The spliceosome is a collection of protein and non-coding RNA subunits, interacting to bind, cleave, and ligate RNA. Alternative splicing can contribute to cancer development through the expression of novel protein isoforms. Oncogenic driver genes that modulate splicing arise from mutations, over-expression, and chromosomal translocations in leukemias, carcinomas, and sarcomas. EWS-FLI1 is one such oncogenic fusion protein derived from a tumor-specific chromosomal translocation in Ewing sarcoma (ES). Previous investigations have described the modulation of transcription by EWS-FLI1 and connected transcription with its oncogenic potential, yet some mutants that do not bind DNA still have oncogenic activity. This suggests EWS-FLI1 has oncogenic capacity outside of transcriptional regulation. We have found that EWS-FLI1 interacts with spliceosomal proteins and significantly alters the isoform landscape in ES. Others have shown some splicing factors are critical for EWS-FLI1 oncogenesis. Yet, the contribution of splicing to ES oncogenesis as well as the role of EWS-FLI1-interacting proteins in the spliceosome, remain unknown. EWS-FLI1 was often termed an `undruggable' target. To develop alternative strategies for therapeutic targeting of EWS-FLI1, we identified compounds that directly bind to EWS-FLI1 and inhibit its interaction with specific partners. In 2009 we reported one such compound, named YK-4-279, that blocks the EWS- FLI1 binding to a key protein partner. ES cells treated with YK-4-279 show altered splicing patterns that mimic EWS-FLI1 loss. An analog of YK-4-279, TK216, is now in phase I clinical trials in ES patients. We therefore hypothesize that regulation of RNA splicing of a small number of critical genes is a rate- limiting oncogenic mechanism of EWS-FLI1 in addition to its canonical role as a transcription regulator. We focus this proposal on three aims. (1) We will determine the relative effects of EWS-FLI1 mutants on transcription and splicing through characterizing key domains and residues. We will then determine the effects of these mutants on oncogenesis. (2) We will define interactions between EWS-FLI1 and splicing factors required for differential splicing. (3) We will investigate how EWS-FLI1-induced splice isoform switching of target genes contributes to oncogenesis. We demonstrated that EWS-FLI1 induces differential splicing of a number of target genes in human mesenchymal stem cells (hMSC); now we will identify key domains and residues in EWS-FLI1 that induce differential splicing. Our approach will also answer whether EWS-FLI1 creates de novo splice variants that are uniquely found in ES or whether EWS- FLI1 is part of a pathway that leads to a spliceosome with novel splicing activities similar to those occurring in myelodysplastic syndromes. Detailed knowledge of splicing drivers that are altered in specific tumors will enhance our understanding of oncogenesis, lead to stratification markers for personalized medicine, and inform approaches to new anti-cancer targets.
New targeted therapies are needed for cancer patients that improve survival and decrease side- effects of therapy by improving recognition of tumor as distinct from host. We show novel data that supports EWS-FLI1 as having a significant role is splicing from cell lines to patient tumors. The modeling of splicing from normal tissue to ES tumor will both inform the process of oncogenic splicing and potentially provide novel therapeutic targets.
