Synthetic siRNA targeting the breakpoint of EWS/Fli-1 inhibits growth of Ewing sarcoma xenografts in a mouse model

Authors


  • The authors declare that they have no competing interests.

Abstract

The EWS/Fli-1 fusion gene, a product of the translocation t(11;22, q24;q12), is detected in 85% of Ewing sarcomas and primitive neuroectodermal tumors. It is thought to be a transcriptional activator that plays a significant role in tumorigenesis. In this study, we developed a novel EWS/Fli-1 blockade system using RNA interference and tested its application for inhibiting the proliferation of Ewing sarcoma cells in vitro and the treatment of mouse tumor xenografts in vivo. We designed and synthesized a small interfering RNA (siRNA) possessing an aromatic compound at the 3′-end targeting the breakpoint of EWS/Fli-1. As this sequence is present only in tumor cells, it is a potentially relevant target. We found that the siRNA targeting EWS/Fli-1 significantly suppressed the expression of EWS/Fli-1 protein sequence specifically and also reduced the expression of c-Myc protein in Ewing sarcoma cells. We further demonstrated that inhibition of EWS/Fli-1 expression efficiently inhibited the proliferation of the transfected cells but did not induce apoptotic cell death. In addition, the siRNA possessing the aromatic compound at the 3′-end was more resistant to nucleolytic degradation than the unmodified siRNA. Administration of the siRNA with atelocollagen significantly inhibited the tumor growth of TC-135, a Ewing sarcoma cell line, which had been subcutaneously xenografted into mice. Moreover, modification of the 3′-end with an aromatic compound improved its efficiency in vivo. Our data suggest that specific downregulation of EWS/Fli-1 by RNA interference is a possible approach for the treatment of Ewing sarcoma.

The Ewing sarcoma family of tumors (ESFTs) is a group of highly malignant neoplasms that most often affect children and young adults in the first two decades of life. It is the second most common malignant bone tumor and accounts for approximately 10% of all primary bone tumors. Despite aggressive treatment strategies (chemotherapy, radiation therapy and surgery), the long-term disease-free survival rate of patients with ES is still disappointingly low, particularly in poor-risk patients with metastasis. Therefore, identification of new therapeutic targets is urgently needed.

The EWS/Fli-1 fusion gene, a product of the translocation t(11;22, q24;q12), is detected in 85% of ESs and primitive neuroectodermal tumors. The EWS/Fli-1 translocation is formed by the N-terminal domain of the RNA-binding protein EWS and the DNA-binding domain of the ETS family transcriptional factor Fli-1. Identification of several breakpoints for both EWS and Fli-1 has demonstrated that the EWS/Fli-1 fusion genes are heterogeneous. Breakpoints of Type 1 (Exon 7 of EWS/Exon 6 of Fli-1) and Type 2 (Exon 7 of EWS/Exon 5 of Fli-1) are most commonly detected in affected patients.1–3 We have reported that the EWS/Fli-1 fusion protein may be a transcriptional activator that plays a significant role in the tumorigenesis of ESFTs.4–7 Although substantial studies have reported that antagonism of the EWS fusion gene reduces tumorigenicity and clonogenicity,8–10 its detailed biological targets remain unknown. Our previous studies have indicated that the EWS/Fli-1 gene regulates telomerase reverse transcriptase (TERT),11 phospholipase D2,12 Aurora A, Aurora B13 and vascular endothelial growth factor (VEGF).14

RNA interference (RNAi) is a process of sequence-specific posttranscriptional gene silencing triggered by double-stranded RNAs (dsRNAs). These dsRNAs are processed by the enzyme Dicer to generate duplexes of 21 to 23 nucleotides (nt). These duplexes, which contain a 2-nt overhang at the 3′-end of each strand, are termed short interfering RNAs (siRNAs). These siRNAs associate with the RNA-induced silencing complex (RISC), which is then guided to catalyze the sequence-specific degradation of the target mRNA.15, 16 RNAi technologies offer the means to rationally design gene-specific inhibitors and are currently the most widely used techniques in functional genomic studies. However, application of siRNAs in vivo remains difficult because of problems associated with their stability, delivery and therapeutic efficacy.

Here, we demonstrate that silencing of EWS/Fli-1 with RNAi impairs both cell proliferation of ES cell lines and tumor growth in a mouse xenograft model. For our study, we used synthetic siRNA possessing an aromatic compound at the 3′-end and unmodified siRNA, targeting the breakpoint of the EWS/Fli-1 type 1 fusion transcript. Targeting of the EWS/Fli-1 chimeric oncogene using RNAi technology will set a paradigm for the validation of siRNA-based applications for treatment of ES, as EWS/Fli-1 is present only in tumor cells and absent in normal cells. Our results suggest that targeting of EWS/Fli-1 could be a possible therapeutic option for ES.

Material and Methods

The siRNAs

The siRNA was designed on the basis of the target sequence at the breakpoint of EWS/Fli-1 type 1. The unmodified siRNA (siEF) was synthesized by and purchased from Dharmacon (Lafayette, CO). The sense and antisense sequences were as follows: 5′-GCAGAACCCUUCUUAUGACdTdT-3′ (sense) and 5′-GUCAUAAGAAGGGUUCUGCdTdT-3′ (antisense). By using a DNA/RNA synthesizer by the phosphoramidite method, we designed and chemically synthesized siRNA possessing the aromatic compound pyridine (p) at the 3′-end (siEFp; Fig. 1a).17, 18 The sequences were 5′-GCAGAA CCCUUCUUAUGACdTdTp-3′ (sense) and 5′-GUCAUAAG AAGGGUUCUGCdTdTp-3′ (antisense). In addition, we synthesized siRNA with a 3′-end modification targeting Renilla luciferase (siCONT) as a negative control. The sequences were 5′-GGCCUUUCACUACUCCUCAdTdTp-3′ (sense) and 5′-GUACGAGUAGUGAAAGGCCdTdTp-3′ (antisense). A BLAST search against EST libraries was performed to confirm that no other human gene was targeted. All siRNAs were resuspended in RNase-free water to prepare a 20-μM stock solution.

Figure 1.

Characterization of the siRNAs. (a) Structures of the siRNAs. (b) Synthetic siRNAs targeting the breakpoint of EWS/Fli-1 Type 1 reduce the expression of EWS/Fli-1 protein in Ewing sarcoma cells. Western blot analysis of EWS/Fli-1 protein in TC-135, A673 and SK-ES-1 cells treated with siEFp, siEF, or siCONT at 48 hr after transfection. β-actin was used as a loading control. The results are representative of three independent experiments.

Cell culture and transfection

TC-135, A673 and SK-ES-1 are ES cell lines carrying EWS/Fli-1. TC-135 and A673 have the EWS/Fli-1 Type 1 fusion, whereas SK-ES-1 has the Type 2 fusion. TC-135 was kindly supplied by Dr. T.J. Triche (University of Southern California, Los Angeles, CA). A673 and SK-ES-1 were purchased from the American Type Culture Collection (Manassas, VA). TC-135 cells were maintained in RPMI 1640 medium (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum (FBS) at 37°C with a 5% CO2 atmosphere. A673 cells were cultured in DMEM with 10% FBS at 37°C under a 10% CO2 atmosphere. SK-ES-1 cells were cultured in McCoy's 5A medium with 10% FBS at 37°C under a 5% CO2 atmosphere. Transfection was carried out in 60-mm dishes (at 40% confluency) using lipofectamine 2000 (Invitrogen), as recommended by the manufacturer. Total protein and mRNA were collected for Western blot analysis and real-time RT-PCR, respectively.

Western blot analysis

Western blot analysis was carried out exactly as described previously.7, 12 All proteins were determined by immunoblotting. The EWS/Fli-1 fusion protein (68 kDa) was sensitively detected by Western blotting using anti-Fli1 antibody (C-19, 1:200 dilution; Santa Cruz Biotechnology, Santa Cruz, CA). Mouse monoclonal antibodies against c-Myc (9E10, 1:200 dilution) and VEGF (C-1, 1:200 dilution) were obtained from Santa Cruz Biotechnology. Mouse monoclonal antibody against TERT (2C4, 1:1000 dilution) was obtained from Novus Biologicals (Littleton, CO). Rabbit polyclonal antibody against poly(ADP-ribose) polymerase (PARP) (9542, 1:1000 dilution) was obtained from Cell Signaling Technology (Boston, MA). Mouse monoclonal antibody against β-actin (JLA20, 1:5000 dilution) was obtained from Calbiochem (San Diego, CA). Quantitative changes in luminescence were estimated by LAS1000 UV mini and Multi Gauge Ver. 3.0 (Fuji Film, Tokyo, Japan).

Real-time quantitative RT-PCR

Total RNA was isolated using a RNeasy mini kit (in vitro) and a midi kit (in vivo; Qiagen, Hilden, Germany). Two micrograms of total RNA was reverse transcribed using a High-Capacity cDNA Reverse Transcription Kit with RNase inhibitor (Applied Biosystems, Foster City, CA). Quantitative real-time PCR analysis using the fluorescent SYBR green method (Bio-Rad, Richmond, CA) was performed in accordance with the manufacturer's instructions. The primers 5′-AGTTACCCACCCCAAACTGG-3′ (forward) and 5′-CCAAGGGGAGGACTTTTGTT-3′ (reverse) were used to amplify EWS/Fli-1, and the primers used for detection of GAPDH were 5′-TCCCATCACCATCTTCCA-3′ (forward) and 5′-ACTCACGCCACAGTTTCC-3′ (reverse). The PCR program consisted of enzyme activation at 95°C for 10 min followed by amplification for 40 cycles (95°C for 30 sec, 61°C for 30 sec, 72°C for 30 sec). Data were generated from each reaction, subjected to gene expression analysis using an iCycler iQ Real-Time PCR Detection System (Bio-Rad) and normalized against GAPDH.

Cell viability assay

Cell proliferation was determined by WST-8 assay using a Cell Counting Kit (Dojin, Kumamoto, Japan). Experiments were carried out in accordance with the manufacturer's recommended procedures. Briefly, cells (5 × 103 cells/well) were incubated overnight in a 96-well plate and then transfected with siRNA duplex. After incubation for the indicated time, the Cell Counting Kit reagents were added to the culture. After a further 1 hr of incubation, the absorbance at 450 nm was measured with a microplate reader. All experiments were performed at least four times.

Analysis of DNA synthesis by labeling with 5-bromo-2′-deoxyuridine

DNA synthesis was analyzed using a 5-bromo-2′-deoxyuridine (BrdU) Labeling and Detection Kit I (Roche Diagnostics, Mannheim, Germany). Experiments were carried out in accordance with the manufacturer's recommended procedures. Briefly, TC-135 cells grown on glass coverslips were treated with siRNAs (50 nM) as described above. After 72 hr of transfection, the cell culture medium was removed, and the BrdU-containing medium was added to the cells for 1 hr. The cells were then fixed with 70% ice-cold ethanol, and the BrdU was detected by immunofluorescence using anti-BrdU antibody. The cells were subsequently labeled with propidium iodide (PI) stain before being examined using a fluorescence microscope (BIOREVO BZ-9000; KEYENCE, Osaka, Japan).

Apoptosis detection

Apoptosis was assessed by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay using an In Situ Cell Death Detection Kit (Roche). Experiments were carried out in accordance with the manufacturer's recommended procedures. Briefly, siRNA (50 nM)-treated TC-135 cells grown on glass coverslips were fixed with 4% paraformaldehyde in PBS, permeabilized with 0.2% Triton X-100 in PBS and stained with the TUNEL reaction mixture. The cells were finally labeled with Hoechst stain before being examined using a fluorescence microscope.

Serum stability

The siRNAs (5 μM) were incubated at 37°C in 10% FBS (Invitrogen) diluted in PBS. Aliquots of the reaction mixtures were collected at different times. The nuclease reactions were stopped by adding RNase inhibitor (Applied Biosystems). All samples were subjected to electrophoresis in 15% polyacrylamide-TBE under nondenaturing conditions and visualized by staining with GelRed (Biotium, Hayward, CA).

Partial hydrolysis of oligoribonucleotide with snake venom phosphodiesterase

Each oligoribonucleotide (ON) (300 pmol) labeled with fluorescein at the 5′-end was incubated with snake venom phosphodiesterase (SVPD) (5 × 10−3 units) in a buffer containing 37.5 mM Tris-HCl (pH 8.0) and 50 mM MgCl2 (total 100 μL) at 37°C. At appropriate time points, aliquots of the reaction mixture were separated and added to a solution of 7 M urea. The solutions were analyzed by 20% polyacrylamide gel containing 7 M urea. The labeled ON in the gel was visualized by a Bio-imaging analyzer (LAS-4000; Fuji Film).

Tumor therapy

Male BALB/c athymic (nu/nu) nude mice (7 weeks old) were obtained from Japan SLC (Tokyo, Japan). The mice were housed in the animal facilities of the Division of Animal Experiment, Life Science Research Center, Gifu University. A total of 3.0 × 106 TC-135 cells in 0.1 mL of serum-free RPMI1640 were inoculated subcuaneously through a 26-gauge needle into the posterior flank. Tumor diameters were measured with digital calipers, and the tumor volume in mm3 was calculated using the formula: volume = (width)2 × length/2. Once tumors had reached a volume of 50–60 mm3 (Day 0), the tumor-bearing nude mice were treated with siEFp or siEF together with atelocollagen (Atelogene Local Use; Koken, Tokyo, Japan). The final concentration of atelocollagen was 1.75% and that of the siRNA was 500 pmol/tumor (10 μM in a 50-μL injection volume for each tumor). The dose of the siRNA was chosen on the basis of our previous study using VEGF siRNA mixed with atelocollagen against ES xenografts.14 As a control, PBS mixed with atelocollagen was injected. Each therapeutic reagent was injected into the tumors on Days 0, 1, 3, 7, 14 and 21. Tumor growth was measured every 3 days for a period of up to 4 weeks. The mice were then sacrificed, and the tumors were removed at specified times during treatment or at the end of the experiment. Part of each tumor was snap frozen in liquid nitrogen for determination of EWS/Fli-1 mRNA levels by quantitative real-time RT-PCR, and another part was fixed in formalin for immunohistochemical analysis. Animal experiments in this study were performed in compliance with the guidelines of the Institute for Laboratory Animal Research, Gifu University Graduate School of Medicine, and the UCCCR guidelines for the Welfare of Animals in Experimental Neoplasia.

Immunohistochemistry

Tumor sections were dewaxed and rehydrated. Endogenous peroxidase activity was blocked with 3% H2O2 for 30 min. For Ki-67 staining, sections were blocked with 2% bovine serum albumin in PBS for 60 min and then incubated overnight at 4°C in a 1:100 dilution of anti-human Ki-67 monoclonal mouse antibody (Dako, Copenhagen, Denmark) in serum block. Subsequently, the sections were incubated with biotinylated secondary antibodies (LSAB2 kit; Dako) for 30 min, followed by incubation with peroxidase-labeled streptavidin (LSAB2 kit; Dako) for 30 min. The sections were developed with 3,3′-diaminobenzidine (DAB) and then counterstained with hematoxylin.

Statistical analysis

Statistical analyses were carried out using GraphPad Prism Version 5.01 (GraphPad Software, CA). The data were analyzed using ANOVA, and differences at P < 0.05 were considered statistically significant.

Abbreviations

ActD: actinomycin D; BrdU: 5-bromo-2′-deoxy-uridine; dsRNAs: double-stranded RNAs; ESFTs: Ewing sarcoma family of tumors; FBS: fetal bovine serum; nt: nucleotides; ON: oligoribonucleotide; PI: propidium iodide; PARP: poly(ADP-ribose) polymerase; RISC: RNA-induced silencing complex; RNAi: RNA interference; siRNA: small interfering RNA; SVPD: snake venom phophodiesterase; TERT: telomerase reverse transcriptase; TUNEL: terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling; VEGF: vascular endothelial growth factor

Results

Synthetic siRNAs targeting the breakpoint of EWS/Fli-1 type 1 reduce the expression of EWS/Fli-1 protein in TC-135 and A673 cells but not in SK-ES-1 cells

We designed two siRNAs targeting the breakpoint of EWS/Fli-1 Type 1: a siRNA possessing the aromatic compound pyridine in the 3′-overhang region (siEFp) and an unmodified siRNA (siEF). To investigate their sequence specificity, these siRNAs were used to treat TC-135 and A673 cells, which carry the Type 1 EWS/Fli-1 fusion, and SK-ES-1 cells, which carry the Type 2 fusion. Both siRNAs reduced the level of EWS/Fli-1 protein expression in TC-135 and A673 cells but had no apparent effect on that of SK-ES-1 cells (Fig. 1b). The siRNA targeting luciferase (siCONT) showed no effect.

Effects of synthetic siRNAs targeting the breakpoint of EWS/Fli-1 type 1 in TC-135 cells

The potency of the siRNA modified with the aromatic compound pyridine at the 3′-end was investigated by analyzing the levels of EWS/Fli-1 mRNA and protein in TC-135 cells at different time points after transfection. The siEFp (50 nM) and siEF (50 nM) significantly downregulated the expression of EWS/Fli-1 mRNA at 24 hr after transfection. However, at 72 hr after transfection, we found that the downregulation of EWS/Fli-1 mRNA by siEFp was significantly greater than that by siEF (Fig. 2a). Decreased mRNA levels were reflected at the protein levels. Figure 2b shows the results of immunoblot analysis and the relative intensities of EWS/Fli-1 protein expression standardized against β-actin. It was found that, at 48 hr after transfection, siEFp (50 nM) and siEF (50 nM) reduced the expression of EWS/Fli-1 protein to a similar extent. On the other hand, at 96 hr after transfection, siEFp downregulated the expression of EWS/Fli-1 protein to a greater degree than siEF. These inhibitory effects were dependent on the concentration of the siRNAs, and maximal suppression of EWS/Fli-1 protein was detected when 50 nM was used (Fig. 2c).

Figure 2.

Effects of synthetic siRNAs targeting the breakpoint of EWS/Fli-1 Type 1 in TC-135 cells. TC-135 cells were transfected with siEFp, siEF, siCONT or lipofectamine alone (mock). (a) Decreased expression of EWS/Fli-1 mRNA at 24 hr (left) and 72 hr (right) after transfection with various siRNAs (50 nM). Knockdown efficiency was measured quantitatively by real-time reverse transcription-PCR (RT-PCR) analysis. Levels of EWS/Fli-1 mRNA expression were obtained subsequent to normalization with constitutively expressed GAPDH mRNA. Each bar represents the mean ± SD (n = 4 dishes). ***P < 0.001 vs. siCONT and mock. ##P < 0.01 vs. siEF. (b) Western blot analysis of EWS/Fli-1 protein performed on a lysate of TC-135 cells at 48 hr (left) and 96 hr (right) after transfection with siRNAs (50 nM). β-actin was used as a loading control. The bands of EWS/Fli-1 were quantified, and their ratios relative to β-actin were calculated. Each bar represents the mean ± SD (n = 4 dishes). ***P < 0.001 vs. siCONT and mock. #P < 0.05 vs. siEF. (c) Inhibition of EWS/Fli-1 protein expression in TC-135 cells transfected with the indicated amounts of siRNAs. The cells were harvested 48 hr after transfection, and cell extracts were analyzed by immunoblotting. β-actin was used as a loading control. Each bar represents the mean ± SD (n = 4 dishes). (d) siRNAs targeting EWS/Fli-1 (50 nM) downregulated the protein levels of c-Myc, TERT and VEGF, as shown by Western blot analysis. β-actin was used as a loading control. The results are representative of three independent experiments.

Because EWS/Fli-1 is known to activate the c-Myc promoter,19 we tested whether siRNAs targeting EWS/Fli-1 had any effect on the level of c-Myc protein expression in TC-135 cells. It was found that the downregulation of EWS/Fli-1 protein caused by siEFp and siEF paralleled the downregulation of c-Myc protein (Fig. 2d). Our previous studies had shown that EWS/Fli-1 regulates TERT and VEGF.11, 14 Therefore, we checked whether EWS/Fli-1 inhibition affected the expression of these proteins. The results (Fig. 2d) showed that siEFp and siEF decreased the level of both TERT and VEGF.

EWS/Fli-1 knockdown correlates with decreased proliferation of ES cells

To study the cellular effects of EWS/Fli-1 knockdown in ES cell lines, TC-135, A673 and SK-ES-1 cells were transfected with the nonfunctional siCONT as a negative control or with the functional siEFp and siEF to specifically knockdown EWS/Fli-1. Cell proliferation was examined using a Cell Counting Kit over different time periods. Both siRNAs targeting EWS/Fli-1 Type 1 efficiently inhibited the proliferation of TC-135 and A673 cells but had no remarkable effect on SK-ES-1 cells (Fig. 3a). We also found that these growth-inhibitory effects were dependent on the concentration of the siRNAs (Fig. 3b). Thus, siEFp and siEF appear to exert antiproliferative activity against ES cells carrying the Type 1 EWS/Fli-1 fusion gene.

Figure 3.

Effect of EWS/Fli-1 knockdown on proliferation of Ewing sarcoma cells. (a) TC-135, A673 and SK-ES-1 cells were treated with siEFp (50 nM), siEF (50 nM), siCONT (50 nM) or lipofectamine alone (mock). Forty-eight and 96 hr later, cell viability was determined with a Cell Counting Kit. The results represent the mean ± SD of four independent experiments, each performed in triplicate. ***P < 0.001 vs. siCONT and mock. (b) TC-135 cells were treated for 96 hr with the indicated amounts of siRNAs, and cell viability was determined with a Cell Counting Kit. The results represent the mean ± SD of cell viability relative to the viability of cells treated with mock.

Knockdown of EWS/Fli-1 by synthetic siRNAs in TC-135 cells impairs BrdU incorporation but does not induce apoptotic cell death

To investigate the mechanism of action of synthetic siRNAs targeting EWS/Fli-1, we used BrdU labeling to examine DNA synthesis during mitosis. In TC-135 cells treated for 72 hr with siEFp and siEF, we observed a significant decrease of BrdU-positive cells in comparison with siCONT- and mock-treated cells (Fig. 4a).

Figure 4.

Knockdown of EWS/Fli-1 by synthetic siRNAs in TC-135 cells impairs BrdU incorporation but does not induce apoptotic cell death. (a) TC-135 cells were treated with siEFp, siEF, siCONT or lipofectamine alone (mock) for 72 hr. The results were representative immunofluorescent images after staining with BrdU (green) and PI (DNA, red). Quantitative data were obtained by counting BrdU-positive cells in five different fields per slide. The results represent the mean ± SD. ***P < 0.001 vs. siCONT and mock. (b) Similarly treated TC-135 cells were stained using an In Situ Cell Death Detection Kit, followed by Hoechst labeling. The number of TUNEL-positive cells was counted and the ratio calculated in five different fields per slide. As positive controls, cells incubated for 6 hr with 50 ng/mL actinomycin D (ActD) were analyzed. (c) TC-135 cells were treated with siRNAs for 96 hr and lysed, and protein samples were then separated by SDS-PAGE. The intact (116 kDa) and catalyzed (89 kDa) forms of PARP protein were detected by using anti-PARP antibody. As positive controls, protein samples from the cells treated with 50 ng/mL ActD for 6 hr were analyzed. β-actin was used as a loading control. The results are representative of three independent experiments.

We then investigated whether the observed siRNA-induced reduction of viability occurred via induction of apoptosis, using TUNEL and Hoechst staining to measure nuclear condensation and fragmentation. TC-135 cells treated for 72 hr with siRNAs were stained with the In Situ Cell Death Detection Kit as described in “Materials and Methods.” The results revealed no variations in the percentage of TUNEL-positive cells after any of the treatments (Fig. 4b).

An early transient burst of poly(ADP-ribosyl)ation of nuclear proteins has been shown to be required for apoptosis to proceed in various cell lines, followed by cleavage of PARP, catalyzed by caspase-3.20, 21 Although treatment of TC-135 cells with siEFp and siEF downregulated the expression of EWS/Fli-1, this did not result in accumulation of detectable levels of cleaved PARP (Fig. 4c). These results show that knockdown of the EWS/Fli-1 fusion protein by siEFp and siEF resulted in inhibition of cell proliferation, although apoptotic cell death was not induced.

Effect of 3′-end modification with an aromatic compound on siRNA stability

We examined the effect of nucleases on the stability of siRNA in FBS by modifying the 3′-end with an aromatic compound. As shown in Figure 5a, the unmodified siEF was fully degraded within 12 hr, whereas the modified siEFp showed only weak signs of degradation after 0.5, 1 and 3 hr. Increased stability was also observed with siEFp, where little full-length product remained at 12 hr.

Figure 5.

Analysis of siRNA stability. (a) siEF and siEFp were incubated in 10% FBS at 37°C for 0, 0.5, 1, 3, 6, 12 or 24 hr, and aliquots were analyzed on 15% polyacrylamide gels. (b) 20% PAGE of 5′-fluorescein-labeled ONs hydrolyzed by SVPD. ONs were incubated with SVPD for 0, 1, 5, 10, 15, 30 min and 1 and 3 hr.

Next, the susceptibility of the ON to SVPD, a 3′-exonuclease, was examined. The antisense strands of siEF and siEFp labeled at the 5′-end with fluorescein were incubated with SVPD. The reactions were analyzed with PAGE. Figure 5b shows the results. The unmodified siEF was hydrolyzed randomly after 5 min of incubation, whereas the modified siEFp was resistant to enzyme. Taken together, our data demonstrate that 3′-end modification with an aromatic compound on siRNA can increase nuclease resistance.

Treatment of TC-135 xenografts with siRNAs targeting EWS/Fli-1

We next investigated the therapeutic effectiveness of siRNAs targeting EWS/Fli-1. We established a xenograft model of TC-135 cells as described in “Materials and Methods” and examined intratumoral treatment with siRNAs targeting EWS/Fli-1 together with atelocollagen. When the tumors reached a volume of 50–60 mm3, the animals were randomized into four groups—siEFp, siEF, siCONT and PBS—and tumor growth was followed up over a period of up to 4 weeks. As shown in Figure 6a, siEFp and siEF markedly suppressed tumor growth in comparison with siCONT or PBS. Moreover, we found that siEFp significantly inhibited tumor growth in comparison with siEF. No death, loss of body weight or gross adverse effects occurred in the mice as a result of treatment with the siRNAs. Furthermore, application of siRNAs targeting EWS/Fli-1 decreased the production of EWS/Fli-1 mRNA in the tumors (Fig. 6b). We also examined the expression of Ki-67, considered an indicator of cell proliferation, using immunohistochemistry. As shown in Figure 6c, the percentage of Ki-67-positive cells was significantly reduced in tumors from mice treated with siEFp and siEF.

Figure 6.

Inhibition of tumor growth by synthetic siRNAs targeting EWS/Fli-1 in the TC-135 xenograft model. (a) Tumor growth curves after treatment with siEFp, siEF, siCONT or PBS. Each therapeutic reagent was injected into the tumors on days 0, 1, 3, 7, 14 and 21 (arrows). The results represent mean ± SE (n = 6 tumors). *P < 0.05; ***P < 0.001, when compared with siCONT and PBS. #P < 0.05 when compared with siEF. (b) Levels of EWS/Fli-1 mRNA in tumors were quantified by real-time quantitative RT-PCR. Each bar represents the mean ± SD (n = 4 tumors). **P < 0.01 vs. siCONT and PBS. (c) Representative micrographs of immunohistochemical detection of Ki-67 in tumors induced by injection into nude mice of TC-135 cells treated as indicated (left). Ki-67-positive cells were counted in each of five independent areas. The percentage of Ki-67-positive cells was then calculated. The results represent the mean ± SD. **P < 0.01 versus siCONT. Data are those for tumors excised 3 days into the treatment schedule.

Discussion

ESFTs are very aggressive pediatric malignancies. Despite the use of multimodal therapy, their prognosis remains poor, with an overall 5-yr survival of 55–65% in patients with localized disease, or even lower in poor-risk patients.22, 23 These tumors are characterized by the presence, in over 85% of cases, of a chromosomal translocation that results in the generation of the chimeric gene, EWS/Fli-1. We have already studied EWS/Fli-1 and established some of the effects of this oncoprotein.4–10, 12–14 The results obtained so far indicate that EWS/Fli-1 has potential as a unique therapeutic target for the treatment of ESFTs. Inhibition of the expression of EWS/Fli-1 in ES cells using antisense oligonucleotide results in decreased proliferation, suggesting a potential therapeutic intervention directed at this oncogene.24–26 Reduction of EWS/Fli-1 by siRNA has recently been investigated by other groups.27–29 However, to our knowledge, no previous report has indicated that chemically synthesized siRNA targeting the breakpoint of EWS/Fli-1 with non-viral delivery inhibits the growth of human ES subcutaneous xenografts. This study demonstrated that knockdown of EWS/Fli-1 using synthetic siRNAs inhibited both the proliferation of ES cells and the growth of human ES tumor xenografts in a mouse model.

The silencing of gene expression by siRNA is a powerful tool for genetic analysis of mammalian cells and has the potential for further development into a specific, potent and safe treatment for human disease. However, application of siRNAs in vivo and their possible use for therapy still has several critical hurdles that have not yet been comprehensively addressed. For instance, the delivery, stability and pharmacokinetics of siRNA are major problems. So far, many types of siRNAs modified at the base or carrying sugar or phosphate moieties have been synthesized, and their nuclease-resistant properties and RNAi-inducing activities have been studied.30–33 Recently, we have developed and synthesized siRNAs possessing aromatic compounds at the 3′-end.17, 18 In this study, we first confirmed that siEFp significantly downregulated the expression of EWS/Fli-1 mRNA and protein and decreased the proliferation of ES cells. We then demonstrated that siEFp significantly inhibited tumor growth in our xenograft model. Furthermore, it was of considerable interest that siEFp was more potent than siEF against tumor xenografts in vivo despite the lack of significant changes observed in the in vitro proliferation. To test whether these modifications were capable of increasing the stability of siRNA, we incubated siEFp and siEF in bovine serum or SVPD, followed by separation on polyacrylamide gels. It was found that 3′-end modification with the aromatic compound was effective for improving the nuclease resistance of the siRNAs. These characteristics potentially improved the efficacy of siEFp, especially in vivo, where sustained delivery will be one major obstacle. In addition, to overcome the problem related to siRNA delivery, we used atelocollagen as a carrier of siRNAs in an in vivo experiment, as described previously.14, 34–38 It is known that atelocollagen has the ability to extend the half-life of siRNA and to keep it intact when embedded in the body. Atelocollagen is already used clinically and considered to be innocuous. Therefore, we consider that the clinical application of chemically synthesized siRNA with atelocollagen represents a simple and an attractive delivery system for siRNA in vivo.

Argonaute2, a key component of RISC, is responsible for mRNA cleavage in the RNAi pathway.39, 40 It is composed of PAZ, Mid and PIWI domains. X-ray structural analysis and nuclear magnetic resonance studies have revealed that the 3′-overhanging region of the guide strand (antisense strand) of siRNA is recognized by the PAZ domain and is accommodated into its hydrophobic binding pocket.41–43 We hypothesized that the introduction of a lipophilic aromatic compound at the 3′-end of the siRNA would improve the affinity of the 3′-end for the PAZ domain, thus improving the effectiveness of the siRNA in comparison with the unmodified form.

The c-Myc, a proto-oncogene, is expressed at high levels in most human cancers and encodes a transcription factor implicated in various cellular processes such as cell growth, proliferation and loss of differentiation.44 Downregulation of c-Myc could, therefore, play a role in a potential therapeutic strategy against human cancers.45, 46 Previous studies have indicated that c-Myc transcription is strongly upregulated by EWS/Fli-1.19, 47 In this study, we found that EWS/Fli-1 knockdown by siRNA resulted in downregulation of c-Myc protein expression (Figure 2c). These results indicate that administration of siRNAs targeting EWS/Fli-1 may induce suppression of ES cell growth through the downregulation of c-Myc.

The tumor growth-inhibitory effect of synthetic siRNAs resulted from specific silencing of EWS/Fli-1. A decrease of EWS/Fli-1 mRNA expression was indeed observed by real-time quantitative RT-PCR of RNA extracted from siRNA-treated tumors. In this study, we also demonstrated that sequence-specific downregulation of EWS/Fli-1 impaired cell proliferation but did not induce apoptosis. It is noteworthy that the chemically synthesized siRNAs used in this study targeted the breakpoint of EWS/Fli-1 Type 1, and we found that these siRNAs reduced both the expression of EWS/Fli-1 protein and the proliferation of ES cells carrying EWS/Fli-1 Type 1 but not cells carrying Type 2. Moreover, our data showed that suppression of EWS/Fli-1 expression resulted in downregulation of the EWS/Fli-1-downstream targets, c-Myc, TERT and VEGF. Although off-target effects of siRNA cannot be completely ruled out, our current results indicate that EWS/Fli-1 inhibition is an attractive strategy for the treatment of ES, as this target sequence is present only in tumor cells.

In summary, we have shown that synthetic siRNAs targeting EWS/Fli-1 downregulate the expression of EWS/Fli-1 protein sequence specifically and also reduce the expression of c-Myc protein. We have also shown that inhibition of EWS/Fli-1 expression efficiently inhibits the proliferation and tumor growth of ES cells. Moreover, we have demonstrated that modification of siRNA with the aromatic compound pyridine at the 3′-end enhances the efficacy of treatment in vivo. These results suggest that specific downregulation of EWS/Fli-1 by synthetic siRNA is a possible approach for the treatment of ES. Simultaneous use of chemotherapy might also be effective. Further preclinical studies are warranted to explore the applicability of EWS/Fli-1 targeting for therapy of ES.

Acknowledgements

The authors thank Dr. T.J. Triche (University of Southern California, Los Angeles, CA) for providing the Ewing sarcoma cells and Takatoshi Yamamoto, Ayako Taguchi and Remi Nakashima for useful discussion and assistance.

Ancillary