Fidelity of translesional synthesis past benzo[a]pyrene diol epoxide-2′-deoxyguanosine DNA adducts: Marked effects of host cell, sequence context, and chirality

We have used a site-specific approach to investigate the mutagenic potential of (+)- and (-)-trans-anti-benzo[a]pyrene diol epoxide (BPDE) DNA adducts. Oligodeoxyribonucleotides (^5′TCCTCCTG₁G₂-CCTCTC), modified at the exocyclic amino groups of G₁ or G₂, were incorporated into a single-stranded shuttle vector and introduced into Escherichia coli or simian kidney (COS) cells. This experimental system permits translesional synthesis to proceed in the absence of DNA repair. The presence of (+)- or (-)-BPDE-N²-dG adducts strongly inhibited translesional synthesis in E. coli; induction of cellular SOS functions reduced this blocking effect. Vectors containing (+)-BPDE adducts at G₁ or G₂ generated mutation frequencies of 19% and 3%, respectively; these values were not altered significantly by induction of SOS functions. In COS cells, (+)-BPDE-modified vectors generated mutation frequencies of 13% at G₁ and 45% at G₂. In E. coli, the (-)-BPDE adduct generated mutation frequencies of ≤2% at G₁ and G₂ and, in COS cells, 13% at G₁ and 21% at G₂. The predominant mutations in E. coli and COS cells were G→T transversions targeted to the site of the lesion; however, when G₂ was modified, a significant number of targeted G→A and G→C mutations were observed in COS cells. We conclude from this study that (+)- and (-)-BPDE-N²-dG adducts pair preferentially to dCMP and dAMP during translesional synthesis in a process that is strongly influenced by the stereochemistry of the adduct, by the bases flanking the lesion, and by host cell factors.