Most amino acid substitutions in the central domain of the p53 protein impart increased tumorigenic potential, somatic selective advantage, to the host mutant somatic cell lineage. If this tumorigenic potential varies considerably from one codon to the next, then the p53 mutational spectrum in the tumor data base will be selection driven. If the tumorigenic potential varies little from one codon to the next, then the p53 mutation spectrum in the tumor data base will be mutagenesis driven and correlated with patterns of mutagen accessibility and lesion repair rates along the p53 gene. To test the relative contributions of these two effects, we will prepare three databases, one from the tumor database (selection bias for tumorigenicity), one of base substitutions fixed during vertebrate evolution of the active p53 gene (selection bias against tumorigenicity) and one of base substitutions which were fixed in p53 pseudogenes (unbiased selection). To measure the effect of selection on the pattern of base substitutions sampled into these three p53 databases, we will apply two novel selection-sensitive analytical tools. These tools are based on properties of the genetic code and measure the average magnitude of the protein perturbation imparted by a base change. Preliminary results indicate that the codon-to-codon variations of mutation frequency in the p53 tumor database are mutagenesis driven. They also show that for mCpG->TpG transitions accumulated during vertebrate evolution, the putative the mC->G transition usually involves the C on the transcribed strand rather than the one on the nontranscribed strand because the former situation on the average effects a less drastic protein perturbation.
|Rodin, Sergei N; Rodin, Andrew S (2002) On the origin of p53 G:C --> T:A transversions in lung cancers. Mutat Res 508:1-19|
|Rodin, S N; Rodin, A S (2000) Human lung cancer and p53: the interplay between mutagenesis and selection. Proc Natl Acad Sci U S A 97:12244-9|
|Rodin, S N; Rodin, A S (1998) Strand asymmetry of CpG transitions as indicator of G1 phase-dependent origin of multiple tumorigenic p53 mutations in stem cells. Proc Natl Acad Sci U S A 95:11927-32|