Role of cytidine deaminases in epigenetic reprogramming and demethylation of DNA

All genes in the genome of humans and other mammals are now known. However we have little knowledge about which signals are needed for genes to switch on and off during development of a new organism from embryo to adult. An important signal are instructions that are attached to DNA in the form of small chemical molecules (methyl-groups). These are usually attached to genes when they need to be off. When genes need to be switched on, the methyl-groups need to be removed from the DNA; this is particularly important at the beginning of development in the very early embryo. When this switching is defective it can give rise to diseases of embryonic development and many cancers have aberrant ?epigenetic? switching of this type. Our current work has identified a mechanism for the removal of the methyl-signals from DNA, and in work under this grant we will gain a better understanding how this mechanism works in embryos, in order to harness its activities for stem cell ?regenerative? medicine and cures for cancer.

DNA methylation of CpG dinucleotides is an important mechanism of epigenetic regulation of genome function in mammals, and has key roles in genomic imprinting, X chromosome inactivation, silencing of transposons, genome stability, and developmental gene expression. Aberrant regulation of DNA methylation is implicated in many human diseases, particularly cancers. DNA methylation is reprogrammed genome wide in the zygote and preimplantation embryo, and in primordial germ cells. Reprogramming involves active demethylation, and is important for the erasure of genomic imprints, and likely for the return of the embryonic genome to totipotency. We have shown that cytidine deaminases such as Aid and Apobec1 can deaminate 5 methylcytosine (5meC) in DNA, leading to T:G mismatches which can be repaired by the base excision repair pathway. These genes are part of a pluripotency gene cluster and are expressed in oocytes and germ cells in which demethylation occurs. We have recently targeted Aid in transgenic mice to a highly methylated region of the genome, and find that this region is efficiently and substantially (approx 90%) demethylated without incurring sequence mutations. Preliminary work shows that Aid deficient zygotes have a deficit in the demethylation of the paternal genome. Having obtained proof-of-principle evidence that Aid can initiate demethylation in vivo, we are now building upon this work to fully understand the mechanisms involved as well as the physiological relevance. The aim of the proposed work is to elucidate in depth the role of Aid/Apobec family deaminases in epigenetic reprogramming and demethylation of DNA in vivo. The specific objectives are 1. the full characterisation of the gain-of-function experiments in which Aid and Apobec1 are targeted to methylated regions in the mouse genome in vivo 2. the characterisation of the pathways downstream of Aid which resolve base mismatches, using Xrcc1 and TDG conditional knockouts 3. the completion of a conditional Aid-Apobec1 knockout, followed by phenotypic characterisation and its combination with Apobec2 and -3 knockouts and 4. the analysis of endogenous DNA methylation including by methylDIP in these knockouts. This work will not only elucidate the mechanisms of epigenetic reprogramming and demethylation, but it will also have direct applications in regenerative medicine and cancer therapy approaches, some of which are already being persued in industrial collaborations.