The present studies provide direct evidence for the importance of the transcription factor YY1 in mediating both basal and ET-dependent human BNP promoter activity. We have demonstrated that YY1 binds to a specific element in the proximal promoter of the hBNP gene, both in vitro and in the context of the intact cell. YY1 has been shown to function as a stimulator, inhibitor or initiator of transcription in different contexts. In cardiac myocytes, YY1 seems to function as an activator of BNP gene transcription. This activation may, in part, be mediated through direct association with the histone deacetylase HDAC2.
YY1 has been shown to be increased in models of heart disease and in patients with heart failure 11, 30
, suggesting a role for YY1 in mediating the transcriptional changes associated with cardiac dysfunction. Indeed, YY1 has been shown to both positively and negatively regulate several genes important in the hypertrophic program. In the cardiac myocyte, YY1 functions as a negative regulator of alpha myosin heavy chain (αMHC) 12
, a key component of sarcomeric structure. In rodent models of cardiac hypertrophy, αMHC is down regulated in favor of the β isoform, which is thought to result in more energy-efficient sarcomeric activation and decreased contractility 1
. YY1 has also been shown to reduce expression of the skeletal α-actin gene 31
, which is increased in hypertrophic states 32
. A mutation which selectively blocks YY1 binding to that promoter resulted in increases in both basal and TGFβ-dependent skeletal α-actin gene transcription implying that YY1 is a constitutive inhibitor of this gene. These changes in cardiac gene expression, taken together with the results in the present study and those of Bhalla et al. 14
with the rat BNP promoter, suggest that YY1 may serve as a point of molecular integration to coordinate expression of the fetal gene program.
YY1 activity can be modified through changes in YY1 expression or YY1 activation, through increased translocation of YY1 from the cytoplasmic to the nuclear compartment 33
or through post-translational acetylation of the YY1 protein 34
. YY1 has been shown to be acetylated by pCAF and p300 and deacetylated by members of the HDAC gene family. In this study, we have shown that ET-1 treatment results in a reduction in acetylated YY1 (Data Supplement, please see http://hyper.ahajournals.org
.) which correlates with transcriptional activation of the BNP promoter. Whether it is this deacetylation (vs. others involving proteins associated with the BNP promoter) that leads the increase in transcriptional activity remains to be defined.
Recently, several members of the HDAC family have been shown to be expressed in the heart and associated with cardiac hypertrophy. In mammals, HDACs are subdivided into three classes termed class I, II and III. Classically, HDACs are thought to associate with and deacetylate histones and, thereby, promote transcriptional repression through condensation of chromatin. Several transcription factors have also been shown to be targets of HDACs, including the myocyte enhancer factor 2 (MEF2) which associates with class II HDACs 35
and YY1 which has been shown to associate with both class I and II HDACs 34, 36
Both class I and II HDAC enzymatic activity is inhibited pharmacologically by the inhibitor, TSA. TSA has been shown to block the development of cardiac hypertrophy induced by thoracic aortic banding (a model of pressure induced hypertrophy), angiotensin II 26, 27
and isoproterenol 25
suggesting a key role for at least some HDACs in the regulation of hypertrophy. In our study, TSA treatment of cardiac myocytes resulted in a dose dependent inhibition of BNP reporter activity, a recognized marker for activation of the hypertrophic gene transcription program. Gene deletion studies of class II HDACs suggest that HDAC5 and HDAC9 act as antagonists of cardiac hypertrophy induced by either aortic banding or constitutive calcineurin signaling 37, 38
. Interestingly, HDAC5 and HDAC9 null mice did not display an enhanced hypertrophic phenotype in response to isoproterenol mediated hypertrophy, raising the possibility of distinct signaling mechanisms associated with the hypertrophic response to β adrenergic stimulation. However, TSA treatment does block isoproterenol mediated hypertrophy. This taken together with the apparent anti-hypertrophic activity of HDAC5 and HDAC9, suggests that class I HDACs may be both pro-hypertrophic and likely targets of TSA treatment in the heart. In support of this, recent studies of HDAC2 null mice have revealed the importance of this class I HDAC in both normal heart development and in hypertrophy 39, 40
. HDAC2 null mice are resistant to isoproterenol mediated hypertrophy and seem to be protected from hypertrophy associated with overexpression of the pro-hypertrophic transcription factor Hod (also known as Hop) 39
. In addition, mice expressing a transgene for HDAC2 in the heart demonstrate increased hypertrophy 39
We have shown that HDAC2 physically associates with the BNP promoter and that the association is inhibited by TSA. This result suggests that HDAC2 participates in the activation of the BNP promoter directly. Of note, ANP gene expression, which closely tracks with BNP expression, is blunted in HDAC2 null mice 29
. In addition, HDAC5 and HDAC9 gene deletion result in enhanced expression of BNP, especially in the setting of enhanced calcineurin signaling. Thus, although YY1 is capable of binding class I and II HDACs, it seems likely that interaction with class I HDACs, specifically HDAC2, is the more probable link to hypertrophy-dependent BNP gene transcription.