Smoking, lung cancers and their TP53 mutations.

Abstract

The TP53 suppressor gene possesses several properties that have facilitated its use as a reporter of genotoxic exposure (Biggs et al., 1993). In particular, it is mutated in a high proportion of many types of human cancer and it sustains a broad spectrum of mutations that can vary in both their position and type. In lung tumours from smokers, the high frequency of G→T transversions (30% compared with 10% in cancers without tobacco aetiology) has been attributed to DNA adducts of polycyclic aromatic hydrocarbons (PAHs), such as benzo[a]pyrene, that are present in tobacco smoke (Hainaut and Pfeifer, 2001). Benzo[a]pyrene is activated to form benzo[a]pyrene diol epoxide (BPDE), which reacts with DNA predominantly at the N2-position of guanine, and both in vitro and in vivo studies have demonstrated that BP and BPDE can, in common with other PAHs, induce G→T transversions (Wei et al., 1991; DeMarini et al., 1995; Schiltz et al., 1999; Wijnhoven et al., 2000). A single point mutation in the TP53 gene results in a base change in both DNA strands with, for example, a G→T mutation in one strand corresponding to a C→A change in the complementary strand (G:C→T:A). The question then arises whether such a change was induced by modification of G in one strand or of the C in the complementing strand. Recently, Rodin and Rodin (2000) examined the spectra of mutations that they believed were induced by damage of each of the two DNA strands of TP53 and concluded that when each strand was examined individually, there were no significant differences between the spectra in lung adenocarcinomas in smokers and non-smokers. The plot of mutations that represents the core of their argument and around which other arguments and conclusions in their paper are based is shown in Figure 1A, where all 12 possible base pair substitution mutations are grouped into six pairs, each consisting of two complementary substitutions. In this figure the mutations above the line represent the changes presumed to be induced by damage in the non-transcribed strand (NTS) of TP53 whereas those below the line represent the same changes induced in the transcribed strand (TS). For example, on the left-hand side, the G→T above the line represents a mutation induced in the NTS, whereas C→A below the line represents the G→T mutation induced in the TS. (Note that by convention all mutations are referred to by the mutation found in the NTS.) The data as presented show: (i) that above the central line the ratio of the mutation frequency in smokers to that of the same mutations in non-smokers remains approximately constant; (ii) that a similar relationship, albeit with a different value of the ratio,