Targeting of DNA Methylation in Mammary Epithelial Cells

Institution: Salk Institute for Biological Studies
Investigator(s): David Liston, Ph.D. -
Award Cycle: 2004 (Cycle 10) Grant #: 10FB-0042 Award: $87,291
Award Type: Postdoctoral Fellowship
Research Priorities
Biology of the Breast Cell>Biology of the Normal Breast: the starting point



Initial Award Abstract (2004)
Breast cancer cells are characterized by changes in DNA structure that lead to the inactivation of genes critical for control of cell growth. The best described of these alterations are mutations which cause changes in the DNA sequence of a particular gene. In recent years, epigenetic alterations have become recognized as an important class of DNA lesions that contribute to the inactivation of critical genes in breast tumors. Epigenetic traits can be inherited, yet are not based on a change in the DNA sequence of a gene. A chemical change to DNA called DNA methylation is one of the best characterized epigenetic traits. It plays an important role in regulating gene expression, where it serves as a marker for inactive regions of the genome. In breast cancers DNA methylation is the most common cause of inactivation of the p16 gene, an important regulator of cell division. In addition, recent studies have found cells with a methylated p16 gene in substantial number of healthy, cancer-free women. This suggests that p16 methylation may be an important early event in the transition of normal breast cells into a pre-cancerous state. Despite the importance of DNA methylation in the development of breast cancer, the molecular mechanisms by which it is targeted to a specific gene remain unknown.

The goal of this proposal is to define the molecular mechanisms by which DNA methylation is specifically targeted to the p16 gene in breast cells. Our aims are: (1) To isolate the protein complexes in breast cells that are responsible for methylation of the p16 gene. (2) To identify and characterize the proteins responsible for bringing these complexes to the p16 gene.

We will examine the methylation of the p16 gene in breast cells isolated from cancer-free women that have a methylated p16 gene. To isolate protein complexes from these cells, we will biochemically separate protein extracts and test the resulting fractions for methylation activity. In cells, DNA is associated with proteins to form a structure called chromatin. In the lab we can assemble DNA into this structure for use in our assays. We will use this technique to assemble the p16 gene into chromatin and test whether the complexes we isolate are still able to methylate the DNA. If not, we will test the ability of known proteins to bring the complex to the DNA. In addition, we will look for novel factors by biochemically fractionating cell extracts. In this way we will define the proteins required for methylation of the p16 gene.

Previous studies of DNA methylation in cancer have used tumor cell lines which have many genetic aberrations. In this proposal, we will use non-tumor forming breast cells, which we believe will more accurately model the early events underlying the transition of a breast cell from a normal to premalignant state. Studies from our lab have shown that important aspects of gene regulation can be reconstituted in the test tube using DNA templates assembled into chromatin. As far as we know, no one has used this approach to study the targeting of DNA methylation. By defining the proteins involved in methylation of p16, we will gain significant insight into how this process is regulated. We also may identify important targets for chemotherapy.


Final Report (2006)
Many breast cancers contain genetic derangements that inactivate critical cell growth control genes. These abnormalities can be mutations, which alter the DNA sequence of a particular gene, or they can be epigenetic alterations, which silence the expression of a gene without changes to the DNA sequence. In cells, DNA is associated with histone proteins to form a structure called chromatin. These epigenetic alterations are mediated by chemical changes to either the DNA or histone proteins, primarily through the addition of methyl groups. DNA and histone methylation play an important role in regulating gene expression, and serve as marks for inactive regions of the genome. In many types of tumors, including breast cancer, the expression of key cell cycle control proteins, such as p16 and p15, is frequently silenced by DNA methylation. The goal of this project was to define the molecular mechanisms by which DNA methylation is specifically targeted to these growth control genes.

We have been using the p15 gene as a model, since it is epigenetically silenced in leukemia cells which are much easier to grow in large amounts than other types of cells. Although p15 itself is not silenced in breast cancer cells, it is very similar to another growth control gene, p16, which is frequently silenced in breast cancers. To start, we partially purified both DNA and histone methyltransferases from cancer cells. We also isolated fractions containing relevant DNA-binding proteins as our hypothesis was that these proteins may act to target the methylation activity. When we combined these protein fractions in various combinations with a chromatin-assembled p15 gene, we did not detect specific methylation of either DNA or histones. Since in vivo, gene silencing activities would likely act on a transcriptionally active gene, we reasoned that an active p15 gene might be required in our in vitro experiments. So far, using various transcription factors we have not been able to reactive transcription of a chromatin-assembled p15 gene in vitro.

We realized that we need more information about how p15 is normally regulated to guide our experiments. TGF¬beta is a critical cytokine for normal mammary gland development and it causes growth arrest in normal but not cancerous breast cells. The induction of p15 by TGF-beta is crucial for this growth arrest. Therefore, we are currently investigating the mechanism by which TGF-beta induces p15. So far, we have identified three transcription factors which bind to the p15 gene and may cooperate with TGF-beta. We have also found that before TGF-beta addition, much of the transcription machinery is already bound to the p15 gene. We are now investigating how TGF-beta causes the pre-bound transcription machinery to become active. These studies will not only help guide our in vitro studies of p15 silencing, but will also provide important insights into how TGF-beta regulates its target genes. Since the failure to stop growing in response to TGF-beta is a key abnormality in breast cancer cells, understanding this response at the molecular level should provide important clues as to how this derangement occurs and affects cancer progression.


Symposium Abstract (2005)
Co-investigator: Beverly M. Emerson, Regulatory Biology Laboratory, Salk Institute for Biological Studies, La Jolla, CA 92037.

Breast cancer cells are characterized by changes in DNA structure that lead to the inactivation of genes critical for control of cell growth. The best described of these alterations are mutations which cause changes in the DNA sequence of a particular gene. In recent years, another important class of DNA lesions that contribute to the inactivation of critical genes in breast tumors has been identified. They are called epigenetic alterations. Epigenetic alterations are passed from cell to cell as they divide, yet are not based on a change in the DNA sequence of a gene. A chemical change to the DNA called DNA methylation is one of the best characterized epigenetic alterations. It plays an important role in regulating which genes are turned on or off and serves as a marker for inactive regions of the genome. In many types of tumors, including breast cancer, expression of key cell cycle regulatory proteins, such as p15 and p16, is frequently silenced by DNA methylation. Despite its importance in the development of tumors, the molecular mechanisms by which DNA methylation-associated silencing is targeted to these genes remain unclear.

The goal of this project is to define the mechanisms by which epigenetic silencing is specifically targeted to the p15 and p16 genes. To do this we have partially purified DNA methyltransferases (DNMTs) from cancer cells and set up an in vitro assay for methylation of these promoter sequences. One hypothesis is that specific DNA binding proteins may be involved in targeting DNMTs. Therefore, we have also isolated protein fractions containing relevant transcription factors and are using these fractions along with the DNMTs in our methylation assay. In addition to DNA methylation, histone modifications, in particular methylation of histone H3 at lysine 9, are also associated with epigenetic silencing. We have isolated protein fractions containing histone methyltransferases (HMTases) in order to look for recruitment of these activities to chromatin-assembled promoters. Since in vivo a transcriptionally active promoter may be required to recruit silencing activities, we have also established an in vitro transcription system for these genes using chromatin-assembled promoters.

Ultimately, we aim to reconstruct both the activation and silencing of these genes in the laboratory. Since epigenetic silencing appears to be an important early event in the transition of normal breast cells into a pre-cancerous state, an understanding of the molecular details regulating this process will give important insights into how this transition occurs.