Understanding Herceptin Resistance with Dual Function Array

Institution: University of California, San Francisco
Investigator(s): Tsui-Ting Ching, Ph.D. -
Award Cycle: 2003 (Cycle IX) Grant #: 9FB-0009 Award: $49,612
Award Type: Postdoctoral Fellowship
Research Priorities
Biology of the Breast Cell>Pathogenesis: understanding the disease

Initial Award Abstract (2003)
Increasing evidence shows the importance of epigenetic mechanisms (e.g. CpG island methylation) in the transcriptional down-regulation of critical tumor suppressors and growth regulatory genes in breast cancer. The patterns of aberrant methylation are non-random and tumor type-specific, and can be distinct with tumor types such as breast tumors. Further, it is also well-known that chromosomal deletions are a major mechanism of gene inactivation in human cancer. The extent to which these two mechanisms affect the same or different sets of genes in breast cancer is mostly unexplored. Herceptin is a monoclonal antibody developed by Genentech and used in the treatment of HER2-overexpressing metastatic breast cancer. Although this therapy is quite successful in treating some patients, approximately 5% of tumors from these patients are resistant to the treatment.

Our interest is to understand the epigenetic and genetic differences which confer Herceptin resistance in patients. This project will examine whether resistance to Herceptin therapy in breast cancers is mediated by the combined effects of aberrant DNA methylation and deletion, particularly on genes of the growth receptor tyrosine kinase (RTK) pathways. Our aim is to develop a new high throughput gene-array method for analysis of both methylation and deletion, taking advantage of existing array CGH (comparative genomic hybridization) technology. We will focus our analysis on the relative contribution of aberrant gene methylation and gene copy number changes in causing specific dysfunctions of the receptor kinase pathways. A key component of our analysis will be comparing the HER2-positive/Herceptin-sensitive with the HER2 -positive/Herceptin-resistant breast cancer cell lines. We have access to custom-made BAC (bacterial artificial chromosome) DNA arrays, which contain 500 known genes involved in RTK pathways. This approach allows us to detect specific aberrant methylation and gene deletions. Finally, we will compare the biological functions with the genomic profiles in these cell lines to select gene candidates related to Herceptin sensitivity. We can confirm our resistant cells with DNA methyltransferase inhibitors or reintroduction of the target genes into the resistant cells.

Aberrant gene methylation and gene deletions are predominant mechanisms of gene silencing in breast tumors, yet the extent of their interaction is unknown. The new dual-function array will be capable of assessing the contribution of both of these mechanisms. This will lead to a substantially improved understanding of breast tumorigenesis that is not biased towards genetic events, as has been done in the past. Most importantly, the application of this method to elucidating the mechanism of Herceptin resistance would lead to new combination treatments to overcome the resistance and potentially would increase the success rate of Herceptin therapy.

Final Report (2004)
Note: the PI resigned the fellowship after 15-months to relocate to Michigan.

The underlying basis of human cancer involves genetic alterations that activate oncogenes and inactivate tumor suppressor genes, culminating in unregulated cell growth. The existing paradigm of tumorigenesis are based on the selection of subsets of genes contributing to this process, and cancer diagnosis/prognosis models are based on this genetic approach. However, epigenetic alterations such as changes in DNA methylation are also involved in tumorigenesis and tumor progression, but are not detected by genetic screening methods. In order to overcome the limitation, we have adapted an existing BAC-array (i.e., large DNA insert libraries of human chromosomal segments) system to provide a versatile assessment of both methylation and deletion genome-wide.

The technical basis for this project is the use of comparative genomic hybridization (CGH), which involves the hybridization of differentially-labeled total genomic tumor and normal DNA. This provides a map of genomic changes in complex tumor genomes. The application of CGH allows the identification of aberrant regions of interest, and allows the rapid identification of new cancer genes. Our aim was to expand this approach to include the comparison of methylation patterns in tumor cells. DNA methylation is a type of epigenetic (i.e, modifications in gene expression that are controlled by heritable but potentially reversible changes in chromosome structure) regulation. It is a post-cell replication process by which cytosine residues in CpG sequences are methylated, forming gene-specific methylation patterns. For example, “housekeeping genes” that possess CpG-rich islands at the promoter region are unmethylated in all cell types, whereas tissue-specific genes are methylated in all tissues except the tissue where the gene is expressed. The whole process by which cell lineages become epigenetically methylated is termed “imprinting.” We wanted to expand our use of CGH and BAC analysis to study methylation patterns; in particular to understand how breast cancer cells become modified to allow resistance to anti-Her-2 drug, Herceptin.

We have successfully developed and validated a novel dual-fuction CGH array. It has been applied to screen DNA methylation changes in normal tissues and breast cancer cell lines. During the screening of normal tissues, we made the unexpected discovery that a large number of CpG islands (i.e., sites of DNA methylation) exhibit tissue-specific methylation. We found differential CpG island methylation and expression of the DKFZP566K1924 and SHANK3 genes in normal human tissues. To validate this new integrated array to uncover the epigenomic and genomic alterations, we aligned the methylation maps with copy number maps of 2,500 loci from array comparative genome hybridization (array CGH) in breast cancer cell lines, such as MCF-7, BT474, and MDA-MB361. The primary result clearly shows this novel array provides a new capability for detection of aberrant methylation and deletion alterations on a global scale. We are in the process of extending this work to study the basis for Herceptin-resistance, and our results suggest that specific genes can be identified to explain this resistance. Our research on this topic is “in press” in Nature Genetics.