We used MK-912, a potent new selective alpha 2-adrenergic receptor antagonist that is active orally, to study the effect of short-term, selective alpha 2-blockade on fasting plasma glucose (FPG) and pancreatic islet function in non-insulin-dependent diabetes (NIDDM). Ten asymptomatic patients with NIDDM received either a single oral dose of MK-912 (2 mg) or placebo in a double-blind, cross-over study. B-cell function was measured by the acute insulin response (AIR) to glucose (1.66 mmol/kg intravenously [IV]) and by the AIR to arginine (5 g IV) during a hyperglycemic glucose clamp at a mean glucose level of 32.1 mmol/L to provide an estimation of maximal B-cell secretory capacity. A-cell function was estimated by the acute glucagon response (AGR) to arginine during the glucose clamp. Effective alpha 2-adrenergic blockade was apparently achieved, as there were substantial increases of plasma norepinephrine (NE) (P less than .01) and both systolic blood pressure (SBP) (P less than .01) and diastolic blood pressure (DBP) (P less than .05) after treatment with MK-912, but not after placebo. MK-912 caused a significant (P less than .05) although modest decrease of FPG that was associated with a small increase of fasting plasma insulin (P less than 0.01), C-peptide (P less than .05), and glucagon (P less than .01). FPG and hormone levels remained unchanged after placebo. MK-912 tended to increase the AIR (P = .06) and the C-peptide response (P = .07) to glucose compared with placebo.(ABSTRACT TRUNCATED AT 250 WORDS)