The role of microRNAs in the EWS-FLI1 gene regulatory network of Ewing´s sarcoma

This project aims at identifying mechanisms and targets of gene regulation by EWS-FLI1 modulated microRNAs (miRNAs) in Ewing`s sarcoma family tumors (ESFT). It builds on extensive pre-existing genomic mRNA and miRNA data and will apply the latest technology in miRNA research to achieve its goals. EWS-FLI1 is a chimeric ETS transcription factor produced in ESFT as a result of chromosomal translocation. There is ample experimental evidence for EWS-FLI1 driving the pathogenesis of this aggressively malignant cancer which arises from a tissue closely related to mesenchymal stem cells (MSC). We have previously demonstrated that binding of EWS-FLI1 to gene promoter regions leads predominantly to transcriptional activation while distal binding of EWS-FLI1 is associated mostly with a silencer effect. The majority of EWS-FLI1 activated genes annotate to proliferation related functions while EWS-FLI1 repressed genes were found to play a role in cell communication, differentiation and development. As the mechanisms of gene repression by EWS-FLI1 remain unknown, we have recently started to explore the possibility that miRNAs may be involved in EWS-FLI1 mediated modulation of genes lacking proximal EWS-FLI1 binding. To study the role of EWS-FLI1 regulated miRNAs in ESFT pathogenesis, we performed a genome-wide screen for miRNAs that are affected by RNAi-mediated modulation of EWS-FLI1 in 5 ESFT cell lines, and differentially expressed between primary ESFT and MSC. The resulting miRNA signature was found to contain similar numbers of activated and repressed miRNAs. In silico target prediction suggested that EWS-FLI1 activated miRNAs are involved in cell cycle regulation, while EWS-FLI1 repressed miRNAs are predicted to target genes in development and cell signaling. Interestingly, among the top thirty EWS-FLI1 regulated miRNAs only three (hsa-mir-22-3p, hsa-mir-27-3p, and the hsa-mir-199/214 cluster) showed EWS-FLI1 binding to their promoters. Thus, for the majority of miRNAs, the mode of regulation by EWS-FLI1 remains unknown. In this project we will i) identify genes controlled by EWS-FLI1-regulated miRNAs based on their association with the RISC complex in ESFT and in MSC, ii) validate candidate target genes predicted to be the substrates for more than one EWS-FLI1 regulated miRNA, iii) validate in silico predicted targets for the directly EWS-FLI1 regulated miRNA cluster hsa-mir-199/214, iv) explore the mechanism(s) of indirect miRNA regulation by EWS-FLI1. The results of this study will add an important layer to our understanding of aberrant gene expression in ESFT. Since miRNAs and their regulation are considered promising diagnostic, prognostic and therapeutic targets, this project may identify novel treatment options for ESFT.

In this project we developed a paradigmatic experimental approach to the unbiased identification of microRNA regulated genes using Ewing sarcoma as a model. MicroRNAs are small (usually 21 to 23 nucleotides long) non-coding RNAs, which play an important role in gene regulation. They function predominantly as modulators of gene expression. Based on partial sequence complementarity at so called seed regions in the 3untranslated region of messenger RNA (mRNA), they guide the RNA induced silencing complex (RISC) to target mRNAs leading to protein synthesis arrest and to mRNA degradation. The identification of target genes for specific microRNAs in a specific cellular context still poses a significant problem. A single microRNA targets many different mRNAs which compete for microRNA binding. In addition, many mRNAs are targeted by several different microRNAs. Further, microRNA/mRNA interactions depend on a variety of physical parameters which all together complicate reliable microRNA target predictions. Here, we developed and validated a novel approach to genome-wide microRNA target identification by combining high throughput screening for RISC associated mRNA/microRNA complexes with RNA expression analysis by high throughput sequencing upon depletion of specific microRNAs. This way, we succeeded in defining the full complement of microRNA- bound mRNAs in a Ewing sarcoma model cell line. Consistent with microRNAs from the hsa- miR-17-92 cluster being most highly up-regulated by the Ewing sarcoma specific oncogene EWS-FLI1, we observed enrichment of seed sequences for this microRNA family among microRNA-bound mRNAs. Focusing on hsa-miR-17-92, we identified 7 known and 80 novel targets. Mutation analysis of seed sequences and reporter assays of selected targets validated our results and demonstrated superiority of our approach to common in silico target prediction algorithms. Finally, we discovered that about 25% of hsa-miR-17-92 targets functionally annotate to TGFB/BMP signaling. Thus, this project identified an important role of EWS-FLI1 regulated hsa-miR-17-92 in mesenchymal differentiation arrest and reprograming of developmental signaling cascades.