Resveratrol supplementation abrogates pro-arteriogenic effects of intramyocardial vascular endothelial growth factor in a hypercholesterolemic swine model of chronic ischemia

Abstract

Background: Clinical trials of therapeutic angiogenesis with vascular endothelial growth factor (VEGF) have been disappointing, owing likely to endothelial dysfunction. We used a swine model of chronic ischemia and endothelial dysfunction to determine whether resveratrol coadministration would improve the angiogenic response to VEGF therapy.

Methods: Yorkshire swine fed a high-cholesterol diet underwent left circumflex ameroid constrictor placement, and were given either no drug (high cholesterol control [HCC], n = 8), perivascular VEGF (2 μg sustained release [high cholesterol VEGF-treated; HCV], n = 8), or VEGF plus oral resveratrol (10 mg/kg, [high cholesterol VEGF- and resveratrol-treated; HCVR], n = 8). After 7 weeks, myocardial contractility, perfusion, and microvessel reactivity in the ischemic territory were assessed. Tissue was analyzed for vessel density, oxidative stress, and protein expression.

Results: Myocardial perfusion was significantly improved in the HCV group compared with the HCC group; resveratrol coadministration abrogated this improvement. There were no differences in regional myocardial contractility between groups. Endothelium-dependent microvessel relaxation was improved in the HCVR group, and endothelium-independent relaxation response was similar between groups. Arteriolar density was greatest in the HCV group, whereas capillary density was similar between groups. Expression of Akt and phospho-endothelial nitric oxide synthase were increased in the HCVR group. Total protein oxidative stress and myeloperoxidase expression were reduced in the HCVR group, but so was the oxidative-stress dependent phosphorylation of vascular endothelial cadherin (VE-cadherin) and β-catenin.

Conclusion: Although resveratrol coadministration decreases oxidative stress and improves endothelial function, it abolishes improvements in myocardial perfusion and arteriolar density afforded by VEGF treatment alone. This effect is due likely to inhibition of the oxidative stress-dependent phosphorylation of VE-cadherin, an essential step in the initiation of arteriogenesis.

Fig 1

Measurements of global and regional function. Shown are heart rate (HR), mean arterial pressure (MAP), developed left ventricular pressure (DLVP), derivative of LVP over time (dP/dt), and vertical segmental shortening (VSS) and horizontal segmental shortening (HSS) in the AAR. Measurements of global LV function were decreased in the HCVR group (A–D), whereas there was no difference in regional function in the AAR between groups (E, F). *P < .05; †P < .01; ‡P < .001. Data are presented as mean values ± SEM.

Fig 2

Myocardial perfusion. Myocardial perfusion was significantly higher in the HCV group compared with both other groups at rest (A) and under ventricular pacing to 150 beats/min (B). *P < .05; †P < .01.

Fig 3

Microvessel function: Dose–response curves to endothelium-dependent vasodilator ADP (A) and endothelium-independent vasodilator sodium nitroprusside (B). Vessel relaxation response to each drug was measured and expressed as a percentage of the preconstricted vessel diameter. The relaxation response to ADP was significantly better in the HCVR group compared with both other groups. *P < .05; †P < .01; ‡P < .001.

Fig 4

Arteriolar density. Sections from the AAR were stained for endothelium-specific CD31 (green), smooth muscle actin (red), and nucleus-specific DAPI (blue). Shown are representative sections from (A) HCC, (B) HCV, and (C) HCVR animals. Arterioles were defined as structures co-staining for both CD-31 and smooth muscle actin (white arrows). Arteriolar density was significantly greater in the HCV group compared with both other groups. *P < .05; †P < .01.

Fig 5

Protein oxidative stress. Immunoblotting was used to quantify total protein oxidative stress in the AAR. Representative lanes are shown. Densitometry was used to measure signal intensity of the entire lane, which is expressed as arbitrary units of optical density (OD). The HCVR group demonstrated significantly lower oxidative stress than both other groups. †P < .01.

Fig 6

Protein expression. Shown are representative bands and quantification of protein expression in the AAR expressed in arbitrary units of optical density (OD). Expression of myeloperoxidase was significantly lower in the HCVR group (A), whereas pro-angiogenic and vasodilatory proteins Akt and phospho-eNOS were more highly expressed in the HCVR group (B, C). Expression of phospho-VE cadherin and β-catenin was significantly lower in the HCVR group (D, E).