Novel Pharmacogenomic Approaches for Understanding & Targeting Cancer

Oncogenic signaling to the translation machinery.

Decades of research have implicated a specialized network of transcripts that control cancer development. An outstanding question is whether translationally regulated nodes of gene expression can also direct cancer initiation and progression, thereby mirroring the cooperativity that has mainly been observed at the level of transcription control. To address this question, our research is delineating the effects of one of the most important oncogenic signaling pathways, PI3K-AKT-mTOR, in controlling the translational landscape of gene expression within a cancer cell.

By generating the first mouse models for core components of the translation machinery downstream of PI3K-AKT-mTOR, we find that overexpression of eIF4E leads to tissue-specific cancer development. These findings were seminal in demonstrating that hyperactivation of the core translation machinery is oncogenic in vivo. These findings also enabled us to answer a long-standing question of why mTOR inhibitors, such as rapamycin, exhibit surprisingly limited clinical efficacy in the treatment of human cancers.

How does the translational landscape downstream of oncogenic PI3K signalling contribute to cancer?

The translational landscape of the cancer genome is very poorly defined. We employed ribosome profiling to delineate for the first time at a codon-by-codon resolution the immediate effects of mTOR inhibition downstream of PI3K-Akt oncogenic signaling in decoding the prostate cancer genome into functional proteins. This revealed a translationally controlled gene expression signature critical for prostate cancer metastasis, for which presently there is no cure. Our current work is testing the hypothesis that cancer cells rely on translation control as a means to maintain high expression levels of specific mRNAs during the unfavorable conditions that these cells encounter when they bypass tissue barriers, migrate, and invade distal secondary sites.

Another outstanding question we seek to define is whether translationally regulated mRNAs possess specialized cis-acting translational elements that may confer specificity downstream of mTOR signaling. Strikingly, we identified a unique signature within the 5’UTR of these mRNAs, the presence of which we have termed a pyrimidine-rich translational element (PRTE). These findings were instrumental in revealing a regulatory “code” of translational control in which specific regulatory elements fine-tune protein abundance, culminating in unique readouts with important biological significance. Our short-term goal is to biochemically characterize the trans-acting factors that may bind to the PRTE element to promote translation initiation of subsets of mRNAs.