Desmoplastic small round cell tumor (DSRCT) is a rare and aggressive soft-tissue malignancy with a poor overall survival and no effective therapeutic options. The tumor is believed to be dependent on the continued activity of the oncogenic EWS-WT1 transcription factor. However, the dependence of the tumor on EWS-WT1 has not been well established. In addition, there are no studies exploring the downstream transcriptional program across multiple cell lines. In this study, we have developed a novel approach to selectively silence EWS-WT1 without impacting either wild-type EWSR1 or WT1. We show a clear dependence of the tumor on EWS-WT1 in two different cell lines, BER and JN-DSCRT-1. In addition, we identify and validate important downstream target pathways commonly dysregulated in other translocation-positive sarcomas, including PRC2, mTOR, and TGFB. Surprisingly, there is striking overlap between the EWS-WT1 and EWS-FLI1 gene signatures, despite the fact that the DNA-binding domain of the fusion proteins, WT1 and FLI1, is structurally unique and classified as different types of transcription factors. This study provides important insight into the biology of this disease relative to other translocation-positive sarcomas, and the basis for the therapeutic targeting of EWS-WT1 for this disease that has limited therapeutic options.
Desmoplastic small round cell tumor (DSRCT) is a rare and aggressive soft-tissue malignancy. It typically occurs in adolescent and young adult males with a male-to-female ratio of 4:11. DSRCT commonly presents with widely metastatic disease throughout the abdomen, as well as diffuse peritoneal carcinomatosis2. There is currently no standard recommendation for management, and despite aggressive multi-agent chemotherapy, and surgical resection, the median survival is 17 months with a 5-year survival rate of less than 20%1.
DSRCT is defined by the t(11;22)(p13;q12) translocation that fuses the N-terminal domain of EWSR1 to the C-terminal DNA-binding domain of WT1, creating the aberrant transcription factor, EWS-WT1, which is diagnostic for the disease3. Apart from this translocation, the disease is believed to have a low somatic mutation frequency, a finding similar to other translocation-positive pediatric sarcomas4. As evidence of this, a screen of 24 DSRCT samples using a panel of 29 genes did not identify any recurrent mutations4. Previous single-patient case reports have described amplification of AURKB and MCL1, as well as variants of unknown significance in ARID1A and RUNX1, but there have been no consistent recurrent mutations observed5,6.
The EWS-WT1 breakpoint can occur in introns 7–10 of EWSR1 and invariably between exons 7 and 8 of WT12. This fuses the potent transcriptional activation domain of EWSR1 with the DNA- binding domain of WT12. It is believed that DSRCT cells depend on the continued activity of EWS-WT1, but there are no studies clearly establishing a dependence of the tumor on this fusion protein. Several studies have published individual downstream-induced targets of EWS-WT1, including PDGFA, IL2RB, and BAIAP3 based on inducible expression in various cell lines7,8,9. However, none of these studies led to cellular transformation following EWS-WT1 expression or established a comprehensive list of downstream EWS-WT1 targets in multiple patient-derived cell lines3.
In this study, we demonstrate the dependence of DSRCT on EWS-WT1 for cell survival. We identify a comprehensive list of downstream targets of the fusion protein in BER and JN-DSRCT-1 cells, and features of these two DSRCT cell lines that reflect the phenotype of the tumor10. Surprisingly, we identify a number of common features with other oncogenic fusion proteins, and EWS-FLI1 in particular, despite differences in the DNA-binding domain of the two proteins. Together, this study provides insight into the molecular pathogenesis of DSRCT and establishes EWS-WT1 as a bona fide drug target for this disease.
Results and discussion
DSRCT is dependent on the activity of EWS-WT1 for continued proliferation
In order to establish the dependence of DSRCT cells on EWS-WT1, we selectively silenced EWS-WT1 using siRNA. The challenge is that both wild-type EWSR1 and WT1 are expressed in the cell. To overcome this, we used two siRNA sequences (one on each side of the EWS-WT1 breakpoint) and titrated the concentration of the pooled siRNA to selectively suppress EWS-WT1 expression in BER and JN-DSRCT-1 without impacting wild-type EWSR1 or WT1 expression (Fig. 1a). Cell-line identity and the presence of the translocation was confirmed by FISH (Supplementary Fig. 1). EWS-WT1 expression was lost by 48 h, which correlated with a loss of proliferation in both cell lines (Fig. 1b–d). These data show that EWS-WT1 is both necessary and sufficient to drive DSRCT cell proliferation as knockdown of this fusion suppresses proliferation.
Loss of EWS-WT1 activity leads to morphologic changes and growth arrest
We next explored the consequence of EWS-WT1 knockdown on the cellular phenotype. Forty-eight hours of siRNA silencing of EWS-WT1 led to growth arrest and a striking morphological change with the cells becoming enlarged and flattened relative to control (Fig. 2a). With continued suppression, there was clear induction of apoptosis as measured by the cleavage of PARP by western blot analysis by 60–72 h (Fig. 2b). This induction of apoptosis was confirmed using a fluorescent marker of cleaved caspase 3/7, which showed an increase in focal fluorescence following loss of EWS-WT1 compared with control cells (Fig. 2c). Importantly, apoptosis occurred long after growth arrest, indicating that this is likely a secondary effect rather than a direct effect of EWS-WT1 loss.
EWS-WT1 loss alters the transcriptome of DSRCT cells
Although a number of individual targets of EWS-WT1 have been reported in the literature, the gene signature of the fusion protein and a comprehensive list of downstream targets has not been established in more than one cell line5,6. RNA sequencing was performed after 48 h of siRNA silencing of EWS-WT1 to ensure complete silencing while minimizing secondary effects on gene expression that would occur with prolonged silencing (Fig. 1b). We identified a total of 1879 genes across both cell lines that changed in expression by >2 log-fold change (FC) (Fig. 3a). Importantly, there were more large-magnitude gene expression changes that were unique to a given cell line than common in both models (Supplementary Tables 1–3). Furthermore, consistent with observations in other sarcomas, more genes were repressed than induced by EWS-WT1 (Fig. 3b)12. We defined a core gene signature consisting of 68 genes induced and 223 genes repressed by EWS-WT1 in both cell lines (Supplementary Table 4). Importantly, we found a number of targets previously identified in the literature, including FGFR4, JAK3, IGF signaling, and members of the Wnt pathway that showed a >2 log FC in both cell lines. Other previously identified targets of the fusion protein, such as PDGFA and BAIAP3, demonstrated >1.5 log FC in both cell lines9,13,14,15. Finally, we validated suppression of the protein expression of both an induced (ERG) and a repressed (CEBPD) target (Fig. 3c). This is the first comprehensive analysis of the EWS-WT1 gene signature in two different DSRCT cell lines.
Functional GSEA suggests important features of the tumor
In order to better characterize the EWS-WT1 transcriptome, we used functional gene set enrichment analysis (fgsea) to evaluate the biologic pathways altered by EWS-WT111. Importantly, we identified pathways that reflected known features of DSRCT. This tumor is known to be chemorefractive and is found primarily in males. Consistent with these observations, the fusion protein represses estrogen signaling, suggesting a dependence on androgen. In addition, fgsea identified strong pro-proliferative signatures, including the MYC signature identified in both cell lines (Fig. 4a). Finally, the tumor modulates the DNA damage response, both suppressing TP53 signaling and driving the expression of gene sets associated with the DNA damage response, suggesting a direct link between the fusion protein itself and resistance to chemotherapy.
Dysregulation of the DNA damage response is a prominent feature of other FET family fusions such as EWS-FLI1, the oncogenic driver of Ewing sarcoma. This correlation between EWS-WT1 and EWS-FLI1, despite different DNA-binding domains, suggests common mechanisms of tumorigenesis with these FET family fusions. Additional common features included an alteration in the E2F family of transcription factors, which has been previously described with EWS-FLI1, as well as modulation of other pathways dysregulated by EWS-FLI1, including TGFB and IGF signaling/mTOR16,17,18. It is notable that mTOR and IGF signaling have been previously identified as important in this disease15,18. An important observation is that many well-established EWS-FLI1 target genes were found to change in expression >0.5 log FC with EWS-WT1 silencing in DSRCT cells (Fig. 4b)19.
Finally, fgsea also suggested that EWS-FLI1 and EWS-WT1 might share common mechanisms of gene expression dysregulation. Most notably, we found a PRC2 signature that is known to be dysregulated by EWS-FLI1 in Ewing sarcoma cells20. In addition, EWS-WT1 upregulates the expression of ERG which is a close ETS family member of FLI1. It is possible that ERG is driving the expression of these targets, making this tumor an ETS-like tumor (Fig. 3c). Consistent with this idea, silencing of ERG leads to a striking loss of cell viability and induction of apoptosis that parallels the direct silencing of EWS-WT1 (Fig. 4c, d, Supplementary Figs. 2 and 3, Supplementary Table 5). Indeed, it appears that ERG is a critical component of the EWS-WT1 downstream transcriptional program, although more needs to be done to establish this idea.
In summary, we have used transcriptional profiling and functional gene set enrichment analysis to comprehensively characterize the transcriptome of EWS-WT1, thus providing the basis for ongoing investigations into the biology and therapeutic targeting of this tumor. We clearly demonstrate the dependence of DSRCT on the EWS-WT1 transcription factor for cell proliferation by selectively silencing EWS-WT1 in two primary DSRCT cell lines. We show reversal of proliferation and a late apoptotic cell death. In addition, we establish a comprehensive list of EWS-WT1 downstream targets, thus providing insight into the biology of the tumor, and highlighting important therapeutic vulnerabilities, including FGFR4, JAK3, mTOR, PDGF, ERG, and TGFB. This list of targets captures a number of therapeutic candidates that are both novel and previously identified by others using complementary approaches.
This study also shows that the fusion protein drives the expression of the DNA damage response as we identified a gene set called “Hallmark_DNA_repair” with a NES of −1.8 and a P = 0.0023. This gene set is known to have multiple members of both the nucleotide excision repair and transcription-coupled nucleotide excision repair pathways. In addition, EWS-WT1 directly represses TP53 signaling. These data may explain why this tumor is chemorefractory in the clinic. More importantly, it suggests that a therapeutic vulnerability may exist in one of the DNA damage response pathways in this tumor, either at baseline or with suppression of EWS-WT1.
Interestingly, this study also makes the novel observation that there is common biology shared with other fusion proteins, and EWS-FLI1 in particular, despite structural differences in the DNA-binding domain of the two proteins. While the mechanism is not clear, EWS-WT1 upregulates the expression of ERG which is a close family member to FLI1. This may be the reason why so many common targets were identified. Importantly, the cells show a striking dependence on ERG expression, suggesting that these gene expression changes are fundamentally important in the dependence of DSRCT on EWS-WT1. However, there may also be common epigenetic mechanisms that have been well-described for EWS-FLI1, such as the dysregulation of PRC2 as was suggested by the fgsea analysis.
Most importantly, this study definitively shows a dependence of these cells on EWS-WT1 for continued proliferation in two different models to establish the fusion protein as a therapeutic target for DSRCT.
The sequencing data have been deposited on GEO under the accession number GSE137561.
Hayes-Jordan, A. & Anderson, P. M. The diagnosis and management of desmoplastic small round cell tumor: a review. Curr. Opin. Oncol. 23, 385–389 (2011).
Gerald, W. L. et al. Clinical, pathologic, and molecular spectrum of tumors associated with t(11;22)(p13;q12): desmoplastic small round-cell tumor and its variants. J. Clin. Oncol. 16, 3028–3036 (1998).
Gerald, W. & Haber, D. The EWS-WT1 gene fusion in desmoplastic small round cell tumor. Semin. Cancer Biol. 15, 197–205 (2005).
Shukla, N. et al. Oncogene mutation profiling of pediatric solid tumors reveals significant subsets of embryonal rhabdomyosarcoma and neuroblastoma with mutated genes in growth signaling pathways. Clin. Cancer Res. 18, 748–757 (2012).
La Starza, R. et al. Multiple EWSR1-WT1 and WT1-EWSR1 copies in two cases of desmoplastic round cell tumor. Cancer Genet. 206, 387–392 (2012).
Silva, J. G. et al. Clinical next generation sequencing of pediatric-type malignancies in adult patients identifies novel somatic aberrations. Oncoscience 2, 187–192 (2015).
Lee, S. B. et al. The EWS-WT1 translocation product induces PDGFA in desmoplastic small round cell tumour. Nat. Genet. 17, 309–314 (2015).
Wong, J. C. et al. Induction of the interleukin-2/15 receptor beta chain by the EWS-WT1 translocation product. Oncogene 21, 2009–2019 (2002).
Palmer, R. E. et al. Induction of BAIAP3 by the EWS-WT1 chimeric fusion implicates regulated exocytosis in tumorigenesis. Cancer Cell 2, 497–505 (2002).
Markides, C. S. et al. Desmoplastic small round cell tumor (DSRCT) xenografts and tissue culture lines: establishment and initial characterization. Oncol. Lett. 5, 1453–1456 (2012).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Kauer, M., et al. A Molecular function map of Ewing’s sarcoma. PLoS ONE 4, e5415 (2009).
Saito, T. et al. EWSWT1 chimeric protein in desmoplastic small round cell tumor is a potent transactivator of FGFR4. J. Cancer Sci. Ther. 4, 335–340 (2012).
Devecchi, A. et al. The genomics of desmoplastic small round cell tumor reveals the deregulation of genes related to DNA damage response, epithelial-mesenchymal transition, and immune response. Cancer Commun. 38, 70 (2018).
Werner, H. et al. A novel EWS-WT1 gene fusion product in desmoplastic small round cell tumor is a potent transactivator of the insulin-like growth factor-I receptor (IGF-IR) gene. Cancer Lett. 247, 84–90 (2007).
Bilke, S. et al. Oncogenic ETS fusions deregulate E2F3 target genes in Ewing sarcoma and prostate cancer. Genome Res. 23, 1797–1809 (2013).
Im, Y. H. et al. EWS-FLI1, EWS-ERG, and EWS-ETV1 oncoproteins of Ewing tumor family all suppress transcription of transforming growth factor beta type II receptor gene. Cancer Res. 60, 1536–1540 (2000).
Subbiah, V. et al. Morphoproteomic profiling of the mammalian target of rapamycin (mTOR) signaling pathway in desmoplastic small round cell tumor (EWS/WT1), Ewing’s sarcoma (EWS/FLI1) and Wilms’ tumor(WT1). PLoS ONE 8, e68985 (2013).
Harlow, M. L. et al. Lurbinectedin inactivates the Ewing Sarcoma oncoprotein EWS-FLI1 by redistributing it within the nucleus. Cancer Res. 76, 6657–6668 (2016).
Richter, G. H. S. et al. EZH2 is a mediator of EWS/FLI1 driven tumor growth and metastasis blocking endothelial and neuroectodermal differentiation. Proc. Natl Acad. Sci. USA 106, 5324–5329 (2009).
Funding was provided by internal funds at Van Andel Institute (P.J.G.), Helen DeVos Children’s Hospital (J.M.G.), and Hyundai Hope on Wheels Young Investigator Award (J.M.G.). The authors would like to thank Marie Adams for technical support and library preparations, and the Bioinformatics and Statistics Core of the Van Andel Research Institute (Grand Rapids, MI). We also like to thank Julie Koeman and the Genomics Core for technical support and FISH reports.
Conflict of interest
The authors declare that they have no conflict of interest.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Gedminas, J.M., Chasse, M.H., McBrairty, M. et al. Desmoplastic small round cell tumor is dependent on the EWS-WT1 transcription factor. Oncogenesis 9, 41 (2020). https://doi.org/10.1038/s41389-020-0224-1