Effects of a novel hydrogen sulfide prodrug in a porcine model of acute limb ischemia

Abstract

Objective: Previous studies have shown that hydrogen sulfide (H₂S) exerts potent proangiogenic properties under in vitro conditions and in rodent models. We sought to determine whether a novel H₂S prodrug promotes peripheral revascularization in a swine model of acute limb ischemia (ALI).

Methods: ALI was induced in 17 female miniswine via intravascular occlusion of the external iliac. At day 7 after ALI induction, miniswine (n = 17) were randomized to received placebo or the H₂S prodrug, SG-1002 (800 mg per os twice a day), for 35 days. At day 35 SG-1002 increased circulating levels of H₂S (5.0 ± 1.2 μmol/L vs 1.8 ± 0.50 μmol/L; P < .05), sulfane sulfur (10.6 ± 2.3 μmol/L vs 2.6 ± 0.8 μmol/L; P < .05), and nitrite (0.5 ± 0.05 μmol/L vs 0.3 ± 0.03 μmol/L; P < .005) compared with placebo. SG-1002 therapy increased angiographic scoring in ischemic limb vessel number (27.6 ± 1.6 vs 22.2 ± 1.8; P < .05) compared with placebo. Treatment with SG-1002 preserved existing capillaries in ischemic limbs (128.3 ± 18.7 capillaries/mm² vs 79.0 ± 9.8 capillaries/mm²; P < .05) compared with placebo. Interestingly, treatment with SG-1002 also improved coronary vasorelaxation responses to bradykinin and substance P in miniswine with ALI.

Conclusions: Our results suggest that daily administration of the H₂S prodrug, SG-1002, leads to an increase in circulating H₂S and nitric oxide signaling and preserves vessel number and density in ischemic limbs. Furthermore, SG-1002 therapy improved endothelial-dependent coronary artery vasorelaxation in the setting of ALI. Our data demonstrate that SG-1002 preserves the vascular architecture in ischemic limbs and exerts vascular protective effects in the coronary vasculature in a model of peripheral vascular disease.

Keywords: Acute limb ischemia; Hydrogen sulfide; Nitric oxide; Peripheral artery disease; Swine.

Figure 1.. Large animal model of acute limb ischemia.

Pigs underwent intravascular induction of acute limb ischemia (ALI) and placebo or SG-1002 (1600 mg PO) administration starting 7 days post-ALI and continued for 35 days. Weekly ABI measurements were performed and blood samples obtained. At day 42 follow-up arteriograms were acquired, animals euthanized and nonischemic and ischemic limb tissues collected (Figure 1A). Under fluoroscopic guidance, contrast arteriography was performed to evaluate baseline vessel architecture (Figure 1B). ALI was generated by deployment of an AMPLATZER^™ Vascular Plug II inside a GORE^® VIABAHN^® Endoprosthesis into the external iliac artery followed by an arteriogram to verify occlusion (Figure 1C).

Figure 2.. Ankle-brachial index unchanged in ischemic limbs.

Ankle-brachial index (ABI) for nonischemic (Figure 2A) and ischemic hindlimbs (Figure 2B) was calculated from systolic pressures measured weekly in each of the forelimbs and hindlimbs. ABI improved but continued to demonstrate a persistent ischemic state with values below 0.35 at day 42 in both groups. No difference in ABI between the placebo and SG-1002 groups was observed at any time point. Values are means ± SEM.

Figure 3.. SG-1002 increases circulating levels of H₂S and sulfane sulfur.

Baseline plasma was obtained before ALI induction, day 7 (prior to SG-1002 administration) and post-ALI days 14, 21, 28 and 35. Weekly circulating levels of H₂S (Figure 3A) and sulfane sulfur (Figure 3C) were similar between the placebo and SG-1002 groups on days 0, 7, 14, 21, and 28. At day 35, levels of H₂S (Figure 3B) and sulfane sulfur (Figure 3D) were higher in the SG-1002 group compared to placebo. Values are means ± SEM with number of animals per group shown. NS, not significant. *P < 0.05 vs. baseline.

Figure 4.. SG-1002 increases circulating levels of nitrite and reduces oxidative stress.

Baseline plasma was obtained before ALI induction, day 7 (prior to SG-1002 administration) and post-CLI days 14, 21, 28 and 35. Weekly circulating levels of nitrite were increased in the SG-1002 group at days 21 and 35 compared to placebo (Figure 4A). Circulating nitrite levels were similar between groups on days 0, 7, 14, and 28. At day 35, levels of nitrite were higher in the SG-1002 group compared to placebo (Figure 4B). Plasma levels of the oxidative stress marker, 8-isoprostane, were similar between the placebo and SG-1002 groups on days 0, 7, 14, 21, and 28 (Figure 4C). At day 35, levels of 8-isoprostane were lower in the SG-1002 group compared to placebo (Figure 4D). Values are means ± SEM with number of animals per group shown. NS, not significant.

Figure 5.. SG-1002 therapy increases collateral vessel number in ischemic hindlimbs.

Contrast angiographic images were acquired at day 42 post-ALI in placebo (Figure 5A) and SG-1002 (Figure 5B) treated animals. Quantitative angiographic scoring of contrast-opacified vessels was higher in ischemic limbs of SG-1002 treated pigs compared to placebo ischemic limbs (Figure 5C). There was no difference in collateral vessel number in the nonischemic limbs between the placebo and SG-1002 groups. Values are means ± SEM with number of animals per group shown. NS, not significant.

Figure 6.. SG-1002 therapy preserves capillary density in ischemic hindlimbs.

Representative micrographs shown of CD31 immunostained muscle tissue obtained from placebo nonischemic (Figure 6A), ischemic (Figure 6B) and SG-1002 nonischemic (Figure 6C), ischemic (Figure 6D) hindlimbs at day 42 post-ALI. Arrows denote CD31 -positive staining of capillaries and capillary density quantified (Figure 6E). Values are means ± SEM with number of animals per group shown. NS, not significant.

Figure 7.. SG-1002 improves coronary vascular reactivity by an NO-dependent mechanism.

Left anterior descending (LAD) and left circumflex (LCX) coronary arteries were isolated from placebo and SG-1002 treated groups for isometric tension experiments. Coronary arteries were precontracted with PGF_2a and relaxation responses to endothelial-dependent bradykinin (Figure 7A and 7B) and substance P (Figure 7D and 7E). Endothelial-independent relaxation was determined by sodium nitroprusside (Figure 7G and 7H). Half maximal effective concentrations (EC₅₀) were calculated (Figure 7C, 7F and 7I). Values are means ± SEM with number of animals per group shown.