DNA interstrand cross-links (ICLs) are a common lesion generated in DNA spontaneously or induced by chemical agents e.g. chemotherapeutics such as Cisplatin, nitrogen mustard, psolaren. In ICLs two opposite strands of double helix are covalently linked blocking DNA replication and transcription [1]. Presence of ICLs in genomic DNA contributes to aging, neurodegeneration, and cancer. Apurinic/apyrimidinc (Ap) sites are form of DNA damage occurring frequently and also spontaneously in genomic DNA. Two equilibrating forms of Ap site are cyclic hemiacetal and ring-opened aldehyde. These unstable moieties can easily undergo amino-catalyzed or spontaneous strand cleavage generating an electrophilic 2,3-didehydro-2,3-dideoxyribose sugar remnant. A ring-opened aldehyde form of Ap sites can react with nucleophiles, for example adenine residues on the opposing DNA strand formin ICL with relatively high yields (20%). This can generally at sequences 5‘-ApT/5‘-AA and under physiologicaly relevant conditions [2, 3]. Therefore it is feasible that Ap-ICLs are generated within the genetic material of living cells. Recently, it has been shown that N-glycosidic bond forming the ICL is cleaved by Endonuclease VIII-like 3 (NEIL3), leading to unhooked crosslink in S phase. A critical step is programmed collapse of replication forks stalled by the ICL. Replication fork converge on the Ap-ICL, promoting NEIL3 dependent glycosylase cleavage of the N-glycosidic bond and following the nucleotide insertion across from the unhooked lesion, leading to strand extension beyond the unhooked lesion [4].
The focus of this work is the preparation of Ap-ICL, its characterization and mechanisms of its formation. In order to understand the molecular mechanisms of the repair we employed crystallization of purified Ap-ICL with and without DNA repair protein. Additionally, we have set up mechanistic experiments in vitro to reveal and calculate the probability of Ap-ICl formation and its stability in vivo. We characterize how environmental conditions affect these processes. Our work suggest that Ap-ICL is relatively a stable lesion in comparison with Ap site itself that is easily fragmented. Our experiments extended the understanding of ICL formation, repair and rough estimates of occurrence of this lesion in vivo.
The project was supported by Project InterBioMed LO1302 from Ministry of Education of the Czech Republic.