| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
T3-1 Gonadotrope Cells Is Stimulated by Cyclic 3',5'-Adenosine Monophosphate and Protein Kinase A, and Is Modulated by Steroidogenic Factor-1 and Nur77
Department of Biochemistry, University of Stellenbosch, Stellenbosch 7600, South Africa
Address all correspondence and requests for reprints to: Dr. Janet Hapgood, Department of Biochemistry, University of Stellenbosch, Stellenbosch 7600, South Africa. E-mail: jhap{at}sun.ac.za.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T3-1 cells are stimulated by forskolin and 8-bromo-cAMP. We also show that the orphan nuclear receptor Nur77 is expressed in T3-1 cells, and that both SF-1 and Nur77 bind to the –15/-7NRS and –244/-236NRS in vitro. We show that the activity of the proximal (-579/+1) mouse GnRHR promoter is up-regulated by protein kinase A, via a mechanism that is modulated by SF-1, both positively and negatively, through binding to the –244/-236NRS or the –15/-7NRS, respectively. Nur77 appears to be capable of acting as a negative regulator of this response, via the –15/-7NRS. Furthermore, we show that forskolin up-regulates SF-1 mRNA levels in T3-1 cells, indicating that the levels of SF-1 play a role in modulating the protein kinase A response. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T3-1 cells (6), gonadotrope-specific regulatory elements are present in a 1.2-kb 5'-flanking genomic fragment (-1226/+12 relative to the start codon) of the mouse GnRHR gene (7) (all numbering from here on will be relative to the translation start site). A tripartite basal gonadotropic-specific enhancer occurs within the same promoter fragment and includes a binding site for the orphan nuclear receptor, steroidogenic factor-1 (SF-1), at –244/-236, a consensus activator protein-1 (AP-1) site at –336/-330, and a noncanonical key element termed GnRHR-activating sequence at –391/-380 (8, 9, 10). Cell-specific and hormonal regulation of the mouse GnRHR gene in T3-1 cells partially converge at the tripartite enhancer element, with GnRH responsiveness primarily conferred through protein kinase C (PKC) activation of the AP-1 element (11), and activin responsiveness conferred through the GnRHR-activating sequence element (12).
There is evidence that both the protein kinase A (PKA) and PKC pathways are involved in stimulation of mouse GnRHR gene transcription. However, results obtained in different model cell lines appear to differ, most likely reflecting the different repertoires of G proteins and/or transcription factors. Using rat pituitary GGH3 cells, a GH3-derived cell line stably expressing the rat GnRHR (13), it was shown that the response of the mouse GnRHR gene to GnRH involves stimulation by cAMP and the PKA pathway (14, 15), stimulation by Ca2+-dependent pathways and the PKC pathway (16), as well as inhibition of GnRH-stimulated activity by the MAPK pathway (16). A cAMP response element (CRE) at –107/-100 mediates the response to both cAMP and GnRH in GGH3 cells (15). In contrast, experiments in T3-1 cells have shown that GnRH stimulation of the expression of a –600 mouse GnRHR promoter-reporter construct is mediated in full by the PKC pathway and via the AP-1 site at –336/-330, but not via the PKA pathway (11). These results were supported by Norwitz et al. (17), who also identified another element at –354/-347 that is required for the full GnRH response. These results in T3-1 cells are consistent with the report that the endogenous receptor in T3-1 cells is coupled exclusively to Gq/11 proteins (18). In contrast, the endogenous GnRHR in LßT2 cells, another mouse gonadotrope cell line, is coupled to both Gq/11 and Gs proteins (19). Interestingly, cultured rat pituitary tissue responds to GnRH stimulation by accumulating cAMP intracellularly (20), indicating that in primary cells, the GnRHR is coupled to Gs proteins. Thus, although it has been consistently demonstrated that transcription of the mouse GnRHR gene in T3-1 cells is up-regulated by GnRH via a mechanism involving the PKC pathway, but not the PKA pathway, whether the mouse gene is regulated by the PKA pathway in response to other signals seems less clear. White et al. (11) found that activation of the PKA pathway for 4 h by forskolin had no effect on the basal activity of a -600 mouse GnRHR promoter fragment. In contrast, for the rat GnRHR gene, in which both the SF-1 site (-244/-237) and the imperfect CRE (-110/-103) are conserved (see Fig. 1
), it has been shown that pituitary adenylate cyclase-activating polypeptide (PACAP) up-regulates promoter activity via PKA in T3-1 cells after 16 h via these elements (21). Very few studies have investigated the effects of activation of kinase pathways on endogenous GnRHR mRNA levels. Although Norwitz et al. (17) showed that the endogenous GnRHR mRNA levels in T3-1 cells are up-regulated by GnRH, this result is in contrast to the findings of Tsutsumi et al. (22), who reported that mRNA activity, but not mRNA levels, was up-regulated in response to GnRH in these cells. Whether cAMP regulates endogenous GnRHR mRNA levels in T3-1 cells has not been previously investigated to our knowledge.
|
), referred to as the –15/-7 nuclear receptor site (-15/-7NRS). The –15/-7NRS does not appear to be necessary for high basal and cell-specific expression (8) or for responsiveness to GnRH stimulation (17) in T3-1 cells. Nevertheless, the sequence of the –15/-7NRS (TGTCCTTGG) closely resembles that of the established SF-1 binding site at –244/-236 (TGGCCTTCA) in the mouse GnRHR promoter. The binding of T3-1 nuclear proteins to the –15/-7NRS has not been previously investigated. In this paper we investigated the role of –15/-7NRS and –244/-236NRS and their cognate binding proteins in basal and cAMP-responsive gene regulation in T3-1 cells. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Plasmids
pGL2-basic promoterless luciferase expression vector, pGL2-basic vector-specific primers GL1 (sense) and GL2 (antisense), and pSV-ß-galactosidase expression vector were purchased from Promega Corp. (Madison, WI). The cytomegalovirus promoter (pCMV)-SF-1 expression construct containing the full-length (2-kb) human SF-1 cDNA was obtained from Dr. Parker (University of Texas, Dallas, TX) (23). For in vitro transcription-translation, the full-length cDNA was excised with EcoRI and cloned into pSPT19 (Roche, Indianapolis, IN). Mouse Nur77 cDNA (2 kb) in the pCMX-Nur77 expression vector was obtained from Dr. Drouin (Institut de Recherches Cliniques de Montréal, Montréal, Canada). The pFC-PKA catalytic subunit expression vector was a gift from Dr. Vanden Berghe (Royal University of Ghent, Ghent, Belgium). The full-length (2.1-kb) human fibroblast cytoplasmic ß-actin cDNA in the Okayama-Berg expression vector (24) was obtained from Dr. Parker (University of Cape Town, Cape Town, South Africa). Mouse GnRHR cDNA (2.1 kb) (25) in the pcDNAI/Amp vector (Invitrogen, San Diego, CA) as well as clone 111 containing 1.1 kb of the mouse GnRHR gene in pBluescript SK were obtained from Dr. Sealfon (Mt. Sinai Medical School, New York, NY). The ATG start codon of clone 111 in pBluescript SK was mutated to a BglII site using oligonucleotide 1S to yield plasmid pGB. Plasmid pGBM, where the two central Cs of the -15/-7NRS region were mutated to two Ts, as well as the ATG start codon mutated to a BglII site were also created from clone 111 in pBluescript SK by a similar strategy, using oligonucleotide 2S. Thereafter, 591-bp BglII/BamHI fragments were excised from pGB and pGBM and cloned into the BglII site of pGL2-basic to yield pLG and pLGM1, respectively. The sequence in the region of the -15/-7NRS for pLG is as follows: GCCTGTCCTTGGAGAAAATAgatctaag, where capital letters denote the wild-type GnRHR sequences, bold letters denote the original position of the ATG codon, and the underlined region denotes the wild-type -15/-7NRS region. The –244/-236NRS was mutated to a PstI site by PCR using overlapping primers –244/-236mps and –244/-236mpa, containing the modified –244/-236NRS sequence. Using pLG as template, two separate PCR reactions were performed, with either GL1 and –244/-236mpa or GL2 and –244/-236mps as primers, generating products of approximately 400 and 300 bp, respectively. Subsequently, these products were used in a PCR fusion reaction with primers GL1 and GL2, generating a product of approximately 700 bp containing the –579/+1 promoter sequence with the modified –244/-236NRS. The product was subsequently cloned back into SacI and HindIII sites of pGL2-basic, to yield pLGM2. All reporter constructs were sequenced for confirmation. For oligonucleotide sequences, refer to Table 1.
|
T3-1 mouse pituitary gonadotrope cells. Cells were cultured in high glucose (4.5 g/ml) DMEM (Life Technologies, Inc., Gaithersburg, MD). Medium was supplemented with 10% fetal calf serum (Highveld Biologicals, Johannesburg, South Africa), 50 IU/ml penicillin, and 50 µg/ml streptomycin (Life Technologies, Inc.) unless stated otherwise. Cultures were maintained in 75-cm2 culture flasks (TPP, Switzerland) at 37 C in an atmosphere of 90% humidity and 5% CO2.
RNA isolation
T3-1 cells were plated in 100-mm petri dishes at a density of 2 x 106 cells/dish. For Northern blotting, test compounds were added to fresh medium supplemented with 10% serum. Incubations were performed as described in the figure legends. Subsequently, the medium was removed, and the cells were harvested in 1 ml Tri-reagent (Sigma-Aldrich Corp.)/dish. Total RNA was isolated according to the Tri-reagent protocol (26).
Rapid amplification of 5' cDNA ends (5'RACE)
Total RNA from male mouse pituitaries or T3-1 cells was subjected to 5'RACE using a 5'RACE system (Life Technologies, Inc.) according to the protocol supplied (catalogue no. 18347-025). Briefly, this involved single-stranded cDNA synthesis using a gene-specific antisense primer (GSP2) and homopolymeric tailing, followed by amplification of the dC-tailed cDNA by PCR using an anchor primer and a nested GSP2. A second nested PCR was performed using a universal anchor primer and an internally nested GSP3. Thereafter, PCR products were analyzed by agarose gel electrophoresis, cloned, and subjected to DNA sequence analysis. For oligonucleotide sequences, refer to Table 1.
Northern blots
The cDNA probes (25 ng) were labeled with 50 µCi [-32P]deoxy-CTP (AEC, Amersham Pharmacia Biotech, Arlington Heights, IL) using the Megaprime DNA Labeling System (Amersham Pharmacia Biotech). Total RNA (20 µg/well) was separated on a 1% denaturing formaldehyde agarose gel in 1x morpholinepropanesulfonic acid buffer, transferred overnight onto a Hybond N+ membrane (Amersham Pharmacia Biotech) in 20x standard saline citrate (SSC) buffer, and cross-linked to the membranes. The membranes were prehybridized in DIG EasyHyb solution (Roche) at 50 C for 30 min. Labeled DNA probes (108–109 dpm/µg DNA) were denatured, and hybridization (25 ng DNA/100 cm2) was performed overnight at 50 C. After hybridization, membranes were washed twice for 5 min each time at room temperature in a 2x SSC buffer containing 0.1% sodium dodecyl sulfate, followed by two washes at 50 C in a 0.1x SSC buffer containing 0.1% sodium dodecyl sulfate. The washed membranes were exposed to Hyperfilm (Amersham Pharmacia Biotech) at -80 C. Membranes were stripped, followed by probing for ß-actin. Quantification of signals was performed by densitometric analysis of the autoradiograms.
Transient transfections
T3-1 cells were plated in 12-well culture plates (Nunc, Naperville, IL) at 2 x 105 cells/well in a volume of 1 ml DMEM containing 10% fetal calf serum and antibiotics as described above. Twenty-four hours after plating, medium was replaced with fresh antibiotics-free medium. Transfections were subsequently performed according to the Fugene 6 (Roche) product protocol. One microgram of luciferase reporter construct, 0.5 µg simian virus promoter-ß-galactosidase, and expression constructs, as indicated in relevant figure legends, together with Fugene 6 reagent at a ratio of 1 µg DNA/2 µl Fugene 6 in 100 µl serum-free, high glucose DMEM were added per well. Thirty hours after transfection, the medium was replaced with fresh antibiotic-free medium, and cells were incubated with various test compounds as indicated in the figure legends. Cells were harvested in 100 µl lysis buffer (Galacto-Star assay system, Tropix, Bedford, MA)/well. The ß-galactosidase assays (Galacto-Star assay system) and luciferase assays (Promega Corp. luciferase assay system) were performed with 20 µl cell extract in black 96-well plates in a Luminoskan RS luminometer (Labsytems, Chicago, IL). The luciferase assay values were divided by the ß-galactosidase assay values to correct for differences in transfection efficiency, and these values were then expressed relative to the pLG control value, which was taken as 1.
Nuclear extracts
Twenty 75-cm2 flasks of T3-1 cells were grown to confluence and washed twice with PBS. Nuclear extracts were prepared at 4 C essentially as described previously (27). The cells were harvested in PBS, and a crude nuclear pellet was obtained as described. This was resuspended in 4 vol buffer A [10 mM HEPES (pH 7.9), 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol (DTT), and 0.05 mM phenylmethylsulfonylfluoride (PMSF)]. Nuclei were collected by centrifugation at 25,000 x g for 20 min and resuspended in 500 µl buffer C [20 mM HEPES (pH 7.9), 1.5 mM MgCl2, 1 mM EDTA, 25% (vol/vol) glycerol, 0.5 mM DTT, and 0.05 mM PMSF]. Nuclei were lysed in 250 µl 1 M KCl while gently rocking for 30 min. Subsequently, the insoluble debris was pelleted at 25,000 x g for 30 min. The supernatant was dialyzed three times for 2 h each time in 200 vol buffer D [10 mM HEPES (pH 7.9), 5 mM MgCl2, 50 mM KCl, 1 mM EDTA, 10% (vol/vol) glycerol, 3 mM DTT, and 0.3 mM PMSF] and stored in aliquots at -80 C.
EMSAs
Probes (10 pmol double-stranded oligonucleotides) were end-labeled with 100 µCi [
-32P]ATP (Amersham Pharmacia Biotech) and 10 U Escherichia coli polynucleotide kinase (Roche). Protein-DNA incubations were performed for 10 min at room temperature in a final volume of 5 µl in assay buffer containing 100 mM NaCl, 7.5 mM HEPES (pH 7.9), 10 mM Tris-HCl, 8% (vol/vol) glycerol, 1.25 mM MgCl2, 1 mM EDTA (pH 8.0), 0.25 mM EGTA (pH 7.0), 0.67 mM PMSF, and 3.5 mM DTT, containing 0.5 µg poly(dI-dC) (Roche) and 6 µg BSA (Roche, molecular biology grade). Labeled probe (0.1 pmol; 150,000–250,000 cpm) and 1.5 µl in vitro transcribed-translated protein (diluted 1:7 in 1x assay buffer), CCRF-CEM nuclear extracts (diluted 1:9 in 1x assay buffer, unless indicated otherwise in figure legends), or T3-1 nuclear extracts (1–2 µg protein) were used for each assay. In vitro transcription-translation of SF-1 cDNA (2 µg) was performed according to the Promega Corp. TnT Quick Coupled Transcription/Translation protocol. For assays containing anti-SF-1 (undiluted), anti-Nur77 (Perlmann; diluted 1:1.5 in 1x assay buffer), or preimmune antisera (diluted 1:29 in 1x assay buffer), reactions were performed as described above, followed by further incubation for 10 min at room temperature after the addition of 1 µl specific antibody or rabbit preimmune serum. For competition assays, double-stranded oligonucleotide competitor DNA, labeled probe, and proteins were added sequentially to the assay mix and incubated for 10 min at room temperature. EMSAs were performed using 4% polyacrylamide gels (29:1 acrylamide/bis-acrylamide) at 100 V in 1x Tris acetate EDTA buffer for 3.5 h at room temperature. The gels were transferred to blotting paper, dried, and exposed to Hyperfilm at -80 C. For oligonucleotide sequences, refer to Table 1.
SDS-PAGE and Western blots
Protein samples containing either 5 µl undiluted in vitro transcribed-translated protein or varying amounts of nuclear extracts were separated on an 8% SDS-PAGE gel for 5 h at 120 V in 25 mM Tris-HCl, 250 mM glycine (Merck \|[amp ]\| Co., Inc., Rahway, NJ; reagent grade), and 0.1% sodium dodecyl sulfate, pH 8.3 (28). Proteins were transferred onto a nitrocellulose membrane (BA 85, Schleicher \|[amp ]\| Schuell, Inc., Keene, NH) by electroblotting for 16 h at 120 mA in 25 mM Tris-HCl, 192 mM glycine, and 20% (vol/vol) methanol, pH 8.3. Membranes were blocked with casein buffer (154 mM NaCl, 0.5% casein, 10 mM Tris base, and 0.02% thiomersol) at room temperature for 1 h and incubated overnight with anti-SF-1 antiserum (1:3000) at 4 C or blocked overnight in casein buffer at 4 C and incubated with anti-Nur77 antibody (1:2000) at 37 C for 2 h. Incubations with the secondary and peroxidase-antiperoxidase antibodies were each performed at 37 C for 90 min at 1:1000 and 1:5000 dilutions, respectively. All antibody dilutions were made in casein buffer. After each antibody incubation, the membrane was washed for 15 min in PBS and for 5 min in PBS containing 0.1% (vol/vol) Tween at room temperature. Protein-antibody complexes were visualized by enhanced chemiluminescence detection (RPN2106, Amersham Pharmacia Biotech).
Statistical analysis
Results were analyzed with PRISM software (version 3.1, GraphPad Software, Inc., San Diego, CA) using one-way ANOVA with either Dunnett’s posttest (when comparing all values to a single control) or Bonferroni’s posttest (when comparing all values to each other).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T3-1 cells and mouse pituitary tissue, upstream from the –15/-7NRST3-1 cells and mouse pituitary tissue was analyzed by 5'RACE. The gene-specific primer used in the final PCR reaction (corresponding to position +32/+57) revealed a single distinct band at about 170 bp for both T3-1 and pituitary RNA upon electrophoretic analysis, corresponding to a start site around -60 relative to the start codon. No larger PCR products corresponding to further upstream sites were detected for several repeat experiments. Cloning of the 170-bp product and sequencing of several clones revealed that the major transcription start site for both T3-1 cells and mouse pituitary tissue occurs at the A residue at position -62 relative to the start codon.
Activators of the PKA pathway result in an increase in endogenous GnRHR mRNA
To determine whether the endogenous GnRHR gene in T3-1 cells is responsive to elevated levels of cAMP and forskolin, the cells were incubated for 16 h in the absence and presence of either forskolin or 8-bromo-cAMP (8-Br-cAMP). Total RNA was isolated and probed by Northern blotting with radiolabeled mouse GnRHR cDNA (Fig. 2A). The results presented in Fig. 2 show that both 8-Br-cAMP and forskolin resulted in a significant (
2.5-fold) increase in GnRHR mRNA (4.5 kb) levels relative to levels in unstimulated cells. Furthermore, forskolin induced the expression of a second, smaller mRNA isoform (1.8 kb) in a concentration-dependent fashion, with 100 µM forskolin resulting in markedly increased expression of this isoform compared with expression at 10 µM.
|
T3-1 nuclear extracts and recognize the -15/-7NRS in vitro, A and B) when T3-1 nuclear extracts were incubated with a radiolabeled double-stranded probe containing the 20 bp of promoter sequence encompassing the –15/-7NRS (Fig. 3A, lane 4, and Fig. 3B, lane 5). Complex I was supershifted in the presence of anti-SF-1 antiserum, indicating that SF-1 protein in T3-1 nuclear extracts binds to –15/-7NRS (Fig. 3A, lanes 5 and 6). However, no change in the intensity of complex II was observed in the presence of anti-SF-1 antibody.
|
, lanes 6 and 7). The anti-Nur77 antiserum itself did not form any complexes with the –15/-7NRS probe (Fig. 3B, lane 4). Interestingly, a consistent increase in the intensity of complex I was also observed in these lanes, most likely due to stabilization of the complex by serum proteins. In vitro transcribed-translated SF-1 protein (Fig. 3A, lanes 1–3) and Nur77-containing CCRF-CEM nuclear extracts (Fig. 3B, lanes 1–3) were used as positive controls. The control protein-DNA complexes exhibited similar mobility on EMSAs as the respective complexes obtained with T3-1 nuclear extract proteins (Fig. 3, A and B). The SF-1 and Nur77 antisera did not cross-react with Nur77 and SF-1 control proteins, respectively (not shown). A pattern of complex formation similar to that seen with T3-1 nuclear extract proteins was observed when SF-1 and Nur77 control proteins were mixed and incubated with the –15/-7NRS probe (Fig. 3C), providing further evidence that complexes I and II formed with T3-1 nuclear extract proteins contain SF-1 and Nur77, respectively (a decrease in the intensity of the Nur77 complex caused by the presence of rabbit reticulocyte lysate proteins was consistently observed). A third complex, above complex II, was also sometimes observed in EMSAs with T3-1 nuclear extract proteins, but the intensity of this complex was not reproducibly affected by the presence of SF-1 or Nur77 antiserum (Fig. 3, A and B).
To serve as further confirmation that the protein in complex II was indeed Nur77, the binding of T3-1 nuclear extract proteins to an established Nur77-binding site was investigated. As with the –15/-7NRS probe, several DNA-protein complexes were detected (Fig. 3D) when T3-1 nuclear extracts were incubated with a radiolabeled double-stranded probe containing the Nur77-binding site from the human proopiomelanocortin promoter (NBRE; Fig. 3D, lane 4) (33). One of these complexes, designated complex III, had similar mobility to that of one of the complexes formed between the NBRE probe and CCRF-CEM nuclear extracts (Fig. 3D, lane 1). The intensities of both of these complexes decreased in the presence of anti-Nur77 antiserum (Fig. 3D, lanes 2 and 5). Although a decrease in intensity of the CCRF-CEM nuclear protein complex was consistently observed in the presence of preimmune serum, this decrease was less pronounced than that seen for anti-Nur77 antiserum. The intensity of complex III was not influenced by anti-SF-1 antiserum (Fig. 3D, lane 6). In fact, it appears that complex IV, with higher mobility than complex III, contains SF-1 protein, as its intensity was diminished by the addition of anti-SF-1 antiserum. These results strongly indicate that complex III contains Nur77 protein from T3-1 nuclear extracts. Furthermore, binding of the protein in complex III was competed for by the addition of a 25-fold molar excess of nonradiolabeled –15/-7NRS oligonucleotide (Fig. 3D, lane 8), providing additional evidence that Nur77 protein from T3-1 nuclear extracts can recognize the –15/-7NRS.
The presence of both SF-1 and Nur77 proteins in T3-1 nuclear extracts was confirmed by Western blotting (Fig. 3E). Once again, no cross-reactivity was observed for the SF-1 antiserum with the Nur77 control protein or the Nur77 antiserum with the SF-1 control protein (not shown).
SF-1 and Nur77 proteins in T3-1 nuclear extracts differ in their sequence requirements for binding to the –15/-7NRS and –244/-236NRS
The sequence requirements for binding of SF-1 (complex I) and Nur77 (complex II) from T3-1 nuclear extracts to the –15/-7NRS were tested in an EMSA with radiolabeled –15/-7NRS oligonucleotide as probe (Fig. 4A) in the absence and presence of varying concentrations of nonradiolabeled competitor DNA containing either the wild-type –15/-7NRS sequence or a mutated version thereof. As the two adjacent cytosines within a NRS had previously been shown to be essential for binding of SF-1 protein (34), the corresponding bases in the –15/-7NRS were changed to two thymidines to yield the oligonucleotide –15/-7NRSm. Although a 5-fold excess of nonradiolabeled –15/-7NRS oligonucleotide can effectively compete for SF-1 binding to radiolabeled –15/-7NRS (Fig. 4A, lanes 2 and 3), the –15/-7NRSm oligonucleotide was unable to compete for SF-1 binding even at a 125-fold molar excess (lanes 4–6). In contrast, the –15/-7NRSm oligonucleotide was able to compete for Nur77 protein binding. However, Nur77 protein seemed to have a lower affinity for –15/-7NRSm than for wild-type –15/-7NRS, because complete competition for Nur77 binding by –15/-7NRSm was observed at a 25-fold molar excess (Fig. 4A, lane 5) compared with a 5-fold molar excess observed for wild-type –15/-7NRS (Fig. 4A, lane 2).
|
show that a 25-fold molar excess of the –244/-236NRS oligonucleotide competed for binding of both SF-1 and Nur77 proteins to –15/-7NRS (Fig. 4B, lane 2). When the –244/-236NRS sequence was modified to yield a PstI restriction digestion site, the resulting oligonucleotide (-244/-236NRSm) failed to compete for SF-1 binding to the –15/-7NRS probe (Fig. 4B, lanes 4–6), even at a 125-fold molar excess (Fig. 4B, lane 6). Once again, the –244/-236NRSm oligonucleotide competed very effectively for Nur77 binding, with complete competition observed at a 5-fold molar excess. Therefore, these results show that SF-1 and Nur77 proteins bind to both the –15/-7NRS and the –244/-236NRS, and that the Nur77 protein recognizes mutated versions of these sites that are not recognized by SF-1.
These experiments were subsequently repeated using the –244/-236NRS oligonucleotide as radiolabeled probe (Fig. 4, C and D). Complexes I and II, as formed by the –15/-7NRS probe, were also observed for the –244/-236NRS probe, and the proteins in the complexes were confirmed by specific antibodies to be SF-1 and Nur77, respectively (data not shown). Once again, a 5-fold molar excess of both –15/-7NRS (Fig. 4C, lane 2) and –244/-236NRS (Fig. 4D, lane 2) was able to completely compete for binding of both SF-1 and Nur77. As described previously, a 125-fold molar excess of either mutated competitor was unable to compete for binding of SF-1 (Fig. 4C, lane 6, and Fig. 4D, lane 6), whereas a 5-fold molar excess was sufficient to compete for Nur77 binding (Fig. 4C, lane 4, and Fig. 4D, lane 4). Finally, the competition results for Nur77 binding were confirmed using the NBRE oligonucleotide as probe (Fig. 4E) in the absence and presence of each of the four competitor oligonucleotides. Both the wild-type and mutated oligonucleotides competed for Nur77 binding (Fig. 4E, lanes 2–5).
The –579/+1 GnRHR promoter is responsive to stimulation by forskolin, PACAP, and GnRH
To investigate the responses of the proximal mouse GnRHR promoter to activators of the PKA pathway, we prepared a promoter construct, designated pLG, by inserting 580 bp of the mouse GnRHR gene, from position –579 to +1 relative to the translation start site, into the promoterless pGL2-basic luciferase reporter vector. T3-1 cells were transfected with pLG, followed by incubation for varying times with 10 µM forskolin (Fig. 5A). No significant effect was observed on promoter activity until 16 h of forskolin stimulation, at which time an approximately 2-fold increase in activity was obtained. To determine whether PACAP has a stimulatory effect on the mouse GnRHR proximal promoter similar to that shown for the rat promoter (21), a similar experiment was performed with PACAP on the mouse pLG construct (Fig. 5B). A significant increase in promoter activity of pLG was observed after only 30 min of stimulation with PACAP, with maximum activity (5-fold) observed after 8 h (Fig. 5B). After 16 h, the activity was still significantly higher (2-fold) than that of uninduced pLG (Fig. 5B). To compare the responses of the mouse promoter to forskolin and PACAP with the response to GnRH, a similar experiment was performed on pLG with GnRH (Fig. 5C). Although the promoter responded to GnRH with a maximal increase of about 1.5-fold after 4 h, no significant effect was observed after 8 or 16 h.
|
. The mutations introduced were shown in Fig. 4 to result in loss of SF-1 binding in vitro. The constructs were successively transfected into T3-1 cells, followed by overnight incubation (16 h) with either 10 µM forskolin in 0.1% dimethylsulfoxide (DMSO; vol/vol) or 0.1% DMSO (vol/vol) alone (Fig. 6B). Mutating –15/-7NRS did not have any significant effect on the basal transcriptional activity of the –579/+1 promoter fragment. However, when –244/-236NRS was mutated, the –579/+1 promoter fragment did not retain any basal activity. This is consistent with results obtained by others (9, 10), demonstrating that –244/-236NRS is necessary for basal GnRHR promoter activity.
|
To confirm that the forskolin-induced increase in transcriptional activity of the –579/+1 promoter fragment is the result of activation of the PKA pathway by elevated cAMP levels, the same constructs described above were successively transfected into T3-1 cells together with a construct expressing the PKA catalytic subunit. To investigate the role of SF-1 protein in the response of the –579/+1 promoter fragment to PKA, a construct expressing SF-1 cDNA was also cotransfected in some experiments. The wild-type –579/+1 GnRHR promoter fragment (pLG) responded strongly to PKA (Fig. 7A). Although overexpression of SF-1 had no effect on basal –579/+1 promoter activity, it substantially (2-fold) enhanced the PKA response of pLG. As already shown in Fig. 6B, mutating –15/-7NRS such that SF-1 could no longer bind there (pLGM1) had no effect on basal levels of transcriptional activity of the –579/+1 promoter fragment. Furthermore, overexpression of SF-1 had no effect on basal pLGM1 activity. Similar to the result obtained for forskolin, the PKA response of pLGM1 was substantially increased (Fig. 7B) compared with that of pLG, i.e. about 32-fold compared with about 13-fold. This is consistent with the result presented in Fig. 6B, indicating that binding of SF-1 at –15/-7NRS negatively modulates the PKA response. Overexpression of SF-1 increased the PKA response of pLGM1 to about 100-fold above basal pLG activity or about 3-fold above that seen in the absence of overexpressed SF-1, as found for pLG.
|
, mutation of –244/-236NRS so as to prevent SF-1 protein from binding resulted in the loss of basal –579/+1 promoter activity (Fig. 7C). pLGM2 retained its ability to respond to PKA, as it did with stimulation by forskolin (Fig. 6B). However, unlike pLG and pLGM1, overexpression of SF-1 did not result in any increase in the response of pLGM2 to PKA. This would be consistent with a mechanism by which occupation of –244/-236NRS by SF-1 in the wild-type promoter results in enhancement of the PKA response. The results with pLGM1 indicate that concomitant binding of SF-1 at site 1 is not required for this effect. As found for forskolin, pGL2-basic was not responsive to overexpression of PKA or SF-1 (not shown).
The PKA response of the –579/+1 GnRHR promoter can be repressed by Nur77
To investigate the role of Nur77 protein in basal activity and in the response of the –579/+1 promoter fragment to PKA, experiments similar to those described above were performed with a construct expressing the Nur77 cDNA (Fig. 7). Overexpression of Nur77 had no effect on the basal activity of all three of the promoter constructs. Unlike SF-1, Nur77 had no influence on the PKA response of pLG. In experiments performed with pLGM2, in which responsiveness to SF-1 was lost, Nur77 overexpression also had no effect on the PKA response (Fig. 7C). However, an unexpected and very interesting result was obtained for the pLGM1 construct, where it was found that overexpression of Nur77 resulted in a substantial (3-fold) attenuation of the PKA response (Fig. 7B). This result indicates that in the absence of SF-1 binding at –15/-7NRS, Nur77 is able to repress the PKA response.
The effect of overexpression of Nur77 in the presence of both overexpressed PKA and SF-1 was also investigated for pLG and pLGM1 (Fig. 8). As overexpressed SF-1 could not enhance the PKA response of pLGM2, this construct was not investigated. As shown previously in Fig. 7, the PKA response of both pLG and pLGM1 was enhanced by overexpressing SF-1 (Fig. 8). Overexpressed Nur77 completely abrogated the enhancing effect of SF-1 on the PKA response of pLG. A small reduction in the SF-1-enhanced PKA response of pLGM1 was also consistently observed in the presence of overexpressed Nur77, although this effect was not statistically significant.
|
T3-1 cells. Total T3-1 RNA was isolated after overnight incubation in the absence or presence of increasing concentrations of forskolin and was probed for SF-1 mRNA by Northern blotting. The results presented in Fig. 9 clearly show that SF-1 mRNA (4 kb) levels increase in the presence of forskolin in a concentration-dependent manner (Fig. 9A), whereas ß-actin mRNA levels were unchanged (Fig. 9B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T3-1 cells and mouse pituitary tissue to be at –62 relative to the ATG start codon. This places the –15/-7NRS in both sources in the 5'-untranslated region and supports the use of the T3-1 cell line as a model cell line for investigating the role of –15/-7NRS in gonadotropes. The position of the major start site in T3-1 cells is consistent with the results obtained by others (7, 8), but is in contrast to the finding by Clay et al. (8) of a second start site at about –245.
Although it has been established that SF-1 protein binds to the mouse GnRHR –244/-236NRS in vitro (9), the binding of T3-1 nuclear extract proteins to the -15/-7NRS has not been previously investigated. We report here for the first time that both SF-1 and Nur77 bind specifically to the –15/-7NRS. We also show that Nur77 protein binds to the –244/-236NRS. We report that SF-1 and Nur77 proteins have different DNA binding specificities, with Nur77, unlike SF-1, still being able to recognize the mutated NRS. The detection of Nur77 protein and mRNA has not previously been reported in T3-1 cells, although it is well known that Nur77 protein is expressed in pituitary corticotropes (33, 37). However, the presence of Nur77 transcripts and their up-regulation by GnRH have been reported in LßT2 cells, another gonadotrope cell line (38). This together with our findings suggests that Nur77 plays a role in regulation of gene expression via NRS sites in pituitary gonadotropes.
We report for the first time that in T3-1 cells the endogenous GnRHR gene is up-regulated by forskolin and 8-Br-cAMP. The major mRNA transcript (4.5 kb) is up-regulated by about 2.5-fold by both compounds. The more marked up-regulation of a smaller transcript of approximately 1.8 kb by forskolin is intriguing and may be due to the use of an alternative transcription start site or alternative RNA processing regulated by the PKA pathway. The up-regulation observed for the endogenous 4.5-kb transcript is likely to be mediated predominantly via transcriptional regulation, as indicated by the similar 2- to 2.5-fold increase in the transcriptional activity of the wild-type -579/+1 promoter fragment when stimulated with forskolin in transient transfection assays. Our promoter assays clearly show that the –579/+1 mouse GnRHR promoter is up-regulated by PKA, supporting the argument that the forskolin-induced increase in mRNA levels and promoter activity was due to activation of the PKA pathway. Our promoter time-course studies with forskolin clearly show that this regulation is only observed after more than 8 h, with a significant increase at 16 h. We did not observe any effect with forskolin after 4 h, indicating that the PKA effects are most likely acting indirectly on the GnRHR promoter and require prior regulation of another gene. In support of our results, the rat (21, 35) and human (39) GnRHR promoters have also been shown by others to be responsive to activators of the PKA pathway in T3-1 cells. Interestingly, Cheng et al. (39) found a time course for forskolin induction of the human promoter similar to our result with the mouse promoter, with maximal induction occurring about 12 h or later. Our results are consistent with those of White et al. (11), who found that a similar construct of the mouse GnRHR promoter was unaffected after 4 h of forskolin treatment in T3-1 cells. These researchers did not, however, investigate longer times of incubation.
Our transfection results show that the –15/-7NRS is not required for basal or PKA-stimulated transcription of the mouse GnRHR gene. We have also found that this site is not involved in gonadotrope-specific expression (not shown), in agreement with previously published results (8). However, our results suggest that the levels of SF-1 protein and the binding of SF-1 at the proximal and distal NRS sites play a modulatory role in regulating the extent of the PKA stimulatory response. SF-1 protein, and intact SF-1-binding sites at the –15/-7NRS and –244/-236NRS do not appear to be required for the PKA response. Nevertheless, our combined data strongly support a role for SF-1 protein in positive modulation of the PKA response via the –244/-236NRS and negative modulation via the –15/-7NRS. It is difficult to envisage at this stage how the same protein could have both positive and negative effects on PKA stimulation. It is possible that promoter context and architecture determine the outcome of the response. For example, PKA-activated SF-1 binding at the –15/-7NRS may recruit a corepressor, whereas SF-1 binding at the distal NRS may recruit a coactivator due to interaction with different factors at the proximal and distal sites. SF-1 has been shown to interact with both transcriptional repressors, such as N-CoR (via DAX-1) (40), and activators, such as Sp1 (41) and SRC-1 (42). The positioning of the –15/-7NRS in the 5'-untranslated region may also play a role in its actions as a negative modulator of the PKA response by interfering with components of the basal transcriptional machinery. The fact that endogenous SF-1 mRNA is up-regulated by forskolin, but that overexpression of SF-1, in the absence of overexpressed PKA, has no effect on promoter activity suggests that the modulatory effects of SF-1 on the PKA response require the direct or indirect enzymatic modification of SF-1 by PKA.
We have shown that Nur77 protein, like SF-1, binds to both the proximal and distal NRS sites. Furthermore, Nur77 has a different DNA binding specificity to SF-1, as indicated by its binding to the mutated NRS sites. Transfection experiments with overexpressed proteins clearly show that Nur77 does not regulate basal promoter activity. However, our results with the pLGM1 mutant strongly suggest that endogenous Nur77 could act as a repressor of the PKA response via the proximal NRS under conditions where SF-1 is not bound to the proximal NRS. In this experiment, Nur77 would most likely be bound at the proximal NRS. The result for the pLGM2 mutant construct, showing that overexpression of Nur77 does not prevent the PKA response, would indicate that Nur77 binding at the distal NRS does not inhibit the PKA response. The finding that overexpression of Nur77 can completely blunt enhancement of the PKA/SF-1-induced reporter activity on pLG is consistent with a negative modulatory role for Nur77. A similar trend, albeit not significant, is seen with pLGM1. Our combined results suggest that modulation of the relative levels and activity of SF-1 and Nur77 could play an important role in the response of the GnRHR promoter to activation of the PKA pathway.
Nur77 has also been implicated in playing a role in the response to the PKA pathway for other genes. Nur77 expression in corticotropes is up-regulated by CRH during the response of the hypothalamic-pituitary-adrenal axis to chronic stress (33), and this up-regulation can be mimicked by forskolin in the AtT20 corticotrope cell line (43). Similarly, ACTH stimulates Nur77 expression in Y1 adrenocortical cells via increased cAMP levels (44). Further experiments are necessary to determine the effect of activation of the PKA pathway on Nur77 levels and activity in T3-1 cells. Our results showing that SF-1 modulates up-regulation of the mouse GnRHR gene by PKA activators are consistent with the role established for SF-1 for several other genes. In the case of certain adrenal steroidogenic enzymes, SF-1 sites appear to act together with CREs to mediate a positive response to cAMP (45, 46), whereas in other cases, SF-1 appears to be the only direct (47, 48) or indirect (49, 50) target for the PKA pathway. Our results for the mouse GnRHR, showing up-regulation of SF-1 and positive modulation by SF-1 via the distal NRS, suggest that similar mechanisms are involved compared with some of these other genes.
Our finding that the proximal and distal NRS sites are not necessary for the PKA response raises the question of which cis-elements and transcription factor(s) are essential for this response. Both the rat (21) and the human (39) GnRHR promoters have been shown to respond to PKA pathway activators via similar imperfect cAMP response elements in T3-1 cells. Both the –15/-7NRS and –244/-236NRS as well as the imperfect CRE are conserved in rat and mouse GnRHR promoters. It thus seems likely that the imperfect cAMP response element in the mouse GnRHR promoter is essential for the PKA response in T3-1 cells, and that a factor(s) binding there interacts directly or indirectly with SF-1 at the proximal and distal NRS sites. Further experiments are in progress to test these hypotheses.
In T3-1 cells, the PKA response of the mouse GnRHR promoter is likely to be involved in the response to PACAP, but not GnRH. We show that the mouse GnRHR promoter is up-regulated by PACAP, similar to the results obtained for the rat (21, 35) and human (39) promoters in T3-1 cells. We detected a significant increase in promoter activity in response to both forskolin and PACAP after 16 h, consistent with a role for PACAP in indirect up-regulation of the GnRHR promoter via the PKA pathway in these cells. However, up-regulation by PACAP was also observed after only 30 min, unlike the case with forskolin, indicating that additional pathways may be involved in PACAP directly regulating the GnRHR promoter (51). The approximately 2-fold induction after 16 h for both forskolin and PACAP, as opposed to the about 5-fold increase after 8 h with PACAP, would also be consistent with a role for PKA in the later, but not the earlier, response to PACAP. Our finding that GnRH up-regulates the GnRHR promoter after 4 h, but that this effect is not seen after 16 h, indicates that PKA is not involved in the GnRH response in T3-1 cells, because no effect is seen after 4 h in our promoter studies with forskolin. In addition, this result is consistent with the Northern blot results reported by Norwitz et al. (17), showing that GnRHR mRNA levels in T3-1 cells are maximally induced after 4 h with GnRH. The lack of activation of the PKA pathway by GnRH in T3-1 cells is most likely due to the fact that GnRHR is not coupled to Gs proteins in these cells (18). In summary, we have shown that the mouse GnRHR promoter responds to activators of the PKA pathway in T3-1 cells. Given that there is strong evidence that both GnRH (20) and PACAP (51) activate this pathway in primary pituitary gonadotropes, our results most likely reflect a physiologically relevant mechanism for the effects of both of these hormones in vivo. In addition, our results suggest that other signals that activate the PKA pathway in pituitary gonadotropes in vivo regulate the expression of the GnRHR gene.
|
Acknowledgments |
|---|
T3-1 cells. Dr. Stuart Sealfon kindly supplied clone 111. Ken Morohashi and Thomas Perlmann very kindly supplied us with antibodies. We thank Johann Riedemann, Emerentia Hutchinson, and Donita Africander for expert technical assistance. |
Footnotes |
|---|
Abbreviations: AP-1, Activator protein-1; 8-Br-cAMP, 8-bromo-cAMP; CRE, cAMP response element; DMSO, dimethylsulfoxide; DTT, dithiothreitol; EMSA, electrophoretic mobility shift assay; GnRHR, GnRH receptor; GSP2, gene-specific antisense primer; NRS, nuclear receptor half-site; PACAP, pituitary adenylate cyclase-activating polypeptide; pCMV, cytomegalovirus promoter; PKA, protein kinase A; PKC, protein kinase C; PMSF, phenylmethylsulfonylfluoride; 5'RACE, rapid amplification of 5' cDNA ends; SF-1, steroidogenic factor-1; SSC, standard saline citrate.
Received August 21, 2002.
Accepted for publication January 29, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T3-1 cells is associated with unchanged receptor messenger RNA (mRNA) levels and altered mRNA activity. Mol Endocrinol 7:1625–1633-, ß- and -actin mRNAs: skeletal but not cytoplasmic actins have an amino-terminal cysteine that is subsequently removed. Mol Cell Biol 3:787–795-subunit gene is regulated by a gonadotrope-specific protein. Mol Cell Biol 12:2143–2153-hydroxylase/c17–20 lyase). Mol Endocrinol 10:147–158This article has been cited by other articles:
 |
|
 |
|
  A. Kotitschke, H. Sadie-Van Gijsen, C. Avenant, S. Fernandes, and J. P. Hapgood Genomic and Nongenomic Cross Talk between the Gonadotropin-Releasing Hormone Receptor and Glucocorticoid Receptor Signaling Pathways Mol. Endocrinol., November 1, 2009; 23(11): 1726 - 1745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Vaudry, A. Falluel-Morel, S. Bourgault, M. Basille, D. Burel, O. Wurtz, A. Fournier, B. K. C. Chow, H. Hashimoto, L. Galas, et al. Pituitary Adenylate Cyclase-Activating Polypeptide and Its Receptors: 20 Years after the Discovery Pharmacol. Rev., September 1, 2009; 61(3): 283 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kanasaki, S. Mutiara, A. Oride, I. N. Purwana, and K. Miyazaki Pulse Frequency-Dependent Gonadotropin Gene Expression by Adenylate Cyclase-Activating Polypeptide 1 in Perifused Mouse Pituitary Gonadotroph LbetaT2 Cells Biol Reprod, September 1, 2009; 81(3): 465 - 472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Grafer, R. Thomas, L. Lambrakos, I. Montoya, S. White, and L. M. Halvorson GnRH Stimulates Expression of PACAP in the Pituitary Gonadotropes via Both the PKA and PKC Signaling Systems Mol. Endocrinol., July 1, 2009; 23(7): 1022 - 1032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Janssens, M. Boussemaere, S. Wagner, K. Kopka, and C. Denef {beta}1-Adrenoceptors in Rat Anterior Pituitary May Be Constitutively Active. Inverse Agonism of CGP 20712A on Basal 3',5'-Cyclic Adenosine 5'-Monophosphate Levels Endocrinology, May 1, 2008; 149(5): 2391 - 2402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A Ferris, H. E Walsh, J. Stevens, P. C Fallest, and M. A Shupnik Luteinizing Hormone Beta Promoter Stimulation by Adenylyl Cyclase and Cooperation with Gonadotropin-Releasing Hormone 1 in Transgenic Mice and LBetaT2 Cells Biol Reprod, December 1, 2007; 77(6): 1073 - 1080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Harada, H. Kanasaki, S. Mutiara, A. Oride, and K. Miyazaki Cyclic Adenosine 3',5'Monophosphate/Protein Kinase A and Mitogen-Activated Protein Kinase 3/1 Pathways Are Involved in Adenylate Cyclase-Activating Polypeptide 1-Induced Common Alpha-Glycoprotein Subunit Gene (Cga) Expression in Mouse Pituitary Gonadotroph LbetaT2 Cells Biol Reprod, October 1, 2007; 77(4): 707 - 716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. Moore Jr, L. L. Burger, A. C. Dalkin, and S. J. Winters Pituitary Adenylate Cyclase Activating Polypeptide Messenger RNA in the Paraventricular Nucleus and Anterior Pituitary During the Rat Estrous Cycle Biol Reprod, September 1, 2005; 73(3): 491 - 499. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Luo, L. Ding, J. Xu, R. S. Williams, and N. Chegini Leiomyoma and Myometrial Gene Expression Profiles and Their Responses to Gonadotropin-Releasing Hormone Analog Therapy Endocrinology, March 1, 2005; 146(3): 1074 - 1096. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G S Harrison, M E Wierman, T M Nett, and L M Glode Gonadotropin-releasing hormone and its receptor in normal and malignant cells Endocr. Relat. Cancer, December 1, 2004; 11(4): 725 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Y. Hong, J. H. Park, R. S. Ahn, S. Y. Im, H.-S. Choi, J. Soh, S. H. Mellon, and K. Lee Molecular Mechanism of Suppression of Testicular Steroidogenesis by Proinflammatory Cytokine Tumor Necrosis Factor Alpha Mol. Cell. Biol., April 1, 2004; 24(7): 2593 - 2604. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
