ews-wt1 oncoprotein activates neuronal reprogramming...

EWS-WT1 Oncoprotein Activates Neuronal Reprogramming Factor ASCL1 and Promotes

Neural Differentiation

Hong-Jun Kang1, Jun Hong Park

1, WeiPing Chen

2, Soo Im Kang

1,4, Krzysztof Moroz

1, Marc Ladanyi

3

and Sean Bong Lee1

Affiliations: 1Tulane University School of Medicine, Department of Pathology and Laboratory

Medicine, 1430 Tulane Ave., New Orleans, LA 70112, USA; 2Genomics Core Facility, National

Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, 9000 Rockville

Pike, Bethesda, MD 20892, USA; 3Department of Pathology, Memorial Sloan-Kettering Cancer Center,

New York, NY 10065, USA.

4Present address: Institute for Cancer Genetics, Department of Pathology and Cell Biology, Columbia

University Medical Center, New York, NY 10032, USA.

Running title: EWS-WT1 activates ASCL1 and induces neural differentiation

Key Words: EWS-WT1, DSRCT, ASCL1, neural gene expression, neural reprogramming

Financial support: This research was supported in part by the Intramural Research Program of the

NIH, NIDDK (S.B.L.), and by the Tulane Startup Fund (S.B.L.).

Corresponding author: Sean Bong Lee, Ph.D., Tulane University School of Medicine, Department of

Pathology and Laboratory Medicine, 1700 Tulane Ave. Room 808, New Orleans, LA 70112; Tel: (504)

988-1331; Fax: (504) 988-7389; E-mail: [email protected]

Conflicts of interest: The authors declare no potential conflicts of interest.

Word count: 5,062 words (4,840 words excluding the title page)

Total number of figures and tables: 6 main figures, 6 supplement figures and 2 supplement tables

Research. on February 13, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on June 16, 2014; DOI: 10.1158/0008-5472.CAN-13-3663

mailto:[email protected]

http://cancerres.aacrjournals.org/

Abstract

The oncogenic fusion gene EWS-WT1 is the defining chromosomal translocation in desmoplastic small

round cell tumors (DSRCT), a rare but aggressive soft tissue sarcoma with a high rate of mortality.

EWS-WT1 functions as an aberrant transcription factor that drives tumorigenesis, but the mechanistic

basis for its pathogenic activity is not well understood. To address this question, we created a transgenic

mouse strain that permits physiologic expression of EWS-WT1 under the native murine Ews promoter.

EWS-WT1 expression led to a dramatic induction of many neuronal genes in embryonic fibroblasts and

primary DSRCT, most notably the neural reprogramming factor ASCL1. Mechanistic analyses

demonstrated that EWS-WT1 directly bound the proximal promoter of ASCL1, activating its

transcription through multiple WT1-responsive elements. Conversely, EWS-WT1 silencing in DSRCT

cells reduced ASCL1 expression and cell viability. Notably, exposure of DSRCT cells to neuronal

induction media increased neural gene expression and induced neurite-like projections, both of which

were abrogated by silencing EWS-WT1. Taken together, our findings reveal that EWS-WT1 can

activate neural gene expression and direct partial neural differentiation via ASCL1, suggesting agents

that promote neural differentiation might offer a novel therapeutic approach to treat DSRCT.




Introduction

Desmoplastic small round cell tumor (DSRCT) is a rare but aggressive tumor occurring predominantly

in adolescents and young adults (1). DSRCT is poorly understood and highly lethal, resulting in 85%

mortality within 5 years despite aggressive multimodal therapy (2, 3). The majority of tumors are found

in serous membrane of abdominal or pelvic cavity without an apparent involvement of any organ

systems and form multiple nests of malignant cells embedded in dense desmoplastic stroma (1). A

distinct immunophenotypic feature of DSRCT is a multi-lineage expression of epithelial, mesenchymal,

neuronal and muscle markers (3). Despite the presence of these multi-lineage markers, the tumor cells

appear poorly differentiated. To date, there is no molecular rationale for the multi-lineage expression

and the tumor cell of origin remains undefined.

DSRCT is caused by a balanced chromosomal translocation t(11;22)(p13;q12) that results in a fusion of

the N-terminal domain (NTD) of Ewing sarcoma gene, EWSR1 (termed EWS) to the C-terminal domain

(CTD) of Wilms Tumor gene, WT1 (4, 5). EWS encodes a RNA/ssDNA binding protein that play

multiple roles in diverse cellular processes, such as maturation of pre B lymphocytes, meiosis,

hypersensitivity to DNA damage, prevention of premature senescence of fibroblasts (6) and

hematopoietic stem cells (7), mitosis (8), cell-fate determination of classical brown fat (9), regulation of

microRNAs (10) and regulation of genotoxic-induced alternative splicing (11, 12). The NTD of EWS

contains multiple degenerate repeats of SYGQQS motif that functions as a potent transcriptional

activation domain (13, 14). WT1 is inactivated in 10-15% of Wilms tumors, a childhood kidney cancer

(15). WT1 encodes a transcription factor with four Cys2-His2 zinc fingers at the C-terminus, and

undergoes an alternative splicing involving only three amino acids (Lys, Thr and Ser, termed KTS)

between the zinc fingers 3 and 4 (16). This splicing produces two isoforms that either lacks (-) or

contains (+) the KTS which alters the DNA binding specificity. In DSRCT, only the last three zinc

fingers of WT1 are fused to EWS and the KTS splicing is preserved (5), resulting in the production of

two isoforms: EWS-WT1-KTS (herein termed E-KTS) and EWS-WT1+KTS (E+KTS).

The presence of EWS-WT1 translocation in all DSRCT points to the fusion products as the initiator of

this tumor. Interestingly, the two isoforms differ in their oncogenic activities as only the E-KTS, but

not E+KTS, has the potential to transform NIH3T3 cells in vitro (17). Therefore, the majority of studies




have focused on identifying the target genes of E-KTS (18-22), while only two genes have been

identified as direct targets of E+KTS (22, 23). These studies also revealed different DNA recognition

sequences of each isoform: E-KTS binds to either a GC-rich, 5’-GXG(T/G)GGGXG-3’ (X is any base)

(17, 20), or TCCn-repeats (n>3) (18), while E+KTS recognizes 5’-GGAGG(A/G)-3’ (23). Despite these

findings, how EWS-WT1 drives oncogenesis remains poorly understood; consequently, the prognosis

of DSRCT remains grim and the development of effective therapy is urgently needed.




Materials and Methods

Cell culture and reagents

HEK293 (CRL-1573) and U2OS (HTB-96) cells were purchased from American Type Culture

Collection and MEFs were generated as described (6). These cells were grown in DMEM with 10%

FBS, 100U/ml penicillin and 100g/ml streptomycin (Invitrogen). JN-DSRCT-1 cells (24),

authenticated by the presence of EWS-WT1 translocation and free of mycoplasma, were grown in

DMEM/F-12 media with 10% fetal bovine serum (FBS). Tamoxifen, 4-hydroxytamoxifen (4-HT) and

G418 (Sigma), and N2 supplement (Invitrogen) were purchased.

Generation of EWS-WT1 knock-in mice and animal care

To generate a conditional EWS-WT1 mouse, we followed the strategy that was used to generate a

conditional EWS-FLI1 mouse (25). Briefly, we inserted a loxP-flanked transcriptional ‘STOP’ cassette

(26) in an antisense direction into Ews intron 6 (Figure 1A). A human WT1(-KTS) cDNA (exons 8-10)

was fused to Ews exon 7 to create an EWS/WT1(-KTS) ‘knock-in’ allele. The targeting construct was

used to generate correctly targeted mouse ES cells (TC-1 (27)) as determined by Southern blot analysis

(25). Positive ES clones were injected into C57BL6 blastocysts (NIDDK Mouse Knockout Core) and

the resulting chimeras were crossed to C57BL6 females to achieve germ-line transmission. A

conditional EWS-WT1(+KTS) knock-in mouse was generated previously (6). A transgenic mouse

constitutively expressing the CreER allele (B6.Cg-Tg(CAG-cre/Esr1)5Amc/J) was purchased (Jackson

Laboratory). The primers for genotyping EWS-WT1 or CreER mice are described in Supplemental

information. All animal procedures were approved and handled according to the guidelines provided by

the Tulane Institutional Animal Care and Use Committee and by the NIH Animal Research Advisory

Committee.

Microarray analysis

MEFs harboring E-KTS/CreER+, E+KTS/CreER

+ or CreER

+ were either untreated or treated with 1

M 4-HT. At 24h, total RNA was prepared using RNeasy Kit (Qiagen). Gene expression profiling with

or without (reference) 4-HT was performed using Affymetrix Mouse Genome 430 2.0 arrays. Three

biological replicates were analyzed for each sample. The data were analyzed using an Affymetrix RMA

algorithm. Genes with greater than 1.5-fold difference and P value <0.05 were selected by ANOVA

using Partek Pro (Partek, St. Charles, MO). The heat maps were generated by using either RMA raw




signal values or fold change values from ANOVA lists by Partek. Microarray data have been submitted

to GEO database (GSE53301). GO analysis of primary tumors were performed using DAVID v6.7

(The Database for Annotation, Visualization and Integrated Discovery) (28, 29).

Real-time qRT-PCR

Total RNA was converted to cDNA using SuperScript III Reverse Transcriptase (Invitrogen). Each

sample was analyzed by real-time quantitative RT-PCR (qRT-PCR) using a TaqMan probe for Ascl1

(Mm03058063_m1) (Applied Biosystems). The relative transcript quantity was calculated by

comparative Ct method normalized against Gapdh.

Promoter-reporter assays

A human ASCL1 promoter (-1043 to +233) luciferase reporter construct was purchased (GeneCopoeia)

and subcloned into pGL3Basic vector (Promega). The ASCL1 promoter-luciferase construct, along with

Renilla luciferase plasmid, was transfected into JN-DSRCT-1 and HEK293 cells by Lipofectamine

2000 (Invitrogen). At 48h post-transfection, luciferase activities were measured using Dual Luciferase

Assay Kit (Promega). Site-specific mutations (M1-M3) in the ASCL1 promoter were generated by

DNA synthesis (Blue Heron Biotechnology) and confirmed by sequencing.

Chromatin immunoprecipitation

ChIP assay was performed as described (30) using anti-WT1 (C19, Santa Cruz) or anti-RNA Pol II

(Millipore) antibodies. The primers used to amplify the ASCL1 promoter are listed in Supplemental

information.

Colony formation assay

Cells were transduced with lentiviruses expressing shWT1 (shWT1-2 or shWT1-3), shASCL1

(shASCL1-1 or shASCL1-2) or scrambled control and cultured in presence of puromycin for 2 weeks.

After staining with crystal violet, colonies were counted and photographed. Three independent

experiments were performed in triplicates. The shWT1 and shASCL1 sequences (Sigma) are listed in

Supplemental information.

Immunofluorescence




Cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.5% Triton X-100 for 10

min, blocked with 5% goat serum (Sigma) and incubated with the following primary antibodies: mouse

Tuj1 (1:1,000) or rabbit anti-MAP2 (1:1,000). Alexa Fluor 488 and 594 (Invitrogen) were used as

secondary antibodies. Confocal microscopy was performed using an inverted laser scanning confocal

microscope (Zeiss Axiovert 200M) and analyzed with the LSM 510 confocal software (Zeiss).




Results

Generation of mice harboring conditional EWS-WT1 knock-in alleles

To determine the physiological mechanisms underlying EWS-WT1 driven tumorigenesis, we decided to

express EWS-WT1 under the control of native Ews transcriptional network. We fused the WT1(-KTS)

cDNA encoding the last three zinc fingers of WT1 lacking the KTS to the mouse Ews exon 7 (Fig. 1A),

recreating the exact fusion transcript found in DSRCT. Similar to EWS-FLI1 mice (25), the loxP-

flanked transcriptional ‘STOP’ was inserted in the antisense direction, which was essential for the

successful targeting of EWS-WT1(-KTS). The STOP inserted in the sense direction completely

prevented homologous recombination (data not shown), suggesting that, similar to EWS-FLI1, a leaky

EWS-WT1(-KTS) expression impedes ES cell growth. Successfully targeted ES cells were used to

generate a conditional EWS-WT1(-KTS) knock-in mouse. We have previously generated a conditional

EWS-WT1(+KTS) knock-in mouse (6). We herein designate heterozygous mice carrying EWS-WT1(-

KTS) as E-KTS and EWS-WT1(+KTS) as E+KTS. E-KTS and E+KTS mice appeared healthy and were

backcrossed to C57BL6 mice for more than 8 generations.

Inducible expression of E-KTS and E+KTS

We first crossed E-KTS or E+KTS mice with EIIa-Cre transgenic mice (general deleter line), but we

only obtained E+KTS;Cre+ mice (data not shown), suggesting that constitutive expression of E-KTS

caused lethality. Therefore, we crossed E-KTS or E+KTS mice with a transgenic mouse constitutively

expressing Cre recombinase fused to mutated estrogen receptor (CreER), allowing an inducible

expression of EWS-WT1 with tamoxifen (Supplement Fig. 1A). To test the efficiency, we

intraperitoneally (i.p.) administered 3 doses of tamoxifen (3mg/40g body weight, given every 3 days)

and examined CreER-mediated excision of STOP by PCR. We observed varying degrees of

recombination in different tissues, with the highest recombination observed in the kidney and pancreas,

intermediate recombination in the brain, heart and lung, and very little recombination in the liver and

spleen (Supplement Fig. 1B). Tamoxifen did not induce recombination in any of the tissues in E-KTS

mice without CreER. To determine the cytotoxicity of E-KTS or E+KTS, we administered 3 doses of

tamoxifen and monitored the mice daily. Notably, all E-KTS;CreER+ mice (n=4, age 6-8 weeks) died

by 10-11 days from the first tamoxifen injection, while tamoxifen-treated E+KTS;CreER+ (n=4) or E-

KTS;CreER- (n=4) mice lived over 16 months. These observations indicate that global expression of E-

KTS, but not E+KTS, causes lethality in mice.




Expression of E-KTS, but not E+KTS, leads to inhibition of cell growth

To examine the cytotoxic effects of E-KTS in detail, we derived MEFs from E-KTS;CreER+ and

E+KTS;CreER+ embryos. We observed near complete excision of STOP in both MEFs following 4-

hydroxytamoxifen (4-HT) treatment (Fig. 1B). There was no recombination in the absence of 4-HT or

without CreER expression. Similar to EWS-FLI1 MEFs (25), an aberrantly spliced transcript (EWS-

STOP-WT1) containing 147-bp derived from the antisense STOP is present in either MEFs in the

absence of 4-HT (Fig. 1C). Upon 4-HT-mediated excision of STOP, expression of the correctly spliced

E-KTS and E+KTS transcripts (EWS-WT1) emerged at 6h and reached maximal levels at 24h (Fig.

1C). Expression of E-KTS and E+KTS proteins was detectable at 12h and reached maximal levels at

24h, but the expression of E-KTS was substantially lower than E+KTS (Fig. 1D). As previously

demonstrated for EWS-FLI1, no detectable E-KTS protein was generated in the absence of 4-HT (Fig.

1D), likely due to a rapid degradation of the proteins generated from the aberrantly spliced transcripts

(25). Strikingly, expression of E-KTS resulted in a rapid cessation of cell growth while E+KTS

expressing cells grew normally (Fig. 2A). To gain insights into the E-KTS-induced growth arrest, we

examined expression of various cell cycle proteins by immunoblotting. Expression of key cell cycle

regulators, Cyclin A and Cyclin D1, was markedly reduced in cells expressing E-KTS at 48h compared

to CreER MEFs (Fig. 2B). Notably, phosphorylated AKT was absent in E-KTS expressing cells even

though total AKT levels were comparable. However, expression of other cell cycle proteins and various

CDK inhibitors (Fig. 2B, right panels) was unaltered. MEFs expressing E-KTS did not undergo

apoptosis as determined by PARP cleavage (data not shown).

Genome-wide gene expression profiling

To gain insights into the E-KTS- and E+KTS-mediated transcriptional regulation, we performed whole-

genome expression analysis. Interrogation of over 39,000 probe sets with RNA isolated from E-KTS or

E+KTS expressing MEFs revealed that 3,228 transcripts (2,051 induced and 1,177 repressed) showed

significant expression changes (p<0.05 and >1.5-fold change) following E-KTS expression while 1,557

transcripts (780 induced and 777 repressed) showed significant alterations upon E+KTS expression

(Supplement Fig. 2A). About 300 transcripts were significantly altered by 4-HT treatment in CreER

MEFs, demonstrating a relatively small effect of 4-HT. Notably, there was a very little overlap of genes

regulated by the two isoforms, suggesting that each isoform is recruited to distinct promoter/enhancer




regions and regulate different genes. Gene Ontology (GO) analysis revealed that a large number of

neural-related genes were induced by E-KTS but not by E+KTS (Supplement Table 1 and Fig. 3A).

This was surprising since EWS-WT1 has never been shown to regulate neural genes. However, it is

well established that most primary DSRCTs express neural markers such as Neuron Specific Enolase

(NSE) and S100 protein (3). GO analysis revealed that expression of E+KTS resulted in a repression of

genes involved in DNA replication and repair pathways (Supplement Table 1).

Neural genes are overexpressed in primary DSRCT

To determine whether neural genes are enriched in primary DSRCT, we analyzed expression profile of

28 primary DSRCT performed previously in comparison to four other tumor types (31). GO analysis of

genes that are up-regulated (>2-fold) in DSRCT compared to either Ewing sarcoma (ES), alveolar

rhabdomyosarcoma (ARMS) or alveolar soft part sarcoma (ASPS) revealed an enrichment of neural

pathways (Supplement Table 2), but not when compared to synovial sarcoma (SS). The enrichment of

neural pathways was not evident by genes that were repressed in DSRCT (data not shown). We next

examined the expression of all the genes listed in the identified GO neural pathways in 137 primary

tumors. Remarkably, a large number of these neural genes were uniquely overexpressed in 28 DSRCT

compared to other tumors (Supplement Fig. 3 and Supplement Table 2). A smaller subset of neural

genes was also uniquely overexpressed in SS, which might explain the failure of our GO analysis to

reveal neural pathway enrichment when DSRCT was compared to SS. Notably, a subset of neural

genes that were highly activated by E-KTS in MEFs was also enriched in primary DSRCT, including

ASCL1, PLXNB1 (Plexin B1) and NTRK3 (Neurotrophic Tyrosine Kinase, Receptor 3) (Fig. 3B). These

results suggest that neural gene expression in DSRCT might be directly regulated by E-KTS.

ASCL1 is directly activated by E-KTS

ASCL1 is one of the neural reprogramming factors that directly converts fibroblasts into neurons (32,

33). Given the enrichment of neural genes in DSRCT, we next determined whether E-KTS directly

activated ASCL1 transcription. 4-HT-mediated expression of E-KTS, but not E+KTS or CreER,

resulted in a massive increase in ASCL1 mRNA (>150-fold) and protein expression in MEFs (Fig. 4A

and B). Expression of E-KTS in a heterologous osteosarcoma cells (U2OS) also resulted in a robust

induction of ASCL1 compared to E+KTS or empty vector control (Supplement Fig. 2B). ASCL1

expression was most abundant in a DSRCT cell line, JN-DSRCT-1, as compared to U2OS or three




Ewing sarcoma cell lines (Fig. 4C). Depletion of EWS-WT1 in JN-DSRCT-1 cells with lentiviruses

containing shRNAs against the 3’ region of WT1 (shWT1-2 or shWT1-3 but not with shWT1-1)

resulted in a successful knockdown of EWS-WT1 (Fig. 4D). Concomitantly, silencing EWS-WT1 led to

almost complete inhibition of ASCL1 expression, but not in controls. We note that endogenous WT1 is

not expressed at detectable levels in JN-DSRCT-1 cells by immunoblotting (data not shown).

Collectively, these results demonstrate that E-KTS is directly responsible for the ASCL1 transcription in

MEFs and JN-DSRCT-1 cells.

EWS-WT1 or ASCL1 expression is critical for DSRCT cell survival

EWS-WT1 is the defining oncogene in DSRCT but it is not known whether continued expression of

EWS-WT1 is necessary to sustain tumor cell growth. Depletion of EWS-WT1 or ASCL1 by two

independent shRNAs in JN-DSRCT-1 cells resulted in a complete loss of tumor cell growth as revealed

by colony formation assay (Fig. 4E-G). These results suggest that persistent expression of EWS-WT1 is

required for tumor cell growth. Interestingly, expression of ASCL1 was also essential for DSRCT

tumor cell growth.

Identification of E-KTS responsive elements in the ASCL1 promoter

To identify the regulatory sequences responsive to E-KTS, we tested a proximal (1.2-kb) human ASCL1

promoter in a luciferase reporter assay. Expression of E-KTS, but not E+KTS, in HEK293 cells

resulted in a concomitant increase in the ASCL1 promoter-driven luciferase activity in a dose-dependent

manner (Fig. 5A). Transfection of the ASCL1 promoter-reporter in JN-DSRCT-1 cells also resulted in a

modest but significant induction of luciferase reporter compared to U2OS or A4573 cells, while a small

(1.5-fold) increase was observed in CHP100 cells (Supplement Fig. 4A). Inspection of human ASCL1

proximal promoter sequences (-1043 to +233) revealed a presence of 12 potential GC-rich E-KTS

binding sites (E-KTS-BS): two upstream (between -951 to -820) and ten within 400-bp of the

transcriptional start site (Fig. 5B). Chromatin immunoprecipitation (ChIP) analysis of JN-DSRCT-1

cells with an antibody against the C-terminus of WT1 demonstrated that EWS-WT1 is mostly bound to

the proximal (-408 to -56) ASCL1 promoter but not in the region (-765 to -384) harboring no E-KTS-

BS. A weak recruitment was detected in the upstream region (-1078 to -743) (Fig. 5B). Consistent with

this, the D2 promoter (-406 to +233) containing the 10 potential E-KTS-BS was fully responsive to E-

KTS as the 1.2-kb promoter (Fig. 5C). Alignment of the human and mouse ASCL1 proximal promoter




regions (-400 to +1) revealed a high conservation of the entire region (86% identity), including the

multiple potential E-KTS-BS (Supplement Fig. 4B).

To determine the precise E-KTS-responsive elements in ASCL1 promoter, we introduced substitution

mutations at 4 distal (M1), 6 proximal (M2) or all 10 (M3) potential E-KTS-BS in the D2 promoter

(Supplement Fig. 4C) and tested them in reporter assays. Destroying the 4 distal sites (M1) resulted in

about 88% reduction in the luciferase activity while abolishing the 6 proximal sites (M2) resulted in

95% inhibition of E-KTS transcription (Fig. 5D). The M3 promoter, which destroyed all E-KTS-BS,

completely (99%) lost the ability to mediate transcription by E-KTS, demonstrating that multiple E-

KTS-BS are necessary for the full activation by E-KTS.

Partial reprogramming of fibroblasts to neuron-like cells by E-KTS

ASCL1, along with POU3F2 (POU domain, class 3, transcription factor 2, also called BRN2) and

MYT1L (Myelin Transcription Factor 1-like), have been shown to directly reprogram fibroblasts into

excitatory neurons (32). Direct reprogramming of fibroblasts into other types of neurons, such as

dopaminergic or motor neuron, can be achieved by a combination of different neural transcription

factors, but ASCL1 is required for nearly all direct neural reprogramming (33). Furthermore, ASCL1,

when expressed alone, has been shown to induce a partial reprogramming of fibroblasts into neuron-

like cells (32). Thus, we tested whether E-KTS, which directly activates ASCL1 expression, could lead

to a partial neural reprogramming of MEFs. Following induction of E-KTS or E+KTS with 4-HT for 2

days, cells were cultured in neural induction media (DMEM/F12 + N2 supplement (1X)) for 10 days

(fresh media given every 2 days). Remarkably, immature neuron-like cells appeared in the E-KTS

expressing MEFs displaying elongated bi- or tri-polar projections that resembled neurites (Fig. 6A).

Immunofluorescence with antibodies against neuron-specific -III-Tubulin (TUBB3 recognized by

Tuj1 antibody) and Microtubule-Associated Protein 2 (MAP2) revealed that about 20% of cells

expressed these neural markers in E-KTS expressing cells (Fig. 6B and Supplement Fig. 5A). There

were less than 5% of Tuj1-positive cells in E-KTS MEFs without 4-HT and no Tuj1-positive cells

appeared in E+KTS MEFs regardless of 4-HT. Extended neural induction period (up to 18 days) did

not increase the number of neuron-like cells nor result in more complex neuronal morphology in E-

KTS MEFs (data not shown). These results demonstrate that E-KTS expression leads to a partial neural

reprogramming of MEFs, likely via ASCL1.




Partial differentiation of JN-DSRCT-1 cells to neuron-like cells

We next examined whether JN-DSRCT-1 cells could be differentiated towards neural lineage by the

endogenous expression of ASCL1. When JN-DSRCT-1 cells were cultured with the neuronal induction

media for 3 days, about 30% of cells displayed short, but occasionally long, bi- or tri-polar Tuj1+

neurite-like projections (Fig. 6C and D). TUBB3 expression was only observed in N2-treated JN-

DSRCT-1 cells but not in the absence of N2 or in U2OS cells with or without N2 (Fig. 6C). Induction

of TUBB3 was observed as early as 24h and continued to increase to 72h (Fig. 6E). Notably, siRNA-

mediated acute depletion of EWS-WT1 led to nearly complete inhibition of Tuj1+ neurite projections

and to marked decrease in TUBB3 expression in JN-DSRCT-1 cells (Supplement Fig. 5B-D),

demonstrating that EWS-WT1 is responsible for the partial neural differentiation.




Discussion

In this study, we report the generation of two conditional EWS-WT1 knockin mice, each expressing a

specific isoform under the control of endogenous EWS transcriptional network. A previous study has

shown that expression of E-KTS, but not E+KTS, was sufficient to transform NIH3T3 cells (17).

Consistent with this, expression of E+KTS in MEFs had no effects on cell growth and mice expressing

E+KTS did not develop any spontaneous tumor (data not shown). However, our study does not

preclude a supporting role for E+KTS in DSRCT tumorigenesis. In contrast, expression of E-KTS

resulted in a rapid cessation of cell growth and rapid death of mice. Though the exact cause of death is

unknown, Ki-67 immunostaining showed that expression of E-KTS caused a significant reduction in

the number of proliferating cells in crypts of intestinal epithelium (Supplement Fig. 6), suggesting a

block in proliferation of gastrointestinal epithelium and malabsorption might be responsible. In support

of this, mice expressing E-KTS lost weight by an average of 3.9g (n=4) in 4 days while the control

mice gained an average of 0.8g (n=3).

Other mouse models expressing oncogenic fusion genes also showed lethal effects (25, 34-39),

suggesting that expression of chimeric oncogenes in primary cells often leads to negative cellular

effects. Therefore, we postulate that in DSRCT, EWS-WT1 translocation must occur in a permissive

cell type(s) or in the context of permissive microenvironment that will allow persistent EWS-WT1

expression. These cells will likely be the tumor cells of origin in DSRCT. An analogous situation exists

in Ewing sarcoma where expression of EWS-FLI1 is toxic to most cell types (25, 40, 41) but its

expression is permitted in mesenchymal stem cells (42, 43) or in neural crest stem cells (44). Therefore,

identifying a cell type that tolerates E-KTS expression is critical to understanding DSRCT

tumorigenesis and for the development of a mouse model.

One of the hallmarks of DSRCT is that tumor cells express neural and other lineage markers (3), but

the rationale for the multilineage expression has not been provided. Our findings demonstrate that E-

KTS induces a number of neural genes including ASCL1, which will likely induce additional neural-

related genes. Thus, this study provides a molecular rationale for the neural gene expression in DSRCT.

Prior to this study, it was not known whether continued expression of EWS-WT1 is required for tumor

cell growth. Our findings showed that persistent EWS-WT1 expression is essential for tumor cell




growth, making it an ideal target for therapy. Surprisingly, depletion of ASCL1 also led to inhibition of

JN-DSRCT-1 cell growth, revealing another potential therapeutic target. When JN-DSRCT-1 cells

were stimulated to undergo neural differentiation, however, ASCL1 expression was further induced

along with its target genes (Supplement Fig. 5E) and partial neural differentiation ensued. Thus,

ASCL1 appears to have dual functions: (1) maintaining tumor cell growth under a proliferative signal

and (2) promoting neural differentiation under a neural differentiation signal.

Direct reprogramming of fibroblasts into neurons requires the expression of three factors: ASCL1,

POU3F2 and MYT1L (32). When we attempted to induce full neural reprogramming in JN-DSRCT-1

cells by expressing POU3F2, MYT1L or NEUROD1, either alone or in combination, expression of any

of these factors under neural differentiation condition resulted in a rapid cell death (data not shown),

demonstrating that JN-DSRCT-1 cells are incompatible with full neural reprogramming. However, a

partial neural differentiation was achieved in JN-DSRCT-1 cells, likely due to E-KTS-activated

ASCL1. These findings suggest a possibility that induction of partial neural differentiation in DSRCT

patients might lead to an arrest or delay of aggressive tumor cell growth, providing a potentially novel

and effective therapy against this incurable disease.




Acknowledgments

We thank Cuiling Li and Chuxia Deng (NIDDK Mouse Knockout Core) for the mouse ES cell injection

and Chithra Keembiyehetty (NIDDK Genomics Core) for performing the microarray analysis and Yun-

Ping Wu (NIDDK) for confocal microscopy. This work was supported in part by the NIDDK

Intramural Research Program and by the Tulane Startup Fund (S.B.L.).




References

1. Gerald WL, Miller HK, Battifora H, Miettinen M, Silva EG, Rosai J. Intra-abdominal desmoplastic

small round-cell tumor. Report of 19 cases of a distinctive type of high-grade polyphenotypic

malignancy affecting young individuals. Am J Surg Pathol. 1991;15:499-513.

2. Gerald WL, Haber DA. The EWS-WT1 gene fusion in desmoplastic small round cell tumor. Semin

Cancer Biol. 2005;15:197-205.

3. Gerald WL, Ladanyi M, de Alava E, Cuatrecasas M, Kushner BH, LaQuaglia MP, 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. 1998;16:3028-36.

4. Ladanyi M, Gerald W. Fusion of the EWS and WT1 genes in the desmoplastic small round cell

tumor. Cancer Res. 1994;54:2837-40.

5. Gerald WL, Rosai J, Ladanyi M. Characterization of the genomic breakpoint and chimeric

transcripts in the EWS-WT1 gene fusion of desmoplastic small round cell tumor. Proc Natl Acad

Sci U S A. 1995;92:1028-32.

6. Li H, Watford W, Li C, Parmelee A, Bryant MA, Deng C, et al. Ewing sarcoma gene EWS is

essential for meiosis and B lymphocyte development. J Clin Invest. 2007;117:1314-23.

7. Cho J, Shen H, Yu H, Li H, Cheng T, Lee SB, et al. Ewing sarcoma gene Ews regulates

hematopoietic stem cell senescence. Blood. 2011;117:1156-66.

8. Azuma M, Embree LJ, Sabaawy H, Hickstein DD. Ewing sarcoma protein ewsr1 maintains mitotic

integrity and proneural cell survival in the zebrafish embryo. PloS one. 2007;2:e979.

9. Park JH, Kang HJ, Kang SI, Lee JE, Hur J, Ge K, et al. A multifunctional protein, EWS, is essential

for early brown fat lineage determination. Dev Cell. 2013;26:393-404.

10. Kim KY, Hwang YJ, Jung MK, Choe J, Kim Y, Kim S, et al. A multifunctional protein EWS

regulates the expression of Drosha and microRNAs. Cell Death Differ. 2013.

11. Dutertre M, Sanchez G, De Cian MC, Barbier J, Dardenne E, Gratadou L, et al. Cotranscriptional

exon skipping in the genotoxic stress response. Nat Struct Mol Biol. 2010;17:1358-66.

12. Paronetto MP, Minana B, Valcarcel J. The Ewing sarcoma protein regulates DNA damage-induced

alternative splicing. Mol Cell. 2011;43:353-68.

13. Ng KP, Potikyan G, Savene RO, Denny CT, Uversky VN, Lee KA. Multiple aromatic side chains

within a disordered structure are critical for transcription and transforming activity of EWS family




oncoproteins. Proceedings of the National Academy of Sciences of the United States of America.

2007;104:479-84.

14. May WA, Lessnick SL, Braun BS, Klemsz M, Lewis BC, Lunsford LB, et al. 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. 1993;13:7393-8.

15. Lee SB, Haber DA. Wilms tumor and the WT1 gene. Exp Cell Res. 2001;264:74-99.

16. Haber DA, Sohn RL, Buckler AJ, Pelletier J, Call KM, Housman DE. Alternative splicing and

genomic structure of the Wilms tumor gene WT1. Proceedings of the National Academy of

Sciences of the United States of America. 1991;88:9618-22.

17. Kim J, Lee K, Pelletier J. The desmoplastic small round cell tumor t(11;22) translocation produces

EWS/WT1 isoforms with differing oncogenic properties. Oncogene. 1998;16:1973-9.

18. Lee SB, Kolquist KA, Nichols K, Englert C, Maheswaran S, Ladanyi M, et al. The EWS-WT1

translocation product induces PDGFA in desmoplastic small round-cell tumour. Nat Genet.

1997;17:309-13.

19. Wong JC, Lee SB, Bell MD, Reynolds PA, Fiore E, Stamenkovic I, et al. Induction of the

interleukin-2/15 receptor beta-chain by the EWS-WT1 translocation product. Oncogene.

2002;21:2009-19.

20. Palmer RE, Lee SB, Wong JC, Reynolds PA, Zhang H, Truong V, et al. Induction of BAIAP3 by

the EWS-WT1 chimeric fusion implicates regulated exocytosis in tumorigenesis. Cancer Cell.

2002;2:497-505.

21. Ito E, Honma R, Imai J, Azuma S, Kanno T, Mori S, et al. A tetraspanin-family protein, T-cell

acute lymphoblastic leukemia-associated antigen 1, is induced by the Ewing's sarcoma-Wilms'

tumor 1 fusion protein of desmoplastic small round-cell tumor. Am J Pathol. 2003;163:2165-72.

22. Li H, Smolen GA, Beers LF, Xia L, Gerald W, Wang J, et al. Adenosine transporter ENT4 is a

direct target of EWS/WT1 translocation product and is highly expressed in desmoplastic small

round cell tumor. PLoS One. 2008;3:e2353.

23. Reynolds PA, Smolen GA, Palmer RE, Sgroi D, Yajnik V, Gerald WL, et al. Identification of a

DNA-binding site and transcriptional target for the EWS-WT1(+KTS) oncoprotein. Genes Dev.

2003;17:2094-107.




24. Nishio J, Iwasaki H, Ishiguro M, Ohjimi Y, Fujita C, Yanai F, et al. Establishment and

characterization of a novel human desmoplastic small round cell tumor cell line, JN-DSRCT-1. Lab

Invest. 2002;82:1175-82.

25. Sohn EJ, Li H, Reidy K, Beers LF, Christensen BL, Lee SB. EWS/FLI1 oncogene activates caspase

3 transcription and triggers apoptosis in vivo. Cancer Res. 2010;70:1154-63.

26. Sauer B. Manipulation of transgenes by site-specific recombination: use of Cre recombinase.

Methods Enzymol. 1993;225:890-900.

27. Deng C, Wynshaw-Boris A, Zhou F, Kuo A, Leder P. Fibroblast growth factor receptor 3 is a

negative regulator of bone growth. Cell. 1996;84:911-21.

28. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists

using DAVID bioinformatics resources. Nat Protoc. 2009;4:44-57.

29. Huang da W, Sherman BT, Lempicki RA. Bioinformatics enrichment tools: paths toward the

comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2009;37:1-13.

30. Kim HS, Kim MS, Hancock AL, Harper JC, Park JY, Poy G, et al. Identification of novel Wilms'

tumor suppressor gene target genes implicated in kidney development. J Biol Chem.

2007;282:16278-87.

31. Filion C, Motoi T, Olshen AB, Lae M, Emnett RJ, Gutmann DH, et al. The EWSR1/NR4A3 fusion

protein of extraskeletal myxoid chondrosarcoma activates the PPARG nuclear receptor gene. J

Pathol. 2009;217:83-93.

32. Vierbuchen T, Ostermeier A, Pang ZP, Kokubu Y, Sudhof TC, Wernig M. Direct conversion of

fibroblasts to functional neurons by defined factors. Nature. 2010;463:1035-41.

33. Yang N, Ng YH, Pang ZP, Sudhof TC, Wernig M. Induced neuronal cells: how to make and define

a neuron. Cell Stem Cell. 2011;9:517-25.

34. Codrington R, Pannell R, Forster A, Drynan LF, Daser A, Lobato N, et al. The Ews-ERG fusion

protein can initiate neoplasia from lineage-committed haematopoietic cells. PLoS Biol. 2005;3:e242.

35. Haldar M, Hancock JD, Coffin CM, Lessnick SL, Capecchi MR. A conditional mouse model of

synovial sarcoma: insights into a myogenic origin. Cancer Cell. 2007;11:375-88.

36. Higuchi M, O'Brien D, Kumaravelu P, Lenny N, Yeoh EJ, Downing JR. Expression of a

conditional AML1-ETO oncogene bypasses embryonic lethality and establishes a murine model of

human t(8;21) acute myeloid leukemia. Cancer Cell. 2002;1:63-74.




37. Keller C, Hansen MS, Coffin CM, Capecchi MR. Pax3:Fkhr interferes with embryonic Pax3 and

Pax7 function: implications for alveolar rhabdomyosarcoma cell of origin. Genes & development.

2004;18:2608-13.

38. Lin PP, Pandey MK, Jin F, Xiong S, Deavers M, Parant JM, et al. EWS-FLI1 induces

developmental abnormalities and accelerates sarcoma formation in a transgenic mouse model.

Cancer Res. 2008;68:8968-75.

39. Lagutina I, Conway SJ, Sublett J, Grosveld GC. Pax3-FKHR knock-in mice show developmental

aberrations but do not develop tumors. Molecular and cellular biology. 2002;22:7204-16.

40. Deneen B, Denny CT. Loss of p16 pathways stabilizes EWS/FLI1 expression and complements

EWS/FLI1 mediated transformation. Oncogene. 2001;20:6731-41.

41. Lessnick SL, Dacwag CS, Golub TR. The Ewing's sarcoma oncoprotein EWS/FLI induces a p53-

dependent growth arrest in primary human fibroblasts. Cancer Cell. 2002;1:393-401.

42. Castillero-Trejo Y, Eliazer S, Xiang L, Richardson JA, Ilaria RL, Jr. Expression of the EWS/FLI-1

oncogene in murine primary bone-derived cells Results in EWS/FLI-1-dependent, ewing sarcoma-

like tumors. Cancer Res. 2005;65:8698-705.

43. Riggi N, Cironi L, Provero P, Suva ML, Kaloulis K, Garcia-Echeverria C, et al. Development of

Ewing's sarcoma from primary bone marrow-derived mesenchymal progenitor cells. Cancer Res.

2005;65:11459-68.

44. von Levetzow C, Jiang X, Gwye Y, von Levetzow G, Hung L, Cooper A, et al. Modeling initiation

of Ewing sarcoma in human neural crest cells. PLoS One. 2011;6:e19305.




Figure Legends

Figure 1. Generation of inducible EWS-WT1 MEFs. (A) Schematics of conditional EWS/WT1

alleles. pA, poly-adenylation signal; Neo, neomycin-resistance gene. Arrows indicate PCR primers. (B)

CreER-mediated recombination analyzed by PCR of genomic DNA isolated from MEFs cultured with

or without 1μM 4-HT. (C) Inducible expression of EWS/WT1 transcripts was analyzed by RT-PCR in

MEFs treated for the indicated times with 4-HT. EWS-STOP-WT1 indicates the aberrantly spliced

transcript. Gapdh was used as a control. (D) MEFs treated as in (C) were immunoblotted with anti-

WT1 or anti-Actin.

Figure 2. E-KTS induces cell growth arrest in MEFs. (A) E-KTS and E+KTS MEFs were grown

with or without 1μM 4-HT and cell number was counted daily. Three independent experiments were

performed in triplicates. (B) Whole cell lysates from CreER and E-KTS MEFs grown with or without

4-HT were analyzed by immunoblotting with the indicated antibodies.

Figure 3. E-KTS activates neural genes in MEFs and in primary tumors. (A) A heat map of

significantly altered genes in neural pathways in E-KTS expressing MEFs. (B) A heat map of some

neural genes that are enriched in primary DSRCT (n=28) compared to other tumors (31). ARMS:

Alveolar Rhabdomyosarcoma (n=23), ES: Ewing sarcoma (n=28), SS: Synovial sarcoma (n=46) and

ASPS: Alveolar Soft Part sarcoma (n=12).

Figure 4. E-KTS activates ASCL1 expression. (A) MEFs cultured with or without 1μM 4-HT for 24h

were analyzed for expression of Ascl1 by qRT-PCR. Three independent experiments were performed in

triplicates. *P<0.05, Student’s t-test. (B) Whole cell lysates from CreER or E-KTS MEFs grown with

or without 4-HT were immunoblotted with anti-ASCL1 or anti-Actin. (C) U2OS, CHP100, A4573,

RD-ES and JN-DSRCT-1 cells were immunoblotted with anti-WT1, anti-ASCL1 or anti-Actin. The

arrow indicates EWS/WT1. *nonspecific band. Note that EWS-WT1 migrates higher (62kDa) in JN-

DSRCT-1 cells due to a different EWS translocation breakpoint (24). (D) JN-DSRCT-1 cells were

transduced with lentivirus carrying three independent shWT1 or control and analyzed by

immunoblotting with anti-EWS, anti-ASCL1 or anti-Actin. (E) JN-DSRCT-1 cells were transduced

with lentiviruses carrying control, two independent shWT1 or two independent shASCL1 and the total

number of colonies was counted and compared to the control (set to 100%). Three independent




experiments were performed in triplicate. *P<0.05, Student’s t-test. (F) Representative images from (E)

are shown. (G) JN-DSRCT-1 cells were transduced with lentiviruses carrying scrambled or two

independent shASCL1 and analyzed by immunoblotting with anti-ASCL1 or anti-Actin.

Figure 5. E-KTS activates ASCL1 via multiple E-KTS-responsive elements. (A) HEK293T cells

were transfected with empty vector or increasing amounts (0.1 g, 0.2 g and 0.6 g) of E-KTS or

E+KTS along with the 1.2kb ASCL1 promoter construct and luciferase activity was measured. Three

independent experiments were performed in triplicates. **

P<0.01, ***

P<0.001, Student’s t-test. (B) ChIP

analysis of ASCL1 promoter in JN-DSRCT-1 cells. Crosslinked chromatin was immunoprecipitated

with rabbit IgG or anti-WT1 (C19) antibody and were amplified by PCR with the indicate primers

(arrows). RNA Pol II antibody was used as a positive control. Circles represent putative E-KTS binding

sites. TSS: Transcription start site (+1). (C) ASCL1 promoter-reporter deletion constructs or a pGL3-

Basic (control) were transfected in JN-DSRCT-1 cells and luciferase activity was measured. (D) The

D2 promoter or mutated (M1, M2 and M3) promoter-reporter constructs were transfected in JN-

DSRCT-1 cells and luciferase activity was measured. The X-circles represent mutated E-KTS binding

sites. Three independent experiments were performed in triplicates. ***

P<0.001.

Figure 6. E-KTS induces partial neural reprogramming. (A) E-KTS and E+KTS MEFs cultured

with or without 1μM 4-HT were cultured for 10 days in N2-containing media. Cells were

immunostained with Tuj1 antibody and DAPI. Scale bar=20 m. (B) The total number of Tuj1-positive

cells was counted from 26 randomly selected fields (n>790). (C) Immunostaining of JN-DSRCT-1 and

U2OS cells with Tuj1 antibody following 3 days of culture with or without N2 media. Scale bar=20 m.

(D) The total number of Tuj1-positive JN-DSRCT-1 cells was counted from 26 randomly selected

fields (n>290). (E) The N2-treated JN-DSRCT-1 cells were immunoblotted with Tuj1 or anti-Lamin

A/C.




Actin




Published OnlineFirst June 16, 2014.Cancer Res Hong-Jun Kang, Jun Hong Park, WeiPing Chen, et al. ASCL1 and promotes neural differentiationEWS-WT1 oncoprotein activates neural reprogramming factor

Updated version

10.1158/0008-5472.CAN-13-3663doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3663.DC1

Access the most recent supplemental material at:

Manuscript

Authoredited. Author manuscripts have been peer reviewed and accepted for publication but have not yet been

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/early/2014/06/14/0008-5472.CAN-13-3663To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-3663

http://cancerres.aacrjournals.org/content/suppl/2014/06/16/0008-5472.CAN-13-3663.DC1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/early/2014/06/14/0008-5472.CAN-13-3663


ews-wt1 oncoprotein activates neuronal reprogramming...

Documents