Targeting DNA repair pathways for cancer therapy

Accumulation of genomic mutations is the consequence of failure in DNA repair as well as increased exposure to endogenous/environmental mutagens. DNA repair pathways safeguard the human genome from such mutagens, and thereby suppress the multi-step process of carcinogenesis. DNA repair pathways that protect the genome from ROS (reactive oxygen species)-induced lesions are attractive anti-cancer targets, as their inhibition may render combinatorial sensitization of tumor cells to both DNA damage and oxidative stresses, known as non-oncogenic addictions of cancer. The aim of this thesis was to validate such DNA repair factors as anti-cancer targets and to develop their inhibitors for potential therapeutic applications. In paper I, we assessed the addiction of cancer cells to MTH1, a nudix hydrolase eliminating oxidized purine nucleotides from the dNTP pool. MTH1 depletion resulted in exclusive accumulation of 8-oxo-dG lesions and cellular toxicity in transformed cells. MTH1 suppression, impaired tumor growth in the xenografts of SW480 cells. We developed potent MTH1 inhibitors (TH278 and TH588), which exhibited target engagement and selective toxicity in transformed cells. Treatment with MTH1 inhibitors caused increased 8-oxo-dG levels in cancer cells, and inhibited the growth of xenografts in vivo. Taken together, our findings revealed the dependency of tumors to MTH1 that can be targeted for cancer therapy. The study in paper II aimed to explore functional cooperation between MTH1 and MUTYH, a DNA glycosylase that removes deoxyadenines paired with 8-oxo-dG. Using stable cell lines expressing inducible shRNA constructs, we showed that combined depletion of MTH1 and MUTYH was more toxic to cells compared to individual knock-downs. In addition, overexpression of nuclear MUTYH could attenuate cell death induced by loss of MTH1. Collectively, this study provided supportive evidence for a protective role of MUTYH. In paper III, we described TH5487 as a novel selective inhibitor of OGG1, a DNA glycosylase that excises 8-oxo-dG opposite deoxycytidine. TH5487 inhibited binding of OGG1 to its substrate and increased thermal stability of the purified protein through interactions with residues in the active site. Moreover, TH5487 engaged with its intended target, increased 8-oxo-dG level, and impaired recruitment of OGG1 to the damage site in cells. Treatment with TH5487 resulted in prolonged S phase, which was similar to the effect of OGG1 depletion using shRNAs. In addition, non-transformed cells could tolerate TH5487 treatment while cancer cells were more sensitive. In sum, this study highlighted the phenotypic lethality of OGG1 inhibition with tumors, by introducing TH5487 as a cell-active OGG1 inhibitor. Overall, our results increased the knowledge about dependency of cancer cells to DNA repair pathways of ROS-induced lesions that can be employed for the development of promising anti-tumor therapies