Among other factors, cytochrome P450 (CYP) enzyme activity determines polychlorinated biphenyl (PCB) bioaccumulation, biotransformation, and toxicity in exposed species. We measured the oxidative metabolism in vitro of 12 PCB congeners, representing structural groups based on the number and position of the chlorine atoms, by the hepatic microsomes of one Baltic grey seal (Halichoerus grypus). Microsomal metabolism was observed for several PCBs with vicinal H atoms exclusively in the ortho and meta positions and without any ortho-Cl substituents (CB-15 [4,4'-Cl2] and CB-77 [3,3',4,4'-Cl4]), vicinal meta and para-H atoms (CB-52 [2,2',5,5'-Cl4], and -101 [2,2',4,5,5'-Cl5]) or with both characteristics in combination with either only one ortho-Cl (CB-26 [2,3',5-Cl3], CB-31 [2,4',5-Cl3]) or two ortho-Cl substituents (CB-44 [2,2',3,5'-Cl4]). To allocate PCB biotransformation to specific CYPs, the inhibitive effect of compounds with known CYP-specific inhibition properties was assessed on in vitro PCB metabolism and on regio- and stereospecific testosterone hydroxylase activities. Metabolic inhibition was considered relevant at concentrations < or = 1.0 microM because these inhibitors became decreasingly selective at higher concentrations. At < 1.0 microM, ellipticine (CYPIAI/2 inhibitor) selectively inhibited CB-15, -26, -31, and -77 metabolism, with no significant inhibition of CB-44, -52, and -101 metabolism. Inhibition of CB-52 and -101 metabolism by chloramphenicol (CYP2B inhibitor) started at 1.0 microM and maximized at about 100% at 10 microM. Ketoconazole (CYP3A inhibitor) appeared to selectively inhibit CB-26, -31, and -44 metabolism relative to CB-15, -77, and -52 at concentrations < or = 1.0 microM. Major testosterone metabolites formed in vitro were 2beta-(CYP3A), 6beta- (CYP3A, CYPIA), and 16beta- (CYP2B) hydroxytestosterone and androstenedione (CYP2B, CYP2C11). The CYP forms indicated are associated with the specific metabolism of testosterone in laboratory animals. Inhibition of 2beta- and 6beta-hydroxytestosterone formation at ellipticine and ketoconazole concentrations < or = 1.0 microM suggested that both inhibitors were good substrates of CYP3A-like enzymes in grey seal. Chloramphenicol (model for CYP2B) is apparently not a good inhibitor of CYPI A and CYP3A activities in grey seal because the chemical did not inhibit any metabolic route of testosterone at concentrations from 0.1 to 10 microM. Our findings demonstrated that at least CYP1A- and CYP3A-like enzymes in the liver of grey seals are capable of metabolizing PCBs with ortho-meta and/or meta-para vicinal hydrogens. A CYP2B form might also be involved, but this could not be proven by the results of our experiments. Defining the profiles of CYP enzymes that are responsible for PCB biotransformation is necessary to fully understand the bioaccumulation, toxicokinetics, and risk of PCB exposure in seals and other free-ranging marine mammals.