We propose a systems biology approach to understand how suites of mutations, identified by sequencing the transcriptomes of human tumor specimens, collaborate to re-wire signaling pathways and thus contribute to the malignant phenotype. Several recently published cancer sequencing projects suggest that each tumor carries a unique set of low frequency mutations. These data make it difficult to determine which mutations are responsible for the malignant phenotype (drivers) and which are merely generated in an unstable genome (passengers). Additionally, this heterogeneity may mask higher order similarities that can be identified and exploited for cancer therapy. Resolving these issues is essential to develop information generated by sequencing into novel cancer therapeutics. We believe that nature has given us some clues to commonalities in this complexity: tumor subtypes that have similar morphology and biological behavior will harbor a similar set of mutations. However, it may be that the similarities are not at the level of individual mutations;there may be several means to the same end. To develop a framework for developing next generation sequencing data into novel therapeutic targets, we propose the following:
Aim 1 : Sequencing transcriptomes from 24 triple negative breast carcinomas and 24 ER positive breast carcinomas and bioinformatic analysis to identify mutations and gene fusions. A bioinformatic pipeline we have built will identify mutations and novel fusion genes.
Aim 2. Network Analysis of Signaling in Triple Negative Breast Carcinomas. We will use Solexa sequencing to identify mutations in triple negative and ER positive cancer cell lines. We will use a novel high throughput Western blot approach to measure levels hundreds of phophsoproteins in these cell lines in response to growth factors. Differences in responses can be associated with differences mutational patterns.
Aim 3. Analysis of mutations and gene fusions in breast cancer cell line models. We will identify cell lines with a suite of mutations similar to that seen in our tumors. Using a knockdown/overexpression format, we will determine the importance of these mutations to the cancer phenotype. This research will significantly impact the future of breast cancer treatment by identifying novel targets for therapeutic intervention in triple negative breast carcinomas. More generally, we will establish a framework for developing next generation sequencing information into clinically relevant information.