Abstract

Azo colorants may undergo in vivo reduction of the azo group to release amines, some of which, through subsequent metabolic activation, are considered to be carcinogenic. Thus, all azo colorants whose metab olism can potentially liberate a carcinogenic amine are suspected of having a similar carcinogenic potential. Chemical methods used to assess the likely production of such amines use reductants such as sodium dithionite or Raney Nickel. A recent investigation of the applicability of the use of these strong chemical reductants to force azo fission questions their suitability for certain classes of azo colorants. In particular, potentially non-carcinogenic aniline-based azo dyes were found to release trace amounts of the known human carcinogen 4-aminodiphenyl, even though this amine is absent before reduction. To assess the potential for the enzymatic formation of 4-aminodiphenyl in vii’o, we have compared the reductive biotransformation of the aniline-azo-linked dye Coriacide Black SB with that of a 4-aminodiphenyl-azolinked analogue in rats. 4-Aminodiphenyl could not be detected in the urine of rats administered Coriacide Black SB, whereas this amine was detected in the urine of rats administered the 4-aminodiphenyl-azo-linked analogue. This finding supports evidence that the 4-aminodiphenyl detected following the chemical reduction of dyes incorporating aniline as the azo component is an artefact of the reduction conditions and would not occur in vivo.