically active in LßT2 cells. The oFSHß regulatory region (5.5 kb) linked to the luciferase reporter gene (oFSHß-Luc; 3 µg) was cotransfected with the internal control, RSV-ßgal (0.5 µg), into pituitary-derived LßT2 (differentiated gonadotrope), {alpha}T3–1 (gonadotrope precursor), and AtT20 (corticotrope) cells and into nonpituitary-derived GT1–7 (hypothalamic), CV1 (kidney), and NIH-3T3 (fibroblast) cells. To normalize for differences in transfection efficiencies and transcription rates between distinct cell lines, a parallel control, RSV-Luc plasmid (0.5 µg), was transfected into each cell line with the internal control RSV-ßgal (0.5 µg). The ratio between the activities of these two plasmids was set at 100 in each cell line to normalize between the distinct cell lines. Forty-eight hours after transfection, cells were harvested for luciferase and ß-galactosidase assays. Results are presented as the mean ± SEM of at least three independent transfection experiments, each performed in triplicate. The bar marked with an asterisk (LßT2 cells) is statistically different from all other bars (P < 0.0001).

 
The transfected FSHß regulatory region responds to the activin/follistatin system specifically in LßT2 cells
We observed that the level of endogenous FSHß mRNA in LßT2 cells appears to increase upon activin treatment, as detected by RT-PCR (Fig. 1Go). It has been demonstrated that follistatin decreases steady state FSHß mRNA levels in primary pituitary cell cultures by destabilizing FSHß RNA (31). To analyze the transcriptional effects of activin and/or follistatin treatments on the oFSHß regulatory region, we transiently transfected the oFSHß-Luc reporter plasmid into LßT2, {alpha}T3–1, and NIH-3T3 cells. RSV-ßgal was used as an internal control, and RSV-Luc was transfected into each cell line in parallel to allow for normalization and comparison between distinct cell lines. At the time of transfection, cells were subjected to follistatin pretreatment to block endogenous activin. One set of plates was transfected and not pretreated to allow observation of autocrine activin effects. Sixteen hours after transfection, additional treatments with activin and/or follistatin were performed for 24 h, after which the cells were harvested. As shown in Fig. 3Go, activin stimulation of transcription from the oFSHß 5'-flanking region was only observed in LßT2 cells. In these cells reporter gene expression was approximately 3- to 4-fold higher in activin-treated cells compared with cells pretreated with follistatin (P < 0.001). As expected, follistatin treatment alone did not have a statistically significant effect on transcription from the 5'-flanking region of the oFSHß gene compared with cells that only received follistatin pretreatment. When activin and follistatin were combined, follistatin completely blocked the stimulatory effect of activin. In untreated cells, oFSHß activity was approximately 2-fold higher than in the pretreated cells, indicating that LßT2 cells synthesize endogenous activin, which is blocked by pretreatment with follistatin (P < 0.001). The oFSHß 5.5-kb region did not respond to activin in either NIH-3T3 or {alpha}T3–1 cells despite the fact that {alpha}T3–1 cells respond to activin by regulating both the GnRH-R gene (19) and the {alpha}-subunit gene (20). This suggests that members of the activin signal transduction pathway involved in FSHß gene stimulation might be present only in the more differentiated gonadotrope LßT2 cells. Alternatively, activin induction of FSHß gene expression might be dependent upon additional cell type- specific transcription factors present only in cells that naturally express FSH.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. The ovine FSHß regulatory region responds to the activin/follistatin system specifically in LßT2 cells. Transient transfections with the oFSHß-Luc reporter plasmid (0.44 pmol or 3 µg) were performed in LßT2 (differentiated gonadotrope), {alpha}T3–1 (gonadotrope precursor), and NIH-3T3 (fibroblast) cells to evaluate the responsiveness of the oFSHß promoter to activin. RSV-ßgal plasmid (0.5 µg) was cotransfected as an internal control. Transfected cells were grown in DMEM and 10% FBS with or without follistatin (250 ng/ml) for 16 h. Cells were then washed and subsequently incubated in DMEM and 1% FBS supplemented with activin (Act; 100 ng/ml), follistatin (FS; 250 ng/ml), neither, or a combination of both (Act+FS). After 48 h, cells were harvested for luciferase and ß-galactosidase assays. To allow comparisons between distinct cell lines, 0.5 µg RSV-Luc plasmid was transfected in parallel into each cell line combined with 0.5 µg RSV-ßgal, as described in Fig. 2Go. The value of RSV-Luc/RSV-ßgal in each cell line was set at 100 (not shown). Results are the mean ± SEM of three independent experiments, each performed in triplicate. Bars marked with a single asterisk (FS+Act) and double asterisks (-) are statistically different from all other bars (P < 0.05), and are also different from each other (P = 0.0022).

 
Activin stimulates the glycoprotein hormone {alpha}-subunit, LHß, and GnRH-R promoters in LßT2 cells
To determine whether activin responsiveness is unique to the FSHß gene, we compared a series of gonadotrope- specific regulatory regions in LßT2 cells. It has been reported that the mouse glycoprotein hormone {alpha}-subunit ({alpha}-GSU) promoter is repressed by activin in transient transfections in the gonadotrope precursor {alpha}T3–1 cells (19) whereas the GnRH-R promoter is stimulated by activin in these cells, effects that are blocked by follistatin (20). Transient transfections with luciferase reporter plasmids containing the oFSHß regulatory region, -507 to +46 region of the mouse {alpha}-GSU promoter ({alpha}GSU-Luc), the 1.2-kb mouse GnRH-R promoter (GnRH-R-Luc), or 1.8 kb of the rat LHß promoter (rLH-Luc), were performed in LßT2 cells. TK-Luc and a promoterless luciferase vector (pGL3) were used as negative controls (Fig. 4Go). Cells were pretreated with follistatin for 16 h, followed by a 24-h treatment with activin, follistatin, or a combination of both. To facilitate interpretation of the data, the values for the follistatin-pretreated condition were set at 1 for each plasmid tested. The oFSHß-Luc plasmid reproduced the results shown in Fig. 3Go, in which activin treatment stimulated the promoter 4-fold, and treatment with both follistatin and activin prevented the induction of reporter gene expression. The GnRH-R promoter was stimulated 6-fold by activin, compared with the pretreated condition, and this effect was completely abolished by follistatin (Fig. 4Go). Follistatin alone decreased GnRH-R (-) expression by 65%, indicating that endogenous activin plays a role in the expression of the GnRH-R promoter. Surprisingly, the LHß promoter showed a statistically significant 60% increase (P < 0.0001) upon activin treatment, and this stimulatory effect of activin was completely blocked upon cotreatment with follistatin. However, the LHß promoter was not affected by pretreatment with follistatin, indicating that LßT2 endogenous activin is not sufficient to increase the transcription of this promoter. In the case of the {alpha}-GSU promoter, an approximately 60% decrease in the activity of the promoter in pretreated cells (P < 0.001) was observed compared with that in the untreated group, demonstrating that the expression of this promoter is maintained by endogenous activin in LßT2 cells. Moreover, activin treatment led to a 40% increase in transcription of the {alpha}-GSU promoter (P < 0.01), an effect that was abolished when activin was combined with follistatin. As expected, the TK promoter does not respond to any of the activin/follistatin treatments, nor does pGL3, demonstrating that the activin/follistatin responses observed in the LßT2 cells are promoter specific.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 4. Regulation by the activin/follistatin system is promoter specific in LßT2 cells. Transient transfections with equimolar amounts (0.44 pmol) of GnRH-R-Luc, {alpha}GSU-Luc, LHß-Luc, oFSHß-Luc, TK-Luc, and pGL3, each cotransfected with 0.5 µg RSV-ßgal as an internal control, were performed in LßT2 cells. Transfected cells were grown in DMEM and 10% FBS with or without follistatin (250 ng/ml) for 16 h. Cells were then washed and subsequently incubated in DMEM and 1% FBS supplemented with activin (Act; 100 ng/ml), follistatin (FS; 250 ng/ml), neither, or a combination of both (Act+FS). After 48 h, cells were harvested for luciferase and ß-galactosidase assays. To facilitate comparisons between promoter responses, the follistatin-pretreated values were set at 1. Results are the mean ± SEM of three independent experiments, each performed in triplicate. ANOVA and post-hoc Fisher’s statistical analyses were performed between treatments for each promoter (not between different promoters). Within each group (representing the same promoter), bars marked with single and double asterisks are statistically different from all other bars (P < 0.05) and also different from each other.

 
Activin, follistatin, and inhibin mRNAs are expressed in LßT2 cells
The stimulatory effect of activin on endogenous FSHß RNA levels and on the gonadotrope-specific regulatory regions transfected into LßT2 cells combined with the blockade of these effects by follistatin (Figs. 1Go, 3Go, and 4Go), indicate that a functional activin/follistatin signaling system is present in these cells. To investigate whether activin, inhibin, and follistatin mRNAs are expressed in LßT2 cells, we performed Northern blots with poly(A)+ RNA extracted from LßT2, {alpha}T3–1, and NIH-3T3 cells and probed them with rat cDNA fragments coding for activin ßB-subunit, follistatin, and inhibin {alpha}-subunit. Activin ßB- subunit transcripts (4.2 and 3.4 kb) were observed (Fig. 5Go) in both LßT2 and {alpha}T3–1 cells, whereas NIH-3T3 cells did not express any of the ßB-subunit RNA forms. These results indicate that an activin autocrine loop is probably present in the LßT2 and {alpha}T3–1 cell cultures. As a positive control, we used ovary RNA, in which we observed only the 4.2 kb ßB-subunit form, in agreement with previous observations (32). As pituitary gonadotropes have been shown to synthesize activin ßB-subunit RNA (33), the presence of endogenous activin in LßT2 cells models mature gonadotropes in vivo and validates the use of this cell for the study of activin regulation of gonadotropin gene expression. Two follistatin mRNA species (2.5 and 1.6 kb) were observed in all three cell lines tested (Fig. 5Go). It is not surprising that the NIH-3T3 fibroblast-derived cells also synthesize follistatin mRNA, as it has been described in a wide variety of tissues, including pituitary (34). Inhibin {alpha}-subunit mRNA (1.6 kb) is observed only in LßT2 cells (Fig. 5Go). Surprisingly, inhibin {alpha}-subunit RNA was not detectable in the less mature gonadotrope cell line, {alpha}T3–1, suggesting that inhibin is either expressed at very low levels or is absent from {alpha}T3–1 cells. In support of this idea, in rat embryos, inhibin {alpha}-subunit is detected by in situ hybridization only later in development, between embryonic days 16 and 20 (35).



View larger version (38K):
[in this window]
[in a new window]
  		Figure 5. Activin, follistatin, and inhibin mRNAs are expressed in LßT2 cells. Approximately 2 µg poly(A)+ mRNA from LßT2, {alpha}T3-1, and NIH-3T3 cells and 10 µg total mouse ovary RNA as a positive control were run on the same agarose gel, then blotted to a filter. Northern blots were probed with radiolabeled rat cDNA fragments coding for activin ßB-subunit, follistatin, and inhibin {alpha}-subunit. GAPDH cDNA was used as a probe to allow visualization of RNA quantities in each lane. Hybridizations were carried out at 55 C overnight, and blots were washed at 65 C, then exposed to x-ray film for 16–72 h.

 
Activin receptor mRNAs are expressed in LßT2 cells
The activin signal is propagated via cell surface serine/threonine receptors that exist in two subtypes: type I (Act RI and Act RIB) and type II (Act RII and Act RIIB) (36). Activin binds to the type II receptors causing association with, and phosphorylation of, the type I receptors (37). To characterize the expression of activin receptors in LßT2 cells, we performed Northern blots with poly(A)+ RNA from LßT2, {alpha}T3–1, and NIH-3T3 cells and probed with mouse cDNA fragments from each of the activin receptors: Act RI, Act RIB, Act RII, and Act RIIB (Fig. 6Go). The expected 3.5-kb mRNA coding for mouse Act RI (38) was observed in all three cell lines tested, as was the 4.4-kb RNA transcript coding for activin receptor type RIB (38). LßT2, {alpha}T3–1, and NIH-3T3 cells also expressed both the 6.9- and 2.1-kb RNA species corresponding to Act RII and the 14-, 11-, 3.8-, and 2.4-kb RNA forms described for Act RIIB (39). Thus, the pituitary-derived cells, LßT2 and {alpha}T3–1, and the fibroblast-derived NIH-3T3 cells synthesize RNA transcripts corresponding to all four types of activin receptors, indicating that these cell lines are capable of responding to activin stimulation. Swiss 3T3 fibroblast cells have been shown to undergo mitogenesis upon activin treatment (40); therefore it is not surprising that activin receptor RNAs are present in NIH-3T3 fibroblasts.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 6. Activin receptor mRNAs are expressed in LßT2 cells. Northern blots with approximately 2 µg poly(A)+ mRNA from LßT2, {alpha}T3-1, and NIH-3T3 cells were probed with radiolabeled mouse cDNA fragments coding for Act RI, Act RIB, Act RII, and Act RIIB. The GAPDH cDNA was used to allow visualization of the quantity of RNA loaded in each lane. Hybridizations were carried out at 55 C overnight, and blots were washed at 65 C, then exposed to x-ray film for 16–72 h. In the case of Act RI, only the 3.5-kb mRNA was detected. A 0.8-kb Act RI short-form mRNA present in spleen, heart, and lung, but not in brain, liver, kidney, or testis, was not detected in any of the cell lines tested. The Act RI receptor short form is only detected when using a probe encompassing the ectodomain region of Act RI (38 ). The cytoplasmic region probe does not cross-hybridize with the short form, suggesting that it could code for a truncated receptor, but this has not been further investigated. The probe we employed contains the full-length cDNA, suggesting that the 0.8-kb form is not present in the cell lines tested. For Act RII, we observed 6.9- and 2.1-kb RNA species. Attardi et al. (20 ) reported an Act RII short-form (0.5-kb) transcript in {alpha}T3-1 cells, which we have not observed. Such a small band may represent a degradation product of the larger mRNAs.

 
The FSHß regulatory region responds to GnRH in LßT2 cells
GnRH is the key releasing hormone controlling the synthesis and secretion of both LH and FSH. LßT2 cells have previously been shown to express GnRH-R (16) and to respond to GnRH (17), as have {alpha}T3–1 cells (41). Furthermore, the rat LHß 5'-flanking region responds to GnRH after transfection into LßT2 cells (42). To investigate the responsiveness of the FSHß regulatory region in LßT2 cells, we transfected the oFSHß-Luc expression vector and treated cells with 1 nM GnRH for 6 h, the optimal time and concentration for induction of this gene (Vasilyev, V., F. Pernasetti, and P. L. Mellon, in preparation). We compared the response of the oFSHß regulatory region to those of the LHß and TK promoters in both LßT2 and {alpha}T3–1 cells (Fig. 7Go). Both oFSHß-Luc and rLHß-Luc were induced by GnRH in LßT2 cells, by 2- and 1.6-fold, respectively. The minimal TK promoter did not respond to GnRH in either of the cells tested. Remarkably, neither of the ß-subunit promoters responded to GnRH treatment in the gonadotrope precursor {alpha}T3–1 cells despite the known GnRH responsiveness of this cell line (21, 41), indicating that under these conditions, the GnRH response of these regulatory regions is LßT2 cell specific.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 7. The GnRH response of the ß-subunit genes is LßT2 cell specific. LßT2 and {alpha}T3–1 cells were transiently transfected with equimolar amounts (0.44 pmol) of the following plasmids: oFSHß-Luc, LHß-Luc, and TK-Luc. RSV-ßgal vector (0.5 µg) was used as an internal control. Forty hours after transfection, the appropriate plates (GnRH) received treatment with 1 nM GnRH for 6 h. Cells were then harvested for luciferase and ß-galactosidase assays. The values of the control untreated samples were set at 1 to facilitate comparisons. Results shown are the mean ± SEM of at least four independent experiments, each performed in duplicate. Bars marked with asterisk are statistically different from the untreated condition (P < 0.05) for the same promoter in the same cell line.

 
Follistatin blocks GnRH induction of both FSHß and LHß gene expression
GnRH and the activin system are the primary physiological regulators of FSH synthesis and secretion, and as we have shown in this study, both up-regulate the ovine FSHß promoter after transfection into LßT2 cells. To investigate the role of endogenous secretion of activin in GnRH regulation of the FSHß gene, transfected LßT2 cells were treated with GnRH in the presence or absence of follistatin. As activin significantly induced the rLHß-Luc plasmid in LßT2 cells (Fig. 4Go), and this promoter is induced by GnRH in LßT2 cells (Fig. 7Go), we also tested whether the response of the LHß promoter to GnRH in these cells is affected by follistatin. All transfected cells were pretreated with follistatin to eliminate the effects of endogenously secreted activin and then tested for GnRH response with or without continued follistatin treatment. GnRH alone stimulates the oFSHß reporter approximately 2-fold, whereas follistatin completely inhibits the stimulatory effect of GnRH (Fig. 8Go), bringing the activity of the oFSHß promoter to the same level as that with follistatin alone. These data strongly suggest that endogenously secreted activin acts within an autocrine loop that is essential for the GnRH response of the FSHß promoter in LßT2 cells.



View larger version (27K):
[in this window]
[in a new window]
  		Figure 8. Follistatin inhibits the GnRH responsiveness of the FSHß and LHß regulatory regions. LßT2 cells were transfected with equimolar amounts (0.44 pmol) of the oFSHß-Luc and LHß-Luc reporter plasmids, each in combination with 0.5 µg RSV-ßgal as the internal control, and were incubated in DMEM and 10% FBS supplemented with follistatin (250 ng/ml) for 16 h. Subsequently, all plates were changed to medium with 1% FBS, but only half of the plates received continued treatment with follistatin (FS; 250 ng/ml) for an additional 24 h. GnRH (1 nM) was added during the last 6 h of this latter 24-h period, before harvest. Luciferase activity was normalized to ß-galactosidase activity, and the follistatin-pretreated condition values were normalized to 1 to facilitate comparisons. Results are the mean ± SEM of at least four independent experiments, each performed in triplicate. The bars marked with a single asterisk (GnRH) are statistically different from all other treatment bars for each promoter, and the bar marked with a double asterisk is statistically different from all other treatment bars within the LHß promoter (P < 0.0017).

 
The LHß promoter exhibits an approximately 2-fold induction in the presence of GnRH (as observed in Fig. 7Go), which is largely blocked (by 70%) by follistatin (Fig. 8Go). This inhibition of GnRH induction is only partial, as the value for the GnRH plus FS is statistically different from the pretreated only condition and the pretreated plus follistatin condition, indicating that follistatin is capable of interfering with the GnRH response of the LHß promoter as well, but to a lesser extent than the oFSHß promoter.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Studies of the regulation of the FSHß gene have progressed slowly largely due to the ponderous nature of primary culture and transgenic mice studies and the limits on information obtainable from transfections of heterologous cells. Previously, we used targeted oncogenesis in transgenic mice to develop a series of pituitary cell lines representing various specific stages of differentiation in specific lineages. The {alpha}T3–1 cells display functional GnRH-R and express both SF-1 and glycoprotein hormone {alpha}-subunit RNAs, indicating that they reside in the gonadotrope lineage (29). As they do not express the FSHß or LHß subunits, we proposed that this cell line represents a fetal gonadotrope progenitor arrested between embryonic days 13.5 and 16.5 of development within the period when GnRH receptors are detectable (43), but before the expression of the LHß gene (44). Our previous observation that LßT2 cells express the LH ß- subunit gene in addition to the gonadotrope markers expressed by the {alpha}T3–1 cells had led us to conclude that LßT2 represented gonadotrope cells arrested between embryonic days 16.5 and 17.5 of mouse development (16), before the expression of the FSH ß-subunit gene (44). Here we have demonstrated that the LßT2 cells do express FSHß mRNA, using a sensitive detection method [RT-PCR on poly(A)+ mRNA], and that the transfected FSHß regulatory region recapitulates gonadotrope specificity. Furthermore, LßT2 cells express the components of the activin and follistatin system as well as responding to activin and GnRH by regulating the gonadotropin ß-subunit genes. Thus, we now conclude that the LßT2 cell represents a gonadotrope with the major hallmarks of the fully differentiated state in the adult mouse pituitary.

FSH ß-subunit mRNA is detectable in untreated LßT2 cells by RT-PCR. It is strongly induced by activin and rendered undetectable by follistatin. In a previous study FSHß mRNA was not observed in untreated LßT2 cells by ribonuclease protection (45), although it increased in a dose-dependent manner in response to activin (45). Our detection of FSHß mRNA in the absence of exogenously added activin is probably due to the acute sensitivity of the RT-PCR assay. Alternatively, different cell culture conditions could account for the different observations by affecting the endogenous production of activin or steroid levels. For example, Graham et al. (45) cultured the cells in 10% charcoal-stripped serum supplemented with dexamethasone (20 nM) in DMEM/Ham’s F-12 (1:1), whereas we used 1% nonstripped serum in DMEM.

Transcription from the ovine FSHß regulatory region is significantly higher in the LßT2 gonadotrope cell line than in other cell types. Expression levels are intermediate in the less differentiated gonadotrope cell line, {alpha}T3–1. Although very little information is available regarding the transcription factors involved in the expression of FSHß, SF-1 is known to play a key role in the expression of the LHß gene, interacting with pituitary homeobox 1 (Ptx1) and early growth response protein 1 (Egr-1) (46). Despite the importance of these proteins in the expression of the LHß gene, SF-1, Ptx-1, and Egr-1 cannot entirely account for the gonadotrope-specific expression of either the LH or FSH ß-subunit genes because they are all present in {alpha}T3–1 cells, which do not express either of the ß-subunit mRNAs (46). Targeted disruption of the Egr-1 gene in mice results in specific loss of LHß gene expression, yet FSHß gene expression is maintained (47). In contrast, a null mutation in the SF-1 gene selectively reduces expression of both gonadotropins in transgenic mice (48), whereas Ptx-1 null mice have decreased numbers of gonadotrope cells and lower levels of LHß and FSHß (49). These observations suggest that both common and distinct proteins are likely to regulate the expression of the LHß and FSHß genes in LßT2 cells.

Expression of activin ßB-subunit and all four activin receptor RNA forms together with the ability to respond appropriately to activin signaling indicate that LßT2 cells accurately recapitulate activin regulation of the FSHß gene in the natural gonadotrope. Our transient transfection experiments with activin show that the oFSHß regulatory region is induced by activin treatment in LßT2 cells. This result is in agreement with experiments performed in primary pituitary cells in which activin causes an increase in FSHß primary transcripts as detected by RT-PCR (50). Cotreatment with the transcriptional inhibitor actinomycin D blocked activin stimulation in primary cells, indicating that the activin response involves transcriptional activation of the FSHß gene (50). Interestingly, the oFSHß regulatory region is not regulated by activin in the gonadotrope precursor cell, {alpha}T3–1. This lack of response does not appear to be due to the absence of activin receptors in {alpha}T3–1 cells, as we have demonstrated that these cells synthesize mRNA for all four activin receptor types. Furthermore, the GnRH-R (19) and {alpha}-subunit (20) genes are regulated by activin in {alpha}T3–1 cells. Thus, the lack of responsiveness of FSHß to activin in {alpha}T3–1 cells implies the involvement of LßT2 cell-specific partners for the downstream activin-signaling molecules that are not present or active in {alpha}T3–1 cells. Smad 2 and Smad 3, members of the novel Smad family of cytoplasmic proteins (51), are downstream targets of activin. Scanning of the oFSHß 5.5-kb regulatory region of the oFSHß gene for Smad protein consensus DNA-binding sites (5'-GTCTAGAC-3') (52) revealed at least 23 sites with 6 or 7 of 8 bases conserved, suggesting a possible role for these proteins in the activin regulation of oFSHß gene expression. The cell specificity of activin regulation of the oFSHß gene may involve Smad proteins present only in more mature gonadotropes, or perhaps interaction of Smad proteins with specific transcription factors that may be present in LßT2 cells, but not in the less mature {alpha}T3–1 cell type. Alternatively, there could be repressor molecules in {alpha}T3–1 cells that are not present in LßT2 cells.

Our experiments comparing the activin responsiveness of several gonadotrope-specific promoters in LßT2 cells demonstrate that these responses are not only cell specific, but also promoter specific. The GnRH-R promoter is strongly stimulated by activin in these cells (6-fold), similar to the approximately 3-fold induction found in {alpha}T3–1 cells (19). Interestingly, activin also induces the {alpha}-GSU promoter by 40% in LßT2, and this induction is blocked by the addition of follistatin. These data are in agreement with observations in primary rat pituitary cultures, where activin had a small, but significant (10–20%), stimulatory effect on {alpha}-subunit synthesis and secretion after 48–72 h (53). However, this is in contrast to previous observations in {alpha}T3–1 cells, where activin was reported to decrease transcription of the {alpha}-GSU promoter by approximately 50% (20). These differences in responsiveness of the {alpha}-GSU promoter may be due to the presence of distinct activin response mechanisms in each cell line, suggesting distinct activin transcriptional regulation of the {alpha}-subunit gene at different stages of gonadotrope development. Attardi et al. (20) suggested that activin might suppress {alpha}-subunit gene expression in the embryonic pituitary until the specific ß-subunit genes are expressed. Our observation that the {alpha}-subunit promoter is stimulated by activin in LßT2 cells, which express both ß-subunits, supports this hypothesis. Activin regulation of {alpha}-subunit gene expression may recapitulate a developmental program in the pituitary gonadotropes and deserves further exploration. Another possibility is that these different observations result from distinct experimental conditions. We performed our experiments in DMEM with 1% FBS, pretreating the LßT2 cells with 250 ng/ml follistatin and treating with 100 ng/ml activin, whereas the {alpha}T3–1 cells (20) were grown in Eagle’s MEM with 10% dextran-coated charcoal-stripped FBS, no pretreatment was performed, and only 10 ng/ml activin was added. The different results obtained in stripped vs. whole serum might indicate a role for steroid hormones or other growth factors in this regulatory process.

We also observed that the LHß promoter is significantly stimulated by exogenous activin in LßT2 cells, and this effect is blocked by follistatin (Fig. 4Go). The negative controls, TK-Luc and pGL3, did not respond to any of the treatments, supporting the specificity of the LHß promoter responses under these experimental conditions. Interestingly, the pretreatment with follistatin to remove endogenous activin does not have a significant repressive effect on the LHß promoter, indicating that higher concentrations of activin might be necessary to observe regulation of LHß transcription. To our knowledge, these data are the first to demonstrate an effect of activin on transcription of the LHß gene. Several reports have failed to detect an effect of activin on LH secretion (54, 55). However, Stouffer et al. (56) have shown acute and sustained increases in LH secretion in response to activin in female rhesus monkeys. These researchers observed a 2-fold increase in LH within 8 h, and the levels remained 10-fold higher than control values after 4–5 days. McLachlan and co-workers (57) reported that 2-day activin infusion in adult male monkeys significantly increased LH as well as FSH release in response to GnRH injection. It has also been previously demonstrated that treatment of primary pituitary cell cultures with activin for 24 h results in small, but significant, increases in LH secretion and cell content as well as in steady state LHß mRNA levels (53). These data indicate that activin might be a physiological regulator of LHß transcription and that the LßT2 cells can serve as a model in which to further investigate the regulatory pathways involved in this response.

Our transient transfection experiments show that continuous GnRH treatment regulates oFSHß in LßT2 cells, with a maximal 2-fold increase in reporter activity at 1 nM GnRH after a 6-h incubation. This is in agreement with transient transfection experiments in heterologous HeLa and COS-7 cells cotransfected with expression vectors for the GnRH-R, in which the 5.5-kb regulatory region of oFSHß is stimulated 3.1- and 2.7-fold, respectively, by treatment with 100 nM GnRH for 12 h (14).

Surprisingly, we observed that FSHß regulation by GnRH is LßT2 cell specific (Fig. 7Go) compared with the less differentiated gonadotrope, {alpha}T3–1. We have previously shown that {alpha}T3–1 cells synthesize GnRH-R and are capable of responding to GnRH by inducing {alpha}-GSU transcription (41). The LßT2 cell specificity of the GnRH response suggests that this effect most likely involves factors present only in this highly differentiated gonadotrope cell line. We observed the same cell specificity in the activin response of oFSHß (Fig 3Go), reinforcing the hypothesis that the LßT2 cell line represents a highly differentiated stage in gonadotrope development, and that these cells synthesize specific sets of transcription factors involved in the molecular pathways of both activin and GnRH responses.

Interestingly, the response of the LHß promoter to GnRH is also LßT2 cell specific under our experimental conditions. Previous work in {alpha}T3–1 cells by Weck and co-workers (58) showed transcriptional regulation by GnRH of a transiently transfected fragment of the rat LHß promoter, containing only the -617 to -245 bp region fused to a minimal TK promoter. This region was induced approximately 2-fold by a 6-h treatment with 10 nM GnRH, whereas the TK promoter did not respond under the same conditions. These observations suggest that the full LHß promoter might contain additional GnRH regulatory regions, including possible repressive elements, that are not present in the small promoter region from -617 to -245 bp. Thus, we hypothesize that in the context of the precursor gonadotrope {alpha}T3–1 cells, which do not synthesize gonadotropin ß-subunits, the intact rat LHß promoter is not induced by GnRH, perhaps due to inhibitory factors expressed in these cells that are not present or active in LßT2 cells. Alternatively, {alpha}T3–1 cells could lack key factors that allow the LHß and FSHß regulatory regions to respond to GnRH or, for that matter, activin.

The observation that follistatin blocks GnRH stimulation of the regulatory region of the ß-subunit genes in transient transfections in LßT2 cells suggests that activity of the activin autocrine loop is required for GnRH stimulation of these genes. These data are in agreement with the observation that GnRH stimulation of FSHß endogenous mRNA is dependent on activin and blocked by follistatin both in perifusion of primary pituitary cells and in vivo (59). However, follistatin has also been shown to bind to inhibin (60), bone morphogenic protein-4 (BMP4), and BMP7 (61), albeit with much lower affinity than that for activin. We have previously described the presence of BMP7 mRNA transcripts in LßT2 cells, by RT-PCR, whereas BMP4 and BMP2 transcripts were not detected (62). However, although BMP7 is capable of binding to Act RI, Act RII, and Act RIIB, with affinities 3-fold lower than activin, it did not stimulate FSH secretion from primary rat pituitary cells (63). These findings do not exclude the possibility that BMPs and transforming growth factor-ß family members play a role in FSHß gene transcription, but strongly suggest an important role for the interaction between activin and follistatin in this process.

The follistatin blockade of GnRH responsiveness of both the FSHß and LHß promoters (Fig. 8Go) suggests two possible mechanisms. The first involves a direct effect of the activin/follistatin system on the signaling pathways employed by GnRH to exert its activity, and the second mechanism involves regulation of the GnRH-R itself, and hence GnRH sensitivity, by the activin/follistatin system. Taken together, our data provide evidence that both of these mechanisms may play roles in follistatin blockade of the GnRH response. The 6-fold stimulation by activin of the GnRH-R regulatory region is completely blocked by follistatin. In addition, this regulatory region is inhibited 2-fold by follistatin treatment alone, indicating that levels of GnRH-R may be reduced after follistatin treatment alone, probably resulting in a reduction of the responsiveness to GnRH. These data tend to support the second mechanism. On the other hand, although the GnRH response of the FSHß promoter is completely blocked by follistatin, the LHß promoter response to GnRH is only reduced by 70% in the presence of follistatin. This suggests that the inhibition of GnRH-R synthesis in these cells might not be the only pathway through which follistatin is interfering with the GnRH response. Thus, at least in the case of LHß gene regulation, cross-talk between the activin and GnRH signal transduction pathways might play a role in the dependence of GnRH action on the activin/follistatin system.

In conclusion, LßT2 cells are a unique gonadotrope cell model in which to study regulation of the LHß and FSHß genes. Our studies place these cells in the category of a fully differentiated gonadotrope, expressing the full spectrum of characteristic markers for these cells: {alpha}-subunit, SF-1, activin ßB-subunit, follistatin, inhibin, activin receptors, GnRH-R, and LHß and FSHß subunits. The oFSHß regulatory region is specifically active in LßT2 cells, compared with other pituitary and nonpituitary cell lines, and its response to FSH’s main physiological regulators, activin and GnRH, is LßT2 cell specific. We have also revealed a requirement for an endogenous activin autocrine loop for GnRH regulation of the gonadotropin ß-subunit genes, indicating interaction between GnRH action and the activin/follistatin system in LßT2 cells.

Identification of the transcriptional elements and proteins involved in the regulation of both FSHß and LHß genes is now feasible using the homologous gonadotrope LßT2 cell model system. Such studies will shed further light on the mechanisms of action and likely interactions between activin and GnRH signaling pathways in the gonadotrope. Understanding the regulation of LH and FSH gene expression in gonadotrope cells will contribute critical knowledge for clinical applications in a variety of endocrine and reproductive conditions.


   Acknowledgments
 
The authors thank Sandra Holley for discussions and reading of the manuscript; Scott Anderson, Melinda Merrill, Kevin Rufner, and Greg DeFuria for technical support; Yuka Kimura and Sandra Holley for the construction of the GnRH-R promoter plasmid (pGnRH-R-Luc); and Karen Kartun for unpublished information about activin receptors in pituitary cell lines. We thank A. F. Parlow, from the National Hormone and Pituitary Program (Torrance, CA) and Shunichi Shimasaki for kindly providing activin and follistatin, Sylvia Evans for the pT81 plasmid, James Roberts for the rat LHß promoter fragment, and Richard Maurer for the mouse {alpha}-subunit promoter plasmid. Probes for the activin receptors, follistatin, and activin/inhibin subunits were kindly provided by W. Vale, S. Shimasaki, and V. Roberts.


   Footnotes
 
1 This work was supported by the NICHHD/NIH through Cooperative Agreement U54-HD-12303 as part of the Specialized Cooperative Centers Program in Reproduction Research. This work was also supported by NIH Grants R37-HD-20377 (to P.L.M.) and R01-HD-34863 (to W.L.M.). Back

2 Supported by an American Heart Association postdoctoral fellowship and a fellowship from the Lalor Foundation. Back

3 Supported by the NIH Training Grant T32-DK-07541. Back

4 V.V.V. and S.B.R. contributed equally to this work. Back

5 Supported by the NIH Training Grant T32-GM-08666. Back

Received October 27, 2000.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Pierce JG, Parsons TF 1981 Glycoprotein hormones: structure and function. Annu Rev Biochem 50:465–495[CrossRef][Medline]
  2. Vale W, Rivier C, Brown M 1977 Regulatory peptides of the hypothalamus. Annu Rev Physiol 39:473–527[CrossRef][Medline]
  3. Shupnik MA 1996 Gonadotropin gene modulation by steroids and gonadotropin-releasing hormone. Biol Reprod 54:279–286[Abstract]
  4. Gregg DW, Schwall RH, Nett TM 1991 Regulation of gonadotropin secretion and number of gonadotropin-releasing hormone receptors by inhibin, activin-A, and estradiol. Biol Reprod 44:725–732[Abstract]
  5. Braden TD, Conn PM 1992 Activin-A stimulates the synthesis of gonadotropin-releasing hormone receptors. Endocrinology 130:2101–2105[Abstract]
  6. Ling N, Ying S-Y, Ueno N, Shimasaki S, Esch F, Hotta O, Guillemin R 1986 Pituitary FSH is released by a heterodimer of the ß-subunits from the two forms of inhibin. Nature 321:779–782[CrossRef][Medline]
  7. Carroll RS, Corrigan AZ, Gharib SD, Vale W, Chin WW 1989 Inhibin, activin, and follistatin: regulation of follicle-stimulating hormone messenger ribonucleic acid levels. Mol Endocrinol 3:1969–1976[CrossRef][Medline]
  8. Vale W, Hseuh A, Rivier C, Yu J 1990 The inhibin/activin family of hormones and growth factors. In: Sporn M, Roberts A (eds) Peptide Growth Factors and Their Receptors: Handbook of Experimental Pharmacology. Springer-Verlag, Berlin, vol 95:211–248
  9. Mathews LS 1994 Activin receptors and cellular signaling by the receptor serine kinase family. Endocr Rev 15:310–324[CrossRef][Medline]
  10. Meunier H, Rivier C, Evans RM, Vale W 1988 Gonadal and extragonadal expression of inhibin {alpha}, ßA, and ßB subunits in various tissues predicts diverse functions. Proc Natl Acad Sci USA 85:247–51[Abstract/Free Full Text]
  11. Yins S-Y 1988 Inhibins, activins, and follistatins: gonadal proteins modulating the secretion of follicule-stimulating hormone. Endocr Rev 9:267–293[CrossRef][Medline]
  12. Shimasaki S, Koga M, Esch F, Cooksey K, Mercado M, Koba A, Ueno N, Ying S-Y, Ling N, Guillemin R 1988 Primary structure of the human follistatin precursor and its genomic organization. Proc Natl Acad Sci USA 85:4218–4222[Abstract/Free Full Text]
  13. Strahl BD, Huang HJ, Pedersen NR, Wu JC, Ghosh BR, Miller WL 1997 Two proximal activating protein-1-binding sites are sufficient to stimulate transcription of the ovine follicle-stimulating hormone-beta gene. Endocrinology 138:2621–2631[Abstract/Free Full Text]
  14. Strahl BD, Huang HJ, Sebastian J, Ghosh BR, Miller WL 1998 Transcriptional activation of the ovine follicle-stimulating hormone ß-subunit gene by gonadotropin-releasing hormone: involvement of two activating protein-1-binding sites and protein kinase C. Endocrinology 139:4455–4465[Abstract/Free Full Text]
  15. Kumar TR, Low MJ 1995 Hormonal regulation of human follicle-stimulating hormone-ß subunit gene expression: GnRH stimulation and GnRH-independent androgen inhibition. Neuroendocrinology 61:628–637[Medline]
  16. Alarid ET, Windle JJ, Whyte DB, Mellon PL 1996 Immortalization of pituitary cells at discrete stages of development by directed oncogenesis in transgenic mice. Development 122:3319–3329[Abstract]
  17. Turgeon JL, Kimura Y, Waring DW, Mellon PL 1996 Steroid and pulsatile gonadotropin-releasing hormone (GnRH) regulation of luteinizing hormone and GnRH receptor in a novel gonadotrope cell line. Mol Endocrinol 10:439–450[Abstract]
  18. Webster JC, Pedersen NR, Edwards DP, Beck CA, Miller WL 1995 The 5'-flanking region of the ovine follicle-stimulating hormone-ß gene contains six progesterone response elements: three proximal elements are sufficient to increase transcription in the presence of progesterone. Endocrinology 136:1049–1058[Abstract]
  19. Fernandez-Vazquez G, Kaiser UB, Albarracin CT, Chin WW 1996 Transcriptional activation of the gonadotropin-releasing hormone receptor gene by activin A. Mol Endocrinol 10:356–366[Abstract]
  20. Attardi B, Klatt B, Little G 1995 Repression of glycoprotein hormone {alpha}-subunit gene expression and secretion by activin in {alpha}T3–1 cells. Mol Endocrinol 9:1737–1749[Abstract]
  21. Schoderbek WE, Kim KE, Ridgway EC, Mellon PL, Maurer RA 1992 Analysis of DNA sequences required for pituitary-specific expression of the glycoprotein hormone {alpha}-subunit gene. Mol Endocrinol 6:893–903[Abstract]
  22. Nordeen SK 1988 Luciferase reporter gene vectors for analysis of promoters and enhancers. BioTechniques 6:454–458[Medline]
  23. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning: A Laboratory Manual, Ed 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor
  24. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159[Medline]
  25. Kumar TR, Kelly M, Mortrud M, Low MJ, Matzuk MM 1995 Cloning of the mouse gonadotropin ß-subunit-encoding genes. I. Structure of the follicle-stimulating hormone ß-subunit-encoding gene. Gene 166:333–334[CrossRef][Medline]
  26. Maurer RA 1987 Molecular cloning and nucleotide sequence analysis of complementary deoxyribonucleic acid for the ß-subunit of rat follicle stimulating hormone. Mol Endocrinol 1:717–723[CrossRef][Medline]
  27. Feinberg AP, Vogelstein B 1983 A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6–13[CrossRef][Medline]
  28. Alarid ET, Holley S, Hayakawa M, Mellon PL 1998 Discrete stages of anterior pituitary differentiation recapitulated in immortalized cell lines. Mol Cell Endocrinol 140:25–30[CrossRef][Medline]
  29. Windle JJ, Weiner RI, Mellon PL 1990 Cell lines of the pituitary gonadotrope lineage derived by targeted oncogenesis in transgenic mice. Mol Endocrinol 4:597–603[CrossRef][Medline]
  30. Mellon PL, Windle JJ, Goldsmith P, Pedula C, Roberts J, Weiner RI 1990 Immortalization of hypothalamic GnRH neurons by genetically targeted tumorigenesis. Neuron 5:1–10[CrossRef][Medline]
  31. Attardi B, Winters SJ 1993 Decay of follicle-stimulating hormone-ß messenger RNA in the presence of transcriptional inhibitors and/or inhibin, activin, or follistatin. Mol Endocrinol 7:668–680[Abstract]
  32. Feng Z-M, Bardin CW, Chen C-LC 1989 Characterization and regulation of testicular inhibin ß-subunit mRNA. Mol Endo 3:939–948[CrossRef][Medline]
  33. Roberts V, Meunier H, Vaughan J, Rivier J, Rivier C, Vale W, Sawchenko P 1989 Production and regulation of inhibin subunits in pituitary gonadotropes. Endocrinology 124:552–554[Abstract]
  34. Michel U, Farnworth P, Findlay JK 1993 Follistatins: more than follicle-stimulating hormone suppressing proteins. Mol Cell Endocrinol 91:1–11[CrossRef][Medline]
  35. Roberts VJ, Barth SL 1994 Expression of messenger ribonucleic acids encoding the inhibin/activin system during mid- and late-gestation rat embryogenesis. Endocrinology 134:914–923