br Materials and methods br Results

Materials and methods
Results
Discussion DNA repair pathways have evolved for a long time to act independently of one another. However, over the past decade several overlaps and crosstalks between these pathways were identified, showing that the DNA repair system is more complex than previously thought (Veen and Tang, 2015). Nagorska et al. (2012) demonstrated the existence of an interaction network between Kifunensine involved in oxidative damage repair in Neisseria meningitidis and found that the MutM and MutY enzymes exhibit overlapping activities and functional redundancies, features that may reflect the adaptive success of this species. A recent work from our group has shown that Corynebacterium pseudotuberculosis has two mutM in its genome (Arantes et al., 2016). Curiously, one of them (MutM1) was able to recognize and excise the 8-oxoguanine lesion from an oligonucleotide while the other (MutM2) did not present specific DNA glycosylase activity in this substrate. Therefore we have proposed that CpMutM2 could act in the GO system recognizing and repairing other sorts of oxidative lesions in the DNA. The enzymes involved in oxidative lesion recognition (MutM, MutY, and MutT) were found in the C. pseudotuberculosis genome and, interestingly, these genes are conserved in genomes of both pathogenic and non-pathogenic Corynebacterium species, revealing the importance of this system for organisms subjected to oxidative pressure as hosts that during colonization must survive the blistering attack of the host immune system, even as for free-living organisms (Resende et al., 2011). Apart from structural similarities and preservation of the same domains and motifs, functional similarities between MutY from C. pseudotuberculosis and E. coli were also observed by performing heterologous functional complementation assays (Fig. 2, Fig. 3). CpMutY was able to complement the hypermutator phenotype of mutY-gene-deficient E. coli and its mutation rate returned to wild-type levels. Similar results were observed in the functional characterization of MutY from other pathogens, such as P. gingivalis (Robles et al., 2011), H. pylori (Eutsey et al., 2007, Huang et al., 2006), Neisseria sp. (Davidsen et al., 2005), Pseudomonas aeruginosa (Oliver et al., 2002), and the free-living bacterium Deinococcus radiodurans (Li and Lu, 2001), all of which were also capable of complementing the MutY-mutant E. coli phenotype. In some of these bacteria, deficiency in the mutY gene also increased the rate of spontaneous mutation and enhanced sensitivity to hydrogen peroxide (Robles et al., 2011, Sanders et al., 2009). Complementation assay using E.coli as model is a well established method widely used in heterologous-genes functional studies. This strategy is important when the chosen organism does not allow a direct mutant production, as observed in organisms that presented several genes copies of or even due to the difficulty in cultivating pathogenic organism (Choi et al., 2016, Verissimo-Villela et al., 2016). The ability of CpMutY to suppress the mutator phenotype of E. coli (mutY), whose ability to repair 8-oxoG:A mismatches is impaired, suggests functional conservation of CpMutY. Under oxidative stress conditions, we verified that CpMutY conferred a significant advantage for mutY-gene-deficient E. coli to grow 180 and 270min after H2O2 exposure that corresponding to 4 and 5.5h after IPTG induction (Fig. 3). Our findings suggest that the CpMutY enzyme may be engaged in preventing incorporation of damage caused by products of H2O2 decomposition, thus precluding the cytotoxic effect observed in cells lacking the mutY gene. Robles et al. (2011) observed that sensitivity to H2O2 is significantly increased in P. gingivalis cells deficient in the mutY gene and mutations in this gene have also been found to increase the sensitivity of P. aeruginosa to H2O2, ultimately leading to cell death (Sanders et al., 2009). After cell exposure to oxidative agents, the AP site produced by DNA glycosylase activity is mutagenic and citotoxic to cells, however some studies have evidenced that E.coli MutY protein remained tightly bound to the AP site, produced after 8-oxoG:A removal. This action most likely occurs to avoid the deleterious effect of AP site and to promote the recruitment of the specific enzymes involved in this DNA repair pathway, avoiding the cell death, as observed in our in vivo experiments (Oliveira et al., 2014).