Adele Murrell scientific summary

Imprinted genes as heritable markers for cancer predisposition

Previous and current research

The Insulin-like growth factor 2 (IGF2) is a well studied imprinted gene that is normally expressed from the paternal allele. DNA methylation, histone modification and secondary chromatin structure are the key components contributing to epigenetic gene regulation.

Loss of IGF2 imprinting (i.e., bi-allelic expression) accompanied by DNA methylation changes at the locus occurs in congenital disease syndromes and many cancers. We have previously shown that genetic variation in IGF2 is associated with loss of methylation at this locus in patients with a congenital cancer predisposing disease known as Beckwith Wiedemann syndrome, suggesting that genetic and epigenetic variation may act synergistically.

Recently we have completed a prospective high throughput methylation study in breast and colorectal cancer patients in order to detect methylation changes at IGF2 that are present prior to cancer onset. Our results indicate that constitutive DNA methylation aberrations at IGF2 are rare and that loss of methylation is acquired as a result of cancer rather than present as a predetermining risk of developing cancer. We are currently examining the extent to which loss of methylation at the IGF2 gene is indicative of cancer and whether it has value as a diagnostic or prognostic biomarker.

In mice, we and others have shown that imprinted genes at this locus are partitioned into active and silent domains through chromatin loops which have different conformation on the maternal and paternal chromosomes. Chromatin looping may explain how different DNA regulatory elements can interact over long distances and also how genetic variation can influence chromatin structure at a distance.

To address the mechanisms whereby imprinting is lost in cancer, we are isolating and characterising proteins that interact specifically with methylated or unmethylated DNA at the imprinting control region at the IGF2 locus. We are also examining the role that these proteins play in maintaining chromatin conformation at the locus and how the conformation is affected by the epigenetic changes that a cancer cell undergoes.

Future projects

Since our recent research strongly indicates that methylation changes at the IGF2 locus are due to disruption of the maintenance of imprinting rather than aberrant establishment of germline imprinting, we are focusing our attention on factors that affect the stability of epigenetic memory. Thus what are the genomic and epigenomic features that determine whether a gene will maintain its imprinted state or lose it?

We have begun to expand our high throughput methylation assay to detect methylation changes in colorectal, breast and lung cancer in a larger panel of imprinted genes. These studies are being done in conjunction with chromatin profiles such as histone modifications and expression assays in order to compare epigenetically stable and labile regions.

Using this approach we hope to identify sequence and protein binding characteristics that may predict whether a gene will become susceptible to epigenetic reprogramming during neoplasia. We also have some interesting chromatin modifying candidates that bind directly to regulatory regions at the IGF2 gene locus which we shall functionally characterise to assess their roles in the maintenance of imprinting.