Role of ATM (ataxia telangiectasia mutated) and Artemis proteins for the repair of DNA double-strand breaks by homologous recombination in mammalian cells

Abstract

DNA double-strand breaks (DSB) represent the most dangerous type of DNA damage which, if unrepaired, can lead to cell death. Mis-repaired damage can result in genomic instability, deletions, translocations and mutations, and further to the inactivation of tumor suppressor or activation of proto-oncogenes, all of which can drive carcinogenesis. DSBs can not only arise exogenously by mutagenic chemicals or ionizing irradiation (IR), they can also occur endogenously as by-products of normal oxidative metabolism (reactive oxygen species). Well conserved repair mechanisms for DSBs are present in all living organisms. In mammalian cells, DSBs are mostly repaired by two fundamentally different processes, non-homologous end-joining (NHEJ), which can act throughout the cell cycle, or homologous recombination (HR), which relies on the presence of a homologous sequence (e.g. sister chromatid) and is therefore restricted to the late S- or G2-phase. Inherited or somatic mutations in any of the key proteins involved in DSB repair generally predispose to malignancy. Understanding the involvement and actions of DNA repair proteins and processes can identify new or improve existing strategies in cancer therapy. This work focuses on two proteins, namely ATM (ataxia telangiectasia mutated) and Artemis. Defects in either the ATM or the Artemis gene lead to pronounced sensitivity to IR, which has previously been ascribed to a common defect in NHEJ. However, it is not known whether depletion of the two proteins affects HR in the same way, since cell cycle progression is markedly different in both deficient cells. The aim of this thesis was to understand the functions of the ATM and Artemis proteins in DNA DSB repair. Three main questions were addressed: (1) How does cell cycle progression influence DNA repair? (2) Are ATM and Artemis involved in HR during the G2- and S-phase? (3) Do they function epistatically for HR during the G2- and also the S-phase? Human fibroblast lines defective in either ATM (AT) or Artemis were studied in addition to a wild-type (WT) line. DNA damage was monitored by immunohistochemistry, detecting nuclear focus formation of either gH2AX as a general marker for DSBs or Rad51 as a marker for recombination activity using immunofluorecent microscopy. Cell cycle distribution by flow cytometry analysis and differential staining of S- (5-ethynyl-2’-deoxyuridine, EdU) and G2-phase cells (CenpF) revealed a robust IR-induced G1 block in Artemis cells, while AT cells migrated through the S-phase and accumulated in the G2-phase before mitosis. In these G2-phase AT cells, all DSBs were additionally decorated with the Rad51 protein, indicating recombination activity. By contrast, in Artemis cells, only 60% of gH2AX foci were also Rad51-positive, hinting at differences in HR. Using a chromosomal reporter construct designed to specifically monitor HR, both Artemis depletion (siRNA) and ATM-inhibition (KU55933) lead to substantial HR defects (ATM>Artemis) hinting again at differences in HR. Corresponding protein expression and phosphorylation was controlled using Western blot analysis. Immunohistochemistry analysis of recombination activity specifically in G2- and S-phase cells showed similar defects in Rad51 focus formation in the G2-phase and no evidence of repair by HR in either of the deficient strains. Surprisingly, only ATM but not Artemis is required for the HR of radiation-induced DSBs during the S-phase. In contrast to WT and Artemis fibroblasts, numerous Rad51 foci form continuously in AT cells after irradiation, indicating a recruitment process that is independent of ATM-mediated functions such as the resection of DSB ends. The Rad51 recruitment to DSBs, however, needs functional ATR/Chk1. ATR activation may occur when a progressing replication fork encounters radiation-induced single-strand damage. Despite successful initiation of recombination (recruitment of Rad51 recombinase), HR repair cannot be completed without ATM. Abrogation of ATM function in Artemis cells further reduced their survival, but only in those cells that actively replicated in the S-phase. In conclusion, we describe important differences in HR between AT and Artemis cells during the S-phase, but a common recombination defect in the G2-phase. We have identified ATM as a core component in the HR of directly and indirectly-induced DSBs downstream of DNA end resection and Rad51 filament formation processes, thus introducing new possibilities for cancer therapies in tumors with compromised expression of the ATM protein.