Formation of diastereomeric benzo[a]pyrene diol epoxide-guanine adducts in p53 gene-derived DNA sequences.

Abstract

G --> T transversion mutations in the p53 tumor suppressor gene are characteristic of smoking-related lung tumors, suggesting that these genetic changes may result from exposure to tobacco carcinogens. It has been previously demonstrated that the diol epoxide metabolites of bay region polycyclic aromatic hydrocarbons present in tobacco smoke, e.g., benzo[a]pyrene diol epoxide (BPDE), preferentially bind to the most frequently mutated guanine nucleotides within p53 codons 157, 158, 248, and 273 [Denissenko, M. F., Pao, A., Tang, M., and Pfeifer, G. P. (1996) Science 274, 430-432]. However, the methodology used in that work (ligation-mediated polymerase chain reaction in combination with the UvrABC endonuclease incision assay) cannot establish the chemical structures and stereochemical identities of BPDE-guanine lesions. In the present study, we employ a stable isotope-labeling HPLC-MS/MS approach [Tretyakova, N., Matter, B., Jones, R., and Shallop, A. (2002) Biochemistry 41, 9535-9544] to analyze the formation of diastereomeric N(2)-BPDE-dG lesions within double-stranded oligodeoxynucleotides representing p53 lung cancer mutational hotspots and their surrounding DNA sequences. (15)N-labeled dG was placed at defined positions within DNA duplexes containing 5-methylcytosine at all physiologically methylated sites, followed by (+/-)-anti-BPDE treatment and enzymatic hydrolysis of the adducted DNA to 2'-deoxynucleosides. Capillary HPLC-ESI(+)-MS/MS was used to establish the amounts of (-)-trans-N(2)-BPDE-dG, (+)-cis-N(2)-BPDE-dG, (-)-cis-N(2)-BPDE-dG, and (+)-trans-N(2)-BPDE-dG originating from the (15)N-labeled bases. We found that all four N(2)-BPDE-dG diastereomers were formed preferentially at the methylated CG dinucleotides, including the frequently mutated p53 codons 157, 158, 245, 248, and 273. The contributions of individual diastereomers to the total adducts number at a given site varied between 70.8 and 92.9% for (+)-trans-N(2)-BPDE-dG, 5.6 and 16.7% for (-)-trans-N(2)-BPDE-dG, 2.1 and 8.5% for (-)-cis-N(2)-BPDE-dG, and 0.5 and 8.3% for (+)-cis-N(2)-BPDE-dG. The relative yields of the minor N(2)-BPDE-dG stereoisomers were elevated at the sites of inefficient adduction, while the major (+)-trans-BPDE lesion was even more dominant at the frequently adducted sites. The introduction of 5-methyl groups at adjacent cytosine bases increased the yields of N(2)-BPDE-dG diastereomers, probably a result of favorable hydrophobic interactions between BPDE and 5-methylcytosine. The targeted formation of N(2)-BPDE-dG at (Me)CG dinucleotides within the p53 gene is consistent with the high prevalence of G --> T transversions at these sites in smoking-induced lung cancer.