A polyphenol extract from a Corbières (France) red wine (P, 200 mg/kg), ethanol (E, 1 mL/kg), or a combination of both (PE) was administered by daily gavage for 6 weeks to healthy control or streptozotocin (60 mg/kg i.v.)-induced diabetic rats (180-200 g). Treatment groups included C or D (untreated control or diabetic) and CP, CE, or CPE (treated control) or DP, DE, or DPE (treated diabetic). P treatment induced a reduction in body growth, food intake, and glycemia in both CP and DP groups. In DP, hyperglycemia was reduced when measured 1 h after daily treatment but not at sacrifice (no treatment on that day). The hyperglycemic response to the oral glucose tolerance test (OGTT) and plasma insulin at sacrifice were impaired similarly in DP and D groups. In contrast, in DE or DPE, body growth was partially restored while hyperglycemia was reduced both during treatment and at sacrifice. In addition, hyperglycemia response to OGTT was reduced and plasma insulin was higher in DE or DPE than in D animals, indicating a long-term correction of diabetes in ethanol-treated animals. Morphometric studies showed that ethanol partially reversed the enlarging effect of diabetes on the mesenteric arterial system while the polyphenolic treatment enhanced it in the absence of ethanol. In summary, our study shows that (i). a polyphenol extract from red wine ("used at a pharmacological" dose) reduces glycemia and decreases food intake and body growth in diabetic and nondiabetic animals and (ii). ethanol ("nutritional" dose) administered alone or in combination with polyphenols is able to correct the diabetic state. Some of the effects of polyphenols were masked by the effects of ethanol, notably in diabetic animals. Further studies will determine the effect of "nutritional" doses of polyphenols as well as their mechanism of action.