Cloning of an ovarian cancer tumor suppressor gene on 8p

Research project P 14138 Cloning of an ovarian cancer tumor suppressor gene on 8p Michael KRAINER 24.01.2000 Loss of function of tumor suppressor genes (TSG) through inactivating mutations, methylation or other as yet undescribed disturbances of transcription is one of the driving forces behind cancer development. A first step to identify regions in the genome, which might harbor putative TSGs is the identification of loss of genetic material by comparing tumor DNA to germline DNA from the same patient. By analyzing loss of heterozygosity (LOH) on chromosome 8p we have initially identified three minimal regions of loss in ovarian cancer One of these regions characterized by loss of microsatellite markers D8SI992 and D8S261 occured mainly in early tumors. This area has also been implicated in tumorigenesis of various other cancers, in particular prostate cancer, where a homozygous loss in this particular region was described previously. By focusing on this area through intensive datamining and reprobing our previous patient population with additional microsatelite markers we were able to integrate the minimal region of loss into the Genemap99 and the G3 Stanford Radiation Hybrid Map and thus to narrow it down to roughly 90OKb. We are currently analyzing ESTs and Genes mapping to this region as potential candidate genes. We will furthermore use probes from the region to isolate bacterial artificial chromosomes (BACs) to achieve genomic coverage of the minimal deletion. These BACs will also be used for in situ hybridization experiments to eventually identify possible hornozygous, deletions in the area. Laser mi . crodissection of additional archival samples will serve the same purpose. Potential coding sequences will be isolated from these BACs by exon amplification. In addition to the end sequences of the isolated BACs, these potential exons will help us to put the BACs in physical order, to isolate additional BACs and eventually to achieve full genomic coverage of the minimal region of loss. Physically isolated transcripts will be evaluated in parallel with ESTs and genes deposited in the databases by mutational analysis in cell lines, tumor samples and eventually the germ line of ovarian cancer patients. After having successfully identified an ovarian carcinoma TSG we will set out to support our mutational analysis with functional data.

Carcinogenesis is driven by the gain of function of oncogenes and the loss of function of so-called tumor suppressor genes (TSGs). Recently, we and others could show that a loss of genetic material on the short arm of chromosome 8 occurs in ovarian cancer frequently. This observation is a strong indicator that this chromosomal region harbours potential TSGs which might play a causative role in ovarian cancer. In the previous years, we have carefully mapped smallest regions of overlap (SRO) in this region and have isolated bacterial artificial chromosomes (BAC) from human DNA libraries to physically cover one SRO on 8p21-22. Consequently, we have used powerful molecular genetic and bioinformatic tools to isolate potential coding sequences. Now, we are working on potential genes predicted by bioinformatic tools from sequences of the region available from the Human Genome Project, on expressed sequence tags (ESTs) mapping to the region and on exons isolated by exon trapping. The evaluation of upcoming TSG candidate genes is challenging since the demonstration of mutations or other forms of inactivation like methylation in the relevant genes is labour intensive. One option is to search systematically for homozygous deletions (HD). HD are usually smaller than regions where one allel is lost and are -most of the time- a prerequisite for the successful isolation of a TSG by reducing the number of potential candidate genes which have to be screened. While we are developing new high-throughput-tools to define regions of homozygous loss by chip technology (microarray CGH), we were lucky by identifying a region of homozygous loss in a pancreatic carcinoma cell line. Covering a region of only 685 kilo bases it will be relatively straightforward to isolate and characterize a defined number of genes from this comparatively small area. Besides the search for novel genes, we are also working candidate genes mapping to the more extended region like the receptors for TNF-related apoptosis inducing ligand (TRAIL). In combination, these approaches should shed light on the underlying cause for the frequent loss of genetic material from 8p21-22 in human tumours, leading to the identification of novel genes involved in oncogenesis, a better understanding of tumorigenesis and ultimatively to better diagnostic and therapeutic options for patients with ovarian cancer.