NMNAT3 depletion decreased mitochondrial NAD level and SIRT3 enzyme activity

Since NMNAT3 is the rate-limiting enzyme in NAD biosynthesis in the mitochondria, NMNAT3 might probably affect the activation of SIRT3 by altering NAD synthesis. To test this hypothesis, NMNAT3 was knockdown by siRNA, then the mitochondrial NAD level and SIRT3 enzyme activity were investigated in cultured cardiomyocytes. As shown in Fig. 8A & B, knockdown of NMNAT3 decreased mitochondrial NAD level, as well as SIRT3 enzyme activity. In contrast, NAD level in cytoplasm was not changed (Fig. 8C). Additionally, the mRNA levels of ANF and BNP were significantly increased by NMNAT3 depletion (Fig. 8D).

SIRT3 did not interact with or deacetylate NAMPT

NAMPT is an important NAD biosynthesis rate-limiting enzyme, however, its presence and functions within the mitochondria is controversial [30, 31]. To investigate the subcellular distribution of NAMPT in NRCMs, western blot was used to analyze the protein expression of NAMPT in the mitochondria. As shown in Fig. 9A, NAMPT was predominantly expressed in the cytosol, but not in the mitochondria. Co-immunoprecipitation experiments were conducted to test whether SIRT3 interacts with NAMPT. As shown in Fig. 9B, SIRT3 was detected by western blot after NAMPT was immunopurified from whole cell extracts or mitochondria extracts, suggesting that there was no direct interaction between SIRT3 and NAMPT. The acetylation levels of NAMPT were assessed using an acetylated lysine antibody. As shown in Fig. 9C, acetylation level of NAMPT remain unchanged in NRCMs treated with Ang II. Transfection with either SIRT3-Flag or H248Y mutant, did not affect the acetylation level of NAMPT (Fig. 9C). These data suggested that SIRT3 specifically interacted with and deacetylated NMNAT3, but not NAMPT, in cultured cardiomyocytes.


Permanent contractility of the heart requires continuous energy production, which is supplied mainly by mitochondrial respiration [32]. Regulation of mitochondrial function is crucial to maintain the cardiac contractility. SIRT3, a mitochondria-localized sirtuin, has been reported to regulate almost every major aspect of mitochondrial biology, including detoxification of reactive oxygen species (ROS), nutrient oxidation, ATP generation, mitochondrial dynamics, and the mitochondrial UPR (unfolded protein response) [33-35]. Recent reports have shown that SIRT3 has multiple protective effects in cardiomyocytes. Downregulation of SIRT3 has been observed in animal models of cardiac dysfunction [18, 36]. Deficiency of SIRT3 leads to spontaneous cardiac hypertrophy in non-stressed hearts and accelerates heart failure in response to pressure overload [15, 16]. The present study has demonstrated that the mRNA level, the protein expression and the enzyme activity of SIRT3 were reduced in hypertrophic cardiomyocytes stimulated by Ang II and in the hearts of Ang II-treated mice (Fig.1B-D & Fig.2). These observations were in line with Benigni’s observations that Ang II downregulated mRNA level of SIRT3 and candesartan, an Ang-II type I (AT1) receptor antagonist, prevented this downregulation [37]. Our observations are also consistent with the findings observed by Mohsen et al. who showed that SIRT3 protein expression in left ventricular ischemic tissue were reduced during acute myocardial ischemia reperfusion and that administration of losartan could normalize SIRT3 protein level in the ischemic heart [38]. However, the present findings do not support the observations in a previous study showing that Ang II induced SIRT3 expression in human endothelial cells [39]. It is possible that this discrepancy arises from the differences of cell types and experimental settings. Moreover, overexpression of SIRT3 attenuated Ang II-induced cardiomyocyte hypertrophy, as evidenced by the decreased expression of hypertrophic markers (ANF and BNP), as well as the decrease of cell surface area (Fig. 3C and D). On the contrary, SIRT3 depletion or deficiency aggravated the hypertrophic response induced by Ang II (Fig. 4B-D). Bioavailable NMN

Several mechanisms are proposed to mediate the anti-hypertrophic effect of SIRT3, including the activation of FOXO3a and LKB1, and prevention of aberrant ROS production [18]. However, the downstream targets of SIRT3, especially those associated with cardiovascular diseases, are not yet fully understood. Since SIRT3 is a NAD-dependent mitochondrial deacetylase, intracellular level of NAD, especially mitochondrial level of NAD, is essential for the activation of SIRT3. Recently, it has been reported that pathological cardiac hypertrophy was associated with depletion of cellular NAD, and NAD supplementation blocked Ang II-induced cardiac hypertrophy via activation of SIRT3 [28]. Thus, enzymes involved in NAD biosynthetic pathways may potentially affect the activation of SIRT3.

NAMPT catalyzes the first reversible step in NAD biosynthesis and nicotinamide (NAM) salvage. It is considered as an important enzyme to regulate NAD consumption, notably, the aging-associated histone deacetylase SIRT1 [40]. However, whether NAMPT is present in the mitochondria remains controversial [30, 41, 42]. The present study aimed to investigate the subcellular distribution of NAMPT in NRCMs. As shown in Fig. 9A, the enzyme was mainly present in the cytosol, but was absent in the mitochondria. Moreover, NAMPT was not involved in the anti-hypertrophic effect of SIRT3 in Ang II-induced cardiomyocyte hypertrophy, as supported by the following observations: (1) no physical interaction between SIRT3 and NAMPT was observed in NRCMs (Fig. 9B); (2) the acetylation level of NAMPT was not changed in Ang II-induced cardiomyocyte hypertrophy (Fig. 9C); (3) overexpression of SIRT3 did not affect the acetylation level of NAMPT (Fig. 9C).