| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Pharmacology (D.C.), Physiology (H.A.F.), and Internal Medicine (M.G., M.A.S.), University of Virginia, Charlottesville, Virginia 22908; and Institute of Biomedicine (M.H., O.A.J., J.J.P.) and Department of Clinical Chemistry (O.A.J.), University of Helsinki and Helsinki University Central Hospital, FIN-00014 Helsinki, Finland
Address all correspondence and requests for reprints to: Margaret A. Shupnik, Ph.D, Department of Internal Medicine/Endocrinology, PO Box 800578 Health Sciences Center, University of Virginia, Charlottesville, Virginia 22908. E-mail: mas3x{at}virginia.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and LHß, with the LHß subunit regulated most dramatically. Two enhancer regions, distal and proximal, on the rat LHß gene promoter cooperate for full basal expression and GnRH stimulation. It has been hypothesized that the transcription factors binding to these regions, Sp1, Egr-1, and steroidogenic factor 1 (SF-1), may interact directly or indirectly via a coactivator. One such coactivator may be small nuclear RING finger protein (SNURF), which is expressed in pituitary tissue and the LßT2 gonadotrope cell line. In transfection experiments in LßT2 cells, SNURF stimulated basal expression of LHß and increased overall GnRH stimulation. SNURF specifically stimulated LHß, with no effect on the -subunit promoter. SNURF interacts with Sp1 and SF-1, but not Egr-1, in pull-down experiments. Point mutations or deletions of SNURF functional domains demonstrated that Sp1 and SF-1 interactions with SNURF are required for SNURF stimulatory effects on the LHß promoter. Endogenous SNURF is associated with the LHß promoter on native chromatin, suggesting that it plays a physiological role in LHß gene expression. SNURF also binds the androgen receptor, and SNURF overexpression overcomes androgen suppression of GnRH-stimulated LHß but not subunit promoter activity. SNURF mutations that disrupt Sp1 or SF-1 binding eliminate rescue by SNURF. We conclude that SNURF may mediate interactions between the distal and proximal GnRH response regions of the LHß promoter to stimulate transcription and can also protect the promoter from androgen suppression. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
LH consists of two subunits, LHß and -subunit, and expression of LHß is limiting for overall LH synthesis. GnRH pulses tightly regulate expression of the LH subunit genes, and only intermediate frequency pulses stimulate LHß expression (2). Two enhancer regions of the rat LHß promoter are required for complete basal expression and GnRH stimulation, including a distal region comprised of an overlapping Sp1/CArG box and second Sp1 site, and a proximal region with two bipartite binding sites for Egr-1 and the orphan nuclear receptor steroidogenic factor 1 (SF-1) separated by a homeodomain binding site that corresponds to a Ptx1 consensus site (3, 4, 5, 6, 7, 8, 9). GnRH treatment of clonal gonadotrope cell lines or rat pituitary cells increases expression of SF-1 and Egr-1 and may increase Egr-1 transcriptional activity (5, 10, 11, 12, 13).
Expression of Egr-1 and SF-1 are critical for fertility and development as well as for expression of LH, and a complicated relationship exists between the Sp1, SF-1, and Egr-1 sites on the LHß promoter (3, 6, 14, 15). Mutations of one but not both Sp1 binding sites in the distal GnRH responsive region of the LHß promoter severely reduces GnRH stimulation of the promoter, even though the proximal region is still intact (3). In addition, a change in the number of bases between the distal and proximal regions attenuates GnRH stimulation (6). These data have led to a hypothesis that the DNA between the two responsive regions of the LHß promoter may enable either a direct interaction between the transcription factors binding to these regions, or an indirect one mediated by a coactivator (3, 6, 16).
Coactivators are generally divided into two categories based on function: proteins that direct and recruit the transcription complex, and factors, such as TATA-binding protein (TBP)-associated factors, and enzymes that modify the structure of chromatin including those with histone acetyltransferase activity (17). In considering possible coactivators that may play a role in basal and GnRH-stimulated expression of the LHß gene, we tested coactivators from these two general classes. We examined examples of chromatin remodeling coactivators with histone acetyltransferase activity, CREB binding protein (CBP) and p300, and the novel coactivator small nuclear RING finger protein (SNURF) that may promote the assembly of the transcription complex (18).
SNURF was identified in yeast two-hybrid assays using the androgen receptor (AR) DNA binding domain (DBD) and part of the hinge region as bait (18). SNURF was found to serve as a coactivator for steroid receptor-dependent and independent promoters, potentially through binding to TBP and acting as a bridge between transcription factors and the transcription complex. Overexpression of SNURF enhances not only AR-dependent transcription but also that of other steroid receptors, including the glucocorticoid, progesterone, and estrogen receptors (18, 19). SNURF also enhances transcription of promoters containing GC box elements (20, 21). The RING finger domain of SNURF is critical for its interaction with Sp1 and activation of the latter promoters (18, 21). In addition, the positively charged N-terminal amino acid residues of SNURF that enable direct interaction with DNA are linked with the ability of SNURF to enhance transcription of minimal promoters containing Sp1 binding sites (20). Because the Sp1 sites in the LHß promoter are critical for both basal and GnRH-stimulated expression, we investigated the role of SNURF in LHß expression. The ability of SNURF to bind AR was also intriguing, as androgens suppress GnRH stimulation of the rat and bovine LHß promoters (16, 22).
Steroid feedback has been implicated in the regulation of gonadotropins at the hypothalamic and pituitary levels (16, 22, 23, 24, 25, 26, 27). Androgens can regulate GnRH expression at the level of the hypothalamus (23) and suppress transcription of the LH subunit genes, LHß and -subunit, in vivo (24). This suppression occurs, at least in part, by direct androgen actions on the pituitary, but not necessarily by direct AR binding to DNA. For example, AR suppresses basal transcription of the human -subunit by AR interference with binding of activating transcription factor 1 and c-Jun to the promoter, rather than binding to the androgen response element (27, 28). Similarly, we and others have shown that AR is required for androgen suppression of GnRH stimulation of the rat and bovine LHß genes (16, 22). AR interferes with transcription factors required for full GnRH stimulation of LHß and thereby represses transcription of the LHß promoter (16, 22).
Here we investigated the role that SNURF plays in the pituitary gonadotrope LßT2 cell line and, in particular, how it activates the Sp1-driven rat LHß promoter. Our results indicate that SNURF is a coactivator for the LHß gene and interacts with transcription factors found in both GnRH-responsive regions of the LHß promoter. Through these interactions, SNURF can also prevent AR from disrupting the recruitment of the transcription complex after androgen treatment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit genes, as well as receptors for GnRH, estrogen, and androgen (29). Cells were transfected with the coactivators CBP or SNURF and the rat LHß promoter construct containing the two GnRH-responsive regions of the LHß promoter [–617 to +44 nucleotides (nt), hereafter termed –617LHß] fused to luciferase. Overexpression of SNURF increased both untreated (3.0-fold) and GnRH-treated (3.2-fold) LHß reporter activity when compared with reporter alone (Fig. 1A
). Overexpression of CBP had no significant effect on basal or GnRH-stimulated LHß reporter activity in agreement with previous studies (6), and similar experiments with p300 did not alter LHß promoter activity (data not shown). CBP was functional in these studies, however, as in parallel wells the forskolin-stimulated expression of a cAMP response element (CRE)-containing reporter was stimulated more than 5-fold (Fig. 1A).
|
-subunit promoter or –617LHß luciferase reporter construct alone or in combination with the SNURF expression plasmid. Both -subunit and LHß promoter reporters were stimulated more than 5-fold by GnRH treatment (Fig. 1B
). However, only LHß promoter activity was stimulated for both basal and GnRH responses when cotransfected with SNURF.
SNURF Is Expressed in Pituitary Tissue and LßT2 Cells
SNURF protein has been found in several tissues and cell types (18) but had not been specifically shown in pituitary or gonadotrope cells. We examined pituitary tissue from male and female mice, and several different cell lines, by immunoblotting with SNURF antiserum (Fig. 2). As controls, we used Cos-1 cells, which express no or extremely low levels of SNURF, with or without transfection with a SNURF expression vector. We easily detected SNURF in pituitaries from male and female mice and found that LßT2 gonadotrope cells expressed significant amounts of SNURF. SNURF expression was not altered by GnRH treatment. GH3 somatolactotrope cells expressed no or very low levels of SNURF. Although there was some variation in SNURF expression levels among individual pituitaries examined, there were no significant differences between male and female glands. In LßT2 cells, and individual pituitary glands with higher levels of SNURF expression, there were often fainter bands above and below the main SNURF band. This could represent splice variants or other modifications of SNURF, but these have not yet been identified. These data show that SNURF is expressed in pituitary and the LßT2 gonadotrope cells and could play a physiological role in LHß expression.
|
). These two regions must cooperate for full GnRH stimulation (3, 6). We performed transfection studies to identify the relationship between SNURF coactivation and the two GnRH response regions. We used the –617LHß reporter that contains both distal and proximal response regions, and a truncated –245LHß reporter containing only the proximal response region. To determine the importance of the distal region in the SNURF response, a construct containing both response regions but with mutations in both SF-1 sites (2SF1-Mut) was also used. Transfection of SNURF leads to stimulation of each LHß promoter construct in both untreated and GnRH-treated cells (Fig. 3B). As noted previously, GnRH stimulation of the –245LHß construct by GnRH is significantly less than stimulation of the larger construct (3). SNURF stimulation of basal (3.9-fold) and GnRH-stimulated (4-fold) activity of the –617LHß promoter containing both GnRH response regions was greater than SNURF stimulation of basal (2.1-fold) and GnRH-stimulated (2.0-fold) activity of the –245LHß promoter. SNURF stimulation of basal (2.5-fold) and GnRH-stimulated (2.3-fold) activity of the 2SF1-Mut construct was also less that that of the intact –617LHß promoter, suggesting that both Sp1 and SF-1 binding sites are required for the full response to SNURF.
|
). Two transcription factors that bind to the proximal GnRH response region, Egr-1 and SF-1, were also used in pull-down experiments. SNURF interacted with GST-Egr-1 very weakly if at all, whereas GST-SNURF showed a strong interaction with in vitro translated 35S-labeled SF-1 (Fig. 3C). Thus, SNURF interacts directly with transcription factors that bind to both proximal and distal GnRH response regions of the LHß promoter.
SNURF Interaction with Sp1 Is Important for Coactivator Activity
SNURF can activate steroid-dependent and -independent gene transcription through its interaction with steroid receptors and Sp1 (18, 19, 21). It has also been shown to bind DNA directly through several positively charged amino acid residues in its amino terminus (20). To test which of these characteristics of SNURF is most important in enhancing basal and GnRH-stimulated LHß promoter activity, we used several mutants of SNURF (Fig. 4A) in transfection assays. These included SNURF-CS1, in which the RING finger domain and Sp1 binding has been disrupted by amino acid substitutions C136S and C139S; SNURF-R8–11A/K9A, in which the positively charged amino acids of the N terminus of SNURF have been changed to alanines to eliminate DNA binding; and SNURF
31–65, in which a portion of the AR interacting domain has been deleted (18, 20, 21). All SNURF mutants were expressed at levels equivalent to levels of wild-type SNURF in transfected cells (Fig. 4D). The –617LHß and –245LHß promoter reporters were used in transfection experiments in LßT2 cells along with the SNURF mutants. SNURF-R8–11A/K9A enhanced basal and GnRH-stimulated –617LHß promoter activity to a similar extent as wild-type SNURF, whereas SNURF-CS1 did not alter –617LHß promoter activity in the absence or presence of GnRH, demonstrating that the Sp1 binding ability of SNURF is crucial under both conditions (Fig. 4B). Similarly, SNURF31–65 had no significant effect on –617LHß promoter activity (Fig. 4B). Similar results were found using the –245LHß construct (Fig. 4C).
|
31–65 Interacts with Sp1 But Not SF-1 shows that SNURF31–65 also interacts with GST-Sp1, although less efficiently than with wild-type SNURF in the same experiments. We next performed studies to identify SNURF regions important for interactions with SF-1 (Fig. 5B). SNURF and SNURF-R8–11A/K9A strongly interact with SF-1, but SNURF31–65 does not. Thus, although SNURF31–65 interacts with Sp1, it cannot bind SF-1 on the proximal promoter region, explaining its low activity on both the –617LHß and –245LHß constructs (Fig. 4, B and C). Through five separate experiments, SNURF-CS1 bound SF-1, but with somewhat diminished ability compared with wild-type SNURF. This weaker binding to SF-1 may explain, at least in part, the lack of effect of this mutant on the –245LHß reporter and the reduced activity on the –617LHß construct.
|
). In some experiments, PCR products from the linear range of amplification were also displayed on gels (Fig. 6B). Antibodies to SF-1 and SNURF precipitated DNA sequences for the LHß, but not insulin, promoter, demonstrating the specificity of the assay. SNURF and SF-1 both occupy the LHß promoter under basal conditions, as might be expected from our results showing strong stimulation of basal promoter activity in transfection assays (Figs. 1 and 3). Thus, SNURF could play a role in the physiological regulation of LHß transcription from native chromatin.
|
). The relative levels of AR and SNURF were important, as overexpression of AR resulted in lower GnRH stimulation even in the presence of SNURF (data not shown). GnRH stimulation of the rat -subunit promoter is also suppressed by androgens, but SNURF cannot rescue the promoter from androgen suppression, suggesting that simple SNURF titration of AR does not completely explain the results with the LHß promoter.
|
). Neither SNURF-CS1 nor SNURF31–65 was able to significantly stimulate promoter activity in cells treated with both GnRH and DHT (Fig. 8), suggesting that both Sp1 and SF-1 binding were required for this effect. On the basis of these data, a model for the relationships between SNURF, transcription factor binding, and AR on the LHß promoter is presented in Fig. 9.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). SNURF contains multiple functional domains, including an N-terminal region that binds DNA and steroid receptors, and a C-terminal RING finger that interacts with Sp1 and TBP. SNURF does not possess intrinsic transcription activity but does bind to TBP, thereby serving as a conduit between factors binding upstream of the promoter and the transcriptional machinery (18). Our data suggest that SNURF acts on the LHß promoter via sets of protein-protein interactions with both promoter enhancer regions.
The rat LHß promoter enhancer regions, distal and proximal, are required for basal and GnRH-regulated transcription (3, 4, 5, 6, 7, 8). The distal enhancer region (–456 to –375 bp) contains a nonconsensus Sp1 site that overlaps with a CArG box region and a region with several overlapping Sp1 consensus sites, whereas the proximal enhancer region (–121 to –50 bp) contains two bipartite binding sites for Egr-1 and SF-1 separated by a Ptx1 consensus site. Ptx-1 is a homeodomain protein and can bind to the bovine promoter (22); however, recent evidence suggests that in LßT2 cells an Otx-related protein may bind with higher affinity to this region (9). Egr-1, like Sp1, is a zinc finger protein that binds to GC-rich elements, and SF-1 is an orphan nuclear receptor that binds DNA as a monomer (14, 30, 31, 32, 33). Egr-1 and SF-1 proteins play a critical role in LHß gene expression, as animals in which either gene is disrupted do not express LH and are infertile (13, 33). The two enhancer regions act synergistically for full GnRH stimulation (3, 6). Mutation of a single Sp1 consensus site in the distal response region prevents GnRH stimulation, and altered spacing between the two regions reduces GnRH stimulation by up to 50% (3, 6). Cooperation between the two regions could occur by direct interaction between transcription factors binding to these regions or via intermediary proteins. Sp1 and SF-1 have been shown to interact directly to modulate CYP11A gene transcription, but in this case the two factors bind within 30–50 nt of each other on the promoter (33). Here, we demonstrate the coactivator SNURF can stimulate basal and thus the overall GnRH-stimulated activity of the rat LHß promoter by interacting with transcription factors that bind to the distal (Sp1) and proximal (SF-1) GnRH-responsive regions (Fig. 9). SNURF did not interact with Egr-1, even in the presence of SF-1 (Curtin, D., unpublished data). The effect of SNURF is specific, as CBP and p300 did not enhance promoter activity, and the steroid coactivator, steroid receptor coactivator 1, has also been reported to have no effect (6). Actions of SNURF are also context dependent, as both the LHß and -subunit promoter contain SF-1 binding sites, but only LHß is stimulated by SNURF.
SNURF was previously characterized by its interaction with steroid receptors, such as AR and ER, and its ability to activate steroid-dependent transcription (18, 19, 21). Here we show for the first time that SNURF can also bind to and stimulate transcription via an orphan nuclear receptor, SF-1, also called Ad4BP and officially designated NR5A1 (35). Unlike other steroid receptor coactivators, SNURF does not bind through LXXLL motifs. Instead, AR-SNURF interaction occurs via an N-terminal region on SNURF (amino acids 31–65) and the DBD and hinge region of the AR (18). Similarly, SNURF stimulation of ligand-bound estrogen receptor- (ER) transcription from estrogen response element (ERE) model promoters requires multiple regions of SNURF, including both N- and C-terminal regions, and multiple regions of the ER, including the DBD (19). Deletion of amino acids 31–65 of SNURF abolished its ability to enhance AR-dependent transcription from an androgen-response element (18, 21), and this mutation also diminished the response of ER on an estrogen response element (19). SF-1 also requires amino acids 31–65 of SNURF for binding, and the SNURF31–65 mutant could not activate the full-length or truncated LHß promoter constructs containing SF-1 binding sites. SF-1 has approximately 50–60% amino acid homology to AR and ER in the DBD region (35), which appears to be important for SNURF interactions with the nuclear receptor family. SNURF does not influence AR binding to its cognate DNA response element, but is suggested to act as a bridging factor, coordinating signals from the AR to the transcriptional machinery. Interestingly, the Drosophila melanogaster nuclear receptor, FTZ-F1, a homolog of SF-1, interacts through its DBD with a 16-kDa multiprotein bridging factor 1, suggesting that similar mechanisms may occur for similar proteins in other species (36).
SNURF also interacts with Sp1 via its RING finger and can stimulate steroid-independent transcription through this mechanism (21). SNURF appears to enhance protein binding to GC boxes, and it may augment transcription by recruiting Sp1 either directly or through additional protein partners (21). Interestingly, in contrast to previous results on a minimal GC box-containing promoter (19), mutation of positively charged N-terminal amino acids that bind DNA did not influence SNURF’s ability to act on the more complex LHß promoter and emphasizes the importance of protein-protein interactions between SNURF and the transcription factors binding to this promoter.
SNURF enhancement of LHß promoter activity has several unique features from previously described systems. First, SNURF stimulation of the LHß promoter is more complex, requiring direct interaction with two different types of transcription factors, the nuclear receptor SF-1 and Sp1. Moreover, a major effect of SNURF on LHß is on basal transcription rather than just on promoter activation, as observed with steroid-regulated model promoters (18, 21). In this respect, it is important to note that both SF-1 and Sp1 are present in and could bind to the gene under basal or unstimulated conditions as well as after stimulation with GnRH. SNURF is basally associated with the LHß promoter on chromatin in LßT2 cells, suggesting that it could play a physiological role in LHß promoter activity. Finally, SNURF enhancement of basal promoter activity results in overall enhancement of activity in the presence of GnRH. GnRH stimulation of the LHß promoter may involve phosphorylation or other posttranscriptional modifications of existing transcription factors, such as Sp1 and Egr-1, and stimulation of Egr-1 transcription (12, 13). Interestingly, SNURF can also be phosphorylated, and it remains to be determined whether the modification can affect its function (18). Clearly, SNURF can stimulate basal and consequently the overall GnRH-stimulated LHß promoter activity, and this effect is greatest on the –617LHß construct that contains both GnRH-responsive regions and thus binds both Sp1 and SF-1.
We cannot definitively deduce from these studies whether more than one SNURF molecule operates with the LHß promoter at the same time, or whether one SNURF molecule binds to both Sp1 and SF-1 on the promoter simultaneously. Because SNURF can bind to Sp1 via its RING finger domain and to SF-1 via its N-terminal region, simultaneous binding is possible. Only SNURF forms that bind both Sp1 and SF-1 have full stimulatory activity on the LHß promoter. SNURF and the DNA-binding mutant SNURF-R8–11A/K9A, both of which bind Sp1 and SF-1, stimulate basal and GnRH-stimulated transcription of the LHß promoter. In contrast, even though the SF-1 binding mutant SNURF31–65 can interact with Sp1, it is unable to enhance basal or GnRH-stimulated transcription (Fig. 4B). Similarly, the Sp1 binding mutant, SNURF-CS1, which still retains some SF-1 binding ability, does not enhance activity of the –617LHß promoter (Fig. 4B). This mutant also lacks TBP binding ability. The failure of CS1 to activate the –617 and –245LHß promoter constructs may be due, in part, to the loss of this property, as the ability to act as a bridging factor to TBP appears to be critical for SNURF action on AR and ER and is likely to be important for SF-1 (18, 19, 20, 21). Looping of chromatin enabling two distantly spaced promoter regions to interact has been postulated for interactions between AR and Sp1 on the cyclindependent kinase inhibitor p21 gene (37) and has been demonstrated in chromatin for Pit-1 and ER binding enhancer regions of the rat PRL gene after estrogen treatment (38, 39, 40). Other coactivator proteins that can bind multiple proteins and have been shown to possess multiple different functions include E6-AP, an E3 ubiquitin protein ligase (41), high mobility group-1/2, chromatin high mobility group proteins (42), Zac1, a zinc finger protein (43), and SRA, a steroid receptor coactivator that is an RNA molecule (44). Thus, SNURF is one of several coactivators that may integrate several transcription functions, with potentially important tissue and gene-specific roles.
Given that SNURF was originally identified through its interaction with AR, and androgens suppress GnRH stimulation of LHß, we tested the effects of SNURF on this suppression. Previously, we demonstrated that AR represses the promoter, not via direct binding to DNA but by interference with transcription factor binding to the promoter (16) and that holo-AR inhibits binding of Sp1 to the LHß distal enhancer region. Other investigators have found that AR prevents SF-1 binding to the bovine promoter, and that the degree of this interaction is influenced by Egr-1 and Pitx1 (23). We show in this work that overexpression of SNURF rescues LHß promoter activity from androgen suppression. SNURF interacting with Sp1 as well as with SF-1 may protect these factors bound to the promoter from AR, or SNURF may alternatively bind to AR and sequester it (Fig. 9). Two findings argue against this as the primary mechanism of action. First, the SNURF-CS1 mutant can still bind AR, but did not restore GnRH stimulation of the promoter in the presence of DHT. Second, the -subunit promoter is suppressed by DHT, and titration of AR by SNURF should then prevent or rescue suppression by DHT. These data suggest that SNURF interaction with Sp1 and SF-1 on DNA, either by steric hindrance or changes in protein/chromatin conformation, blocks the access of AR to the transcription factors. The mechanism that is most important in the cell may depend on relative amounts of each factor or relative affinity of SNURF for AR, SF-1, and Sp1 bound to chromatin.
Overall, our results indicate that SNURF can function as a bridging factor on a complex physiological promoter. Enhancement of LHß promoter activity requires coordinate interaction of SNURF with two separate transcription factors, Sp1 and the nuclear orphan receptor SF-1, which bind to different specific SNURF domains. These transcription factors bind the promoter on enhancer regions positioned over 250 nt apart, suggesting a novel mechanism by which SNURF may act as an integrator for enhancer cooperation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit as well as GnRH, estrogen, and androgen receptors (29), and was originally obtained from Dr. Pamela Mellon (University of California San Diego). Cells were treated in phenol red-free DMEM with charcoal-stripped 5% newborn calf serum and antibiotics as described (16). Cells were plated in 12-well plates at a density of 5 x 105 cells per well 16–20 h before CaPO4 transfection. Each well was transfected with 1.5 µg of reporter vector as indicated for 16 h, washed, and treated with 10 nM GnRH for 6 h before collection and measurement of luciferase activity. Steroids or forskolin (1 µM; Sigma Chemical Co., St. Louis, MO) were included as indicated during the transfection period and the GnRH treatment period, for a total treatment of 24 h. In several experiments, a cytomegalovirus (CMV)-ß-galactosidase vector (0.3 µg/well) was included to normalize for luciferase activity, and in all cases luciferase activity was also normalized for protein. For studies with mutant LHß promoter constructs, normalized luciferase activity was expressed relative to the full-length –617LHß construct, which was set at 1.0. After transfection, cells were washed with PBS and lysates were collected in 150 µl lysis buffer (Promega Corp., Madison, WI), vortexed, centrifuged for 1 min, and assayed in a Turner 20e luminometer (Turner Designs, Mountain View, CA). Protein concentrations were determined by the colorimetric assay from Bio-Rad Laboratories (Hercules, CA). Hormones were obtained from Sigma Chemical Co.
LHß-luciferase reporter vectors have been described elsewhere (3). For most studies, the construct –617LHß containing the promoter region from –617 to +44 nt relative to the rat LHß gene transcriptional start site, and both GnRH-responsive elements, was used. To test the relative importance of the distal region in the SNURF response, a mutant construct containing both GnRH-response regions, but with mutations in both SF-1 sites in the proximal response regions (2SF1-Mut) was used. In some experiments the deletion construct –245LHß corresponding to nt –45 to +44 and containing only the proximal GnRH response element was also tested. The -subunit promoter luciferase reporter vector (–411) contains –411 to +77 nt of the -subunit promoter (3, 45). The pcDNA3.1(+)-FLAG-SNURF constructs encoding the following SNURF forms were used: wild-type SNURF, SNURF-CS1 in which the cysteines 136 and 139 are substituted with serines creating a mutant unable to bind Sp1 (21), SNURF-R8–11A/K9A in which Arg8, Lys9, Arg10, and Arg11 were changed to alanines yielding a DNA binding-deficient mutant (20), and SNURF31–65 in which amino acid residues 31–65 have been deleted, creating a mutant that cannot activate AR-dependent transcription (21). CMV-hAR (46) and pcDNA-CBP (47) expression vectors have been described. For studies using wild-type or mutant SNURF, total DNA in each transfection reaction was balanced with empty vector. A SNURF dose-response curve was performed to test maximally effective concentrations and, unless otherwise indicated, SNURF constructs were added at 1.5 µg/500,000 cells. To test CBP function, we used a CRE-luciferase reporter construct containing six tandem CRE elements fused to the TATA box of the rat PRL gene promoter, kindly provided by Dr. Richard N. Day (University of Virginia, Charlottesville, VA).
Immunoblotting
Expression of transfected wild-type and mutant SNURF-FLAG constructs was evaluated in LßT2 cells 24 h posttransfection by immunoblotting. Approximately 50 µg of transfected cell lysate were examined on 12% SDS-PAGE, followed by transfer to a nylon membrane, incubation with M5 FLAG primary antibody (1:2000; Sigma Chemical Co.), horseradish peroxidase-labeled goat antimouse IgG secondary antibody (1:10,000; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) followed by detection with PicoWest (Pierce Chemical Co., Rockford, IL). Expression of native SNURF protein was examined in pituitary tissue from mature male and female mice and in clonal cell lines. GH3 cells were cultured as described (48). Cos-1 cells were transfected as described (49) with 1.5 µg of SNURF expression vector and collected after 24 h. LßT2 cells were treated with and without 100 nM GnRH for 2 h, which stimulates Egr-1 expression. Tissue or cell lysates (25 µg) were examined on 10% SDS-PAGE and transferred to a nylon membrane. Polyclonal anti-SNURF antibody was generated in rabbits against full-length SNURF protein produced in insect cells (18). SNURF was detected with this antiserum at a dilution of 1:20,000, followed by incubation with horseradish peroxidase-labeled donkey antirabbit IgG secondary antibody (1:20,000, Amersham Biosciences, Buckinghamshire, UK) followed by detection with PicoWest. For both types of studies, ß-actin was used as a loading control, as previously described (48). Duplicate samples were tested from three to four independent experiments, and samples from the same experiment were run simultaneously on parallel gels in the same apparatus.
Statistical Analysis
Data for normalized luciferase activity are presented as the mean ± SEM for six wells per group, compared with untreated controls, and each experiment was performed between three and six times with equivalent results. Statistical significance was evaluated by two-way ANOVA comparing control and GnRH treatment groups, and wild-type and mutant SNURF or CBP expression, and analyzed post hoc by Tukey’s wholly significant difference test. Statistical significance was taken to be at least P < 0.05, or as shown in the figures.
GST Pull-Down Experiments
BL21 bacterial cells were transformed with constructs expressing GST, GST-Sp1, GST-Egr-1, GST-SF-1, or GST-SNURF. Luria Broth (100 ml) containing 50 µg/ml ampicillin was inoculated with 1 ml of bacteria and incubated in an orbital shaker at 37 C. Bacteria were grown to A600 = 0.5, induced with 0.1 mM isopropyl ß-thiogalactopyranoside, and shaken overnight at room temperature. The bacterial pellet was resuspended in 5 ml of buffer containing 50 mM Tris (pH 7.5), 0.5 mM EDTA, 300 mM NaCl, 10 mg/ml lysozyme, and 1 mM dithiothreitol. One hundred microliters of 10% Nonidet P-40 were added, and after 10 min, the lysate was frozen at –70 C in an ethanol bath. Lysate was thawed at room temperature and then incubated for 1 h in 5 ml of buffer containing 1.5 M NaCl, 12 mM MgCl2, 5 µg DNase I, 10 µg/ml leupeptin, 1 µg/ml pepstatin A, and 0.1 mM phenylmethylsulfonylfluoride. Lysates were passed through a 20-gauge needle and centrifuged for 30 min at 7500 x g. Supernatants were incubated with glutathione beads (Sigma) overnight at 4 C. Beads were extensively washed with PBS, and bound protein concentrations were analyzed by electrophoresis on 12% SDS-PAGE followed by Coomassie Blue staining and immunoblotting with appropriate antibodies [Sp1 and Egr-1 antibody from Santa Cruz Biotechnology, Inc., Santa Cruz, CA; SF-1 antibody from Upstate Biotechnology, Lake Placid, NY; and anti-SNURF antiserum (17)]. For pull-down experiments, approximately 1 µg of GST fusion protein was used in each sample incubation. BSA (20 µg/ml) was added to each incubation containing [35S]methionine-labeled (0.04 mCi/50-µl reaction) in vitro translated proteins (TNT Rabbit Reticulocyte Transcription/Translation Kit; Promega). Labeled proteins included SNURF, SNURF-CS1, SNURF R8–11A/K9A, SNURF 31–65, Egr-1, SF-1, and Sp1. For SNURF studies, approximately equal molar quantities of SNURF or SNURF mutant proteins were used in reactions. Total volume was adjusted to 150 µl with GST wash buffer (10 mM MgCl2; 150 mM KCl; 20 mM HEPES, pH 7.6; 10% glycerol; and 0.12% Nonidet P-40). Beads and proteins were incubated for 1.5 h at 4 C including GST alone with labeled proteins for control experiments, and then centrifuged and washed four times in GST wash buffer. Beads were resuspended in 10 µl of sodium dodecyl sulfate (SDS) loading buffer and boiled for 5 min. Proteins were electrophoresed on SDS containing 10% acrylamide gels at 150 V, along with standard molecular weight markers (Benchmark, Life Technologies, Inc., Gaithersburg, MD). Gels containing [35S]methionine-labeled proteins were dried and exposed to film for 24–72 h at –70 C. Films were scanned and subjected to densitometric analysis. Each experiment was performed a minimum of five to six times.
Chromatin Immunoprecipitation (ChIP) Assays
ChIP assays were performed essentially as described by Chakarabarti et al. (50). LßT2 proteins were cross-linked to genomic DNA by addition of formaldehyde to a final concentration of 1%, in PBS. Chromatin was sonicated to an average length of 1 kb in 1% Triton X-100, 0.1% deoxycholate, 50 mM Tris, pH 8.1, 150 mM NaCl, 5 mM EDTA. After the lysates was clarified, 1 µg of control CMV-luciferase plasmid and 100 µg of BSA were added. Chromatin lysates were then diluted with sonication buffer plus protease inhibitors (0.2 mg/ml leupeptin and aprotinin, 20 mM phenylmethylsulfonylfluoride), divided into aliquots, and incubated at 4 C overnight with no antibody, or antibodies to SNURF (5 µl) or SF-1 (5 µl), and then precipitated with protein A beads (Santa Cruz Biotechnology, Inc.). Protein DNA complexes were washed three times in buffer without protease inhibitors and once in TE, and cross-linking was reversed by addition of NaCl to 100 mM and incubation at 65 C. Proteins were eluted from the beads by incubation with elution buffer (1% SDS, 0.1 M NaHCO3, 0.01 mg/ml herring sperm DNA) and were digested with Proteinase K. DNA was extracted with phenol-chloroform, and mouse LHß promoter sequences were detected with primers that flank the sequence between –102 bp (5'-CTGTGTCTCGCCCCCAAAGAGATTA-3') and –1 bp (5'-CCTGGCTTTATACCTGCGGGGTT-3'). Sequences were detected by real-time PCR (iCycler, Bio-Rad Laboratories, Hercules, CA), incorporating SYBR green, as previously described (50). Insulin gene promoter primers were used as a specificity control, and luciferase gene primers were used as an internal control for sample handling (50). For some studies, PCR was performed for 28 cycles, within the linear range of the curve, and the products were subjected to electrophoresis. For PCR quantification, previously described protocols were followed (50), using a total number of 40 cycles, with continuous SYBR green monitoring. Briefly, relative proportions of immunoprecipitated promoter fragments were determined based on the threshold cycle (Tc) value for each PCR reaction, which is determined automatically and is the cycle at which fluorescence rises 10 times above the mean SD of background in all reaction wells. The Tc is calculated by correction of immunoprecipitated sample fluorescence for luciferase plasmid and for signal with control samples (no antibody). All samples were normalized for input chromatin. Values are calculated for each treatment, and fold changes were then calculated by raising 2 to the Tc power. Calculations are summarized as the fold-difference (SNURF ChIP vs. Control ChIP) = 2[Tc(control)-Tc(SNURF)], where Tc = Tc(immunoprecipitated sample)Tc(input). Each promoter fragment was analyzed in duplicate on at least two separate occasions and from three independent immunoprecipitations.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: AR, Androgen receptor; CBP, CREB-binding protein; ChIP, chromatin immunoprecipitation; CMV, cytomegalovirus; CRE, cAMP response element; DBD, DNA-binding domain; DHT, dihydrotestosterone; ER, estrogen receptor; GST, glutathione-S-transferase; nt, nucleotide; SDS, sodium dodecyl sulfate; SF-1, steroidogenic factor 1; SNURF, small nuclear RING finger; TBP, TATA-binding protein; Tc, threshold cycle.
Received for publication June 7, 2003. Accepted for publication February 17, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
in ZR-75 human breast cancer cells by SNURF and TATA-binding protein. J Biol Chem 277:2485–2497 glycoprotein hormone subunit gene through protein-protein interactions with cJun and activation transcription factor 2. Mol Endocrinol 15:1496–1504-subunit gene by androgen receptor occurs independently of DNA binding but requires the DNA-binding and ligand-binding domains of the receptor. Mol Endocrinol 11:1497–1506 protein levels and estrogen responsiveness. Endocrinology 141:2174–2184NURSA Molecule Pages Link:
This article has been cited by other articles:
 |
|
 |
|
  H. E. Walsh and M. A. Shupnik Proteasome Regulation of Dynamic Transcription Factor Occupancy on the GnRH-Stimulated Luteinizing Hormone {beta}-Subunit Promoter Mol. Endocrinol., February 1, 2009; 23(2): 237 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-F. Mouillet, X. Yan, Q. Ou, L. Jin, L. J. Muglia, P. A. Crawford, and Y. Sadovsky DEAD-Box Protein-103 (DP103, Ddx20) Is Essential for Early Embryonic Development and Modulates Ovarian Morphology and Function Endocrinology, May 1, 2008; 149(5): 2168 - 2175. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Hong, H. Sun, C. H. Jin, M. Chapman, J. Hu, W. Chang, K. Burnett, J. Rosen, A. Negro-Vilar, and J. N. Miner Cell-Specific Activation of the Human Skeletal {alpha}-Actin by Androgens Endocrinology, March 1, 2008; 149(3): 1103 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kowase, H. E. Walsh, D. S. Darling, and M. A. Shupnik Estrogen Enhances Gonadotropin-Releasing Hormone-Stimulated Transcription of the Luteinizing Hormone Subunit Promoters via Altered Expression of Stimulatory and Suppressive Transcription Factors Endocrinology, December 1, 2007; 148(12): 6083 - 6091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A Ferris, H. E Walsh, J. Stevens, P. C Fallest, and M. A Shupnik Luteinizing Hormone Beta Promoter Stimulation by Adenylyl Cyclase and Cooperation with Gonadotropin-Releasing Hormone 1 in Transgenic Mice and LBetaT2 Cells Biol Reprod, December 1, 2007; 77(6): 1073 - 1080. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. Waring and J. L. Turgeon Estradiol Inhibition of Voltage-Activated and Gonadotropin-Releasing Hormone-Induced Currents in Mouse Gonadotrophs Endocrinology, December 1, 2006; 147(12): 5798 - 5805. [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]
|
|
|
|
 |
|
 M. Luo, M. Koh, J. Feng, Q. Wu, and P. Melamed Cross Talk in Hormonally Regulated Gene Transcription through Induction of Estrogen Receptor Ubiquitylation Mol. Cell. Biol., August 15, 2005; 25(16): 7386 - 7398. [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 |
