Cooperative Interaction between the Basic Helix-loop-helix Transcription Factor dHAND and Myocyte Enhancer Factor 2C Regulates Myocardial Gene Expression*

Cardiac-restricted transcription factors dHAND and myocyte enhancer factor 2C are expressed in the developing heart and activate several cardiac promoters. However, their regulatory mechanisms are still to be understood. To elucidate their exact regulatory functions, we have developed an RNA interference strategy to specifically inhibit dHAND and myocyte enhancer factor 2C protein production in H9c2 cells, which are derived from rat embryonic heart. Expression of endogenous cardiac genes atrial natriuretic peptide and (cid:1) -my-osin heavy chain was down-regulated in H9c2 cells lack-ing both dHAND and myocyte enhancer factor 2C, indicating that these factors are required for the main-tenance of the cardiac genetic program. Consistent with these, expression of atrial natriuretic peptide and (cid:1) -my-osin heavy chain was up-regulated in H9c2 cells, which overexpressed dHAND and myocyte enhancer factor 2C. In addition, dHAND and myocyte enhancer factor 2C interact to synergistically activate atrial natriuretic peptide and (cid:1) -myosin heavy chain transcription. Fur-thermore, chromatin immunoprecipitation analysis in (300 m M sodium chloride and 300 m M sodium citrate), and then immobilized by incubating the membrane for 2 h at 80 °C. Blots were hybridized with random prime-labeled rat cDNA probes for ANP, (cid:1) -MHC, and glyceraldehyde-3-phosphate dehy-drogenase (GAPDH). The hybridization signal obtained with the GAPDH probe was used to correct differences in loading and/or transfer efficiencies. The probes were labeled with [ 32 P]dCTP by a Prime-a-Gene labeling system (Promega). The hybridization and washing were performed according to standard methods. The membranes were then exposed to x-ray film for 24 h at (cid:1) 70 °C. The ANP probe was obtained by using the specific oligonucleotide primers 5 (cid:2) -AACCAGAGAGT- GAGCCGAGACAGCAAA-3 (cid:2) and 5 (cid:2) -TGCTCGAGCAGATTTGGCTGT-TATCTT-3 (cid:2) , which are based on the nucleotide sequence of rat ANP cDNA. The primers for (cid:1) -MHC probe were 5 (cid:2) -TGGCGCCAAGCA-GAAAATGCACGA-3 (cid:2) and 5 (cid:2) -ACAGGCAAAGTCAAGCATTCATAT-3 (cid:2) . Immunoprecipitation and Western Blot Analysis— H9c2 cells were lysed at 4 °C with a gentle rotation in lysis buffer (50 m M Tris-HCl, pH 7.5, 150 m M NaCl, 0.5% Triton-100) containing protease inhibitors (1 m M phenylmethylsulfonyl fluoride, 1 (cid:2) g/ml leupeptin, 1 (cid:2) g/ml pepsta-tin, and 1 (cid:2) g/ml aprotinin). Lysates were cleared by centrifugation at 12,000 (cid:4) g for 10 min. Lysate proteins were immunoprecipitated

Cardiac-restricted transcription factors dHAND and myocyte enhancer factor 2C are expressed in the developing heart and activate several cardiac promoters. However, their regulatory mechanisms are still to be understood. To elucidate their exact regulatory functions, we have developed an RNA interference strategy to specifically inhibit dHAND and myocyte enhancer factor 2C protein production in H9c2 cells, which are derived from rat embryonic heart. Expression of endogenous cardiac genes atrial natriuretic peptide and ␣-myosin heavy chain was down-regulated in H9c2 cells lacking both dHAND and myocyte enhancer factor 2C, indicating that these factors are required for the maintenance of the cardiac genetic program. Consistent with these, expression of atrial natriuretic peptide and ␣-myosin heavy chain was up-regulated in H9c2 cells, which overexpressed dHAND and myocyte enhancer factor 2C. In addition, dHAND and myocyte enhancer factor 2C interact to synergistically activate atrial natriuretic peptide and ␣-myosin heavy chain transcription. Furthermore, chromatin immunoprecipitation analysis in H9c2 cells treated with phenylephrine showed that dHAND and myocyte enhancer factor 2C protein complex bind to the A/T sequence on atrial natriuretic peptide promoter. Taken together, these results not only suggest that the complex cis-trans interaction of dHAND, myocyte enhancer factor 2C, and the target gene may fine-tune gene expression in cardiac myocytes but also provide a molecular paradigm to elucidate the mechanisms of action of dHAND and myocyte enhancer factor 2C in the developing heart.
The basic helix-loop-helix (bHLH) 1 transcription factors are a large family of genes that play important roles in determina-tion and differentiation of diverse cell types including skeletal cells, hematopoietic cells, some neuronal cells, and cardiomyocytes. The bHLH proteins contain the basic region and the HLH domain. The basic region can bind to the E-box consensus sequence (CANNTG), and the HLH domain can mediate homodimerization or heterodimerization. This family can be divided into several subgroups based either on their structure, DNA binding activity, or expression pattern during development. Tissue-specific bHLH proteins compose the largest subgroup and bind the E-box element (CANNTG) of target gene by dimerization with a ubiquitously expressed bHLH factors, the E protein. In addition, transcriptional activity of the tissuespecific bHLH proteins requires association with other bHLH proteins or non-bHLH proteins (1). For example, in skeletal muscle cells the muscle bHLH protein MyoD regulates skeletal muscle genes through direct interaction with MEF2A, a member of myocyte enhancer factor 2 (MEF2) family (2). dHAND is a bHLH protein expressed in deciduum, heart, autonomic nervous system and neural crest derivatives. Several laboratories cloned it in 1995 and gave it different names such as HAND2, Th2, and exd. The phenotype of dHAND null mice demonstrates the essential role of dHAND in the formation of the right ventricle, the trabeculae, and the neural crestderived aortic arches. dHAND is also implicated in the regulation of chamber specification, cardiac looping, and cardiac neural crest (3).
Atria1 natriuretic peptide (ANP) is an important cardiac hormone expressed during heart development, and it regulates the blood volume and pressure. Reduced ANP expression in dHAND Ϫ/Ϫ mice suggests that dHAND is potentially important for the regulation of ANP. The ANP promoter harbors four functional E-box elements that might bind dHAND (4).
Another transcription factor that is required for cardiogenesis is myocyte enhancer binding factor 2C (MEF2C), which belongs to the MADS (MCM1, agamous, deficiens, and serum response factor) box family. In mammals, the MADS-box family is composed of four members, MEF2A, MEF2B, MEF2C, and MEF2D. These transcription factors contain an almost identical N-terminal region necessary to mediate DNA binding, and they are divergent in their C-terminal region, which is implicated in transcription activation. The MEF2 proteins can form homo-and heterodimers to bind the consensus DNA sequence (T/C)TA(A/T) 4 TA(G/A) to regulate gene expression. Inactivation of MEF2C in mouse leads to cardiac morphogenetic defects, vascular abnormalities, and embryonic lethality, which demonstrates an essential role of MEF2C in heart development (5).
Transcriptional regulation of cardiac development requires the coordinated expression of several factors in a temporally and spatially defined manner. These cardiac-restricted transcription factors form multiprotein complexes around cardiacrestricted genes through interaction to regulate target genes. But the mechanism by which these genes regulate has yet to be determined. Here we demonstrate that transcriptional factors dHAND and MEF2C interact to synergistically activate expression of cardiac gene promoters in H9c2 cells. Inhibiting expression of dHAND and MEF2C in H9c2 cells result in downregulation of ANP and ␣-myosin heavy chain (␣-MHC). Moreover, overexpression of dHAND and MEF2C in H9c2 cells result in up-regulation of ANP and ␣-MHC. In addition, the functional interaction on ANP promoter is mediated through the MEF2C-binding site but not dHAND-binding sites. This interaction suggests that dHAND protein was recruited by MEF2C to A/T sequence on ANP promoter to activate transcription.

MATERIALS AND METHODS
Plasmids-The pcDNA3-MEF2C expression vector was a gift of Dr. Coralie Poizat and Larry Kedes (University of Southern California). The pcDNA3-His2B-dHAND expression vector, the ANP reporter plasmid (Ϫ638 base pairs upstream from the transcriptional start site), and the ␣-MHC-luciferase (Ϫ330 base pairs upstream from the transcriptional start site) were gifts from Jeffery D. Molkentin and Dr. Yan-Shan Dai (Cincinnati, OH).
RNAi Plasmids Construct-The U6/dHAND and U6/MEF2C RNAi constructs were designed following pSilencer neo™ Instruction Manual (Ambion). Briefly, 21-nucleotide-long inverted repeats separated by a 9-nucleotide linker were inserted downstream of the U6 promoter. Six thymidines were inserted downstream of the antisense strand to provide a stop signal for the RNA polymerase III. The sense strand of the hairpin was homologous to a 21-nucleotide region in the target mRNA (MEF2C or dHAND) and were analyzed by BLAST research to ensure that they did not have significant sequence homology with other genes. The sense strand of the MEF2C small interfering RNA (siRNA) was designed to be homologous to a region covering the MEF2C-coding sequence (nucleotides 538 -558 bp) from the start codon of the mouse sequence (accession number L13171.1). The dHAND siRNA was designed to target a region near the C terminus of the dHAND-coding sequence (nucleotides 529 -549 bp of the rat sequence NM_022696). Nucleotide overhangs to the BamHI and HindIII restriction sites were added to the 5Ј and 3Ј ends of the DNA oligonucleotides, respectively. Sequences containing DNA templates for the synthesis of siRNAs are listed below. siRNA top strand of dHAND is 5Ј-GATCC AAA GAG GAG AAG AGG AAG AAA TTC AAG AGA TTT CTT CCT CTT CTC CTC TTT TTT TTTGGA AA-3Ј, and siRNA bottom strand of dHAND is 5Ј-AGC TTT TCC AAA AAA AAA GAG GAG AAG AGG AAG AAA TCT CTT GAA TTT CTT CCT CTT CTC CTC TTTG-3Ј. siRNA top strand of MEF2C is 5Ј-GATCC AAT AGT ATG TCT CCT GGT GTA TTC AAG AGA TAC ACC AGG AGA CAT ACT ATT TTT TTT GGA AA 3Ј, and siRNA bottom strand of MEF2C is 5Ј-AGC TTT TCC AAA AAA AAT AGT ATG TCT CCT GGT GTA TCT CTT GAA TAC ACC AGG AGA CAT ACT ATTG-3Ј. These oligonucleotides were chemically synthesized and mixed in equimolar amounts, heated for 5 min at 95°C, then gradually cooled to 37°C for 1 h. The annealed siRNA templates were then ligated into pSilence2.1-U6 vector. pSilence2.1-U6 was purchased from Ambion.
Transient Transfection Assays-H9c2 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM glutamine, streptomycin (10 g/liter), and penicillin (10 g/liter). All transfections were performed with Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol in six-well plates. 24 h after plating, each well was transfected with 1.0 g of reporter DNA ANP-Luc or ␣-MHC-Luc and 0.5 g of expression vectors pcDNA3-MEF2C or/and pcDNA3-HisB-dHAND. ␤-Galactosidase (20 ng in each well) was used as an internal control. The total amount of DNA was kept constant using pcDNA3.1 empty vector. Luciferase activity was measured 48 h after transfection in a luminometer using FOLAR star reporter assay system (BMG) and normalized to ␤-galactosidase activity. Each value presented is the average of triplicate samples and is representative of multiple independent experiments (n Ն 3). The data were statistically analyzed with Student's t test.
Northern Blot Analysis-After transfection with dHAND and MEF2C or RNAi plasmid of dHAND and MEF2C or empty vector pcDNA3 and pSilencer2.1-U6, total cellular RNA was isolated from H9c2 with Trizol reagent (Invitrogen). Twenty micrograms of total RNA were size-fractionated on formaldehyde-agarose gel (1%), transferred to a nylon membrane (Hybond N ϩ ; Amersham Biosciences) by capillary blotting in the presence of 20ϫSSC (300 mM sodium chloride and 300 mM sodium citrate), and then immobilized by incubating the membrane for 2 h at 80°C. Blots were hybridized with random prime-labeled rat cDNA probes for ANP, ␣-MHC, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The hybridization signal obtained with the GAPDH probe was used to correct differences in loading and/or transfer efficiencies. The probes were labeled with [ 32 P]dCTP by a Prime-a-Gene labeling system (Promega). The hybridization and washing were performed according to standard methods. The membranes were then exposed to x-ray film for 24 h at Ϫ70°C. The ANP probe was obtained by using the specific oligonucleotide primers 5Ј-AACCAGAGAGT-GAGCCGAGACAGCAAA-3Ј and 5Ј-TGCTCGAGCAGATTTGGCTGT-TATCTT-3Ј, which are based on the nucleotide sequence of rat ANP cDNA. The primers for ␣-MHC probe were 5Ј-TGGCGCCAAGCA-GAAAATGCACGA-3Ј and 5Ј-ACAGGCAAAGTCAAGCATTCATAT-3Ј.
Immunoprecipitation and Western Blot Analysis-H9c2 cells were lysed at 4°C with a gentle rotation in lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% Triton-100) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, 1 g/ml pepstatin, and 1 g/ml aprotinin). Lysates were cleared by centrifugation at 12,000 ϫ g for 10 min. Lysate proteins were immunoprecipitated overnight at 4°C with anti-dHAND antibody by a gentle rotation. The immune complex (lysate proteins and antibody) was added to the spin cup containing the equilibrated immobilized protein A. Then it was incubated for at least 10 min, the tube was centrifuged, and 3 washes were done with binding/washing buffer (0.14 M NaCl, 0.008 M sodium phosphate, 0.002 M potassium phosphate, and 0.01 M KCl, pH 7.4). The immunoprecipitated protein was recovered using elution buffer and centrifugation. The eluted protein was mixed with sample buffer and loaded on a SDS-polyacrylamide electrophoresis gel. Resolved proteins were transferred on Hybond-polyvinylidene difluoride membrane (Amersham Biosciences), and Western blots were performed. The blot was incubated with MEF2C polyclonal antibody (Santa Cruz Biotechnology) and detected with ECL chemiluminescence reagents (Santa Cruz) following the recommended protocol. The seize classic (A) immunoprecipitation kit was purchased from Pierce.
Chromatin Immunoprecipitation (ChIP) Assay-H9c2 Cells were grown to 95% confluence in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Cells were collected and crosslinked with 1% formaldehyde at 37°C for 10 min, then were rinsed with ice-cold phosphate-buffered saline twice and centrifuged for 5 min at 2000 ϫ g. Cells were then resuspended in 0.2 ml of lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-Cl, pH 8.1, protease inhibitor) and sonicated 12 times for 10 s each followed by centrifugation for 10 min. Supernatants were collected and diluted in buffer (1.1% Triton X-100, 0.01% SDS, 1.2 mM EDTA, 167 mM NaCl, 16.7 mM Tris-HCl, pH 8.1) followed by immunoclearing with 80 l of salmon sperm DNA/protein A-agarose for 2 h at 4°C. Immunoprecipitation was performed overnight at 4°C with dHAND antibody or MEF2C antibody (Santa Cruz). After immunoprecipitation, 60 l of salmon sperm DNA/protein A-agarose was added, and the incubation was continues for another hour. Precipitates were washed sequentially in the following three different washing buffers for 5 min each; a low salt immune complex washing buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 150 mM NaCl), a high salt immune complex washing buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl), and a LiCl immune complex washing buffer (0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA,10 mM Tris-HCl, pH 8.1). Precipitates were washed two times with Tris-EDTA buffer then used to PCR assay for their binding sequences. For PCR assays precipitates were extracted 3 times with 1% SDS, 0.1 M NaHCO 3 . Elutions were pooled and heated at 65°C for at least 6 h to reverse the formaldehyde cross-linking. DNA fragments were purified; 1 l from 50 l of DNA extract was reserved for 25-30 cycles of PCR amplification. Chromatin immunoprecipitation assay kit was purchased from Upstate Biotechnology. The primers for the E-box were 5Ј-TCCACCCACGAGGCCAATGAAT-3Ј (sense primer) and 5Ј-CCGCTCGAGGATGTTTGCTGTCTCGGCTC-3Ј (antisense primer). The primers for the A/T site were 5Ј-GACCTGTATCATGTTG-GCTTCCTGG-3Ј (sense primer) and 5Ј-ATTGGCCTCGTGGGTGGAC-CTCTGG-3Ј (antisense primer).

RESULTS
Synergistic Activation of the ANP and ␣-MHC Promoter by dHAND and MEF2C-Both dHAND and MEF2C regulate cardiac promoters and are expressed in cardiomyocytes. To gain great insight into the transcriptional mechanism regulation of cardiac-specific gene expression by dHAND and MEF2C, we investigated whether these two transcription factors could functionally interact to induce the activation of the ANP and ␣-MHC promoters. We employed H9c2 cells to address this issue. The H9c2 cell line is derived from 13.5-embryonic day BDIX rat heart tissue and was cloned by Kimes and Brandt in 1976 (6). ANP, ␣-MHC, dHAND, and MEF2C are all endogenously expressed in H9c2 cells.
Cotransfection of both dHAND and MEF2C in H9c2 cells leads to a 7-and 5-fold increase in synergistic activation of the ANP and ␣-MHC promoter respectively (Fig. 1). These data suggest that the ANP and ␣-MHC genes are synergistically activated by co-expression of dHAND and MEF2C. Collectively, these results indicate that the transcription factors dHAND and MEF2C can functionally synergize to activate the promoters of the ANP and ␣-MHC gene.
dHAND and MEF2C Synergistically Activate Expression of Endogenous ANP and ␣-MHC in H9c2 Cells-To further assess the function of dHAND and MEF2C in the activation ANP and ␣-MHC, we examined the effects of overexpression of dHAND and MEF2C on the expression of the endogenous ANP and ␣-MHC gene in H9c2 cells. As shown in Fig. 2, expression of the cardiac-specific ANP gene was enhanced in H9c2 cells transfected with dHAND and MEF2C expression vector compared with H9c2 cells transfected with the empty vector pcDNA3.1. Furthermore, when repressing dHAND and MEF2C expression by transfecting the H9c2 cells with RNAi plasmid of dHAND and MEF2C, the expression of the ANP gene was reduced compared with the transfection of H9c2 cells with pSi-lencer2.1-U6 ( Fig. 2A). Similar results were obtained for the ␣-MHC gene ( Fig. 2A). Importantly, endogenous expression of dHAND and MEF2C was dramatically reduced in H9c2 cells transfected with the RNAi plasmid of dHAND and MEF2C, respectively (Fig. 2B). Moreover, both expression of dHAND

FIG. 2. Northern blot analysis of ANP and ␣-MHC mRNA levels and Western blot analysis of the dHAND and MEF2C expression in transfected H9c2 cells. A, Northern blot analysis of ANP and
␣-MHC mRNA levels in transfected H9c2 cells. H9c2 cells were transfected with dHAND and MEF2C to enhance their expression, or the cells were transfected with RNAi plasmid to squelch dHAND and MEF2C expression. H9c2 cells transfected with pcDNA3.1 and pSi-lencer2.1-U6 served as control. 48 h after transfection the cells were harvested for the isolation of total RNA. Total RNA was resolved on a 1% agarose gel and transferred to nylon membranes, blots were hybridized with 32 P-labeled ANP and ␣-MHC and normalized to the signal from the GAPDH probe. a, the effects of overexpression and inhibition of dHAND and MEF2C on ANP mRNA levels were examined. The results showed that ANP mRNA levels were enhanced in H9c2 cells, which overexpressed dHAND and MEF2C. Moreover, the level of ANP mRNA was reduced by the inhibition of dHAND and MEF2C. b, Northern blot analysis showing that ␣-MHC mRNA levels were regulated by dHAND and MEF2C. Overexpression of dHAND and MEF2C up-regulated ␣-MHC mRNA levels, whereas inhibiting the expression of dHAND and MEF2C down-regulated ␣-MHC mRNA levels. c, GAPDH mRNA levels were used to correct the differences in loading and/or transfer efficiencies. B, Western blot assays showing that RNAi dHAND and RNAi MEF2C blocks, respectively, the expression of endogenous dHAND and MEF2C protein. On the other hand, the expression of dHAND and MEF2C is enhanced in H9c2 cells transfected with dHAND and MEF2C. H9c2 cells were transfected with the RNAi plasmid of dHAND and MEF2C or RNAi vector control pSilencer2.1-U6 or the plasmid vector of dHAND and MEF2C. Protein lysates from transfected H9c2 cells were run through an SDS-polyacrylamide electrophoresis, then transferred onto a Hybond-polyvinylidene difluoride membrane, and a Western blot was performed. The blot was incubated with an antibody and detected with ECL chemiluminescence reagents. ␤-Actin served as a loading control. and MEF2C were increased in transfected H9c2 cells with dHAND and MEF2C. These findings suggest that dHAND and MEF2C maintain transcription of ANP and ␣-MHC gene.
dHAND and MEF2C Physically Interact in Vivo-Our previous studies indicate that dHAND directly interacts with fulllength MEF2C in vitro (7). To examine whether dHAND and MEF2C interact in vivo, co-immunoprecipitation experiments were performed in H9c2 cells. Protein lysates from H9c2 cells were first incubated with dHAND antibody, then protein A was used to precipitate dHAND from protein lysates, which was resolved in a SDS-PAGE and subjected to Western blot with anti-MEF2C antibody. The data demonstrated that MEF2C interacted with dHAND in vivo but not with preimmune goat serum (Fig. 3). These data suggest that MEF2C and dHAND physically interact in vivo to form a protein complex.
dHAND⅐MEF2C Protein Complex Bind to the A/T Sequence of ANP Promoter in H9c2 Cells Treated with Phenylephrine (PE)-Our previous studies show that the dHAND⅐MEF2C protein complex binds to the A/T sequence of ANP promoter in unstimulated H9c2 cells (7), which also suggests that in the H9c2 cell line the interaction of dHAND and MEF2C takes place in the A/T sequence of ANP promoter. To further address whether the dHAND and MEF2C complex change their binding sites on ANP promoter under PE stimulation, a serial of ChIP experiments was performed in H9c2 cells treated with PE. PE is a well known stimulator of cell enlargement (hypertrophy) in rat ventricular myocytes. This hypertrophic response induces the expression of ANP (8). To further examine whether dHAND⅐MEF2C protein complexes are assembled on one or both sites of ANP promoter, we divided the soluble chromatin into two aliquots; one was immunoprecipitated with dHAND antibodies and the other with MEF2C antibodies. Both of the eluted immunoprecipitates were amplified by the two site sequences of the ANP promoter (Fig. 4). The data demonstrate that both dHAND and MEF2C can occupy the DNA fragment from Ϫ628 to Ϫ494 bp, which harbors the A/T sequence (Fig. 4A), whereas they do not assemble on the DNA fragment from Ϫ511 to ϩ62 bp, which harbors the E-box sequence (Fig. 4B). All these results demonstrate that the dHAND⅐MEF2C protein complex still binds to the A/T sequence of ANP promoter in H9c2 cells treated with PE. DISCUSSION In this study we have obtained the following results. 1) dHAND and MEF2C interact to synergistically activate transcription of ANP and ␣-MHC in H9c2 cells, 2) ANP and ␣-MHC mRNA levels are reduced when inhibiting the expression of dHAND and MEF2C, and 3) dHAND⅐MEF2C synergy is dependent on MEF2C DNA binding, and dHAND protein is recruited by MEF2C to the A/T sequence on the ANP promoter to activate ANP transcription.
Roles of MEF2C and dHAND in Regulating Heart Development-MEF2C is a member of the MEF2 transcriptional factors family that contains MADS and MEF2 domains to mediate DNA binding and dimerization. These MEF2 transcriptional factors have been shown to be expressed at high level in all muscle cells, but they are also found in brain and lymphoid tissue (9). They are important regulators of muscle-specific gene expression and differentiation of all three muscle lineages. For example, during myogenesis, MEF2A and muscle bHLH proteins cooperatively activate skeletal muscle genes. Thus, skeletal myogenesis is mediated by MEF2A, which can initiate the developmental cascade. In fact, the MEF2 family directly associates with cell-specific transcriptional factors to regulate the determination, proliferation, and differentiation of cardiac muscle, smooth muscle, and neural cells. They control the development of multiple tissues including heart, vasculature, neural tubes, and skeletal muscle (2). MEF2C is expressed in heart, skeletal muscle, spleen, and brain. In the developing mouse heart, MEF2C is initially expressed in the precardiogenic mesoderm beginning at embryonic day 7.75 (E 7.75). Consistent with its expression pattern, inactivation of the MEF2C gene in mice leads to cardiac morphogenetic defects, vascular abnormalities, and lethality by E 9.5, suggesting that MEF2C functions as a critical regulator of cardiac development. Indeed, MEF2C plays a critical role in the looping of the cardiac tube, development of the right ventricle, and expression of a subset of cardiac muscle genes (10).
The direct transcriptional targets of MEF2C in regulation of diverse cell fates are largely unknown. Four members of the MEF2 transcriptional factor family bind to a DNA sequence element, referred to as an A/T-rich sequence, which consists of the consensus DNA sequence (T/C)TA(A/T) 4 TA(G/A). MEF2C forms homodimers and heterodimers that interact with the A/T-rich sequence element to directly promote transcriptional activation of the target gene. Here we demonstrated that MEF2C functions in concert with dHAND as a mechanism of enhancing cardiac-specific ANP and ␣-MHC gene transcription. This synergy between MEF2C and dHAND was independent of the ability of dHAND to bind the E-box on ANP promoter, and it required the dHAND⅐MEF2C protein complex to bind the A/T-rich sequence element on the ANP promoter. These observations suggest that MEF2C can participate in programming the cardiac gene expression through combinatorial interactions with other transcription factors such as dHAND. These results also indicate that dHAND can function in regulating ANP gene expression independently of the E-box sequence elements.
Like MEF2C, dHAND is also thought to function in regulating cell type-specific gene expression through combinatorial interactions with other transcriptional factors. dHAND is a member of a large transcription factors family that contains the bHLH domain which mediates DNA binding and dimerization. These transcription factors are expressed in a wide array of eukaryotic organisms from yeast to humans. For example, dHAND is expressed in numerous tissues including the heart, limbs, and multiple neural crest-derivatized tissues during development. Most notably, dHAND forms both homodimers with itself and heterodimers with the ubiquitously expressed E12/E47 bHLH proteins or with non-bHLH proteins to enhance transcriptional activation (4,11). Here we demonstrated that dHAND was recruited by MEF2C to potentiate cardiac-specific ANP gene expression. dHAND has also been reported to inter- FIG. 3. Immunoprecipitation analysis showing that dHAND and MEF2C physically interacted in vivo to form a protein complex in H9c2 cells. Anti-dHAND antibody immunoprecipitated (IP) lysates from H9c2 cells. The immunoprecipitate was resolved by SDS-PAGE, transferred on a Hybond-polyvinylidene difluoride membrane, and probed with anti-MEF2C antibody. Proteins were visualized using a rabbit anti-goat secondary antibody conjugated to horseradish peroxidase and a ECL chemiluminescence detect system. The result showed that dHAND could interact with MEF2C, but preimmune goat serum alone failed to interact with MEF2C. act with transcriptional factors NKX2.5 and GATA4, which together are co-expressed only in the myocardium. All these combinatorial interactions result in synergistic activation of the ANP promoter.
We also did RNAi and Northern blot analysis to address the roles of dHAND and MEF2C on transcriptional activation of ANP and ␣-MHC. RNAi is the process of gene-silencing whereby siRNAs hybridize to their cognate mRNA to induce the specific degradation of the target mRNA (13). U6 promoterdriven RNAi vectors of dHAND and MEF2C were used to induce hairpin RNA-triggered RNAi and characterize the role of the two transcription factors in H9c2 cells. We found that expression of dHAND and MEF2C hairpin RNA leads to the efficient inhibition of endogenous dHAND and MEF2C protein expression in H9c2 cells. Meanwhile, ANP and ␣-MHC mRNA levels were down-regulated when suppressing the expression of dHAND and MEF2C. Accordingly, our data reveal that dHAND and MEF2C play critical roles in regulating the activity of ANP and ␣-MHC promoters.
Mechanism for MEF2C and dHAND Transcriptional Regulation of ANP-ANP and MHC play important roles during heart development. For example, ANP is a cardiac peptide hormone that possesses significant diuretic, natriuretic, and vasodilatory activities and plays a critical role in the maintenance of blood pressure and sodium balance. During embryonic development, the ANP gene is expressed in both the atrium and the ventricle, but after birth, the expression of ANP is mainly in the cardiac atria. However, if the ventricle is subjected to diverse stimuli causing ventricular hypertrophy, including growth factors, cytokines such as phenylephrine or angiotensin II, mechanical stretch, and activators of protein kinase C, the ventricular ANP gene is stimulated and starts to be expressed. ANP, once released into the circulation, cause natriuresis, diuresis, and vasodilation. Thus, ANP plays a potentially important role in hemodynamic regulation during hypertension. Because expression of ANP is abundant in ventricular cells during embryogenesis and subsequently is extinguished during normal adult development, the reexpression of ANP within hypertrophied ventricular myocytes has been considered to be representative of the induction of an embryonic gene program (14).
␣-MHC, another cardiac-specific gene, encodes a gene for a cardiac contractile protein and plays an important role in the speed of heart contraction because it has a characteristic adenosine triphosphatase (ATPase) activity, which is correlated with the energy transduction responsible for the generation of force in muscle. The myosin heavy chains are encoded by a highly conserved multigene family, and their expression is tissue-and developmental stage-specific. In cardiac muscle, two distinct MHC genes, ␣and ␤-MHC isoforms, have been characterized. Both cardiac MHC genes are coexpressed in the early heart tube when septation and the cardiac compartments form; the ␣-MHC gene is expressed only in the atria, whereas ␤-MHC expression is restricted to the ventricles. At birth or in the adult the ␣-MHC mRNA becomes the predominant MHC transcript in the ventricle (15).
Although ANP and ␣-MHC play important roles during heart development, little is understood in the mechanisms of regulation in cardiomyocytes. Delineation of the ANP and ␣-MHC regulatory elements has further yielded insight into the transcriptional regulation of a cardiac-specific gene. ANP promoter harbors four copies of the E-box sites and an A/T-rich sequence. ␣-MHC promoter harbors two copies of the E-box sites and an A/T-rich sequence. Interactions between these cis-elements and cardiac-specific transcriptional factors finetune transcription of ANP and ␣-MHC genes. Here we demonstrated that dHAND interacts with MEF2C to potentiate cardiac-specific expression of ANP and ␣-MHC. Most notably, dHAND protein was recruited by MEF2C to an A/T sequence on the ANP promoter to activate ANP transcription in H9c2 FIG. 4. Chromatin immunoprecipitation assay showing that dHAND⅐MEF2C protein complex binds an A/T-rich sequence of ANP promoter in H9c2 cells. A, occupancy of the A/T element of ANP promoter by dHAND and MEF2C was measured by ChIP analysis. Soluble chromatin from H9c2 cells treated with phenylephrine was immunoprecipitated (IP) with antibodies against dHAND or against MEF2C. The final DNA extractions were amplified using pairs of primers that cover the regions of the A/T-rich sequence of ANP promoter as indicated. The results show that both dHAND and MEF2C are assembled on this region. B, ChIP experiments were done to examine whether the dHAND⅐MEF2C complex is assembled on the E-box sequence of the ANP promoter. ChIP experiments were performed as described above (A); DNA extractions were amplified using primers that cover the regions of E-box element of ANP promoter. The results show that neither dHAND nor MEF2C is assembled on this region. cells treated with PE. Consistent with our results, Thattaliyath et al. (4) reported that dHAND functions as a transcriptional activator partially independent on DNA binding. Indeed, interaction between transcription factors may take place at one of their binding sites. For example, among these cardiac-specific transcription factors, GATA4 recruits MEF2C (5), NKX2.5 recruits dHAND (4), and NKX2.5 recruits GATA4 (15). All these recruitments lead to an increase of ANP promoter activity. On the other hand, interaction between transcription factors may take place at both of their binding sites. For example, the synergistic activation of cardiac promoters by NKX2.5-TBX5 interaction has been shown to be based on stable ternary complexes composed of NKX2.5-TBX5 interaction as well as the simultaneous binding of each protein to its binding site (17,18). Accordingly, our results provide a molecular paradigm for elucidating the regulation mechanisms of dHAND and MEF2C on heart development.
During cardiogenesis and cardiac differentiation, many transcription factors play important roles on ANP transcription. For example, dHAND interacts with MEF2C, GATA4, or NKX2.5 (4,11), and GATA4 interacts with MEF2C or NKX2.5 (5,16). All these interactions contribute to enhance the activity of ANP promoter. These transcription factors and their target genes generate a unique network of protein-protein and protein-DNA interactions to form a cardiac-specific enhanceosome, which coordinates expression of target gene. The enhanceosome is made of stable multiprotein complexes that promote the cooperative recruitment of coactivators and RNA polymerase II complex to activate transcription. Assembly of the enhanceosome is regulated by specific extracellular events and intracellular signals (19 -22).
Although we have shown that MEF2C forms a complex with dHAND, it is not clear how complex formation is regulated. A challenge for future investigations is to discover the temporal and spatial regulation of these interactions during heart development. It is reported that p38 MAPK (mitogen-activated protein kinase) activates MEF2 in cardiac muscle (23)(24)(25) and dHAND has been implicated in epithelial 1-induced transcrip-tion (12), so the analysis of interaction between dHAND and MEF2C in response to different stimuli in cardiomyocytes at various development stages will be important to understand the mechanisms of enhanceosome formation on ANP and ␣-MHC transcription regulation.