The high mutator activity of the dnaQ49 allele of Escherichia coli is medium-dependent and results from both defective 3′→5′ proofreading and methyl-directed mismatch repair

Abstract

The Escherichia coli dnaQ49 mutator allele maps at the dnaQ locus, the structural gene for the epsilon subunit of the DNA polymerase III holoenzyme. Epsilon, when bound to the alpha subunit, provides the 3'-->5' exonuclease activity (proofreading) that removes 3' mismatched terminal nucleotides from the nascent DNA strand during replication. The temperature sensitive dnaQ49 allele lacks this catalytic activity which results in mutation frequencies 10(4)-10(5)-fold above wild-type values at 37 degrees C. At 30 degrees C dnaQ49 mutation frequencies are much lower but still higher than wild-type levels. We found that dnaQ49, like mutD5, another strong mutator allele of dnaQ, is medium-dependent with mutation frequencies ranging from 12 to nearly 1000-fold higher in rich media (L-broth) than in minimal media. In minimal media dnaQ49 retains modest mutator activity. In addition the base-pair substitution mutational spectrum of dnaQ49 was medium-dependent. Unlike mutD5 the addition of thymidine to minimal medium did not enhance dnaQ49 mutator activity. We also constructed dnaQ49mutL double mutator strains and compared mutator frequencies with single dnaQ49 and mutL strains. The mutL allele results in inactive methyl-directed mismatch repair. Double and single dnaQ49 mutators had similar mutation frequencies at 37 degrees C in L-broth suggesting that dnaQ49 strains are defective in mismatch repair as well as 3'-->5' exonuclease proofreading activity. In contrast in minimal media at 37 degrees C and in L-broth at 30 degrees C dnaQ49 mutL mutation frequencies were much higher than dnaQ49 values indicating the presence of active mismatch-repair activity in the latter strain. In addition at 37 degrees C dnaQ49mutL mutation frequencies were about 100-fold higher in L-broth than in minimal media. We conclude from this result that the rich media effect with dnaQ49 involves an actual increase in replication errors rather than a medium-dependent modulation of mismatch repair activity.