It is now appreciated that both coding and non-coding RNAs contain a large variety of post-transcriptional nucleotide modifications that change the chemical composition of RNA. The RNA component of the ribosome, ribosomal RNA (rRNA) is the most abundant type of RNA in all domain of life and contains the highest amount of base modifications, however the function of these modifications and whether they are subject to regulation remains poorly understood. Thus, an outstanding question is whether individual pseudouridine modifications influence the ability of the ribosome to translate the transcriptome and more importantly whether the pattern of Ψ modifications on rRNA is differentially regulated. To this end we sought to determine the landscape of H/ACA snoRNA expression during the earliest cellular responses to oncogenic insult such as DNA damage. Unexpectedly, upon profiling H/ACA small RNAs in primary human fibroblasts, we observed coordinated regulation of specific subsets of H/ACA snoRNAs guiding modifications within similar regions of rRNA following DNA damage. The observed coordinated regulation in H/ACA snoRNA expression is functionally important as we observed a corresponding change in site-specific rRNA Ψ modifications within specific regions of the ribosome. Furthermore, we uncovered that loss of function of one snoRNA, SNORA24, and inhibition of its target rRNA Ψ modifications alters the translation of mRNAs implicated in lipid metabolism both in vitro and in vivo to maintain senescence, a stress response that acts as a critical barrier against cellular transformation. Rewired lipid metabolism results in increased lipid synthesis allowing for bypass of senescence in response to DNA damage in primary human fibroblasts and upon activation of oncogenic N-RasV12 in mouse liver. We further probed whether increased lipid synthesis upon loss of SNORA24 contributes to cellular transformation and tumor development. Strikingly, we observe that loss of function of SNORA24 cooperates with oncogenic N-RasV12 to develop hepatocellular carcinoma with significant accumulation of lipid droplets. These exciting findings provide the first evidence that the pattern of RNA base modifications within the ribosome are dynamically regulated in response to oncogenic insult and suggests a new layer of ribosome-mediated control in safeguarding the genome against cellular transformation.

Malignant progression in glioma and other cancers is associated with the acquisition of a mesenchymal phenotype and with poor prognosis. To test if radiotherapy, as standard of care in glioma, promotes a mesenchymal transition, we irradiated proneural tumors arising in a genetically engineered mouse model for high-grade glioma. Cranial ionizing radiation induced a robust and durable proneural-to-mesenchymal transition in tumors in vivo. Radiation of primary proneural glioma cells derived from mouse and human tumors also induced a cell-intrinsic and sustained proneural-to-mesenchymal transition in vitro. Following irradiation, glioma cells undergoing a mesenchymal transition displayed increased resistance to alkylating chemotherapy, and showed increased invasion and motility. The transcription factor STAT3 has been identified as a master regulator of the mesenchymal transcriptional network in high-grade glioma tumors. We demonstrate that cell-intrinsic activation of STAT3 activation mediates a proneural-to- mesenchymal transition that confers radiation resistance in glioma. Pretreatment with inhibitors of JAK2, an upstream kinase that activates STAT3, blocked this mesenchymal transition. Our results suggest that therapeutic blockade of the JAK2-STAT3 pathway concurrent with current therapy in patients with proneural high-grade glioma may block the emergence of therapy-resistant mesenchymal tumors at relapse.

Although it was first observed nearly a century ago that transformed cells are marked by enlarged nucleoli, where ribosomal RNA synthesis and ribosome assembly occur, it has only recently become clear that deregulated protein synthesis plays a causal role in the development and progression of cancer. Indeed, aberrant mRNA translation has emerged as a hallmark of oncogenic transformation, underlying many cellular functions critical for tumorigenesis. Oncogenic signaling has been shown to enhance and alter the protein synthesis capacity of cancer cells to directly contribute to their survival, proliferation, metastasis, and genome instability. Therefore, inhibiting enhanced protein synthesis may represent a highly relevant strategy for the treatment of human cancers. For example, Myc is one of the most commonly deregulated oncogenes in human cancer, yet therapies directly targeting Myc hyperactivation are not presently available in the clinic. Although the ability to modulate protein synthesis control is critical to the Myc oncogenic program, components of the translation machinery that can be exploited as therapeutic targets for Myc-driven cancers remain poorly defined. Recently, the translation initiation factor eIF4E, which is considered the quantitatively rate limiting factor for translation initiation, has become a central focus for therapeutic approaches seeking to inhibit oncogenic activation of translation. Despite this, the requirements for eIF4E in normal mammalian development and physiology are largely unknown. Moreover, we still do not understand the specific roles that eIF4E plays in translating the genome of both normal and cancer cells.

Here, we uncover a surprising and important functional link between Myc and mammalian target of rapamycin (mTOR)-dependent phosphorylation of eukaryotic translation initiation factor 4E binding protein-1 (4EBP1), a master regulator of eIF4E-dependent protein synthesis control. Using a pharmacogenetic approach, we find that mTOR-dependent phosphorylation of 4EBP1 is required for cancer cell survival in Myc-dependent tumor initiation and maintenance. We further show that a clinical mTOR active site inhibitor, which is capable of blocking mTOR-dependent 4EBP1 phosphorylation, has remarkable therapeutic efficacy in Myc-driven hematological cancers. Moreover, we demonstrate the clinical implications of these results by delineating a significant link between Myc, mTOR-dependent phosphorylation of 4EBP1, and therapeutic response in human lymphomas. Together, these findings identify an important mTOR substrate hyperactivated downstream of Myc oncogenic activity, which promotes tumor cell survival and confers synthetic lethality, thereby revealing a unique therapeutic approach to render Myc druggable in the clinic.

Additionally, we have generated a novel Eif4e haploinsufficient mouse, allowing the requirements for eIF4E dose to be defined at the organismal level for the first time. Surprisingly, we find that a 50% reduction in eIF4E is compatible with normal development and global protein synthesis. Strikingly, however, Eif4e+/- cells and mice are remarkably resistant to oncogenic transformation. Unbiased genome-wide translational profiling of mRNAs induced by oncogenic transformation reveals eIF4E dosage is critical for translating key mRNA networks, including redox balance, signaling, and proteasome control, which are demarcated by a novel 5'UTR signature. In particular, eIF4E dose is essential for translating a novel class of mRNAs regulating reactive oxygen species (ROS), a hallmark of oncogene-induced stress. Translational control of intracellular ROS fuels cancer cell survival and underlies Eif4e+/- resistance to cellular transformation. Therefore, mammalian cells have evolved surplus eIF4E levels that are usurped by cancer cells to drive a translational program supporting tumor growth and survival.

Protein translation is necessary for cell function, but it is an incredibly energy demanding process, and is therefore tightly regulated by the metabolic state of the cell. There are a plethora of translation control mechanisms that are only recently being elucidated. My thesis research has investigated how perturbing the metabolic state of the cell, both subtly via autophagy inhibition and with a sledge-hammer of acute amino acid starvation, impacts translation rates on both a global and mRNA by mRNA basis. Overall, I found that these stresses do not repress translation as expected, indicating the identification of novel mechanisms of protein translation regulation.

The majority of my thesis focused on the role of autophagy in regulating protein translation. Autophagy, a cellular sorting, degradation and recycling system, is crucial for the survival of cells under stress and has been demonstrated to play a role in the progression of many human diseases, including cancer and neurodegeneration. By promoting protein degradation, autophagy is proposed to maintain amino acid pools to sustain protein synthesis during metabolic stress. I utilized ribosome profiling to delineate the effects of acute genetic ablation of autophagy on protein translational control. Instead of shaping overall global rates of cap dependent translation, autophagy supports the translation of specific mRNAs, most notably targets involved in cell cycle control and DNA damage repair, by modulating the availability of RNA binding proteins to interact with mRNAs. Specifically, by enabling the protein translation of the DNA damage repair protein BRCA2, autophagy is functionally required to attenuate DNA damage as well as promote cell survival in response to PARP inhibition. This helps to explain the reported increased DNA damage in autophagy deficient cells, and is an important consideration for autophagy inhibitors as adjuvant chemotherapies, which are being tested now.

I have also uncovered a novel mechanism of protein translation regulation following acute amino acid starvation. Although mTORC1 signaling indicates repressed translation, 35S-methionine incorporation rates more than double following amino acid withdrawal. This increase in translation rates can be prevented by addition of leucine, although the molecular mechanisms controlling this novel process remain to be identified.