Our laboratory has discovered a novel type of genetic alteration in cancer, loss of imprinting (LOI). Genomic imprinting is a modification of a specific parental allele of a gene in the germline that causes silencing of that allele in the offspring. LOI of the insulin-like growth factor-II (IGF2) gene leads to biallelic rather than monoallelic expression of this important autocrine growth factor. The current grant period has been highly productive, and it includes the discovery that LOI arises in both the tumors and normal tissue of patients with colorectal cancer, and that LOI is specifically linked to those patients with microsatellite instability in their tumors. We will now determine whether loss of imprinting is a hereditary trait, and whether it is linked to cancer risk in families. We also discovered that LOI identifies a specific pathological subtype of Wilms tumors, the first time a particular category of tumor has been defined by abnormal imprinting. Furthermore, we found that there are distinct mechanisms for LOI in Wilms tumor and colorectal cancer, involving abnormal DNA methylation affecting the H19 and IGF2 genes. We will now determine the cis-acting changes, trans-acting factors, and alterations in gene expression that distinguish cancers with LOI, and their relationship to microsatellite instability. A major limitation to mechanistic studies of genomic imprinting in cancer has been the relatively small number of known imprinted genes. We have overcome this obstacle with a novel strategy for identifying normally methylated CpG islands throughout the genome, allowing us to identify many new imprinted genes, and to identify common features among them and among CpG islands whose methylation is altered in tumors. Finally, we are identifying genes that may modify genomic imprinting and DNA methylation in cancer. These studies should continue to provide new insights into the causes and consequences of this surprisingly common alteration in cancer. In addition, since LOI represents the first common genetic abnormality of any type found in the normal cells of cancer patients, these studies may have a substantial impact on cancer surveillance and mortality.
Showing the most recent 10 out of 55 publications
