Analysis of several hundred glioblastoma samples compiled by the TCGA (The Cancer Genome Atlas) produced an extensive transcriptomic map, identified prevalent chromosomal alterations and defined important driver mutations. However, as of today, clinical trials based on these results have not delivered an improvement on outcome. Therefore, we decided to characterize other regulatory routes known for playing a role in tumor relapse and response to treatment.
We selected splicing regulation for the following reasons:
- Alternative splicing affects 90% of the transcriptome and is an important source for transcript variation and gene regulation;
- Numerous genes involved in apoptosis, proliferation, migration and DNA repair display cancer specific splicing isoforms with functions distinct from the ones in normal tissue;
- Mutations and alterations in splicing factors are highly prevalent in multiple cancers and can act as tumor drivers;
- Genomic instability, a common characteristic of cancer, can be induced by splicing defects;
- The splicing machinery is targetable: there are numerous examples of drugs that either inhibit splicing factors or promote changes in splicing;
- Importantly, no comprehensive studies have been performed to study splicing regulation in GBM.
Using resources from TCGA (The Cancer Genome Atlas) and GTex (Genotype-Tissue Expression), we have analyzed the expression profile of splicing regulators in normal and brain tumor tissues and produced detailed maps of splicing alterations in cancer cells. Our current plans are to link splicing regulators to oncogenic signals required for transformation, identify critical splicing isoforms and evaluate their contribution to gliomagenesis.
Genome-scale knowledge of translational regulation has lagged behind that of transcription, despite a central role determining cell phenotype, and major implications for numerous diseases and cancer. The assay of choice for global gene expression profiling is RNA-seq, which makes sense for understanding transcriptional control. But levels of mRNA in a cell explain only a fraction of observed protein levels, and much of the remainder is due to translational control. Ribosome profiling (RP) or Ribo-seq is a novel genomic approach that delivers quantitative information on the number and behavior of ribosomes, and gives profiles of gene expression much more closely linked to actual protein levels. We are using Ribo-seq aligned with computational tools to identify alterations in translation regulation in cancer cells and study its behavior upon drug and radiation treatments. We are also interested in examining the role of aberrantly expressed ribosomal proteins in tumor development and determine if they affect translation in specific fashions.