Losses and gains of multiple portions of the genome occur in tumor cells, and in some cases amplifications, or high level copy number increases of portions of chromosomes are observed. Alterations in gene expression that result from these copy number aberrations provide a selective advantage to the tumor cells and specific copy number aberrations have been associated with inactivation of tumor suppressor genes or overexpression of oncogenes. The driver genes for many regions of recurrent amplification have not been identified and methods to facilitate their identification would be broadly applicable to increasing knowledge of the genetic events involved in cancer development. Recent implementation of a high resolution form of comparative genomic hybridization (CGH) using microarrays allows quantitative mapping of copy number across an amplicon such that the magnitude of the copy number increase and the extent of the amplicon can be mapped relative to the physical map of the human genome. The goal of this project is to further develop our understanding of the information that can be obtained from array CGH copy number profiles. In particular, we will determine if the copy number maxima as mapped by array CGH can be used to identify the driver gene(s) for an amplicon. This investigation was motivated by the observation that array CGH copy number profiles displayed peaks mapping to the locus of candidate oncogenes amplified at 20q13 in breast cancer, suggesting that these peaks in the copy number profile may result from selection for highest copy number of the genes conferring growth advantage. Therefore, quantitative mapping of amplicon structure by array CGH may provide a powerful new way to identify oncogenes in regions of recurrent amplification in tumors. In order to establish the generality of these observations, in this project, we will use array CGH to quantitatively map the amplicon structure of three well-established oncogenes, CMYC, ERBB2 and CCND1 in different tumor types. We will also map amplicons arising in cells in culture in response to selection for drug resistance. We will determine (a) the extent to which amplicon boundaries are clustered in the genome around these genes, (b) where amplification maxima occur in relation to the locus of the oncogene and (c) the similarities of differences in the amplicons as a function of tissue type. Characterization of the copy number profiles of amplicons containing these known oncogenes and drug resistance genes will provide the information required for interpretation of amplicon structure in other regions of the genome and facilitate identification of the critical genes they contain.
