amphiphilic toothbrushlike copolymers based on poly

Amphiphilic Toothbrushlike Copolymers Based onPoly(ethylene glycol) and Poly(ε-caprolactone) as Drug Carriers

with Enhanced Properties

Wenlong Zhang,† Yanli Li,‡ Lixin Liu,*,‡ Qiquan Sun,§ Xintao Shuai,*,§ Wen Zhu,† andYongming Chen*,†

State Key Laboratory of Polymer Physics and Chemistry, Joint Laboratory of Polymer Sciences andMaterials, Institute of Chemistry, The Chinese Academy of Sciences and Beijing National Laboratory for

Molecular Science, Beijing 100190, China, College of Life Sciences, Graduate University of ChineseAcademy of Sciences, Yuquanlu 19A, Beijing, 100049, China, and Center of Biomedical Engineering,School of Chemistry and Chemical Engineering, Sun Yat-Sen University, Guangzhou 510275, China

Received February 2, 2010; Revised Manuscript Received April 7, 2010

Amphiphilic poly(ethylene glycol)-b-poly(2-hydroxyethyl methacrylate-g-poly(ε-caprolactone)) (PEG-b-P(HEMA-g-PCL)) toothbrushlike copolymers were synthesized and evaluated as drug delivery carriers. Two toothbrushlikepolymers were synthesized via ring-opening polymerization of ε-caprolactone (CL) initiated by poly(ethyleneglycol)-b-poly(2-hydroxyethyl methacrylate) (PEG-b-PHEMA) macromolecular initiators, and their molecularstructures and physical properties were characterized using 1H NMR, gel permeation chromatography (GPC),and differential scanning calorimetric analysis (DSC). The melting points and crystallizable temperature havebeen decreased obviously, implying that the PCL cores of PEG-b-P(HEMA-g-PCL) toothbrushlike copolymermicelles with shorter PCL segments were unlikely to crystallize at room temperature for drug delivery application.Also the micellization properties of toothbrushlike copolymers in aqueous solution were investigated by fluorescencespectroscopy, dynamic light scattering (DLS), and transmission electron microscopy (TEM). Compared with themicelles from linear PEG-b-PCL block copolymers, the micelles of PEG-b-P(HEMA-g-PCL)s exhibited higherloading capacity to the anticancer drug, doxorubicin (DOX), and the drug-loaded micelles were highly stable inaqueous solution. In vitro DOX release data and confocal laser scanning microscopy (CLSM) studies showedthat DOX-loaded toothbrushlike copolymer micelles could be effectively internalized by bladder carcinoma EJcells, and the DOX could be released into endocytic compartments and finally transported to the nucleus. Suchtoothbrushlike copolymer micelles can be analogues of linear PEG-b-PCL diblock copolymers, but demonstratedbetter properties of loading and release due to their hydrophobic PCL cores do not crystallize at delivery conditions.

Introduction

Over the past two decades, polymeric nanoparticles includingmicelles formed from amphiphilic block copolymers havereceived significant attention as nanocarriers to deliver drugsto the target sites via the enhanced permeability and retentioneffect.1-5 Lipophilic drug molecules can be incorporated intothe hydrophobic core of polymeric micelles by physical entrap-ment, while the hydrophilic shell composed of flexible polymersprovides steric protection and helps these nanoparticles to escapefrom the reticuloendothelial system (RES) uptake after intra-venous administration.6,7 A wide variety of natural and syntheticpolymers have been used in the drug delivery system.8 Amongthem, amphiphilic PEG-b-PCL block copolymers have beenwidely investigated as drug carriers due to their biocompatibility,biodegradability, and nonimmunogenicity.9-12 The polymericnanoparticles prepared from the self-assembled PEG-b-PCLsexhibit a core-shell architecture, in which the biodegradablePCL segments aggregate to form a core to entrap hydrophobicdrugs and the PEG segments used as hydrophilic shell toenhance the circulation time in blood.

Besides traditional linear block polymers, polymers with otherarchitectures such as dendrimers,13 linear-b-dendritic blockcopolymers,14 star block copolymers,15-18 and brush copoly-mers19,20 have also investigated as carriers of drug deliverysystems. Numerous studies have revealed that these branchedpolymers exhibited significantly different physical propertiesfrom linear polymers, such as melt rheology, mechanicalbehavior, and solution properties.21 The architecture of thepolymers also has a great influence on the physicochemicalproperties of the polymer itself in vivo application as well asother aspects of the drug delivery, including drug loadingefficiency, drug release kinetics, biodistribution, and eveninteraction with specific tissues or cells in vivo.22 Therefore,more extensive work is required to optimize polymer architec-ture to fabricate novel drug delivery systems.

The performance of polymeric micelles as drug carriers isdecided by the loading capacity, size, circulation time, stability,release kinetics, and biodistribution. Drug loading content (DLC)and drug encapsulation efficiency of polymeric micelles wereaffected by the affinity of the loaded drugs with the micellarcores, the volume of hydrophobic cores, and the solubility ofdrugs in water.23 The physical entrapment of hydrophobic drugsinto polymeric micelles is driven by the hydrophobic interactionsbetween the drugs and the hydrophobic segments of polymers.In general, the micelle cores constructed with longer PCLsegments can encapsulate more drug molecules, so the DLC

* To whom correspondence should be addressed. E-mail: [email protected](L.L.); [email protected] (X.S.); [email protected] (Y.C.).

† The Chinese Academy of Sciences and Beijing National Laboratoryfor Molecular Science.

‡ Graduate University of Chinese Academy of Sciences.§ Sun Yat-Sen University.

Biomacromolecules 2010, 11, 1331–1338 1331

10.1021/bm100116g 2010 American Chemical SocietyPublished on Web 04/20/2010

increased with an increase of the PCL length.24 However, longerPCL segments lead to a higher degree of crystallinity, resultingin low DLC and poor degradability in body because only theamorphous PCL phase is likely to accommodate drug mol-ecules.25 So, the higher DLC may be expected by decrease ofthe crystallization ability of the PCL segments. However, asfar as we know, little efforts have been paid, except bycopolymerization of CL and lactide.26,27

It is known that the PCLs with either low polymerizationdegree or grafted structure show a decreased crystallizationtemperature.28-30 In this work, we described the synthesis andmicellar characterization of novel amphiphilic toothbrushlikecopolymers poly(ethylene glycol)-b-poly(2-hydroxyethyl meth-acrylate-g-poly(ε-caprolactone)) (PEG-b-P(HEMA-g-PCL). Thesetoothbrushlike copolymers were synthesized by ring-openingpolymerization (ROP) of ε-caprolactone initiated by PEG-b-PHEMA. These toothbrushlike copolymers have the similar EG/CL ratio to traditional linear PEG-b-PCL, however, the PCLblocks are composed of many densely grafted short PCL grafts,hence, their crystallization is restricted. Differential scanningcalorimetric (DSC) analysis showed that these toothbrushlikecopolymers had no crystallinity at room temperature. Comparedto the linear PEG-b-PCL with comparable EG/CL ratio, micellesformed from the toothbrushlike copolymers exhibited higherDLC of anticancer doxorubicin (DOX), and the loaded drugsshowed extraordinary stability in an aqueous solution. Mean-while, both the number and the length of PCL segments of thesetoothbrushlike copolymers can be easily controlled duringsynthesis, thus, the crystallinity of PCL and DLC can be tuned.The cytotoxicity and cellular uptake of the DOX-loadedtoothbrushlike copolymer micelles against bladder carcinomaEJ cells were also investigated using MTT assay and confocallaser scanning microscopy (CLSM).

Experimental Section

Materials. 2-Hydroxyethyl methacrylate (HEMA; 99%, Aldrich) waspurified by passing through a column filled with basic alumina toremove inhibitor. ε-Caprolactone (CL, Aldrich) was dried over CaH2

and distilled before use. Stannous octoate (SnOct2) and 2-bromoio-sobutyryl bromide (BrBBr) were purchased from Sigma-Aldrich.Copper(I) bromide and 2,2′-bipyridine (bpy) were used as received.Monomethoxy poly(ethylene glycol) with Mn ) 5000 g/mol (PEG113,Fluka) was dried by azeotropic distillation in the presence of toluene.N,N-dimethylformamide (DMF) was distilled from CaH2. Doxorubincinhydrochloride (DOX ·HCl) was purchased from Zhejiang Haizheng Co.,China, and was deprotonated at pH 9.6 to obtain the hydrophobic DOXwith sodium hydroxide solution. Toluene, diethyl ether, tetrahydrofuran(THF), dichloromethane, and triethylamine (TEA) were purchased fromBeijing Corporation of Chemical Reagent and used as received. LinearPEG-b-PCL diblock copolymer was synthesized using PEG113 asinitiator of ROP of CL, and the degree of polymerization (DP) of PCLwas 102 calculated from its 1H NMR spectrum with the PCL weightfraction (fwPCL) of 0.70, and it was further named as PEG113-b-PCL102.31

PEG113-Br macroinitiator was synthesized by a reaction between PEG113

and BrBBr according the literature.32,33

Synthesis of PEG-b-PHEMA Diblock Polymer. A typical polym-erization procedure was as follow. A dry Schlenk flask with a magneticstirrer was charged with PEG113-Br (2.06 g, 0.4 mmol), HEMA (2.6 g,20 mmol), bpy (124 mg, 0.8 mmol), and 5 mL of methanol as solvent.The flask was degassed with three freeze-evacuate-thaw cycles; CuBr(57.4 mg, 0.4 mmol) was added into the reaction system under theprotection of nitrogen atmosphere. Next, the polymerization wasperformed at 40 °C for 1 h. After being rapidly cooled to roomtemperature, the reaction flask was opened to air, and the crude productwas precipitated in an excess of diethyl ether, and then redissolved in

5 mL of methylene chloride and passed through a neutral oxide aluminacolumn to remove the copper catalysts. The polymer was obtained byprecipitation in diethyl ether and dried in vacuum for 48 h until constantweight.

Synthesis of PEG-b-P(HEMA-g-PCL) Copolymers. A typicalpolymerization procedure was as follows. In a dried polymerizationtube, PEG113-b-PHEMA13 (1.00 g, 1.86 mmol -OH), CL (1.92 g, 16.8mmol), Sn(Oct)2 (12 mg, 0.03 mmol), and 1 mL of anhydrous DMFwere added, and the air was exchanged with nitrogen three times. Thetube was immersed into an oil bath at 115 °C with vigorous stirringfor 24 h. The resulting product was dissolved in 5 mL of dichlo-romethane and precipitated twice from diethyl ether. PEG-b-P(HEMA-g-PCL) copolymer was dried in vacuum for 48 h until constant weight.

Preparation of DOX-Loaded Micelles and DOX-Free Micelles.DOX-loaded micelles were prepared by a dialysis method. Briefly,

the copolymer (10 mg) and DOX (2 mg) were dissolved in a THF (1mL) and DMSO (0.1 mL) mixed solvent in a glass vial at roomtemperature. After that, the polymer solution was added dropwise into10 mL of deionized water under vigorous stirring. The beaker was thenexposed to air for 12 h, allowing slow evaporation of THF andformation of micelles. The residual THF was then completely removedby vacuum distillation with a rotary evaporator. The micelle solutionwas transferred into dialysis bag (MWCO 14000, Viskase Co.) anddialyzed against deionized water for 48 h to remove free DOX dissolvedin the micelle solution. DOX-free copolymer micelles were preparedin the similar procedure without using DOX.

Characterization of Copolymers and Micelles. 1H NMR spectrumwas recorded on a Bruker AV 400 MHz proton NMR spectrometerwith DMSO-d6 or CDCl3 as solvent. Molecular weight distributions ofcopolymers were determined by gel permeation chromatography (GPC)using a series of three linear Styragel columns (HT2, HT4, HT5)calibrated by polystyrene standards. DMF with LiBr (1 g/L) was usedas eluent solvent at a flow rate of 0.8 mL/min at 35 °C. DSCmeasurements were carried out on a DSC thermal analysis system(Diamond from Perkin-Elmer). Samples were first heated from roomtemperature to 100 °C to erase thermal history at a heating rate of 10°C/min under nitrogen atmosphere, followed by cooling to -50 at 10°C/min after stopping at 100 °C for 1 min and finally heating to 100at 10 °C/min. Transmission electron microscopy (TEM) studies werecarried out on Hitachi H-800 instrument microscope operating at anaccelerating voltage of 100 kV. Samples were prepared by dropping afew microliters of the aqueous solution (1.00 g/L) onto the copper gridscoated with a sustaining film and allowing the samples to dry in airbefore measurements. Samples were stained with 1 wt % phospho-tungstic acid before measurements. The size of micelles was measuredby dynamic light scattering (DLS) on ALV/DLS-5022F equipped witha multi-τ digital time correlator (ALV5000) and a cylindrical 22 mWUNIPHASE He-Ne laser source (λ0 ) 632 nm). The measurementswere carried out at 90° scattering angle at 25 °C. The stock solutionswere filtered through filters of 0.45 µm pore size into the scatteringcells with diameter of 10 mm. The obtained results were analyzed byusing the commercial CONTIN software provided by ALV.

Determination of DOX DLC. The DLC was measured by deter-mining the absorbance at 485 nm using a Shimadzu UV-1601spectrophotometer. An aliquot of the DOX-loaded micelle solution waslyophilized to yield the solid micelle sample. Then the samples wereredissolved in a mixture of chloroform and DMSO (1/1, v/v) for theUV measurement. DOX solutions with various concentrations wereprepared, and the absorbances at 485 nm were used to generate acalibration curve for the DLC calculations from DOX-loaded micelles.

In Vitro Drug Release. A total of 10 mg of DOX-loaded micelleswere diluted to 1 mg/mL in phosphate buffered saline (PBS, pH 7.4)and transferred into a dialysis bag with a molecular weight cutoff of3600 Da (Viskase Co.). The dialysis bag was then immersed into 90mL PBS solution with gentle shaking (100 rpm) at 37 °C in a ZhiChengZHWY-200B shaker. At predetermined time intervals, 2.0 mL buffersolution outside the dialysis bag was extracted, and it was replaced by

1332 Biomacromolecules, Vol. 11, No. 5, 2010 Zhang et al.

fresh PBS to remain the sink condition. The DOX concentration wasdetermine by UV spectrophotometer based on the absorbance intensityat 485 nm.

Cytotoxicity Study. Cytotoxicities of DOX-loaded micelles and freeDOX were measured against bladder carcinoma EJ cells by MTT assay.The cells harvested in a logarithmic growth phase were seeded in a96-well plates with a density of 5000 cell per well. Cells were incubatedin DMEM (Gibco) supplemented with 10% FBS (Gibco) at 37 °C for1 day in a humidified atmosphere with 5% CO2. Then the cells wereincubated with the medium containing free DOX or DOX-loadedmicelles. The DOX concentrations of each formulation were preparedby serial dilution with DMEM medium. After treatment for 48 h, 100µL of fresh medium containing 10% of MTT of 5 mg/mL stock wasreplaced to each well. The plates were incubated for 4 h; then 150 µLof DMSO (Sigma) was added to each well to dissolve intracellularMTT formazan crystals, followed by absorbance at 570 nm using amicroplate reader. Experiments were done in triplicate and wererepeated at least twice.

Confocal Laser Scanning Microscopy (CLSM). Bladder carcinomaEJ cells were seeded on coverslips in a 24-well plate, free DOX (3 µM)and DOX-loaded micelles (DOX concentration: 3 µM) were incubatedwith EJ cells for different times before CLSM measurement. Cells werestained with Hoechst 33342 to mark the nuclei. Before the measurement,the cells were washed three times with PBS and fixed with 4%formaldehyde. Images were obtained with an Olympus PV1000-IX81Confocal Microscope. Hoechst 33342 and DOX were excited at 352 and485 nm with emissions at 455 and 595 nm, respectively.

Results and Discussions

Synthesis of PEG-b-PHEMA Diblock Polymers. ATRP waschosen to synthesize PEG-b-PHEMA block copolymers in thispaper due to its well control over molecular weight and chainuniformity.34 As shown in Scheme 1, the monofunctionallizedPEG113 macroinitiator was used for the ATRP of HEMA inmethanol. The feed ratios of PEG-Br/HEMA were adjusted toachieve different chain length of PHEMA. The resulted PEG-b-PHEMAs gave a monomodal GPC trace, indicating that thePEG-Br showed high efficiency of initiation. The compositionand the molecular weight of the product were estimated by the1H NMR spectrum. Figure 1A shows the spectrum of the diblockcopolymer PEG113-b-PHEMA13; the polymerization degree was

calculated by comparing the signals at 3.9 ppm (COOCH2) ofHEMA and the signals at 3.5 ppm (OCH2CH2) of EG unit. Thepeaks at 4.8 ppm (OH) represent the hydroxyl groups of HEMAunits. The molecular weight obtained by 1H NMR agrees withthe theoretical molecular weight calculated by monomer conver-sion. The DPs of the PHEMA blocks in two samples werecalculated to be 13 and 26, and the block copolymers werefurther named as PEG113-b-PHEMA13 and PEG113-b-PHEMA26,respectively, to synthesize different PCL brushes. The conditionof ATRP and the characterization of PEG-b-PHEMA blockcopolymer were listed in Table 1.

Synthesis of PEG-b-P(HEMA-g-PCL) Copolymers. PEG-b-P(HEMA-g-PCL) copolymers were synthesized by ROP usingthe hydroxyl groups in the side chains of PEG-b-PHEMApolymers to initiate the CL polymerization in dried DMF underthe catalysis of Sn(Oct)2 at 115 °C for 24 h. To obtaintoothbrushlike copolymers with the similar fwPCL of PEG113-b-PCL102, the feed molar ratio of CL and hydroxyl groups ofPEG113-b-PHEMA13 copolymer was 9, while the ratio of CLand hydroxyl groups of PEG113-b-PHEMA26 was 6 to synthesizethe toothbrushlike copolymer with higher fwPCL. The monomerconversion was 81 and 78%, respectively. 1H NMR spectrumof one sample of PEG-b-P(HEMA-g-PCL) copolymer is shownin Figure 1B. Assuming that all of the hydroxyl groups ofPHEMA chain took a part in initiation, the average DP of PCLin PEG-b-P(HEMA-g-PCL) copolymer can be calculated bycomparing the proton signals at 2.3 ppm (OCOCH2) of PCLand the signals at 3.6 ppm (OCH2CH2) of PEG113. The averageDPs of PCLs in the toothbrushlike copolymers were 7.7 and5.4, respectively. As listed in Table 2, these toothbrushlikecopolymers were named as PEG113-b-P(HEMA-g-PCL8)13 withfwPCL of 0.70 and PEG113-b-P(HEMA-g-PCL5)26 with fwPCL of0.76. All of these results indicated that PEG-b-P(HEMA-g-PCL)toothbrushlike copolymers could be easily synthesized andtailored by varying the segment lengths of PHEMA in PEG-b-PHEMA and the feed ratios of CL and hydroxyl groups ofPEG-b-PHEMA.

DSC Characterization. The melting and crystallizationbehaviors of these PEG-b-P(HEMA-g-PCL) toothbrushlikecopolymers were investigated by DSC. The crystallizationtemperature (Tc) was obtained from the cooling run, and themelting temperature (Tm) was obtained from the second heatingrun. The obtained DSC curves were shown in Figure 2 and thatof PEG113-b-PCL102 polymer was used for comparison. Withthe same fwPCL of 0.70, as listed in Table 2, Tc,PCL of the PEG113-b-P(HEMA-g-PCL8)13 toothbrushlike copolymer was -7.0 °C,while Tc,PCL of the PEG113-b-PCL102 linear diblock polymer was27.6 °C. An obvious decrease of Tc should be attributed to thecrystalline imperfection of the short PCL grafts. Moreover, thetoothbrushlike structure of these polymers should make anothercontribution to the imperfection. Further shortening the chainlength of PCL segments, the Tc,PCL of PEG113-b-P(HEMA-g-PCL5)26 toothbrushlike copolymer decreased to -17.4 °C eventhough it had a higher fwPCL of 0.76. These results suggestedthat crystallization properties of these PEG-b-P(HEMA-g-PCL)toothbrushlike copolymers could be easily adjusted with thenumber and length of PCL segments. Meanwhile, Tc,PEGs of

Scheme 1. Synthesis of Amphiphilic PEG-b-P(HEMA-g-PCL)Copolymers

Table 1. Experimental Conditions and Characterization Results of PEG-b-PHEMA Polymers

polymers conditionsa time conversion (%) DPPHEMAb Mn

b Mw/Mnc

PEG113-b-PHEMA13 25:1:1:2 0.5 h 48.2 13 6840 1.11PEG113-b-PHEMA26 50:1:1:2 1 h 47.8 26 8530 1.13

a Feed: [HEMA]/[PEG-Br]/[CuBr]/[bpy], T ) 40 °C. b Calculated based on 1H NMR. c Determined by GPC analyses using DMF as the eluent.

Amphiphilic Toothbrushlike Copolymers Biomacromolecules, Vol. 11, No. 5, 2010 1333

these PEG-b-P(HEMA-g-PCL)s also were greatly reduced,indicating that the brushlike blocks would also influence thecrystallization behaviors of the PEG blocks. Generally, themicelles made from amphiphilic copolymers used in drugdelivery were performed at room temperature. So the cores ofthese toothbrushlike copolymer micelles would have no crystal-linity that would benefit drug loading and delivering and alsoadjust biodegradation behavior of PCL segments.

Preparation and Characterization of Copolymer Micelles.The amphiphilic toothbrushlike copolymer self-assembled intomicelles in aqueous solution using the dialysis method. The

critical micelle concentrations (CMC) of these toothbrushlilkecopolymers were analyzed by fluorescence spectra using pyreneas a hydrophobic probe. In the excitation spectra, several aspectsof the pyrene spectroscopic properties were observed as thecopolymer concentration increased. The intensity ratios at bandat 372 and 383 nm (I1/I3) of the pyrene excitation spectra versusthe logarithm of copolymer concentration were shown in Figure3. The I1/I3 remained almost constant at low copolymerconcentrations and increased sharply when the copolymerconcentrations reached a value, indicating formation of copoly-mer micelles and the pyrene was trapped into the micelle core

Figure 1. 1H NMR spectra of (A) PEG113-b-PHEMA13 in DMSO-d6 and (B) PEG113-b-P(HEMA-g-PCL8)13 in CDCl3.

Table 2. Synthesis and Characterization of PEG-b-P(HEMA-g-PCL) Polymers

polymers [CL]/[OH] DPPCLa Mn

a PDI b fwPCLa Tc,PCL

c (°C)

PEG113-b-P(HEMA-g-PCL8)13 9:1 8 18200 1.21 0.70 -7PEG113-b-P(HEMA-g-PCL5)26 6:1 5 24500 1.26 0.76 -17PEG113-b- PCL102 105 102 16700 1.24 0.70 27.6

a Calculated based on 1H NMR. b Determined by GPC analyses using DMF as the eluent. c Determined by DSC analyses.

Figure 2. DSC curves of PEG113-b-PCL102, PEG113-b-P(HEMA-g-PCL8)13, and PEG113-b-P(HEMA-g-PCL5)26 copolymers in the cooling run (solidlines) and in the second heating run (dotted lines).


with more hydrophobic environment. From the sigmoidal curves,CMCs of PEG113-b-P(HEMA-g-PCL8)13 and PEG113-b-P(HEMA-g-PCL5)26 were determined as 2.5 and 2.9 mg/L in aqueoussolution, which were similar with the CMC of linear PEG113-b-PCL102 block copolymer (2.5 mg/L).35

To further evaluate the properties of toothbrushlike copoly-mers micelles, both the morphology and the mean size of thesePEG-b-P(HEMA-g-PCL) micelles were studied by TEM andDLS. As shown in Figure 4 and Table 3, the micelles of thesetoothbrushlike copolymers exhibited a uniform spherical mor-phology. From the DLS experiments, the average diameters ofmicelles were about 47 nm, with a polydispersity index of 0.13for PEG113-b-P(HEMA-g-PCL8)13 and 79 nm with a polydis-persity index of 0.32 for PEG113-b-P(HEMA-g-PCL5)26, respec-tively. The size of these micelles measured from TEM imageswere 54 and 98 nm that were comparable to that from DLS.

Drug Loading and Encapsulation Efficiency. To assess theinfluence of toothbrushlike copolymer structure on drug incor-poration, DOX was used as model anticancer drug to evaluatethe drug loading and release properties. DOX-loaded PEG-b-P(HEMA-g-PCL) micelles were prepared by the dialysis

method. The size and size distribution of DOX-loaded micellesmeasured by DLS showed that the diameters of these DOX-loaded micelles were larger than their parent micelles withoutDOX as summarized in Table 3. The drug loading contents ofPEG113-b-P(HEMA-g-PCL8)13 and PEG113-b-P(HEMA-g-PCL5)26 were 7.9 and 6.9%, respectively. However, under thesame fwPCL and loading conditions, the DLC of the linear PEG113-b-PCL102 was 6.2%. The DLC of the PEG113-b-P(HEMA-g-PCL8)13 copolymers increased about 27% compared to that ofthe linear PEG113-b-PCL102. It is well-known that physicaltrapping of hydrophobic drugs into polymeric micelles is drivenby the hydrophobic interaction between the drug and thehydrophobic segments of copolymers. Generally, long PCLblocks with more hydrophobic favor DOX-encapsulating intothe micelles. However, long PCL segments results in a highercrystallinity, which lead to less drug loading in the micelles.Compared with the linear PEG113-b-PCL102 diblock copolymer,these toothbrushlike copolymer analogues with shorter graftedPCL segments have no crystallinity at room temperature, sotheir micelles could entrap more drug molecules.

In vitro release profiles of the loaded DOX in micelles ofPEG-b-P(HEMA-g-PCL) copolymers were studied and com-pared to that of the linear PEG113-b-PCL102 diblock copolymermicelles under a simulated physiological condition PBS (0.01M, pH 7.4) at 37 °C. As illustrated in Figure 5, the DOXrelease behaviors of the novel micellar carriers and theircomparison had the similar typical two-phase release profile.In the first stage, about 19-28% DOX encapsulated inmicelles released within 24 h, followed by a sustained andslow release over a prolonged time. It is expected that themost of DOX remains in the micelle cores for a considerabletime when these micelles circulate in the plasma at normalphysiological conditions. However, the faster releases ofDOX from toothbrushlike copolymer micelles were observed.This is because PEG-b-P(HEMA-g-PCL) toothbrushlikecopolymers with the amorphous short and grafted PCLbranches under the delivery condition would facilitate thedrug molecules diffusing out from the micelle cores. PEG113-b-P(HEMA-g-PCL5)26 had shorter PCL segments than PEG113-b-P(HEMA-g-PCL8)13 did, so the DOX in PEG113-b-P(HEMA-

Figure 3. Plot of the I372/I383 ratio against log C of polymeric micelles.

Figure 4. TEM photographs of the self-assembled nanoparticles from (A) PEG113-b-P(HEMA-g-PCL8)13 and (B) PEG113-b-P(HEMA-g-PCL5)26 inaqueous solution: 1 mg/mL.

Table 3. Influence of Copolymer Architecture and Compositions on Micellar Properties

DOX-free micelles DOX-loaded micelles

sample diameter (nm)a PDIa diameter (nm) a PDIa DLC (%)

PEG113-b-P(HEMA-g-PCL8)13 46.9 ( 1.5 0.13 ( 0.01 132.2 ( 3.2 0.18 ( 0.01 7.9PEG113-b-P(HEMA-g-PCL5)26 78.9 ( 2.4 0.32 ( 0.04 151.1 ( 4.3 0.17 ( 0.01 6.9PEG113-b- PCL102 44.2 ( 1.3 0.16 ( 0.02 90.6 ( 3.4 0.25 ( 0.03 6.2

a Measured by DLS analyses, all the measurements were performed in triplicate.


g-PCL5)26 micelles was released faster even though thePEG113-b-P(HEMA-g-PCL8)13 micelles had a higher DLC.All these results suggested that these toothbrushlike copoly-

mers with a different PCL number and length provide astraight way to adjust the DOX release kinetics. It isnoteworthy that the DOX-loaded toothbrushlike copolymermicelles in PBS exhibited no deposits for 6 months, but thecompared DOX-loaded PEG113-b-PCL102 micelles produceda lot of floccules in 4 days, as shown in Figure 6. Thisphenomenon should be contributed by the architecture oftoothbrushlike copolymers. So the toothbrushlike copolymermicelles with drugs have enhanced stability in aqueoussolution than that of the linear PEG-b-PCL polymer micelles.This is very important in practical application of drug deliverysystem.

Cellular Uptake and Cytotoxicity. To investigate themechanism of internalization of DOX-loaded toothbrushlikecopolymers micelles and intracellular drug delivery, bladdercarcinoma EJ cells were observed by CLSM microscopy atdifferent time. As shown in Figure 7, intracellular distributionof the DOX-loaded micelles was different from that of freeDOX. After 4 h of cell incubation with the free DOX, strongfluorescence was observed in the nucleus of the cell. In contrast,DOX was observed mainly in the cytoplasm of the cells insteadof the nucleus when the cells were incubated with the DOX-loaded toothbrushlike copolymer micelles for 4 and 10 h.

Figure 5. In vitro DOX release profiles from DOX-load micelles inPBS (0.01M, pH 7.4) at 37 °C.

Figure 6. Stability properties of DOX-loaded micelles of (1) PEG113-b-P(HEMA-g-PCL8)13, (2) PEG113-b-P(HEMA-g-PCL5)26, and (3) PEG113-b-PCL102 in PBS (0.01 M, pH 7.4) at 37 °C in 96 h.

Figure 7. CLSM images of EJ cells incubated with DOX-loaded PEG113-b-P(HEMA-g-PCL8)13 micelles for (A) 4 h and (B) 24 h, DOX-loadedPEG113-b-P(HEMA-g-PCL5)26 micelles for (C) 4 h and (D) 24 h, DOX-loaded PEG113-b-PCL102 micelles for (E) 4 h and (F) 24 h, and DOX for (G)4 h and (H) 24 h. For each panel, images from left to right show the cells with nuclear staining by Hoechst 33342, with DOX fluorescence, andoverlays of both images; scale bars correspond to 20 µm in all images.


However, there was a significant DOX accumulation in thenucleus when the cells were incubated with the DOX-loadedPEG-b-P(HEMA-g-PCL) micelles for 24 h, which differed fromDOX-loaded PEG113-b-PCL102 micelles that DOX still mainlystayed in the cytoplasm after 24 h incubation. These resultssuggested that the novel DOX-loaded micelles may have beentaken up by the cells through a nonspecific endocytosismechanism; and then the DOX molecules were released inendocytic compartments (i.e., endosomes and later lysosomes)and finally rapidly reached to the nucleus.

To investigate cytotoxicity of the free blank micelles andDOX-loaded micelles, bladder carcinoma EJ cells were exposedto a series of different concentrations of the blank micelles, freeDOX, and DOX-loaded micelles for 48 h, and the percentagesof viable cells were quantified using the MTT method. As shownin Figure 8, these blank toothbrushlike copolymer micelles havethe similar cytotoxicity to that of micelles formed by PEG113-b-PCL102. No obvious cytotoxicity against EJ cells was observedeven the copolymer micelles concentration reached 500 µg/mL.The cytotoxicity of DOX-loaded PEG113-b-P(HEMA-g-PCL8)13

and PEG113-b-P(HEM-g-PCL5)26 copolymer micelles comparedwith that of free DOX was shown in Figure 9. The concentrationof DOX in micelles that cause 50% cell killing was relativelyhigher than that of free DOX. DOX in the toothbrushlike

copolymers micelles exhibited much lower cytotoxicity whencompared with free DOX at the same dose. The lowercytotoxicity of DOX in the micelles was likely due to a time-consuming DOX release from micelles and delayed nuclearuptake in EJ cells, as shown by the in vitro DOX release andinternalization studies by CLSM.

Conclusions

In conclusion, PEG-b-P(HEMA-g-PCL) toothbrushlike co-polymers of different molecular weights and compositions weresynthesized by a combination of ATRP and ROP. Because theseshort and grafted PCL branches were difficult to crystallize, themicelles prepared from these toothbrushlike copolymers pos-sessed higher DOX loading capability than the micelles formedby linear PEG-b-PCL polymer did. The DOX-loaded tooth-brushlike copolymer micelles also exhibited extraordinarystability in aqueous solution than that of DOX-loaded linearPEG-b-PCL polymer micelles. In vitro release data indicatedthat the DOX-release from toothbrushlike copolymer micelleswas faster than that in linear block copolymer micelles. CLSMstudies showed that DOX-loaded toothbrushlike copolymermicelles could be effectively internalized by bladder carcinomaEJ cells and DOX could be released in endocytic compartmentsand then reached to the nucleus. With these enhanced propertiesrelative to that of linear PEG-b-PCL, the toothbrushlikecopolymer micelles composed of PEG and PCL have potentialapplication as a versatile nanocarrier of various liposolubilitydrugs.

Acknowledgment. Financial support from NSF China(20534010, 20625412, and 50830107) and the Chinese Academyof Sciences (Knowledge Innovation Program, KJCX2-YW-H19)is gratefully acknowledged.

References and Notes

(1) Peer, D.; Karp, J. M.; Hong, S.; Farokhzad, O. C.; Margalit, R.; Langer,R. Nat. Nanotechnol. 2007, 2, 751–760.

(2) Davis, M. E.; Chen, Z.; Shin, D. M. Nat. ReV. Drug DiscoVery 2008,7, 771–782.

(3) Kwon, G. S.; Kataoka, K. AdV. Drug DeliVery ReV. 1995, 16, 295–309.

(4) Gaucher, G.; Dufresne, M. H.; Sant, V. P.; Kang, N.; Maysinger, D.;Leroux, J.-C. J. Controlled Release 2005, 109, 169–188.

(5) Maeda, H.; Wu, J.; Sawa, T.; Matsumura, Y.; Hori, K. J. ControlledRelease 2000, 65, 271–284.

(6) Gref, R.; Minamitake, Y.; Peracchia, M. T.; Trubetskoy, V.; Torchilin,V.; Langer, R. Science 1994, 263, 1600–1603.

(7) Gref, R.; Luck, M.; Quellec, P.; Marchand, M.; Dellacherie, E.;Harnisch, S.; Blunk, T.; Muller, R. H. Colloids Surf., B 2000, 18,301–313.

(8) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M.Chem. ReV. 1999, 99, 3181–3198.

(9) Allen, C.; Yu, Y. S.; Maysinger, D.; Eisenberg, A. Bioconjugate Chem.1998, 9, 564–572.

(10) Savic, R.; Luo, L.; Eisenberg, A.; Maysinger, D. Science 2003, 300,615–618.

(11) Ghoroghchian, P. P.; Li, G.; Levine, D. H.; Davis, K. P.; Bates, F. S.;Hammer, D. A.; Therien, M. J. Macromolecules 2006, 39, 1673–1675.

(12) Park, E. K.; Kim, S. Y.; Lee, S. B.; Lee, Y. M. J. Controlled Release2005, 109, 158–168.

(13) Medina, S. H.; El-Sayed, M. E. H. Chem. ReV. 2009, 109, 3141–3157.

(14) Yang, Y.; Hua, C.; Dong, C.-M. Biomacromolecules 2009, 10, 2310–2318.

(15) Chen, S.; Zhang, X. Z.; Cheng, S. X.; Zhuo, R. X.; Gu, Z. W.Biomacromolecules 2008, 9, 2578–2585.

(16) Wiltshire, J. T.; Qiao, G. G. Aust. J. Chem. 2007, 60, 699–705.(17) Wei, H.; Zhang, X.; Cheng, C.; Cheng, S. X.; Zhuo, R. X. Biomaterials

2007, 28, 99–107.

Figure 8. Viability of EJ cells as a function of polymer type andconcentration after a 48 h incubation (n ) 3).

Figure 9. Cytotoxicity of free DOX and DOX-loaded micelles to EJcells after a 48 h incubation (n ) 3).


(18) Prabaharan, M.; Grailer, J. J.; Pilla, S.; Steeber, D. A.; Gong, S.Biomaterials 2009, 30, 3009–3019.

(19) Yan, Q.; Yuan, J.; Zhang, F.; Sui, X.; Xie, X.; Yin, Y.; Wang, S.;Wei, Y. Biomacromolecules 2009, 10, 2033–2042.

(20) Du, J. Z.; Tang, L. Y.; Song, W. J.; Shi, Y.; Wang, J. Biomacromol-ecules 2009, 10, 2169–2174.

(21) McKee, M. G.; Unal, S.; Wilkes, G. L.; Long, T. E. Prog. Polym.Sci. 2005, 30, 507–539.

(22) Qiu, L.; Bae, Y. Pharm. Res. 2006, 23, 1–30.(23) Allen, C.; Maysinger, D.; Eisenberg, A. Colloids Surf., B 1999, 16,

3–27.(24) Shuai, X.; Merdan, T.; Schaper, A. K.; Xi, F.; Kissel, T. Bioconjugate

Chem. 2004, 15, 441–448.(25) Shuai, X.; Ai, H.; Nasongkla, N.; Kim, S.; Gao, J. J. Controlled Release

2004, 98, 415–426.(26) Hu, Y.; Jiang, X.; Ding, Y.; Zhang, L.; Yang, C.; Zhang, J.; Chen, J.;

Yang, Y. Biomaterials 2003, 24, 2395–2404.

(27) Zhang, L.; Hu, Y.; Jiang, X.; Yang, C.; Lu, W.; Yang, Y. H. J.Controlled Release 2004, 96, 135–148.

(28) Yuan, W.; Yuan, J.; Zhang, F.; Xie, X.; Pan, C. Macromolecules 2007,40, 9094–9102.

(29) Sosnik, A.; Cohn, D. Polymer 2003, 44, 7033–7042.(30) Zhou, S.; Deng, X.; Yang, H. Biomaterials 2003, 24, 3563–3570.(31) He, C.; Sun, J.; Ma, J.; Chen, X.; Jing, X. Biomacromolecules 2006,

7, 3482–3489.(32) Du, J. Z.; Chen, Y. M. J. Polym. Sci., Part A: Polym. Chem. 2004,

42, 2263–2271.(33) Wang, X. S.; Armes, S. P. Macromolecules 2000, 33, 6640–6647.(34) Matyjaszewski, K.; Xia, J. Chem. ReV. 2001, 101, 2921–2990.(35) Forrest, M. L.; Won, C. Y.; Malick, A. W.; Kwon, G. S. J. Controlled

Release 2006, 110, 370–377.

BM100116G


amphiphilic toothbrushlike copolymers based on poly

Documents