This study was designed to evaluate the potential initiating effects of transplacental 3'-azido-2',3'-dideoxythymine (AZT) and the role of ras mutational activation in skin tumors induced in a two-stage mouse skin model. In addition, mouse liver and lung tumors from a transplacental AZT tumorigenicity study reported elsewhere (Olivero et al., J Natl Cancer Inst 89:1602-1608, 1997) were examined for evidence of ras activation. For both tumor studies, pregnant CD-1 mice were given either vehicle or 25 mg of AZT daily on days 12-18 of gestation. In the 1997 study, the offspring were given no further exposure and were killed at 1 yr of age. For the skin tumor study, all mice received twice-weekly topical 12-O-tetradecanoyl-phorbol-13-acetate (TPA) treatment from weeks 5-35; half of the mice had been exposed to AZT in utero. At weeks 16-18, 30, 31, and 34-41, the skin tumor incidences in mice given AZT and TPA were significantly higher than in mice given TPA alone (P < or = 0.05). At week 41, the average numbers of tumors per mouse were 1.44+/-0.36 (mean +/- standard error of the mean) and 0.57+/-0.13 for mice given AZT plus TPA and TPA alone, respectively (P = 0.006). Mutagenesis in ras exons I and II was determined by polymerase chain reaction (PCR) and dye-terminator cycling sequencing of PCR products. Ha-ras exon I codons 12 and 13 were mutated in 11 of 19 tumors (58%) from mice given AZT and TPA and in one of 15 tumors (7%) from mice given TPA alone (P= 0.004). The only mutation in Ha-ras codon 12 (four in four tumors examined) was a G-->A transition in the second base, and the major mutation in codon 13 (six in seven tumors examined) was a G-->T transversion in the second base. In skin tumors, AZT exposure did not increase the number of Ha-ras codon 61 mutations, and no Ki-ras mutations were observed. Analysis of ras mutations in liver and lung tumors from mice exposed to AZT in utero (Olivero et al., J Natl Cancer Inst 89:16021608, 1997) with no TPA promotion showed no significant AZT-related increases.