Several lines of evidence support the hypothesis that ATP-sensitive K+ channels (K+(ATP)) participate in the brain's regulation of peripheral glucose homeostasis. In testing this hypothesis we conducted a series of in vivo experiments using albino rats and bilateral intrahypothalamic injections of K+(ATP) channel blockers, glibenclamide and repaglinide. The results show that 0.2 and 2.0 nM injections of glibenclamide lowered blood glucose in a dose-dependent manner. During mild insulin-induced hypoglycemia, hypothalamic glibenclamide delayed recovery to normoglycemia. The impaired recovery was associated with a reduction in plasma norepinephrine (P<0.001), though circulating epinephrine and glucagon were not reduced. In a separate experiment, 2-deoxy-D-glucose (200 mg/kg) was intraperitoneally administered to produce neuroglucopenia. Hypothalamic injections of either glibenclamide or repaglinide significantly blunted compensatory hyperglycemic responses (P<0.01). In a feeding study, 2.0, but not 0.2 nM of hypothalamic glibenclamide, reduced chow intake over a 2-h period (P<0.01). The results support the hypothesis that hypothalamic K+(ATP) channels participate in central glucose-sensing and glucose regulation.