A hypoxia-inducible vigilant vector system for activating therapeutic genes in ischemia

Article metrics


Hypoxia represents an endogenous pathophysiological signal underlying cell growth, adaptation and death in a variety of diseases, including ischemic heart diseases, stroke and solid tumors. A vigilant vector system depends on a gene switch which can sense the hypoxia signal occurring in ischemic events and turn on/off protective gene expressions when necessary. This system uses the oxygen-dependent degradation domain derived from hypoxia-inducible factor 1α as the hypoxia sensor and a double-vector system as signal amplifier. For treating ischemic heart diseases, a cardiac-specific MLC-2v promoter is used to deliver transgenes specifically to the heart. When tested in cardiomyocyte cultures, it produced a rapid and robust gene induction upon exposure to low oxygen. In a mouse model for myocardial infarction, the vigilant vectors turned on therapeutic genes such as heme oxygenase-1 in response to ischemia, significantly reduced apoptosis in the infarct area and improved cardiac functions. The hypoxia-regulated gene transfer afforded by the vigilant vectors may provide a powerful tool for delivering therapeutic proteins specifically to ischemic tissues with optimal physiological control.


The devastating consequences of ischemic diseases including acute myocardial infarction, stroke, lower limb ischemia and solid tumors have stimulated research in efficient drug delivery to ischemic cells. Hypoxia (low oxygen) is a common feature of these pathological conditions, leading to cellular injury and adaptation.1, 2 Uncontrolled expression of therapeutic proteins,3 for instance, vascular endothelial growth factor (VEGF), can cause adverse effects such as hemangioma, retinopathy and occult tumor growth.4, 5 Therefore, physiologically regulated gene delivery by low oxygen levels offers great potential for safer gene therapy for ischemic diseases.

In order to target genes to hypoxic cells, several hypoxia-sensitive gene switches have been developed, all of which are based on hypoxia response elements (HRE).6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 Typically, multiple copies of HRE sequences are placed immediately upstream of a basal promoter (eg minimal SV40 promoter) to drive genes of interest. HRE is a cis-acting element residing in the enhancer of many hypoxia-stimulated genes, including VEGF, erythryopoietin and several glycolytic enzymes. Low oxygen upregulated hypoxia-inducible factor 1 (HIF-1) by stabilizing HIF-1α, which accumulates in the cytoplasm and forms functional heterodimers with HIF-1β, which subsequently translocate to the nuclei and bind to HRE to transactivate gene expression. Therefore the oxygen sensitivity of these HRE-based systems indeed depends on the accumulation and functionality of HIF-1 under hypoxic conditions.

Myocardial ischemia, caused by occlusion of coronary artery, is a leading cause for mortality and morbidity worldwide. It results in acute cardiac damage and progressive remodeling that eventually culminate into chronic heart failure. Gene therapy strategies have been designed to protect the heart from ischemic injury. However, one of the major challenges facing these approaches is to precisely control transgene expression in a spatial, temporal and physiological manner. Therefore, attempts have been made to deliver therapeutic genes specifically to ischemic heart tissues, however, with marginal success. Thus far, only one study reported promising results, wherein Prentice et al15 constructed a vector containing HRE and a muscle-specific α-MHC promoter that demonstrated hypoxia-regulated transgene expression in cardiac myocytes. However, in our previous efforts16, 17, 19 in developing such a system by ligating HRE to a cardiac-specific MLC-2v promoter, the results showed that while HRE-SV40 was capable of mediating hypoxia-inducible gene expression, HRE-MLC lost its oxygen sensitivity. Only when exogenous HIF-1 was co-expressed in cells, the responsiveness to hypoxia could be observed.19 A possible explanation is that unfavorable interaction between HRE and MLC-2v promoter might interfere with the binding of HIF-1 to HRE. Moreover, our finding manifested the fact that the hypoxia-regulated gene switch based on HRE is primarily dependent on regulation of HIF-1, because HRE itself cannot sense oxygen change and has to rely on interaction with HIF-1 for hypoxia responsiveness.20 In addition, as HIF-1 is modulated not only by oxygen-dependent but also oxygen-independent pathways,21, 22, 23, 24 HRE-based gene switches may also potentially respond to stimuli other than hypoxia. Therefore, we set out to develop a new gene transfer system that can provide sensitive and robust gene induction in response to low oxygen and can be easily designed to target specific tissues.

The oxygen-dependent degradation (ODD) domain is located in the central region of HIF-1α, which independently mediates rapid degradation of HIF-1α under normal oxygen levels via a ubiquitin–proteosome pathway.25, 26 Under hypoxic conditions, cleavage of the ODD domain is suppressed, thus leading to stabilization of HIF-1α. It has been shown that the ODD domain alone can confer oxygen-dependent instability when fused to another protein.25 Therefore, a chimeric transcriptional factor composed of DNA-binding domain, ODD domain and activation domain was developed to respond directly to changes in oxygen tension and activate gene expression. We have proposed the concept of the vigilant vector, whereby a single administration of a vector with tissue-specific promoter and hypoxia gene switch would lie in wait for a pathological signal, for example, hypoxia.17 Here we report a functional, stable system for hypoxia-induced gene therapy in the heart.


Design of the hypoxia-inducible vigilant vector system

The hypoxia-inducible vigilant vector system contains two vectors, that is, a sensor plasmid (pS) vector and an effector plasmid (pE) vector (Figure 1). The sensor vector encodes a chimeric transcriptional factor (GAL4-ODD-p65) composed of three components, that is, DNA-binding domain of yeast GAL4 (GAL4 DBD), ODD domain of HIF-1α and activation domain of human NF-κB p65 subunit (p65 AD). The effector vector contains six copies of GAL4 upstream activation sequence (GAL4 UAS) linked to E1b TATA box driving the gene of interest. As the ODD domain of HIF-1α is degraded under normoxia while stabilized under hypoxia,25, 26 GAL4-ODD-p65 only accumulates in hypoxic conditions and subsequently transactivates the effector through the interaction between GAL4 DBD and GAL4 UAS. Thus the effector is transcriptionally silent until stimulated by the stabilized and functionally active GAL4-ODD-p65 in hypoxic tissues. Such strategy of designing a chimeric transcriptional factor has been successfully used to generate a regulatory system in response to progesterone administration.27

Figure 1

Diagram of the hypoxia-regulated vigilant vector. The system contains two vectors, a sensor vector and an effector vector. The sensor vector expresses a chimeric transcriptional factor composed of GAL4 DNA-binding domain (GAL4 DBD), HIF-1α ODD domain and p65 activation domain (p65 AD). The effector vector contains six copies of GAL4 upstream activation sequence (GAL4 UAS) and a TATA box driving genes of interest. The chimeric transcriptional factor is degraded under normoxia, but accumulates under hypoxia and thus activates the effector vector. β-Gal, and SEAP, respectively.

In vitro hypoxia-inducible vigilant vector system

To confirm that GAL4-ODD-p65 is degraded under normoxia via a proteosome-mediated pathway, a sensor vector driven by CMV promoter was tested in HEK 293 cells. Western blot analysis demonstrated that GAL4-ODD-p65 was barely detectable at 20% O2, while exposure to 0.5% O2 significantly elevated its concentration by 11±2-fold (P<0.01). Addition of Cbz-LLL (a proteosome-specific inhibitor) to cells cultured at 20% O2 led to accumulation of GAL4-ODD-p65, indicating its degradation under normoxia is proteosome-dependent (Figure 2a).

Figure 2

Upregulation of transgenes delivered by vigilant vector under hypoxia in vitro. (a) Western blot and quantification of GAL4-ODD-p65 under normoxia, 0.5% O2 and proteasome inhibitor Cbz-LLL treatment. Paxillin served as internal control. HEK 293 cells were transfected with 0.5 μg/well of pS-CMV. Then the cells were subjected to 0.5% O2 or 20% O2 for 43 h, or were added with 10 μM Cbz-LLL for 16 h before preparation of cell lysate. Mock cell was only added with Lipofectamine and incubated under normoxia. (b) Dose response of the vigilant vectors in embryonic rat cardiomyocytes (H9c2). In all, 1–4 μg of pS-MLC was transfected into H9c2 cells along with 0.5 μg/well pE-Luc and 20 ng/well pRL-CMV control plasmid. The cells were incubated at either 20% O2 or 1% O2 for 24 h before preparation of cell lysates. Relative luciferase activity represents the ratio of firefly luciferase activity versus Renilla luciferase activity (n=3). (c) Hypoxia responsiveness and expression potency of MLC-Luc (not hypoxia-regulated), HRE-SV40-Luc (HRE-based hypoxia-regulated) and pS-MLC+pE-Luc vigilant vectors (ODD-based hypoxia-regulated) were compared in H9c2 cells. Cells were transfected with 0.5 μg MLC-Luc plasmid, or 0.5 μg HRE-SV40-Luc plasmid, or the mixture of 1 μg pS-MLC and 0.5 μg pE-Luc. pRL-CMV plasmid (20 ng/well) was added as transfection control. Then cells were incubated at either 20% O2 or 1% O2 for 24 h. *P<0.01, 20% O2 versus 1% O2. (d) Selective expression of the MLC-2v vigilant vectors in cardiomyocytes. Cells were transfected with 1 μg pS-MLC and 0.5 μg pE-Luc along with 20 ng pRL-CMV control plasmid prior to incubation at either 20% O2 or 1% O2 for 24 h. *P<0.01, 20% O2 versus 1% O2.

In order to apply the hypoxia-regulatable vectors to ischemic heart diseases we constructed a pS driven by a cardiac-specific MLC-2v promoter17, 28 (pS-MLC) and an effector plasmid encoding luciferase (pE-Luc). To examine the dose response of the double vectors, we co-transfected H9c2 cardiomyoblast cells with increasing amounts of pS (1–4 μg) while keeping a constant amount of pE (0.5 μg). After incubation at 1% O2 or 20% O2 for 24 h, luciferase activities in the hypoxic cells were three- to seven- fold higher than those under normoxia (Figure 2b). The hypoxia inductivity decreased with increasing amounts of the pS due to an elevated basal activity. The optimal sensor-to-effector ratio was identified as 2.0. In addition, the hypoxia-induced expression can be attributed to the ODD domain, because the fusion protein containing only GAL4 DBD and p65 AD did not increase reporter expression under hypoxia (data not shown).

Then we evaluated the hypoxia responsiveness and expression intensity of three expression systems, that is, MLC-Luc (a single plasmid with MLC-2v promoter driving luciferase), HRE-SV40-Luc (a HRE-based hypoxia-inducible vector)6, 10, 11, 13, 14, 16 and pS-MLC+pE-Luc (the hypoxia-inducible vigilant vectors) in H9c2 cells (Figure 2c). As expected, hypoxia at 1% O2 did not stimulate luciferase expression from MLC-Luc. The inductive potency of the OOD-based vigilant system upon hypoxia was comparable to that of HRE-based system, as the luciferase activity in response to 1% O2 was induced by 6.94±1.01-fold by pS+pE-Luc and 7.12±1.52-fold by HRE-SV40-Luc. However, the robust transactivation capacity of GAL4-ODD-p65 increased the power of MLC-2v promoter by over 100-fold, comparing the MLC-Luc single vector and pS-MLC+pE-Luc vigilant vectors. Thus, it permitted the hypoxia-induced expression levels to approach those achieved by powerful promoters such as CMV promoter,16, 19 at values considerably greater than HRE-SV40-Luc. When tested in different cell lines, the vigilant vectors displayed a hypoxia induction ratio of 16±5-fold in primary culture of adult cardiomyocytes but very low expression in MPV endothelial cells (Figure 2d), demonstrating its cardiac selectivity under hypoxia.

Time course of gene expression

Time is a key factor determining the extent of myocardial salvage by clinical intervention.29, 30 Thrombolytic therapy initiated within 2 h of symptom onset significantly improved left ventricular function and reduced infarct size in myocardial infarction patients than later treatment.29. Thus, timely activation of therapeutic genes is necessary for efficient cardiac protection during ischemia. In order to investigate how fast the ODD-based vigilant system can switch on gene expression in response to hypoxia, the time course of hypoxia-induced luciferase activities was studied in H9c2 cells (Figure 3a). Within 2 h, the treated cells (1% O2) demonstrated a 62±32% increase in luciferase activity (P<0.05) higher than that in the control (20% O2). This elevation continued over time and, by 22 h of hypoxia, the activity was 315±23% over control (P<0.01). Further studies indicated that the onset of luciferase activation was within 0.2–1.4 h (data not shown), which falls well within the therapeutic window for cardioprotective genes to act after myocardial ischemia.

Figure 3

Switching on and off of the vigilant vectors. (a) Time course of switch-on function in response to hypoxia. H9c2 cells were transfected with 1 μg/well of pS-MLC, 0.5 μg/well pE-Luc and 20 ng/ well pRL-CMV control plasmid, then were exposed to 1% or 20% O2 for 0–22 h. The cells were retrieved from different time points for luciferase assay. The increased luciferase activity in hypoxia-treated group was expressed as increase percent over that in the control cells (mean±s.d.; n=6–9) (b) Time course of switch-off function after re-oxygenation. H9c2 cells were transfected with 1 μg/well of pS-MLC, 0.5 μg/well pE-Luc and 20 ng/well pRL-CMV control plasmid. The transfected cells in the treated group were first exposed at 1% O2 for 8 h, then returned to 20% O2 for another 21 h. The cells in the control group were continuously incubated at 20% O2. The increased luciferase expression in the treated group is expressed as increase percent over control. (c) Oxygen tension-dependent expression of MLC-OST system. Results from three independent experiments. (d) Simultaneous expression of multiple reporter genes in response to hypoxia. In all, 1 μg/well of PS (pS-MLC) and 0.5 μg/well of three effector plasmids (pE-Luc, pE-SEAP and pE-Lacz, respectively) were co-transfected into H9c2 cells. At 24 h after transfection, the cells were subjected to either 0.5% or 20% O2 for another 24 h before preparing cell lysate. The expression under hypoxia was expressed as X-fold that under normoxia, which was arbitrarily set as 1 (n=3).

To serve as a functional gene switch, the vigilant vectors should be able to turn on and off gene expression in response to changes in oxygen levels. To test the on-and-off function of the vigilant vectors, we exposed transfected H9c2 cells to 1% O2 for 8 h, followed by 20% O2 for 21 h. Compared to the control group that was exposed to a constant level of 20% O2, the treated cells displayed a significant increase in luciferase expression after 8 h of hypoxia (P<0.05), which declined to normal levels after 21 h of re-oxygenation (Figure 3b).

Oxygen concentration levels for induction

To further define the hypoxia-induced gene expression in response to different O2 concentrations, we performed a series of experiments in which H9c2 cells were exposed to 0.5, 0.7, 1.5, 4 or 20% O2 for 24 h. The luciferase activities did not show substantial increase until the oxygen level dropped to 4%, and rose exponentially thereafter between 4% and 0.5% O2 (Figure 3c). The functional range of the vigilant vectors between 4% O2 (30.4 mmHg) and 0.5% O2 (3.8 mmHg) coincides with the therapeutic window of oxygen tension in ischemic hearts and solid tumors, varying from 3.1 to 3.9% in normal tissues to 0–1.3% in the infarct and tumor core zones.31, 32

Hypoxia-induced cell injury involves many mechanisms. Therefore, it is desirable to provide multiple protective transgenes at the same time. To test if one pS could activate multiple genes simultaneously, we co-transfected one pS with three effector plasmids encoding luciferase (Luc), β-galactosidase (β-Gal) and secreted alkaline phosphotase (SEAP), respectively. The cells were exposed to either 20 or 0.5% O2. All three reporter genes showed equally significant increases (16.2–18.5-fold) in their expression under hypoxia (Figure 3d).

In vivo hypoxia-inducible vigilant vector system

To evaluate if the vigilant vector could be switched on by the myocardial ischemic signals in vivo, we injected the mixture of pS and pE-SEAP plasmids into the anterior wall of left ventricle in BALB/c mice (n=4), followed by surgically induced myocardial infarction (MI) one week later. At 7 days after MI, the expression of alkaline phosphatase was clearly visible in the peri-infarct zone, but absent in the noninfarct area (Figure 4a).

Figure 4

In vivo expression of the vigilant vectors in mouse heart infarction model. (a) Expression of SEAP in ischemic hearts. The mixture of 20 μg pS-MLC and 20 μg pE-SEAP was injected into the anterior wall of left ventricle of BALB/C mice followed by MI. The expression of alkaline phosphate was evaluated in the heart sections 7 days after MI. (b) Immunohistochemical localization of expression of vigilant hHO-1 in ischemic hearts. Left: sham operation, right: MI. (c) Western blot analysis of hHO-1. (d) Attenuation of apoptosis in ischemic hearts pretreated with vigilant vector-hHO-1 assessed by TUNEL assay (green). Magnification × 100. *P<0.01, n=6. (e) Vigilant hHO-1 improved left ventricular function post MI. MI (black bar), MI+vigilant HO-1 (gray bar). *P<0.01, **P<0.001, n=6 per group.

We next investigated the cardioprotective effects of therapeutic genes such as heme oxygenase-1 (HO-1),33 delivered by the vigilant vectors in ischemic hearts. The plasmid mixture of pS and pE-hHO-1 (coding human HO-1) was injected into the left anterior ventricular wall of mice 3 days prior to MI induction (n=13). Two control groups were performed, one injected with the same plasmid mixture followed by sham operation (n=8), and the other injected with saline followed by MI operation (n=10). At 7 days after MI, the heart sections were evaluated for hHO-1 expression by immunofluorescent staining. The expression of hHO-1 was markedly activated in the infarct area in the treatment group, but at very low levels in the controls (Figure 4b). This result was further confirmed by Western blot analysis, which demonstrated more than two-fold increase in hHO-1 intensity in the infarct hearts versus noninfarct hearts that were both treated with vigilant vectors carrying hHO-1 (Figure 4c). Interestingly, the expression of hHO-1 was substantially stronger in peri-infarct zone than in infarct core area, where cardiomyocytes could not survive, and therefore might not express exogenous genes anymore. In addition, we evaluated apoptosis in the peri-infarct zone by in situ TUNEL assay (Figure 4d). The quantitative image analysis of apoptotic nuclei indicated that hHO-1 delivered by vigilant vector significantly reduced apoptosis index in the experimental group relative to the control group that was injected with saline and subjected to MI (4.0±0.69 versus 7.78±1.15/hpf, P<0.01, n=6). Furthermore, hHO-1 stimulation in ischemic hearts significantly improved left ventricular function as assessed by considerable increases in maximum dP/dt (P<0.001) and left ventricular developed pressure (LVDP) (P<0.01) 1 week following MI (Figure 4e). It also significantly reduced infarct size from 44.8±2.5% in the control group to 26.7±2.1% in the hHO-1 group (P<0.001).


In the current study, we have developed and characterized a new regulatable system that confers hypoxia-inducible gene transfer. When combined with a cardiac-specific promoter, it efficiently delivered transgenes to hypoxic cardiomyocytes in vitro and ischemic myocardium in vivo. Our results demonstrated that therapeutic gene expression of HO-1 mediated by this system provided significant functional protection to the hearts subjected to myocardial ischemia.

There are several major differences between the ODD-based vigilant vector system and the previous HRE-based system.6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 First, the vigilant vector system is built upon the unique feature of the ODD domain. Therefore, the chimeric sensor GAL4-ODD-p65 is stable and functional and directly regulated by hypoxia, whereas the HRE-based system relies on endogenous HIF-1 to mediate hypoxia responsiveness. Second, SV40 promoter is the only promoter that has consistently demonstrated hypoxia responsiveness when linked to HRE. Substitution with other promoters has not been successful, whereas the ODD-based oxygen sensor driven by MLC-2v promoter is highly sensitive to hypoxia in our study. The problem with HRE might be explained by the phenomenon that placing two cis-acting elements (such as HRE and promoters) in close proximity could interfere with the optimal binding of transcriptional factors to one or both elements.19 Thereby utilization of ODD as oxygen sensor would avoid the likely, yet complex, interactions between HRE, promoters and transcriptional factors. As a result, alteration of promoters to achieve targeted gene transfer would be simplified in the vigilant vector system. Finally, the sensor–effector double vectors adopted by the vigilant vector system amplify gene expression via the powerful p65 activation domain. Indeed, our results indicated that the gene expression level induced by the vigilant vectors was almost 10-fold higher than that of the HRE-based system (Figure 2c).

Our results demonstrated that the vigilant vector can switch on transgenes within 2 h, which falls well within the time window for cardioprotective genes to act after MI.29, 30 Moreover, the system is sensitive to changes in oxygen tension below 4% O2, a range that corresponds to the oxygen gradient within ischemic heart tissues31 and solid tumors.32 This observation was further supported by our in vivo data that the expression of hHO-1 was strongest in the peri-infarct zone. Cardiomyocytes could not survive in the infarct core zone (0% O2) and therefore would not express any proteins.

The in vivo functionality of the ODD-based vigilant vector system was illustrated first by expression of a reporter protein SEAP in the ischemic heart tissues following myocardial infarction, but not in the oxygenated areas. Further, the functionality was corroborated by the selective upregulation of hHO-1 in the peri-infarct areas. This was accompanied by significant improvements in the post-MI cardiac function. HO-1 is an anti-oxidant enzyme with documented cytoprotective activities. Constitutive or regulated overexpression of HO-1 in hearts or organ transplants is capable of reducing cell death caused by ischemia–reperfusion injury.18, 33, 34, 35, 36 This notion is in agreement with our recent report that hypoxia-inducible expression of HO-1 conferred significant protection to the heart following myocardial ischemia.37 The vigilant vector system targets tissue selectively, unlike other HRE-SV40-based gene switch systems which are not tissue selective. Using the ODD-based system in combination with the MLC-2v promoter, we can achieve heart-selective and hypoxia-responsive expression of HO-1 for optimal cardiac protection.37

The ultimate goal of this study was to achieve long-lasting therapeutic effects by delivering hypoxia-inducible gene expression with recombinant adeno-associated virus (rAAV). To establish the model, we first tested it with AAV-based plasmid. To overcome the relatively small packaging limit of rAAV (4.5 kb), we divided the system into two vectors, that is, the sensor vector and the effector vector. Testing with plasmid demonstrated functional interaction between the two vectors in vitro and in vivo. For rAAV, both vectors would need to be administered simultaneously and co-infect efficiently. Results from previous studies have demonstrated highly efficient co-infection of rAAV vectors in vitro and in vivo.38, 39, 40, 41. Indeed, strategies that rely on co-infection and subsequent intermolecular combination of two split rAAV vectors have been successfully used to expand the restricted packaging capacity of rAAV vectors.39, 40 Only rAAV has the limited packaging size that requires double-vector system. In viral vectors that have higher packaging capacity including lentivirus and adenovirus, the double-vector system is unnecessary and can be integrated into one.

In summary, we have developed a novel hypoxia-responsive vigilant vector for regulated gene transfer into hypoxic tissues. It can sense oxygen changes and produce timely and robust gene stimulation when hypoxia occurs. We have shown that myocardial ischemia induced the expression of therapeutic genes carried by the vigilant vector, which in turn provided cardiac protection to ischemic hearts. This gene transfer approach may be beneficial as preventive therapy for patients with or at risk of developing coronary ischemic events. Furthermore, this system is not limited to ischemia in the heart. It also represents a safe and efficient means for delivery of therapeutic proteins to a variety of other clinical manifestations in which hypoxia is a key factor, including stroke, cancer, peripheral vascular diseases and organ transplantation.

Materials and methods

Vector construction

The plasmid pCMV (kindly provided by Dr Sullivan) expresses a chimeric transcription factor consisting of the yeast GAL4 DBD (amino acids 1–93) and the human NF-κB p65 AD (amino acids 283–551) under the control of CMV promoter. The CMV promoter in pCMV was replaced by a 281 bp MLC-2v promoter17 (−264 to +17, Genebank: U26708) to generate pMLC. The ODD domain25 (amino acids 394–603) was amplified by polymerase chain reaction (PCR) from pCEP4/HIF-1alpha (kindly provided by Dr Semenza) and fused in frame between the coding sequence of GAL4 DBD and p65 AD in pMLC to generate the sensor vector pS-MLC. The ODD domain was inserted into pCMV to generate pS-CMV. The effector vector pE-LacZ (Invitrogen) encodes β-galactosidase driven by six copies of 17 bp GAL4 UAS and an adenovirus E1b TATA box. The lacZ coding sequence was replaced by firefly luciferase or SEAP cDNA to generate pE-Luc and pE-SEAP, respectively. SV40 promoter of pGL-SV40 (Promega) was replaced by 281 bp MLC-2v promoter to generate pMLC-Luc. HRE-SV40-Luc plasmid was created by inserting a 68 bp human enolase HRE sequence (-416 to –349, Genebank: X16287) into the 5′ flank of the SV40 promoter of pGL-SV40. All sequences were confirmed by sequencing analysis.

Cell cultures and hypoxic treatment

H9c2, MPV and HEK 293 cells were cultured in Dulbecco's modified Eagle's medium supplemented with sodium pyruvate and 10% fetal bovine serum. Normoxia was defined as 20% O2, 5% CO2 and 75% N2, and hypoxia was defined as 0.5 or 1% O2, 5% CO2 and 94% N2. The oxygen level in the medium was monitored by an OxyLite probe (Oxford Optronix).

Primary cultures of adult rat cardiomyocytes

Adult cardiomyocytes were isolated from the hearts of Sprague–Dawley rats (Harlan). Briefly, the heart was perfused retrogradely using a Langendorff apparatus with perfusion buffer containing collagenase (Type II, Worthington) at 38°C for 1 h. Then the ventricles were minced and incubated at 37°C for 10 min with gentle swirling. Following three low-speed centrifugation and four sedimentation steps, myocytes were resuspended in M199 (Invitrogen) containing 10% fetal bovine serum. The myocytes were plated on laminin-coated tissue culture plates and placed in a 37°C, 5% CO2 incubator. After 1 h the media was removed and replaced with fresh growth medium.

Transient transfection and reporter gene assays

Cells were transfected at a confluence of 50–60%. Transfection was performed with Lipofectamine (Invitrogen) according to the manufacturer's protocol. pRL-CMV or pRL-TK (Promega) encoding Renilla luciferase was used as an internal control to normalize the transfection efficiency. Luciferase assays were performed with dual luciferase assay system (Promega). β-Gal and SEAP activity were assayed with Galacto-light plus and phospha-light system (Applied Biosystems), respectively. Results were quantified with a Monolight 3010 luminometer (BD Biosciences). All experiments were repeated at least twice and measured in triplicate each time.

Mouse MI

MI was induced in male BALB/C mice (7 weeks of age) by ligation of left anterior descending coronary artery (LAD). Mice were anesthetized, intubated and mechanically ventilated. A thoracotomy incision was made in the fourth intercostal space and the proximal LAD was surgically occluded. In sham mice, ligation was placed beside the coronary artery. At 3 days prior to infarction, the thorax was opened and 60 μl vigilant vectors complexed in Polymer (kindly provided by Dr Sullivan) were injected in the anterior wall of left ventricle. Control group received saline injection. Animals were killed 7 days after MI.

Heart function assessment

Mice were anesthetized and a Mikro-tip pressure catheter transducer (Model SPR-671, Millar Instruments) was cannulated into the LV chamber through right carotid artery. The LV pressure was acquired and digitized (PowerLab/8sp, AD Instruments). After stabilization, dP/dt and LVDP were recorded in the close-chest preparation.

Immunohistochemical and immunofluorescent analyses

The hearts were stained for SEAP in X-Phos/NBT Reaction Solution. hHO-1 was detected by immunostaining with mouse monoclonal anti-hHO-1 (BD Biosciences). The 5 μm heart cryosections were stained with fluorescent TUNEL kit (Upstate) for apoptosis assessment. Apoptotic cells were visualized and acquired by confocal microscope (Bio-Rad 1024 ES). TUNEL-positive cells were scored per high-powered field (hpf) in peri-infarct regions.

Western blot analysis

GAL4-ODD-p65 was probed with polyclonal anti-Gal4 antibody (Santa Cruz), and Paxillin probed with an antibody (BD Biosciences). Anti-hHO-1 antibody was purchased from BD Biosciences and GAPDH antibody from Chemicon. The antigen–antibody complexes were visualized by enhanced chemiluminescence (Amersham).

Statistical analysis

Results are presented as means±s.d. Significance between two measurements was determined by Student’s t-test, and in multiple comparisons was evaluated by the one-way ANOVA. Statistical significance was defined as P<0.05.

Accession codes




  1. 1

    Semenza GL et al. Hypoxia, HIF-1, and the pathophysiology of common human diseases. Adv Exp Med Biol 2000; 475: 123–130.

  2. 2

    Acker T, Plate KH . A role for hypoxia and hypoxia-inducible transcription factors in tumor physiology. J Mol Med 2002; 80: 562–575.

  3. 3

    Khan TA, Sellke FW, Laham RJ . Gene therapy progress and prospects: therapeutic angiogenesis for limb and myocardial ischemia. Gene Therapy 2003; 10: 285–291.

  4. 4

    Lee RJ et al. VEGF gene delivery to myocardium: deleterious effects of unregulated expression. Circulation 2000; 102: 898–901.

  5. 5

    Springer ML et al. VEGF gene delivery to muscle: potential role for vasculogenesis in adults. Mol Cell 1998; 2: 549–558.

  6. 6

    Binley K et al. Hypoxia-mediated tumour targeting. Gene Therapy 2003; 10: 540–549.

  7. 7

    Greco O et al. Novel chimeric gene promoters responsive to hypoxia and ionizing radiation. Gene Therapy 2002; 9: 1403–1411.

  8. 8

    Su H, Arakawa-Hoyt J, Kan YW . Adeno-associated viral vector-mediated hypoxia response element-regulated gene expression in mouse ischemic heart model. Proc Natl Acad Sci USA 2002; 99: 9480–9485.

  9. 9

    Ido A et al. Gene therapy targeting for hepatocellular carcinoma: selective and enhanced suicide gene expression regulated by a hypoxia-inducible enhancer linked to a human alpha-fetoprotein promoter. Cancer Res 2001; 61: 3016–3021.

  10. 10

    Boast K et al. Characterization of physiologically regulated vectors for the treatment of ischemic disease. Hum Gene Ther 1999; 10: 2197–2208.

  11. 11

    Binley K et al. An adenoviral vector regulated by hypoxia for the treatment of ischaemic disease and cancer. Gene Therapy 1999; 6: 1721–1727.

  12. 12

    Ruan et al. A hypoxia-regulated adeno-associated virus vector for cancer-specific gene therapy. Neoplasia 2001; 3: 255–263.

  13. 13

    Bainbridge JW et al. Hypoxia-regulated transgene expression in experimental retinal and choroidal neovascularization. Gene Therapy 2003; 10: 1049–1054.

  14. 14

    Binley K et al. Long-term reversal of chronic anemia using a hypoxia-regulated erythropoietin gene therapy. Blood 2002; 100: 2406–2413.

  15. 15

    Prentice H et al. Regulated expression of a foreign gene targeted to the ischaemic myocardium. Cardiovasc Res 1997; 35: 567–574.

  16. 16

    Tang Y, Jackson M, Qian K, Phillips MI . Hypoxia inducible double plasmid system for myocardial ischemia gene therapy. Hypertension 2002; 39: 695–698.

  17. 17

    Phillips M et al. Vigilant vector: heart-specific promoter in an adeno-associated virus vector for cardioprotection. Hypertension 2002; 39: 651–655.

  18. 18

    Pachori AS et al. Hypoxia-regulated therapeutic gene as a preemptive treatment strategy against ischemia/reperfusion tissue injury. Proc Natl Acad Sci USA 2004; 101: 12282–12287.

  19. 19

    Tang Y et al. Vigilant vectors: adeno-associated virus with a biosensor to switch on amplified therapeutic genes in specific tissues in life-threatening diseases. Methods 2002; 28: 259–266.

  20. 20

    Semenza GL . HIF-1 and mechanisms of hypoxia sensing. Curr Opin Cell Biol 2001; 13: 167–171.

  21. 21

    Salnikow K, Su W, Blagosklonny MV, Costa M . Carcinogenic metals induce hypoxia-inducible factor-stimulated transcription by reactive oxygen species-independent mechanism. Cancer Res 2000; 60: 3375–3378.

  22. 22

    Richard DE, Berra E, Pouyssegur J . Nonhypoxic pathway mediates the induction of hypoxia-inducible factor 1alpha in vascular smooth muscle cells. J Biol Chem 2000; 275: 26765–26771.

  23. 23

    Sandau KB, Fandrey J, Brune B . Accumulation of HIF-1alpha under the influence of nitric oxide. Blood 2001; 97: 1009–1015.

  24. 24

    Stroka DM et al. HIF-1 is expressed in normoxic tissue and displays an organ-specific regulation under systemic hypoxia. FASEB J 2001; 15: 2445–2453.

  25. 25

    Huang LE, Gu J, Schau M, Bunn HF . Regulation of hypoxia-inducible factor 1alpha is mediated by an O2-dependent degradation domain via the ubiquitin-proteasome pathway. Proc Natl Acad Sci USA 1998; 95: 7987–7992.

  26. 26

    Sutter CH, Laughner E, Semenza GL . Hypoxia-inducible factor 1alpha protein expression is controlled by oxygen-regulated ubiquitination that is disrupted by deletions and missense mutations. Proc Natl Acad Sci USA 2000; 97: 4748–4753.

  27. 27

    Wang Y, O'Malley Jr BW, Tsai SY, O'Malley BW . A regulatory system for use in gene transfer. Proc Natl Acad Sci USA. 1994; 91: 8180–8184.

  28. 28

    O'Brien TX, Lee KJ, Chien KR . Positional specification of ventricular myosin light chain 2 expression in the primitive murine heart tube. Proc Natl Acad Sci USA 1993; 90: 5157–5161.

  29. 29

    Mathey DG, Sheehan FH, Schofer J, Dodge HT . Time from onset of symptoms to thrombolytic therapy: a major determinant of myocardial salvage in patients with acute transmural infarction. J Am Coll Cardiol 1985; 6: 518–525.

  30. 30

    Heyndrickx GR et al. Depression of regional blood flow and wall thickening after brief coronary occlusions. Am J Physiol 1978; 234: H653–H659.

  31. 31

    Rumsey WL, Pawlowski M, Lejavardi N, Wilson DF . Oxygen pressure distribution in the heart in vivo and evaluation of the ischemic ‘border zone’. Am J Physiol 1994; 266: H1676–H1680.

  32. 32

    Dekker M . Oxygenation of solid tumours. In: Teicher BA (ed). Drug Resistance in Oncology. Marcel Dekker, Inc.: New York, 1993.

  33. 33

    Melo LG et al. Gene therapy strategy for long-term myocardial protection using adeno-associated virus-mediated delivery of heme oxygenase gene. Circulation 2002; 105: 602–607.

  34. 34

    Vulapalli SR et al. Cardioselective overexpression of HO-1 prevents I/R-induced cardiac dysfunction and apoptosis. Am J Physiol Heart Circ Physiol 2002; 283: H688–H694.

  35. 35

    Yet SF et al. Absence of heme oxygenase-1 exacerbates atherosclerotic lesion formation and vascular remodeling. FASEB J 2003; 17: 1759–1761.

  36. 36

    Yet SF et al. Cardiac-specific expression of heme oxygenase-1 protects against ischemia and reperfusion injury in transgenic mice. Circ Res 2001; 89: 168–173.

  37. 37

    Tang YL et al. Protection from ischemic heart injury by a vigilant heme oxygenase-1 plasmid system. Hypertension 2004; 43: 746–751.

  38. 38

    Duan D, Yue Y, Yan Z, Engelhardt JF . A new dual-vector approach to enhance recombinant adeno-associated virus-mediated gene expression through intermolecular cis activation. Nat Med 2000; 6: 595–598.

  39. 39

    Sun L, Li J, Xiao X . Overcoming adeno-associated virus vector size limitation through viral DNA heterodimerization. Nat Med 2000; 6: 599–602.

  40. 40

    Nakai H, Storm TA, Kay MA . Increasing the size of rAAV-mediated expression cassettes in vivo by intermolecular joining of two complementary vectors. Nat Biotechnol 2000; 18: 527–532.

  41. 41

    Rendahl KG et al. Regulation of gene expression in vivo following transduction by two separate rAAV vectors. Nat Biotechnol 1998; 16: 757–761.

Download references


We thank Dr Sean M Sullivan for providing pCMV and advice on the double vector system; Dr Gregg L Semenza for providing pCEP4/HIF-1alpha. This work was supported by the NIH MERIT award HL 27334 to MIP, Postdoctoral Fellowship from American Heart Association to Yao Liang Tang (0325378B) and Predoctoral Fellowship from American Heart Association to Yi Tang (0110140B).

Author information

Correspondence to M I Phillips.

Rights and permissions

Reprints and Permissions

About this article


  • hypoxia
  • regulatable gene delivery
  • ischemic diseases
  • heart
  • myocardial ischemia

Further reading