Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Feb 16.
Published in final edited form as: Cell Rep. 2016 Jan 14;14(3):598–610. doi: 10.1016/j.celrep.2015.12.063

Functional genomic screening reveals splicing of the EWS-FLI1 fusion transcript as a vulnerability in Ewing sarcoma

Patrick J Grohar 1,2,3,10, Suntae Kim 4,10, Guillermo O Rangel Rivera 4,5, Nirmalya Sen 4, Sara Haddock 4, Matt L Harlow 3, Nichole K Maloney 1, Jack Zhu 6, Maura O’Neill 7, Tamara L Jones 4, Konrad Huppi 4, Magdalena Grandin 4, Kristen Gehlhaus 4, Carleen A Klumpp-Thomas 8, Eugen Buehler 8, Lee J Helman 9, Scott E Martin 8, Natasha J Caplen 4,11
PMCID: PMC4755295  NIHMSID: NIHMS747179  PMID: 26776507

Summary

Ewing sarcoma cells depend on the EWS-FLI1 fusion transcription factor for cell survival. Using an assay of EWS-FLI1 activity and genome-wide RNAi screening, we have identified proteins required for the processing of the EWS-FLI1 pre-mRNA. We show Ewing sarcoma cells harboring a genomic breakpoint that retains exon 8 of EWSR1 require the RNA-binding protein HNRNPH1 to express in-frame EWS-FLI1. We also demonstrate the sensitivity of EWS-FLI1 fusion transcripts to the loss-of-function of the U2 snRNP component, SF3B1. Disrupted splicing of the EWS-FLI1 transcript alters EWS-FLI1 protein expression and EWS-FLI1 driven expression. Our results show that the processing of the EWS-FLI1 fusion RNA is a potentially targetable vulnerability in Ewing sarcoma cells.

Introduction

Chromosomal translocations that generate fusion genes encoding oncogenic transcription factors are associated with the initiation and maintenance of many cancers, including leukemias and epithelial and mesenchymal solid tumors (Mitelman et al., 2007). Ewing sarcoma (ES) is an aggressive cancer of the bone and soft tissue (Hawkins et al., 2010). The primary oncogenic event in ~85% of ES tumors is a t(11:22)(q24:q12) translocation (Delattre et al., 1992). This translocation generates a fusion gene containing the 5’ end of the EWSR1 gene and the 3’ end of the FLI1 gene referred to as EWS-FLI1 (Delattre et al., 1992; May et al., 1993a). The EWS-FLI1 transcript encodes the transcription factor EWS-FLI1 that is responsible for malignant transformation and is necessary for ES cell survival (Bailly et al., 1994; May et al., 1993a; May et al., 1993b). Molecules that either suppress the expression or inhibit the activity of an oncogenic transcription factor have the potential to block cancer cell growth selectively. In this study, we used a genome-wide RNAi screen of a cell-based reporter assay to identify proteins required for EWS-FLI1 activity. This screening strategy revealed the sensitivity of ES cells to the reduced expression of proteins required for the processing and maturation of the EWS-FLI1 fusion transcript.

The genomic breakpoints within the EWSR1 and FLI1 genes that give rise to the expression of EWS-FLI1 vary (Zucman et al., 1992; Zucman et al., 1993; Zucman-Rossi et al., 1998). The breakpoint cluster region (BCR) in EWSR1 is small (~5 kb) but spans several exons (exons 7 to 11). The BCR in FLI1 is much larger (over 30 kb), extending from exons 4 to 9. The EWS-FLI1 transcript observed most frequently consists of a fusion of exons 1–7 of EWSR1 to exons 6–9 of FLI1, referred to as a 7/6 or type 1 fusion. Another, rarer EWS-FLI1 transcript consists of the fusion of EWSR1 exons 1–7 to FLI1 exons 5–9, referred to as a 7/5 or type 2 fusion. The breakpoints observed in ES tumors occur typically within introns of EWSR1 or FLI1 and expression of EWS-FLI1 requires the splicing machinery to generate an in-frame transcript. Specifically, translocations that retain exon 8 of EWSR1 generate an out-of-frame transcript unless this exon is removed (Berger et al., 2013; Crompton et al., 2014; Patocs et al., 2013; Zoubek et al., 1994; Zucman et al., 1993). A recent study showed 15 of 42 ES tumors harbored translocations in which the EWSR1 exon 8 must be spliced out to express an in-frame EWS-FLI1 transcript (Berger et al., 2013). Here, we demonstrate that the heterogeneous nuclear ribonucleoprotein H1 (HNRNPH1) is required for the splicing of EWS-FLI1 transcripts expressed in ES cells in which the breakpoint retains EWSR1 exon 8. We also show ES cell lines harboring 7/6 or 7/5 EWS-FLI1 fusions are sensitive to the inhibition of the core splicing factor, SF3B1. This study establishes splicing as a vulnerability that could be exploited for the development of therapeutic strategies for ES.

Results

RNA processing proteins required for EWS-FLI1 activity are identified by genome-wide RNAi screening

Parallel genome-wide RNAi screens were conducted in TC32 ES cells, expressing a luciferase (luc) reporter protein driven by either the promoter of the EWS-FLI1 regulated gene NR0B1 (TC32-NR0B1-luc) or the CMV promoter (TC32-CMV-luc) (Grohar et al., 2011) (Figure S1A). Luciferase activity was assessed 48 hour (h) post siRNA-transfection. Assay optimization and screen quality control data are presented in Figure S1B–D. Genome-wide TC32-NR0B1-luc and TC32-CMV-luc RNAi screening data were normalized and z-score transformed (Chung et al.,2008). Next, the miRNA-like seed sequence of each siRNA was determined and used to generate an adjusted z-score value (Z) (Buehler et al., 2012). Comparison of the Z values for each siRNA in each screen distinguished siRNAs affecting transcription in a non-specific manner (e.g. POLR2A) from those siRNAs that target the EWS-FLI1 fusion transcript directly (Figure 1A and 1B). To prioritize genes for confirmation, we focused on genes for which the median seed adjusted Zdiff (ZNR0B1-ZCMV) for the 3 siRNAs per gene was either < −1.5 (the top 1% of candidates) or > 2 (the top 0.25% of candidates). We selected 183 genes (plus EWSR1 and FLI1) for a confirmatory screen, 137 genes that when silenced selectively reduced the TC32-NR0B1-luc signal (Group 1) (Figure 1C) and 46 genes that selectively increased the TC32-NR0B1-luc signal (Group 2) (Figure S1E) relative to the TC32-CMV-luc signal. In most cases, an additional 4 siRNAs per gene were tested (Table S1). We defined a gene as a priority candidate if at least 2 further siRNAs mediated a Zdiff <-1.5 for Group 1 genes or a Zdiff > 1.5 for Group 2 genes in the reporter assay. Based on these criteria, 26 Group 1 genes and 16 Group 2 genes were supported by 2 or more siRNAs. The top gene ontology terms associated with the 26 Group 1 genes were mRNA splicing (p-value [Bonferroni corrected] = 1.42−08) and mRNA processing (p-value [Bonferroni corrected] = 2.32−07). RNA processing genes identified included SF3B1, SF3A1, SNRPD1, and SNRPD2 that encode components of the U2 snRNP. The transcription elongation factor and histone chaperone, SUPT6H, and a non-core alternative splicing factor, the heterogeneous nuclear ribonucleoprotein, HNRNPH1, were also identified as required for EWS-FLI1 activity. Further analysis of the effects of silencing SF3B1, SUPT6H, or HNRNPH1 in the TC32 reporter assay (6 siRNAs per gene, see Supplemental Experimental Procedures) confirmed that the depletion of these proteins induces a selective decrease in the NR0B1-luc reporter (Figure 1D).

Figure 1. Identification of proteins required for EWS-FLI1 activity by genome-wide RNAi screening.

Figure 1

Open in a new tab

(A) A genome-wide RNAi screen of EWS-FLI1 activity. Data are shown as the median Z-score value for the 3 siRNAs corresponding to each gene (~21,000 genes) except for POL2RA (light blue circles), EWSR1(magenta diamonds) and FLI1 (dark blue squares) for which the Z-score value of each of the 3 siRNAs targeting each gene are shown.

(B) The Zdiff (ZNR0B1-ZCMV) for the EWSR1 and FLI1 siRNAs included in the screen and a schematic of the sites in EWSR1, FLI1, and, EWS-FLI1 targeted by these siRNAs.

(C) The median Zdiff (ZNR0B1-ZCMV) (3 siRNAs per gene) for the 139 genes, selected for follow up analysis, that exhibited a selective decrease in the TC32-NR0B1-luc reporter when silenced.

(D) Ratio of the TC32-NR0B1-luc and TC32-CMV-luc reporter signals, 72 h post siRNA-transfection (mean ± standard deviation (SD), n=5). ** p<0.01; *** p<0.001 compared to siNeg.

See also Figure S1 and Table S1.

EWS-FLI1 driven expression is altered by the silencing of SF3B1, SUPT6H, or HNRNPH1

To investigate the mechanistic basis for the identification of SF3B1, SUPT6H, or HNRNPH1 as required for the activity of EWS-FLI1 we confirmed their silencing by multiple siRNAs (Figure S2A) and then selected 3 siRNAs per gene to generate expression profiles of silenced TC32 cells (Figure S2B–D). To determine if silencing SF3B1, SUPT6H, or HNRNPH1 results in specific changes in expression of genes that are deregulated in ES or a broad, pleiotropic effect because of their importance in the processing of RNA, we conducted gene set enrichment analysis (GSEA) using a set of genes previously shown to exhibit altered expression in ES (Figure 2A). For this analysis, we used a list of ~1200 genes assembled using multiple datasets that encompass both genes regulated by EWS-FLI1 directly and indirectly (Table S2) (Hancock and Lessnick, 2008; Kauer et al., 2009). The transcriptome of SF3B1-silenced TC32 cells showed no enrichment for alterations in the expression of genes specifically deregulated in ES (Figure 2A, upper panels). In contrast, the transcriptome of HNRNPH1-silenced TC32 cells was highly enriched for the down-regulation of genes over-expressed in ES and up-regulation of repressed genes (FWER p-value <0.001 for all analyses performed) (Figure 2A, lower panels). These findings were confirmed using additional datasets of EWS-FLI1 regulated genes (Bilke et al., 2013; Riggi et al., 2014) (Figure S2E). The transcriptome of SUPT6H-silenced TC32 cells also indicated reversal in the expression of genes exhibiting altered expression in ES (Figure 2A, middle panels).

Figure 2. EWS-FLI1 driven expression in TC32 cells is altered by the silencing of SF3B1, SUPT6H, or HNRNPH1.

Figure 2

Open in a new tab

(A) GSEA of the transcriptome of SF3B1, SUPT6H, or HNRNPH1-silenced TC32 cells (48 h) using a set of genes up-regulated (751) or down-regulated (494) in ES. NES = normalized enrichment score, FDR = false discovery rate, FWER = family-wise error rate.

(B) Heat-map representation of the fold-change (Log2) in the expression of genes deregulated in ES following silencing of SF3B1, SUPT6H, or HNRNPH1 in TC32 cells (48 h).

(C) qPCR assessment of EWS-FLI1 expression in HNRNPH1, SF3B1, or SUPT6H–silenced TC32 cells (48 h). Data are expressed relative to siNeg-transfected cells (mean ± SD, n=3).

(D) Immunoblot analysis of whole-cell lysates prepared from ES cells 48 h post transfection of HNRNPH1, SF3B1, or SUPT6H siRNAs using antibodies against the proteins indicated.

* p<0.05; ** p<0.01; *** p<0.001compared to siNeg.

(E) Relative viability of TC32 cells 48 and 72 h post-siRNA transfection (siNeg median normalized, mean ± SD, n=5).

See also Figure S2 and Tables S2 and S3.

We next focused on those genes that exhibited differential expression in HNRNPH1-, SUPT6H-, or SF3B1-silenced TC32 cells. In total 2,840 genes were considered differentially expressed (fold-change > ±1.5, FDR < 0.05, 2 or more siRNAs per gene; Table S3 and Figure S2F), of which 652 exhibited altered expression following silencing of 2 or more genes (Figure S2G). To identify genes related to ES biology within the set of 2,840 differentially expressed genes, we overlapped our data with the results of a study detailing genes deregulated in ES (Kauer et al., 2009) (Figure 2B). This analysis demonstrated reversal of the expression of many genes deregulated by EWS-FLI1, including decreased expression of NR0B1 in SUPT6H or HNRNPH1-silenced TC32 cells (Figure S2D). Importantly, though the transcriptome of SF3B1-silenced TC32 cells showed no enrichment for the altered expression of ES specific genes or decreased expression of NR0B1, several genes recently reported as activated by EWS-FLI1 (Riggi et al., 2014) were down-regulated following silencing of SF3B1 (Figure S2H). Overall, these data suggest that the silencing of SF3B1, SUPT6H, or HNRNPH1 results in some selective changes in the expression of genes that are deregulated in ES. We next assessed whether each protein is required for the expression of the EWS-FLI1 fusion transcript itself using qPCR primers spanning the fusion breakpoint (EWSR1 exon 7/FLI1 exon 7, E7/F7). Depletion of SF3B1 mediated a 5.5-fold decrease in EWS-FLI1 expression, while depletion of HNRNPH1 resulted in a 3.75-fold decrease and depletion of SUPT6H a 1.9-fold decrease (Figure 2C).

Analysis of EWS-FLI1 protein (Figure 2D) showed silencing of HNRNPH1 in TC32 cells mediated an almost complete elimination of oncoprotein expression. Silencing of SUPT6H also decreased EWS-FLI1 protein levels, but to a lesser extent. Interestingly, while qPCR detected a decrease in EWS-FLI1 RNA levels following silencing of SF3B1, immunoblot analysis of EWS-FLI1 detected multiple products, leading us to speculate that silencing of SF3B1 results in expression of EWS-FLI1 transcript variants undetected by the E7/F7 primers. Overall, these observations are consistent with the reduced expression of SF3B1, SUPT6H, or HNRNPH1 leading to altered expression of the EWS-FLI1 mRNA, the EWS-FLI1 protein and thus its activity. ES cells depend on the activity of EWS-FLI1 for survival and so disruption of the maturation of the EWS-FLI1 transcript should result in a reduction in viability of ES cells. This was borne out by analysis of SF3B1-, SUPT6H-, or HNRNPH1-silenced TC32 cells that exhibited a similar reduction in viability as EWS-FLI1-silenced cells (Figure 2E).

HNRNPH1 is required for the splicing of EWS-FLI1 in Ewing sarcoma cell lines with a genomic breakpoint that retain EWSR1 exon 8

The chromosome 22 translocation breakpoint in the TC32 cell line is within intron 8 of EWSR1 (May et al., 1993a). Thus, to generate an in-frame EWS-FLI1 transcript, exon 8 of EWSR1 must be excised by splicing (Berger et al., 2013; May et al., 1993a). To determine if the role of HNRNPH1, SF3B1, and SUPT6H in the maturation of the EWS-FL11 transcript applied to just TC32 cells or whether these proteins are required for the processing of the EWS-FLI1 transcript more broadly, we silenced each gene in additional ES lines representing different ES translocations and transcript types (Figure 3A).

Figure 3. HNRNPH1 is required for the splicing of EWS-FLI1 in Ewing sarcoma cell lines with a genomic breakpoint that retains EWSR1 exon 8.

Figure 3

Open in a new tab

(A) Schematic of the organization of the EWSR1 and FLI1 genes and EWS-FLI1 fusion transcripts in ES cell lines. EWSR1 exons are indicated in magenta and FLI1 exons in blue. The EWSR1 and FLI1 exon count are based on NM_05243 and NM_002017 reference sequences.

(B) Expression of SF3B1 and EWS-FLI1 (E7/F7 primer pair) in SF3B1-silenced ES cells (48 h). Data are expressed relative to siNeg-transfected cells (mean ± SD, n=3).

(C) Expression of HNRNPH1 and EWS-FLI1 (E7/F7 primer pair) in HNRNPH1-silenced ES cells (48 h). Data are expressed relative to siNeg-transfected cells (mean ± SD, n=3).

(D) PCR analysis of the splicing of EWS-FLI1 using E6/F6 primer pair and sequence analysis of the amplified products labeled a and b focusing on the junction of EWSR1 exons 7 and 8 and EWSR1 exon 8 and FLI1 exon 7.

(E) Immunoblot analysis of whole-cell lysates prepared from TC32 cells 48 h post siRNA-transfection using antibodies against the proteins indicated.

** p<0.01; *** p<0.001compared to siNeg.

See also Figure S3.

EWS-FLI1 expression was reduced in all 4 ES cell lines following silencing of SF3B1 (Figure 3B), and to a lesser extent following silencing of SUPT6H (Figure S3A). In contrast, silencing of HNRNPH1 in SKNMC cells mediated a decrease in the expression of EWS-FLI1, similar to TC32 cells but induced minimal changes in EWS-FLI1 expression in the other ES cell lines (Figure 3C). SKNMC cells harbor a similar chromosome 22 translocation as TC32 cells, and so these results imply HNRNPH1 is specifically required for the processing of EWS-FLI1 pre-mRNAs containing EWSR1 exon 8, while all ES cells require SF3B1 and SUPT6H to process EWS-FLI1. Because of their defined roles in splicing, we focused our subsequent studies on HNRNPH1 and SF3B1, beginning with analysis of the putative function of HNRNPH1 in the splicing of EWSR1 exon 8 to generate an in-frame EWS-FLI1 transcript in ES cells with a chromosome 22 translocation downstream of this exon.

To determine if HNRNPH1 is required for the splicing of EWSR1 exon 8 from the EWS-FLI1 pre-mRNA expressed in TC32 or SKNMC ES cells, we performed analysis using an EWSR1 exon 6 forward and FLI1 exon 6 reverse primer pair (E6/F6). Analysis of siNeg-transfected cells generated a single PCR product of the predicted size (Figure 3D) and sequence at the fusion breakpoint (Figure S3B). We also detected a single PCR product following silencing of HNRNPH1 in TC71 or RDES cells (Figure 3D). In contrast, silencing of HNRNPH1 in TC32 and SKNMC cells resulted in detection of an additional PCR product of the size expected if exon 8 of EWSR1 is retained (a and b in Figure 3D). Retention of EWSR1 exon 8 was confirmed by sequencing. The failure to exclude EWSR1 exon 8 is predicted to result in the generation of an out-of-frame transcript (Figure S3C) that does not encode the EWS-FLI1 oncoprotein (Figure 3E). This finding explains the altered expression of EWS-FLI1 regulated genes observed following whole transcriptome analysis of HNRNPH1-silenced TC32 cells (Figure 2A and B) and the identification of this protein as required for EWS-FLI1 activity by the RNAi screen.

The silencing of HNRNPH1 reverses the expression of genes deregulated by EWS-FLI1

To confirm the selective effect of the silencing of HNRNPH1 on the expression of EWS-FLI1 regulated genes in ES cells with a chromosome 22 translocation within EWSR1 intron 8, we examined the expression of 12 genes regulated by EWS-FLI1 (see Supplemental Experimental Procedures). All 12 genes exhibited significant changes in expression in HNRNPH1-silenced SKNMC cells (Figure 4A, upper panel). Critically, we observed no changes in the expression of the same genes when HNRNPH1 was silenced in TC71 cells (Figure 4A, lower panel). The selective effect of silencing of HNRNPH1 on EWS-FLI1 expression and thus its activity in TC32 and SKNMC cells was further validated when we examined EZH2, PRKCB, and VRK1 protein expression in these cell lines following silencing of HNRNPH1 versus HNRNPH1-silenced TC71 or RD-ES cells (Figure 4B). Importantly, the silencing of HNRNPH1 in TC32 cells resulted in a highly significant decrease in cell viability, similar to the direct silencing of EWS-FLI1, while the silencing of HNRNPH1 in TC71 resulted in minimal changes in cell viability (Figure 4C).

Figure 4. HNRNPH1 directly binds the EWS-FLI1 pre-mRNA expressed in TC32 cells.

Figure 4

Open in a new tab

(A) Expression of EWS-FLI1 regulated genes in HNRNPH1-silenced SKNMC or TC71 cells. Data are expressed relative to untransfected cells (mean ± SD, n=3). * p<0.05; ** p<0.01; *** p<0.001 compared to siNeg.

(B) Immunoblot analysis of whole-cell lysates prepared from ES cells 48 h post siRNA-transfection using antibodies against the proteins indicated.

(C) The viability of TC32 and TC71 cells 72 h post siRNA-transfection (siNeg median normalized, mean ± SD, n=10). p-values are compared to siNeg.

(D) UV-cross-linked RNA from TC32 cells (nuclear fraction) were subjected to RNA immunoprecipitation using an HNRNPH1 antibody or a control IgG isotype control. The fold-enrichment of co-precipitating RNA (HNRNPH1-bound versus IgG control) was determined by qPCR across regions of the EWS-FLI1 transcript (see Figure S4A). Data are shown as the mean ± standard error of the mean (SEM) of 3 experiments.

(E) Sequence of EWSR1 exon 8 indicating putative G-rich HNRNPH1 binding sites.

(F) The fold-enrichment for HNRNPH1-bound RNA oligomers determined by comparison of chemiluminiscent signals from the pull-down performed using a HNRNPH1 antibody and an IgG isotype control. Data are shown as the mean ± SEM of 3 experiments. p-values are compared to no oligomer/no antibody controls.

See also Figure S4.

HNRNPH1 binds the EWS-FLI1 pre-mRNA expressed in TC32 and SKNMC Ewing sarcoma cells

We next used an RNA pull-down strategy to determine if HNRNPH1 binds the EWS-FLI1 pre-mRNA expressed in TC32 cells directly (Figure S4A). In the nuclear fraction, we observed enrichment of EWS-FLI1 pre-mRNA in HNRNPH1 pull-down samples over the IgG pull-down control at EWSR1 exons 5 and exon 8 (Figure 4D). There was no enrichment of EWS-FLI1 RNA in the cytosol fraction (Figure S4B). The results of the RNA pull-down prompted us to next examine EWSR1 exon 8 for potential HNRNPH1 binding sites. The EWSR1 exon 8 sequence indicated 2 G-rich sequences consistent with the reported binding-motif of HNRNPH1 (Huelga et al., 2012) (Figure 4E). An in vitro protein-RNA-oligomer binding assay (Figure S4C) confirmed HNRNPH1 is able to bind to either of these sequences (Figure 4F). These results demonstrate a sequence-specific, breakpoint-dependent vulnerability in ES cells.

Physiological levels of SF3B1 are required for expression of EWS-FLI1

A decrease in EWS-FLI1 mRNA levels was observed in SF3B1-silenced ES cell lines representing different breakpoints and fusion types (Figure 3B). However, the presence of multiple variant protein in SF3B1-depleted ES cells (Figure 5A) suggests that the effects of SF3B1 silencing on the splicing of the fusion transcript are complex. One possibility is that SF3B1 silencing results in both reduced expression of full length EWS-FLI1 and the expression of mis-spliced EWS-FLI1 variant transcripts that are in-frame. To test this hypothesis, we performed PCR analysis of RNA from siRNA-transfected SKNMC cells using the E7/F7 primer pair and a second primer pair consisting of the same forward primer, but a FLI1 exon 9 reverse primer (E7/F9) (Figure 5B). As expected, we observed a substantial decrease in EWS-FLI1 expression in HNRNPH1-silenced SKMNC cells using either PCR primer pair. The E7/F7 primer pair detected a similar reduction in EWS-FLI1 expression in SF3B1-silenced SKNMC cells, but the E7/F9 primer pair detected only a partial loss of EWS-FLI1 expression, indicating the expression of a variant EWS-FLI1 transcript that excludes FLI1 exon 7, but includes FLI1 exon 9. These results suggest that physiological levels of SF3B1 are required for the generation of an EWS-FLI1 transcript that includes all exons and disruption of spliceosome activity can result in the generation of variant EWS-FLI1 transcripts with altered exon usage.

Figure 5. Depletion of SF3B1 results in mis-splicing of EWS-FLI1.

Figure 5

Open in a new tab

(A) Immunoblot analysis of whole cell lysates prepared from ES cells 48 h post siRNA-transfection using antibodies against the proteins indicated.

(B)EWS-FLI1 expression (E7/F7 or E7/F9 primer pairs) following silencing of HNRNPH1 or SF3B1 in SKNMC cells. Data are expressed relative to siNeg-transfected cells (mean ± SD, n=3). Expression of RPL27 is shown as a control for assessing non-specific splicing effects. * p<0.05; ** p<0.01; *** p<0.001 compared to siNeg.

(C) PCR analysis of the splicing of EWS-FLI1 using primers corresponding to EWSR1 exon 7 and FLI1 exon 8 and representative sequence chromatograms for the PCR products indicated in each gel (a - f).

(D) Expression of EWS-FLI1 in TC32 cells treated with increasing concentrations of PlaB (1.25 – 10 nM, 6 h). Data are expressed relative to 0.1% DMSO treated cells (mean ± SD, n=3). * p<0.05; ** p<0.01; *** p<0.001 compared to V.

(E) PCR analysis of the splicing of EWS-FLI1 using primers corresponding to EWSR1 exon 7 and FLI1 exon 8 and representative sequence chromatograms for the PCR product indicated *. TC32 cells were treated with 5 nM PlaB for 6 h.

(F) Immunoblot analysis of whole-cell lysates prepared from TC32 cells treated with increasing concentrations of PlaB for 24 h using antibodies against the proteins indicated.

(G) HCD Fragmentation spectra of peptide ([M+2H]2+-H2O, m/z 657.62) from SKMNC cells treated with 5 nM PlaB for 24 h. The protein band corresponding to EWS-FLI1 was excised and digested with trypsin and the resulting peptides were analyzed on a Thermo Fusion Orbitrap.

(H) PCR analysis of the splicing of EWS-FLI1 using a FLI1 exon 6 forward and a FL11 exon 9 reverse primer and representative sequences of the FLI1 exon 6/8 junction for the indicated product *.

See also Figure S5.

Exons adjacent and downstream of the fusion in EWS-FLI1 are vulnerable to disruption of spliceosome activity

To determine if depletion of SF3B1 results in altered splicing of EWS-FLI1, we conducted PCR analysis of the exons closest to the breakpoint (Figure 5C). Using an EWSR1 exon 7 - FLI1 exon 8 PCR primer pair (E7/F8) and cDNA generated from siNeg-transfected cells, we amplified single products of the expected sizes. In contrast, analysis of RNA from SF3B1-silenced cells (TC32, SKNMC, TC71, and RD-ES) showed the presence of multiple PCR products. The sizes of these PCR products were consistent with mis-splicing of exons close to and downstream from the fusion breakpoint, irrespective of fusion type. Sequencing of PCR products derived from SF3B1-silenced TC32 or SKNMC cells showed the splicing of EWSR1 exon 8 to FLI1 exon 8 (a and c in Figure 5C). This product is out-of-frame and establishes that TC32 and SKNMC cells also require SF3B1 to exclude EWSR1 exon 8 from EWS-FLI1, a finding confirmed using the E6/F6 primer pair (Figure S5A). Other PCR products are, however, predicted to be in-frame, though still disruptive of protein structure, including products in which EWSR1 exon 7 (b, d and e in Figure 5C) or FLI1 exon 5 are spliced to FLI1 exon 8 (f in Figure 5C). The silencing of other U2 snRNP components identified by the RNAi screen mediated a similar decrease in the expression and mis-splicing of EWS-FLI1 (Figure S5B and S5C). These findings suggest that expression of full-length EWS-FLI1 expressed in ES cell lines representing different fusion types, requires optimal activity of the U2 snRNP.

To more fully understand the effect of inhibiting U2 snRNP function on the splicing of EWS-FLI1, we turned to a pharmacological inhibitor of the SF3b spliceosome subunit, Pladienolide B (PlaB) (Kotake et al., 2007; Yokoi et al., 2011). First, we assessed the effects of PlaB on EWS-FLI1 expression by treating ES cells with PlaB (1.25 – 10 nM, 6 h) (Figures 5D and S5D). Using the E7/F7 and E7/F9 primer pairs, we detected concentration-dependent differences in EWS-FLI1 expression that phenocopied those observed following SF3B1 silencing. To confirm we were detecting the same EWS-FLI1 transcript variants as observed in SF3B1-silenced ES cells, we assessed the splicing of EWS-FLI1 in ES cells treated with 5 nM PlaB (6h). Following PlaB treatment, we observed the failure of TC32 cells to exclude EWSR1 exon 8 when processing EWS-FLI1 (Figure S5E) and mis-splicing of EWS-FLI1 at the exons adjacent to the breakpoint (Figures 5E and S5F). The consequence of PlaB treatment on the splicing of EWS-FLI1 was also evident when we examined EWS-FLI1 protein, as at the two highest concentrations of PlaB (24 h) we detected multiple protein bands (Figure 5F and S5G). In order to begin to characterize the variant protein products of mis-spliced EWS-FLI1 transcripts we immunoprecipitated EWS-FLI1 from either DMSO or 5 nM PlaB (24 h) treated TC32 or SKNMC cell lysates and performed mass spectrometry analysis. Under these conditions, we identified one unique peptide present in PlaB treated samples not present in control samples (Figure 5G). Though this peptide differs by only one amino acid (N, Asn) from a peptide observed in DMSO and PlaB treated cells (D, Asp), we were intrigued that this is precisely the sequence change predicted if either FLI1 exon 7 is excluded or FLI1 exons 7 and 8 are excluded (Figure 5G). PCR analysis of the splicing of FLI1 exons following PlaB treatment of ES cells (or the silencing of SF3B1) detected transcripts in which FLI1 exon 7 is excluded (Figure 5H, Figure S5H) suggesting that mis-spliced EWS-FLI1 transcripts may have the potential to express variant EWS-FLI1 proteins, though further study will be required to confirm this.

The inhibition of SF3b reverses the expression of genes activated by EWS-FLI1

Several studies have assessed the cytotoxic activity of PlaB on cancer cells, reporting IC50 values ranging between ~0.5 – 8.5 nM (Kotake et al., 2007; Sato et al., 2014; Yokoi et al., 2011). The PlaB IC50 values for ES cells (at 48 h) ranged between 1.5 and 2.5 nM; values comparable with those determined for two non-fusion driven prostate cancer cell lines, PC3 and LNCaP (Figure 6A). However, we observed that the responses of ES cells to PlaB differed from the prostate cancer cell lines and the published responses of two other cancer cell lines, HeLa and DLD-1 (Kotake et al., 2007; Yokoi et al., 2011). Specifically, the cytotoxic activity of PlaB on non-ES cells plateaued without inducing a complete reduction in cell viability, whereas, ES cell lines exhibited near complete loss of cell viability (Figure S6A). A possible explanation for this difference could be the disruption of EWS-FLI1 splicing mediated by PlaB and the resulting alteration in the expression and activity of EWS-FLI1. To establish if PlaB treatment results in an alteration in EWS-FLI1 activity we profiled the transcriptome of ES cells treated with 5 nM PlaB for 24 h.

Figure 6. Inhibition of SF3b function reverses expression of genes activated by EWS-FLI1.

Figure 6

Open in a new tab

(A) The viability of Ewing sarcoma (TC32, SKNMC, TC71, and RD-ES) and prostate cancer (PC3 and LNCaP) cell lines exposed to increasing concentrations of the spliceosome inhibitor Pladienolide B (PlaB) for 48 h, mean ± SEM, n=3 at each PlaB concentration normalized to vehicle (V, 0.1% DMSO); CI = Confidence Interval.

(B) GSEA of the transcriptome of PlaB treated ES cells (5 nM, 24 h) using a set of 751 genes up-regulated in ES and a set of ~100 genes activated by EWS-FLI1 (Riggi et al., 2014).

(C) Heat-map of the fold-change (Log2) in the expression of EWS-FLI1 regulated genes (Kauer et al., 2009) and genes considered direct targets of EWS-FLI1 (Riggi et al., 2014) following exposure of ES cells to PlaB (5 nM for 24 h).

(D) Expression of genes activated by EWS-FLI1 in 5 nM PlaB treated (24 h) TC32 or SKNMC cells. Data are expressed relative to DMSO treated cells (mean ± SD, n=3). * p<0.05; ** p<0.01; *** p<0.001 compared to siNeg or V.

(E) Expression of genes activated by EWS-FLI1 in SF3B1-silenced TC32 or SKNMC cells (48 h). Data are expressed relative to siNeg-transfected cells (mean ± SD, n=3). * p<0.05; ** p<0.01; *** p<0.001 compared to siNeg.

(F) Immunoblot analysis of whole-cell lysates prepared from ES cells treated with either 0.1% DMSO or 5 nm PlaB (24 h) or siNeg or SF3B1-silenced (48 h) ES cells using antibodies against the proteins indicated.

(G) ChIP analysis of DNA from TC32 cells (DMSO 0.1% or 5nM PlaB treated, 24 h) using either a FLI1 antibody or an IgG antibody. Fold-enrichment of co-precipitating DNA (PlaB compared to DMSO) was determined by qPCR for the indicated EWS-FLI1 bound GGAA microsatellites; the regions targeted and the PCR primers used are detailed in Supplemental Experimental Procedures, mean ± SEM, 3 independent experiments (triplicate samples for each treatment in each experiment).

See also Figure S6 and Tables S2 and S4

GSEA of the expression profiles of PlaB-treated ES cells showed highly significant enrichment for reduced expression of genes that are specifically disrupted in ES (NES −1.85 to −2.43, FWER p<0.001 and NES −2.11 to −2.22, FWER p<0.001) (Figure 6B). There was no enrichment for the increased expression of genes that are down-regulated in ES cells following PlaB treatment (Figure S6B). Examination of specific genes that exhibited differential expression following PlaB treatment (Table S4) confirmed the decreased expression of many EWS-FLI1 activated genes (Figure 6C). These data are consistent with PlaB treatment resulting in the disruption of the activating function of EWS-FLI1 that requires binding of EWS-FLI1 multimers at GGAA microsatellite sequences within cis-regulatory elements (Gangwal et al., 2008; Riggi et al., 2014), but not the repressive function of EWS-FLI1, that one recent study has proposed is as a result of the displacement of the binding of other transcription factors at canonical ETS binding sites (Riggi et al., 2014). To further examine the effect of disrupting spliceosome activity on the expression of EWS-FLI1 activated genes we conducted qPCR analysis of the expression of NR0B1, PRKCB, VRK1, GRK5, EZH2, FCGRT, and NKX2.2 following PlaB treatment (4 ES cell lines, 5 nM, 24 h) (Figure 6D and Figure S6B) or the silencing of SF3B1 (TC32 and SKNMC cells, 48 h) (Figure 6E). Overall, we observed a substantial decrease in the expression of each EWS-FLI1 activated gene following either PlaB treatment or silencing of SF3B1. Importantly, with the exception of EZH2 (Figure S6C) we observed no direct effect of PlaB or the silencing of SF3B1 on the splicing of these genes (data not shown). Extending our analysis to assessment of protein, we observed variable decreases in the expression of NR0B1 following PlaB and silencing of SF3B1, but we observed complete depletion of PRKCB protein expression (Figure 6F). Consistent with this observation using chromatin immunoprecipitation (ChIP) analysis we observed a substantial reduction in the fold-enrichment of EWS-FLI1 binding at a reported enhancer sequence within the PRKCB locus following PlaB treatment (Figure 6G). The fold-enrichment of EWS-FLI1 binding was also reduced at enhancer sites within 4 other genes (VRK1, GRK5, EZH2, and FCGRT) following PlaB treatment. Taken together, these data suggest that the identification by RNAi screening of multiple members of the U2 snRNP as required for EWS-FLI1 activity is because the EWS-FLI1 transcript is particularly sensitive to disruption of spliceosome function. In particular, this altered splicing of the EWS-FLI1 transcript is sufficient to disrupt the ability of the EWS-FLI1 to activate the expression of EWS-FLI1 target genes including genes associated with ES tumorigenesis.

Discussion

Ewing sarcoma is dependent on the EWS-FLI1 transcription factor for cell survival and maintenance of the malignant phenotype. Unfortunately, EWS-FLI1 has proven to be a challenging drug target (Grohar and Helman, 2013; Kovar, 2014). Therefore, the goal of this study was to identify therapeutic vulnerabilities in EWS-FLI1 activity more amenable to drug development. In order to identify these vulnerabilities, we employed a genome-wide RNAi screen to identify genes that modulate EWS-FLI1 activity. We subsequently linked lead candidate genes identified by the screen to the biogenesis of EWS-FLI1 itself. We report that the EWS-FLI1 transcript expressed in ~85% of ES tumors is vulnerable to the loss of specific RNA processing proteins, in particular factors required to ensure the splicing of exons at, and downstream, of the fusion breakpoint. We establish that splicing at these exons is susceptible to mis-splicing when genes encoding components of the U2 snRNP are silenced or spliceosome activity is inhibited (Figure 7 upper panel). We also identified that HNRNPH1 is required for the removal of EWSR1 exon 8 from the EWS-FLI1 transcript expressed in cells with a chromosome 22 breakpoint downstream of EWSR1 exon 8 (Figure 7 lower panel). Disruption of the splicing of EWS-FLI1 alters the expression of the EWS-FLI1 protein and reverses the expression of a significant proportion of the genes EWS-FLI1 deregulates, including many of the genes required for the survival of ES cells. This study opens up a potential strategy for the treatment of ES through disruption of the processing of the EWS-FLI1 fusion transcript itself.

Figure 7. The altered splicing of EWS-FLI1 results in disruption of its activity as an oncogenic transcription factor.

Figure 7

Open in a new tab

Schematic illustrating the splicing of the EWS-FLI1 pre-mRNA by SF3B1 and HNRNPH1.

HNRNPH1 contains 3 repeats of the quasi-RRM domain that binds to RNA at G triplet (GGGn) tracts typically positioned in close proximity to 5’ or 3’ splice sites (ss) (Caputi and Zahler, 2001; Huelga et al., 2012; Xiao et al., 2009). HNRNPH1 can either enhance or inhibit the use of an alternatively spliced exon, depending on the length of the G tracts, the intronic versus exonic position, and the strength of the 5’ ss (Xiao et al., 2009). HNRNPH1 and its binding sites in exon 8 of EWSR1 are intriguing putative drug targets. No small molecule inhibitor of HNRNPH1 has been described, but a logical next step will be to screen for such a compound or for an RNA-binding molecule that blocks the interaction of HNRNPH1 with the EWS-FLI1 pre-mRNA. Importantly, a molecule that blocks the excision of EWSR1 exon 8 from the EWS-FLI1 pre-mRNA could have the advantage of sequence and structure specificity. This will not be applicable in all cases of ES, but a recent study estimated that ~35% of ES tumors harbor a breakpoint that retains EWSR1 exon 8 (Berger et al., 2013) and a whole genome sequencing study of 6 ES tumors showed that 4 harbor EWSR1 intron 8 breakpoints (Brohl et al., 2014).

In contrast to HNRNPH1, the requirement for SF3B1 activity is breakpoint independent, with the inhibition of SF3B1 function resulting in altered splicing of EWS-FLI1 fusion transcripts irrespective of the position of the fusion breakpoint. The challenge of SF3B1-directed therapies will be exploiting the vulnerability of the EWS-FLI1 fusion transcript while minimizing general effects on splicing. The importance of SF3B1 in tumorigenesis has recently become appreciated with the identification of mutations in SF3B1 (Scott and Rebel, 2013). These findings have stimulated interest in the development of compound inhibitors of the spliceosome (Bonnal et al., 2012; Webb et al., 2013). In this study, we showed that the effect of silencing of SF3B1 on the splicing of EWS-FLI1 is phenocopied by a small molecule inhibitor of the spliceosome, Pladienolide B. The toxicity of this compound precluded in vivo evaluation, but other splicing inhibitors are in development (Eskens et al., 2013; Folco et al., 2011; Hong et al., 2014). As these compounds become available, their evaluation in ES is warranted. An alternative approach will be to consider other components of the spliceosome and other RNA processing proteins identified by RNAi screening as required for EWS-FLI1 activity as one or more of these may provide a more effective and less toxic target.

The aberrant splicing of EWS-FLI1 resulting from inhibition of spliceosome function also provided insight into the biology of ES. The analysis of the downstream consequences of PlaB treatment on ES cells clearly distinguished the two modes by which EWS-FLI1 regulates gene expression. It could be that activated targets of EWS-FLI1 are more vulnerable to a reduction in expression of the full length EWS-FLI1 oncoprotein because of the need for binding of multimers of EWS-FLI1 to GGAA microsatellite sequences, or the variant EWS-FLI1 proteins encoded by mis-spliced EWS-FLI1 transcripts could have different binding affects at GGAA microsatellites than at repressive ETS consensus sites. Further analysis of the transcriptional activity of variant proteins generated following the mis-splicing of EWS-FLI1 may give insight into how the oncoprotein deregulates the expression of specific genes. This study also has implications for other tumor types dependent upon the expression of a fusion gene for cell survival. For example, as for the EWSR1 exon 8 containing EWS-FLI1 pre-mRNA, there is evidence that the first NUT exon must be spliced out to generate the in-frame BRD4–NUT fusion observed in nuclear protein in testis (NUT)-midline carcinomas harboring t(15;19) translocations (Thompson-Wicking et al., 2013). The application of a functional genomic approach akin to that used in this study has the potential to identify proteins required for specific steps in the maturation of other fusion transcripts and thereby provide new strategies for the treatment of fusion driven tumors.

Experimental procedures

RNAi screen

The RNAi screen of EWS-FLI1 activity was conducted using a siRNA library targeting ~21000 human genes, 3 siRNAs per gene and TC32-NR0B1-luc and TC32-CMV-luc reporter lines (Grohar et al., 2011). Parallel transfections were conducted in each cell line, and luciferase expression was assayed 48 h later. The raw luciferase data for each screen were normalized and z-score-MAD transformed, and a seed-adjusted z-score value (Z) was calculated for each siRNA (Buehler et al., 2012; Marine et al., 2012). The difference between the Z value for each siRNA in each screen (Zdiff = ZNR0B1-ZCMV) was used to identify siRNAs that alter EWS-FLI1 activity. The confirmatory screen was conducted under analogous assay conditions using ≥4 siRNAs per gene. Cell viability was measured using the CellTiter Glo assay(Promega).

Gene expression analysis

Gene expression was assayed by quantitative reverse transcription-PCR (qPCR) using Fast SYBR Green Master Mix and StepOne Plus Real time PCR System (Applied Biosystems) or using Human HT-12 v4 BeadChip Illumina arrays (Illumina). A gene was considered differentially expressed if probes had an average Log2 fold change of ± 0.6 (~1.5 fold change on a linear scale) and a q-value (FDR) <0.05. For analysis of splicing, RNA was reverse transcribed using iScript™ Reverse Transcription Supermix (Bio-Rad) and PCR products were amplified using Platinum® PCR SuperMix High Fidelity (Invitrogen). For sequencing, PCR products were further amplified using T7-forward and T3-reverse tagged primers and, following gel electrophoresis and extraction, were sequenced using an ABI 3130 XL Genetic Analyzer (Applied Biosystems).

RNA-pull-down and binding assays

Cells were irradiated and harvested by scraping, and the cytoplasmic and nuclei fractions were separated (CelLytic NuCLEAR Extraction kit, Sigma-Aldrich). The protein–RNA complexes were immunopurified with an anti-HNRNPH1 antibody (A300-511A, Bethyl Laboratories, Inc.) or a rabbit IgG (mock control, Molecular Probes) immobilized on protein A-coated Dynabeads (Novex). Cross-linked RNAs were released using proteinase K (Roche Life Sciences), purified, reversed transcribed and amplified using primers spanning different regions of EWS-FLI1. The enrichment ratio was calculated by comparison of the quantification of normalized HNRNPH1-pull-down RNA to mock pull-down RNA.

For the RNA-binding assay, an IgG antibody-coated 96-well plate (Pierce) was washed and then incubated with an anti-HNRNPH1 antibody or rabbit IgG. Whole cell lysate and biotin-labeled oligomer were combined and UV-cross-linked. Following washing to remove unbound oligomer, the protein-oligomer complex was added to each well of the antibody coated plate, and the plate was incubated for 1 h. Unbound protein was removed by washing, and the bound protein-oligomer complex was detected using Streptavidin-HRP (Pierce).

Immunoblotting, Mass spectrometry, and Chromatin immunoprecipitation analysis

Immunoblotting was performed using standard procedures. For mass spectrometry, EWS-FLI1 protein was immunoprecipitated using a FLI1-antibody (Abcam), and samples were separated in a Novex 4–20% Tris-Glycine protein gel at 140V and subsequently coomassie stained. Bands corresponding to EWS-FLI1 were excised from the gel, and in-gel tryptic digestion was performed. Mass spectrometry experiments were performed on a Thermo Scientific Orbitrap Fusion. Peptides were identified using proteome discoverer with a significance minimum Xcorr score of 2.2. Chromatin immunoprecipitation was carried out using a magnetic bead-based kit (Millipore).

Computational and statistical analysis

The analysis of the EWS-FLI1 regulated genes used data reported by Hancock and Lessnick, 2008 and Kauer et al., 2009; see Table S3. Gene based analytical tools were accessed through http://david.abcc.ncifcrf.gov/ or www.ingenuity.com. Statistical analysis was conducted using a Student (heteroscedastic) t-test in Excel; p<0.05 was considered significant. Concentration curves, IC50 values, and Confidence Intervals were generated using Graph Pad Prism 6.

Supplementary Material

1
2
3
4
5
6

Acknowledgments

This research was supported by the Intramural Research Program of the National Cancer Institute, Center for Cancer Research NIH, DHHS, federal funds from the National Cancer Institute, NIH, under contract no. HHSN261200800001E, and by funding to P.J.G. from the Lily's Garden Foundation, the Sarcoma Alliance for Research Through Collaboration, and the Turner-Hazinski Award, Vanderbilt University. We thank Christiane Olivero (Colgate University, Hamilton, New York), Zhili Zheng (Surgery Branch, CCR, NCI), Amy McCalla (Pediatric Oncology Branch, CCR, NCI), and Thorkell Andersson (Protein Characterization Lab, Cancer Research Technology Program at NCI-Frederick, NCI) for technical assistance and Javed Khan for SKNMC cells. We also thank members of the Genetics Branch, CCR, NCI, in particular, Javed Khan, Ashish Lal, Shile Zhang, John Shern, Young Song, Sean Davies, and Josh Waterfall for helpful discussion.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Accession numbers

Data accompanying this paper have been deposited into PubChem under accession number 1159506 https://pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?aid=1159506 and GEO under accession numbers GSE67736 and GSE67548 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=ijsjcccwbnmhnqh&acc=GSE67736 and http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=ihezmggytdytrej&acc=GSE67548).

Supplemental Information

Supplemental Information includes Supplemental Experimental Procedures, four tables, and six figures.

Author contributions

Conceptualization, P.J.G., L.J.H., and N.J.C; Experimentation, S.K., P.J.G., G.O.R.R., S.H., N.S., M.H., N.K.M., J.Z., M.O., T.L.J., M.G., K.G., and C.K.-T.; Data analysis, K.H., E.B., S.E.M, S.K., and N.J.C.; Writing, N.J.C; Editing, P.J.G., S.K., N.S., S.E.M., and L.J.H.; Study supervision, N.J.C.

References

  1. Bailly RA, Bosselut R, Zucman J, Cormier F, Delattre O, Roussel M, Thomas G, Ghysdael J. DNA-binding and transcriptional activation properties of the EWS-FLI-1 fusion protein resulting from the t(11;22) translocation in Ewing sarcoma. Mol Cell Biol. 1994;14:3230–3241. doi: 10.1128/mcb.14.5.3230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Berger M, Dirksen U, Braeuninger A, Koehler G, Juergens H, Krumbholz M, Metzler M. Genomic EWS-FLI1 fusion sequences in Ewing sarcoma resemble breakpoint characteristics of immature lymphoid malignancies. PLoS One. 2013;8:e56408. doi: 10.1371/journal.pone.0056408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bilke S, Schwentner R, Yang F, Kauer M, Jug G, Walker RL, Davis S, Zhu YJ, Pineda M, Meltzer PS, et al. Oncogenic ETS fusions deregulate E2F3 target genes in Ewing sarcoma and prostate cancer. Genome Res. 2013;23:1797–1809. doi: 10.1101/gr.151340.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bonnal S, Vigevani L, Valcarcel J. The spliceosome as a target of novel antitumour drugs. Nat Rev Drug Discov. 2012;11:847–859. doi: 10.1038/nrd3823. [DOI] [PubMed] [Google Scholar]
  5. Brohl AS, Solomon DA, Chang W, Wang J, Song Y, Sindiri S, Patidar R, Hurd L, Chen L, Shern JF, et al. The genomic landscape of the Ewing Sarcoma family of tumors reveals recurrent STAG2 mutation. PLoS Genet. 2014;10:e1004475. doi: 10.1371/journal.pgen.1004475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Buehler E, Khan AA, Marine S, Rajaram M, Bahl A, Burchard J, Ferrer M. siRNA off-target effects in genome-wide screens identify signaling pathway members. Sci Rep. 2012;2:428. doi: 10.1038/srep00428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Caputi M, Zahler AM. Determination of the RNA binding specificity of the heterogeneous nuclear ribonucleoprotein (hnRNP) H/H'/F/2H9 family. J Biol Chem. 2001;276:43850–43859. doi: 10.1074/jbc.M102861200. [DOI] [PubMed] [Google Scholar]
  8. Chung N, Zhang XD, Kreamer A, Locco L, Kuan PF, Bartz S, Linsley PS, Ferrer M, Strulovici B. Median absolute deviation to improve hit selection for genome-scale RNAi screens. J Biomol Screen. 2008;13:149–158. doi: 10.1177/1087057107312035. [DOI] [PubMed] [Google Scholar]
  9. Crompton BD, Stewart C, Taylor-Weiner A, Alexe G, Kurek KC, Calicchio ML, Kiezun A, Carter SL, Shukla SA, Mehta SS, et al. The genomic landscape of pediatric ewing sarcoma. Cancer Discov. 2014;4:1326–1341. doi: 10.1158/2159-8290.CD-13-1037. [DOI] [PubMed] [Google Scholar]
  10. Delattre O, Zucman J, Plougastel B, Desmaze C, Melot T, Peter M, Kovar H, Joubert I, de Jong P, Rouleau G, et al. Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours. Nature. 1992;359:162–165. doi: 10.1038/359162a0. [DOI] [PubMed] [Google Scholar]
  11. Eskens FA, Ramos FJ, Burger H, O'Brien JP, Piera A, de Jonge MJ, Mizui Y, Wiemer EA, Carreras MJ, Baselga J, et al. Phase I pharmacokinetic and pharmacodynamic study of the first-in-class spliceosome inhibitor E7107 in patients with advanced solid tumors. Clin Cancer Res. 2013;19:6296–6304. doi: 10.1158/1078-0432.CCR-13-0485. [DOI] [PubMed] [Google Scholar]
  12. Folco EG, Coil KE, Reed R. The anti-tumor drug E7107 reveals an essential role for SF3b in remodeling U2 snRNP to expose the branch point-binding region. Genes Dev. 2011;25:440–444. doi: 10.1101/gad.2009411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gangwal K, Sankar S, Hollenhorst PC, Kinsey M, Haroldsen SC, Shah AA, Boucher KM, Watkins WS, Jorde LB, Graves BJ, et al. Microsatellites as EWS/FLI response elements in Ewing's sarcoma. Proc Natl Acad Sci U S A. 2008;105:10149–10154. doi: 10.1073/pnas.0801073105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Grohar PJ, Helman LJ. Prospects and challenges for the development of new therapies for Ewing sarcoma. Pharmacol Ther. 2013;137:216–224. doi: 10.1016/j.pharmthera.2012.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grohar PJ, Woldemichael GM, Griffin LB, Mendoza A, Chen QR, Yeung C, Currier DG, Davis S, Khanna C, Khan J, et al. Identification of an inhibitor of the EWS-FLI1 oncogenic transcription factor by high-throughput screening. J Natl Cancer Inst. 2011;103:962–978. doi: 10.1093/jnci/djr156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hancock JD, Lessnick SL. A transcriptional profiling meta-analysis reveals a core EWS-FLI gene expression signature. Cell Cycle. 2008;7:250–256. doi: 10.4161/cc.7.2.5229. [DOI] [PubMed] [Google Scholar]
  17. Hawkins DS, Bölling T, Dubois S, Hogendoorn PCW, Jürgens H, Paulussen M, Randall RL, Lessnick SL. Ewing Sarcoma. In: Pizzo PA, Poplack DG, editors. Principles and Practice of Pediatric Oncology. Philadelphia: Lippincott Williams & Wilkins; 2010. [Google Scholar]
  18. Hong DS, Kurzrock R, Naing A, Wheler JJ, Falchook GS, Schiffman JS, Faulkner N, Pilat MJ, O'Brien J, LoRusso P. A phase I, open-label, single-arm, dose-escalation study of E7107, a precursor messenger ribonucleic acid (pre-mRNA) splicesome inhibitor administered intravenously on days 1 and 8 every 21 days to patients with solid tumors. Invest New Drugs. 2014;32:436–444. doi: 10.1007/s10637-013-0046-5. [DOI] [PubMed] [Google Scholar]
  19. Huelga SC, Vu AQ, Arnold JD, Liang TY, Liu PP, Yan BY, Donohue JP, Shiue L, Hoon S, Brenner S, et al. Integrative genome-wide analysis reveals cooperative regulation of alternative splicing by hnRNP proteins. Cell Rep. 2012;1:167–178. doi: 10.1016/j.celrep.2012.02.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kauer M, Ban J, Kofler R, Walker B, Davis S, Meltzer P, Kovar H. A molecular function map of Ewing's sarcoma. PLoS One. 2009;4:e5415. doi: 10.1371/journal.pone.0005415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kotake Y, Sagane K, Owa T, Mimori-Kiyosue Y, Shimizu H, Uesugi M, Ishihama Y, Iwata M, Mizui Y. Splicing factor SF3b as a target of the antitumor natural product pladienolide. Nat Chem Biol. 2007;3:570–575. doi: 10.1038/nchembio.2007.16. [DOI] [PubMed] [Google Scholar]
  22. Kovar H. Blocking the road, stopping the engine or killing the driver? Advances in targeting EWS/FLI-1 fusion in Ewing sarcoma as novel therapy. Expert Opin Ther Targets. 2014;18:1315–1328. doi: 10.1517/14728222.2014.947963. [DOI] [PubMed] [Google Scholar]
  23. Marine S, Bahl A, Ferrer M, Buehler E. Common seed analysis to identify off-target effects in siRNA screens. J Biomol Screen. 2012;17:370–378. doi: 10.1177/1087057111427348. [DOI] [PubMed] [Google Scholar]
  24. May WA, Gishizky ML, Lessnick SL, Lunsford LB, Lewis BC, Delattre O, Zucman J, Thomas G, Denny CT. Ewing sarcoma 11;22 translocation produces a chimeric transcription factor that requires the DNA-binding domain encoded by FLI1 for transformation. Proc Natl Acad Sci U S A. 1993a;90:5752–5756. doi: 10.1073/pnas.90.12.5752. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. May WA, Lessnick SL, Braun BS, Klemsz M, Lewis BC, Lunsford LB, Hromas R, Denny CT. The Ewing's sarcoma EWS/FLI-1 fusion gene encodes a more potent transcriptional activator and is a more powerful transforming gene than FLI-1. Mol Cell Biol. 1993b;13:7393–7398. doi: 10.1128/mcb.13.12.7393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mitelman F, Johansson B, Mertens F. The impact of translocations and gene fusions on cancer causation. Nat Rev Cancer. 2007;7:233–245. doi: 10.1038/nrc2091. [DOI] [PubMed] [Google Scholar]
  27. Patocs B, Nemeth K, Garami M, Arato G, Kovalszky I, Szendroi M, Fekete G. Multiple splice variants of EWSR1-ETS fusion transcripts co-existing in the Ewing sarcoma family of tumors. Cell Oncol (Dordr) 2013;36:191–200. doi: 10.1007/s13402-013-0126-8. [DOI] [PubMed] [Google Scholar]
  28. Riggi N, Knoechel B, Gillespie SM, Rheinbay E, Boulay G, Suva ML, Rossetti NE, Boonseng WE, Oksuz O, Cook EB, et al. EWS-FLI1 utilizes divergent chromatin remodeling mechanisms to directly activate or repress enhancer elements in Ewing sarcoma. Cancer Cell. 2014;26:668–681. doi: 10.1016/j.ccell.2014.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Sato M, Muguruma N, Nakagawa T, Okamoto K, Kimura T, Kitamura S, Yano H, Sannomiya K, Goji T, Miyamoto H, et al. High antitumor activity of pladienolide B and its derivative in gastric cancer. Cancer Sci. 2014;105:110–116. doi: 10.1111/cas.12317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Scott LM, Rebel VI. Acquired mutations that affect pre-mRNA splicing in hematologic malignancies and solid tumors. J Natl Cancer Inst. 2013;105:1540–1549. doi: 10.1093/jnci/djt257. [DOI] [PubMed] [Google Scholar]
  31. Thompson-Wicking K, Francis RW, Stirnweiss A, Ferrari E, Welch MD, Baker E, Murch AR, Gout AM, Carter KW, Charles AK, et al. Novel BRD4-NUT fusion isoforms increase the pathogenic complexity in NUT midline carcinoma. Oncogene. 2013;32:4664–4674. doi: 10.1038/onc.2012.487. [DOI] [PubMed] [Google Scholar]
  32. Webb TR, Joyner AS, Potter PM. The development and application of small molecule modulators of SF3b as therapeutic agents for cancer. Drug Discov Today. 2013;18:43–49. doi: 10.1016/j.drudis.2012.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Xiao X, Wang Z, Jang M, Nutiu R, Wang ET, Burge CB. Splice site strength-dependent activity and genetic buffering by poly-G runs. Nat Struct Mol Biol. 2009;16:1094–1100. doi: 10.1038/nsmb.1661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Yokoi A, Kotake Y, Takahashi K, Kadowaki T, Matsumoto Y, Minoshima Y, Sugi NH, Sagane K, Hamaguchi M, Iwata M, et al. Biological validation that SF3b is a target of the antitumor macrolide pladienolide. FEBS J. 2011;278:4870–4880. doi: 10.1111/j.1742-4658.2011.08387.x. [DOI] [PubMed] [Google Scholar]
  35. Zoubek A, Pfleiderer C, Salzer-Kuntschik M, Amann G, Windhager R, Fink FM, Koscielniak E, Delattre O, Strehl S, Ambros PF, et al. Variability of EWS chimaeric transcripts in Ewing tumours: a comparison of clinical and molecular data. Br J Cancer. 1994;70:908–913. doi: 10.1038/bjc.1994.419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Zucman J, Delattre O, Desmaze C, Plougastel B, Joubert I, Melot T, Peter M, De Jong P, Rouleau G, Aurias A, et al. Cloning and characterization of the Ewing's sarcoma and peripheral neuroepithelioma t(11;22) translocation breakpoints. Genes Chromosomes Cancer. 1992;5:271–277. doi: 10.1002/gcc.2870050402. [DOI] [PubMed] [Google Scholar]
  37. Zucman J, Melot T, Desmaze C, Ghysdael J, Plougastel B, Peter M, Zucker JM, Triche TJ, Sheer D, Turc-Carel C, et al. Combinatorial generation of variable fusion proteins in the Ewing family of tumours. EMBO J. 1993;12:4481–4487. doi: 10.1002/j.1460-2075.1993.tb06137.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zucman-Rossi J, Legoix P, Victor JM, Lopez B, Thomas G. Chromosome translocation based on illegitimate recombination in human tumors. Proc Natl Acad Sci U S A. 1998;95:11786–11791. doi: 10.1073/pnas.95.20.11786. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS747179-supplement-1.xlsx (130.3KB, xlsx)
2
NIHMS747179-supplement-2.xlsx (60.2KB, xlsx)
3
NIHMS747179-supplement-3.xlsx (375.6KB, xlsx)
4
NIHMS747179-supplement-4.xlsx (287.8KB, xlsx)
5
NIHMS747179-supplement-5.pdf (17MB, pdf)
6
NIHMS747179-supplement-6.docx (77.9KB, docx)

RESOURCES