The potassium channel openers could share a common mechanism of action with the opiates in the production of antinociception because both classes of drugs have been shown to enhance potassium efflux. The i.t. administration of the potassium channel openers diazoxide, minoxidil, and lemakalim (BRL38227) produced antinociception as measured in the tail-flick test. The ED50's were 122, 184 and 35 micrograms/mouse, respectively. Minoxidil and lemakalim (BRL 38227) were full agonists, whereas diazoxide was a partial agonist. Diazoxide-induced antinociception was blocked by the potassium channel blockers glyburide, apamin and charybdotoxin. Minoxidil- and lemakalim (i.t.)-induced antinociception was blocked totally by only glyburide. The antinociception produced by the potassium channel openers (i.t.) was blocked differentially by opiate antagonists (i.t.). The antinociceptive effects of diazoxide were blocked by nor binaltorphimine, ICI 174,864 and naloxone. Minoxidil- and lemakalim-induced antinociception was blocked by naloxone and ICI 174,864, but not by nor-binaltorphimine. Naloxone (s.c.) shifted the dose-effect curve for minoxidil to the right in a parallel manner. Morphine-induced antinociception was partially blocked by glyburide and apamin, whereas that produced by DPDPE was blocked totally by apamin. U50, 488H-induced antinociception was blocked partially by apamin. The potassium channel openers (i.t.) were not cross tolerant to morphine when measured in the tail-flick test. Apamin and glyburide precipitated "withdrawal-like" symptoms in morphine-tolerant mice. The interaction of potassium channel openers and opioids probably does not represent a direct interaction of these two classes of drugs at a similar receptor, but rather may occur via an interaction with a common second messenger system such as calcium.