In vitro studies were conducted to identify the hepatic cytochrome P450 (CYP) forms involved in the oxidative metabolism of [14C]zileuton (ABT-077) and its N-dehydroxylated metabolite, [14C]Abbott-66193, by human liver microsomes. The two compounds were metabolized by parallel pathways to form the corresponding ring-hydroxylated and diastereomer sulfoxide metabolites. Results suggested that whereas the metabolism of zileuton and Abbott-66193 were mediated by the same CYP forms, the CYP forms responsible for hydroxylation (CYP1A2 and CYP2C9/10) were distinct from those involved in sulfoxidation (CYP3A > CYP2C9/10). Sulfoxidation (zileuton, Km = 0.82 +/- 0.40 mM, Vmax = 39.1 +/- 21.8 pmol/min/mg; Abbott-66193, Km = 0.23 +/- 0.06 mM, Vmax = 507 +/- 215 pmol/min/mg; mean +/- SD, N=3) was highly correlated with the CYP3A-specific erythromycin N-demethylase activity (r=0794-0.856; p<0.01, N=11) in human microsomes and was inhibited (32-67%) by ketoconazole and troleandomycin. In addition, purified recombinant human CYP3A4/rat NADPH-P450 reductase fusion protein catalyzed only the sulfoxidation of zileuton and Abbott-66193; no hydroxylated metabolites were detected. On the other hand, hydroxylation of the two compounds (zileuton, Km = 0.34 +/- 0.25 mM, Vmax = 17.8 +/- 5.58 pmol/min/mg; Abbott-66193,Km = 0.39 +/- 0.14 mM, Vmax = 1061 +/- 220 pmol/min/mg) was significantly correlated with 7-ethoxyresorufin O-deethylase (CYP1A2; r=0.652-0.762; p<0.01, N=11) and tolbutamide methyl hydroxylase (CYP2C9/10; r=0.863-0.935; p<0.01, N=10) activity in human liver microsome, and was inhibited (26-51%) by well-known CYP1A2 inhibitors (furafylline and alpha-naphthoflavone). Furthermore, microsomes from human B-lymphoblastoid cells expressing CYP1A2 catalyzed only the hydroxylation of zileuton and Abbott-66193; sulfoxide were not formed. Abbott-66193 was a better substrate for CYP2C9/10, when compared with zileuton: 1) the effect of sulfaphenazole on hydroxylation in human liver microsomes was more pronounced for Abbott-66193 than zileuton (56% vs. 9% inhibition); and 2) the rate of Abbott-66193 hydroxylation by purified CYP2C9 was almost 30-fold greater than that of zilueton.