Differences in the metabolic activation of 3-nitrobenzo[a]pyrene and 3-aminobenzo[a]pyrene: studies in the rat, hamster and human lung and liver fractions. 1. The in vitro activation of 3-nitro[a]pyrene (3-Np) and 3-aminobenzo[a]pyrene (3-ABP) was studied in the 7,8-benzoflavone (BNF)-induced hamster embryo system and 7,8-benzoflavone (BNF)-induced rat lung, liver and kidney fractions. 2. Both compounds were less effective than benzo[a]pyrene in increasing lung weights and the activity of the BNF-dependent P450 (P450) and aryl hydrocarbon hydroxylase (AHH) enzymes. In rat liver, 3-ABP was 5.5-fold more effective than 3-Np in enhancing microsomal P450 activity while in the kidney 3-Np was more active. 3. 3-Np was a weaker inducer of AHH than 3-ABP in the kidney and only slightly more effective than 3-ABP in the lung. In hamster, the two compounds induced AHH activity in the liver at the same level and in the kidney at 1.4-1.8 times lower levels than 3-ABP. 4. These results indicate that the activation of 3-Np to its ultimate carcinogenic form (3-amino-7,8-dihydroxy-9,10-epoxide) is less effective than the activation of 3-ABP. The in vitro data obtained in the hamster embryo system, in vivo experiments in the lung and the results of studies of covalent binding strongly suggest that these compounds require additional conversion to benzo[a]pyrene-trans-7,8-dihydrodiol and/or benzo[a]pyrene-3,6-quinone, for metabolism by enzymes such as P450. These putative enzymes may be involved in a detoxication pathway. 5. Comparison of the results from the various experimental systems is hampered by differences in the methods used for metabolic activation and also by the limitations of the techniques used for quantitating the enzyme activity and the DNA binding. However, these limitations do not diminish the fact that the differences between the compounds observed in these systems do indeed exist and that the presence of the nitro group in the 3-position of benzo[a]pyrene results in a greater decrease in carcinogenic activity.