Defects in replicative DNA polymerases linked to cancer predisposition and tumour development

Our genetic material is constantly subjected to damage. Much of this damage comes not from external sources such as carcinogens or radiation, but from the intrinsic biological processes that DNA has to perform to allow our cells to function. One key biological function of DNA involves its replication, so that each of the two daughter cells produced during cell division receives a complete and accurate copy of all the genetic information. Even in normal cells, this copying process is not 100% perfect, and the resulting errors can lead to mutations that alter cell function, potentially leading to cell death or other changes that can lead to cancer.

Recent findings have shown that some cancers are associated with defects in the polymerase enzymes that copy DNA. DNA polymerases are normally highly accurate, but defects in these polymerases can increase the error rate when the genome is copied, and this can accelerate the development of some types of cancer in tissues where there is a high rate of cell multiplication. In this grant we will study these defective polymerases in detail, to understand which changes found in human cells are likely to be pathogenic. We will study how the polymerase malfunctions, the consequences for the process of chromosome replication and the types of copying mistakes which result. Understanding the properties of the cancer-associated DNA polymerase should help to explain how these defects lead to tumour formation. It is also possible that drugs might be developed that target the replication of tumour cells that express these variant polymerases.

Germline mutations in the exonuclease domain of replicative DNA polymerases delta and epsilon have been recently shown to be associated with a predisposition to colorectal cancer. In addition a wider range of exonuclease domain mutations have been found as somatically variants in colorectal and endometrial cancer, suggesting that changes in polymerases may more widespread as a step leading to genome instability in tumourigenesis. The exonuclease domain contains the active site for a 3'-5' exonuclease activity required for proofreading, capable of removing misincorporated nucleotides that are not correctly base paired to the template. In this grant we will characterize the polymerase variants associated with colorectal cancer predisposition and endometrial tumours by constructing equivalent mutations in a model organism, fission yeast, and by purifying human polymerases containing the relevant mutations. We will compare the phenotype of the strains constructed and the biochemical properties of the purified polymerases with exonuclease null version of the polymerases, where key catalytic residues in the exonuclease active site have been changed, to clarify the extent to which the phenotype is due to loss of exonuclease activity. We will grow strains expressing polymerases with exonuclease domain mutations for many generations and perform genomic sequencing to determine the mutational spectra and whether specific regions of the genome are prone to mutagenesis. We will establish why some exonuclease domain mutants require an intact checkpoint for survival in yeast, and whether this is relevant to the mechanism of mutagenesis. This will allow insight into the mechanism whereby polymerases with exonuclease domain defects promote tumour development in humans.

The local beneficiaries of this project will be the post-doctoral research assistants working on the project who, through experience gained, will be in a better position to secure future jobs in the healthcare/research sector. The PIs' groups also have a steady flow of graduate students and MRes/undergraduate research project students who will be able to participate in aspects of the research.

The academic DNA replication and repair community will benefit from the research outputs which should clarify the consequences of human polymerase defects on S phase execution, and genome stability.

The clinical community will benefit from clarification of which human DNA polymerase mutations are likely to have a phenotype in human cells, rather than just being polymorphisms or passenger mutations, enhancing the usefulness of any genetic screening for germline mutations and possibly aiding the prognostic value of these changes for tumour classification.

The general public will benefit in that those individuals affected by polymerase mutations stand to receive better health care as a result of improved understanding of the nature of the phenotypic consequence of the mutation. This might include frequent screening for individuals at risk from germline mutations, or better prognostic information using DNA polymerase mutations in tumour classification. In the long term drugs exploiting tumour phenotypes resulting from polymerase mutations may have value for treatment of specific cancers.