LETTERS p53-induced inhibition of Hif-1 causes cardiac dysfunction during pressure overload

Cardiac hypertrophy occurs as an adaptive response to increased workload to maintain cardiac function. However, prolonged cardiac hypertrophy causes heart failure, and its mechanisms are largely unknown. Here we show that cardiac angiogenesis is crucially involved in the adaptive mechanism of cardiac hypertrophy and that p53 accumulation is essential for the transition from cardiac hypertrophy to heart failure. Pressure overload initially promoted vascular growth in the heart by hypoxia-inducible factor-1 (Hif-1)-dependent induction of angiogenic factors, and inhibition of angiogenesis prevented the development of cardiac hypertrophy and induced systolic dysfunction. Sustained pressure overload induced an accumulation of p53 that inhibited Hif-1 activity and thereby impaired cardiac angiogenesis and systolic function. Conversely, promoting cardiac angiogenesis by introducing angiogenic factors or by inhibiting p53 accumulation developed hypertrophy further and restored cardiac dysfunction under chronic pressure overload. These results indicate that the anti-angiogenic property of p53 may have a crucial function in the transition from cardiac hypertrophy to heart failure. During the development of cardiac hypertrophy, it has been postulated that a mismatch between the number of capillaries and the size of cardiomyocytes develops, leading to myocardial hypoxia. There are various reports indicating a potential relationship between cardiac angiogenesis, cardiac hypertrophy and cardiac function. We thus proposed that cardiac angiogenesis might contribute to the development of cardiac hypertrophy and that its impairment might induce heart failure. We first established a hypertrophy model that shows cardiac dysfunction at chronic stage by performing a severe transverse aorta constriction (TAC). In this model, cardiac hypertrophy gradually developed, reached a peak on day 14 after TAC and decreased afterwards (Fig. 1a–d). Fractional shortening was preserved until day 14 and significantly decreased on day 28 with left ventricular dilation and increased cardiac fibrosis (Fig. 1d, e, and Supplementary Fig. 1a, b). These results suggest that pressure overload initially induced ‘adaptive’ hypertrophy (days 1–14) with preserved cardiac function; however, this adaptive mechanism could not protect the hypertrophied heart against sustained pressure overload, resulting in systolic dysfunction (days 14–28). The number of microvessels per cardiomyocyte increased until day 14 and decreased thereafter (Fig. 1b, c). The number of bromodeoxyuridine (BrdU)-positive endothelial cells, but not that of BrdU-positive cardiomyocytes, was significantly increased (Supplementary Fig. 1c). Consistent with these results was our observation that the expression of angiogenic factors such as vascular endothelial growth factor (VEGF) and angiopoietin-1 (Ang-1) was upregulated in the early phase and decreased in the late phase (Fig. 1f, g). To explain the role of angiogenesis in the development of cardiac hypertrophy, we examined the effects of TNP-470, an inhibitor of angiogenesis, on cardiac hypertrophy. TNP-470 suppressed the increase in the number of microvessels in the heart of mice that had undergone TAC (Fig. 2a). TAC-induced hypertrophy was almost completely inhibited by the treatment with TNP-470 (Fig. 2a, b). Administration of TNP-470 to mice significantly impaired cardiac function at 2 weeks after TAC (Fig. 2b). Similar inhibitory effects of TNP-470 were observed in other models of cardiac hypertrophy such as an angiotensin II infusion model (Supplementary Fig. 2a–d). These results suggest that cardiac angiogenesis is crucially involved in preserving cardiac function as well as in developing cardiac hypertrophy. To examine whether promoting angiogenesis prevents the transition from cardiac hypertrophy towards heart failure, we introduced adenoviral vectors encoding VEGF and Ang-1 directly into the heart and produced pressure overload. Introduction of angiogenic factors enhanced an increase in the number of microvessels compared with that of LacZ after TAC (Fig. 2c). Cardiac hypertrophy was further developed and its function was preserved in the VEGF/Ang-1 group at 4 weeks after TAC (Fig. 2c, d). Conversely, the introduction of a soluble form of Flt-1, an inhibitor of angiogenesis, into the thigh muscle markedly reduced cardiac hypertrophy as well as the number of microvessels compared with that of LacZ, and suppression of this adaptive response caused a further decline in cardiac systolic function at 4 weeks after TAC (Fig. 2c, d). These results indicate that cardiac angiogenesis, which is induced in the early adaptive phase, may be sufficient to maintain cardiac function and that the angiogenesis becomes insufficient to keep the function of hypertrophied hearts in the maladaptive phase, presumably because of decreased expression of angiogenic factors. Cardiomyocyte hypertrophy has been thought to increase diffusion distance, resulting in reduced oxygen supply in the myocardium. Neovascularization associated with cardiac hypertrophy may be attributable to angiogenic factors in cardiomyocytes being upregulated by hypoxia. We therefore examined the expression of Hif1a, a key transcription factor for the hypoxic induction of angiogenic