The effect of ethanol on testosterone-5 alpha-A-ring reductase activity was studied in the whole homogenate and/or subcellular fractions (microsomes and cytosol) of buccal mucosa, gingiva, liver and prostate from intact ethanol-fed (36% of dietary calories were given as ethanol for up to 360 days) and pair-fed control rats. An increased enzyme activity was found in the hepatic, prostatic and gingival homogenates from ethanol-fed rats. No difference in the enzyme activity was seen in the homogenates of buccal mucosa. However, when the buccal mucosal microsomal fraction was used a significant (P less than .005) increase in the enzyme activity was found in the ethanol-fed rats. It was determined that the lack of an increase in 5 alpha-reductase activity in the buccal mucosal homogenates from ethanol-fed rats was due to the presence of a cytosolic inhibitor of this enzyme. Enzyme kinetics showed a decrease in the velocity in correlation with the increasing concentration of cytosolic inhibitor (Vmax control = 1.9 and Vmax ethanol = 1.7, 1.45 and 1.0 nmol, respectively). The apparent Km (0) and Kp (1--3) values were similar for all combinations (4.5 x 10(-5) M). In addition, a similar Ki constant (2.2 mg of cytosolic protein) was found for different testosterone concentrations. These results suggest that the ethanol-induced cytosolic inhibitor in buccal mucosa combines with the enzyme, independent of the substrate and inhibits it by an allosteric mechanism. Studies using dialysis, heating and tryptic digestion suggests that the inhibitor is a protein.