Genes for the 290 amino acid, 33-34 kDa cytosolic acetyltransferases (NAT1* and NAT2*) from rat and hamster were cloned and expressed in Escherichia coli. Active clones were selected by a simple visual test for their ability to decolorize 4-aminoazobenzene in bacterial medium by acetylation. These recombinant acetyltransferases were analyzed for: (i) N-acetyltransferase, which was assayed by the rate of acetyl coenzyme A-dependent N-acetylation of 2-aminofluorene (2-AF) or 4-aminoazobenzene (AAB); (ii) arylhydroxamic acid acyltransferase, assayed by N,O-acyltransfer with N-hydroxy-N-acetyl-2-aminofluorene. Both NAT2s showed first order increases in N-acetylation rates with increasing 2-AF or AAB concentrations between 5 and 100 microM, with apparent K(m) values of 22-32 and 62-138 microM respectively. Although under the same conditions the N-acetylation rates for the two NAT1s declined by > 50%, below 5 microM 2-AF or AAB, the NAT rate data fit Michaelis-Menten kinetics, and the apparent K(m) values were 0.2-0.9 microM. For N,O-acyltransferase, the apparent K(m) values of the NAT1s were approximately 6 microM, while the K(m) values of the NAT2s were approximately 20- to 70-fold higher. SDS-PAGE/Western blot analysis of the recombinant acetyltransferases gave apparent relative molecular weights (MWr) of approximately 31 kDa for both NAT1s and rat NAT2 and approximately 29 kDa for hamster NAT2. Comparable MWr values were observed for native hamster liver NAT1 and NAT2 and for rat NAT1 under the same conditions. Although we did not detect NAT2-like activity in rat liver cytosol previously, the present data show that the rat NAT2* gene does code for a functional acetyltransferase, with properties similar to those of hamster liver NAT2. The data also indicate that at low substrate concentrations, NAT1 would apparently play the predominant role in vivo in N-acetylation and N,O-acyltransfer of aromatic amine derivatives, including their metabolic activation to DNA-reactive agents.