| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Subunit Promoter by Gonadotropin-Releasing Hormone
Department of Biomedical Sciences (J.X., S.P.B., M.S.R.), Cornell University, Ithaca, New York 14853; Department of Biomedical Sciences (T.M.N.), Colorado State University, Fort Collins, Colorado 80523; and Department of Neurology (B.J.E., S.C.S.), Mount Sinai School of Medicine, New York, New York 10029
Address all correspondence and requests for reprints to: Mark S. Roberson Ph.D., Associate Professor of Physiology, T3-004d Veterinary Research Tower, Department of Biomedical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853. E-mail: msr14{at}cornell.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
T3-1 gonadotrope cells, ATF3 mRNA and protein were induced by GnRH in a manner consistent with in vivo observations. Pharmacological studies implicated a combined role for the activities of protein kinase C isozymes, ERK and c-Jun N-terminal kinase, in modulating ATF3 expression. The role of ATF3 was further investigated in the activation of the human glycoprotein hormone -subunit gene promoter. GnRH induced the -subunit promoter-luciferase reporter approximately 16-fold, and this induction was completely abolished with mutations in the dual cAMP response elements (CREs) or the combined inhibition of GnRH-induced ERK and c-Jun N-terminal kinase. GnRH induced recruitment of ATF3, c-Jun, and c-Fos to the dual CREs. Overexpression and specific knockdown of ATF3 by small inhibitory RNA implicate a functional role for ATF3 in mediating activation of the -subunit gene promoter. These studies provide clear evidence that ATF3 is a key immediate early gene induced by GnRH administration in vivo and in the T3-1 gonadotrope cell model. These studies support the conclusion that the dual CREs of the human -subunit promoter are the target of GnRH-induced MAPK regulation through ATF3. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit and the two separate ß-subunits of LH and FSH (recently reviewed in Refs.2 and 3). Gonadotropic stimulation of gonadal steroid and peptide hormones modulates negative and positive feedback mechanisms that serve to control follicular development, maturation, and ovulation within reproductive cycles (4). Disruption of the hypothalamic-pituitary axis either surgically or by genetic mutation results in reduced gonadotropin subunit gene expression and infertility due to hypogonadotropism/hypogonadism (5, 6, 7, 8, 9). Due to their complexity, the gonadotropin subunit genes have served as important models for understanding the integration of cell signaling via multiple intracellular pathways induced by GnRH and cell-specific transcriptional regulation based upon their restricted pattern of expression.
Regulation of gonadotrope function by GnRH requires modulation of multiple signaling pathways that effectively serve to integrate a gene regulatory network comprising early, intermediate, and late response genes. The signaling pathways induced by GnRH are initially regulated by engagement of the heterotrimeric G protein Gq, activation of phospholipase Cß, and the elaboration of the second messengers, inositol 1,4,5-trisphosphate and diacylglycerol (DAG) (4, 10, 11). Inositol 1,4,5-trisphosphate mediates release of calcium from internal stores necessary for gonadotropin secretion whereas DAG, in turn, mediates activation of protein kinase C (PKC) isozymes. GnRHR activation leads to stimulation of several MAPK cascades, including the ERK, p38 MAPK, and c-Jun N-terminal kinase (JNK) pathways, which mediate different aspects of gene induction ranging from activation of classical immediate early genes such as c-Fos and c-Jun, intermediate genes such as the dual specificity MAPK phosphatase MKP-2, and late genes such as the gonadotropin subunit and GnRHR genes (12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28). This interconnected signaling system provides an important link for the transmission of signals from the cell surface to the nucleus and plays a role in the integration of a gene network induced by GnRH.
Analysis of the gene network induced by GnRH has been recently investigated in the gonadotrope cell line, LßT2 (29, 30). Several gonadotrope-like cell lines, including LßT2 and T3-1 cells, were derived from targeted oncogenesis in transgenic mice (31, 32, 33). The importance of these cell lines comes from the observation that they express many of the differentiated cell markers, including the glycoprotein hormone common -subunit (LßT3 and T3-1), the ß-subunits to LH and FSH (LßT2), and the GnRHR (LßT2 and T3-1) that define the gonadotrope cell lineage. The initial studies using transcript profiling by gene arrays and high-throughput quantitative real-time PCR (qPCR) using mRNA from LßT2 cells treated with GnRH revealed marked changes in the expression of multiple genes subdivided into important classes such as cytoskeletal elements, transcription factors, signaling molecules, and regulators of ion channel activity (29). These studies focused on analysis of gene products regulated in the initial 1–6 h after administration of GnRH. Early growth response (Egr) protein-1 and LRG21/activating transcription factor (ATF)3 were the two most robustly up-regulated genes after 1 h of GnRH treatment as measured by gene array and by qPCR (29). Egr-1 appears to integrate multiple intermediate and late genes including the dual specificity MAPK phosphatase, MKP-2 (20), and the ß-subunit to LH (34, 35, 36, 37, 38). The Egr-1 knockout mouse was found to be infertile, likely due to diminished expression of LHß, which confirms the overall importance of Egr-1 in the hypothalamic-pituitary-gonadal axis (35). Little is known regarding a potential role for ATF3 within the GnRH pathway. The present studies demonstrate that ATF3 was markedly up-regulated in vivo and in the T3-1 gonadotrope cell model. Using GnRH-treated T3-1 cells, our studies define the signaling pathways necessary for GnRH-induced ATF3 expression and implicate ATF3 as an important transcriptional regulator of the human glycoprotein hormone -subunit gene promoter in response to GnRH administration via dual cAMP response elements (CREs).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
T3-1 Cells
12 fold), and was diminished by 120 min after dAla6-GnRH administration (Fig. 1
).
|
T3-1 cells were nearly identical to in vivo responses peaking at 60 min and declining thereafter (Fig. 2A). ATF3 protein levels were easily detectable by 60 min after GnRHa administration and remained elevated for several hours (Fig. 2B). GnRHa-induced ATF3 expression was completely abolished when the specific GnRH receptor antagonist, Antide, was administered concurrently with GnRHa, providing evidence for the specificity of this response (Fig. 2C).
|
T3-1 cell model, we examined the signaling intermediates involved in the expression of ATF3 by GnRHa. Acute administration of phorbol ester was sufficient to mimic the effects of GnRHa on ATF3 in a similar time course, suggesting the possibility of PKC isozyme involvement in this mechanism (Fig. 3A). However, inhibition of PKC isozymes by chronic (18 h) exposure to phorbol ester (Fig. 3B) or pretreatment with the specific PKC inhibitor bisindolylmaleimide (GFX, Fig. 3C) only partially inhibited GnRH-induced expression of ATF3. We have previously demonstrated that the conditions and doses of PKC inhibitors used in the current studies were sufficient to block GnRH-induced ERK activation and increased expression of MKP-2 in T3-1 cells (15, 16, 21).
|
T3-1 cells with the specific MAPK kinase 1 inhibitor PD98059 (Fig. 4A). In contrast to the regulation of the ERK pathway, studies from our laboratory suggest that activation of the JNK cascade by GnRHa occurs via Cdc42 and is independent of DAG-dependent PKC activity (16). Interestingly, inhibition of JNK activity using SP600125 resulted in a dose-dependent decline in GnRHa-induced ATF3 expression as well (Fig. 4B). The effects of the SP600125 compound appeared to be specific as GnRH-induced ERK phosphorylation was unaffected or modestly enhanced by this drug (Fig. 4B). When these two drugs were administered in combination, GnRHa-induced ATF3 levels were inhibited to a greater degree than by either drug alone (Fig. 4C). These studies support the conclusion that both ERK and JNK activities are required for GnRHa-induced ATF3 expression.
|
-Subunit Promoter Requires Multiple DNA Elements Including the Dual cAMP Response Elements-subunit promoter required for activation by GnRH. In those studies, deletion of the –346 to –244 region of this promoter appeared to reduce responsiveness to GnRH; despite this reduction, significant GnRH-induced transcriptional activity remained even in the presence of the deletion, suggesting additional elements downstream of –244 within the -subunit promoter were involved in GnRH regulation. We examined this possibility in the context of the –846 and –204 deletion mutants of the human -subunit reporter gene (Fig. 5). The wild-type –846 reporter construct was induced 9-fold by GnRHa in T3-1 cells, and deletion of the dual CREs (–147 to –116) in the context of the –846 promoter reduced this response (Fig. 5A). Deletion of the reporter from –846 to –204 resulted in lower basal activity; however, response to GnRHa remained robust (Fig. 5B). Thus, loss of the –346 to –244 element did not completely abolish responsiveness to GnRHa consistent with previous reports (40). In the context of the –204 -subunit promoter, partial loss of GnRH responsiveness was evident with individual mutations in either the up- or downstream CREs (M1 and M2, respectively). In addition, both of these CRE mutants had a marked effect on basal expression of the reporter gene. When both CREs were deleted (
CRE), response to GnRHa was completely abolished, providing convincing evidence for the requirement of the dual CREs in mediating responsiveness to GnRHa within the –204 -subunit reporter gene.
|
-subunit reporter also required the combined modulation of both the ERK and JNK cascades for full activation by GnRHa (Fig. 6). Pretreatment of transfected T3-1 cells with PD98059 or SP600125 at doses known to suppress ATF3 expression resulted in incremental inhibition of GnRHa-induced -subunit reporter gene activation. When these two drugs were administered in combination, GnRHa-induced -subunit promoter activity was reduced to near baseline. The involvement of these signaling pathways in the regulation of -subunit promoter activity via the CREs was consistent with a role for ATF3 in transcriptional regulation in this system.
|
-subunit promoter. Using nuclear extracts from unstimulated T3-1 cells, Heckert and Nilson (42) previously reported that CREB, CRE modulator, c-Jun, ATF1, and ATF2 were putatively present within the CRE binding complex. Interestingly, c-Jun was identified as a target immediate early gene in the transcript profiling studies conducted in LßT2 cells; however, neither CREB, CRE modulator, ATF1, nor ATF2 were identified in this screen (29). Our laboratory has also identified c-Fos and c-Jun (AP-1) as possible regulators of CRE-dependent transcription of the -subunit promoter in a human choriocarcinoma cell model using epidermal growth factor (43). To investigate CRE binding activities in T3-1 cells, we used a pull-down approach with biotinylated CREs coupled to streptavidin (SA) agarose beads to examine the components of the CRE binding complex after GnRHa administration. In nuclear extracts from control cells, ATF3, c-Jun, and c-Fos were not present in binding complexes, consistent with the notion that these transcription factors were found to be GnRH-responsive immediate early genes in LßT2 cells (29) and in vivo (Fig. 1
). In contrast, CREB appeared to bind the dual CREs regardless of cell stimulation. In nuclear extracts obtained from T3-1 cells after a 2-h exposure to GnRHa, ATF3, c-Jun, c-Fos, and CREB were present within the CRE binding complex (Fig. 7A). The specificity of binding of these basic leucine zipper (bZip) proteins was supported by competition studies using homologous and heterologous competitors (Fig. 7A). Using immunoprecipitation (IP; Fig. 7B), ATF3-c-Jun and c-Fos-c-Jun (data not shown) formed heterodimers after GnRHa treatment in T3-1 cells, suggesting that these bZip proteins may participate within the -subunit dual CREs as heterodimeric complexes. Technical difficulties with the ATF3 antibody precluded immunoprecipitation with this antibody; thus we cannot rule out an additional possibility of an ATF3 homodimer binding the dual CREs as well. The question that remained was the functional significance of any of these potential bZip hetero- or homodimers within the CRE complex in the regulation of the human -subunit promoter.
|
-subunit reporter. Expression of c-Jun protein was partially reduced with pretreatment with PD98059 or SP600125, and the combination of drugs reduced c-Jun levels to near baseline (Fig. 7C). Phosphorylation of c-Jun was entirely dependent upon GnRHa-induced JNK activity. In contrast, c-Fos expression and CREB phosphorylation did not correlate with ATF3/c-Jun expression or regulation of the -subunit promoter. GnRHa-induced c-Fos required ERK activity but was not remarkably affected by JNK inhibition. GnRHa induced a marked increase in CREB phosphorylation; however, inhibition of neither ERK nor JNK has an appreciable impact on CREB phosphorylation. These pharmacological studies provide evidence to rule out a role for CREB and reduce the likelihood for a role for c-Fos in this transcriptional model. Importantly, these studies implicate ATF3 and c-Jun as possible targets of GnRH action, leading to the regulation of the -subunit promoter based upon their shared signaling requirements and their ability to participate in a CRE binding complex.
ATF3 Was Sufficient and Required for Full CRE-Dependent Transcriptional Regulation of the -Subunit Promoter by GnRHa
To determine the functional significance of ATF3 expression within the GnRH pathway, we sought to determine the impact of ATF3 overexpression on -subunit reporter gene activity. Overexpressed ATF3 was detected by the ATF3 antibody as a higher molecular weight form due to the addition of the epitope tag at the amino termini of the overexpressed protein (Fig. 8A). Thus, a direct comparison could be made between endogenous ATF3 expression induced by GnRHa and ATF3 derived from overexpression. Overexpression of HA-ATF3 resulted in increased ATF3 protein levels comparable to ATF3 levels after GnRHa administration (Fig. 8A), suggesting these overexpression studies produced responses within a reasonable physiological range. Interestingly in the absence of GnRHa stimulation, HA-ATF3 overexpression induced -subunit reporter gene activity in this system (Fig. 8B). Further, both the effects of GnRHa and HA-ATF3 were completely abrogated with the deletion of the dual CREs, supporting the conclusion that ATF3 was functioning through this element. These studies supported the prediction that the combined action of ATF3 overexpression and addition of GnRHa in transfected cells may not alter the response to GnRHa alone in this system. Data in Fig. 8C supported this prediction. ATF3 overexpression again increased expression of the -subunit reporter as seen previously; however, the combined actions of ATF3 overexpression and GnRHa did not alter fold induction over GnRHa alone. Recombinant ATF3 alone or in combination with c-Jun was sufficient to bind to the dual CREs of the -subunit promoter in DNA binding studies (Fig. 9), providing an explanation for the ability of ATF3 overexpression (in the absence of GnRHa stimulation) to induce expression of the -subunit reporter.
|
|
-subunit gene regulation, we used two methodologies. First, a dominant negative form of c-Fos (acidic- or A-Fos) have been used to successfully interfere with c-Fos-related heterodimers such as AP-1 (43, 44). Within the A-Fos mutant, acidic amino acids were substituted for the basic residues within the DNA binding domain of c-Fos. This mutation provided a mutant form of c-Fos that could interact with c-Jun but effectively block DNA binding, thus interfering with AP-1-mediated transcription or potentially serving to sequester c-Jun away from other interacting partners such as ATF3. In T3-1 cells, A-Fos overexpression inhibited basal expression of the -subunit reporter, providing evidence for the efficacy of A-Fos activity; however, A-Fos overexpression did not affect GnRH-mediated induction of the -subunit promoter (Fig. 10A). One interpretation of these data is that c-Fos interacting partners such as c-Jun may not play an important role in GnRHa-induced -subunit promoter regulation.
|
). Inclusion of the control siRNA in these studies demonstrated specificity of the ATF3 siRNA. Moreover, neither siRNA preparation affected ß-actin expression nor did they alter GnRHa-induced c-Jun protein levels (data not shown). Under identical conditions, ATF3 knockdown using siRNA decreased GnRHa-induced -subunit reporter gene activity from 14.6- to 8.9-fold compared with control levels (Fig. 10C), a level consistent with knockdown of ATF3 protein (50%). Basal activity of the reporter gene was unaffected by the siRNA treatment. These knockdown studies support the conclusion that ATF3 was required for full inducibility of the -subunit reporter by GnRHa. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
T3-1 gonadotrope cell model, we demonstrated the combined requirements for the ERK and JNK cascades for ATF3 expression using pharmacological approaches. GnRHa-induced ERK and JNK activities were also required for CRE-dependent -subunit reporter gene activity. In these studies, we investigated the regulation of the human -subunit because the promoter region of the primate -subunit gene contains dual consensus CREs; the presence of the dual CREs in the human gene are in contrast to other mammalian promoters (including mouse) for the -subunit that contain only a single (likely nonfunctional) CRE-like sequence (47). The dual CREs within the human -subunit promoter made this a likely target gene for ATF3 action. Consistent with this hypothesis, ATF3 was shown to bind to and transactivate the -subunit reporter dependent upon the dual CREs. Importantly, specific knockdown of GnRHa-induced ATF3 expression using an siRNA approach resulted in reduction in GnRHa-induced -subunit transcriptional activation supporting the conclusion that GnRHa regulation of ATF3 was instrumental in the regulation of late response genes such as the human glycoprotein hormone -subunit. Up-regulation of ATF3 has been described in other systems using gene array or profiling of inducible transcriptional regulators, primarily associated with cell stress responses (recently reviewed in Refs.45 and 46)). For example, recent array studies demonstrate response(s) to low-dose UV-A light results in dramatically increased expression of ATF3 associated with activation of the JNK cascade in human fibroblasts (48). During stress-induced neuronal cell death, ATF3 mRNA expression increased and was thought to play a survival role, mediating the expression of heat shock protein 27, effectively rescuing PC12 cells from apoptosis (49). ATF3 expression was also greatly enhanced, associated with the endoplasmic reticulum stress and amino acid deprivation in mouse embryonic fibroblasts (50). In those studies, expression of ATF3 was preceded by up-regulation of ATF4, which appeared to be required for ATF3 expression, mediating the cascade associated with the stress response. The mechanism(s) for regulation of ATF3 in the context of GnRH action does not appear to include regulation by ATF4 because ATF4 was not found to be regulated by GnRH within the immediate early gene network defined in the LßT2 cell model (29).
Interesting overlap exists in the immediate early gene responses to stress when compared with responses induced by GnRHa. Most notably during the initiation of acute pancreatitis, ATF3 mRNA was strongly induced along with MKP-1 and Egr-1, all markers potentially useful in the prediction of severity of inflammatory disease (51). All three of these genes are also up-regulated by GnRH (29), albeit not associated with inflammation but with differentiated cell function. In the pancreatitis model, the role of ATF3 is unknown; however, Egr-1 appears to be a principal response gene linked to increased expression of proinflammatory cytokines. Pancreatitis severity was reduced when similar studies were conducted in Egr-1 null mice, suggesting an important role for Egr-1 during pancreatic inflammatory disease. In contrast, Egr-1 in pituitary gonadotropes was clearly linked to the expression of the LH ß-subunit (35) and MKP-2 promoters (20), providing for a very different outcome when compared with the role of Egr-1 within the stress response pathway. These studies underscore the importance of understanding the cadre of immediate early response genes and activated signaling pathways associated with normal physiological and pathophysiological stimuli.
The cell signaling associated with GnRH-induced ATF3 gene expression suggests that both the ERK and JNK pathways contribute in a coordinated manner. In the present studies, inhibition of the ERK cascade by PD98059 reduced, but did not completely abolish, ATF3 protein accumulation. A similar result was observed with JNK inhibition by SP600125. The combined action of both drugs was additive. These results supported speculation that both MAPK pathways were independently involved in ATF3 expression in T3-1 cells. These studies were further supported by the observation that combined inhibition of GnRH-induced ERK and JNK was necessary for full blockade of CRE-dependent -subunit gene activation. Other studies have shown dual regulation of ATF3 by the ERK and JNK cascades. TNF strongly induced ATF3 mRNA and protein in vascular endothelial cells (52). JNK activity was strongly induced by TNF, which was linked to increased expression of ATF3. Concurrent activation of the ERK pathway, however, was sufficient to inhibit TNF/JNK-mediated ATF3 expression. Thus, unlike the GnRH pathway, ERK and JNK play competing roles in TNF regulation of ATF3.
In addition to the role of MAPKs on ATF3 gene regulation, both of these pathways could potentially play a role in the regulation of ATF3 activity as a transcriptional regulator. ATF3 has several consensus proline-directed phosphorylation sites (PxxS/TP or TP; x = any amino acid); however, the kinase/substrate specificity of phosphorylation at these sites remains to be elucidated (53). Regardless, phosphorylation at this site may alter DNA binding, dimerization, transcriptional activation of ATF3, and/or recruitment of coactivators or corepressors, depending on the cellular context. ATF3 has been shown to be a transcriptional repressor when present as a homodimer whereas transcriptional stimulating activity was characteristic of ATF3-c-Jun heterodimers (45, 46). In the context of GnRH signaling, ATF3 was shown to be capable of heterodimer formation with c-Jun; however, overexpression of ATF3 alone (absent GnRHa stimulation) stimulated transcriptional activation of the -subunit promoter via the dual CREs. Although speculative, the mechanism for how overexpression of ATF3 alone resulted in transcriptional activation of the -subunit reporter gene could be linked to a novel role for ATF3 homodimers as transactivators. This possibility is supported by the observation that recombinant ATF3 alone could bind to the dual CREs of the -subunit promoter. The likelihood that ATF3-c-Jun heterodimers were critical in this mechanism was lessened based upon overexpression studies using the A-Fos dominant negative mutant. A-Fos could putatively titrate c-Jun away from participation in a functional ATF3-c-Jun heterodimer, again leaving open the possible role for the ATF3 homodimer. Additional studies will be necessary to elucidate the absolute role and mechanism(s) for ATF3 homodimers serving as transcriptional activators in this system.
Basal and agonist-induced regulation of the human -subunit promoter has long been recognized as complex, involving multiple transcriptional activators functioning in a combinatorial manner. Reports by several groups, including our own, have clarified elements within this combinatorial code regarding cell-specific and basal regulation of this gene in both pituitary gonadotropes and in human placental trophoblasts (42, 43, 54, 55, 56, 57, 58, 59). In placental trophoblasts, the glycoprotein hormone -subunit is a component of human chorionic gonadotropin, an important luteotropin in early pregnancy. Expression of the -subunit in this cell type requires an apparent placental-specific combinatorial code comprised of cis elements binding AP-2, Dlx3, and probably CREB/AP-1 (56, 59, 60). In the pituitary gonadotrope, upstream elements such as the pituitary glycoprotein hormone basal element, basal element, and the gonadotrope-specific element all appear to be important in the context of a functional dual CRE for pituitary specific regulation (61). Few studies have investigated this promoter in the context of GnRH action. As mentioned previously, Kay and Jameson (40) identified a GnRH-responsive region of the -subunit promoter upstream of the dual CREs. These data were consistent with other later reports (62, 63), and these authors concluded that the dual CREs were less important when compared with more distal regions of the human -subunit promoter regarding responsiveness to GnRH. Interestingly, deletion mutants lacking the GnRH-responsive region, but containing the dual CREs, still displayed appreciable GnRH responsiveness (40, 62). In contrast, the present studies support the role of the dual CREs in mediating GnRH responsiveness of the -subunit when considering this more proximal promoter region. An important difference between the present studies and those of others (40, 62) is the duration of GnRH treatment before analysis of luciferase activity. The present studies examined luciferase activity 6 h after GnRHa administration because the ATF3 expression profile was maximal during this timeframe. Because previous studies examined responses to GnRH at 20–24 h (40, 62), it is reasonable that early response mechanisms involving the dual CREs and ATF3 may have been overlooked. The present studies also implicate a role for both the ERK and JNK cascades in the regulation of the -subunit promoter by GnRH. Others have suggested a role for the ERK pathway using this promoter but ruled out a role for the JNK cascade (63). In these studies, JNK cascade involvement was evaluated using putative dominant interfering expression vectors that may have been less effective when compared with the use of SP600125. The present studies validate the utility and specificity of SP600125 and provide support for the role for JNK activation in these studies. The requirement of both the ERK and JNK cascades for ATF3 expression was highly correlated with regulation of the human -subunit reporter by GnRHa. Moreover, the use of ATF3-specific siRNA for knockdown was instrumental in supporting the conclusion that ATF3 was involved in GnRH-induced human -subunit reporter gene activity. Worth noting was the utility of the siRNA approach for knockdown of an immediate early gene. In unstimulated T3-1 cells, ATF3 mRNA levels were very low. Upon GnRHa administration, ATF3 mRNA abundance increased to high levels in a very rapid manner. Successful knockdown with siRNA in the face of a rapidly expanding mRNA population is probably not trivial. The ATF3 siRNA reduced ATF3 protein expression, and this was correlated with a reduction in GnRHa responsiveness, supporting the conclusion that ATF3 was required for full response to GnRHa in this system. The possibility certainly exists that ERK and JNK activities may also coordinate additional transcriptional regulation not accounted for by inhibition of ATF3 alone. In the present studies, the dual CREs function as the target for integrated cell signaling via multiple inputs including, but not restricted to, MAPK activities and ATF3.
In summary, our studies provide a functional link between regulation of immediate early gene expression and a molecular mechanism governing the activation of a key gene leading to the production of a gonadotropic hormone subunit. ATF3 is an immediate early response gene most closely linked to stress responses in a number of experimental paradigms. However, in the pituitary gonadotrope, the role of ATF3 is clearly different, serving to support differentiated cell function. Although speculative, increased expression of ATF3 in the appropriate context may be instrumental in effectively coordinating the regulation of a number of CRE-dependent genes by GnRH. This is reminiscent of the integrated role of Egr-1 in coordinating the regulation of LHß and the MKP-2 gene promoters within the GnRH-induced gene network.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
SYBR green-based real-time PCR (qPCR) was performed as previously reported (30, 65). In brief, total RNA was converted into cDNA and approximately 250 pg was used for 40-cycle three-step PCR in an ABI Prism 7900HT (PE Applied Biosystems, Foster City, CA) in 20 mM Tris, pH 8.4; 50 mM KCl; 3 mM MgCl2; 200 M deoxynucleoside triphosphates; 0.5 SYBR Green I (Molecular Probes, Inc., Eugene, OR); 200 nM each primer; and 0.5 U Platinum Taq (Invitrogen, Carlsbad, CA). Amplicon size and reaction specificity were confirmed by agarose gel electrophoresis. The relative levels of expression were interpolated from detection threshold (CT) values normalized relative to the CT of RPS11, an unregulated housekeeping gene. Each transcript in each sample was assayed twice, and the average CT values were used to calculate the fold-change ratios between experimental and control samples for each gene used in the analysis.
Cell Culture and Plasmids
T3-1 cells, an immortalized mouse pituitary cell line of the gonadotrope lineage (generously provided by Pamela Mellon, University of California, San Diego), were cultured as described previously (43, 47). The human glycoprotein hormone -subunit luciferase reporters ( –846- and –204 luciferase, including the M1, M2, and CRE mutations) have been described previously (43, 47). ATF3 cDNA (GenBank accession no. NM_007498) was generated by PCR using the 5'-primer (5'-ACCGTGGATCCATGATGCTT CAACATCCAGGCCAG) and 3'-primer (5'-CAGTCGAATTCTTAGCTCTGCAATGTTCCTTC) with BamHI and EcoRI restriction sites for inserting into the mammalian expression vector pKH3. The resulting cDNA was confirmed by nucleotide sequence analysis. The pKH3 vector contains three copies of the HA epitope in frame with ATF3 cDNA. This cloning step generated the expression construct referred to as HA-ATF3.
Northern and Western Blot Analyses
For Northern blot analysis, cells were cultured in 100-mm plates and grown to approximately 70% confluence before use. Cells were serum starved in DMEM for 2 h before receiving hormone. The GnRHa buserelin (des-GLY10 [D-Ser(t-But)6 ]-LH-RH Ethylamide; referred to as GnRHa; Sigma Chemical Co., St. Louis, MO) was applied to the cells at 10 nM for various lengths of time before extraction of RNA. At the end of the incubation period, total cellular RNA was extracted using Triazol reagent (Invitrogen) according to the manufacturer’s instructions. Total RNA (20 µg) was resolved in denaturing formaldehyde agarose gels. Ethidium bromide staining of the 28s and 18s rRNAs demonstrated equal sample loading in these gels. The RNA was transferred to nitrocellulose filters after which the filters were baked at 80 C in a vacuum for 2 h. Hybridization (using an ATF3 cDNA probe) and posthybridization washes were carried out as described previously (66). Northern blots were completed twice using independently prepared RNA samples; each replicate had identical results.
For Western blot studies, cells were cultured in 60-mm plates and grown to approximately 70% confluence before use. Cells were serum starved in DMEM for 2 h before receiving agonist or antagonist treatments. GnRHa was applied to the cells at 10 nM for various lengths of time before being collected. Antide (used at 10 µM) and phorbol ester [phorbol 12-myristate 13-acetate (PMA); used at 1 µM]) were purchased from Sigma. For acute PMA studies, PMA was administered in a manner identical to GnRHa. For chronic PMA studies, cells were treated with PMA (1 µM) in serum-free DMEM for 18 h before GnRHa was applied for the designated time course. Control-treated cells in these studies were also serum starved for 18 h before GnRHa treatment. Bisindolylmaleimide (used at 2 µM), PD98059 (used at 50 µM), and SP600125 (used at 10 and 25 µM) were purchased from Calbiochem (San Diego, CA). Inhibitors were prepared as stock solutions in dimethylsulfoxide and applied to the cells in DMEM 30 min before the treatment of GnRHa. Cells were never exposed to greater than 0.1% dimethylsulfoxide, and the concentration of vehicle had no effect on responses of T3-1 cells. At the end of each experiment, cells in culture dishes were placed on ice and washed once with ice-cold Dulbecco’s PBS (Invitrogen). Cell lysates were prepared using a radioimmunoprecipitation assay (RIPA) buffer containing 20 mM Tris-HCl (pH 8.0), 137 mM NaCl, 10% glycerol, 1% Nonidet P-40, 0.1% sodium dodecyl sulfate (SDS), 0.5% deoxycholate, 2 mM EDTA, 5 mM sodium vanadate, 0.2 mM phenylmethylsulfonyl fluoride, 5 mM benzamidine. After centrifugation to remove cell debris, the protein amounts in the whole-cell lysates were analyzed by Bradford’s assay. Protein (30 µg) from individual samples were resolved by SDS-PAGE and transferred to polyvinylidene difluoride membranes by electroblotting. The blots were blocked with 5% nonfat dry milk (NFDM) in TBST (10 mM Tris-HCl, pH 7.5; 150 mM NaCl; 0.05% Tween 20) and then probed with primary and secondary antibodies. The antibodies for ATF3, ß-actin, phospho-c-Jun, and ERK (all obtained from Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were diluted 1:500 in 5% NFDM/TBST. The phospho-specific ERK antibody (Cell Signaling, Beverly MA) was diluted 1:1000 in 5% NFDM/TBST. The secondary antibodies were species specific and coupled to horseradish peroxidase (1:5000 dilution in 5% NFDM/TBST; Santa Cruz Biotechnology). Protein bands were visualized using enhanced chemiluminescence (PerkinElmer, Boston, MA). In some cases, membranes were stripped at 56 C for 30 min in buffer containing 100 mM ß-mercaptoethanol, 2% SDS, and 62.5 mM Tris-HCl (pH 6.7); the blots were then washed extensively and reprobed with relevant antibodies such as ß-actin to standardize lane loading. All Western blot studies were completed independently on at least three different occasions. The results of these replicate studies were similar.
Transient Transfection and Luciferase Assay
For transfection studies, all plasmids were prepared by two cycles through cesium chloride using standard methods. T3-1 cells were transiently cotransfected with the luciferase reporter constructs (e.g. human –204-Luc) and expression vector for ß-galactosidase (as an internal control for transfection efficiency) using calcium phosphate precipitation methodologies. For studies in Figs. 5, 6, 8, and 10, cells were washed with Dulbecco’s PBS 16 h after cotransfection, fresh serum-containing medium was added, and cells were treated with control solution or GnRHa for 6 h before collection. In some experiments, the MAPK kinase 1 inhibitor PD98059 and the JNK pathway inhibitor SP600125 were administered to cells 30 min before GnRHa administration. In other studies, overexpression of HA-ATF3 was carried out using cotransfection of HA-ATF3 expression vector (4 µg) with luciferase reporters. Overexpressed HA-ATF3 displayed a retarded electrophoretic mobility on Western blot analysis using the ATF3 antibody, providing a mechanism for detection of overexpressed ATF3. The triple HA epitope increased the mass of ATF3 by approximately 3500 Da, accounting for the shift in electrophoretic mobility. All transient transfection experiments were conducted in the presence of culture medium containing 10% fetal bovine serum. After completion of a given study, luciferase activity was determined as previously described (66) in samples standardized for ß-galactosidase activity (BD Biosciences, Palo Alto, CA). All transfection studies were conducted in triplicate on at least three separate occasions with similar results. Data shown are reported as means (n = 3) ± SEM.
For the ATF3 siRNA studies, cells were cultured in 35-mm plates and grown to approximately 50–60% confluence before use. Cells were transfected using Lipofectamine 2000 (Invitrogen) according to manufacturer’s instruction. Equal amounts of double-stranded control (scrambled sequence) or specific ATF3 siRNA (Superarray Biosciences Corp., Fredrick, MD) were transfected into cells. Medium was changed 24 h later, and the cells were allowed to incubate an additional 24 h. For protein expression experiments, the cells were serum starved for 2 h followed by GnRHa for 1 or 2 h. The cells were then collected as whole-cell lysates in RIPA buffer and analyzed by Western blot for ATF3 as described above. For reporter gene/luciferase studies using siRNA, cells were transfected using a dual transfection methodology. Initially, siRNAs were transfected using lipofection for 24 h, the medium was changed, and cells were incubated for an additional 24 h as described above. On the morning of the second day, the human –204-Luc reporter and the expression vector for ß-galactosidase were cotransfected into cells using calcium phosphate precipitation for a 3-h period. Treatment with GnRHa (6 h) then followed this second transfection. Cells were collected for determination of luciferase/ß-galactosidase activity as described above. The siRNA studies were completed three times with equivalent results.
Preparation of Nuclear Extracts, Recombinant ATF3, and c-Jun and DNA Binding Assays
For preparation of nuclear extracts, T3-1 cells were serum starved for 2 h followed by administration of control solution or GnRHa for a 2-h period. Nuclei were purified from cells using a sucrose density centrifugation methodology described previously (13). Protein concentration was determined by the Bradford assay. Nuclear extracts were aliquoted and stored at –80 C until later use. Recombinant ATF3 and c-Jun were prepared in vitro using a coupled transcription and translation system with rabbit reticulocyte lysates (Promega Corp., Madison, WI) according to the manufacturer’s instructions.
The CRE pull-down assay has been described previously (43). Briefly, annealed dual CRE oligonucleotides were bound to streptavidin agarose (SA) beads and incubated with nuclear extracts (control and GnRHa treated) from T3-1 cells. CRE binding reactions were carried out in a binding buffer containing 10 mM HEPES (pH 7.4), 2.5 mM MgCl2, 3.6 mM KCl, 150 mM NaCl, 10% glycerol, 1 mM dithiothreitol, 5 mM benzamadine, and 0.2 mM phenylmethylsulfonyl fluoride with 50 pmol of SA-bound CRE, 10 µg of sonicated salmon sperm DNA, and 100 µg of nuclear protein in a total volume of 500 µl. Binding reactions were incubated for 2 h at 4 C with constant rocking. In some cases, nuclear extracts were bound to SA beads alone as a negative control and in other cases, 20-fold molar excess of nonbiotinylated -subunit CREs, the SGII single consensus CRE, or CCAAT oligonucleotides were added as homologous and heterologous competitors. Bound complexes were washed and resuspended in a buffer (100 mM Tris; pH 6.8; 4% SDS; 5% glycerol; 0.01% bromophenol blue; 2x SDS loading buffer) and boiled. The binding complexes were resolved by SDS-PAGE and examined by Western blot analysis using ATF3 or c-Jun antibodies. These experiments were carried out with three separate sets of nuclear extracts prepared independently on separate occasions. Results were similar with all three sets of nuclear extracts.
For EMSAs, 1 ng 32P-labeled dual CRE oligonucleotide from the -subunit promoter was incubated for 30 min at room temperature with recombinant ATF3 (1 µl), c-Jun (5 µl), or the combination of ATF3 and c-Jun in a volume of 25 µl of binding buffer containing (final concentration) 50 mM KCl, 2.5 mM MgCl2, 1 mM EDTA, 1 mM dithiothreitol, 10 mM Tris-HCl (pH 7.5), 10% glycerol, and 1 µg polydeoxyinosinic deoxycytidylic acid. Some reactions contained ATF3, c-Jun, or CREB antibodies or preimmune normal rabbit serum. DNA-protein complexes were resolved by electrophoresis through a native 6% polyacrylamide gel and dried. Dried gels were exposed to film at –70 C with intensifying screens to visualize DNA-protein complexes.
For immunoprecipitation studies, the c-Jun antibody was incubated with nuclear extracts (total protein, 500 µg; precleared for 1 h with protein A/G agarose in the absence of primary antibody followed by removal of the agarose beads) for 2 h in RIPA buffer with constant rocking at 4 C. Protein A/G agarose beads were added for an additional 2 h at 4 C. The immunoprecipitates were washed five times with RIPA buffer and then eluted from the agarose beads by boiling in 2x SDS loading buffer and resolved by SDS-PAGE. Western blot analyses were carried out as described above. This experiment was conducted using three independently prepared nuclear extracts and all provided similar results.
|
FOOTNOTES |
|---|
First Published Online June 16, 2005
Abbreviations: A-Fos, Acidic Fos; AP-1, activator protein 1; ATF, activating transcription factor; bZip, basic leucine zipper; CRE, cAMP response element; CREB, CRE-binding protein; CT, detection threshold; DAG, diacylglycerol; Egr, early growth response; GnRHa, GnRH agonist; GnRHR, GnRH receptor; HA, hemagglutinin; JNK, c-Jun N-terminal kinase; NFDM, nonfat dry milk; PKC, protein kinase C; qPCR, quantitative real-time PCR; RIPA, radioimmunoprecipitation; SA, streptavidin agarose; SDS, sodium dodecyl sulfate; siRNA, small inhibitory RNA; TBST, Tris-buffered saline-Tween 20.
Received for publication January 24, 2005. Accepted for publication June 6, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit, growth hormone, and prolactin after hypothalamic pituitary disconnection in ovariectomized ewes. Endocrinology 119:2704–2710-subunit promoter by gonadotropin-releasing hormone. Mol Cell Biol. 15:3531–3539
T3-1 cell line is mediated by protein kinase C, c-Src, and CDC42. Mol Endocrinol. 12:815–824
T3-1 cells. Endocrinology. 144:839–849
-subunit promoter. Mol Endocrinol. 6:1767–1773
subunit gene bind similar proteins in trophoblasts and gonadotropes but have distinct functional sequence requirements. J Biol Chem. 271:31650–31656
subunit gene by epidermal growth factor and forskolin. Mol Cell Biol. 20:3331–3344
-subunit gene. Mol Endocrinol. 6:893–903
-induced ATF3 expression is bidirectionally regulated by the JNK and ERK pathways in vascular endothelial cells. Genes Cells 9:59–70-subunit gene occurs through several distinct mechanisms. Mol Endocrinol. 4:573–582
subunit gene. J Biol Chem. 265:21874–21880
- and ß-subunit promoters by AP-2. J Biol Chem. 272:15405–15412
-subunit gene in pituitary gonadotropes. Mol Endocrinol. 8:878–885
-subunit gene in the placenta and pituitary gland. Mol Cell Biol. 14:5592–5602
subunit gene in choriocarcinoma cells. J Biol Chem. 276:10016–10024
-subunit glycoprotein hormone promoter with trophoblast specificity and maximal activity. Mol Endocrinol. 11:1669–1680
-subunit transcription in T3-1 gonadotropes. Endocrinology 139:1731–1737-subunit promoter. Endocrinology 144:612–622-subunit promoter by a LIM-homeodomain transcription factor. Mol Cell Biol. 14:2985–2993This article has been cited by other articles:
 |
|
 |
|
  T. B. Salisbury, A. K. Binder, J. C. Grammer, and J. H. Nilson GnRH-Regulated Expression of Jun and JUN Target Genes in Gonadotropes Requires a Functional Interaction between TCF/LEF Family Members and {beta}-Catenin Mol. Endocrinol., March 1, 2009; 23(3): 402 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. I. Mayer, V. Dexheimer, E. Nishida, S. Kitajima, and G. Thiel Expression of the Transcriptional Repressor ATF3 in Gonadotrophs Is Regulated by Egr-1, CREB, and ATF2 after Gonadotropin-Releasing Hormone Receptor Stimulation Endocrinology, December 1, 2008; 149(12): 6311 - 6325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. L. Burger, D. J. Haisenleder, K. W. Aylor, and J. C. Marshall Regulation of Intracellular Signaling Cascades by GNRH Pulse Frequency in the Rat Pituitary: Roles for CaMK II, ERK, and JNK Activation Biol Reprod, November 1, 2008; 79(5): 947 - 953. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Xie, K. H. Allen, A. Marguet, K. A. Berghorn, S. P. Bliss, A. M. Navratil, J. L. Guan, and M. S. Roberson Analysis of the Calcium-Dependent Regulation of Proline-Rich Tyrosine Kinase 2 by Gonadotropin-Releasing Hormone Mol. Endocrinol., October 1, 2008; 22(10): 2322 - 2335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Sasson, S. H. Luu, V. G. Thackray, and P. L. Mellon Glucocorticoids Induce Human Glycoprotein Hormone {alpha}-Subunit Gene Expression in the Gonadotrope Endocrinology, July 1, 2008; 149(7): 3643 - 3655. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. B. Salisbury, A. K. Binder, and J. H. Nilson Welcoming {beta}-Catenin to the Gonadotropin-Releasing Hormone Transcriptional Network in Gonadotropes Mol. Endocrinol., June 1, 2008; 22(6): 1295 - 1303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Beleza-Meireles, V. Tohonen, C. Soderhall, C. Schwentner, C. Radmayr, I. Kockum, and A. Nordenskjold Activating transcription factor 3: a hormone responsive gene in the etiology of hypospadias Eur. J. Endocrinol., May 1, 2008; 158(5): 729 - 739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Xie and M. S. Roberson 3', 5'-Cyclic Adenosine 5'-Monophosphate Response Element-Dependent Transcriptional Regulation of the Secretogranin II Gene Promoter Depends on Gonadotropin-Releasing Hormone-Induced Mitogen-Activated Protein Kinase Activation and the Transactivator Activating Transcription Factor 3 Endocrinology, February 1, 2008; 149(2): 783 - 792. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Haisenleder, L. L. Burger, H. E. Walsh, J. Stevens, K. W. Aylor, M. A. Shupnik, and J. C. Marshall Pulsatile Gonadotropin-Releasing Hormone Stimulation of Gonadotropin Subunit Transcription in Rat Pituitaries: Evidence for the Involvement of Jun N-Terminal Kinase But Not p38 Endocrinology, January 1, 2008; 149(1): 139 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. Bliss, A. M. Navratil, M. Breed, D. C. Skinner, C. M. Clay, and M. S. Roberson Signaling Complexes Associated with the Type I Gonadotropin-Releasing Hormone (GnRH) Receptor: Colocalization of Extracellularly Regulated Kinase 2 and GnRH Receptor within Membrane Rafts Mol. Endocrinol., February 1, 2007; 21(2): 538 - 549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Ferris and M. A. Shupnik Mechanisms for Pulsatile Regulation of the Gonadotropin Subunit Genes by GNRH1 Biol Reprod, June 1, 2006; 74(6): 993 - 998. [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 |
