Proteomic comparison of the EWS-FLI1 expressing cells EF with NIH-3T3 and actin remodeling effect of (R/W)9 cell-penetrating peptide

(1)

HAL Id: hal-01365335

https://hal.archives-ouvertes.fr/hal-01365335

Submitted on 8 Dec 2016

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives| 4.0 International License

Proteomic comparison of the EWS-FLI1 expressing cells EF with NIH-3T3 and actin remodeling effect of

(R/W)9 cell-penetrating peptide

Séverine Clavier, Françoise Illien, Sandrine Sagan, Gérard Bolbach, Emmanuelle Sachon

To cite this version:

Séverine Clavier, Françoise Illien, Sandrine Sagan, Gérard Bolbach, Emmanuelle Sachon. Pro- teomic comparison of the EWS-FLI1 expressing cells EF with NIH-3T3 and actin remodeling ef- fect of (R/W)9 cell-penetrating peptide. EuPA Open Proteomics, Elsevier, 2016, 10, pp.1-8.

�10.1016/j.euprot.2015.10.002�. �hal-01365335�

(2)

Proteomic comparison of the EWS-FLI1 expressing cells EF with

NIH-3T3 and actin remodeling effect of (R/W)

9

cell-penetrating peptide

Séverine Clavier

^a,b

, Françoise Illien

^b

, Sandrine Sagan

^b

, Gérard Bolbach

^a,b

, Emmanuelle Sachon

^a,b,

*

aSorbonneUniversité,UPMC-UnivParis6,EcoleNormaleSupérieure,PSLResearchUniversity,DépartementdeChimie,CNRS,UMR7203Laboratoiredes BioMolécules,4PlaceJussieu,ParisCedex05,75252Paris,France

bSorbonneUniversité,UPMC—UnivParis6,PlateformedeSpectrométriedeMasseetProtéomique-IBPS,cc41,7-9QuaiSaintBernard,ParisCedex05,75252 Paris,France

ARTICLE INFO

Articlehistory:

Received1June2015

Receivedinrevisedform25August2015 Accepted25October2015

Availableonline30October2015

Keywords:

EWS-FLI1

Actincytoskeletonremodeling Passivedissemination Cell-penetratingpeptide

SILACquantitativeproteomicapproach

ABSTRACT

EWS-FLI1expressioninNIH-3T3ﬁbroblastshasaprofoundimpactonthephenotype,resultinginthe cytoskeletonandadhesivecapacitydisorganization(EFcells).Besidesthis,(R/W)9,acell-penetrating peptide (CPP),hasanintrinsicactin remodelingactivity inEFcells.Toevaluate theimpactofthe oncogenicproteinEWS-FLI1onproteinsexpressionlevels,aquantitativecomparisonoftumoralEFand non-tumoral3T3 proteomeswasperformed.Thentoseeifwe couldlinktheEWS-FLI1oncogenic transformationtothephenotypereversioninducedby(R/W)9,(R/W)9inﬂuenceonEFcellsproteomewas assessed.Toourknowledgenosuch“CPPomic”studyhasbeenperformedbefore.

Biologicalsigniﬁcance:Uptonowveryfewglobalquantitativeproteomicstudieshavebeenpublishedto helpunderstandtheoncogenictransformationinducedbyEWS-FLI1fusionproteinandleadingtoEwing sarcomadevelopmentanddissemination.Thecomparisonwedidinthisstudybetweenamodeltumoral celllineEFanditsnon-tumoralcounterpart(3T3)allowedustohighlightseveralfeatureseithercommon tomosttumortypesorspeciﬁctoEwingsarcoma.Particularly,lackofactincytoskeletonorganization couldverylikelybeexplainedbythedown-regulationofmanyimportantactinbindingproteins.These resultsareinaccordancewiththehypothesisofapassive/stochasticmodeofdisseminationconferring Ewingsarcomatumoralcellahighmetastaticpotential.

ã2015TheAuthors.PublishedbyElsevierB.V.onbehalfofEuropeanProteomicsAssociation(EuPA).This isanopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/licenses/by-nc-nd/

4.0/).

1.Introduction

Ewingtumors,ﬁrstdescribedbyJamesEwingin1921[1],are thesecondmostfrequenttumortypeandaregenerallydeveloping aroundorinbonesbutcanalsoaffectsofttissues(extra-osseous Ewingsarcoma).James Ewingﬁrstsuggestedan epithelialcells originasEwing’stumorswereshowingsimilaritieswithangio- endothelial tumors [1]. Other origins were further proposed:

hematopoieticcells[2],ﬁbroblasticcells[3]ormesenchymalstem cells[4].Today,thequestionoftheoriginofEwingsarcomatumor cellsremainsopen.

Ewing sarcoma presents a remarkable characteristic: its oncogenesis is generally accepted to be initiated by a simple

genetic event that is a chromosomal translocation between chromosomes11and22whichfusestheEWSgeneofchromosome 22totheFLI1geneofchromosome11.Thet(11;22)chromosomal translocationjuxtaposesthe5⁰ sequencesfromEWSwiththe3⁰ sequence fromamember oftheETS transcriptionfactor family (FLI1genein85%ofcases)[5].FusionofEWStoFLI1geneyieldsan oncoproteinEWS-FLI1.EWS-FLI1actsasatranscriptionalregulator tomodulateexpressionofhundredstothousandsgenes[6].

The oncogenic transformation resulting from EWS- FLI1expressionwasreproducedinaNIH-3T3fibroblastscellline [7]. These NIH-3T3 fibroblasts were stably transformed by the transductionofEWS-FLI1fusiongene,givingbirthtotheso-called EFcellline. EWS-FLI1expressionhasaprofoundimpactoncell phenotypeasitiscausingalossoftheircytoskeletonorganization andadhesivecapacity.Cellcytoskeletonismadeofactinmicro- filaments,intermediatefilamentsandmicrotubules.Actinmicro- filaments result from the polymerization of monomeric actin molecules(actin-G)andarecalledactin-F.Theactincytoskeletonis ahighlydynamicassemblyregulatedbyahugenumberofactin

* Corresponding author at: SorbonneUniversité, UPMC-Univ Paris 6, Ecole NormaleSupérieure, PSL ResearchUniversity, Départementde Chimie,CNRS, UMR7203LaboratoiredesBioMolécules,4PlaceJussieu,ParisCedex05,75252 Paris,France.

E-mailaddress:[email protected](E.Sachon).

http://dx.doi.org/10.1016/j.euprot.2015.10.002

2212-9685/ã2015TheAuthors.PublishedbyElsevierB.V.onbehalfofEuropeanProteomicsAssociation(EuPA).ThisisanopenaccessarticleundertheCCBY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/).

ContentslistsavailableatScienceDirect

EuPA Open Proteomics

j o u r n a lh o m e p ag e : w w w . e l s e vi e r . c o m / l o c a t e / e u p r o t

(3)

bindingproteins(ABP).Uptonowabout150proteinshavebeen showntohaveanactinbindingdomainandtobeabletoinﬂuence its polymerization state [8]. The different classes of proteins involvedintheregulationofactin polymerizationinclude actin monomer sequestration proteins [9] such as thymosin

b

⁴ ^and

profilin, nucleation proteins such as Arp2/3 and WASP family [10,11],cappingproteinssuchasgelsolin[12]aswellasfilament depolymerizingor,onthecontrary,filamentstabilizingproteins suchastropomyosinorcaldesmon[13,14].

Asaresultoftheiractincytoskeletondisorganization,EFcells displaysmall,roundcellmorphology. The strikinglossof actin stress ﬁbers and focal adhesion prevent the formation of protrusions(ﬁlapodia,invadopodia...)commonlyusedforthe spreadingoftumoralcells[15–17].ThisleadChaturvedietal.tothe hypothesisthat,contrary tomosttumoralcells, cellsexpressing EWS-FLI1maydisseminateviaa“passive/stochastic”model[18].

ThisoriginalbehaviorandphenotypeofEwingsarcomatumoral cellsishowevernotyetwellcharacterizedattheproteomelevel.

Itwasparticularlyinterestingtoperformthisproteomestudy as a previous study reported that (R/W)9, a polycationic and amphiphilic cell-penetrating peptide (CPP) of sequence NH2- RRWWRRWRR-CONH2 (amidated at the C-terminus), able to internalizeubiquitouslyineukaryoticcells,displaysactinremod- elingactivityinEFcells[19].Afterafewhoursofincubationwith (R/W)₉peptide,achangeinEFcellsmorphology(alreadyvisible usingan optical microscope) and a significant reappearance of stressfibers were indeed observed by immunofluorescence. In addition, a decrease in EF cells motility was observed by videomicroscopy.Finally, theabilityof EFtumoralcellstogrow withoutanchoragewasalsoshowntobeimpairedinthepresence ofthepeptide.TheseresultsindicateapossiblereversionoftheEF cellstumoralphenotypeinthepresenceof(R/W)9CPP[19].

TohelpﬁllingthegapinproteomicdataforEwingsarcomaand toprogressintheunderstandingof(R/W)9modeofactiononEF tumoralcells,weperformedtwosetsofquantitativedifferential proteomicexperiments.First,wecomparedtumoralEFandnon- tumoral3T3cellstoseehowtheoncogenicproteinEWS-FLI1was affectingproteins expressionlevels. Thentoseeif, in terms of protein expression, we could link the EWS-FLI1 oncogenic transformation to the phenotype reversion induced by (R/W)9

CPP,weassessed(R/W)9inﬂuenceonEFcellsproteome.

The comparisons between EF and 3T3 cells proteomes and betweenEFcells treatedby(R/W)9 anduntreatedEFcells were performedusinga robust globaldifferential proteomicstrategy basedonSILACquantificationtechnique[20].Oneshouldnotice here,thattheCPP(R/W)9usedintheworkofDelarocheetal.(NH2- RRWWRRWRR-CONH2)wasslightlymodifiedinthepresentstudy to incorporate a benzophenone moiety: Biot(O₂)-G₅-K(pBz)- RRWWRRWRR-CONH2 andthereforerenamedphoto(R/W)9.The photoactivablebenzophenoneshouldallowcomplementarystud- iessuchasthesearchofCPPinteractingpartners.Thebiotinshould helppurificationof thesepartners. Boththebiotintag andthe photoactivableprobewillnotbeofinterestforthepresentwork butthesemodificationshaverequireda biologicalvalidation to ensurethatphoto(R/W)9behavedsimilarlyto(R/W)9,inparticular regardingtheactinremodelingactivity.

2.Experimental

2.1.Photo(R/W)9peptidesynthesisandbiologicalvalidation 2.1.1.Peptidesynthesis

Biot(O2)-G5-K(pBz)-RRWWRRWRR-CONH2 (photo(R/W)9)was automaticallysynthesized(Fmocstrategy)aspreviouslypublished [21] with some modifications. Briefly, the biologically active (R/W)9 peptide sequence NH2-RRWWRRWRR-CONH2 was first

synthesized with a peptide synthesizer (Applied Biosystems, Darmstadt,Germany,model433A)usinga solidphase butylox- icarbonyl(Boc)chemistryinthe0.1mmolscale,startingfromap- methylbenzhydrylamine resin(MBHA resin).Standard protocols wereused,withDCC/HOBtactivation(10eq.excessforstandard Bocaminoacids).AftertheremovalofthelastN

a

^-Boc^protecting

group, the synthesis was manually completed by adding the protectedLysineresidue(Boc-Lys(Fmoc)-OH).Theresinwasdried invacuoandseparatedintotwobatchesforthemanualadditionof the5Boc-Gly-OHfollowedbytheadditionofthebiotinylsulfone group.Thisbiotinisoxidizedtobiotinsulfone(biot(O2))toavoid further oxidation during the sample preparation which would dilutetheMSsignal(morespecieswithlowerintensityeach).The biotinmoietyisseparatedfromthe(R/W)9sequenceviaastringof 5Glytoensureflexibilityofthespacerarm.TheLysine(Boc-Arg (Tosyl)-OH)andtryptophan(Boc-Trp(Formyl)-OH)residueswere then deprotectedusing20% piperidin.Then, thecarboxybenzo- phenonewascoupledtothelateralchainofthedeprotectedlysine residue. Finally, both peptidyl-MBHA-resins were treated with liquidHFat0C(2h30)understirring,inthepresenceofanisole anddimethylsulfide.AfterevaporationofHFandofthesolventsin vacuo, theresinswerewashedthree timeswithEt2O andthen subsequentlyextractedthreetimeswith10%AcOH.Lyophilization oftheextractsgavecrudepeptides,whichwerepurifiedbyHPLC andpuritywascheckedbyUV(>95%)andbyMALDI-TOFMS.

2.1.2.Cytotoxicityevaluation

Thecytotoxicityofphoto(R/W)9forEFcellswastestedwitha cell counting kit (CCK-8) commercialized by Dojindo (Dojindo Laboratories,Kumamoto,Japan).20,000cellswereharvestedina 96wellsplate24hbeforethebeginningofthepeptidetreatment.

Cellswereincubatedwithphoto(R/W)9or (R/W)9withconcen- trationsrangefrom1to30

m

^M.^The^percentage^of^cell^viability^is

directly related to the absorbance at 450nm (Fluostar optima microplatereader,BMGlabtech,Offenburg,Germany)oftheWST- 8 formazan solutionand was evaluatedafter2h, 3h30 and 5h incubation with peptides. Triplicates were performed for each condition (concentration and type of peptide), which were repeatedindependentlyatleasttwice.

2.1.3.Internalizationmeasurements

After a 75min incubation at 37C with a extracellular concentrationof 5 or 7.5

m

^M, internalizationcapacity of photo (R/W)9 was quantiﬁed witha mass spectrometryquantiﬁcation protocolusingthedeuteratedversionofthephoto(R/W)9asthe internalstandardaspreviouslydescribed[22].

2.1.4.Biologicalactivity

Theactinremodelingactivityofphoto(R/W)9wasassessedby immunoﬂuorescence as previously described [19]. 10,000 cells wereharvestedonthin12mmglassslidescoatedwithgelatin24h beforetreatmentwiththepeptide.Thesecellsweresubmittedtoa 16hincubationwith5

m

^M^photo(R/W)⁹^.^Cells^were^then^washed

withPBS, ﬁxedwithparaformaldehyde 3%,permeabilized with TritonX-1000.4%andincubatedwithphalloïdinFITCandDAPIto labelactinstressﬁbers andcellnuclei(chromatin)respectively.

Cells were observed using ﬂuorescence microscopy (Eclipse, TE2000-S,Nikon,Tokyo,Japan).

2.2.3T3andEFcelllineproduction

3T3cellsareembryonicmouseﬁbroblasts.EFcellscorrespond toamonoclonalcelllineobtainedbytransfectionof3T3bythe cDNAencodingtheproteinEWS-FLI1undertheinﬂuenceofthe LTRinternalpromoterofthemurineleukemiaMoloneyvirusina retroviralvectorp-BABE-purocontainingEWS-FLI1cDNAand a

2 S.Clavieretal./EuPAOpenProteomics10(2016)1–8

(4)

puromycin resistant gene allowing for the selection of the transformed cells. The cell line NIH 3T3 transformed by the EWS-FLI1 oncoprotein (EF cells) as well as the corresponding controlcelllinetransfectedwithanemptyvector(3T3cells)have beenprovidedbyDr.J.Ghysdaël(InstitutCurie/CNRSUMR3306/

INSERMU1005).

3T3andEFcellswereculturedinDulbecco’smodiﬁedEagle’s mediumsupplementedwith10%newborncalfserum(Invitrogen), penicillin (100,000IU/l), and streptomycin (100,000IU/l) (PAA Laboratories)andselectedwith25

m

^g/ml^puromycin^(Sigma).

2.3.Stableisotopelabelingwithaminoacidsincellculture(SILAC) EFand3T3cellsweregrowninA14431-DMEM(highglucose, noglutamine,nolysine,noarginine)mediawith10%dialyzedNCS (new-born calfserum),1%penicillin-streptomycin, 4mM gluta- mineandeither146mg/lL-lysine-2HCland84mg/lL-arginine-HCl, or146mg/l¹³C6L-lysine-2HCland84mg/l¹³C615

N4L-arginine-HCl forsixpassages,andincorporationefﬁciencywasdeterminedby massspectrometricanalysis.AllculturesmaterialswerefromLife Technologiesandamino-acids (isotopicallylabeledornot)were fromPierce(BothThermoFisherScientiﬁcbrands,Waltham,MA, USA).

2.4.Samplepreparation

Cellsweregrownin10cmplatesmakingsurethatthetwocell populationswereatthesamedensityandthatbothwereatabout 80%conﬂuenceonthesamplepreparationday.EFcellstreatment withphoto(R/W)9CPPwasperformedduringthenightpreceding thesamplepreparationadding25

m

^lôfâ¹^mM^solutionôf^photo

(R/W)9CPPinordertohavea5

m

^Mﬁnalconcentrationinthe5ml of culture medium. Cells were washed, trypsinized, counted (glassticslide10withgrids,HycorBiomedical,Indianapolis,IN, USA)andmixedataratioof1:1(2.10⁶cellsofeachcondition).Cells werewashed,spinneddownandlysedusing100

m

^l^of^lysis^buffer

(100mMTris–HCl,100mMDTT, 4%SDS).Protein concentration wasmeasuredusingaBCAassayfromPierce.100

m

^g^of^proteins

were separated by 1D-SDS polyacrylamide gel electrophoresis (10%)and stainedwithCoomassieblue.Thegelwas slicedinto 10gelbandspriortoreduction,alkylationandovernighttrypsin digestionat37C(1:30(w:w)protease-to-proteinratio).

2.5.NanoLC–ESI–MS/MSanalysis

Protein digests were analyzed by nano-LC (Ultimate 3000, Dionex,ThermoFisherScientiﬁc)coupledtoanESI-LTQ-Orbitrap (LTQ Orbitrap XL, Thermo Scientiﬁc, Bremen, Germany) mass spectrometer.Trypticpeptideswereinjectedbytheautosampler and concentrated ona trapping column (Pepmap,C18, 300

m

^m

5mm,5

m

^m,¹⁰⁰^Å,^Dionex)^with^water^containing^2%^acetoni-

trile (ACN) and 0.1% formic acid (solvent A). After 10min, the peptideswereelutedontotheseparation column(Pepmap,C18, 75

m

^m500mm, 2

m

^m ¹⁰⁰^Å, ^Dionex) equilibrated with 98%

solventA.Peptideswereseparatedwithagradient0–50min2– 40% solventB(98% ACN+0.1% formic acid), 50–60min 40–60%

solventB,and60–70min60%solventBataﬂowrateof200nl/min.

TheLTQ-OrbitrapmassspectrometerisoutﬁttedwithananoESI interface.Electrosprayemitterswere360/20

m

^m^o.d.10

m

^m^i.d.

fused-silicatips(PicoTipEmitter,StandardCoatedSilicaTip,New Objective,Woburn,MA,USA).Theheatedcapillary temperature andsprayvoltagewere200C and1.5kV,respectively.Orbitrap spectra(automatedgaincontrol(AGC)2.10⁵)werecollectedfrom m/z 500–2000 at a resolution of 30,000 in the proﬁle mode followedbydatadependentsequentialCIDMS/MSspectraofthe tenmostintenseionswithanormalizedenergyof35.Adynamic

exclusiontimeof60swasusedtodiscriminateagainstpreviously analyzedions.

2.6.Datatreatment,statisticalanalysisandinterpretation

ThepeptideswereidentifiedandquantifiedusingtheProteome Discoverer 1.3software (ThermoScientific)withcarbamidome- thylation(C),oxidation(M),¹³C6(K)and¹³C615N4(R)asvariable modifications.Thedatabaseusedforproteinidentificationisthe UniprotproteindatabaseforthetaxonomyMusmusculus(down- loaded on http://www.uniprot.org/, 30th October 2013, 74,249 entries). False Discovery Rate (FDR) calculated at the peptidelevelwithadecoydatabase(reversedpeptidessequences) wassetto1%,MStolerancewas10ppmandMS/MStolerancewas 0.8Da.Proteinratioswerecorrectedusingthemedianproteinratio normalization.

Thestatisticalanalysisoftheproteinslistsobtainedforeach experiment was performed with MyProMS software [23]. The thresholdvaluesforabsolutefoldchangeinproteinexpressionand p-valuefortheratiocalculationweredeterminedwithacontrol experiment mixing identical EF cells grown in light and heavy mediumataratioof1:1.Proteinswerefoundtobesigniﬁcantly over-orunder-expressedfromanabsolutefoldchangeof1.3anda p-value<0.05. Three biological replicates of the experiment comparing (1) EF versus 3T3 and (2) EF treated with photo (R/W)9versusEFuntreatedwereperformed.

Only proteins with a ratio corresponding to a fold change greaterthan1.5,withap-value<0.05andquantiﬁedinatleasttwo ofthethreereplicateswithatleasttwopeptideswerekeptfordata interpretation.

Analysis of the protein networks for signiﬁcantly over- or under-expressed proteins was performed using the STRING softwareavailableonline(http://string-db.org/)[24,25].

3.Resultsanddiscussion

3.1.Biologicalvalidationoftheactinremodelingactivityof photo(R/W)₉peptideintumoralEFcells

Cytotoxicityassayswithphoto(R/W)9onEFcellshaveshown thatthisCPPaffectsthecellviabilityabove7.5

m

^Mextracellular concentration(Fig.S1-A).Photo(R/W)9ismorecytotoxicthan(R/

W)9forwhichnocellviabilitydecreasewasobserveduntil30

m

^M

(not shown)[19]. However it should be noted that this higher cytotoxicity can very likely be explained by a much higher internalizationcapacity,aboutsixtimesmoreefﬁcientforphoto (R/W)9 compared to(R/W)9 (3pmolversus 0.5pmol,respectively at 5

m

^M extracellular concentration) (Fig. S1-B). For the comparisonofEFcellstreatedwithphoto(R/W)9CPPuntreatedEF cells, a 5

m

^M extracellular concentration was further used to preventanycytotoxicityeffect.(Forphoto(R/W)₉thiscorresponds toabout2.10⁶CPPmoleculesinternalizedineachcell,takinginto accountanestimatedcellvolumeof1.5pl[22]).Finallyimmuno- ﬂuorescence experiments indicated that the addition of the photoprobe and thebiotin tagto the(R/W)9 sequencedid not affectitsactinremodelingactivity.Photo(R/W)9thusprovedtobe a suitable model to investigate the actin remodeling effect previouslyobservedfor(R/W)9[19](Fig.S2).

3.2.Comparisonoftumoral(EF)andnon-tumoral(3T3)cell proteomes,usingaSILACexperiment

EF cells have beenlabeledwithheavy Argand Lys whereas 3T3cellsweregrownwithlightamino-acids.Quantificationofthe proteinlevels(heavy/light)wasdoneusingthesoftwareProteome Discoverer 1.3(Thermo FisherScientific)(Supplementalfile S9)

(5)

andthestatisticalanalysisofthelistsofunder-orover-expressed proteins obtained for each quantification experiment was performed using MyProMS software [23]. Based on the filtering criteriadeterminedwithcontrolexperiments(foldchange>1.5,p- value<0.05andpresenceinatleasttwoofthethreereplicates experiments)wewereabletoobtainalistof94proteins(among 1700proteinsquantified)withasignificantvariationofexpression levelbetweenEFand3T3.Amongthese94proteins,43wereover- expressedand 51under-expressed inEF compared to3T3. The volcano plot obtained for the comparison of EF versus 3T3 proteomeis presentedin Fig.1 whilethelists ofproteins with significantlymodifiedexpressionlevelsarepresentedinFig.3-A and-B.

ThesedifferencesarecompiledinFig.2whichsummarizesthe resultsobtainedand describedin detailsinthefollowing para- graphs.Thecommonfeaturesoftumoralcellsarehighlightedas wellasfeaturesspeciﬁctoEwingsarcomatumors.

3.3.Proteinssigniﬁcantlyover-expressedinEFcellscomparedto 3T3cells

3.3.1.Glycolysisincrease

For the upregulated proteins network, the most densely populatedinteractionnodecorresponds toproteins involvedin glycolyticprocesses(Fig.S4).

This increase of glycolysis is a quasi-universal property for primarytumorormetastaticcells[26,27].Actually,inadditionto theATPnecessaryforhomeostaticcellactivity,cancercellsneed supplementalenergytosustaintheirrapidgrowth.Moreoverithas beenshown that tumoral cells, insteadof producing ATPvia a classicalaerobicandefﬁcienttwo-stepsprocessinvolvingglycol- ysisandtheKrebscycle(36ATPmoleculesproducedperglucose molecule),areperforminganaerobicglycolysis(10ATPmolecules

produced per glucose molecule). This metabolic phenotype is knownastheWarburgeffect[28].

Therefore, to compensate this energy loss and to ensure a greaterandfasterATPproduction,tumoralcellshavetoconsumea largerquantity of glucose, which explainsthe up-regulationof severalproteinsinvolvedintheglycolysisprocess.

3.3.2.Increaseofbiosyntheticprocesses

Amongthe43 signiﬁcantlyup-regulatedproteins(Fig.S3-A), 13aredirectlyinvolvedinbiosyntheticprocesses(monosacchar- ides, nucleotides, amino acids biosynthesis pathways...) (Fig. S5). Tumoral cells which are characterized by a greater growthrateactuallyneedmore“buildingblocks”tosustaintheir proliferation [29]. Increase in biosynthetic processes requiring highamountofenergyisthusdirectlylinkedwiththeincreased glycolysis.

3.3.3.Oncogenicproteins

Onecanalsonoticetheup-regulationofproteinsknowntobe involved in oncogenic processes such as moesin whose over- expressionwas found necessaryto theepithelial-mesenchymal transition(EMT)oftumoralcellsleadingtometastasisformation [30].Anotherexampleistheinsulinlikegrowthfactor2(IGF-2),a mitogenicpeptideof7.5kDaover-expressedinseveralcancersand generallyassociatedwithapoordiagnosis[31,32].Theproliferat- ing cell nuclear antigen (PCNA) which is a marker of cell proliferation [33] and the Y-box protein 1 also known as nuclease-sensitiveelement-bindingprotein1,amarkeroftumor aggressiveness [34] are also interesting markers to distinguish tumoralEFfromnon-tumoral3T3cells.

3.3.4.Up-regulationofanactinstressﬁberdissolutingprotein TheproteinSTE20likeserine/threonineproteinkinase(SLK) playsaroleinremodelingofactincytoskeleton[35].Itwasshown

Fig.1.VolcanoplotobtainedusingMyProMSsoftware[21]forthecomparisonoftumoralcellsEF(heavy=H)andtheirnon-tumoralcounterparts3T3(light=L).Eachspot correspondstoaprotein.Thethreecolours,orange,pinkandblackarereferringtothethreebiologicalreplicates.(H)and(L)indicatethatthecellshavebeengrownwith heavylabelledaminoacids(¹³C615N4 L-arginineand¹³C6 L-lysine)orclassicalaminoacids,respectively.ThexaxiscorrespondstotheH/Lratioofquantiﬁedproteinsinlog2

scalewhiletheyaxiscorrespondstothep-valuefortheproteinH/LratiodeterminedusingthedifferentH/Lpeptidesratiosforthisprotein.Verticalgreenlinescorrespondto the1.5foldchangeinproteinexpressionlevel.Thehorizontalredlinecorrespondstoap-value=0.05.Thesethresholdvalueshavebeendeterminedwithcontrolexperiments (notshown)forwhichnoproteinwasfoundwithap-value<0.05andafoldchangeabove1.3.Outsidethezonedelimitedbytheselinesareproteinswithasigniﬁcantly modiﬁedexpressionbetweenEFand3T3cells:ontheleftareproteinsunder-expressedintumoralEFcellsandontherighttheover-expressedones.

(6)

thatover-expressionofthisproteincausestheretractionofcells fromtheirsubstrateandthusinhibitstheiractivemigrationalong this surface. This decreased adhesion is likely due to the dissolution of actin stress ﬁbers by SLK [36]. The SLK over- expression observed in EF compared to 3T3 is thus in good agreementwiththe“passive/stochastic”wayofdisseminationof cellsexpressingEWS-FLI-1describedbyChaturvedi[18].

3.4.Proteinssigniﬁcantlyunder-expressedinEFcellscomparedto 3T3cells

3.4.1.Decreaseoftheoxidation-reductionandrespirationprocesses Among the 51 under-expressed proteins, 11 are directly involvedinoxidation-reductionprocesses(Fig.S6).Tumoralcells comparedtonormalcellsareunderoxidativestressconditionsdue toanincreasedmetabolic activityneglectingmitochondriaand respirationprocesses[37], asalready described in theprevious section.Thisoxidativestressischaracterizedbyanincreasedlevel ofreactiveoxygenspecies(ROS).ROSincludefreeradicalssuchas superoxides(O2),hydroxylradicals(HO)aswellasnon-radical moleculessuchasH2O2.

It was shown that this increase in ROS level could have numerousbeneﬁcialconsequencesforthetumoralcells.Actuallyit appears that ROSstimulate cell proliferation and contribute to mutationsand geneticinstabilities responsible for cancer drug treatmentresistance[38].However,themechanismsresultingin oxidativestressobservedinnumerouscancersarenotyetprecisely understood.ItisespeciallydifﬁculttoassessthethresholdofROS concentration necessary to maintain a favorable situation for tumoralproliferationandonethreateningcellsurvival.

Expressionlevelmodiﬁcationsofproteins involvedinoxidation-reductionprocessesareacommonfeaturefortumoralcells.

[Cu–Zn] and [Mn] superoxide dismutases (SOD), glutathione-S- transferaseMu1andMu2,cytoplasmicandmitochondrialNADPH isocitratedeshydrogenasesorperoxiredoxin-2areproteinswith anti-oxidantspropertiesweidentiﬁedwithsigniﬁcantlydecreased expressionlevelsinEFcells(Fig.S3-B).Inamajorityofcancers, expressionlevelsofantioxidantproteinssuchasSODsarerather increased, likely to prevent cells from reaching a cytotoxic concentrationofROS[39,40].Howeverinothertypesofcancer, expressionlevelsof SOD are decreased [41,42] as weobserved herein.ThissuggeststhatthewaycancercellsregulatetheirROS levelisvaryingfromonecelltypetoanother,andthatinEwing sarcomathelevelofROSnecessarytopreservethecancercellsis

reachedbydecreasingtheexpressionlevelofproteinswithanti- oxidantproperties.

Onthecontrary,itisinterestingtonoticethatthehypoxiaup- regulated protein 1 is signiﬁcantly over-expressed in EF cells compared to 3T3 indicating that EF cells are under hypoxic conditions.Severalstudies[26,43]haveshownthattumoralcells canadapttolowoxygenconditions,asituationtheyhavetofaceto developininvadedtissuesbeforethesetupofnewbloodvessels (angiogenesis).

3.4.2.Down-regulationofactinbindingproteins

Actincytoskeletonhasafundamentalroleinthecelllifeand cycle (cell division, migration, signals transmission...) and mechanisms regulatingactin dynamics rely ona large number ofactinbindingproteins.Theexpressionlevelsoftheseproteins areusuallymodiﬁedduringoncogenesis[44,45].Howevertoour knowledge, these modiﬁcations in actin binding protein levels have not yetbeendescribed in details in theliterature for the oncogenictransformationcausedbyEWS-FLI1.

A careful analysis of the list of under-expressed proteins (Fig.S3-B)allowedustoidentifyatotalof12proteinsknowntobe involvedinactindynamics.TheseproteinsaregatheredinTable1 aswellastheirexpressionlevelvariationsandbiologicalprocesses implications[13,14,46–52].

Withtheexceptionofdestrinandgelsolincontributingtothe maintenanceofamonomericactinpool[48,50],alltheotheractin bindingproteinsidentifiedwithdecreasedexpressionlevelsinEF cells compared to 3T3 play a role in the stabilization of actin filaments(tropomyosin,caldesmon)[13,14],intheirassemblyinto network orbundles (filamin,prelaminA/C, laminB1,myosinII subunit,calponin-3)[47,49,51,52]or inthemembraneadhesion (LASP-1)[46].

Most of these proteins are linked to “actin ﬁlament based process” gene ontology (GO) term [53] that is signiﬁcantly enrichedforthislistofproteinsasdepictedinFig.3.

ConsequentlythiscomparisonoftumoralEFwithnon-tumoral 3T3cellsindicatesthattheoncogenictransformationinducedby the presence of the fusion protein EWS-FLI1 is leading to a signiﬁcant decrease of several key actin binding proteins. This resultisagainstwhatiscurrentlyseenfortumoralcellswhichtend to reinforce their actin network in order to disseminate by migrationviatheformationofﬁlipodia,lamellipodia,etc.[15–17].

Nonetheless,thisconclusionisingoodagreementwithChaturvedi etal.study[3]whohaveevidencedanoriginalbehaviorofEwing Fig.2. SchemesummarizingkeyfeaturesofEFcells(anEwingsarcomamodelcellline)evidencedbyproteinsexpressionlevelvariationsexistingbetweentumoralEFand non-tumoral3T3cells.ThesefeaturesareeithercommontoothertumortypesorspeciﬁctoEwingsarcomacells.

(7)

sarcomacellswithapassive/stochasticdisseminationratherthan anactivemigration.Thus, EFcell phenotypecharacterized bya disorganizedactincytoskeletonlackingstressﬁbers,bya round morphologyandtheabsenceoffocaladhesionpoint appearsto result from the down-regulation of several key actin binding proteinsastheonesweidentiﬁedinthisstudy.

3.5.Assessmentofphoto(R/W)9treatmentonEFcellsproteome Aspreviouslymentionedintheintroductionsection,thecell- penetratingpeptide(CPP)(R/W)₉hasbeenreported[19]toreverse thetumoralphenotype of theEF cells to3T3 non-tumoral-like phenotypewiththereappearanceofactinstressﬁbersinEFcells aftera fewhoursincubationwiththisCPP.In ordertoexamine whether we can correlate the proteomic differences found

betweenEFand3T3cellswiththephenotypechangesobserved byimmunoﬂuorescence,wecomparedtheproteomeof EFcells grown in a light medium treated overnight with photo(R/W)9

(5

m

^M)^to^theôneôfûntreatedÊF^cells^grown^with^heavyÂrgând

Lysmedium.

AsforthecomparisonofEF versus3T3cells,threebiological replicateswerestudiedtoassesstheeffectofphoto(R/W)9CPPon theEFcellsproteome(SupplementalfileS10).Againthethreshold valuesfortheabsolutefoldchangeandp-valuedeterminedinthe controlexperimentwereused.Howeverinthiscase,amongthe 1745 proteins identified (1% FDR) and quantified (2 peptides minimum),nonewasfoundwithanexpressionlevelsignificantly andreproduciblymodified(Fig.S8).

Therefore,thestrongphenotypechangeofEFcellsobservedby immunoﬂuorescence (Fig. S2) and also simply with an optical

Fig.3. Interactionnetworkgeneratedwiththelistofsigniﬁcantlyover-expressedproteinsinEFversusT3usingtheopensourceSTRINGsoftware[24,25].Spherescolouredin redarecorrespondingtoproteinsknowntobeinvolvedinactindynamicsandgatheredundertheactinﬁlament-basedprocessGOtermwhichisoneofthemostlyenriched terminthislist(p-value1.10³).

Table1

Listofactinbindingproteins(ABP)identiﬁedamongthelistofproteinsunder-expressedintumoralEF(H)cellscomparedtonon-tumoral3T3(L)cells.Thesecondcolumn indicatestheaverageH/Lratioandtheassociatedcoefﬁcientofvariation(CVinpercentage)obtainedforthethreebiologicalreplicates(FigureS7givesdetailsaboutthe differentH/Lvaluesobtained).Thethirdcolumndescribesthefunctionoftheseproteinsonactincytoskeletondynamics.

Actinbindingproteins(ABP’s) AverageH/Lratio andassociated CV

Function References

LIMandSH3domainprotein1(LASP-1) 1/1.9(4%) Regulationofcellularfunctionsassociatedwithactincytoskeletonreorganization atthemembrane.Partoffocaladhesionpointsandassociatedwithzyxin

[46]

Filamin(Flna) 1/2.1(16%) Actinﬁlamentscross-linkingandorthogonalactinnetworksbuildingblocks [47]

Destrin(Dstn)(orActindepolymerizingfactorADF) 1/2.1(15%) Sequestrationofactinﬁlamentsandbindingtoactinmonomers [48]

PrelaminA/C 1/2.4(5%) Roleinactinbundling [49]

LaminB1 1/1.5(4%) Roleinactinbundling [49]

Gelsolin(Gsn) 1/2.6(18%) Actinﬁlamentscappingandmonomerssequestration [50]

Myosin(myosinregulatorylightchain12B (Myl12b)+myosin-9(Myh9)+myosinlight polypeptide6(My16))

1/1.5(7%) 1/1.5(4%) 1/1.5(3%)

PartofMyosinIIcomplexlinkedtoactinandpivotalroleincellularadhesion, migrationanddivision.Down-regulationofmyosinregulatorylightchainsAorB (MYL12B/12A)inducesimportantcellmorphologychangesanddisappearanceof actinstressﬁbers

[51]

Tropomyosin 1/1.5(15%) Lateralstabilizationofactinﬁlaments [13]

Caldesmon(Cald1) 1/1.9(24%) Lateralstabilizationofactinﬁlaments [14]

Calponin-3(Cnn-3) 1/2.2(22%) Roleinactinstressﬁbersformation [52]

(8)

microscopeafter18htreatmentwithphoto(R/W)9peptide,does notresultfromimportantmodiﬁcationsoftheEFcellproteome.

In SILACbased quantitative proteomicexperimentswehave access to the relative expression levels quantification for a significantpartofthecellproteins.Howeveroneshouldkeepin mindthatevenwitheffortsmadeinthefractionationofthecell lysatesaswellasin theMSanalysiswearestill limitedin the numberofproteinsweareabletoidentify,notablybecauseofthe greatdynamicrangeofproteinexpressionlevels.Consequentlywe aremostlylookingattheproteinsproducedinlargequantitiesby thecells.Inaddition,evenwithisotopelabeling,systematicerror inpeptiderelativequantificationduetobiologyvariabilityledusto onlyconsider importantexpressionchanges (>1.5fold change).

The energetic cost for cells to increase the production of an abundantproteinby50%canbemoreimportantthantheoneto doubleortriplethequantityoflessabundantproteins.Unfortu- natelytheseweaklyabundantproteinsremaindifﬁculttodetectin globalproteomicapproaches.Beyondtheselimitationsofprotein abundance and range of expression variations, it has been demonstrated that PTMs are crucial and can quickly boost a proteinactivity[54] at constantexpressionleveland mayeven compensate a decreasein expressionlevel. Consequentlyan in depth study of PTMs such as phosphorylation might help understandingphoto(R/W)9modeofaction.

Another important option to consider in order to explain phenotypechangesatconstantproteinlevelsisproteinorganiza- tionin thecells. Protein-protein interactionsplay a keyrole in regulatingsignalingpathwaysandthuscellularfunctions.At5

m

^M

extracellularconcentrationabout2.10⁶photo(R/W)9moleculesare internalized per cell. Although the number of photo(R/W)9

moleculesinternalizedis abouttwo ordersofmagnitude lower than the number of actin monomers (estimated to be 5.10⁸ moleculespercell)[56],photo(R/W)9CPPmoleculescoulddisrupt some actin related protein complexes or stabilize others. The presenceof photo(R/W)9 moleculescouldthus compensate the down-regulation of key actin binding proteins evidenced in EF versus3T3andleadtoactinstressﬁbersformationthroughdirect interaction with actin [55] and to subsequent morphological changes.

4.Conclusions

Thegapbetweengenomeandphenotypehasbeenevidenced alreadyawhileagoandproteomicswasshowntobeanextremely usefultooltohelpﬁllingthisgap[56].Thequantitativeproteomic experiments realized to compare tumoral EF and non-tumoral 3T3actuallyhelpustogaininsightinthemolecularchangesdrove by EWS-FLI1 oncogene and offer a set of proteomic data to biologistsworkingonEwingsarcoma.Wecanalreadysayfrom thesedatathatEFcellsexpressingEWS-FLI1proteinexhibitsome characteristicfeaturesoftumoralcellattheproteomelevelsuchas metabolicchanges (bioenergetics,biosynthesis,redoxstatus).In addition, the down-regulation of several key actin-binding proteinsdistinguishesEwingsarcomacellsfromthegreatmajority of tumoral cells and is here proposed to explain the peculiar phenotype of EF cells as well as their particular mode of dissemination.

Thesecondobjectiveofthestudy,thatwastounderstandthe photo(R/W)9modeofactiononEFcells,revealedthataphenotype change can not necessarily easily be seen using a global quantitative proteomicapproach.Moreover this studyindicates that even if proteomics has allowed a big step forwardin the correlationbetweenamolecularperturbationandtheassociated phenotypeitisstillnotstraightforwardtolinkaphenotypechange and proteome. Especially, withthese experimentswe saw that similarphenotypescancorrespondtoclearlydifferentproteomes

andthatadrasticphenotypechangeisnotnecessarilyrelatedto importantchangesinproteinsexpressionlevels.

Conﬂictofinterest None.

Associatedcontent

Supporting Information Available: eight additional ﬁgures (S1–S8)arepresentedin theelectronicsupplementarymaterial.

ExcelspreadsheetsexportedfromP.D1.3softwareandcontaining theidentiﬁcationandquantiﬁcationdataforEFcomparisonwith 3T3 (3biological replicates) andfor EF treatedby photo(R/W)9

comparison withuntreated EF (3 biological replicates)are also provided(S9andS10).

Acknowledgements

TheauthorswishtothankJ.GhysdaëlforthekindgiftofEFand 3T3celllines,R.Marquantforpeptidesynthesisandpuriﬁcation, as well asP. Poullet and G. Arras who provided theirhelp for MyProMS software installation and training. The authors also thankG.Clodicforhelpandtechnicalsupport.

AppendixA.Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound, in theonline version, at http://dx.doi.org/10.1016/j.euprot.2015.

10.002.

References

[1]J.Ewing,Classicsinoncology.Diffuseendotheliomaofbone,in:JamesEwing (Ed.),ProceedingsoftheNewYorkPathologicalSociety,1921.CACancerJ.Clin.

22(1972)95–98.

[2]M.E.Kadin,K.G.Bensch,OntheoriginofEwing’stumor,Cancer27(1971) 257–273.

[3]P.S.Dickman,L.A.Liotta,T.J.Triche,Ewing’ssarcoma:characterizationin establishedculturesandevidenceofitshistogenesis,Lab.Investig.J.Tech.

MethodsPathol.47(1982)375–382.

[4]J.J.Navas-Palacios,R.Aparicio-Duque,M.D.Valdés,Onthehistogenesisof Ewing’ssarcoma.Anultrastructural,immunohistochemical,andcytochemical study,Cancer53(1984)1882–1901.

[5]O.Delattre,J.Zucman,B.Plougastel,C.Desmaze,T.Melot,M.Peter,etal.,Gene fusionwithanETSDNA-bindingdomaincausedbychromosometranslocation inhumantumours,Nature359(1992)162–165,doi:http://dx.doi.org/10.1038/

359162a0.

[6]A.Prieur,F.Tirode,P.Cohen,O.Delattre,EWS/FLI-1silencingandgene proﬁlingofEwingcellsrevealdownstreamoncogenicpathwaysandacrucial roleforrepressionofinsulin-likegrowthfactorbindingprotein3,Mol.Cell.

Biol.24(2004)7275–7283,doi:http://dx.doi.org/10.1128/mcb.24.16.7275- 7283.2004.

[7]W.A.May,M.L.Gishizky,S.L.Lessnick,L.B.Lunsford,B.C.Lewis,O.Delattre, etal.,Ewingsarcoma11;22translocationproducesachimerictranscription factorthatrequirestheDNA-bindingdomainencodedbyFLI1for transformation,Proc.Natl.Acad.Sci.U.S.A.90(1993)5752–5756.

[8]C.G.Remedios,C.dos,D.hhabra,M.Kekic,I.V.Dedova,M.Tsubakihara,D.A.

Berry,etal.,Actinbindingproteins:regulationofcytoskeletalmicroﬁlaments, Physiol.Rev.83(2003)433–473,doi:http://dx.doi.org/10.1152/

physrev.00026.2002.

[9]L.G.Tilney,S.Hatano,H.Ishikawa,M.S.Mooseker,Thepolymerizationofactin:

itsroleinthegenerationoftheacrosomalprocessofcertainechinoderm sperm,J.CellBiol.59(1973)109–126.

[10]M.D.Welch,J.Rosenblatt,J.Skoble,D.A.Portnoy,T.J.Mitchison,Interactionof humanArp2/3complexandtheListeriamonocytogenesActAproteininactin ﬁlamentnucleation,Science281(1998)105–108.

[11]R.D.Mullins,J.A.Heuser,T.D.Pollard,TheinteractionofArp2/3complexwith actin:nucleation,highafﬁnitypointedendcapping,andformationof branchingnetworksofﬁlaments,Proc.Natl.Acad.Sci.U.S.A.95(1998) 6181–6186.

[12]J.A.Cooper,T.D.Pollard,Effectofcappingproteinonthekineticsofactin polymerization,Biochem.(Mosc.)24(1985)793–799.

[13]B.W.Bernstein,J.R.Bamburg,TropomyosinbindingtoF-actinprotectstheF- actinfromdisassemblybybrainactin-depolymerizingfactor(ADF),CellMotil.

2(1982)1–8.