Patients with dihydralazine hepatotoxicity have been found to have circulating autoantibodies that react with liver microsomes (anti-liver microsome antibodies) and that are clearly distinct from anti-liver and kidney microsomal antibodies observed in patients with tienilic acid-induced hepatitis and in some patients with autoimmune hepatitis. The authors show that anti-liver microsome antibodies present in the sera of five patients with dihydralazine-induced hepatitis specifically react on immunoblots with a 53 kD protein. They further conclude that this target antigen is the phase I drug metabolizing enzyme termed P-450IA2 based on the following observations: (a) immunoblots of a battery of human liver microsomes produced the identical pattern of relative staining whether the blots were developed with anti-rat P-450IA2 IgG, anti-human P-450IA2 IgG or each of the five patients' sera; (b) P-450IA2 catalytic activity was selectively inhibited when human liver microsomes were preincubated with anti-liver microsome-positive sera; (c) anti-liver microsome-positive sera identified purified human P-450IA2 on immunoblots. Anti-liver microsome antibodies appeared to be specific for dihydralazine hepatitis because they were not present in sera obtained from 28 other patients including patients receiving dihydralazine without a toxic response and patients with other significant liver diseases. Finally, the authors demonstrated that dihydralazine could competitively inhibit catalytic activity characteristic of P-450IA2 in human liver microsomes, suggesting that P-450IA2 may be involved in the metabolism of dihydralazine.