In vitro and in vivo comparison of two K+ channel openers, diazoxide and cromakalim, and their inhibition by glibenclamide.

Diazoxide caused an increase in 86Rb+ efflux from the rat aorta and portal vein and inhibited spontaneous activity of the latter at concentrations 100 times higher than the K+ channel opener cromakalim. In the rabbit aorta both drugs inhibited vasoconstrictor responses to angiotensin II, noradrenaline and low concentrations (less than or equal to 30 mM) of KCl in a similar manner, the antivasoconstrictor activities being abolished in vessels depolarized with greater than or equal to 35 mM K+. In vivo cromakalim was about 100 times more potent than diazoxide at lowering blood pressure in rats. Diazoxide (30 mg/kg) caused a more than 2-fold increase in plasma glucose in rats and prevented any return toward base line within 1.5 hr after a glucose load. Cromakalim had minimal effects upon glucose homeostasis at equihypotensive doses. Glibenclamide, a potent blocker of ATP-dependent K+ channels, inhibited the stimulation by cromakalim and diazoxide of 86Rb+ efflux from the portal vein and aorta (IC50 approximately 0.1 microM), antagonized their vasorelaxant effects in vitro and in vivo (20-30 mg/kg i.v.) and reversed the diazoxide-induced changes in plasma glucose and insulin levels. These results provide evidence that diazoxide, like cromakalim, is able to open 86Rb+-permeable K+ channels in vascular smooth muscle. This action is likely to be responsible for the in vitro and in vivo vasodilator activity of these two drugs. However, there would seem to be pharmacological differences between the K+ channels affected by these drugs in vascular smooth muscle and the (ATP-sensitive) K+ channels of pancreatic beta-cells, which are thought to be responsible for the effects of diazoxide on plasma glucose.