warwick.ac.uk/lib-publications

Original citation: Hartlieb, Matthias, Floyd, Thomas, Cook, Alexander B., Sanchez-Cano, Carlos, Catrouillet, Sylvain, Burns, James A. and Perrier, Sébastien. (2017) Well-defined hyperstar copolymers based on a thiol–yne hyperbranched core and a poly(2-oxazoline) shell for biomedical applications. Polymer Chemistry. Permanent WRAP URL: http://wrap.warwick.ac.uk/87063 Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes this work of researchers of the University of Warwick available open access under the following conditions. Copyright © and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable the material made available in WRAP has been checked for eligibility before being made available. Copies of full items can be used for personal research or study, educational, or not-for-profit purposes without prior permission or charge. Provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way. Publisher statement: First published by Royal Society of Chemistry 2017 http://dx.doi.org/10.1039/C7PY00303J A note on versions: The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher’s version. Please see the ‘permanent WRAP URL’ above for details on accessing the published version and note that access may require a subscription. For more information, please contact the WRAP Team at: wrap@warwick.ac.uk

PolymerChemistry

ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx Polym.Chem.,2016,00,1-3|1

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Received00thJanuary20xx,Accepted00thJanuary20xx

DOI:10.1039/x0xx00000x

www.rsc.org/

Well-definedhyperstarcopolymersbasedonathiol-ynehyperbranchedcoreandapoly(2-oxazoline)shellforbiomedicalapplicationsMatthiasHartlieb,aThomasFloyd,aAlexanderB.Cook,aCarlosSanchez-Cano,aSylvainCatrouillet,a,†JamesBurns,cSébastienPerriera,b,*

Well defined ‘hyperstar’ copolymers were synthesized by combining hyperbranched polymers produced by thiol-ynechemistrywithpoly(oxazoline)s.ThehyperbranchedcorewaspreparedusinganAB2monomerandatrifunctionalalkene,applyingamonomer feedingapproach.Thedegreeofbranchingwashigh (0.9)whilemaintaining lowdispersities (1.3).Poly(2-ethyl-2-oxazoline) (PEtOx) functionalized by a thiol end groupwas coupled to the surface of the hyperbranchedstructureaccessingterminalalkyneunits.PEtOx-SHwasproducedbyterminationofthe livingpolymerizationwithethylxanthate and subsequent conversion to thiol under alkaline conditions. The degree of polymerization was variedproducingPEtOxwith23or42repeatingunits,respectivelywithadispersityaround1.1.Afterconjugationofthepolymerarms,hyperstarcopolymerswerecharacterizedbySEC,NMRspectroscopy,lightscattering,andAFM.Thepolymerswereabletoencapsulatethehydrophobicdyenile redwithinthecoreof thestructurewith loadingefficienciesbetween0.3and 0.9wt%. Cytotoxicity of the hyperstarswas assessed using A2780 human ovarian carcinoma cells resulting in IC50values around 0.7mgml-1. Successful internalization and colocalizationwith lysosomal compartmentswas observed byconfocalmicroscopystudies.

IntroductionTheuseofpolymerbaseddrugdeliverysystemshasreceivedincreasing interest in the lastdecades.1Besidenanoparticels2most prominently, polymeric micelles, composed ofamphiphilicdiblockcopolymershavebeenextensivelyusedtodeliverpayloads.3Thesepolymersareabletophasesegregatein aqueous solution forming hydrophobic compartmentsprotected by a hydrophilic shell. Drug molecules can beencapsulatedwithinthecoreofthemicelleandreleasedupondeliverytotheintendedtargettissues,bypassive(EPReffect)4or active targeting methods.5 However, micelles suffer fromtheir dynamic nature; they disassemble below the criticalmicelle concentration, which can lead to the prematurerelease of their payload upon administration.6 This issue canbe addressed by the use of unimolecular micelles, such asdendrimerspossessingahydrophilicshell,7astheyarenotabletodisassembleevenathighdilutions.However, thesynthesisand upscaling of these materials is highly demanding whencomparedtopolymericmicelles.Hyperbranched polymers are a promising alternative for

dendrimers, as they offer high degrees of branching (DB,usually ranging between 0.4 and 0.6) while being able to beproduced by relatively simple and scalable syntheticapproaches.8 Higher DB values can be achieved using AB2 orAB3monomers,multifunctional coremolecules, and efficientpolymerization reactions such as esterification9 or morerecently radical thiol-yne addition10 and copper mediatedazide-alkynecycloadditions(CuAAc).11,12A major drawback of hyperbranched polymers is theirrelatively high dispersity (Ð > 3.0).13 By using thiol-ynechemistry and a monomer feeding approach our group hasrecently developed a method to produce hyperbranchedpolymerswithveryhighdegreesofbranching(DB>0.8)whilemaintaining lowdispersities (Ð < 1.3).14However, inorder toimprove the application of these architectures as a drugdelivery vehicle, a hydrophilic shell can be added.15 As thesurfaceofhyperbranchedpolymers isdenselypopulatedwithfunctionalgroups,endfunctionalizedpolymerscanbecoupledtothedendriticpolymer,resultinginhyperstarcopolymers.Poly(2-oxazoline)s(POx)areapromisingpolymerclassforthesynthesisofhyperstarcopolymers.Smallsidechainderivativessuch as poly(2-methyl-oxazoline) or poly(2-ethyl-oxazoline)(EtOx)aregenerallyregardedasbiocompatible16-18andshowastealth effect comparable to poly(ethylene glycol).19-21 PEtOxeven shows higher circulation times and lower unspecificaccumulation in the body compared to PEG.22 POx can beproduced by cationic ring opening polymerization (CROP) ofoxazolinesofferingamultitudeofpotentialfunctionalities.23,24

ARTICLE JournalName

2 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Scheme1:Schematicrepresentationofthesynthesisofthiol-yneandPOxbasedhyperstarcopolymers.

The beneficial properties of POx based surface coatings,25micelles26,27starshapedcopolymers,28aswellasmicrogels,21,29 nanogels30-32 or capsules33-35 in a biomedical context havebeendemonstratedinmultiplereports.Thecombinationofdendrimersandhyperbranchedpolymersdecorated with POx arms has previously been described inliterature.36Generally,suchhyperstarscanbesynthesizedinacore-first approach,where themultiple functionalities of thesurfaceof thebranched structureareusedasan initiator forthe CROP. This was demonstrated by the use ofhyperbranched polymer based on poly(glycidol),37 poly vinylbenzyl chloride,38 or hydroxyphenyl valeric acid.39Alternatively,linearpolymerchainscanbeattachedinanarm-firstapproachasshownforglycogen,40poly(propylene imine)dendrimers,41orpoly(urea)dendrimers42byanendcappingofthelivingCROP.Furthermore,POxwithfunctionalitiesreactivetowards amines were used to decorate branchedpoly(ethylene imine),43 poly(amidoamine),44 and lysinedendrimers.45 Recently, it was shown that POx-basedhyperstars show great promise in vivo in the treatment ofcancer.46In this article, we describe the combination of the beneficialpropertiesofPOxwithnarrowdispersityandhighDBofthiol-yne based hyperbranched polymers as an easy way tohyperstar copolymers (Scheme 1). Thiol end functionalizedPOxwas produced using a two-step procedure involving endcapping of the living polymerization with ethyl xanthate andsubsequentaminolysis.AlthoughfunctionalizationofPOxwiththiol end groups has been described previously usingtermination with NaSH47 or thioacetate,48, 49 andfunctionalizationbyCuAAc,29theendcappingofalivingCROPwithaxanthate50followedbyaminolysis51offeresanovelandefficeint route to thiol functionalized polymers. Synthesizedhyperstarswere characterizedvia light scattering and atomicforce microscopy techniques. The potential as drug deliveryvectors of the resulting unimolecular nanoparticles wasillustratedbyloadingthemwiththehydrophobicnilered,usedas a drugmodel. The resultingmaterialswere found tohavemoderatecytotoxicityandcellinternalisationstudiesrevealedan endocytosis pathway, resulting in colocalization withlysosomal compartments. In contrast to previously describedhyperstar systems, the herein presented polymer does not

require tedious dendrimer synthesiswhile the product is stillwelldefined.

ResultsandDiscussion

HyperbranchedPolymersbythiol-ynechemistry

Thethiol-ynemonomer,prop-2-yn-1-yl3-mercaptopropanoate(PYMP), used to produce the hyperbranched core of thehyperstar copolymerswas synthesizedaccording to literatureprocedure.14 Briefly, 3,3-dithiodipropionic acid was activatedusing N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide (EDC)and 4-Dimethylaminopyridine (DMAP). Propargyl alcohol wasadded to the solution to yield the protected form of themonomer, di(prop-2-yn-1-yl) 3,3’-disulfanediyldipropionate(1). Deprotection was accomplished under reductiveconditions using dithiothreitol (DTT), yielding the thiol-ynebearing PYMP monomer (2). Both, precursor and monomerwerecharacterizedusingNMRspectroscopy(FigureS1).The synthesis of hyperbranched copolymers using PYMPwaspreviouslyoptimizedinourgrouptoyieldsystemswithahighdegreeofbranching,aswell as lowdispersities.14Key factorsare the use of an alkene core (triallyl 1,3,5-Benzenetricarboxylic acid) anda lowmonomer concentrationaccomplished by a feeding approach. In the presence of aphoto initiator, 2,2-dimethoxy-2-phenylacetophenone(DMPA),andasourceofUVlight,apolymerwithhighdegreeofbranching(0.9)andalowdispersitiy(1.29)wassynthesised.The degree of branching was determined by 1H-NMRspectroscopycomparingthesignalsofbranched,terminalandlinearunits(FigureS2).Size exclusion chromatography (SEC) measurements in DMFshowed a mono-modal distribution for refractive index andlight scattering detectors indicating the absence of crosslinking(FigureS3).APMMAcalibrationwasusedtodeterminethemolarmassofthepolymer(Mn=6,000gmol-1).Duetotheglobularstructureofthehyperbranchedpolymer,asignificantdifferenceinthehydrodynamicvolumecomparedtothelinearstandard used to calibrate themeasurements was expected.However,themeasurementillustratestheabsenceofcouplingreactionsduringthepolymerization.To determine the absolute molar mass of poly(PYMP) (3),staticlightscattering(SLS)inDMFwasemployed(FigureS4).

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|3

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Table1:CharacterizationdataofPEtOxwithvaryingDPvaluesafterendfunctionalizationusingethylxanthateandafteraminolysisandoxidation.

Sample Composition DPa Conv.a(%) DF(%)a,bMn,NMR

a

(gmol-1)Mn,SEC

c

(gmol-1)Ðc

4 PEtOx23-Xan 23 97 88 2,400 3,200 1.095 PEtOx42-Xan 42 94 89 4,300 5,900 1.126 PEtOx23-S-S-PEtOx23 - - - 4,800 4,900 1.107 PEtOx42-S-S-PEtOx42 - - - 8,600 11,700 1.11

a)Determinedfrom1HNMRspectroscopy.b)DF=DegreeofFunctionalization.c)DeterminedbySECinDMFusingaDRIdetectorandPMMAstandard.

Figure1:A)ESI-massspectrumofPEtOx23terminatedwithethylxanthate(4).B)Overlayofsimulatedandmeasured isotopicpattern forsinglechargedspecies.C) Overlay of simulated and measured isotopic pattern for double chargedspecies.

Usingtherefractiveindexincrement,dn/dCalsodeterminedinDMF (dn/dC = 0.102), a molar mass of 11,500gmol-1 wasdetermined. The difference between the values obtained bySLS and SEC illustrate the branched nature of synthesizedpolymer.After purification, the polymer was found to cross-link andgelatewithindayswhenstoredinbulk,potentiallyduetothepresence of residual photo initiator causing alkyne-alkynecoupling. This hampered the preparation of large batches ofthematerial for PEtOx conjugation. The issuewas addressedby removing residual photo initiator by precipitation, andstoring the polymer in solution (DMF, 20 mg mL-1) at -20˚C.Thepolymerprovedtobestableforatleastonemonthwhenstoredundertheseconditions(FigureS5).ThiolendfunctionalizedPoly(2-oxazoline)s

InordertoproducePEtOxabletobecoupledtothesurfaceofpoly(PYMP), thePOxhadtobe funcionalisedwitha thiolendgroup. To ensure quantitative end functionalization and adefined degree of polymerization (DP) the kinetics of thepolymerization had to be investigated. Knowledge about therate constant enables the prediction of conversion for aspecific polymerization. The polymerization of EtOx wascarried out in a schlenk-flask in acetonitrile under nitrogenatmosphereat78°C,aimingforaDPof50.Both,initiatorandmonomer were distilled to dryness prior to use. During thepolymerization,samplesweretakenatspecific timepoints toanalyseconversion,molarmassanddispersity.Theconversionwas determined byNMR spectroscopy comparing the signalsof theoxazoline ringand thepolymerbackbone.Molarmassdistribution was determined by SEC using chloroform as aneluent and poly(styrene) as a calibration standard. Theevolution of ln(M0/Mt) over time and the increase of molarmassasa functionof theconversionwere foundtobe linearindicating a living polymerization process (Figure S6). A rateconstantof5.6×10-3Lmol-1wasobtained.ThefunctionalizationofPEtOxwiththiolswasachievedviathetermination of the living polymerization process usingpotassiumethyl xanthate,whichwas used as aprecursor fortheaimed functionality (Scheme1).Afteradditionof the saltto the polymerizationmixture under a nitrogen atmosphere,immediate precipitation of potassium tosylate was observedindicatingasuccessfulendfunctionalization.

ARTICLE JournalName

4 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Figure2:SECtracesofPEtOx-Xanthate(4,5)aswellaspolymersafteraminolysisandoxidativecoupling(6,7).PMMAwasusedasacalibrationstandardforthemeasurements.

In order to study the influence of the chain length on thesolubilisationofhyperstars inwaterandtheirperformanceasdrugdeliveryvehicles,twodifferentPEtOx-xanthatepolymerstargetingaDPof25and50wereproduced(4,5).Aconversionof95%was targeted inorder toavoid side reactionsprior toendcappingand,atthesametime,reachthedesiredDP.TheconversionwasdeterminedbyNMRspectroscopybeforeendcapping by comparing the signals of the backbone with thesignalsofresidualEtOx.The DP, as well as the degree of functionalization (DF) withethyl xanthate was determined via NMR spectroscopy afterprecipitation, by comparison of the signal of the polymerbackbone with the peak of the methyl group at the α-chainend or with the methyl group of the xanthate, respectively(FigureS7). The obtained DP values (23 and 42) were lowerthan the targeted values (24 and 47.5), due to incompletemonomer conversion and experimental error in the analysis.The DFwas found to be higher than 85% for both polymers(Table1).To prove the insertion of the desired end group, ESI massspectrometry of polymerswithDP25wasperformed (Figure1). Two main distributions were observed, corresponding tothe

Figure3:A)ESI-massspectrometryofPEtOx23afteraminolysis(8).B)Overlayofsimulated and measured isotopic pattern for single charged. C) Overlay ofsimulatedandmeasuredisotopicpatternfordoublechargedspecies.

desired species with one or two charges, respectively. Theoverlay of the measured spectrum with simulated isotopicpatternprovedtheexistenceofpolymersinitiatedbyamethylgroupandhavingaxanthategroupattheω-chainend.Smallerdistributions in the enhanced region correspond to specieswithdifferentcounterionsand/orstateofcharge.Themolarmass,determinedbySECmeasurementswasintheexpectedrange,althoughaPMMAcalibrationwasused(Table1). Obtained dispersities were low, indicating a livingpolymerizationprocess.The xanthategroupwas converted intoa thiolvia aminolysisusing dimethylamine in ethanol at 40°C for 3h. The reactionresulted in the formation of a thiourethane and a thiol endfunctionalized polymer chain. The formed PEtOx-SH werefound to be susceptible to oxidation at ambient conditions.

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|5

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Table2:Characterizationofhyperbranchedpolymerandhyperstarcopolymers.

Sample SLSinDMF SLSinwater SEC NMR NileredloadingMa

a(gmol-1-)

Naggb Rh

(nm)Ma

a(gmol-1-)

Naggb Rh

(nm)Mn Ð Arms Mn

wt%

mol%

3 11,500 - 8 - - - 6,000 1.29 - - - -10 254,000 3.7 41 311,000 4.5 6 13,800 1.30 24 69,000 0.9 211 259,000 2.4 45 445,000 4.1 8 24,200 1.28 23 110,400 0.3 0.7

[a]ApparentmolecularweightdeterminedbytheSLS.[b]DeterminedbycomparisonofMa(SLS)andMn(NMR).

Figure 4: SEC measurements of poly(PYMP) (3), PEtOx (6, 7) and hyperstarcopolymers(10,11)inDMFusingaPScalibrationstandard.

SECmeasurementsafterpurificationshowed increasedmolarmassvaluesas compared to theprecursorpolymer.Thiswasattributed to oxidative chain coupling during the extractionprocess,whichwas supportedby thedoubledmolarmass ofPEtOx as indicated by SEC measurements. The mono-modaldistribution and low dispersities suggest a near quantitativethiolfunctionalizationofPEtOx(Figure2).ForPEtOx42,asmalltail at lowmolecularweight indicates the presence of eitherunfunctionalized or non-oxidized chains. As the control overthe polymerization decreases with increasing monomerinitiatorratios,thisbehaviourwasexpected.Nevertheless,themajorityofchainswerethiolend-functionalizedforbothDPs.NMR spectra clearly showed the disappearance of signalsattributed to ethyl xanthate, supporting a quantitativeaminolysis(FigureS7).To enable thiol-yne based coupling between thehyperbranched polymer and PEtOx chains, oxidized polymerswere reduced using DTT in dichloromethane. The reducingagent was removed via extraction with water under oxygenfreeconditions,andtheresultingpolymerswereanalysedvia

ESI mass spectrometry (Figure 3). Again, one single and onedouble charged main distribution were observed, bothattributed to thiol end functionalized PEtOx. Smallerdistributions in the enhanced region correspond to specieswithdifferentcounterionsand/orstateofchargeSynthesisandcharacterizationofhyperstarcopolymers

The conjugation of PEtOx and poly(PYMP) was performed inDMF using DMPA under UV light as a radical source. Thereaction was performed under nitrogen atmosphere topreventoxidationofPEtOx-SHandtheconsumptionofradicalsbyoxygen.Thenumberofalkynefunctionalitiesforanindividualcorewasdetermined by NMR spectroscopy and SLS. The Ma of thehyperbranched core was found to be 11,500gmol-1 by lightscattering. Subtracting the weight of the tri-allyl core (10%)results in a DP of 72 including terminal, linear and branchedunits. As 44 % of the monomer units were terminal, asdeterminedbyNMR,thetotalnumberofalkynefunctionalitiesper poly(PYMP) is 32. For coupling reactions, a 1:1 ratio ofPEtOx to alkyne functionality was used. This value shouldresult ina sufficientsurfacecoverageofhyperstarcopolymerto ensure water solubility. Remaining alkyne or alkenefunctionscanpotentiallybeusedfor further functionalizationreactions. The reaction mixture was diluted with water andfiltered to remove any insufficiently reacted poly(PYMP).ResidualPEtOxwasremovedbydialysisusingcentrifugefiltertubesusingamembranewitha10kDacutoff.TheobtainedhyperstarcopolymerswereanalysedusingSECinDMF (Figure 4, Table 2). The dispersities were found to besimilartothehyperbranchedstartingmaterialwhilethemolarmass increased significantly. A higher molar mass wasobtainedwhenusingPEtOxwithaDPof42ascomparedtoDP23.Thisindicatesthatasimilarnumberofpolymerarmswereattached in each case. The lowmolecularweight shoulderofpolymer 11 could be attributed to residual linear polymerwhichwasnotremovedquantitativelybedialysis.NMR spectroscopy was used to analyse the ratio betweenpoly(PYMP) and PEtOx by comparison of the monomermethylenesignalsat4.23and3.82ppmwiththesignalsofthePEtOx polymer backbone at 3.46ppm (Figure S8).

ARTICLE JournalName

6 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Figure5:AFMimagesofhyperstarsdrop-castedfromasolutionof0.1mgmL-1hyperstarinTHF.A)Compound10,5μmscansize,B)Compound10,1μmscansize,C)Heightprofileofcompound10,D)Compound11,5μmscansize,E)Compound11,1μmscansize,F)Heightprofileofcompound11.

AstheDPvalueisknownforbothinitialpolymers,thenumberof arms per hyperstar could be calculated. Both hyperstarshaveasimilarnumberofarms(24armsforcompound10and23 arms for compound 11), which are both lower than thetargetedvalueof32.This canbeexplainedbyan incompleteconversionofthethiol-ynereaction,possiblycausedbysterichindrance or by the presence of unfunctionalized PEtOx. Infact, some of the alkyne functionalities might be located onthe inside of the hyperbranched structure, inaccessible forpolymer chains. Themolarmassof the conjugates calculatedfrom these valueswas 69,000 and 110,000gmol-1,markedlyhigher than the molar mass obtained by SEC, which can beexplained by the hyperbranched nature of the polymer andthelackofahyperbranchedSECstandard.Staticlightscatteringwasemployedtodeterminetheabsolutemolarmassofthesynthesizedhyperstarcopolymers.Inorderto minimize possible aggregation, measurements were firstperformedinDMF,aspoly(PYMP),aswellasPEtOxaresolublein this solvent. Both hyperstar polymers were analysed atconcentrations between 10 and 0.5mgmL-1 (Figure S9 andS10). The resultsasdisplayed inTable2 showeda significantdifference compared to the values obtained by NMRspectroscopy. Both hyperstars were found to have a molarmassaround250,000gmol-1.InthecaseofPEtOx23,thisvaluewould correspond to a number of arms above 100, which ishigher than the available number of conjugation sites perpoly(PYMP).InthecaseofPEtOx42,quantitativeconversionoftheallalkynefunctionalitieswouldberequiredtoreachsuchamolecular weight. Therefore, the values are likely to be theresultofaggregationofmultiplehyperstars.This issupportedby the hydrodynamic radius of 40nm, which is larger thanexpected for non-aggregated hyperstars. Taking the molarmasscalculatedfromNMRasabasis,anumberofaggregation(Nagg)of4and2,respectively,werecalculated.Apossibleexplanation for thisbehaviourcanbe found in thepurification protocol (dialysis in aqueous medium) of the

hyperstarcopolymers.Sincetheconjugationwasperformedina1:1stoichiometryofalkyneandthiol,andgiventheresultingvinyl thiolether has an increased reactivity towards thiols ascomparedto thealkynemoieties,a reactionof twopolymerspersiteispossibledespitesterichindrance.Thiscouldleadtothe presence of hydrophobic patches on the surface of thehyperbranched structure, leading to a self-assembly inaqueoussolutionduetothehydrophobiceffectandapossibleentanglement, which persists even after re-solubilisation inDMF. This assumption is supported by a higher number ofaggregationforcompound10,whichhasshorterPEtOxarms,compared to compound 11. This steric stabilization is lesspronounced leading to a higher tendency to aggregate. Themono-modal size distribution in SEC can be explained by thehighshearforcesonthecolumn,breakingaparttheassembledstructures.SLSinwaterwasperformedtoelucidatethemolecularweightof the system in a polar solvent (Figure S11 and S12).Molecular weight, as well as Nagg were found to increase ascompared to measurements in DMF indicating an increasedtendency of assemble in water. Furthermore, thehydrodynamic radius is drastically decreased in water, whichcould be attributed to a collapse of the hydrophobic core,whileinDMFpoly(PYMP)swellsleadingtoanincreasedradius.DLSmeasurementsofthehyperstars(FigureS13)showasmallportion of large aggregates (around 50 nm) for sample 10whereasonlyonesharpdistribution isdetected forhyperstar11.This is inaccordancewithan increasedshieldingabilityofDP50PEtOxcomparedtotheshorterpolymer.Itisimportanttoemphasisethatinbothcasesassembledstructuresarewelldefinednano-aggregatesshieldedbyaPEtOxshell.Toillustratesizeandsizedistribution,atomic-forcemicroscopy(AFM)wasutilized to imagehyperstar copolymers (Figure5).SamplesweredissolvedinTHF,whichisexpectedtosolubilizeboth,coreandshell,anddropcastedontosilicon.Afterdrying,theheightprofilewasmeasuredusingtappingmode.Forboth

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|7

Pleasedonotadjustmargins

Pleasedonotadjustmargins

polymers the profile shows a height between 6 and 8 nm,which is consistentwith theexpectations fora softhyperstar

structuredepositedonasurface.Figure 6: Nile red loading of hyperstar copolymers. A) Absorbance (solid) andfluorescence (dashed)spectraofNile redsolubilizedbyvaryingconcentrationsofhyperstar10.B)Absorbance(solid)andfluorescence(dashed)spectraofNilered solubilized by varying concentrations of hyperstar 11. C and D)Determination of nile red loading per hyperstar by weight (C) or molar (D)concentrationusingtheabsorbanceat520nm.

From the size of the imaged structures, ranging between 20and50nm,minoraggregationoffewhyperstarmoleculesasaresultofthedryingprocesscanbeconcluded.

Dyeloadingofhyperstarcopolymers

The ability of the hyperstar copolymers to encapsulatehydrophobic molecules was investigated using nile red, afluorophore typically used as a model drug of intermediatelipophilicity(logP∼3–5)suchasthepotentchemotherapeuticPaclitaxel. As hyperstars are essentially unimolecularmicellesthe loading process could be conducted in THF, whichsolubilizes both fluorophore and hyperstar. To study theloading capacity of hyperstars varying amounts of polymerwere added to a constant amount of nile red. By slowevaporation of the solvent, nile red was forced into thehydrophobic domains (poly(PYMP)). Subsequently,waterwasadded to dissolve the fluorophore loaded hyperstars. As thePEtOxdecoratedpolymersare, incontrast tonile red, readilywater soluble, non-encapsulated nile red could be separatedbyfiltration.Todetermine theexactamountof loadeddye, sampleswerefreezedriedtoremovethewaterandre-dissolvedinTHF.Thiswasnecessary toexclude solvato-chromiceffectsaltering theabsorptionandemissionprofileasaresultoftheenvironmentwithin the hyperbranched core. To determine the amount ofencapsulated nile red, absorbance intensities were used andcomparedtoacalibrationoffluorophorealoneinTHF(FigureS14).Theabsorbanceat520nmwasrecordedindependenceof the concentration of hyperstar polymer, able to solubilizethedye(Figure6AandB).By plotting the weight concentration of nile red versus theweight concentration of hyperstar a loading capacity can bedetermined using the slope of the linear fit (Figure 6C).

Loading capacities of 0.9 and 0.3 wt% were obtained forhyperstars10and11,respectively.AsthecontentofPEtOxinpolymer 11 is higher as compared to polymer 10 a loweramountofencapsulateddye(inwt%)wasexpectedasalsothepercentage of hydrophobic compartment per hyperstar isdecreased.Based on the molecular weight determined by NMRspectroscopyandSLSamolar loadingcapacitywascalculated(2mol%forpolymer10and0.7mol%forpolymer11),whichisin perfect agreement with reports on other nile red loadedhyperstarcopolymers.52-54Bothhyperstarscopolymersshowasimilar loading efficiency, which was expected as thehydrophobiccompartmentisidentical.

Figure 7: Cell viability of A2780 human ovarian carcinoma cells treated withhyperstars10and11for72hat37˚C.

Biologicalbehaviourofhyperstarcopolymers

Synthetic polymer systems have to meet a number of keyrequirementstobeusedasdrugcarrier.Mostimportantly,thedelivery vector should show no or a low cytotoxicity againstmammalian cells. The ability of the hyperstar copolymers toinhibitthecellgrowthofA2780humanovariancarcinomacellswastestedusingtheSRBassay,andIC50valuesobtainedfor10and11wereabove0.5mgmL-1(Figure7),whichishigherthanconcentrations relevant for drug delivery applications. Theability of both hyperstar molecules to inhibit cell growth atconcentrations above 0.5mgmL-1, could be attributed toincomplete coverage of the hydrophobic surface of thehyperbranched polymer by the PEtOx. Exposed hydrophobicdomainscouldpotentiallyinteractwithcellularmembranesorinterfere with processes within the cell. Surprisingly, bothhyperstar copolymers inhibit the cell growth of A2780similarly,whereas the ratioofbiocompatiblePEtOx is slightlyhigherforcompound11.Additionally, confocal microscopy was used to assess thecellular uptake of dye loaded hyperstar copolymers. A2780cells were incubated for two hours with non-inhibitingconcentrations (0.1mg/mL-1) of 10 and 11. Nile redfluorescence within the cell showed that the hyperstars aretakenupreadilybycells(FigureS15).Fluorescencemicroscopy

ARTICLE JournalName

8 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

withdifferentorganellesstainingagentswasusedtoevaluatecellular localization.

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|9

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Figure8:ConfocalimagesofA2780humanovariancarcinomacellstreatedwithnileredloadedhyperstars(10,11)for2hat37˚Cataconcentrationof0.1mgmL-1.LysosomalcompartmentswerestainedusingLysoTracker®greenDND-26,nucleiwerestainedusingHoechst33258.

Figure9:ConfocalimagesofA2780humanovariancarcinomacellstreatedwithnileredloadedhyperstars(10,11)for2hat37˚Cataconcentrationof0.1mgmL-1.MitochondriawerestainedusingMitoTracker®greenFM,nucleiwerestainedusingHoechst33258.

Nuclei were stained using Hoechst 33258, and eitherlysosomes or mitochondria were labelled with LysoTracker®Green (Figure 8) or MitoTracker® Green (Figure 9),respectively.These experiments clearly show that hyperstars are able topass through the cellular membrane, but do not reach the

nucleus,beingconcentratedinthecytoplasmofthecell.Whentheirlocalizationiscomparedwithlysosomalstaining,astrongcolocalizationofgreenandredfluorescencewasobserved.Onthecontrary,MitoTracker®andnile redwasnot found in thesamecellularareas.These results suggestanaccumulation inthe lysosomal compartments and not in themitochondria of

ARTICLE JournalName

10 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

treated cells, and indicates an energy dependent uptakemechanism (e.g. endocytosis), which was expected from apolymerofthissize.Nevertheless, although the dye is not covalently coupled tothehyperstars,notallowingtomakeanyconclusivestatementabout the fate of10 and11, both hyperstar systems are notexpected to degrade under physiological conditions within arelevant time frame . The monomer units of thehyperbranchedcorecontainesterbonds,butthehydrophobicnatureof the structure shouldprevent anefficient lysosomaldegradation. However, as the hydrophobic compartment ofthe hyperstar is small compared to common nanoparticlesystems a translocation of the drug to hydrophobic domainswithincellularcompartmentsismorelikely.Theissuecouldbeaddressedinthefuturebyincludingresponsivebuildingblockstothepolymertointroduceastimuliresponsivedrugrelease.

ExperimentalpartMaterialsandInstrumentation

Propargyl alcohol, 3,3- dithiodipropionic acid, dithiothreitol(DTT), 4-(dimethylamino)- pyridine (DMAP), and 2,2-dimethoxy-2-phenylacetophenone(DMPA)wereallpurchasedfrom Sigma-Aldrich. 1-Ethyl-3-(3-(dimethylamino)propyl)carbodiimide hydrochloride (EDC·HCl)waspurchasedfromIrisBiotech.TriethylaminewaspurchasedfromFischerScientific.Triallyl1,3,5-benzenetricarboxylatewaspurchased from Acros. All other materials were purchasedfromFisherScientific,Sigma-Aldrich,Merck,Fluka,andAcros.2-Ethyl-2-oxazoline (EtOx) andmethyl tosylate (MeTos) weredistilled to dryness prior to use. EtOx was dried using CaHbeforedistillation.1H-NMR spectra were measured using a Bruker DPX-300 orDPX-400 NMR spectrometer which operated at 300.13 and400.05 MHz, respectively. The residual solvent peaks wereusedasinternalreferences.Deuteratedchloroform(CDCl3)(δH=7.26ppm)wasusedassolventforallmeasurements.SECmeasurementswereperformedonanAgilent390-LCMDSinstrument equipped with differential refractive index (DRI),viscometry (VS), dual angle light scatter (LS) and dualwavelengthUVdetectors. The systemwasequippedwith2 xPLgelMixedDcolumns(300x7.5mm)andaPLgel5µmguardcolumn. The eluent is DMF with 5 mmol NH4BF4 additive.Samples were run at 1 ml/min at 50˚C. Poly(methylmethacrylate) standards (Agilent EasyVials) were used forcalibration.ForSECmeasurements inchloroform,anAgilent390-LCMDSinstrumentwithdifferential refractive index (DRI), viscometry(VS), dual angle light scatter (LS) and two wavelength UVdetectors. The systemwas equippedwith 2 x PLgelMixed Dcolumns(300x7.5mm)andaPLgel5µmguardcolumn.TheeluentisCHCl3with2%TEA(triethylamine)additive.Sampleswere run at 1mLmin-1 at 30 °C. Poly(methylmeth-acrylate),and polystyrene standards (Agilent Easy Vials) were used forcalibration.

Analytesampleswerefilteredthroughanylonmembranewith0.22μmporesizebeforeinjection.Respectively,experimentalmolarmass(Mn,SEC)anddispersity(Đ)valuesofsynthesizedpolymers were determined by conventional calibration usingAgilentGPC/SECsoftware.Electrospray Ionisation (ESI) measurements were obtainedusingaBrukerMicroToFandtheresultsanalysedusingBrukerData Analysis. Samples were dissolved in methanol at aconcentrationof1µgmL-1.AFM pictures were taken on an Asylum Research MFP-3DStandAloneatomicforcemicroscopewithextendedz-rangeof40μm,with closed loop scanning in x and y over a range of90μm.Tappingmodewasusedforallmeasurements.Di(prop-2-yn-1-yl)3,3’-disulfanediyldipropionate(1)3,3-dithiodipropionic acid (10.0g, 47.56mmol), N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride(EDC HCl) (21.88g, 114.14mmol) and 4-Dimethylaminopyridine (DMAP) (1.14 g, 9.52 mmol) wereadded to a round bottom flask containing 250 mL ofdimethylformamide (DMF) cooled in an ice bath and left tostir.Propargylalcohol (6.4g,114.14mmol)wasslowlyaddedto the solution and the RBF removed from the ice bath andallowed to reach roomtemperaturebefore leaving to stir for70 hr. The DMFwas removed and the concentrated productwas re-dissolved inDCM (250mL). TheDCMphasewas thenwashedwithwater(3x100mL)followedbywasheswithHCl(1x100mL),NaOH(1x100mL)andwater(1x100mL)beforedryingoverNa2SO4.DCMwasremovedbyrotaryevaporationandtheproductpurifiedbycolumnchromatographytoyieldayellow-orange oil, di(prop-2-yn-1-yl) 3,3’-disulfanediyldipropionate(9.8g,34.2mmol,72.1%).1HNMR(400MHz,CDCl3)δppm:4.74(s,2H,O−CH2≡C),2.97(t, 2H, C(O)−CH2−CH2), 2.82 (t, 2H, CH2−CH2−S), 2.51 (s, 1H,C≡CH).13CNMR(100MHz,CDCl3)δppm:170.91(C=O),76.84(CH2−C≡CH), 75.25 (C≡CH), 52.32 (O−CH2−C≡), 33.86(C(O)−CH2−CH2),32.79(CH2−CH2−S).Prop-2-yn-1-yl3-mercaptopropanoate(2)Di(prop-2-yn-1-yl) 3,3’-disulfanediyldipropionate (0.5 g, 1.75mmol)wasdissolvedindichloromethane(DCM)(7mL)beforeaddition of dithiothreitol (DTT) (1.17 g, 7.66 mmol) andtrimethylamine (1.12 mL, 8.00 mmol). The solution wasdeoxygenatedbypurgingwithnitrogenfor10minutesbeforebeing left to stir at room temperature for 2 hr. The solutionwasthenwashedwithHCl (1x10mL)andwater (1x10mL)andthendriedoverNa2SO4andthesolventremovedresultingin a yellowoil, prop-2-yn-1-yl 3-mercaptopropanoate (0.14 g,0.97mmol,28%), thatwasstoredundernitrogentopreventformationofdisulfides.1HNMR(400MHz,CDCl3)δppm:4.71(s,2H,O−CH2−C≡),2.79(m,2H,C(O)−CH2−CH2),2.71(m,2H,CH2−CH2−SH),2.49(s,1H,C≡CH),1.66(t,1H,CH2−SH).

13CNMR(100MHz,CDCl3)δppm:170.81 (C=O), 76.84 (CH2−C≡CH), 75.20 (C≡CH), 52.15(O−CH2−C≡),38.16(C(O)−CH2−CH2),19.56(CH2−CH2−SH).

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|11

Pleasedonotadjustmargins

Pleasedonotadjustmargins

Preparationofbranch-poly(prop-2-yn-1-yl3-mercaptopropanoate)(3)Prop-2-yn-1-yl 3-mercaptopropanoate (100mg, 0.69mmol)and 2,2-dimethoxy-2-phenylacetophenone (DMPA) (2mg,16.7µmol) were dissolved in dimethylformamide (DMF)(318μL). The solutionwasdeoxygenatedbynitrogenpurgingand added to a 250μL Hamilton Gastight glass syringe withstainless steel cannula and wrapped in aluminium foil. In asecond vial, 1,3,5-Benzenetricarboxylic acid triallyl ester(11.46mg, 34.7µmol) and DMPA (1mg, 8.4µmol) weredissolvedinDMFanddeoxygenatedbynitrogenbubbling.TheHamiltonsyringewasplacedintoanautomatedpumpandtheneedlefedintothesecondvialcontainingthecore,whichwasthenplacedinanaluminiumfoillinedboxandcoveredwitha365nmUV lamp (UVP,UVGL-55, 6watt). Themonomerwasthen fed into the vial at a rate of 10.45μLmin-1. After thefeedingfinishedtheneedlewasremovedandthevialwasleftunder UV light for a further 1hour. The solution was thenprecipitatedintodiethyletherandcentrifugedtoseparatethepolymerfromsolvent.Thediethyletherwasdecantedandthepolymerdriedwithnitrogentoremoveanyresidualsolvent.Astock solution of the hyperbranched core indimethylformamide(20mgmL-1)waspreparedandstoredinafreezer.SEC(DMF,NH4BF4):Mn=6,000gmol-1,Ð=1.29;Kineticstudyofpoly(oxazoline)polymerisationDry ethyl oxazoline (EtOx), methyl tosylate (MeTos) andacetonitrilewereaddedtoaschlenkflaskundernitrogenandleft to stir in an oil bath at 80 °C. At pre-determined timeintervals,aliquotsofthereactionmixturewereremovedusinga pre heated needle and analysed by SEC and NMRspectroscopy.Conversions,aswell asSECdataarepresentedinTable3.

Table3:AnalyticaldataofPEtOXkineticsamples.

Time(min) Conversion(%)Mn,SEC

(gmol-1)Ð

10 4.3 500 1.0430 9.1 700 1.1260 28.0 1,200 1.17120 61.5 2,000 1.17180 76.0 2,400 1.22240 93.1 3,000 1.09300 93.9 2,700 1.24

PreparationofXanthateterminatedPoly(oxazoline)s(4,5)Dry ethyl oxazoline (EtOx), methyl tosylate (MeTos) andacetonitrile (Table 4) were added to a Schlenk flask undernitrogen and left to stir in an oil bath at 80 °C. After apredetermined time, the solution was removed from the oilbath,asamplefordeterminationoftheconversionwastaken,and potassium ethyl xanthate was added to terminate thepolymer chain. The product was left to stir for 1 hr beforebeingdilutedwithchloroform(100mL).Theorganiclayerwasthenwashedwithsat.Na2CO3(3x100mL)andbrine(3x100mL) thendriedoverMgSO4. The chloroformwas removed to

leaveacolourlessoil.Theoilwasre-dissolvedinDCM(10mL)and precipitated into ether and the polymer collected bygravityfiltrationasawhitesolid.Thewhitesolidwasagainre-dissolved intoDCMwhichwasevaporated.Subsequently, thepolymerwasdriedundervacuumtoyieldawhitesolid.

Table4:CompositionofthereactionsmixturesforthesynthesisofxanthateendfunctionalizedPEtOx.

SampleTarget

DP

EtOx

(mL)

MeTos

(mL)

Acetonitrile

(mL)

Ethyl

Xanthate

(mg)

Time

(min)

4 25 4.058 0.243 7.699 309 186

5 50 4.038 0.121 5.841 462 311

1H NMR (5, 300MHz, CDCl3) δ ppm: 3.75 – 3.13 (m, 168 H,backbone),3,10 -2.92 (m,3H,Methylgroup (α-end)),2.54 -2.13 (m, 87 H, CH2 side chain), 1.44 (t, 2.7 H,Methyl group(xanthate)),1.23–0.98(m,124H,CH3sidechain);SEC(CHCl3,trimethylamine,PScalibration):Mn=5,900gmol-1,Ð=1.12;ESI-ToF (4): measured: 1843.039 m/z (M+K+), simulated:1843.130m/z;PreparationofPEtOx-S-S-PEtOx(6,7)PEtOx-xanthate (1.4 g, 53.8 mmol) was dissolved indiethylamine (33% in EtOH) (20mL) and stirred at 40°C for 3hr. The reactionmixturewas then poured over a solution ofsulphuric acid (H2SO4) (20 mL) and ice-water (200 mL). Thepolymer was extracted by washing with CHCl3 (3 x 50 mL),followed bywashing of the CHCl3 layerwith sat.Na2CO3 (3 x100mL) and brine (3 x 100mL). The organic layerwas thendriedoverMgSO4and filteredbefore thesolvent removedtoyielddisulfide-poly(oxazoline).1H NMR (7, 300MHz, CDCl3) δ ppm: 3.75 – 3.13 (m, 166 H,backbone),3,10 -2.92 (m,3H,Methylgroup (α-end)),2.54 -2.13(m,88H,CH2sidechain,1.23–0.98(m,123H,CH3sidechain);SEC(CHCl3,trimethylamine,PScalibration):Mn=11,700gmol-1,Ð=1.11;Preparationofthiol-terminatedpoly(oxazoline)(8,9)Disulfide-poly(oxazoline)(130mg,26.3µmol)wasdissolvedinDCM(5mL)beforeadditionofDTT(17.8mg,115.7µmol)andtriethylamine (16.8µL, 120.9µmol). The solution wasdeoxygenatedbypurgingwithN2for10minutes,beforebeingleft to stir for 2hr. The solution was then washed withdeoxygenated HCl (1 x 10 mL), sat. Na2CO3 (1 x 10 mL) andbrine(1x10mL)andthendriedoverNa2SO4andthesolventremoved, yielding the thiol terminated poly(oxazoline). ThepolymerwasstoredunderN2topreventoxidationofthethiolbacktodisulfide.ESI-ToF (4): measured: 2151.313 m/z (M+K+), simulated:2151.424m/z;Preparationofbranch-poly(prop-2-yn-1-yl3-mercaptopropanoate)-star-poly(oxazoline)(10,11)

ARTICLE JournalName

12 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

A stock solution of poly(oxazoline) in DMF (200mg/mL) waspreparedandaddedtothehyperbranchedcorestocksolution(20mgmL-1)ata0.66:1ratioofpoly(oxazoline):alkynegroups.DMPA was added (Table 5) and the resulting solutiondeoxygenated, by N2 purging, for 10 minutes before beingplaced under a 365nm UV lamp (UVP, UVGL-55, 6 watt) for24hr. The solution was then precipitated into diethyl etherand the resulting precipitate recovered before beingredissolvedinDMFandprecipitatedintowatertoremoveanyresidual hyperbranched core. The excess poly(oxazoline) wasremovedbyusinganappropriatelysizedcentrifugemembranefilter(10kDaMWCOforDP25POx,30KDaMWCOforDP50).The solution was passed through the membrane 5 times toensure complete removal of the poly(oxazoline) and thenplacedonthefreeze-dryerovernighttoremovethewaterandleavetheresultinghyperstarpolymer.SEC(10,DMF,NH4BF4):Mn=13,800gmol-1,Ð=1.30;

Table 5: Composition of reaction mixtures for the synthesis of hyperstarcopolymers.

Sample Poly(PYMP)(mg)

PEtOx(mg)

DMPA(mg)

DMF(mL)

10 48.8 300 19.7 3.94

11 22.8 280 12.7 2.54

StaticlightscatteringThe incremental refractive index, dn/dC, was determined bymeasuringtherefractiveindexofthepolymeroverarangeofconcentrations. The RI was determined using a Shodex RIdetector,operatingatawavelengthof632nm.Multiplyingthegradient,of theplotofRI vs conc.,by the refractive indexofthesolvent(water=1.3325)anddividingbytheRIconstantoftheinstrument(-1,398,000)givesthedn/dCofthepolymer.

Table6:Incrementalrefractiveindicesofpoly(PYMP)andhyperstarcopolymers.

Sample dn/dCinwater dn/dCinDMF

3 - 0.102

10 0.17 0.093

11 0.16 0.101

Light scattering measurements were obtained using an ALV-CGS3 system operating with a vertically polarized laser withwavelengthλ=632nm.Themeasurementsweretakenat20°Coverarangeofscatteringwavevectors(q=4πnsin(θ/2)/λ,withθ theangleofobservationandn the refractive indexofthesolvent).TheRayleighratio,Rθ,wasdeterminedusingeq.1,

(1)

whereIsolution,IsolventandItoluenearethescatteringintensitiesofthesolution,solventandreference(toluene)respectively,n isthe refractive index (nwater = 1.333, ndmf = 1.431, ntoluene =

1.496)andRtoluenetheRayleighratiooftoluene(Rtoluene=1.35x10-5cm-1forλ=632.8nm).The optical constant, K, is defined by eq. 2, where Na isAvogadro number and dn/dC is the incremental refractiveindex.

(2)

Inallcasesitwasverifiedthattheapparentradiusofgyrationofthesystemsverifiedq×Rg<1.TheZimmapproximationcanthusbeusedtoobtaineq.3.PlottingKC/Rθasafunctionofq2foreachconcentrationyieldedtheapparentradiusofgyrationRgofthescatterersaswellastheirapparentmolecularweightextrapolated to zero angle, Ma. Representative plots areshownintheSupportingInformation.

(3)

AtagivenconcentrationtheRayleighratio,Rθ,isrelatedtotheapparentmolecularweightofthesample,givenbyeq.3. It isonly at infinite dilutions, where the interactions betweenscatteringparticlesarenegligible,thattheapparentmolecularweight is equal to the true molecular weight.55 Multipleconcentrationsweremeasuredandaplotof linearregressionused to determine the apparent molecular weight at aconcentrationof0mgmL-1.FluorescenceandAbsorbancemeasurementsFluorescence spectra were measured using a Cary EclipseFluorescence Spectrophotometer fromAgilent. Sampleswereexcited at a wavelength of 520nm and the fluorescenceintensity was recorded between 530 and 700 nm in 2nmintervals. Absorbance spectrawere obtained using a Cary 60UV-VisspectrometerfromAgilent.Toprobenile reduptakeofhyperstarsnile redwasdissolvedin THF at a concentration of 0.796mgmL-1. 40μL of thatsolution(100nmol)wereaddedtoavialcontaininghyperstarinTHF(1.96mL)resultinginafinalnileredconcentrationof50μmol L-1. The concentration of hyperstar was varied startingwith an overall concentration of 1mgmL-1. Differentconcentrations were obtained performing a serial 2 folddilution. The THF was left to evaporate under ambientconditionsover24h.Subsequently,water(2mL)wasaddedtoeachvialandstirredfor2htosolubilizehyperstarcopolymers.Thewaterwasfilteredusing0.2μmsyringefiltersand1.5mLwerefreezedried.TheresultingsolidwasdissolvedinTHFandabsorbanceaswellasfluorescencewasrecorded.Nileredcalibrationwasperformedusinga50μmolL-1solutionof the dye in THF and a 2 fold serial dilution approach. Theintensitywasdeterminedat520nm.CellCultureA2780 human ovarian carcinoma cells were used betweenpassages 5 and 25 and grown in Roswell Park MemorialInstitutemedium(RPMI-1640)supplementedwith10%offetal

2( ) ( )

( )solution solvent solvent

toluenetoluene toluene

I I nR RI nθ

θ θθ

⎛ ⎞−= ⋅ ⋅⎜ ⎟

⎝ ⎠

22 2

4

4 solvent

a

n nKN C

πλ

∂⎛ ⎞= ⎜ ⎟∂⎝ ⎠

2 21 13g

a

q RKCR Mθ

⎛ ⎞⋅= ⋅ +⎜ ⎟⎜ ⎟

⎝ ⎠

JournalName ARTICLE

Thisjournalis©TheRoyalSocietyofChemistry20xx J.Name.,2013,00,1-3|13

Pleasedonotadjustmargins

Pleasedonotadjustmargins

calf serum, 1% of 2mM glutamine and 1%penicillin/streptomycin. The cells were grown as adherentmonolayersat310K ina5%CO2humidifiedatmosphereandpassagedatapproximately70-80%confluence.InvitrogrowthinhibitionassaysThe antiproliferative activity of polymers 10 and 11 wasdetermined in A2780 ovarian cancer cells. Briefly, 96-wellplateswereusedtoseed5000cellsperwell.Theplateswereleft topre-incubatewithdrug-freemediumat310K for24hbefore adding different concentrations of the compounds tobe tested (2mgml-1-200ngml-1). A drug exposure period of72hwas allowed. The SRB assaywas used to determine cellviability. GI50 values, as the concentration which causes 50%celldeath,weredeterminedasduplicatesoftriplicatesintwoindependentsetsofexperimentsandtheirstandarddeviationswerecalculated.ConfocalmicroscopyA2780 ovarian carcinoma cells were seeded in 8-wellmicroscopy chambers (15000 cells/well; 200 µl phenol redfreeDMEMmedium),lefttoattachfor24h,andthentreatedforanother2hwith0.1mg/mlofthehyperbranchedpolymersloadedwithNileRed(λex=520nm/λem=590nm).CellnucleiwerestainedusingHoechst33258(2.5µg/ml;30minat37C;λex = 352 nm/λem = 461 nm), and either lysosomes ormitochondriawere labelledwith LysoTracker®GreenDND-26(50 nM; 30 min at 37C; λex = 504 nm/λem = 511 nm) orMitoTracker® Green FM (400 nM; 30 min at 37C; λex = 490nm/λem=516nm), respectively.Cellswere thenwashedwithPBS and fresh phenol red free medium added. Cells wherethen imaged using a Leica SP5 or a Zeiss LSM 880 confocalmicroscope. Images were open and processed using FIJIImageJpackage.

Conclusion&OutlookIn summary,wewereable toproducewell-definedhyperstarcopolymers by the combination of thiol-yne and POxchemistries. The core was based on narrow dispersity (1.29)thiol-yne hyperbranched polymer with a high degree ofbranching (0.9), and the shell produced from water solublePoly(2-Ethyl-2-oxazoline) (PEtOx) chains bearing thiol endgroups.Inordertostudytheinfluenceofthehydrophilicshellon the properties of the resulting hyperstar, the DP of thePEtOxwasvaried.The thiolendgroupwasgenerated fromaxanthate moiety, which was used to terminate the livingpolymerization of EtOx. Conjugation was performed using aradicalbasedthiol-yneadditionprocess.Hyperstar copolymers were characterized using SEC, NMRspectroscopyandSLSinordertoassessstructuralinformation.MinoraggregationwasfoundtobepresentinDMFaswellasin water for PEtOx shells with DP 25 and 50 likewise, whichwas ascribed to the presence of remaining hydrophobicpatches on the hyperstar as well as to entanglement during

purification. For both DPs of PEtOx the number of arms perhyperbranchedcorewasaround20.Theencapsulationofnileredintothehyperbranchedcoreofthepolymerswasstudiedand found to be dependent on the nature of the shell withvaluesbetween1and2mol%.Inordertoevaluatethepotentialoftheproducedpolymersasdrug delivery vectors, cytotoxicity was assessed using A2780humanovarian carcinoma cells. IC50 valueswere found tobearound 0.7mgmL-1 regardless of the length of the PEtOxconstituting the shell. This concentration is well above thedoseofpolymerusedinadrugdeliveryapplication.Hyperstarswerefoundtobeinternalizedintocellsandcolocalizedwithinlysosomalcompartments.Future studieswill focus on the implementation of cleavableunitswithinthehyperbranchedcore,abletotriggerarelease,as well as on further functionalization of the system usingresidualalkynesoralkenes.

AcknowledgementsMHgratefullyacknowledgestheGermanResearchFoundation(DFG,GZ:HA7725/1-1)forfunding.TheRoyalSocietyWolfsonMerit Award (WM130055; S.P.), Monash-Warwick Alliance(S.P.), and Syngenta (A.C.) are acknowledged for financialsupport. CSC gratefully acknowledges Cancer Research UK(C53561/A19933).

References1. T.L.DoaneandC.Burda,Chem.Soc.Rev.,2012,41,2885-

2911.2. M. E. Davis, Z. Chen and D. M. Shin, Nat. Rev. Drug

Discov.,2008,7,771-782.3. H. Cabral and K. Kataoka, J. Control. Release, 2014,190,

465-476.4. H.Maeda,K.GreishandJ.Fang, inPolymerTherapeutics

II, eds. R. Satchi-Fainaro and R. Duncan, Springer BerlinHeidelberg,2006,vol.193,ch.26,pp.103-121.

5. L.Dai, J. Liu, Z. Luo,M. Li andK.Cai, J.Mater. Chem.B,2016,4,6758-6772.

6. V.P.Torchilin,J.Control.Release,2001,73,137-172.7. P. Kesharwani, K. Jain and N. K. Jain, Prog. Polym. Sci.,

2014,39,268-307.8. C.GaoandD.Yan,Prog.Polym.Sci.,2004,29,183-275.9. E. Malmstroem, M. Johansson and A. Hult,

Macromolecules,1995,28,1698-1703.10. D. Konkolewicz, A. Gray-Weale and S. Perrier, J. Am.

Chem.Soc.,2009,131,18075-18077.11. Y. Shi, R.W.Graff, X. Cao, X.Wang andH.Gao,Angew.

Chem.Int.Ed.,2015,54,7631-7635.12. L.Zou,Y.Shi,X.Cao,W.Gan,X.Wang,R.W.Graff,D.Hu

andH.Gao,PolymerChemistry,2016,7,5512-5517.13. C. Gao, D. Yan and H. Frey, inHyperbranched Polymers,

John Wiley & Sons, Inc., 2011, DOI:10.1002/9780470929001.ch1,pp.1-26.

14. A. B. Cook, R. Barbey, J. A. Burns and S. Perrier,Macromolecules,2016,49,1296-1304.

15. M. C. Lukowiak, B. N. S. Thota and R. Haag, Biotechnol.Adv.,2015,33,1327-1341.

ARTICLE JournalName

14 |J.Name.,2012,00,1-3 Thisjournalis©TheRoyalSocietyofChemistry20xx

Pleasedonotadjustmargins

Pleasedonotadjustmargins

16. J. Kronek, Z. Kroneková, J. Lustoň, E. Paulovičová, L.Paulovičová andB.Mendrek, J.Mater. Sci.Mater.Med.,2011,22,1725-1734.

17. J.Kronek,E.Paulovičová,L.Paulovičová,Z.KronekováandJ.Lustoň,J.Mater.Sci.Mater.Med.,2012,23,1457-1464.

18. R. Luxenhofer, G. Sahay, A. Schulz, D. Alakhova, T. K.Bronich,R.JordanandA.V.Kabanov,J.Control.Release,2011,153,73-82.

19. M.C.Woodle,C.M.EngbersandS.Zalipsky,BioconjugateChem.,1994,5,493-496.

20. S. Zalipsky, C. B. Hansen, J. M. Oaks and T. M. Allen, J.Pharm.Sci.,1996,85,133-137.

21. M.Platen,E.Mathieu,S.Lück,R.Schubel,R.JordanandS.Pautot,Biomacromolecules,2015,16,1516-1524.

22. L.Wyffels,T.Verbrugghen,B.D.Monnery,M.Glassner,S.Stroobants, R. Hoogenboom and S. Staelens, J. Control.Release,2016,235,63-71.

23. B. Guillerm, S. Monge, V. Lapinte and J.-J. Robin,Macromol.RapidCommun.,2012,33,6000-6016.

24. K. Lava, B. Verbraeken andR.Hoogenboom,Eur. Polym.J.,2015,65,98-111.

25. M. N. Leiske, M. Hartlieb, C. Paulenz, D. Pretzel, M.Hentschel,C.Englert,M.GottschaldtandU.S.Schubert,Adv.Funct.Mater.,2015,25,2458-2466.

26. Z.He,X.Wan,A.Schulz,H.Bludau,M.A.Dobrovolskaia,S.T.Stern,S.A.Montgomery,H.Yuan,Z.Li,D.Alakhova,M. Sokolsky, D. B. Darr, C. M. Perou, R. Jordan, R.Luxenhofer and A. V. Kabanov, Biomaterials, 2016, 101,296-309.

27. J.Li,Y.Zhou,C.Li,D.Wang,Y.Gao,C.Zhang,L.Zhao,Y.Li,Y.LiuandX.Li,BioconjugateChem.,2015,26,110-119.

28. K.Knop,D.Pretzel,A.Urbanek,T.Rudolph,D.H.Scharf,A. Schallon,M.Wagner, S. Schubert,M.Kiehntopf,A.A.Brakhage, F. H. Schacher and U. S. Schubert,Biomacromolecules,2013,14,2536-2548.

29. S. Lück, R. Schubel, J. Rüb, D. Hahn, E. Mathieu, H.Zimmermann,D. Scharnweber, C.Werner, S. Pautot andR.Jordan,Biomaterials,2016,79,1-14.

30. M. Hartlieb, D. Pretzel, M. Wagner, S. Hoeppener, P.Bellstedt, M. Gorlach, C. Englert, K. Kempe and U. S.Schubert,J.Mater.Chem.B,2015,3,1748-1759.

31. C.Legros,A.-L.Wirotius,M.-C.DePauw-Gillet,K.C.Tam,D. Taton and S. Lecommandoux, Biomacromolecules,2015,16,183-191.

32. V.M. Gaspar, P. Baril, E. C. Costa, D. deMelo-Diogo, F.Foucher,J.A.Queiroz,F.Sousa,C.PichonandI.J.Correia,J.Control.Release,2015,213,175-191.

33. K.Kempe,S.L.Ng,S.T.Gunawan,K.F.NoiandF.Caruso,Adv.Funct.Mater.,2014,24,6187-6194.

34. S. T.Gunawan, K. Kempe, T. Bonnard, J. Cui, K.Alt, L. S.Law, X. Wang, E. Westein, G. K. Such, K. Peter, C. E.Hagemeyer and F. Caruso,Adv.Mater., 2015, 27, 5153-5157.

35. M.Hartlieb,K.KempeandU.S.Schubert,J.Mater.Chem.B,2015,3,526-538.

36. P. Wilson, P. Chun Ke, T. P. Davis and K. Kempe, Eur.Polym. J., DOI:http://dx.doi.org/10.1016/j.eurpolymj.2016.09.011.

37. A.Kowalczuk,J.Kronek,K.Bosowska,B.TrzebickaandA.Dworak,Polym.Int.,2011,60,1001-1009.

38. F. Däbritz, A. Lederer, H. Komber and B. Voit, J. Polym.Sci.,PartA:Polym.Chem.,2012,50,1979-1990.

39. R.Weberskirch, R. Hettich, O. Nuyken, D. SchmaljohannandB.Voit,Macromol.Chem.Phys.,1999,200,863-873.

40. A.Pospisilova,S.K.Filippov,A.Bogomolova,S.Turner,O.Sedlacek, N.Matushkin, Z. Cernochova, P. Stepanek andM.Hruby,RSCAdvances,2014,4,61580-61588.

41. H. M. L. Lambermont-Thijs, M. W. M. Fijten, U. S.Schubert and R. Hoogenboom,Aust. J. Chem., 2011,64,1026-1032.

42. R. B. Restani, J. Conde, R. F. Pires, P. Martins, A. R.Fernandes,P.V.Baptista,V.D.B.BonifácioandA.Aguiar-Ricardo,Macromol.Biosci.,2015,15,1045-1051.

43. C. Zhang, S. Liu, L. Tan, H. Zhu and Y. Wang, J. Mater.Chem.B,2015,3,5615-5628.

44. S. Park Yong, S. Kang Yong and J. Chung Dong, Journal,2002,2,211.

45. R.M.England,J.I.Hare,P.D.Kemmitt,K.E.Treacher,M.J.Waring,S.T.Barry,C.AlexanderandM.Ashford,Polym.Chem.,2016,7,4609-4617.

46. R.M.England,J.I.Hare,J.Barnes,J.Wilson,A.Smith,N.Strittmatter, P. D. Kemmitt,M. J.Waring, S. T. Barry, C.Alexander and M. B. Ashford, J. Control. Release, 2017,247,73-85.

47. Y.Shimano,K.SatoandS.Kobayashi,J.Polym.Sci.,PartA:Polym.Chem.,1995,33,2715-2723.

48. O. Koshkina, D. Westmeier, T. Lang, C. Bantz, A.Hahlbrock, C. Würth, U. Resch-Genger, U. Braun, R.Thiermann,C.Weise,M.Eravci,B.Mohr,H.Schlaad,R.H.Stauber,D.Docter,A.BertinandM.Maskos,Macromol.Biosci.,2016,16,1616-5195.

49. G.-H. Hsiue, H.-Z. Chiang, C.-H. Wang and T.-M. Juang,BioconjugateChem.,2006,17,781-786.

50. V. R. de la Rosa, Z. Zhang, B. G. De Geest and R.Hoogenboom,Adv.Funct.Mater.,2015,25,2511-2519.

51. M. Le Neindre, B.Magny and R. Nicolay, Polym. Chem.,2013,4,5577-5584.

52. C.S.Popeney,M.C.Lukowiak,C.Böttcher,B.Schade,P.Welker,D.Mangoldt,G.Gunkel,Z.GuanandR.Haag,ACSMacroLetters,2012,1,564-567.

53. E. Burakowska and R. Haag,Macromolecules, 2009, 42,5545-5550.

54. G.ChenandZ.Guan,J.Am.Chem.Soc.,2004,126,2662-2663.

55. W. Brown, Light Scattering: Principles and Development,ClarendonPress,1996.