The Role of 5-Hydroxymethylcytosine in Genome Instability

Abstract

5-hydroxymethylcytosine (5hmC) is a modified base generated in vivo via the oxidation of 5-methylcytosine (5mC) by members of the Ten Eleven Translocation (TET) dioxygenase enzyme family. Iterative oxidation of 5hmC to 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC) is also TET-dependent. 5mC represents a critical epigenetic modification, and DNA methylation is involved in many biological processes. DNA methylation state is not static; hence, DNA demethylation mechanisms exist to remove 5mC from the genome. This is critical during developmental periods of widespread epigenetic reprogramming in the early zygote and primordial germ cells. DNA demethylation mechanisms may be passive, relying on dilution of 5mC with successive rounds of DNA replication, or active. Active DNA demethylation is hypothesised to involve TET-mediated 5mC oxidation and excision of the resultant modified base(s) by DNA repair processes. 5hmC represents the most abundant form of oxidised 5mC derivative, and its abundance varies significantly between cell types and developmental stages. However, it is not clear how 5hmC is tolerated by the cell. Specifically, it remains to be determined whether 5hmC represents an essential intermediate in the process of active DNA demethylation, or if it is recognised as a form of base damage necessitating repair. The results presented in this thesis identify a critical role of the Base Excision Repair (BER) pathway in the tolerance of exogenously supplied 5hmC, the absence of which results in genotoxicity and cell death following 5hmC treatment. They also identify that exogenous 5hmC-mediated genotoxicity depends on its deamination within the nucleotide pool, prior to its incorporation within genomic DNA. This is validated by the observation that loss of the SMUG1 DNA glycosylase, which excises genomic 5hmU, rescues the hypersensitivity of BER-deficient Polb-/- cells to exogenous 5hmC. I also present preliminary data suggesting that endogenously generated 5hmC, produced via overexpression of the TET1 catalytic domain, is unlikely to threaten genome stability via deamination in the nucleotide pool based on the observation that loss of the TDG DNA glycosylase, which excises 5fC and 5caC, rescues the sensitivity of Polb-/- cells to TET1-CD overexpression, but loss of SMUG1 does not. Finally, this thesis also explores the role of protamine in mediating paternal genome stability. The protamines, PRM1 and PRM2, displace canonical histones during the final stages of sperm development, a process that must be correctly achieved to maintain male fertility. PRM2 is subject to proteolytic processing during spermiogenesis, in which the N-terminus is cleaved following DNA binding. Using a PRM2 N-terminus-deficient mouse model, I find that the cleaved region of PRM2 is essential for sperm genome stability and male fertility.