We examined how acute diabetes mellitus and acute ethanol intoxication modulate factors that mediate immune responses as a basis for explaining the increased susceptibility to infection in these two conditions. Our working hypothesis is that ethanol intoxication in diabetes compromises host defense mechanisms to a greater extent than observed in each condition alone. Male and female rats were made diabetic with streptozotocin (65 mg/kg, i.p.). Forty-eight hours after administration of streptozotocin, rats either received no treatment (control group) or were treated with (1) ethanol (bolus injection of 1.75 g/kg, followed by a 3-h infusion at the rate of 300 mg/kg/h), (2) lipopolysaccharide [(LPS); 0.9 mg/kg], or (3) a combination of LPS+ethanol. At the end of 3 h, rats were killed, and the livers were digested by perfusion with collagenase-containing Hanks' balanced salt solution to isolate hepatocytes and Kupffer cells. To measure chemokine generation, hepatocytes (2.5x10(5) cells per well) and Kupffer cells (1x10(6) cells per well) were cultured for 20 h, and the supernatant was used to measure cytokine-induced neutrophil chemoattractant (CINC) and regulated on activation, normal T-cell expressed and secreted (RANTES) chemokines. Phagocytosis by Kupffer cells was measured by flow cytometry and expressed as mean channel fluorescence intensity (MCF). Induction of diabetes as well as treatment of nondiabetic rats with LPS, ethanol, or LPS+ethanol caused depression of MCF values of Kupffer cells. However, treatment of the diabetic male and female rats with LPS and LPS+ethanol increased the MCF values relative to those of Kupffer cells obtained from untreated diabetic rats, but administration of ethanol to diabetic rats did not have a similar effect. The induction of diabetes caused an increase in CINC generation by Kupffer cells obtained from male rats, but not from female rats. This diabetes-induced elevation of chemoattractant factor was decreased when diabetic animals were treated with LPS, ethanol, or LPS+ethanol, and the sex difference was obliterated. Thus, the induction of diabetes as well as treatment with LPS, ethanol, or LPS+ethanol in nondiabetic rats depressed the phagocytic capability of Kupffer cells, whereas the presence of endotoxemia (administration of the endotoxin LPS) or administration of LPS+ethanol reversed the diabetic effect, but ethanol intoxication did not. These findings seem to indicate a persistence of depression of host defense capacity in the ethanol-intoxicated diabetic condition. This is further reinforced by the depression of the diabetes-induced enhancement of chemotaxis when the diabetic rats became intoxicated.