It is known that diabetic mice are less sensitive to the analgesic effect of morphine. Some factor(s) derived from mononuclear cells, e.g. interleukin-1beta (IL-1beta), may be responsible for the diminished analgesic effect of morphine in diabetic mice. Therefore, we examined direct effects of IL-1beta, intracerebroventricularly (i.c.v.), on morphine-induced analgesia, subcutaneously (s.c.), in diabetic and control mice by using the tail-flick test. Morphine at doses of 1, 2 and 5 mg/kg (s.c.) produced dose-dependent analgesia in diabetic and control mice but diabetic mice were less sensitive to the analgesic effect of morphine when compared to the controls. IL-1beta at a dose of 0.1 ng/mouse produced analgesia in control mice but not in diabetics, whereas IL-1beta at a dose of 10 ng/mouse produced a hyperalgesic effect both in diabetic and control mice. IL-1beta at a dose of 1 ng/mouse has neither an analgesic nor a hyperalgesic effect in control and diabetic mice. Administration of a neutral (neither analgesic nor hyperalgesic) dose of IL-1beta, 1 ng/mouse (i.c.v.), just prior to administration of morphine (s.c.) abolished the analgesic effect of morphine at doses of 1, 2 and 5 mg/kg in control mice and the analgesic effect of morphine became similar to that in diabetics. The diminished analgesic effect of morphine in diabetes was attenuated further with IL-1beta at a dose of 1 ng/mouse (i.c.v.). These results suggest that the decreased analgesic effect of morphine in diabetes may be related to IL-1beta.