Summary DNA interstrand cross-links (ICLs) covalently link the two strands of the DNA double helix and are extremely cytotoxic. Widely used chemotherapeutics (e.g. nitrogen mustards and cisplatin compounds) act by generating ICLs, but tumors almost invariably become resistant to these agents, often by upregulating repair. ICLs are also created by endogenous metabolites (e.g. reactive aldehydes, abasic sites), and failure to repair endogenous ICLs causes human disease. Thus, mutation of 22 different ‘FANC’ genes renders cells sensitive to ICLs and causes Fanconi anemia (FA), a human disease characterized by bone marrow failure and predisposition to leukemia and other cancers. ICL repair is coupled to transcription and DNA replication, but how these processes promote repair is poorly understood. To elucidate mechanisms of ICL repair, we replicate plasmids containing site-specific ICLs in frog egg extract, which allows us to resolve DNA repair intermediates at high resolution. We test the hypotheses generated in extracts in mammalian cells. We showed previously that ICL repair in egg extracts is initiated when replication forks converge on the lesion. Fork convergence triggers replicative CMG helicase ubiquitylation by the E3 ubiquitin ligase TRAIP, which promotes CMG unloading by the p97 ATPase and initiation of the FA ICL repair pathway. Although TRAIP associates constitutively with replisomes, it only ubiquitylates CMGs when they have converged at an ICL (“trans” ubiquitylation); it does not ubiquitylate single CMGs before convergence (“cis” ubiquitylation). This injunction against cis ubiquitylation is critical to avoid premature replisome disassembly and fork collapse but its mechanistic basis is unknown. In Aim 1, we test the hypothesis that TRAIP’s interaction with CDC45 at the replisome constrains TRAIP’s RING domain so that it only ubiquitylates proteins in trans. Once CMG is unloaded, the FANCI-FANCD2 dimer recruits a complex of the SLX4 scaffold and the XPF-ERCC1 nuclease, which incises the ICL, but the molecular basis of this recruitment has remained elusive. In Aim 2, we test the hypothesis that FANCI interacts directly with SLX4 to recruit XPF. We also test the idea that SNM1A makes the second incision to fully unhook the ICL. ICL repair is also triggered by RNA polymerase II, but the mechanism is unknown. We have recapitulated transcription-coupled nucleotide excision repair in egg extracts. In Aim 3, we will use this novel, cell-free transcription system to also study the mechanism of transcription- coupled ICL repair. Many of the hypotheses tested throughout the proposal are based on the in silico protein- protein interaction screening pipeline we recently established using AlphaFold-Multimer. Our studies will lend fundamental insights into some of the most pressing questions in ICL repair. As such, our work will help create a solid foundation for the development of more effective cancer chemotherapies, as well as drugs to mitigate ...