A consolidated classification system is described for prokaryotic and eukaryotic N-acetyltransferases in accordance with the international rules for gene nomenclature. The root symbol (NAT) specifically identifies the genes that code for the N-acetyltransferases, and NAT* loci encoding proteins with similar function are distinguished by Arabic numerals. Allele characters, denoted by Arabic numbers or by a combination of Arabic numbers and uppercase Latin letters, are separated from gene loci by an asterisk, and the entire gene-allele symbols are italicized. Alleles at the different NAT* loci have been numbered chronologically irrespective of the species of origin. For designation of genotypes at a single NAT* locus, a slash serves to separate the alleles; in phenotype designations, which are not italicized, alleles are separated by a comma.

A genetic polymorphism at the NAT2 gene locus, encoding for polymorphic N-acetyltransferase (NAT2), segregates individuals into rapid, intermediate or slow acetylator phenotypes. Both rapid and slow acetylator phenotypes have been associated with increased incidence of cancer in certain target organs related to arylamine exposure, suggesting a role for acetylation in both the activation and deactivation of arylamine carcinogens. A second gene (NAT1) encodes for a different acetyltransferase isozyme (NAT1) that is not subject to the classical acetylation polymorphism. In order to assess the relative ability of NAT1 and NAT2 to activate and deactivate arylamine carcinogens, we tested the capacity of recombinant human NAT1 and NAT2, expressed in Escherichia coli XA90 strains DMG100 and DMG200 respectively, to catalyze the N-acetylation (deactivation) and Oacetylation (activation) of a variety of carbocyclic and heterocyclic arylamine carcinogens. Both NAT1 and NAT2 catalyzed the N-acetylation of each of the 17 arylamines tested. Rates of N-acetylation by NAT1 and NAT2 were considerably lower for heterocyclic arylamines such as 2-amino-3-methyl-imidazo[4, 5-f]quinoline (IQ), particularly those (e.g. IQ) with steric hindrance to the exocydic amino group. For carbocyclic arylamines such as 4-aminobiphenyl and β-naphthylamine, the apparent affinity was significantly (P < 0.05) higher for NAT2 than NAT1. NAT1/NAT2 activity ratios and clearance calculations suggest a significant role for the polymorphic NAT2 in the N-acetylation of carbocyclic arylamine carcinogens. Both NAT1 and NAT2 catalyzed acetyl coenzyme A-dependent O-acetylation of Af-hydroxy-2-aminofluorene and N-hydroxy-4-aminobiphenyl to yield DNA adducts. NAT1 catalyzed paraoxon-resistant, intramolecular N, O-acetyl-transferase-mediated activation of N-hydroxy-2-acetylamiio-fluorene and N-hydroxy-4acetylaminobiphenyl at low rates; catalysis by NAT2 was not readily detectable in the presence of paraoxon. In summary these studies strongly suggest that the human acetylation polymorphism influences both the metabolic activation (O-acetylation) and deactivation (N-acetylation) of arylamine carcinogens via polymorphic expression of NAT2. These findings lend mechanistic support for human epidemiological studies suggesting associations between both rapid and slow acetylator phenotype and cancers related to arylamine exposure.