In order to characterize patterns of global transcriptional deregulation in primary colon carcinomas, we performed gene expression profiling of some 300 rectal and colon carcinomas, and, as expected, identified comprehensive lists of deregulated genes. Genes that are upregulated are potentially required for the viability of colorectal cancer cells, and can therefore be considered oncogenes. Our systematic comparison of colon and rectal carcinomas also revealed a significant overlap of genomic imbalances and transcriptional deregulation, including activation of the Wnt/beta-catenin signaling cascade, suggesting similar pathogenic pathways. The functional validation of novel targets was performed in cell lines established from colorectal carcinomas that recapitulate the genomic and gene expression changes that we have previously observed in primary colorectal cancers. The relevance of this project with respect to treatment response is described in Project 4: ZIA BC 010837. 1) Identification of dynamics of chromosome segregation errors and clonal evolution Despite ongoing chromosomal instability and an enormous degree of intratumor heterogeneity the distribution of chromosomal gains and losses, and with that, the pattern of genomic imbalances is maintained when the bulk of the tumor is analyzed; therefore, such tumor-type specific imbalances must be considered as an evolutionary bottleneck for the continued division of tumor cells and therefore drivers of tumorigenesis. This has been shown by us and others over the past decade, mainly through profiling of tumor genomes by CGH; the fact that entire chromosome arms or chromosomes are recurrently gained or lost, and that these imbalances are maintained throughout the life of the tumor cells, therefore, strongly suggests that there must be genes other than the obvious candidates (i.e., MYC on chromosome arm 8q) that are necessary for tumorigenesis in a particular tissue. We propose to test this hypothesis by asking the question whether continued propagation of CRC cells requires the maintenance of chromosomal aneuploidies that define this cancer. Cells that would be forced to eliminate chromosomes that are usually gained would have a negative selective pressure, i.e., would be eliminated from the population. In collaboration with Dr. Michael Kuehl from our Branch we propose to tag the commonly gained chromosomes (and control chromosomes that have the same modal number as the cell line as a whole) with herpes simplex thymidine kinase (HSV-TK) and then ask whether the addition of ganciclovir to the medium would result in the loss of these chromosomes. This will be performed initially in two CRC cell lines, DLD1 and SW480. We propose tagging two chromosomes each, i.e., chromosomes 3 and 7. Chromosome 3 is copy number neutral in both cell lines, while chromosome 7 is neutral in DLD1 but gained in the population of SW480 cells. Using extensive interphase FISH analysis, we established that 96.5% of the cells of SW480 have copy numbers of chromosome 7 above the modal number of the cell (i.e., a gain), and chromosome 7 is never below the modal number. Cytogenetic analysis after selection would indicate whether this colon cancer cell can survive if chromosome 7 would be eliminated, and potentially determine whether certain regions of this chromosome are responsible for its continued maintenance in the tumor (i.e., identify regions that contain driver genes). In our opinion, these experiments would answer fundamental question on the biology of epithelial cancers because it could explain the preponderance of tissue specific chromosomal aneuploidies. 2) Cellular consequences of silencing transcription from aneuploid chromosomes utilizing genome editing Recently, Lawrence and colleagues showed that it is possible to silence transcription from the extra copy of chromosome 21 in cells from patients with Down's syndrome using Zinc finger based genome editing to introduce the Xist gene into chromosome 21. We now propose using this approach to reduce or eliminate transcriptional activity of genes from aneuploid chromosomes in cancer cells (and, as controls, from cells with artificial chromosomal aneuploidies that we generated using microcell mediated chromosome transfer). We will initially target chromosome 7 in the cell line SW480, because of its central role in CRC. First, gene expression profiling will be used to test the successful silencing of the aneuploid chromosome 7. Second, we will analyze whether silencing will result in reduction of (i) proliferative activity, (ii) invasiveness and (iii) tumorigenicity in nude mice, and (iv) alterations of the 3D architecture of the chromosome 7 territory. It also affords us the possibility to query whether silencing of chromosome 7 would require the acquisition of additional, compensatory cytogenetic changes. Analogous experiments will be performed in our mouse model of spontaneous transformation, in which chromosome 15 is recurrently gained. We expect that these experiments will be critical to further evaluate the role of chromosome wide copy number changes in tumorigenesis and can therefore help to interpret results from related experiments described in this Project. We have discussed these experiments with Dr. Jeanne Lawrence, who will collaborate with us and provide the Xist construct that she used for silencing chromosome 21. We have also consulted with Dr. Rafael Casellas, NIAMS, who has considerable expertise in using genome editing techniques. He will assist us in identifying suitable chromosomal integration sites.
Showing the most recent 10 out of 43 publications
