The present study was conducted to see the effect of probucol on streptozotocin diabetes in rats. After 2 weeks of a 1% probucol diet, 35 or 50 mg/kg of streptozotocin were intravenously injected into male Wistar rats. All the rats became diabetic 2 days after treatment. Thereafter, in order to see the effect of probucol on spontaneous recovery from streptozotocin diabetes, 25 mg/kg of streptozotocin was injected into rats after two weeks of probucol diet and the diet was continued for additional two weeks. All the rats with a standard diet (group CS, n = 13) and 12 of 13 rats with probucol diet (group PS) became diabetic 2 days after streptozotocin injection. One rate from group PS did not develop diabetes. Two weeks after injection, only 4 of 13 rats in groups CS showed recovery, while 11 of 12 rats in group PS showed recovery from streptozotocin diabetes (p less than 0.05). The average blood glucose levels in group PS were significantly lower than group CS (10.5 +/- 4.6 vs 18.5 +/- 0.6 mM, p less than 0.05). In addition, the pancreatic insulin content of group PS was 8 times greater than that of group CS (0.75 +/- 0.24 vs 0.09 +/- 0.03 mmol/pancreas, p less than 0.01). Thus, the in vivo diabetogenic action of streptozotocin could not be reduced by pretreatment with probucol. However, recovery from streptozotocin diabetes was induced by subsequent treatment with probucol. The precise mechanisms for this phenomenon were not known; but the present findings suggest the protective effect of probucol on beta-cell damage induced by small dose of streptozotocin.