FUS DDIT3 Fusion Protein-Driven IGF-IR Signaling is a ... · Cancer Therapy: Preclinical FUS–DDIT3 Fusion Protein-Driven IGF-IR Signaling is a Therapeutic Target in Myxoid Liposarcoma

Cancer Therapy: Preclinical

FUS–DDIT3 Fusion Protein-Driven IGF-IRSignaling is a Therapeutic Target in MyxoidLiposarcomaMarcel Trautmann1, Jasmin Menzel1, Christian Bertling1, Magdalene Cyra1,Ilka Isfort1, Konrad Steinestel1, Sandra Elges1, Inga Gr€unewald1, Bianca Altvater2,Claudia Rossig2, Stefan Fr€ohling3,4,5, Susanne Hafner6, Thomas Simmet6,Pierre Åman7, Eva Wardelmann1, Sebastian Huss1, and Wolfgang Hartmann1

Abstract

Purpose: Myxoid liposarcoma is an aggressive disease withparticular propensity to develop hematogenic metastases. Over90% of myxoid liposarcoma are characterized by a reciprocalt(12;16)(q13;p11) translocation. The resulting chimericFUS–DDIT3 fusion protein plays a crucial role in myxoidliposarcoma pathogenesis; however, its specific impact ononcogenic signaling pathways remains to be substantiated. Wehere investigate the functional role of FUS–DDIT3 in IGF-IR/PI3K/Akt signaling driving myxoid liposarcoma pathogenesis.

Experimental Design: Immunohistochemical evaluation ofkey effectors of the IGF-IR/PI3K/Akt signaling axiswas performedin a comprehensive cohort of myxoid liposarcoma specimens.FUS–DDIT3 dependency and biological function of the IGF-IR/PI3K/Akt signaling cascade were analyzed using a HT1080 fibro-sarcoma-based myxoid liposarcoma tumor model and multipletumor–derived myxoid liposarcoma cell lines. An establishedmyxoid liposarcoma avian chorioallantoic membrane modelwas used for in vivo confirmation of the preclinical in vitro results.

Results: A comprehensive subset of myxoid liposarcomaspecimens showed elevated expression and phosphorylationlevels of various IGF-IR/PI3K/Akt signaling effectors. InHT1080 fibrosarcoma cells, overexpression of FUS-DDIT3induced aberrant IGF-IR/PI3K/Akt pathway activity, whichwas dependent on transcriptional induction of the IGF2 gene.Conversely, RNAi-mediated FUS–DDIT3 knockdown in myx-oid liposarcoma cells led to an inactivation of IGF-IR/PI3K/Akt signaling associated with diminished IGF2 mRNA expres-sion. Treatment of myxoid liposarcoma cell lines with severalIGF-IR inhibitors resulted in significant growth inhibitionin vitro and in vivo.

Conclusions:Our preclinical study substantiates the funda-mental role of the IGF-IR/PI3K/Akt signaling pathwayin myxoid liposarcoma pathogenesis and provides a mecha-nism-based rationale for molecular- targeted approaches inmyxoid liposarcoma cancer therapy. Clin Cancer Res; 23(20); 6227–38.�2017 AACR.

IntroductionAccounting for approximately 5% to 10% of all soft tissue

sarcomas, myxoid liposarcoma (MLS) represent �20% of allmalignant adipocytic tumors (1). In the majority of cases, MLS

arise in younger adults, defining the most frequent liposarcomasubtype in patients <20 years of age. Clinically, MLS arecharacterized by a high rate of local recurrences and develop-ment of metastases affecting in total �40% of patients (2).Morphologically, MLS comprise a large spectrum ranging frompaucicellular myxoid tumors to hypercellular round cell sarco-mas associated with a more aggressive clinical course (3).Genetically, the vast majority of MLS is characterized by achromosomal t(12;16)(q13;p11) translocation, juxtaposingthe FUS and DDIT3 genes. About 5% of all patients with MLSdisplay an alternative chromosomal t(12;22) rearrangementleading to an EWSR1–DDIT3 gene fusion (4). The resultingFUS–DDIT3 and EWSR1–DDIT3 fusion proteins are thought toplay an essential role in MLS pathogenesis, acting as transcrip-tional (dys-) regulators (5–8); however, the functional detailsand the specific impact of the chimeric fusion protein ononcogenic signaling pathways known to be activated in MLSis incompletely understood.

It has been shown that MLS are characterized by EGFR,PDGFRB, RET, MET, and VEGFR1 activation sustained by auto-crine/paracrine loops and receptor tyrosine kinase (RTK) cross-talk, resulting in activation of the downstream PI3K/Akt signalingpathway (9, 10). PI3K/Akt signaling is a central hub in the

1Gerhard-Domagk-Institute of Pathology, University Hospital M€unster, M€unster,Germany. 2Department of Pediatric Hematology and Oncology, UniversityChildren's Hospital M€unster, M€unster, Germany. 3Department of TranslationalOncology, National Center for Tumor Diseases (NCT) Heidelberg and GermanCancer Research Center (DKFZ), Heidelberg, Germany. 4Section for Personal-ized Oncology, Heidelberg University Hospital, Heidelberg, Germany. 5GermanCancer Consortium (DKTK), Heidelberg, Germany. 6Institute of Pharmacology ofNatural Products & Clinical Pharmacology, Ulm University, Ulm, Germany.7Sahlgrenska Cancer Center, University of Gothenburg, Gothenburg, Sweden.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author:Marcel Trautmann andWolfgang Hartmann, UniversityHospital M€unster, Domagkstr. 17, M€unster 48149, Germany. Phone: +49 (0) 25183-55440 and -58479; Fax: +49 (0) 251 83 57559; E-mail:[email protected], [email protected]

doi: 10.1158/1078-0432.CCR-17-0130

�2017 American Association for Cancer Research.

ClinicalCancerResearch

www.aacrjournals.org 6227

on January 8, 2021. © 2017 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 21, 2017; DOI: 10.1158/1078-0432.CCR-17-0130

http://crossmark.crossref.org/dialog/?doi=10.1158/1078-0432.CCR-17-0130&domain=pdf&date_stamp=2017-10-10

http://clincancerres.aacrjournals.org/

transduction of different RTK inputs involving diverse growth-controlling effectors such as GSK-3b, p70 S6 kinase (p70S6K),ribosomal S6 protein (11–13), and the cell-cycle regulator CyclinD1 (14–16). In line with the data presented by Negri andcolleagues (9), Barretina and colleagues (17) did not detectsomatic RTK mutations in MLS; however, they were the first todescribe a relatively high frequency (18% of cases) of activatingpoint mutations in the PIK3CA gene encoding the catalytic PI3Ksubunit which was associated with shorter disease-specific sur-vival. Pointing to alternative activation mechanisms of the PI3K/Akt signaling pathway, Demicco and colleagues reported loss ofPTEN or strong overexpression of the insulin-like growth factor-Ireceptor (IGF-IR) in subsets of MLS, which were shown to bemutually exclusive or to occur only very rarely and simultaneouslywith PIK3CA mutations (18). Results previously presented byCheng and colleagues (19) indicate that overexpression of IGF-IR(which was predominantly detected in the prognostically unfa-vorable round cell component) is associated with an aggressiveclinical course in MLS.

Current therapeutic approaches in high-grade MLS comple-ment radical surgery with radiotherapy and/or conventionalchemotherapy, conventionally based on anthracyclines/ifosfa-mide and recently supplemented with novel agents such asTrabectedin or Eribulin (20–22). However, althoughMLS displayhigher chemotherapy sensitivity than other liposarcoma sub-types, the high rate of recurrences and metastases in MLS under-lines the urgent need of novel therapeutic options.

Overall, previously published data suggest a particularimportance of IGF-IR/PI3K/Akt signaling in the pathogenesisand progression of MLS (18, 19). Although the biologicalimpact of PIK3CA and PTEN alterations is intuitive, it remainsopen in which way IGF-IR contributes to MLS oncogenesis andwhether pharmacologic IGF-IR inhibition might result in favor-able therapeutic effects. This study was performed to explorethe functional relevance of IGF-IR/PI3K/Akt signaling in MLS,including its molecular dependence on the pathognomonic

FUS–DDIT3 fusion protein, and to test a molecularly targetedapproach using small molecule IGF-IR inhibitors in a preclin-ical setting.

Materials and MethodsPatients, tumor specimens, and tissue microarray

In summary, 60 myxoid liposarcoma tumor specimens wereincluded (25 women, 35 men; median age at diagnosis was 48years, range, 24–78 years of age). Median tumor size was10 cm (range, 1.5–29 cm). According to the current WHOclassification of tumors of soft tissue and bone (23), alldiagnoses were reviewed by two experienced pathologists(E.W., W.H.) based on clinical information, morphologicalcriteria, and DDIT3 break-apart FISH or reverse-transcriptionalPCR (RT-PCR) analysis, demonstrating the pathognomonictranslocations as previously described (24). Clinicopathologiccharacteristics of the cohort are summarized in Table 1. MLStissue microarrays (TMA) were prepared from formalin-fixed,paraffin-embedded (FFPE; with two representative 1-mmcores) tissue specimens selected from the archival files of theGerhard-Domagk-Institute of Pathology, University HospitalM€unster, M€unster, Germany. Two areas within each tumorwere selected by two experienced pathologists (W.H., S.H.) forthe TMA in order to represent potential heterogeneity, forexample, with regard to the round cell content. Occasionallyoccurring necrobiotic areas and their neighborhood wereexcluded from TMA sampling to avoid the detection of sec-ondary (e.g., hypoxia-induced) alterations. The study wasapproved by the Ethical Committee of the University ofM€unster (2015-548-f-S) and conducted in accordance withcurrent ethical standards (Declaration of Helsinki, 1975).

Table 1. Clinicopathological characteristics of patients with myxoidliposarcoma (n ¼ 60)

Age (years)Mean (�SD) 48.5 (�12.5)Median (range) 48 (24–78)<48 28 (46.7%)�48 32 (53.3%)

TypePrimary tumor 37 (61.7%)Metastasis 9 (15%)Recurrence 9 (15%)ND 5 (8.3%)

MorphologyMyxoid 37 (61.7%)Round cell 23 (38.3%)

Size (cm)Mean (�SD) 10.3 (�5.6)Median (range) 10 (1.5–29)<10 26 (43.3%)�10 21 (35%)ND 13 (21.7%)

SexFemale 25 (41.7%)Male 35 (58.3%)

FISHDDIT3 positive 59 (98.3%)ND 1 (1.7%)

t(12;16) translocation typeFUS–DDIT3 (type 1; exon 7-2) 13 (21.7%)FUS–DDIT3 (type 2; exon 5-2) 27 (45%)ND 20 (33.3%)

Abbreviation: ND, not determined.

Translational Relevance

IGF-IR overexpression has been shown to be associatedwith an unfavorable clinical course in myxoid liposarcomas.However, the molecular contribution of IGF-IR to the path-ogenesis of myxoid liposarcoma as well as its specificmechanism of activation has not been understood so far.We here document a specific, to date unknown molecularlybased mechanism of IGF-IR/PI3K/Akt cascade activation inmyxoid liposarcoma through FUS-DDIT3-dependent IGF2induction. We provide a rational proof of a cell-autono-mous stimulation of myxoid liposarcoma cells involving anIGF-II/IGF-IR transactivation loop and demonstrate highefficacy of a IGF-IR–directed therapeutic approach in vitroand in vivo for myxoid liposarcoma cancer therapy. Ourpreclinical evaluation substantially contributes to theunderstanding of myxoid liposarcoma pathogenesis under-lining the molecular and clinical relevance of actionabletyrosine kinase signals, either based on activating PIK3CAmutations or transmitted via the IGF-IR as induced by thespecific FUS–DDIT3 fusion protein.

Trautmann et al.

Clin Cancer Res; 23(20) October 15, 2017 Clinical Cancer Research6228




Cell culture and cell linesThe MLS cell lines MLS402-91 (FUS-DDIT3 type 1; exon 7-2)

and MLS1765-92 (FUS-DDIT3 type 8; exon 13-2) were contrib-uted by Pierre Åman (25). For the purpose of cell line authen-tication, presence of the pathognomonic t(12;16) translocationwas confirmed by RT-PCR and Sanger sequencing using specificprimers for the translocation subtypes. All monolayer cell cul-tures were grown under standard incubation condition (37�C,humidified atmosphere, 5% CO2) and maintained in RoswellPark Memorial Institute medium 1640 (RPMI; MLS402-91 andMLS1765-92), Dulbecco's Modified Eagles' medium (DMEM;A673 and HT1080) or Iscove's Modified Dulbecco's medium(IMDM; Capan-1), supplemented with 10% FBS (Life Technol-ogies). Mycoplasma testing was performed quarterly by stan-dardized PCR, and cells were passaged for a maximum of 25 to35 culturing cycles between thawing and use in the describedexperiments. To study the effects of increasing concentrations(0.125–2 mmol/L) of NVP-AEW541 (26, 27), BMS-754807 (28,29) and Picropodophyllin (PPP; 30, 31), MLS cells were grownin medium supplemented with 2% FBS. Cell lysis, proteinextraction, and immunoblotting were performed 15 to 72 hoursafter treatment as previously described (32).

ImmunohistochemistryIGF-IR (polyclonal rabbit, 1:100, #3027), phospho-Akt (S473,

monoclonal rabbit, clone D9E, 1:50, #4060), phospho-GSK-3b(S21/9, polyclonal rabbit, 1:50, #9331), phospho-S6 (S240/244,monoclonal rabbit, clone D68F8, 1:100, #5364), phospho-p44/42 MAPK (T202/Y204, monoclonal rabbit, clone D13.14.4E,1:150, #4370), and Cyclin D1 (monoclonal rabbit, clone92G2, 1:50, #2978) antibodies were purchased from Cell Signal-ing Technology, IGF-II (monoclonal mouse, 1:50, clone S1F2,#05-166) from Merck Millipore, MIB-1/Ki-67 (monoclonal rab-bit, clone 30-9, #790-4286) from Roche, and PTEN (monoclonalrabbit, clone SP218, 1:50, #M5180) from Spring Bioscience.Immunohistochemical staining was performed with a Bench-Mark ULTRA Autostainer (VENTANA/Roche) on 3-mmMLS TMAsections. In brief, the staining procedure included: (i) heat-induced epitope retrieval (HIER) pretreatment using Tris-Borate-EDTA buffer (pH 8.4; 95–100�C, 32–72 min) followedby (ii) incubation with respective primary antibodies for 16 to120 min, and (iii) employment of the OptiView DAB IHCDetection Kit (VENTANA/Roche) according to the manufac-turer's instructions. Loss of PTEN protein was assessed by immu-nohistochemical studies according to a standardized algorithmpreviously described (18). Positive and negative control stain-ings using an appropriate IgG subtype (DCS) were included.Immunoreactivity was assessed using a semiquantitative score(0, negative; 1, weak; 2, moderate; and 3, strong) defining thestaining intensity in the positive control (invasive breast cancer,NST) as strong. Only TMA tissue cores with at least moderatestaining (semiquantitative score �2) were considered positivefor the purposes of the study. The IHC readers (W.H., M.T., S.H.,S.E.) were blinded to outcome data, the score cutpoint (positive¼ semiquantitative score �2) was prespecified without prioranalyses of the clinical course.

FUS–DDIT3 fusion protein overexpression in HT1080 cellsGeneration of expression plasmids for FUS–DDIT3, FUS, and

DDIT3was previously described (5). Human HT1080 fibrosarco-ma cells were grown in six-well plates and transfected with 2.5 mg

of plasmid DNA using Lipofectamine 2000 (Life Technologies)according to the manufacturer's instructions. Transient vector-based expression was confirmed 24 hours after transfection byimmunoblotting. As control, HT1080 cells were transfectedwith the peGFP-N1 control plasmid (Clontech Laboratories/Takara Bio Inc.).

Promoter-specific IGF2 RT-PCRPromoter-specific transcription of IGF2was determined using a

competitive RT-PCR assay as previously described (33). Briefly,total RNA was extracted from eGFP, FUS–DDIT3, FUS, orDDIT3-transfected HT1080 cells (RNeasy Plus Kit, Qiagen), reverse-transcribed (SuperScript IV First-Strand Synthesis Super Mix, LifeTechnologies), and PCR-amplified (FastStart Taq PolymerasedNTP Pack, Roche) with specific primer sets for the different IGF2transcripts (P1, P2, P3, and P4). Primer sequences and PCRconditions were previously published (34). Ribosomal 28S rRNAtranscript levels were used as reference (35).

Cell viability assay (MTT)The MTT Cell Proliferation Kit (Roche) was applied according

to themanufacturer's instructions. In brief,MLS402-91 (2� 103),MLS1765-92 (1.5 � 103), and control cells (A673: Ewing'ssarcoma and Capan-1: pancreatic ductal adenocarcinoma;refs. 28, 36, 37) were seeded in 96-well plates (100 mL of mediumsupplemented with 2% FBS) and exposed to increasing concentra-tions of NVP-AEW541, BMS-754807, and PPP (0.125–2 mmol/L)for 72 hours. An appropriate DMSO control was included. At leastthree independent experiments were performed (each in quintu-plicates), and results were calculated as mean � SEM.

RNA interferenceTo target the constant DDIT3 portion of FUS–DDIT3 by RNA

interference (RNAi), a set of prevalidated duplex oligos was used:VHS40607 (siRNA#1), VHS40605 (siRNA#2; Life Technologies),and siRNA#3 (5'-GGAAGUGUAUCUUCAUACAdTdT-3'), pre-viously published as TLS-CHOP siRNA (38). Nontargeting neg-ative control siRNA (BLOCK-iT Alexa Fluor Red FluorescentControl) was purchased from Life Technologies. MLS402-91 andMLS1765-92 cells were cultured in 25 cm2 cell culture flasks(medium supplemented with 2% FBS) and transfected withindicated siRNA (25 pmol; cell density of 50%) using Lipofecta-mine RNAiMAX (Life Technologies). After incubation for72 hours, siRNA-transfected cells were lysed and knockdownefficiency was documented by immunoblotting and/or RT-PCR.

Immunoblot analysisFollowing primary antibodies were used according to the

manufacturer's instructions: IGF-IR, phospho-IGF-IR (Tyr1135/1136), Akt, phospho-Akt (Ser473), GSK-3b, phospho-GSK-3b(Ser21/9), mTOR, phospho-mTOR (Ser2448), p70S6K, phos-pho-p70S6K (Thr389), S6, phospho-S6 (Ser235/236 andSer240/244), p44/42 MAPK, phospho-p44/42 MAPK (Thr202/Tyr204), and Cyclin D1 (all obtained from Cell SignalingTechnology); b-actin (Sigma-Aldrich); DDIT3/GADD153, andFUS/TLS (both obtained from Santa Cruz Technology). TheFUS–DDIT3 fusion protein was detected with an antibodytargeting the N-terminus of DDIT3 (which is retained in theFUS–DDIT3 fusion oncoprotein). Secondary antibody labeling(Bio-Rad Laboratories) and immunoblot development wasperformed using an enhanced Chemiluminescence Detection

IGF-II in myxoid liposarcoma

www.aacrjournals.org Clin Cancer Res; 23(20) October 15, 2017 6229




Kit (SignalFire ECL Reagent; Cell Signaling Technology) and theMolecular Imager ChemiDoc system (Image Lab Software; Bio-Rad Laboratories).

Flow cytometryEffects of NVP-AEW541 and PPP onMLS apoptotic andmitotic

rates were assessed by flow cytometric analyses. Briefly, MLS cellswere grown in 175 cm2 cell culture flasks (medium supplementedwith 2% FBS) and treated with NVP-AEW541 and PPP(0.75–1.5 mmol/L; 72 hours). Adherent cells were detached using0.025% Trypsin (Life Technologies), fixed in 2% paraformalde-hyde (10 minutes on ice), washed in PBS, collected by centrifu-gation and incubated in PBS (supplementedwith 0.25%Triton X-100) for 5 minutes on ice. After an additional washing step, cellswere resuspended in PBS containing following antibodies:cleaved PARP (Asp214; BD Biosciences; phycoerythrin-labeled)and phospho-histone H3 (Ser10; Cell Signaling Technology;Alexa Fluor 647-labeled) followed by incubation for 60 minutesat roomtemperature. Fluorescence intensitywasmeasuredusing aFACSCanto II analytical flow cytometer and cytometric data wereanalyzed using the FACSDiva software (both BD Biosciences).Each experiment was performed at least in duplicates.

Next-generation sequencingA customized GeneRead DNAseq Mix-n-Match V2 panel

(Qiagen) was used to amplify the exonic region of PIK3CA.Target enrichment was processed by means of the GeneReadDNAseq Panel PCR V2 Kit (Qiagen), following the manufac-turer's instructions. All purification and size selection steps wereperformed utilizing Agencourt AMPure XP magnetic beads(Beckman Coulter). End repair, A-addition, and ligation toNEXTflex-96 DNA barcodes (Bioo Scientific) were carried outusing the GeneRead DNA Library I Core Kit (Qiagen). Ampli-fication of adapter-ligated DNA was conducted using NEXTflexprimers (Bioo Scientific) and the HiFi PCR Master Mix (Gene-Read DNA I Amp Kit, Qiagen). Next-generation sequencing(NGS) was performed applying 12.5 pmol/L library pools (2%PhiX V3 control) and the MiSeq Reagent v2 chemistry (Illu-mina, Inc.). NGS data analysis was performed by means of theCLC Biomedical Genomics Workbench software (CLC bio,Qiagen) as described before (39). Validation by Sangersequencing was conducted according to standard proceduresusing the BigDye Terminator v3.1 Cycle Sequencing Kit (LifeTechnologies).

In vivo efficacy of NVP-AEW541 and PPP in MLS chickenchorioallantoic membrane studies

For in vivo experiments, chorioallantoic membrane (CAM)assays were performed as previously described (40). In brief,MLS402-91 and MLS1765-92 cells (1 � 106 cells/egg; dissolvedinmedium/Matrigel 1:1, v/v) were xenografted onto the egg CAM(7 days after fertilization) and incubated at 37�C with 60%relative humidity. On day 8 of incubation, NVP-AEW541, PPP(1 mmol/L) or control (0.2% DMSO in NaCl 0.9%) were appliedtopically. The identical treatment protocol was recapitulated ontwo consecutive days. Three days after treatment initiation,tumor-containing CAM xenografts were explanted, fixed in 5%PFA, and processed for histopathologic examination. Tumorvolume (mm3) was calculated according to the formula: TV ¼length (mm) � width2 (mm) x p/6 (41). All in vivo studies were

performed in accordance with the standards of the National andEuropean Union guidelines.

CompoundsThe IGF-IR kinase inhibitors NVP-AEW541 (hydrochloride;

C27H29N5O * 2HCl; CAS#: 475489-16-8, IGF-IR ATP antagonist),BMS-754807 (C23H24FN9O;CAS#: 1001350-96-4, IGF-IR/IR ATPantagonist), and picropodophyllin (PPP; C22H22O8; CAS#:477-47-4, IGHF-IR non-ATP antagonist; 26–31) were purchasedfrom Biomol and dissolved in DMSO (Sigma-Aldrich). The finalDMSO concentration did not exceed 0.1% (v/v) for all in vitro andin vivo applications.

Statistical analysisImmunohistochemical staining results and Kaplan–Meier

survival/event-free correlations were statistically analyzed byGehan–Breslow–Wilcoxon test (GraphPad Software). Two-groupcomparisons were analyzed by unpaired Student t test (GraphPadSoftware). Experimental results ofMTTassays andflow cytometricanalyses are represented as mean � SEM from n independentexperiments (n � 3). Statistical differences were consideredsignificant at P < 0.05 (�). The concentration of NVP-AEW541,BMS-754807, and PPP required for 50% growth inhibition(IC50 value) was calculated by nonlinear regression analysis usingthe GraphPad Prism (GraphPad Software).

ResultsExpression of IGF-IR and PI3K/Akt/GSK-3b signalingcomponents in human MLS tumor tissues and MLS cell lines

To determine the involvement of IGF-IR- and PI3K/Akt/GSK-3b–mediated signal transduction in myxoid liposarcomatumorigenesis, expression of IGF-IR, IGF-II, Akt (Ser473),GSK-3b (Ser21/9), S6 (Ser240/244), Cyclin D1, and PTEN wereexamined in a comprehensive set of 60 MLS specimens usingimmunohistochemistry. In addition, p44/42 MAPK (Thr202/Tyr204) was analyzed as an indicator of activated RAS/RAF/ERKsignaling (summarized in Supplementary Table S1). Positivestaining for IGF-IR and IGF-II was detected in 49.1% and 70.2%of all cases, respectively (Fig. 1A and B), whereas six of 60specimens (10%) showed no membranous IGF-IR immunore-activity. Consistently, several phosphorylated signaling com-ponents were highly expressed in MLS including Akt (Ser473),GSK-3b (Ser21/9), S6 (Ser240/244), and Cyclin D1; p44/42MAPK (Thr202/Tyr204) was also detected at relevant levels(Fig. 1C–G). Loss of PTEN protein expression was detected in9% of all studied MLS cases. In total, 26.3% of MLS specimensdisplayed moderate to strong phosphorylation levels of Akt(Ser473), 34.5% were moderately/strongly positive for GSK-3b(Ser21/9), 34.8% for S6 (Ser240/244), and 53.4% for p44/42MAPK (Thr202/Tyr204). Strong Cyclin D1 expression levelswere detected in 10.3% of tumors, whereas 36.2% displayedmoderate and 41.4% weak Cyclin D1 expression (summarizedin Fig. 2A). Moderate to strong staining for MIB-1 was detectedin 21.1% of all cases (Fig. 1H). An overlap of positive IGF-IR/IGF-II immunoreactivity and phosphorylation of Akt (Ser473),GSK-3b (Ser21/9), and S6 (Ser240/244) was detected in 33% ofMLS specimens (Fig. 2B). Expression of IGF-IR and PI3K/Akt/GSK-3b signaling components did not correlate with thepatients' age, gender, translocation subtype, and/or tumor size.No statistically significant differences in overall/event-free

Trautmann et al.





Figure 1.

Activation of the IGF-IR and PI3K/Akt/GSK-3b signaling axis in representativecases of myxoid liposarcoma. A and B,Immunohistochemical staining shows strongexpression of IGF-IR and IGF-II. C–F,Elevated phosphorylation levels of Akt(S473), GSK-3b (S21/9), and S6 (S240/244)indicate PI3K/Akt signaling pathway activitywith p44/42 MAPK (T202/Y204) beingactivated as well. G and H, Elevatedexpression levels of Cyclin D1 and MIB-1(original magnification: 10�, inset 20�).






survival were detected for IGF-IR (P ¼ 0.305), IGF-II (P ¼0.971), p-Akt (Ser473; P ¼ 0.162), p-GSK-3b (Ser21/9; P ¼0.607), or Cyclin D1 (P ¼ 0.269) IHC-positive subgroups. Inaccordance with the immunohistochemical results in MLStissue specimens, elevated protein expression and phosphory-lation levels were demonstrated in total protein extracts (Fig.2C) and immunostainings of MLS cell lines (SupplementaryFig. S1). As oncogenic mutations in the PIK3CA gene might be

responsible for constitutive activation of Akt signaling, weanalyzed the entire PIK3CA coding region by targeted NGS. Innine of 60 (15%) MLS specimens, SNVs in the PIK3CA genecould be detected (summarized in Supplementary Table S1),whereas no PIK3CA alterations were identified in both MLS celllines (Supplementary Fig. S2). Tumors with PIK3CA mutationsdisplayed a statistically significant (P ¼ 0.0003) worse overallsurvival (OS) and disease-free survival (DFS).

MLS402-91

MLS1765-92

Akt

p-Akt(Ser473)

GSK-3β

p-GSK-3β(Ser21/9)

IGF-IR

p-IGF-IR(Tyr1135/1136)

p70S6K

p-p70S6K (Thr389)

S6

p-S6(Ser235/236)

p-S6(Ser240/244)

β-Actin

Cyclin D1

mTOR

p-mTOR(Ser2448)

FUS-DDIT3 (type 1)165 bp

FUS-DDIT3 (type 8)197 bp

28S rRNA130 bp

MLS402-91

MLS1765-92

p44/42 MAPKp-p44/42 MAPK(Thr202/Tyr204)

CA

B

Figure 2.

Activation of IGF-IR and PI3K/Akt/GSK-3b signaling in myxoid liposarcoma tissue specimens and cell lines. A, Immunohistochemical spectrum of tumortissue specimens summarized as bar chart. B, Venn diagram indicating the concordance of positive IGF-IR/IGF-II immunoreactivity and phosphorylationof Akt (S473), S6 (S240/244), and/or GSK-3b (Ser21/9) in 33% of tumor specimens. In total, five cases were quintuple-negative for IGF-IR, IGF-II, Akt(S473), S6 (S240/244), or GSK-3b (S21/9) expression. C, Immunoblotting results demonstrate elevated expression and phosphorylation levels of IGF-IRand PI3K/Akt/GSK-3b signaling components in total protein extracts of MLS402-91 and MLS1765-92 cells (b-actin was used as loading reference).Detection of t(12;16) FUS–DDIT3 fusion gene transcripts in MLS402-91 (type 1; exon 7-2) and MLS1765-92 (type 8; exon 13-2) cells by RT-PCR (28S rRNAwas used as loading reference).

Trautmann et al.





Activation of IGF-IR and PI3K/Akt/GSK-3b signaling is inducedby the chimeric FUS–DDIT3 fusion protein

To evaluate whether signal transduction via IGF-IR andPI3K/Akt/GSK-3b activation is dependent on the expression ofFUS-DDIT3, HT1080 cells were transiently transfected withFUS–DDIT3, FUS, DDIT3, or eGFP expression vectors. First, wedetermined the molecular regulation of promoter-specific IGF2expression, demonstrating induction of P2-dependent IGF2 tran-scripts uponFUS–DDIT3 expression (Fig. 3A, top). The stimulatedP2 promoter-dependent IGF2 transcript levels in FUS–DDIT3expressing HT1080 cells were comparable to levels inMLS402-91and MLS1765-92 cell lines (Fig. 3A, bottom). Upon vector-basedFUS-DDIT3 expression, immunoblot analyses showed signifi-cantly increased phosphorylation levels of IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), and mTOR (Ser2448)compared to cells overexpressing FUS, DDIT3, or eGFP (Fig. 3B).No relevant changes in baseline protein levels were detected,confirming FUS-DDIT3-triggered stimulation of the IGF-IR andPI3K/Akt/GSK-3b pathway activation as indicated by enhancedphosphorylation levels. Stimulation of MLS402-91 andMLS1765-92 cells with recombinant human IGF-II protein(200 ng/mL; 15 minutes) was associated with enhanced phos-phorylation levels of IGF-IR (Tyr1135/1136), Akt (Ser473),GSK-3b (Ser21/9), and mTOR (Ser2448), suggesting IGF-IR–mediated signals as a functionally relevant mechanism leadingto PI3K/Akt/GSK-3b activation (Fig. 3C). As shown in Supple-mentary Fig. S3, IGF-II stimulation was able to rescue PI3K/Akt/GSK-3b pathway activation in FUS-DDIT3-depleted myxoid lipo-sarcoma cells. A minor induction of phosphorylation wasobserved upon eGFP control expression; however, this activationwas not associated with IGF2 transcriptional induction (Fig. 3A,top) or elevated expression levels of Cyclin D1 (Fig. 3B).

FUS–DDIT3 knockdown affects IGF2 mRNA transcription andIGF-IR/Akt/GSK-3b phosphorylation levels in MLS cell lines

To further analyze the functional contribution of FUS–DDIT3fusion protein to oncogenic IGF-IR and PI3K/Akt/GSK-3b medi-ated signaling by a nonpharmacological approach, MLS402-91andMLS1765-92 cells were transfected with siRNA. Consistently,knockdown of FUS–DDIT3 significantly reduced levels of (i) P2promoter-dependent IGF2 transcripts, (ii) phosphorylation ofIGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), andmTOR (Ser2448), combined with (iii) reduced Cyclin D1 expres-sion (Fig. 3D and Supplementary Fig. S4). The knockdownefficiency for a set of prevalidated and published siRNAduplex oligos (38) targeting the DDIT3 portion of the chimericFUS–DDIT3 fusion gene (which is retained in the FUS–DDIT3fusion oncoprotein) was validated on protein level (Supplemen-tary Fig. S4). These results confirmed that the FUS–DDIT3 fusionprotein is involved in the regulation of IGF-IR and PI3K/Akt/GSK-3b signaling activity through modulation of IGF2 mRNA expres-sion. Consistent with elevated phosphorylation levels of IGF-IR,Akt, GSK-3b, andmTORupon transient FUS-DDIT3 expression inHT1080 cells (Fig. 3A), phosphorylation and activation wasinversely suppressed compared to nontargeting negative controlsiRNA (Fig. 3D and Supplementary Fig. S4).

NVP-AEW541, BMS-754807, and PPP reduce cell viability ofMLS cell lines in vitro

To investigate the biological effects of pharmacological inhi-bition of IGF-IR, MLS and control cell lines (A673 and Capan-1;

Supplementary Fig. S5) were exposed to increasing concen-trations (0.125–2 mmol/L) of small molecule (i) IGF-IR ATPantagonists (NVP-AEW541 and BMS-754807) and (ii) the IGF-IRnon-ATP antagonist PPP. In MTT assays, all three IGF-IR inhibi-tors were effective in suppressing MLS and A673 cell viabilitywith IC50 values ranging from 0.36 to 1.35 mmol/L, showinga dose-dependent mode of action. MLS402-91 (fusion type 1;exon 7-2) cells were more sensitive to IGF-IR inhibitioncompared to MLS1765-92 (fusion type 8; exon 13-2) cells.Capan-1 control cells expressing low levels of IGF-IR showedonly minor responses (Fig. 4A, Supplementary Fig. S5, andTable 2).

NVP-AEW541 and PPP inhibits IGF-IR and PI3K/Akt/GSK-3bsignal transduction activity

Suppressive effects of treatmentwith the IGF-IR inhibitorsNVP-AEW541 and PPP on signal transduction activity in MLS cell lineswere assessed by immunoblotting. In two MLS cell lines, signif-icant dose-dependent reduction of FUS-DDIT3-induced phos-phorylation levels were demonstrated for IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), p70S6K (Thr389), andS6 (Ser235/236 and Ser240/244; Fig. 4B and SupplementaryFig. S6). Cyclin D1 showed a dose- (0.5–1 mmol/L) and time-dependent downregulation in MLS402-91 andMLS1765-92 cells(Supplementary Fig. S7).

NVP-AEW541 and PPP reduce cell viability by inducingapoptosis and decreasing mitotic activity in MLS cell lines

Performing flow cytometric analyses, PARP (Asp214) cleav-age was used as a marker for apoptosis and phospho-histoneH3 (Ser10) was used as a marker for mitotic activity. Aftertreatment with increasing concentrations of NVP-AEW541 orPPP (0.75–1.5 mmol/L; 72 hours), MLS402-91 cells showeda significantly increased rate of apoptosis, accompanied by adecrease of the mitotic fraction (Fig. 4C and SupplementaryFig. S6). Similar results were observed in MLS1765-92 cells(Supplementary Fig. S8).

In vivo efficacy of NVP-AEW541 and PPP in aCAMmodel ofMLSTo verify the efficacy of IGF-IR inhibition on tumor growth and

progression in an in vivo model of human MLS, we xenograftedMLS402-91 and MLS1765-92 cells onto a chick CAM to initiateMLS tumor formation. Topical NVP-AEW541 and PPP adminis-tration (1 mmol/L) resulted in a significant reduction of tumorvolume compared to the DMSO vehicle control group (Fig. 4Dand Supplementary Fig. S6; �, P < 0.05). Representative H&Estainings of CAM tumor specimens are included in Supplemen-tary Fig. S9.

DiscussionMyxoid liposarcoma is a malignant lipogenic soft tissue

neoplasia with a particular propensity to develop distant metas-tases. High histologic grade, generally defined as >5% roundcell component is the major predictor of an unfavorable out-come (3, 42). Although there is an established role for con-ventional radiotherapy and cytotoxic therapies in MLS (20),molecularly targeted therapeutic approaches are still missing.The vast majority of MLS are characterized by the FUS–DDIT3gene fusion encoding an aberrant transcriptional regulator thathas the potential to transform mesenchymal stem cells to form






Figure 3.

Myxoid liposarcoma associated FUS–DDIT3 fusion protein stimulates IGF-IR and PI3K/Akt/GSK-3b pathway signal transduction. A, Induction of promoterP2-dependent IGF2 transcripts in FUS–DDIT3 expressing HT1080 cells (top); comparable IGF2 levels in FUS–DDIT3-tranfected HT1080 and MLS celllines (bottom). B, HT1080 cells were transfected with indicated FUS–DDIT3, FUS, DDIT3, or eGFP control expression vectors to study IGF-IR and PI3K/Akt/GSK-3b–mediated signal transduction. FUS–DDIT3 expression significantly increased phosphorylation of IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b(Ser21/9), and mTOR (Ser2448), confirming pathway induction and activity. Elevated target protein levels of Cyclin D1 in HT1080 cells expressing theFUS–DDIT3 fusion protein. C, Enhanced phosphorylation of IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), and mTOR (Ser2448) uponstimulation with IGF-II (200 ng/mL; 15 minutes). D, In MLS402-91 and MLS1765-92 cells, siRNA-mediated knockdown of FUS-DDIT3 significantly reducedlevels of P2-promoter-dependent IGF2 transcripts (RT-PCR) and phosphorylation of IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), and mTOR(Ser2448). b-Actin and 28S rRNA were used as loading references (NTC, no template control).

Trautmann et al.





Figure 4.

In vitro and in vivo evaluation of NVP-AEW541, BMS-754807, and PPP in two myxoid liposarcoma cell lines. A, Cell viability of MLS402-91 and MLS1765-92 cellswas significantly reduced by treatment with increasing concentrations of NVP-AEW541, BMS-754807, and PPP in MTT assays. A673 (Ewing's sarcoma) andCapan-1 (pancreatic ductal adenocarcinoma) cells were included as sensitive and/or resistant controls to IGF-IR inhibition, respectively. At least threeindependent experiments were performed (each in quintuplicates); results are expressed as mean � SEM. B, NVP-AEW541 suppressed phosphorylation levelsof IGF-IR (Tyr1135/1136), Akt (Ser473), GSK-3b (Ser21/9), p70S6K (Thr389), and S6 (Ser235/236 and Ser240/244) in both MLS cell lines. Changes in Cyclin D1expression levels were determined by immunoblotting. C, In flow cytometric analyses, significantly increased rates of apoptosis (cleaved PARP) anddecreased mitotic fractions (phospho-histone H3) were detected upon treatment with NVP-AEW541 (0.75–1.5 mmol/L; DMSO was used as control). D, MLScells were xenografted on the CAM of chick eggs. Tumor-bearing eggs were randomized and treated with NVP-AEW541 (1 mmol/L) or control (0.2%DMSO in NaCl 0.9%). Significantly reduced tumor volumes � SEM (NVP-AEW541–treated; � , P < 0.05) and representative explants are shown (H&E stainingof CAM tumor specimens are included in Supplementary Fig. S9).






MLS in mice (8). As in other sarcomas driven by specific genefusions, MLS characteristically display only few additionalgenetic alterations; however, 14% to 18% of MLS were reportedto carry activating mutations in the PIK3CA gene encoding thecatalytic PI3K subunit which occur predominantly in the moreaggressive round cell variant of MLS. As alternative mechanismsof PI3K/Akt signaling pathway activation, rare biallelic losses ofPTEN, and overexpression of the IGF-IR (in �25% of the casesand occurring only very rarely simultaneously with PI3KCAmutations) have been described (17, 18). As reported, highprevalence of IGF-IR overexpression in the round cell variant ofMLS (18) fits well with previous data showing that IGF-IRoverexpression in MLS is associated with a poor metastasis-free survival (19). Because therapeutic targeting of the chimericfusion protein represents a particular challenge, it appearsreasonable to therapeutically address a signaling pathwaywhich is known to be significantly associated with a moreaggressive MLS phenotype. We therefore set out to analyze indetail IGF-IR-related signals mediated through the PI3K/Aktsignaling cascade, putting a particular focus on the functionaldependency on the chimeric FUS–DDIT3 fusion protein.

In accordance with previous studies (18, 19), we detectedmoderate to strong expression levels of IGF-IR and IGF-II in alarge subset of MLS including the major subfraction of tumorswith a significant round cell component. In immunohistochem-ical analyses, strong IGF-IR/IGF-II expression was associated withphosphorylation of Akt, GSK-3b, and/or S6 in 33% of the cases,indicating downstream activation of PI3K/Akt signals (Figs. 1and 2). In contrast to what was previously reported by Cheng andcolleagues in a series of 32 MLS, we were unable to confirm asignificant prognostic impact of IGF-IR overexpression in ourcohort of MLS (19). However, in line with data presented byBarretina and colleagues (17), we found PIK3CAmutations to beassociated with a negative prognostic impact (summarized inSupplementary Table S1).Confirming the essential biological roleof IGF-IR-dependent signals in MLS lacking PIK3CA or PTENalterations, the used MLS cells (all wild type for PIK3CA; Sup-plementary Fig. S2) responded to IGF-II stimulation with asignificant induction of phosphorylation of IGF-IR and PI3K/Aktdownstream effectors (Fig. 3C and Supplementary Fig. S3). Tocomprehensively understand the oncogenic mechanisms leadingto aberrant activation of IGF-IR signaling in MLS in detail, weexplored the molecular dependence of IGF-IR signals on thepathognomonic FUS–DDIT3 fusion protein. In response toFUS–DDIT3 knockdown, MLS cells showed loss of phosphory-lation of IGF-IR, Akt, GSK-3b, and a significant reduction of totalCyclin D1 protein levels (Fig. 3D and Supplementary Fig. S4).These alterations were associated with decreased expression levelsof IGF2 promoter P2-dependent transcripts (Fig. 3D). To evaluatethe functional role of the chimeric fusion protein in a MLS-independent cell context, we overexpressed the FUS–DDIT3fusion protein in HT1080 fibrosarcoma cells and thus confirmed

specific regulation of IGF2 promoter P2 transcription and subse-quent activation of downstream PI3K/Akt effectors through thechimeric oncogenic driver ofMLS (Fig. 3A and B).Our finding of afunctional connection of the MLS-specific FUS–DDIT3 genefusion and the activation of IGF signaling is an essential contri-bution to the understanding of the role of IGF-IR in MLS andmakes strong proof for the presence of a cell-autonomous stim-ulation of MLS cells. Thus, our finding provides a missing linkin the concept of MLS pathogenesis in which a functional con-nection between the chimeric transcriptional (dys-) regulatorFUS-DDIT3 and the activation of IGF-IR-dependent PI3K/Aktsignals (occurring in a large subset of tumors) is not known.Based on our findings, the IGF-IR signaling cascade now qualifiesas a specificmolecularly based target for therapeutic approaches inMLS. The pattern of activated IGF signals in MLS resemblesfindings reported for other fusion-driven sarcomas.We andothersreported transcriptional regulation of IGF2 through the oncogenicSS18–SSX fusion protein and subsequent activation of the PI3K/Akt signaling cascade in synovial sarcoma (43, 44), and thePAX3–FKHR fusion protein was shown to induce IGF2 in inalveolar rhabdomyosarcoma (45). In Ewing's sarcoma, theEWSR1-FLI1 oncoprotein was shown to activate IGF-IR signalsthrough the transcriptional induction of IGF1 (46, 47), and thisfinding led to the successful exploration of IGF-IR targeted ther-apeutic approaches in Ewing's sarcoma (48, 49). The high prev-alence of IGF-IR transmitted signals in different fusion-drivensarcomas implies a particular importance of inputs transmittedthrough this cascade for these soft tissue tumors carrying only fewmutations apart from the pathognomonic gene fusions. From amolecular diagnostic point of view, fusion-driven sarcomas there-fore represent challenging paradigmatic malignancies in whichpure high-throughput genomics canoftennot provide elementarymutations qualifying as therapeutic targets but in which func-tional precision oncology approaches are needed.

To investigate if IGF-IR–directed approaches might be of ther-apeutic benefit in MLS, we treated MLS and control cells withdifferent small molecule IGF-IR inhibitors (26–31). Our datashowa significant dose-dependent reduction ofMLSproliferationand viability (Fig. 4A), associated with the expected decrease inphosphorylation of PI3K/Akt effectors (Fig. 4B and Supplemen-tary Fig. S6). Consistent with these in vitro results, administrationof NVP-AEW541 and PPP to xenografted MLS402-91 andMLS1765-92 cells led to an inhibition of tumor growth in vivo(Fig. 4D and Supplementary Figs. S6 and S9). The panel ofdifferent substances available for inhibition of the IGF signalingsystem has considerably increased during the recent years andnow includes, apart from small molecule tyrosine kinase inhibi-tors, different monoclonal antibodies to the IGF-IR, and anti-bodies to IGF-I and IGF-II (50). In contrast to relatively disap-pointing results in early clinical trials with other solid tumors,sustained success of IGF-IR–directed therapies was observed indefined subsets of sarcomas (48, 49). However, the major

Table 2. IC50 values for IGF-IR inhibitors in myxoid liposarcoma, Ewing's sarcoma, and pancreatic ductal adenocarcinoma cell lines

IC50 (mmol/L)Compound MLS402-91 MLS1765-92 A673 Capan-1

NVP-AEW541 1.16 1.35 0.70 NDBMS-754807 0.36 0.50 0.21 NDPicropodophyllin (PPP) 0.66 0.75 0.57 1.38

Cytotoxic effects on myxoid liposarcoma (MLS402-91 and MLS1765-92), Ewing's sarcoma (A673), and pancreatic adenocarcinoma (Capan-1) cells were assessed inMTT assays (72 hours). Results are represented as mean of at least three separate experiments performed in quintuplicates (ND, not determined).

Trautmann et al.





challenge of therapeutic approaches addressing the IGF systemremains the identification of appropriate predictive biomarkers.In MLS (absence of a), PIK3CA mutation and IGF-II overexpres-sion as well as IGF-IR phosphorylation might be tested as suchpredictive indicators.

In conclusion, the results of this study imply that activationof the IGF-IR/PI3K/Akt signaling system is a common pattern inMLS which appears to be transcriptionally controlled, at least inpart by induction of IGF2 gene transcription in a FUS–DDIT3-dependent manner. Disruption of IGF-IR–mediated signals viaa small molecule inhibitor may provide an effective therapeuticapproach for advanced MLS. The present preclinical testing ofan IGF-IR–directed targeted approach shows potent effects bothin vitro and in vivo, qualifying the IGF-IR/PI3K/Akt signalingpathway as a specific therapeutic target in MLS.

Disclosure of Potential Conflicts of InterestE. Wardelmann reports receiving speakers bureau honoraria from Eli Lilly

and Company, The Menarini Group, Nanobiotix, and Novartis and is aconsultant/advisory board member for New Oncology, Novartis, and Pfizer.No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: M. Trautmann, C. Rossig, S. Huss, W. HartmannDevelopment of methodology: M. Trautmann, C. Bertling, K. Steinestel, B.Altvater, S. Hafner, W. HartmannAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. Trautmann, J. Menzel, C. Bertling, M. Cyra,

I. Isfort, S. Elges, I. Gr€unewald, B. Altvater, S. Hafner, P. Åman, E. Wardelmann,S. Huss, W. HartmannAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): M. Trautmann, J. Menzel, C. Bertling, M. Cyra,I. Isfort, K. Steinestel, S. Elges, B. Altvater, S. Hafner, S. Huss, W. HartmannWriting, review, and/or revision of the manuscript: M. Trautmann,K. Steinestel, I. Gr€unewald, B. Altvater, C. Rossig, S. Fr€ohling, S. Hafner,P. Åman, E. Wardelmann, S. Huss, W. HartmannAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M. Trautmann, J. Menzel, C. Bertling, S. Huss,W. HartmannStudy supervision: M. Trautmann, W. HartmannOther (supervision of the CAM xenograft experiments): T. Simmet

AcknowledgmentsThe authors thank Charlotte Sohlbach and Inka Buchroth for excellent

technical support.

Grant SupportThis study was supported in part by grants from "Innovative Medical

Research" of the University of M€unster Medical School (M.T. and S.H.;#HU121421), the Deutsche Forschungsgemeinschaft (DFG) (K.S.; #STE2467/1-1), and the Wilhelm Sander-Stiftung (W.H., M.T., and K.S.;#2016.099.1).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received January 17, 2017; revised May 6, 2017; accepted June 14, 2017;published OnlineFirst June 21, 2017.

References1. Fletcher CDM, Unni K, Mertens F. Pathology and genetics of tumours of

soft tissue and bone. WHO Classification Tumours 2002:200–4.2. Dei Tos AP. Liposarcomas: diagnostic pitfalls and new insights. Histopa-

thology 2014;64:38–52.3. Antonescu CR, Tschernyavsky SJ, Decuseara R, Leung DH, Woodruff JM,

Brennan MF, et al. Prognostic impact of P53 status, TLS-CHOP fusiontranscript structure, and histological grade in myxoid liposarcoma: amolecular and clinicopathologic study of 82 cases. Clin Cancer Res 2001;7:3977–87.

4. Panagopoulos I, HoglundM, Mertens F, Mandahl N, Mitelman F, Aman P.Fusion of the EWS and CHOP genes in myxoid liposarcoma. Oncogene1996;12:489–94.

5. Engstrom K, Willen H, Kabjorn-Gustafsson C, Andersson C, Olsson M,Goransson M, et al. The myxoid/round cell liposarcoma fusion oncogeneFUS-DDIT3 and the normal DDIT3 induce a liposarcoma phenotype intransfected human fibrosarcoma cells. Am J Pathol 2006;168:1642–53.

6. Kuroda M, Ishida T, Takanashi M, Satoh M, Machinami R, Watanabe T.Oncogenic transformation and inhibition of adipocytic conversion ofpreadipocytes by TLS/FUS-CHOP type II chimeric protein. Am J Pathol1997;151:735–44.

7. Perez-Losada J, Pintado B, Gutierrez-Adan A, Flores T, Banares-Gonzalez B,del Campo JC, et al. The chimeric FUS/TLS-CHOP fusion protein specificallyinduces liposarcomas in transgenic mice. Oncogene 2000;19:2413–22.

8. Riggi N, Cironi L, Provero P, SuvaML, Stehle JC, Baumer K, et al. Expressionof the FUS-CHOP fusion protein in primarymesenchymal progenitor cellsgives rise to amodel ofmyxoid liposarcoma.Cancer Res 2006;66:7016–23.

9. Negri T, Virdis E, Brich S, Bozzi F, Tamborini E, TarantinoE, et al. Functionalmapping of receptor tyrosine kinases in myxoid liposarcoma. Clin CancerRes 2010;16:3581–93.

10. Andersson MK, Goransson M, Olofsson A, Andersson C, Aman P. Nuclearexpression of FLT1 and its ligand PGF in FUS-DDIT3 carrying myxoidliposarcomas suggests the existence of an intracrine signaling loop. BMCCancer Jun 1;10:249. doi: 10.1186/1471-2407-10-249.

11. Martin-P�erez J, Thomas G. Ordered phosphorylation of 40S ribosomalprotein S6 after serum stimulation of quiescent 3T3 cells. Proc Nat Acad Sci1983;80:926–30.

12. Meyuhas O.Chapter two-ribosomal protein S6 phosphorylation: fourdecades of research. Int Rev Cell Mol Biol 2015;320:41–73.

13. Wettenhall R, Erikson E, Maller J. Ordered multisite phosphorylation ofXenopus ribosomal protein S6 by S6 kinase II. J Biol Chem 1992;267:9021–27.

14. Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3bregulates cyclin D1 proteolysis and subcellular localization. Genes Dev1998;12:3499–511.

15. Osaki M, Oshimura M, Ito H. PI3K-Akt pathway: its functions and altera-tions in human cancer. Apoptosis 2004;9:667–76.

16. Vivanco I, Sawyers CL. The phosphatidylinositol 3-Kinase AKT pathway inhuman cancer. Nat Rev Cancer 2002;2:489–501.

17. Barretina J, Taylor BS, Banerji S, Ramos AH, Lagos-Quintana M, DecarolisPL, et al. Subtype-specific genomic alterations define new targets for soft-tissue sarcoma therapy. Nat Genet 2010;42:715–21.

18. Demicco EG, Torres KE, Ghadimi MP, Colombo C, Bolshakov S, HoffmanA, et al. Involvement of the PI3K/Akt pathway in myxoid/round cellliposarcoma. Mod Pathol 2012;25:212–21.

19. Cheng H, Dodge J, Mehl E, Liu S, Poulin N, van de Rijn M, et al. Validationof immature adipogenic status and identification of prognostic biomarkersin myxoid liposarcoma using tissue microarrays. Hum Pathol 2009;40:1244–51.

20. Ratan R, Patel SR. Chemotherapy for soft tissue sarcoma. Cancer 2016;122:2952–60.

21. Grosso F, Jones RL, Demetri GD, Judson IR, Blay J-Y, Le Cesne A, et al.Efficacy of trabectedin (ecteinascidin-743) in advanced pretreated myxoidliposarcomas: a retrospective study. Lancet Oncol 2007;8:595–602.

22. Sch€offski P, Chawla S, Maki RG, Italiano A, Gelderblom H, Choy E, et al.Eribulin versus dacarbazine in previously treated patients with advancedliposarcoma or leiomyosarcoma: a randomised, open-label, multicentre,phase 3 trial. Lancet 2016;387:1629–37.

23. Fletcher CD, Organization WH. WHO classification of tumours of softtissue and bone:[this book reflects the views of a working group thatconvened for a consensus and editorial meeting at theUniversity of Zurich,Switzerland, 18-20 April 2012]. International Agency for Research onCancer; 2013.






24. Powers MP, Wang WL, Hernandez VS, Patel KS, Lev DC, Lazar AJ, et al.Detection of myxoid liposarcoma-associated FUS-DDIT3 rearrangementvariants including a newly identified breakpoint using an optimized RT-PCR assay. Mod Pathol 2010;23:1307–15.

25. Aman P, Ron D, Mandahl N, Fioretos T, Heim S, Arheden K, et al.Rearrangement of the transcription factor gene CHOP in myxoidliposarcomas with t(12;16)(q13;p11). Genes Chromosomes Cancer1992;5:278–85.

26. García-Echeverría C, Pearson MA, Marti A, Meyer T, Mestan J, Zimmer-mann J, et al. In vivo antitumor activity of NVP-AEW541—a novel,potent, and selective inhibitor of the IGF-IR kinase. Cancer Cell 2004;5:231–39.

27. Manara MC, Landuzzi L, Nanni P, Nicoletti G, Zambelli D, Lollini PL,et al. Preclinical in vivo study of new insulin-like growth factor-Ireceptor–specific inhibitor in Ewing's sarcoma. Clin Cancer Res 2007;13:1322–30.

28. Carboni JM, Wittman M, Yang Z, Lee F, Greer A, Hurlburt W, et al. BMS-754807, a small molecule inhibitor of insulin-like growth factor-1R/IR.Mol Cancer Thera 2009;8:3341–49.

29. Wittman MD, Carboni JM, Yang Z, Lee FY, Antman M, Attar R, et al.Discovery of a 2, 4-disubstituted pyrrolo [1, 2-f][1, 2, 4] triazine inhibitor(BMS-754807) of insulin-like growth factor receptor (IGF-1R) kinase inclinical development. J Med Chem 2009;52:7360–63.

30. Girnita A, Girnita L, del Prete F, Bartolazzi A, Larsson O, Axelson M.Cyclolignans as inhibitors of the insulin-like growth factor-1 receptor andmalignant cell growth. Cancer Res 2004;64:236–42.

31. Vasilcanu D, Girnita A, Girnita L, Vasilcanu R, Axelson M, Larsson O. Thecyclolignan PPP induces activation loop-specific inhibition of tyrosinephosphorylation of the insulin-like growth factor-1 receptor. Link to thephosphatidyl inositol-3 kinase/Akt apoptotic pathway. Oncogene 2004;23:7854–62.

32. Trautmann M, Sievers E, Aretz S, Kindler D, Michels S, Friedrichs N, et al.SS18-SSX fusion protein-induced Wnt/beta-catenin signaling is a thera-peutic target in synovial sarcoma. Oncogene 2014;33:5006–16.

33. Hartmann W, Waha A, Koch A, Albrecht S, Gray SG, Ekstrom TJ, et al.Promoter-specific transcription of the IGF2 gene: a novel rapid, non-radioactive and highly sensitive protocol for mRNA analysis. VirchowsArch 2001;439:803–7.

34. Hartmann W, Waha A, Koch A, Albrecht S, Gray SG, Ekstr€om TJ, et al.Promoter-specific transcription of the IGF2 gene: a novel rapid, non-radioactive and highly sensitive protocol for mRNA analysis. VirchowsArch 2001;439:803–07.

35. Thellin O, Zorzi W, Lakaye B, De Borman B, Coumans B, Hennen G, et al.Housekeeping genes as internal standards: use and limits. J Biotechnol1999;75:291–5.

36. Martins AS, Ord�o~nez JL, Amaral AT, Prins F, Floris G, Debiec-Rychter M,et al. IGF1R signaling in Ewing sarcoma is shaped by clathrin-/caveolin-dependent endocytosis. PloS One 2011;6:e19846.

37. Patel M, Gomez NC, McFadden AW, Moats-Staats BM, Wu S, Rojas A, et al.PTEN deficiency mediates a reciprocal response to IGFI and mTOR inhi-bition. Mol Cancer Res 2014;12:1610–20.

38. Oikawa K, Tanaka M, Itoh S, Takanashi M, Ozaki T, Muragaki Y, et al. Anovel oncogenic pathway by TLS-CHOP involving repression of MDA-7/IL-24 expression. Br J Cancer 2012;106:1976–9.

39. Gr€unewald I, Trautmann M, Busch A, Bauer L, Huss S, Schweinshaupt P,et al. MDM2 and CDK4 amplifications are rare events in salivary ductcarcinomas. Oncotarget 2016;7:75261–72.

40. Isachenko V, Mallmann P, Petrunkina AM, Rahimi G, Nawroth F, HanckeK, et al. Comparison of in vitro- and chorioallantoic membrane (CAM)-culture systems for cryopreserved medulla-contained human ovarian tis-sue. PLoS One 2012;7:e32549.

41. TomaykoMM, Reynolds CP.Determination of subcutaneous tumor size inathymic (nude) mice. Cancer Chemother Pharmacol 1989;24:148–54.

42. Smith TA, Easley KA, Goldblum JR. Myxoid/round cell liposarcoma of theextremities. A clinicopathologic study of 29 cases with particular attentionto extent of round cell liposarcoma. Am J Surg Pathol 1996;20:171–80.

43. de Bruijn DR, Allander SV, van Dijk AH, Willemse MP, Thijssen J, vanGroningen JJ, et al. The synovial-sarcoma-associated SS18-SSX2 fusionprotein induces epigenetic gene (de)regulation. Cancer Res 2006;66:9474–82.

44. Michels S, TrautmannM, Sievers E, Kindler D, Huss S, Renner M, et al. SRCsignaling is crucial in the growth of synovial sarcoma cells. Cancer Res2013;73:2518–28.

45. Khan J, Bittner ML, Saal LH, Teichmann U, Azorsa DO, Gooden GC, et al.cDNA microarrays detect activation of a myogenic transcription programby the PAX3-FKHR fusion oncogene. Proc Natl Acad Sci U S A 1999;96:13264–9.

46. Cironi L, Riggi N, Provero P, Wolf N, Suva ML, Suva D, et al. IGF1 is acommon target gene of Ewing's sarcoma fusion proteins in mesenchymalprogenitor cells. PLoS One 2008;3:e2634.

47. Herrero-Martin D, Osuna D, Ordonez JL, Sevillano V, Martins AS, Mack-intosh C, et al. Stable interference of EWS-FLI1 in an Ewing sarcomacell line impairs IGF-1/IGF-1R signalling and reveals TOPK as a new target.Br J Cancer 2009;101:80–90.

48. Juergens H, Daw NC, Geoerger B, Ferrari S, Villarroel M, Aerts I, et al.Preliminary efficacy of the anti-insulin-like growth factor type 1 receptorantibody figitumumab in patients with refractory Ewing sarcoma. J ClinOncol 2011;29:4534–40.

49. Pappo AS, Patel SR, Crowley J, Reinke DK, Kuenkele KP, Chawla SP, et al.R1507, amonoclonal antibody to the insulin-like growth factor 1 receptor,in patients with recurrent or refractory Ewing sarcoma family of tumors:results of a phase II sarcoma alliance for research through collaborationstudy. J Clin Oncol 2011;29:4541–7.

50. Iams WT, Lovly CM. Molecular pathways: clinical applications and futuredirection of insulin-like growth factor-1 receptor pathway blockade. ClinCancer Res 2015;21:4270–7.


Trautmann et al.




2017;23:6227-6238. Published OnlineFirst June 21, 2017.Clin Cancer Res Marcel Trautmann, Jasmin Menzel, Christian Bertling, et al. Target in Myxoid Liposarcoma

DDIT3 Fusion Protein-Driven IGF-IR Signaling is a Therapeutic−FUS

Updated version

10.1158/1078-0432.CCR-17-0130doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2017/07/28/1078-0432.CCR-17-0130.DC3

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/23/20/6227.full#ref-list-1

This article cites 47 articles, 15 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/23/20/6227.full#related-urls

This article has been cited by 6 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/23/20/6227To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-17-0130

http://clincancerres.aacrjournals.org/content/suppl/2017/07/28/1078-0432.CCR-17-0130.DC3

http://clincancerres.aacrjournals.org/content/23/20/6227.full#ref-list-1

http://clincancerres.aacrjournals.org/content/23/20/6227.full#related-urls

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/23/20/6227


Documents

FUS DDIT3 Fusion Protein-Driven IGF-IR Signaling is a ... · Cancer Therapy: Preclinical FUS–DDIT3 Fusion Protein-Driven IGF-IR Signaling is a Therapeutic Target in Myxoid Liposarcoma