Molecular Endocrinology, doi:10.1210/me.2004-0519
Molecular Endocrinology 19 (11): 2780-2797
Copyright © 2005 by The Endocrine Society
Gonadotropin-Releasing Hormone (GnRH) Positively Regulates Corticotropin-Releasing Hormone-Binding Protein Expression via Multiple Intracellular Signaling Pathways and a Multipartite GnRH Response Element in 
T3-1 Cells
Nicole J. Westphal and
Audrey F. Seasholtz
Neuroscience Program and Molecular and Behavioral Neuroscience Institute, University of Michigan, Ann Arbor, Michigan 48108
Address all correspondence and requests for reprints to: Audrey F. Seasholtz, University of Michigan, Molecular and Behavioral Neuroscience Institute, 205 Zina Pitcher Place, Ann Arbor, Michigan 48108. E-mail: aseashol{at}umich.edu
 |
ABSTRACT
|
|---|
CRH-binding protein (CRH-BP) binds CRH with high affinity and inhibits CRH-mediated ACTH release from anterior pituitary corticotrope-like cells in vitro. In female mouse pituitary, CRH-BP is localized not only in corticotropes, but is also expressed in gonadotropes and lactotropes. To investigate the functional significance of gonadotrope CRH-BP, we examined the molecular mechanisms underlying GnRH-regulated CRH-BP expression in 
T3-1 gonadotrope-like cells. CRH-BP is endogenously expressed in T3-1 cells, and quantitative real-time RT-PCR and ribonuclease protection assays demonstrate that GnRH induces a 3.7-fold increase in CRH-BP mRNA levels. GnRH also induces intracellular CRH-BP (2.0-fold) and secreted CRH-BP (5.3-fold) levels, as measured by [125I]CRH:CRH-BP chemical cross-linking. Transient transfection assays using CRH-BP promoter-luciferase constructs indicate that GnRH regulation involves protein kinase C-, ERK- and calcium-dependent signaling pathways and is mediated via a multipartite GnRH response element that includes activator protein 1 and cAMP response element (CRE) sites. The CRE site significantly contributes to GnRH responsiveness, independent of protein kinase A, representing a unique form of multipartite GnRH regulation in T3-1 cells. Furthermore, EMSAs indicate that T3-1 nuclear proteins specifically bind at activator protein 1 and CRE sites. These data demonstrate novel regulation of pituitary CRH-BP, highlighting the importance of the pituitary gonadotrope as a potential interface between the stress and reproductive axes.
|
INTRODUCTION
|
|---|
CRH IS THE KEY NEUROENDOCRINE modulator of the mammalian stress response. In response to stress, CRH is secreted from hypothalamic paraventricular nucleus neurons and stimulates the synthesis and release of pituitary ACTH, which in turn circulates peripherally to stimulate glucocorticoid release from the adrenals (1). In addition to its neuroendocrine role, CRH localizes to other central nervous system sites where it is thought to act as a neurotransmitter to mediate behavioral, reproductive, autonomic, metabolic, and immune responses to stress (1, 2). Several other mammalian CRH-like ligands, urocortin (3), urocortin II/stresscopin-related peptide (4, 5), and urocortin III/stresscopin (5, 6) were recently discovered, adding to the complexity of this family of peptides.
The CRH family of peptides signals via two distinct seven transmembrane G-protein coupled receptors, CRH receptor type I (CRH-R1) and CRH receptor type II (CRH-R2) (7). In addition to the two receptors, CRH also binds to CRH-binding protein (CRH-BP), a 37-kDa secreted glycoprotein (8). CRH-BP binds CRH with an equal or greater affinity than the CRH receptors and has been shown to be an important modulator of CRH activity (9). Rodent CRH-BP is expressed in pituitary and brain (10), and in many cases CRH-BP localizes to sites of CRH expression or CRH target sites, including anterior pituitary corticotropes.
Although the function of CRH-BP may vary with cellular context, data support an inhibitory role for CRH-BP at the anterior pituitary corticotrope (11, 12, 13, 14). Based on both in vitro and in vivo data, pituitary CRH-BP may act as a local homeostatic regulator of CRH activity, binding free CRH and targeting it for clearance or degradation. Recent in vivo work from our laboratory has shown pituitary expression of CRH-BP to include not only corticotropes, but also gonadotropes and lactotropes in female mice (15). Its expression was only recently discovered in these cell types, so little is known about its regulation or function. Interestingly, both the pituitary gonadotrope and lactotrope are classically associated with endocrine regulation of various aspects of reproductive function. Because pituitary CRH-BP has been shown to be an important negative regulator of CRH activity at the corticotrope, the localization of CRH-BP in additional anterior pituitary cell types suggests that these cells may represent additional sites of interaction between the stress and reproductive axes.
Both experimental studies and clinical observations document the deleterious effects of stress on reproductive function (16, 17). Although much of this inhibition has been attributed to central mechanisms (16, 18), the recent characterization of CRH receptors on rat gonadotropes (19) and on mouse T3-1 cells (20), a gonadotrope-like cell-line, and our evidence of CRH-BP expression in gonadotropes in vivo (15), suggest that the pituitary gonadotrope should be reexamined as a potential site of interaction between the stress and reproductive systems. To begin to address this interaction, we examined the regulation of CRH-BP expression by GnRH, the primary hypothalamic peptide signaling at gonadotropes, and we performed a detailed analysis of the underlying molecular mechanisms.
GnRH-signaling pathways have been extensively studied and reviewed in a number of cell systems (21, 22, 23). T3-1 cells were used in the present study because they have been widely used as a model system for investigating GnRH signaling and endogenously express GnRH receptors, CRH-BP, and the CRH receptors (20, 24, 25, 26). In T3-1 cells, the GnRH receptor couples exclusively to Gq/G11 G proteins (27). GnRH binding activates phospholipase C, increasing inositol 1,4,5-triphosphate and diacylglycerol formation and activating downstream protein kinase C (PKC) signaling pathways. GnRH also increases intracellular calcium levels, from both extracellular influx and mobilization of intracellular stores. Furthermore, in addition to PKC and calcium, GnRH receptors activate downstream ERK, jun-N-terminal kinase (JNK), p38MAPK, big MAPK, and protein kinase A (PKA) signaling within pituitary gonadotropes and gonadotrope-like cells (22, 23, 25). The previously documented regulation of CRH-BP by PKA, PKC, and MAPK signaling pathways in neuronal and astrocyte cultures (28, 29, 30) led us to hypothesize that GnRH positively regulates CRH-BP expression in gonadotropes utilizing similar pathways and mechanisms of action. In this study we describe novel regulation of CRH-BP expression by GnRH within pituitary gonadotrope-like cells.
|
RESULTS
|
|---|
GnRH Positively Regulates Endogenous CRH-BP Expression in T3-1 Cells
Previous studies from our laboratory demonstrated that CRH-BP is expressed in murine gonadotropes in vivo. RT-PCR analysis confirmed that the T3-1 gonadotrope-like cell line expressed CRH-BP transcript whereas the LßT2 cell line (another gonadotrope-like cell line) did not express detectable levels of CRH-BP transcript (data not shown). As T3-1 and LßT2 cell lines are of independent clonal origin, this result is consistent with the observation that CRH-BP is expressed in only a subset of gonadotropes in vivo. GnRH-induced changes in CRH-BP steady-state mRNA levels in T3-1 cells were examined using real time RT-PCR and ribonuclease protection assays (RPAs) (Fig. 1
). Results obtained from real time RT-PCR analysis demonstrated that GnRH stimulated a significant 3.75-fold increase in CRH-BP steady-state mRNA expression when compared with the control (Fig. 1B). CRH-BP levels were normalized with succinate dehydrogenase complex, subunit C (SDHC) and TATA-binding protein gene expression with similar results and corrected for efficiency (see Materials and Methods). SYBR green was used to detect amplification of the PCR products. Products were run on a 2% agarose gel to confirm that a single product of expected size was produced (Table 1 and Fig. 1A). Melt curve analysis produced similar results (data not shown). Figure 1C is a representative RPA image of the CRH-BP-protected fragments, clearly illustrating GnRH-stimulated increases in CRH-BP mRNA as compared with the control. CRH-BP-protected fragments are present in mouse cortex and absent in yeast RNA, as expected. Consistent with the real time RT-PCR experiments, quantitation of the normalized RPA hybrid bands demonstrated a 3.7-fold increase in CRH-BP steady-state mRNA levels in GnRH-treated cells compared with the control (Fig. 1D).
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 1. GnRH Positively Regulates CRH-BP Steady-State mRNA Levels in T3-1 Cells as Measured by Real-Time RT-PCR and RPAs
A, Real time RT-PCR products from control (top panel) and GnRH-treated (100 nM, 6 h; bottom panel) were run on a 2% agarose gel to confirm that all primer pairs produced a single specific band of expected size (See Table 1 for expected sizes). The control +RT reactions reached the threshold cycle (Ct) at 28.61 ± 0.59 cycles whereas the GnRH treated +RT reactions reached Ct at 26.65 ± 0.65 cycles (n = 3). Abbreviations: mBP, mouse CRH-BP; TBP, TATA binding protein; Cont, control. Numbers to the left of the panel in Fig. 1A are DNA size standards in base pairs. Data shown are from a representative experiment. B, Histogram of normalized fold inductions in CRH-BP mRNA after GnRH treatment from real time RT-PCR experiments. GnRH (100 nM, 6 h) induced CRH-BP mRNA expression in T3-1 cells by 3.75-fold (*, P < 0.05, as compared with control). Assays were performed in triplicate and normalized with SDHC and TBP gene expression. Expression ratios were calculated with PCR efficiency corrections (all efficiencies were greater than 95%, see Materials and Methods). Data are represented as the mean ± SEM of three independent experiments, each performed in triplicate. Statistical significance was reached when P < 0.05 (95% confidence level). C, Representative phosphoimage from RPA, demonstrating that GnRH (100 nM, 6 h) treatment positively regulated endogenous CRH-BP steady-state mRNA levels in T3-1 cells. Yeast RNA (yst) was used as a negative control and mouse cortex RNA (mCtx) was used as a positive control for the CRH-BP riboprobe. Total T3-1 RNA (20 µg) was used for each hybridization reaction. The panel in C depicts CRH-BP-protected hybrids from a 24-h exposure. D, Histogram of fold GnRH-induced CRH-BP expression, normalized by L3 gene expression (internal control), from RPA experiment. GnRH (100 nM, 6 h) induced CRH-BP expression by 3.7-fold in T3-1 cells. CRH-BP hybrid densities were divided by L3 hybrid densities to normalize for RNA concentration, recovery, and loading. Data are represented as normalized fold induction over control of a representative experiment, repeated twice.
|
|
To examine whether GnRH stimulated a concomitant increase in CRH-BP protein expression and secretion, [125I]CRH:CRH-BP cross-linking assays were performed on cellular media and lysates harvested from control T3-1 cells or cells stimulated with GnRH. Representative cross-linking assays are depicted in Fig. 2A. GnRH treatment resulted in a significant 2.0-fold induction in intracellular levels of CRH-BP when compared with the control (Fig. 2B). GnRH also increased CRH-BP secretion 5.3-fold when compared with the control (Fig. 2C). As GnRH is known to activate PKC signaling pathways in T3-1 cells, we also tested whether CRH-BP protein expression and secretion is regulated by PKC. Cells were treated with phorbol 12-myristate 13-acetate (TPA), an activator of PKC, and changes in CRH-BP levels were measured by cross-linking. Similar to GnRH, TPA treatment resulted in significant increases in intracellular levels of CRH-BP protein, as shown in Fig. 2B (1.7-fold over control), and increases in secreted protein, as shown in Fig. 2C (4.8-fold over control). These results demonstrate that both GnRH and activators of PKC signaling positively regulate intracellular and secreted CRH-BP levels and suggest that PKC signaling pathways may be an important mediator of GnRH-induced CRH-BP expression. Together, these results clearly demonstrate basal CRH-BP expression and GnRH-induced increases in CRH-BP mRNA, intracellular protein, and secretion in T3-1 cells.
View larger version (60K):
[in this window]
[in a new window]
|
Fig. 2. GnRH Positively Regulates Intracellular and Secreted CRH-BP Protein Expression in T3-1 Cells as Measured by [I125]CRH:CRH-BP Chemical Cross-Linking
A, Representative phosphoimage from [125I]CRH:CRH-BP cross-linking assays from T3-1 cells treated with GnRH and TPA. Both treatments increased intracellular CRH-BP levels (Cellular Lysate) and secreted CRH-BP levels (Culture Media), as demonstrated by increased [125I]CRH: CRH-BP complex. Protein content was determined by Bio-Rad protein assay, and equivalent amounts of protein (150 µg) were cross-linked in each reaction. B, Histogram of the cellular lysate data represented in panel A. GnRH (100 nM, 24 h) and TPA (100 nM, 24 h) treatment significantly increased intracellular CRH-BP protein expression. Data are represented as the mean ± SEM of three independent experiments. *, P < 0.015; **, P < 0.002 as compared with untreated control. Statistical significance was reached when P < 0.0167 (95% confidence level). C, Histogram of the culture media data represented in panel A. GnRH (100 nM, 18 h) and TPA (100 nM, 18 h) treatment resulted in a 5.3-fold and 4.8-fold increase in secreted CRH-BP levels. Data are represented as normalized fold induction over control of a single representative experiment, repeated twice.
|
|
The CRH-BP Promoter Is Responsive to GnRH in T3-1 Cells
T3-1 cells exhibit GnRH regulation of endogenous CRH-BP expression and provide an excellent model cell system in which to examine the underlying molecular mechanisms. We examined GnRH regulation of a luciferase reporter vector containing 2.5 kb of the proximal mouse (m) CRH-BP promoter (m2.5BP-pGL3) in transiently transfected T3-1 cells. GnRH or GnRH receptor agonist (GnRHa) treatment alone resulted in a significant increase in CRH-BP promoter activity (Fig. 3; P < 0.0001). This result demonstrates that the DNA sequences required for GnRH regulation of CRH-BP transcription are contained within the 2.5-kb region of 5'-flanking DNA. GnRH also stimulated a small, but significant, increase in activity from the promoterless vector (pGL3-basic); however, the 1.5-fold increase was insignificant when compared with the 18-fold increase in GnRH-stimulated activity from the m2.5BP-pGL3 vector. To demonstrate that this effect was GnRH receptor dependent, cells were treated with GnRH or GnRHa, and a GnRH receptor antagonist (Antide; Fig. 3). Antide cotreatment inhibited GnRH- and GnRHa-induced CRH-BP promoter activity to levels not different than control levels (P < 0.0001) and had no effect alone. These results demonstrate that GnRH-induced increases in CRH-BP promoter activity are GnRH receptor dependent and likely involve downstream GnRH-activated intracellular signaling pathways.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 3. GnRH-induced CRH-BP Promoter Activity Is GnRH-Receptor Dependent
Treatment of m2.5BP-pGL3-transfected T3-1 cells with GnRH (100 nM, 12 h) or the GnRH-receptor agonist des-Gly10, [D-Ala6]-LH-RH ethylamide (GnRHa; 1 µM, 12 h) significantly induced m2.5BP-pGL3 promoter activity. Pretreatment with the GnRH receptor-specific antagonist, Antide (Ant; 1 µM, 30 min) completely abolished both GnRH and GnRHa-induced m2.5BP-pGL3 promoter activity. The relative promoter activity is represented as fold induction over control. T3-1 cells were also transfected with the promoterless pGL3-basic vector to test for nonspecific GnRH regulation (100 nM, 12 h). Values represent the mean ± SEM, n = 3. Each experiment was performed three independent times. *, P < 0.0001, as compared with GnRH and GnRHa, respectively; #, P < 0.0001 as compared with Control or Antide. Statistical significance was reached when P < 0.0033 (95% confidence level). **, P < 0.02 as compared with pGL3-basic control. Statistical significance was reached at P < 0.05 (95% confidence level).
|
|
GnRH-Induced CRH-BP Promoter Activity Is Mediated by Protein Kinase C
In amygdalar and/or astrocyte cultures, significant increases in rat CRH-BP mRNA were observed in response to forskolin, CRH, TPA, and IL-6 (28, 29, 30). Combined, these results indicate that multiple intracellular signaling pathways, including PKA-, PKC-, and MAPK-activating pathways, regulate CRH-BP expression. GnRH is known to activate many of the same pathways in gonadotropes, suggesting potential mechanisms for regulating CRH-BP expression. To elucidate the individual contributions of each pathway to GnRH-induced CRH-BP promoter activity, T3-1 cells were transiently transfected with m2.5BP-pGL3 and then treated with GnRH and specific pharmacological inhibitors.
We have shown that TPA, a direct activator of PKC, regulated endogenous CRH-BP protein levels in T3-1 cells using cross-linking assays (Fig. 2). To examine whether PKC also mediates GnRH-induced CRH-BP promoter activity, transiently transfected cells were cotreated with GnRH and PKC inhibitors. As Fig. 4A illustrates, GnRH significantly induced CRH-BP activity (P < 0.0001). Pretreatment with PKC-specific inhibitors, Bisindolylmaleimide I (Bis I) or Go6983, significantly inhibited GnRH-induced activity (P < 0.0001) to levels not different than control, Bis I, or Go6983 alone (P < 0.0001). As expected, Bis I and Go6983 had no effect alone. Pretreatment with excess TPA (1 µM, TPA-D), to deplete intracellular PKC stores, had no effect alone but significantly reduced GnRH-induced activity (P < 0.0001, Fig. 4B) to levels not different than control. All three methods of PKC inhibition resulted in significant reductions in GnRH-stimulated CRH-BP promoter activity to levels not different than the control, strongly supporting a major role for PKC-signaling pathways in mediating GnRH-induced CRH-BP activity.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 4. GnRH-Induced Increases in CRH-BP Promoter Activity Are PKC Dependent
A, GnRH (100 nM, 12 h) treatment of m2.5BP-pGL3-transfected T3-1 cells significantly induced CRH-BP activity. Pretreatment (45 min) with the PKC inhibitors, Bis I (µM) or Go6983 (1 µM) attenuated GnRH-induced increases in CRH-BP promoter activity to levels not statistically different than control. Bis I and Go6983 had no effect alone. Normalized luciferase activity is expressed as mean fold induction over control ± SEM. Each experiment was performed three independent times. a, P < 0.0001 as compared with control, Bis I, or Go6983; b, P < 0.0001 as compared with GnRH. Statistical significance was reached at P < 0.0014 (95% confidence level). B, T3-1 cells transfected with m2.5BP-pGL3 were pretreated with TPA (TPA-down-regulated, TPA-D; 1 µM, 16 h) or without TPA before GnRH stimulation (100 nM, 12 h), in the presence of TPA. Control and GnRH alone did not receive TPA treatment. Normalized luciferase activity is expressed as mean fold-induction over control ± SEM of triplicate samples. *, P < 0.0001 as compared with GnRH; #, P < 0.0001 as compared with control and TPA-D. Each experiment was performed three independent times. Statistical significance was reached at P < 0.0083 (95% confidence level).
|
|
MAPK-Signaling Pathways Mediate GnRH Activation of the CRH-BP Promoter
The intracellular signaling pathways activated by GnRH have been well studied in T3-1 cells and include PKC-dependent activation of MAPK pathways including ERK, JNK, p38MAPK, and big MAPK (23). We used the MAPK kinase 1/2-specific inhibitor, PD98059, to determine the relative involvement of ERK in mediating GnRH-induced CRH-BP activity. T3-1 cells transiently transfected with m2.5BP-pGL3 were treated with PD98059 in the presence or absence of GnRH. GnRH consistently induced CRH-BP promoter activity (Fig. 5, P < 0.0001). Treatment with PD98059 (50 µM) caused a small, but significant, increase in luciferase activity alone. Even so, PD98059 significantly inhibited GnRH-induced activity (P < 0.0001). These results suggest that the ERK signaling pathway mediates part of the GnRH-induced regulation of CRH-BP. To confirm the specificity of PD98059, Western blots using phosphospecific MAPK antibodies were performed on cellular lysates treated with the same inhibitor/GnRH combinations as described in Fig. 5A and then normalized with ß-tubulin expression to determine whether levels had changed (Fig. 5B). As expected, GnRH caused a robust increase in phosphorylated ERK (P-ERK), whereas 50 µM PD98059 inhibited GnRH-induced P-ERK (P-ERK blot). To confirm that PD98059 did not affect other phosphorylated MAPK levels, blots were stripped and reprobed with phosphospecific antibodies to p38MAPK and JNK. Treatment with PD98059 alone or in combination with GnRH did not significantly affect basal or GnRH-induced levels of phosphorylated p38MAPK (P-p38MAPK, P-p38blot). GnRH treatment increased phosphorylated JNK (P-JNK) levels that were not inhibited by PD98059 treatments (P-JNK blot). Data from the Western blots confirm that 50 µM PD98059 completely inhibited GnRH-induced P-ERK levels without significantly affecting the levels of P-p38MAPK or P-JNK and support an important role for ERK in mediating a portion of the GnRH-induced CRH-BP activity.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 5. GnRH-Induced CRH-BP Promoter Activity Is Dependent on ERK Signaling in T3-1 Cells
A, T3-1 cells were transfected with m2.5BP-pGL3 and pretreated (30 min) with PD98059 (PD, 50 µM) before addition of GnRH (100 nM, 12 h). a, P < 0.0001 as compared with control; b, P < 0.0001 as compared with GnRH or PD. Data were normalized by protein content and represented as mean fold induction over control ± SEM of triplicate samples. Data are representative of three independent experiments. Statistical significance was reached at P < 0.0083 (95% confidence level). B, To confirm inhibitor specificity and function, Western blots were performed on T3-1 lysates from cells treated with the same inhibitor combinations represented in panel A to detect phosphorylated MAPK family members, including p42/44 MAPK (P-ERK), p38MAPK (P-p38), and p46/54 MAPK (P-JNK). All inhibitors were added 30 min before GnRH treatment (100 nM, 30 min), and cells were harvested as described in Materials and Methods. Protein lysates were used for Western blot analysis, and equivalent amounts of protein (40 µg) were loaded in each lane. Blots were stripped and anti-ß-tubulin staining (ß-tub) was performed to normalize for differences in protein loading.
|
|
GnRH-Induced CRH-BP Promoter Activity Is Dependent on Extracellular Calcium
In addition to activating PKC and downstream MAPK pathways, GnRH stimulates increases in intracellular calcium levels (from both intracellular stores and extracellular influx). Because calcium signaling is a major arm of the downstream GnRH-activated signaling cascade, we investigated whether calcium was necessary for GnRH-induced CRH-BP activity. Cotreatment of cells with GnRH and the extracellular calcium chelator, EGTA, significantly inhibited GnRH-induced CRH-BP activity (Fig. 6A, P < 0.0001) to levels not different than the control. EGTA alone had no significant effect. The calcium ionophore, ionomycin, did not induce CRH-BP activity alone and had no effect on GnRH-induced levels. Pretreatment with the L-type calcium channel blocker, nifedipine, suggests that GnRH-induced regulation is dependent on functional L-type calcium channels (Fig. 6B); however, treatment with BayK8644, an L-type calcium channel agonist, was unable to induce significant CRH-BP promoter activity (data not shown). Combined, these data suggest that although calcium alone is not sufficient to induce CRH-BP promoter activity, calcium is necessary for GnRH-induced CRH-BP promoter activity in T3-1 cells.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 6. GnRH-Induced CRH-BP Promoter Activity Is Dependent on Extracellular Calcium and L-Type Calcium Channels in T3-1 Cells
A, T3-1 cells were transiently transfected with m2.5BP-pGL3 and cotreated with EGTA, an extracellular calcium chelator (2 mM, 12 h) and GnRH (100 nM) or ionomycin, a calcium ionophore (Iono; 500 nM, 12 h) and GnRH (100 nM). Data were normalized by protein content and are represented as mean fold induction over control ± SEM of triplicate samples. Data are representative of three independent experiments. a, P < 0.0001 as compared with control, EGTA, or ionomycin; b, P < 0.0001 as compared with GnRH. Statistical significance was reached at P < 0.0018 (95% confidence level). B, T3-1 cells that were transiently transfected with m2.5BP-pGL3 were pretreated with nifedipine (Nif, 1 µM, 45 min) before GnRH treatment (100 nM, 12 h). Data were normalized by protein content and are represented as mean fold induction over control ± SEM of triplicate samples. Data are representative of three independent experiments. a, P < 0.0001 as compared with control or nifedipine; b, P < 0.0001 as compared with GnRH. Statistical significance was reached at P < 0.0083 (95% confidence level).
|
|
GnRH-Induced CRH-BP Activity Is Not Dependent on PKA Signaling in T3-1 Cells
Studies from our laboratory and others demonstrated that CRH-BP gene expression is positively regulated by cAMP and PKA signaling pathways in mixed neuronal and astrocyte cultures (29, 30, 31). In T3-1 cells, the GnRH receptor does not couple to Gs proteins, and GnRH binding does not cause a measurable increase in intracellular cAMP (27). However, significant cross-talk between PKA and PKC signaling pathways does exist. We used PKI (protein kinase inhibitor-) to test whether PKA signaling pathways were involved in GnRH-induced increases in CRH-BP activity, either directly or indirectly (Fig. 7). PKI proteins are highly potent and specific competitive inhibitors of PKA and have been used to investigate cAMP-signaling pathways and cAMP-responsive promoters (32). As predicted, both forskolin and GnRH caused a significant increase in CRH-BP promoter activity (P < 0.002). Consistent with forskolin activation of PKA-signaling pathways, cotransfection with the PKI construct had no effect on basal activity but significantly inhibited forskolin-induced CRH-BP promoter activity (P < 0.0025). However, PKI expression had no effect on GnRH-induced CRH-BP promoter activity, confirming that GnRH-regulated CRH-BP expression is independent of PKA signaling in T3-1 cells.
The Proximal 377 bp of the Mouse CRH-BP Gene Contains the Necessary DNA Sequences Required for GnRH Regulation
To identify the DNA sequences conferring GnRH responsiveness, deletion mutagenesis of m2.5BP-pGL3 was performed to create a 662-bp (m662BP-pGL3) and 377-bp (m377BP-pGL3) promoter-luciferase construct for transient transfection in T3-1 cells. m2.5BP-pGL3, m622BP-pGL3, and m377BP-pGL3 all exhibited strong GnRH responsiveness (Fig. 8, P < 0.0001). These data suggest that the elements necessary for GnRH responsiveness are located within the proximal 377 bp of the promoter. To identify potential regulatory regions within this sequence, the proximal 377-bp promoter of three vertebrate species, human, rat, and mouse, was analyzed, and a number of highly conserved potential transcription factor-binding sites were identified using mVISTA and Transfac sequence analysis programs (Fig. 9). Earlier experiments indicated that PKC- and MAPK-signaling pathways are important in mediating GnRH-induced CRH-BP activity. Three putative activator protein 1 (AP-1) sites, which are known to be downstream targets of PKC- and MAPK-signaling pathways and have been shown to be important members of multipartite GnRH-responsive elements in other genes, were identified within the 377-bp upstream promoter. Within the same region, a conserved consensus CRE binding site was also identified. This CRE site has been shown to be important in regulating cAMP-induced rat CRH-BP reporter activity in other cellular contexts and could play a role in mediating GnRH responsiveness in murine T3-1 cells, in a PKA-independent fashion. Furthermore, two putative SP-1 sites were identified within the same region. SP-1 elements have been shown to be important regulatory elements in other GnRH-responsive genes and could contribute to GnRH regulation of CRH-BP activity.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 8. GnRH Responsive Regions Are Located in the Proximal 377 bp of the 5'-Flanking Sequence of the CRH-BP Gene
T3-1 cells were transfected with 5'-deletion constructs or the pGL3-basic vector and RSV-ß-gal. Cells were treated with GnRH (100 nM, 12 h). ß-Gal normalized data are represented as mean fold induction over control ± SEM of triplicate samples. Each experiment was performed three independent times. *, P < 0.0001 as compared with each respective control; #, P < 0.0001 as compared with pGL3 + GnRH. Statistical significance was reached at P < 0.0018 (95% confidence level).
|
|
View larger version (43K):
[in this window]
[in a new window]
|
Fig. 9. Sequence Homology of the 5'-Flanking DNA of the Mouse (M), Rat (R), and Human (H) CRH-BP Genes
Sequence homology between mouse, rat, and human 377-bp CRH-BP promoter fragments is shown. Homologous and divergent nucleotides are represented by upper- and lowercase letters, respectively. Conserved transcription factor-binding sites are bolded and labeled above the site. The * indicates the CRH-BP gene transcription start site, as identified in human liver. The human, rat, and mouse CRH-BP gene sequences were taken from Behan et al. (69 ), Cortright et al. (33 ), and Lesh, S., J. Woodworth, and A. Seasholtz, unpublished data, respectively.
|
|
Multiple AP-1-Binding Sites and a CRE-Binding Site Contribute to the GnRH Responsiveness of the CRH-BP Gene Promoter
Site-directed mutagenesis was performed on the three putative AP-1 sites, the CRE binding site, and the two putative SP-1 sites to determine the relative contributions of these sites alone or in combination. Site-directed mutants were constructed in the m377BP-pGL3 vector and used for transient transfection into T3-1 cells (see Materials and Methods for primer sequences). As Fig. 10A illustrates, GnRH did not significantly affect the activity of the vector alone but induced a significant increase in the activity of the nonmutated m377BP-pGL3 construct (P < 0.0001). Mutation of the most distal AP-1 site (AP-1A) did not affect GnRH regulation. However, mutation of the AP-1B and -C sites, alone or in combination, resulted in a significant reduction in GnRH-induced activity (P < 0.0001). The GnRH-induced activity of the AP-1B mutant remained statistically greater than its matched control (P < 0.001), whereas mutation of the AP-1C site alone and the AP-1B and -C sites together reduced GnRH-induced activity to levels not statistically different from their matched controls. Mutation of all three AP-1 sites (AP-1ABC) caused a reduction in GnRH-induced activity similar to that observed with the AP-1BC mutant. These data suggest that both the AP-1B and -C sites are functionally important for GnRH-regulated expression of CRH-BP, consistent with the importance of AP-1 sites in mediating GnRH signaling in other GnRH-responsive promoters. Mutation of the CRE binding site also caused a significant reduction in GnRH-induced activity (P < 0.0001), but GnRH-induced values remained significantly greater than control levels (P < 0.001). The reduction was similar to that observed with the AP-1B mutation. This is consistent with the functional importance of the CRE binding site in regulating CRH-BP promoter activity both here and in other cellular contexts (33). Furthermore, mutation of the SP-1A and -B sites, alone or in combination, had no effect on GnRH-induced CRH-BP activity (Fig. 10). Thus, these sites are not likely involved in mediating GnRH responsiveness of the CRH-BP promoter.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 10. The AP-1 and CRE Binding Sites Are Necessary for GnRH-Induced CRH-BP Promoter Activity in T3-1 Cells
Transient transfections were performed in T3-1 cells with either wild-type m377BP-pGL3 or mutant constructs with mutations in the AP-1, CRE, and/or SP-1 predicted binding sites. Cells were cotransfected with RSV-ßgal for normalization of transfection efficiency between constructs and treated with GnRH (100 nM, 12 h). Data are represented as normalized mean fold induction relative to untreated paired control ± SEM of triplicate samples. Data are representative of three independent experiments. a, P < 0.0001 as compared with untreated paired control for each construct; b, P < 0.0001 as compared with GnRH-treated wild-type m377BP-pGL3. Statistical significance was reached at P < 0.0002 (95% confidence level).
|
|
GnRH has been shown to signal via multipartite response elements in a number of other GnRH-responsive genes (34, 35, 36, 37, 38). Mutations in the two proximal putative AP-1 sites (B and C) and the CRE site all showed reduced GnRH-induced activity, and it is possible that the sites are members of a multipartite GnRH response element. Whereas mutation of the CRE alone or AP-1B alone did not reduce GnRH-induced activity to control levels, mutation of the CRE in conjunction with the three AP-1 sites (CRE/AP1) significantly inhibited GnRH-induced activity as compared with the GnRH-treated CRE mutant or AP-1B mutant (P < 0.0002). The GnRH-induced activity of the CRE/AP1 mutant was not statistically different from its paired control or the GnRH-treated AP-1BC and AP-1ABC mutations. These mutation data suggest that the two proximal AP-1 sites and the CRE binding site contribute significantly to the GnRH responsiveness of the CRH-BP promoter in T3-1 cells.
T3-1 Nuclear Proteins Specifically Bind the mCRH-BP AP-1B and AP-1C DNA Elements
Gel retardation competition assays were performed to determine whether proteins from T3-1 nuclear extracts could specifically bind the AP-1B or AP-1C sequences in the mCRH-BP gene promoter. Incubating a radiolabeled synthetic oligonucleotide containing the wild-type AP-1B element and flanking sequence with T3-1 nuclear extracts (8 µg) resulted in the formation of a single predominant complex (Fig. 11A). Specificity of the complex was confirmed by the strong competition obtained with 10-, 50-, or 200-fold molar excess unlabeled probe (lanes 4–6), whereas a similar amount of excess mutated probe (AP1B-M) was unable to reduce complex formation to the same degree (lanes 7–9). When a radiolabeled probe containing the wild-type AP-1C and flanking sequence was used, two major complexes were formed (Fig. 11B). The larger complex (C-1) was successfully competed away with 10-, 50-, and 200-fold molar excess unlabeled probe, whereas the mutant AP-1 C probe (AP1C-M) did not compete as efficiently. Complex C-2 appeared less abundant than C-1 and exhibited nonspecific competition at 200-fold excess by both the wild-type and mutant AP-1C probes. A radiolabeled probe containing a consensus AP-1 (AP1 CON) sequence was used to determine whether unlabeled AP-1B or AP-1C probes could compete for the same nuclear proteins that bind a consensus AP-1 element. Two major complexes were formed (Fig. 11C, C-1 and C-2), and strong competition was exhibited for both complexes with 10- or 200-fold molar excess of homologous probe (lanes 3–4). Competition with AP-1B probe resulted in diminished C-1 formation and complete abrogation of C-2 formation at 200-fold excess (lanes 5–6). Interestingly, the AP-1B mutant probe was unable to successfully reduce C-1 formation but easily reduced C-2 formation at 200-fold excess (lanes 7–8). Competition with unlabeled AP-1C probe reduced C-1 formation at both 10- and 200-fold excess but had only a minor effect on C-2 formation (lanes 10–11). Competition with the AP-1C mutant probe did not effect C-1 formation and had a minor effect on C-2 formation at 200-fold excess (lanes 12–13). These data suggest that the nuclear proteins comprising C-1 and C-2 are AP-1 specific; however, only those proteins that form complex C-1 can also specifically bind the AP-1 C and, with lower affinity, the AP-1B sequences. To test whether the protein-DNA complexes that formed with each probe were of the same size, binding reactions using the radiolabeled AP-1B, AP-1C, and AP-1 CON probes were analyzed in the same gel. Four major complexes were formed; however, the major specific complex for each probe did not appear to be of the same size (Fig. 11D; major specific complex: AP-1B, band C; AP-1C, band A; AP-1 CON, bands B and D). These size differences suggest that whereas AP-1 C and AP-1B probes can compete for T3-1 nuclear proteins that bind a consensus AP-1 element, a different combination of AP-1 protein family members is likely binding at the sequences endogenously.
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 11. The AP-1B and AP-1C Elements Specifically Bind T3-1 Nuclear Proteins
A, Nuclear (Nuc) extracts from T3-1 cells (8 µg) were incubated with a 27-bp radiolabeled probe containing the AP-1B element and flanking sequence from the mCRH-BP gene promoter. Specific binding to nuclear proteins was assessed by competition with 10-, 50-, and 200-fold molar excess of unlabeled AP-1B DNA (AP1B) or DNA with the AP-1B element mutated (AP1B-M). B, T3-1 nuclear extracts were incubated with a 25-bp radiolabeled probe containing the AP-1 C element and flanking sequence from the murine CRH-BP gene promoter. Specific binding to nuclear proteins was assessed by competition with 10-, 50-, and 200-fold molar excess of unlabeled AP-1 C DNA (AP1C) or DNA with the AP-1 C element mutated (AP1C-M). C, T3-1 nuclear extracts were incubated with a 23-bp radiolabeled probe containing a consensus AP-1 element (sequence from Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Specific binding to nuclear proteins was assessed by competition with 10- and 200-fold molar excess of unlabeled AP-1 consensus DNA (AP1 CON). To determine whether either AP-1B or AP-1C sequences described in panel A and B, respectively, could compete for proteins binding to the AP-1 consensus sequence, 10- and 200-fold molar excess of unlabeled AP-1B, AP-1B mutant, AP-1C, or AP-1C Mutant DNA sequences were added to the binding reactions. D, T3-1 nuclear extracts were incubated with the AP-1B, AP-1 C, and AP-1 consensus radiolabeled probes used in panels A, B, and C, respectively, to determine whether the different AP-1 probes formed DNA-nuclear protein complexes of similar size. Size differences in complex formation cannot be accounted for by differences in probe size (27 bp, 25 bp, and 23 bp, respectively). All binding reactions were subjected to electrophoresis through 5% nondenaturing polyacrylamide gels as described in Materials and Methods.
|
|
T3-1 Nuclear Proteins Specifically Bind the Murine CRH-BP Putative CRE
To investigate the binding capacity of the putative CRE, T3-1 nuclear extracts were incubated with a radiolabeled synthetic oligonucleotide containing the putative CRE and flanking sequence from the mCRH-BP gene promoter (CRE) or containing a consensus CRE and analyzed by gel retardation. When using the putative CRE probe, two major complexes and one minor complex were identified (Fig. 12A). Specificity of binding was confirmed by strong competition with 10-, 50-, and 200-fold molar excess of unlabeled homologous probe (lanes 3–5). A probe containing a mutated CRE sequence was unable to effectively reduce complex formation (lanes 6–8). Interestingly, competition with a consensus CRE probe abrogated complex 1 and 2 at similar levels of molar excess but had only a minor effect on complex 3 (lanes 9–11), suggesting that similar proteins could bind both the consensus CRE probe and the putative CRE sequence. Incubation of T3-1 nuclear extracts with a radiolabeled consensus CRE probe also resulted in the formation of three complexes, although C-2 was the most abundant (Fig. 12B). Specificity was confirmed by strong competition with 10- and 200-fold excess homologous probe (lanes 2–4). Similarly, competition with the putative mCRH-BP CRE probe was able to reduce complex formation whereas the mutant CRE probe had no effect (lanes 5–8). Furthermore, Fig. 12 C illustrates that the proteins binding each probe formed complexes of similar size and supports the hypothesis that similar proteins bind both the consensus CRE sequence and the putative mCRH-BP CRE sequence.
View larger version (69K):
[in this window]
[in a new window]
|
Fig. 12. The mCRH-BP CRE Specifically Binds T3-1 Nuclear Proteins That Competitively Bind to a Consensus CRE Sequence
A, Nuclear (Nuc) extracts from T3-1 cells (8 µg) were incubated with a 24-bp radiolabeled probe containing the CRE and flanking sequence from the murine CRH-BP gene promoter. Specific binding to nuclear proteins was assessed by competition with 10-, 50-, and 200-fold molar excess of unlabeled CRE DNA (CRE), DNA with the CRE mutated (CRE-M), or DNA containing a consensus CRE binding site (CRE-C; sequence from Santa Cruz Biotechnology). B, T3-1 nuclear extracts were incubated with a 27-bp radiolabeled probe containing a consensus CRE (CRE-C). Specific binding to nuclear proteins was assessed by competition with 10- and 200-fold molar excess of unlabeled CRE consensus DNA (CRE-C). To determine whether the CRE sequence described in panel A could compete for proteins binding to the CRE consensus sequence, 10- and 200-fold molar excess of unlabeled CRE DNA or mutated CRE DNA (CRE-M) were added to the binding reactions. C, T3-1 nuclear extracts were incubated with the CRE and CRE consensus (CRE-C) radiolabeled probes used in panels A and B, respectively, to determine whether the different CRE probes formed DNA-nuclear protein complexes of similar size. All binding reactions were subjected to electrophoresis through 5% nondenaturing polyacrylamide gels as described in Materials and Methods.
|
|
|
DISCUSSION
|
|---|
CRH-BP expression is highly regulated by multiple intracellular signaling pathways (13, 28, 29, 31, 33, 39, 40). Whereas numerous studies have focused on the regulatory role of CRH, steroid hormones, and stress, few have investigated additional endocrine regulators of pituitary CRH-BP expression. Our recent in vivo data demonstrating pituitary expression of CRH-BP, not only in corticotropes but also in lactotropes and gonadotropes, suggests that the hypothalamic releasing hormones signaling at these cell types may also regulate CRH-BP expression. We demonstrate, for the first time, that GnRH robustly regulates CRH-BP mRNA, intracellular protein, and secretion and stimulates CRH-BP promoter activity in vitro.
It is well known that GnRH activates a myriad of intracellular signaling pathways to regulate gene expression in gonadotropes. Regulation of the GnRH receptor gene by GnRH has been shown to include PKC/MAPK and calcium pathways, as well as PKA in certain cellular contexts (41). GnRH regulation of the glycoprotein hormone -subunit and the LHß and FSHß subunit genes involves PKC, MAPK, and calcium-dependent signaling pathways (42, 43, 44, 45). Similarly, our data demonstrate the involvement of a number of GnRH-activated signaling pathways in regulating CRH-BP promoter activity. Studies with pharmacological inhibitors show that GnRH regulation of the CRH-BP promoter is dependent on PKC signaling. Downstream of PKC activation, the ERK pathway contributes significantly to GnRH regulation of CRH-BP activity. However, in addition to ERK, it is possible that some of the GnRH-induced regulation is mediated via additional MAPK pathways. GnRH activates p38MAPK in T3-1 cells; however, the percent involvement of this pathway was obscured by the consistent and robust increase in levels of phosphorylated ERK after cotreatment with GnRH and the p38MAPK inhibitor, SB203580 (data not shown). Finally, the JNK-signaling pathway may also contribute because activation of this pathway has been shown to activate transcription factors known to bind AP-1 and CRE sites in other cellular contexts (46, 47).
In addition to activating PKC, Gq/11-coupled GnRH receptors classically increase intracellular calcium levels from both extracellular influx and intracellular stores. Although calcium is clearly important for gonadotropin secretion, numerous studies have shown that calcium and PKC differentially regulate the gonadotropin subunits gene expression (43, 44). Our results with EDTA, nifedipine, ionomycin, and BayK 8644 suggest that extracellular calcium is required, but not sufficient, for GnRH-regulated CRH-BP promoter activity. Although calcium cannot directly stimulate CRH-BP gene expression, substantial cross-talk exists between the calcium- and PKC-signaling pathways in T3-1 cells and could account for the calcium requirement (48). Recent data demonstrate that GnRH-dependent PKC activation of MAPK/ERK is calcium dependent and that calcium influx is downstream of PKC (49, 50). Because the CRH-BP promoter is highly regulated by PKC-signaling pathways, the calcium dependency could also be a result of a calcium-dependent PKC subtype or the involvement of a calcium-binding protein downstream of PKC (50).
Earlier studies from our laboratory illustrated that the rat CRH-BP promoter is responsive to PKA signaling in a number of cell types (33). Our present data demonstrate that the mouse CRH-BP promoter is responsive to forskolin stimulation, and hence PKA signaling, in T3-1 cells. Direct activation of PKA signaling by GnRH receptors is highly dependent on cellular context. In LßT2 gonadotropes, GGH3 gonadotropes, and primary pituitary cultures, the GnRH receptor has been shown to couple to Gs, activate downstream PKA signaling (25, 51), and regulate GnRH-responsive genes (52). In contrast, the endogenous GnRH receptor couples exclusively to Gq/11 proteins in T3-1 cells. However, due to the large degree of cross-talk between intracellular signaling pathways, including PKA and PKC (53), it remained possible that GnRH could indirectly activate PKA signaling to regulate CRH-BP activity in T3-1 cells. The PKI transfection studies demonstrate that although the CRH-BP promoter is responsive to PKA signaling, GnRH regulation of the CRH-BP promoter is independent of PKA-signaling pathways in T3-1 cells. Because cellular context is critical, it remains to be determined whether GnRH-activated PKA signaling could regulate CRH-BP promoter activity within additional cellular contexts or in vivo.
The GnRH regulation of GnRH-responsive genes, including CRH-BP, involves not only multiple intracellular signaling pathways, but also numerous cis-acting DNA elements. AP-1 DNA elements are well-established downstream targets of several signal transduction cascades, the most notable being PKC signaling. GnRH signals largely via PKC pathways in T3-1 cells, and AP-1 sites have been shown to be important elements for GnRH regulation in GnRH-responsive genes, including the GnRH receptor (41) and the FSHß subunit (54). Mutation of the two proximal AP-1 sites in the CRH-BP promoter caused a significant reduction in GnRH-stimulated reporter activity, demonstrating the importance of AP-1 sites in mediating GnRH responsiveness. Furthermore, gel retardation data illustrate that AP-1B and AP-1C probes could specifically bind T3-1 nuclear proteins and compete effectively for proteins binding a consensus AP-1 probe whereas their mutated counterparts could not. Although the transcription factors binding to these sites in vivo are unknown, transcription is initiated at AP-1 sites by the binding of heterodimers of immediate early gene products of the c-fos and c-jun family. GnRH treatment has been shown to increase both c-fos/c-jun in T3-1 cells (55). Together, these data suggest that the large family of b-zip proteins that bind AP-1 sites are candidate transcription factors regulating GnRH-induced CRH-BP expression.
CRE sites are also important for the regulation of gene expression in gonadotropes (34, 44, 46, 52). In T3-1 cells, the secretogranin II promoter is regulated by GnRH in a PKA-dependent fashion and involves a functional CRE site (56). We have shown that GnRH regulation of the CRH-BP promoter is PKA independent in T3-1 cells; however, our mutagenesis data strongly support the involvement of a functional CRE in the upstream promoter. CRE sites provide the binding site for the transcription factor CREB (CRE-binding protein) and in T3-1 cells, GnRH treatment increases levels of phosphorylated CREB (57). Although PKA signaling pathways are classically associated with CREB phosphorylation (58), other kinases have been shown to directly phosphorylate CREB including PKC (59), calcium-calmodulin-dependent kinase II/IV (60), ERK, and p38MAPK (61), suggesting additional mechanisms for CREB activation and CRE involvement in T3-1 cells. Furthermore, the gel retardation data demonstrated that the CRH-BP CRE site specifically bound T3-1 nuclear proteins and competed effectively for proteins binding a consensus CRE sequence, whereas its mutated counterpart could not. Together, these data provide a unique example of a CRE site contributing to GnRH responsiveness in a PKA-independent manner in T3-1 cells and suggest that CREB is a candidate transcription factor regulating GnRH-induced CRH-BP expression via binding at the CRE site.
Furthermore, we propose that the two proximal AP-1 sites and the consensus CRE site form a multipartite GnRH-response element. Examples of multipartite GnRH response elements in GnRH-responsive gene promoters are abundant (34, 35, 36, 37, 38). Although Sp1 sites have been shown to be members of multipartite response elements and important for GnRH responsiveness in other GnRH-regulated genes (35), they did not contribute to the GnRH responsiveness of the CRH-BP promoter. Mutation of the two proximal AP-1 sites and CRE sites alone or in combination resulted in dramatic reductions in GnRH-induced CRH-BP promoter activity, strongly indicating that each site is an important component of a multipartite GnRH response element.
The primary physiological function of CRH-BP is to modulate the biological activity of CRH, thereby acting as a local homeostatic regulator of CRH activity. Although the potential physiological significance of GnRH-regulated CRH-BP is unknown, it is plausible that regular GnRH pulses function to maintain a constant level of CRH-BP in the pituitary and provide a regular, but robust, mechanism controlling the bioavailability of CRH. When applied to the known function of pituitary CRH-BP, this novel means of regulating CRH-BP expression in gonadotropes may be a consistent source of CRH-BP necessary for returning the hypothalamic-pituitary-adrenal axis to homeostasis after a stressor.
Perhaps a more intriguing suggestion is that GnRH-regulated CRH-BP could have a gonadotrope-specific function, attenuating CRH signaling and thereby serving a protective role at the gonadotrope. Numerous studies have demonstrated that stress, and the actions of CRH specifically, can impair reproductive function. Although the central involvement of CRH has been carefully examined, studies have also shown that CRH can act at the pituitary to inhibit reproductive function. Although species-specific differences do exist (16), peripheral infusion of CRH into rhesus monkeys rapidly decreases LH levels in vivo (62), and treatment of primary rat pituitary cultures with CRH attenuates basal LH release (63, 64). The mechanisms and sites of action are unknown, but the recent characterization of CRH receptors on rat gonadotropes (19) and on mouse gonadotrope-like T3-1 cells (20) lends additional evidence that the gonadotrope is a target for CRH signaling. Because CRH-BP is known to attenuate CRH activity at pituitary corticotropes, GnRH-induced increases in gonadotrope CRH-BP, particularly resulting from the GnRH surge and immediately preceding ovulation, could buffer the system from CRH inhibition of gonadotropin secretion. It is likely that central mechanisms could override such an effect during times of severe or chronic stress; however, GnRH regulation of pituitary CRH-BP levels could play an important role in attenuating the effects of CRH following low-level stressors. Future studies in primary cultures and in vivo will investigate this potential regulation. Our data demonstrating robust GnRH-regulated CRH-BP expression in T3-1 cells support a novel means by which CRH-BP levels are regulated in the pituitary, highlighting the importance of the pituitary gonadotrope as a potential interface between the stress and reproductive axes.
|
MATERIALS AND METHODS
|
|---|
Cell Culture
T3-1 cells (24) were obtained from Dr. Pamela Mellon (University of California, San Diego, CA). These cells were maintained in DMEM (Invitrogen, Carlsbad, CA), 10% fetal calf serum (FCS, Hyclone, Logan UT), and gentamicin (50 µg/ml, Invitrogen) at 37C in 5% CO2.
Isolation of RNA
Cultures of T3-1 cells were plated on 10-cm plates (107 cells per plate) in phenol red-free DMEM + 10% charcoal stripped FCS + 50 µg/ml gentamicin and grown to 
80% confluence. GnRH (10–6 M, Sigma Chemical Co., St. Louis, MO) was added 6 h before harvest, and total cellular RNA was isolated using Trizol (Invitrogen) (according to manufacturer’s recommendations).
Real Time RT-PCR
Ten micrograms of total RNA from each sample were treated with RNase-free DNase according to the manufacturer’s recommendations (DNA free protocol; Ambion, Austin, TX). First-strand cDNA was synthesized using random hexamer primers and Superscript II reverse transcriptase [according to the manufacturer’s recommendations (Invitrogen)]. PCRs (50 µl) contained 1 µl of cDNA template, 25 µl 2x SYBR Green I Master Mix (Bio-Rad Laboratories, Inc., Hercules, CA), and 125 nM forward and reverse primer. Reactions were carried out in a Bio-Rad iCycler. The cycling conditions were as follows: polymerase activation 3 min at 95 C and 34 cycles at 95 C for 10 sec, 62 C for 20 sec, and 74 C for 30 sec followed by melt curve analysis and agarose gel (2%) detection, to determine whether primer dimers or multiple products had formed. All reactions were performed in triplicate, and each experiment contained–RT reactions for each sample and a no-template control for each primer set. PCR efficiencies for all primer pairs were determined by performing a standard curve of four serial dilution points of T3-1 cDNA, and all were above 95%. mCRH-BP gene-specific expression was normalized in parallel reactions with two different housekeeping genes, SDHC and TATA-binding protein. These genes were not regulated by GnRH treatment in T3-1 cells. Primer pairs for housekeeping genes were obtained from Primer Bank (65). Primer pairs for mCRH-BP were designed using Primer3 platform. SDHC and mCRH-BP primer pairs span at least one intron-exon boundary. Changes in gene expression levels of mCRH-BP were quantitated with efficiency correction using the following mathematical formula, ratio = (Etarget) 
CPtarget/(Eref) CPref in which E represents efficiency, CP is threshold crossing point, target is target gene, and ref stands for reference gene (66), which accounts for differences in primer efficiencies. Data represent the average of three separate experiments, and statistical significance was determined using Student’s t test.
RPAs
The templates used to produce the mouse CRH-BP-specific riboprobe and mouse L3 ribosomal subunit internal control riboprobe used in these experiments have been previously described (15). The linearized CRH-BP template produced a 268-base riboprobe that protected 248 bases of the CRH-BP coding region and 3'-untranslated region (UTR) whereas the linearized L3 template produced a 202-base riboprobe and protected a 104-base fragment of the L3 transcript.
Total RNA (20 µg) was used in each experimental RPA hybridization reaction. In a separate control reaction, 5 µg of mouse cortex total RNA + 15 µg yeast tRNA was used as a positive control for CRH-BP expression. RPAs were completed as previously described (15) with the following modifications. Briefly, the CRH-BP and L3 riboprobe transcription reactions contained 100 µCi and 25 µCi [32P]uridine 5'-triphosphate (>3000 µCi/mmol, ICN Pharmaceuticals, Inc.,) respectively. Each hybridization reaction received 500,000 cpm of the CRH-BP riboprobe, and 50,000 cpm of the L3 riboprobe (14 h, 48 C). RNA hybrids were digested using a mixture of RNase A/RNase T1 (250 U/ml RNase A; 10,000 U/ml RNase T1) diluted 1:50 in RNA Digestion Buffer Bx with 1% Glyco Blue for RNA pellet visualization (Ambion, Inc., Austin, TX). Hybrid bands were separated on 6% sequencing gels. Data were collected using PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA) and quantitated with ImageQuant software (Amersham Biosciences, Piscataway, NJ). In all RPAs, multiple bands of similar size were observed due to breathing at the ends of the RNA hybrids.
[I125]CRH Chemical Cross-Linking
Cultures of T3-1 cells were plated as described for RNA Isolation. Two hours before the addition of GnRH (10–6 M, 24 h), media were changed to phenol-free DMEM + 1x insulin-transferrin-selenium-A (Invitrogen) + gentamicin (50 µg/ml). Media from nearly confluent T3-1 cells were collected, and cells were harvested in cold 1x PBS (pH 7.4), pelleted, and lysed (0.5 M Tris (pH 8.0), 0.5% Triton-X100, 1x Protease Inhibitor cocktail (Sigma)]. Lysates were centrifuged at 10,000 rpm for 10 min at 4 C, and supernatants were collected and frozen until use. Media were centrifuged at 2000 rpm for 5 min to remove detached cells and concentrated 10x at 4 C (JM-10; Millipore Corp., Bedford, MA). Equivalent amounts of concentrated media were used in each cross-linking reaction. Lysate protein content was normalized using the Bio-Rad microtiter plate protein assay (Bio-Rad Laboratories). Changes in CRH-BP protein expression and secretion were measured by chemically cross-linking equivalent amounts of protein lysate (150 µg) to (2-[I125]iodohistidyl32) human CRH (Amersham Biosciences UK Limited, Little Chalfont, Buckinghamshire, UK) using disuccinimidyl suberate (Pierce Chemical Co., Rockford, IL) as previously described (11). Cross-linked products were separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gels and exposed to Biomax MS film with intensifying screens or phosphoimager screens for 2–7 d. This method has previously been shown to be quantitative with nonsaturating levels of protein (67). Changes in CRH-BP-[125I]CRH cross-linked product were quantitated using ImageQuant analysis software (Molecular Dynamics, Inc., Sunnyvale, CA). Lysate data represent the average of three separate experiments whereas media data are representative of two independent experiments. Statistical analysis was performed using a one-way ANOVA followed by a multiple comparison Bonferroni-Dunn post hoc test, where all selected groups were analyzed simultaneously. Statistical significance was reached when P < 0.0167 (95% confidence level). Analysis was performed using Statview (SAS Institute, Cary, NC).
Transient Transfections
Cultures of T3-1 cells were plated on six-well plates (300,000 cells per well) in phenol red-free DMEM + 10% charcoal stripped FCS + gentamicin (50 µg/ml). Cells were transfected the following morning with 1 µg luciferase reporter DNA (100 ng PKI as indicated) using Fugene (Roche, Indianapolis, IN) as recommended by the manufacturer. When multiple reporter plasmids were compared in a single experiment, cells were cotransfected with 0.5 µg Rous sarcoma virus-ß-galactosidase to normalize for transfection efficiency.
A time course profile was completed and demonstrated that the maximal GnRH stimulation of CRH-BP promoter activity in T3-1 cells was at 12 h of drug treatment. For all studies, 100 nM GnRH (Sigma) or 1 µM of the GnRH agonist des-Gly10, [D-Ala6]-LH-RH ethylamide (GnRHa, Sigma), 100 nM TPA (phorbol 12-myristate 13-acetate, Sigma), 1 µM forskolin (Calbiochem, San Diego, CA), 2 mM EGTA (Sigma), 500 nM ionomycin (Calbiochem), and 1 µM BayK 8644 (Calbiochem) were applied at 36 h posttransfection, 12 h before harvest (at 48 h posttransfection). Inhibitors including 1 µM Antide (Sigma), 1 µM Bis I (Calbiochem), 1 µM Go6983 (Calbiochem), 1 µM nifedipine (Calbiochem), or 50 µM PD98059 (Sigma), were applied 45 min before GnRH addition and remained in the media until harvest. Cells were harvested 48 h post transfection in cold 1x PBS, pelleted, and lysed in 150 µl lysis buffer [0.25 M Tris (pH 8.0), 0.1 M EDTA, 15 mM MgSO4, 1 mM dithiothreitol (DTT), and 1% Triton X-100] and incubated on ice 10 min. Lysates were centrifuged at 10,000 rpm for 10 min at 4 C, and 25 µl of each supernatant was added to 100 µl of luciferase assay buffer (33) and assayed 30 sec in a MGM Optocomp I (MGM Instruments, Inc., Hamden, CT) or Turner 20/20 (Promega, Madison, WI) luminometer. ß-Galactosidase activity was assayed as previously described (33). When multiple reporter plasmids were not being compared in a single experiment, Bio-Rad protein assays were used to normalize intersample variation. Normalized data are represented as fold induction over untreated vehicle control (untreated controls = 1.0). All experiments were performed in triplicate, all samples were assayed in duplicate, and each experiment was repeated at least three independent times. Statistical analysis was performed using a one-way ANOVA followed by a multiple comparison Bonferroni-Dunn post hoc test, where all selected groups were analyzed simultaneously. Analysis was performed using Statview software. P values that reached 95% confidence levels are included in each figure legend. Tukey post hoc analysis yielded similar significant differences.
Reporter Vectors
Transient transfection assays were performed using 2.5 kb, 662 bp, and 377 bp of 5'-flanking region from the mCRH-BP gene, including 12 bp of the 5'-UTR subcloned into the pGL3 basic vector (Promega Corp.). The start site was determined by homology to the human and rat transcription start sites. The 377-bp and 662-bp fragments were amplified by PCR with Tgo proofreading polymerase. EcoRV restriction sites were designed into the ends of the primer sequences, and the PCR products were subcloned into the pCR-Blunt II-TOPO vector (Invitrogen), isolated by digestion with EcoRV, and ligated into the SmaI site of pGL3-basic. The m2.5CRHBP-pGL3 vector was created by subcloning a 2.5-kb SacI DNA fragment containing approximately 2.5 kb of the proximal mCRH-BP gene promoter region and 12 bp of 5'-UTR into the pGL3 basic vector. Cells were transfected with the pGL3-basic vector in at least one parallel experiment for each drug treatment to test for regulation originating from the vector sequence. The control vector used to test for transfection efficiency contained the Rous sarcoma virus promoter linked to the cDNA encoding ß-galactosidase). For studies investigating the role of PKA signaling pathways, an expression construct encoding PKI under the control of a cytomegalovirus (CMV) promoter was transfected [pCMV.PKI3HA (32)]. Control plates were transfected with an equal amount of parental pCMV.neo plasmid DNA.
Deletion Analysis and Site-Directed Mutagenesis
To perform a more detailed analysis of the transcription factor-binding sites mediating GnRH-regulated CRH-BP expression, site-directed mutagenesis was performed on several of the conserved cis-binding sites in the upstream 377-bp mCRH-BP promoter region (QuikChange II XL Site-Directed Mutagenesis kit; Stratagene, La Jolla, CA). All mutations were introduced into the 377-bp mCRH-BP-pGL3 reporter vector. The complementary oligonucleotides used in each reaction are listed in Table 2.
Underlined nucleotides represent mismatches with the actual mCRH-BP sequence. The mutation in the CRE site has previously been shown to destroy the cAMP responsiveness of the functional CRE site in the rat CRH-BP gene (33). All mutations were confirmed by sequencing.
Western Blot Analysis
Western blots were performed on cell lysates treated with PD98059 to confirm that inhibitor concentrations used were sufficient to inhibit the ERK pathway without causing nonspecific inhibitory effects on other MAPK pathways. Cells were cultured as described and plated on 6-cm plates (500,000 cells per well) and grown to confluence. Cells were treated with 50 µM PD98059 for 30 min before GnRH (10–6 M, 30 min) addition. Cells were washed with 1x Dulbecco’s PBS (pH 7.4) and harvested with 200 µl phosphorylated Western lysis buffer [20 mM HEPES (pH 7.6), 10 mM EGTA, 40 mM ß-glycerophosphate, 2.5 mM MgCl2, 1 mM DTT, 2 mM sodium orthovanadate, 1% Triton-X-100, 1:50 dilution of Protease Inhibitor Cocktail (Sigma)]. Cells were triturated 20 times, rocked 10 min at 4 C, and centrifuged at 13,000 rpm for 10 min. Protein content of supernatant was determined by the Bio-Rad protein assay. Equivalent amounts of protein (40 µg) + 5x protein SDS-sample loading dye (25 mM Tris, pH 6.8; 0.008% bromophenol blue; 5% ß-mercaptoethanol; 35% glycerol; 2% SDS) was boiled for 5 min, iced, spun briefly, and loaded on a 10% SDS-polyacrylamide denaturing gel and run for 30 min at 60 V and 1.5 h at 165 V on ice. Gels were equilibrated 30 min in cold transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol) and transferred to Immobilon-P polyvinylidine difluoride membrane (0.45 µm, Millipore Corp., Billerica, MA) using the Semi-Dry transfer system according to the manufacturer’s recommendations (Bio-Rad) for 20 min at 15 V. Blots were air dried and blocked with 5% milk powder in 1x TBS-T (20 mM Tris-HCl; 150 mM NaCl; 0.1% Tween 20, pH 7.5) for 1 h at room temperature (RT) and incubated 24–72 h at 4 C in 5% BSA, 1x TBS-T, with the specified primary antibody. To detect changes in phosphorylated p38MAPK activity, a phosphospecific antibody to Thr180/Try182 (1:250) was used (New England Biolabs Cell Signaling, Beverly, MA). Similarly, to detect changes in phosphorylated ERK1/2 or JNK/stress-activated protein kinase, phosphospecific antibodies to Thr202/Tyr204 (1:1000) and Thr183/Tyr185 (1:250), respectively, were used (New England Biolabs Cell Signaling). Blots were probed with an horseradish peroxidase-conjugated secondary antibody (G
R-HRP; Sigma A0545, 1:160,000) in 1x TBS-T + 3% BSA for 2 h at RT with gentle rocking. Blots were visualized with chemiluminescence (ECL-plus; Amersham Pharmacia Biotech, Piscataway, NJ) and exposed to Hyperfilm (Amersham). To confirm equivalent protein loading between lanes, blots were stripped according to the manufacturer’s recommendations (ECL-Plus Detection Reagents; Amersham) and reprobed for ß-tubulin expression [Monoclonal Anti-ß-tubulin clone TUB 2.1; Sigma (1:10,000)].
EMSAs
Nuclear extracts were isolated from nearly confluent T3-1 cells, cultured in the conditions previously described. Cells were harvested, and the nuclear extracts were isolated as described (
TOP
ABSTRACT
INTRODUCTION
