Silencing of Transcription of the Human Luteinizing Hormone Receptor Gene by Histone Deacetylase-mSin3A Complex*

Modification of chromatin structure by histone acetylases and deacetylases is an important mechanism in modulation of eukaryotic gene transcription. The present study investigated regulation of the human luteinizing hormone receptor (hLHR) gene by histone deacetylases. Inhibition of histone deacetylases (HDACs) by trichostatin A (TSA) increased hLHR promoter activity by 40-fold in JAR cells and markedly elevated endogenous hLHR mRNA levels. Acetylated histones H3 and H4 accumulated in TSA-treated cells and associated predominantly with the hLHR promoter. Furthermore, TSA significantly enhanced the recruitment of RNA polymerase II to the promoter. One of the two Sp1 sites essential for basal promoter activity was identified as critical for the TSA effect, but the binding of Sp1/Sp3 to this site remained unchanged in the absence or presence of TSA. A multiprotein complex was recruited to the hLHR promoter via interaction with Sp1 and Sp3, in which HDAC1 and HDAC2 were docked directly to Sp1-bound DNA and indirectly to Sp3-bound DNA through RbAp48, while mSin3A interacted with both HDACs. HDAC1 and HDAC2 were shown to potently repress the hLHR gene transcription, and mSin3A potentiated the inhibition mediated by HDAC1. Our studies have demonstrated that the HDAC-mSin3A complex has an important role in the regulation of hLHR gene transcription by interaction with Sp1/Sp3 and by region-specific changes in histone acetylation and polymerase II recruitment within the hLHR promoter.


Modification of chromatin structure by histone acetylases and deacetylases is an important mechanism in modulation of eukaryotic gene transcription. The present study investigated regulation of the human luteinizing hormone receptor (hLHR) gene by histone deacetylases. Inhibition of histone deacetylases (HDACs) by trichostatin A (TSA) increased hLHR promoter activity by 40-fold in JAR cells and markedly elevated endogenous hLHR mRNA levels. Acetylated histones H3 and H4 accumulated in TSA-treated cells and associated predominantly with the hLHR promoter.
Furthermore, TSA significantly enhanced the recruitment of RNA polymerase II to the promoter. One of the two Sp1 sites essential for basal promoter activity was identified as critical for the TSA effect, but the binding of Sp1/Sp3 to this site remained unchanged in the absence or presence of TSA. A multiprotein complex was recruited to the hLHR promoter via interaction with Sp1 and Sp3, in which HDAC1 and HDAC2 were docked directly to Sp1-bound DNA and indirectly to Sp3-bound DNA through RbAp48, while mSin3A interacted with both HDACs. HDAC1 and HDAC2 were shown to potently repress the hLHR gene transcription, and mSin3A potentiated the inhibition mediated by HDAC1. Our studies have demonstrated that the HDAC-mSin3A complex has an important role in the regulation of hLHR gene transcription by interaction with Sp1/Sp3 and by region-specific changes in histone acetylation and polymerase II recruitment within the hLHR promoter.
The luteinizing hormone receptor (LHR) 1 is an essential G protein-coupled receptor located on the plasma membrane of gonadal cells. It mediates gonadotropin signals and triggers intracellular responses that participate in maturation and function of the gonads as well as the regulation of steroidogenesis and gametogenesis (1,2). The LHR gene is also expressed in several non-gonadal tissues, including the uterus and placenta, where its functions have not been determined (for a review, see Ref. 2).
Characterization of the LHR gene promoter from different species has provided insights into regulatory mechanism of LHR gene transcription. Our previous studies showed that the LHR gene is TATA-less and contains multiple transcriptional start sites (3)(4)(5)(6)(7). The minimal promoter domain resides within the 180-bp region 5Ј to ATG (ϩ1) and contains multiple elements with potential binding activity for transcriptional factors Sp1 and Sp3. Two Sp1/Sp3 binding domains of central importance for basal promoter activity have been identified in human and rodents (for a review, see Ref. 2). Furthermore, nuclear orphan receptors EAR2 and EAR3 were shown to repress LHR gene transcription through binding to an imperfect direct repeat motif (8). In contrast, activation of the promoter activity at the same site by another orphan receptor, TR4, was only observed for the human LHR gene (hLHR). This difference was attributed to a single nucleotide base pair substitution at the core motif and absence of a guanine in the 3Ј adjacent sequence of the rat LHR promoter (8,9). Recent evidence have demonstrated that modification of nucleosomal histones by two opposing enzymes, histone acetyltransferases and histone deacetylases (HDACs), plays an active role in transcriptional regulation of a number of target genes (10 -13). Histone acetyltransferase-catalyzed acetylation of the conserved lysine residues at N-terminal region of histones reduces interaction between chromosomal DNA and histone tails, rendering an accessible chromatin environment for initiation of target gene transcription. In contrast, HDACs promote formation of a compact promoter and transcriptional repression (for reviews, see Refs. 14 and 15). In the present study, an investigation of the modulation of hLHR gene transcription demonstrated that the histone deacetylase-mSin3A complex has a major role in silencing of Sp1/Sp3-driven hLHR gene transcription.

Reporter Gene Constructs and Expression
Vectors-All plasmids were constructed by standard recombinant DNA techniques. The wild type hLHR promoter construct or the promoters harboring the mutant Sp1-1, Sp1-2, or orphan receptor binding site (direct repeat) have been described previously (7,8). Other mutant promoter/reporter gene constructs were designed in order to disrupt any DNA binding activity of Sp1/Sp3 transcriptional factors.
For construction of HDAC1 and HDAC2 expression plasmids, the appropriate DNA fragment containing HCAC1 or HDAC2 coding sequence with the stop codon removed was generated by PCR amplification using Quick-Clone cDNA of human ovary (CLONTECH, Palo Alto, CA) as template, followed by ligation into pcDNA 3.1 vector (Invitrogen, Carlsbad, CA) in frame with C-terminal V 5 /His tag. The fidelity of the clones was verified by DNA sequencing. The expression of HDACs in transfection assays was confirmed by Western blot using V 5 antibody (Invitrogen). For construction of the expression plasmid of mSin3A, a 4.721-kb fragment harboring full-length mSin3A coding sequence as well as 78-bp 5Ј-untranslated region and 793-bp 3Ј-untranslated region was released from pBluescript SK-mSin3A plasmid (kindly provided by Drs. Halleck and Schlegel, Department of Biochemistry and Molecular Biology, Penn State University) and ligated into pcDNA3.1 vector at the EcoRI site.
Transfection of the various constructs into JAR cells was carried out using LipofectAMINE Plus reagent, following the procedures recommended by the manufacturer. The luciferase activities were normalized to light units/g of protein and expressed as mean Ϯ S.E. All experiments were performed at least three times in triplicate wells.
RNA Isolation and RT-PCR-JAR cells were cultured to 90% confluence followed by treatment with vehicle (ethanol) or with 100 ng/ml TSA (330 nM) for various times as indicated. At different time points, cells were collected, and total RNA was isolated using an RNeasy kit (Qiagen, Valencia, CA). Any possible co-purified DNA was removed during the preparation using an RNase-free DNase set (Qiagen).
RT-PCR was carried out using Qiagen One-Step RT-PCR kit (Qiagen), in which 600 ng of total RNA of each sample was used as template. The primers used for amplification of a 474-bp fragment of human LHR coding region were 5Ј-GGAAACCACTCTCTCACAAGT-3Ј and 5Ј-GGT-GGATTGAGAAGGCTTATTTG-3Ј. Amplification of a 620-bp fragment of human ␤-actin gene was carried out as an internal control. The specific primers used for h␤-actin were 5Ј-CCTCGCCTTTGCCGATC-C-3Ј and 5Ј-GGATCTTCATGAGGTAGTCAGTC-3Ј (Invitrogen).
Chromatin Immunoprecipitation-Cells were grown to 90% confluence before TSA was added, and cells continued to be cultured for 24 h. Cells were washed twice with phosphate-buffered saline and then crosslinked with 1% formaldehyde at room temperature for 10 min. Cells were rinsed with ice-cold phosphate-buffered saline twice and collected into 1 ml of ice-cold phosphate-buffered saline. Cells were harvested by brief centrifugation and then washed once with phosphate-buffered saline. Cells were resuspended in 0.3 ml of lysis buffer (1% SDS, 5 mM EDTA, 50 mM Tris-HCl, pH 8.1, 1ϫ protease inhibitor mixture (Roche Molecular Biochemicals)) and incubated on ice for 10 min. Sonication of cells was performed 5 times for 10 s each followed by centrifugation at 4°C for 10 min. Supernatants were collected and diluted 1:10 in dilution buffer (1% Triton X-100, 2 mM EDTA, 150 mM NaCl, 20 mM Tris-HCl, pH 8.1, 1ϫ protease inhibitor mixture) followed by immunoclearing with 2 g of sheared salmon sperm DNA, 20 l of preimmune serum and 30 l of protein A-agarose (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) for 2 h at 4°C. Immunoprecipitation was carried out for overnight at 4°C with 3-5 l of specific antibodies as indicated. The antibodies for acetylated forms of histone H3 or H4 were obtained from Upstate Biotechnology, Inc. (Lake Placid, NY), and the RNA polymerase II antibody was purchased from Santa Cruz Biotechnology. After immunoprecipitation, 30 l of protein A-agarose and 2 g of salmon sperm DNA were added, and the sample was incubated for another 1 h. Precipitates were recovered by centrifugation at 2500 rpm at 4°C for 5 min and washed sequentially for 10 min each in TSE I (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 150 mM NaCl), 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% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl, pH 8.1). Precipitates were then washed three times with TE buffer and extracted three times with 150 l of 1% SDS, 0.1 M NaHCO 3 . Eluates were pooled and heated at 65°C for 6 h to reverse the formaldehyde cross-linking. The samples were then treated with protease K (100 g/ml) at 50°C for 45 min followed by phenol extraction and ethanol precipitation facilitated by adding Pellet Paint TM co-precipitant reagent (Novagen, Madison, WI). DNA pellet was dissolved in 40 l of TE buffer. 1 l of DNA was used for PCR with 25 cycles of amplification.
Nuclear Extract Preparation and Electrophoretic Mobility Shift Assay-Nuclear extracts from TSA-treated or vehicle-treated (ethanol) JAR cells were prepared as described previously (8,16). For EMSA, [ 32 P]ATP end-labeled double-stranded DNA oligomer corresponding to the Sp1-1 site (5Ј-AGCCAAGGGGCGGGGAGAGGG-3Ј) of the hLHR promoter was incubated with 5-g nuclear extracts on ice for 15 min. In supershift assays, specific antibodies for Sp1, Sp3, or normal rabbit IgG (Santa Cruz Biotechnology) were preincubated with proteins for 30 min at 4°C before the probe was added. DNA-protein complexes were resolved by electrophoresis on 5% native polyacrylamide gel.
DNA Affinity Precipitation Assay (DAPA)-5Ј biotin end-labeled sense and antisense oligonucleotides corresponding to the wild type (5Ј-AGCCAAGGGGCGGGGAGAGGG-3Ј) or mutant (5Ј-AGCCAAatctg-caGcAGAGGG-3Ј; mutation shown in lowercase type) Sp1-1 binding site of the hLHR promoter were custom made by Gene Probe Technology, Inc. (Gaithersburg, MD). The oligomers were annealed and gel-purified by 12% PAGE gel. 50 g of nuclear extracts isolated from TSA-treated or vehicle-treated JAR cells were incubated with 0.25 g of wild type or mutant oligonucleotide probe in 400 l of binding buffer (60 mM KCl, 12 mM HEPES, pH 7.9, 4 mM Tris-HCl, 5% glycerol, 0.5 mM EDTA, 1 mM dithiothreitol, 1ϫ protease inhibitor mixture) on ice for 45 min. The DNA-protein complexes were then incubated with 40 l of Tetralink TM Avidin Resin (Promega, Madison, WI), which was preequilibrated in the binding buffer for 1 h. The incubation was continued for 1 h at 4°C with gentle rotation. DNA-protein complexes were then washed five times with the binding buffer. 36 l of 2ϫ protein sample buffer (Invitrogen) was added to the avidin-precipitated DNA-protein complex, which was then boiled for 5 min to dissociate the complexes. The proteins were resolved on PAGE followed by Western blot detection using specific antibodies as indicated.
Western Blot and Coimmunoprecipitation-The nuclear proteins (30 g/lane) isolated from TSA-or vehicle-treated JAR cells were subject to Western blot for detection of various transcription factors as shown. The antibodies for acetylated forms of histone H3 or H4 were obtained from Upstate Biotechnology. Antibodies against HDAC-1, HDAC-2, mSin 3A, RbAp46 or RbAp48, and actin were purchased from Santa Cruz Biotechnology.
Coimmunoprecipitation was performed by incubating 100 g of nuclear extracts with 0.5 g of the appropriate immunoprecipitation antibody in 200 l of immunoprecipitation buffer (20 mM Tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, 0.5% Nonidet P-40, 1ϫ protease inhibitor mixture) for 1 h at 4°C with gentle rotation. 30 l of protein A-agarose beads was then added, and the incubation was continued for overnight. Protein A-precipitated protein complex was recovered by brief centrifugation followed by washing for five times with immunoprecipitation buffer. The beads were resuspended in 36 l of 2ϫ protein sample buffer and were boiled for 5 min to release the bound proteins, and the proteins were analyzed by Western blot using the antibodies as indicated.

Histone Deacetylase Inhibitors Activated the Transcription of
Human LHR Gene-To investigate whether transcription of the hLHR gene was subject to modulation through chromatin modification related to histone acetylation and deacetylation, transient transfection studies using reporter gene analysis of the hLHR promoter activity was carried out in JAR cells treated with or without two HDAC inhibitors, TSA and NaB. TSA is the most specific of the HDAC inhibitors yet identified, whereas butyrate affects DNA methylation and other enzyme systems (17,18). Since, in some studies, both TSA and NaB were reported to display pleiotropic effects as cell differentiation inducers in cell lines tested (19), hydroxyurea, a known potent cell differentiation inducer (20), was also evaluated. The hLHR promoter activity was markedly increased in the cells incubated with either TSA or NaB (Fig. 1A). The magnitude of the activation observed in the presence of 100 ng/ml (330 nM) TSA was 40-fold, which more than doubled that elicited by NaB at a much higher concentration (2.5-5.0 mM). In contrast, hydroxyurea only marginally elevated the hLHR promoter activity over the control, and the increase was not dose-related. Thus, the results made TSA a preferred HDAC inhibitor to employ in the subsequent experiments. Further analyses showed a dose-dependent activation of the hLHR promoter activity by TSA (Fig. 1B). A potent induction of the hLHR promoter was observed at 50 ng/ml TSA, and the maximal activation was achieved at the concentration of 100 ng/ml. In temporal studies, TSA caused a major induction of hLHR promoter activity at 6 h (13 Ϯ 1.2-fold) that reached maximal levels at 12 h (36.7 Ϯ 1.8-fold), and were maintained at the 24-h time point (38.4 Ϯ 3.5-fold; Fig. 1C).
In further studies, we evaluated whether the endogenous expression of the hLHR gene governed by its natural promoter could be regulated by TSA. RT-PCR analysis of RNA from TSA-treated or -untreated cells revealed a strong activation of endogenous hLHR gene transcription following a 6-h incubation of the cells with 100 ng/ml of TSA (7-fold induction; Fig.  1D). The hLHR mRNA level was further increased at 12 h after treatment (11-fold), and similar -fold-induction was observed at 24 h. In contrast, the expression of the human ␤-actin gene was not affected. Our results showed clear correlation between the transcriptional responses of the hLHR gene to TSA, exogenously and endogenously. These studies have demonstrated that hLHR gene transcription was markedly induced when histone deacetylase activity was inhibited by TSA and that histone deacetylation caused significant repression of the hLHR gene transcription in these cells.
Specific Localization of Acetylated Histones and Increased Recruitment of RNA Polymerase II to the hLHR Gene Promoter-The initial studies led us to propose that TSA-mediated changes in the balance of histone acetylation and deacetylation might result in local chromatin modification or remodeling of the hLHR gene, which in turn would elicit activation/derepression of the hLHR gene transcription. In order to test this proposal, the effect of TSA treatment on the degree of acetylated histone proteins present in JAR cells was first examined ( Fig. 2A). Both acetylated histones H3 and H4 were undetectable in the absence of TSA and no increases were noted in the cells incubated with TSA for 3 or 6 h. However, at 12 h of treatment with TSA, a strong immunoreactive band of acetylated H4 and a weaker band of acetylated H3 were observed. Further increases were observed at 24 h, when the intensities of the bands for acetylated H3 and H4 were similar. In contrast, the level of actin (used as normalizing protein) was not affected by TSA treatment. In order to determine assembly of the acetylated histones throughout the 5Ј-flanking region of the hLHR gene, chromatin immunoprecipitation was performed. Occupancy of four regions of the 5Ј-flanking sequence by the acetylated histones was analyzed, including one that covered the entire 176-bp hLHR gene promoter as well as a small part of the coding region (ϩ1 to ϩ27), and three other nonoverlapping segments 5Ј to the promoter domain (Fig. 2C). TSA treatment caused a profound induction in the binding of both acety-lated H3 and H4 to the region 1 (Fig. 2B, b and c). Also, weak binding of both acetylated histones to region 2 was detected in the presence of TSA, whereas no binding was noted in its absence. In contrast, no binding to regions 3 and 4 was observed regardless of TSA treatment. The results showed that TSA-induced acetylated histones were associated predominantly with the hLHR gene promoter rather than distributed randomly at its 5Ј-flanking region. This promoter-specific localization of the histone acetylation could provide a more accessible promoter environment for recruitment of component(s) of general transcriptional machinery and RNA polymerase II. In this regard, specific recruitment of RNA polymerase II to the hLHR gene promoter was significantly increased upon TSA treatment (Fig. 2B, d). Taken together, the results indicated a direct link between the inhibition of HDAC activity and the increase of RNA polymerase II recruitment to the hLHR promoter in activation/derepression of the hLHR gene transcription.
Identification of the Sp1-1 Site as Critical for TSA-mediated Activation/Derepression of hLHR Gene Transcription-Previous studies have revealed that the highly GC-rich hLHR promoter harbored two Sp1-activating domains (Sp1(I) and Sp1(II)) and a repressive/activating nuclear orphan receptor binding site (direct repeat; DR) (Fig. 3A, 7 and 8). Moreover, four additional elements bearing weak Sp1/Sp3 binding activities but without detectable function were also identified (domains A 1 , A 2 , A 3 , and S 3 ). 2 In order to identify cis-element(s) responsible for TSA-mediated induction of the hLHR gene promoter activity, reporter gene analyses were performed utilizing the hLHR wild type promoter (WT) and its various mutant constructs. These included the constructs that were devoid of binding of Sp1/Sp3 or of the three nuclear orphan receptors (EAR2, EAR3, and TR4). As shown in Fig. 3, the wild type promoter activity was markedly increased after TSA treatment (construct 2), whereas the Basic vector as a control was only marginally elevated (construct 1). Mutation of the direct repeat element resulted in activated basal promoter activity due to release of the inhibition at this site (construct 3). Moreover, TSA significantly induced the activity of this mutant construct (-fold induction, 35.3 Ϯ 2.36) similarly as for the wild type promoter (-fold induction, 38.8 Ϯ 2.95). Mutation at the four non-functional Sp1/Sp3 binding sites (A 1 , A 2 , A 3 , and S 3 ) did not affect promoter activities in the absence or presence of TSA when compared with the wild type construct (constructs 4 and 5). Thus, these domains did not participate in basal promoter activity or the TSA response. When constructs containing either mutated Sp1-1 or Sp1-2 or double mutations were analyzed, as expected, mutation at either site decreased the basal promoter activity (constructs 6 -8). However, only when the Sp1-1 site was mutated was the induction of the hLHR promoter activity by TSA markedly decreased. The largely impaired TSA responses due to Sp1-1 site mutation were most evident when presented as -fold induction from control (Fig.  3B). Similar results were obtained in analyses using the constructs with simultaneous mutations at all Sp1 binding sites (constructs 9 and 10). Taken together, these studies demonstrated that the Sp1-1 site is critical for activation/derepression of the hLHR gene transcription by HDAC inhibitor TSA.

Binding of Transcription Factors Sp1/Sp3 to the Sp1-1 Site of the hLHR Promoter Was Not
Changed by TSA Treatment-We next investigated whether the Sp1-1 site-dependent activation of the hLHR promoter activity could be attributed to a change of Sp1/Sp3 binding activities to this site or to formation of novel DNA-protein complex(s) after TSA treatment. Electrophoretic mobility shift assay analyses demonstrated that incubation of Sp1-1 site probe with the nuclear extracts from control or TSA-treated JAR cells formed Sp1 and Sp3 binding complexes (Fig. 4A, lanes 1 and 7). The binding specificity was confirmed by cold competition (lanes 2 and 8) and antibody supershift assays (lanes 3-5 and 9 -11) using normal rabbit IgG as a control (N-IgG, lanes 6 and 12). As previously shown (7), the binding was competed by the unlabeled Sp1-1 oligomer DNA (100-fold) and supershifted by Sp1 or Sp3 antibody. Furthermore, the binding of Sp1/Sp3 to the Sp1-1 site was not affected by TSA, since similar binding was observed in the absence or presence of TSA. Also, no additional DNAprotein complex was detected.
Recruitment of Histone Deacetylase Complexes to the hLHR Promoter-The observation that activation of hLHR gene transcription occurred when HDAC-induced deacetylation of the promoter region was inhibited by TSA treatment indicated that HDAC(s) recruited to the hLHR gene caused an unfavorable promoter environment. DAPAs were used to determine whether Sp1/Sp3 were possible candidates to target the non-

FIG. 2. Acetylated H3 and H4 accumulated only in TSA-treated JAR cells and predominantly bound to the human LHR gene promoter.
A, whole cell extracts prepared from JAR cells treated with or without 100 ng/ml TSA for 0, 3, 6, 12, or 24 h were subjected to Western blot analyses using antibodies against the acetylated form of histone H3 or H4. B, soluble chromatin prepared from the cells treated with or without 100 ng/ml TSA for 24 h was precipitated with antibodies against acetylated H3, H4, or RNA polymerase II (RNA Pol II) or normal rabbit IgG, followed by PCR analyses (a-e). Numbers 1-4 refer to regions of the promoter/5Ј-flanking region covered by PCR amplification following the immunoprecipitation. Results of amplification of soluble chromatin before precipitation are shown as control (Input). C, schematic representation of the human LHR gene promoter (Ϫ176 to ϩ1) and its 5Ј-flanking region. The inverted arrows show the regions (1-4) analyzed by PCR using specific pairs of primers (B), which are indicated with numbers relative to the translation initiation codon (ATG, ϩ1).
DNA-binding protein HDACs to the hLHR promoter. The expected binding of Sp1 and Sp3 to the WT but not the mutated Sp1-1 probe (Mutant) was shown in this assay (Fig. 5, right). Histone deacetylases HDAC1 and HDAC2 were specifically pulled out by the wild type Sp1-1 but not by the mutant probe, revealing association of both proteins to the Sp1-1 complex. Also, the possible inclusion in the complex of two corepressor proteins, mSin3A and NcoR, was investigated, since HDAC1/2 were previously reported to associate differentially with these cofactors to regulate target gene expression (21,22). mSin3A was identified as a component of the complex, whereas association of NcoR was not detected. Moreover, when two reported HDAC-interacting proteins, RbAp46 and RbAp48, were evaluated, only RbAp48 bound to the complex. The formation of the complex was sequence-specific, since it was abolished when the Sp1-1 site was mutated. Western blot analyses revealed that all of the relevant proteins evaluated by DAPA assays, including NcoR and RpAp46, were expressed in JAR cells (Fig. 5, left). Thus, the lack of participation of NcoR or RbAp46 was not related to absence of protein expression. It was also noted that the binding of the complex and the expression level of the individual complex component was not affected by the TSA treatment. In addition, no recruitment to the hLHR promoter of several transcription factors with histone acetyltransferase activity, including p300, CBP, or pCAF, was observed (data not shown). Taken together, our findings have indicated the formation of a multiprotein complex including DNA-binding protein(s) and at least four non-DNA-binding cofactors at the Sp1-1 site. More significantly, it was revealed that two histone deacetylases and the corepressor mSin3A but not NcoR were specifically recruited to the hLHR gene promoter.
To determine the order in which the identified components of the complex associated with each other, coimmunoprecipitation assays were performed. This study showed that HDAC1 could be co-precipitated by specific antibody against Sp1 but not by Sp3 antibody (Fig. 6A). A similar observation was obtained for The two repeated arrows show the imperfect direct repeat binding site for the nuclear orphan receptors EAR2, EAR3, and TR4. The wild type hLHR promoter (construct 2) and its various mutants (filled symbols, constructs [3][4][5][6][7][8][9][10] were transiently expressed in JAR cells. Promoterless vector was used as a control (Basic, construct 1). The cells were treated with or without 100 ng/ml TSA for 24 h before termination. Right, relative luciferase activities were normalized to light units/g of protein and expressed as mean Ϯ S.E. of at least 3-6 independent experiments of triplicate wells. B, relative luciferase activities are represented as -fold induction of the activity in the presence of TSA over that observed in the absence of TSA for the constructs indicated. X represents mutation of the indicated element.

FIG. 4. Electrophoretic mobility shift assay analysis of Sp1/Sp3
binding to the Sp1-1 site of the hLHR promoter in the absence or presence of TSA. 32 P-Labeled double-stranded DNA oligomer containing the Sp1-1 site of the hLHR promoter was incubated with nuclear extracts (NE) prepared from JAR cells treated with or without TSA (100 ng/ml) for 24 h. The reaction was carried out in the absence (lanes 1 and  7) or presence of an unlabeled 100-fold excess of wild type Sp1-1 oligomer (lanes 2 and 8) or in the presence of specific antibodies of Sp1 (lanes 3 and 9), Sp3 (lanes 4 and 10), or both (lanes 5 and 11) or in the presence of normal rabbit IgG (N-IgG, lanes 6 and 12).
HDAC2 and RbAp48, whereas a weak interaction between Sp3 and RbAp48 was also detected. Furthermore, mSin3A was not revealed in the presence of Sp1 or Sp3 antibody, indicating that there was no direct protein-protein interaction between mSin3A and the Sp1/Sp3 factors. However, direct association between mSin3A and either HDAC1 or HDAC2 was shown, since it could be co-precipitated by both HDAC antibodies (Fig.  6B). In addition, coimmunoprecipitation using RbAp48 antibody showed that RbAp48 was able to directly interact with both HDAC1 and HDAC2 (Fig. 6B). Taken together, the results indicated that recruitment of HDAC1/2 to the hLHR promoter was mediated through a double-docking action: a direct recruitment by Sp1 but not Sp3 and an indirect binding via their association with RbAp48, which interacted with both Sp1 and Sp3. Targeting of mSin3A to the Sp1-1 site was achieved indirectly through binding of mSin3A to both HDAC1 and HDAC2.
Repression of the hLHR Gene Transcription by HDACs-mSin3A Complex-To elucidate whether the HDACs-mSin3A complex could actively participate in the regulation of the hLHR gene transcription, cotransfection studies of the hLHR promoter and the HDAC1 and HDAC2 constructs were carried out (Fig. 7A). As was shown above (Fig. 1B), TSA dose-dependently activated the hLHR promoter activity in the absence of cotransfected HDACs. Coexpression of HDAC1 with the hLHR promoter in the presence of 10 ng/ml TSA did not cause significant difference when compared with the control group. In contrast, the promoter activity significantly activated with higher doses of TSA (20, 50, 100 ng/ml) was strongly inhibited by HDAC1 (51%). Similar results were observed with cotransfection of HDAC2, where the promoter activity induced by 20 -100 ng/ml of TSA was inhibited by 56 -68%. To determine the putative function of mSin3A, the hLHR promoter construct was cotransfected with mSin3A and HDAC1 or HDAC2 constructs into JAR cells (Fig. 7B). Transfection of mSin3A alone with the hLHR promoter caused a small but significant inhibition of the hLHR promoter activity (12%; *, p Ͻ 0.05). This effect of mSin3A was probably due to its cooperation with endogenous HDACs. In contrast, coexpression of HDAC1 or HDAC2 with the hLHR promoter caused more pronounced reduction of promoter activity, by 45 and 39%, respectively. Furthermore, when mSin3A was cotransfected with HDAC1, the repression of hLHR promoter activity by HDAC1 was significantly enhanced (to 70%). However, cotransfection of mSin3A and HDAC2 did not cause further inhibition of hLHR promoter activity when compared with that elicited by HDAC2 alone. Taken together, these results have shown that both HDAC1 and HDAC2 are potent repressors of the hLHR gene transcription and that mSin3A is a corepressor protein that potentiates the inhibitory function of HDAC1.

DISCUSSION
The present studies have demonstrated that inhibition of HDAC activities by TSA caused potent induction of both hLHR gene promoter activity and its endogenous mRNA expression level in JAR cells. The acetylation of histones induced by TSA was predominantly associated with the hLHR promoter region, to which the recruitment of RNA polymerase II was markedly increased. Furthermore, the Sp1-1 site that is essential for basal promoter activity has been identified as critical for the TSA response, and recruitment of the HDAC1/HDAC2-mSin3A complex to the hLHR promoter was mediated primarily by Sp1 and also by Sp3. Overexpression of HDAC1 or HDAC2 in JAR cells caused repression of the hLHR promoter activity and counteracted the TSA effect. mSin3A associated with HDAC1/2 functions as corepressor for HDAC1. Our studies have demonstrated that the human LHR gene is transcriptionally regulated by histone deacetylase-corepressor complex.
LHR gene expression undergoes prominent changes during gonadal cell maturation and steroid biosynthesis, and the regulatory mechanism(s) whereby this gene is controlled have not been completely elucidated. The marked activation of hLHR promoter activity by TSA revealed that HDACs or HDAC complexes cause significant repression of hLHR gene transcription in JAR cells. The induction/derepression of endogenous hLHR gene expression by TSA, in contrast to the unchanged level of the human ␤-actin gene mRNA, indicates that the TSA-mediated regulation is a gene-selective effect. Results from at least two reports showed that expression of ϳ2% of genes (8 of 340 genes tested) was changed upon TSA treatment in human lymphoid cell lines (23), and 10 and 21 genes of 588 candidates (ϳ5% in total) were up-or down-regulated by TSA in human lung adenocarcinoma cells (24). The notable magnitude of activation of the hLHR promoter activity by TSA treatment indicates that hLHR gene may belong to a small fraction of cellular genes, whose expression is uniquely sensitive to the degree of histone acetylation in chromatin.
The restriction of acetylation of histones to the hLHR promoter region suggests the existence of sequence-specific binding proteins that dictate such promoter-selective localization. Furthermore, the highly enhanced parallel recruitment of RNA polymerase II demonstrates that local chromatin decondensation contributes to the hLHR gene activation/derepression. This finding is consistent with the notion that transcriptionally competent chromatin correlates with increased accessibility of protein complexes of the basal transcriptional machinery (25,26).
The evidence that the Sp1-1 site is responsible for mediating the TSA effect (Fig. 2) illustrates a novel function for this Sp1/Sp3 binding domain besides its central contribution to the basal promoter activity of the hLHR gene. Recognized primarily as a constitutive regulatory element in modulation of many target gene expression, Sp1 sites have been found to be involved in tissue-specific gene expression (27,28) and in control of transcription in response to a number of different stimuli (29 -31). A well known example is illustrated in regulation of the Cdk (cyclin-dependent kinase) inhibitor p21 WAF1/Cip1 gene, where Sp1 sites are central in integrating the promoter responses of diverse signals including transforming growth factor-␤, phorbol esters, and HDAC inhibitors (32)(33)(34)(35), indicating that Sp1 site-dependent transcription could be highly regulated. On the other hand in our studies, the unchanged binding pattern to the Sp1-1 site in the presence or absence of TSA (Fig.  4) indicates that activation of the hLHR promoter activity is not related to increased DNA binding activity of Sp1/Sp3, or to formation of a new complex(es). Moreover, reversible acetylation and deacetylation of nonhistone proteins including some transcription factors has provided insights into histone acetyltransferase/HDAC-regulated gene expression (36). Modification of Sp factor(s) by this process could be a mechanism to cause a change in its activity and/or its interaction with other proteins and constituents of the basic transcriptional complex. Although very little is known about modification of Sp factors aside from phosphorylation and glycosylation, a recent report has described regulation of transcriptional activity of Sp3 by acetylation (37).
DAPA analyses have shown that a multiprotein complex containing HDAC1, HDAC2, mSin3A, and RbAp48 was specifically recruited to the hLHR gene promoter region. These results have provided a molecular basis to explain the HDAC1/ 2-mSin3A complex-induced repression of hLHR gene transcription in the functional studies as well as in the TSAinduced promoter-selective histone acetylation. Furthermore, the observation that no effect of NcoR was detected indicates that this corepressor does not participate in HDAC-mediated repression of the hLHR gene. Corepressors NcoR, SMRT, and mSin3A have been shown to associate with HDAC1 to potentiate the function of HDAC1 in repression of several hormoneresponsive and Mad-Max target genes (22, 38 -41). However, there is some evidence showing that NcoR or SMRT were not endogenously associated with HDAC1-containing complexes in several cell lines tested (42,43). On the other hand, it has been found that in leukemia cells, NcoR and SMRT interact directly with class II histone deacetylases HDAC4/5 instead of HDAC1 (44). Therefore, differential interaction between individual corepressors and different HDAC complexes may provide alternative ways for repression of specific target genes. In addition, the finding in this study that the HDAC1/2-interacting protein RbAp48 was recruited to the hLHR promoter is consistent with the notion that this protein is a component of a histone deacetylase core complex, which includes HDAC1 and HDAC2 (45,46). The weak interaction observed between RbAp48 and Sp3 also indicates a possible functional role for this protein either singly or by tethering the HDAC complex to the Sp3 factor, of which no direct interaction was detected (Fig. 6). However, it remains to be determined whether RbAp48 contributes to the hLHR gene regulation.
Recruitment of HDAC1/2-mSin3A complex to the hLHR gene promoter through direct interaction between HDAC1/2 and Sp1 but not Sp3 demonstrates that Sp1 and Sp3 contribute differentially in HDAC-mediated repression of the hLHR gene, although both have similar binding activities to the Sp1-1 site and are effective transcriptional activators of the hLHR gene promoter activity (7). Differential participation of Sp1 and Sp3 has been implicated in modulation of target gene expression in response to multiple signals (for a review, see Ref. 47). It was reported that Sp1 but not Sp3 was responsible for transforming growth factor-␤ mediated activation of the p21 gene (32). Whereas both Sp1 and Sp3 were required for regulation of the p21 gene by a HDAC inhibitor, suberoylanilide hydroxamic acid, in NIH3T3 cells (48), Sp3 but not Sp1 mediates the TSA response of the same gene in the human osteosarcoma cell line MG63 (49). The observed differences were attributed to the various cell systems and HDAC inhibitors used in the experiments. However, different mechanisms may also be implicated.
This study has identified the hLHR gene as a target for regulatory repression by the HDACs-mSin3A complex that is operative in the control of hLHR gene transcription. Expression of the LHR gene is subject to tonic repression by deacetylation of its gene promoter. The regulated derepression of such control of the LHR gene, through as yet unidentified signal inputs, may provide functional control during induction and cyclical changes in the differentiation, growth, and development of gonadal cells. Histone acetylation/deacetylation integrates the modulation of an array of genes that are actively involved in cell proliferation and differentiation. The finding that hLHR gene expression is subject to HDAC-mediated regulation extends our understanding of the essential role of the LHR in its contribution to gonadal cell development and differentiation.