Abstract
The excision‐repair of damaged DNA in bacteria and other systems probably requires at least three enzymes to carry out the following steps in sequence: (1) Recognition of a structural distortion in the DNA and the production of an endonucleolytic cleavage of the damaged strand near the lesion. (2) The simultaneous peeling back of the damaged strand and resynthesis of the excised region, with eventual cleavage of the damaged segment from the DNA. (3) The rejoining of the newly synthesized strand to contiguous parental DNA. Evidence for all three steps has been obtained from in vivo studies. The E. coli DNA polymerase has been shown to carry out step # 2 in vitro [1] and the polynucleotide ligase has the required specificity for step # 3[2–4]. An enzyme responsible for step # 1 has been purified from Micrococcus lysodeikticus [5,6] but not from E. coli, although a class of u.v. sensitive mutants in E. coli has been shown to be defective in this step in the repair sequence. In such mutants the release of pyrimidine dimers from the damaged DNA is not observed during post‐irradiation growth of u.v. irradiated cultures [7]. It would be predicted, as a consequence, that the next step, non‐conservative repair replication, would not be seen in these mutants. Hanawalt and Petti‐john showed this to be true for the double mutant E. coli B8‐1 that includes a deficiency in dimer excision [8]. In the present study we have looked more closely at an E. coli K‐12 strain that has only the uvrA6 deficiency that results in inability to excise pyrimidine dimers.