RING1B contributes to Ewing sarcoma development by repressing the Na(V)1.6 sodium channel and the NF-kappa B pathway, independently of the fusion oncoprotein

Oncotarget1 www.impactjournals.com/oncotarget www.impactjournals.com/oncotarget/ Oncotarget, Advance Publications 2016 RING1B contributes to Ewing sarcoma development by repressing the Na V 1.6 sodium channel and the NF-κB pathway, independently of the fusion oncoprotein Inmaculada Hernandez-Muñoz1, Elisabeth Figuerola2,*, Sara Sanchez-Molina2,*, Eva Rodriguez2, Ana Isabel Fernández-Mariño3,6, Carlos Pardo-Pastor3, María Isabel Bahamonde3, José M. Fernández-Fernández3, Daniel J. García-Domínguez4, Lourdes Hontecillas-Prieto4, Cinzia Lavarino2, Angel M. Carcaboso2, Carmen de Torres2, Oscar M. Tirado5, Enrique de Alava4, Jaume Mora2 1Fundació Institut Hospital del Mar d’Investigacions Mèdiques (FIMIM), 08003-Barcelona, Spain 2Developmental Tumor Biology Laboratory, Department of Pediatric Hematology and Oncology, Hospital Sant Joan de Déu, 08950-Barcelona, Spain 3Laboratori de Fisiologia Molecular, Departament de Ciències Experimentals i de la Salut, Universitat Pompeu Fabra, 08003-Barcelona, Spain 4Department of Pediatric Hematology and Oncology, Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen del Rocio/CSIC/Universidad de Sevilla, 41013-Seville, Spain 5Sarcoma Research Group, Laboratori d’Oncología Molecular, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), L’Hospitalet de Llobregat, 08908-Barcelona, Spain 6Present Affiliation: Department of Neuroscience and Biomolecular Chemistry, School of Medicine and Public Health, University of Wisconsin, Madison-53705, USA *These authors have contributed equally to this work Correspondence to: Inmaculada Hernandez-Muñoz, email: [email protected] Jaume Mora, email: [email protected] Keywords: RING1B, Ewing sarcoma, voltage-gated sodium channel, NF-κB, FGFR/SHP2/STAT3 Received: December 10, 2015 Accepted: May 28, 2016 Published: June 15, 2016 ABSTRACT Ewing sarcoma (ES) is an aggressive tumor defined by EWSR1 gene fusions that behave as an oncogene. Here we demonstrate that RING1B is highly expressed in primary ES tumors, and its expression is independent of the fusion oncogene. RING1B-depleted ES cells display an expression profile enriched in genes functionally involved in hematological development but RING1B depletion does not induce cellular differentiation. In ES cells, RING1B directly binds the SCN8A sodium channel promoter and its depletion results in enhanced Nav1.6 expression and function. The signaling pathway most significantly modulated by RING1B is NF-κB. RING1B depletion results in enhanced p105/p50 expression, which sensitizes ES cells to apoptosis by FGFR/ SHP2/STAT3 blockade. Reduced NaV1.6 function protects ES cells from apoptotic cell death by maintaining low NF-κB levels. Our findings identify RING1B as a trait of the cell-of-origin and provide a potential targetable vulnerability. INTRODUCTION Ewing sarcoma (ES) is an aggressive and poorly differentiated tumor, typically arising from bone and soft tissues in children and young adults. It is characterized by reciprocal translocations that result in almost all cases in the fusion of the EWS RNA binding protein 1 (EWSR1) to an ETS transcription factor, being EWSR1–FLI1 the most common chimera [1, 2]. ES tumors display a high degree of genomic stability with very few recurrent mutations besides the pathognomonic fusion, and are among the most genetically normal cancers [3–5]. This strikingly unaltered somatic landscape highlights the role of EWSR1-FLI1 as the unique trigger of the oncogenic transformation in an otherwise yet unidentified cell-of-origin harboring key features that will likely contribute to the eventual development of ES. Meta-analysis of data coming from EWSR1-FLI1 gain- of-function approaches revealed that the genes up-regulated Oncotarget2 www.impactjournals.com/oncotarget by the fusion in heterologous cell systems are more numerous and display more similarities among different experimental models than the genes down-regulated. Since the cell-of- origin of ES remains elusive, gain-of-function models have been carried out expressing the oncogene in a variety of poorly or undifferentiated heterologous cell types. Genomewide analysis using high-throughput sequencing technologies have identified a plethora of EWSR1-FLI1 direct targets and shown that EWSR1-FLI1 primarily up-regulates gene expression through the interaction with GGAA repeats present in satellite DNA within the genome [6]. In contrast, data obtained by depleting EWSR1-FLI1 in ES cells revealed that many more genes resulted down-regulated by the fusion oncogene than up-regulated, suggesting that gene repression may be more prevalent than transcriptional activation [7]. However, many of these EWSR1-FLI1 repressed targets are divergent and highly dependent on the cellular background [8]. Since EWSR1-FLI1 directly binds to promoters of a small subset of repressed targets [7], the lack of consistency among the different sets of repressed genes is likely due to a variety of both direct and indirect mechanisms used by EWSR1-FLI1 for gene silencing. EZH2 is a direct EWSR1-FLI1 target that belongs to the Polycomb (PcG) family of epigenetic regulators and blocks endothelial and neuro-ectodermal differentiation of ES cells [9]. PcG proteins form two major families of complexes, the Polycomb-repressive complex (PRC) 1 and 2. PRC2 comprises EED, SUZ12 and EZH2, which catalyzes the K27 trimethylation of histone H3 (H3K27me3). Mammalian PRC1 includes BMI1, MEL18, and RING1B, which catalyzes H2A K119 ubiquitination (ubH2K119) [10, 11]. PRC1 and PRC2 mostly differ in their genomic localization with a small subset of PRC1 co-localizing with H3K27me3. Adding complexity, six major groups of PRC1 subcomplexes with specific developmental functions and mutually exclusive PRC1 subunits have been described, being RING1B the unique common feature [12]. Importantly, it has recently been reported that RING1B catalytic activity results in gene repression, consistent with the classic repressive function of the Polycomb complexes, whereas catalytic-independent association of RING1B with UTX, an H3K27 demethylase, and p300, leads to transcriptional activation [13]. Despite the important role of the epigenetic landscape in ES, studies addressing the PcG contribution to ES tumorigenesis have been restricted to EZH2 and BMI1. Here we investigate the expression and function of RING1B in ES, a protein with unique abilities among the PcG family of epigenetic regulators. RESULTS Ewing sarcoma tumors express high levels of RING1B ES tumors express high EZH2 mRNA levels [9]. To better characterize PcG expression we analyzed EZH2 and RING1B protein expression in ES primary tumors. EZH2 was detected in all the tumor samples, most of them with variable EZH2 expression patterns (Figure 1, right). Particularly poor EZH2 expression was found in largely hemorrhagic tumors, blood clots and tumors infiltrating the adipose tissue (Figure 1, J-N). In contrast, RING1B was highly expressed and uniformly distributed throughout the tumor in most samples, reaching the maximum score (Figure 1, left; Supplementary Figure S1A). Of note, RING1B was expressed in endothelial cells of tumor blood vessels and in the adipose tissue (Figure 1C, 1G), whereas RING1B expression was observed in sparse cells of blood clots (Figure 1F). In these tissues EZH2 expression was low. Importantly, RING1B expression in ES was found to be significantly higher than in other developmental tumors such as rhabdomyosarcoma, synovial sarcoma and Wilms tumor (Supplementary Figure S1B). Since it has been postulated that neural crestderived progenitors could be the cell-of-origin [14] and ES tumors display a continuum of different degrees of neural differentiation [15], RING1B expression was also investigated in neural tissues. Moderate RING1B expression was found in cerebellum and peripheral sympathetic ganglia, whereas low or no expression was found in the brain and spinal cord (Supplementary Figure S1C). In addition, RING1B could not be detected in undifferentiated neuroblastic aggregates of the normal fetal adrenal medulla, a derivative of the embryonic neural crest (Supplementary Figure S1C). Next we analysed the expression of BMI1 and MEL18, two PRC1 proteins. Some specimens displayed low BMI1 and MEL18 levels (Supplementary Figure S2A and S2B), accordingly to reported BMI1 expression in many, but not all, ES primary tumors [16]. In embryonic neural tissues BMI1 expression pattern resembled that of RING1B, with low to moderate expression in most of the embryonic neural tissues (Supplementary Figure S2C). RING1B expression is independent of EWSR1- FLI1 To test whether RING1B expression was modulated by the oncogene, EWSR1-FLI1 was introduced in 293T and HeLa cells. In both cell lines EZH2 expression was enhanced, as previously reported [9], whereas RING1B levels were not affected (Figure 2A). Conversely, down-regulation of EWSR1-FLI1 in A673 cells resulted in diminished EZH2 levels, but RING1B levels remained unaffected (Figure 2B). Accordingly, endogenous EWSR1-FLI1 binds specifically to a conserved ETS recognition sequence of the EZH2 promoter [17]. In contrast, the genomic region of RING1B was not reported to be an EWSR1-FLI1- direct binding domain [6, 18] nor RING1B was a Oncotarget3 www.impactjournals.com/oncotarget Figure 1: RING1B and EZH2 expression in primary Ewing sarcoma (ES) tumors. Immunohistochemistry analysis of RING1B (A-G) and EZH2 (H-N) expression in serial sections of primary ES tumors. Right columns, higher magnifications of the left panel fields (scale bars 50μm). A, B, H, I. samples with intense and homogeneous RING1B and EZH2 staining; C, D, E, J, K, L. hemorrhagic samples with intense RING1B and low EZH2 staining. Blood lakes can be extensively observed in these tumors; F, M. blood clots in an ES sample; G, N. adipose tissue in an ES sample. Oncotarget4 www.impactjournals.com/oncotarget differentially expressed gene in EWSR1-FLI1-depleted cells [19, 20]. We suspected that EWSR1-FLI1- independency of RING1B expression might be reflecting the high levels of expression of the cell-of-origin where the oncogenic translocation occur, likely regulating key functions of this cell. Therefore, we decided to study the role of RING1B in ES cells by depleting RING1B protein levels. Gene expression profile in RING1B-depleted ES cells is unique and enriched in genes functionally involved in hematological development To identify genes whose expression is regulated by RING1B in ES tumors we analyzed four different ES cell lines stably depleted of RING1B by short hairpin RNA (shRNA) (Figure 3A). Gene expression analysis Figure 2: RING1B expression is independent of EWSR1-FLI1. A. EWSR1-FLI1 was ectopically expressed in 293T (left panel) and HeLa (right panel) cells and RING1B and EZH2 levels were determined by western blot. Actin and Tubulin, loading controls. B. A673 cells were transiently transfected with mock or FLI1 short interfering RNA (siRNA) and RING1B and EZH2 levels were determined 48 hours later by western blot (left panel) and qRT-PCR (right panel). Numbers at the bottom represent protein band intensities normalized to Actin and relative to control cells, as described in “Methods”. Graph shows mean ± SD of three replicate samples from one representative experiment out of three. Oncotarget5 www.impactjournals.com/oncotarget Figure 3: Characterization of the transcriptome in RING1B-depleted ES cells. A. Efficiency of stable Ring1B knockdown (short hairpin, sh) in ES cells determined by qRT-PCR (left panel) and western blot (right panel). *, P < 0.05; **, P < 0.01; ***, P < 0.005. Numbers at the bottom, RING1B expression normalized to Lamin B and relative to RING1B levels in each control cell line. B. Cluster dendrogram derived from array expression data of the four ES parental cell lines (left panel) and functions of the genes differentially regulated in the A4573 cell line versus A673/SK-ES-1/TC-1 cell lines, identified by the Ingenuity Pathway Analysis (IPA) software (right panel). C. Primary heat map of genes differentially expressed between shMock and shRING1B cells. Right panel, heat map depicting the first 30 annotated and up-regulated genes in shRING1B cells. D. Top functional annotations enriched among RING1B-regulated genes, identified by IPA software. Array data deposited at the Gene Expression Omnibus (GEO, National Center for Biotechnology Information) with accession number GSE71007. Oncotarget6 www.impactjournals.com/oncotarget followed by unsupervised hierarchical clustering of the most variably expressed genes stratified parental cell lines into one main group formed by A673, SK-ES-1 and TC71 and segregated the A4573 cell line (Figure 3B). Of note, A4573 cells also display low BMI1 protein levels [21]. Despite substantial differences in A4573 gene expression profile, linear models for microarrays (LIMMA) analysis was performed using the four cell line data in an attempt to identify common traits of RING1B gene regulation in ES. Logarithmic fold change for each shRING1B/shMock sample pair revealed that RING1B depletion resulted in altered expression of 1430 sequences (Figure 3C). Among them, 523 corresponded to annotated genes, 259 up-regulated (49%) and 264 downregulated (51%), a ratio that parallels that of RING1B- transcriptionally regulated genes in melanoma [13]. The genes differentially expressed in RING1B-depleted cells were then analyzed at the functional level using the IPA software. This analysis revealed “tissue morphology” and “hematological system development and function” as the top two biological functions altered, while the top canonical pathways identified included “heme biosynthesis” and “haematopoiesis from pluripotent stem cells” (Figure 3D). To compare RING1B with other PcG proteins, BMI1 was depleted in A673 and TC71 cells (Supplementary Figure S3A). Gene expression analysis revealed that BMI1 or RING1B depletion resulted in 734 and 838 differentially expressed sequences, respectively (Supplementary Figure S3B), 196 overlapping, and among these, 70 were annotated sequences (Supplementary Figure S3C). Most genes, however, were differentially regulated by each PRC1 protein in ES cells. EZH2 downregulation in ES cells results in the induction of genes important for neural and endothelial differentiation [9]. In contrast, few endothelial and neural genes were identified among the annotated transcripts in RING1B-depleted cells (Figure 4A). Furthermore, RING1B depletion only resulted in enhanced expression of three genes, MS4A3, CDRT1 and TLE4, out of the most significantly EZH2 downregulated genes in ES cells [9] (Figures S4, S5). All-trans retinoic acid (ATRA)-induced neural stem cell differentiation results in EZH2 down-regulation and decreased binding of EZH2 to ATRA-inducible target genes [22]. Particularly important for ES biology is the differentiation blockade that EZH2 imposes in ES cells, which can undergo neuroectodermal differentiation with butylated hydroxyanisole (BHA) only upon EZH2 depletion [9]. ATRA and BHA treatment of SK-ES-1 cells resulted in reduced EZH2, but not RING1B, mRNA levels (Figure 4B). Furthermore, neither ATRA nor BHA were able to induce differentiation in shRING1B cells since mRNA levels of neural and endothelial differentiation genes remained similar to those of the untreated cells (Figure 4C). RING1B regulates the NaV1.6 sodium channel in Ewing sarcoma cells FGF14 and SCN8A were among the top 30-upregulated annotated sequences upon RING1B depletion (Figure 3C), and these increases where validated by qRT-PCR in RING1B-depleted cells (Figures 5A, S6A). SCN8A encodes the α subunit of the voltage-gated sodium channel type VIII, or NaV1.6. This protein forms a sodium-selective channel essential for rapid membrane depolarization in response to voltage differences in excitable neurons. FGF14 is a member of the intracellular fibroblast growth factor subfamily (iFGF) that co-localizes and interacts with the NaV channel α subunits, being NaV1.6 the most sensitive to FGF14 modulation [23]. While RING1B was not detected at the FGF14 promoter (data not shown), it was found to bind to the promoter region of the SCN8A gene in ES cells (Figure 5B). RING1B depletion using an independent shRNA construct [11] confirmed that SCN8A is repressed by RING1B in ES cells (Supplementary Figure S6B). SCN8A released expression in RING1B-depleted cells resulted in enhanced NaV1.6 protein levels (Figure 5C). In contrast, RING1B depletion in 293T did not affect NaV1.6 levels (Figure 5C) while in neuroblastoma SK-N-SH cells caused subtle increases in SCN8A mRNA levels (Figures S6C), suggesting that RING1B regulation of the Na V 1.6 channel is specific of ES cells. Next we investigated NaV1.6 channel regulation by other PcG proteins. BMI1 depletion resulted in SCN8A down-regulation while EZH2 depletion lead to minor changes in SCN8A mRNA levels (Supplementary Figure S7). Furthermore, treatment with the histone methylation inhibitor DZNep did not affect SCN8A mRNA levels (Supplementary Figure S7D). All these data suggest that NaV1.6 channel repression in ES is specifically set by RING1B. To check the functional impact of the enhanced expression of NaV1.6 upon RIING1B depletion we used the voltage-clamp technique in the whole-cell configuration to record voltage-gated sodium currents. In response to depolarizing pulses, significant fastactivating and fast-inactivating inward sodium currents, with a persistent component (consistent with NaV1.6 electrophysiological properties [24, 25]), were only recorded in shRING1B cells but not in shMock control cells (Figure 5D). Since NaV1.6 knockout in dorsalroot ganglion cells results in migration towards the ventral region and acquisition of a parasympathetic phenotype [26] we investigated the migratory abilities of ES cells. A4573 and A673 control cells displayed migratory abilities that were attenuated in RING1B depleted cells (Figure 5E). Since RING1B depletion affects the expression of very few neural genes and did not alleviate ES cells from their neural differentiation blockade, this result suggests that targeted NaV1.6 Oncotarget7 www.impactjournals.com/oncotarget repression might be of particular relevance for ES biology. RING1B modulates the NF-κB pathway in Ewing sarcoma cells To identify molecular pathways affected by RING1B-modulated targets we conducted IPA. On the basis of the four RING1B-depleted cell line signature, the IPA analysis revealed NF-κB as the most significantly altered node (Figure 6A). To explore this pathway we performed western blot analyses of the five members of the NF-κB transcription factor family: NF-κB1 (p50 and its precursor p105), NF-κB2 (p52 and its precursor p100), c-Rel, RelA/p65 and RelB. Levels of p105 and p50 were enhanced to variable degrees in shRING1B cells when compared to their corresponding shMock controls (Figure 6B). RING1B depletion using an independent Figure 4: RING1B depletion does not result in global alterations of developmental transcription programs in ES cells. A. Heat map showing endothelial (red) and neural (black) genes up-regulated in RING1B-depleted cells, identified by manual curation of the microarray annotated sequences. B. RING1B and EZH2 mRNA levels determined by qRT-PCR in SK-ES-1 cells treated with differentiating agent butylated hydroxyanisole (BHA) and all trans-retinoic acid (ATRA). C. SK-ES-1 cells, either shMock or shRING1B, were treated with BHA or ATRA and mRNA levels of different neural or endothelial genes were analyzed by qRT-PCR. B and C. graphs show mean ± SD from two independent experiments performed with three replicate samples. Oncotarget8 www.impactjournals.com/oncotarget Figure 5: RING1B represses the NaV1.6 channel in ES cells. A. SCN8A mRNA levels in RING1B-depleted cells determined by qRT-PCR and referred to shMock control mRNA levels. *, P < 0.05; **, P < 0.01. B. Relative enrichment of RING1B occupancy at SCN8A promoter, measured by ChIP-qPCR. Graph displays percentage of RING1B recruitment to two different SCN8A promoter amplicons, referred to enrichment from the IgG control, set as 1. Data express mean ± SD of three independent experiments. *, P < 0.05; **, P < 0.01. C. NaV1.6 protein levels in cytoplasmic (A4573) and total (A673, SK-ES-1, TC-71, 293T) extracts from control or RING1B-depleted cells, analyzed by immunoblot. Numbers at the bottom, band intensities normalized to Actin or Lamin B and relative to control cells. D. Representative current traces obtained from shMock and shRING1B SK-ES-1 cells clamped at -90 mV and pulsed for 40 ms to the indicated depolarizing voltage (top) and average current density-voltage (I-V) relationships for shMock (open circles, n = 5) and shRING1B (filled cyan circles, n = 5) cells (bottom). Fast-activating inward currents with both transient and persistent components, consistent with NaV1.6 currents, were only observed in shRING1B SK-ES-1 cells (peak current was 4.2 ± 0.4 pA/pF) and the corresponding V1/2 act, kact and Vrev values after fitting the I-V curve to a Boltzmann equation (see Methods for further details) were (in mV): -21 ± 4.1 mV (as reported for NaV1.6 channels heterologous expression in the ND7/23 neuronal cells, either in the absence or presence of FGF14 (54;61)), 10.8 ± 2.1 and 84.5 ± 16.3, respectively. E. Cell migration speed measurements (in micron/sec) during wound healing from shMock (open) and shRING1B (black filled) A4573 and A673 cell lines (obtained in three different experiments). Symbols represent individual cells. Red bars depict mean ± SEM speed. Suppressing RING1B expression, which correlates with increased SCN8A expression, decreased cell speed in A4573 and A673 cells. **, P<0.01; ***, P<0.001. Oncotarget9 www.impactjournals.com/oncotarget Figure 6: RING1B represses NF-κB in ES cells. A. Top-scoring multigene network associated by the IPA software to microarray expression data in RING1B-depleted cells. Overexpressed genes, red; downregulated genes, green. B. Levels of NF-κB proteins, detected by immunoblot performed using total extracts of shMock and shRING1B cells. Numbers at the bottom, band intensities normalized to Actin and relative to their own control cells. C. Flow cytometry analysis of A673 cells treated with 25 μM SU5402 or 50 μM NSC87877 for 24h. Percentage of cells in the sub-G1 region is indicated. DNA histograms of 10,000 propidium iodide stained cells. Oncotarget16 www.impactjournals.com/oncotarget 13. Rai K, Akdemir KC, Kwong LN, Fiziev P, Wu CJ, Keung EZ, Sharma S, Samant NS, Williams M, Axelrad JB, Shah A, Yang D, Grimm EA, et al. Dual Roles of RNF2 in Melanoma Progression. Cancer Discov. 2015; 5: 1314-27. 14. Staege MS, Hutter C, Neumann I, Foja S, Hattenhorst UE, Hansen G, Afar D, and Burdach SE. DNA microarrays reveal relationship of Ewing family tumors to both endothelial and fetal neural crest-derived cells and define novel targets. Cancer Res. 2004; 64: 8213-21. 15. Suh CH, Ordonez NG, Hicks J, and Mackay B. Ultrastructure of the Ewing's sarcoma family of tumors. Ultrastruct Pathol. 2002; 26: 67-76. 16. Cooper A, van Doorninck J, Ji L, Russell D, Ladanyi M, Shimada H, Krailo M, Womer RB, Hsu JH, Thomas D, Triche TJ, Sposto R, and Lawlor ER. Ewing tumors that do not overexpress BMI-1 are a distinct molecular subclass with variant biology: a report from the Children's Oncology Group. Clin Cancer Res. 2011; 17: 56-66. 17. Burdach S, Plehm S, Unland R, Dirksen U, Borkhardt A, Staege MS, Muller-Tidow C, and Richter GH. Epigenetic maintenance of stemness and malignancy in peripheral neuroectodermal tumors by EZH2. Cell Cycle. 2009; 8: 1991-6. 18. Patel M, Simon JM, Iglesia MD, Wu SB, McFadden AW, Lieb JD, and Davis IJ. Tumor-specific retargeting of an oncogenic transcription factor chimera results in dysregulation of chromatin and transcription. Genome Res. 2012; 22: 259-70. 19. Bilke S, Schwentner R, Yang F, Kauer M, Jug G, Walker RL, Davis S, Zhu YJ, Pineda M, Meltzer PS, and Kovar H. Oncogenic ETS fusions deregulate E2F3 target genes in Ewing sarcoma and prostate cancer. Genome Res. 2013; 23: 1797-809. 20. Kauer M, Ban J, Kofler R, Walker B, Davis S, Meltzer P, and Kovar H. A molecular function map of Ewing's sarcoma. PLoS One. 2009; 4: e5415. 21. Douglas D, Hsu JH, Hung L, Cooper A, Abdueva D, van Doorninck J, Peng G, Shimada H, Triche TJ, and Lawlor ER. BMI-1 promotes ewing sarcoma tumorigenicity independent of CDKN2A repression. Cancer Res. 2008; 68: 6507-15. 22. Surface LE, Thornton SR, and Boyer LA. Polycomb group proteins set the stage for early lineage commitment. Cell Stem Cell. 2010; 7: 288-98. 23. Laezza F, Lampert A, Kozel MA, Gerber BR, Rush AM, Nerbonne JM, Waxman SG, Dib-Hajj SD, and Ornitz DM. FGF14 N-terminal splice variants differentially modulate Nav1.2 and Nav1.6-encoded sodium channels. Mol Cell Neurosci. 2009; 42: 90-101. 24. Burbidge SA, Dale TJ, Powell AJ, Whitaker WR, Xie XM, Romanos MA, and Clare JJ. Molecular cloning, distribution and functional analysis of the NA(V)1.6. Voltage-gated sodium channel from human brain. Brain Res Mol Brain Res. 2002; 103: 80-90. 25. Wittmack EK, Rush AM, Hudmon A, Waxman SG, and Dib-Hajj SD. Voltage-gated sodium channel Nav1.6 is modulated by p38 mitogen-activated protein kinase. J Neurosci. 2005; 25: 6621-30. 26. Wright MA, Mo W, Nicolson T, and Ribera AB. In vivo evidence for transdifferentiation of peripheral neurons. Development. 2010; 137: 3047-56. 27. Mohammadi M, McMahon G, Sun L, Tang C, Hirth P, Yeh BK, Hubbard SR, and Schlessinger J. Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science. 1997; 276: 955-60. 28. Chen L, Sung SS, Yip ML, Lawrence HR, Ren Y, Guida WC, Sebti SM, Lawrence NJ, and Wu J. Discovery of a novel shp2 protein tyrosine phosphatase inhibitor. Mol Pharmacol. 2006; 70: 562-70. 29. Hadari YR, Kouhara H, Lax I, and Schlessinger J. Binding of Shp2 tyrosine phosphatase to FRS2 is essential for fibroblast growth factor-induced PC12 cell differentiation. Mol Cell Biol. 1998; 18: 3966-73. 30. Banasiak KJ, Burenkova O, and Haddad GG. Activation of voltage-sensitive sodium channels during oxygen deprivation leads to apoptotic neuronal death. Neuroscience. 2004; 126: 31-44. 31. Lessnick SL, and Ladanyi M. Molecular pathogenesis of Ewing sarcoma: new therapeutic and transcriptional targets. Annu Rev Pathol. 2012; 7: 145-59. 32. Koppens M, and van Lohuizen M. Context-dependent actions of Polycomb repressors in cancer. Oncogene. 2015; 35: 1341-52. 33. Ewing J. The Classic: Diffuse endothelioma of bone. Proceedings of the New York Pathological Society. 1921; 12:17. Clin Orthop Relat Res. 2006; 450: 25-27. 34. van der Schaft DW, Hillen F, Pauwels P, Kirschmann DA, Castermans K, Egbrink MG, Tran MG, Sciot R, Hauben E, Hogendoorn PC, Delattre O, Maxwell PH, Hendrix MJ, et al. Tumor cell plasticity in Ewing sarcoma, an alternative circulatory system stimulated by hypoxia. Cancer Res. 2005; 65: 11520-8. 35. Maniotis AJ, Folberg R, Hess A, Seftor EA, Gardner LM, Pe'er J, Trent JM, Meltzer PS, and Hendrix MJ. Vascular channel formation by human melanoma cells in vivo and in vitro: vasculogenic mimicry. Am J Pathol. 1999; 155: 739-52. 36. Aurora AB, Biyashev D, Mirochnik Y, Zaichuk TA, Sanchez-Martinez C, Renault MA, Losordo D, and Volpert OV. NF-kappaB balances vascular regression and angiogenesis via chromatin remodeling and NFAT displacement. Blood. 2010; 116: 475-84. 37. Javelaud D, and Besancon F. NF-kappa B activation results in rapid inactivation of JNK in TNF alpha-treated Ewing sarcoma cells: a mechanism for the anti-apoptotic effect of NF-kappa B. Oncogene. 2001; 20: 4365-72. 38. Lagirand-Cantaloube J, Laud K, Lilienbaum A, Tirode F, Delattre O, Auclair C, and Kryszke MH. EWS-FLI1 inhibits TNFalpha-induced NFkappaB-dependent transcription Oncotarget17 www.impactjournals.com/oncotarget in Ewing sarcoma cells. Biochem Biophys Res Commun. 2010; 399: 705-10. 39. White DE, and Burchill SA. BAY 11-7082 induces cell death through NF-kappaB-independent mechanisms in the Ewing's sarcoma family of tumours. Cancer Lett. 2008; 268: 212-24. 40. Sturla LM, Westwood G, Selby PJ, Lewis IJ, and Burchill SA. Induction of cell death by basic fibroblast growth factor in Ewing's sarcoma. Cancer Res. 2000; 60: 6160-70. 41. Chotani MA, and Chiu IM. Differential regulation of human fibroblast growth factor 1 transcripts provides a distinct mechanism of cell-specific growth factor expression. Cell Growth Differ. 1997; 8: 999-1013. 42. Xie Y, Zhou S, Chen H, Du X, and Chen L. Recent research on the growth plate: Advances in fibroblast growth factor signaling in growth plate development and disorders. J Mol Endocrinol. 2014; 53: T11-34. 43. Calo V, Migliavacca M, Bazan V, Macaluso M, Buscemi M, Gebbia N, and Russo A. STAT proteins: from normal control of cellular events to tumorigenesis. J Cell Physiol. 2003; 197: 157-168. 44. Behjati S, Basu BP, Wallace R, Bier N, Sebire N, Hasan F, and Anderson J. STAT3 Regulates Proliferation and Immunogenicity of the Ewing Family of Tumors In Vitro. Sarcoma. 2012; 2012: 987239. 45. Lai R, Navid F, Rodriguez-Galindo C, Liu T, Fuller CE, Ganti R, Dien J, Dalton J, Billups C, and Khoury JD. STAT3 is activated in a subset of the Ewing sarcoma family of tumours. J Pathol. 2006; 208: 624-32. 46. Jiang Y, Janku F, Subbiah V, Angelo LS, Naing A, Anderson PM, Herzog CE, Fu S, Benjamin RS, and Kurzrock R. Germline PTPRD mutations in Ewing sarcoma: biologic and clinical implications. Oncotarget. 2013; 4: 884-9. doi: 10.18632/oncotarget.1021. 47. Aggarwal BB, Kunnumakkara AB, Harikumar KB, Gupta SR, Tharakan ST, Koca C, Dey S, and Sung B. Signal transducer and activator of transcription-3, inflammation, and cancer: how intimate is the relationship? Ann N Y Acad Sci. 2009; 1171: 59-76. 48. Anderson JL, Titz B, Akiyama R, Komisopoulou E, Park A, Tap WD, Graeber TG, and Denny CT. Phosphoproteomic profiling reveals IL6-mediated paracrine signaling within the Ewing sarcoma family of tumors. Mol Cancer Res. 2014; 12: 1740-54. 49. Yan S, Li Z, and Thiele CJ. Inhibition of STAT3 with orally active JAK inhibitor, AZD1480, decreases tumor growth in Neuroblastoma and Pediatric Sarcomas In vitro and In vivo. Oncotarget. 2013; 4: 433-45. doi: 10.18632/oncotarget.930. 50. Ryland KE, Svoboda LK, Vesely ED, McIntyre JC, Zhang L, Martens JR, and Lawlor ER. Polycomb-dependent repression of the potassium channel-encoding gene KCNA5 promotes cancer cell survival under conditions of stress. Oncogene. 2015; 34: 4591-600. 51. Coles EG, Lawlor ER, and Bronner-Fraser M. EWS-FLI1 causes neuroepithelial defects and abrogates emigration of neural crest stem cells. Stem Cells. 2008; 26: 2237-44. 52. Hu-Lieskovan S, Zhang J, Wu L, Shimada H, Schofield DE, and Triche TJ. EWS-FLI1 fusion protein up-regulates critical genes in neural crest development and is responsible for the observed phenotype of Ewing's family of tumors. Cancer Res. 2005; 65: 4633-44. 53. Tirode F, Laud-Duval K, Prieur A, Delorme B, Charbord P, and Delattre O. Mesenchymal stem cell features of Ewing tumors. Cancer Cell. 2007; 11: 421-9. 54. Cavazzana AO, Miser JS, Jefferson J, and Triche TJ. Experimental evidence for a neural origin of Ewing's sarcoma of bone. Am J Pathol. 1987; 127: 507-18. 55. Franchi A, Pasquinelli G, Cenacchi G, Della RC, Gambini C, Bisceglia M, Martinelli GN, and Santucci M. Immunohistochemical and ultrastructural investigation of neural differentiation in Ewing sarcoma/PNET of bone and soft tissues. Ultrastruct Pathol. 2001; 25: 219-25. 56. Rettig WJ, Garin-Chesa P, and Huvos AG. Ewing's sarcoma: new approaches to histogenesis and molecular plasticity. Lab Invest. 1992; 66: 133-7. 57. Navarro S, Gonzalez-Devesa M, Ferrandez-Izquierdo A, Triche TJ, and Llombart-Bosch A. Scanning electron microscopic evidence for neural differentiation in Ewing's sarcoma cell lines. Virchows Arch A Pathol Anat Histopathol. 1990; 416: 383-91. 58. Rorie CJ, Thomas VD, Chen P, Pierce HH, O'Bryan JP, and Weissman BE. The Ews/Fli-1 fusion gene switches the differentiation program of neuroblastomas to Ewing sarcoma/peripheral primitive neuroectodermal tumors. Cancer Res. 2004; 64: 1266-77. 59. Teitell MA, Thompson AD, Sorensen PH, Shimada H, Triche TJ, and Denny CT. EWS/ETS fusion genes induce epithelial and neuroectodermal differentiation in NIH 3T3 fibroblasts. Lab Invest. 1999; 79: 1535-43. 60. Olsen BR, Reginato AM, and Wang W. Bone development. Annu Rev Cell Dev Biol. 2000; 16: 191-220. 61. Takashima Y, Era T, Nakao K, Kondo S, Kasuga M, Smith AG, and Nishikawa S. Neuroepithelial cells supply an initial transient wave of MSC differentiation. Cell. 2007; 129: 1377-88. 62. Torchia EC, Jaishankar S, and Baker SJ. Ewing tumor fusion proteins block the differentiation of pluripotent marrow stromal cells. Cancer Res. 2003; 63: 3464-8. 63. Kopen GC, Prockop DJ, and Phinney DG. Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains. Proc Natl Acad Sci U S A. 1999; 96: 10711-6. 64. Woodbury D, Schwarz EJ, Prockop DJ, and Black IB. Adult rat and human bone marrow stromal cells differentiate into neurons. J Neurosci Res. 2000; 61: 364-70. 65. Miyagawa Y, Okita H, Nakaijima H, Horiuchi Y, Sato B, Taguchi T, Toyoda M, Katagiri YU, Fujimoto J, Hata J, Oncotarget18 www.impactjournals.com/oncotarget Umezawa A, and Kiyokawa N. Inducible expression of chimeric EWS/ETS proteins confers Ewing's family tumorlike phenotypes to human mesenchymal progenitor cells. Mol Cell Biol. 2008; 28: 2125-37. 66. Riggi N, Suva ML, Suva D, Cironi L, Provero P, Tercier S, Joseph JM, Stehle JC, Baumer K, Kindler V, and Stamenkovic I. EWS-FLI-1 expression triggers a Ewing's sarcoma initiation program in primary human mesenchymal stem cells. Cancer Res. 2008; 68: 2176-85. 67. Tanaka M, Yamazaki Y, Kanno Y, Igarashi K, Aisaki K, Kanno J, and Nakamura T. Ewing's sarcoma precursors are highly enriched in embryonic osteochondrogenic progenitors. J Clin Invest. 2014; 124: 3061-74. 68. Cornaz-Buros S, Riggi N, DeVito C, Sarre A, Letovanec I, Provero P, and Stamenkovic I. Targeting cancer stem-like cells as an approach to defeating cellular heterogeneity in Ewing sarcoma. Cancer Res. 2014; 74: 6610-22. 69. Sohn EJ, Li H, Reidy K, Beers LF, Christensen BL, and Lee SB. EWS/FLI1 oncogene activates caspase 3 transcription and triggers apoptosis in vivo. Cancer Res. 2010; 70: 1154-63. 70. Torchia EC, Boyd K, Rehg JE, Qu C, and Baker SJ. EWS/ FLI-1 induces rapid onset of myeloid/erythroid leukemia in mice. Mol Cell Biol. 2007; 27: 7918-34. 71. Lin PP, Pandey MK, Jin F, Xiong S, Deavers M, Parant JM, and Lozano G. EWS-FLI1 induces developmental abnormalities and accelerates sarcoma formation in a transgenic mouse model. Cancer Res. 2008; 68: 8968-75. 72. Martinez-Romero C, Rooman I, Skoudy A, Guerra C, Molero X, Gonzalez A, Iglesias M, Lobato T, Bosch A, Barbacid M, Real FX, and Hernandez-Munoz I. The epigenetic regulators Bmi1 and Ring1B are differentially regulated in pancreatitis and pancreatic ductal adenocarcinoma. J Pathol. 2009; 219: 205-13. 73. Riccardi C, and Nicoletti I. Analysis of apoptosis by propidium iodide staining and flow cytometry. Nat Protoc. 2006; 1: 1458-61.