In vivo imaging of DNA lipid nanocapsules after systemic administration in a melanoma mouse model

ContentslistsavailableatScienceDirect

International Journal of Pharmaceutics

j o u r n al ho m e p age :w w w . e l s e v i e r . c o m / l o c a t e / i j p h a r m

In vivo imaging of DNA lipid nanocapsules after systemic administration in a melanoma mouse model

Stephanie David

, Nathalie Carmoy

, Pauline Resnier

, Caroline Denis

, Laurent Misery

, Bruno Pitard

, Jean-Pierre Benoit

, Catherine Passirani

, Tristan Montier

aLUNAMUniversité–IngénieriedelaVectorisationParticulaire,F-49933Angers,France

bINSERMU646,F-49933Angers,France

cINSERMUMR915–UniversitédeNantes,8quaiMoncousu,F-44000Nantes,France

dINSERMU613–UniversitédeBrest,5avenueduMaréchalFoch,F-29200Brest,France

eLaboratoiredeNeurobiologiedelaPeau,ServicedeDermatologie,CHUdeBrest,F-29609Brest,France

a r t i c l e i n f o

Articlehistory:

Received16February2011

Receivedinrevisedform15June2011 Accepted17June2011

Available online 24 June 2011

Keywords:

Skincancer DNALNCs Real-timeimaging pHSV-tk Ganciclovir Intravenousinjection

a b s t r a c t

ThebiodistributionofintravenouslyinjectedDNAlipidnanocapsules(DNALNCs),encapsulatingpHSV-tk, wasanalysedbyinvivoimagingonanorthotopicmelanomamousemodelandbyasubsequenttreatment withganciclovir(GCV),usingthegene-directedenzymeprodrugtherapy(GDEPT)approach.Luminescent melanomacells,implantedsubcutaneouslyintherightflankofthemice,allowedustofollowtumour growthandtumourlocalisationwithinvivobioluminescenceimaging(BLI).Inparallel,DNALNCsor PEGDNALNCs(DNALNCsrecoveredwithPEG2000)encapsulatingafluorescentprobe,DiD,allowedus tofollowtheirbiodistributionwithinvivobiofluorescenceimaging(BFI).TheBF-imagesconfirmeda prolongedcirculation-timeforPEGDNALNCsaswaspreviouslyobservedonanectotopicmodelof glioma;comparisonwithBL-imagesevidencedthecolocalisationofPEGDNALNCsandmelanomacells.

Afterthesepromisingresults,treatmentwithPEGDNALNCsandGCVonafewanimalswasperformed andthetreatmentefficacymeasuredbyBLI.Thefirstresultsshowedtumourgrowthreductiontendency and,onceoptimised,thistherapystrategycouldbecomeanewoptionformelanomatreatment.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Inhomeostaticconditions,melanocytesproducemelaninand thuscontributetothepigmentationofskinandhair,protectthe skinfromdamagebyultravioletradiation,andpreventskincancer (LinandFisher,2007).Buttheyarealsoprecursorsofmelanoma, themostdeadlyformofskincancer,followingmutationsofcrit- icalgrowthregulatorygenes,theproductionofautocrinegrowth factors,and/oralossofadhesionreceptors(Gray-Schopferetal., 2007).Theprimarysiteofmelanomaistheskin,butotherpig- mentedtissues,suchastheeyesortheintestine,canalsobeat theoriginofmelanoma.Thetumour-specific,10-yearsurvivalfor melanomais75–85%whendiagnosedasprimarytumourswithout anyevidenceofmetastasis,butmelanomametastasisdramatically reducesthispercentageto20–70%,thisbeingdependentonthe

∗Correspondingauthorat:INSERMU646,IBS-CHU,4rueLarrey,49933Angers Cedex9,France.Tel.:+33244688534;fax:+33244688546.

∗∗Correspondingauthorat:INSERMU613,HopitalMorvan–CHUBrest,5avenue duMaréchalFoch,29200Brest,France.Tel:+33298018080;fax:+33298467910.

E-mailaddresses:catherine.passirani@univ-angers.fr(C.Passirani), Tristan.Montier@univ-brest.fr(T.Montier).

metastasistype(Garbeetal.,2010).Ifdiagnosedearly,localsurgical resectioncancuremelanomain80%ofcases,contrarytometastatic melanomawhichislargelyrefractorytoexistingtherapies(Tawbi andNimmagadda,2009).

For thesereasons,newtherapies needtobedeveloped, and genetherapy,whichisanemergingfieldincancertreatment,rep- resents apromising option.Gene therapycanbeusedwiththe aim of treating the tumour (destructiveapproach) or with the aimofrevertingitsmalignantphenotype(correctiveapproach).

Onetumour-treatmentpossibilityisthe‘genesuicide’approach, alsocalledgene-directedenzymeprodrugtherapy(GDEPT).This approachisbasedonthedeliveryofagene whichcodesforan enzyme,and isable toconverta nontoxicprodrug intoa cyto- toxicmetabolite(Gutzmerand Guerry,1998; Portsmouthetal., 2007).Thefirstproof-ofprincipleofGDEPTwasmadewiththe enzymeHSV-tk(Herpessimplexthymidinekinase)andthepro- drugganciclovir(GCV)(Moolten,1986),whichisstillwidelyused inclinicalandexperimentalapplicationsindifferenttumourmod- els (Portsmouthet al., 2007). The efficacy of this concept was forexampleshownin2004inarandomisedcontrolledstudyon patientswithoperableprimaryorrecurrentmalignantgliomaand allowedasignificantincreaseinpatientsurvivaltime(71weeks inthetreatedgroupcomparedto39weeksinthecontrolgroup) 0378-5173/$–seefrontmatter© 2011 Elsevier B.V. All rights reserved.

doi:10.1016/j.ijpharm.2011.06.031

(Immonenet al.,2004).Totransfer a geneeffectively, different parameters,suchastheadministrationroute,thelengthofgene expressiontime,theanimalmodeland/ortheneedofadelivery vehicle,havetobefixed.Localadministrationdoesnotnecessar- ilyneed adeliveryvehicleand low dosesare usuallyrequired;

however,thetargetedtissuehastobeaccessibletousethisadmin- istrationroute.Systemicadministrationallowsaccessibilitytothe differenttissuesandhastheadvantageofrapidactionandofallow- ingrepeatedadministrationbutitneedsadeliveryvehicleinorder tobeeffective.Moreover,inclinicalapplications,manydiseases requireintravenous(i.v.)orintraperitoneal(i.p.)injectiontreat- mentregimens.Deliveryvehiclesforsystemicadministrationcan beclassifiedintotwomaingroups:viralandnon-viraldeliverysys- tems.Althoughviralvectorsarevery effectiveintermsof gene deliveryandexpression,themajordrawbacksaretheirpotential riskassociatedwithreplication-competentviruses,immunogenic- ityandhighcost(Chowdhury,2009;Collinsetal.,2008;Jinand Ye,2007).Non-viralvectorsdonotshowthesedrawbacks;they presenttheadvantagesofbeingabletocarrylargeinsertsandto beeasilyformulated(JinandYe,2007;Kreissetal.,1999;Morille etal.,2008);theycanalsobeadaptedtopassiveoractivetumour targeting(Huynhet al.,2010; Kangetal., 2010; Wagneret al., 2004).

In recent literature,nanocarriers (lipidor polymer) used in various treatment strategies, carrying different agents such as plasmids,siRNA,ODNorpharmacologicalinhibitors,andapplied byseveraldifferentadministrationroutes(intravenous,intratu- moural,topical,etc.),havebeenshowntoincreasemelanomacell deliveryandtreatmentefficacy(deCamposetal.,2010;Stoneetal., 2009;WeissandAplin,2010;Zhengetal.,2009).

Lipid nanocapsules (LNCs) developed in our laboratory (Heurtaultetal.,2002)havealreadybeenshowntobeefficient forinvitroandinvivotransfection(Morilleetal.,2009a,2010).

TheformulationprocessofLNCsrequiresneitherorganicsolvents norhighenergy, andis basedonthephase-inversion tempera- ture(PIT)method.Itconsistsofusingnon-ionicpolyethoxylated surfactantstoallowthephaseinversionofanemulsion(Shinoda andSaito,1968,1969)andtheapplicationofatemperature-cycle treatmentaroundthePITtoobtainsmallsizeswithalowpoly- dispersityindex(PDI)(Antonetal.,2007).Asuddencoolingand dilutionofthemicro-emulsionatthePITleadstotheobtentionof LNCs,whichconsistofalipid,liquidcoreoftriglycerides,andarigid shelloflecithinandshortchainsofpoly(ethyleneglycol)(PEG₆₆₀) (Heurtault et al., 2002). Thisformulation process allows fragile moleculessuchasnucleicacidstobeencapsulated.Toencapsu- latehydrophilicDNAinthelipidcore,DNAwascomplexedwith cationiclipidspriortoencapsulation(Vonarbourgetal.,2009).As alreadyobserved(Morilleetal.,2009a;Vonarbourgetal.,2009),the encapsulationoftheselipoplexesshouldprovideanefficientlossof toxicityallowinghigherdosestobeinjectedinvivo.Furthermore,in ordertoenhancestealthpropertiesofDNALNCs,PEGwasused(see thereviewofHuynhetal.(2010)describingtheadvantagesand drawbacksofthispolymer).ThecoatingoftheDNA-LNCsurface withlongerPEGchains(PEG2000)improvednanocapsuleinvivocir- culationtimeonamousemodelofsubcutaneouslyinjectedglioma cells(Morilleetal.,2009a)

In this study, non-viral LNCs were used to deliver pHSV-tk inmelanoma tumourcells after IVinjectioninto mice.Consid- eringthe melanomamice model, someluc+ humanmelanoma cellswereengraftedonnudemice.Thus,thelocalisationandthe growthofthetumourscouldbefollowedbyluminescence.Once thetumourswereestablished,thebiodistributionandlocalisation ofintravenouslyinjectedDNALNCsinthisorthotopicmelanoma mousemodelweretracedviainvivofluorescenceimaging.Biolu- minescenceandfluorescenceimageswerethencomparedinorder tolocaliseDNALNCsversusPEGDNALNCs,andatreatmentwith

ganciclovir(GCV)wascarriedoutinordertoevaluatetheefficiency ofthegenesuicideapproach.

2. Materialsandmethods

2.1. Preparationofnanocarriers 2.1.1. Liposomesandlipoplexes

Solutions of DOTAP (1.2-DiOleoyl-3-TrimethylAmmonium- Propane) and DOPE (1.2-DiOleyl-sn-glycero-3- PhosphoEthanolamine) in chloroform (20mg/ml) (Avanti Polar Lipids, Inc., Alabaster, USA) were firstdried by anevaporation processundervacuumandtheformedlipidfilmwasthenhydrated withdeionisedwaterovernightat4^◦C.Thenextday,liposomes weresonicatedfor20minandlipoplexeswereprepared.Fortheir preparation,DOTAP/DOPE(1/1,M/M)liposomesweremixedwith 660␮gofHSV-tkencodingplasmid(pORF-TK-CpG;InvivoGen;

4.35kb)atachargeratioof5(+/−)in150mMNaCl.

2.1.2. DNA-loadedlipidnanocapsules(DNALNCs)

LNCswerecomposedoflipophilicLabrafac^® WL1349(Gate- fosséS.A.,Saint-Priest, France),a mixtureofcaprylicand capric acid triglycerides, oleic Plurol^® (polyglyceryl-6 dioleate) which was kindly provided by Gatefossé S.A. (Saint-Priest, France), NaCl(Prolabo,Fontenay-sous-Bois,France),water(obtainedfrom a Milli-Q-plus^® system, Millipore, Paris, France) and Solutol^® HS-15(30%offreepolyethyleneglycol660and70%ofpolyethy- leneglycol660hydroxystearate(HS-PEG))(BASF,Ludwigshafen, Germany).Theformulationprocessisbasedonphase-inversions ofamicroemulsiondescribedbyHeurtaultetal.(2002).Briefly,all thecomponents,inawelldefinedratio(3.9%ofoleicPlurol^®(w/w), 5.9%ofSolutol^® (w/w),9.9%ofLabrafac^® (w/w),78.9%ofwater (w/w)and 1.4% ofNaCl), were mixedtogether undermagnetic stirring and temperature cycles around the phase-inversion- temperature(PIT)wereperformed.Inthelaststep,coldwaterwas added(inaratio1:1.96)todilutetheemulsionandtoformtheLNCs.

ToformulatefluorescentDNALNCs,lipoplexeswereaddedinstead oftheformulationwater(Vonarbourgetal.,2009),andamixtureof 1,1-dioctadecyl-3,3,3,3-tetramethylindodicarbocyanineperchlo- rate(DiD,em.=644nm;exc.=665nm)(Invitrogen,Cergy-Pontoise, France) and Labrafac^® replaced Labrafac^® alone. To obtainthe Labrafac^®–DiDmixture,asolutionofDiDinacetoneat0.6%(w/w) wasprepared,incorporatedinLabrafac^® inaratioof1:10(w/w) andacetonewasevaporatedbeforeuse(Garcionetal.,2006).

2.1.3. Preparationofcoatednanocapsulesbypost-insertion Coatednanocapsules were preparedas previously described (Morille et al., 2009a). Briefly, fluorescent DNA LNCs were purified after theirformulation, using PD10 Sephadex columns (Amersham Biosciences Europe,Orsay, France). To compensate for the dilution of our formulation and the desalting effect causedbythis purificationstep,anultrafiltrationstepwasper- formedwithMilliporeAmicon^®Ultra-15centrifugalfilterdevices (Millipore,St. Quentin-Yvelines, France) andthe salt-and LNC- concentrationwerereadjustedafterwardstoobtainaphysiologic concentrationof NaCl(150mM)andtheinitialconcentrationof LNCs (152g/l). These purified fluorescent DNA LNCs were then mixed with 1,2-DiStearoyl-sn-glycero-3-PhosphoEthanolamine- N-[methoxy(polyethyleneglycol)-2000] (DSPE-mPEG2000) (Mean MolecularWeight(MMW)=2805g/mol)(AvantiPolarLipids,Inc., Alabaster,USA)toobtainafinalpolymerconcentrationof10mM.

Themixture wasincubatedfor4hat30^◦Cand vortexedevery 15min.

2.2. Characterisationofnanocarriers

2.2.1. PhysicochemicalcharacteristicsofDNALNCs

Coatedand non-coated fluorescent DNA LNCs were charac- terised using a Malvern Zetasizer^® (Nano Series ZS, Malvern InstrumentsS.A.,Worcestershire,UK).Theformulationwasdiluted ina1:100ratiowithdeionisedwaterandsizeandzetapotential measurementswereperformed withthesamesample at25^◦C.

Allmeasurementswereperformedintriplicateandwithsimilar conductivityvalues.

2.2.2. Agarosegelelectrophoresis

The encapsulationand integrity of theDNA molecules after theprocessofnanocapsuleformulationandpost-insertion,were evaluatedbyagarosegelelectrophoresisasdescribedpreviously (Vonarbourgetal.,2009).Briefly,atreatmentwithTriton^®×100 (Sigma,Saint-QuentinFallavier,France)wasperformedtodestroy theLNCs.AvolumeofLNCsequivalentto0.2␮gofDNAbeforeand afterthistreatmentwasmixedwithgel-loadingsolution(Sigma, Saint-QuentinFallavier,France)anddepositedineachwellof1%

agarosegelcontainingethidiumbromide(Sigma, Saint-Quentin Fallavier,France).Themigrationtimewasabout30min,at100V.

2.3. Invivofluorescenceandbioluminescenceimaging 2.3.1. Orthotopicmousemodelofmelanoma

Six-tonine-week-oldfemale,nudeNMRImice(ElevageJanvier, France)werehousedandmaintainedattheUniversityanimalfacil- ity;theywereprocessedinaccordancewiththeLaboratoryAnimal CareGuidelines(NIHPublication 85-23,revised1985)and with theagreementoftheregionalveterinaryservices(authorisation FR;29-024).

Tumour bearing mice were prepared by injecting subcuta- neouslya suspension of3×10⁶ SK-Mel28 luc+melanoma cells (n^◦HTB-72-ATCC)in100␮lofPBS1Xintotherightflankofathymic nude NMRI mice(6-week-old females, 20–24g). The SK-Mel28 celllinehadpreviouslybeentransfectedwithaplasmidencoding theluciferase,andbearingtheresistancegenetotheneomycine (pTG11033;Transgene; 9.6kb).Under neomycinepressure, one clonewasselectedduetoitshighandstableexpressionofluciferase anditsproliferativecapacities.Thus,thestableluminescenceof thesecellsallowedthelocalisationandgrowthofthetumoursto bemeasured. Concerningimmunohistologicalanalyses,tumours werePS100+,AgHMB45+andAgMelan-A+.Allthesecriteriaindi- catedthemalignityandthemelanomicnatureofthetumours.

Threeweeksaftertumourimplantation,150␮lofDNALNCsor PEGDNALNCswereadministeredbyintravenousinjectionintothe tailveinofthemice.

2.3.2. Invivobioluminescenceimaging

In orderto followthe tumourcell growthof theluciferase- expressingmelanomacells,non-invasivebioluminescenceimaging (BLI)wasperformedafter19days,immediatelybeforeinjectionof thenanoparticles,and24hand48hpost-injection.

Mice to be imaged first received an intraperitoneal injec- tion of highly purified synthetic d-luciferin (4mg in 200␮L of water;Interchim, France).Five minuteslater, theanimalswere anaesthetised with a 4% air–isofluorane blend and maintained witha2% air–isofluranmixturethrougha nosecone.Tenmin- utesafterluciferininjection,luminescenceimageswereacquired usinganinvivoimagingsystem(NightOWLII;BertholdTechnolo- gies,Germany) andassociatedsoftware(WinLight32;Berthold, Germany)withabinningof8×8andexposuretimeof4min.Lumi- nescenceimageswerethensuperimposedontostillimagesofeach

mouse.Thesignalswerequantifiedwithintheregionsofinterest inunitsofphotonspersecond.

2.3.3. Invivobiofluorescenceimaging

Non-invasivefluorescentimaging(BFI)wasperformed1h,3h, 5h, 24h and 48h post-injection, using the BFI system of the NightOWL II (Berthold Technologies, Germany) equipped with cooled,slow-scanCCDcameraanddrivenwiththeWinLight32 software(BertholdTechnology,Germany).Consideringthefluo- rescentcharacteristicsoftheDiDfluorescenttagusedtolocalise thenanoparticles,the590nmexcitationand655nmemissionfil- terswereselected.Inparallel,thelightbeamwaskeptconstantfor eachfluorescentmeasurement,whichwasidealwiththeringlight, epi-illumination.Astheringlightwasalwayssetatthesameheight, theexcitationenergyonthesamplewouldalwaysbethesame.

Eachmousewasanaesthetisedwitha 4%air-isofluranblend.

Onceplacedintheacquisitionchamber,theanaesthesiaofthemice wasmaintainedwitha2%air-isofluranmixturethroughout the experimentasdescribedabove.WiththeBFIsystem,thefluores- centacquisitiontimewas3sandthefluorescentsignalwasthen overlaidonapictureofeachmouse.

2.3.4. Treatmentwithganciclovir(GCV)

Toevaluatethe treatmentefficacy,nudeNMRI mice(n=10) bearing subcutaneous melanomas were prepared as described above. Five weeks after tumour implantation, 150␮lPEG DNA LNCswereadministratedbyintravenous injectionandthemice weretreatedtwiceadaywith150␮lganciclovir[concentration 50mg/ml; InvivoGen]for 4days. Bioluminescenceimaging was performedonceadaytofollowandquantifytumourgrowthas describedabove.Thecontrolgroup(n=5)waspreparedinthesame wayfortumourestablishment,butreceivednotreatment.

3. Resultsanddiscussion

3.1. PreparationandcharacterisationofDNALNCs

DNA LNCs encapsulating the fluorescent probe DiDand the plasmidHSV-tkwereprepared,purifiedandthenhalfofthefor- mulationwascoveredwithDSPE-PEG₂₀₀₀chainsbypost-insertion.

Afterwards, agarose gel electrophoresis experiments were per- formedanddemonstratedtheencapsulationoftheHSV-tkplasmid (Table1).ThefirstlaneshowsLNCswithouttreatment.Asverylow fluorescenceisvisible,weconsiderthatthemajorpartoftheplas- midisencapsulated.ThesecondlaneshowsLNCswithatreatment ofTritontodestabilisetheLNCsandhenceliberatetheencapsu- latedplasmid.Here,thefluorescenceindicatesthattheplasmidis intactsinceonlyonedistinctlineisvisibleandcorrespondstothat oftheplasmidalone(datanotshown).Thecomparisonofbothelec- trophoresisgelsshowsnoinfluenceofthepost-insertionprocess ontheDNAencapsulation,sincetheplasmidiswellencapsulated beforeandafterpost-insertion.Theseresultsalsoevidencethat ouroriginalencapsulationmethodoflipoplexesinLNCsisvalid fordifferentkindsofplasmids.Indeed,thisHSV-tkcodingplasmid hasbeenencapsulatedforthefirsttimeinLNCsandshowssimi- larcharacteristicstoDNALNCsencapsulatingaluciferase-coding plasmid(Vonarbourgetal.,2009).

Bothtypesof LNCswerethencharacterisedbysizeand zeta potentialmeasurements(Table1).DNALNCsaresmallparticles withasizeof90nm,andalowpolydispersityindexof0.18.This resultofPDI,inferiorto0.3indicatesanarrowsizedistribution of the system.The plasmidHSV-tk is complexed withcationic lipidsprior toencapsulationwhichresultsinpositivelycharged DNA LNCswitha zetapotentialof+31mV.Afterpost-insertion withDSPE-PEG₂₀₀₀,sizemeasurementsofPEGDNALNCsshowed aslightincreaseinsizeofabout6nm.Analmostdoublingofthe

Table1

SizeandzetapotentialofDNALNCsandPEGDNALNCsencapsulatingtheplasmidHSV-tkandthefluorescentprobeDiD.Agarosegelelectrophoresisshowsthefluorescence ofDNALNCsorPEGDNALNCsbefore(firstlane)andafter(secondlane)destructionwithTriton.

HSV-tkDiDLNCs DNALNCs PEGDNALNCs

Size 90nm 96nm

Polydispersityindex 0.18 0.34

Zetapotential 31mV −6mV

polydispersityindex(0.34)wasobserved,provingtherealinser- tionofsomelongchainsofPEGatthesurfaceofthenanocapsules.

Incontrast,thezetapotentialdiminishedby37mVfrompositive tonear-neutralparticles, aspreviously observed (Morille et al., 2009a).These resultscan be explained by the fact that DSPE- PEG₂₀₀₀chainscarrynegative,dipolarcharges(Vonarbourgetal., 2005)andaretherebyabletomaskthepositivesurfacechargesdue totheencapsulationofthepositivelychargedlipoplexes.Theposi- tivesurfacechargeofDNA-LNCsbeforepost-insertioncouldhelpto interactwithnegativelychargedcellmembranesandleadtobetter internalisationandtransfectioneffectsinvitroasshown,forexam- ple,onHela,H1299orHEK293␤3cells(Morilleetal.,2009b,2010).

However,positivesurfacechargecanalsointeractwithnegatively chargedproteinsinthebloodanddramaticallyreducethecircula- tiontimeinbloodincomparisontoPEGDNALNCs(Morilleetal., 2009a).In conclusion,thecoatingbylongerPEGchainsyielded smallpegylatedneutralparticlesandthusoptimisedthecharacter- isticsofthePEGDNALNCsrequiredforintravenousadministration (Violaetal.,2010;Vonarbourgetal.,2006).

3.2. Themelanomamousemodel

Luciferase-expressingmelanomacells wereinjectedsubcuta- neouslyintotherightflankofnudemiceand,asthetumourgrowth wasquiteheterogeneous,thegroupsweredrawnbychance.After tumourestablishment,thetumoursweredissectedandhistological analysiswascarriedoutonfreshtumoursintheanatomopatho- logicalserviceoftheacademicalhospital(CHU)ofBrest.There,the tumoursweresoakedinbufferedformalinsolutionbeforetheir inclusioninparaffin.Forstandardexamination,3␮msliceswere preparedandcolouredwithhematoxyline–eosin–safran(HES).For immunohistochemicalanalysis, 5␮m slices were prepared and depositedonglassslides.Afterrehydrationoftheprobes,theywere taggedwithanti-PS100antibodies(amarkerofvariouscelltypes:

melanocytes,nervecells,sweatcells,etc.),anti-HMB45antibodies, anti-Melan-Aantibodies(markerofmelanocytes)andKi67(cell proliferationmarker).Theseanalyseswerecarriedoutonseveral tumours.

Instandardhistology,atumouralinfiltration,constitutedofcells presentingvariousatypicalcharacteristicssuchasnucleiwithvolu- minousnucleoli,andinfiltratingthemusclesincertainareas,was observed(seeFig.1).In immunohistochemicalanalysis,positive labelling,toagreaterorlesserextent,oftheproteinS100,butalsoof theantigenHMB45andMelan-A,wasobserved(seeFig.1).Allthese criteriawerecompatiblewiththediagnosisofmalignanttumours ofmelanocyticorigin.

3.3. FollowingthebiodistributionofDNALNCsbyinvivo fluorescenceimaging

TheformulatedDNALNCs andPEGDNALNCs wereinjected threeweeksaftercellinjectionbyintravenousinjectioninthetail veinofthemice.Biofluorescenceimaging(BFI)wasperformedat differenttimesafterLNCadministration(1h,3h,5h,24h,and48h)

tofollow thetissue distributionofLNCs thankstotheencapsu- latedNIRfluorescentprobeDiD(seeFig.2).BFIisafast,simpleand low-costmethod,andallowsthenumberofsacrificedanimalsto bereduced,sinceitcanbecarriedoutonlivinganimalsatdiffer- enttimes,butitisnotanabsolutelyquantitativemethodasthe absorptionanddiffusionofthevisiblelightdependsonthetissues (Goutayeretal.,2010).TheNIRdyeDiDhasusuallybeenusedtofol- lowthebiodistributionoflabelledmoleculesornanocapsules,since thedyeisnotfluorescentinaqueousmedia,andthefluorescentsig- nalcannotcomefromfreeDiD(Texieretal.,2009).Imageswere takenfromlateralanddecubitusdorsalviewstoobtainageneral viewofthebiodistribution,andtobetteridentifytheorgansand subjacenttissueswhicharemoreorlessvisiblebyBFI,depending ontheposition.

ImagesofmicewhichreceivedDNALNCsshowedanintense fluorescencesignalintherightflank,wherethetumourcellshad beengrafted.Alowerfluorescentsignalwasobservedintheliver anduterus/ovaries,andpersistedforonlyafewhours.Indeed,it washardlyvisible5hafterLNCinjection,whereasanintensefluo- rescentsignalpersistedintherightflankand,wasstillvisible48h afterLNCadministration.ImagesofmicethatreceivedPEGDNA LNCsrevealedanintensefluorescentsignalintheliver,theright flank,andtheuterus/ovarieswhichwasstillvisible48hafterLNC injection.Thisisinconcordancewiththefact,thatthesetissues arehighlyvascular,haveahighdegreeofmicrocirculation,andare sufficientlypresentatthebodysurface.However,contrarytothe fluorescentsignalofDNALNCs,whichpresentedamaximumat1h afterLNCadministrationanddiminishedwithtime,thefluorescent signalofPEGDNALNCswasrelativelylowat1hafterLNCadminis- trationandthenincreasedwithamaximumat3hand5hafterLNC administration.Thefluorescentsignalseemstobemoreimportant forPEGDNALNCsthanforDNALNCs.Inconclusion,thecirculation timeofPEGDNALNCsisprolongedcomparedtothatofDNALNCs sinceintensefluorescencesignalsareobtainedupto48hpost-PEG DNALNC-injection.Evenifthetumourwasdifferent(gliomaversus melanoma),asimilarphenomenonhasalreadybeenobservedasa directconsequenceofpegylation(Maedaetal.,2000;Morilleetal., 2009a).

3.4. Followingthetumourcellsbyinvivobioluminescence imagingandcomparisonofBLIandBFIimages

Inparallel,tumour establishmentandgrowthwerefollowed byinvivobioluminescenceimaging(BLI)andBL-andBF-images werecompared asshown inFig.3,tolookmoreclosely atthe localisationoftheLNCsversustheluciferase-expressingmelanoma cells.BLIisbasedontheproductionoflightfromlivingluciferase cellsduetoachemicalreactionbetweenthesubstrateluciferin, injectedintraperitoneallybeforeimaging, and ATP.Thislight is thencapturedexternallybyacooled,charge-coupleddevice(CDD) camera (Hardy et al., 2001).The advantages of BLI are, as BFI, the sensitivity of the technology, the cost-effective instrumen- tation, the simple procedure and the significant reduction in

Fig.1.Immunohistochemicalanalysisofmelanoma.Standardtreatmentwithhematoxyline–eosin–safran(HES)allowsdifferentiatingthetumourcellsfromothercells.

TreatmentwithPS100antibody(PS100),Melan-AantibodyorHMB45antibody,evidencesthemalignityofthemelanomatumours.

Fig.2.Biofluorescenceimages(BFI)atdifferenttimesafterinjectionofDNALNCsandPEGDNALNCsinamousemodelofmelanomafollowingthebiodistributionofthe injectedvectors.ImagesrepresentonemouseafterinjectionofDNALNCsandPEGDNALNCs(firstandsecondtwolinesofimagesrespectively)(lateralanddecubitusdorsal viewsofthemice)at1h,3h,5h,24hand48h.

thenumber of sacrificed animals (Roda etal., 2009).The loca- tion and growth of luciferase expressing melanoma cells can be monitored in real-time and, in contrast to BFI, quantified asphotons per second. BL-images clearly showthe location of luciferase-expressingmelanomacellsintherightflankofthemice and quantification ofthe luciferase expressionshowed a mean luciferaseexpressionof3.3×10⁶±4.0×10⁶photons/sbeforeLNC injectionforthegroupofmicethatreceivedDNALNCs(n=4)and 5.1×10⁶±8.2×10⁶photons/sforthemicereceivingthePEGDNA LNCs(n=4).

Asseenintheprevioussection,PEGDNALNCshadaprolonged circulationtimecomparedtoDNALNCs,andprovokedanintense fluorescentsignalinthewholebodyat5h,andamorespecificlocal- isationinthetumoursat24h.AcomparisonoftheBF-imagesand BL-imagesshowedaclearcolocalisationofluciferase-expressing melanomacellsandfluorescentLNCsat24hafterLNCadministra- tion.TheseobservationsconfirmthatDNALNCsandPEGDNALNCs attainthetumoursiteafterintravenousadministrationviapassive targeting,probablyduetotheenhancedpermeabilityandretention (EPR)effect,whichisknowntobeduetothecombinationofthe leakytumourvasculatureandthelowlymphaticdrainageobserved inthetumoursincomparisontothehealthytissues(Maedaetal., 2000;Maeda,2001).

3.5. IntravenousadministrationofPEGDNALNCsand subsequenttreatmentwithganciclovir

After these promising results of colocalisation of PEG DNA LNCsandluciferase-expressingmelanomacells,thefollowingstep

oftheexperimentwasbasedonganciclovir(GCV)treatmentin ordertoevaluateifthetumourcells weretransfected withthe PEG DNA LNCs and if it could influence the tumour size. The gene-suicideapproach,alsocalledgene-directedenzymeprodrug therapy(GDEPT),isbasedontheintroductionofagene,inourcase, thegenefortheherpessimplexvirusencodingthymidinekinase (HSV-tk)viathePEGDNALNCs,whichwillsensitisethetumour cellstothesubsequentlydeliveredGCV.Todothis,HSV-tkcon- vertstheinertprodrugGCVintotheactivetriphosphorylatedGCV (GCVTP)whichwillactasachainterminatorafterincorporation intotheDNAofdividingcells(Altaner,2008;GutzmerandGuerry, 1998;Portsmouthetal.,2007).

Nude NMRI mice, bearing subcutaneous melanoma cells, receivedoneintravenousinjectionofPEGDNALNCsanda GCV treatmenttwiceadayfor 4days. Thistime periodwaschosen becausethetransfectedtumourcellswillinevitablydiewiththe GCV treatmentand thereforelimit thetimeperiod ofthether- apeutic effect. Tumour establishment was followed by in vivo bioluminescenceimagingasinthepreviousexperimentsandthe PEGDNALNCsfollowedviainvivofluorescentimaging.

Toseethetreatmentefficacy,invivobioluminescenceimaging ofluciferase-expressingmelanomacellswasperformedonceaday duringthetreatmentperiod,and2daysafter;theevolutionofthe luciferase-expressionwasquantifiedasphotonspersecond(Fig.4).

OnDay1,theluciferase-expressiondiminishedslightlyforallthe mice.Afterwards,untilDay3,theluciferase-expressionincreased withtime.InthegroupofmicetreatedwithLNCsandGCV,the increaseseemstobelowerthaninthecontrolgroup.AtDay4, asignificantdifferenceappearsbetweenthetwo groupswitha

Fig.3. BF-imagesat5hand24hafterLNCinjectionandBL-imagesat24hof3miceinjectedwithDNALNCsandPEGDNALNCsbyintravenousinjection.BF-imagesshow LNCdistributionandBL-imagesmelanomacelllocalisation.

Fig.4. TreatmentefficacyofPEGDNALNCs,encapsulatingaplasmidcodingforHSV- tk,incombinationwithGCVofmelanomabearingmiceincomparisontomelanoma bearingmicewithoutLNCadministrationandtreatment(control,n=5).Micewere treatedwithoneinjectionofPEGDNALNCsatDay0andafterwardstwiceadayfor 4dayswithGCV(n=10)byintraperitonealinjection.

decreaseofluciferase-expressionforthetreatedgroup.Thisdif- ferenceseemstoaccentuatetwodaysafterthetreatmentperiod.

However,consideringthestandarddeviationscertainlyduetothe heterogeneityofthetumourgrowth,anothersetofexperiments includingalargenumberofanimalswithsimilartumourvolumes needstobecarriedouttoconfirmornotthistherapeuticeffect.

Moreover,anothergroupofcontrolmicereceivingonlyPEGDNA LNCswithouta GCVtreatmentcouldbeof interest,toevaluate thepotentialofapossibleanti-tumoureffectduetoLNCs.Agroup receivingonlyGCVwasrealisedbutwasnotdifferentfromthecon- trolgroup(datanotshown)confirmingthatGCVisonlyaprodrug andnecessitatesthepresenceofHSV-tktobeactive.

Intheliterature,thegene-suicidetherapyisoftenassociated withabystandereffect.Thisbystandereffectcaneitherbedirect onnearbycellsordistant,bytheinductionofanimmuneresponse by natural killer cells or T-cells (Altaner, 2008). Although this bystandereffectisapossibility,itisconditionedbytheviabilityof thetransfectedcells.Ifthesecellsarekilledtooearly,thebystander effectbasedonthetranscytosisofGVC-monophosphatebypassive diffusionorthroughgapjunctionswillbelimited.Therefore,an importantfactortotakeintoaccountcouldbethetimebetween theLNC administration and theGCV treatment.Here, the GCV treatmentwasgiven8haftertheLNCinjection,butaGCVadmin- istrationoneortwodaysaftertheLNCtreatmentdidnotshowany treatmentefficacy(datanotshown).Anotherpossibilitytoaug- mentthetreatmentefficacycouldbetherepeatedadministration ofPEGDNALNCsafterthegene-suicideeffect.

Finally,asthepassivetumourtargetingbyPEGLNCswasclearly demonstratedhere,weplannowtotesttheseformulationsona metastaticluc+tumourmouse,increasingthenumber ofexper- imental and control animals.In a metastaticmelanoma mouse model,asystemicadministrationofthetreatmentwouldbehighly requiredand,insuchconditions,PEGLNCscouldrepresentanew promisingapproach.

4. Conclusion

LNCsarepromising non-viralgene deliverysystemsbecause theyarewellcharacterised,easytoformulate,adaptivetothetar- getedtissueandhavepreviouslyprovedtobeefficientintermof transfection.Here,ourfirstaimwastostudythecapacityofLNCs totargetspecificallyamelanomatumour engraftedonamouse model,especiallywhenPEGmotifswereincorporatedintothefor- mulation,afterIVinjectioninordertobenefitoftheperturbations ofthestromalmicro-vascularityandtheEPReffect.Asasecond

goal,wewereinterestedtoknowiftumourcellswerewelltrans- fectedandcouldbeledtodeathwhereasthePEGDNALNCswere colocalisedintothetumours.

UsinginvivoBioluminescenceandBiofluorescenceimagingsys- tems,theimagesshowedontheonehand,prolongedcirculation time of DNALNCs coated withlongPEGchains afteri.v. injec- tioncomparedtonon-coatedDNALNCs,andontheotherhand, for thefirst time,colocalisationofnanocapsules andluciferase- expressingmelanomacellsin micefor bothtypesof DNALNCs tested.Theseresultsconfirmedthepreviousresultsobtainedon anectopicgliomamousemodelbutshowedalsotheirvalidityon anorthotopicmelanomamousemodel.

Then,thetreatmentduring4dayswithganciclovirfollowing thetransfectionofpHSV-tkcarriedoutbyPEGDNALNCsshowed, fromDay3,aninterestingdecreaseofphotonsemittedbytheluc+

tumours,reflectingprobablythetransfectionofsomecellsandthe actionofGCV.Theseresultsareencouraging,eveniffurtherexper- imentshavetobecarriedouttoconfirmthisconclusion.

Acknowledgements

TheauthorswouldliketothankMarieMorille(InsermU646, Angers,France)forherhelpandexperienceinthe(PEG)DNALNC formulationandtheexperimentalsettingaswellastheplatform SynNanoVect.ThisworkissupportedbygrantsfromRégionPays delaLoire(CIMATH),Biogenouest,RégionBretagne,Liguecontre lecancer29andCanceropoleGrandOuest.

References

Altaner,C.,2008.Prodrugcancergenetherapy.CancerLett.270,191–201.

Anton,N.,Gayet,P.,Benoit,J.P.,Saulnier,P.,2007.Nano-emulsionsandnanocapsules bythePITmethod:aninvestigationontheroleofthetemperaturecyclingon theemulsionphaseinversion.Int.J.Pharm.344,44–52.

Chowdhury,E.H.,2009.Nucleartargetingofviralandnon-viralDNA.ExpertOpin.

DrugDeliv.6,697–703.

Collins,S.A.,Guinn,B.A.,Harrison,P.T.,Scallan,M.F.,O’Sullivan,G.C.,Tangney,M., 2008.Viralvectorsincancerimmunotherapy:whichvectorforwhichstrategy?

Curr.GeneTher.8,66–78.

deCampos,V.E.,Teixeira,C.A.,daVeiga,V.F.,RicciJr.,E.,Holandino,C.,2010.l- Tyrosine-loadednanoparticlesincreasetheantitumoralactivityofdirectelectric currentinametastaticmelanomacellmodel.Int.J.Nanomed.5,961–971.

Garbe,C.,Peris,K.,Hauschild,A.,Saiag,P.,Middleton,M.,Spatz,A.,Grob,J.J.,Malvehy, J.,Newton-Bishop,J.,Stratigos,A.,Pehamberger,H.,Eggermont,A.,2010.Diag- nosisandtreatmentofmelanoma:Europeanconsensus-basedinterdisciplinary guideline.Eur.J.Cancer46,270–283.

Garcion,E.,Lamprecht,A.,Heurtault,B.,Paillard,A.,Aubert-Pouessel,A.,Denizot, B.,Menei,P.,Benoit,J.P.,2006.Anewgenerationofanticancer,drug-loaded, colloidalvectorsreversesmultidrugresistanceingliomaandreducestumor progressioninrats.Mol.CancerTher.5,1710–1722.

Goutayer,M.,Dufort,S.,Josserand,V.,Royere,A.,Heinrich,E.,Vinet,F.,Bibette, J.,Coll,J.L.,Texier,I.,2010.Tumortargetingoffunctionalizedlipidnanoparti- cles:assessmentbyinvivofluorescenceimaging.Eur.J.Pharm.Biopharm.75, 137–147.

Gray-Schopfer,V.,Wellbrock,C.,Marais,R.,2007.Melanomabiologyandnewtar- getedtherapy.Nature445,851–857.

Gutzmer,R.,Guerry,D.t.,1998.Genetherapyformelanomainhumans.Hematol.

Oncol.Clin.NorthAm.12,519–538.

Hardy,J.,Edinger,M.,Bachmann,M.H.,Negrin,R.S.,Fathman,C.G.,Contag,C.H.,2001.

Bioluminescenceimagingoflymphocytetraffickinginvivo.Exp.Hematol.29, 1353–1360.

Heurtault,B.,Saulnier,P.,Pech,B.,Proust,J.E.,Benoit,J.P.,2002.Anovelphase inversion-basedprocessforthepreparationoflipidnanocarriers.Pharm.Res.

19,875–880.

Huynh,N.T.,Roger,E.,Lautram,N.,Benoit,J.P.,Passirani,C.,2010.Theriseand riseofstealthnanocarriersforcancertherapy:passiveversusactivetargeting.

Nanomedicine(Lond.)5,1415–1433.

Immonen,A.,Vapalahti,M.,Tyynela,K.,Hurskainen,H.,Sandmair,A.,Vanninen, R.,Langford,G.,Murray,N.,Yla-Herttuala,S.,2004.AdvHSV-tkgenetherapy withintravenousganciclovirimprovessurvivalinhumanmalignantglioma:a randomised,controlledstudy.Mol.Ther.10,967–972.

Jin,S.,Ye,K.,2007.Nanoparticle-mediateddrugdeliveryandgenetherapy.Biotech- nol.Prog.23,32–41.

Kang,J.H.,Toita,R.,Katayama,Y.,2010.Bioandnanotechnologicalstrategiesfor tumor-targetedgenetherapy.Biotechnol.Adv.28,757–763.

Kreiss,P.,Cameron,B.,Rangara,R.,Mailhe,P.,Aguerre-Charriol,O.,Airiau,M.,Scher- man,D.,Crouzet,J.,Pitard,B.,1999.PlasmidDNAsizedoesnotaffectthe

physicochemicalpropertiesoflipoplexesbutmodulatesgenetransferefficiency.

NucleicAcidsRes.27,3792–3798.

Lin,J.Y.,Fisher,D.E.,2007.Melanocytebiologyandskinpigmentation.Nature445, 843–850.

Maeda,H.,Wu,J.,Sawa,T.,Matsumura,Y.,Hori,K.,2000.Tumorvascularperme- abilityandtheEPReffectinmacromoleculartherapeutics:areview.J.Control.

Release65,271–284.

Maeda,H.,2001.Theenhancedpermeabilityandretention(EPR)effectintumor vasculature:thekeyroleoftumor-selectivemacromoleculardrugtargeting.

Adv.EnzymeRegul.41,189–207.

Moolten,F.L.,1986.Tumorchemosensitivityconferredbyinsertedherpesthymidine kinasegenes:paradigmforaprospectivecancercontrolstrategy.CancerRes.46, 5276–5281.

Morille,M.,Passirani,C.,Vonarbourg,A.,Clavreul,A.,Benoit,J.P.,2008.Progressin developingcationicvectorsfornon-viralsystemicgenetherapyagainstcancer.

Biomaterials29,3477–3496.

Morille,M.,Montier,T.,Legras,P.,Carmoy,N.,Brodin,P.,Pitard,B.,Benoit,J.P.,Passir- ani,C.,2009a.Long-circulatingDNAlipidnanocapsulesasnewvectorforpassive tumortargeting.Biomaterials31,321–329.

Morille,M.,Passirani,C.,Letrou-Bonneval,E.,Benoit,J.P.,Pitard,B.,2009b.Galacto- sylatedDNAlipidnanocapsulesforefficienthepatocytetargeting.Int.J.Pharm.

379,293–300.

Morille,M.,Passirani,C.,Dufort,S.,Bastiat,G.,Pitard,B.,Coll,J.L.,Benoit,J.P.,2010.

TumortransfectionaftersystemicinjectionofDNAlipidnanocapsules.Bioma- terials.

Portsmouth,D.,Hlavaty,J.,Renner,M.,2007.Suicidegenesforcancertherapy.Mol.

AspectsMed.28,4–41.

Roda,A.,Guardigli,M.,Michelini,E.,Mirasoli,M.,2009.Bioluminescenceinanalytical chemistryandinvivoimaging.TrACTrendsAnal.Chem.28,307–322.

Shinoda,K.,Saito,H.,1968.Theeffectoftemperatureonthephaseequilibriaand thetypesofdispersionsoftheternarysystemcomposedofwater,cyclohexane, andnonionicsurfactant.J.ColloidInterfaceSci.26,70–74.

Shinoda,K.,Saito,H.,1969.ThestabilityofO/Wtypeemulsionsasfunctionsof temperatureandtheHLBofemulsifiers:theemulsificationbyPIT-method.J.

ColloidInterfaceSci.30,258–263.

Stone,G.W.,Barzee,S.,Snarsky,V.,Santucci,C.,Tran,B.,Langer,R.,Zugates,G.T., Anderson,D.G.,Kornbluth,R.S.,2009.Nanoparticle-deliveredmultimericsolu- bleCD40LDNAcombinedwithToll-LikeReceptoragonistsasatreatmentfor melanoma.PLoSOne4,e7334.

Tawbi,H.,Nimmagadda,N.,2009.Targetedtherapyinmelanoma.Biologics3, 475–484.

Texier,I.,Goutayer,M.,Silva,A.D.,Guyon,L.,Djaker,N.,Josserand,V.,Neumann, E.,Bibette,J.,Vinet,F.,2009.Cyanine-loadedlipidnanoparticlesforimproved invivofluorescenceimaging.J.Biomed.Opt.14,054005.

Viola,J.R.,El-Andaloussi,S.,Oprea,I.I.,Smith,C.I.,2010.Non-viralnanovectorsfor genedelivery:factorsthatgovernsuccessfultherapeutics.ExpertOpin.Drug Deliv.7,721–735.

Vonarbourg,A.,Saulnier,P.,Passirani,C.,Benoit,J.P.,2005.Electrokineticproperties ofnonchargedlipidnanocapsules:influenceofthedipolardistributionatthe interface.Electrophoresis26,2066–2075.

Vonarbourg,A., Passirani, C.,Saulnier,P., Benoit,J.P., 2006. Parametersinflu- encingthestealthinessofcolloidaldrugdeliverysystems.Biomaterials27, 4356–4373.

Vonarbourg,A.,Passirani,C.,Desigaux,L.,Allard,E.,Saulnier,P.,Lambert,O.,Benoit, J.P.,Pitard,B.,2009.TheencapsulationofDNAmoleculeswithinbiomimetic lipidnanocapsules.Biomaterials30,3197–3204.

Wagner, E., Kircheis, R., Walker, G.F., 2004. Targeted nucleic acid delivery into tumors:new avenuesfor cancertherapy.Biomed. Pharmacother.58, 152–161.

Weiss,M.B.,Aplin,A.E.,2010.Paying“particle”attentiontonovelmelanomatreat- mentstrategies.J.Invest.Dermatol.130,2699–2701.

Zheng,D.,Li,X.,Xu,H.,Lu,X.,Hu,Y.,Fan,W.,2009.Studyondocetaxel-loaded nanoparticleswithhighantitumorefficacyagainstmalignantmelanoma.Acta Biochim.Biophys.Sin.(Shanghai)41,578–587.