Base selection, proofreading, and mismatch repair during DNA replication in Escherichia coli.

The accuracy by which organisms duplicate their DNA is of considerable interest. At least three mechanisms operate, serially, to secure high fidelity: base selection, exonucleolytic proofreading, and postreplicative mismatch correction. To obtain insights into the efficiency and specificity of these steps in the bacterium Escherichia coli, we have performed DNA sequence analysis of mutations occurring in the bacterial lacI gene in a series of strains genetically disabled in one or more of these error avoidance pathways. The base selection efficiency was estimated from mutagenesis occurring in a mutDmutL strain, which is deficient in both proofreading (mutD5) and mismatch repair (mutL). The proofreading efficiency was derived comparing the mutD5 mutL strain to the mismatch repair-deficient mutL strain. The efficiency of mismatch repair was derived comparing the mutL strain to the wild-type strain. The results show that base selection discriminates against errors by 200,000-2,000,000-fold, proofreading by 40-200-fold, and mismatch repair by 20-400-fold, each depending on the type of error. Base selection and proofreading act more strongly against transversions than transitions, whereas mismatch repair does the opposite. The data are based on 866 sequenced lacI mutations in a target that allows the scoring of at least 127 different mutations in 76 distinct DNA sequence contexts in vivo. They may therefore have general significance.