RESUMEN
When testing new products, potential new products, or their impurities for genotoxicity in the Ames test, the quantity available for testing can be a limiting factor. This is the case for a dye repository of around 98,000 substances the Max Weaver Dye Library (MWDL). Mutagenicity data on dyes in the literature, although vast, in several cases is not reliable, compromising the performance of the in silico models. In this report, we propose a strategy for the generation of high-quality mutagenicity data for dyes using a minimum amount of sample. We evaluated 15 dyes from different chemical classes selected from 150 representative dyes of the MWDL. The purity and molecular confirmation of each dye were determined, and the microplate agar protocol (MPA) was used. Dyes were tested at the limit of solubility in single and concentration-response experiments using seven strains without and with metabolic activation except for anthraquinone dyes which were tested with eight strains. Six dyes were mutagenic. The most sensitive was YG1041, followed by TA97a > TA98 > TA100 = TA1538 > TA102. YG7108 as well as TA1537 did not detect any mutagenic response. We concluded that the MPA was successful in identifying the mutagenicity of dyes using less than 12.5 mg of sample. We propose that dyes should be tested in a tiered approach using YG1041 followed by TA97a, TA98, and TA100 in concentration-response experiments. This work provides additional information on the dye mutagenicity database available in the literature.