Apoptosis accompanies myocardiocyte injury induced by ischemia and hypoxia [38]. Cardiomyocyte apoptosis is common to almost all types of heart diseases [38–42]. To alleviate the effect or avoid the occurrence of cardiomyocyte apoptosis, especially the extensive damage caused by cardiomyocyte apoptosis under hypoxia injury, extensive studies were conducted. Cobalt chloride was used to simulate hypoxia in cardiomyocytes at an early stage, and NRF-1 expression was detected in cardiomyocytes. NRF-1 overexpression was found to alleviate cardiomyocyte apoptosis induced by chemical hypoxia. To further investigate effects of NRF-1 on cardiomyocyte apoptosis, especially under actual hypoxic conditions, we used three-gas incubators to reduce the oxygen concentration to 1%. The results showed that the overexpression and inhibition of NRF-1 could affect cell growth and proliferation under normoxic conditions. While NRF-1 overexpression promoted cell growth, its inhibition suppressed cell proliferation. These results indicate that NRF-1 can promote cell proliferation, and NRF-1 inhibition affects cell division and growth, similar to the finding that NRF-1 knockout is lethal in mouse embryos [43]. This result established that NRF-1 plays an important role in the cell viability and growth of individual organisms. Subsequently, we studied the effect of NRF-1 on cardiomyocyte proliferation under different durations of hypoxia. The results showed that in the first few hours (6 h) under hypoxia, the cells maintained a certain proliferation level. This observation may be attributed to the residual oxygen in the solution. However, as oxygen was progressively depleted with the increase in hypoxia duration, cardiomyocyte proliferation decreased in each group. Notably, the overexpression of NRF-1 significantly slowed the decline in cell proliferation, whereas the cell proliferation level decreased significantly upon NRF-1 inhibition. These results indicate that NRF-1 can protect cardiomyocytes from hypoxia and highlight the anti-apoptotic mechanism adopted by NRF-1.
The decrease in cell proliferation under hypoxia may be caused by cell necrosis, apoptosis, and autophagy. To further dissect the role of apoptosis in hypoxia, we observed the level of apoptosis-related molecules. As mentioned above, caspase-3 is the primary executor of apoptosis. In transgenic mice, caspase-3 overexpression increased the infarct size of cardiomyocytes caused by oxygen stress and increased the likelihood of death in mice [44]. Conversely, the downregulation of caspase-3 reduced the apoptotic index and improved cardiac function after myocardial infarction [45, 46]. We first assessed the activity of caspase-3 in different groups of cardiomyocytes under hypoxia to observe whether the decrease in cardiomyocyte proliferation was caused by apoptosis. The results showed that with the increase in the duration of hypoxia, caspase-3 activity gradually increased. In addition, NRF-1 overexpression could alleviate the high levels of caspase-3 activity. After NRF-1 was inhibited, caspase-3 activation became more pronounced. Subsequently, the real-time dynamic observation of cell growth showed that with the increase in caspase-3 activity, the nuclear red fluorescence became more evident. As the severity of nuclear membrane damage increased, nuclear condensation increased as well. However, NRF-1 overexpression significantly inhibited this process. These results further indicate that hypoxia-induced damage caused to cardiomyocytes is partly triggered by apoptosis, and since NRF-1 can alleviate apoptosis, it can prevent the damage caused to cardiomyocytes under hypoxia.
Previous studies on NRF-1 have primarily focused on its effect on mitochondrial function and showed that NRF-1 could regulate the expression of mitochondrial respiratory chain complex gene family members, affecting mitochondrial biogenesis, and increase mitochondrial ATP production [47, 48]. Reportedly, NRF-1 can significantly improve the mitochondrial membrane potential in cardiomyocytes under hypoxia and enhance the mitochondrial respiratory capacity to increase cardiomyocyte viability [5]. Therefore, the anti-apoptotic effect of NRF-1 may be achieved by regulating apoptosis-related proteins associated with the mitochondria. Most of these molecules belong to the BCL-2 family, including some pro-apoptotic molecules, such as Bax, Bak, and Bid, and anti-apoptotic molecules, including BCL-2 and BCL-xL. At high levels, Bax can form dimers with other pro-apoptotic molecules, such as Bak and Bad, and form pore channels in the mitochondrial membrane, which results in cytochrome c release and a change in the mitochondrial membrane potential, eventually triggering apoptosis. BCL-2 and BCL-xL can competitively bind to Bax to reverse the effects of Bax binding to other apoptotic triggers, which helps achieve an anti-apoptotic effect. The interaction between anti-apoptotic and pro-apoptotic proteins of the BCL-2 family can directly determine the fate of different cardiac pathological processes, including myocardial infarction, dilated cardiomyopathy, and ischemic heart disease [49]. For example, BCL-2 can significantly reduce the infarct size caused by apoptosis [16, 50], whereas BCL-xL can inhibit the expression of the pro-apoptotic molecules Bax and Bid through different mechanisms [51]. Our results showed that decreased NRF-1 expression caused the expression of the anti-apoptotic molecules BCL-2 and BCL-xL to also decrease, while that of the pro-apoptotic molecule Bax was not affected. Since NRF-1 expression can be upregulated via human intervention, this increase can induce subsequent upregulation of BCL-2 and BCL-xL expression, thereby preventing or delaying their downregulation induced by hypoxia. Although the level of Bax did not increase significantly with hypoxia, NRF-1 could inhibit its expression. The above results indicate that the anti-apoptotic effect of NRF-1 on cardiomyocytes under hypoxia is achieved by promoting the expression of BCL-2/BCL-xL and inhibiting Bax expression.
These results indicate that NRF-1 exerts an anti-apoptotic effect in hypoxia-induced cardiomyocyte injury. The specific mechanism or the molecular control method adopted by NRF-1 for regulating the process remains to be studied. The central element in hypoxia is the reduction in oxygen concentration. The study of specific genes related to oxidative stress and the regulation of cardiomyocyte apoptosis by proteins encoded by these genes may help understand the regulatory mechanism. As mentioned above, proteins of the HIF family are extremely sensitive to oxygen. Under normoxia, they are degraded by proteases, while they remain stable under hypoxia. Among them, HIF-1 (HIF-1α and HIF-1β) is the most extensively studied molecule. It can affect apoptosis under hypoxia and in cardiac diseases in various ways [52, 53]. Firstly, we assessed whether NRF-1 targets HIF-1α. Transcription factors generally modulate molecular regulation by controlling the transcription levels of genes. The results showed that NRF-1 could inhibit Hif1a mRNA expression and led us to determine whether the effect is directly inhibited by binding or mediated by other regulatory processes. As mentioned above, NRF-1 exhibits a competitive relationship with methylation; however, researchers later found that NRF-1 could promote the expression of DNMT-1 [35] to maintain the level of methylation and further regulate spermatogenesis [13]. Besides maintaining methylation levels, DNMT-1 initiates methylation [52, 54]. Since NRF-1 inhibits the expression of HIF-1α, we attempted to evaluate the effect of the increase in methylation levels caused by methyltransferase-1 on the above relationship. The results showed that NRF-1 could bind to the promoter region of Dnmt1 and regulate its expression, consistent with previous findings [13]. Subsequent studies showed that DNMT-1 expression decreases under prolonged hypoxia, whereas NRF-1 overexpression delayed this decline. Notably, according to an existing prediction website (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3) and the contributions of previous reports [13], an NRF-1-binding sequence is found in human and mouse Dnmt1. Our analysis revealed that the same sequence also exists in rats, which indicates the conservation of the binding sequence and highlights the important regulatory role of NRF-1 in DNMT-1 expression. However, subsequent experiments showed that even though the promoter region of Hif1a predictably possessed a methylation site, there was no significant change in the methylation status before and after hypoxia, as shown by bisulfite sequencing PCR. Possibly, methylation is not the primary regulatory event affecting HIF-1α expression. Based on the above results, we conclude that NRF-1 negatively regulates HIF-1α expression by directly binding to the promoter region of Hif1a.
Previous studies have shown that HIF-1α acts as a protective molecule in cardiomyocytes under stress [55, 56]. It enhances cardiac tolerance to hypoxia in various ways, for instance, by enhancing anaerobic respiration and nucleotide metabolism and reducing cellular oxidative stress [57–60]. However, reportedly, HIF-1α can induce apoptosis, increase the myocardial infarct area, and promote damage [61, 62]. Therefore, the effect of HIF-1α on the heart has yet to be completely deciphered. To confirm the effect of HIF-1α on the apoptosis of rat cardiomyocytes under hypoxia, we performed the relevant experimental studies. The results showed that HIF-1α inhibition reduced cardiomyocyte proliferation under normoxia; however, following hypoxia, the proliferation of cardiomyocytes in the HIF-1α inhibition group decreased more significantly, and caspase-3 activity increased. These results indicate that HIF-1α exerts a protective effect on cardiomyocytes under hypoxia. To elucidate the negative regulatory effect of NRF-1 on HIF-1α and the effect of NRF-1 on cardiomyocyte apoptosis, we specifically inhibited NRF-1 expression in coordination with HIF-1α inhibition. Our data showed that under normoxia, the inhibition of both molecules reduced cardiomyocyte proliferation; however, compared with that in the HIF-1α inhibition group, the reduction in cardiomyocyte proliferation level was alleviated in the NRF-1 inhibition group, and caspase-3 activity was suppressed. Further analysis showed that HIF-1α inhibition increased the expression of the anti-apoptotic molecules BCL-2 and BCL-xL, and the levels of BCL-2/BCL-xL were relatively lower in the NRF-1 and HIF-1α inhibition group than in the HIF-1α inhibition group. Interestingly, HIF-1α inhibition led to an increase in NRF-1 expression, which has not been previously reported. This novel finding could also explain the simultaneous increase in BCL-2/BCL-xL expression with HIF-1α and NRF-1 inhibition. Superficially, HIF-1α inhibition leads to cardiomyocyte injury and apoptosis under hypoxia, which indicates that HIF-1α serves as an anti-apoptotic molecule under hypoxia. However, under the simultaneous inhibition of NRF-1, the cell morphology appeared qualitatively better than that in the HIF-1α inhibition group; this may be related to the fact that NRF-1 inhibition can partially restore HIF-1α levels and alleviate, to a certain extent, cardiomyocyte hypoxia-induced injury. The results are concordant with the protective effect of HIF-1α on cardiomyocytes previously reported and suggest that HIF-1α plays a more significant anti-apoptotic role than NRF-1. The specific inhibition of HIF-1α expression was accompanied by an increase in BCL-2/BCL-xL expression, whereas after the simultaneous inhibition of NRF-1, BCL-2/BCL-xL expression decreased with a relative increase in HIF-1α expression; these findings were consistent with those reported previously by Choi et al. [63], Menrad et al. [64], and Zhao et al. [62], and indicated that BCL-2/BCL-xL is primarily affected by NRF-1 rather than by HIF-1α. The above results suggest that apoptosis caused by NRF-1-mediated inhibition of HIF-1α may not involve BCL-2 family proteins but by hitherto unknown mechanisms.
Since NRF-1 and HIF-1α exert the same anti-apoptotic effect, it is worth investigating why they exhibit mutual inhibition. It remains unknown whether the mutual inhibition is associated with the similar anti-apoptotic effects and the antagonism between the two molecules. HIF-1α regulates downstream genes by forming a dimer with HIF-1β. It has yet to be determined whether NRF-1 is involved in the formation of this dimer. However, our results indicated that NRF-1 could also bind to HIF-1β, and the binding levels gradually decrease under prolonged hypoxia, which reciprocally affects the binding with HIF-1α. In addition, the inhibition of NRF-1 expression led to an increase in the combination of HIF-1α and HIF-1β, suggesting a possible competitive binding between HIF-1α and HIF-1β. We used DMOG and BAY 87-2243 to verify our hypothesis and found that these two drugs did not affect HIF-1β expression and could positively or negatively regulate HIF-1α expression, respectively. The results showed that DMOG promoted HIF-1α expression, inhibited NRF-1 expression, and suppressed the binding between NRF-1 and HIF-1β. However, when BAY 87-2243 was used to inhibit HIF-1α, a contrasting yet consistent phenomenon was observed—HIF-1α inhibition increased NRF-1 expression, as well as increased the binding levels of NRF-1 and HIF-1β. This phenomenon may be related to the adaptation of cells to hypoxic stress. The above results indicate that under normal conditions, NRF-1 acts as a key transcription factor that exerts multiple effects on cardiomyocyte molecular regulation. In the absence of external stimulation, NRF-1 and other nuclear molecules, including HIF-1β, form a key complex and participate in the regulation of mitochondrial function, cell growth, and metabolism. However, under prolonged hypoxia, the expression of NRF-1 decreases gradually, which leads to the loss of cell function. As an adaptation to hypoxia and to promote survival, HIF-1α accumulates, gradually replacing NRF-1, which promotes the binding between HIF-1α and HIF-1β. Certain key molecules, such as CD39, CD73, p53, and LDHA, are expressed, facilitating cell tolerance to hypoxia. This could explain why HIF-1α can promote hypoxia tolerance in cardiomyocytes and aggravate cardiomyocyte injury under different hypoxic conditions [65, 66]. The exact reason for aggravation of cardiomyocyte injury still needs to be investigated.
Lastly, we examined the role of PGC-1α. Several studies have indicated the involvement of PGC-1α in processes related to apoptosis regulation, such as p53 gene-mediated apoptosis, enhancement of mitochondrial recovery to reduce apoptosis, and regulation of the expression of apoptosis-related molecules [67–69]. Results showed that PGC-1α levels gradually decreased with the increase of hypoxia duration, and this change was not affected by NRF-1 inhibition, indicating that PGC-1α is a transcription factor operating upstream of NRF-1. This finding was consistent with previous reports [70–71]. Our results also showed that NRF-1 and PGC-1α could form stable dimers, and the binding levels of NRF-1 and PGC-1α did not change as the duration of hypoxia was prolonged. Next, upon stimulating the cells with DMOG, the same effect observed for NRF-1 was observed. When PGC-1α expression was promoted upon ZLN005 treatment, the levels of NRF-1 and PGC-1α increased and the levels of NRF-1 and HIF-1β binding increased during the initial stages of hypoxia (0 and 6 h), whereas the binding between HIF-1α and HIF-1β was suppressed. However, the enhanced binding effect was gradually reversed by HIF-1α because the levels of the two molecules decreased with the increase in the duration of hypoxia. These results suggest that PGC-1α, as an upstream transcription regulator of NRF-1, is also involved in regulating the competitive binding between NRF-1 and HIF-1α. However, whether PGC-1α directly binds to HIF-1α or indirectly binds to HIF-1β by binding with NRF-1 needs to be studied further.