Dynamic flexibility of DNA repair pathways in growth arrested Escherichia coli

Abstract

The DNA of all organisms is constantly damaged by exogenous and endogenous agents. Base excision repair (BER) is important for the removal of several non-bulky lesions from the DNA, however not much is known about the contributions of other DNA repair pathways to the processing of non-bulky lesions. Here we utilized a luciferase reporter system to assess the contributions of transcription-coupled repair (TCR), BER and nucleotide excision repair (NER) to the repair of two non-bulky lesions, 8-oxoguanine (8OG) and uracil (U), in vivo under non-growth conditions. We demonstrate that both TCR and NER are utilized by Escherichia coli to repair 8OG and U. Additionally, the relative level of recognition of these lesions by BER and NER suggests that TCR can utilize components of either pathway for lesion removal, depending upon their availability. These findings indicate a dynamic flexibility of DNA repair pathways in the removal of non-bulky DNA lesions in prokaryotes, and reveal their respective contributions to the repair of 8OG and U in vivo.

Fig. 1. DNA Repair Pathways Involved in Removal of Non-Bulky Lesions from DNA in vivo

WT and DNA repair mutant strains (Table 2) were transformed with damage-containing construct, and the level of TM (measured via luciferase activity and expressed as relative light units per 10⁶ cells) was measured at 0, 45, and 120 minutes following IPTG induction using TM-LAS methodology (Supplementary Fig. 1). Each data point represents the mean of at least three replicates ± SEM. Statistical significance was calculated using a Student’s t-test. Distributions were considered to be significantly different when p < 0.05. (A) U/Stop construct. * Denotes points at which the single mutant, ung, is significantly different from the double mutant ung mfd. All points for ung and ung mfd are significantly different from the wild type and mfd strains. (B) U/Stop construct. All points for ung and ung uvrA(kan) are significantly different from the wild type strain. At t=45 and 120 minutes uvrA(kan) is significantly different from the wild type strain. (C) U/Stop construct. * Denotes points at which the single mutant, mfd, is significantly different from the double mutant uvrA(Tn10) mfd. All points for uvrA(Tn10) and uvrA(Tn10) mfd are significantly different from the wild type strain, but do not significantly differ from each other at any point. (D) 8OG/Stop construct. *Denotes points at which the single mutant, mutM is significantly different from the double mutant mutM uvrA(kan). All points after t=0 for mutM and mutM uvrA(kan) are significantly different from the wild type strain. At t=45 the uvrA(kan) strain is significantly different from the wild type strain. (E) 8OG/Stop construct. * Denotes points at which the single mutant, uvrA(Tn10), is significantly different from the double mutant uvrA(Tn10) mfd. All points after t=0 minutes for uvrA(Tn10), mfd, and uvrA(Tn10) mfd are significantly different from the wild type strain.

Fig. 2. Dynamic Flexibility of Interacting DNA Repair Pathways in E. coli

Non-bulky lesions in DNA (small green box) are primarily repaired by the BER pathway. When these lesions are present on the template strand of a gene and are encountered by RNAP, at least two possible outcomes can result: 1. TM via insertion of an incorrect ribonucleotide opposite the lesion on the nascent mRNA (red arrow), or 2. Mfd-mediated TCR of the lesion (green arrows). Mfd removes RNAP from the lesion and subsequently acts to block incoming RNAP from bypassing the lesion in a mutagenic manner. Mfd leaves the damaged base accessible to BER. However, if the capacity of BER is overwhelmed by excessive DNA damage elsewhere, Mfd can recruit NER to repair the lesion, thus complementing the reduced BER capacity for the repair of small lesions.