Yin Yang 1 is increased in human heart failure and represses the activity of the human alpha-myosin heavy chain promoter.

Yin Yang 1 (YY1) is a transcription factor that can repress or activate transcription of the genes with which it interacts. In this report we show that YY1 is a negative regulator of the alpha-myosin heavy chain (alphaMyHC) gene, which, with betaMyHC are the molecular motors of the heart. AlphaMyHC mRNA and protein levels are down-regulated in hypertrophy and heart failure, and this is thought to be detrimental for cardiac contractility. We show that YY1 specifically interacts with the alphaMyHC promoter and that overexpression of YY1 in cardiac cells represses the activity of the alphaMyHC promoter. We also show that the 170-200-amino acid region of YY1, important for its interaction with histone acetyl transferases and histone deacetylases, is important for its repressive activity and that YY1 deleted in this region is an activator of the alphaMyHC promoter. Moreover, we show that YY1 levels and DNA binding activity are increased in failing human left ventricles and in a mouse model of hypertrophic cardiomyopathy, where alphaMyHC levels are decreased. These results suggest that YY1 is a negative regulator of alphaMyHC gene expression.

The cardiac myosin heavy chains (MyHCs), 1 ␣ and ␤, differ in the rate of ATP hydrolysis, with ␣MyHC having an ϳ2 fold greater rate than ␤MyHC (1). The contractile velocity of cardiac muscle correlates with MyHC isoform composition, suggesting a causal relationship between MyHC isoform composition and cardiac contractility (1). Cardiac myosin composition varies among species, with small rodents expressing predominantly ␣MyHC and large mammals expressing predominantly ␤MyHC. Independent of species, antithetical transitions in cardiac MyHC isoform expression have been observed in response to various pathological and physiologic stimuli. Many of these stimuli result in cardiac hypertrophy accompanied by changes in gene expression and increases in cardiac myocyte size. In rodent models, pathological stimuli result in down-regulation of ␣MyHC and up-regulation of ␤MyHC (2). In the normal human heart, ϳ20 -30% of total MyHC mRNA consists of ␣MyHC mRNA, whereas in the failing heart, ␣MyHC expression represents less than 2% of total MyHC (3)(4)(5). At the protein level, ␣MyHC in normal hearts constitutes ϳ11% of total MyHCs, but it is undetectable in the failing heart (6). Moreover, in humans, increases in ␣MyHC isoform expression are closely associated with improvements in left ventricular function and chamber remodeling, in contrast to the expression of several of the other genes known to regulate contractile function or pathological hypertrophy (7). Therefore, rodent and human models of hypertrophy and failure have directionally similar MyHC isoform shifts.
In contrast to pathological stimuli, the hypertrophy induced by exercise (physiological hypertrophy) is accompanied by an induction of ␣MyHC and repression of ␤MyHC in rats (8) and in C57/B16 mice (9). Furthermore, this hypertrophy is beneficial and is associated with cardiovascular conditioning (9,10). Recent evidence suggests that even a small difference in MyHC isoform composition in the heart can have important consequences for contractility. Transgenic mice expressing 12% of their cardiac myosin as ␤MyHC have decreased contractility and myofibrillar ATPase (11). In contrast, isolated cardiac myocytes that express only 12% of their MyHCs as ␣ have 52% greater power output than those expressing only ␤ (12). Thus, the decrease in ␣MyHC in human heart failure may play an important role in the reduction of contractility observed in the failing heart and is likely to be attributable, at least in part, to transcriptional regulation.
Through transfection, gene injection, and transgenic experiments, it has been possible to identify a number of regulatory sites in the rodent ␣MyHC promoter region (13,14,reviewed in Ref. 15). Although most of these elements act positively, there is one report characterizing an element that acts as a repressor of the rat ␣MyHC promoter. This element, located in the first intron of the gene, is an ets binding site (16) but is not present in either the proximal or distal promoter elements of the human ␣MyHC promoter. 2 Relatively little is known about the regulatory elements of the human ␣MyHC gene. However, the proximal promoter regions (Ϫ340/ϩ20 bp) of the human and rat ␣MyHC genes show 80% sequence identity, including most of the potential DNA-binding sites (Fig. 1). Previous studies have shown that the Ϫ340/ϩ20 bp promoter region of the rat ␣MyHC gene is sufficient to direct cardiac-specific expression in cells (17). This fragment contains binding sites for GATA4, MEF2C, SRF, TEF, and thyroid hormone (15).
Yin Yang 1 (YY1) is a transcription factor that has been shown to regulate a variety of promoters (reviewed in Ref. 18). In cardiac myocytes, it has been shown to act largely as a repressor of transcription, with one report showing that it can act as an activator (19 -21). YY1 has been shown to interact with histone deacetylases (HDACs) 1, 2, and 3 in HeLa cells, consistent with its role as a negative regulator of transcription (22,23). Other studies have shown that YY1 interacts with the histone acetyltransferase (HATs) pCAF and p300, possibly explaining its mechanism in transcription activation (24,25). Recently, YY1 has been shown to have a polycomb-like function in Drosophila, acting as a repressor of developmentally regulated genes (26). The activity of YY1 itself is regulated through acetylation by p300 and PCAF and deacetylation by HDACs (27). Acetylation of YY1 augments its repressor activity, possibly by facilitating a tighter interaction with HDACs (27).
To investigate the molecular mechanisms whereby ␣MyHC is decreased in heart failure, we have analyzed the activity of the human ␣MyHC promoter and proximal regulatory region with respect to the effect of YY1. We show that YY1 binds to the ␣MyHC promoter region and is a negative regulator of transcription in neonatal cardiac myocytes (neonatal rat ventricular myocytes, NRVM) but is a positive regulator of the same promoter in non-cardiac cells. Additionally, we show that YY1 protein abundance is significantly increased in pathological models where ␣MyHC is decreased, failing human hearts and in a transgenic mouse model of hypertrophic cardiomyopathy. Conversely, YY1 levels are decreased in exercise-induced hypertrophy in which expression of ␣MyHC is induced. Our data suggest that YY1 is an important contributor to the repression of ␣MyHC expression seen in cardiac hypertrophy and failure.

MATERIALS AND METHODS
Antibodies-YY1 antibody (SC-7341X) was purchased from Santa Cruz Biotechnology. The alkaline phosphatase anti-mouse (115-053-146) was purchased from Jackson Laboratories. The ␣MyHC antibody was a gift from Dr. Kathy Schreiber, Myogen.
Plasmid Construct-The Ϫ454/ϩ32 bp fragment of the ␣MyHC promoter was cloned into the pGL3 basic vector (Promega). Mutations were created in the YY1 binding sites by generating oligonucleotides (see "EMSA") containing the mutation of interest and amplifying the fragments containing the mutation by PCR. The YY1 expression construct was a gift from Dr. Michael Atchison (University of Pennsylvania). The YY1 170-200del construct was a gift of Dr. Edward Seto (University of South Florida). The DNA constructs were purified using the Qiagen method.
Cell Culture and Transfection-Neonatal rat ventricular myocytes (NRVMs) were prepared according to the method described in Waspe et al. (28). Briefly, 150,000 cells/well were plated in 12-well tissue culture plates coated with gelatin. Eighteen h later, the medium was changed to minimum Eagle's medium supplemented with Hank's salt and Lglutamine. 20 mM Hepes pH 7.5, penicillin, vitamin B 12 , bovine serum albumin, insulin, and transferrin were added to the medium. Transfections were carried out by the FuGENE 6 (Roche Applied Science) method according to the manufacturer's recommendations; 0.75 l of FuGENE/0.25 g of plasmid DNA was transfected in each well. In the cotransfection experiments, the total amount of DNA was kept constant by the addition of a plasmid containing the cytomegalovirus (CMV) promoter not driving the expression of any gene. H9C2 cells were maintained according to American Type Culture Collection recommendation. Transfection in H9C2 cells was done by the FuGENE 6 method; 1.8 l of FuGENE/0.6 g of DNA was transfected in each well on a 24-well plate. Retinoic acid (RA) differentiation of H9C2 cells was done by treating the cells with 10 mM RA every day for 5-7 days in medium with 1% fetal bovine serum.
Preparation of Protein Extracts-Protein extracts from human normal and failing (idiopathic cardiomyopathy) left ventricles and transgenic hypertrophic cardiomyopathy (HCM) mice and littermate controls, 4 months of age, were prepared according to Molkentin and Marklan (29) with minor modifications. 0.5 g of tissue was homogenized with a Teflon homogenizer attached to a drill (SKIL-PA6-GF30) at 50% of maximum power. The resulting preparation was sonicated in a cell disruptor (Ultrasonics, W185F) at 50% of maximum power for 15 s. After cell lysis, the proteins were precipitated by the slow addition of the same volume of 4 M NH 4 SO 4 . These results were placed in a 50% final concentration of NH 4 SO 4 , allowing a greater protein recovery when compared with the 30% final concentration of NH 4 SO 4 as described in the original method. Nuclear extracts from NRVMs were prepared according to Gupta et al. (30). HeLa cell nuclear extracts were prepared by the method of Dignam et al. (31).
The resulting complex was resolved in a non-denaturing 4% acrylamide gel in 0.5ϫ Tris-borate-EDTA.
Western Blots-Forty g of protein extracts were resolved in a 7.5% acrylamide gel. The resulting gel was transferred to a polyvinylidene difluoride membrane for 1 h in 10% methanol at 400 mA. The membrane was blocked for 1 h in 1ϫ PBS containing 10% low fat dry milk and 0.1% Tween. YY1 antibody was diluted 1:200 in 1ϫ PBS containing 3% bovine serum albumin, 2% normal goat serum, and 0.1% Tween and incubated with the blot for 1 h at room temperature. The mouse secondary antibody conjugated to alkaline phosphatase was diluted 1:2500 in 1ϫ PBS containing 5% low fat dry milk and 0.1% Tween and incubated with the blot for 1 h at room temperature. We used alkaline phosphatase-conjugated secondary antibodies, the Vistra (Amersham

YY1 Is Increased in Human Heart Failure
Biosciences) enhanced chemiluminescence reagent, an ABI STORM PhosphorImager, and ImageQuant software to visualize and quantify bound antibody.
Coomassie Staining of Polyvinylidene Difluoride Membranes-After Western blot experiments, the membrane was stripped of the antibodies and stained with Coomassie for 1 h. The membrane was than destained for at least 12 h. The proteins on the membrane were quantified on the LiCor scanner.
Exercised-induced Hypertrophy-HCM and wild-type mice were exercised in a voluntary wheel running model for 2 weeks (10).

RESULTS
The YY1 Transcription Factor Binds to Several Sites in the Human ␣MyHC Promoter Proximal Region-YY1 has a loose consensus binding site, with the 5ЈCCAT3Ј core sequence being essential for binding (33). The Ϫ452-bp fragment of the ␣MyHC promoter was scanned for potential YY1 binding sites using the Vector NTI Suite v. 6.0 (InforMax). Three regions containing the 5ЈCCAT3Ј core were found (Ϫ425/Ϫ413 (YY1-1), Ϫ370/ Ϫ360 (YY1-2), and Ϫ325/Ϫ310 (YY1-3)). To test whether these regions were capable of interacting with the YY1 transcription factor, 30-bp probes containing the putative YY1 binding sites were created and used in EMSAs with nuclear extracts prepared from NRVMs (see "Materials and Methods" for probe description). Fig. 2A shows the result of EMSA experiments using the three potential YY1 binding regions in NRVMs. A prominent complex was observed for all three probes ( Fig. 2A,  lanes 2, 5, and 8). To confirm that this complex contains YY1, an anti-YY1 antibody added to the reaction prevented the formation of the complex ( Fig. 2A, lanes 3, 6, and 9). The YY1 antibody was used against GATA4 and Sp1 binding complexes, and no alteration to their binding ability was observed, consistent with the specificity of the antibody (data not shown). To characterize further the interaction of YY1 and the ␣MyHC promoter, mutations in the YY1-1, YY1-2, and YY1-3 binding sites that should disrupt YY1 binding were designed (see "Ma-terials and Methods"). The mutant oligonucleotides were labeled and used as probes against NRVM nuclear extracts. The mutants failed to interact with any proteins in the extract (Fig.  2B, lanes 2, 4, and 6). Specificity of this binding was tested using the wild-type and the mutant double-stranded oligonucleotides as cold competitors. As seen in Fig. 2C (lanes 3, 4, and 5), increasing amounts of the YY1-3 wild-type site as a cold competitor decreased YY1 binding in a concentration-dependent manner. Addition of the mutant cold competitor did not affect YY1 binding, suggesting that the binding is specific (Fig.  2D, lanes 3 and 4).
The YY1 Transcription Factor Is a Repressor of ␣MyHC Promoter Activity-YY1 has been shown to act as a positive or negative regulator of transcription (18). To test the effect of YY1 on the activity of the ␣MyHC promoter, cotransfection experiments were performed in NRVMs. As shown in Fig. 3A, Ϫ452 bp of the ␣MyHC promoter driving the expression of the luciferase gene was cotransfected with increasing amounts of a YY1 expression construct. Dose-dependent repression of promoter activity was observed with 0.04 to 0.2 g of CMV-YY1 DNA transfected into the cells (Fig. 3A). Because YY1 can act as a positive or negative regulator of transcription depending on the promoter and/or cell context, we tested whether its repressive effect on ␣MyHC promoter activity was specific for NRVMs. H9C2 is a rat atrial embryonic cell line that has characteristics of both skeletal and cardiac muscle (34). Differentiation of H9C2 cells with RA has been shown to activate a more cardiac phenotype (34). Undifferentiated H9C2 cells were cotransfected with the ␣MyHC promoter and CMV-YY1. As shown in Fig. 3B, YY1 significantly increases the activity of the ␣MyHC promoter in H9C2 cells. Interestingly, upon differentiation of H9C2 cells by RA, YY1 becomes a repressor of the ␣MyHC promoter. Cotransfection of a YY1 expression construct, and the ␣MyHC promoter-luciferase construct into HeLa cells also resulted in an increase in the activity of the promoter (data not shown). These results suggest that repression of ␣MyHC promoter activity by YY1 occurs in a cell typedependent manner.
EMSA experiments showed that the ␣MyHC promoter has three sites that can interact with YY1. To test whether abolishing the interaction of the YY1 transcription factor with the ␣MyHC promoter would result in an up-regulation of this promoter in cells, point mutations in all three YY1 binding sites and all possible combinations were generated in the context of the ␣MyHC promoter. The resulting constructs were transfected into NRVMs. As shown in Fig. 4A, each of the mutant constructs showed 2-4-fold higher activity when compared with the wild-type promoter construct. Combinations of mutations in the different sites did not result in an increase in the activity of the promoter when compared with mutations in single sites, suggesting that a single YY1 site is sufficient for complete repression by YY1 (see "Discussion").
To show further that the effect of YY1 on the ␣MyHC promoter is attributable to a direct interaction of YY1 with the promoter, cotransfection experiments using the YY1 cDNA and the wild-type ␣MyHC promoter, as well as promoter constructs containing each of the sites mutated individually, or a combination of the three sites mutated, were done in NRVMs. As shown in Fig. 4B, the wild-type and singly mutant constructs were repressed by YY1, but the construct containing all three sites mutated did not respond to YY1. These results suggest that YY1 repression of the ␣MyHC promoter is not the result of a non-specific squelching effect attributable to the presence of an exogenous DNA into the transfection experiments. To the contrary, an ␣MyHC promoter construct that cannot bind YY1 has increased activity in NRVMs.

YY1 Repression of the ␣MyHC Promoter Is Suppressed upon Deletion of the 170 -200-Amino Acid Region-The region of YY1
encompassing 170 -200 amino acids has been shown to be important for its interactions with HDACs and HATs (27). A YY1 construct deleted for this region was used in cotransfection experiments with the ␣MyHC promoter. As shown in Fig. 5, this construct not only fails to repress the activity of the ␣MyHC promoter but increases the activity of the promoter 2-3-fold. These results suggest that the 170 -200-amino acid region of the YY1 transcription factor is of fundamental importance to its function as a repressor of ␣MyHC promoter activity in NRVMs. Because this is the region that has been shown to be important in the interaction with HDACs and HATs in HeLa cells, it is possible that the function of YY1 may be regulated by these proteins in NRVMs.

YY1 Levels Are Increased in Failing Human Left Ventricle and in
Mouse HCM-Considering that ␣MyHC expression (mRNA and protein) is decreased significantly in myocardial pathological hypertrophy and failure, we next tested the hypothesis that the amount of YY1 would be increased in these settings. Western blots were performed using protein extracts from non-failing and failing human left ventricle as well as from cardiac tissue from wild-type mice and a transgenic mouse model of HCM (35). YY1 levels are increased 2-fold in failing human left ventricle (Fig. 6A) and in the hearts from the HCM transgenic mouse model (Fig. 7A). Six different nonfailing and failing left ventricle samples were used, and the experiment was repeated twice. The failing left ventricle samples came from patients 39, 66, 54, 49, 55, and 54 years of age, and the non-failing left ventricle samples came from patients 50, 19, 20, 75, 56, and 55 years of age. Four different heart samples were used for the wild-type and HCM samples. Because hypertrophy increases the general transcription/translation activity of cardiac myocytes, the loading control for the Western blots was done by measuring total protein in each lane through Coomassie staining of the polyvinylidene difluoride membranes (see "Materials and Methods"). To test whether YY1 binding activity correlated with the increase in protein levels observed in the failing hearts, the YY1-2 probe from the ␣MyHC gene was used as a probe in EMSAs against normal and failing extracts. As shown in Fig. 6B, binding of YY1 was increased 3-fold in the failing heart samples. There is, therefore, an antithetic relationship between the amount and binding activity of the YY1 transcription factor and ␣MyHC mRNA and protein. These results are consistent with the hypothesis that YY1 up-regulation in failing hearts could be contributing to the down-regulation of human ␣MyHC gene expression in hypertrophy and heart failure.
YY1 Is Decreased in Exercise-induced Hypertrophy-As described earlier, exercise induces hypertrophy and has been shown to be associated with an increase in ␣MyHC gene expression (8,34). Exercise has also been shown to counteract the deleterious effects of hypertrophy (36). Considering that YY1 decreases the activity of the ␣MyHC promoter in NRVMs and its levels are increased in human failing hearts, we hypothesized that the elevated YY1 levels in HCM mice would be decreased by exercise. Sedentary HCM mouse hearts have also been shown to have elevated ␤MyHC and decreased ␣MyHC (37). To test this hypothesis, male HCM mice, 4 months of age, were exercised on a voluntary cage wheel for 2 weeks. Wildtype mice ran an average of 5.7 km/24 h, and the HCM mice ran an average of 5.4 km/24 h. Protein extracts were prepared, and YY1 levels were assessed by Western blot. YY1 levels were indistinguishable in sedentary and exercised wild-type mice, correlating with ␣MyHC levels (wild-type young adult mouse hearts express 100% of their MyHC as ␣; therefore, there can theoretically be no increase). As hypothesized, YY1 levels in exercised HCM mice returned to wild-type levels (Fig. 7A). To determine whether ␣MyHC protein levels were up-regulated in exercised HCM mice, a Western blot using an antibody specific for the ␣ isoform was performed. As expected, ␣MyHC levels in exercised HCM mice were up-regulated when compared with sedentary HCM mice (Fig. 7B). These results suggest a strong link between YY1 and ␣MyHC levels. Hearts that show an up-regulation of YY1 protein levels also show a decrease in ␣MyHC protein levels and vice versa, suggesting that YY1 plays an important role in the regulation of ␣MyHC gene expression in physiological and pathological states. DISCUSSION In this study, we attempted to gain an understanding of the molecular mechanisms whereby ␣MyHC is decreased in path-ological hypertrophy and heart failure. The transcription factors that have been shown to be important for the regulation of the rat ␣MyHC promoter, i.e. GATA4, MEF2C, NFAT3, and SRF among others, are positive regulators of transcription (15). There is only one report of an ets binding region in the rat ␣MyHC promoter that has been shown to act as a repressor (16), but this region is not present in the human ␣MyHC proximal promoter region (data not shown). Here we show that YY1 binds to the ␣MyHC promoter in a sequence-specific manner and represses its transcription in cardiac cells. We also show that YY1 is increased in the failing human heart and in a transgenic mouse model of HCM but decreased in exercisedinduced hypertrophy.
Cotransfection experiments in NRVMs suggest that YY1 is a strong repressor of human ␣MyHC promoter activity. Squelching of transcription in the presence of an excess of a transcription factor is a phenomenon that has been extensively described (38). In our studies, we have three experiments that show that the repression mediated by YY1 is not attributable to a non-specific squelching effect. First, point mutations in the YY1 binding sites of the human ␣MyHC promoter prevented the binding of YY1 to this promoter. Mutation of these sites also resulted in its up-regulation, suggesting again that the effects of YY1 on the ␣MyHC promoter are as a repressor. Second, cotransfection of the ␣MyHC promoter and the YY1 cDNA in undifferentiated H9C2 cells and in HeLa cells resulted in an up-regulation of activity of this promoter, and upon differentiation of H9C2 cells into a phenotype that is more cardiac-like, YY1 became a repressor of the ␣MyHC promoter, demonstrating that YY1 can have a dual activity even on the same promoter, depending on the cell type context or differentiation state. The cause for this dual activity of YY1 in different cell types is not known, but it is likely attributable to the factors that interact with YY1 and possibly change its function.
Transfection experiments using the constructs containing single sites mutated as well as combinations of these sites mutated suggest that the repressive effect of YY1 on the ␣MyHC promoter is not additive but equivalent for all three sites. Combinations of mutations in the three different sites did not result in a further increase in promoter activity when compared with constructs containing single mutations. At this point, we do not know the reason for this finding, but it could be attributable to the fact that repression by YY1 requires all three sites and that disruption of a single site abolishes YY1 repression activity.
Why YY1 represses transcription of the ␣MyHC promoter in cell types in which this promoter is normally active is not yet known. One hypothesis is that YY1 is a general repressor of transcription in cardiac cells. However, YY1 can activate tran-scription of brain naturietic peptide in cardiac cells (21). Another hypothesis is that there is an equilibrium between YY1 (as a repressor) and activators of the ␣MyHC promoter. In a normal physiological state, YY1 repression of the promoter is counteracted by positive regulators, but in a pathological state, when ␣MyHC activity is repressed, YY1 is capable of overcoming the activators and effectively represses the promoter. The mechanism by which YY1 represses or activates transcription is not clear. Various groups have attempted to characterize functional domains of YY1 that are responsible for its opposing effects. The reports in the literature are controversial, and some regions that are characterized as activators by one group are characterized as repressors by other groups (18). Recently, YY1 has been shown to be itself acetylated and deacetylated by HATs and HDACs, respectively. These studies were done in HeLa cells and two HATs, pCAF and p300, were found to acetylate the 170 -200-amino acid region of YY1, whereas the C-terminal zinc finger was acetylated only by pCAF. Acetylation of this central region increases the repressive activity of YY1 and results in a more efficient interaction with HDACs (27). Acetylation of the C-terminal DNA-binding region most likely decreases the repression activity of YY1 by decreasing its affinity for DNA (27). Our results suggest that the 170 -200-amino acid region plays a fundamental role in the repression of ␣MyHC promoter activity mediated by YY1 in NRVMs. However, preliminary studies from our laboratory suggest that in NRVMs neither pCAF nor p300 is capable of reversing the repressor activity of YY1 (data not shown). YY1 has been shown to be deacetylated by class I HDACs (1, 2, and 3) in HeLa cells. In addition to Class I HDACs, cardiac cells also express the class II HDACs (4, 5, and 6). It is unknown whether the class II HDACs play a role in the repression of YY1 in cardiac cells. In these cells, the best characterized example of HDAC/HAT regulation is the MEF2C-HDAC-calcium calmodulin-dependent kinase pathway (39), and it is possible that YY1 regulation might follow a similar pathway in this context. It is also possible that the regulation of YY1 in cardiac cells is not mediated by acetylation/deacetylation.
The increase in YY1 protein levels in human failing heart extracts and in the HCM mouse model, where ␣MyHC levels are very low, is consistent with the hypothesis that it is a repressor of ␣MyHC gene expression in vivo. Our data are in agreement with a previous report that YY1 levels are increased in failing human heart and in animal models submitted to left coronary occlusion (40). In both cases, a limitation of the technique is that YY1 levels are not measured in individual myocytes. The decrease in YY1 levels in exercised HCM mice is of particular interest. In physiological hypertrophy in wild-type mice, there is a 10% increase in heart weight, no change in ␣MyHC expression, and YY1 levels do not change. However, in HCM, where there is pathological hypertrophy and a decrease in ␣MyHC, exercise reverses the induction of YY1 levels and increases ␣MyHC levels. Together, these results suggest that ␣MyHC levels are tightly linked to YY1.
In conclusion, our results show that YY1 and ␣MyHC are regulated in a antithetical manner and suggest that transcriptional activity of the ␣MyHC promoter is dependent on low levels of YY1 and vice versa. Our results also show for the first time a transcription factor that represses the activity of the human ␣MyHC promoter. Although further investigation is required, YY1 could be part of a pathway that functions to repress the ␣MyHC gene expression in cardiac hypertrophy and failure, contributing to the pathology of human heart failure.