A ternary complex of transcription factors, Nishéd and NFATc4, and co-activator p300 bound to an intronic sequence, intronic regulatory element, is pivotal for the up-regulation of myosin light chain-2v gene in cardiac hypertrophy.

Transcriptional up-regulation of the myosin light chain-2 (MLC-2v) gene is an established marker for hypertrophic response in cardiomyocytes. Despite the documentation on the role of several cis-elements in the MLC-2v gene and their cognate proteins in transcription, the mechanism that dictates the preferential increase in MLC-2v gene expression during myocardial hypertrophy has not been delineated. Here we describe the properties of a cardiac specific intronic activator element (IRE) that shares sequence homology with the repressor element, the cardiac specific sequence, in the chicken MLC-2v gene. The transcription factor, Nishéd, that recognizes both IRE and the cardiac specific sequence potentiates the transcription of the MLC-2v gene via interaction with another transcription factor, nuclear factor of activated T cells, and the co-activator p300 at the IRE site. Angiotensin II (Ang II), a potent agonist of hypertrophy, causes induction of the MLC-2v gene transcription, which correlates well with the enhanced binding of Nishéd-nuclear factor of the activated T cells-p300 complex to IRE in the gel mobility shift assay. Losartan, an antagonist of Ang II receptor (AT1), abolishes the agonist-dependent stimulation of IRE/protein interaction and the consequent increase in MLC-2v gene transcription. These results together have thus established a transcriptional role of IRE as a direct target sequence of Ang II-mediated signaling that appears to be pivotal in the mechanism underlying the up-regulation of the MLC-2v gene during cardiac hypertrophy.

Transcriptional up-regulation of the myosin light chain-2 (MLC-2v) gene is an established marker for hypertrophic response in cardiomyocytes. Despite the documentation on the role of several cis-elements in the MLC-2v gene and their cognate proteins in transcription, the mechanism that dictates the preferential increase in MLC-2v gene expression during myocardial hypertrophy has not been delineated. Here we describe the properties of a cardiac specific intronic activator element (IRE) that shares sequence homology with the repressor element, the cardiac specific sequence, in the chicken MLC-2v gene. The transcription factor, Nishé d, that recognizes both IRE and the cardiac specific sequence potentiates the transcription of the MLC-2v gene via interaction with another transcription factor, nuclear factor of activated T cells, and the co-activator p300 at the IRE site. Angiotensin II (Ang II), a potent agonist of hypertrophy, causes induction of the MLC-2v gene transcription, which correlates well with the enhanced binding of Nishé d-nuclear factor of the activated T cells-p300 complex to IRE in the gel mobility shift assay. Losartan, an antagonist of Ang II receptor (AT 1 ), abolishes the agonist-dependent stimulation of IRE/protein interaction and the consequent increase in MLC-2v gene transcription. These results together have thus established a transcriptional role of IRE as a direct target sequence of Ang II-mediated signaling that appears to be pivotal in the mechanism underlying the up-regulation of the MLC-2v gene during cardiac hypertrophy.
The ventricular myosin light chain-2 (MLC-2v) 1 gene serves as a paradigm for understanding the complexities of eukaryotic transcriptional regulation in development and disease, as it displays a strict cardiac tissue specificity (1,2) and also serves as a marker for cardiac hypertrophy (3)(4)(5)(6). Cardiac hypertrophy, an adaptive response, is characterized by an increase in cardiac cell size (7,8), sarcomeric reorganization (3)(4)(5)(6)9), reexpression of embryonic genes (10), and induction of other hypertrophy-associated genes (11); among them, MLC-2v gene transcription is preferentially stimulated in response to agonist-induced hypertrophy (3,12). To understand the mechanism(s) that dictates the transcriptional regulation of the MLC-2v gene during development and disease, several regulatory cis-elements and their cognate proteins in both the avian (13)(14)(15)(16)(17) and murine (18) MLC-2v genes have been identified. Previous studies on chicken MLC-2v gene from our laboratory (1,(13)(14)(15)(16)(17) and on murine MLC-2v gene of others (18) have revealed that the basal promoter architecture of the MLC-2v gene is conserved in both genomes and shares the sequence motifs of the HF-1 box containing the CArG, MEF2, and AP-2 elements (18). In the chicken MLC-2v gene, there is an additional upstream sequence, cardiac specific sequence (CSS) (17,19), that represses the chicken MLC-2v gene expression in skeletal muscle cells. Recently, we have identified and partly characterized a transcription factor, Nishéd, as a putative repressor that interacts with CSS (17). In studies using transgenic mice, the HF-1 box in the 250-bp rat promoter region of the rat MLC-2v gene was assigned a role in cardiac tissue specificity (20).
Several inducible regulatory sequence elements have been identified in hypertrophy-responsive genes. For example, the serum-responsive element and SP-1 confer ␣-adrenergic stimulation to the atrial natriuretic factor (ANF) promoter (21,22). The AP-1 and/or cAMP-responsive elements cause induction of ANF expression in pressure overload hypertrophy (22). Similarly, the GATA-binding element and its interaction with GATA-4, a zinc finger protein, mediate the induction of ␤-myosin heavy chain in left ventricular hypertrophy resulting from aortic constriction (23)(24)(25)(26). More recently, the nuclear factor of activated T cells (NFAT) family of transcription factors was reported to induce the expression of target genes associated with hypertrophy via the formation of a protein complex involving GATA-4 (27), MEF2 (28), and AP-1 (29). The HF-1 box implicated in tissue specificity above also confers the ␣-adrenergic inducible activation of the rat MLC-2v gene (3), but the role of the regulatory proteins, HF-1b (30) and EF1 A /YB-1 (31), that interact with the HF-1b/MEF2 site and the CArG box in HF-1, respectively, has not been defined. Thus, the mechanism of transcriptional regulation of MLC-2v gene in hypertrophy response remains elusive.
Here we provide evidence of a new cardiac specific cis-regulatory element, intronic regulatory element (IRE), located in the first intron of the chicken MLC-2v gene. IRE shares a * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 1 The abbreviations used are: MLC-2v, myosin light chain-2 ventricular; AP-1, AP-2, activator protein-1, 2; MEF2, myocyte enhancer factor 2; NFAT, nuclear factor of activated T cells; IRE, intronic regulatory element; CSS, cardiac specific sequence; Ang II, angiotensin II; GMSA, gel mobility shift assay; SHR, spontaneously hypertensive rat; ANOVA, analysis of variance; ORF, open reading frame; PMSF, phenylmethylsulfonyl fluoride; GMSA, gel electrophoretic mobility shift assays; PBS, phosphate-buffered saline; PIPES, 1,4-piperazinediethanesulfonic acid. common core motif 5Ј-GAAG-CTTC-3Ј with the upstream negative CSS element and plays a dual role, i.e. it overrides the repressive effect of CSS via its binding to a common transcription factor, Nishéd, and in hypertrophy, IRE responds to signals triggered by the hypertrophic ligand, angiotensin II (Ang II), to recruit activator proteins that induce MLC-2v gene transcription. Mutating the IRE sequence in the MLC-2v-luciferase reporter totally abolishes the Ang II-induced transcriptional up-regulation in transient transfection assays. In ovo treatment of the chick embryos with Ang II results in a selective increase in IRE-Nishéd complex formation in which Nishéd associates with NFATc4 and the co-activator p300, the major components of the endogenous IRE/protein binding activity in cardiac muscle cells. This ternary complex formation is enhanced in response to Ang II and is therefore likely to play a key role in the transcriptional up-regulation of the MLC-2v gene in hypertrophy.

EXPERIMENTAL PROCEDURES
Primary Cardiac Cell Culture and Treatment-Hearts of 11-day-old White Leghorn chicken embryos (Spafas) were collected and washed in phosphate-buffered saline (PBS), and the cells were dissociated from the ventricular tissues by successive enzymatic digestion using 0.1% pancreatin (Difco) at 37°C. Dissociated cells from the first treatment were discarded, and those from the subsequent three treatments were combined, washed, and resuspended in growth medium (M199 with 5% fetal bovine serum, 100 units/ml penicillin/streptomycin, and 2 nmol/ liter L-glutamine). The cells were differentially plated for 1 h in cell culture to remove contaminating non-myocytes and then plated at a density of 1 ϫ 10 6 cells/well of 6-well cell culture dishes or 3 ϫ 10 6 cells in 100-mm tissue culture plates and incubated overnight in M199 medium supplemented with 5% fetal bovine serum, penicillin/streptomycin (100 units/ml), and L-glutamine (2 nmol/liter). The following day the cells were washed in PBS and cultured in serum-free M199 medium, containing penicillin/streptomycin, and L-glutamine as above and 5 g/ml insulin/selenium/transferrin (Invitrogen). Primary cardiac cells were treated with Ang II (Sigma) at 10 Ϫ7 M, which was added when needed, and cells were incubated for 2 h. Losartan (10 Ϫ7 M) (kindly provided by DuPont) was added 1.5 h prior to Ang II.
Construction of Reporter Genes-Construction of pLC106CAT and pLC2.4CAT was described earlier (32). pMLC2.1Luc is a 2.1-kb SmaI/ StuI fragment of the MLC-2v gene cloned into the promoter-less vector, pGL2Basic (Promega, Madison, WI). A partial HindIII digest of the MLC-2v gene yields a 1.4-kb fragment that was cloned into pGL2-Basic to yield the pMLC106Luc reporter gene construct. The IRE sequence in pMLC2.1Luc was mutated (GAAGCTTC3 GtAcCccg) by using the Gene Editor site-directed mutagenesis kit (Promega, Madison, WI).
Transient Transfections and Luciferase Assay-Primary chicken cardiomyocytes were transiently transfected 24 h after plating, using Fu-GENE 6 reagent (Roche Applied Science) in 6-well plates with 1 g of reporter gene construct and 10 ng of thymidine kinase promoter-driven Renilla luciferase-thymidine kinase vector, which was used to normalize the transfection efficiency. The promoter-less reporter vector, pGL2-Basic, was used as control. Luciferase assays were performed with the dual luciferase assay kit (Promega, Madison, WI).
RNA Extraction, Northern Hybridization-Total RNA was extracted from cardiomyocytes using Trizol reagent (Invitrogen). For Northern hybridization, 5-10 g of RNA was size-fractionated on a 1.4% formaldehyde-agarose gel, transferred to a nylon membrane, and hybridized with either radiolabeled chicken MLC-2v or rat ␣-skeletal actin cDNA probes as described previously (1,14).
Screening of cDNA Expression Library-Chicken skeletal muscle cDNA expression library cloned in the ZAP expression vector (Stratagene) was screened with IRE sequence as probe, and positive clones were isolated and characterized as described previously (17).
Hybrid Selection Assay and cDNA Synthesis-Poly(A) ϩ RNA was isolated from skeletal muscle using the Fast Track kit (Invitrogen). Ten g of purified Nishéd cDNA insert was denatured, immobilized on a nylon membrane, and incubated with 10 g of poly(A) ϩ RNA in hybridization buffer (42% formamide, 0.01 M PIPES, pH 6.4, 0.4 M NaCl) overnight at 42°C. The filter was washed repeatedly with 1 ml of wash buffer (0.2ϫ SSC, 0.2% SDS) at 55°C followed by three rinses in 2 mM EDTA. The selected RNA was eluted after boiling for 60 s in 300 l of 1 mM EDTA and precipitated (33). The RNA thus obtained was subjected to cDNA synthesis using the SUPERSCRIPT choice system for cDNA synthesis (Invitrogen), cloned into the plasmid BlueScript SK, and subsequently sequenced.
Partial Purification of Nishéd Protein-Nishéd cDNA insert was obtained from the phagemid Bluescript by digesting with BamHI and XhoI and cloned into the bacterial expression vector pET29b (Novagen). BL(21)DE3 bacterial cells were transformed with this construct, and cells were induced to make the recombinant protein according to the manufacturer's instructions (Novagen).
Ang II Treatment of Chicken Embryos-Chicken eggs were incubated for 7 days at 37°C in a humidified incubator. Ang II at a final concentration of 10 Ϫ7 M in PBS was injected daily for 12 days into the air sac of the embryos.
Histological Analysis-The hearts were harvested from the embryos, rinsed in PBS, blotted dry, and weighed. Hearts were fixed overnight at 4°C in 4% formaldehyde buffered with phosphate-buffered saline and embedded in Tissue-Tek O.C.T Compound (Sakura Finetek) embedding medium for histological sectioning. Longitudinal 10-m cryostat sections were cut and stained with hematoxylin and eosin.
Western Blot Analysis-Proteins were fractionated by SDS-PAGE (15-18%) and transferred onto nitrocellulose membrane in electroblotting buffer (20 mM Tris, 150 mM glycine, 20% (v/v) methanol) for 1 h at 4°C. Polyclonal rabbit antibodies to glutathione S-transferase-fused Nishéd protein were generated by Pocono Rabbit Farm & Laboratory Inc. The blots were blocked in Tris-buffered saline, 0.1% Tween 20 (TBST) with 3% bovine serum albumin and 5% milk powder for 1 h at room temperature. Glutathione S-transferase-depleted Nishéd antibodies were used as primary antibody at 1:4000 dilution in TBST overnight at 4°C. Secondary detection was done using horseradish peroxidecoupled anti-rabbit antibody in the ECL enhanced chemiluminescence assay as per the supplier's recommendations (Amersham Biosciences).
Statistical Analysis-Values are expressed as mean Ϯ S.E. of n experiments. Values are given as percent activity relative to the activity of the wild type MLC-2v promoter, pMLC2.1Luc. Differences between experimental groups were evaluated for statistical significance using one-way ANOVA tests, and Tamhane's tests were performed post-hoc to test for significant differences. Significance was determined at the level of p values Ͻ0.05.

RESULTS
The Role of Intron 1 in MLC-2v Gene Transcription-It was reported previously (19,34) that the MLC-2v promoter/reporter constructs, containing the 5Ј-flanking and intragenic region inclusive of the first intron, produce maximal transcription activity upon transient transfection in cardiac cells in culture compared with constructs without the intragenic sequence, suggesting that the first intronic sequence contains an activator element. This was consistent with the presence of a cardiac specific DNase I-hypersensitivity region located in the first intron of the gene (35). In order to identify the precise intronic sequence with a possible role in MLC-2v gene transcription, 3Ј deletion of the MLC-2v-reporter constructs were generated (Fig. 1A). Plasmid pLC2.4CAT (2400 bp) contains 1307 bp of upstream promoter and a 1093-bp intragenic sequence containing exon 1, the first intron and part of exon 2. p⌬7LCCAT (1575 bp) has the identical upstream promoter sequence as in pLC2.4CAT and 268 bp of intragenic region. Plasmid pLC106CAT (1461 bp) is identical to pLC2.4CAT but contains 154 bp of intragenic sequence. When examined for the promoter activity by transient transfection in chicken primary cardiac cells in culture, pLC2.4CAT and p⌬7LCCAT show almost identical promoter activity reaching the level of pSV2CAT, whereas pLC106CAT was relatively inactive (22%) (Fig. 1B), suggesting that the region between ϩ154 to ϩ268 bp contains an activator element. An examination of the sequence in this region revealed the presence of a palindromic sequence, 5Ј-GAAGCTTC-3Ј at ϩ151 to ϩ158 bp within the first intron that shares the GAAG or CTTC sequence in the core motif of the previously characterized negative element (CSS) upstream in the MLC-2v promoter (17) (Fig. 1C). To assess the role of this putative regulatory sequence, hereafter referred to as IRE, a substitution mutation was introduced (GAAGCTTC3 GTAC-CCCG) in the core IRE sequence, and the resulting plasmid, pMutIRELuc, was examined for the promoter activity in primary cardiac cells in culture (Fig. 2). The introduction of mutation in the IRE palindrome caused a significant loss (80%) compared with the parent plasmid pMLC2.1Luc. pMLC106Luc, where only one-half of the IRE palindrome was deleted, also showed a similar low level activity, suggesting that the wild type palindromic sequence within IRE is essential for the maximal activation of the MLC-2v gene. We speculated that because the activator IRE shares the palindromic motif with the negative element CSS, it is likely that IRE and CSS binding trans-acting factors play a cooperative function in the overall regulation of MLC-2v gene transcription.
Gel Mobility Shift Analysis-We performed gel electrophoretic mobility shift assays (GMSA) with nuclear extracts from the chick embryonic heart and skeletal muscle and radiolabeled IRE synthetic oligonucleotides (ϩ149 to ϩ175 bp) as probe. Nuclear extracts from both tissues produced several DNA-protein complexes that were more abundant in skeletal muscle extracts (Fig. 3). The IREBP1 complex is distinctly visible in GMSA with skeletal muscle nuclear extracts and is much reduced in cardiac extracts, possibly because of differences in stability of proteins that form the IREBP1 complex in skeletal and cardiac tissues. Competition with 15-and 30-fold molar excess of unlabeled IRE oligonucleotide (Fig. 3, I) reduced all complex formation (Fig. 3). CSS (Fig. 3, C) oligonucleotide competed away the IREBP2 complex but not others indicating that IREBP2 recognizes the common core motif (Fig.  3). Thus, it appears that the two functionally disparate elements, CSS and IRE, share common DNA-binding proteins.
Isolation of Recombinant Nishéd-Because it appeared that IRE and CSS binding complexes have a common protein(s), an expression screening of an adult chicken muscle cDNA library was done using both IRE and CSS oligonucleotides separately as probes, as reported previously (17). The cDNA clone, Nishéd, thus isolated contains an insert of 1167 bp with an ORF of 132 amino acids but lacks the 5Ј end of the ORF. Therefore, we performed hybrid selection to obtain the Nishéd selected mRNA, which was used as template for synthesizing the fulllength cDNA. The cDNA was cloned into pBlueScript and sequenced, which confirmed that it contains the portion of the 5Ј-untranslated region, the conserved Kozak sequence, and a region encoding the first five amino acids that were missing in Nishéd. The full-length Nishéd cDNA has an ORF of 137 amino acids that encodes a protein of ϳ15 kDa (Fig. 4).
To evaluate the DNA binding potential of the recombinant Nishéd protein, its cDNA was introduced into the bacterial expression vector, pET29b, and the partially purified protein was used in GMSA with end-labeled IRE oligonucleotide. A single Nishéd-IRE protein complex was formed that was competed out by IRE and as well as CSS, but not by the B and S elements containing the heterologous sequence of the MLC-2v promoter (16,36) (Fig. 5A). To identify the IRE-protein complex(es) that contains Nishéd, cardiac nuclear extracts were preincubated with anti-Nishéd antibody (see "Experimental Procedures") and examined by GMSA. Preincubation with Nishéd antibody disrupted the IREBP2 complex (Fig. 5B)  a role in induction of MLC-2v expression in response to cardiac hypertrophic signals. We initially examined the upregulation of the MLC-2v gene in Ang II-stimulated cardiomyocytes. When primary cardiomyocytes in culture isolated from 11-day chick embryonic heart were treated with Ang II (10 Ϫ7 M), MLC-2v mRNA increased significantly (ϳ2-fold) compared with untreated cells (Fig. 6A). We then tested the possibility that IRE-binding protein(s) also respond positively to Ang II administration. Cardiac cells isolated as above were stimulated in culture with different concentrations (1, 10, and 100 nM) of Ang II, and the nuclear extracts were examined by GMSA. A short IRE DNA (ϩ148 to ϩ167 bp) that contains the core motif (see "Experimental Procedures") and yields only the two IREBP1 and -2 complexes was used as probe. One of these two complexes (IREBP2) contains Nishéd (see Fig. 5B). A concentration-dependent increase in two IRE-binding proteins (IREBP1 and -2) was observed (Fig.  6B). Competition with unlabeled IRE effectively abolished the binding in all Ang II-stimulated extracts.
The positive role of IRE in MLC-2v transcription was further demonstrated by the transcriptional activity of pMLC2.1Luc and pMutIRELuc in transient transfection assays using primary cardiac cells treated with Ang II (10 Ϫ7 M). There was a 1.8-fold increase in transcription of pMLC2.1Luc due to Ang II treatment (Fig. 7). Because the actions of Ang II in the cardiovascular system are almost exclusively mediated by the AT 1 receptor (37), we examined whether the promoter activity of pMLC2.1Luc is susceptible to inhibition of AT 1 signaling by losartan, a documented antagonist of AT 1 receptor (38). Cardiac cells treated with Ang II in the presence of losartan exhibited total inhibition of the Ang II-stimulated pMLC2.1Luc promoter activity. The plasmid, pMutIRELuc, with mutated IRE did not respond to Ang II or losartan treatment. Thus, the promoter containing the known functional upstream elements with the exception of native IRE sequence rendered the MLC-2v gene nonresponsive to Ang II stimulation, suggesting that it may be the sole target sequence for Ang II-induced signaling in the up-regulation of MLC-2v in hypertrophy. This also supports the possibility that Ang II/AT 1 -mediated signaling is directed to the transcriptional apparatus of the MLC-2v gene and involves IRE and probably its cognate protein Nishéd as targets.
Role of IRE in Ang II-induced Hypertrophy Response of MLC-2v in Ovo-Ang II (100 nM) or the carrier PBS was injected daily in ovo in chick embryos starting on day 7 as described under "Experimental Procedures" (39). On day 19, Ang II-treated embryos displayed a 34% increase in heart weight/body weight ratio as compared with the PBS-treated embryos (n ϭ 6, in triplicates) (Fig. 8A). Hearts of Ang IItreated embryos were distinctly larger in size and exhibited changes, such as thickening of the inter-ventricular septum and narrowing of the left ventricular cavity (Fig. 8B). Northern hybridization of total RNA isolated from PBS and Ang IItreated hearts with radiolabeled probe for either chicken MLC-2v or rat ␣-skeletal actin, documented hypertrophy markers, showed increased levels of transcripts in Ang II-treated hearts (Fig. 8C). These findings confirmed that Ang II is a potent agonist of cardiac hypertrophy in chicken heart. To examine if IRE-binding proteins are activated in vivo, cardiac nuclear extracts from embryos treated with Ang II were examined by GMSA with the IRE oligonucleotide probe used above (see Fig. 3). The IREBP2 complex was selectively increased in nuclear extracts from hypertrophied hearts (Fig. 9). When the extracts were preincubated with anti-Nishéd antibody, a pronounced loss of the IREBP2 complex was exhibited, as expected.
Nishéd Interacts with the Co-activator p300 -It is known that p300 is a ubiquitous co-activator that interacts with a diverse family of transcription factors (40). To test the possible involvement of p300 in the IRE-protein complex(es) formation, cardiac muscle nuclear extracts were incubated with anti-p300 antibody. Fig. 10A shows that p300 antibody disrupted the IRE/Nishéd interaction (IREBP2) in heart extract suggesting the possible physical interaction between Nishéd and p300. The involvement of p300 in the IRE functional complex was not observed in skeletal muscle extracts. To examine whether Nishéd solicits the recruitment of p300 in the Ang II-induced hypertrophy response, we examined the physical interaction of FIG. 6. Angiotensin II treatment stimulates MLC-2v mRNA level. A, total RNAs from cardiac cells in culture isolated from 11-dayold chicken embryos were treated with Ang II (100 nM), or PBS was isolated and size-fractionated on a 1.2% formaldehyde-agarose gel. MLC-2v mRNA levels were examined by Northern blotting with MLC-2v cDNA probe. 18 S RNA loading was used as loading control. B, IRE-binding proteins from Ang II-treated cardiomyocytes in culture. Primary cardiac cells from 11-day-old chicken embryos were cultured and treated with 1, 10, and 100 nM Ang II, as described above. Gel mobility shift analysis was performed with nuclear extracts of Ang II-treated and -untreated cardiac cells and 32 P-labeled DNA probe containing the ϩ148 to ϩ167-bp region of the IRE oligonucleotide of MLC-2v gene. FIG. 7. Stimulation of MLC-2v gene transcription by Ang II is mediated by IRE. pMLC2.1Luc and the mutant IRE-containing plasmid pMutIRELuc were transfected into primary cardiac cells isolated from 11-day-old chicken embryos. The transfected cells were maintained in defined medium for 48 h and stimulated with 100 nM angiotensin II for 2 h. Cells were pretreated with 100 nM losartan for 1.5 h prior to Ang II stimulation when indicated. The luciferase reporter activity was assayed as above. Renilla luciferase expression was used for normalization. Asterisks indicate significant differences (one-way ANOVA, unequal variances assumed, with Tamhane's test, p Ͻ 0.05 versus pMLC2.1Luc. Mean Ϯ S.E., n ϭ 3 or greater performed in triplicate).
Nishéd with p300 in a co-immunoprecipitation assay where the anti-p300 antibody was used for immunoprecipitation followed by Western blotting with anti-Nishéd antibody. The data in Fig. 10B show that the complex immunoprecipitated by anti-p300 antibody contains Nishéd and that it is increased in extracts from the Ang II-treated embryos. Immunoprecipitation with anti-Nishéd antibody showed no increase in total Nishéd protein. By contrast, the IRE-bound Nishéd was increased in Ang II-treated extracts (see Fig. 9).
NFATc4 Participates in Hypertrophy-induced IRE-Nishéd Complex Formation-Recently, NFAT activity has been implicated in cardiac hypertrophy (41,42). Transgenic mice expressing NFATc4 in a cardiac selective manner developed hypertrophy readily (43). The possibility of NFATc4 participation in the IRE/Nishéd interaction is strengthened by the existence of the NFAT DNA binding consensus sequence (5Ј-WGGAA-3Ј) that overlaps the 5Ј region of the IRE palindrome (Fig. 11A). Therefore, we examined the possibility that NFATc4 is employed in the Ang II-induced up-regulation of the MLC-2v gene. When GMSA was done using Ang II-treated cardiac extracts, as above, with antibody against NFATc4, there was a distinct reduction of IREBP2 complex formation (Fig 11B). The Ang II-mediated enhancement of NFATc4 participation in the ternary complex formation is also shown in Fig. 11C. As in Fig.  10B, Ang II-induced extracts were immunoprecipitated with anti-NFATc4 antibody followed by Western blotting with anti-Nishéd antibody. Clearly, there was an increase in NFATc4/ Nishéd interaction in Ang II-treated cardiac extracts (Fig.  11C). These results were also corroborated with extracts isolated from the hypertrophic hearts of spontaneously hypertensive rat (SHR), which showed a prominent increase in NFATc4 interaction with Nishéd compared with extracts from the hearts of the age-matched normotensive rat strain WKY (Fig. 11D).
Taken together, these data demonstrate that IRE, a potent activator in MLC-2v gene, plays, along with Nishéd, NFATc4 and p300, an important role in the regulation of MLC-2v expression in myocardial hypertrophy.

DISCUSSION
Transcription of genes is known to be regulated by distal sequence elements, illustrated by the fact that the proximal promoter alone rarely supports correct cell type-specific expression of linked reporter genes in transgenic animals. Distal regulatory regions spanning large segments of DNA are often needed (44,45). In transgenic mice studies (46,47) the presence of rat MLC-2v 5Ј-flanking sequence alone does not display an expression profile similar to that of the endogenous gene. To date, only three cis-elements located at the proximal promoter FIG. 8. Cardiac hypertrophy induction in chicken embryos treated in ovo with Ang II. A, PBS or Ang II (1-100 nM) was injected daily into 7-day-old chick embryos. At day 19, hearts were isolated and weighed. The ratio of heart weight to body weight is shown as mean Ϯ S.E. (n ϭ 3, six embryos per treatment). B, histological analysis of Ang II-treated embryonic hearts. Hearts of Ang II-treated and control embryos were fixed in 4% paraformaldehyde, 2 mM EGTA for overnight at 4°C. Cryostat sections (10 m) were stained with hematoxylin and eosin. C, total RNA was isolated from hearts of embryos treated with either PBS or Ang II as described above. Total RNA (10 g) was electrophoresed on a 1.4% formaldehyde-agarose gel, and Northern blot was performed with radiolabeled chicken MLC-2v or rat ␣-skeletal actin cDNA probes.18 S RNA served as loading control.
FIG. 9. IRE-binding proteins are induced by Ang II treatment in ovo. Chicken eggs were treated with either PBS or Ang II, as described in the legend to Fig. 8. Nuclear extracts were prepared from the isolated hearts, and GMSA was performed with radiolabeled IRE oligonucleotide. Arrow indicates the DNA-protein complex. Preincubation of cardiac nuclear extracts with anti-Nishéd antibody shows the complex that contains Nishéd is induced by Ang II treatment in ovo.
FIG. 10. Co-activator p300 participates in IRE-protein complex. A, GMSA was done with 32 P-labeled IRE oligonucleotide and nuclear extracts from cardiac or skeletal muscle tissue. Incubation with anti-p300 antibody effectively disrupted the IREBP2 complex in cardiac nuclear extracts (see arrow). B, Ang II treatment induces Nishéd interaction with p300. 500 g of cardiac cell lysates from either PBS-or Ang II-treated embryos from two separate experiments were co-immunoprecipitated with anti-Nishéd or anti-p300 antibodies. Top panel shows extracts immunoprecipitated (IP) with p300 antibody and immunoblotted (WB) with Nishéd antibody, and bottom panel shows extracts immunoprecipitated and blotted with Nishéd antibody.
have been characterized in the rat MLC-2v gene. They alone are insufficient to support the regulation of its expression in development and hypertrophy. Our present work utilizes the chicken MLC-2v gene promoter, extensively characterized in our laboratory (13)(14)(15)(16)(17), and documents the role of IRE in the cardiac hypertrophic response resulting in its preferential upregulation. IRE or IRE-like sequences are also present in other mammalian genes. The murine small heat shock protein/␣Bcrystallin gene that is expressed, in addition to the lens, in cardiac myocytes at high levels (48,49) contains a sequence similar to IRE, the ␣BE-4 cis element (5Ј-GAAGATTC-3Ј), in its enhancer. Like IRE, ␣BE-4 is a heart-specific regulatory element required for the maximal expression of this gene in myocardial cells (50) and has been implicated in the induction of the ␣B-crystallin gene in response to mitogen-activated protein kinase-induced stress in cardiac myocytes (51). However, there is no evidence on the identity of the ␣BE-4-binding protein raising the possibility that Nishéd or an analogous factor is functional at this site. IRE-like sequence (5Ј-GAAGCTCC-3Ј), identified by the web-based program VISTA, is also present at an analogous position (ϩ155 bp) in the rat MLC-2v gene, although its potential in the regulation of transcription has not been investigated.
IRE shares the core sequence motif within the upstream CSS element identified previously as a repressor in the MLC-2v promoter (17). Although Nishéd binds both CSS and IRE, these two elements perform distinct functions; CSS is a repressor and IRE an activator. The functional activity of these elements may be dictated by the ability of Nishéd to recruit specific co-activators or co-repressors at IRE and CSS sites, respectively. In a parallel study, 2 we have shown that Nishéd recruits a corepressor mSin3A at the CSS site to afford repression of the cardiac MLC-2v gene in skeletal muscle cells. We demonstrate here that the recruitment of the common transcription factor, Nishéd, and its association with NFAT and the co-activator p300 at IRE is likely to play a critical role in the regulation of MLC-2v gene transcription in cardiac cells.
We also show here that the chicken embryonic heart is susceptible to the hypertrophic agonist Ang II. The Ang II receptor, AT 1 , is known to express in developing chick heart as early as day 4 corresponding to Hamburger Hamilton stage 12 (52), and Ang II is a documented stimulus of hypertrophic effects in cardiomyocytes in cell culture resulting in increased protein synthesis and cell growth (38,53). In this context, the chicken embryo offers a convenient experimental model for studies on hypertrophy-induced gene regulation.
The regulated association of hypertrophic agonists with transcriptional activation of genes, such as MLC-2v, in the hypertrophic response has been well documented (53,54), but the underlying mechanism remains unsolved. Our data shows that the Nishéd-bound IRE complex serves as the target for Ang II-induced signals generated in both cultured cardiac myocytes and in the intact heart tissue. IREBP1 complex is distinct in nuclear extracts from cardiac cells in culture than from whole tissue, underscoring the differences in cell culture and in vivo systems. The observation that the AT 1 receptor antagonist losartan abolishes the induction of transcription and that the mutated IRE was insensitive to Ang II establishes the role of IRE as the sole sensor of AT 1 signaling. We propose that a critical role of Nishéd is to mediate its association with other hypertrophy associated factors NFATc4 and p300 in response to AT 1 -induced signaling.
The increased interaction of Nishéd with p300 in cardiac cells upon Ang II stimulation is relevant in the light of the known role of p300 in cardiac hypertrophy. p300 has been shown to serve as an adaptor for hypertrophy-responsive factors, such as GATA-4 (55) and MEF2 (56,57). The expression of p300 and its histone acetyltransferase activity is increased during hypertrophy (58,59). Phosphorylation of p300 influences the histone acetyltransferase activity of p300 (60). It has been reported that p44 MAP kinase (61) and calcium/calmodulindependent kinase (62) mediate the phosphorylation of p300, thereby causing an induction of p300-mediated transcription. Ang II activates, either directly or indirectly, both these signaling cascades (63,64) and thus is likely to be involved in the post-translational modification of p300 resulting in the increased interaction between Nishéd and p300 in cardiomyocytes. We also observed that p300 does not participate in the DNA/protein assembly at IRE in skeletal muscle extracts. One possible explanation is that in skeletal muscle the post-translational modification status of p300 or the absence of the interacting protein(s) may preclude its participation in the IREBP2 DNA-protein complex formation. It is also possible that a co-2 S. Mathew, E. Mascareno, and M. A. Q. Siddiqui, unpublished data. FIG. 11. NFATc4 is a component of IREBP2 complex. A, DNA sequence of IRE and the NFATc consensus sequence showing the overlap of the palindrome at the 5Ј end. B, GMSA was done with 32 P-labeled IRE oligonucleotide and cardiac nuclear extracts isolated from PBS-or Ang II-treated embryos incubated with anti-NFATc4 antibody. The anti-NFATc4 antibody significantly reduced the intensity of the IREBP2 complex (see arrow). C, increased interaction of Nishéd with NFATc4 in Ang II-treated nuclear extracts. Cardiac cell lysates from either PBS-or Ang II-treated embryos were co-immunoprecipitated with anti-Nishéd or anti-NFATc4 antibody. Top panel shows extracts immunoprecipitated (IP) with NFATc4 antibody and immunoblotted (WB) with Nishéd antibody, and bottom panel indicates extracts immunoprecipitated and blotted with Nishéd antibody. D, increased interaction of Nishéd with NFATc4 in cardiac extracts from spontaneously hypertensive rat (SHR). Cardiac cell lysates from SHR and agematched normotensive WKY rats were co-immunoprecipitated with anti-Nishéd or anti-NFATc4 antibodies. Top panel shows extracts immunoprecipitated with NFATc4 antibody and immunoblotted with Nishéd antibody, and bottom panel indicates extracts immunoprecipitated and blotted with Nishéd antibody. repressor in the skeletal muscle extracts interacts with IRE at a higher affinity than p300. Indeed, we have observed that Nishéd interacts with N-CoR, a co-repressor protein of similar molecular weight as p300 in skeletal muscle (data not shown).
The participation of NFAT in the development of cardiac hypertrophy has also been reported (43,(65)(66)(67). NFAT isoforms are present in ventricular myocytes (41) and serve as mediators in the recruitment of p300 (68) to promote a cooperative transcriptional activation of several target genes (41,69). Ang II stimulates the DNA binding activity of NFAT proteins and the NFAT/calcineurin-dependent transcriptional activation of target genes (70,71). We observed the NFAT-binding site overlaps the IRE sequence, and this prompted us to examine its presence in the IRE-Nishéd complex. Our data suggest that NFATc4 contributes to hypertrophy response in conjunction with the Ang II-induced Nishéd/IRE interaction. Most surprisingly, however, heart-targeted disruption of NFATc4 gene in mice did not show a defect in the ability to mount a hypertrophy response (42). The functional hierarchy of different NFAT isoforms has not been clearly established. For example, despite the lack of response in NFATc4 null animals for hypertrophy, forced expression of the same isoform in cardiomyocytes rendered the cells less susceptible to stress-induced apoptosis (66,72). Also, NFATc4 was shown to be responsible for inducing pro-survival signaling in cardiac cells (73,74). Therefore, we believe that, consistent with its abundance in the heart (41), NFATc4 is functionally involved in the Ang IIinduced ternary complex formation in MLC-2v up-regulation in cardiac hypertrophy.
In conclusion, we postulate that the induction of MLC-2v gene transcription during Ang II-mediated cardiac hypertrophy is via the induction of IRE-binding proteins like Nishéd and its association with p300 and NFATc4, both known to participate in hypertrophy response. One can envision that the co-activator, p300, facilitates the anchorage of the Nishéd-p300-NFATc4 complex, resident in the first intronic region, to the basal transcriptional complex in modulation of the regulated MLC-2v gene in hypertrophy.