The effect of acetyl-CoA supplementation on the mutagenicity of benzidines in the Ames assay.

The conventional Ames assay metabolising system was confirmed to be deficient in its ability to N-acetylate. This may render the test less sensitive to compounds which normally have an acetylation step during their in vivo activation to carcinogens. The addition of acetyl-coenzyme A to the S9 mix in the Ames assay increased the mutagenicity of benzidine in Salmonella typhimurium strains TA98 and TA1538 4-5-fold. This was consistent with the observation that benzidine is N-acetylated prior to DNA binding in vivo in rat liver. Two 3,3'-disubstituted benzidines, o-tolidine and o-dianisidine, were also tested. A smaller increase in o-tolidine mutagenicity, compared to that observed with benzidine, occurred with the addition of acetyl-coenzyme A. However, the production of acetylated metabolites from o-tolidine was only 37% of that from benzidine. The mutagenicity of o-dianisidine was unaffected by acetyl-coenzyme A. Acetylation of o-dianisidine was only 16% of that observed with benzidine, and the N-acetyl derivatives of o-dianisidine showed lower mutagenicity than the parent amine. The differing responses of benzidine, o-tolidine and o-dianisidine to addition of acetyl-coenzyme A suggests it may not be possible to simply infer the metabolism of 3,3'-disubstituted benzidines to DNA binding species from data on benzidine itself.