p53 mutations are common in lung cancer. In smoking-associated lung cancer,the occurrence of G:C to T:A transversions at hotspot codons, e.g., 157, 248, 249,and 273, has been linked to the presence of carcinogenic chemicalsin tobacco smoke including polycyclic aromatic hydrocarbons suchas benzo(a)pyrene (BP). In the present study, we have used a highly sensitive mutation assay to determine the p53 mutation load in nontumorous human lung and to study the mutability of p53 codons 157, 248, 249, and 250 to benzo(a)pyrene-diol-epoxide (BPDE), an active metabolite of BP in human bronchial epithelial BEAS-2B cells. We determined the p53 mutational load at codons 157, 248, 249, and 250 in nontumorous peripheral lung tissue either from lung cancer cases among smokers or noncancer controls among smokers and nonsmokers. A 5-25-fold higher frequency of GTC(val) to TTC(phe) transversions at codon 157 was found in nontumorous samples (57%) from cancer cases (n = 14) when compared with noncancer controls (n = 8; P < 0.01). Fifty percent (7/14) of the nontumorous samples from lung cancer cases showed a high frequency of codon 249 AGG(arg) to AGT(ser) mutations (P < 0.02). Four of these seven samples with AGT(ser) mutations also showed a high frequency of codon 249 AGG(arg) to ATG(met) mutations, whereas only one sample showed a codon 250 CCC to ACC transversion. Tumor tissue from these lung cancer cases (38%) contained p53 mutations but were different from the above mutations found in the nontumorous pair. Noncancer control samples from smokers or nonsmokers did not contain any detectable mutations at codons 248, 249, or 250. BEAS-2B bronchial epithelial cells exposed to doses of 0.125, 0.5, and 1.0 microM BPDE, showed G:C to T:A transversions at codon 157 at a frequency of 3.5 x 10(-7), 4.4 x 10(-7), and 8.9 x 10(-7), respectively. No mutations at codon 157 were found in the DMSO-treated controls. These doses of BPDE induced higher frequencies, ranging from 4-12-fold, of G:C to T:A transversions at codon 248, G:C to T:A transversions and G:C to A:T transitions at codon 249, and C:G to T:A transitions at codon 250 when compared with the DMSO-treated controls. These data are consistent with the hypothesis that chemical carcinogens such as BP in cigarette smoke cause G:C to T:A transversions at p53 codons 157, 248, and 249 and that nontumorous lung tissues from smokers with lung cancer carry a high p53 mutational load at these codons.