Preconditioning in cultured cardiomyocytes elevates the expression of several protective genes including Glut-4 and heat shock protein (HSP)70. Hypoxia-inducible factor-1 (HIF-1) is known to mediate the transcriptional activation of hypoxia-responsive genes. In this study, we examined the effect of adenovirus-mediated expression of constitutively stable hybrid forms of HIF-1α on cardiomyocyte viability and gene expression. Cultured neonatal rat cardiomyocytes were subjected to simulated ischemia-reperfusion with or without preinfection with recombinant adenoviral vectors [Ad2/HIF-1α/herpes simplex virus protein VP16 and Ad2/HIF-1α/nuclear factor-κB (NF-κB)]. Cellular viability and mRNA levels of several cardioprotective genes were measured. We demonstrated that infection with Ad2/HIF-1α/VP16 and Ad2/HIF-1α/NF-κB mimicked the upregulation of the mRNA levels of vascular endothelial growth factor (VEGF), Glut-1, Glut-4, HSP70, and inducible NO synthase (iNOS) and the protection of cultured neonatal rat cardiomyocytes by late-phase preconditioning against simulated ischemia-reperfusion. The same dose of a control viral vector expressing no transgene had no effect. Preconditioning also elevated HIF-1α protein levels. These results suggest that adenovirus-mediated expression of HIF-1α/VP16 or HIF-1α/NF-κB, a constitutively stable hybrid transcriptional factor, protected cultured neonatal cardiomyocytes against simulated ischemia-reperfusion injury by inducing multiple protective genes.
- ischemic preconditioning
- myocardial protection
ischemic preconditioning elicits a biphasic myocardial protection (27). An early phase of protection that develops within minutes after the preconditioning stimulus lasts for 2–3 h, whereas a late phase of protection that occurs 12–24 h later lasts for 3–4 days (8, 21, 23). A number of chemical signals that trigger the development of the late phase of preconditioning, including adenosine, NO, reactive oxygen species, and opioid receptor agonists, have been identified (8, 23). Activation of protein kinase C (4), protein tyrosine kinases (16), and mitogen-activated protein kinases (33) by these triggers leads to increased expression of the mediators of the late phase of preconditioning. Myocardial protein synthesis of several mediators (or effectors), including inducible NO synthase (iNOS), cyclooxygenase-2 (COX-2), aldose reductase, antioxidant enzymes, and heat shock proteins (HSPs), and activation of ATP-sensitive K+ (KATP) channels play a critical role in late-phase preconditioning (8, 15, 21, 23, 24, 29). It has been proposed that activator protein 1 (AP-1) and nuclear factor-κB (NF-κB) may mediate the transcriptional activation of iNOS, COX-2, aldose reductase, and other genes (22, 36).
A critical component of myocardial ischemia is hypoxia, which triggers a wide range of profound cellular responses, including rapid posttranslational modifications of proteins and regulation of gene expression. The paradigm for gene regulation in response to hypoxia involves oxygen sensing followed by a series of signal transduction mechanisms that lead to induction or repression of transcription and changes in mRNA stability or the rate of translation (10). Hypoxia-inducible factor-1 (HIF-1), a heterodimeric basic helix-loop-helix-PAS (bHLH-PAS) domain transcription factor, mediates the transcriptional activation of hypoxia-responsive genes (32). HIF-1 is composed of two subunits, HIF-1α and HIF-1β (aryl hydrocarbon nuclear translocator). Whereas HIF-1β is constitutively expressed, the stability, DNA binding capability, and transcriptional activity of the HIF-α subunit are directly controlled by the intracellular oxygen concentration (9, 12, 32). Recent studies have demonstrated that protection of rat brain and retinal tissues by prior exposure to an episode of sublethal hypoxia is associated with elevated levels of the HIF-1α protein (6, 7, 14, 30). In the neonatal rat brain HIF-1 target genes including vascular endothelial growth factor (VEGF) and Glut-1 have been shown to be upregulated by preconditioning stimulus. These results imply that HIF-1 may play a role in preconditioning of neural tissues. Indeed, several mediators and other cardioprotective genes including iNOS, HSP70, erythropoietin (EPO), and VEGF are also known hypoxia-responsive or HIF-1 target genes. Recently, we demonstrated (34, 37) that expression of HIF-1α/VP16, a constitutively stable hybrid of HIF-1α and herpes simplex virus (HSV) VP16 protein, mimics the angiogenic response to hypoxia in vitro and in vivo by inducing multiple growth factors. In human fetal cardiac cells, adenovirus-mediated expression of HIF-1α/VP16 upregulated several known HIF-1-inducible genes, including VEGF, Glut-1, Glut-3, and glycolytic enzymes (18). Insulin growth factor binding protein-3 (IGFBP-3), gp130, survivin, and TNF-α-inducible protein were also upregulated (18). These genes may directly or indirectly promote cell survival by various mechanisms. However, it is not clear whether adenovirus-mediated expression of HIF-1α/VP16 is capable of protecting cardiomyocytes against ischemia-reperfusion injury.
In this study, we investigated the effects of adenovirus-mediated expression of hybrid forms of HIF-1α on ischemia-reperfusion injury and the mRNA levels of known preconditioning mediators. We also measured the cellular protein levels of HIF-1α and the mRNA levels of several HIF-1 target genes in a cultured neonatal rat cardiomyocyte model of preconditioning (2, 13, 23, 28, 35, 38). Our results suggest that expression of a constitutively stable hybrid HIF-1α protects cultured neonatal cardiomyocytes against simulated ischemia-reperfusion injury by inducing multiple protective genes.
MATERIALS AND METHODS
Construction of recombinant adenoviral vectors.
Ad2/HIF-1α/VP16, Ad2/GFP, and Ad2/CMVEV (empty adenoviral vector), which encode the HIF-1α/VP16 hybrid, green fluorescent protein (GFP), and no transgene, respectively, were constructed as described previously (11, 18). The HIF-1α/VP16 hybrid was composed of the DNA-binding and dimerization domains from HIF-1α and the transactivation domain of HSV VP16 protein (34). The HIF-1α/NF-κB hybrid was constructed as HIF-1α/VP16, except that the transactivation domain of human NF-κB was used.
The investigation conforms to the NIH Guide for the Care and Use of Laboratory Animals (NIH Pub. No. 86-23). Neonatal rat ventricular cardiomyocytes were isolated and cultured as described previously (11). Briefly, the cells were seeded at a density of 2.0 × 105 cells/cm2 on rat type I collagen-coated dishes and maintained in Dulbecco's modified Eagle's medium supplemented with 10% (vol/vol) fetal bovine serum for 48–72 h before use in the studies.
Effect of preconditioning on cardiomyocyte death.
As described previously (23, 35), preconditioning was created by placing the cultured neonatal rat ventricular cardiomyocytes in a hypoxic chamber (1% O2, 5% CO2; 37°C) for 3 h and then returned to control conditions (21% O2, 5% CO2; 37°C) for 24 h, followed by simulated lethal ischemia (depletion of serum and glucose and hypoxia) for 2 h. The cardiomyocyte cultures were then maintained under control conditions for an additional 15 h. The cellular level of 3-(4,5-dimethylthiazol-2-yl-5)-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS), indicative of the mitochondrial function in living cells and cell viability, and lactate dehydrogenase (LDH) levels in the culture medium were measured with the CellTiter96 Aqueous One kit (Promega) and the TOX-7 assay kit (Sigma), respectively (11).
Measurement of mRNA levels.
The mRNA levels were measured with TaqMan 5′ nuclease fluorigenic quantitative PCR, as described previously (5). Briefly, primers and probes (Table 1) were designed and synthesized according to ABI-Perkin Elmer guidelines (Foster City, CA) and synthesized by Qiagen-Operon (Alameda, CA). Standard curves for each gene of interest were performed in duplicate. The mRNA levels of the gene of interest were normalized to 18S rRNA and expressed as fold increase over control.
Measurement of endogenous HIF-1α protein levels.
Endogenous HIF-1α protein levels were detected by Western blotting. Briefly, total protein (60 μg) from cardiomyocyte lysates was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to nitrocellulose membranes. The filters were blocked with 5% milk, followed by incubation with a mouse anti-human HIF-1α monoclonal antibody (1:500 dilution; BD Bioscience), and finally developed with chemiluminescence (Santa Cruz Scientific).
Data are expressed as means ± SE. The number of samples examined is indicated by n. Data were analyzed by ANOVA, followed by a modified Student's t-test. A probability value of <0.05 was considered statistically significant.
Effect of preconditioning on HIF-1α mRNA and protein levels.
As expected (2, 13, 28, 35, 38), subjecting cultured cardiomyocytes to an episode of sublethal hypoxia decreased cell death induced by subsequent simulated lethal ischemia-reperfusion, as measured by the MTS and LDH assays (Fig. 1). To examine the involvement of HIF-1 in preconditioning, both mRNA and protein levels of HIF-1α were measured. Exposure of cultured cardiomyocytes to an episode of sublethal hypoxia for 20 min did not affect HIF-1α mRNA levels, as measured by TaqMan analysis (Fig. 2). In contrast, HIF-1α protein levels were transiently increased after preconditioning, as detected by Western blotting (Fig. 3). The elevation of HIF-1α protein levels by preconditioning was comparable to that induced by desferrioxamine, an iron antagonist that inhibits the prolyl hydroxylation of HIF-1α and consequently causes a well-characterized “hypoxia-mimetic” effect (9, 12, 32). These results suggest that HIF-1 may be involved in preconditioning of cultured neonatal cardiomyocytes.
Effect of preconditioning on mRNA levels of protective HIF-1 target genes.
To examine the role of HIF-1 target genes in the protection of cultured cardiomyocytes by preconditioning, the mRNA levels of several known cardioprotective genes were measured by TaqMan analysis. The mRNA levels of VEGF, Glut-1, Glut-4, and HSP70 were significantly upregulated 1 h after exposure to sublethal hypoxia (Fig. 4). The elevation of Glut-4, VEGF, and HSP70 mRNA levels persisted to 24 h after preconditioning, whereas Glut-1 expression was transient and returned to basal levels. These results suggest that in cultured neonatal cardiomyocytes HIF-1 may play a role in preconditioning by activating the expression of hypoxia-responsive genes.
Protection of cardiomyocytes against ischemia-reperfusion injury by infection with Ad2/HIF-1α/VP16 or Ad2/HIF-1α/NF-κB.
To assess whether expression of exogenous HIF-1α protein exhibits protection of cultured cardiomyocytes, a constitutively stable hybrid form of HIF-1α was expressed with adenoviral vectors. As previously reported (11), infection with Ad2/GFP at 30 and 100 multiplicities of infection (MOI) transduced 70 ± 3% and 95 ± 1% of the cultured cardiomyocytes, respectively, with little effect on cell viability. Infection of cultured cardiomyocytes with Ad2/HIF-1α/VP16 (30 and 100 MOI) 48 h before simulated ischemia-reperfusion effectively reversed cardiomyocyte death, whereas Ad2/CMVEV (100 MOI) had no effect (Fig. 5). These results suggest that adenovirus-mediated expression of constitutively stable forms of HIF-1α mimics the protection of cultured cardiomyocytes by preconditioning. Ad2/HIF-1α/NF-κB exhibited a similar protection of cultured cardiomyocytes, indicating that the protection of cultured neonatal cardiomyocytes by adenovirus-mediated expression of hybrid HIF-1α is independent of the origin of the transactivation domain.
Effect of infection with Ad2/HIF-1α/VP16 or Ad2/HIF-1α/NF-κB on expression of protective genes.
To assess the molecular mechanism by which the constitutively stable forms of HIF-1α mimic the later phase of preconditioning, the mRNA levels of several known mediators of ischemic preconditioning were measured in cardiomyocytes infected with Ad2/HIF-1α/VP16 or Ad2/HIF-1α/NF-κB. Adenovirus-mediated expression of HIF-1α/VP16 or HIF-1α/NF-κB increased the expression of Glut-1, Glut-4, iNOS, and HSP70 in a dose-dependent manner (Fig. 6). VEGF expression was also significantly upregulated. Infection with Ad2/CMVEV had no effect, suggesting that adenoviral infection per se does not upregulate the expression of these genes. However, infection with Ad2/HIF-1α/VP16 or Ad2/HIF-1α/NF-κB did not change endothelial NO synthase (eNOS) mRNA levels (data not shown). Whereas VEGF, Glut-1, Glut-4, and iNOS are known HIF-1 target genes, our studies are the first to demonstrate that hypoxic induction of HSP70 is mediated by HIF-1. These results suggest that in cultured neonatal cardiomyocytes the constitutively stable forms of HIF-1α mimic the later phase of preconditioning by inducing multiple protective genes.
Protection of cultured neonatal cardiomyocytes by adenovirus-mediated expression of constitutively stable hybrid of HIF-1α.
In this study, we demonstrated that in cultured neonatal cardiomyocytes, infection with Ad2/HIF-1α/VP16 reduced ischemia-reperfusion injury in a dose-dependent manner. Similar protection of cultured neonatal cardiomyocytes was achieved by infection with Ad2/HIF-1α/NF-κB. These results suggest that adenovirus-mediated expression of a constitutively stable hybrid form of HIF-1α, namely, HIF-1α/VP16 or HIF-1α/NF-κB, is capable of protecting cultured neonatal cardiomyocytes against simulated ischemia-reperfusion, which is independent of the transactivation domain of the hybrid transcriptional factor.
In response to cellular hypoxia, cardiomyocytes adapt to consuming less oxygen by shifting ATP production from mitochondrial fatty acid β-oxidation to glycolysis (3, 10). Glucose-insulin-potassium infusion improves the survival of patients with ischemic heart disease, partially because of enhanced glucose uptake and glycolysis by the ischemic myocardium (1). Studies in transgenic mice overexpressing HSP70 and in rats that had received HSP70 expression vectors also indicate that this protein protects myocardium against ischemia-reperfusion injury (17, 24, 26, 29, 31). Recently we reported (18) that in human fetal cardiac cells, adenovirus-mediated expression of HIF-1α/VP16 upregulated VEGF, Glut-1, Glut-3, IGFBP-3, gp130, survivin, TNF-α-inducible protein, and glycolytic enzymes. In the present study, we observed that adenovirus-mediated expression of either HIF-1α/VP16 or HIF-1α/NF-κB significantly increased mRNA levels of Glut-1, Glut-4, and HSP70. Upregulation of these genes may partially explain the protection of cultured neonatal cardiomyocytes against simulated ischemia-reperfusion by adenovirus-mediated expression of either HIF-1α/VP16 or HIF-1α/NF-κB. It is known that HIF-1 upregulates expression of glucose transporters and glycolytic enzymes at the transcriptional level by binding to the hypoxia-responsive elements in the promoter of these genes (32). It remains to be determined whether HIF-1α/VP16 or HIF-1α/NF-κB activates HSP70 transcription by binding to its promoter.
NO plays a dual role in the pathophysiology of the late phase of preconditioning (8, 23). Enhanced biosynthesis of NO by eNOS is essential to trigger the late phase of ischemia-induced or exercise-induced preconditioning. Enhanced NO production by iNOS is obligatorily required to mediate the antistunning and anti-infarct actions of the late phase of preconditioning elicited by different stimuli. The generation of NO associated with an ischemic stress initiates a cascade of events that involves the activation of several kinases and transcriptional factors and culminates in the upregulation of the iNOS gene.
It is known that hypoxic activation of iNOS transcription is mediated by HIF in cardiomyocytes (19, 25). The increase of iNOS mRNA levels by the infection of cultured neonatal cardiomyocytes with Ad2/HIF-1α/VP16 or Ad2/HIF-1α/NF-κB implies that the upregulation of this protein is an important protective mechanism by expression of a constitutively stable hybrid HIF-1α in this particular experimental setting. Together, our results suggest that expression of a constitutively stable hybrid HIF-1α protected cultured neonatal cardiomyocytes against simulated ischemia-reperfusion injury by inducing multiple protective genes.
It has been reported that ischemic preconditioning upregulates VEGF expression and neovascularization, which in return reduces infarct size on subsequent lethal ischemia (20). HIF-1 activates VEGF transcription by binding to the hypoxia-responsive element of its promoter (32). Recently we reported (5, 18, 34) that HIF-1α/VP16 induced VEGF expression in various cells including rat ventricular myocytes. In the present study, we demonstrated that adenovirus-mediated expression of either HIF-1α/VP16 or HIF-1α/NF-κB significantly increased VEGF mRNA levels, suggesting that the upregulation of VEGF expression by constitutively stable hybrid forms of HIF-1α is independent of the transactivation domain. It is well known that administration of selective adenosine receptor (A1 and A3) agonists or pharmacological NO donors in the absence of ischemia induces a delayed protective effect on cardiomyocytes against lethal ischemia-reperfusion that is similar to that observed during ischemic preconditioning in vivo, presumably through protein kinases/AP-1- and/or NF-κB-mediated gene expression (8). Adenosine receptor agonists and NO can also increase myocardial oxygen availability through their vasodilator effect. Whether VEGF can directly act on cardiomyocytes and exhibit protective effects remains to be determined. One may speculate that VEGF expression induced by a constitutively stable hybrid form of HIF-1α may improve myocardial perfusion in vivo through increased vascular permeability and angiogenesis.
Involvement of HIF-1 in protection of cultured neonatal cardiomyocytes by preconditioning.
The late phase of preconditioning requires simultaneous activation of multiple mediator/effector genes (8). It has been suggested that AP-1 and NF-κB may mediate the transcriptional activation of several mediators that are involved in ischemic preconditioning (22, 36). It is also known that transcription of iNOS, an important mediator/effector of myocardial preconditioning (8), is directly activated by HIF-1 in response to hypoxia (9, 12, 19, 32). Protection of rat brain and retinal tissues against subsequent lethal ischemia by prior exposure to sublethal hypoxia is associated with elevated levels of the HIF-1α protein (6, 7, 14). In this study, we demonstrated that in isolated neonatal rat cardiomyocytes the protein levels of HIF-1α were elevated by preconditioning. The mRNA levels of several hypoxia-responsive genes including iNOS, HSP70, Glut-1, and Glut-4 were upregulated in response to preconditioning. Significant elevation of HSP70 mRNA levels persisted up to 24 h after preconditioning. It is conceivable that upregulation of Glut-1, Glut-4, and iNOS is mediated by HIF because they are known HIF-1 target genes (19, 25, 32). Furthermore, expression of a constitutively stable hybrid HIF-1α protected cultured neonatal cardiomyocytes against simulated ischemia-reperfusion injury and induced a similar set of protective genes. These results suggest that HIF-1 is involved in the development of late-phase preconditioning in cultured neonatal cardiomyocytes by inducing multiple protective genes. It should be noted that the involvement of HIF in preconditioning of cultured cardiomyocytes against ischemia-reperfusion injury has been demonstrated by overexpression experiments only. The cause-effect relationship remains to be established by eliminating the expression of the endogenous α-subunits of HIF in cultured cardiomyocytes in vitro and/or myocardium in vivo.
Limitations of studies.
In our studies, cultured neonatal cardiomyocytes were subjected to simulated ischemia-reperfusion (23). Adult cardiomyocytes die or dedifferentiate in long-term culture, which is required for manipulating gene expression and elucidating the molecular mechanisms underlying the late phase of preconditioning. Neonatal cardiomyocytes reasonably recapitulate the key features of ischemic preconditioning, including simulated ischemia, the temporal relationship between initiating and lethal simulated ischemia, the involvement of ligands to G protein-coupled receptors, and protein kinase C dependence (23). However, cultured neonatal cardiomyocytes are phenotypically different from working adult cardiomyocytes in the whole heart in vivo. In addition, these cells undergo isolation and culture manipulation and cannot be subjected to true ischemia-reperfusion. Because of these limitations, the molecular mechanisms observed in the cultured neonatal cardiomyocyte model may not reflect those in the aged working heart in vivo.
In summary, we have demonstrated that adenovirus-mediated expression of HIF-1α/VP16 or HIF-1α/NF-κB, constitutively stable hybrid transcriptional factors, protects cultured neonatal cardiomyocytes against simulated ischemia-reperfusion injury by inducing multiple protective genes. Several lines of evidence suggest the involvement of HIF-1 in late-phase preconditioning in cultured neonatal cardiomyocytes. Our study provides new insights into the mechanism underlying the protection of cultured neonatal cardiomyocytes by simulated ischemic preconditioning.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society