Soluble epoxide hydrolase inhibition improves coronary endothelial function and prevents the development of cardiac alterations in obese insulin-resistant mice

Abstract

This study addressed the hypothesis that inhibiting the soluble epoxide hydrolase (sEH)-mediated degradation of epoxy-fatty acids, notably epoxyeicosatrienoic acids, has an additional impact against cardiovascular damage in insulin resistance, beyond its previously demonstrated beneficial effect on glucose homeostasis. The cardiovascular and metabolic effects of the sEH inhibitor trans-4-[4-(3-adamantan-1-yl-ureido)-cyclohexyloxy]-benzoic acid (t-AUCB; 10 mg/l in drinking water) were compared with those of the sulfonylurea glibenclamide (80 mg/l), both administered for 8 wk in FVB mice subjected to a high-fat diet (HFD; 60% fat) for 16 wk. Mice on control chow diet (10% fat) and nontreated HFD mice served as controls. Glibenclamide and t-AUCB similarly prevented the increased fasting glycemia in HFD mice, but only t-AUCB improved glucose tolerance and decreased gluconeogenesis, without modifying weight gain. Moreover, t-AUCB reduced adipose tissue inflammation, plasma free fatty acids, and LDL cholesterol and prevented hepatic steatosis. Furthermore, only the sEH inhibitor improved endothelium-dependent relaxations to acetylcholine, assessed by myography in isolated coronary arteries. This improvement was related to a restoration of epoxyeicosatrienoic acid and nitric oxide pathways, as shown by the increased inhibitory effects of the nitric oxide synthase and cytochrome P-450 epoxygenase inhibitors l-NA and MSPPOH on these relaxations. Moreover, t-AUCB decreased cardiac hypertrophy, fibrosis, and inflammation and improved diastolic function, as demonstrated by the increased E/A ratio (echocardiography) and decreased slope of the end-diastolic pressure-volume relation (invasive hemodynamics). These results demonstrate that sEH inhibition improves coronary endothelial function and prevents cardiac remodeling and diastolic dysfunction in obese insulin-resistant mice.

Keywords: cardiac function; endothelium; insulin resistance; soluble epoxide hydrolase.

Fig. 1.

Average food intake (A), caloric intake (B), and weight gain (C) in control mice (n = 30), nontreated high-fat diet (HFD) mice (n = 28), and HFD mice treated with glibenclamide (GLI; n = 30) or with t-AUCB (n = 36) from week (W)0 to W16. *P < 0.05 vs. control.

Fig. 2.

Evolution of glycemia during glucose (A), pyruvate (B), and insulin (C) tolerance tests expressed as absolute values or as area under the curve (AUC) at W16 in control mice (n = 4–8), nontreated HFD mice (n = 4–10), and HFD mice treated with GLI (n = 5–12) or with t-AUCB (n = 5–11). *P < 0.05 vs. control; †P < 0.05 vs. nontreated HFD; ‡P < 0.05 vs. HFD + GLI.

Fig. 3.

Plasma levels of free fatty acids (A), LDL cholesterol (B), triglycerides (C), and histological scores of lipid content in the liver (D) at W16 in control mice (n = 4–9), nontreated HFD mice (n = 4–8), and HFD mice treated with GLI (n = 5–11) or with t-AUCB (n = 5–10). *P < 0.05 vs. control; †P < 0.05 vs. nontreated HFD; ‡P < 0.05 vs. HFD + GLI.

Fig. 4.

Coronary endothelium-dependent relaxations to acetylcholine (ACh) after vessel contraction with 10⁻⁵ M serotonin (HT) under basal conditions (A: mean values and representative recordings), inhibitory effect of N^ω-nitro-l-arginine (l-NA), N-methylsulfonyl-6-(2-propargyloxyphenyl)-hexanamide (MSPPOH), l-NA + MSPPOH, and apamin + TRAM34 on the relaxations to 3 × 10⁻⁵ M acetylcholine (B), endothelium-independent relaxations to sodium nitroprusside (C), and relaxations to NS309 (D) and NS1619 (E), at W16 in control mice, nontreated HFD mice, HFD mice treated with GLI or with t-AUCB (n = 4 to 5 per condition). *P < 0.05, nontreated HFD vs. control; †P < 0.05, HFD + t-AUCB vs. nontreated HFD; ‡P < 0.05, HFD + t-AUCB vs. HFD + GLI; §P < 0.05, l-NA vs. control conditions; δP < 0.05, MSPPOH vs. control conditions; ||P < 0.05, l-NA + MSPPOH vs. l-NA.

Fig. 5.

Western-blot analysis of left descending coronary artery protein expressions of endothelial nitric oxide synthase (eNOS), soluble epoxide hydrolase (sEH), and large-conductance calcium-activated potassium (BK_Ca) channels, normalized to smooth muscle actin at W16 in control mice (n = 6–8), nontreated HFD mice (n = 5 to 6), and HFD mice treated with GLI (n = 5 to 6) or with t-AUCB (n = 6).

Fig. 6.

Assessment of left ventricular end-diastolic (LVED) anterior wall thickness (A), LVED posterior wall thickness (B), LVED diameter (C), E/A ratio (D: mean values and representative recordings), cardiac output (E), and ejection fraction (F) in control mice (n = 6–9), nontreated HFD mice (n = 9–11), and HFD mice treated with GLI (n = 6–9) or with t-AUCB (n = 8–10) at W0, W8, and W16. *P < 0.05 vs. W0; †P < 0.05 vs. control; ‡P < 0.05 vs. W8; §P < 0.05 vs. nontreated HFD; ||P < 0.05 vs. HFD + GLI.

Fig. 7.

Left ventricular (LV) weight-to-tibia length ratio (A), LV collagen density (B), LV infiltration (C), LV protein expression of phospho-Akt (pAkt) normalized to Akt, NF-κB (dotted line indicates gel splicing), pIκB normalized to IκB, sEH, NOX4, catalase, and SOD normalized to actin (D), and LV mRNA expression of CYP2J5 and CYP2C9 normalized to GAPDH and β₂-microglobulin, at W16 in control mice (n = 6–16), nontreated HFD mice (n = 6–15), and HFD mice treated with GLI (n = 6–15) or with t-AUCB (n = 6–18). *P < 0.05 vs. control; †P < 0.05 vs. nontreated HFD; ‡P < 0.05 vs. HFD + GLI.

Fig. 8.

Markers of diastolic function (A): left ventricular end-diastolic pressure (LVEDP), LV minimal change in pressure over time (dP/dt_min), LV relaxation constant τ, LV end-diastolic pressure-volume relation (LVEDPVR); and markers of systolic function (B): LV end-systolic pressure (LVESP), LV maximal change in pressure over time (dP/dt_max), LV end-systolic pressure-volume relation (LVESPVR) obtained by invasive hemodynamics (mean values and representative recordings) at W16 in control mice (n = 6), nontreated HFD mice (n = 7), and HFD mice treated with GLI (n = 4) or with t-AUCB (n = 8). *P < 0.05 vs. control; †P < 0.05 vs. nontreated HFD.