Endothelial-specific CYP4A2 overexpression leads to renal injury and hypertension via increased production of 20-HETE

Abstract

We have previously reported that adenoviral-mediated delivery of cytochrome P-450 (CYP) 4A2, which catalyzes the synthesis of 20-hydroxyeicosatetraenoic acid (20-HETE), results in endothelial dysfunction and hypertension in Sprague-Dawley (SD) rats (Wang JS, Singh H, Zhang F, Ishizuka T, Deng H, Kemp R, Wolin MS, Hintze TH, Abraham NG, Nasjletti A, Laniado-Schwartzman M. Circ Res 98: 962-969, 2006). In this study, we targeted the vascular endothelium by using a lentivirus construct expressing CYP4A2 under the control of the endothelium-specific promoter VE-cadherin (VECAD-4A2) and examined the effect of long-term CYP4A2 overexpression on blood pressure and kidney function in SD rats. A bolus injection of VECAD-4A2 increased blood pressure (P < 0.001) by 26, 36, and 30 mmHg 10, 20, and 30 days postinjection, respectively. Arteries from VECAD-4A2-transduced rats produced increased levels of 20-HETE (P < 0.01), expressed lower levels of endothelial nitric oxide synthase (eNOS) and phosphorylated eNOS (p-eNOS) (P < 0.05), generated higher levels of superoxide anion, and displayed decreased relaxing responsiveness to acetylcholine (P < 0.05). Proteinuria increased by twofold in VECAD-4A2-transduced rats compared with controls. Treatment of VECAD-4A2-transduced rats with HET0016, an inhibitor of 20-HETE biosynthesis, not only attenuated the increase in blood pressure (P < 0.05) but also improved vascular function (acetylcholine-induced relaxations) and reduced plasma creatinine and proteinuria. HET0016 treatment decreased oxidative stress and increased the phosphorylated state of key proteins that regulate endothelial function, including eNOS, AKT, and AMPK. Collectively, these findings demonstrate that augmentation of vascular endothelial 20-HETE levels results in hypertension, endothelial dysfunction, and renal injury, which is offset by HET0016 through a reduction in vascular 20-HETE coupled with a lessening of oxidative stress and the amplification of pAKT, pAMPK, and p-eNOS levels leading to normalization of endothelial responses.

Fig. 1.

A: Western blot and densitometric analysis of green fluorescent protein (GFP) in cultured HEK cells and EC cells transduced with the VECAD-GFP lentivirus. Blots are representative of 3 separate experiments. Data are shown as mean band density normalized relative to β-actin (means ± SE; n = 3). *P < 0.01 vs. control. B: immunofluorescence analysis of GFP, CD31, and CYP4A2 in renal cortical sections from rats transduced with VECAD-GFP (top) and VECAD-4A2 (bottom). GFP and CYP4A2, green; CD31, red; 4,6-diamidino-2-phenylindole, blue.

Fig. 2.

A: Western blot and densitometric analysis of CYP4A2 in kidney and aorta of rats transduced with the VECAD-GFP (control) or VECAD-CYP4A2 lentiviral vectors with and without treatment with HET0016 (HET). Blots are representative of 3 separate experiments. Data are shown as mean band density normalized relative to β-actin (means ± SE; n = 3). *P < 0.01 vs. control. ^#P < 0.05 vs. CYP4A2. B: 20-HETE biosynthesis in renal interlobar arteries. Values are means ± SE; n = 4. *P < 0.05 vs. control. ^#P < 0.05 vs. CYP4A2.

Fig. 3.

Systolic blood pressure (A), plasma creatinine (B), and proteinuria (C) in VECAD-GFP (control)- or VECAD-4A2-transduced rats with and without treatment with HET0016 (HET). Values are means ± SE; n = 4. *P < 0.01 vs. control. ^#P < 0.05 vs. CYP4A2.

Fig. 4.

Histology of kidneys from rats transduced with the VECAD-GFP (control) or VECAD-4A2 lentiviral vectors with and without HET0016 (HET) treatment. A: hematoxylin and eosin (H&E) staining in cortical and medullary sections. G, glomerulus; PT, proximal tubules; DT, distal tubules; HL, Henle's loop; VR, vasa recta; arrows, inflammatory cells. B: collagen staining in cortical and medullary sections. Pictures are representatives of 4 separate experiments. C: quantitative analysis of Sirius Red staining. Values are means ± SE expressed as the percentage of stained area. *P < 0.05 vs. controls. ^#P < 0.05 vs. CYP4A2 10 days.

Fig. 5.

Acetylcholine-induced relaxation in femoral arteries of VECAD-GFP (control)- or VECAD-4A2-transduced rats with and without treatment with HET0016 (HET). A: concentration-dependent relaxations to acetylcholine. B: EC₅₀ of acetylcholine. Values are means ± SE; n = 4. *P < 0.05 vs. control. ^#P < 0.05 vs. CYP4A2. C: effect of N-methylsulfonyl- 12,12-dibromododec-11-enamide (DDMS) and 20-HETE on acetylcholine-induced relaxations in femoral arteries from VECAD-CYP4A2-transduced rats. D: EC₅₀ of acetylcholine. Values are means ± SE; n = 4. *P < 0.01 vs. CYP4A2. ^#P < 0.05 vs. CYP4A2+DDMS. **P < 0.05 vs. CYP4A2.

Fig. 6.

Superoxide levels in kidney (A) and aorta (B) from VECAD-GFP (control)- or VECAD-4A2-transduced rats with and without treatment with HET0016 (HET). Values are means ± SE; n = 4. *P < 0.01 vs. control. ^#P < 0.05 vs. CYP4A2. **P < 0.05 vs. CYP4A2+HET0016.

Fig. 7.

Western blot and densitometric analyses of pAKT and AKT (A), pAMPK and AMPK (B), eNOS (C), and phosphorylated eNOS (p-eNOS; D) in kidney and aorta of VECAD-GFP (control)- or VECAD-4A2-transduced rats with and without treatment with HET0016 (HET). Blots are representative of 3 separate experiments. Values are shown as mean band density normalized relative to AKT, AMPK, or β-actin (means ± SE; n = 3). *P < 0.01 vs. control. ^#P < 0.01 vs. CYP4A2. ‡P < 0.01 vs. CYP4A2+HET0016.