Analysis of a method for testing azo dyes for mutagenic activity in Salmonella typhimurium in the presence of flavin mononucleotide and hamster liver S9.

A protocol for assessing the mutagenic activity of azo dyes derived from mutagenic or potentially mutagenic aromatic amines was evaluated, using 4 model compounds. This protocol is based upon one developed in Sugimura's laboratory with modifications, including the use of flavin mononucleotide (FMN) rather than riboflavin to reduce the azo compounds to free amines, and hamster liver S9 rather then rat liver S9 for metabolic activation. The protocol developed differs from the standard Ames Salmonella plate incorporation assay in 5 ways: (1) uninduced hamster liver S9 rather than Aroclor 1254-induced rat liver S9 is used; (2) 150 microliters of S9 is used rather than the maximum of 50 microliter of S9 used in the standard assay; (3) FMN is added to the cofactor mix; (4) the cofactor mix is modified to include exogenous glucose 6-phosphate dehydrogenase, NADH, and 4 times the standard amount of glucose 6-phosphate; and (5) a 30-min "pre-incubation" step is used before addition of top agar. We found that each of these 5 changes is necessary for optimal mutagenic activity of azo dyes derived from the mutagenic aromatic amines benzidine, o-tolidine or o-dianisidine. The use of hamster liver S9 rather than rat liver S9 was also required for optimal mutagenic activity of benzidine itself. Rat liver S9 inhibited the ability of hamster S9 to activate benzidine to a mutagen. The presence in rat liver S9 of an inhibitor of the metabolic activation of benzidine may account for the failure of benzidine and a benzidine dye (Congo red) to be strongly mutagenic when tested with this type of S9.