novel thermosensitive polymeric micelles for docetaxel delivery

Novel thermosensitive polymeric micellesfor docetaxel delivery

Mi Yang,1 Yitao Ding,2 Leyang Zhang,3 Xiaoping Qian,1 Xiqun Jiang,3 Baorui Liu1

1Department of Oncology, Affiliated Drum Tower Hospital, Medical School of Nanjing University,Nanjing 210008, People’s Republic of China2Department of Hepatobiliary Surgery, Affiliated Drum Tower Hospital, Medical School of Nanjing University,Nanjing 210008, People’s Republic of China3Department of Polymer Science and Engineering, College of Chemistry and Chemical Engineering,Nanjing University, Nanjing 210093, People’s Republic of China

Received 4 April 2006; revised 6 August 2006; accepted 6 September 2006Published online 18 January 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.31129

Abstract: Targeted delivery of antitumor drugs triggeredby hyperthermia has significant advantages in clinical ap-plications, since it is easy to implement and side effects arereduced. To release drugs site-specifically upon local heat-ing often requires the drugs to be loaded into a thermosen-sitive polymer matrix with a low critical solution tempera-ture (LCST) between 37 and 428C. However, the LCSTs ofmost thermosensitive materials were below 378C, which lim-its their application in clinic because they would precipitateonce injected into human body and lost thermal targetingfunction. Herein, we prepared a novel thermosensitive co-polymer (poly(N-isopropylacrylamide-co-acrylamide)-b-poly(DL-lactide)) that exhibits no obvious physical change up to418C when heated. Docetaxel loaded micelles made of suchthermosensitive polymer were prepared by dialysis method

and the maximum loading content was found to be up to27%. The physical properties, such as structure, morphology,and size distribution of the micelles with and without doce-taxel were investigated by NMR, X-ray diffraction, dynamiclight scattering, atomic force microscopy, etc. The efficacy ofthis drug delivery system was also evaluated by examiningthe proliferation inhibiting activity against different cell linesin vitro. After hyperthermia, the cytotoxicity of docetaxel-loaded micelles increased prominently. Our results demon-strated that this copolymer could be an ideal candidate forthermal targeted antitumor drug delivery. � 2007 WileyPeriodicals, Inc. J Biomed Mater Res 81A: 847–857, 2007

Key words: thermosensitive material; micelles; docetaxel;targeted drug delivery; antitumor

INTRODUCTION

Even though the pharmaceutical industry has beensuccessful in discovering many new cytotoxic drugsthat are potential candidates for the treatment of cancer,this life-threatening disease still causes over 6 milliondeaths annually around the world and the number isstill growing.1 Nevertheless, conventional chemothera-peutical drugs, such as docetaxel, cisplatin, and soforth, are nonspecific against both normal and malig-nant cells. Therefore, the clinical use of traditionalchemotherapy is limited because of its intolerable sideeffects on treated patients. The side effects includebone marrow depression, gastrointestinal tract reac-tion, edema, anaphylaxis, and skin toxicity.2–4 Thus, itis highly desirable in clinical practice to develop new

approaches that can maintain drugs’ efficacy in treat-ing cancer, and at the same time minimize their cyto-toxicity against normal tissues. During the last severalyears, targeted drug delivery has been developed sig-nificantly and made anticancer agents release andaccumulate to a high concentration only at sites wheretumors reside, and thus considerately reducing theirtoxicity on normal tissues.

Various strategies, such as antibody-, receptor-, mag-netic-, and thermo-target, can be used for targeted drugdelivery.5–7 With the invention of large-scale hyperther-mia instruments and the wide use of hyperthermia, thethermal targeting has become much easy to implementand precise to control.8 Unlike other techniques such asactive targeting, thermal targeting, which relies on localheating to confine the release of drugs, can be appliedon a wide spectrum of cancer types, notably improvingtheir clinical applications.9 Hyperthermia also increasesthe permeability of tumor vasculature preferentiallywhen compared with that of normal vasculature, fur-ther promoting the delivery of drugs to tumors.10 Inaddition, hyperthermia has a particular advantage of

Correspondence to: B. Liu; e-mail: [email protected] grant sponsor: Natural Science Foundation of

Jiangsu Province; contract grant number: H200349.

' 2007 Wiley Periodicals, Inc.

synergetic effects to kill malignant tumor cells whencombined with chemotherapies.11,12

In terms of practical applications of thermal targetingtreatment, one of the key issues is to find suitable thermalsensitive materials for packaging of anticancer drugs. Inthe past, poly(N-isopropylacrylamide) (PIPAAm), a well-known thermosensitive polymer that undergoes a lowcritical solution temperature (LCST) phase transition,has been extensively used as a drug carrier.13 LCST isa critical physical parameter to evaluate the perform-ance and usefulness of a thermosensitive material forits application in drug delivery. Below the LCST, thepolymer is well soluble in water due to extensive for-mation of hydrogen bonds between polymer andwater molecules. At a temperature above the LCST,however, the network of hydrogen bonds collapses toexclude water molecules from the polymer, eventuallyleading to aggregation and precipitation of the poly-mer.14 This unique character of PIPAAm makes it anideal choice as a drug vehicle and many attempts havebeen made to prepare various forms of this material,such as crosslinked hydrogels,15 micelles,16,17 and linear-polymers,18 to examine its properties. Polymeric micellesreflect a kind of ordered dimensional structure of AB-type block copolymers with a hydrophilic corona/shell in the aqueous solution.16 The micelles can es-cape non-selective scavenging by the reticuloendothe-lial system (RES) and selectively accumulate at a solidtumor by a passive targeting mechanism due to theirhydrophilic shells and appropriate sizes (20–200 nm).13,17

Therefore, polymeric micelles have become one of themost noteworthy candidates as drug carriers for thetreatment of cancer as reviewed by Kataoka et al.19

However, the LCST of reported thermosensitivemicelles were almost below physiological body tem-perature (378C), which limited their use in clinic be-cause they would precipitate once injected into humanbody and lose thermal targeting function. In the pres-ent study, we report the copolymerization of N-isopro-pylacrylamide with acrylamide (AAm) to increase theLCST of the copolymer to a level between physiologi-cal body temperature (378C) and that used in localhyperthermia (about 428C). This kind of copolymer iswell soluble after injection so it is able to escapeuptake by RES and circulate longer in vivo. Uponbeing heated locally, this copolymer precipitates site-specifically to release drugs wrapped inside, enrichingthe concentration of anticancer agents there signifi-cantly. To prepare a useful form of this material suita-ble for drug delivery, we further conjugated this co-polymer with a hydrophobic polymer, poly(DL-lactide)(PDLLA), to construct a block copolymer that canform micelle structure in aqueous media. Thus, themicelle is composed of a hydrated outer shell formedby thermosensitive copolymer and a hydrophobicinner core formed by PDLLA. The nature of thishydrated outer shell prevents micelles from being ag-

gregated and enables them to escape from nonselec-tive scavenging by the RES to gain a longer plasmahalf-life at the physiological temperature. Once circu-lated into heated malignant tissue sites where localtemperature is above its LCST, the outer shell of thesemicelles transits into a hydrophobic structure and sub-sequently is absorbed into cells mediated by hydropho-bic interaction. As a result, anticancer drugs packedinside can accumulate at malignant tissues to a levelhigh enough to kill the cancer cells.

PDLLA is a biodegradable material and exhibits notoxicity after hydrolyzation.20 Therefore, these poly-meric micelles will eventually dissociate into the blockcopolymers and the hydrophilic segments less than150,000 Da that can be easily excreted by glomerularfiltration, eliminating the danger of accumulation ofthis substance in vivo.

We prepared micelles with this novel thermosensitivecopolymer as a carrier for docetaxel, a widely applicableantitumor drug in the clinical practice.21–23 Particle size,size distribution, and morphology of the obtained micelleswere characterized. In vitro cytotoxicity was also in-vestigated using human umbilical vein endothelial cell(HUVEC) and three different tumor cell lines.

MATERIALS AND METHODS

Materials

Docetaxel was kindly provided by Hengrui Pharmaceuti-cal, Jiangsu, China and used as received. N-isopropylacryla-mide (IPAAm, ACROS, NJ) was purified by recrystallizationfrom n-hexane (Nanjing, China). 2,20-azobisbutyronitrile (AIBN,Aldrich, MO) was recrystallized in hexane-benzene (70:30,v/v)(Nanjing, China). RPMI 1640 (Gibco, NY), EGM (Clonetics,CA), fetal bovine serum (Amersco, SF), calf blood serum(Amersco), dimethylthiazoly-2,5-diphenyltetrazolium bro-mide (MTT, Amersco), Acrylamide (AAm, FLUKA, Switzer-land), 2-hydroxyethanethiol (SIGMA, St. Louis), fluorescein-iso-thiocyanate (FITC, Acros, Belgium), DL-lactide (TOKYO,Japan), tin(II)2-ethylhexanoate (SIGMA,) were used as re-ceived. Acetonitrile (Merck, Germany) was of chromatogramgrade. All other chemicals were of analytical grade and usedwithout further purification.

Poly(IPAAm-co-AAm) synthesis

Hydroxyl-terminated copolymers of poly(IPAAm-co-AAm)were synthesized by free radical polymerization of N-isopro-pylacrylamide and AAm in ethanol, using 2,20-azobisbutyro-nitrile and 2-hydroxyethanethiol as initiator and chain-trans-fer reagent, respectively. Polymerization reaction was per-formed at 708C for 20 h under nitrogen. The copolymerswere obtained by precipitation of the reaction solution intodiethyl ether and dried in vacuo. The synthetic scheme isshown in Figure 1(A). The dried polymer was dissolved inwater and filtered through an ultrafiltration membrane with

848 YANG ET AL.

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

10,000 molecular weight cut-off (Stirred Cells 8050, UltracelPL, 10.000 NMWL, Millipore).

Poly(IPAAm-co-AAm)-b-PDLLA synthesis

The AB-block copolymers were obtained by ring-openingpolymerization of DL-lactide using the terminal hydroxylgroup of poly(IPAAm-co-AAm) with tin(II)2-ethylhexa-noate as a catalyst. Polymerization was proceeded at 1508Cfor 20 h under a nitrogen atmosphere [Fig. 1(B)]. Thecopolymers obtained were dissolved in ethanol, filteredthrough medium speed filter paper, and dried in vacuo.

Micelles preparation

The AB-block copolymers were dissolved in acetonitrile(1 mg/mL). The solution was dialysed against pure waterusing a dialysis membrane with a cut-off molecular weightof 12,000 (Sorua, Germany). The resultant micelles werefreeze dried and stored at 48C.

Docetaxel loading into micelles

20 mg docetaxel and 50 mg poly(IPAAm-co-AAm)-b-PDLLA were dissolved in 5 mL acetonitrile. After drasti-cally stirred for 2 h, the mixture was immediately heated to408C by soaking into a water bath and then slowly cooleddown to room temperature. The non-entrapped, precipi-tated docetaxel was removed by filtration through a 0.45 mmfilter. The obtained micellar solutions were frozen and ly-ophilized by freeze dryer system.

Molecular weight measurements

The number-average molecular weight (Mn), weight-aver-age molecular weight (Mw), and molecular weight distribu-tion of the copolymers were measured by GPC (Waters 244,Differential Refractometer R401) with tetrahydrofuran (THF)as eluent at a flow rate of 1 mL/min. Calibration was accom-

plished with monodispersed polystyrene standards with amolecular weight ranging from 800 to 124,000 g/mol.

LCST measurements

Poly(IPAAm-co-AAm) and poly(IPAAm-co-AAm)-b-PDLLAwere dissolved in phosphate-buffered saline (PBS) respec-tively at a concentration of 5 mg/mL. Optical transmittance ofthe solutions at various temperature was measured at 500 nmwith a UV-Vis spectrometer (UV-2401, Shimadzu, Japan).Sample cells were thermostated with a temperature-controller(ELLY4, TOKYO Rikaaikai, Japan). The heating rate was0.18C/min. The LCST values of polymer solutions were deter-mined at the temperatures showing an optical transmittanceof 50%.

Nuclear magnetic resonance measurements

The 1H-NMR and 13C-NMR spectra of the poly(IPAAm-co-AAm)-b-PDLLA in deuterated chloroform solution weremeasured by a MSL-300 spectrometer (Bruker, Germany) todetermine the chemical structure of the obtained copolymer.

Dynamic light scattering measurements

The mean particle size and size distribution of micelleswere determined by dynamic light scattering (DLS) using a90 Plus Particle Size Analyzer (Brookhaven InstrumentsCorporation). Each analysis lasted for 3 min and was per-formed from 25 to 458C at a step of 18C. The sample waskept at each temperature for 10 min. All the measurementswere repeated three times. Average size and size distribu-tion were determined. Values reported are the mean diame-ter 6 SD for two replicate samples.

Atomic force microscope measurements

Atomic force microscope (AFM) (SPI3800, Seiko Instru-ments, Japan) was used to investigate the morphology of

Figure 1. Scheme of copolymer synthesis. (a) Synthesis of poly(IPAAm-co-AAm) with one hydroxyl end-group. (b) Synthe-sis of poly(IPAAm-co-AAm) with DL-lactide.

THERMOSENSITIVE POLYMERIC MICELLES FOR DOCETAXEL DELIVERY 849


docetaxel-loaded micelles in detail. One drop of properlydiluted docetaxel-loaded micelle suspension was placed onthe surface of a clean silicon wafer and dried at 378C (belowthe LCST) or at 438C (above the LCST). The AFM observa-tion was performed with a 20 mm scanner in tapping mode.

X-ray diffraction measurements

X-ray diffraction (XRD) patterns of docetaxel, emptymicelles, and docetaxel-loaded micelles were obtained withan X’Pert Pro (PANalytical, Netherlands) diffraction meteremploying Ni-filtered CuKa radiation (0.15418 nm) at roomtemperature. The X-ray tube was operated at 40 kV and150 mA. The scanning rate was 0.048.

Fourier-transfer infrared spectrum measurements

Fourier-transfer infrared (FTIR) spectra were recorded on aVector-22 (Bruker, Germany) type spectrometer. In each ex-periment, docetaxel, empty micelles, and docetaxel-loaded mi-celles were first mixed with KBr at a ratio of 1/99 (by weight).A self supporting wafer with a 1.3-cm diameter was preparedby pressing 30 mg of the mixture and was then loaded into anIR cell with BaF2 windows. The spectra were recorded atroom temperature by accumulating 20 scans at a spectral reso-lution of 4 cm�1, using the empty cell as background.

Drug loading content and entrapment efficiency

The drug loading content and entrapment efficiency weredetermined, using high-performance liquid chromatography

(HPLC) (Waters, 600E, US). The HPLC system was equippedwith a Dikma Dimansil C18 column (4.6 mm � 150 mm,5 mm). The mobile phase consisted of acetonitrile and water(47:53 v/v). The flow rate was set at 1.0 mL/min, and thedetection wavelength was 230 nm. Sample solution wasinjected at a volume of 20 mL.

The drug loading content and entrapment efficiencywere calculated by:

Drug loading content ð%Þ¼weight of drug inmicelles

weight ofmicelles�100;

Entrapment efficiency ð%Þ

¼ weight of drug inmicelles

weight of drug fed initially�100:

In vitro release studies were carried out as follows. Drug-loaded micellar suspensions (5 mL) (corresponding to 1.0 mgof docetaxel) were placed in a dialysis membrane bag withcut-off molecular weight of 12,000, tied, and immersed into200 mL of PBS. The entire system was kept at 378C with con-tinuous magnetic stirring. At selected time intervals, 0.5 mLof the release media was withdrawn. The concentration ofdocetaxel was determined by HPLC. In the same way, thein vitro release of drug at 438C was investigated by raisingthe system temperature from 37 to 438C.

In vitro cytotoxicity studies

For in vitro cytotoxicity study, HUVEC and three differenttumor cell lines (human gastric carcinoma cell line BGC823,human hepatic carcinoma cell line HepGII, and human col-orectal adenocarcinoma cell line LoVo) were used.

In vitro cytotoxicity of copolymers was determined bystandard MTT assays. HUVEC cells were seeded in a 96-well plate at a density of 5,000 cells per well and incubatedat 378C in a humidified atmosphere with 5% CO2. The cul-ture medium was endothelial cell growth medium (EGM)supplemented with 10% fetal bovine serum and changedevery other day until 80% confluence were reached. Themedium was then replaced with 200 mL medium with freedocetaxel, poly(IPAAm-co-AAm), and poly(IPAAm-co-AAm)-b-PDLLA of different concentrations. One row of 96-well

Figure 2. 1H-NMR spectra of poly(IPAAm-co-AAm)-b-PDLLA.

TABLE IMajor Characteristics of Copolymers

Mw (Da) Mn (Da) Dp LCST (8C)

Poly(IPAAm-co-AAm) 8,700 5,500 1.58 50Poly(IPAAm-co-AAm)-b-

PDLLA 22,600 9,600 2.35 41

Mw, weight-averaged molecular weight; Mn, number-averaged molecular weight; Dp, polydispersity.

Figure 3. 13C NMR spectra of poly(IPAAm-co-AAm)-b-PDLLA.

850 YANG ET AL.


plates was used as control with 200 mL culture medium only.The cells were incubated at 438C for 30 min to simulateatmosphere of hyperthermia and then were incubated at 378Cfor 24 h. The cell viability was determined by means of MTTenzyme assay,24 using an ELISA reader (Huadong, DG-5031,Nanjing). In vitro cytotoxicity without hyperthermia was alsoevaluated in a similar manner except incubating cells at 438Cafter replacing the medium.

In vitro cytotoxic effects of free docetaxel, docetaxel-loadedmicelles, and empty micelles were measured using three car-cinoma cell lines. The culture medium is RPMI 1640 supple-mented with 10% calf blood serum. Different concentrationsof docetaxel-loaded micelles, nonloaded micelles, and doce-taxel were used to find the cytotoxic activity above or belowthe LCST by the same MTT assays as hereinbefore. To deter-mine the interaction between docetaxel and hyperthermia, theWebb coefficient was applied.25 Predicted value (C) was cal-culated according to the equation C ¼ A � B/100, where (A)and (B) indicate survival values with drugs or hyperthermia.Synergism in drugs and hyperthermia was indicated byobserved survival which was less than the predicted survival.

Fluorescence microscopy measurement

About 100 mg poly(IPAAm-co-AAm)-b-PDLLA was dis-solved in 5 mL pure water. Then, 0.2 mL DMSO solution of

FITC (1 mg/mL) was added in drops and stirred for 2 h indark place. The solution was dialyzed against pure water usinga cut-off molecular weight of 8,000 (Sorua, Germany) to removefree FITC and form micelles. The purified polymeric micelleslabeled with FITC were kept in dark place before use.

Approximately 5 � 105 BGC823 cells were seeded in a25 mL cell culture flask with RPMI 1640 supplemented with10% calf blood serum. After incubation at 378C in a humidi-fied atmosphere with 5% CO2 for 24 h, cells were replacedwith 10 mL medium containing 0.2 mL FITC labeled micellessolution. The cells were incubated at 438C for 30 min to simu-late the atmosphere of hyperthermia and then were incubatedat 378C for 23.5 h. Detached from culture flask by trypsin, thecells were washed with PBS for three times to remove extrac-ellular/nonassociated micelles. Next, the cells were smearedon a slide with cell smear centrifugal apparatus. The cellswere observed under a fluorescence microscopy (Olympus,BX41-32H02-FLB3, Japan). As a contrast, cells added withFITC labeled micelles were observed in a similar mannerwithout incubating cells at 438C after replacing the medium.

RESULTS AND DISCUSSION

Copolymer synthesis

Compositions of poly(IPAAm-co-AAm) and poly(IPAAm-co-AAm)-b-PDLLA were confirmed by 1H-NMR and 13C-NMR spectra. The molecular ratio of poly(IPAAm-co-AAm):PDLLA in the copolymer was ap-proximately 1.3:1, which was measured from peakarea of methine protons from PIPAAm (4.0 ppm) andPDLLA (5.1 ppm), respectively, in 1H-NMR spectrumas displayed in Figure 2. There were three peaks withchemical shift around 180 ppm in 13C-NMR spectrumrecorded with inverse gated decoupled pulse sequence(Fig. 3), representing three different carbonyl carbon inthe polymer (PDLLA, 169.7 ppm; PIPAAm, 175.0 ppm;AAm, 180.4 ppm) and the areas of them were 8.27:9.9:1,which represent the ratios of three monomers. The ratioof PIPAAm:PDLLA in 13C-NMR spectra was calculated

Figure 4. Particle size of micelles at different temperature.

Figure 5. AFM images of micelles on the mica dried at 378C (A) and 438C (B). [Color figure can be viewed in the onlineissue, which is available at www.interscience.wiley.com.]



to be 1.2:1, which was close to the results in 1H-NMRspectrum.

The molecular weight of two copolymers was mea-sured by GPC, and the calculated results were sum-marized in Table I. From the table, it can been seenthat the molecular weight of poly(IPAAm-co-AAm)block is 8700 Da that can be easily excreted by glomer-ular filtration, eliminating the danger of accumulationof this substance in vivo.

A primary criterion to judge whether a thermosensi-tive polymer is suitable as a drug carrier for thermo-targeted delivery is to examine its LCST, which shouldbe above physiological temperature (378C), but belowthat used in hyperthermia (about 428C). Thus, it canbe easily used in clinical hyperthermia without induc-ing deleterious side effects in surrounding healthy tis-sue. Because the LCST of PIPAAm is only 328C, wecopolymerized N-isopropylacrylamide with AAm, ahydrophilic monomer, to increase the LCST from 32 to508C. After an additional polymerization step withPDLLA, the LCST of the copolymer brought downfrom 50 to 418C. Chilkoti et al. reported a kind of

amine-terminated poly(IPAAm-co-AAm) whose ratioof N-isopropylacrylamide to AAm was 84:16 similarto ours.10 The LCST of their copolymer is 408C, a valueclose to the LCST of our copolymer poly(IPAAm-co-AAm)-b-PDLLA. One possible explanation is that thehydroxyl in the terminal of poly(IPAAm-co-AAm)increased the LCST of copolymer. In our case, aftercoupled with PDLLA block and then formed micelles,the poly(IPAAm-co-AAm) moiety which formed theouter hydrated shell would behave its own characterwhich determined the LCST as 418C.

Morphology and size distribution

Since particle size has a crucial impact on the in vivofate of a particular in drug delivery system, the controlover the particle size is of great importance for drugcarriers. We found that the sizes of empty micelleswere around 80 nm and temperature-independentfrom 35 to 398C (Fig. 4). Polydispersity of the micelleswas less than 0.1, indicating the uniform size of themicelles. Initially, when the temperature increased, theparticle size decreased a little first in the range of 40–418C, and then dramatically jumped after temperaturereached 428C, accompanying a much larger polydisper-sity. This result suggested that the LCST of preparedcopolymer was about 418C (the inflexion of the curve)as the micelles became inhomogeneous at this tempera-ture. We speculated that the small decrease of the parti-cle size from 40 to 418C was caused by the shrink of thelong hydrophilic chains in outer shell when it becamemore hydrophobic at higher temperature. Once abovethe LCST, the micelles started to aggregate because ofthe absence of hydrophobic shell in micelles, leading tocontinuous increase of particle size.

The docetaxel-loaded micelles showed similar sizesand changes with temperature, which indicated thatdrug incorporation did not affect particle size and thechange of particle size in LCST range. Because thepore sizes in solid tumor vasculature vary from 100

Figure 6. Cumulative in vitro release profiles of docetaxel-loaded micelles at 378C (a) and 438C (b).

Figure 7. XRD patterns of docetaxel (a), empty micelles(b), and docetaxel-loaded micelles (c).

Figure 8. FTIR spectra of docetaxel (a), empty micelles (b),and docetaxel-loaded micelles (c).

852 YANG ET AL.


to 780 nm,26,27 which are much larger than the junc-tions in normal tissue where the gaps are usually lessthen 6 nm,28 these micelles could easily permeate totumor sites, but hardly access to normal tissues. Also,the nonselective scavenging of micelles by RES wasreduced, because of the hydrophilic poly(IPAAm-co-AAm) shell in micelle, which may guarantee theirlong circulation in vivo.

To investigate the morphology of micelles, AFMobservations of docetaxel-loaded micelles were con-ducted at both 378C [Fig. 5(A)] and 43 8C [Fig. 5(B)].As shown in Figure 5(A), all of the micelles had anear-spherical shape with the diameter around 100nm, and there were no free drug crystals on the sur-face of the micelles to be detected. In Figure 5(B), themicelles contacted with each other and also retainedspherical shapes which supported that the micellesdid not fuse, but aggregated via surface-surface inter-action instead above the LCST.

Drug loading, encapsulation efficiency and stateof the drug incorporation in micelles

The maximum loading content of docetaxel was27.1% and the entrapment efficiency was 57.8% as mea-sured by HPLC analysis. We investigated the cumula-tive in vitro release profile of docetaxel loaded micellesin PBS at both 378C [Fig. 6(a)] and 438C [Fig. 6(b)]. Therelease percentage at both temperatures exhibited aburst at the initial stage and then slowed down gradu-ally. It should be noticed that the release rate at bothtemperatures was similar for the first 10 h, suggestingthat heating for a short time could not accelerate drugrelease from micelles. The release rate at 378C wasslower than that at 438C and the difference becamelarger with time. We assumed that the hydrophilicouter shell became hydrophobic that enhanced ad-sorption of docetaxel to micelles and kept them frombeing released to aqueous medium. Therefore, thetotal release rate of docetaxel did not change obviouslyat the initial stage although high temperature quick-ened up drug release. Both at 37–438C the release ratewas very low and the drug release was continuousin vitro over 72 h. The sustained release is a very im-portant character of drug delivery system, which leadsto longer circulatory time and less toxicity to normaltissues.

The XRD patterns of different samples are pre-sented in Figure 7. The empty micelles were amor-phous, so their XRD patterns just gave a broad peak[Fig. 7(b)]. Free docetaxel crystals showed several dis-tinct sharp peaks [Fig. 7(a)], which disappeared uponbeing entrapped into micelles [Fig. 7(c)], suggestingmolecular or amorphous dispersion of docetaxel in themicelles.29

Figure 8 showed the FTIR spectra over the range of400–2000 cm�1 for docetaxel, empty micelles, anddocetaxel-loaded micelles. Docetaxel had characteris-tic bands at 1715, 1248, and 710 cm�1 [Fig. 8(a)]. Thesebands were not shown in empty micelles [Fig. 8(b)],but still in docetaxel-loaded micelle samples withoutdistinct shift [Fig. 8(c)], which indicated that doce-taxel was physically entrapped in the polymer matrix

TABLE IIIC50 (lg/ml) of Different Cell Lines Treated by Poly(IPAAm-co-AAm), Empty Micelles, Docetaxel-Loaded

Micelles, and Free Docetaxel With and Without Hyperthermia

HUVEC HepG II BGC823 Lovo

378Ca 438Cb 378C 438C 378C 438C 378C 438C

Poly(IPAAm-co-AAm) 757 327 – – – – – –Empty micelles 521 275 1207 796 1012 858 1075 858Docetaxel-loaded micelles – – 0.152 0.021 0.128 0.015 0.115 0.013Free docetaxel 0.007 0.004 0.037 0.015 0.014 0.010 0.028 0.017

IC50, inhibitory concentration of 50% cell inhibition or death.aWithout incubating cells at 438C after replacing the medium.bIncubating cells at 438C for 30 min after replacing the medium.

Figure 9. In vitro cytotoxicity of docetaxel, poly(IPAAm-co-AAm) and poly(IPAAm-co-AAm)-b-PDLLA on HUVECwithout hyperthermia (A) and with hyperthermia (B).



and there was no chemical interaction between doce-taxel and polymer.

In vitro cytotoxicity studies

To determine in vitro cytotoxicity of poly(IPAAm-co-AAm) and poly(IPAAm-co-AAm)-b-PDLLA, HUVECcells were used. The concentrations required to inhibitgrowth by 50% (IC50 values) were evaluated from thedose-response curves (Fig. 9) and showed in Table II.The results obtained from both cytotoxicity assayswere confirmed by repeating the experiment on threeindependent occasions and testing in triplicate eachtime. With or without hyperthermia, poly(IPAAm-co-AAm) and poly(IPAAm-co-AAm)-b-PDLLA showed

significantly (p < 0.01) lower toxicity than free doce-taxel. The viability of HUVEC cells with hyperthermia(438C) was lower than cells without hyperthermia (p <0.05), which indicated that hyperthermia itself had cer-tain toxicity to HUVEC cells. Cytotoxicity of copoly-mers increased because the copolymers became hydro-phobic and enhanced their nonspecific adsorption tocells mediated by hydrophobic interactions betweencells and polymeric material at a temperature higherthan the LCST. But either at 378C or at 438C, such lowtoxicity of copolymers to HUVEC ensured their highsafety in vivo.

Tables II–V and Figure 10 reflected the cytotoxicityto carcinoma cells directly. The results of all the threecell lines indicated the same trend, in which hyperther-mia itself also had certain cytotoxicity to carcinoma

TABLE IIICytotoxic Action of Docetaxel-Loaded Micelles and Free Docetaxel in BGC823 Cell Lines

Docetaxel Concentration(ng/ml)

Survival Values (%) of BGC 823 Cell Lines

Docetaxel-Loaded Micelles Free Docetaxel

378Ca 438Cb Predicted Valuec 378Ca 438Cb Predicted Valuec

0 0.910d 0.910d

2.5 0.880 0.820 0.801e 0.920 0.860 0.8375 0.830 0.710 0.755e 0.910 0.785 0.828e

10 0.880 0.560 0.801e 0.676 0.560 0.615e

20 0.870 0.470 0.792e 0.500 0.438 0.455e

40 0.840 0.360 0.764e 0.340 0.294 0.309e

80 0.790 0.210 0.719e 0.235 0.214 0.214e

160 0.357 0.130 0.325e 0.155 0.126 0.141e

320 0.256 0.030 0.233e 0.038 0.004 0.035e

aWithout incubating cells at 438C after replacing the medium.bIncubating cells at 438C for 30 min after replacing the medium (observed cell viability).cPredicted survival values, evaluated by the Webb method.25dTreat with hyperthermia only.eSynergism in drugs and hyperthermia (indicated by observed survival less than the predicted survival).

TABLE IVCytotoxic Action of Docetaxel-Loaded Micelles and Free Docetaxel in HepGII Cell Lines

Docetaxel Concentration(ng/ml)

Survival Values (%) of HepGII Cell Lines



0 0.870d 0.884d

2.5 0.937 0.835 0.815e 0.967 0.843 0.855e

5 0.925 0.734 0.805e 0.932 0.832 0.82410 0.912 0.620 0.793e 0.772 0.640 0.682e

20 0.887 0.492 0.772e 0.540 0.473 0.477e

40 0.874 0.403 0.760e 0.492 0.429 0.435e

80 0.833 0.340 0.725e 0.485 0.425 0.428e

160 0.452 0.252 0.393e 0.391 0.324 0.346e

320 0.268 0.130 0.233e 0.168 0.071 0.149e


854 YANG ET AL.


cells. Evaluated by the Webb coefficient method,25 freedocetaxel showed more toxicity at 438C than at 378C.The observed survival was less than predicated sur-vival (Tables III–V), which implied possible synergeticeffect of docetaxel in combination with hyperthermia.Drug-loaded micelles showed less toxicity than freedocetaxel itself (p < 0.01) at 378C, but demonstratedsimilar toxicity at 438C (p > 0.05). Because heating in ashort time could not facilitate drug release, this en-hancement of cytotoxicity was apparently caused bythe fact that the micelles became hydrophobic andactively interacted with cells to deposit drugs locallyat a temperature higher than the LCST, while the hy-drated form stayed in aqueous solution when temper-ature was below the LCST. As a result, the cytotoxicityof docetaxel-loaded micelles was approaching to thatof free docetaxel after heated at 438C for 30 min alt-hough only 6.2% docetaxel was released. This resultsuggested that the docetaxel-loaded micelles can causethe specific cytotoxicity by local heating, making itpossible to achieve thermal targeted delivery for anti-tumor drugs.

Interaction between micelles and cells

The internalization of polymeric micelles into cellshas been reported previously.30–32 In this study, the in-teraction between micelles labelled with FITC andBGC823 cells was observed by fluorescence microscopy(Fig. 11). Cells with hyperthermia at 438C for 30 min[Fig. 11(D)] showed stronger fluorescent signals in cellsthan those without hyperthermia [Fig. 11(B)]. Becausecells have been washed with PBS to remove extracellu-lar/nonassociated micelles, the fluorescent signals rep-resented micelles interaction with cells. Stronger fluo-rescent signals indicated stronger interaction between

TABLE VCytotoxic Action of Docetaxel-Loaded Micelles and Free Docetaxel in LoVo Cell Lines

DocetaxelConcentration (ng/ml)

Survival Values (%) of LoVo Cell Lines



0 0.873d 0.897d

2.5 0.943 0.810 0.823e 0.954 0.817 0.856e

5 0.924 0.720 0.807e 0.914 0.757 0.820e

10 0.918 0.576 0.801e 0.763 0.633 0.684e

20 0.912 0.444 0.796e 0.572 0.486 0.513e

40 0.883 0.384 0.771e 0.472 0.429 0.42380 0.808 0.305 0.705e 0.417 0.369 0.374e

160 0.335 0.230 0.292e 0.280 0.271 0.251320 0.188 0.117 0.164e 0.141 0.124 0.126e


Figure 10. In vitro cytotoxicity of docetaxel, empty micellesand docetaxel-loaded micelles on different carcinoma celllines with and without hyperthermia. The reference was re-placed with culture medium only. A: Gastric carcinoma cellline BGC823. B: Hepatic carcinoma cell line HepGII. C: Col-orectal adenocarcinoma cell line LoVo.



micelles and cells after hyperthermia. Thus drug en-trapped by micelles would be internalized to tumorcells easier and be more effective. This explained thecytotoxicity results that docetaxel-loaded micellesshowed much more cytotoxicity with hyperthermiathan without hyperthermia although their release rateswere similar in a short time.

CONCLUSIONS

The copolymer reported here exhibits an LCST of418C, making it a perfect candidate as the coating ma-terial for anticancer compounds. Micelles preparedfrom this copolymer not only exist as a highly solubleform at physiological temperature, but also may evadefrom nonselective RES scavenging because of its smallsize approximately at 80 nm in diameter. Docetaxel,the cytotoxic drug used in this study, is molecularly oramorphously dispersed inside micelles and can mea-sure up to 27% of total mass. When heated at an ele-vated temperature like 438C, this formulation of doce-taxel demonstrates similar efficacy as docetaxel itself,indicating transition of polymer coat. More impor-tantly, our results indicate this micelle-wrapped doce-taxel shows far less toxicity than free drug at physiolog-ical temperature, promising to reduce side effectsgreatly in vivo. Furthermore, once the drug is removed,both the micelle itself and its decomposed fragmenthave low toxicity to vein endothelial cells.

With the above-mentioned properties, this kind ofmaterial would have broad applications in thermal tar-geting drug delivery. In the further study, the in vivoactivities of the micelles will be investigated to assessits value in the clinical practice.

The authors also thank Jinlin Pharmaceutical Com-pany for HPLC measurements and Dr. Shuwei Li forproofreading.

References

1. Andresen TL, Jensen SS, Jorgensen K. Advanced strategies inliposomal cancer therapy: Problems and prospects of activeand tumor specific drug release. Prog Lipid Res 2005;44:68–97.

2. Valero V, Holmes FA, Walters RS, Theriault RL, Esparza L,Fraschini G, Fonseca GA, Bellet RE, Buzdar AU, Hortobagyi GN.Phase II trial of docetaxel: A new, highly effective antineo-plastic agent in the management of patients with anthracy-cline-resistant metastatic breast cancer. J Clin Oncol 1995;13:2886–2894.

3. Ferraresi V, Milella M, Vaccaro A, D’Ottavio AM, Papaldo P,Nistico C, Marsella A, Carpino A, Giannarelli D, Terzoli E,Cognetti F. Toxicity and activity of docetaxel in anthracycline-pretreated breast cancer patients: A phase II study. Am J ClinOncol Cancer Clin Trials 2000;23:132–139.

4. Esmaeli B, Valero V, Ahmadi MA, Booser D. Canalicular steno-sis secondary to docetaxel (Taxotere)—A newly recognized sideeffect. Ophthalmology 2001;108:994–995.

5. Schrump DS, Nguyen DM. Novel molecular targeted therapyfor esophageal cancer. J Surg Oncol 2005;92:257–261.

6. Nobuto H, Sugita T, Kubo T, Shimose S, Yasunaga Y, MurakamiT, Ochi M. Evaluation of systemic chemotherapy with magneticliposomal doxorubicin and a dipole external electromagnet. IntJ Cancer 2004;109:627–635.

7. Kim GJ, Nie S. Targeted cancer nanotherapy. Mater Today2005;8:28–33.

8. Reisen JV, Polley GT, Verheijen PJ. Structural targeting forheat integration retrofit. Appl Therm Eng 1998;18:283–294.

9. Kong G, Dewhirst MW. Hyperthermia and liposomes. IntJ Hyperthermia 1999;15:345–370.

10. Chilkoti A, Dreher MR, Meyer DE, Raucher D. Targeted drugdelivery by thermally responsive polymers. Adv Drug Deliv Rev2002;54:613–630.

11. Dewhirst MW, Prosnitz L, Thrall D, Prescott D, Clegg S,Charles C, MacFall J, Rosner G, Samulski T, Gillette E, LaReu S.Hyperthermic treatment of malignant diseases: Current statusand a view toward the future. Semin Oncol 1997;24:616–625

12. Riehemann K, Schmitt O, Ehlers EM. The effects of thermoche-motherapy using cyclophosphamide plus hyperthermia on themalignant pleural mesothelioma in vivo. Ann Anat 2005;187:215–223.

13. Kohori F, Sakai K, Aoyagi T, Yokoyama M, Sakurai Y, Okano T.Preparation and characterization of thermally responsive blockcopolymer micelles comprising poly(N-isopropylacrylamide-b-DL-lactide). J Control Release 1998;55:87–98.

14. Schild HG. Poly(N-isopropylacrylamide): Experiment, theoryand application. Prog Polym Sci 1992;17:163–249.

15. Jeong B, Kim SW, Bae YH. Thermosensitive sol-gel reversiblehydrogels. Adv Drug Deliv Rev 2002;54:37–51.

16. Zhang JX, Qiu LY, Jin Y, Zhu KJ. Thermally responsive poly-meric micelles self-assembled by amphiphilic polyphosphazenewith poly(N-isopropylacrylamide) and ethyl glycinate as sidegroups: Polymer synthesis, characterization, and in vitro drugrelease study. J Biomed Mater Res A 2006;76:773–780.

Figure 11. Fluorescence mcroscopy image of BGC 823 in-cubated with FITC labeled micelles for 24 h at different tem-peratures. A, B: without hyperthermia at 438C (A: Brightfield, B: Fluorescence field). C, D: with hyperthermia at 438Cfor 30 min (C: Bright field, D: Fluorescence field). Cells at438C for 30 min showed stronger fluorescent signals in cellsthan those without hyperthermia. Stronger fluorescent sig-nals indicated stronger interaction between micelles andcells after hyperthermia. [Color figure can be viewed in theonline issue, which is available at www.interscience.wiley.com.]

856 YANG ET AL.


17. Chung JE, Yokoyama M, Yamato M, Aoyagi T, Sakurai Y,Okano T. Thermo-responsive drug delivery from polymericmicelles constructed using block copolymers of poly(N-isopro-pylacrylamide) and poly(butylmethacrylate). J Control Release1999;62:115–127.

18. Meyer DE, Shin BC, Kong GA, Dewhirst MW, Chilkoti A.Drug targeting using thermally responsive polymers and localhyperthermia. J Control Release 2001;74:213–224.

19. Kataoka K, Harada A, Nagasaki Y. Block copolymer micellesfor drug delivery: Design, Characterization and biological sig-nificance. Adv Drug Deliv Rev 2001;47:113–131.

20. Bostman O, Pihlajamaki H. Clinical biocompatibility of biode-gradable orthopaedic implants for internal fixation: A review.Biomaterials 2000;21:2615–2621.

21. Kosmas C, Tsavaris N, Mylonakis N, Kalofonos HP. An over-view of current results with the gemcitabine and docetaxelcombination as initial and salvage chemotherapy regimen inadvanced non-small cell lung cancer. Crit Rev Oncol Hematol2003;45:265–275.

22. Belani CP, Eckardt J. Development of docetaxel in advancednon-small-cell lung cancer. Lung Cancer 2004;46:S3–S11.

23. Berry WR. The evolving role of chemotherapy in androgen-in-dependent (hormone-refractory) prostate cancer. Urology 2005;65:2–7.

24. Fonseca C, Simoes S, Gaspar RE. Paclitaxel-loaded PLGAnanoparticles: Preparation, physicochemical characterization

and in vitro anti-tumoral activity. J Control Release 2002;83:273–286.

25. Yeh YA, Herenyiova M, Weber G. Quercetin: Synergistic actionwith carboxyamidotriazole in human breast carcinoma cells. LifeSci 1995;57:1285–1292.

26. Jang SH, Wientjes MG, Lu D, Au JLS. Drug delivery and trans-port to solid tumors. Pharm Res 2003;20:1337–1350.

27. Hobbs SK, Monsky WL, Yuan F, Roberts WG, Griffith L,Torchilin VP, Jain RK. Regulation of transport pathways intumor vessels: Role of tumor type and microenvironment.Proc Natl Acad Sci USA 1998;95:4607–4612.

28. Drummond DC, Meyer O, Hong KL, Kirpotin DB, Papahadjo-poulos D. Optimizing liposomes for delivery of chemothera-peutic agents to solid tumors. Pharmacol Rev 1999;51:691–743.

29. Zhang LY, Hu Y, Jiang XQ, Yang CZ, Lu W, Yang YH. Campto-thecin derivative-loaded poly(caprolactone-co-lactide)-b-PEG-b-poly(caprolactone-co-lactide) nanoparticles and their biodistri-bution in mice. J Control Release 2004;96:135–148.

30. Savie R, Luo L, Eisengerg A, Mayslinger D. Micellar nanocon-tainers distribute to difined cytoplasmic organelles. Science 2003;300:615–618.

31. Shuai X, Merdan T, Schaper AK, Xi F, Kissel T. Core-cross-linked polymeric micelles as paclitaxel carriers. Bioconjug Chem2004;15:441–448.

32. Rapoport N. Combined cancer therapy by micellar-encapsulateddrug and ultrasound. Int J Pharm 2004;277:155–162.



novel thermosensitive polymeric micelles for docetaxel delivery

Documents