Cyclic Adenosine 3',5'-Monophosphate (cAMP) Enhances cAMP-Responsive Element Binding (CREB) Protein Phosphorylation and Phospho-CREB Interac
http://www.100md.com
内分泌学杂志 2005年第3期
Department of Biochemistry and Molecular Biology and The Center for Genetics and Molecular Medicine, University of Louisville School of Medicine, Louisville, Kentucky 40292
Address all correspondence and requests for reprints to: Dr. Barbara J. Clark, Department of Biochemistry and Molecular Biology, University of Louisville School of Medicine, Louisville, Kentucky 40292. E-mail: bjclark@louisville.edu.
Abstract
Steroidogenic acute regulatory protein (StAR) transcription is regulated through cAMP-protein kinase A-dependent mechanisms that involve multiple transcription factors including the cAMP-responsive element binding protein (CREB) family members. Classically, binding of phosphorylated CREB to cis-acting cAMP-responsive elements (5'-TGACGTCA-3') within target gene promoters leads to recruitment of the coactivator CREB binding protein (CBP). Herein we examined the extent of CREB family member phosphorylation on protein-DNA interactions and CBP recruitment with the StAR promoter. Immunoblot analysis revealed that CREB, cAMP-responsive element modulator (CREM), and activating transcription factor (ATF)-1 are expressed in MA-10 mouse Leydig tumor cells, yet only CREB and ATF-1 are phosphorylated. (Bu)2cAMP treatment of MA-10 cells increased CREB phosphorylation approximately 2.3-fold within 30 min but did not change total nuclear CREB expression levels. Using DNA-affinity chromatography, we now show that CREB and ATF-1, but not CREM, interact with the StAR promoter, and this interaction is dependent on the activator protein-1 (AP-1) cis-acting element within the cAMP-responsive region. In addition, (Bu)2cAMP-treatment increased phosphorylated CREB (P-CREB) association with the StAR promoter but did not influence total CREB interaction. In vivo chromatin immunoprecipitation assays demonstrated CREB binding to the StAR proximal promoter is independent of (Bu)2cAMP-treatment, confirming our in vitro analysis. However, (Bu)2cAMP-treatment increased P-CREB and CBP interaction with the StAR promoter, demonstrating for the first time the physical role of P-CREB:DNA interactions in CBP recruitment to the StAR proximal promoter.
Introduction
STEROIDOGENESIS IN THE adrenals and gonads is regulated by gonadotropin trophic hormone release from the anterior pituitary and is controlled in two temporal phases, beginning within minutes for the acute phase and continuing longer term for the chronic phase. The chronic phase is mediated by an increase in gene expression for the steroid hydroxylases enzymes involved in metabolizing cholesterol into tissue-specific steroid hormones (1), whereas the acute phase of steroidogenesis is characterized by the rapid uptake of cholesterol into the mitochondria (2, 3, 4). The steroidogenic acute regulatory protein (StAR) regulates the acute phase of steroid biosynthesis by mediating the cholesterol transfer from the outer- to the inner-mitochondrial membrane, in which it is subsequently metabolized into pregnenolone by the cytochrome P450 side-chain cleavage enzyme (5, 6). This action of cholesterol transfer has been shown to be the rate-limiting step in acute steroid biosynthesis thereby providing a mechanism by which tissues can control steroid hormone output by regulating StAR protein levels and thus function (7).
Previous studies have shown that StAR is regulated at the transcriptional level via a cAMP-protein kinase A (PKA) signaling cascade within the gonads (8, 9, 10, 11, 12, 13, 14). Transactivation of target genes by the cAMP-PKA pathway involves the binding of cAMP-responsive element binding protein (CREB) family members to consensus cAMP-response elements (CRE; 5'-TGACGTCA-3') within the gene promoter. Within the classical cAMP-PKA biochemical pathway, activated PKA phosphorylates CREB on Ser133, which increases association with the CREB binding protein (CBP) coactivator, resulting in histone modification and increased transcription (15, 16). Although the StAR promoter lacks a canonical CRE, we previously localized the cAMP-responsive region of the mouse StAR promoter region between –105 and –60 bp upstream of the transcriptional start site and identified functional elements for steroidogenic factor-1 (SF-1) (–95), CCAAT/enhancer binding protein-?/nonconsensus activating protein-1/nuclear receptor half-site [(CAN) region –79 bp] and GATA-4 (–68 bp) (17). Mutation of the activator protein-1 (AP-1)-like element in the CAN region resulted in decreased basal promoter activity and abolished protein DNA-interactions within this region as determined by reporter gene assay and EMSA, respectively, whereas SF-1 binding was shown to stabilize protein interactions with the CAN region (17). Recently CREB family members were shown to directly associate with the StAR promoter via this AP-1-like element within the CAN region, and CREB overexpression was shown to increase the cAMP-dependent transactivation of StAR promoter (18). Furthermore, overexpression of both CREB and SF-1 resulted in a synergistic activation after cAMP treatment, suggesting a functional cooperation between these two factors (19). More importantly, the CREB-dependent transcriptional response was shown to be dependent on CREB phosphorylation because overexpression of phosphorylation mutants led to a loss in transactivation capability as well as functional interactions with SF-1 (18, 19).
These studies suggest that a component of the cAMP-mediated StAR transactivation involves the classical cAMP-PKA signaling pathway of CREB phosphorylation with possible subsequent recruitment of a coactivator. Indeed, recent reports of increased histone acetylation, a marker for coactivator activity, on the StAR gene promoter as well as CBP association within the proximal region of the mouse StAR promoter in MA-10 cells supports a classical cAMP-PKA signaling pathway for StAR transcriptional activation (20, 21). However, there are contradictory reports within the literature on the effects of cAMP on CREB-DNA associations with various cAMP-responsive promoters. Investigators examining the liver-specific tyrosine aminotransferase or the corticotropin-releasing factor gene promoters using genomic footprinting or in vivo cross-linking, respectively, concluded that CREB-DNA interactions with specific CREs were increased by cAMP treatment (22, 23). In contrast, studies on other gene promoters, such as the T-cell receptor, phosphoenolpyruvate carboxykinase, proliferating cell nuclear antigen, cyclin A, and c-fos showed no influence on CRE occupation with cAMP treatment in an in vivo chromatin background (24, 25, 26, 27). We previously reported a new protein-DNA complex is present in EMSA analysis using nuclear extract from (Bu)2cAMP-treated MA-10 cells, compared with control, suggesting recruitment of a cAMP-responsive factor to the StAR promoter (17).
The purpose of the present study was to examine the role of cAMP treatment on endogenous CREB family member phosphorylation and the role of phospho-CREB in subsequent CREB-DNA interactions as well as the in situ effects of CREB, phospho-CREB, and CBP association with the StAR promoter. We now demonstrate that (Bu)2cAMP stimulation of MA-10 mouse Leydig tumor cells increased CREB phosphorylation 2.3-fold within 30 min. However, the total CREB-DNA interaction was not influenced by cAMP treatment but rather that phospho-CREB association and CBP recruitment to the StAR proximal promoter were increased. Therefore, we conclude that even in the absence of a canonical CRE, the classical cAMP-PKA signaling cascade activates phospho-CREB and CBP binding to the CAN region and up-regulates transcription of the mouse StAR gene.
Materials and Methods
Materials
Waymouth’s MB/752, N6,2'-O-(Bu)2cAMP, and streptavidin-conjugated agarose solution were purchased from Sigma (St. Louis, MO). Gentamicin reagent and Dulbecco’s PBS were purchased from Life Technologies, Inc. (Rockville, MD). The protein assay kit was supplied by Bio-Rad Life Sciences (Hercules, CA). Biotin sense and antisense oligonucleotides for DNA-affinity chromatography and PCR primers for chromatin immunoprecipitation (ChIP) analysis were purchased from Midland Certified Reagent Co. (Midland, TX). T4-polynucleotide kinase was purchased from Promega Corp. (Madison, WI). [-32P]ATP was purchased from NEN Life Science Products (Boston, MA). Anti-CREB (sc-186), anti-CREM-1 (sc-440) and anti-c-Jun (sc-44) antibodies were supplied by Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Anti-phospho-CREB (06–519), anti-CBP (06–294), and the ChIP assay kit were purchased from Upstate USA, Inc. (Charlottesville, VA).
Cell culture
The MA-10 mouse Leydig tumor cell line was a gift from Dr. M. Ascoli (Department of Pharmacology, University of Iowa, Iowa City, IA). MA-10 cells were cultured in Waymouth’s MB/752 media supplemented with 15% heat-inactivated horse serum and 40 μg gentamicin sulfate per milliliter. Treatment of MA-10 cells with (Bu)2cAMP was performed in serum-depleted Waymouth’s MB/752 for the indicated time points.
Whole-cell and nuclear extract preparations
For whole-cell lysates, MA-10 cells were plated at 1 x 106 in 60-mm dishes. Forty-eight hours after seeding, cells were serum starved for 24 h before treatment in the presence or absence of (Bu)2cAMP for 0, 30, 60, 120, or 240 min in serum-free Waymouth’s MB/752 media. After treatment, cells were washed once with PBS(+) and collected by scraping in 1.5 ml PBS(+). Cells were pelleted by centrifugation at 10,000 rpm for 5 min, and resuspended in 300 μl Tris-sucrose-EDTA [10 mM Tris-HCl (pH 7.4), 250 mM sucrose, and 1 mM EDTA] plus protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml leupeptin) and phosphatase inhibitors [100 nM okadaic acid plus phosphatase inhibitor cocktail II (Sigma)]. Cells were lysed by repeated freeze-thaw (three times) followed by passage through an 18.5-gauge needle (15 times). After centrifugation at 2000 rpm for 10 min, supernatant was removed and protein concentration was determined by Bradford protein assay. Nuclear extract preparations from untreated or treated MA-10 cells for 0, 30, 60, 120, or 240 min with (Bu)2cAMP were performed as previously described (17). As with the whole-cell lysates, a protein assay was performed to determine protein concentrations, and all samples were stored at –80 C until further analysis.
DNA-affinity chromatography
Oligonucleotide annealing and purification.
A biotinylated sense strand (5'-biotin-TGCACAATGACTGATGACTTTTT-3') corresponding to the 23-bp CAN region of the StAR promoter was annealed to its respective antisense strand (5'-AAAAAGTCATCAGTCATTGTGCA-3'). Alternatively, sense and antisense strands with mutations generated within the AP-1 site [(sense: 5'-biotin-TGCACAATAGATCTTGAC TTTTT-3'), (antisense: 5'-AAAAAGTCAAGATCTATTGTGCA-3')] were used in the annealing reactions. Equal molar amounts of sense and antisense strands were mixed in annealing buffer [10 mM Tris-HCl (pH 7.5), 50 mM NaCl, and 1 mM EDTA] and boiled for 10 min with subsequent cooling to room temperature. To examine double-stranded annealing, 2.5 pmol of annealing reaction was 5' end-labeled with [-32P]ATP (NEN Life Science Products) using T4 polynucleotide kinase. Excess nucleotide was removed from the labeling reactions by a nucleotide removal kit (Qiagen, Santa Clarita, CA). The labeling reaction (50 fmol) was mixed with 5 μl EMSA dye [25 mM Tris-HCl (pH 8.0), 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, 20% glycerol, and 2.5% bromophenol blue/xylene canol] and separated by 15% nondenaturing PAGE. The gel was dried and exposed to phosphorscreen, and oligonucleotide detection was done using the Storm PhosphorImager (Molecular Dynamics, Amersham Biosciences, Sunnyvale, CA). Because 100% oligonucleotide annealing was never observed, a double-stranded oligonucleotide purification procedure was performed. Briefly, 50 fmol of radiolabeled oligonucleotide sample were added to their respective annealing reactions. Entire annealing reaction was separated by 15% nondenaturing PAGE, and the gel was exposed to x-ray film. The autoradiograph was placed back onto the gel, and the bands corresponding to the double-stranded oligonucleotides were excised. Gel slices were incubated with 0.3 M NaOAc for 16 h at 37 C with rotation. Acrylamide was pelleted by brief centrifugation, and supernatant was removed and EtOH precipitated. After purification, sample aliquots were again separated on a 15% nondenaturing PAGE to ensure the presence of only double-stranded oligonucleotides. DNA concentration was determined using the Ultraspec 3000 (Pharmacia Biotech, Uppsala, Sweden).
Column generation and chromatography procedure.
Biotin-wild-type CAN (1 nmol) or biotin-mutant CAN double-stranded oligonucleotide was incubated with 1 ml of a 50% streptavidin-conjugated agarose solution that had been equilibrated in the 1x Promega binding buffer [25 mM Tris-HCl (pH 8.0), 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, and 10% glycerol]. The DNA-avidin mixture was rotated at 4 C for 1 h and added to a 1-ml syringe with glass wool stopper (approximately 500 μl column matrix). Nuclear extract (1 mg) from MA-10 cells treated in the presence or absence of 1 mM (Bu)2cAMP for 2 h was incubated with the column matrix for 2 h at 4 C. Column fractions were collected in 500-μl volumes. The void volume [termed flow thru (FT)] was collected followed by three 0.1 M KCl binding buffer washes to eliminate nonbinding nuclear extract proteins. Subsequently, three 0.2 M KCl washes were performed to elute specific DNA-binding proteins (bound proteins) from the affinity columns. Fractions were stored at –80 C until further analysis. The salt concentrations for the wash and elution were determined based on preliminary experiments using step KCl gradients followed by monitoring with Coomassie-stain gel, Western blot, and EMSA for DNA binding of the fractions.
Batch-binding DNA-affinity chromatography.
For repeat experiments of the column DNA-affinity chromatography, a batch binding method was used. Briefly, 100 μg of MA-10 nuclear extract were incubated with 100 pmol of double-stranded biotinylated CAN oligonucleotides on ice for 45 min. Subsequently, 100 μl of the 50% streptavidin-linked agarose bead slurry (50 μl bead pellet) preequilibrated in 0.1 M KCl Promega binding buffer was added and rotated at 4 C for 1 h. Streptavidin beads were pelleted by brief centrifugation, and the void volume [termed FT] was collected. The streptavidin beads were then washed three times with 0.1 M KCl binding buffer, and DNA-associated proteins were eluted by adding 50 μl of 2x sodium dodecyl sulfate (SDS)-Laemmli buffer and boiling at 95 C for 10 min. Samples were then immediately analyzed by the immunoblot procedure.
Immunoblot analysis
Fifty micrograms of whole-cell lysates, 10 μg nuclear extract, or indicated volumes of the nuclear extract (NE), column FT, 0.1 M KCl, and 0.2 M KCl fractions from the DNA-affinity chromatography procedures were analyzed. The protein samples were separated by 10% SDS-PAGE, transferred to polyvinyl difluoride membrane, and then the membranes were blocked in PBS containing 4% nonfat dry milk and 4% Tween 20 (PBS-Tween) overnight. The membrane was incubated with primary antibody (anti-phospho-CREB, anti-CREB, or anti-CREM1 rabbit polyclonal antibodies as indicated) in PBS-Tween with 2% nonfat dry milk and immunoblot analysis performed using a horseradish peroxidase-conjugated donkey antirabbit secondary antibody. Specific immunoreactive bands were detected using the Western lightning chemiluminescent kit (Amersham, Piscataway, NJ). After immunoblot analysis with the anti-phospho-CREB antibody, the membranes were stripped using the Restore Western blot stripping buffer (Pierce, Rockford, IL), reblocked with 4% nonfat dry milk/PBS Tween, and reprobed with anti-CREB antibody for total CREB protein detection. The Un-Scan-It software (version 5.1, Silk Scientific Corp., Orem, UT) was used to quantitate the integrated ODs (IODs) for phospho-CREB, phospho-activating transcription factor (ATF)-1, CREB, and ATF-1 immunoreactive bands. The IOD values were determined from multiple exposure times to obtain values within the linearity of response for each experiment. The ratios of phosphorylated CREB (P-CREB) to CREB (total CREB) and phosphorylated ATF-1 (P-ATF-1) to ATF-1 (total ATF-1) were determined for NE and bound-protein (BND = 0.2 M KCl fraction or SDS-Laemmli buffer-eluted protein) fractions from the DNA-affinity columns. The data (ratios) were analyzed using ANOVA analysis followed by Dunnett’s posttest or the unpaired Student’s t test (GraphPad Prism, GraphPad Software, San Diego, CA). P < 0.05 was considered a statistically significant difference.
ChIP assay
ChIP assays were performed with modifications according to the ChIP assay kit purchased from Upstate Biotechnology. Briefly, MA-10 cells were plated at 2.5 x 106 cells in 60-mm dishes. Thirty-six hours after seeding, cells were treated in the absence or presence of 1 mM (Bu)2cAMP for 30 or 120 min. After treatment, 1% formaldehyde was added and the cells incubated for 10 min at 37 C. Cells were washed two times with ice-cold PBS(+) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml pepstatin A). Cells were collected, pelleted, and resuspended in provided SDS-lysis buffer. Sonication was performed using a 550 Dismembrator sonicator (Fisher Scientific, Pittsburgh, PA) for 20 cycles of 10-sec pulses, and chromatin fragments were determined to be less than 500 bp in size. Sheared chromatin was immunoprecipitated with IgG, anti-CREB, anti-c-Jun, anti-P-CREB, or anti-CBP overnight with rotation at 4 C. Immunocomplexes were collected using salmon sperm/protein A-agarose slurry and subsequently washed one time each with provided low salt, high salt, LiCl, and Tris/EDTA buffers following the manufacturer’s instructions. Protein-DNA complexes were eluted from protein A-agarose beads by addition of elution buffer (1% SDS, 0.1 M NaHCO3) and rotation at room temperature. Formaldehyde cross-links were reversed by the addition of 200 mM NaCl and heating at 65 C for 4 h. DNA was purified using proteinase K treatment followed by phenol extraction and EtOH precipitation. PCR was done using 0.5 μl input chromatin sample and 3 μl immunoprecipitated DNA sample with primer pairs specific for the proximal (–299 bp/–41 bp) [–299 forward: 5'-TGATGCACCTCAGTTACTGG-3'; –41 reverse: 5'-GCTGTGCATCATCACTTGAG] or distal (–3584 bp/–3428 bp) [StAR-3584 forward: 5'-CATACGTGCACTGTCTTAGC-3'; StAR-3428 reverse: 5'-ACTCCTCCA GTAACTCCTTC-3'] regions of the StAR promoter (20). PCR was carried out using Accuprime Pfx supermix (Invitrogen Life Technologies, Carlsbad, CA) at 95 C for 5 min, followed by 39 cycles of 95 C for 20 sec, 57 C for 30 sec, and 68 C for 30 sec. A 2% agarose gel with ethidium bromide was used to separate and examine PCR products.
Results
cAMP treatment increases CREB phosphorylation in MA-10 cells
To identify members of the CREB family expressed in MA-10 cells, a CREB antibody (-CREB) with cross-reactivity to CREB p43, ATF-1, and CREM1 was used for immunoblot analysis of isolated NEs. As shown in Fig. 1A, the -CREB detected CREB (43 kDa), ATF-1 (38 kDa), and CREM (35 kDa) family members (-CREB). Only two immunodetectable bands corresponding to CREB and ATF-1 were detected when the samples were probed using a Ser 133 P-CREB/ATF-1/CREM-specific antibody, indicating that only CREB and ATF-1 are phosphorylated (Fig. 1A, -P-CREB). Analysis of the P-CREB to CREB ratio or P-ATF-1 to ATF-1 ratios indicated that treatment of the cells with (Bu)2cAMP for 30, 60, 120, or 240 min significantly increased P-CREB (43 kDa) 2.3-fold within 30 min of (Bu)2cAMP treatment, and this increase was sustained over the 4-h treatment time (Fig. 1B). However, (Bu)2cAMP treatment had no effect on the P-ATF-1 (38 kDa) levels (Fig. 1C), and a phospho-CREM immunoreactive band was not detected (Fig. 1A).
FIG. 1. (Bu)2cAMP-dependent increase in CREB phosphorylation in MA-10 mouse Leydig tumor cells. MA-10 cells were treated with 1 mM (Bu)2cAMP for the indicated time and NE was isolated. A, Ten micrograms of NE were separated by SDS-PAGE and P-CREB protein was detected by immunoblot analysis as described in Materials and Methods using an anti-P-CREB antibody (-P-CREB). The membrane was stripped and reprobed with an anti-CREB antibody (-CREB). Shown is an immunoblot of NE samples, and the experiment was repeated two additional times using whole-cell lysates. Arrows indicate immunoreactive phosphorylated proteins P-CREB(p43) and P-ATF-1(p38), and total proteins CREB1(p43), ATF-1(p38), and CREM(p35). B and C, The IOD ratios for P-CREB to CREB or P-ATF-1 to ATF-1 were determined as described in Materials and Methods, and shown are the mean values ± SEM from three independent experiments except for the 0- and 120-min time points that have an n = 7. The asterisks indicate a significant difference between treated and control samples based on ANOVA analysis followed by Dunnett’s posttest. *, P < 0.05 and **, P < 0.01.
CREB and ATF-1, but not CREM, bind to the AP-1 element in the CAN region of the StAR promoter
DNA-affinity chromatography was used to examine which CREB family members associate with the CAN region of the StAR promoter. As shown in Fig. 2, only CREB and ATF-1, but not CREM, were detected in the 0.2 M KCl elution fraction from the CAN wild-type affinity column. No CREB/ATF-1 protein was retained by the DNA-affinity column containing point mutations within the AP-1 element (Fig. 2B). These data confirm the importance of the AP-1 site for CREB and ATF-1 binding within this region (17, 18) and suggest that only the CREB family members detected as phosphoproteins (Fig. 1A, -P-CREB) bind to the StAR promoter.
FIG. 2. CREB and ATF-1 associate with the AP-1 half-site within the CAN region of the StAR promoter. DNA-affinity columns using a single copy of the 23-bp wild-type (CANwt) or AP-1 mutant [CANmt (AP-1)] CAN region of the StAR promoter were prepared and column fractions collected as outlined in Materials and Methods. One percent of total input NE, 2% of the FT volume, and 4% of 0.1 M KCl, 0.2 M KCl fractions were separated by SDS-PAGE, and CREB detection was performed as described in Materials and Methods. A and B, CREB family members detected using an anti-CREB antibody (-CREB) show only CREB and ATF-1 recovered in the 0.2 M KCl fraction from CANwt column (A) and no binding to the CANmt column (B). C, Duplicate gel as shown in A except anti-CREM1 antibody (-CREM1) was used for the immunoblot analysis. A–C, Shown are representative blots from two independent experiments using two different NE preparations. A, Similar results were obtained in four additional experiments, each using new NE preparations performed using the batch-binding procedure for DNA-affinity chromatography as described in Materials and Methods. Arrows indicate immunoreactive proteins CREB(p43), ATF-1(p38), and CREM(p35).
Because the anti-CREB antibody used in these studies recognizes CREB, ATF-1, and CREM, we probed the CREB protein profile with an additional antibody (-CREM) that has greater specificity for CREM1 relative to other family members. An immunodetectable band corresponding to the 35-kDa CREM protein was detected in NE (Fig. 2C). No immunoreactive bands were detected in the 0.2 M KCl fraction from the wild-type CAN affinity column using the CREM antibody (Fig. 2C). Together these results indicate that only CREB and ATF-1 bind to the CAN region of the StAR promoter and that the AP-1 element is essential for this interaction.
cAMP treatment increases P-CREB association with the CAN region of the StAR promoter
DNA-affinity chromatography was again used to determine the cAMP effect on CREB-DNA association. NE was isolated from untreated (control) and (Bu)2cAMP-treated (30 or 120 min) MA-10 cells, and the binding of CREB to the CAN wild-type affinity column was determined by immunoblot analysis. No significant difference was observed for total CREB and ATF-1 binding to the CAN region of StAR promoter among control, 30, or 120 min (Bu)2cAMP-treated NE (Fig. 3, A and B, compare BND lanes in panel 2; Fig. 3C). On the other hand, only P-CREB binding to the AP-1 element in the CAN region was significantly increased after (Bu)2cAMP treatment (Fig. 3, A and B, compare BND lanes in panel 1; Fig. 3C). These data indicate that either 30 or 120 min (Bu)2cAMP treatment resulted in a 2-fold increase in P-CREB association with the StAR promoter, although total CREB levels bound to the CAN element remained constant (Fig. 3C).
FIG. 3. The effect of (Bu)2cAMP stimulation on CREB association with the CAN region of the StAR promoter. A and B, Batch-binding DNA-affinity chromatography was performed using control, 30, and 120 min (Bu)2cAMP-treated MA-10 NE as described in Materials and Methods. Collected fractions were separated by SDS-PAGE and P-CREB (-P-CREB) and CREB (-CREB) detected by immunoblot analysis. The indicated lanes contain 10% of input NE, 10% of column FT volume, 20% of 0.1 M KCl wash, and total bound protein eluted with SDS-sample buffer (BND). The membrane was probed with an anti (Ser133)-P-CREB (-P-CREB) antibody and then stripped and reprobed for total CREB using an anti-CREB (-CREB) antibody. A, Shown are representative blots from three (A) or four (B) independent experiments using different NE preparations. Arrows indicate P-CREB(p43) and P-ATF-1(p38), and total CREB(p43), ATF-1(p38), and CREM(p35). C, IOD values for total and P-CREB and P-ATF-1 in the bound fractions (BND), and the ratios for P-CREB to CREB or P-ATF-1 to ATF-1 were determined for control, 30, or 120 min (Bu)2cAMP-treated NE as described in Materials and Methods. The (Bu)2cAMP-treated values were normalized to the control values [(Bu)2cAMP IOD/control IOD], and shown are the mean values ± SEM. An unpaired Student’s t test was used to compare the difference between the (Bu)2cAMP and control samples for total CREB and ATF-1 binding and P-CREB (P-CREB/CREB) and P-ATF-1 (P-ATF-1/ATF-1) binding from three (30 min) or four (120 min) independent experiments. *, P < 0.05 between treated and control samples.
cAMP stimulation increases P-CREB and CBP association with the StAR proximal promoter in vivo
To determine the effect of (Bu)2cAMP treatment of MA-10 cells on CREB and P-CREB association with the StAR promoter under more physiologically relevant conditions, ChIP assays were performed. Cross-linked, sheared chromatin from 30 or 120 min (Bu)2cAMP-treated MA-10 cells was immunoprecipitated with anti-CREB (-CREB) antibodies, and recovered DNA was subjected to PCR using primers specific to the proximal region (–299 bp/–41 bp) of the StAR promoter encompassing the CAN region. As shown in Fig. 4, a 258-bp PCR product specific for StAR proximal promoter was amplified from -CREB-immunoprecipitated DNA samples. In contrast, little or no amplification product was observed with DNA recovered using IgG for immunoprecipitation or when primers for the StAR distal promoter region were used. These data indicate specificity for association of CREB within the proximal region of the StAR promoter. Furthermore, the amount of StAR promoter association with CREB was not increased with either 30- or 120-min (Bu)2cAMP treatment (Fig. 4). These data suggest that total CREB association with the proximal region of the StAR promoter is not qualitatively changed after treatment of the cells with (Bu)2cAMP for 30 or 120 min, consistent with the DNA-affinity chromatography results.
FIG. 4. CREB and c-Jun association with the proximal region of the StAR promoter is not affected by cAMP treatment in MA-10 cells. MA-10 cells were treated in the absence or presence of 1 mM (Bu)2cAMP for 30 or 120 min, and ChIP assay was performed as described in Materials and Methods. Cross-linked, sheared chromatin was immunoprecipitated with IgG, anti-CREB, or anti-c-Jun antibodies. Recovered chromatin was subjected to PCR analysis using primers (–299 bp/–41bp) encompassing the CAN region of the StAR promoter and primers approximately 3.5 kb upstream of the CAN region (–3584 bp/–3428 bp). A, Schematic representation of StAR promoter with location of primer pairs indicated. B and C, Ethidium bromide-stained gel of PCR products. Primers for the proximal and distal region were expected to amplify a 258- and 156-bp fragment, respectively. Shown is a representative ChIP analysis from three independent experiments.
In parallel experiments, c-Jun association with the StAR proximal promoter was also examined by ChIP because recent reports have implicated a functional AP-1 interaction via the same AP-1 element in the CAN region that binds CREB (28). A PCR product specific for StAR proximal promoter, but not the distal promoter region, was amplified from -c-Jun-immunoprecipitated DNA samples. As observed with CREB, 30- or 120-min (Bu)2cAMP-treatment did not change c-Jun interaction with the StAR proximal promoter because similar levels of end point PCR product were amplified from control and treated cells. Together these data reveal that CREB and AP-1 associate with the StAR proximal promoter in intact chromatin in MA-10 cells and that this association is not influenced by (Bu)2cAMP treatment.
On the other hand, in the absence of (Bu)2cAMP-treatment, ChIP analysis using P-CREB or CBP antibodies indicated low to absent association of these factors with the StAR proximal promoter (Fig. 5). However, the presence of the PCR product specific for StAR proximal promoter was observed after 30 min and continuing up to 120 min of (Bu)2cAMP stimulation, indicating an increase in P-CREB and CBP interaction at the StAR promoter. Again, the specificity of association of P-CREB and CBP with the proximal region of the StAR promoter was demonstrated by the absence of PCR product when IgG was used for immunoprecipitation or primers designed to amplify the distal region of the StAR promoter were used. It is important to note that the P-CREB antibody used in these experiments does not distinguish between P-CREB and P-ATF-1. However, the in vitro DNA-affinity chromatography results in Fig. 3 strongly suggest that the increase in phospho-protein association with the StAR promoter is due to P-CREB and not P-ATF-1. In sum, these data suggest that (Bu)2cAMP-treatment promotes an increase in P-CREB and CBP interaction with the StAR proximal promoter.
FIG. 5. Association of P-CREB and CBP with the proximal region of the StAR promoter is increased on cAMP treatment in MA-10 cells. ChIP analysis was performed as described in Materials and Methods on MA-10 cells treated with or without 1 mM (Bu)2cAMP for 30 or 120 min. Cross-linked, sheared chromatin was immunoprecipitated (IP) with either anti-P-CREB or anti-CBP antibodies (CBP), and recovered chromatin was subjected to PCR analysis using primers encompassing the CAN region of the StAR promoter or distal region as shown in Fig. 4A. A and B, Shown is an ethidium bromide-stained gel of PCR products that is representative of ChIP analysis from three (A) or four (B) independent experiments.
Discussion
The transcriptional regulation of the murine StAR gene is complex with numerous factors involved in both the basal and cAMP-mediated effects on StAR transcription (9, 17, 18, 19, 29, 30, 31, 32, 33, 34). These studies show not only the significance of each factor alone but also recently the functional cooperation of these factors and the cis-acting elements that are required for StAR transcription (9, 19, 30, 31, 34). Of these, even in the absence of a canonical CRE, CREB family members have been shown to interact with the StAR promoter, and that their transactivation ability is dependent on phosphorylation (18). We have now examined the extent of endogenous CREB phosphorylation in MA-10 cells after cAMP stimulation in a time-dependent manner and demonstrate that promoter occupancy by P-CREB is required for CBP recruitment.
In the present study, we have identified three immunoreactive CREB bands in MA-10 whole-cell and NEs that correspond to CREB(p43), ATF-1(p38), and CREM(p35) (35, 36, 37). Our data are consistent with previous studies that indicated multiple CREB family members are expressed in MA-10 cells (18). Of the isoforms present in MA-10 cells, we now demonstrate that only CREB and ATF-1 are phosphorylated and that only CREB phosphorylation increased in response to (Bu)2cAMP stimulation. CREB phosphorylation was shown to occur within 30 min of (Bu)2cAMP treatment and remained elevated up to 4 h. It has been reported that the cAMP-PKA signaling cascade in context of CREB phosphorylation occurs via burst-attenuation kinetics in the NIH 3T3 cell line. Forskolin treatment resulted in a rapid increase in PKA-dependent phosphorylation of CREB (Ser133) after 30 min (burst phase) with a subsequent decline after 2 h (attenuation phase) by a protein phosphatase 1-dependent mechanism (38). Our results suggest that this may be tissue or promoter specific because CREB phosphorylation in MA-10 cells was still elevated after 4 h of (Bu)2cAMP stimulation. One possibility for this difference may be that the burst-attenuation kinetics are prolonged in MA-10 Leydig cells. Alternatively, cAMP analogs may mimic intracellular cAMP levels at higher concentrations and for longer time periods in comparison with forskolin treatment, as used in the NIH 3T3 studies.
Using DNA-affinity chromatography, we demonstrate that both CREB and ATF-1, but not CREM, bind to the cAMP-responsive region of the StAR promoter. DNA binding did not appear to be dependent on phosphorylation, although cAMP treatment increased the amount of P-CREB that is associated with the StAR promoter. A limitation of the in vitro DNA-binding assay is that the data do not distinguish between whether DNA-associated CREB is phosphorylated in response to PKA activation or a dynamic model in which (Bu)2cAMP treatment of MA-10 cells increases the overall phosphorylation state of CREB that is then able to associate with the StAR promoter. In contrast, our ChIP data indicate that P-CREB is not significantly associated with the StAR promoter before (Bu)2cAMP treatment, supporting a dynamic model for P-CREB association with StAR promoter in which the increase in nuclear P-CREB after (Bu)2cAMP treatment results in more P-CREB binding to the CAN region of the StAR promoter without increasing total amount of bound CREB.
The lack of CREM binding to the StAR promoter in our study contrasts previous data that indicated CREM as the CREB family member that bound the StAR promoter based on protein-DNA complex supershift EMSA (18). The antibody source is different between these two studies and may attribute to the differences because we were unable to demonstrate CREB or CREM binding to the CAN region by EMSA analysis using our antibodies (Ref.17 , and data not shown). Nevertheless, using DNA-affinity chromatography, which provides a more powerful molecular tool for examining protein-DNA interactions by allowing immunoblot protein detection and direct estimation of bound protein molecular weights, we demonstrated that CREB and ATF-1, but not CREM, bind to the StAR promoter.
CREB binding is dependent on the AP-1-like element in the cAMP-responsive region of StAR promoter. Our previous data suggest that this site is essential for basal activity of the StAR promoter (17, 18). However, mutation of the AP-1 site in context of StAR luciferase reporter constructs had no significant effect on the fold increase in StAR transcription after (Bu)2cAMP treatment, even though the overall strength of the promoter was diminished (17). The lack of effect on the cAMP response may be attributed to protein binding to surrounding functional cis-acting elements (SF-1 and/or GATA-4), which contribute to the overall stimulated response and therefore may stabilize the cAMP-dependent complex even in the presence of the AP-1 mutation (17, 18). Because the affinity columns contained only 23 bp of the previously reported CAN region and lacked the SF-1 and GATA-4 elements (17), future studies can be focused to investigate whether CREB association with the cAMP-responsive region would be observed in the presence of the AP-1 mutation and intact SF-1 and GATA-4 elements.
CREB and P-CREB association with the StAR proximal promoter was confirmed using the more physiological relevant ChIP approach. We observed no qualitative difference in CREB interaction with the StAR promoter between control and 30 or 120 min (Bu)2cAMP treatment. These data are also consistent with a recent report on protein-DNA interactions with the StAR promoter using ChIP analysis that showed CREB association with the proximal region of the StAR promoter was increased 50% after 30 min of 8-bromoadenosine-cAMP treatment but was comparable with control levels after 2 h, suggesting a temporal association of CREB with the StAR promoter (21). The discrepancy between our data and the previous study at the 30-min time point may be due to difference in detection used for the ChIP analysis between real-time PCR vs. end point ethidium bromide staining. Possibly the increased phosphorylation of CREB at this short time point may contribute to the previous observed increase in total CREB association using ChIP. However, we show CREB phosphorylation remains elevated for an extended time; therefore, the apparent time-dependent association most likely reflects other factors in addition to CREB phosphorylation.
ChIP analysis was also used to investigate the classical cAMP-PKA signaling cascade leading to CREB phosphorylation and CBP recruitment to StAR proximal promoter in MA-10 cells. Our results demonstrate an increase in P-CREB association with the StAR proximal promoter after 30 or 120 min treatment with (Bu)2cAMP, compared with control cells, again consistent with the in vitro DNA-affinity chromatography studies. The results of the ChIP analysis for CBP association with the StAR proximal promoter paralleled the data with P-CREB in that a significant association was observed only after (Bu)2cAMP treatment. Recently ChIP analysis on the cAMP-dependent temporal association of SF-1, GATA-4, CREB, and CBP with the murine StAR demonstrated that alterations in histone modifications (acetylation and methylation) paralleled CBP association with the mouse StAR proximal promoter in response to 8-bromoadenosine-cAMP treatment. These studies support the premise that the activity of the coactivator CBP contributes to a more transcriptionally active chromatin structure (20, 21). Our data support these results and extend these findings to demonstrate for the first time the physical role of phosphorylated CREB-DNA interactions in CBP recruitment and the cAMP-stimulation of StAR transcription.
In addition to CREB binding to the AP-1-like element within the CAN region of StAR promoter, recent studies have shown that overexpression of the AP-1 family member c-Jun increased StAR reporter gene activity in transiently transfected MA-10 and Y1 cells (28). The transcriptional activation by c-Jun was dependent on the AP-1 element, although no direct binding of c-Jun to the StAR promoter was detected by EMSA. The authors suggested that c-Jun may associate with the StAR cAMP-responsive region, possibly through a tethering mechanism with c-fos or Fra-2 (28). Therefore, we tested for a c-Jun-StAR proximal promoter interaction in vivo by ChIP. Our data confirm that c-Jun associates with the murine StAR proximal promoter and that the association remains constant in response to 30 or 120 min (Bu)2cAMP treatment, suggesting c-Jun, as well as CREB, has the ability to constitutively bind to the proximal StAR promoter. However, we cannot conclude that the CREB and c-Jun associations observed in our ChIP analysis are from the same chromatin templates or represent binding to the same element. Although the functional relevance for CREB and c-Jun individually on StAR promoter activity had previously been reported (18, 19, 28), our ChIP analysis would support a model that both CREB and AP-1 associate with the StAR proximal promoter under basal conditions without disrupting the cAMP-PKA transactivation as defined by the simultaneous association of c-Jun, P-CREB and CBP with StAR promoter. Although we cannot exclude the possibility that c-Jun association with StAR promoter may also enhance CBP recruitment with (Bu)2cAMP treatment, the finding that CREB phosphorylation was increased by (Bu)2cAMP treatment and P-CREB and CBP promoter association in vivo occurred concurrently and only after (Bu)2cAMP treatment support P-CREB as functionally relevant in coactivator recruitment.
In sum, our data indicate that the classical cAMP-PKA-mediated CREB phosphorylation and CBP recruitment signaling cascade is involved in the cAMP-stimulated response of the murine StAR gene. CREB is constitutively bound to the AP-1 site within the cAMP-responsive region, and cAMP-treatment enhances the P-CREB to CREB ratio within the nucleus, thereby allowing for an increase in P-CREB interaction at the AP-1 cis-acting element. Previous studies implicated CREB family members and the importance of CREB phosphorylation in cAMP-dependent StAR transcription and CBP association with the StAR promoter (18, 21). Our data provide the link between these observations and show that StAR promoter occupancy by CBP occurs concurrently with P-CREB and not CREB or c-Jun. These data are consistent with the model for enhanced P-CREB association with the AP-1 element that results in recruitment of CBP, thereby providing a mechanism for histone acetylation and increased StAR transcription.
Acknowledgments
The authors thank Terry Wright for his contribution to the ChIP analysis and Drs. Carolyn Klinge and Doug Darling (University of Louisville, Louisville, KY) for their review of this manuscript.
References
Parker KL, Schimmer BP 1995 Transcriptional regulation of the genes encoding the cytochrome P-450 steroid hydroxylases. Vitam Horm 51:339–370
Jefcoate CR, DiBartolomeis MJ, Williams CA, McNamara BC 1987 ACTH regulation of cholesterol movement in isolated adrenal cells. J Steroid Biochem 27:721–729
Ohno Y, Yanagibashi K, Yonezawa Y, Ishiwatari S, Matsuba M 1983 A possible role of "steroidogenic factor" in the corticoidogenic response to ACTH; effect of ACTH, cycloheximide and aminoglutethimide on the content of cholesterol in the outer and inner mitochondrial membrane of rat adrenal cortex. Endocrinol Jpn 30:335–338
Simpson ER, Boyd GS 1967 The cholesterol side-chain cleavage system of bovine adrenal cortex. Eur J Biochem 2:275–285
Privalle CT, Crivello JF, Jefcoate CR 1983 Regulation of intramitochondrial cholesterol transfer to side-chain cleavage cytochrome P-450 in rat adrenal gland. Proc Natl Acad Sci USA 80:702–706
Stocco DM, Clark BJ 1996 Regulation of the acute production of steroids in steroidogenic cells. Endocr Rev 17:221–244
Crivello JF, Jefcoate CR 1980 Intracellular movement of cholesterol in rat adrenal cells. Kinetics and effects of inhibitors. J Biol Chem 255:8144–8151[Free Full Text]
Clark BJ, Soo SC, Caron KM, Ikeda Y, Parker KL, Stocco DM 1995 Hormonal and developmental regulation of the steroidogenic acute regulatory protein. Mol Endocrinol 9:1346–1355
Sugawara T, Saito M, Fujimoto S 2000 Sp1 and SF-1 interact and cooperate in the regulation of human steroidogenic acute regulatory protein gene expression. Endocrinology 141:2895–2903
Clark BJ, Combs R 1999 Angiotensin II and cyclic adenosine 3',5'-monophosphate induce human steroidogenic acute regulatory protein transcription through a common steroidogenic factor-1 element. Endocrinology 140:4390–4398
Sandhoff TW, Hales DB, Hales KH, McLean MP 1998 Transcriptional regulation of the rat steroidogenic acute regulatory protein gene by steroidogenic factor 1. Endocrinology 139:4820–4831
Rust W, Stedronsky K, Tillmann G, Morley S, Walther N, Ivell R 1998 The role of SF-1/Ad4BP in the control of the bovine gene for the steroidogenic acute regulatory (StAR) protein. J Mol Endocrinol 21:189–200
Sugawara T, Kiriakidou M, McAllister JM, Kallen CB, Strauss 3rd JF 1997 Multiple steroidogenic factor 1 binding elements in the human steroidogenic acute regulatory protein gene 5'-flanking region are required for maximal promoter activity and cyclic AMP responsiveness. Biochemistry 36:7249–7255
Caron KM, Ikeda Y, Soo SC, Stocco DM, Parker KL, Clark BJ 1997 Characterization of the promoter region of the mouse gene encoding the steroidogenic acute regulatory protein. Mol Endocrinol 11:138–147
De Cesare D, Sassone-Corsi P 2000 Transcriptional regulation by cyclic AMP-responsive factors. Prog Nucleic Acids Res Mol Biol 64:343–369
Montminy M 1997 Transcriptional regulation by cyclic AMP. Annu Rev Biochem 66:807–822
Wooton-Kee CR, Clark BJ 2000 Steroidogenic factor-1 influences protein-deoxyribonucleic acid interactions within the cyclic adenosine 3,5-monophosphate-responsive regions of the murine steroidogenic acute regulatory protein gene. Endocrinology 141:1345–1355
Manna PR, Dyson MT, Eubank DW, Clark BJ, Lalli E, Sassone-Corsi P, Zeleznik AJ, Stocco DM 2002 Regulation of steroidogenesis and the steroidogenic acute regulatory protein by a member of the cAMP response-element binding protein family. Mol Endocrinol 16:184–199
Manna PR, Eubank DW, Lalli E, Sassone-Corsi P, Stocco DM 2003 Transcriptional regulation of the mouse steroidogenic acute regulatory protein gene by the cAMP response-element binding protein and steroidogenic factor 1. J Mol Endocrinol 30:381–397
Christenson LK, Stouffer RL, Strauss 3rd JF 2001 Quantitative analysis of the hormone-induced hyperacetylation of histone H3 associated with the steroidogenic acute regulatory protein gene promoter. J Biol Chem 276:27392–27399
Hiroi H, Christenson LK, Chang L, Sammel MD, Berger SL, Strauss 3rd JF 2004 Temporal and spatial changes in transcription factor binding and histone modifications at the steroidogenic acute regulatory protein (stAR) locus associated with stAR transcription. Mol Endocrinol 18:791–806
Weih F, Stewart AF, Boshart M, Nitsch D, Schutz G 1990 In vivo monitoring of a cAMP-stimulated DNA-binding activity. Genes Dev 4:1437–1449
Wolfl S, Martinez C, Majzoub JA 1999 Inducible binding of cyclic adenosine 3',5'-monophosphate (cAMP)-responsive element binding protein (CREB) to a cAMP-responsive promoter in vivo. Mol Endocrinol 13:659–669
Sibinga NE, Wang H, Perrella MA, Endege WO, Patterson C, Yoshizumi M, Haber E, Lee ME 1999 Interferon--mediated inhibition of cyclin A gene transcription is independent of individual cis-acting elements in the cyclin A promoter. J Biol Chem 274:12139–12146
Tommasi S, Pfeifer GP 1999 In vivo structure of two divergent promoters at the human PCNA locus. Synthesis of antisense RNA and S phase-dependent binding of E2F complexes in intron 1. J Biol Chem 274:27829–27838
Nichols M, Weih F, Schmid W, DeVack C, Kowenz-Leutz E, Luckow B, Boshart M, Schutz G 1992 Phosphorylation of CREB affects its binding to high and low affinity sites: implications for cAMP induced gene transcription. EMBO J 11:3337–3346
Dey A, Nebert DW, Ozato K 1991 The AP-1 site and the cAMP- and serum response elements of the c-fos gene are constitutively occupied in vivo. DNA Cell Biol 10:537–544
Manna PR, Eubank DW, Stocco DM 2004 Assessment of the role of activator protein-1 on transcription of the mouse steroidogenic acute regulatory protein gene. Mol Endocrinol 18:558–573
Sugawara T, Kiriakidou M, McAllister JM, Holt JA, Arakane F, Strauss 3rd JF 1997 Regulation of expression of the steroidogenic acute regulatory protein (StAR) gene: a central role for steroidogenic factor 1. Steroids 62:5–9
Reinhart AJ, Williams SC, Clark BJ, Stocco DM 1999 SF-1 (steroidogenic factor-1) and C/EBP? (CCAAT/enhancer binding protein-?) cooperate to regulate the murine StAR (steroidogenic acute regulatory) promoter. Mol Endocrinol 13:729–741
Silverman E, Eimerl S, Orly J 1999 CCAAT enhancer-binding protein ? and GATA-4 binding regions within the promoter of the steroidogenic acute regulatory protein (StAR) gene are required for transcription in rat ovarian cells. J Biol Chem 274:17987–17996
Christenson LK, Johnson PF, McAllister JM, Strauss 3rd JF 1999 CCAAT/enhancer-binding proteins regulate expression of the human steroidogenic acute regulatory protein (StAR) gene. J Biol Chem 274:26591–26598
Tremblay JJ, Viger RS 2001 GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory elements. Endocrinology 142:977–986
Tremblay JJ, Hamel F, Viger RS 2002 Protein kinase A-dependent cooperation between GATA and CCAAT/enhancer-binding protein transcription factors regulates steroidogenic acute regulatory protein promoter activity. Endocrinology 143:3935–3945
Montminy MR, Bilezikjian LM 1987 Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature 328:175–178
Liu F, Thompson MA, Wagner S, Greenberg ME, Green MR 1993 Activating transcription factor-1 can mediate Ca(2+)- and cAMP-inducible transcriptional activation. J Biol Chem 268:6714–6720
Behr R, Weinbauer GF 1999 Germ cell-specific cyclic adenosine 3',5'-monophosphate response element modulator expression in rodent and primate testis is maintained despite gonadotropin deficiency. Endocrinology 140:2746–2754
Canettieri G, Morantte I, Guzman E, Asahara H, Herzig S, Anderson SD, Yates 3rd JR, Montminy M 2003 Attenuation of a phosphorylation-dependent activator by an HDAC-PP1 complex. Nat Struct Biol 10:175–181(Brian F. Clem, Elizabeth )
Address all correspondence and requests for reprints to: Dr. Barbara J. Clark, Department of Biochemistry and Molecular Biology, University of Louisville School of Medicine, Louisville, Kentucky 40292. E-mail: bjclark@louisville.edu.
Abstract
Steroidogenic acute regulatory protein (StAR) transcription is regulated through cAMP-protein kinase A-dependent mechanisms that involve multiple transcription factors including the cAMP-responsive element binding protein (CREB) family members. Classically, binding of phosphorylated CREB to cis-acting cAMP-responsive elements (5'-TGACGTCA-3') within target gene promoters leads to recruitment of the coactivator CREB binding protein (CBP). Herein we examined the extent of CREB family member phosphorylation on protein-DNA interactions and CBP recruitment with the StAR promoter. Immunoblot analysis revealed that CREB, cAMP-responsive element modulator (CREM), and activating transcription factor (ATF)-1 are expressed in MA-10 mouse Leydig tumor cells, yet only CREB and ATF-1 are phosphorylated. (Bu)2cAMP treatment of MA-10 cells increased CREB phosphorylation approximately 2.3-fold within 30 min but did not change total nuclear CREB expression levels. Using DNA-affinity chromatography, we now show that CREB and ATF-1, but not CREM, interact with the StAR promoter, and this interaction is dependent on the activator protein-1 (AP-1) cis-acting element within the cAMP-responsive region. In addition, (Bu)2cAMP-treatment increased phosphorylated CREB (P-CREB) association with the StAR promoter but did not influence total CREB interaction. In vivo chromatin immunoprecipitation assays demonstrated CREB binding to the StAR proximal promoter is independent of (Bu)2cAMP-treatment, confirming our in vitro analysis. However, (Bu)2cAMP-treatment increased P-CREB and CBP interaction with the StAR promoter, demonstrating for the first time the physical role of P-CREB:DNA interactions in CBP recruitment to the StAR proximal promoter.
Introduction
STEROIDOGENESIS IN THE adrenals and gonads is regulated by gonadotropin trophic hormone release from the anterior pituitary and is controlled in two temporal phases, beginning within minutes for the acute phase and continuing longer term for the chronic phase. The chronic phase is mediated by an increase in gene expression for the steroid hydroxylases enzymes involved in metabolizing cholesterol into tissue-specific steroid hormones (1), whereas the acute phase of steroidogenesis is characterized by the rapid uptake of cholesterol into the mitochondria (2, 3, 4). The steroidogenic acute regulatory protein (StAR) regulates the acute phase of steroid biosynthesis by mediating the cholesterol transfer from the outer- to the inner-mitochondrial membrane, in which it is subsequently metabolized into pregnenolone by the cytochrome P450 side-chain cleavage enzyme (5, 6). This action of cholesterol transfer has been shown to be the rate-limiting step in acute steroid biosynthesis thereby providing a mechanism by which tissues can control steroid hormone output by regulating StAR protein levels and thus function (7).
Previous studies have shown that StAR is regulated at the transcriptional level via a cAMP-protein kinase A (PKA) signaling cascade within the gonads (8, 9, 10, 11, 12, 13, 14). Transactivation of target genes by the cAMP-PKA pathway involves the binding of cAMP-responsive element binding protein (CREB) family members to consensus cAMP-response elements (CRE; 5'-TGACGTCA-3') within the gene promoter. Within the classical cAMP-PKA biochemical pathway, activated PKA phosphorylates CREB on Ser133, which increases association with the CREB binding protein (CBP) coactivator, resulting in histone modification and increased transcription (15, 16). Although the StAR promoter lacks a canonical CRE, we previously localized the cAMP-responsive region of the mouse StAR promoter region between –105 and –60 bp upstream of the transcriptional start site and identified functional elements for steroidogenic factor-1 (SF-1) (–95), CCAAT/enhancer binding protein-?/nonconsensus activating protein-1/nuclear receptor half-site [(CAN) region –79 bp] and GATA-4 (–68 bp) (17). Mutation of the activator protein-1 (AP-1)-like element in the CAN region resulted in decreased basal promoter activity and abolished protein DNA-interactions within this region as determined by reporter gene assay and EMSA, respectively, whereas SF-1 binding was shown to stabilize protein interactions with the CAN region (17). Recently CREB family members were shown to directly associate with the StAR promoter via this AP-1-like element within the CAN region, and CREB overexpression was shown to increase the cAMP-dependent transactivation of StAR promoter (18). Furthermore, overexpression of both CREB and SF-1 resulted in a synergistic activation after cAMP treatment, suggesting a functional cooperation between these two factors (19). More importantly, the CREB-dependent transcriptional response was shown to be dependent on CREB phosphorylation because overexpression of phosphorylation mutants led to a loss in transactivation capability as well as functional interactions with SF-1 (18, 19).
These studies suggest that a component of the cAMP-mediated StAR transactivation involves the classical cAMP-PKA signaling pathway of CREB phosphorylation with possible subsequent recruitment of a coactivator. Indeed, recent reports of increased histone acetylation, a marker for coactivator activity, on the StAR gene promoter as well as CBP association within the proximal region of the mouse StAR promoter in MA-10 cells supports a classical cAMP-PKA signaling pathway for StAR transcriptional activation (20, 21). However, there are contradictory reports within the literature on the effects of cAMP on CREB-DNA associations with various cAMP-responsive promoters. Investigators examining the liver-specific tyrosine aminotransferase or the corticotropin-releasing factor gene promoters using genomic footprinting or in vivo cross-linking, respectively, concluded that CREB-DNA interactions with specific CREs were increased by cAMP treatment (22, 23). In contrast, studies on other gene promoters, such as the T-cell receptor, phosphoenolpyruvate carboxykinase, proliferating cell nuclear antigen, cyclin A, and c-fos showed no influence on CRE occupation with cAMP treatment in an in vivo chromatin background (24, 25, 26, 27). We previously reported a new protein-DNA complex is present in EMSA analysis using nuclear extract from (Bu)2cAMP-treated MA-10 cells, compared with control, suggesting recruitment of a cAMP-responsive factor to the StAR promoter (17).
The purpose of the present study was to examine the role of cAMP treatment on endogenous CREB family member phosphorylation and the role of phospho-CREB in subsequent CREB-DNA interactions as well as the in situ effects of CREB, phospho-CREB, and CBP association with the StAR promoter. We now demonstrate that (Bu)2cAMP stimulation of MA-10 mouse Leydig tumor cells increased CREB phosphorylation 2.3-fold within 30 min. However, the total CREB-DNA interaction was not influenced by cAMP treatment but rather that phospho-CREB association and CBP recruitment to the StAR proximal promoter were increased. Therefore, we conclude that even in the absence of a canonical CRE, the classical cAMP-PKA signaling cascade activates phospho-CREB and CBP binding to the CAN region and up-regulates transcription of the mouse StAR gene.
Materials and Methods
Materials
Waymouth’s MB/752, N6,2'-O-(Bu)2cAMP, and streptavidin-conjugated agarose solution were purchased from Sigma (St. Louis, MO). Gentamicin reagent and Dulbecco’s PBS were purchased from Life Technologies, Inc. (Rockville, MD). The protein assay kit was supplied by Bio-Rad Life Sciences (Hercules, CA). Biotin sense and antisense oligonucleotides for DNA-affinity chromatography and PCR primers for chromatin immunoprecipitation (ChIP) analysis were purchased from Midland Certified Reagent Co. (Midland, TX). T4-polynucleotide kinase was purchased from Promega Corp. (Madison, WI). [-32P]ATP was purchased from NEN Life Science Products (Boston, MA). Anti-CREB (sc-186), anti-CREM-1 (sc-440) and anti-c-Jun (sc-44) antibodies were supplied by Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Anti-phospho-CREB (06–519), anti-CBP (06–294), and the ChIP assay kit were purchased from Upstate USA, Inc. (Charlottesville, VA).
Cell culture
The MA-10 mouse Leydig tumor cell line was a gift from Dr. M. Ascoli (Department of Pharmacology, University of Iowa, Iowa City, IA). MA-10 cells were cultured in Waymouth’s MB/752 media supplemented with 15% heat-inactivated horse serum and 40 μg gentamicin sulfate per milliliter. Treatment of MA-10 cells with (Bu)2cAMP was performed in serum-depleted Waymouth’s MB/752 for the indicated time points.
Whole-cell and nuclear extract preparations
For whole-cell lysates, MA-10 cells were plated at 1 x 106 in 60-mm dishes. Forty-eight hours after seeding, cells were serum starved for 24 h before treatment in the presence or absence of (Bu)2cAMP for 0, 30, 60, 120, or 240 min in serum-free Waymouth’s MB/752 media. After treatment, cells were washed once with PBS(+) and collected by scraping in 1.5 ml PBS(+). Cells were pelleted by centrifugation at 10,000 rpm for 5 min, and resuspended in 300 μl Tris-sucrose-EDTA [10 mM Tris-HCl (pH 7.4), 250 mM sucrose, and 1 mM EDTA] plus protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml leupeptin) and phosphatase inhibitors [100 nM okadaic acid plus phosphatase inhibitor cocktail II (Sigma)]. Cells were lysed by repeated freeze-thaw (three times) followed by passage through an 18.5-gauge needle (15 times). After centrifugation at 2000 rpm for 10 min, supernatant was removed and protein concentration was determined by Bradford protein assay. Nuclear extract preparations from untreated or treated MA-10 cells for 0, 30, 60, 120, or 240 min with (Bu)2cAMP were performed as previously described (17). As with the whole-cell lysates, a protein assay was performed to determine protein concentrations, and all samples were stored at –80 C until further analysis.
DNA-affinity chromatography
Oligonucleotide annealing and purification.
A biotinylated sense strand (5'-biotin-TGCACAATGACTGATGACTTTTT-3') corresponding to the 23-bp CAN region of the StAR promoter was annealed to its respective antisense strand (5'-AAAAAGTCATCAGTCATTGTGCA-3'). Alternatively, sense and antisense strands with mutations generated within the AP-1 site [(sense: 5'-biotin-TGCACAATAGATCTTGAC TTTTT-3'), (antisense: 5'-AAAAAGTCAAGATCTATTGTGCA-3')] were used in the annealing reactions. Equal molar amounts of sense and antisense strands were mixed in annealing buffer [10 mM Tris-HCl (pH 7.5), 50 mM NaCl, and 1 mM EDTA] and boiled for 10 min with subsequent cooling to room temperature. To examine double-stranded annealing, 2.5 pmol of annealing reaction was 5' end-labeled with [-32P]ATP (NEN Life Science Products) using T4 polynucleotide kinase. Excess nucleotide was removed from the labeling reactions by a nucleotide removal kit (Qiagen, Santa Clarita, CA). The labeling reaction (50 fmol) was mixed with 5 μl EMSA dye [25 mM Tris-HCl (pH 8.0), 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, 20% glycerol, and 2.5% bromophenol blue/xylene canol] and separated by 15% nondenaturing PAGE. The gel was dried and exposed to phosphorscreen, and oligonucleotide detection was done using the Storm PhosphorImager (Molecular Dynamics, Amersham Biosciences, Sunnyvale, CA). Because 100% oligonucleotide annealing was never observed, a double-stranded oligonucleotide purification procedure was performed. Briefly, 50 fmol of radiolabeled oligonucleotide sample were added to their respective annealing reactions. Entire annealing reaction was separated by 15% nondenaturing PAGE, and the gel was exposed to x-ray film. The autoradiograph was placed back onto the gel, and the bands corresponding to the double-stranded oligonucleotides were excised. Gel slices were incubated with 0.3 M NaOAc for 16 h at 37 C with rotation. Acrylamide was pelleted by brief centrifugation, and supernatant was removed and EtOH precipitated. After purification, sample aliquots were again separated on a 15% nondenaturing PAGE to ensure the presence of only double-stranded oligonucleotides. DNA concentration was determined using the Ultraspec 3000 (Pharmacia Biotech, Uppsala, Sweden).
Column generation and chromatography procedure.
Biotin-wild-type CAN (1 nmol) or biotin-mutant CAN double-stranded oligonucleotide was incubated with 1 ml of a 50% streptavidin-conjugated agarose solution that had been equilibrated in the 1x Promega binding buffer [25 mM Tris-HCl (pH 8.0), 50 mM KCl, 6.25 mM MgCl2, 0.5 mM EDTA, and 10% glycerol]. The DNA-avidin mixture was rotated at 4 C for 1 h and added to a 1-ml syringe with glass wool stopper (approximately 500 μl column matrix). Nuclear extract (1 mg) from MA-10 cells treated in the presence or absence of 1 mM (Bu)2cAMP for 2 h was incubated with the column matrix for 2 h at 4 C. Column fractions were collected in 500-μl volumes. The void volume [termed flow thru (FT)] was collected followed by three 0.1 M KCl binding buffer washes to eliminate nonbinding nuclear extract proteins. Subsequently, three 0.2 M KCl washes were performed to elute specific DNA-binding proteins (bound proteins) from the affinity columns. Fractions were stored at –80 C until further analysis. The salt concentrations for the wash and elution were determined based on preliminary experiments using step KCl gradients followed by monitoring with Coomassie-stain gel, Western blot, and EMSA for DNA binding of the fractions.
Batch-binding DNA-affinity chromatography.
For repeat experiments of the column DNA-affinity chromatography, a batch binding method was used. Briefly, 100 μg of MA-10 nuclear extract were incubated with 100 pmol of double-stranded biotinylated CAN oligonucleotides on ice for 45 min. Subsequently, 100 μl of the 50% streptavidin-linked agarose bead slurry (50 μl bead pellet) preequilibrated in 0.1 M KCl Promega binding buffer was added and rotated at 4 C for 1 h. Streptavidin beads were pelleted by brief centrifugation, and the void volume [termed FT] was collected. The streptavidin beads were then washed three times with 0.1 M KCl binding buffer, and DNA-associated proteins were eluted by adding 50 μl of 2x sodium dodecyl sulfate (SDS)-Laemmli buffer and boiling at 95 C for 10 min. Samples were then immediately analyzed by the immunoblot procedure.
Immunoblot analysis
Fifty micrograms of whole-cell lysates, 10 μg nuclear extract, or indicated volumes of the nuclear extract (NE), column FT, 0.1 M KCl, and 0.2 M KCl fractions from the DNA-affinity chromatography procedures were analyzed. The protein samples were separated by 10% SDS-PAGE, transferred to polyvinyl difluoride membrane, and then the membranes were blocked in PBS containing 4% nonfat dry milk and 4% Tween 20 (PBS-Tween) overnight. The membrane was incubated with primary antibody (anti-phospho-CREB, anti-CREB, or anti-CREM1 rabbit polyclonal antibodies as indicated) in PBS-Tween with 2% nonfat dry milk and immunoblot analysis performed using a horseradish peroxidase-conjugated donkey antirabbit secondary antibody. Specific immunoreactive bands were detected using the Western lightning chemiluminescent kit (Amersham, Piscataway, NJ). After immunoblot analysis with the anti-phospho-CREB antibody, the membranes were stripped using the Restore Western blot stripping buffer (Pierce, Rockford, IL), reblocked with 4% nonfat dry milk/PBS Tween, and reprobed with anti-CREB antibody for total CREB protein detection. The Un-Scan-It software (version 5.1, Silk Scientific Corp., Orem, UT) was used to quantitate the integrated ODs (IODs) for phospho-CREB, phospho-activating transcription factor (ATF)-1, CREB, and ATF-1 immunoreactive bands. The IOD values were determined from multiple exposure times to obtain values within the linearity of response for each experiment. The ratios of phosphorylated CREB (P-CREB) to CREB (total CREB) and phosphorylated ATF-1 (P-ATF-1) to ATF-1 (total ATF-1) were determined for NE and bound-protein (BND = 0.2 M KCl fraction or SDS-Laemmli buffer-eluted protein) fractions from the DNA-affinity columns. The data (ratios) were analyzed using ANOVA analysis followed by Dunnett’s posttest or the unpaired Student’s t test (GraphPad Prism, GraphPad Software, San Diego, CA). P < 0.05 was considered a statistically significant difference.
ChIP assay
ChIP assays were performed with modifications according to the ChIP assay kit purchased from Upstate Biotechnology. Briefly, MA-10 cells were plated at 2.5 x 106 cells in 60-mm dishes. Thirty-six hours after seeding, cells were treated in the absence or presence of 1 mM (Bu)2cAMP for 30 or 120 min. After treatment, 1% formaldehyde was added and the cells incubated for 10 min at 37 C. Cells were washed two times with ice-cold PBS(+) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml pepstatin A). Cells were collected, pelleted, and resuspended in provided SDS-lysis buffer. Sonication was performed using a 550 Dismembrator sonicator (Fisher Scientific, Pittsburgh, PA) for 20 cycles of 10-sec pulses, and chromatin fragments were determined to be less than 500 bp in size. Sheared chromatin was immunoprecipitated with IgG, anti-CREB, anti-c-Jun, anti-P-CREB, or anti-CBP overnight with rotation at 4 C. Immunocomplexes were collected using salmon sperm/protein A-agarose slurry and subsequently washed one time each with provided low salt, high salt, LiCl, and Tris/EDTA buffers following the manufacturer’s instructions. Protein-DNA complexes were eluted from protein A-agarose beads by addition of elution buffer (1% SDS, 0.1 M NaHCO3) and rotation at room temperature. Formaldehyde cross-links were reversed by the addition of 200 mM NaCl and heating at 65 C for 4 h. DNA was purified using proteinase K treatment followed by phenol extraction and EtOH precipitation. PCR was done using 0.5 μl input chromatin sample and 3 μl immunoprecipitated DNA sample with primer pairs specific for the proximal (–299 bp/–41 bp) [–299 forward: 5'-TGATGCACCTCAGTTACTGG-3'; –41 reverse: 5'-GCTGTGCATCATCACTTGAG] or distal (–3584 bp/–3428 bp) [StAR-3584 forward: 5'-CATACGTGCACTGTCTTAGC-3'; StAR-3428 reverse: 5'-ACTCCTCCA GTAACTCCTTC-3'] regions of the StAR promoter (20). PCR was carried out using Accuprime Pfx supermix (Invitrogen Life Technologies, Carlsbad, CA) at 95 C for 5 min, followed by 39 cycles of 95 C for 20 sec, 57 C for 30 sec, and 68 C for 30 sec. A 2% agarose gel with ethidium bromide was used to separate and examine PCR products.
Results
cAMP treatment increases CREB phosphorylation in MA-10 cells
To identify members of the CREB family expressed in MA-10 cells, a CREB antibody (-CREB) with cross-reactivity to CREB p43, ATF-1, and CREM1 was used for immunoblot analysis of isolated NEs. As shown in Fig. 1A, the -CREB detected CREB (43 kDa), ATF-1 (38 kDa), and CREM (35 kDa) family members (-CREB). Only two immunodetectable bands corresponding to CREB and ATF-1 were detected when the samples were probed using a Ser 133 P-CREB/ATF-1/CREM-specific antibody, indicating that only CREB and ATF-1 are phosphorylated (Fig. 1A, -P-CREB). Analysis of the P-CREB to CREB ratio or P-ATF-1 to ATF-1 ratios indicated that treatment of the cells with (Bu)2cAMP for 30, 60, 120, or 240 min significantly increased P-CREB (43 kDa) 2.3-fold within 30 min of (Bu)2cAMP treatment, and this increase was sustained over the 4-h treatment time (Fig. 1B). However, (Bu)2cAMP treatment had no effect on the P-ATF-1 (38 kDa) levels (Fig. 1C), and a phospho-CREM immunoreactive band was not detected (Fig. 1A).
FIG. 1. (Bu)2cAMP-dependent increase in CREB phosphorylation in MA-10 mouse Leydig tumor cells. MA-10 cells were treated with 1 mM (Bu)2cAMP for the indicated time and NE was isolated. A, Ten micrograms of NE were separated by SDS-PAGE and P-CREB protein was detected by immunoblot analysis as described in Materials and Methods using an anti-P-CREB antibody (-P-CREB). The membrane was stripped and reprobed with an anti-CREB antibody (-CREB). Shown is an immunoblot of NE samples, and the experiment was repeated two additional times using whole-cell lysates. Arrows indicate immunoreactive phosphorylated proteins P-CREB(p43) and P-ATF-1(p38), and total proteins CREB1(p43), ATF-1(p38), and CREM(p35). B and C, The IOD ratios for P-CREB to CREB or P-ATF-1 to ATF-1 were determined as described in Materials and Methods, and shown are the mean values ± SEM from three independent experiments except for the 0- and 120-min time points that have an n = 7. The asterisks indicate a significant difference between treated and control samples based on ANOVA analysis followed by Dunnett’s posttest. *, P < 0.05 and **, P < 0.01.
CREB and ATF-1, but not CREM, bind to the AP-1 element in the CAN region of the StAR promoter
DNA-affinity chromatography was used to examine which CREB family members associate with the CAN region of the StAR promoter. As shown in Fig. 2, only CREB and ATF-1, but not CREM, were detected in the 0.2 M KCl elution fraction from the CAN wild-type affinity column. No CREB/ATF-1 protein was retained by the DNA-affinity column containing point mutations within the AP-1 element (Fig. 2B). These data confirm the importance of the AP-1 site for CREB and ATF-1 binding within this region (17, 18) and suggest that only the CREB family members detected as phosphoproteins (Fig. 1A, -P-CREB) bind to the StAR promoter.
FIG. 2. CREB and ATF-1 associate with the AP-1 half-site within the CAN region of the StAR promoter. DNA-affinity columns using a single copy of the 23-bp wild-type (CANwt) or AP-1 mutant [CANmt (AP-1)] CAN region of the StAR promoter were prepared and column fractions collected as outlined in Materials and Methods. One percent of total input NE, 2% of the FT volume, and 4% of 0.1 M KCl, 0.2 M KCl fractions were separated by SDS-PAGE, and CREB detection was performed as described in Materials and Methods. A and B, CREB family members detected using an anti-CREB antibody (-CREB) show only CREB and ATF-1 recovered in the 0.2 M KCl fraction from CANwt column (A) and no binding to the CANmt column (B). C, Duplicate gel as shown in A except anti-CREM1 antibody (-CREM1) was used for the immunoblot analysis. A–C, Shown are representative blots from two independent experiments using two different NE preparations. A, Similar results were obtained in four additional experiments, each using new NE preparations performed using the batch-binding procedure for DNA-affinity chromatography as described in Materials and Methods. Arrows indicate immunoreactive proteins CREB(p43), ATF-1(p38), and CREM(p35).
Because the anti-CREB antibody used in these studies recognizes CREB, ATF-1, and CREM, we probed the CREB protein profile with an additional antibody (-CREM) that has greater specificity for CREM1 relative to other family members. An immunodetectable band corresponding to the 35-kDa CREM protein was detected in NE (Fig. 2C). No immunoreactive bands were detected in the 0.2 M KCl fraction from the wild-type CAN affinity column using the CREM antibody (Fig. 2C). Together these results indicate that only CREB and ATF-1 bind to the CAN region of the StAR promoter and that the AP-1 element is essential for this interaction.
cAMP treatment increases P-CREB association with the CAN region of the StAR promoter
DNA-affinity chromatography was again used to determine the cAMP effect on CREB-DNA association. NE was isolated from untreated (control) and (Bu)2cAMP-treated (30 or 120 min) MA-10 cells, and the binding of CREB to the CAN wild-type affinity column was determined by immunoblot analysis. No significant difference was observed for total CREB and ATF-1 binding to the CAN region of StAR promoter among control, 30, or 120 min (Bu)2cAMP-treated NE (Fig. 3, A and B, compare BND lanes in panel 2; Fig. 3C). On the other hand, only P-CREB binding to the AP-1 element in the CAN region was significantly increased after (Bu)2cAMP treatment (Fig. 3, A and B, compare BND lanes in panel 1; Fig. 3C). These data indicate that either 30 or 120 min (Bu)2cAMP treatment resulted in a 2-fold increase in P-CREB association with the StAR promoter, although total CREB levels bound to the CAN element remained constant (Fig. 3C).
FIG. 3. The effect of (Bu)2cAMP stimulation on CREB association with the CAN region of the StAR promoter. A and B, Batch-binding DNA-affinity chromatography was performed using control, 30, and 120 min (Bu)2cAMP-treated MA-10 NE as described in Materials and Methods. Collected fractions were separated by SDS-PAGE and P-CREB (-P-CREB) and CREB (-CREB) detected by immunoblot analysis. The indicated lanes contain 10% of input NE, 10% of column FT volume, 20% of 0.1 M KCl wash, and total bound protein eluted with SDS-sample buffer (BND). The membrane was probed with an anti (Ser133)-P-CREB (-P-CREB) antibody and then stripped and reprobed for total CREB using an anti-CREB (-CREB) antibody. A, Shown are representative blots from three (A) or four (B) independent experiments using different NE preparations. Arrows indicate P-CREB(p43) and P-ATF-1(p38), and total CREB(p43), ATF-1(p38), and CREM(p35). C, IOD values for total and P-CREB and P-ATF-1 in the bound fractions (BND), and the ratios for P-CREB to CREB or P-ATF-1 to ATF-1 were determined for control, 30, or 120 min (Bu)2cAMP-treated NE as described in Materials and Methods. The (Bu)2cAMP-treated values were normalized to the control values [(Bu)2cAMP IOD/control IOD], and shown are the mean values ± SEM. An unpaired Student’s t test was used to compare the difference between the (Bu)2cAMP and control samples for total CREB and ATF-1 binding and P-CREB (P-CREB/CREB) and P-ATF-1 (P-ATF-1/ATF-1) binding from three (30 min) or four (120 min) independent experiments. *, P < 0.05 between treated and control samples.
cAMP stimulation increases P-CREB and CBP association with the StAR proximal promoter in vivo
To determine the effect of (Bu)2cAMP treatment of MA-10 cells on CREB and P-CREB association with the StAR promoter under more physiologically relevant conditions, ChIP assays were performed. Cross-linked, sheared chromatin from 30 or 120 min (Bu)2cAMP-treated MA-10 cells was immunoprecipitated with anti-CREB (-CREB) antibodies, and recovered DNA was subjected to PCR using primers specific to the proximal region (–299 bp/–41 bp) of the StAR promoter encompassing the CAN region. As shown in Fig. 4, a 258-bp PCR product specific for StAR proximal promoter was amplified from -CREB-immunoprecipitated DNA samples. In contrast, little or no amplification product was observed with DNA recovered using IgG for immunoprecipitation or when primers for the StAR distal promoter region were used. These data indicate specificity for association of CREB within the proximal region of the StAR promoter. Furthermore, the amount of StAR promoter association with CREB was not increased with either 30- or 120-min (Bu)2cAMP treatment (Fig. 4). These data suggest that total CREB association with the proximal region of the StAR promoter is not qualitatively changed after treatment of the cells with (Bu)2cAMP for 30 or 120 min, consistent with the DNA-affinity chromatography results.
FIG. 4. CREB and c-Jun association with the proximal region of the StAR promoter is not affected by cAMP treatment in MA-10 cells. MA-10 cells were treated in the absence or presence of 1 mM (Bu)2cAMP for 30 or 120 min, and ChIP assay was performed as described in Materials and Methods. Cross-linked, sheared chromatin was immunoprecipitated with IgG, anti-CREB, or anti-c-Jun antibodies. Recovered chromatin was subjected to PCR analysis using primers (–299 bp/–41bp) encompassing the CAN region of the StAR promoter and primers approximately 3.5 kb upstream of the CAN region (–3584 bp/–3428 bp). A, Schematic representation of StAR promoter with location of primer pairs indicated. B and C, Ethidium bromide-stained gel of PCR products. Primers for the proximal and distal region were expected to amplify a 258- and 156-bp fragment, respectively. Shown is a representative ChIP analysis from three independent experiments.
In parallel experiments, c-Jun association with the StAR proximal promoter was also examined by ChIP because recent reports have implicated a functional AP-1 interaction via the same AP-1 element in the CAN region that binds CREB (28). A PCR product specific for StAR proximal promoter, but not the distal promoter region, was amplified from -c-Jun-immunoprecipitated DNA samples. As observed with CREB, 30- or 120-min (Bu)2cAMP-treatment did not change c-Jun interaction with the StAR proximal promoter because similar levels of end point PCR product were amplified from control and treated cells. Together these data reveal that CREB and AP-1 associate with the StAR proximal promoter in intact chromatin in MA-10 cells and that this association is not influenced by (Bu)2cAMP treatment.
On the other hand, in the absence of (Bu)2cAMP-treatment, ChIP analysis using P-CREB or CBP antibodies indicated low to absent association of these factors with the StAR proximal promoter (Fig. 5). However, the presence of the PCR product specific for StAR proximal promoter was observed after 30 min and continuing up to 120 min of (Bu)2cAMP stimulation, indicating an increase in P-CREB and CBP interaction at the StAR promoter. Again, the specificity of association of P-CREB and CBP with the proximal region of the StAR promoter was demonstrated by the absence of PCR product when IgG was used for immunoprecipitation or primers designed to amplify the distal region of the StAR promoter were used. It is important to note that the P-CREB antibody used in these experiments does not distinguish between P-CREB and P-ATF-1. However, the in vitro DNA-affinity chromatography results in Fig. 3 strongly suggest that the increase in phospho-protein association with the StAR promoter is due to P-CREB and not P-ATF-1. In sum, these data suggest that (Bu)2cAMP-treatment promotes an increase in P-CREB and CBP interaction with the StAR proximal promoter.
FIG. 5. Association of P-CREB and CBP with the proximal region of the StAR promoter is increased on cAMP treatment in MA-10 cells. ChIP analysis was performed as described in Materials and Methods on MA-10 cells treated with or without 1 mM (Bu)2cAMP for 30 or 120 min. Cross-linked, sheared chromatin was immunoprecipitated (IP) with either anti-P-CREB or anti-CBP antibodies (CBP), and recovered chromatin was subjected to PCR analysis using primers encompassing the CAN region of the StAR promoter or distal region as shown in Fig. 4A. A and B, Shown is an ethidium bromide-stained gel of PCR products that is representative of ChIP analysis from three (A) or four (B) independent experiments.
Discussion
The transcriptional regulation of the murine StAR gene is complex with numerous factors involved in both the basal and cAMP-mediated effects on StAR transcription (9, 17, 18, 19, 29, 30, 31, 32, 33, 34). These studies show not only the significance of each factor alone but also recently the functional cooperation of these factors and the cis-acting elements that are required for StAR transcription (9, 19, 30, 31, 34). Of these, even in the absence of a canonical CRE, CREB family members have been shown to interact with the StAR promoter, and that their transactivation ability is dependent on phosphorylation (18). We have now examined the extent of endogenous CREB phosphorylation in MA-10 cells after cAMP stimulation in a time-dependent manner and demonstrate that promoter occupancy by P-CREB is required for CBP recruitment.
In the present study, we have identified three immunoreactive CREB bands in MA-10 whole-cell and NEs that correspond to CREB(p43), ATF-1(p38), and CREM(p35) (35, 36, 37). Our data are consistent with previous studies that indicated multiple CREB family members are expressed in MA-10 cells (18). Of the isoforms present in MA-10 cells, we now demonstrate that only CREB and ATF-1 are phosphorylated and that only CREB phosphorylation increased in response to (Bu)2cAMP stimulation. CREB phosphorylation was shown to occur within 30 min of (Bu)2cAMP treatment and remained elevated up to 4 h. It has been reported that the cAMP-PKA signaling cascade in context of CREB phosphorylation occurs via burst-attenuation kinetics in the NIH 3T3 cell line. Forskolin treatment resulted in a rapid increase in PKA-dependent phosphorylation of CREB (Ser133) after 30 min (burst phase) with a subsequent decline after 2 h (attenuation phase) by a protein phosphatase 1-dependent mechanism (38). Our results suggest that this may be tissue or promoter specific because CREB phosphorylation in MA-10 cells was still elevated after 4 h of (Bu)2cAMP stimulation. One possibility for this difference may be that the burst-attenuation kinetics are prolonged in MA-10 Leydig cells. Alternatively, cAMP analogs may mimic intracellular cAMP levels at higher concentrations and for longer time periods in comparison with forskolin treatment, as used in the NIH 3T3 studies.
Using DNA-affinity chromatography, we demonstrate that both CREB and ATF-1, but not CREM, bind to the cAMP-responsive region of the StAR promoter. DNA binding did not appear to be dependent on phosphorylation, although cAMP treatment increased the amount of P-CREB that is associated with the StAR promoter. A limitation of the in vitro DNA-binding assay is that the data do not distinguish between whether DNA-associated CREB is phosphorylated in response to PKA activation or a dynamic model in which (Bu)2cAMP treatment of MA-10 cells increases the overall phosphorylation state of CREB that is then able to associate with the StAR promoter. In contrast, our ChIP data indicate that P-CREB is not significantly associated with the StAR promoter before (Bu)2cAMP treatment, supporting a dynamic model for P-CREB association with StAR promoter in which the increase in nuclear P-CREB after (Bu)2cAMP treatment results in more P-CREB binding to the CAN region of the StAR promoter without increasing total amount of bound CREB.
The lack of CREM binding to the StAR promoter in our study contrasts previous data that indicated CREM as the CREB family member that bound the StAR promoter based on protein-DNA complex supershift EMSA (18). The antibody source is different between these two studies and may attribute to the differences because we were unable to demonstrate CREB or CREM binding to the CAN region by EMSA analysis using our antibodies (Ref.17 , and data not shown). Nevertheless, using DNA-affinity chromatography, which provides a more powerful molecular tool for examining protein-DNA interactions by allowing immunoblot protein detection and direct estimation of bound protein molecular weights, we demonstrated that CREB and ATF-1, but not CREM, bind to the StAR promoter.
CREB binding is dependent on the AP-1-like element in the cAMP-responsive region of StAR promoter. Our previous data suggest that this site is essential for basal activity of the StAR promoter (17, 18). However, mutation of the AP-1 site in context of StAR luciferase reporter constructs had no significant effect on the fold increase in StAR transcription after (Bu)2cAMP treatment, even though the overall strength of the promoter was diminished (17). The lack of effect on the cAMP response may be attributed to protein binding to surrounding functional cis-acting elements (SF-1 and/or GATA-4), which contribute to the overall stimulated response and therefore may stabilize the cAMP-dependent complex even in the presence of the AP-1 mutation (17, 18). Because the affinity columns contained only 23 bp of the previously reported CAN region and lacked the SF-1 and GATA-4 elements (17), future studies can be focused to investigate whether CREB association with the cAMP-responsive region would be observed in the presence of the AP-1 mutation and intact SF-1 and GATA-4 elements.
CREB and P-CREB association with the StAR proximal promoter was confirmed using the more physiological relevant ChIP approach. We observed no qualitative difference in CREB interaction with the StAR promoter between control and 30 or 120 min (Bu)2cAMP treatment. These data are also consistent with a recent report on protein-DNA interactions with the StAR promoter using ChIP analysis that showed CREB association with the proximal region of the StAR promoter was increased 50% after 30 min of 8-bromoadenosine-cAMP treatment but was comparable with control levels after 2 h, suggesting a temporal association of CREB with the StAR promoter (21). The discrepancy between our data and the previous study at the 30-min time point may be due to difference in detection used for the ChIP analysis between real-time PCR vs. end point ethidium bromide staining. Possibly the increased phosphorylation of CREB at this short time point may contribute to the previous observed increase in total CREB association using ChIP. However, we show CREB phosphorylation remains elevated for an extended time; therefore, the apparent time-dependent association most likely reflects other factors in addition to CREB phosphorylation.
ChIP analysis was also used to investigate the classical cAMP-PKA signaling cascade leading to CREB phosphorylation and CBP recruitment to StAR proximal promoter in MA-10 cells. Our results demonstrate an increase in P-CREB association with the StAR proximal promoter after 30 or 120 min treatment with (Bu)2cAMP, compared with control cells, again consistent with the in vitro DNA-affinity chromatography studies. The results of the ChIP analysis for CBP association with the StAR proximal promoter paralleled the data with P-CREB in that a significant association was observed only after (Bu)2cAMP treatment. Recently ChIP analysis on the cAMP-dependent temporal association of SF-1, GATA-4, CREB, and CBP with the murine StAR demonstrated that alterations in histone modifications (acetylation and methylation) paralleled CBP association with the mouse StAR proximal promoter in response to 8-bromoadenosine-cAMP treatment. These studies support the premise that the activity of the coactivator CBP contributes to a more transcriptionally active chromatin structure (20, 21). Our data support these results and extend these findings to demonstrate for the first time the physical role of phosphorylated CREB-DNA interactions in CBP recruitment and the cAMP-stimulation of StAR transcription.
In addition to CREB binding to the AP-1-like element within the CAN region of StAR promoter, recent studies have shown that overexpression of the AP-1 family member c-Jun increased StAR reporter gene activity in transiently transfected MA-10 and Y1 cells (28). The transcriptional activation by c-Jun was dependent on the AP-1 element, although no direct binding of c-Jun to the StAR promoter was detected by EMSA. The authors suggested that c-Jun may associate with the StAR cAMP-responsive region, possibly through a tethering mechanism with c-fos or Fra-2 (28). Therefore, we tested for a c-Jun-StAR proximal promoter interaction in vivo by ChIP. Our data confirm that c-Jun associates with the murine StAR proximal promoter and that the association remains constant in response to 30 or 120 min (Bu)2cAMP treatment, suggesting c-Jun, as well as CREB, has the ability to constitutively bind to the proximal StAR promoter. However, we cannot conclude that the CREB and c-Jun associations observed in our ChIP analysis are from the same chromatin templates or represent binding to the same element. Although the functional relevance for CREB and c-Jun individually on StAR promoter activity had previously been reported (18, 19, 28), our ChIP analysis would support a model that both CREB and AP-1 associate with the StAR proximal promoter under basal conditions without disrupting the cAMP-PKA transactivation as defined by the simultaneous association of c-Jun, P-CREB and CBP with StAR promoter. Although we cannot exclude the possibility that c-Jun association with StAR promoter may also enhance CBP recruitment with (Bu)2cAMP treatment, the finding that CREB phosphorylation was increased by (Bu)2cAMP treatment and P-CREB and CBP promoter association in vivo occurred concurrently and only after (Bu)2cAMP treatment support P-CREB as functionally relevant in coactivator recruitment.
In sum, our data indicate that the classical cAMP-PKA-mediated CREB phosphorylation and CBP recruitment signaling cascade is involved in the cAMP-stimulated response of the murine StAR gene. CREB is constitutively bound to the AP-1 site within the cAMP-responsive region, and cAMP-treatment enhances the P-CREB to CREB ratio within the nucleus, thereby allowing for an increase in P-CREB interaction at the AP-1 cis-acting element. Previous studies implicated CREB family members and the importance of CREB phosphorylation in cAMP-dependent StAR transcription and CBP association with the StAR promoter (18, 21). Our data provide the link between these observations and show that StAR promoter occupancy by CBP occurs concurrently with P-CREB and not CREB or c-Jun. These data are consistent with the model for enhanced P-CREB association with the AP-1 element that results in recruitment of CBP, thereby providing a mechanism for histone acetylation and increased StAR transcription.
Acknowledgments
The authors thank Terry Wright for his contribution to the ChIP analysis and Drs. Carolyn Klinge and Doug Darling (University of Louisville, Louisville, KY) for their review of this manuscript.
References
Parker KL, Schimmer BP 1995 Transcriptional regulation of the genes encoding the cytochrome P-450 steroid hydroxylases. Vitam Horm 51:339–370
Jefcoate CR, DiBartolomeis MJ, Williams CA, McNamara BC 1987 ACTH regulation of cholesterol movement in isolated adrenal cells. J Steroid Biochem 27:721–729
Ohno Y, Yanagibashi K, Yonezawa Y, Ishiwatari S, Matsuba M 1983 A possible role of "steroidogenic factor" in the corticoidogenic response to ACTH; effect of ACTH, cycloheximide and aminoglutethimide on the content of cholesterol in the outer and inner mitochondrial membrane of rat adrenal cortex. Endocrinol Jpn 30:335–338
Simpson ER, Boyd GS 1967 The cholesterol side-chain cleavage system of bovine adrenal cortex. Eur J Biochem 2:275–285
Privalle CT, Crivello JF, Jefcoate CR 1983 Regulation of intramitochondrial cholesterol transfer to side-chain cleavage cytochrome P-450 in rat adrenal gland. Proc Natl Acad Sci USA 80:702–706
Stocco DM, Clark BJ 1996 Regulation of the acute production of steroids in steroidogenic cells. Endocr Rev 17:221–244
Crivello JF, Jefcoate CR 1980 Intracellular movement of cholesterol in rat adrenal cells. Kinetics and effects of inhibitors. J Biol Chem 255:8144–8151[Free Full Text]
Clark BJ, Soo SC, Caron KM, Ikeda Y, Parker KL, Stocco DM 1995 Hormonal and developmental regulation of the steroidogenic acute regulatory protein. Mol Endocrinol 9:1346–1355
Sugawara T, Saito M, Fujimoto S 2000 Sp1 and SF-1 interact and cooperate in the regulation of human steroidogenic acute regulatory protein gene expression. Endocrinology 141:2895–2903
Clark BJ, Combs R 1999 Angiotensin II and cyclic adenosine 3',5'-monophosphate induce human steroidogenic acute regulatory protein transcription through a common steroidogenic factor-1 element. Endocrinology 140:4390–4398
Sandhoff TW, Hales DB, Hales KH, McLean MP 1998 Transcriptional regulation of the rat steroidogenic acute regulatory protein gene by steroidogenic factor 1. Endocrinology 139:4820–4831
Rust W, Stedronsky K, Tillmann G, Morley S, Walther N, Ivell R 1998 The role of SF-1/Ad4BP in the control of the bovine gene for the steroidogenic acute regulatory (StAR) protein. J Mol Endocrinol 21:189–200
Sugawara T, Kiriakidou M, McAllister JM, Kallen CB, Strauss 3rd JF 1997 Multiple steroidogenic factor 1 binding elements in the human steroidogenic acute regulatory protein gene 5'-flanking region are required for maximal promoter activity and cyclic AMP responsiveness. Biochemistry 36:7249–7255
Caron KM, Ikeda Y, Soo SC, Stocco DM, Parker KL, Clark BJ 1997 Characterization of the promoter region of the mouse gene encoding the steroidogenic acute regulatory protein. Mol Endocrinol 11:138–147
De Cesare D, Sassone-Corsi P 2000 Transcriptional regulation by cyclic AMP-responsive factors. Prog Nucleic Acids Res Mol Biol 64:343–369
Montminy M 1997 Transcriptional regulation by cyclic AMP. Annu Rev Biochem 66:807–822
Wooton-Kee CR, Clark BJ 2000 Steroidogenic factor-1 influences protein-deoxyribonucleic acid interactions within the cyclic adenosine 3,5-monophosphate-responsive regions of the murine steroidogenic acute regulatory protein gene. Endocrinology 141:1345–1355
Manna PR, Dyson MT, Eubank DW, Clark BJ, Lalli E, Sassone-Corsi P, Zeleznik AJ, Stocco DM 2002 Regulation of steroidogenesis and the steroidogenic acute regulatory protein by a member of the cAMP response-element binding protein family. Mol Endocrinol 16:184–199
Manna PR, Eubank DW, Lalli E, Sassone-Corsi P, Stocco DM 2003 Transcriptional regulation of the mouse steroidogenic acute regulatory protein gene by the cAMP response-element binding protein and steroidogenic factor 1. J Mol Endocrinol 30:381–397
Christenson LK, Stouffer RL, Strauss 3rd JF 2001 Quantitative analysis of the hormone-induced hyperacetylation of histone H3 associated with the steroidogenic acute regulatory protein gene promoter. J Biol Chem 276:27392–27399
Hiroi H, Christenson LK, Chang L, Sammel MD, Berger SL, Strauss 3rd JF 2004 Temporal and spatial changes in transcription factor binding and histone modifications at the steroidogenic acute regulatory protein (stAR) locus associated with stAR transcription. Mol Endocrinol 18:791–806
Weih F, Stewart AF, Boshart M, Nitsch D, Schutz G 1990 In vivo monitoring of a cAMP-stimulated DNA-binding activity. Genes Dev 4:1437–1449
Wolfl S, Martinez C, Majzoub JA 1999 Inducible binding of cyclic adenosine 3',5'-monophosphate (cAMP)-responsive element binding protein (CREB) to a cAMP-responsive promoter in vivo. Mol Endocrinol 13:659–669
Sibinga NE, Wang H, Perrella MA, Endege WO, Patterson C, Yoshizumi M, Haber E, Lee ME 1999 Interferon--mediated inhibition of cyclin A gene transcription is independent of individual cis-acting elements in the cyclin A promoter. J Biol Chem 274:12139–12146
Tommasi S, Pfeifer GP 1999 In vivo structure of two divergent promoters at the human PCNA locus. Synthesis of antisense RNA and S phase-dependent binding of E2F complexes in intron 1. J Biol Chem 274:27829–27838
Nichols M, Weih F, Schmid W, DeVack C, Kowenz-Leutz E, Luckow B, Boshart M, Schutz G 1992 Phosphorylation of CREB affects its binding to high and low affinity sites: implications for cAMP induced gene transcription. EMBO J 11:3337–3346
Dey A, Nebert DW, Ozato K 1991 The AP-1 site and the cAMP- and serum response elements of the c-fos gene are constitutively occupied in vivo. DNA Cell Biol 10:537–544
Manna PR, Eubank DW, Stocco DM 2004 Assessment of the role of activator protein-1 on transcription of the mouse steroidogenic acute regulatory protein gene. Mol Endocrinol 18:558–573
Sugawara T, Kiriakidou M, McAllister JM, Holt JA, Arakane F, Strauss 3rd JF 1997 Regulation of expression of the steroidogenic acute regulatory protein (StAR) gene: a central role for steroidogenic factor 1. Steroids 62:5–9
Reinhart AJ, Williams SC, Clark BJ, Stocco DM 1999 SF-1 (steroidogenic factor-1) and C/EBP? (CCAAT/enhancer binding protein-?) cooperate to regulate the murine StAR (steroidogenic acute regulatory) promoter. Mol Endocrinol 13:729–741
Silverman E, Eimerl S, Orly J 1999 CCAAT enhancer-binding protein ? and GATA-4 binding regions within the promoter of the steroidogenic acute regulatory protein (StAR) gene are required for transcription in rat ovarian cells. J Biol Chem 274:17987–17996
Christenson LK, Johnson PF, McAllister JM, Strauss 3rd JF 1999 CCAAT/enhancer-binding proteins regulate expression of the human steroidogenic acute regulatory protein (StAR) gene. J Biol Chem 274:26591–26598
Tremblay JJ, Viger RS 2001 GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory elements. Endocrinology 142:977–986
Tremblay JJ, Hamel F, Viger RS 2002 Protein kinase A-dependent cooperation between GATA and CCAAT/enhancer-binding protein transcription factors regulates steroidogenic acute regulatory protein promoter activity. Endocrinology 143:3935–3945
Montminy MR, Bilezikjian LM 1987 Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature 328:175–178
Liu F, Thompson MA, Wagner S, Greenberg ME, Green MR 1993 Activating transcription factor-1 can mediate Ca(2+)- and cAMP-inducible transcriptional activation. J Biol Chem 268:6714–6720
Behr R, Weinbauer GF 1999 Germ cell-specific cyclic adenosine 3',5'-monophosphate response element modulator expression in rodent and primate testis is maintained despite gonadotropin deficiency. Endocrinology 140:2746–2754
Canettieri G, Morantte I, Guzman E, Asahara H, Herzig S, Anderson SD, Yates 3rd JR, Montminy M 2003 Attenuation of a phosphorylation-dependent activator by an HDAC-PP1 complex. Nat Struct Biol 10:175–181(Brian F. Clem, Elizabeth )