A high priority shipment: How nuclease delivery by SLX4 mediates DNA interstrand crosslink repair

Abstract

DNA interstrand crosslinks (ICLs) are lesions on the DNA that covalently attach both strands of the double helix. ICLs are very toxic to cells since important cellular processes such as DNA replication and transcription depend on separation of the two strands. Although knowledge about endogenous crosslinking agents is currently unfolding, the toxic nature of ICL inducing agents has been exploited in anti-cancer therapies for several decades. Similarly, our understanding of how cells repair ICL damage has long been elusive while significant advances have been made in recent years. Important roles are ascribed to the ‘FA proteins’ that are mutated in patients that have the rare genetic disorder Fanconi anemia (FA). This cancer susceptibility syndrome has been linked to mutations in 22 different genes, the products of which cooperate with several other DNA repair proteins to coordinate ICL repair in the FA pathway. This pathway is initiated when DNA replication forks are blocked by the ICL, resulting in monoubiquitylation of FANCD2 that marks activation of downstream repair processes. A critical step in this process is the nucleolytic ‘unhooking’ of the ICL from one of the strands that allows replication to complete on the other strand. This requires at least the scaffold protein SLX4(FANCP) and an incision by the nuclease complex XPF(FANCQ)-ERCC1. The exact mechanism of this ICL unhooking step is explored in this thesis. To do this, we employed a unique model system of extracts from Xenopus laevis eggs. This cell-free system supports replication of plasmid DNA as well as the repair of a site-specific ICL on that plasmid, thereby recapitulating the repair process. Much of our current understanding of ICL repair stems from studies that used this system. We purified full length proteins from insect cells and added them to SLX4- or XPF-depleted extracts to study the effect of mutations on ICL repair. We investigated why specific mutations in XPF cause FA while other mutations affect its role in the repair of bulky adducts and identify an FA-causing mutation that disrupts interaction with SLX4. Since SLX4 binds several other proteins in addition to XPF-ERCC1, including two nucleases, we investigated the requirement of these interactions for ICL repair. We found that two-third of the protein is dispensable, which includes the binding sites for the other nucleases, while interaction with XPF-ERCC1 is crucial for ICL repair. Recruitment of SLX4 to ICL sites is associated with tandem ubiquitin-binding (UBZ) domains but details are lacking. We show that SLX4 recruitment and ICL repair strongly depend on the first, but not the second, UBZ domain. Our results disfavor a direct interaction between SLX4 and FANCD2, and suggest the existence of an unidentified factor that mediates SLX4 recruitment to the lesion. Collectively, the findings presented in this thesis contribute to our understanding of the molecular mechanism of ICL repair. Detailed knowledge of ICL repair is useful for the development of improved anti-cancer therapies and to fight resistance to existing therapies.