> Research & Grants
> Grants Program
> Research Grants
> Research Grants Awarded
Mouse Models of Breast Cancer Based on Epigenetic Silencing of the Hic-1 and AP-2ƒÑ Genes
Mouse Models of Breast Cancer Based on Epigenetic Silencing of the Hic-1 and AP-2ƒÑ Genes
Background: It is increasingly apparent that epigenetic events, in the form of promoter hypermethylation and transcriptional silencing of genes, provide an alternative to genetic changes for loss of tumor suppressor gene function in cancer. We have recently defined two such epigenetic changes in breast cancer for the transcription factor genes, HIC-1 (hypermethylated ¡Vin¡Vcancer-1) and AP-2ƒÑ. These candidate tumor suppressor genes are not mutated in cancers but epigenetically mediated loss of function may help drive key stages of human breast tumorigenesis. HIC-1 is hypermethylated in 50% or more of such cancers and insertion of this gene into cultured cancer cells lacking its expression reduces soft agar cloning and restores an element of normal branching morphogenesis. AP-2ƒÑ is a critical factor for proper mammary and other epithelial development and 70% of invasive, but only 10% of DCIS, breast tumors have 5¡¦region hypermethylation of this gene and loss of protein. Complete disruption of either of the above genes in mice results in embryonic lethality. Heterozygote knockout mice for either gene develop normally but HIC-1 +/- mice develop carcinomas and lymphomas (40%) by 2 years of age. These tumors lose wild type HIC-1 allele function in association with hypermethylation rather than deletion. HIC-1 +/- mice develop breast glandular hyperplasia and, when crossed with p53+/- mice which also do not get breast tumors, develop breast carcinomas (10-15%) at an early age. AP-2ƒÑ +/- mice is not known to get tumors.
Objective/Hypothesis: We propose to determine whether double heterozygote HIC-1+/- , AP-2ƒÑ+/- mice, in various strains, with and without p53+/- contribution, may provide mouse models for breast cancers dependent on steps mediated by epigenetic gene silencing. The models can be used to test the efficacy of targeting reversal of the epigenetic silencing events for prevention, early intervention, and therapy of this cancer.
Specific Aims: 1) To determine the incidence for breast cancer, and mechanism for loss of the wild type alleles in tumors, for double het HIC-1+/-, AP-2ƒÑ+/- mice in their current non- mammary tumor permissive backgrounds. 2) To determine the above for HIC-1+/-; AP-2ƒÑ +/-; and double het HIC-1+/-, Ap2+/- mice in the mammary cancer permissive FVB background. 3) To determine for the mice in #2, the contribution to breast cancer of breeding the animals with p53+/- mice. 4) To begin determining, for any of the models with high incidence of breast cancers and epigenetic silencing of the HIC-1 and AP-2ƒÑ genes, the efficacy for prevention or treatment of tumors by lowering capacity for hypermethylation through genetic, and drug induced reduction of DNA methyltransferase activity.
Study Design: During the first year of the award, we will breed the HIC-1+/-, AP-2+/- double het mice, and double het AP-2ƒÑ +/-, p53 +/- mice, in the non-mammary tumor permissive background, to begin studying the incidence for breast cancers and the mechanisms for loss of the wild type HIC-1 and AP-2ƒÑ alleles in tumors which arise. We would also, in these strain backgrounds, breed animals to create a triple het situation for the HIC-1+/-, AP-2ƒÑ+/-, and p53+/- gene states to do these same studies. During the second year, we will breed each of the above single het, double het, and triple het mice into the mammary tumor permissive FVB background to see the effects on breast cancer incidence and mechanisms of allelic function loss. Finally, we will cross, as soon as models for breast cancer emerge, tumor prone mice with mice heterozygous for the DNA methyltranferase 1 gene (DNMT1), which reduces overall DNMT activity by 50%. We will also use the DNMT inhihibitor 5-deoxy-azacytidine to assess efficacy for prevention, early intervention, and therapy for the breast cancers.
Potential Outcomes and Benefits of the Research: We have solid evidence that epigenetic silencing of HIC-1 and AP-2ƒÑ may play a key role in the early stages of human breast cancer. Our data, to date, for HIC1+/-, and HIC-1 +/- ,p53+/- double het mice support this possibility. We believe there is an excellent chance that the studies in this proposal will create extremely valuable mouse models for breast cancer based on early steps dependent upon epigenetically mediated gene silencing. If so, our studies to target reversal of this epigenetic gene silencing for prevention, early intervention, and therapy in these models will have great importance for considering these approaches for the human disease.
Mouse Models of Breast Cancer Based on Epigenetic Silencing of the Hic-1 and AP-2 alpha Genes
One of the most powerful steps in the initiation and progression of cancers, including breast carcinomas, is the loss of proper function of so called “tumor suppressor genes”. To have full effects, each of two DNA copies of a given such gene must become inoperative in the tumors. Until some five to seven years ago, the mechanism for this loss of function appeared solely due to mutations, or genetic changes, in the DNA for such genes which results in the production of a faulty protein, or disrupted protein. In the classic situation in the tumors, one of two DNA copies of such a gene would bear this mutation while the second copy is completely deleted by loss of a region of the chromosome where the gene DNA resides. It is now apparent that a second mechanism or “epigenetic” abnormality can also very frequently account for loss of tumor suppressor gene function in cancer. In this process, methyl groups are inappropriately added to the DNA region which turns on the genes to produce the RNA from which protein is made and this “hypermethylation” is associated with loss of RNA and protein in the tumor. The hypermethylation can, not infrequently, occur for both copies of the gene, or be present on one copy while the other is lost by deletion of the portion of the chromosome carrying the gene. A major difference in the epigenetic change from the genetic change is that the former is reversible. In cell cultures of all types of cancer, treatment of the cells with drugs which cause loss of the methylation results in re-initiation of RNA production from the genes and appearance of functional protein. Some of these drugs are actually showing great clinical promise for the treatment of leukemias and pre-leukemia states.
Our laboratory has played a role in defining the above epigenetic mechanism for loss of tumor suppressor gene function in cancer. Most recently we have found this change frequently in human breast cancer for two genes, HIC-1 (hypermethylated in cancer-1) and AP-2 alpha (Activator Protein-2 alpha), which are important in normal embryos and mature tissues for proper cell development including control of cell growth and death. Neither gene has been found to have classic mutations in tumors. We created a mouse model for genetic disruption of one copy of the HIC-1 gene (HIC-1 heterozygotes) in which mice, as they age, develop many tumor types. These animals do not get breast cancer but do have abnormal benign growth of glands in the breast. We have bred these HIC-1 heterozygote mice with mice which have one copy of a major known tumor suppressor gene, p53, genetically disrupted (p53 heterozygotes). Mutations of one copy in association with chromosome loss of the second copy, is a frequent finding for the p53 gene in human breast cancers. Our mice which are both HIC-1 and p53 heterozygotes start to get breast cancers, involving 10 to 15% of the animals, which fully resemble the human disease. Also, in these tumors, the second normal copy of HIC-1 is functionally lost in association with hypermethylation while the remaining normal p53 gene copy is lost with chromosome deletion. These are the precise changes seen in many human breast cancers. Mice with one disrupted copy of the AP-2 alpha gene have also been created by our collaborator, Dr. Trevor Williams, in Colorado but these mice have not been extensively explored for tumor development.
In the current proposal, we will create a series of mice that simultaneously carry loss of one copy of the HIC-1 and AP-2 alpha genes and also mice with one copy loss for the AP-2 alpha and p53 genes plus mice with loss of one copy of all three genes. We predict that one or more of the mouse types will very frequently develop breast cancers and these will have hypermethylation of the remaining normal copies of the HIC-1 and AP-2 alpha genes. If so, we will have created a very valuable model which fully mimics the human disease with respect to important early steps which are mediated by potentially reversible epigenetic silencing of the two genes under study. We can then begin to test the efficacy of reversing this gene silencing for the prevention, early intervention, and therapy of breast cancer. Positive results would have great implications for instituting these approaches for the human disease.