Molecular Endocrinology, doi:10.1210/me.2005-0215
Molecular Endocrinology 20 (1): 147-166
Copyright © 2006 by The Endocrine Society
Adrenocorticotropic Hormone-Mediated Signaling Cascades Coordinate a Cyclic Pattern of Steroidogenic Factor 1-Dependent Transcriptional Activation
Jonathon N. Winnay and
Gary D. Hammer
Department of Molecular and Integrative Physiology (J.N.W., G.D.H.) and Department of Internal Medicine (G.D.H.), Division of Metabolism, Endocrinology and Diabetes, University of Michigan Medical School, Ann Arbor, Michigan 48109-0678
Address all correspondence and requests for reprints to: Gary D. Hammer, M.D., Ph.D., 1150 West Medical Center Drive, Office 5560A MSRB II, Ann Arbor, Michigan 48109-0678. E-mail: ghammer{at}umich.edu.
 |
ABSTRACT
|
|---|
Steroidogenic factor 1 (SF-1) is an orphan nuclear receptor that has emerged as a critical mediator of endocrine function at multiple levels of the hypothalamic-pituitary-steroidogenic axis. Within the adrenal cortex, ACTH-dependent transcriptional responses, including transcriptional activation of several key steroidogenic enzymes within the steroid biosynthetic pathway, are largely dependent upon SF-1 action. The absence of a bona fide endogenous eukaryotic ligand for SF-1 suggests that signaling pathway activation downstream of the melanocortin 2 receptor (Mc2r) modulates this transcriptional response. We have used the chromatin immunoprecipitation assay to examine the temporal formation of ACTH-dependent transcription complexes on the Mc2r gene promoter. In parallel, ACTH-dependent signaling events were examined in an attempt to correlate transcriptional events with the upstream activation of signaling pathways. Our results demonstrate that ACTH-dependent signaling cascades modulate the temporal dynamics of SF-1-dependent complex assembly on the Mc2r promoter. Strikingly, the pattern of SF-1 recruitment and the subsequent attainment of active rounds of transcription support a kinetic model of SF-1 transcriptional activation, a model originally established in the context of ligand-dependent transcription by several classical nuclear hormone receptors. An assessment of the major ACTH-dependent signaling pathways highlights pivotal roles for the MAPK as well as the cAMP-dependent protein kinase A pathway in the entrainment of SF-1-mediated transcriptional events. In addition, the current study demonstrates that specific enzymatic activities are capable of regulating distinct facets of a highly ordered transcriptional response.
|
INTRODUCTION
|
|---|
THE ORPHAN NUCLEAR hormone receptor SF-1 (steroidogenic factor 1) was originally identified as the key regulator of the tissue-specific expression of the cytochrome P450 steroid hydroxylases (1, 2, 3). Subsequently, the use of gene-targeting approaches to produce SF-1-deficient mice or specifically ablate SF-1 in a tissue-specific manner has substantiated a role for SF-1 in both adrenal steroidogenesis and organogenesis (4, 5, 6, 7, 8). Consistent with its role as a regulator of steroidogenic enzyme gene expression, SF-1 has been observed in a variety of steroidogenic tissues including the adrenal cortex, ovarian theca and granulosa cells, as well as testicular Leydig cells (5). SF-1 participates in the regulation of many major steroidogenic enzymes including cytochrome P450 side-chain cleavage enzyme (9, 10, 11), 17
-hydroxylase (P450c17) (12, 13, 14), 3ß-hydroxysteroid dehydrogenase (15), 21-hydroxylase (16, 17), 11ß-hydroxylase (3) as well as several additional genes involved in the steroid biosynthetic pathway including the melanocortin 2 receptor (Mc2r) (18, 19, 20) and the steroidogenic acute regulatory protein (21, 22), both early determinants of steroidogenic potential.
Nuclear receptors comprise a large family of related transcription factors that share a conserved structure and mechanism of activation. Classically, steroid receptors are DNA-dependent transcription factors that are regulated by small lipophilic molecules including adrenal steroids (mineralocorticoids and glucocorticoids), vitamin D3, retinoids (all-trans and 9-cis), gonadal and placental steroids (androgens, progestins, and estrogens), as well as thyroid hormone (23). In addition to these well-characterized hormone receptors, there are a significant number of receptors, referred to as orphan receptors, for which no regulatory ligand has been identified (24). Structural analysis of ligand-binding domains (LBDs) of classical as well as orphan nuclear receptors has provided insight into novel mechanisms by which some orphan receptors adopt active conformations similar to those observed for ligand-bound receptors. The presence of empty LBDs (liver receptor homolog-1, retinoid-related orphan receptor ß 
, estrogen-related receptor ß ), constitutive binding of ubiquitous structural ligand (hepatocyte nuclear factor-), or the complete absence of an LBD (nuclear growth factor I-B, Nor1, Nurr1) suggests that the regulation of some orphan receptors in response to physiological cues, particularly those involved in endocrine homeostasis, may be conferred by acting as downstream effectors of peptide hormone-mediated signaling cascades downstream of cell surface receptors (23, 25, 26). In support of this view, several posttranslational modifications of SF-1 have been observed, including phosphorylation, sumoylation, and acetylation, which regulate diverse aspects of SF-1 function (27, 28, 29, 30, 31, 32).
The proopiomelanocortin-derived peptide ACTH represents the primary hormone responsible for activation of steroid biosynthesis in the adrenal cortex (33). The binding of ACTH to its cognate G protein-coupled receptor promotes the activation of protein kinase A (PKA) and MAPK-dependent signaling cascades that collectively initiate adrenal-specific, steroidogenic transcriptional programs (27, 28, 34, 35, 36, 37). The pharmacological inhibition of either pathway is capable of inhibiting the ACTH-dependent transcriptional response, suggesting a critical role for PKA and MAPK in ACTH-dependent signaling (38, 39). Despite active investigation attempting to identify the downstream signaling events that couple PKA activation with steroidogenic enzyme gene transcription, no adequate mechanism has been proposed that would account for this phenomenon. Studies examining the ACTH-dependent MAPK pathway have implicated the downstream kinases, ERK 1/2, in the regulation of SF-1 phosphorylation (27, 28). Interestingly, the phosphorylation of SF-1 leads to enhanced transcriptional activation of target genes, presumably the effect of phosphorylation-induced association of cofactors with the LBD of SF-1. However, the dephosphorylation of SF-1 through the actions of the dual-specificity, mitogen-activated kinase phosphatase-1 (MKP-1) has also been proposed to activate SF-1 via MKP-1-mediated dephosphorylation of ERK 1/2 (40). Specifically, ACTH has been shown to elicit a robust increase in MKP-1 transcript and protein expression as well as the dephosphorylation of SF-1 and subsequent transcriptional activation of P450c17 in mouse Y1 adrenocortical cells (40, 41). Taken together, these potentially disparate conclusions suggest that context-specific phosphorylation and dephosphorylation of SF-1 both serve as positive influences on target gene transcription. Moreover, a temporal coordination of the phosphorylation and dephosphorylation of SF-1 may together contribute to the dynamic assembly of SF-1 complexes and ultimate transcriptional output.
The transcriptional activation of genes by nuclear hormone receptors requires a coordinated, temporal recruitment of a variety of factors concomitant with alterations of the local chromatin environment, which involves posttranslational alterations of core histones and ATP-dependent nucleosome remodeling (42). SF-1 has been shown to physically and/or functionally interact with a number of transcription factors and transcriptional cofactors on adrenal-specific promoters including ubiquitously expressed transcription factors (transcription factor IIB, surfactant protein 1, CCAAT enhancer binding protein-ß, nuclear factor 1, and Jun) (43, 44), general coactivators [p160/steroid receptor coactivator (SRC) family, receptor-interacting protein 140, cAMP response element-binding protein (CREB)-binding protein, and p300] (39, 45, 46, 47), and the adrenal/gonad-specific transcriptional corepressor, Dax-1(48). To date, the participation of these factors in SF-1-mediated transcription has largely been inferred from in vitro binding assays as well as studies utilizing reporter constructs that cannot faithfully recapitulate the in vivo transcriptional regulation of SF-1 target-genes within the endogenous, chromatinized environment.
In this study, we have performed chromatin immunoprecipitation (ChIP) assays over a 3-h time course after ACTH stimulation to determine the temporal dynamics of regulatory factor recruitment and transcriptional activation of the proximal region of the Mc2r gene promoter that is responsible for conferring ACTH responsiveness (49, 50). In an attempt to correlate the temporal activation of upstream signaling events with recruitment and/or occupancy of factors with the endogenous promoter, we have examined the activation of the PKA and MAPK-dependent pathways and addressed their relative contributions in SF-1-mediated transcriptional activation through the use of specific pharmacological inhibitors of both signaling cascades. In addition, we have examined the global role of histone deacetylases (HDACs) and serine/threonine phosphatases in the transcriptional activation of the Mc2r gene promoter. These studies have revealed that distinct enzymatic activities regulated downstream of ACTH-dependent pathways are capable of drastically altering the amplitude and frequency of transcriptional activation, thereby establishing the overall tone of the transcriptional response.
|
RESULTS
|
|---|
ACTH Stimulation Acutely Activates Mc2r Gene Promoter
To assess the assembly of SF-1 transcription complexes, we developed a ChIP assay to detect changes in promoter occupancy and transcriptional activation of ACTH-dependent target genes. The Y1 adrenocortical cell line was used as these cells are ACTH responsive and express a number of SF-1-dependent target genes including the Mc2r (38, 50, 51, 52, 53). Y1 cells were cultured in serum-free media for 48 h, synchronized by incubation in -amanitin-containing media followed by treatment with or without ACTH. We first assessed the activation of Mc2r gene transcription by performing ChIP assays at baseline and after ACTH stimulation using antibodies recognizing the phosphorylated heptad repeat (YSPTSPS) within the C-terminal domain of RNA polymerase II (RNA Pol II) as a surrogate for transcriptional initiation (54, 55). These experiments revealed a dramatic increase in phospho-RNA Pol II occupancy of the Mc2r proximal promoter region after 80 min of ACTH treatment (Fig. 1
). To validate the observed results, the same experimental samples were subjected to PCR amplification using a primer pair located approximately 6 kb from the transcriptional start site (distal Mc2r) and nonspecific primers amplifying a region of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene promoter. As expected, PCR performed with both primer pairs amplified products from genomic DNA, but failed to amplify products in experimental samples.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1. Effects of ACTH Stimulation on Mc2r Promoter Activation
Diagram of the Mc2r promoter including position of proximal and distal primer pairs used for ChIP analysis and a variety of transcription factors predicted to bind response elements within the proximal promoter region using MatInspector (www.Genomatix.de) (77 ). Y1 cells were serum deprived for 48 h, synchronized with -amanitin followed by stimulation with or without ACTH (1 x 10–8 M) (A). Genomic fragments immunoprecipitated with -phospho RNA Pol II antibodies from Y1 cells untreated or treated with ACTH for 80 min were PCR amplified using proximal primers encompassing the SF-1 RE/cAMP-responsive site, a distal region of the Mc2r promoter, and primers located in the nonspecific GAPDH promoter followed by agarose gel electrophoresis. Genomic DNA served as a positive control for PCR amplification. GR, Glucocorticoid receptor; AP-1, activator protein 1; Sp1, trans-acting transcription factor-1.
|
|
ACTH Elicits a Cyclic Transcriptional Response
Having established that activation of ACTH-dependent signaling cascades leads to transcriptional initiation, we next sought to determine whether ACTH stimulation would elicit a dynamic pattern of SF-1-dependent transcriptional activation similar to that observed for several ligand-dependent, classical nuclear hormone receptors (56, 57, 58). Using real-time PCR to quantify promoter occupancy, we first performed ChIP assays over an extended time course using -acetyl H4 antibodies to precipitate genomic fragments. Interestingly, two transient yet distinct periods of H4 hyperacetylation were observed with peak acetylation occurring at the 40- and 120-min time points (Fig. 2A). To verify that histone H4 hyperacetylation correlated with active transcription and an overall accumulation of Mc2r gene transcript, quantitative RT-PCR was performed at the 0- and 180-min time points. As expected, ACTH elicited a significant increase in Mc2r transcript levels (1.00 ± 0.073 vs. 1.32 ± 0.034; fold of baseline; P < 0.005), suggesting that productive rounds of transcriptional activation occur during this initial period after ACTH stimulation (Fig. 2B, left panel). As anticipated, each round of histone acetylation was followed by maximally active periods of transcriptional initiation at the 80- and 160-min time points as assessed by phospho-RNA Pol II promoter occupancy (Fig. 2B, right panel). Similar patterns of dynamic recruitment were also observed on the steroidogenic acute regulatory protein and P450scc promoters after ACTH stimulation (data not shown), suggesting a common mechanism regulating the ACTH-dependent transcriptional activation of several SF-1-dependent target genes. In contrast, real-time PCR analysis of the same samples using promoter-specific primers for the ACTH-independent GAPDH gene promoter demonstrate the relative absence of phospho-RNA Pol II occupancy, thereby confirming the specificity of the observed results (Fig. 2C). Collectively, these data suggest that the activation of ACTH-dependent signaling cascades is capable of evoking a cyclic transcriptional response similar to that observed after ligand binding and activation of classic steroid receptors.
View larger version (33K):
[in this window]
[in a new window]
|
Fig. 2. Kinetics of ACTH-Dependent Activation of the Mc2r Gene
-Amanitin-synchronized Y1 adrenocortical cells were serum deprived for 48 h followed by ACTH treatment (1 x 10–8 M) over the indicated time course. A, Real-time PCR quantitation of histone H4 acetylation of the Mc2r proximal promoter as a function of time. B, Real-time PCR was used to assess phospho-RNA Pol II occupancy of the Mc2r gene promoter. Data are normalized to values obtained for 1% input controls and results are presnted as percent of baseline values (left panel). Mc2r transcript levels were determined by quantitative real-time PCR. GAPDH primers were used to normalize data, and results are expressed as fold of wild-type values (right panel). C, Real-time PCR was performed using GAPDH promoter-specific primers over the indicated time-course. Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline values. ChIP experiments were performed a minimum of two times with similar results. pPol II, Phospho-RNA Pol II; Acetyl., acetylation.
|
|
Establishment and Characterization of Stable Y1 Cell Lines Expressing Epitope Tagged SF-1
Although a number of studies have delineated the requirements governing SF-1 binding to response elements found within target genes, the recruitment of SF-1 to endogenous promoters has not been thoroughly assessed (59, 60, 61, 62). To examine whether SF-1 is actively recruited to the endogenous Mc2r promoter in response to ACTH stimulation, Y1 stable cell lines expressing an SF-1 fusion protein with tandem myc and hemagglutinin (HA) epitope tags were generated. These Y1 cell lines were established to provide additional antigens for immunoprecipitation with -HA or -myc antibodies. Epitope fusion protein was readily observed in stable cell lines using -myc (Fig. 3A) or -HA antibodies (data not shown). To establish that the epitope-tagged SF-1 fusion protein retained function, luciferase assays were performed in the presence of a mouse Mc2r-luc (–1 kb) reporter and increasing amounts of MycHA-SF-1 expression plasmid in the SF-1-deficient cell line, JEG 3 (Fig. 3B). Transfection of the reporter plasmid alone did not lead to significant activation. In contrast, a dose-dependent increase in MycHA-SF-1 expression resulted in a concomitant increase in luciferase activity. Using this cell line, the pattern of SF-1 recruitment after ACTH stimulation was assessed by performing ChIP assays using -HA antibodies. Interestingly, two distinct periods of maximal SF-1 recruitment were observed at the 40- and 120-min time points, consistent with the transient association of SF-1 with the Mc2r promoter (Fig. 3C). As anticipated, a similar pattern of SF-1 recruitment was observed using -myc antibodies (data not shown).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 3. Creation and Characterization of Stable Y1 Cell Lines Expressing Epitope-Tagged SF-1
A, Immunoblot analysis of mycHA-tagged SF-1 stable Y1 cell line and B, luciferase assays using Mc2r-luc and increasing amounts of pCMVtag2-mycHASF-1 in the SF-1-negative cell line JEG-3. Data represents the mean ± SEM of three independent experiments. C, Real-time quantitation of Mc2r proximal promoter fragments from mycHA-tagged SF-1 stable cell lines immunoprecipitated with -HA antibodies as a function of time. Data are normalized to values obtained for 1% input controls and results are presented as percent of baseline values. IB, Immunoblot; acetyl., acetylation.
|
|
Endogenous SF-1 Is Actively Recruited to the Mc2r Promoter in an ACTH-Dependent Manner
We next examined the recruitment of endogenous SF-1 in native Y1 cells by performing ChIP assays using SF-1-specific antibodies. Again, a similar pattern of recruitment was observed over the time course, confirming the temporal recruitment pattern observed using the epitope-tagged SF-1 cell line (Fig. 4A). Lastly, the recruitment of SF-1 to the Mc2r promoter in an in vivo setting was assessed. Wild-type C57BL/6J mice were administered dexamethasone (2 µg/g body weight) at 2300 h followed by stimulation with ACTH (1 µg/g body weight) 12 h later. Adrenals were removed at the indicated time points, soluble chromatin extracts were prepared, and ChIP assays were performed using -SF-1 antibodies. Once again, SF-1 was actively recruited to the proximal promoter with maximal occupancy at the 60-min time point (Fig. 4b). The association of SF-1 with the proximal promoter returned to baseline levels by the 90-min time point; however, these results may reflect diminished serum ACTH levels due to hormone clearance rather than the regulated removal of SF-1 from the endogenous Mc2r promoter. Collectively, these data demonstrate that SF-1 is actively recruited to the Mc2r promoter within the native chromatinized environment.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 4. ACTH-Stimulated Association of SF-1 with the Proximal Mc2r Promoter
A, Real-time PCR quantitation of ChIP assays performed using -SF-1 antibodies to precipitate endogenous SF-1 and associated genomic fragments from Y1 cells as a function of time. B, Real-time quantitation of ChIP assays performed using chromatinized template derived from dexamethasone-suppressed mice (5 µg/g body weight) after ACTH stimulation (1 µg/g body weight) for the indicated times. SF-1 antibodies were used to precipitate associated genomic fragments. Data were normalized to values obtained for 1% input controls. Data are representative of pooled adrenals from three animals for each time point. acetyl., Acetylation.
|
|
ACTH Stimulation Leads to the Concerted Recruitment of SF-1 Coactivators
Having established that ACTH stimulation leads to the active recruitment of SF-1, acetylation of core histones, and subsequent rounds of active transcription, we assessed the timing of transcription cofactor recruitment and attempted to correlate these events with the patterns of histone acetylation and transcriptional activation observed over the extended time course. Among the coactivator proteins implicated in SF-1-mediated transcription are three distinct members of the p160 family, SRC-1, SRC-2, and SRC-3 (also referred to as: SRC-1, nuclear receptor coactivator 1; SRC-2, glucocorticoid receptor-interacting protein 1 or transcriptional intermediary factor 2; SRC-3, amplified in breast cancer 1, receptor-associated coactivator 3, nuclear receptor coactivator 3, or thyroid hormone receptor activator molecule 1) (63, 64, 65, 66). Using specific antibodies against the SRC proteins to precipitate promoter fragments, a rapid and transient cycling of these factors on the Mc2r promoter was observed (Fig. 5, A–C). Interestingly, the dynamics of SRC occupancy for each family member differed slightly after the initial recruitment event. For example, the second round of SRC-1 recruitment occurred later than the patterns observed for SRC-2 and SRC-3. Additionally, SRC-2 recruitment exhibited a moderate recruitment in the first round of transcription, with a more prominent second round of occupancy. In contrast, although the first recruitment of SRC-3 was robust, subsequent recruitment was diminished, potentially indicating a more pronounced role for SRC-3 in the initial activation events. These data support a specific role for each SRC family member in the transcriptional activation of nuclear receptor target genes.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5. Dynamics of Cofactor Recruitment to the Mc2R Promoter in Response to ACTH Stimulation
Kinetic ChIP experiments were performed using specific antibodies against SRC-1 (A), SRC-2 (B), SRC-3 (C), and GCN5 (D). Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline values as a function of time. ChIP experiments were performed a minimum of two times with similar results.
|
|
Lastly, we examined Mc2r promoter occupancy by GCN5 (general control of amino acid synthesis 5-like 1) over the time course and observed recruitment during each transcription cycle with occupancy increasing approximately 90-fold in the second recruitment period (Fig. 5D
). As anticipated, SF-1 recruitment preceded the recruitment of the SRC family members and GCN5, suggesting that the recruitment of SF-1 is required as an initial step in complex formation. Furthermore, the clearance of SF-1 from the promoter coincided with the reduced occupancy of coactivators on the Mc2r promoter, further supporting the contention that the binding of SF-1 to DNA supports secondary cofactor recruitment and establishment of a transcriptionally competent promoter. Although the timing of SF-1 and GCN5 promoter occupancy were observed to temporally overlap, only an inferred association between these coactivators and SF-1 on the Mc2r gene promoter could be supported. Therefore, we performed re-ChIP experiments to directly examine whether SF-1 and one of these coregulatory proteins, GCN5, existed in separate or shared complexes on the Mc2r gene promoter. As expected, stimulation of Y1 cells with vehicle or ACTH for 40 min promoted the recruitment of both SF-1 and GCN5 to the Mc2r promoter as previously observed (Fig. 6, lanes 1–4). Interestingly, when SF-1 or GCN5 primary immunoprecipitations were eluted and immunoprecipitated with reciprocal antibodies (Fig. 6, lanes 5–8), an ACTH-dependent interaction between SF-1 and GCN5 was readily observed, demonstrating the coexistence of SF-1 and a known coregulatory factor in an ACTH-responsive, promoter-bound complex.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6. Re-ChIP Experiments Reveal That SF-1 and GCN5 Coexist in the Same Complex(es)
Re-ChIP experiments were performed to determine whether SF-1 and a general coactivator (GCN5) associated within a complex on the Mc2r promoter. Soluble chromatin was prepared from Y1 adrenocortical cells treated with ACTH (1 x 10–8 M) or vehicle alone for 40 min. Duplicate immunoprecipitations were performed using SF-1 and GCN5 antibodies for each experimental condition. One set of samples was immunoprecipitated with primary antibody alone (lanes 1–4). The other set was immunoprecipitated with SF-1 or GCN5 antibodies followed by elution and reprecipitation (Re-ChIP) with the reciprocal antibody (lanes 5–8). Data are normalized to 1% input and are expressed as percent of the baseline value for each experimental condition. IP, Immunoprecipitation.
|
|
ACTH Stimulation Elicits a Cyclic Pattern of ERK 1/2 Activation
Activation of the MAPK pathway was assessed over the time course using antibodies recognizing the catalytically active kinases ERK 1/2. ACTH stimulation led to the transient activation of the ERKs 1/2 over the indicated time course as previously observed (37). ERK activation was detected as early as 20 min after stimulation with two distinct periods of elevated ERK phosphorylation between the 40- to 100- and 140- to 180-min time periods (Fig. 7A). Immunoblots performed with -ERK 1/2 antibodies confirmed equal loading of proteins. Quantification of multiple experiments reveals a statistically significant biphasic pattern of transient ERK activation that closely paralleled the temporal dynamics of recruitment and activation observed on the Mc2r promoter (Fig. 7A and Table 1). The ERKs have been shown to phosphorylate SF-1 on a single serine residue (Ser-203) within a major activation domain (activation function 1), leading to alterations in the ability of SF-1 to interact with cofactors (27). Therefore, we used a phospho-specific antibody to assess SF-1 phosphorylation on this residue. Once again, the pattern of SF-1 phosphorylation mimicked the temporal pattern of ERK activation with two distinct rounds of SF-1 phosphorylation (Fig. 7B). Immunoblots were performed using -SF-1 antibodies to confirm that SF-1 levels were unchanged over the indicated time course. Although SF-1 serine 203 resides within a consensus ERK 1/2 site and is phosphorylated in vitro and in response to MAPK activation, it is feasible that alternative kinases mediate this ACTH-dependent phosphorylation event in vivo.
Temporal Dynamics of cAMP Accumulation and PKA Activation in Response to ACTH
Having established that ERK 1/2 activation and phosphorylation of SF-1 on Ser-203 exhibited a biphasic mode of activation/phosphorylation, we examined the activation of the PKA signaling pathway. Determination of cAMP accumulation after ACTH stimulation indicated a robust increase in cAMP levels reaching maximal levels at the 20-min time point, with levels slowly declining over the next 40 min to baseline values for the remainder of the time course (Fig. 8A). Next, we examined the downstream activation of PKA by performing in vitro kinase assays using kemptide as substrate. Activation of PKA was observed within 20 min after ACTH treatment with sustained levels of activation detected throughout the remainder of the time course without any observed pattern of cyclic activation (Fig. 8B).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 8. Activation of the cAMP-Dependent PKA Pathway
cAMP (A) and PKA (B) activity were assessed in -amanitin synchronized Y1 cells over the extended time course using commercially available kits. Data represent the mean ± SEM of three independent experiments.
|
|
Activation of ERK 1/2 and PKA Are Required for ACTH-Dependent Transcription
The observed temporal dynamics of MAPK and PKA-dependent pathway activation led us to examine the effect of pharmacological inhibition of each pathway on ACTH-dependent H4 hyperacetylation and subsequent transcriptional initiation. Pretreatment of cells for 30 min with inhibitors of the MAPK (U0126) or PKA (H-89) pathways completely blocked the hyperacetylation of histone H4 on the Mc2r promoter compared with saline-treated controls (Fig. 9A). The apparent absence of histone acetylation led us to examine whether pharmacological inhibition of ACTH-dependent signaling cascades altered the recruitment and association of SF-1 with the Mc2r gene promoter. Interestingly, pharmacological blockade of the MAPK or PKA pathways precluded the ability of SF-1 to associate with the Mc2r promoter, suggesting an essential role for both pathways in the regulation of early SF-1-dependent transcriptional events, namely, recruitment and physical association with regulated promoter elements (Fig. 9B). Consistent with the lack of H4 hyperacetylation and SF-1 recruitment, no occupancy of phosphorylated RNA Pol II was observed after pharmacological inhibition compared with saline-treated controls (Fig. 9C). Collectively, these data suggest that the MAPK and PKA pathways coordinately regulate critical processes culminating in SF-1 recruitment and binding to cognate response elements.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 9. Effect of Pharmacological Inhibition of MAPK and PKA Pathway Activation on Mc2r Promoter Dynamics
Kinetic ChIP experiments were performed using -acetyl H4 (A), SF-1 (B), and phospho-RNA Pol II antibodies (C) over the 3-h time course. After 2 h of -amanitin synchronization, cells were washed and treated with U0126 (10 µM) or PKA inhibitor H-89 (1 µM) for 30 min before stimulation with ACTH (1 x 10–8 M). Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline value obtained for untreated cells as a function of time. ChIP experiments were performed a minimum of two times with similar results. pPol II, Phospho-RNA Pol II; Acetyl., acetylation.
|
|
ACTH-Dependent Up-Regulation of the MAPK Phosphatase, MKP-1
A number of studies have implicated MKP-1, a nuclear dual-specificity phosphatase, as a downstream target of PKA and a positive mediator of SF-1-dependent steroidogenic enzyme gene expression (12, 40). Given the temporal pattern of ERK 1/2 activation and the significance of the MAPK pathway in the regulation of SF-1 recruitment to the Mc2r, we assessed MKP-1 transcript (Fig. 10A) and protein (Fig. 10B) abundance over the extended time course to determine whether up-regulation of MKP-1 levels coincided with ERK 1/2 inactivation. Increases in MKP-1 mRNA were quantified by performing real-time PCR using MKP-1-specific, intron-spanning primers. GAPDH mRNA levels were used as an internal standard for data normalization. As anticipated, a significant increase (P < 0.01) in MKP-1 mRNA (380.05 ± 92.35) was detected at the 20-min time point when compared with baseline (100.00 ± 0.059), with a slow diminution over the remainder of the time course (40, 41). An increase in MKP-1 protein levels lagged MKP-1 transcription by approximately 20 min, with increased protein levels clearly observed at the 40-min time point. MKP-1 protein levels remained elevated through the 100-min time point with protein levels slowly decreasing through the remainder of the time course. Interestingly, the maximal accumulation of MKP-1 immediately preceded the inactivation of ERK 1/2, consistent with the known role of MKP-1 in the modulation of ERK activation and the observed dephosphorylation of ERK 1/2 in response to ACTH stimulation.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 10. Temporal Dynamics of MKP-1 Transcript and Protein Levels
-Amanitin-synchronized Y1 cells were stimulated over the indicated time course with ACTH (1 x 10–8 M). A, Transcript levels of MKP-1 were assessed by performing quantitative real-time PCR using intron-spanning primers. Data were normalized using GAPDH as an internal standard. Data represents the mean ± SEM of three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001. B, Protein levels of MKP-1 were examined by performing immunoblots with -MKP-1 antibodies. IB, Immunoblot.
|
|
HDAC Recruitment Coincides with Phospho-RNA Pol II Promoter Clearance
Dynamic changes in the posttranslational modification pattern of histones including acetylation, deacetylation, phosphorylation, methylation, and ubiquitination are known to provide gene-specific cues for the appropriate regulation of gene expression by affecting local chromatin to promote or restrict transcription (67, 68). Having established the pattern of histone H4 acetylation over the time course of ACTH stimulation, we sought to determine whether HDACs occupy the Mc2r promoter between active rounds of transcription. A survey of HDAC expression was performed by RT-PCR using specific primers for class I and class II HDACs (data not shown). We identified HDAC 1 and 7 as excellent candidates based upon their relatively high expression in whole mouse adrenals and Y1 adrenocortical cells (Fig. 11A). RT-PCR performed using RNA obtained from embryonic d 18 mice served as a positive control. HDAC 1 and 7 expression in the Y1 cell line was also confirmed by immunoblotting using specific antibodies (data not shown). ChIP assays performed over the extended time course using -HDAC 1 and -HDAC 7 antibodies revealed a single, distinct period of recruitment, spanning the 100- to 140-min time period (Fig. 11, B and C). Interestingly, the timing of HDAC recruitment coincided with RNA Pol II promoter clearance, which follows hypoacetylation of core histones within the proximal promoter region of the Mc2r gene.
HDAC 1 Associates with a Member of the Protein Phosphatase 1 (PP1)/Protein Phosphatase 2A (PP2A) Family
Previously, HDAC 1 was shown to associate with PP1 and promote the dephosphorylation of a serine residue in the cAMP-responsive element binding protein (CREB) which is required for activation of CREB-mediated transcription events (69). The ACTH-dependent association of HDAC 1 with the Mc2r promoter and the established role of phosphatases in steroidogenesis prompted us to examine whether HDAC 1 associated with protein phosphatases in an ACTH-dependent manner to potentially regulate the phosphorylation status of SF-1 (70). ChIP assays were performed using -PP1 antibodies to precipitate genomic fragments. Surprisingly, PP1 recruitment was biphasic with 10-fold and 22-fold increases in promoter occupancy at the 60- and 140-min time points, concomitant with the loss of H4 acetylation observed in the proximal promoter region (Fig. 12A). To determine whether ACTH stimulation altered the association of HDAC 1 with PP1, microcystin agarose was used to precipitate the catalytic subunit of PP1 and PP2A protein phosphatases before or after stimulating Y1 cells with ACTH for 120 min. Associated proteins were resolved by PAGE and immunoblotted with -HDAC antibodies. ACTH specifically induced the association of HDAC 1 but not HDAC 7 with PP1/PP2A (Fig. 12B), suggesting the potential of a functional interaction of these proteins on the Mc2r promoter.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 12. Association of PP1 or PP2A Family Members with the Mc2r Promoter and HDAC 1 and 7 in an ACTH-Dependent Manner
A, Kinetic ChIP experiments were performed using -PP1 antibodies, capable of recognizing PP1 and PP2A family members, over the 3-h time course. After 2 h of -amanitin synchronization, cells were washed and stimulated with ACTH (1 x 10–8 M) for the indicated times. Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline value obtained for untreated cells as a function of time. ChIP experiments were performed a minimum of two times with similar results. B, Microcystin-LR agarose was used to precipitate phosphoprotein phosphatases and associated proteins. Precipitations were washed extensively, resolved by SDS-PAGE, transferred to nitrocellulose, and subsequently immunoblotted with antibodies specifically recognizing HDAC 1 and 7. Equal quantities of protein lysates for each condition were resolved in parallel and immunoblotted with -SF-1 antibodies to confirm that equal quantities of protein had been added to microcystin agarose precipitations. IB, Immunoblot; acetyl., acetylation.
|
|
HDAC Inhibition Alters the Amplitude But Not the Periodicity of H4 Acetylation
To examine the functional significance of HDAC recruitment to the Mc2r promoter, Y1 cells were pretreated for 30 min with or without the potent HDAC inhibitor, trichostatin A (TSA), and the acetylation of the proximal Mc2r promoter was assessed by performing ChIP assays over the remainder of the time course. In the absence of HDAC inhibition, a 5- to 7-fold increase in histone H4 acetylation was observed at the 40- and 120-min time points as observed previously (Fig. 13A; refer to Fig. 1). In contrast, TSA pretreatment resulted in a dramatic increase in the acetylation of H4 on the Mc2r proximal promoter with a 40- to 50-fold increase in H4 acetylation. Interestingly, although the amplitude of this response was drastically higher after TSA pretreatment, no changes were observed in the dynamic periodicity of H4 acetylation over the entire time course. To determine the functional outcome of heightened H4 acetylation, -amanitin synchronized Y1 cells were treated with or without TSA 20 min after ACTH stimulation (before any recruitment) and phospho-RNA Pol II occupancy of the Mc2r gene promoter was assessed. Surprisingly, despite an exaggerated acetylation of histone H4, a state normally associated with heightened gene transcription, there was 1) a complete absence of transcriptional initiation as assessed by phospho-RNA Pol II occupancy of the Mc2r promoter (Fig. 13B); and 2) absence of mRNA accumulation as assessed by quantitative RT-PCR (data not shown). Therefore, although HDAC activity determines the magnitude of the transcriptional response, it is not solely responsible for promoter clearance after ACTH stimulation. Nonetheless, it is absolutely required for any productive transcriptional initiation. These data suggest the presence of nonhistone targets of histone acetyltransferase and deacetylase action, which are required for effective control of gene transcription.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 13. Effect of HDAC Inhibition on ACTH-Dependent Histone H4 Acetylation
Kinetic ChIP experiments were performed on -amanitin-synchronized Y1 adrenocortical cells treated with or without TSA (5 µM) 30 min (A) preceding ACTH stimulation (1 x 10–8 M) or 20 min (B) before the 40-min time point and for all subsequent time points for the 3-h time course. Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline value obtained for untreated cells as a function of time. ChIP experiments were performed a minimum of two times with similar results. pPol II, Phospho-RNA Pol II; Acetyl., acetylation.
|
|
Inhibition of Serine/Threonine Phosphatases Alters the Periodicity, But Not the Amplitude, of Transcriptional Initiation
Having defined a role for HDACs in the regulation of the amplitude of the transcriptional response and identified an ACTH-dependent interaction of HDAC 1 and 7 with an unidentified protein phosphatase precipitated with microcystin agarose, as well as the active recruitment of a protein phosphatase to the proximal promoter region of the Mc2r promoter, we next sought to determine the potential importance of protein phosphatase recruitment within the context of transcriptional activation. Due to the complete absence of complex formation when cells were pretreated with pharmacological inhibitors before stimulation over the entire time course (refer to Fig. 9), the ability to examine the functional role of specific signaling pathways in the temporal regulation of promoter-specific events is limited. To circumvent this limitation, -amanitin-synchronized Y1 cells were treated with serine-threonine phosphatase inhibitor 20 min after ACTH stimulation (before any recruitment), and phosho-Pol II occupancy of the Mc2r gene promoter was assessed. Surprisingly, similar to HDAC inhibition, pharmacological inhibition of serine/threonine phosphatases before recruitment abolishes 1) transcriptional initiation as assessed by phospho-Pol II occupancy of the Mc2r gene promoter (Fig. 14A); and 2) mRNA accumulation as assessed by quantitative RT-PCR (data not shown). To assess the putative role of serine/threonine phosphatases in the regulation of the temporal dynamics observed for transcriptional initiation, pharmacological inhibition of serine/threonine phosphatases was initiated coincident with the first cycle of phospho-Pol II promoter occupancy, allowing for the initiation of a single round of transcription before the blockade of phosphatase action. ChIP assays were performed using -phospho-Pol II antibodies to examine the effect of serine/threonine phosphatase inhibition on the temporal pattern of recruitment and subsequent transcriptional activation of the Mc2r gene. These studies revealed a profound effect on the dynamic periodicity of phospo-RNA Pol II occupancy on the Mc2r promoter with the second round of recruitment occurring 20 min earlier than the normal recruitment of phospho-RNA Pol II (Fig. 14B). This effect upon the periodicity of recruitment was quite specific, with no changes in the amplitude of recruitment observed. Therefore, these data suggest that the temporal dynamics of serine/threonine phosphorylation/dephosphorylation of (an) unknown protein(s) is capable of modulating both transcriptional initiation and the ultimate periodicity of transcriptional initiation. In addition, these results elucidate an important role for serine/threonine phosphatase activity in the early stages of gene transcription.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 14. Inhibition of Serine/Threonine Phosphoprotein Phosphatases Alters the Frequency of Transcriptional Initiation
Kinetic ChIP experiments were performed on -amanitin-synchronized Y1 adrenocortical cells treated with or without serine/threonine phosphatase inhibitor cocktail 20 min before the 40-min (A) or 80-min time point (B) and for all subsequent time points for the remainder of the 3-h time course. Data are normalized to values obtained for 1% input controls, and results are presented as percent of baseline value obtained for untreated cells as a function of time. ChIP experiments were performed a minimum of two times with similar results. pPol II, Phospho-RNA Pol II; Phos Inh., phosphatase inhibitor.
|
|
|
DISCUSSION
|
|---|
The transcriptional activation of eukaryotic genes is a complex process that entails specific response element recognition in the context of a complex genomic environment and the recruitment of multiprotein coregulatory complexes capable of modifying the local chromatinized environment in a manner that favors transcriptional initiation and the ultimate recruitment of the RNA polymerase II holoenzyme complex (71, 72). Traditional models attempting to characterize nuclear receptor-mediated transcription have proposed a static model of transcriptional activation, which suggests that nuclear receptors and their associated complexes remain in contact with promoters until the initiating stimulus is removed. Recently, several laboratories have used the ChIP assay as well as other experimental paradigms to examine the ligand-dependent pattern of nuclear receptor and coregulatory proteins recruitment over time courses encompassing periods of active transcription (57, 73). Collectively, these studies have revealed a very complex model of nuclear receptor action in response to ligand binding, which involves the transient, dynamic, and combinatorial association of factors with promoter elements, events that are reflected by the cyclical association of complexes that are functionally linked with different states of gene transcription (57, 73, 74).
Despite the high degree of structural conservation and shared mechanisms used by both ligand-dependent and orphan members of the nuclear receptor superfamily, a similar mode of cyclic transcriptional activation has not been established for the ligand-independent activation of orphan receptor-mediated transcriptional programs initiated by cell surface, peptide hormone receptors. We hypothesized that such a system would be controlled by the temporal activation of upstream signaling pathways capable of regulating the dynamic recruitment and assembly of transcriptional complexes through enzymatic modification of the nuclear receptor and associated complex components.
In this study, we demonstrate that activation of ACTH-dependent signaling cascades leads to the cyclical recruitment of SF-1, active remodeling of the local chromatin environment characterized by coactivator recruitment, and core histone acetylation followed by the subsequent initiation of multiple rounds of transcription. An assessment of PKA and MAPK signaling dynamics over the same time course indicates that both pathways are essential for SF-1 recruitment and the establishment of transcriptional competence, highlighting the potential existence of cross talk between, or indispensable contributions provided by, each signaling cascade in ACTH-mediated transcriptional activation. Although the absolute requirement for PKA activation in the ACTH-dependent transcriptional response has long been established, the critical effectors downstream of PKA activation have not been identified (75, 76). Therefore, considerable effort has been focused upon identification of downstream effectors of PKA action. More recently, a role for the MAPK signaling pathway has been substantiated in this coordinated transcriptional response in addition to its known role in partially mediating the weak mitogenic actions of ACTH (27, 28, 29, 37, 40, 78). Undoubtedly, the cyclical pattern of transcriptional initiation involves inherent recycling of transcriptional components as well as the modulation of intracellular signaling pathways, which ultimately dictate the formation of coregulatory complexes through a variety of posttranslational modifications. Therefore, the temporal regulation of signaling pathways responsible for regulating the posttranslational modification status of nuclear receptors and associated coregulatory proteins is an important area of active investigation.
We demonstrate that ACTH induces a cyclic phosphorylation of both ERK 1/2 and SF-1 on serine 203, coincident with the cyclic recruitment of SF-1 and ultimate transcriptional activation. This is of interest because we had previously shown that ERK 1/2 mediated phosphorylation of SF-1 on serine 203 served to enhance the recruitment of transcriptional coregulators and subsequent transcriptional output in transient transcription assays. On the contrary, Sewer and Waterman (79) subsequently revealed that the dephosphorylation of SF-1 on serine 203 correlated with an enhanced transcription of P450c17 in similar assays. The studies presented here indicate that coordinated phosphorylation and dephosphorylation of SF-1 may represent a potential determinant affecting transcriptional activation in the context of a chromatinized template. Mechanistically, MKP-1 has previously been implicated as the phosphatase responsible for reducing the phosphorylation of SF-1 on serine 203 (27, 79). We show that ACTH results in de novo synthesis of MKP-1 with maximal protein accumulation at the time of ERK 1/2 and SF-1 dephosphorylation and promoter clearance. Moreover, PKA has previously been shown to directly phosphorylate MKP-1 in vitro, suggesting that MKP-1 may be a putative downstream effector of PKA that is capable of modulating the dynamics of MAPK activation (40, 41, 79) and subsequent clearance of the SF-1 complex(s) from the promoter. Whereas loss of SF-1 recruitment and subsequent transcriptional activation after pretreatment with U0126 and H-89 are consistent with the proposed roles of PKA and the ERKs, it is certainly possible that other signaling pathways inhibited by these compounds may also be involved. We propose the following model to account for the intrinsic cyclicity of SF-1 recruitment. ACTH-stimulated ERK 1/2 activation leads to the phosphorylation of SF-1, resulting in SF-1 binding and coregulator recruitment. Concomitantly, ACTH-mediated activation of PKA results in the generation of MKP-1 leading to the inactivation of ERK 1/2 and subsequent reduction of SF-1 phosphorylation resulting in ultimate promoter clearance between transcriptional cycles (Fig. 15).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 15. Model of ACTH-Mediated Cyclic SF-1 Recruitment and Initiation of Transcriptional Activation
ACTH activates both MAPK and PKA cascades. ERK 1/2 phosphorylates SF-1 and facilitates SF-1 binding and subsequent recruitment of coregulatory proteins such as the SRC family and ultimate activation of the phospho-RNA polymerase holoenzyme. Conversely, PKA transcriptionally up-regulates the MAPK phosphatase, MKP-1, which dephosphorylates ERK 1/2 resulting in a loss of SF-1 phosphorylation and ultimate SF-1 promoter clearance. During this interpeak period of transcriptional repression, ACTH also induces the recruitment of PP1 to the promoter and the subsequent recruitment of HDAC activity before the initiation of transcriptional cycle 2. Ac-H4, Acetyl H4; pPOLII, phospho-RNA Pol II.
|
|
The observed cyclicity of nuclear receptor-mediated transcriptional activation is predicated on the sequential presence of two discrete transcriptional cycles. Temporally, these peaks are only recognized due to the unique absence of promoter occupancy between cycles. Nuclear receptor transcriptional activation requires the coordinate regulation of both chromatin remodeling, binding and assembly of the receptor complex, and ultimate communication with the basic transcription machinery. Promoter clearance would be predicted to involve signals that mediate both nuclear receptor binding and chromatin remodeling. Indeed, in addition to the regulated loss of SF-1 binding in response to cyclic dephosphorylation after the ACTH-mediated activation of phosphatase pathways such as MKP-1, we also observe ACTH-dependent recruitment of HDAC 1 and HDAC 7 to the MC2R promoter during the phospho-Poll 2 interpeak period of promoter clearance. Consistent with the known physical and functional association of protein phosphatases in HDAC activation (70, 80, 81), ACTH induces the association of PP1/2A with HDAC 1 within a time period encompassing the interpeak period of histone H4 deacetylation. Lastly, we demonstrate that the pharmacological inhibition of a serine/threonine phosphoprotein phosphatase leads to an alteration in the frequency of transcriptional initiation. In contrast, the pharmacological inhibition of HDACs alters the amplitude of the transcriptional response, suggesting that multiple enzymatic functions are required for the appropriate entrainment of the ACTH-dependent transcriptional response. Additionally, the fact that HDAC and phosphatase inhibition did not result in loss of interpeak promoter clearance suggests that additional enzymatic activities, such as the Sirtuin family of histone deacetylases and TSA-insensitive HDACs or processes such as nucleosome degradation and recycling, participate in this critical process. How HDACs participate in peak amplitude and PP1/2A in periodicity are areas of active inquiry.
Phylogenetic analysis of the nuclear hormone receptor superfamily suggests that common ancestral nuclear receptors may have been orphans, with ligand-binding capacity acquired independently for each receptor during evolution (24). In support of this view, several classical as well as orphan receptors can be activated in a ligand-independent manner, suggesting the retention of evolutionarily conserved mechanisms of transcription factor activation predating ligand acquisition (82, 83, 84, 85, 86). Coincident with the development of endocrine systems predicated on feed-forward mechanisms of activation and feed-back mechanisms of inhibition was the need to couple peptide-hormone-mediated signaling cascades to nuclear receptor function in endocrine organs. Whereas some nuclear receptors assumed the responsibility of responding to the steroid ligands synthesized in these endocrine glands, other orphans such as SF-1 became responsible for gene activation of the enzymes responsible for steroid production and gained the ability to become activated through posttranslational modifications mediated by the peptide hormones responsible for the feed-forward stimulation of an endocrine organ in a unique endocrine system. Activation of nuclear receptors by peptide hormone-dependent posttranslational modification and ligand binding appear to follow an ancient convention of transcriptional cyclicity. The recent observation of phospholipids molecules in the SF-1 binding pocket of bacterially derived SF-1 protein leaves open the possibility that peptide hormone stimulation might additionally modulate SF-1 transcriptional activity in mammalian systems by regulating ligand availability (87, 88). The common mechanisms by which these two modes of activation functionally coordinate the transcriptional cyclicity and interpeak promoter clearance should reveal fundamental mechanisms that determine the dynamic assembly and disassembly of nuclear receptor transcription complexes.
|
MATERIALS AND METHODS
|
|---|
-Amanitin was purchased from Sigma Chemical Co. (St. Louis, MO). All tissue culture reagents, protein A-agarose, and Trizol were purchased from Invitrogen (Carlsbad, CA). Human ACTH(1–24) was from Calbiochem (La Jolla, CA). Antibodies used were as follows. SF-1, acetylated-H4, and RNA polymerase II (Upstate Biotechnology, Inc., Lake Placid, NY); CTD4H8 (Upstate Biotechnology Inc., Charlottesville, VA); HA Y-11 and MKP-1 (Santa Cruz Biotechnology Inc., Santa Cruz, CA); ERK 1/2, phospho-ERK 1/2 (Thr202/Tyr204), and HDAC 1 and 7 antibodies (Cell Signaling Technology Inc., Beverly, MA). Luciferase activity was measured using the Luciferase Assay System (Promega Corp., Madison, WI), and ß-galactosidase chemiluminescence reagents were from Tropix (Applied Biosystems, Foster City, CA).
Plasmid Construction
The pCMVTag3c-HASF-1 expression plasmid was created by directional cloning of the BamHI/XhoI fragment from pciNeo-HASF-1 (27) into pCMVTag3c (Stratagene, La Jolla, CA). The resultant plasmid was linearized and transfected into Y1 adrenocortical cells and selected in G418 (500 µg/ml) containing media. G418-resistant colonies were isolated and screened for HASF-1 expression. The luciferase reporter plasmid (pGL3-mACTHR) containing 1 kb of the mouse melanocortin 2 receptor promoter was a kind gift of F. Beuschlein (University of Freiburg, Freiburg, Germany).
ChIP Assay
Y1 adrenocortical cells were maintained in DMEM supplemented with 7.5% horse serum, 2.5% fetal bovine serum, and antibiotics at 37 C under a humidified atmosphere of 5% CO2. Assay was performed with minor modifications of the procedure described by Shang et al. (57). Cells were grown to 90% confluence and serum deprived for 48 h in DMEM supplemented with 0.05% BSA followed by treatment with 2.5 µM -amanitin for 2 h. Cells were washed twice with PBS, and fresh serum-free medium supplemented with or without inhibitors was added 30 min before ACTH (1 x 10–8 M) stimulation over the indicated time course. Chromatin was cross-linked by addition of formaldehyde at a final concentration of 1% and incubation at 37 C for 10 min. Cells were washed two times in ice-cold PBS followed by collection in PBS containing 1x protease inhibitor cocktail (Sigma). Crude nuclei were prepared as previously described (89). Nuclei were resuspended in 150 µl of lysis buffer [1% sodium dodecyl sulfate (SDS), 10 mM EDTA, 50 mM Tris-HCl (pH 8.1), 1x protease inhibitor cocktail] per 2 x 106 cells and sonicated four times for 10 sec each at maximum setting (Fisher Sonic Dismembrator, model 300; Fisher Scientific, Hampton, NH) followed by centrifugation at 4 C for 15 min. Supernatants were collected and diluted in buffer [0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl (pH 8.1), 167 mM NaCl, 1x protease inhibitor cocktail] followed by immunoclearing with 2 µg sheared salmon sperm DNA (Invitrogen), 15 µl preimmune serum, and 80 µl protein A-Sepharose [50% slurry in 10 mM Tris-HCl (pH 8.1), 1 mM EDTA, 0.5 mg/ml BSA, 0.05% sodium azide, 200 µg/ml sheared salmon sperm DNA) for 2 h at 4 C. Immunoprecipitation was performed overnight with specific antibodies. Immune complexes were recovered by addition of 40 µl protein A-Sepharose and incubation at 4 C for an additional hour. Precipitates were washed sequentially for 5 min each in buffer TSE I [0.1% SDS, 1%Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 150 mM NaCl], twice in TSE II [0.1% SDS, 1%Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl], and buffer III [0.25 M LiCl,1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl (pH 8.1)], followed by three washes in Tris-EDTA. Immunoprecipitates were extracted two times with 1% SDS, 0.1 M NaHCO3. Eluates were combined and heated overnight at 65 C to reverse cross-linking. DNA fragments were isolated by performing a phenol-chloroform extraction and subsequent precipitation of DNA overnight at –20 C. Primer pairs used for ChIP assays are listed in Table 1.
Re-ChIP
Soluble chromatin was prepared from -amanitin-synchronized Y1 cells treated with vehicle or ACTH (1 x 10–8) for 40 min. Complexes were eluted from primary immunoprecipitations by incubation in 10 mM dithiothreitol for 30 min at 37 C, diluted 1:50 in re-ChIP buffer [1% Triton X-100, 2 mM EDTA, 150 mM NaCl, 20 mM Tris-HCl (pH 8.1)] followed by immunoprecipitation with the second antibodies. Results were quantified by performing quantitative real-time PCR using Mc2r promoter-specific primers. Data are presented as percent of unstimulated for each experimental condition and were normalized to 1% input values.
cAMP Accumulation and PKA Activity
Y1 adrenocortical cells were treated exactly as mentioned for ChIP studies. cAMP accumulation was determined using an immunoassay kit (Sigma). PKA activity was measured in whole-cell lysates using a commercially available kit (Upstate Biotechnology). All assays were performed in triplicate. and results are expressed as mean ± SEM.
Immunoblotting
Total cell lysates were prepared in lysis buffer [50 mM Tris-HCl (pH 7.4), 1% NP-40, 0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM NaF, and protease inhibitor cocktail]. Lysates were allowed to rotate at 4 C for 30 min, and protein contents of the high-speed supernatant were measured using the Bradford protein assay (Bio-Rad Laboratories, Inc., Hercules, CA). Equivalent quantities of protein (20–45 µg) were resolved on polyacrylamide-SDS gels, transferred to nitrocellulose membrane (Bio-Rad), and immunoblotted with specific antibodies. Results were visualized using the Supersignal West Dura Extended Duration Substrate kit (Pierce Chemical Co., Rockford, IL.).
RT-PCR and Real-Time Quantitative PCR
Total RNA was isolated from Y1 cells using the Trizol reagent (Invitrogen) according to the manufacturer’s directions. Isolated RNA was DNase treated using the DNase-Free kit (Ambion, Inc., Austin, TX). One microgram of total RNA was reverse transcribed using the iScript cDNA synthesis kit (Bio-Rad). Real-time quantitative PCR was performed using the QuantiTect SYBR Green PCR kit (QIAGEN, Chatsworth, CA) and the Opticon 2 real-time PCR system (MJ Research, Beverly, MA). All data were normalized to GAPDH as an internal standard. The specific oligonucleotide sequences for MKP-1 and GAPDH are provided in Table 1. Results are expressed as the mean ± SEM of three independent experiments. All primer pairs were validated using delta C(T) analysis as previously described (90).
Luciferase Assay
Y1 adrenocortical cells (2 x 104/well) were plated in 24-well plates and transiently transfected 24 h later with the mouse Mc2r-Luc reporter plasmid and pCMVTag3c-HASF-1 by calcium phosphate coprecipitation. Cotransfection with CMV-ß-gal was used to normalize for transfection efficiency. Cells were lysed for measurement of luciferase and ß-galactosidase activity 48 h after transfection using a Veritas microplate luminometer (Turner Biosystems, Sunnyvale, CA). Data represent the mean ± SEM obtained from three independent experiments, each performed in triplicate.
|
ACKNOWLEDGMENTS
|
|---|
We thank H. Ingraham for providing the phospho-specific SF-1 antibodies.
|
FOOTNOTES
|
|---|
This work was supported by National Institutes of Health Grants R01 DK62027 from the National Institute of Child Health and Human Development (to G.D.H.) and Grant T32 DE007057 from The National Institute of Dental and Craniofacial Research (to J.N.W.)
First Published Online August 18, 2005
Abbreviations: ChIP, Chromatin immunoprecipitation; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HA, hemagglutinin; GCN5, general control of amino acid synthesis 5-like 1; HDAC, histone deacetylase; LBD, ligand-binding domain; Mc2r, melanocortin 2 receptor; MKP-1, mitogen-activated kinase phosphatase-1; P450c17, cytochrome P450 17-hydroxylase; PKA, protein kinase A; PP1, protein phosphatase 1; RNA Pol II, RNA polymerase II; SDS, sodium dodecyl sulfate; SF-1, steroidogenic factor 1; SRC, steroid receptor coactivator; TSA, trichostatin.
Received for publication May 30, 2005.
Accepted for publication August 8, 2005.
|
REFERENCES
|
|---|
- Ikeda Y, Lala DS, Luo X, Kim E, Moisan MP, Parker KL 1993 Characterization of the mouse FTZ-F1 gene, which encodes a key regulator of steroid hydroxylase gene expression. Mol Endocrinol 7:852–860[Abstract]
- Morohashi K, Honda S, Inomata Y, Handa H, Omura T 1992 A common trans-acting factor, Ad4-binding protein, to the promoters of steroidogenic P-450s. J Biol Chem 267:17913–17919[Abstract/Free Full Text]
- Morohashi K, Zanger UM, Honda S, Hara M, Waterman MR, Omura T 1993 Activation of CYP11A and CYP11B gene promoters by the steroidogenic cell-specific transcription factor, Ad4BP. Mol Endocrinol 7:1196–1204[Abstract]
- Val P, Lefrancois-Martinez AM, Veyssiere G, Martinez A 2003 SF-1 a key player in the development and differentiation of steroidogenic tissues. Nucl Recept 1:8[CrossRef][Medline]
- Hammer GD, Ingraham HA 1999 Steroidogenic factor-1: its role in endocrine organ development and differentiation. Front Neuroendocrinol 20:199–223[CrossRef][Medline]
- Luo X, Ikeda Y, Parker KL 1994 A cell-specific nuclear receptor is essential for adrenal and gonadal development and sexual differentiation. Cell 77:481–490[CrossRef][Medline]
- Babu PS, Bavers DL, Beuschlein F, Shah S, Jeffs B, Jameson JL, Hammer GD 2002 Interaction between Dax-1 and steroidogenic factor-1 in vivo: increased adrenal responsiveness to ACTH in the absence of Dax-1. Endocrinology 143:665–673[Abstract/Free Full Text]
- Beuschlein F, Mutch C, Bavers DL, Ulrich-Lai YM, Engeland WC, Keegan C, Hammer GD 2002 Steroidogenic factor-1 is essential for compensatory adrenal growth following unilateral adrenalectomy. Endocrinology 143:3122–3135[Abstract/Free Full Text]
- Hu MC, Hsu NC, Pai CI, Wang CK, Chung B 2001 Functions of the upstream and proximal steroidogenic factor 1 (SF-1)-binding sites in the CYP11A1 promoter in basal transcription and hormonal response. Mol Endocrinol 15:812–818[Abstract/Free Full Text]
- Huang Y, Hu M, Hsu N, Wang CL, Chung B 2001 Action of hormone responsive sequence in 2.3 kb promoter of CYP11A1. Mol Cell Endocrinol 175:205–210[CrossRef][Medline]
- Chau YM, Crawford PA, Woodson KG, Polish JA, Olson LM, Sadovsky Y 1997 Role of steroidogenic-factor 1 in basal and 3',5'-cyclic adenosine monophosphate-mediated regulation of cytochrome P450 side-chain cleavage enzyme in the mouse. Biol Reprod 57:765–771[Abstract]
- Sewer MB, Waterman MR 2002 Transcriptional complexes at the CYP17 CRS. Endocr Res 28:551–558[CrossRef][Medline]
- Bakke M, Lund J 1995 Mutually exclusive interactions of two nuclear orphan receptors determine
TOP
ABSTRACT
INTRODUCTION
