Thioredoxin-1 Gene Therapy Enhances Angiogenic Signaling and Reduces Ventricular Remodeling in Infarcted Myocardium of Diabetic Rats
Background— The present study evaluated the reversal of diabetes-mediated impairment of angiogenesis in a myocardial infarction model of type 1 diabetic rats by intramyocardial administration of an adenoviral vector encoding thioredoxin-1 (Ad.Trx1). Various studies have linked diabetes-mediated impairment of angiogenesis to dysfunctional antioxidant systems in which thioredoxin-1 plays a central role.
Methods and Results— Ad.Trx1 was administered intramyocardially in nondiabetic and diabetic rats immediately after myocardial infarction. Ad.LacZ was similarly administered to the respective control groups. The hearts were excised for molecular and immunohistochemical analysis at predetermined time points. Myocardial function was measured by echocardiography 30 days after the intervention. The Ad.Trx1-administered group exhibited reduced fibrosis, oxidative stress, and cardiomyocyte and endothelial cell apoptosis compared with the diabetic myocardial infarction group, along with increased capillary and arteriolar density. Western blot and immunohistochemical analysis demonstrated myocardial overexpression of thioredoxin-1, heme oxygenase-1, vascular endothelial growth factor, and p38 mitogen-activated protein kinase-β, as well as decreased phosphorylated JNK and p38 mitogen-activated protein kinase-α, in the Ad.Trx1-treated diabetic group. Conversely, we observed a significant reduction in the expression of vascular endothelial growth factor in nondiabetic and diabetic animals treated with tin protoporphyrin (SnPP, a heme oxygenase-1 enzyme inhibitor), even after Ad.Trx1 therapy. Echocardiographic analysis after 4 weeks of myocardial infarction revealed significant improvement in myocardial functional parameters such as ejection fraction, fractional shortening, and E/A ratio in the Ad.Trx1-administered group compared with the diabetic myocardial infarction group.
Conclusions— This study demonstrates for the first time that impairment of angiogenesis and myocardial dysfunction can be regulated by Ad.Trx1 gene therapy in streptozotocin-induced diabetic rats subjected to infarction.
- myocardial infarction
- diabetes mellitus
- oxidative stress
- gene therapy
Received April 22, 2009; accepted December 30, 2009.
Diabetic individuals often experience complications such as coronary artery disease, peripheral artery disease, and stroke that contribute to the morbidity and mortality of these subjects.1 This higher incidence of cardiovascular complications and the unfavorable prognosis among diabetic individuals who develop such complications has been correlated to diabetes-mediated impairment of angiogenesis.2 Inhibited angiogenesis is known to contribute to impaired coronary collateral vessel formation and wound healing.2 Various studies have linked diabetes-mediated impaired angiogenesis to improper degradation of the basement membrane, alterations to the delicate balance of angiogenic growth factors and cytokines that regulate vascular stability, problems in signal transduction, and, to a large extent, diabetes-mediated oxidative stress.2 Evidence suggests that reactive oxygen species (ROS)–mediated oxidative stress is the primary contributor to vascular and myocardial dysfunction, compromising recovery from such conditions.3,4
Clinical Perspective on p 1255
Maintenance of a balanced redox status in the microenvironment might aid in alleviating many diabetes-mediated complications, especially those related to impaired angiogenesis. Balanced and responsive antioxidant systems are vital for proper regulation of the redox status of the cell. Cells are normally able to defend themselves against oxidative stress–induced damage by use of several antioxidant systems.5 The thioredoxin system is one of the main ubiquitously expressed thiol-reducing antioxidant systems. The classic 12-kDa cytosolic thioredoxin-1 (Trx1) is the most studied of the different forms of thioredoxin.5,6 Trx1 has been described as a growth regulator, a transcriptional factor regulator, and a cofactor, apart from its very important antioxidative role.7
Key biological activities of Trx1 include antioxidative, antiapoptotic, and growth-stimulatory properties through its interactions with nuclear factor-κB, apoptosis signal regulation kinase-1, and hypoxia inducible factor-1α.8–10 Induction of apoptosis signal regulation kinase-1 is known to activate the proapoptotic JNK and p38 mitogen-activated protein kinase (MAPK) signaling pathways.11 In addition, it was reported that transgenic mouse hearts overexpressing Trx1 displayed significantly improved postischemic ventricular recovery, reduced myocardial infarct size, and resistance to ischemic reperfusion injury compared with hearts from wild-type mice.7,12 Inhibition of Trx1 expression in the heart has been associated with increased oxidative stress and apoptosis.13 Several reports suggest the contribution of the thioredoxin system in the upregulation of heme oxygenase-1 (HO-1) protein levels and HO-1 promoter activity under conditions associated with inflammation and increased oxidative stress.14,15 HO-1 is known to be a stress-response protein that has important redox regulatory functions.15,16 Some of our recent studies have demonstrated that the mechanism responsible for the enhanced angiogenesis and the cardioprotective effect of resveratrol and sildenafil operates through Trx1, HO-1, and vascular endothelial growth factor (VEGF).16–18 We also observed that resveratrol alleviates cardiac dysfunction in streptozotocin-induced diabetic rats through the induction of Trx1, HO-1, and VEGF.17 We recently reported the possible role of decreased VEGF in the pathogenesis of diabetes-mediated impairment of angiogenesis in the myocardium.19 The process of angiogenesis was shown to be highly impaired in diabetic mouse models of hindlimb ischemia and wound healing owing to decreased expression of VEGF.20,21 Moreover, Trx1 has been shown to promote the expression and activity of hypoxia inducible factor-1α, which leads to elevated expression of VEGF and thus enhanced vascular growth.10
There have been several attempts at the preclinical and clinical levels to induce myocardial angiogenesis by overexpressing angiogenic factors such as VEGF in the peri-infarct zone after myocardial infarction (MI).20,21 Recently, we showed that intramyocardial coadministration of a combination of adenoviral vectors encoding VEGF and angiopoietin-1 induced and stabilized the process of angiogenesis in ischemic diabetic myocardium and reduced ventricular remodeling, which led to cardioprotection.19
In view of the increased functional severity of cardiac failure in diabetic post-MI subjects, the given role of oxidative stress–mediated reduction in myocardial angiogenesis, and the antioxidative, growth regulatory, and angiogenic potential of Trx1 via induction of VEGF, we hypothesized that overexpression of Trx1 alone in the myocardium could prove to be useful for prophylaxis of subsequent heart failure after diabetes-associated MI. Therefore, to address the importance of Trx1 in angiogenesis in vivo during myocardial ischemia in a diabetic scenario and identify a potential therapeutic tool, we investigated the effect of Trx1 gene therapy with an adenoviral vector carrying the Trx1 gene in the MI model of streptozotocin-induced diabetic rats. In the present study, we investigated the effect of Trx1 gene delivery on myocardial fibrosis, oxidative stress, cardiomyocyte and endothelial cell apoptosis, capillary and arteriolar density, and left ventricular remodeling in diabetic rats. We demonstrate for the first time that the myocardium can be rescued from diabetes-related impairment of angiogenesis, severity of functional disorder, and subsequent heart failure by Trx1 gene therapy in streptozotocin-induced diabetic rats.
The present study was performed in accordance with the principles of laboratory animal care formulated by the National Society for Medical Research and in accordance with the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences and published by the National Institutes of Health (publication No. 85–23, revised 1985). The experimental protocol was approved by the Institutional Animal Care Committee of the University of Connecticut Health Center (Farmington, Conn). Male Sprague-Dawley rats (weight 300 to 325 g) were randomly separated into normal and diabetic rats as they received an intraperitoneal injection of vehicle (0.1 mol/L citrate buffer, pH 4.5) alone or streptozotocin at a dosage of 65 mg/kg body weight dissolved in 0.1 mol/L citrate buffer. Five days after streptozotocin injection, hyperglycemia was documented by measuring the glucose content of tail-vein blood with the FreeStyle Flash blood glucose–monitoring system (Abbott Diabetes Care, Alameda, Calif). Rats with blood glucose concentrations ≥300 mg/dL were used for the study.
MI was induced in the diabetic animals 30 days after the induction of diabetes, as described previously.16 Age-matched nondiabetic animals were used as comparable controls. Sham-operated rats underwent thoracotomy and pericardiectomy without MI. The rats were randomized into 8 groups: (1) Control sham (CS), (2) control plus MI (CMI), (3) CMI plus Ad.LacZ (CMI-AdLacZ), (4) CMI plus Ad.Trx1 (CMI-AdTrx1), (5) diabetic sham (DS), (6) diabetic plus MI (DMI), (7) DMI plus Ad.LacZ (DMI-AdLacZ), and (8) DMI + Ad.Trx1 (DMI-AdTrx1).
The adenoviral vector encoding Trx1 (Ad.Trx1, 1×109 plaque-forming units)22 was administered intramyocardially (in 100 μL of PBS, with a 30-g needle) at 4 different sites into the left anterior free wall adjacent to the point of ligation (border zone surrounding the infarct) immediately after an MI was induced. Adeno-LacZ (Ad.LacZ, 1×109 plaque-forming units)22 was used as the control. Ad.Trx1 and Ad.LacZ were generous gifts from Dr J. Sadoshima, New Jersey Medical School, Newark, NJ.13,23
In the present study, we used streptozotocin-induced diabetic rats, and MI was induced 30 days after the induction of hyperglycemia. These rats tend to lose weight, become weak, and show less physical activity and their vital sign indices deteriorate even before the first surgical intervention. There was a significant reduction in heart rate in these animals in the sham group and after MI.19 Overall, there was a higher rate of mortality in the diabetic animals after MI (45% to 52%) than in the nondiabetic MI controls (8% to 12%). Although we considered a second thoracotomy for gene delivery at a later time after the induction of MI, because of the higher rate of mortality and the stress and pain that the diabetic animals have to undergo, we decided to administer the adenoviral vectors immediately after MI.
Another group of nondiabetic and diabetic animals were treated with tin protoporphyrin IX (SnPP, an inhibitor of HO-1 activity; 50 μmol/kg IP) to determine whether HO-1 plays an important role in Trx1-mediated VEGF expression.16 These SnPP-treated animals were then subjected to MI and treated with Ad.Trx1 as described above.
For detailed descriptions of the materials and methods and details on the time point of harvesting the tissue for various experiments, please refer to the online-only Data Supplement.
All data were analyzed with the statistical software GraphPad Prism 5.0′ (San Diego, Calif). Statistical analysis was performed with 1-way ANOVA. Post hoc comparisons between groups were performed by Newman-Keuls multiple comparison test. Results are presented as mean±SEM, with P≤0.05 used to indicate statistical significance.
Transfection Efficiency of Ad.LacZ and Ad.Trx1 in Human Umbilical Vein Endothelial Cells
To examine the transfection efficiency of Ad.LacZ and Ad.Trx1, cytochemistry was performed after 48 hours to examine the extent of β-galactosidase (β-gal) and immunocytochemistry for Trx1 expression in the cells. Figure 1A and 1B represent expression of β-gal and Trx1, respectively, in Ad.LacZ- and Ad.Trx1-treated and untreated human umbilical vein endothelial cells (HUVECs). Significant β-gal staining was also observed in Ad.LacZ-treated HUVECs (Figure 1A). A significant increase in expression of Trx1 was observed in Ad.Trx1-treated cells compared with untreated HUVECs (Figure 1B).
Effect of Ad.Trx1 Gene Transfection on Tubulogenesis in HUVECs
To determine whether Ad.Trx1 gene transfection has angiogenic potential, we performed an in vitro Matrigel assay with HUVECs to analyze the extent of tube formation, as shown in Figure 1C. We observed a significant increase in tubulogenesis in the Ad.Trx1-treated HUVECs compared with the Ad.LacZ-treated control. This Ad.Trx1-mediated tubulogenesis was found to be significantly abolished when HUVECs were transfected with Ad-sh-Trx1 (adeno vector-sh-Trx1 encoding small hairpin RNA that inhibits Trx1 expression) along with Ad.Trx1 (Figure 1C), thereby documenting the angiogenic potential of Trx1.
Ad.Trx1 Transfection Increased the Expression of Trx1, HO-1, and VEGF in HUVECs
The present in vitro Matrigel assay experiments documented the angiogenic potential of Trx1. To examine the molecular basis of the angiogenic potential of Ad.Trx1 treatment in HUVECs, we examined the effect of Ad.Trx1 transfection on the expression levels of HO-1 and VEGF (n=3 per group). We observed a significant increase in the expression of Trx1 (8.7-fold; Figure 1D and 1E), HO-1 (9.2-fold; Figure 1D and 1F), and VEGF (1.8-fold; Figure 1D and 1G) in HUVECs treated with Ad.Trx1 compared with Ad.LacZ-treated cells. However, when the cells were transfected with Ad.Trx1 and treated with SnPP, we observed a significant reduction in VEGF expression (2-fold) even though there was no significant change in Trx-1 or HO-1 expression compared with the Ad.Trx1-treated group (Figure 1D through 1G).
Histochemistry for the Transfection Efficiency of Ad.LacZ and Ad.Trx1 (In Vivo)
After we verified the angiogenic potential of Trx1 and the molecular basis of this angiogenic capacity, we intended to study the effect of Ad.Trx1 gene therapy in the infarcted diabetic myocardium. To examine the efficiency of our gene transfer technique in vivo, we evaluated adenoviral gene expression 4 days after intramyocardial adenoviral gene administration in the sham-operated nondiabetic control groups. We injected the adenoviral vector encoding β-gal (Ad.LacZ) and Trx1 (Ad.Trx1) at 4 different sites (in 100 μL of PBS, using a 30-g needle) into the left anterior wall of the myocardium in nondiabetic sham animals (n=3 per group). We observed robust infection and expression of the respective proteins in the myocardium as assessed by X-gal staining (Figure 1H) and Trx1 immunohistochemical staining (Figure 1I) in the myocardium surrounding the sites of gene transfer. β-Gal expression was not evident in remote areas such as the septum and the right ventricular muscles.
Biological Effects of Trx1 Gene Therapy in Diabetic Ischemic Myocardium
Trx1 Gene Therapy Reduced Myocardial Fibrosis
Four days after MI and gene therapy, hearts (n=3 per group) were harvested, and Masson trichrome staining was performed on paraffin-embedded tissue sections to visualize the infarct scar (Figure 2A). There was no evident difference in infarct scar extension or thickness in Ad.LacZ-treated CMI-AdLacZ and DMI-AdLacZ compared with their respective nontreated controls (CMI and DMI, respectively). After Trx1 gene therapy, the nondiabetic CMI-AdTrx1 group showed a thicker, more muscular infarct, a smaller scar region, and more islands of viable cardiac tissue than the CMI group. Similarly, the diabetic MI group that received Trx1 gene therapy (DMI-AdTrx1) demonstrated an obviously smaller scar region and more islands of viable cardiac tissue than the DMI group.
Thirty days after MI and gene therapy, hearts were harvested, and Masson’s trichrome staining was performed on paraffin-embedded tissue sections to visualize the extent of collagen deposition and myocardial fibrosis (Figure 2B). Infarct scar extension and thinning of the infarct were evident 30 days after MI in the DMI groups compared with the CMI group. There was no marked difference in the extent of collagen deposition or myocardial fibrosis in Ad.LacZ-treated CMI-AdLacZ and DMI-AdLacZ rats compared with their respective nontreated controls (CMI and DMI, respectively). There was a less prominent scar extension, less collagen deposition, and less fibrosis in Ad.Trx1-treated nondiabetic (CMI-AdTrx1) and diabetic (DMI-AdTrx1) rats than in CMI and DMI rats, respectively.
Trx1 Gene Therapy Reduced Oxidative Stress in Myocardium
The amount of ROS-mediated oxidative stress was determined 4 days after MI and gene therapy by evaluation of superoxide anion (O2·−) formation by means of dihydroethidium staining24 (n=3 to 4 per group; Figure 2C and 2D). MI alone increased levels of ROS in the nondiabetic MI (CMI) and nondiabetic Ad.LacZ-treated MI (CMI-AdLacZ) groups compared with the nondiabetic sham (control sham) group. However, this increase in myocardial ROS was exacerbated in all of the diabetic MI groups compared with their respective nondiabetic controls. Myocardial ROS levels increased significantly in the diabetic sham group compared with the nondiabetic sham (control sham) group. The induction of MI significantly intensified the formation of ROS in the DMI and DMI-AdLacZ groups compared with the diabetic sham group. Although there was a reduction in MI-induced ROS levels in the CMI-AdTrx1 group compared with the CMI and CMI-AdLacZ groups, the change was not significant. However, Ad.Trx1 gene therapy significantly reduced ROS levels in the treated DMI-AdTrx1 myocardium compared with the DMI and DMI-AdLacZ myocardium, thereby reducing ROS-mediated oxidative stress (Figure 2C and 2D).
Trx1 Gene Therapy Significantly Reduced Cardiomyocyte and Endothelial Cell Apoptosis
To understand whether Trx1 gene therapy was paralleled by a reduction in cardiomyocyte and endothelial cell apoptosis, we performed a terminal dUTP nick end-labeling (TUNEL) assay followed by antibody staining with either anti-α-sarcomeric actin or anti-von Willebrand factor (Figure 3A through 3D) to measure cardiomyocyte and endothelial cell apoptosis, respectively (n=4 per group). We observed a significant increase in apoptotic cardiomyocytes (520±92.6 versus 306±16.3 cells/100 high-power fields [HPF]) and endothelial cells (650±90.5 versus 369±38.7 cells/100 HPF) in the DMI group compared with the CMI group. There was no significant difference in the number of apoptotic cardiomyocytes or endothelial cells between the Ad.LacZ-treated CMI-AdLacZ or DMI-AdLacZ group and their respective nontreated controls (CMI or DMI). However, Trx1 gene therapy significantly reduced cardiomyocyte (119±21.5 versus 306±16.3 cells/100 HPF) and endothelial cell (163±15.4 versus 369±38.7 cells/100 HPF) apoptosis in the CMI-AdTrx1 group compared with the CMI group. Similarly, Trx1 gene therapy in the DMI-AdTrx1 group significantly reduced cardiomyocyte (204±13.1 versus 520±92.6 cells/100 HPF) and endothelial cell (208±32.8 versus 650±90.5 cells/100 HPF) apoptosis compared with the DMI group (Figure 3A through 3D).
Trx1 Gene Therapy Increased Capillary and Arteriolar Density
We performed immunohistochemical analysis of CD31/PECAM-1 (platelet and endothelial cell adhesion molecule, an endothelial cell marker) for capillary density (counts/mm2), as well as analysis of α-smooth muscle actin for arteriolar density (counts/mm2; n=4 per group; Figure 3E through 3H). There was a significant reduction in capillary density (1773±126 versus 2415±105; Figure 3E and 3F) and arteriolar density (13.4±1.4 versus 23.7±1.3; Figure 3G and 3H) in the DMI group compared with the CMI group. There was no significant difference in capillary or arteriolar density between the Ad.LacZ-treated CMI-AdLacZ or DMI-AdLacZ group and their respective nontreated controls (CMI or DMI). However, Trx1 gene therapy significantly increased capillary (3433±121 versus 2415±49; Figure 3E and 3F) and arteriolar (38.4±1.7 versus 23.7±1.3; Figure 3G and 3H) density in the CMI-AdTrx1 group compared with the CMI group. Similarly, a significant increase in capillary density (2450±153 versus 1773±126; Figure 3E and 3F) and arteriolar density (30.6±1.5 versus 13.4±1.4; Figure 3G and 3H) was observed with Trx1 gene therapy in the DMI-AdTrx1 group compared with the DMI group.
Molecular Basis for the Angiogenic and Cardioprotective Effect of Trx1 Gene Therapy in Diabetic Ischemic Myocardium
Trx1 Gene Therapy Significantly Increased Expression of Trx1, HO-1, and VEGF
Immunohistochemical analysis was performed for protein expression of Trx1, HO-1, and VEGF 4 days after MI and gene therapy (n=3 to 4 per group). We observed an evident reduction in the expression of Trx1 (Figure 4A), HO-1(Figure 4B), and VEGF (Figure 4C) in diabetic sham-operated and MI-induced (DMI and DMI-AdLacZ) groups compared with their respective nondiabetic sham-operated (control sham) and MI-induced (CMI and CMI-AdLacZ) controls. However, there was an evident increase in the expression of Trx1, HO-1, and VEGF in the Ad.Trx1-administered CMI-AdTrx1 group compared with the CMI or CMI-AdLacZ groups. Similarly, Trx1 gene therapy resulted in increased expression of these proteins in the DMI-AdTrx1 group compared with the DMI and DMI-AdLacZ groups (Figure 4A through 4C).
Western blot analysis 4 days after gene therapy (n=3 to 4 per group) also revealed that Trx1 gene therapy in the CMI-AdTrx1 group significantly increased the expression of Trx1 (1.8-fold; Figure 5A and 5B), HO-1 (2.2-fold; Figure 5A and 5C), and VEGF (1.9-fold; Figure 5A and 5D) compared with the CMI group. Ad.LacZ treatment in the CMI-AdLacZ or DMI-AdLacZ group did not reveal any significant difference in the expression of these proteins compared with their respective nontreated controls (CMI and DMI). The expression of Trx1 (2.4-fold), HO-1 (2.5-fold), and VEGF (2.3-fold) was found to be decreased significantly in the DMI group compared with the CMI group. However, Trx1 gene therapy significantly increased the expression of Trx1 (2.8-fold), HO-1 (3.2-fold), and VEGF (2.2-fold) in the DMI-AdTrx1 group compared with the DMI group (Figure 5A through 5D).
SnPP Treatment Significantly Reduced Trx1-Mediated VEGF Expression In Vivo
To examine whether HO-1 mediated the Trx1-induced VEGF expression in Ad.Trx1-treated animals, we intramyocardially administered Ad.Trx1 after MI to SnPP-treated nondiabetic and diabetic animals (n=3 per group). There was no significant difference in the expression of Trx1 and HO-1 between the SnPP-treated and nontreated animals (in both the nondiabetic and diabetic groups) that received Ad.Trx1 gene therapy. However, we observed significant downregulation of VEGF (2.7- and 2.2-fold, respectively) despite the increase in the expression of Trx1 and HO-1 in SnPP-treated nondiabetic (CMI-AdTrx1 plus SnPP) and diabetic (DMI-AdTrx1 plus SnPP) animals even after Ad.Trx1 injections compared with nondiabetic (CMI-AdTrx1) and diabetic (DMI-AdTrx1) animals that were treated with Ad.Trx1 but not SnPP (Figure 5A and 5D).
Trx1 Gene Therapy Increased Expression of Antiapoptotic p38MAPKβ While Reducing Expression of Proapoptotic Phosphorylated JNK and p38MAPKα
In the present study, we evaluated the expression of phosphorylated JNK (p-JNK) and levels of p38MAPK (α and β) proteins (n=3 to 4 per group; Figure 6). Trx1 gene therapy in the CMI-AdTrx1 group significantly increased the expression of p38MAPKβ (1.9-fold; Figure 6A and 6C) and decreased the expression of p38MAPKα (2.3-fold; Figure 6A and 6B) and p-JNK (2.3-fold; Figure 6A and 6D) compared with the CMI group. Ad.LacZ treatment in the CMI-AdLacZ and DMI-AdLacZ groups did not yield any significant difference in the expression of these proteins compared with their respective nontreated controls (CMI and DMI). A significant increase in activation of proapoptotic p-JNK (1.5-fold) and p38MAPKα (1.5-fold) expression and downregulation of antiapoptotic p38MAPKβ (2.5-fold) was observed in the DMI group compared with the CMI group. However, Trx1 gene therapy increased the expression of p38MAPKβ (3.1-fold) and decreased the expression of p-JNK (6.4-fold) and p38MAPKα (2.1-fold) in the DMI-AdTrx1 compared with DMI group (Figure 6A through 6D).
Preservation of Myocardial Function After Acute MI in Diabetic Myocardium Through Trx1 Gene Therapy
Echocardiographic assessment of myocardial function was performed 30 days after surgical intervention, MI, and gene therapy in the experimental animals (n=4 to 5 per group). The representative pictures (Figure 7A through 7F) show the parasternal short-axis view in M-mode. Quantitative representations of echocardiographic measurements are shown in Figure 7B through 7F. There was significant functional disorder in all of the diabetic groups compared with their respective nondiabetic controls. Trx1 gene therapy significantly improved myocardial function in both the nondiabetic and diabetic groups compared with their respective nontreated or Ad.LacZ-treated controls.
Improved contraction along with improved end-systolic and end-diastolic dimensions were observed with Trx1 treatment, which preserves the contractile function of the heart and suggests attenuated remodeling in the treated animals. Increased left ventricular internal diameter in systole (8.7±0.1 versus 7.1±0.1 mm) and left ventricular internal diameter in diastole (10.9±0.1 versus 9.3±0.1 mm) were observed in DMI rats compared with CMI rats (Figure 7B and 7C). Trx1 gene therapy reduced the left ventricular internal diameter in systole (6.6±0.2 versus 8.6±0.1 mm) and diastole (8.6±0.2 versus 10.9±0.1 mm) in the DMI-AdTrx1 group compared with the DMI group, although a significant difference was not observed in the DMI-AdLacZ group, which shows the functional recovery of the ischemic myocardium after Trx1 gene delivery (Figure 7B and 7C). Improved systolic parameters were observed in Trx1-treated rats compared with DMI controls, as assessed by ejection fraction and fractional shortening. Ejection fraction and fractional shortening were improved in DMI-AdTrx1 rats (ejection fraction 45.3±0.9% versus 38.5±0.8%, fractional shortening 24.3±0.4% versus 20±1.4%) compared with DMI controls (Figure 7E and 7F). There was no significant difference in ejection fraction or fractional shortening in the DMI-AdLacZ group compared with the DMI group.
There was also significant diastolic dysfunction as evidenced by a higher E/A ratio in the diabetic groups compared with the respective nondiabetic controls. After 30 days of MI, the DMI-AdTrx1 group showed a significant improvement in diastolic function, as evidenced by the E/A ratio (3.1±0.1 versus 4.3±0.1) compared with the DMI group (Figure 7D).
The present study is the first to document that induction of the expression of Trx1 can rescue diabetic myocardium from diabetes/oxidative stress–related impairment of myocardial angiogenesis by reducing oxidative stress and enhancing the expression of HO-1 and VEGF. In addition, we observed decreased expression of proapoptotic proteins such as p-JNK and p38MAPKα and increased expression of the antiapoptotic protein p38MAPKβ in the DMI-AdTrx1 group compared with the DMI group. Associated with these molecular changes, we also observed reduced myocyte and endothelial cell apoptosis, reduced fibrosis, and increased capillary and arteriolar density, with significant improvement of myocardial function as assessed by echocardiography. The present preclinical results indicate that suppression of oxidative stress by Trx1 induction has beneficial effects in the treatment of diabetes-related cardiac failure.
Preclinical and clinical observations have linked diabetes-mediated oxidative stress to increased myocyte and endothelial cell apoptosis and reduction of antioxidative and angiogenic factors.4,25 In the present study, we observed that Trx1 gene therapy in ischemic diabetic myocardium was able to significantly reduce oxidative stress, as evidenced by the reduction in ROS generation. This reduction in ROS generation can be correlated to the reduction in myocyte and endothelial cell apoptosis and the increase in the expression of angiogenic VEGF.
The proangiogenic effect of resveratrol through the activation of Trx1, HO-1, and VEGF was blocked by SnPP, a specific inhibitor of HO-1 activity.16 HO-1 is a stress-inducible enzyme system that catalyzes the breakdown of prooxidant heme into biliverdin, carbon monoxide, and iron and is well known for its antiapoptotic and antiinflammatory activities.26,27 The expression and activity of HO-1 have been shown to be impaired in a diabetic milieu.25 It was reported that HO-1 can inhibit postmyocardial infarction remodeling and restore ventricular function.28 Evidence suggests that HO-1 promotes neovascularization in the ischemic heart via the induction of VEGF.29 Rivard and colleagues20 reported that the impairment in new blood vessel formation in a hindlimb ischemia model of diabetic mice was a result of reduced expression of VEGF and that VEGF gene therapy could induce neovascularization in this diabetic mice model. It was also shown that adenovirus-mediated gene transfer of VEGF could promote wound healing via the induction of angiogenesis in CD1 diabetic mice.21
With Trx1 gene therapy, we observed a significant increase in the expression of Trx1, HO-1, and VEGF in the DMI-AdTrx1 group compared with the DMI group, which suggests that Trx1 gene therapy is effective in inducing the expression of these proteins, which were downregulated in a diabetic condition. However, in the SnPP-treated, Ad.Trx1-injected nondiabetic and diabetic MI animals, there was significant downregulation of VEGF even though there was no significant difference in the expression of Trx1 and HO-1 compared with the CMI-AdTrx1 and DMI-AdTrx1 groups, respectively, which suggests that the Trx1-induced expression of VEGF is mediated by HO-1.
The fact that Trx1 gene therapy significantly reduced the number of apoptotic cardiomyocytes and endothelial cells in the DMI-AdTrx1 group compared with the DMI group suggests that Trx1 gene therapy significantly aids in the rescue of diabetic myocardium from MI-induced degeneration and also might support the neovascularization process. The decrease in the number of apoptotic cardiomyocytes and endothelial cells in the treatment group can be correlated with the increase in expression of Trx1 and HO-1, both of which have been shown to be antioxidative, antiapoptotic, and growth stimulatory.9,10,28
We also observed decreased expression of p-JNK and p38MAPKα (proapoptotic signal) and increased p38MAPKβ (antiapoptotic signal) in the DMI-AdTrx1 rats that also might have resulted in decreased cardiomyocyte and endothelial cell apoptosis. Several reports have shown that ischemia and doxorubicin-induced cardiomyocyte apoptosis are attenuated by pharmacological p38 inhibition.30,31 Four splice variants of the p38 family (p38MAPKα, p38MAPKβ, p38MAPKγ, and p38MAPKδ) were identified.32 Differential activation of p38MAPKα and p38MAPKβ has been reported.33 Although p38MAPKα has been shown to contribute to the proapoptotic pathway, p38MAPKα activation through ischemic preconditioning has been shown to decrease infarct size, thus demonstrating activation of the antiapoptotic pathway.33 Moreover, p38MAPKα dominant-negative mice were reported to be less susceptible to ischemia-reperfusion injury and subsequent apoptosis.34 In the present study, we observed increased activation of p38MAPKβ after Ad.Trx1 treatment that might have resulted in decreased apoptosis and a subsequent reduction in myocardial fibrosis.
We also documented that capillary and arteriolar density after 30 days of MI was increased significantly with Trx1 gene therapy in the DMI-AdTrx1 group compared with the DMI group. This can be correlated to the increase in expression of Trx1, HO-1, and VEGF in the treatment group. The antiapoptotic and growth-stimulatory Trx1 and HO-1 might have reduced endothelial cell apoptosis, whereas the increase in VEGF should have stimulated the neovascularization process in the treatment group.
The observed reduction in ROS and apoptosis and the improvement in the vascular density in the treatment group might have led to a reduction in infarct scar extension and fibrosis in the Ad.Trx1 treatment groups. This ultimately could explain the fact that there was significant improvement in left ventricular myocardial function (assessed by echocardiogram) in areas such as left ventricular internal diameter in systole or diastole, E/A ratio, ejection fraction, and fractional shortening in the DMI-AdTrx1 group compared with the DMI group. Reports have linked type 1 diabetes mellitus to early diastolic dysfunction, progressive systolic dysfunction, and subsequent heart failure.35 The E/A ratio, an index of diastolic dysfunction, has been shown to be reduced significantly in patients with type I diabetes mellitus.36 However, in diabetes-mediated advanced heart failure, the E/A ratio tends to increase.36 In the present study, the severity of heart failure could be assessed by the significantly higher E/A ratio in all diabetic groups compared with the respective nondiabetic controls. However, Trx1 gene therapy significantly reduced the magnitude of diastolic dysfunction in the diabetic rats. The magnitude and duration of hyperglycemia in a hypoinsulinemic setting are known to determine the extent of progressive systolic dysfunction in subjects with type 1 diabetes mellitus.35 There was significant systolic functional disorder in the diabetic myocardium after MI; however, Trx1 gene therapy preserved systolic functional parameters significantly in treated diabetic MI rats compared with nontreated diabetic MI rats.
In conclusion, the present preclinical findings indicate that a precisely balanced antioxidant system is essential for the induction of a functional neovasculature in diabetic ischemic myocardium. The present data are supportive of the concept that in a diabetic milieu, impaired angiogenesis is not only caused by downregulation of angiogenic growth factors but also is dependent on excessive oxidative stress and the lack of responsive antioxidant systems. Therefore, for the first time, we propose to treat diabetes-mediated impairment of angiogenesis by normalizing and stabilizing the redox microenvironment in the myocardium by means of Ad.Trx1 gene therapy, thereby reducing oxidative stress and cell death and inducing neovessel formation and maturation, subsequently reducing ventricular remodeling after an MI. Our novel and unique approach of Trx1 gene delivery could prove to be a strategic therapeutic modality in the treatment of diabetes-related cardiac failure and may ultimately improve quality of life and mitigate disease progression in affected subjects.
Sources of Funding
This study was supported by National Institutes of Health grants HL-56803 and HL-69910 to Dr Maulik.
Taniguchi Y, Taniguchi-Ueda Y, Mori K, Yodoi J. A novel promoter sequence is involved in the oxidative stress-induced expression of the adult T-cell leukemia-derived factor (ADF)/human thioredoxin (Trx) gene. Nucleic Acids Res. 1996; 24: 2746–2752.
Matthews JR, Wakasugi N, Virelizier JL, Yodoi J, Hay RT. Thioredoxin regulates the DNA binding activity of NF-kappa B by reduction of a disulphide bond involving cysteine 62. Nucleic Acids Res. 1992; 20: 3821–3830.
Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y, Kawabata M, Miyazono K, Ichijo H. Mammalian thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase (ASK) 1. EMBO J. 1998; 17: 2596–2606.
Welsh SJ, Bellamy WT, Briehl MM, Powis G. The redox protein thioredoxin-1 (Trx-1) increases hypoxia-inducible factor 1alpha protein expression: Trx-1 overexpression results in increased vascular endothelial growth factor production and enhanced tumor angiogenesis. Cancer Res. 2002; 62: 5089–5095.
Wiesel P, Foster LC, Pellacani A, Layne MD, Hsieh CM, Huggins GS, Strauss P, Yet SF, Perrella MA. Thioredoxin facilitates the induction of heme oxygenase-1 in response to inflammatory mediators. J Biol Chem. 2000; 275: 24840–24846.
Koneru S, Varma Penumathsa S, Thirunavukkarasu M, Vidavalur R, Zhan L, Singal PK, Engelman RM, Das DK, Maulik N. Sildenafil-mediated neovascularization and protection against myocardial ischaemia reperfusion injury in rats: role of VEGF/angiopoietin-1. J Cell Mol Med. 2008; 12: 2651–2664.
Samuel SM, Akita Y, Paul D, Thirunavukkarasu M, Zhan L, Sudhakaran PR, Li C, Maulik N. Co-administration of adenoviral VEGF and Ang-1 enhances vascularization and reduces ventricular remodeling in the infarcted myocardium of type I diabetic rats. Diabetes. 2010; 59: 51–60.
Koneru S, Penumathsa SV, Thirunavukkarasu M, Zhan L, Maulik N. Thioredoxin-1 gene delivery induces heme oxygenase-1 mediated myocardial preservation after chronic infarction in hypertensive rats. Am J Hypertens. 2009; 22: 183–190.
Singh VP, Le B, Khode R, Baker KM, Kumar R. Intracellular angiotensin II production in diabetic rats is correlated with cardiomyocyte apoptosis, oxidative stress, and cardiac fibrosis. Diabetes. 2008; 57: 3297–3306.
Di Filippo C, Marfella R, Cuzzocrea S, Piegari E, Petronella P, Giugliano D, Rossi F, D'Amico M. Hyperglycemia in streptozotocin-induced diabetic rat increases infarct size associated with low levels of myocardial HO-1 during ischemia/reperfusion. Diabetes. 2005; 54: 803–810.
Wagener FA, van Beurden HE, von den Hoff JW, Adema GJ, Figdor CG. The heme-heme oxygenase system: a molecular switch in wound healing. Blood. 2003; 102: 521–528.
Liu X, Pachori AS, Ward CA, Davis JP, Gnecchi M, Kong D, Zhang L, Murduck J, Yet SF, Perrella MA, Pratt RE, Dzau VJ, Melo LG. Heme oxygenase-1 (HO-1) inhibits postmyocardial infarct remodeling and restores ventricular function. FASEB J. 2006; 20: 207–216.
Kang YJ, Zhou Z-X, Wang G-W, Buridi A, Klein JB. Suppression by metallothionein of doxorubicin-induced cardiomyocyte apoptosis through inhibition of p38 mitogen-activated protein kinases. J Biol Chem. 2000; 275: 13690–13698.
Ma XL, Kumar S, Gao F, Louden CS, Lopez BL, Christopher TA, Wang C, Lee JC, Feuerstein GZ, Yue T-L. Inhibition of p38 mitogen-activated protein kinase decreases cardiomyocyte apoptosis and improves cardiac function after myocardial ischemia and reperfusion. Circulation. 1999; 99: 1685–1691.
Poornima IG, Parikh P, Shannon RP. Diabetic cardiomyopathy: the search for a unifying hypothesis. Circ Res. 2006; 98: 596–605.
Young LH. Diastolic function and type 1 diabetes. Diabetes Care. 2004; 27: 2081–2083.
The higher incidence of post–myocardial infarction angina, larger infarcts, severity of myocardial functional disorder, and unfavorable prognosis among diabetic individuals who develop such complications has been correlated with oxidative stress–mediated impairment of angiogenesis. Moreover, diabetes reflects a challenging condition in which the usual revascularization techniques such as coronary artery bypass surgery and percutaneous transluminal coronary angioplasty tend to fail, thereby leaving many diabetic patients with ischemic heart disease with no options for treatment. Maintenance of a balanced redox status in the microenvironment might aid in alleviating many diabetes-mediated complications, particularly those related to impaired angiogenesis. Our preclinical findings are supportive of the concept that in a diabetic milieu, impaired angiogenesis is not only caused by the downregulation of angiogenic growth factors but is also dependent on excessive oxidative stress and the lack of responsive antioxidant systems. For the first time, our unique results have identified the in vivo angiogenic potential of thioredoxin-1 and substantiated its antioxidative and antiapoptotic properties in the setting of myocardial infarction in a diabetic milieu. We therefore propose to treat diabetes-mediated impairment of angiogenesis by normalizing and stabilizing the redox microenvironment in the myocardium by means of thioredoxin-1 gene therapy, thereby reducing oxidative stress and cell death and inducing neovessel formation and maturation, subsequently reducing ventricular remodeling after a myocardial infarction. Our novel approach of Trx1 gene delivery may prove to be a strategic therapeutic modality in the treatment of diabetes-related cardiac failure and may ultimately improve quality of life and mitigate disease progression in affected individuals.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.109.872481/DC1.
Guest Editor for this article was Robert A. Kloner, MD, PhD.