In vitro metabolism of acrylonitrile to 2-cyanoethylene oxide, reaction with glutathione, and irreversible binding to proteins and nucleic acids.

The metabolism of the suspected carcinogen acrylonitrile was studied using subcellular fractions isolated from rats and humans. Irreversible binding of radioactive label from [1-14C]or [2,3-14C]acrylonitrile to protein and DNA was enhanced by reduced nicotinamide adenine dinucleotide phosphate in the presence of rat liver microsomes or a reconstituted cytochrome P-450 system. During the reduced nicotinamide adenine di nucleotide phosphate-dependent reaction, HCN was pro duced, and the heme of cytochrome P-450 was destroyed. Rat brain microsomes did not produce detectable levels of metab olites. Conclusive evidence of metabolically mediated binding of acrylonitrile to protein and DNA in human systems was not found. With rats, metabolism was induced by pretreatment of animals with either phénobarbital or 5,6-benzoflavone. Labeled 2-cyanoethylene oxide was found to bind irrevers ibly to calf thymus DNA and microsomal protein. The extent of binding was greater in the case of 2,3-14C-labeled than Re labeled material. The relative first-order rate of acrylonitrile binding to calf thymus DNA in rat liver microsomal systems was one to two orders of magnitude less than that for 1,1,2-trichloroethylene and three orders of magnitude less than that for vinyl chloride or vinyl bromide. Rat liver microsomes or a reconstituted cytochrome P-450 system catalyzed the mixedfunction oxidation of acrylonitrile to 2-cyanoethylene oxide. 2Cyanoethylene oxide has a half-life of about 2 hr in neutral buffer at 37°.This epoxide was found to serve as a substrate for microsomal epoxide hydrolase. HCN was released during hydrolysis of 2-cyanoethylene oxide or reaction of the epoxide with reduced glutathione; however, HCN release in either case was not stoichiometric with epoxide disappearance. 2-Cyanoethylene oxide reacted less rapidly with reduced glutathione than did acrylonitrile. Rat liver cytosol preparations contained glutathione S-transferase activity towards acrylonitrile and greater activity with 2-cyanoethylene oxide as a substrate. Cytosol preparations from rat brain and human liver had no detectable glutathione S-transferase activity towards acryloni trile but did exhibit some activity towards 2-cyanoethylene oxide. These studies establish the basic pathways involved in the metabolism of acrylonitrile and should provide a basis for ' This work was supported in part by USPHS Grants ES 00267 and ES 02205 and Contract NO1 CP 85672. 2 Recipient of USPHS Research Career Development Award ES 00041. To whom requests for reprints sbould be addressed. 3 Recipient of a Monsanto Fund Fellowship in Toxicology as postdoctoral fellow. ' Recipient of USPHS Training Grant GM 07628. Received June 2, 1981 ; accepted August 24, 1981. examination of the relevance of these individual steps and their roles in bioactivation and detoxication under more physiologi cal situations.