1,3-Butadiene (BD) has been classified as a probable human carcinogen based on sufficient evidence of a carcinogenic response in B6C3F1 mice and Sprague-Dawley rats and limited human evidence of carcinogenicity. Mice are much more susceptible to BD-induced carcinogenicity than rats. Previous in vitro studies revealed that mouse liver microsomes formed 1,2-epoxy-3-butene (BMO) from BD and 1,2:3,4-diepoxybutane (BDE) from BMO at much higher rates than rat or human microsomes. BDE was also readily quantitated in blood and tissues of mice exposed to BD but could not be detected in rats similarly exposed. These findings suggest a key role for BDE in BD-induced carcinogenicity. The purpose of this study was to characterize the glutathione (GSH) conjugation of BDE by liver and lung cytosol from B6C3F1 mice and Sprague-Dawley rats and human liver cytosol from six different individuals in vitro. BDE and glycine-[2-3H]GSH were incubated, at pH 7.4, with cytosol. 13C NMR and mass spectral analysis indicated formation of two isomeric conjugates, S-(1-(hydroxy-methyl)-2,3-epoxypropyl)glutathione and S-(2-hydroxy-3,4-epoxy--butyl)glutathione, which were rapidly hydrolyzed in cytosol to the corresponding trihydroxy conjugates. Total conjugates were quantitated by HPLC. Conjugation of BDE with GSH followed Michaelis-Menten kinetics in human as well as rat and mouse cytosolic fractions. The conjugation rates in mouse and rat liver cytosol were similar (Vmax 162 +/- 16 and 186 +/- 37 nmol/mg protein/min, respectively) and an order of magnitude higher than in human liver cytosol (Vmax 6.4 +/- 1.9 nmol/mg protein/min). the apparent KM values were lower in human (2.1 +/- 1.4 mM) than mouse (6.4 +/- 1.6 mM) or rat (24 +/- 6 mM) liver. Mouse lung cytosol (Vmax 38.5 +/- 2.5 nmol/mg protein/min, KM 1.70 +/- 0.37mM) is also more efficient in GSH conjugation than rat lung cytosol (Vmax 17.1 +/- 3.0 nmol/mg protein/min, KM +/- 1.7 mM). These results suggest that the higher BDE blood concentrations in mice compared with rats following inhalation exposure to BD are not due to differences in hepatic or pulmonary GSH conjugation of BDE. Also, considering the low oxidation rates of BD to BMO and of BMO to BDE in humans as compared to mice, the relatively low capacity of GSH conjugation of BDE in human liver will not necessarily lead to increased BDE blood levels in humans potentially exposed to BD.