Q11 Therapeutic Discovery

2 IGF-1R/MDM2 Relationship Confers Enhanced Sensitivity3 to RITA in Ewing Sarcoma Cells45 Giusy Di Conza1,4, Marianna Buttarelli1,2, Olimpia Monti1, Marsha Pellegrino1,3, Francesca Mancini1,3,6 Alfredo Pontecorvi3, Katia Scotlandi5, and Fabiola Moretti1

7 Abstract8 Ewing sarcoma is one of themost frequent bone cancers in adolescence. Althoughmultidisciplinary therapy

9 has improved the survival rate for localized tumors, a critical step is the development of new drugs to improve

10 the long-term outcome of recurrent and metastatic disease and to reduce side effects of conventional therapy.

11 Here, we show that the small molecule RITA (Reactivation of p53 and Induction of Tumor cell Apoptosis,

12 NSC652287) is highly effective in reducing growth and tumorigenic potential of Ewing sarcoma cell lines.

13 These effects occur both in the presence of wt-p53 as well as of mutant or truncated forms of p53, or in its

14 absence, suggesting the presence of additional targets in this tumor histotype. Further experiments provided

15 evidence that RITA modulates an important oncogenic mark of these cell lines, insulin-like growth factor

16 receptor 1 (IGF-1R). Particularly, RITA causes downregulation of IGF-1R protein levels. MDM2 degradative

17 activity is involved in this phenomenon. Indeed, inhibition of MDM2 function by genetic or pharmacologic

18 approaches reduces RITA sensitivity of Ewing sarcoma cell lines. Overall, these data suggest that in the cell

19 context of Ewing sarcoma, RITAmay adopt additional mechanism of action besides targeting p53, expanding

20 its field of application. Noteworthy, these results envisage the promising utilization of RITA or its derivative as

21 a potential treatment for Ewing sarcomas. Mol Cancer Ther; 1–10. �2012 AACR.

222324

25 Introduction26 Ewing sarcoma is the second most common tumor of27 bone in children and young adults. Besides surgery, its28 treatment includes highly intensive chemo- and/or radi-29 ation therapies, which are associated with significant30 short- and long-term side effects, of particular relevance31 in consideration of the young age of patients (1, 2). The32 potential application of newly developed strategies or the33 identification of newly targeted therapies is a field of34 intense study in the treatment of Ewing sarcoma.35 New anticancer therapies are mostly residing on the36 development of small chemical entities against specific37 oncogenic targets (3). Ewing sarcoma is molecularly char-38 acterized by the presence of a somatic translocation that39 involves chromosome 11, resulting in an oncogenic fusion

41protein that encompasses the EWS gene (the more abun-42dant translocation gives rise to EWS-FLI-1 chimera ref. 4).43Therefore, approaches have beendeveloped to inhibit this44oncogene, achieved mainly through the inhibition of its45transcript product and have shown their efficacy to limit46cell growth and tumorigenicity (5–8). Strictly related to47EWS-FLI-1, a crucial oncogenic pathway in Ewing sarco-48ma is that one mediated by Insulin-like growth factor49receptor 1 (IGF-1R; refs. 9–12). Indeed, activation of this50receptor is necessary for EWS-FLI-1–induced transforma-51tion of human cells (9). As confirmation of the relevance of52IGF-1R pathway, different therapeutic strategies are53showing their efficacy in preclinical models of the disease54and inEwing sarcomapatients in I or II phase clinical trials55(2, 12).At the same time, increasing evidence suggests that56combination of IGF-1R inhibitory molecules with other57treatments may further improve the efficacy of the first58(13).59Reactivationof theoncosuppressorp53 function is agoal60of different therapeutic approaches. These approaches61can be grossly divided in 2 groups: (i) one aiming to62correct the activity of mutant p53 in those tumors63bearing a mutated gene; (ii) a second one aiming to64reactivate p53 function in those tumors in which it is65still preserved (14). TP53 gene is rarely altered in Ewing66sarcoma (12) and therefore can be a potential target of67the second type of approach. Among strategies for p5368reactivation, a promising molecule is the furanic com-69pound RITA (Reactivation of p53 and Induction of70Tumor cell Apoptosis, also NSC652287; refs. 15, 16). At

AU

Authors' Affiliations: 1Cell Biology and Neurobiology Institute-CNR/Fon-dazione Santa Lucia; 2Department of Surgery, Breast Unit; 3Division ofEndocrinology, Catholic University of Rome, Rome; 4Department of Exper-imental-Clinical Medicine and Pharmacology, University of Messina, Mes-sina; and 5CRSDevelopment of Biomolecular Therapies, Lab ExperimentalOncology, Orthopaedic Rizzoli Institute, Bologna, Italy

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Fabiola Moretti, Cell Biology and NeurobiologyInstitute-CNR/Fondazione Santa Lucia, Via del Fosso di Fiorano 64, 00143.Phone: 39-06-501-703-242; Fax: 39-06-501-703-313; E-mail:mailto:fabiola.moretti@cnr.it

doi: 10.1158/1535-7163.MCT-11-0913

�2012 American Association for Cancer Research.

MolecularCancer

Therapeutics

www.aacrjournals.org 1

73 its discovery, RITA has been identified as a molecule74 able to bind and activate p53 by dissociating it from its75 inhibitor MDM2 (16). Later on, other studies have sug-76 gested the possibility of RITA to bind additional targets,77 as MDM2 itself and mutant p53 (17–19).78 In this work, we investigated the sensitivity to RITA of79 different cell lines derived from Ewing sarcoma and the80 biologic response caused by this molecule. To define the81 role of p53 in RITA sensitivity, we used cell lines charac-82 terized by a different status of p53. Our data show that83 RITA is highly effective in suppressing growth and84 tumorigenic properties of Ewing sarcoma cell lines inde-85 pendently of the status of p53. At themolecular levels, we86 observed a decrease of IGF-1R protein, at least in part,87 mediated by MDM2 degradative activity. Overall, these88 data indicate the ability of RITA to affect additional89 molecules other than p53 and its potential application in90 Ewing sarcoma.

91 Materials and Methods92 Cell cultures, transfections, and treatments93 All ES cell lines were obtained by Orthopaedic Rizzoli94 Institute Bologna, Italy in 2006. Previously, SK-N-MC95 cell line was obtained from the American Type Culture96 Collection, TC-71 and 6647 cell lines were a generous97 gift from TJ. Triche (Childrens Hospital, Los Angeles,98 CA), WE-68 cell line was established and kindly pro-99 vided by F. van ValenQ2 (20), LAP-35 and IOR-BRZ71 cell100 lines were established in Orthopaedic Rizzoli Institute101 (21). In all cell lines, but WE-68, the presence of EWS-102 FLI1 has been verified by Western blot. All cell lines103 have been tested for mycoplasma contamination and104 verified to be Mycoplasma free (last control April 2011).105 SK-N-MC, TC-71, and WE-68 have been further authen-106 ticated by DNA fingerprinting in 2008, using genes107 MPX-2 and MPX-3 kits (22). The status of TP53 gene108 in all cell lines, but SK-N-MC, was confirmed by109 sequencing in 2008. Genetic analysis of the TP53 gene110 was carried out by PCR amplification of exons 1–12 with111 flanking intronic primers, followed by sequencing.112 Genomic DNA was extracted from cell lines according113 to standard procedures. Cells were cultured in Iscove’s114 modified Dulbecco’s medium (IMDM; Gibco), supple-115 mented with 100 units/mL penicillin, 100 mg/mL strep-116 tomycin, and 10% FBS (Invitrogen) and maintained at117 37�C in a humidified 5% CO2 atmosphere.118 SK-N-MC cells were stably transfected with pCMV-119 MDM2, pCMV-MDM2C438L, or pCDNA3.1BusingLipo-120 fectamine Plus Reagent (Invitrogen) and were then main-121 tained in medium supplemented with G418 700 mg/mL.122 Cellswere treatedwithRITA1or 3mmol/L,a-amanitin 10123 mg/mL,MG132 10 or 15 mmol/L andwith sempervirin 0.5124 mg/mL.

125 Cell viability and TUNEL assay126 Cell viability was assessed by collecting both floating127 and adherent cells counted in a hemocytometer after the

129addition of trypan blue. Alternatively, cell viability was130determined using Cell Titer Blue colorimetric assay131according to manufacturer’s instruction (Promega).132Growth curves were previously set up with increasing133cell number (4,000, 6,000, and 8,000 cells per well) for134each cell line.135Apoptotic DNA fragmentation was carried out by ter-136minal deoxynucleotidyl transferase–mediated dUTP nick137end labeling (TUNEL) assays with fluorescent In Situ Cell138Death Detection kit (Roche) following the manufacturer’s139instructions. Alternatively, apoptotic cells were deter-140mined using ApoAlert Annexin V kit (Clontech) accord-141ing to manufacturer’s instruction.

142Immunoprecipitation and Western blot143For immunoprecipitation experiments, cells were lysed144in Saito’s modified buffer (50 mmol/L Tris–HCl, pH 7.4,1450.15 mol/L NaCl, 0.5% Triton-X100, and 5 mmol/L146EDTA) contained a cocktail of protease inhibitors (Boeh-147ringer). Immunoprecipitations were done by preincuba-148tion lysates with protein G-sepharose (Pierce) and then149with the indicated antibody, under gentle rocking at 4�C150overnight. For Western blot, cells were lysed in radio-151immunoprecipitation assay buffer (50 mmol/L Tris–Cl,152pH 7.5, 150 mmol/L NaCl, 1% Nonidet P-40, 0.5% Na153desoxicholate, 0.1% SDS, and 1 mmol/L EDTA) supple-154mentedwith a cocktail of protease inhibitors (Boehringer).155The following primary antibody were used: rabbit156anti-IGF-1R polyclonal antibody C-20 (Santa Cruz), goat157anti-p53 polyclonal antibody FL393 (Santa Cruz), mouse158anti-MDM2 monoclonal antibody (mAb) 2A10, and Ab1159(Calbiochem),mouse anti-a-tubulinmAbDM1A (Sigma),160rabbit anti-FLI-1 polyclonal antibody C-19 (Santa Cruz),161mouse anti-PARP1 mAb C2–10 (Pharmingen), mouse162anti-Hsp70 mAb SPA-820 (StressGen), mouse anti-GFP163mAb7.1 and 13.1 (Roche), andmouse anti-actinmAbC-40164(Sigma).

165Cell-cycle analysis166Cell-cycle profileswere evaluated by fixing 5� 105 cells167in 70% ethanol for 1 hour on ice and staining DNA for 30168minutes at room temperature with 50 mg/mL propidium169iodide in PBS containing 1 mg/mL RNase A. Cell per-170centages in the different phases of the cycle were mea-171sured by flow cytometric analysis of propidium iodide–172stained nuclei using Multicycle Software (Phoenix Flow173System) Epics XL (Coulter).

174Soft-agar foci formation assay175The anchorage-independent growth of LAP-35, TC-17671, and SK-N-MC cell lines was monitored by the soft-177agar colony formation assay. In 60-mm dishes 5 mL of178hard agar layer, 0.7% agar-noble (Difco), reconstituted179in serum-containing medium, were used as substrate180for the Ewing sarcomas cell lines that were plated on a1812.0-mL layer of 0.35% warm agar in serum-containing182medium. Cells were then cultured at 37�C, 5% CO2 for 3183weeks and RITA treatment was renewed every 3 days.

Di Conza et al.

Mol Cancer Ther; 2012 Molecular Cancer Therapeutics2

186 SiRNA187 MDM2 siRNA and control siRNA were generated by188 Invitrogen (Stealth RNAi) and consisted of a mix of 3189 different siRNA. Cells were transfected using RNAiMAX190 reagent (Invitrogen) following the manufacturer’s191 instructions.

192 Real-time PCR193 Total RNA extraction was carried out with TRIzol194 reagent (Invitrogen). For quantitative reverse transcrip-195 tion PCR (qRT-PCR), total RNA (1 mg) was reverse tran-196 scribed using Gene Amp kit (Applied Biosystems) and197 subjected to PCR amplification using SYBR Green PCR198 Master Mix (Applied Biosystems). The following primers199 were used:

200 upper IGF-1R 50-CCATTCTCATGCCTTGGTCT-30;201 lower IGF-1R 50-TGCAAGTTCTGGTTGTCGAG-30;202 upper IGF-1 50-GGTGGATGCTCTTCAGTTCGT-30;203 lower IGF-1 50-CCACGATGCCTGTCTGAGG-30;204 upper IGF-BP3 50-TGTGGCCATGACTGAGGAAA-30;205 lower IGf-BP3 50-TGCCAGACCTTCTTGGGTTT-30;206 upper hTBP 50-GAACATCATGGATCAGAACAACA-30;207 lower hTBP 50-ATAGGGATTCCGGGAGTCAT-30.208209 Each tissue sample mRNA was normalized relative to210 the glyceraldehyde-3-phosphate dehydrogenase mRNA.211 Samples underwent 40 amplification cycles,monitored by212 an ABI Prism 7900 sequence detector (Applied Biosys-213 tems). All amplification reactions were conducted at least214 in duplicate and averages of threshold cycleswere used to215 interpolate standard curves and calculate transcript216 amount using the software SDS version 2.3 (Applied217 Biosystems).

218 Results219 RITA inhibits growth and tumorigenic potential of220 Ewing sarcoma cell lines independently of p53 status221 To ascertain RITA efficacy in Ewing sarcoma, we used222 cell lines characterized by the same chromosomal trans-223 location, t(11;22)(q24;q12), that generates the oncogenic224 proteinEWS/FLI-1. Toverify the relevance of p53 inRITA225 sensitivity, we used cell lines characterized by a different226 p53 status: LAP-35 andWE-68 with wild-type p53, TC-71227 and IOR-BRZ71 with a truncated p53 lacking the tetra-228 merization and DNA-binding domains (p53-213STOP),229 6647 with a point mutant p53 (p53-S241F) and SK-N-MC230 lacking p53 expression (23). The status of TP53 gene in all231 cell lines, but SK-N-MC, was confirmed by sequencing232 (data not shown). At first, we analyzed cell growth of all233 cell lines in the presence or absence of RITA. Because234 analysis of NCI-60 cell panel has evidenced a mean IC50 >235 3.5 mmol/L for RITA in mutant p53 carrying cancer cell236 lines (24),we treatedEwing sarcoma cell lineswith 1 and 3237 mmol/L RITA. Unexpectedly, analysis of cell viability238 evidenced the ability of RITA to reduce this parameter

240in all tested cell lines, independently of the presence and241of the status of p53 (Fig. 1A–F). In wt-p53 cell lines, cell242viability was strongly reduced in WE68 cells, whereas in243LAP-35 it was similar to that observed in mutant p53 cell244lines or even lower in comparison with null p53 SK-N-245MC. Cell growth reduction was enhanced by increased246doses (1 or 3 mmol/L) and treatment times of RITA,247although with a different kinetics among cell lines. To248confirm these data and ascertain the specificity of the249observed effects, we measured viability by analyzing cell250metabolic activity (by reduction of resazurin) and com-251pared the dose of RITA able to reduce this parameter to25250% inEwing cell lines and inH1299, a lung tumor cell line253lacking p53. Although not overlapping with the results254obtained by Trypan blue uptake, RITA confirmed its255efficacy in all Ewing cells in comparison with H1299256cells that resulted unaffected by the used doses of RITA257(Fig. 1G). Particularly, the p53-null SK-N-MC confirmed258the highest sensitivity whereas wt-p53 LAP35 showed a259sensitivity similar to that of truncated-p53 IOR-BRZ71.260HCT116�/�, a colon cancer cell line lacking p53 too, and261insensitive to the drug (16), showed results similar to262those observed in H1299 (data not shown). These data263overall confirmed the specific sensitivity of Ewing sarco-264ma cell lines to RITA.265Fluorescence-activated cell sorting (FACS) analysis evi-266denced a strong alteration of cell cycle caused by RITA as267early as 16 hours after drug treatment in 3 representative268cell lines (Fig. 2A–C) and in the other cell lines (Supple-269mentary Fig. S1A–C). In all cell lines it was evident that a270progressive accumulation of cells in the S/G2 phases of271cell cycle, in agreement to what previously reported272(15, 18); in TC-71 and especially in SK-N-MC, it was also273evident a sub-G1 population indicative of cell death (Fig.2742A–C). LAP-35, although containing wt-p53, showed a275modest increase of sub-G1 at the latest time point (36276hours). To further characterize the loss of viability evi-277denced by FACS analysis, a TUNEL assay was done. The278experiments confirmed a strong increase of apoptotic cells279in TC-71 and SK-N-MC cells upon RITA 3 mmol/L treat-280ment, whereas only amild increase was observed in LAP-28135 in agreement to FACS data (Fig. 2D).282Overall, these data indicated that RITA is effective in283suppressing cell growth of Ewing sarcoma cell lines and284that its effects are mediated by a strong growth suppres-285sive and/or apoptotic activity. Noteworthy, these effects286seem to be not related to the status of p53.287RITA exerts a strong antitumor activity in tumor cell288lines (16, 25). To verify whether RITA is able to affect289tumorigenic properties of Ewing sarcoma cells too, we290analyzed their anchorage-independent growth by a soft-291agar assay using RITA 1 and 3 mmol/L. Different amounts292of cells were plated (20,000 or 100,000 cells/60 mm dish)293and after 48 hours treated with vehicle (CTR) or RITA,294repeating the administration every 3 days. After 3 weeks,295all CTR cells have formed marked colonies, more evident296in TC-71 and SK-N-MC, less evident in LAP-35 in agree-297ment with the reduced tumorigenicity of this cell line

Ewing Sarcoma Sensitivity to RITA

www.aacrjournals.org Mol Cancer Ther; 2012 3

300 (ref. 26; Fig. 3A, CTR row). Strikingly, both doses of RITA301 were able to completely abolish anchorage-independent302 growth independently of the number of seeded cells (Fig.303 3A and data not shown), indicating that RITA is highly304 effective in reducing in vitro tumorigenic potential of these305 cells. Again these effects were irrespective of p53 status,306 suggesting that RITA may affect additional pathways in307 this tumor histotype.

308 RITA downregulates IGF-1R levels309 Anchorage-independent growth requires a combina-310 tion of proliferation and survival signaling strongly affect-311 edby thepresence of a functional IGF-1Rpathway (13, 27).312 To investigate RITA activity in Ewing sarcoma cells, the

314levels of IGF-1Rwere analyzed in 2 cell lines characterized315by a null and a truncated p53, SK-NMC, and TC-71,316respectively. RITA treatment caused a progressive down-317regulation of IGF-1R levels in both cell lines (Fig. 4A and318B). IGF-1R decrease appeared earlier and more evident in319cells that undergo apoptotic response evidenced also by320the progressive appearance of PARP cleavage product321(Fig. 4A and B). Under the same conditions, MDM2 levels322decreased at the later time points as previously reported323(Fig. 4A and B; refs. 28, 29). Similar results were observed324also in the other cell lines (Supplementary Figs. S2 andS3).325P53, wt as well as mutant S241F forms, were not substan-326tially altered or slightly increased in LAP35 and 6647327(Supplementary Fig. S3). Conversely, a strong decrease

Figure 1. RITA effects on cell Q3

growth. RITA reduces cell viabilityin A, LAP-35, B, WE-68, C, TC-71,D, IOR-BRZ71, E, 6647, and F, SK-N-MC.Cellswere treatedwithRITAat the indicated concentrations for24 or 48 hours and then stainedwith Trypan blue for immediatecount analysis. Cell growth wascalculated as number of Trypanblue negative cells. Percentage ofcell viability was obtained bysetting 100% the cell growth ofcontrol vehicle–treated cells (CTR)at each time point and calculatingthe relative cell growth of treatedcells. Data are representative of atleast 3 different experiments. G,cells were plated in 96-well plateand treated in octuplicate withincreasing concentration of RITA(0.5-1-2-3-6 mmol/L) for 24 or 48hours (�). Cell viability wasassessed by Cell Titer blue. Thedose reported in the graph is thedose able to reduce cell viability to50% in comparison with untreatedcells.

Di Conza et al.

Mol Cancer Ther; 2012 Molecular Cancer Therapeutics4

330 of wt-p53 coexistent with IGF-1R was observed in WE68331 (Supplementary Fig. S2), suggesting a specific suscepti-332 bility of this cell line to RITA.333 IGF-1R has been previously described as a transcrip-334 tional target of RITA-mediated p53 transcriptional activ-335 ity (30, 31). Although, in the cell lines here used, but LAP-336 35 and WE68, p53 is transcriptionally inactive or absent,337 we testedwhether IGF-1Rdownregulation is correlated to338 a decrease of its mRNA. Indeed, the levels of IGF-1R339 mRNA were not significantly altered in both cell lines340 (Fig. 4C and D). We therefore tested whether the down-341 regulation of IGF-1R was mediated by increased protein342 turnover. IGF-1R degradation is mediated by proteasome343 and lysosome (32). To analyze the involvement of protea-

345some in RITA-mediated IGF-1R downregulation, TC-71346and SK-NM-C cell lines were treatedwith the proteasome347inhibitor, MG132, or RITA or both drugs. MG132 treat-348ment per se increased the levels of IGF-1R, confirming the349proteasome-mediateddegradation of the receptor in these350cell lines.Noteworthy, addition ofMG132 toRITA-treated351cellswas able to recover the decrease of IGF-1R levels (Fig.3524E and F), indicating that RITA-induced downregulation353was indeed caused by degradation of the receptor.354It has been reported that IGF-1R levels are downregu-355lated by IGF binding protein 3 (IGF-3BP; ref. 33), a protein356acting as inhibitor of IGF-1R activity by sequestering its357ligand IGF-1. IGF-BP3 is a transcriptional target of EWS-358FLI-1, the oncogenic mark of Ewing sarcoma (34).

Figure 2. RITA causes G2 arrest and cell apoptosis. FACS analysis of A, LAP-35, B, TC-71, and C, SK-N-MC carried out at 12, 24, or 30 hours after treatmentwith 3 mmol/L RITA. Cells were grown in the presence or absence of RITA and then stained with PI for the immediate cell-cycle analysis. Each panel reportsthe percentage of cells in each cell-cycle phase. Sub-G1 region (as indicated) has been independently evaluated relative to the total population. Theseanalyses are representative of at least 2 different experiments. D, TUNEL analysis of indicated cells at 16 or 24 hours after treatment with 3 mmol/L RITA. Thepercentage of apoptotic cells was measured as TUNEL-positive cells relative to total cells. The data report the mean of 3 independent counts.

Ewing Sarcoma Sensitivity to RITA

www.aacrjournals.org Mol Cancer Ther; 2012 5

361 Particularly, IGF-BP3 is repressed by EWS-FLI-1. Because362 it has been shown that RITA downregulates various key363 oncogenic pathways (30), we investigated whether IGF-364 1R decrease was related to modification of EWS-FLI-1

366levels and/or activity. Nor substantial modification of367EWS-FLI-1 protein levels (Supplementary Fig. S4A–C) or368of its transcriptional targets IGF-1 and IGF-BP3 mRNA369(Supplementary Fig. S4D)was observed uponRITA treat-370ment in SK-N-MC and TC-71, suggesting that downre-371gulation of IGF-1R by RITA is not mediated by alteration372of EWS-FLI-1. Conversely, in LAP-35 RITA caused a373decrease of EWS-FLI1 (Supplementary Fig. S4C), support-374ing the p53-mediated downregulation of this oncogenic375pathway too (30).

376MDM2mediates RITA-induced IGF-1R degradation377Degradation of IGF-1R ismediated by 2RINGubiquitin378ligases, c-Cbl andMDM2 (35, 36).Althoughp53 is the only379known target of RITA, MDM2 has been reported as an380importantmediator of RITA activity (28, 37).We therefore381investigated whether MDM2 may be involved in IGF-1R382degradation. We interfered MDM2 expression by siRNA383and analyzed IGF-1R levels of siMDM2 interfered cells in384comparison with control cells interfered with scramble385siRNA (siCTR), both in the absence and in the presence of386RITA. Interference of MDM2 expression was indeed able387to counteract RITA-mediated IGF-1R downregulation388(Fig. 5A), supporting the potential involvement ofMDM2389in RITA-mediated IGF-1R degradation. To further ascer-390tain whether MDM2 activity is involved in RITA

Figure 4. RITA downregulatesIGF-1R protein levels. A, SK-N-MCand B, TC-71 cells were incubatedwith RITA 1 or 3 mmol/L for 16 or24 hours.Whole cells extractswereprepared and analyzed byWesternblot for the indicated proteins.Actin was used as loading control.C and D, qReal-time PCR ofIGF-1R in SK-N-MC and TC-71cells, respectively, treated as in Aand B. E, SKNMC were incubatedwith 3 mmol/L RITA and after14 hours 10 mmol/L MG132 wasadded. After additional 4 hours,cells were collected for Westernblot analysis. F, TC71 wereincubated with 3 mmol/L RITA andafter 20 hours, 15 mmol/L MG132was added. After additional4 hours, cells were collected forWestern blot analysis.

Figure 3. RITA inhibits anchorage-independent growth. 100,000 cells per60-mm dish were plated in soft agar and 48 hours after plating, RITA wasadded at the indicated concentrations. Photos here reported arerepresentative of 2 independent experiments conducted in duplicate.

Di Conza et al.

Mol Cancer Ther; 2012 Molecular Cancer Therapeutics6

393 sensitivity, cell viability was analyzed by Annexin stain-394 ing. Indeed, cell apoptosis induced by RITAwas reduced395 in siMDM2 in comparison with siCTR cells, paralleling396 IGF-1R recovery. These effects were more pronounced in397 SK-N-MC cells in agreement with the increased apoptosis398 observed in these cells.399 To ascertain whether MDM2-induced degradation of400 IGF-1R is mediated by transcriptional activation of other401 cofactors by RITA, we induced transcriptional block by402 pretreating cells with the inhibitor of RNA polymerase II,403 a-amanitin (38, 39). In the absence of RITA, a-amanitin404 does not affect IGF-1R levels, in agreement with the long405 half-life of the receptor (24 hours, as reported in 40).406 Importantly, the presence of a-amanitin does not alter407 RITA-induced decrease of IGF-1R levels (Fig. 5C and D),408 suggesting that transcription of other factors is not409 required for RITA-mediated IGF-1R downregulation.410 Functionality of a-amanitin was evidenced by the down-411 regulation of Pol II levels, as previously reported (39).412 MDM2-mediated degradation of IGF-1R ismediated by413 direct association between the 2 proteins (35). To support414 the activity of MDM2 toward IGF-1R, we analyzed the415 binding between IGF-1R and MDM2 in the presence or416 absence of RITA after few hours of RITA treatment. In the417 absence of RITA, association between the 2 proteins was418 barely detectable. Conversely, after 4 hours of RITA treat-419 ment, an evident increase in the association between the 2

421proteins was observed in both SK-N-MC and TC-71 (Fig.4225E and F). This early association is in agreement with the423observation that MDM2-mediated ubiquitination of IGF-4241R precedes subsequent internalization and degradation425of the receptor (36) and supports the hypothesis that RITA426treatment increases the association between MDM2 and427IGF-1R, leading to IGF-1R degradation.

428MDM2-degradation of IGF-1R mediates RITA429sensitivity430To confirm the role of MDM2 activity and of MDM2/431IGF-1R relationship in RITA sensitivity, we interfered432with MDM2 ubiquitination function. It has been recently433described the ability of sempervirine, a plant alkaloidwith434well-known anticancer activities (41), to inhibit ubiquiti-435nation and degradative activity of MDM2 (42). We there-436fore impaired MDM2 function by sempervirine and ana-437lyzed RITA effects. A dose of 0.5 mg/mL of sempervirine438showed its efficacy to inhibitMDM2degradative function439both in TC-71 and in SK-N-MC cells, after 8 or 4 hours of440treatment, respectively (data not shown). Simultaneous441administration of RITA and sempervirine was able to442antagonize downregulation of IGF-1R levels in both443TC-71 and SK-N-MC (Fig. 6A and C). Most importantly,444these effects correlate with an impairment of cell death445induced by RITA (Fig. 6B andD), confirming thatMDM2-446mediated IGF-1R degradation is an important step of

Figure 5. MDM2 is involved in RITA-mediated IGF-1R degradation. A,SK-N-MC and TC-71 cells weretransiently transfected with siMDM2orCTRoligos (siCTR). After 10 hours,interfered cells were treated with 3mmol/L RITA and after additional 24hours collected. Whole cells extractswere prepared and analyzed byWestern blot for the indicatedproteins. B, cell death was assessedby FACS analysis of Annexin-positive cells (top). Bottom panelreports theWestern blot analysis of afraction of cells subjected to FACSanalysis. C, SK-N-MC and D, TC-71cells were treated with 3 mmol/L RITAfor 24 hours or with 10 mg/mLaamanitin for 27 hours or both.Whole cells extracts were preparedand analyzed by Western blot for theindicated proteins. E, SK-N-MC andF, TC-71 cells were treated with RITAat the indicated concentration ortreated with vehicle (CTR) andafter 4 or 8 hours whole cellsextracts collected. MDM2coimmunocomplexes were analyzedafter immunoprecipitation of 800 mgof WCE with a-MDM2 2A10-Ab1antibodies. WCEs indicatesanalysis of 1/25 of whole-cellextracts.

Ewing Sarcoma Sensitivity to RITA

www.aacrjournals.org Mol Cancer Ther; 2012 7

449 RITA activity and supporting the role of MDM2/IGF-1R450 relationship in RITA sensitivity of these tumor cells.451 To confirm the role of MDM2 and of its ubiquitinating452 activity in RITA sensitivity, we stably overexpressed453 wt-MDM2 and a mutant MDM2, MDM2-C438L, in SK-454 N-MC. This mutation disrupts the RING region where455 ubiquitin ligase activity of MDM2 resides and therefore456 abolishes it (43). Upon 15 hours of RITA treatment,457 the overexpression of wt-MDM2 accelerated IGF-1R458 downregulation, whereas this was poorly observed in459 the pcDNA3.1 transfected- (CTR) and the MDM2-460 C438L-expressing cell populations (Fig. 6E).Of note, these461 molecular data correlate with altered survival of SK-N-462 MC. Indeed, the expression of wt-MDM2 potentiates the

464effects of RITA, whereas the mutant MDM2-C438L does465not exhibit such activity and rather slightly reduced cell466sensitivity to the small molecule (Fig. 6F), supporting the467role of MDM2 degradative activity in RITA sensitivity of468Ewing sarcoma cells.

469Discussion470The aim of this study was to evaluate the efficacy of the471furanic compound RITA in Ewing sarcoma. Indeed, the472requirement of new therapeutic tools for this type of473tumor is a field of intense study (2).474The canonical target of RITA is wt-p53, although addi-475tional molecules have been hypothesized (16–18). Here,476wecompared the effects ofRITAamongdifferent cell lines477characterized bywt, mutant, truncated, and null p53. The478first result was that RITA strongly affects the growth and479tumorigenic properties of these cell lines independently of480p53, suggesting a specific sensitivity of this tumor type to481RITA. Particularly, RITA reduced to a similar extent cell482viability of wt-p53 LAP-35, null p53 SK-N-MC, and IOR-483BRZ71 cells expressing a truncated form p53-213STOP484that resembles a null condition. In addition, a significant485cell deathwas observed inwt-p53WE-68 cells aswell as in486p53-213STOP TC-71 and null-p53 SK-N-MC cells. Con-487versely, a reduced cell death was evident in the wt-p53488LAP-35 cell line. Although, this may be attributed to the489reduced tumorigenicity and growth rate of this cell line,490the presence of additional factors affecting RITA sensi-491tivity could contribute as well (28).492RITA has been identified through a cell-based drug493screening strategy (16) based on its biologic effects on494target function, specifically p53. The possibility that it can495affect additional pathways cannot be excluded. Indeed,496RITA activates a DNA damage response (44) and affects497wt p53-related off-targets (28). In addition, cell lines car-498rying mutant p53 have been reported to be sensitive to499RITA, although at higher doses in comparison with cell500lines carrying wt-p53 (24, 45).501In fact, accumulation of cells in the S/G2 phases of cell502cycle, as observed in Ewing sarcoma cell lines, has been503previously reported both inwild-type aswell as inmutant504and null p53-carrying cancer cell lines (16, 18), suggesting505that RITA-mediated alteration of the cell cycle may506depend on additional factors besides p53.507The hallmarks of Ewing sarcoma are the presence of508oncogenic protein EWS-FLI-1 and the autocrine-activa-509tion of IGF-1R. We did not observe substantial modi-510fication of EWS-FLI-1 levels. Conversely, we observed a511consistent decrease of IGF-1R levelsuponRITA treatment.512IGF-1R is a crucial pathway for Ewing sarcoma survival.513Molecular and translational studies have shown that514inhibition of this pathway suppresses tumor growth515(reviewed in refs. 12) and in fact clinical trials with IGF-1R516inhibitors are under evaluation (2). The data here pre-517sented indicate that IGF-1R may be indirectly affected by518a targeted therapy too, opening a new potential field of519anti–IGF-1R therapies.

Figure 6. RITA sensitivity is affected by MDM2 function. A, Western blotanalysis of SKNMC cells treated with vehicle, 3 mmol/L RITA for 22 hoursor 3 mmol/L RITA plus 0.5 mg/mL sempervirine added 4 hours before celllysate collection; B, percentage of cell death after treatment of indicatedcells as in A;C,Western blot analysis of TC-71 cells treatedwith vehicle, 3mmol/L RITA for 24 hours or 3 mmol/L RITA plus 0.5 mg/mL sempervirineadded 8 hours before cell lysate collection; D, percentage of cell deathafter treatment of indicated cells as inC. Thepercentageof cell deathwascalculated as number of dead/total cell number. Columns, average of 3independent experiments. Bars, SD; E, Western blot analysis of SKNMCstably transfected with pcDNA 3.1 (CTR), wt-MDM2, or MDM2C438L.Cells were treated with 3 mmol/L RITA and after 15 hours collected forWestern blot analysis of indicated proteins. F, cell death of previouslyindicated SKNMC populations treated as in E. Cell death was calculatedas fold of increase of cell death of RITA treated cells/untreated cells.Columns, average of 3 independent experiments. Bars, SD

Di Conza et al.

Mol Cancer Ther; 2012 Molecular Cancer Therapeutics8

522 Our experiments indicate that downregulation of IGF-523 1R protein levels is mediated, at least in part, by the E3524 ubiquitin ligase, MDM2. Various studies have shown the525 ability ofMDM2 to control IGF-1R levels (35, 46), although526 it is not completely clear what activates and controls this527 MDM2 function.Moreover, it has been shown thatMDM2528 counteracts the IGF-1R survival function, independently529 of p53 (47). Our data indicate that in Ewing sarcoma cell530 lines, RITA activates MDM2 function. How this happens,531 it is not clear. Cell fractionation experiments did not show532 any difference in MDM2 localization in the presence or533 absence of RITA, excluding a delocalization of MDM2 by534 RITA treatment (data not shown). A study of docking535 simulation indicated that RITA can bind MDM2 (17),536 although with an affinity 415,000 less than p53. The537 possibility that MDM2 modifications alter RITA affinity538 to MDM2 has not been tested. However, the selective539 sensitivity of Ewing sarcoma to RITA suggests that addi-540 tional specific factors mediate and direct MDM2 activity541 toward IGF-1R in this tumor type. Interestingly, a recent542 study reported the overexpression in Ewing sarcoma of a543 protein involved in ubiquitin ligase activity, CDT2, rais-544 ing the hypothesis that deregulation of protein ubiquitin545 machinery may be of relevance in this tumor type (48).546 Finally, downregulation of MDM2 upon RITA treatment,547 as observed in this work and reported by other authors548 (28, 29), but yet its role in RITA function may seem to549 counteract its role in RITA sensitivity. However, the550 crucial role of MDM2-degradative activity in RITA sen-551 sitivity has been previously reported by other authors552 (28, 37), indicating thatMDM2 is important at least for the553 early response to RITA.554 It remains to be ascertained whether MDM2-mediated555 IGF-1R downregulation is the main determinant of RITA556 activity. Unfortunately, experiments of rescue of RITA557 effects by overexpression of IGF-1R failed because of the558 inability of 2 different IGF-1R coding plasmids to sub-559 stantially overexpress IGF-1R levels in the Ewing sarcoma560 cell lines here studied (data not shown). Of note,MDM2 is

562able to alter cell-cycle progression and different studies563have reported the ability of MDM2 to induce a G0/G1

564arrest in specific cell lines, independently of p53 (49).565Therefore, growth control activities of MDM2 too might566be involved in RITA activity in Ewing sarcoma. However,567the complete understanding of the mechanism by which568RITA is effective in this tumor type is a point of relevance569before any consideration of this drug as therapeutic agent570in Ewing sarcoma.571Overall, these data expand the spectrum of antitumor572RITA activity showing its efficacy in Ewing sarcoma.573Interestingly, p53 mutation, rare in Ewing sarcoma (12),574is increased in a subset characterized by high aggressive575behavior and poor chemoresponse (50), suggesting the576useful application of RITA in those cases escaping stan-577dard therapies. Given the low toxicity of this molecule in578normal cells (16), these results pave the way to the poten-579tial application of a new therapeutic tool to Ewing580sarcoma.

581Disclosure of Potential Conflicts of Interest Q4

582No potential conflicts of interest were disclosed.

583Acknowledgments584Q5The authors thank Dr.ssa Igea D’Agnano, Dr.ssa Francesca Siepi, and585Dr.ssa Manuela Natoli for their assistance in FACS analysis; Prof. R.586Baserga and Prof. A. Belfiore for providing IGF-1R expressing plasmids;587and Dr.ssa Ada Sacchi and Prof. F. Trimarchi for their invaluable support588and helpful discussion.

589Grant Support590This work was supported by a grant of Ministero della Salute (RF-IOR-5912006-422755) to K. Scotlandi, and F. Moretti, and Integrato Oncologia5922009/1536 to K. Scotlandi and of AIRC (Italian Association for Cancer593Research) to F. Moretti and K. Scotlandi. O. Monti is recipient of a FIRC594fellowship. G.Di Conza is recipient of a PhD fellowship fromUniversity of595Messina.596The costs of publication of this article were defrayed in part by the597payment of page charges. This article must therefore be hereby marked598advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate599this fact.

600Received November 9, 2011; revised February 13, 2012; accepted March6015, 2012; published OnlineFirst xx xx, xxxx.

602 References603 1. Paulussen M, Frohlich B, Jurgens H. Ewing tumour: incidence,604 prognosis and treatment options. Paediatr Drugs 2001;3:605 899–913.606 2. Subbiah V, Anderson P, Lazar AJ, Burdett E, Raymond K, Ludwig JA.607 Ewing's sarcoma: standard and experimental treatment options. Curr608 Treat Options Oncol 2009;10:126–40.609 3. MaWW, Adjei AA. Novel agents on the horizon for cancer therapy. CA610 Cancer J Clin 2009;59:111–37.611 4. Ordonez JL, Osuna D, Herrero D, de Alava E, Madoz-Gurpide J.612 Advances in Ewing's sarcoma research: where are we now and what613 lies ahead? Cancer Res 2009;69:7140–50.614 5. Kovar H, Aryee DN, Jug G, Henockl C, Schemper M, Delattre O, et al.615 EWS/FLI-1 antagonists inducegrowth inhibitionof Ewing tumor cells in616 vitro. Cell Growth Differ 1996;7:429–37.617 6. Tanaka K, Iwakuma T, Harimaya K, Sato H, Iwamoto Y. EWS-Fli1618 antisense oligodeoxynucleotide inhibits proliferation of human619 Ewing's sarcoma and primitive neuroectodermal tumor cells. J Clin620 Invest 1997;99:239–47.

6227. Lambert G, Bertrand JR, Fattal E, Subra F, Pinto-Alphandary H, Malvy623C, et al. EWS fli-1 antisense nanocapsules inhibits Ewing sarcoma-624related tumor in mice. Biochem Biophys Res Commun 2000;279:401–6256.6268. Erkizan HV, Uversky VN, Toretsky JA. Oncogenic partnerships: EWS-627FLI1 protein interactions initiate key pathways of Ewing's sarcoma.628Clin Cancer Res 2010;16:4077–83.6299. Toretsky JA, Kalebic T, Blakesley V, LeRoith D,Helman LJ. The insulin-630like growth factor-I receptor is required for EWS/FLI-1 transformation631of fibroblasts. J Biol Chem 1997;272:30822–7.63210. Yee D, Favoni RE, Lebovic GS, Lombana F, Powell DR, Reynolds CP,633et al. Insulin-like growth factor I expression by tumors of neuroecto-634dermal origin with the t(11;22) chromosomal translocation. A potential635autocrine growth factor. J Clin Invest 1990;86:1806–14.63611. Scotlandi K, Benini S, Sarti M, Serra M, Lollini PL, Maurici D, et al.637Insulin-like growth factor I receptor-mediated circuit in Ewing's sar-638coma/peripheral neuroectodermal tumor: a possible therapeutic tar-639get. Cancer Res 1996;56:4570–4.

Ewing Sarcoma Sensitivity to RITA

www.aacrjournals.org Mol Cancer Ther; 2012 9

642 12. Toomey EC, Schiffman JD, Lessnick SL. Recent advances in the643 molecular pathogenesis of Ewing's sarcoma. Oncogene 2010;29:644 4504–16.645 13. Chitnis MM, Yuen JS, Protheroe AS, Pollak M, Macaulay VM. The type646 1 insulin-like growth factor receptor pathway. Clin Cancer Res647 2008;14:6364–70.648 14. Cheok CF, Verma CS, Baselga J, Lane DP. Translating p53 into the649 clinic. Nat Rev Clin Oncol 2011;8:25–37.650 15. Nieves-Neira W, Rivera MI, Kohlhagen G, Hursey ML, Pourquier P,651 Sausville EA, et al. DNA protein cross-links produced by NSC 652287,652 a novel thiophene derivative active against human renal cancer cells.653 Mol Pharmacol 1999;56:478–84.654 16. Issaeva N, Bozko P, Enge M, Protopopova M, Verhoef LG, Masucci655 M, et al. Small molecule RITA binds to p53, blocks p53-HDM-2656 interaction and activates p53 function in tumors. Nat Med 2004;657 10:1321–8.658 17. Espinoza-Fonseca LM. Targeting MDM2 by the small molecule RITA:659 towards the development of new multi-target drugs against cancer.660 Theor Biol Med Model 2005;2:38.661 18. Zhao CY, Grinkevich VV, Nikulenkov F, Bao W, Selivanova G. Rescue662 of the apoptotic-inducing function of mutant p53 by small molecule663 RITA. Cell Cycle 2010;9:1847–55.664 19. Girardini JE, Del Sal G. Improving pharmacological rescue of p53665 function: RITA targets mutant p53. Cell Cycle 2010;9:1847–55.666 20. Van Valen F. Ewing's sarcoma family of tumors. In:Masters JRW,667 Palsson B, editors. Human cell culture vol I. The Netherlands: Kluwer668 Academic Publisher; 1999. p. 55–85.669 21. Bagnara GP, Serra M, Giovannini M, Badiali M, Stella M, Montaldi A,670 et al. Establishment and characterization of a primitive neuroectoder-671 mal tumor of bone continuous cell line (LAP-35). Int J Cell Cloning672 1990;8:409–24.673 22. Ottaviano L, Schaefer KL, Gajewski M, Huckenbeck W, Baldus S,674 Rogel U, et al. Molecular characterization of commonly used cell lines675 for bone tumor research: a trans-European EuroBoNet effort. Genes676 Chromosomes Cancer 2010;49:40–51.677 23. Davidoff AM, Pence JC, Shorter NA, Iglehart JD,Marks JR. Expression678 of p53 in human neuroblastoma- and neuroepithelioma-derived cell679 lines. Oncogene 1992;7:127–33.680 24. Liu W, He L, Ramirez J, Ratain MJ. Interactions between MDM2 and681 TP53 genetic alterations, and their impact on response to MDM2682 inhibitors and other chemotherapeutic drugs in cancer cells. Clin683 Cancer Res 2009;15:7602–7.684 25. Rivera MI, Stinson SF, Vistica DT, Jorden JL, Kenney S, Sausville EA.685 Selective toxicity of the tricyclic thiophene NSC 652287 in renal686 carcinoma cell lines: differential accumulation and metabolism. Bio-687 chem Pharmacol 1999;57:1283–95.688 26. Scotlandi K, Benini S, Manara MC, Serra M, Nanni P, Lollini PL, et al.689 Murine model for skeletal metastases of Ewing's sarcoma. J Orthop690 Res 2000;18:959–66.691 27. Larsson O, Girnita A, Girnita L. Role of insulin-like growth factor 1692 receptor signalling in cancer. Br J Cancer 2005;92:2097–101.693 28. Rinaldo C, Prodosmo A, Siepi F, Moncada A, Sacchi A, Selivanova694 G, et al. HIPK2 regulation by MDM2 determines tumor cell response695 to the p53-reactivating drugs nutlin-3 and RITA. Cancer Res696 2009;69:6241–8.697 29. Spinnler C, Hedstr€om E, Li H, de Lange J, Teunisse AF, Verlaan-de-698 Vries M, et al. Abrogation of Wip1 expression by RITA-activated p53699 potentiates apoptosis induction via activation of ATM and inhibition of700 Hdmx. Cell Death Differ 2011;18:1736–45.701 30. Grinkevich VV, Nikulenkov F, Shi Y, Enge M, Bao W, Maljukova A,702 et al. Ablation of key oncogenic pathways by RITA-reactivated703 p53 is required for efficient apoptosis. Cancer Cell 2009;15:704 441–53.705 31. Saha MN, Jiang H, Mukai A, Chang H. RITA inhibits multiple myeloma706 cell growth through induction of p53-mediated caspase-dependent707 apoptosis and synergistically enhances nutlin-induced cytotoxic708 responses. Mol Cancer Ther 2010;9:3041–51.

71032. Sehat B, Andersson S, Vasilcanu R, Girnita L, Larsson O. Role of711ubiquitination in IGF-1 receptor signaling and degradation. PLoS One7122007;2:e340.71333. Mathur RS, Mathur SP. In vitro downregulation of growth factors by714insulin-like growth factor binding protein-3 in cervical cancer. Gynecol715Oncol 2003;91:410–5.71634. Prieur A, Tirode F, Cohen P, Delattre O. EWS/FLI-1 silencing and gene717profiling of Ewing cells reveal downstream oncogenic pathways and a718crucial role for repression of insulin-like growth factor bindingprotein 3.719Mol Cell Biol 2004;24:7275–83.72035. Girnita L, Girnita A, Larsson O. Mdm2-dependent ubiquitination and721degradation of the insulin-like growth factor 1 receptor. ProcNatl Acad722Sci U S A 2003;100:8247–52.72336. Sehat B, Andersson S, Girnita L, Larsson O. Identification of c-Cbl as a724new ligase for insulin-like growth factor-I receptor with distinct roles725from Mdm2 in receptor ubiquitination and endocytosis. Cancer Res7262008;68:5669–77.72737. Enge M, Bao W, Hedstr€om E, Jackson SP, Moumen A, Selivanova G.728MDM2-dependent downregulation of p21 and hnRNP K provides a729switch between apoptosis and growth arrest induced by pharmaco-730logically activated p53. Cancer Cell 2009;15:171–83.73138. Koumenis C, Giaccia A. Transformed cells require continuous activity732of RNA polymerase II to resist oncogene-induced apoptosis. Mol Cell733Biol 1997;17:7306–16.73439. Arima Y, Nitta M, Kuninaka S, Zhang D, Fujiwara T, Taya Y, et al.735Transcriptional blockade induces p53-dependent apoptosis associ-736ated with translocation of p53 to mitochondria. J Biol Chem 2005;280:73719166–76.73840. Higashi Y, Sukhanov S, Parthasarathy S, Delafontaine P. The ubiquitin739ligase Nedd4 mediates oxidized low-density lipoprotein-induced740downregulation of insulin-like growth factor-1 receptor. Am J Physiol741Heart Circ Physiol 2008;295:H1684–9.74241. Beljanski M, Beljanski MS. Three alkaloids as selective destroyers of743cancer cells in mice. Synergy with classic anticancer drugs. Oncology7441986;43:198–203.74542. Sasiela CA, Stewart DH, Kitagaki J, Safiran YJ, Yang Y,WeissmanAM,746et al. Identification of inhibitors for MDM2 ubiquitin ligase activity from747natural product extracts by a novel high-throughput electrochemilu-748minescent screen. J Biomol Screen 2008;13:229–37.74943. FangS, JensenJP, LudwigRL,VousdenKH,WeissmanAM.Mdm2 is a750RING finger-dependent ubiquitin protein ligase for itself and p53. J Biol751Chem 2000;275:8945–51.75244. Yang J, Ahmed A, Poon E, Perusinghe N, de Haven BA, Box G, et al.753Small-molecule activation of p53 blocks hypoxia-inducible factor7541alpha and vascular endothelial growth factor expression in vivo and755leads to tumor cell apoptosis in normoxia and hypoxia. Mol Cell Biol7562009;29:2243–53.75745. Nahi H, Selivanova G, Lehmann S, Mollgard L, Bengtzen S, Concha H,758et al. Mutated and non-mutated TP53 as targets in the treatment of759leukaemia. Br J Haematol 2008;141:445–53.76046. Girnita L, Shenoy SK, Sehat B, Vasilcanu R, Girnita A, Lefkowitz RJ,761et al. {beta}-Arrestin is crucial for ubiquitination anddown-regulation of762the insulin-like growth factor-1 receptor by acting as adaptor for the763MDM2 E3 ligase. J Biol Chem 2005;280:24412–9.76447. Froment P, Dupont J, Christophe-Marine J. Mdm2 exerts pro-apo-765ptotic activities by antagonizing insulin-like growth factor-I-mediated766survival. Cell Cycle 2008;7:3098–103.76748. Mackintosh C, Ord�o~nez JL, García-Domínguez DJ, Sevillano V, Llom-768bart-Bosch A, Szuhai K, et al. 1q gain and CDT2 overexpression769underlie an aggressive and highly proliferative form of Ewing sarcoma.770Oncogene 2012;31:1287–98.77149. Deb SP. Cell cycle regulatory functions of the human oncoprotein772MDM2. Mol Cancer Res 2003;1:1009–16.77350. Huang H-Y, Illei PB, Zhao Z,MazumdarM, Huvos AG, Healey JH, et al.774Ewing sarcomas with p53 mutation or 16/p14ARF homozygous dele-775tion: a highly lethal subset associatedwith poor chemoresponse. JClin776Oncol 2005;23:548–58.

Di Conza et al.

Mol Cancer Ther; 2012 Molecular Cancer Therapeutics10

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES

Q1: Page: 1: Per journal style, genes, alleles, loci, and oncogenes are italicized; proteins areroman. Please check throughout to see that the words are styled correctly.

Q2: Page: 2: AU: Please provide the complete address of F. van Valen.

Q3: Page: 2: Author: Please confirm quality/labeling of all images included within thisarticle. Thank you.

Q4: Page: 9: AU:/PE: Is the disclosure statement correct?

Q5: Page: 9: AU: Please note that author’s contributions have not been included as perinstruction.

AU: Below is a summary of the name segmentation for the authors according to our records.

The First Name and the Surname data will be provided to PubMedwhen the article is indexed

for searching. Please check eachname carefully and verify that theFirstName andSurname are

correct. If a name is not segmented correctly, please write the correct First Name and Surname

on this page and return it with your proofs. If no changes are made to this list, we will assume

that the names are segmented correctly, and the names will be indexed as is by PubMed and

other indexing services.

First Name Surname

Giusy Di Conza

Marianna Buttarelli

Olimpia Monti

Marsha Pellegrino

Francesca Mancini

Alfredo Pontecorvi

Katia Scotlandi

Fabiola Moretti