4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK, 1) and its metabolite, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL, 2) are both potent pulmonary carcinogens in rats. The metabolism of NNK to NNAL is stereoselective and reversible, with (S)-NNAL being the major enantiomer formed from NNK. In rats, (R)-NNAL undergoes facile glucuronidation and is rapidly excreted in urine, whereas (S)-NNAL is preferentially retained in tissues and converted to NNK. We hypothesized that the lung carcinogenicity of NNK in the rat is due in part to the preferential retention of (S)-NNAL in the lung, the reconversion to NNK, and then the metabolic activation of NNK to pyridyloxobutyl (POB)-DNA adducts. We tested this hypothesis by treating male F344 rats with 10 ppm of NNK, (R)-NNAL, or (S)-NNAL in drinking water. After 1, 2, 5, 10, 16, or 20 weeks of treatment, POB-DNA adducts in liver and lung DNA were quantified by HPLC-ESI-MS/MS. At each time point, total adduct levels were higher in the lung than in the liver. O2-[4-(3-pyridyl)-4-oxobut-1-yl]thymidine (O2-POB-dThd, 13) was the major adduct detected. Total adduct levels in the rats treated with (S)-NNAL were 0.6-1.3 times as great as those in the NNK group in the lung and 0.7-1.4 times in the liver, and 6-14 times higher than those in the (R)-NNAL group in the lung and 11-17 times in the liver. These results suggest that (S)-NNAL is stereoselectively retained in tissues. This study demonstrates for the first time the accumulation and persistence of specific POB-DNA adducts in the rat lung and liver during chronic treatment with NNK, (R)-NNAL, and (S)-NNAL and supports the hypothesis that the preferential retention of (S)-NNAL in the lung, followed by reconversion to NNK and then the metabolic activation of NNK is critical for lung carcinogenesis by NNK and NNAL.