To examine whether endothelial dysfunction in hypertension is reversible or not, we studied the effects of imidapril, an angiotensin-converting enzyme inhibitor, on nitric oxide release in stroke-prone spontaneously hypertensive rats (SHR) and deoxycorticosterone acetate (DOCA)-salt hypertensive rats. After a 4-week treatment with imidapril (1 or 10 mg/d SC) or vehicle, acetylcholine-induced vasodilation and nitric oxide release in the isolated kidneys were determined. Nitric oxide release was measured by a chemiluminescence assay. Imidapril lowered blood pressure in stroke-prone SHR in a dose-dependent manner. Untreated stroke-prone SHR exhibited significantly attenuated responses to acetylcholine (10(-8) mol/L) of both renal perfusion pressure (stroke-prone SHR 42+/-4% versus Wistar-Kyoto rats [WKY] 58+/-4% [mean+/-SE], P<.01) and nitric oxide release (stroke-prone SHR +7.6+/-2.1 versus WKY +29.7+/-9.7 fmol/min per gram of kidney wt, P<.01). Imidapril at 10 mg/d significantly increased acetylcholine-induced renal vasodilation and nitric oxide release in stroke-prone SHR (renal perfusion pressure, 56+/-3%; nitric oxide release, +27.1+/-6.4 fmol/min per gram of kidney wt; both P<.01 versus stroke-prone SHR treated with vehicle). On the other hand, imidapril neither decreased blood pressure nor changed nitric oxide release induced by acetylcholine in DOCA-salt hypertensive rats. Staining for endothelial nitric oxide synthase and brain nitric oxide synthase was clearly detected in the kidneys of both stroke-prone SHR and WKY, whereas staining intensity was weaker in DOCA-salt hypertensive rats. Inducible nitric oxide synthase immunoreactivity was barely noticeable in any type of rat. Thus, imidapril restored endothelial damage by pressure-dependent mechanisms. Most of the nitric oxide detected in the perfusate seemed to be derived from constitutive nitric oxide synthase.