On the fidelity of DNA replication. Accuracy of Escherichia coli DNA polymerase

Abstract

The fidelity with which Escherichia coli DNA polymerase I copies poly[d(A-T)] has been studied. The poly[d(A-T)] template was synthesized by E. coli DNA polymerase I in a de nouo reaction using dTTP and dATP in the presence of radioactive dGTP. The poly[d(A-T)] product synthesized in the de nouo reaction contained 1± 0.5 mol of dGMP for every 2 × l0⁶ mol of dAMP and dTMP polymerized. In copying this template, the apparent error rate of E. coli DNA polymerase I is 1 in 80,000 with dGTP and 1 in 8,000 with dCTP as the noncomplementary nucleotide. The product of the reaction with dGTP or dCTP as the noncomplementary nucleotide was analyzed by enzymatic hydrolysis and subsequent chromatography of the nucleotide monomers. With dGTP, the noncomplementary nucleotide can be identified in the poly[d(A-T)] product as dGMP and is invariably incorporated as a single base substitution in place of dAMP. In contrast, with [³H]dCTP, the majority of the incorporated label was identified as dUMP or dTMP suggesting that the high error rate may represent the incorporation of a contaminant. This misincorporation of dGMP requires Mg²⁺, poly[d(A-T)], polymerase, and d'lTP. Omission of dATP results in an enhanced incorporation of dGMP. The frequency of dGTP incorporation into poly[d(A-T)] was markedly reduced by increasing the concentration of dATP but not dTTP. A similar substrate competition is observed with the DNA polymerase from avian myeloblastosis virus with poly[d(A-T)] as a template. However, with both enzymes, there is a lack of substrate competition with homopolymer templates. Thus, the lack of substrate competition in copying homopolymers is determined by the nature of the template and does not provide evidence for multiple catalytic nucleotide binding sites.