HarvardUniversitySimonFraserUniversity1GregoriadisGFEBSLett952NicolauCParafALiposomesdrugsandImmunocompetentCellFunctionsAcademicPressNewYork19813BarronLGMeyerKBSzokaF ID: 212438
Download Pdf The PPT/PDF document "ProductionofUnilamellarVesiclesUsinganIn..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
ProductionofUnilamellarVesiclesUsinganInvertedSophiePautot,BarbaraJ.Frisken,andD.A.Weitz*DepartmentofPhysicsandDEAS,HarvardUniversity,Cambridge,Massachusetts02138andDepartmentofPhysics,SimonFraserUniversity,Burnaby,BCV5A1S6ReceivedJune20,2002.InFinalForm:January2,2003Weinvestigateamethodforthecontrolledassemblyofunilamellarvesiclesconsistingofbilayersassembledoneleafletatatime.Weusewater-in-oilemulsionsstabilizedbythematerialfortheinnerleafletandproducevesiclesbypassingthewaterdropletsthroughasecondoilwaterinterface,wheretheybecomecoatedwiththeouterleaflet.Wehaveusedthistechniquetoformvesiclesfromlipids,mixedlipidandsurfactantsystems,anddiblockcopolymers.Thestabilityoflipid-stabilizedemulsionslimitstherangeofsizesthatcanbeproducedandthevesicleyield;nevertheless,thereareseveraladvantageswiththisemulsion-basedtechnique:Itispossibletomakeunilamellarvesicleswithsizesrangingfrom100nmtom.Moreover,theprocessallowsforefficientencapsulationandensuresthatthecontentsofthevesiclesremainisolatedfromthecontinuousaqueousphase.Toillustratepossibleapplicationsofthistechnique,wedemonstratetheuseofvesiclesasmicroreactorswherewepolymerizeactinthroughtheadditionofmagnesiumandshowthatthepolymerizationkineticsareunaffectedbytheencapsulation.Bilayersoflipidmoleculesthatfullyencloseafixedvolumeoffluidarecalledvesiclesorliposomes.Theyareubiquitousincellswheretheyencapsulateandisolateintracellularfluidsandoftencontainspecificproteinsorothermacromolecules.Vesiclesarealsousedtotransportmacromoleculesthroughthebloodstreamorthroughtheskin,leadingtothewidespreaduseofvesiclesincosmeticsanddrugdelivery.Inaddition,theyserveasamodelsystemforthestudyofthefundamentalpropertiesoflipidbilayermembranes.Ideally,theseapplicationsrequiresimultaneouscontrolofvesiclesize,unilamellarity,en-capsulationyield,biocompatibility,andlipidcomposition,butnoneofthevesicleproductiontechniquescurrentlyusedfulfillseveryrequirement.Sonicationcanbeusedtoproducedvesicleswithradiismallerthan50nm(smallunilamellarvesiclesorSUV)butthisproducesahighlyheterogeneoussizedistribution.Alternatively,largemultilamellarvesicles(MLV),formedspontaneouslyinexcesswater,canbeextrudedthroughanarrowporesizefiltertoproduceamoremonodispersedistributionoflargeunilamellarvesicles(LUV).Bycontrast,productionofgiantunilamellarvesicles(GUV)thatrangeinsizefrom1to20misusuallyaccomplishedbyrehydrationorthinlipidfilmswithwaterorbyelectroforma-theresultantsuspensioncontainsabroaddistribu-tionofsizes,andonlyafewofthevesiclesproducedareunilamellar.Vesiclesproducedbyallofthesetechniquesareformedinthesamemediumthattheyencapsulate,whichlimitstheirutilityforencapsulationofpreciousdrugsorotheractiveagentsavailableinlimitedquantities.Onetechnique,calledreverseevaporation,thisdeficiencybutdependsontheuseoforganicsolvents,traceamountsofwhichremaininthebilayerandcancontaminatetheinternalaqueousphase.Finally,tosuccessfullyproducevesicles,eachtechniquehasspecificrequirementsforeitherthelipidsorthebufferused.Thislackofflexibilityhasnegativelyimpactedtheuseofvesiclesfordrugdelivery;theirroleistypicallylimitedtothatofsimplecarrierratherthanthedynamicandfunctionalcompartmentstheycouldbecome.Tohelpovercomesomeoftheselimitations,wehaveinvestigatedamethodofvesicleproductionthatinvolvestheassemblyoftwoindependentlyformedmonolayersofamphiphilicmoleculesintounilamellarvesicles.Thefirststepisthepreparationofastableinvertedemulsion,asshowninFigure1A.Theaqueoussolutiontobeemulsifiedismixedwiththeoilphasecontainingsurfactantmol-ecules.Asthewaterdropsareformed,thesurfactantmoleculesdispersedintheoiladsorbatthesurfaceofthedropstoformamonolayerthathelpsstabilizetheemulsion,reducingcoalescenceofthedrops.Thismono-layerwillformtheinnerleafletofthebilayerthatwillmakeupthefinalvesiclebilayer.Wethenpourasecondoilphase,alsocontainingsurfactantmolecules,overthetopofanaqueousphase,asshowninFigure1B.Becausetheoilislessdense,itremainsabovetheaqueousphase,andsurfactantmoleculesdispersedintheoilphasediffusefromthebulktothewater/oilinterfacetoformamonolayer.Aftertheinterfaceisfullycovered,wegentlypourtheinvertedemulsionontopoftheoilinterfaceasshowninFigure1C.Thewaterdropletsintheinvertedemulsionareheavierthantheoilphaseandsedimenttowardtheoilbufferinterface.Uponcrossingtheinterface,theemulsiondropletspickupasecondlayeroflipid,formingabilayerandtransformingtheemulsion HarvardUniversity.SimonFraserUniversity.(1)Gregoriadis,G.FEBSLett.,95.(2)Nicolau,C.;Paraf,A.Liposomes,drugsandImmunocompetentCellFunctions;AcademicPress:NewYork,1981.(3)Barron,L.G.;Meyer,K.B.;Szoka,F.C.J.Hum.GeneTherapy,315(4)Gregoriadis,G.PreparationofLiposomes;CRCPress:BocaRaton,FL,1984;Vol.1.(5)Ostro,M.J.;MarcelDekkerInc.:NewYork,1983.(6)Philippot,J.R.;Schubert,F.LiposomesasToolsinBasicResearchandIndustry;CRCPress:BocaRaton,FL,1995.(7)Olson,F.;Hunt,C.A.;Szoka,F.C.;Vail,W.J.;Papahadjopoulos,Biochim.Biophys.Acta,557.(8)Lasic,D.D.;Yechezkel,B.HandbookofNonMedicalApplicationsofLiposomes;CRCPress:BocaRaton,FL,1996.(9)Angelova,M.I.;Dimitrov,D.S.FaradayDiscuss.,303. (10)Szoka,F.;Papahadjopoulos,D.Proc.Natl.Acad.Sci.10.1021/la026100vCCC:$25.002003AmericanChemicalSocietyPublishedonWeb02/19/2003 dropletsintounilamellarvesicles.Aftersedimentation,thevesiclesareinthebottomaqueousbuffer.Anemulsion-basedmethodofvesicleproductionhasseveralobviousadvantages.Duringtheentireprocess,theencapsulatedmaterialremainsdistinctfromthehostingbuffer.Thevesiclesarethusfilledwiththesoluteofinterestdirectlyduringtheirformation,makinghighencapsulationefficiencypossible.Theuseofemulsificationforencapsulationinthefirststepmakesitindependentofthetypeofamphiphilicmoleculeused,ofthesizeorchargeofthemacromoleculestobeencapsulated,oroftheionicstrengthofthebuffer.Itimprovesthefabricationofunilamellarvesiclesbymakingawiderrangeofsizesaccessiblewithasingletechnique.Itprovidesameanstocontrolthecompositionofeachleafletmakingitpossibletouseabroaderchoiceofamphiphilicmoleculestoassemblethebilayer.Italsomakesitpossibletoformasymmetricbilayers.Becauseeverycomponentofthevesiclecanbecontrolled,thistechniqueprovidesthepotentialfornewapplicationsinbiomaterialsengineeringincludingthepossibilityoftransformingthesevesiclesintomicroreactorswherereactivemoleculescanbeefficientlyencapsulatedandisolatedfromthebulkandwhereremotelytriggeredreactionscanoccur.Thismethodhasbeendiscussedbeforebuthasnotbeenimplementedwidely.Itwasdemonstratedasamethodofvesicleproductionusingabenzene:water:egg-PCemul-sion,andthepossibilityofmakingasymmetricvesicleswasproposed.Morerecently,anencapsulationefficiencyof60%wasreportedusingvesiclesproducedfromadecane:water:egg-PCemulsion.Inbothcases,sonicationwasusedtopreparetheinitialemulsionandelectronmicros-copywasusedtodeterminethevesiclesizes;theywereconsistentlylessthan200nmindiameter.Inthispaper,wepresentadetaileddescriptionoftheinverted-emulsionmethodandwediscussitslimitations.Weshowthat,byapplicationofsuitableemulsificationtechniques,itispossibletomakevesiclesranginginsizefrom0.1mto1mwithavarietyofsurfactant-likemolecules.Wedemonstratethatthevesiclesareunila-mellarandthatitispossibletoachieveencapsulationefficienciesashighas98%.Inotherwork,weshowthatitispossibletousethistechniquetomakeasymmetricvesicles.Wealsodiscusslimitationstothistechniquethatoccurwhenlipidsareusedtostabilizetheemulsion;theseincludecontrolofthesizeandnumberofvesiclesformed.Wehaveobservedthatlipidsdonotbehavelikemostsurfactants.Thus,theemulsificationprocessissometimesnotfullycontrolledandtheslowequilibrationoflipidmonolayersatoilwaterinterfaceslimitsreplenishmentofthisinterfaceandthusvesicleyield.Wediscussthebehavioroflipidsinanoilwaterenvironmentandhowthisimpactsthecontrolofthesizeoftheemulsion,thecompositionofeachleafletofthebilayer,andultimatelythenumberofvesiclesproduced.Finally,toillustratethepotentialofthistechnique,wedemonstratetheencap-sulationofavarietyofmoleculeswithoutlossoftheiractivityandshowthatitispossibletousevesiclesasMaterialsandMethodsThelipidsusedincludedegg-PC(egg-phosphatidyl-choline),POPC(1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocho-line),POPS(1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-andDOPS(dioleoyl-sn-glycero-3-phospho--serine).WealsousedfluorescentlylabeledlipidsincludingNBD-PC(1-palmitoyl-2-(6-((7-nitro2-1,3-benzoxadiazol-4-yl)amino)caproyl)-sn-glycero-3-phosphocholine)andNBD-PS(1-palmitoyl-2-(6-((7-nitro2-1,3-benzoxadiazol-4-yl)amino)caproyl)-sn-glycero-3-(phospho-serine))andrhodamine-DHPE(1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine).AlloftheselipidswerepurchasedfromAvantiPolarLipids,Inc.(Alabaster,AL)as99%purechloroformstocksolutionandwereusedwithoutfurtherpurification.Biotin-conjugatedlipid(N-((6-biotinoyl)amino)hexanoyl)-1,2-dihexa-decanoyl-sn-glycero-3-phosphoethanolamine)wasobtainedfromMolecularProbes(Eugene,OR).AdiblockcopolymerPABu-PAMprovidedbyRhodia(Cranbury,NJ)wasusedtopreparepolymersomes.Thisisasymmetricdiblockcopolymercomposedofapolybutylacrilateblockandapolyacrilamideblock,bothwith8000g/mol.Theencapsulationofmacromoleculeswasdemonstratedwitha1wt%solutionofdextran,10000g/mol,taggedwithTexasRed(MolecularProbes,Eugene,OR).Theencapsulatedsolutionusedtoprobethetransportofdivalentcationsacrossthebilayerwasapolymersolutionpreparedbysuspending10000dextrantaggedwithOregonGreen488bapta-1,acalciumsensitivedye(MolecularProbes,Eugene,OR),inanaqueousbufferconsistingof0.1MKClwith10mMTrisatpH8.Thesolutionwasdialyzedovernightagainstabufferfreeofcalciumandmagnesiumtoremovedivalentcationsthatmightbepresentinthepolymerpowder;thefinalconcentrationwasthensetat1mg/mL.Finally,tostudythepolymerizationofactininsidethevesicles,weusedactinfromacanthamoebapurifiedaccordingtotheprotocolestablishedbyGordonetal.WepreparedN-(1-pyrenyl)iodoacetamide-labeledF-actinandused20%labeled (11)Trauble,H.;Grell,E.Neurosci.Res.ProgrammBull. (12)Zhang,L.;Hu,J.;Lu,Z.J.ColloidInterfaceSci.(13)Pautot,S.;Frisken,B.J.;Weitz,D.A.Unpublished.(14)Gordon,D.J.;Eisenberg,E.;Korn,E.D.J.Biol.Chem.,4778(15)Kouyama,T.;Mihashi,K.Euro.J.Biochem.,33 Figure1.Schematicillustrationofthesynthesisofvesiclesfromaninvertedemulsion.(A)Waterisemulsifiedinoilwithlipidasthesurfactant,formingastableinvertedemulsion.(B)Thewaterthatwillreceivethefinalvesiclesisplacedinasecondvial,andlipid-saturatedoilispouredontopofit.Alipidmonolayerformsattheoilwaterinterface.(C)PreparationAisthengentlypouredontopofpreparationB.Theinvertedemulsiondropletsareheavierthantheoilthatcontainsthemandsedimentintothesecondwaterphase.Astheypassthroughtheinterfacewherethesecondlayeroflipidissitting,thebilayeriscompletedandthefinalvesiclesareformed.ProductionofUnilamellarVesiclesLangmuir,Vol.19,No.7,2003 actinand80%nonlabeledactintostudythekineticsofpolym-erizationoftheactinuponadditionofmagnesium.PreparationoftheVesicles.Thechoiceoftheorganicphaseusedtopreparetheinvertedemulsionisdictatedbythebalancebetweenthesolubilityofthelipidinoil,thestabilityoftheinvertedemulsion,andthepossiblepresenceofoilinthefinalbilayer.Unlikepreviousattemptstousesimilartechniques,weuseasufficientlylong-chainalkanetoensurethatthealkanemoleculesarenotincorporatedintothefinalbilayerandthatastable,invertedmacroemulsionisformed.Weobservedthattheuseofalkaneswithchainsshorterthan10carbonsresultedinamicroemulsionwherewewerenolongerabletocontrolthesizeofthewaterdroplets.Wefoundthatdodecane,a12carbonalkane,wasmostsuitableforourapplications,butavarietyofoilscanbeused.Alkaneswithchainlengthsrangingfrom12to17carbonsresultedinstablemacroemulsions,aswassquallene,acomplexcarbonalkanechainofhighmolecularweight.Wehavecheckedthecompositionofthebilayerbymakingathin-layerchromatographyanalysisofthefinalvesiclesandcomparingthistothatofpurelipidsandpurealkanesamples.Wecouldnotdetectthepresenceofalkaneindicatingthat,ifalkanemoleculeswerepresent,theymadeuplessthan5%ofthebilayer.Thisisthesensitivitylimitofthistechnique;othertechniqueswerenotpursued.Thebestalkaneforanoil-freebilayerissquallene,whichhastheadvantagethatitisimmiscibleinlipidbilayerswhilestillprovidingagoodcontinuousphaseforemulsionAwidevarietyofsurfactantscanbeusedtostabilizeaninvertedemulsion.Inthiswork,weusednonionicsurfactants,lipids,anddiblockcopolymers,chosenaccordingtogeometricalsuchthatisthevolumeofthehydrocarbonchain,isthesurfaceareaofthepolarhead,andisthechainlength.Whenthisratioisgreaterthanone,theareaofthehydrophobicpartofthemoleculeislargerthanthatofthehydrophilicpart,whichfavorsthestabilizationofaninvertedemulsion.Lipidshavearatioonlyslightlyhigherorlowerthanonedependingonthepolarheadgroupandsaltconditions;thisfactmakesthemapoorsurfactantwithwhichtostabilizeanemulsion.Theinvertedemulsionwasusuallypreparedbyfirstdispersingthesurfactantintheoilphase.Formulticomponentlipidmixtures,thelipidswerefirstdissolvedtogetherinchloroformtoobtainahomogeneousdispersion.Thechloroformwasthenevaporatedunderastreamofnitrogentoformadrythinfilmonthesurfaceofaglassvial,whichswelledeasilywhendodecanewasadded.Tofullydispersethelipidmoleculesindodecane,thesuspensionwasplacedinasonicationbathovernightat25ÉC.Surfactantsinliquidform,likeSpan80,anonionicsurfactant,wereaddeddirectlytothedodecane.Finally,forsamplespreparedwiththediblockcopolymerPABuPAM(Rhodia,Cranburry,NJ),itwasmoreconvenienttodissolvethepowderdirectlyintheaqueoussolutionat0.1wt%topreparetheinvertedemulsion.Theemulsificationoftheaqueoussolutionandsurfactant-saturateddodecanecanbeachievedbyvariousmethods.amountofshearandthemethodbywhichitisappliedtothesuspensionareresponsibleforthemeansizeandpolydispersityoftheemulsionproduced.Fourdifferentemulsificationtech-niqueswereused:shearbygentlestirring,shearusingamixer,extrusion,andsonication.Forlipidsandnonionicsurfactants,acrudeinvertedemulsionwaspreparedeitherbyvortexingthesolutionforseveralminutesorbygentlystirringthesuspensionwithamagneticstirbarfor13h.Theemulsionproducedwasverypolydispersewithameansizearound1m.Morecontrolledpreparationwasachievedbyusingashearmixerwithvariablerotationspeed(Ultra-TurraxT25,IKA,Wilmington,NC)orextrusionofthesuspensionthroughanarrowporesizepoly-carbonatefilter(PoreticsCorp.,Livermore,CA).Extrusionwasalsousedtoreducethepolydispersityandthesizeofthedropletsinanemulsionproducedbyothertechniques.Furtherpurificationstepsarethenrequiredtoextracttheinvertedemulsion.insomecases,suchasthediblockcopolymer,thebestresultswereobtainedbysonicatingthesuspension.Unlessotherwisespecified,theemulsifiedaqueoussolutionconsistedof100mMNaCland5mMTrisbufferedatpH7.4.Thevolumefractionoftheaqueoussolutionwas0.5%ofthetotalvolume.Toprepareforthefinalstepintheassembly,weplaced3mLofaqueousbufferinacentrifugetube(Falcontube,50mLwitha1-in.diameter)andthengentlypouredabout2mLofdodecanecontainingamphiphilicmoleculesfortheoutermonolayerontopofthebuffer.Someofthesemoleculesdiffusedtocoverthewaterinterface.Formostsurfactants,adsorptionataninterfaceisdiffusionlimitedandfullcoverageshouldoccurinonlyafewminutes.Theadsorptionmechanismforlipidsappearstobemorecomplex;wefoundthatittakesfrom30minforchargedlipidsupto90minforzwitterioniconestoachievefullcoverage.Oncetheinterfacewasfullycovered,theinvertedemulsionwasaddedtotheFalcontube.Thevolumeoftheemulsionaddedwasvariedfrom0.1mLto1mLtoensurethatthetotalareaoftheemulsiondidnotexceedthetotalareaofthemonolayeravailabletosupplytheouterleaflet.Afterthethreelayerswereassembled,thetubewasplacedinaspin-bucketcentrifugeandspunat120gfor510min.Sincewaterisdenserthanoil,thewaterdropletssedimenttowardtheoilbufferinterface.Centrifugationwasusedtoenhancetherateofsedimentation,decreasingthetimerequiredtotransferthedropletsfromseveralhourstoafew(1)Microscopy.Allofourreal-spaceobservationsweredoneusinganinvertedmicroscope(Leica)usedeitherinnormalorconfocalmode.Someofthepictureswereobtainedwitha100xphasecontrastoilimmersionobjectivewhilethefluorescencepictureswereobtainedwitha100xDICoilimmersionobjective.Sincefreelydiffusingvesiclessmallerthan1mmoveinandoutofthefocusplanerapidly,itisdifficulttoimagethemandmeasuretheirsizebymicroscopy.Toovercomethis,weusedstreptavidin/biotinchemistrytofixthevesiclestothecoverslip,confiningthemtoaplaneandimprovingtheimage.Forthispurpose,vesicleswerepreparedwith1%biotin-conjugatedlipid.Afterthevesiclesweremade,theywerefirstincubatedwitha1:1molarratiooffreestreptavidin(Sigma,St.Louis,MO)for15min,resultinginstreptavidin-coatedvesicles,andwerethenincubatedfor10minonacoverglasspreviouslycoatedwithbiotinylatedalbumin(Sigma,St.Louis,MO).Thecoverglasswaswashedfivetimeswithbuffertoeliminateunboundvesicles.Theboundvesiclesdidnotseemtoadheretothecoatedcoverglasswithawidecontactarea;althoughattachedtothesurface,theystillmovedabouttheiranchoringpoint,suggestingthatthebindingwasrelativelyweak.(2)FluorescenceExperiments.Fluorescencemeasurementswereperformedonaluminescencespectrometer(Perkin-ElmerLS50B).Theinstrumentwasusedtomonitorchangesintheintensityofemittedlightofwavelengthforafixedexcitationwhilethelocalenvironmentofthedyewasmodified.Experimentalparameterssuchasexcitationandemissionwavelengthweredeterminedbymeasuringtheexcita-tionandtheemissionspectrumforeachsample.Weusedafluorescencequenchingassaytomeasurethedistributionoftaggedlipidsbetweeninnerandoutermonolayers. (16)Szoka,F.C.;Papahadjopoulos,D.Annu.Rev.Biophys.Bioeng.,467(17)Xiao,Z.D.;Huang,N.P.;Xu,M.H.;Lu,Z.;Wei,Y.Chem.Lett.,225.(18)McIntosh,T.J.;Simon,S.A.;MacDonald,R.C.Biochim.Biophys.,445(19)Israelachvili,J.IntermolecularandSurfaceForces;AcademicPress:London,1992.(20)Hunter,R.J.FoundationsofColloidScience,VolumeII;OxfordSciencePublications:NewYork,1989. (21)Bibette,J.J.ColloidInterfaceSci.,474(22)Ferri,J.K.;Stebe,K.J.Adv.ColloidInterfaceSci.(23)Pautot,S.;Frisken,B.J.;Cheng,J.-X.;Xie,S.;Weitz,D.A.(24)Bayer,E.A.;Wilchek,M.MethodsBiochem.Anal.(25)McIntyre,J.C.;Sleight,R.G.,11819(26)GruberH.J.;Schindler,S.H.Biochim.Biophys.Acta,212 1(1)Langmuir,Vol.19,No.7,2003Pautotetal. Asuspensionofvesicleswaspreparedwith0.5%ofafluorescentlylabeledlipid.Thefluorescenceofthevesiclesolutionwasmeasuredbeforeandafteradditionofaquenchingsolutionofsodiumhydrosulfite(1MNain5mMTESatpHprepareddaily).Wheninthevicinityofthefluorophore,thesodiumhydrosulfiteextinguishesthefluorescencebyreducingthedye.Asthismoleculedoesnotdiffuseacrossthelipidbilayer,theadditionofthequenchertothevesiclesuspensionresultsintheextinctionofonlythedyelocatedontheouterleafletofthebilayer.Thedyeontheinnerleafletwasexposedbyaddingdetergent(TritonX-100reduced)tolysethebilayer.Theexcitationwavelengthforthesemeasurementswassetat470nm,andtheemissionoffluorescencewasmeasuredat550nm.Finally,fluorescencemeasurementswereusedtoprobethepolymerizationofpyrene-labeledactinprotomers.Polymerizationofactincontaining20%actinprotomerlabeledwithpyreneleadstoanincreaseinfluorescenceintensityofthepyrene.increaseinfluorescenceisduetoamodificationofthepyreneenvironmentwhentheactinprotomersassociateduringthepolymerization.Furthermore,thespectralsignatureofpyreneF-actinisslightlydifferentfrompyreneG-actinallowingthepolymerizationtobemonitored.Theexcitationwavelengthused365nm,andthechangeintheemittedfluorescenceintensitywasmonitoredat407nmafterthepolymerizationwasinitiatedbytheadditionofsalt.(3)DynamicLightScattering.Theapparatususedforthelight-scatteringexperimentswasanALVDLS/SLS-5000spectrometer/goniometer(ALV-LaserGmbH,Langen,Germany).Thelightsourcefortheexperimentswasanargonionlaserofwavelength514.5nm(Coherent,CA).Lightscatteredbythesamplewasdetectedatanangleof90Éfromthetransmittedbeam,wheretheeffectsofreflectionareminimized.Thesamplecellwasplacedinatoluenebathmaintainedatatemperatureof25ÉC.Thesizeandsizedistributionoftheinvertedemulsionandthevesicleswerecharacterizedbydynamiclight-scattering(DLS)measurements.InDLS,wemeasurethenormalizedtimeautocorrelationoftheintensityofthescatteredlight),whichisrelatedtothetimeautocorrelationfunctionofthescatteredelectricfieldlightForadilutesuspensionofmonodisperse)decaysexponentiallywithadecayrateisthemagnitudeofthescatteringwavevectorandisthediffusioncoefficient,whichcanberelatedtothehydro-dynamicradiusthoughtheStokesEinsteinrelation.Forapolydispersesample,)isnolongerasingleexponentialbecauseparticleswithdifferentsizeshavedifferentdiffusioncoefficientsandhencedifferentdecayrates.Theeffectofadistributionofdecayrates)on)isgivenbyAregularizedfitroutinesuppliedbytheALVinstrument,whichissimilartoCONTIN2DP,wasusedtoobtainG()whenanexpansionintermofcumulantsfailedtofitthedatabecausethesizedistributionwasmultimodal.Theresultingdistributionshavetobeviewedwithconsiderablecaution,andtheyareusedhereonlytoestimatethepopulationspresentinoursampleswhenothertypesofanalysisarenotpossible.VesicleProduction(1)Visualization.Thevesiclesproducedwiththistechniquecanbevisualizedbymicroscopy.Figure2Ashowsanimage,obtainedwithphasecontrastmicroscopy,ofpolydispersevesiclespre-paredfromaninvertedemulsion.Theemulsionwaspreparedbygentlystirringfor3h1vol%aqueousbufferindodecanecontaining0.05mg/mLPOPC.Theresultingemulsionwaspolydispersewithdropletsizesrangingfrom1to4mindiameter.Figure2Bshowsasuspensionofmonodisperseegg-PCvesiclesobtainedfromanemulsionthatwasextrudedseveraltimesthrougha1-mporefilter.Onemole%ofrhodaminewasmixedwiththeegg-PCtoallowvisualizationbyfluorescencemicroscopy,and1mol%ofbiotin-conjugatedlipidwasaddedsothatthevesiclescouldbeboundtoastreptavidinsubstrate.Theemulsionusedinthispreparationwascomposedofdropsabout500nmindiameter,aswellassomemuchlargerdropsofaqueousphasethatwasnotfullyemulsified,whichwerenotusedinthevesiclepreparation.Thevesiclesthatcouldbemademosteasilywererelativelysmallinsize,lessthan500nmindiameter;wefoundthatthesizedistributionofthesesmallervesicleswasdeterminedbythatoftheinitialinvertedemulsion.Toillustratethis,wecomparetheautocorrelationfunc-tionsmeasuredforavesiclesamplewiththatoftheinvertedemulsionusedtoprepareitinFigure3A.Theinvertedemulsion,stabilizedwithPOPCandpreparedbygentlestirring,wasusedtoformvesiclesafterthelargerdropshadbeenremovedbysedimentation.Afterthetimescalefortheemulsiondatahasbeenscaledbytheratiooftheviscosities,theautocorrelationfunctionfortheinvertedemulsion(triangles)isvirtuallyindistinguishablefromthatofthevesicles(circles).ThedistributionsofdecaytimesareshowninFigure3B.Thedistributionfortheinvertedemulsionisslightlybroader,perhapsbecauseofthepresenceoflipidaggregatesinthesample.Themeanofthetwodistributionsiscomparable;themeanradiusoftheemulsiondistributionis220nmwherethatofthevesicledistributionis170nm.Theseresultsconfirmthatthesizeoftheinvertedemulsiondictatesthesizeofthevesiclesforthesesmalldroplets.(2)DeterminationofVesicleUnilamellarity.Weusedafluorescencequenchingassaytodeterminethelamellarityofthevesicles.Figure4showsthetime-dependenceofthefluorescenceintensityofPOPS:NBD-PSvesiclesmadefromanemulsionpreparedbygentlestirring.Atthefirstarrow,60Lofquenchersolutionwasadded.Asthequencherdiffusestothesurfaceofthevesicles,theintensitydecreasesandfinallyreachesaplateauwhenthefluorescenceofalltheNBD-PSlipidspresentintheouterleafletofthebilayerhasbeenextinguished.Atthesecondarrow,Triton-X-100wasadded.Asthevesicles (27)Berne,B.J.;Pecora,R.DynamicLightScattering;JohnWileyandSons:NewYork,1976.(28)Provencher,S.W.Comput.Phys.Commun.,213(29)Provencher,S.W.Comput.Phys.Commun.,229(30)Koppel,D.E.J.Phys.Chem.,4814 Figure2.(A)Phasecontrastmicroscopyimageoflarger,polydispersePOPCvesiclesassembledusinganinvertedemulsionpreparedwithalow-sheartechnique.(B)Fluorescencemicroscopyimagesofegg-PCvesiclesformedfromanemulsionextrudedthrougha1-mfilter.ProductionofUnilamellarVesiclesLangmuir,Vol.19,No.7,2003 breakupandtheNBD-PSlipidspresentontheinnerleafletareexposedtothequencher,theintensityreachesitsminimumvalue.Tocalculatethelamellarity,weassumethatthefirstintensitydropisproportionaltotheconcentrationofNBD-PSontheouterleafletofthebilayerandthatthedifferencebetweeninitialandfinalvaluesisproportionaltothetotalconcentrationofNBD-PSpresentinthesample.Wecalculatetheratioofthesemeasurementstobe0.48,consistentwithanequaldistributionofdyeontheouterandinnerlayerandvesiclesthatareunilamellar.LimitationsoftheTechnique.Thistechniquereliesontheformationofastableinvertedemulsionandonthepassageoftheemulsiondropletsthroughasecondinterfacetomakethebilayer.Unfortunately,bothofthesestepscreateproblemswhenlipidsareusedandlimitthegeneralityandpracticalityofthistechnique.Theproblemsoriginatefromtheuseoflipidsintheroleofsurfactants.First,lipidsformlargeaggregatestructuresinbothoilandwaterandhencedonotstabilizetheinterfaceoftheemulsionaswellasatraditionalsurfactantcan;thisseverelylimitstheeffectivenessoftheemulsificationandrestrictstheconcentrationofwaterthatcanbeemulsified.Second,lipidsadsorbveryslowlytotheinterfaceandrequireextendedperiodsoftimetofullycoverthesurface;asaresult,theemulsiondropsaremuchlessrobustandrequireanequilibrationtimeofatleast30min.Third,lipidsattheinterfacebetweenoilandwaterapparentlyundergospontaneousemulsification;thisseemstoresultinastronglypreferreddropsize,makingitexceedinglydifficulttoproducedropsofarbitrarilycontrolledsizes.Fourth,theslowequilibrationoftheinterfacemakesitdifficulttoreplenishthesecondinterfaceastheemulsiondropspassthroughit;thisseverelylimitstheefficiencyoftheformationofthevesicles,particularlyasthesizeofthedropletsincreases.Thecombinationoftheseeffectsseverelylimitstheusefulnessofthetechniqueemphasize,however,thatdespitetheselimitations,thismethoddoesallowtheefficientproductionofuniformlysized,unilamellarvesicleswithafargreaterrangeofsizesthanisattainablewithothertechniquesandinamannerthatallowsefficientencapsulation.However,tohelpestablishtherangeofvalidityofthismethod,weelaborateontheselimitations:(1)StabilityoftheEmulsion.Weobservethatthelipidwillpartitionbetweenthedodecaneandtheaqueousphaseforvolumefractionsofwatergreaterthan1%;anexampleisshowninFigure5.Thissampleconsistsofaninvertedemulsionpreparedbysonicationandcontainingdodecane,0.5mg/mLPOPC,and1vol%aqueousbuffer.Thefinalsuspensioncontainsbothinvertedemulsiondrops,showninFigure5A,andlargeremulsiondropscontainingvesicles,showninFigure5B.ThelargeemulsiondropsshowninFigure5Beventuallycoalescewithoneanotherandphaseseparate.Thistendencyoftheemulsiontodestabilizeathighwatercontentswasobservedinawidevarietyoflipidsandforemulsionsmadebydifferenttechniques.Itwasmorelikelytooccurathighertem-peratures.Webelievethatthisbehavioroccursbecausethegeometricalshapeofthelipidismarginallysuitableforemulsionformation,sincethehydrophilicandthehydrophobicareashavecomparablesize.Topreventphaseseparationoftheemulsionatroomtemperature,ouremulsionscontainednomorethan1vol%aqueoussolutionand0.05mg/mLlipid.Insomecases,thesequantitiescanbeincreasedbytheadditionofsmallfractionofasuitablesurfactant,suchasSpan-80.Thispartitioningsetsastringentlimitonthenumberofvesiclesthatcanbeproducedandrequiresaverydelicatebalanceofconcen-trationstoattainsuitableconditionstoproduceastableinvertedemulsion.(2)EquilibrationoftheEmulsion.Thistechniquereliesontheformationofstableoilwaterinterfacesfullycoveredwithlipids.However,theformationofafully Figure3.Comparisonofthedistributionofsizesbetweentheemulsionandvesicleswhentheinvertedemulsioncontainssmall(m)droplets.(A)Theautocorrelationfunctionfunctiong(2)(ô)-1]obtainedbyDLSatascatteringangleof90ÉforastableinvertedemulsionstabilizedbyPOPC(triangles)andfromthevesiclesproducedfromit(circles).Thedatahavebeencorrectedforthedifferenceinviscosity.Weplotin(B)thedistributionofdecaytimesobtainedusingALV'sregularizedfitroutinefortheemulsion(triangles)andforthevesiclesproducedfromit(circles),againcorrectedforthedifferenceinviscosity.Thedistributionfortheemulsionisbroaderthanthatforthevesiclesbuttheirmeanvaluesareessentiallythe Figure4.FluorescencemeasurementsofintactPOPS:NBD-PSvesiclesbeginsattime0.Additionofsodiumhydrosulfiteoccurredatarrow1.Thefluorescenceofthevesiclesdecreasedtoaplateauvalue,whichrepresentsthefluorescenceintensityduetotheinnerleafletonly.Atarrow2,Triton-X-100reducedwasaddedtolysethevesicles.Thefluorescencedecreasedtobackgroundlevelsastheinnerlipidswereexposedtothequencher.Inthisexperiment,48%ofthefluorescentlipidislocatedintheouterleaflet,implyingthatthevesiclesformedwithourtechniqueareunilamellar.Langmuir,Vol.19,No.7,2003Pautotetal. equilibratedinterfacethatiscompletelycoveredbylipidstakesasurprisinglylongtime.Thisismostclearlydemonstratedbythebehaviorofthesurfacetensionoftheoilwaterinterfaceinthepresenceofthelipids.Uponadditionofthelipids,thesurfacetensionoftheinterfacedecreasesoverseveralhours,indicatingtheveryslowequilibrationtime;bycontrast,thesurfacetensionofanwaterinterfacestabilizedbysurfactantsreachesequilibriumwithinseconds.Thisslowequilibrationoftheinterfacecanaffecttheformationofboththeinnerandouterleafletsofthevesicle.Toillustratethis,westudiedtheeffectofadsorptionontheformationoftheinnermonolayerusingemulsionspreparedbygentlestirring.Wepreparedanemulsionbystirring0.5vol%waterinto20mLofdodecanecontaining0.05mg/mLPOPCwith1mol%ofNBD-PC.Asthesuspensionwasstirred,0.5mLofemulsionwascollectedafter60,90,120,and180minandplacedoverpreviouslypreparedinter-mediate/aqueousphasesthathadbeenallowedtoequili-bratefor3h,eachconsistingof3mLofthesamestocksolutionusedfortheemulsionplacedoverseparateaqueousphases.ThesestepsensuredthattheinterfacesbetweentheaqueousandintermediatephasehadreachedequilibriumandthatalltheseinterfaceshadthesameTheyieldofvesiclesproducedfromemulsionsofdifferentagewascomparedusingfluorescencequenchingtech-niques;resultsareshowninFigure6.Thefluorescenceofboththeouter(triangles)andtheinner(circles)leafletsincreaseswiththeageoftheemulsionusedtomakethevesicles.Fortheemulsionthathadequilibratedtheshortesttime,theoutermonolayercontributes100%ofthetotalfluorescenceandthecontributionduetotheinnermonolayerisnegligible.Forthemostfullyequilibratedemulsion,thecontributionoftheinnermonolayerhasincreasedto40%.Theouterleafletismadefromafullyequilibratedmonolayer;theincreaseinfluorescenceofthislayerwithequilibrationtimeindicatesthattheefficiencyofthevesicleproductionisincreasingasthedropletsequilibrate.Theincreaseinfluorescenceoftheinnerlayerwithequilibrationtimeindicatesthatthetaggedlipids,NBD-PC,adsorbattheinterfaceoftheemulsiondropletsevenmoreslowlythantheirundyedcounterparts.Theseresultssuggestthatafullyequili-bratedinterfaceshouldproducethebestyield;however,wehavepreparedgoodqualityvesicleswithhighyieldinsomeinstanceswheretheinterfaceswerenotfully(3)ControllingtheSizeoftheEmulsion.Whensurfac-tantsareusedtostabilizeaninvertedemulsion,themeansizeofthedropletsproducedissetbythemethodofemulsificationusedandtheamountandtypeofshearsuppliedtothesuspension.However,lipid-stabilizedemulsionsappeartobehavesignificantlydifferently.Thisseverelylimitsthepossiblecontrolofthesizeoftheemulsiondropsandhencethesizeofthevesiclesproduced.Toillustratethisproblem,inthissectionwecomparethedistributionofsizesobtainedforthreeinvertedemulsionspreparedatthesamevolumefractionbutusingthreedifferenttechniques:sonication(for5min),highshearat5000rpm(5min),andlowshear(3h).Experiencewithsurfactant-stabilizedemulsionssuggeststhatthesmallestdropsizeisexpectedforsonicatedsamplesandthelargestdropsizeisexpectedwhentheemulsionisformedbylow-sheartechniques.WeusedDLStomeasurethesizeoftheinvertedemulsionsproduced.InFigure7A,weshowtheautocor- Figure5.PhasecontrastmicroscopypicturesofaninvertedemulsionstabilizedbyPOPCpreparedbysonication.Afterseveralminutes,thesuspensionphaseseparates.Picture(A)showstheupperphase,whichcorrespondstotheinvertedemulsionthatremainssuspendedindodecaneandpicture(B)showsbigemulsiondrops,whichhavesedimentedtothebottomofthesamplevialandarefullofvesicles. Figure6.IncreasedfluorescenceintensityforPOPCvesiclesuspensionsproducedwithafixedvolumeofemulsionremovedafterdifferentamountsofstirring.Theincreaseinthefluorescenceoftheouterleaflet(triangles)indicatesanincreaseinthenumberofvesiclesproduced,andtheincreaseinthefluorescenceoftheinnerleaflet(circles)indicatesthatNBD-PCentersthemonolayermoreslowly.ProductionofUnilamellarVesiclesLangmuir,Vol.19,No.7,2003 relationfunctionsmeasuredforinvertedemulsionssta-bilizedwithPOPSandformedbysonication(squares),highshear(circles),orlowshear(triangles).Theyareclearlynotsingleexponentialconfirmingthattheemulsionispolydisperse.ThedistributionsofdecayratesarepresentedinFigure7B.Twopeakscorrespondingtotwodistinctpopulationsofdropletsareobservedinallthreesamples.Thereappearstobeapopulationoflargedropswhosemeansizevarieswiththeamountofshearsuppliedtoemulsifythesuspensionandapopulationofsmalldropswithradiiofabout150nmthatispresentineach,independentoftheemulsificationmethodused.Thefactthatthesmalldropletsoccurindependentlyofthemethodusedsuggeststhatthesedropshavebeenformedbysomeformofspontaneousemulsificationashasbeenobservedinsomesurfactantsystems.Thus,twoemulsificationprocessesaretakingplace:normalemulsificationduetoappliedshearresultinginthelargerdropsandspontane-ousemulsificationleadingtothesmallerdrops.Whenspontaneousemulsificationoccurs,itpresentsaseriouslimitationtoourabilitytocontrolthesizeofthevesicleswhilepreservingtheconcentrationofvesicles.Forex-ample,itmakesitdifficulttoefficientlyproducelargevesicles;todoso,averylowsheartechniquemustbeusedtoproducetheemulsion,followedbyfurtherpurificationstepstoisolatethelargerdroplets.thesepurificationstepscompromisetheefficiencyofencapsulation,limitingtheutilityofthismethod.Alternatively,adirectschemeofdropletproductionmaybeusedtorestricttheformationoftheemulsiondropstothedesiredsize.Forexample,emulsionsformedbyextrusioncanbeusedtoproducelargevesicleswithrelativelynarrowsizedistributions,aswewillshowbelow.(4)TransferoftheEmulsionDropletstotheAqueousTheoutermonolayerisformedastheemulsiondropletspassthroughtheinterfacebetweentheinter-mediateandaqueousphases.Successfulcompletionofthisstepdependsonstabilityoftheemulsionandcoverageoftheinterfacelayer.AsshowninFigure3,weareabletoobtaingoodtransferofsmalldroplets,lessthannmdiameter.Largerdropletspresentamuchgreaterproblemwhentheyaretransferredtothefinalaqueousphase.Figure8showsdataforaninvertedemulsionstabilizedbyPOPSandpreparedbygentlestirringandasolutionofvesiclesformedfromitwithoutfractionation.Figure8Ashowstheautocorrelationfunctionfortheinvertedemulsion(triangles)rescaledtotakeintoaccountthedifferenceinviscosityandtheautocorrelationforthevesiclesproduced(circles).Thetwocurvesdepartfromeachotheratlargecorrelationtimesindicatingthatlargerdropsarepresentintheemulsion.Thecorrespondingdecaytimedistribu-tionsareshowninFigure8B;thetwocurvesconfirmthatthepopulationoflargeobjectsismuchlessimportantin (31)Rang,M.;Miller,C.A.ProgressinColloidandPolymerScienceSpringer-Verlag:Berlin,1998;Vol.109.(32)Rang,M.;Miller,C.A.J.ColloidInterfaceSci.,179.(33)Kawashima,Y.;Hino,T.;Takeuchi,H.;Niwa,T.;Horibe,K.ColloidInterfaceSci.,512 (34)Umbanhowar,P.B.;Prasad,V.;Weitz,D.A.,347(35)Thorsen,T.;Roberts,R.;Arnold,F.;Quake,S.Phys.Rev.Lett.,4163 Figure7.Dynamiclight-scatteringmeasurementsat90ÉforinvertedemulsionsstabilizedbyPOPSandpreparedusingthreedifferenttechniques;(A)[1]forinvertedemulsionspreparedusingsonication(squares),high-shear(circles),andlow-shear(triangles)techniques.(B)Bimodaldistributionofdecaytimes.Thefirstpeakcorrespondstoapopulationofsmalldropletsthathasasimilarsizeinallthreesamplesandthesecondpeakcorrespondstoapopulationoflargedropletswhosesizedependsontheemulsificationtechnique. Figure8.Comparisonofthedistributionofsizesbetweentheemulsionandvesicleswhentheinvertedemulsioncontainslarge(m)droplets.(A)DLSdatatakenat90ÉforaninvertedemulsionstabilizedbyPOPSrescaledtotakeintoaccountthedifferenceinviscosity(triangles)andforthesuspensionofvesiclesformedfromthisemulsion(circles).(B)Correspondingdistributionofdecaytimesforbothsamples.Weobservedifferencesbetweenthedistributionsfoundfortheemulsion(triangles)andthatfoundforthevesicles(circles),particularlythelossofpartofthelargestinvertedemulsiondropletsuponLangmuir,Vol.19,No.7,2003Pautotetal. thevesiclesolutionthanintheinvertedemulsion.Webelievethatonlyafractionofthelargeemulsiondropletspassthroughtheinterfacetoformvesiclesbecausemostbreakattheinterface.Thebreakagemechanismremainsunclear,butexperi-mentalevidencesuggeststhatthesurfaceconcentrationoflipidattheoilwaterinterfaceplaysanimportantrole.Thismonolayerisalsoaffectedbythekineticsofinterfaciallipidadsorptionattheoilwaterinterfacebetweenthetwobottomphases.Wehaveobservedinourstudyofasymmetricvesiclesthatseveralhoursofequilibrationtimearealsorequiredtoachievesufficientlipidcoverageoftheintermediatephasemonolayerforoptimumconver-sionofemulsiondropletsintovesicles,andthatshorterequilibrationtimesleadtoaninsufficientcoverageanduncontrolledlipidcomposition.Unfortunately,whentheinterfaceislefttoequilibrateforamuchlongerperiodoftime,weobserveawhitishcloudattheoilwaterinterfacecorrespondingtoanaccumulationoflipidinmultilayerstructures.Thisverydelicatebalancebetweenequilibra-tion,whichisrequiredfortheoptimumformationofthevesicles,andspontaneousemulsification,whichdegradestheformation,createsasignificantlimitationinthetechnique.Inpractice,weallowthisinterfacetoequilibrateformorethantwo,butfewerthan3h,providingthebestcompromisechoice.Thedepletionoflipidfromtheoilwaterinterfacecanalsoberesponsiblefortransferfailure.Astheemulsiondropletspassthroughtheinterface,theypickuppartofthemonolayerforbilayercompletion.Newlipidsdonotadsorbquicklyenoughtoreplenishtheinterfaceduringtherelativelyfastsedimentationstage.Thissuggeststhattheareaofthelowerinterfaceshouldbecomparabletothetotalareaofbilayerrequiredforoptimumvesicleproduction,whichlimitsvesicleyield.Herewepresentsomeapplicationsthatutilizethismethodofvesicleproduction;theseexamplesexploitsomeoftheadvantagesofthistechnique:thepossibilityofproducingvesiclesusingavarietyofbilayerconstituents,theencapsulationefficiency,andthepreservationofbiologicalactivity.PolymerorSurfactantConstituentMolecules.Thistechniqueisnotlimitedtolipids.Justaselectro-formationhasbeenusedtoformmoreresistantpoly-mersomesfromdiblockcopolymers,thistechniquecanbeadaptedforusewithawidevarietyofamphiphilicmoleculestoproducepolymericbilayersormixedbilayerswherehighencapsulationyieldispreserved.Infact,awideselectionofamphiphilicmoleculescanbeusedaslongastheycanself-assembleattheoilwaterinterfaceandcanstabilizeaninvertedemulsion.Wehavesuc-cessfullymadevesiclesusinginvertedemulsionspreparedwithawidevarietyofphospholipidsbothwithandwithoutaddedcholesterol,syntheticsurfactants(eithercationicornonionic),anddiblockcopolymers.TwoexamplesarepresentedinFigure9.InFigure9A,weshowaphasecontrastmicroscopyimageofvesiclespreparedfromanemulsionstabilizedby80:20Span-80:egg-PCandformedbysonication.Duringthefirst15minfollowingtheirformation,mostofthelargestvesiclespresentinganexcessareafluctuateandbudtoformtwosmallervesicles.Thearrowshowsavesicleabouttobud.Sincewedidnotobservesuchbehaviorinothersurfactants,webelievethatthepresenceofSpan-80makesthesevesiclesmoreflexibleleadingtobudding.Afterthisreorganization,thevesiclesremainstableforseveraldays.WhenthepercentageofSpan-80usedisgreaterthan80%,thevesiclesformedareunstableandbreakintomicelleswithinamatterofhours.Animageobtainedbyconfocalmicroscopyofoneofthevesiclesmadewithadiblock-copolymerPABuPAM,16000g/mol,taggedwith1mol%NBD-PCisshowninFigure9B.Nematicorderingofthemoleculesinthebilayerisobservedinthepolarizationofthefluorescence;thebrighterpolesofthevesiclecorrespondtofluorescentlightpolarizedparalleltotheanalyzer,andthedarkerpolescorrespondtofluorescentlightpolarizedperpen-diculartotheanalyzer,indicatingthatthefluorescentlipidsareinsertedinthepolymerbilayernormaltotheinterfaceasinalipidbilayer.Althoughthisbipolarityislesspronouncedthaninathinnerlipidbilayer,itdoessuggestthatthepolymerbilayerisfairlyordered.Usingthispreparationtechnique,activemoleculescanbeisolatedfromthebulkbyencapsulatingtheminvesicles.Onceencapsulated,mac-romoleculesareunabletodiffuseout,butalimitedquantityofsmallmoleculescandiffusefromthebulkthroughthebilayerintothevesicles.Thusitispossibletodesignvesiclesforuseasmicroreactors,wherethereactionofmacromoleculesoccursinacontrolled,isolatedenvironment.Thiscouldhaveparticularapplicationintheencapsulationofbiologicallyimportantmolecules,providedtheirbiologicalactivitycanbepreservedduringvesiclesynthesis.Todemonstratetheuseofvesiclesasmicroreactors,itisnecessarytoshowthatthevesiclessuccessfullyencapsulatemacromolecules,thatwecancontrolthetransportofinitiatorsacrossthebilayer,andthatwecanretainbiologicalactivityduringthevesicleproductionprocess.Asanexample,wedemonstratethe (36)Discher,B.M.;Won,Y.Y.;Ege,D.S.;Lee,J.C.M.;Bates,F.S.;Discher,D.E.;Hammer,D.A.,1143 Figure9.(A)PhasecontrastmicroscopyimageofSpan80-egg-PC(80:20)vesicles.Thearrowpointstoabuddingvesicle.(B)ConfocalmicroscopyimageofaPABuPAM(diblockcopolymer)vesicletaggedwith1mol%NBD-PC.Thedistributionofthepolarizedfluorescenceintensityconfirmsthatthetaggedlipidsareincorporatedintothepolymerbilayerandareorderednormaltosurface.ProductionofUnilamellarVesiclesLangmuir,Vol.19,No.7,2003 polymerizationofactinmonomerstriggeredbyanincreaseoftheconcentrationofmagnesium.(1)Encapsulation.Todemonstratetheencapsulationofmacromolecules,wemeasuredthefluorescenceof2-vesiclespreparedfromanegg-PC:cholesterolstabilizedemulsionwheretheaqueousphasecontained1wt%ofafluorescentlylabeleddextran(10000g/mol),awater-solublepolysaccharide.Usingfluorescencemicros-copy,weobservedthatthefluorescenceoriginatesonlyintheinteriorofthevesicleswithlittlebackgroundintensity,confirmingthatthedextranremainsencapsulatedwithinthevesicles,asshowninFigure10.Wehavealsobeenabletoencapsulatedextranwith200000g/mol.(2)IonTransportacrosstheBilayer.Todemonstratecontrollediontransport,weencapsulated10000dextranconjugatedwithacalciumsensitivedye,OregonGreen,invesicleswithandwithoutionophores.ionophoreusedwasA23187(Sigma,StLouis);whenpresentinthebilayer,itfacilitatesthetransportofdivalentcationsacrossthebilayer.Weusedfluorescencespec-troscopytoquantitativelymeasurechangesinfluorescenceintensityasafunctionoftheconcentrationofcalciuminthesolution.Asacontrolexperiment,wefirstmeasuredthefluorescenceintensity(488nm,535nm)oftheconjugateddextraninbulkbuffersolutionbeforeencapsulationasafunctionofthecalciumadded.Figure11Ashowsthechangesintheemissionspectrumuponadditionofcalcium.Thesolidtrianglescorrespondtotheintensityoftheemissionspectrumforthisdyewithoutcalciuminsolution.Aftertheadditionof5mMcalciumchloride,theintensityincreasedbyafactorof1.7,asshownbytheopentriangles.Thesamepolymersolutionwasthenemulsifiedandencapsulatedinvesicles.Thevesicleswerepreparedwithegg-PC:DOPS(80:20)withandwithoutionophores.Inthepresenceofionophores,calciumcandiffuseintothevesiclesuntiltheconcentrationgradientiszero,withacorrespondingincreaseinfluorescenceintensitywithtime.Figure11Bshowsthetimedependenceofthefluorescenceintensityobtainedfortheencapsulatedcalcium-sensitivedextran.Attime0,weobservethefluorescenceoftheencapsulatedpolymerwithoutcalcium.Theadditionof5mMcalciumtothevesiclesuspensionresultedinanincreaseofthefluorescencebyafactorof1.7,identicaltothatobservedforthebulksample.Thisconfirmsthatthefinalconcentrationofcalciuminsidethevesiclesisthesameasthebulkconcentration.Incontrast,inacontrolexperimentwherevesicleswerepreparedwithoutiono-phores,wedidnotobserveanydetectablechangesinfluorescenceafteradditionofcalciumtothesuspension,confirmingthationtransportacrossthemembranecanbecontrolledusingionophores.Furthermore,thisalsoquantitativelyconfirmsourpreviousresultthatthereisnomeasurablenonencapsulateddextraninthebulksolutionandthatdextranhasbeensuccessfullyencap-(3)Biocompatibility.Averyimportantfeatureofvesiclesistheirbiocompatibility.However,thisemulsion-basedproductiontechniqueinvolvesseveraltransfersbetweenfluids,notallofwhichmaybecompletelycompatiblewithbiomaterials.Totestthepreservationofbiologicalactivitywithinourvesicles,weinvestigatedtheaggregationofG-actintoformF-actin.WeencapsulatedG-actinmono-mersfromacanthamoebainegg-PCvesicleswithiono-phoresincorporatedinthebilayerandstudiedthepolymerizationkineticsoftheactinfortwoconcentrationsofencapsulatedmonomer:5Mand15M.Tofollowthekineticsofpolymerization,weadded20%N-(1-pyrenyl)-iodoacetamide-labeledG-actintothesolutionofactinmonomer;thefluorescenceofpyrene-labeledactinin-creasessignificantlyuponpolymerization,providingasensitiveprobeofthereaction.Weinitiatedthepolym-erizationbytheadditionofsalttoreachafinalconcen-trationof50mMKCland2mMMgClThefluorescenceintensityfromboththeencapsulatedactinandacontrolsampleofthesameactininbulksolutionduringthepolymerizationprocessfortwoactinconcentrationsisshowninFigure12.Alargedifferenceinabsoluteintensitiesismeasuredbetweenthebulk(symbols)andtheencapsulatedactin(curves),reflectingthedilutionfactor;theencapsulateddataforbothsampleshavebeenmultipliedbyaconstantfactorof18.5torescalethemtothesamelevelasthebulkdata.Figure12showsthat,inallcases,theadditionofsaltgeneratesastrongincreaseinthefluorescentsignalobserved.Theresultsobtainedfor15Mactininbulk (37)Hackl,W.;Barmann,M.;Sackmann,E.Phys.Rev.Lett.,1789. Figure10.Vesiclesformedfromanemulsionstabilizedbyegg-PC:cholesterol(80:20),encapsulating10000dextrantaggedwithTexasRed,andproducedbyextrusionthrougham-diameterfilter.Inthiscase,thevesiclesarenotboundtothesubstrate,sotheirintensityvariesdependingontheirpositioninthefocalplane. Figure11.(A)EmissionspectrumofdextranlabeledwithOregonGreen488bapta-1at488nm.Thesolidtrianglescorrespondtotheemissionspectrumofthedextraninbufferwithoutcalcium,andtheopentrianglesshowthechangesintheemissionspectrumafteradditionof5mMCaClinsolutionandareusedtocalibratethefluorescentresponsefortheadditionoffixedamountofcalcium.(B)Timecourseofthechangeinfluorescenceintensityat535nmfordextranencapsulatedinegg-PC:DOPSvesiclesuponadditionof5mMtothevesiclesuspension.Ionophoreshavebeenincor-poratedintothebilayerstoallowpassivetransportofdivalentcationsacrossthemembrane.ThecurveshowshowthefluorescenceintensitychangesascalciumionsdiffuseintotheLangmuir,Vol.19,No.7,2003Pautotetal. (crosses)andthecorrespondingencapsulatedactin(dottedline)overlap,confirmingthatthetimedependenceofthepolymerizationkineticsisidentical.Similarly,thecurvefor5Mactininbulk(circles)overlapswiththecorre-spondingencapsulatedactin(solidline).Theseresultsconfirmthatthesamepolymerizationreactiontakesplacewithinthevesicleasinbulkandthattheconcentrationandproteinactivityarebothpreservedbytheencapsula-tiontechnique.Theresultsalsoallowameasurementofencapsulationefficiency.Theemulsiondropletsthatcontaintheactinhaveatotalvolumeof0.165mLandaretransferredinto3.0mLoffinalaqueousphase.Thisresultsinadilutionfactorof18.2.Thus,fora100%efficientprocessofencapsulationandtransferthroughtheinterface,thefluorescenceofthevesiclesuspensionisexpectedtobe18.2timessmallerthanthebulksample.Themeasuredscalingfactorof18.5correspondsto98%encapsulationThelipidbehavioratoilwaterinterfacesmakesitdifficulttocontroltheemulsificationprocessandreducesthedomainwherethisemulsion-basedvesicleproductiontechniqueproducesoptimalresults.Someoftheselimita-tionscanbemitigatedbyconsiderationofthephasebehaviorofthelipid/oil/watersystemandthekineticsoflipidadsorption.Onceaprotocolisdesignedthattakestheselimitationsintoaccount,theinvertedemulsiontechniqueprovidesawaytosimultaneouslyimprovetheencapsulationyieldandtocontrolthebilayercomposition,whileproducingunilamellarlipidvesicles.Wehaveshownthatthistechniquecanbeusedtopreparesurfactantvesiclesorpolymersomes;infact,theuseofconventionalsurfactantsordiblockcopolymerimprovesourcontroloftheemulsificationprocess.Wecanencapsulateawidevarietyofmacromoleculesatanydesiredconcentrationwhilepreservingphysiologicalconditionswithencapsula-tionefficienciesofupto98%.Thistechniqueisoneoftheveryfewmeansofconvenientlyencapsulatinglargemacromoleculeswithradiiofgyrationaslargeas1asmayberequiredforpotentialapplicationsuchasdrugdeliveryandgenetherapy.Moreover,thesecarrierscanbeªfunctionalizedºusingbiotin-conjugatedlipids,provid-ingaprimarybindingsiteforstreptavidin.Thesevesiclescanthenreactwiththedesiredbiotinylatedantibodiesandprovideaconvenientwaytodesignimmuno-lipo-whichcanspecificallybindtocellularreceptors.Furthermore,wehavedemonstratedthatthistechniqueiscompatiblewithbiologicalactivityandthatthevesiclesproducedwiththismethodcanbeusedasmicro-biore-actors.Thisprovidesthepossibilityofcontrollingreactionsinaconfinedspace.Thismethodalsooffersanewwaytodesignchemicalsensorswherevesiclescouldbeusedtolocallyprobetheionconcentrationinsolutionaswellasmultiencapsulationdeviceswithvesiclescontainingsub-compartmentsfilledwithdifferentmolecules.Insummary,emulsion-basedvesicleproductionprovidesanewmeansofmakingvesiclesthatmayfindusesinawiderangeofmaterialsscience.WethankE.M.Ostapforas-sistanceinpurifyingG-actinandforuseofhislaboratoryfortheactinexperimentsandG.Whitesidesforuseofhisspectrofluorimeter.WearegratefultoJ.Crocker,D.Discher,andE.Weeksfortheirusefulsuggestions.S.P.isparticularlygratefultoC.GayandM.T.ValentineforhelpinthepreparationofthisarticleandtoP.Poulinforhisencouragementattheearlystageofthisworkandfordiscussionsastheprojectprogressed.WealsothankRhodia(Cranburry,NJ)forsupplyingthediblockcopoly-mers.ThisworkwassupportedbyRhodia,KraftFoods,andbytheNSF(DMR-9971432). (38)Szebeni,J.Crit.Rev.TherapeuticDrugCarrierSystems,57(39)Park,J.W.;Hong,K.;Kirpotin,D.B.;Papahadjopoulos,D.;Benz,C.C.ImmunoliposomesforCancerTreatment;AcademicPress:NewYork,1997;Vol.40.(40)Harokopakis,E.;Hajishengallis,G.;Michalek,S.M.,4299 Figure12.Fluorescenceintensityofpyrene-labeledactinasafunctionoftimeafteradding50mMKCland2mMMgCltoinitiatepolymerization.Thesymbolsrepresentthekineticsofbulkactinatconcentrationsof5M(emptycircles)and15M(plusses),whilethecurvesarethekineticsofactinencapsulatedinegg-PCvesiclesatconcentrationsof5M(solidline)and15M(dottedline).Theencapsulatedactindatahavebeenscaledbyaconstantconsistentwiththedilutionfactor.ProductionofUnilamellarVesiclesLangmuir,Vol.19,No.7,2003