Year: 2012
Type: dissertation
Abstract: Cancer is a disease characterized by the continuous accumulation of somatic cellular aberrations, whether in the DNA or epigenetic changes. During the last decade, many different techniques, including high resolution microarrays as well as exome- and whole genome sequencing, have been developed to comprehensively characterize these changes in cancer cells. In parallel with the development of laboratory techniques, a large variety of bioinformatic methods to analyze data from these have been developed. However, many of these concentrate on the analysis of data from only one laboratory technique, while it is becoming clear that advances in cancer research increasingly depend on integration of multiple different data types for the same tumors. Simultaneously, the recent explosive growth in sequencing data requires the development of new analythical methods. The aim of this thesis was to further characterize the genomic changes in breast cancer, with an emphasis on the development and application of bioinformatic methods to analyze and integrate data from different high throughput analysis techniques. In the first part of this work, the Gene Identification by Nonsense Inhibition method was applied to identify potential tumor suppressor genes in breast cancer. The integration of steady state gene expression, transcript stabilization and array comparative genomic hybridization data for six breast cancer cell lines led to the identification of a nonsense mutation in the RIC8A gene located at 11p15, a region deleted in ~15% of breast tumors. Taken together, our results suggest loss of RIC8A expression may be important in a subgroup of aggressive breast cancers. In study II, we developed a bioinformatic method for highly specific fusion gene identification from paired-end RNA-seq data. Application of the bioinformatic pipeline to data from four breast cancer cell lines led to the identification of 24 novel and three previously published fusion genes, with 95% specificity. In addition to showing that fusion genes are more prevalent in breast cancer than previously thought, several biological characteristics of fusion genes were identified. Most prominently, fusion genes were frequently associated with DNA copy number transitions, particularly high level amplifications, suggesting that most of them are not generated by balanced rearrangements. siRNA knock-down studies furthermore provided evidence for the functional importance of the VAPB-IKZF3 fusion gene in the BT-474 cell line. In the final study, we used aCGH to characterize the size distribution of the ERBB2 amplicon across 71 amplicon carrying tumors and 10 cell lines. To study the possible contribution to cancer of other coamplified genes in the amplicon, 23 genes amplified in 60% of tumors were selected for siRNA screening in two trastuzumab sensitive, two insensitive and one control cell line. In addition to single gene siRNA silencing experiments, five amplicon genes were knocked-down together with ERBB2 to identify synergistic effects. Our results suggest that cancer cells may be dependent on a number of genes in the ERBB2 amplicon besides the primary driver oncogene, a phenomenon termed non-oncogene addiction.
Source: Työväentutkimus Vuosikirja
Author: Henrik Edgren
Language: English
Subfield: Cancer Research
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W1554272132