Utilising Uracil DNA Glycosylase to detect the presence of 5-Methylcytosine

Abstract

DNA is regularly subjected to endogenous and exogenous reagents that cause mutations that can be detrimental to a cell if they are not repaired. One class of enzymes responsible for DNA repair is the family of DNA glycosylases and their role is to remove damaged bases. Uracil DNA Glycosylase (UDG) is a member of this family and is highly specific, removing only uracil, an RNA base, from DNA. Uracil arises in DNA through misincorporation of deoxyuridine monophosphate (dUMP) creating an A.U base pair, or through deamination of cytosine resulting in a G.U base pair. Though UDG acts on A.U pairs, this is not it’s primarily role as A.U pairings are not mutagenic. However the G.U mispair is highly mutagenic and leads to a G.C to A.T transition on subsequent rounds of replication. UDG only reacts with uracil and has no activity at thymine since the 5-methyl group on the base is excluded from the active site. This thesis examines mutants of UDG that can cleave cytosine but not 5-methylcytosine. Methylation of cytosine at CpG sites leads to gene silencing and is an important epigenetic signal. Knowing the methylation state of cytosines will therefore be important for understanding gene control and may be beneficial for treating many diseases. The most common method for detecting cytosine methylation uses a bisulphite reaction followed by normal DNA sequencing methods. This process has several drawbacks and the aim of this work is to create an enzyme that is capable of distinguishing between5-methylcytosine and cytosine. It has been reported that mutation of a critical asparagine in UDG to an aspartate allows the enzyme to accommodate cytosine into its active site; generating a cytosine DNA glycosylase (CDG). Using the natural ability of UDG to distinguish between uracil and thymine due to the presence of the 5-methyl group, we hypothesised that the mutant enzyme should be able to discriminate between5-methylcytosine and cytosine, which differ by the presence or absence of a methyl group in the same position. E. coli and human CDGs were prepared and their ability to remove cytosine or 5-methylcytosine examined when placed in different sequence contexts. hCDG was generated through complete gene synthesis of hUDG followed by the N204D mutation. The corresponding mutation in E.coli (N123D) generates a highly cytotoxic enzyme that cannot even be cloned in pUC19. As L191 aids base flipping, mutation to alanine (L191A) renders the enzyme inactive; activity can then be rescued using a bulky synthetic nucleoside that occupies the base pair and forces the target base into an extrahelical conformation. The L191A mutation was followed by N123D to generate an expressible and functional eCDG, denoted eCYDG. We demonstrate that these mutants have cytosine glycosylase activity when the cytosine is mispaired or unpaired, but not when paired with guanine, and show no activity against5-methylcytosine in any context. The activity of these CDGs varies with the stability of the base pair, with the fastest cleavage rates being obtained with the least stable base pairs, and also varies with the local sequence context. As CDGs are able to discriminate between cytosine and 5-methylcytosine we began development of a real-time PCR assay for detection of 5-methylcytosine. This employed a hexaethylene glycol (HEG) linker opposite the target cytosine, as this produces one of the fastest cleavage rates and cannot be read by a DNA polymerase.

More information

Identifiers

ORCID for Keith R. Fox: orcid.org/0000-0002-2925-7315

Catalogue record

Export record