Contents lists available at SciVerse ScienceDirect
Journal of Controlled Release
j ourna l homepage: www.e lsev ie r .com/ locate / jconre l
NANOMEDICIN
E
Functionalized polymersomes with outlayered polyelectrolyte gels forpotential tumor-targeted delivery of multimodal therapies andMR imaging
Wen-Hsuan Chiang a,1, Wen-Chia Huang b,1, Chien-Wen Chang a, Ming-Yin Shen c, Zong-Fu Shih a,Yi-Fong Huang b, Sung-Chyr Lin b, Hsin-Cheng Chiu a,⁎a Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu 300, Taiwanb Department of Chemical Engineering, National Chung Hsing University, Taichung 402, Taiwanc Department of Surgery, National Taiwan University Hospital—Hsinchu Branch, Hsinchu 300, Taiwan
Article history:Received 22 October 2012Accepted 10 March 2013Available online 4 April 2013
Keywords:Magneto-thermo-chemotherapyPolymersomesSuperparamagnetic iron oxide nanoparticlesTumor-targetingStimuli-triggered drug releaseMR imaging
A novel tumor-targeting polymersome carrier system capable of delivering magnetic resonance imaging (MRI)and chemotherapy is presented in this study. The doxorubicin (DOX)-loaded magnetic polymersomes werefirst attained by the self-assembly of lipid-containing copolymer, poly(acrylic acid-co-distearin acrylate), inaqueous solution containing citric acid-coated superparamagnetic iron oxide nanoparticles (SPIONs), andfollowed by DOX loading via electrostatic attraction. To further functionalize these artificial vesicles with superi-or in vivo colloidal stability, pH-tunable drug release and active tumor-targeting, chitosan and poly(γ-glutamicacid-co-γ-glutamyl oxysuccinimide)-g-poly(ethyleneglycol)-folate (FA) were deposited in sequence onto theassembly outer surfaces. The interfacial nanogel layers via complementary electrostatic interactions andin-situ covalent cross-linking were thus produced. These nanogel-caged polymersomes (NCPs) show excellentanti-dilution and serumproteins-repellent behaviors. Triggerable release of the encapsulatedDOXwas governedby dual external stimuli, pH and temperature. When these theranostic NCPs were effectively internalized byHeLa cells via FA receptor-mediated endocytosis and then exposed to high frequency magnetic fields (HFMF),the combined effects of both pH and magnetic hyperthermia-triggered drug release and thermo-therapyresulted in greater cytotoxicity than the treatment by DOX alone. By virtue of the SPION clustering effect inthe assembly inner aqueous compartments, the SPION/DOX-loaded NCPs displayed an r2 relaxivity value(255.2 Fe mM−1 S−1) higher than Resovist (183.4 Fe mM−1 S−1), a commercial SPION-based T2 contrastagent. The high magnetic relaxivity of the tumor-targeting NCPs coupled with their enhanced cellular uptakeconsiderably promoted the MRI contrast of targeted cancer cells. These results demonstrate the great potentialof the FA-decorated SPION/DOX-loaded NCPs as an advanced cancer theranostic nanodevice.
Development of multifunctional nanodevices combining multimodaltumor therapies with diagnosis has been one of the major interests innanomedicine researches owing to their potential to promote the effica-cies of anticancer treatments [1–10]. By virtue of passive and activetumor targeting, theranostic nanocarriers are capable of delivering anti-cancer drugs anddiagnostic agents to target tumor sites for simultaneousmonitoring of disease progression and biological responses to treat-ments. Because of the capability of non-invasivemagnetic resonance im-aging (MRI) techniques in delivering in vivo tomographic information ofillness tissues in real time, MRI contrast agents have become one of theprominent diagnostic candidates in constructing theranostic systems
886 35718649.).
rights reserved.
[11,12]. A selective transport of contrast agents such as SPIONs orparamagnetic GdIII complexes by tailor-made nanovehicles, includingliposomes [12–15], polymersomes (polymeric vesicles) [11,16–18], andpolymeric micelles [3,9,19–21], to the target site is crucial in impartingthe theranostic system ultrasensitive MRI contrast. SPIONs can alsoserve as a nano-converter for cancer thermotherapy to generatelocal heat from the input of external magnetic energy due to itsmagnetically-induced hyperthermia capability [7,17]. In addition to can-cer thermotherapy, remotely-controlled thermo-triggered drug releasecan thus be achieved by manipulating the external alternative high fre-quency magnetic fields (HFMF) [13,17].
Despite of their great potentials for anticancer treatments, the ap-plications of theranostic nanovehicles are frequently hampered bytheir low in vivo colloidal stability which often results in prematuredrug release and thus serious side effects and low therapeutic efficacy[11,12]. To overcome this issue, several approaches to improve thestructural stability of nanovehicles by covalent cross-linking of the
281W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
preformed nanoassemblies have been widely explored [22–25]. Forexample, through free radical polymerization of acrylate groups ofthe SPION-loaded polymeric vesicles composed of FA-poly(ethyleneglycol) (PEG)-b-poly(glutamate hydrozone DOX)-b-PEG-acrylate,the cross-linked polymeric vesicles showed superior in vivo stability,pH-triggered drug release, and enhanced MR imaging sensitivity [11].Another new strategy has also been proposed to create novelpolymer-caged nanobins (PCNs) with excellent structural stabilityby crosslinking poly(acrylic acid) terminated at both ends in thecholesterol residues incorporated within the bilayer membranes[12]. These PCNs were further equipped with excellent therapeuticand diagnostic functionalities through the encapsulation of antican-cer drug (gemcitabine) and MRI contrast agent (GdIII). Nevertheless,in vivo-stable theranostic platforms integrating hybrid cancer treat-ment and non-invasive biodiagnosis are still not well developed [17].
Herein, we propose a novel stable polymersome system in which thevesicle structure is stabilized by crosslinked-outlayered nanogel. Thenanogel-caged polymersomes (NCPs) were prepared by enclosing thepreformed SPION/DOX-loaded polymersomes with pH-responsiveFA-containing layered polyelectrolyte nanogels to achieve tumor-targeted magneto-thermo-chemotherapy and MRI (Scheme 1). TheSPION-loaded polymersomes were prepared by self-association oflipid-containing copolymer, poly(acrylic acid-co-distearin acrylate)(poly(AAc-co-DSA)), in the citric acid-coated SPION aqueous disper-sion. DOX was then loaded into the SPION-bearing polymersomesthrough its π–π stacking association and electrostatic attraction withthe ionized AAc residues of poly(AAc-co-DSA). By the layer-by-layer ad-sorption technique, chitosan and poly(γ-glutamic acid-co-γ-glutamyl
Scheme 1. Development of versatile FA-conjugated SPION/DOX-loaded NCPs.
oxysuccinimide)-g-poly(ethylene glycol)-folate (poly(γ-GA-co-γ-GAOSu)-g-PEG-FA) were deposited in sequence onto the outer surfacesof SPION/DOX-loaded polymersomes to form the layered polyelectro-lyte gels via electrostatic interactions and in-situ covalent crosslinkingby aminolysis of the primary amino groups of chitosan and the OSumoieties of (poly(γ-GA-co-γ-GAOSu)-g-PEG-FA). The FA-conjugatedSPION/DOX-loaded NCPs exhibit several outstanding characteristicsincluding: (1) highly stable colloidal structure against large volumedilution and protein adsorption, (2) rapid drug release in the intracellu-lar acidic endosomes/lysosomes and/or under HFMF, (3) superioranticancer efficacy by the combined active tumor-targeting andmagneto-thermo-chemotherapy, and (4) enhanced MRI sensitivity.
2. Materials and methods
2.1. Materials
γ-PGA (Mn > 100 KDa) was purchased from Vedan, Taiwan. Thefragmentation and fractionation of γ-PGA and synthesis of citricacid-coated SPIONs are described in detail in Supplementary data.N-Hydroxysuccinimide (NHS), and N,N′-dicyclohexylcarbodiimide(DCC)were obtained from Sigma-Aldrich, USA. Chitosanwas purchasedfrom Fluka, USA. Maleimide-PEG-NH2 (Mw = 5000 g/mol) was ac-quired from JenKem Technology, USA. Folic acid and citric acidmonohydratewere purchased fromFisher, USA. DOX (in the hydrochlo-ride salt form)was obtained from Seedchem, Australia. The human cer-vical tumor cell line, HeLa cell, was purchased from Food IndustryResearch and Development Institute of Taiwan. The Dulbecco's modi-fied Eagle medium (DMEM) and Hoechst 33258 were purchased fromInvitrogen, USA. AlamarBluewas purchased fromAbD Serotec, England.All other chemicals were reagent grade and used as received.
2.2. Preparation of SPION/DOX-loaded polymersomes
Poly(AAc-co-DSA) (DSA 13 mol%; 10.0 mg), prepared and charac-terized as reported in the literature and Supplementary data [26–28],was dissolved in THF (1.0 mL). The copolymer solution (0.1 mL) wasdropped slowly into acetate buffer (pH 5.0, 3.0 mL, ionic strength0.01 M) containing citric acid-coated SPIONs (0.3 mg/mL). Themixturewas then subjected to ultrasonication at 25 °C for 10 min, followed byequilibration at 25 °C for 6 h. After repeated ultrafiltration (Amicon8010 with a Millipore PBMK membrane, MWCO 300,000) againstpH 5.0 acetate buffer to remove THF and then centrifugation(3000 rpm, 20 min) to eliminate unloaded SPIONs, SPION-loadedpolymersomes used in this work were obtained. The aqueous solutionof DOX (10−3 M, pH 5.0, 1.0 mL) was added directly into the aqueoussuspension of the SPION-loaded polymersomes (pH 7.0, 2.0 mL) to afinal concentration of 3.3 × 10−4 M. The suspension was then stirredat 4 °C for 12 h to complete the drug loading process.
2.3. Synthesis of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA
The synthetic route of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA is illus-trated in Scheme 2. Fractured γ-PGA and NHS (50 mol% with respectto the γ-GA residues) were simultaneously dissolved in anhydrousDMSO/pyridine co-solvent (3/1 (v/v)). Upon the addition of DCC,the coupling reaction was allowed to proceed at 4 °C for 48 h. Thegrafting of poly(γ-GA-co-γ-GAOSu) with maleimide-PEG-NH2
(Mw = 5000 g/mol) (10 mol% with respect to the γ-GA residues inγ-PGA) was conducted at 25 °C for 96 h. After the removal of theby-product, N,N′-dicyclohexylurea, by repeated filtration, the poly-mer solution was dialyzed (Cellu Sep MWCO 12,000–14,000) againstDMSO and was then collected by precipitation from diethylether. Poly(γ-GA-co-γ-GAOSu)-g-PEG-maleimide thus obtained wasdissolved with thiol-functionalized folate (FA-SH) in equal molarityto maleimide units in anhydrous DMF and allowed to react at room
Scheme 2. Synthetic route of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA.
282 W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
temperature for 5 h. The copolymer was purified by dialysis againstDMSO and collected by precipitation in diethyl ether. γ-PGA-g-PEG-FA was prepared by hydrolyzing the γ-GAOSu residues ofpoly(γ-GA-co-γ-GAOSu)-g-PEG-FA in DMSO into γ-GA moieties byaddition of tris buffer (0.01 M, pH 8.5).
2.4. Preparation of polyelectrolyte gel-outlayered polymersomes
Suspension of SPION/DOX-loaded polymersomes (pH 7.0, 3.0 mL)was transferred into a chitosan solution (pH 6.0, 1.5 mL, 3.0 mg/mL)and thoroughly mixed by stirring for 2 h. Solution ofpoly(γ-GA-co-γ-GAOSu)-g-PEG-FA (pH 7.4, 0.4 mL, 3.0 mg/mL) wasincorporated into the above solution (4.5 mL) and the mixture afterits pH being adjusted to 7.4 was stirred at room temperature for 24 h.The suspension was then dialyzed (Cellu Sep MWCO 12,000–14,000)against pH 7.4 phosphate buffer. The FA-free SPION/DOX-loaded NCPsand non-crosslinked FA-conjugated SPION/DOX-loaded polymersomeswere prepared by using poly(γ-GA-co-γ-GAOSu)-g-PEG-maleimideand γ-PGA-g-PEG-FA, respectively, following the same procedure de-scribed above.
2.5. Structural characterization
The average molecular weight and polydispersity of the fracturedγ-PGA and poly(γ-GA-co-γ-GAOSu)-g-PEG-FA were determined bysize exclusion chromatography (SEC) (Agilent 1100, PL Aquagel-OHcolumns in series GF083: separation range 100–30 K; GF084: 10 K–200 K and GF086: 200 K–10 M, calibrated with standard PEG ofknown molecular weights with narrow molecular weight distribu-tions). The chemical composition of poly(γ-GA-co-γ-GAOSu)-g-PEG-FAwas determined by 1H NMR (Varian Unity Inova-500 NMR Spectrome-ter) using DMSO-d6 as the solvent. The characteristic absorbance of FAgroup (at 285 nm) of poly(γ-GA-co-γ-GA OSu)-g-PEG-FA was exam-ined byUV–Vis analysis (Hitachi U-2900, Japan) using DMSO as the sol-vent. The particle size distribution as well as the zeta potential ofSPION-loaded polymersomes, SPION/DOX-loaded polymersomes, andFA-decorated SPION/DOX-loaded NCPs were determined by dynamiclight scattering (DLS) (Malvern ZetaSizer Nano Series instrument,USA) with He–Ne laser 4 mW, λ = 633 nm. The particle size ofFA-conjugated SPION/DOX-loaded NCPs and polymersomes withoutany cross-linking after 100-fold dilution with FBS-containing (10%)aqueous medium or phosphate buffered saline (PBS) was alsoexamined. Transmission electronic microscopy (TEM) images ofnanoparticleswere obtained from a JEOL JEM-1200 CXIImicroscope op-erating at an accelerating voltage of 120 kV. Samples were prepared byplacing a few drops of the particle suspensions on a 300-mesh coppergrid covered with carbon and then negatively stained with uranyl ace-tate (2.0 wt.%) for 20 s and dried at ambient temperature for 2 daysprior to measurements. To determine the iron content of the SPION/DOX-loaded NCPs, freeze-dried SPION/DOX-containing NCPs wereadded to a 1 M HCl solution to achieve complete dissolution of SPIONs.Iron concentration was determined by inductively coupled plasmamass spectrometer (ICP-MS) (Agilent 7500ce, Japan). The SPIONloading content was calculated as the ratio of the weight of embeddedSPIONs to the total weight of the lyophilized SPION/DOX-loadedNCPs. To assess the drug loading level, a small portion of SPION/DOX-encapsulated NCPswaswithdrawn and dilutedwithDMF to a vol-ume ratio of DMF/H2O = 9/1. The amount of DOX encapsulated wasdetermined by a fluorescence spectrophotometer (Hitachi F-7500,Japan). The excitationwas performed at 480 nmand the emission spec-trumwas recorded in the range of 500–700 nm. Drug loading efficiency(DLE) and drug loading content (DLC) were calculated according to thefollowing formulas, respectively:
DLE %ð Þ ¼ weight of loaded DOX=weight of DOX in feedð Þ� 100%
DLC %ð Þ ¼�weight of loaded DOX=weight of lyophilized SPION=
DOX�loaded NCPs�
� 100%:
2.6. Evaluation of in vitro stimuli-triggered drug release
The dispersion (1.0 mL) of SPION/DOX-loaded NCPs was dialyzed(Cellu Sep MWCO 12,000–14,000) against succinic acid buffers ofpH 4.7 and 6.3, and phosphate buffer of pH 7.4 (50 mL, ionic strength0.15 M), respectively, at 37 °C. At the prescribed time intervals, thedialysate (1.0 mL) was withdrawn and replaced with an equivalentvolume of fresh media. The concentration of DOX was determinedas described above in aqueous solution of either pH 4.7, 6.3 or 7.4.The effects of magnetic hyperthermia on the release behavior of load-ed DOX were explored by exploiting a HFMF consisting of a powersupply, functional generator, amplifier, and cooling water. The coil(diameter 35 mm) was constructed in a 7-loop structure, with a fre-quency of 37 kHz and a magnetic field strength of 2.5 kA/m. A glasstube containing FA-conjugated SPION/DOX-loaded NCP suspension
Citric acid-covered SPIONs at pH 5.0 Pristine polymersomes at pH 5.0 SPION-loaded polymersomes at pH 5.0 SPION/DOX-loaded polymersomes at pH 7.0 FA-conjugated SPION/DOX-loaded NCPs at pH 7.4
a
-60
-45
-30
-15
0
15
30
45
(IV)
(III)
(II)(I)
Zet
a po
tent
ial (
mV
)
b
10 100 10000
4
8
12
16
243 nm25100 kcps
Inte
nsit
y (%
)
Hydrodynamic diameter (nm)
pH 7.4
pH 4.7c
Dh: 221 nm
Light scattering intensity: 33800 kcps
Fig. 1. (a) DLS particle size distribution profiles of citric acid-coated SPIONs and differ-ent polymersomes in aqueous solutions at 20 °C. (b) Zeta potentials of (I) citricacid-coated SPIONs at pH 5.0, (II) SPION/DOX-loaded polymersomes at pH 7.0, (III)chitosan-coated SPION/DOX-loaded polymersomes at pH 6.0 and (IV) FA-conjugatedSPION/DOX-loaded NCPs at pH 7.4 and 20 °C. (c) DLS particle size distribution profilesof FA-conjugated SPION/DOX-loaded NCPs in aqueous solutions of pH 7.4 and 4.7 at20 °C.
283W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
of pH 4.7 (1.0 mL) was positioned in the center of the HFMF for20 min. The DOX fluorescence intensity of the above solution after1-h incubation at 37 °C was measured.
2.7. Cellular uptake study
HeLa cells (2 × 105 cells/well) in a 6-well plate were treated withfree DOX and FA-conjugated and FA-free SPION/DOX-loaded NCPs, re-spectively, at a DOX concentration of 10 μM at 37 °C. After beingwashed twicewith PBS, cells were detachedwith the trypsine-EDTA so-lution and then dispersed in 0.5 mL of PBS. Drug cellular uptake wasthen analyzed on a FACSCalibur flow cytometer (BD Biosciences, USA).For confocal laser scanning microscopy (CLSM) studies, HeLa cells(4 × 105 cells) seeded onto 22 mm round glass coverslips were placedin a 6-well plate and cultured overnight. The cells were then incubatedwith free DOX and FA-conjugated and FA-free SPION/DOX-loadedNCPs,respectively, at a DOX concentration of 10 μM for 2 h. Cells were subse-quently washed with PBS, fixed with 4% formaldehyde, and stainedwith Hoechst 33258 for 10 min. Cellular uptake of DOX was visualizedat the excitation and emission wavelength of 488 and 590 nm, respec-tively, with a ZESS LSM 510 META.
2.8. Cytotoxicity analysis
HeLa cells, cultured in a 96-well plate at a density of 8 × 103 cells/well in DMEM containing 10% FBS and 1% penicillin, were incubated in100 μL of fresh medium containing free DOX or FA-conjugated orFA-free SPION/DOX-loaded NCPs, or DOX-free SPION-loaded NCP for24 h. AlamarBlue inmedium (10% v/v, 100 μL)was added to cell suspen-sions. After 2-h incubation, the absorbance of the suspension at 570 and600 nm was analyzed using a SpectraMax M5 microplate reader [29].
The effects of HFMF-induced hyperthermia and drug release ofFA-conjugated SPION/DOX-loaded NCPs on cell viability were studies.HeLa cells were seeded in a 24-well culture plate at a density of2 × 104 cells/well and incubated at 37 °C for 24 h. With the milieubeing replaced with fresh medium containing free DOX or NCPs, thecells were then incubated at 37 °C for 4 h. The cells harvested withthe trypsin-EDTA solution were placed into glass tubes. The glasstubes were positioned in the center of the HFMF described abovefor 20 min. The HFMF-treated cells were re-seeded in DMEM(500 μL) in a 24-well plate and incubated at 37 °C for 20 h. The invitro cytotoxicity was evaluated by AlamarBlue assay as describedabove.
2.9. Relaxivity measurement
T2 relaxation times of Resovist and FA-conjugated SPION/DOX-loaded NCPs were measured, respectively, on a Bruker Minispecmq20 at 20 MHz (0.47 T, which is comparable to the magnetic fieldstrength of clinically used MR scanners) and 37 °C. The T2-weightedMRI of NCPs and HeLa cells (2 × 105/well) exposed to NCPs at aniron concentration of 0.36 mM was acquired on a Bruker S300Biospec/Medspec MRI at 7 T using the following parameters: TR:3000 ms; TE: 15–150 ms; FOV: 60 × 60 mm2; matrix: 192 × 192;and slice thickness: 1 mm. The relaxivity values, r2, were evaluatedvia the linear least-squares fitting of 1/T2 relaxation time (s−1) versusthe iron concentration (mM Fe).
3. Results and discussion
3.1. Preparation of SPION/DOX-loaded polymersomes
Results of thermal gravimetric analysis indicate that the citric acidcontent of the citric acid-coated SPIONs is ca. 25 wt.%. The particlesizes of citric acid-coated SPIONs in aqueous solution at pH 5.0 deter-mined by DLS and in dry state observed by TEM are ca. 43 nm
(Fig. 1a) and ca. 8–10 nm (Fig. 2a), respectively. The larger particlesize obtained by DLS is caused primarily by the aggregation of thecitric acid-coated SPIONs in aqueous phase of pH 5.0. When subjectedto the ultrasonication, adopted from our previous study [27],poly(AAc-co-DSA) in aqueous solution of pH 5.0 underwent exten-sive hydrophobic intermolecular packing into polymersomes with aDh of ca. 120 nm and a narrow particle size distribution (Fig. 1a).The encapsulation of citric acid-coated SPIONs into polymersomeswas achieved by the self-assembly of poly(AAc-co-DSA) in aqueoussolution (pH 5.0) containing citric acid-coated SPIONs. The DLS data
Fig. 2. TEM images of (a) citric acid-coated SPIONs, (b and c) pristine polymersomeswith different amplifications, (d) SPION-loaded polymersomes, (e and f)FA-conjugated SPION/DOX-loaded NCPs.
HN C*
C O
OH
O
NH
CC O
O
NO O
O HN C
O
*
C O
NH
NH
O
CO
NOO
CH2
CH2
x y-z z
S
114
FA
a
b
c
d
Fig. 3. 1H NMR spectra of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA (5.0 mg/mL) in DMSO-d6at 20 °C.
284 W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
in Fig. 1a show that the SPION-loaded polymersomes have a Dh of ca.110 nm and mono-modal size distribution with a polydispersityindex (PDI) of 0.21, similar to the pristine polymersomes. As observedin Fig. 2, the morphology and particle size of SPION-encapsulatedpolymersomes are comparable to those of the SPION-free polymersomesand these citric acid-coated SPIONs are largely localized within theinner aqueous lumen of polymersomes. The results suggest that thepolymersome architecture was essentially not altered upon SPION en-capsulation. Due to the pronounced drug concentration gradient devel-oped across the lipid-rich membranes of polymersomes at pH 7.0wheremost AAc residues are ionized, the positively chargedDOX speciesdissolved in the external aqueous phase tended to permeate through thepolymersome membranes and then deposited therein by forminghydrophobic AAc/DOX ionic complexes via electrostatic attraction withthe ionized AAc residues alongwith stacking interactions of DOX species[27,30,31].
3.2. Synthesis and characterization of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA
The average molecular weight and polydispersity of fracturedγ-PGA, determined by SEC, were 15,600 g/mol and 1.46, respectively.From 1H NMR analysis (data not shown), ca. 45 mol% of the γ-GA resi-dues of the γ-PGA segments was activated into reactive OSu moieties.By grafting poly(γ-GA-co-γ-GAOSu) with maleimide-PEG-NH2 andconjugating with FA-SH, poly(γ-GA-co-γ-GAOSu)-g-PEG-FA wasacquired. The PEG content of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA wasestimated to be ca. 4.5 mol%, based on the relative integrals of featuresignals of ethylene protons from PEG (δ 3.5 ppm) and glutamic α pro-tons (δ 4.1 ppm) shown in Fig. 3. The presence of characteristic proton
signals (δ 6.6 and 8.6 ppm) and UV absorbance (285 nm) from the FAmoieties of poly(γ-GA-co-γ-GAOSu)-g-PEG-FA in 1H NMR (Fig. 3) andUV spectra (Fig. S2), respectively, confirms the successful conjugationof FA with the maleimide moiety of poly(γ-GA-co-γ-GAOSu)-g-PEG-maleimide.
3.3. Development and characterization of FA-conjugatedSPION/DOX-loaded NCPs
To further equip the SPION/DOX-loaded polymersomes withenhanced structural stability and active tumor-targeting ability, chi-tosan and poly(γ-GA-co-γ-GAOSu)-g-PEG-FA were deposited in se-quence onto the outer surfaces of polymersomes to formhydrophobic, dense polymeric complex layers via complementaryelectrostatic attractions (Scheme 1). Upon the facile aminolysis ofpartial primary amino groups from chitosan with reactive OSu resi-dues from poly(γ-GA-co-γ-GAOSu)-g-PEG-FA, the polymeric com-plex layers were further covalently cross-linked into the layeredpolyelectrolyte nanogels. An inversion of polymersome surfacecharges in terms of zeta potential (Fig. 1b) during layer-by-layer de-position process confirmed the formation of stable complex layersencircling the polymersomes. As shown in Fig. 1a, compared to theSPION/DOX-loaded polymersomes (ca. 137 nm) at pH 7.0, theFA-decorated SPION/DOX-loaded NCPs at pH 7.4 have a narrower sizedistribution (PDI 0.13) and a significantly larger size (ca. 221 nm) dueto the presence of layered polyelectrolyte gels. Nevertheless, the TEMimages (Fig. 2) show that the morphology of FA-conjugated SPION/DOX-loaded NCPs is similar to that of SPION-loaded polymersomes, sig-nifying the construction of outlayered nanogels on the surfaces ofpayloads-containing polymersomes did not disturb their original archi-tecture. Although a chitosan outlayer per se could endow polymersomeswith thedesired pH-responsive behavior, severe aggregation andprecip-itation of chitosan-coated polymersomes at pH 7.4 occurred due to thegreatly diminished protonation of chitosan segments. By further depos-iting and cross-linking a poly(γ-GA-co-γ-GAOSu)-g-PEG layer, the col-loidal stability of outlayered gel-caged polymersomes was thuseffectively promoted along with the additional steric repulsion propertyof the PEG chain segments. After being subjected to large-volume dilu-tion with PBS, the colloidal size of FA-conjugated SPION/DOX-loadedNCPs remained essentially unchanged whereas significant increases inparticle size and size distribution from the non-cross-linked FA-coated
Polymersomes without outlayered gelsat pH 4.7 at pH 7.4
Cum
ulat
ive
DO
X r
elea
se (
%)
Time (h)
Polymersomes with outlayered gels at pH 4.7 at pH 7.4
a
520 540 560 580 600 620 6400
50
100
150
200
250
300
350
400
450
0 min 1 h without HFMF treatment 1 h with HFMF treatment
DO
X f
luor
esce
nce
inte
nsit
y
Wavelength (nm)
b
Fig. 4. (a) CumulativeDOX release profiles of FA-conjugated SPION/DOX-loadedNCPs andSPION/DOX-loaded polymersomes in aqueous solutions of various pH at 37 °C. (b) DOXfluorescence spectra of FA-conjugated SPION/DOX-loaded NCP dispersion at pH 4.7with HFMF pre-treatment of 20 min or without any HFMF pre-treatment over a periodof 1 h at 37 °C.
285W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
SPION/DOX-loaded polymersomes were observed, presumably due tothe salt-evolved aggregation of polymeric complexes (Fig. S3). Further-more, upon receiving the dilution with FBS-containing (10%) aqueoussolution, the NCPs still maintained unvaried particle size and size distri-bution (Fig. S4). These results suggest that the layered polyelectrolytenanogels and outer hydrophilic PEG chain segments impart thepolymersomes superior colloidal stability and serum protein-repellentproperty. Furthermore, as shown in the in vivo images of Fig. S5, at24 h post-tail-vein-injection with either Cy5.5-loaded NCPs orpolymersomes, the tumor site of Tramp-C1 tumor-bearing micedisplayed considerable fluorescence intensity, illustrating the passive ac-cumulation of nanoparticles in tumors via the enhanced permeabilityand retention effects. Notably, the tumor of mice administrated withNCPs exhibited a significantly higher fluorescence intensity comparedto that of mice treated with uncoated polymersomes. Obviously, theoutlayered gels and outer hydrophilic PEG segments are capable of en-hancing the NCP in vivo stability in blood circulation and thus promotingtheir accumulation at tumor region.
The loading efficiency and content of SPIONs within polymersomeswere estimated to be ca. 40.5% and 10.7 wt.%, respectively. Althoughthe presence of copolymer and drug payload in SPION/DOX-containingNCPs inevitably diluted the concentration of SPIONs, leading to alower saturation magnetization (25 emu g−1) than that of the citricacid-covered SPIONs alone (75 emu g−1), the former still displays asuperparamagnetic behavior with negligible hysteresis (Fig. S6).Compared to other polymersome-based DOX delivery systems reportedelsewhere [16,32], the SPION/DOX-loaded NCPs developed herein pos-sess appreciably higher drug loading efficiency (ca. 73.3%) and content(9.3 wt.%) via the formation of stable AAc/DOX ionic complexes andthe presence of outlayered nanogels which reduce drug leakage duringpurification.
In view of the outlayered gels consisting primarily of pH-responsivechitosan and γ-PGA chain segments in spatial order, the SPION/DOX-loaded NCPs are expected to exhibit the pH-tunable behavior. Asshown in Fig. 1c, with the solution pH being adjusted from 7.4 to 4.7,the particle size of SPION/DOX-loaded NCPs was slightly increasedfrom 221 to 243 nmwith a reduction in light scattering intensity, indi-cating swelling of the NCPs in weak acidic environments. The swellingwas ascribed to the disruption of the chitosan/γ-PGA ionic complexesby the increased unionized γ-GA residues and the ionic osmotic pres-sure gradient developed within the interfacial polysaccharide layers asa result of the increased protonation of the amino groups. ThepH-induced structural variation of SPION/DOX-loadedNCPswas furtherexplored for drug release.
3.4. Stimuli-triggered drug release
For insight into the effects of outlayered gels of NCPs on drug re-lease, the in vitro DOX release performance of SPION/DOX-loadedpolymersomes before and after being coated with nanogel layers atdifferent pH was assessed and showed in Fig. 4a. Notably, at pH 7.4more than 42% of loaded DOX was released from the polymersomesover a period of 10 h, whereas only 20% of DOX was liberated fromNCPs under the same conditions. This indicates that the denseoutlayered gels with rather hydrophobic chitosan/γ-PGA complexescovering the surfaces of cargo-loaded polymersomes can effectivelyprevent the therapeutic payload from premature leakage. By contrast,both the NCPs and the polymersomes without gel coating exhibited aprompt DOX release (>70% within 10 h) at pH 4.7. The rapid releaseof DOX at acidic pH is primarily attributed to the massive disruptionof AAc/DOX ionic complexes, due to the considerable reduction ofthe AAc ionization, and the increase of the DOX solubility at pH 4.7.Note that, although the presence of the nanogel layers reduced thepremature leakage of DOX at 7.4, it did not hinder the release ofDOX at pH 4.7, due to the swelling of the protonated chitosan seg-ments within polysaccharide layers (Fig. 1c) by electrostatic
repulsion [33–36]. Taken together, the SPION/DOX-loaded NCPs ex-hibit a weak acid-triggered rapid drug release behavior with appre-ciably reduced premature leakage and thus the potential to serve asan intracellular-specific drug delivery vehicle to maximize anticancerefficacy and reduce premature drug leakage during blood circulationto minimize side effects toward other healthy tissues.
The response of the SPION/DOX-loaded NCPs in drug releaseto HFMF-induced hyperthermia at pH 4.7 was also studied. Approxi-mately 100% increase in fluorescence intensity from enhanced DOX re-lease was observed for the SPION/DOX-loaded NCPs after the HFMFtreatment (Fig. 4b). Since the bulk water temperaturewasmarkedly in-creasedwith the duration of HFMF exposure (Fig. S7), the enhancementin drug release can be attributed to the increase in polymersome per-meability, resulting from the elevation of local temperature beyondthe transition temperature of hydrophobic lipid bilayerwithin vesicularmembranes due to the magnetic hyperthermia from the embeddedSPIONs under alternative magnetic fields. Similar observations inhyperthermia-induced drug release from SPION-loaded liposomes andpolymersomes have been reported in literature [37–39]. Coupling thepH- and heat-responsive behaviors of the NCPs by performing HFMFtreatment at pH 4.7, enhanced release of DOX can be achieved (Fig. 4b).
3.5. Cellular uptake and intracellular distribution studies
The effect of the conjugated FA moieties on the cellular uptake of theNCPs was characterized, using folate receptor-overexpressing HeLa cellsas model tumor cells. As shown in Fig. 5a, HeLa cells incubated with
a
b
15 min 4 h
Fig. 5. (a) Flow cytometric histogram profiles of HeLa cells incubated with free DOX, FA-conjugated and FA-free SPION/DOX-loaded NCPs at 37 °C (DOX concentration = 10 μM).(b) CLSM images of HeLa cells incubated with free DOX, FA-conjugated and FA-free SPION/DOX-loaded NCPs at 37 °C for 2 h. Cell nuclei were stained with Hoechst 33258.
286 W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
FA-conjugated SPION/DOX-loaded NCPs exhibited a higher DOX fluores-cence intensity than those exposed to FA-free counterparts. Since thecellular uptake of FA-conjugated NCPs was inhibited by free folic acidin a dose-dependent manner (Fig. S8), these results confirm that theefficient cellular uptake of the FA-conjugated SPION/DOX-loaded NCPsis predominantly achieved by the FA receptor-mediated endocytosis.However, when the incubation time was prolonged from 15 min to4 h, the cellular uptake of freeDOXwas significantly increased comparedto that of SPION/DOX-loaded NCPs either with or without FA ligand dec-oration (Fig. 5a). This is primarily attributed to the different cellular up-take pathways for free DOX, passive diffusion, and for NCPs, endocytosis.
The CLSM images in Fig. 5b also show that the DOX fluorescenceintensity of HeLa cells treated with FA-conjugated SPION/DOX-loadedNCPs (for a time period of 2 h) was much higher than that culturedwith FA-free counterparts. Owing to the significant difference in theendocytosis rate between FA receptor-mediated and non-specificpathways, the DOX molecules from the FA-conjugated SPION/DOX-loaded NCPs were found not only in the cytoplasm but also innuclei of HeLa cells in contrast to only in the cytoplasm where thedrug species delivered by FA-free counterparts were mainly localized.
3.6. Cytotoxicity study
The in vitro therapeutic potency of FA-conjugated SPION/DOX-loaded NCPs against HeLa cells was evaluated using AlamarBlueassay. As observed in Fig. 6a, the survival rates of HeLa cells incubated
with free DOX, FA-conjugated and FA-free SPION/DOX-loaded NCPswere reduced proportionally with the increase of the DOX dose, whilethat of HeLa cells incubated with drug-free magnetic NCPs remainedhigh, suggesting that NCPs have virtually no toxicity (Fig. S9). Thedrug dose (0.62 μM) required for 50% cellular growth inhibition (IC50)of FA-conjugated SPION/DOX-loaded NCPs was comparable to that(0.61 μM) of free DOX and appreciably lower than that (1.25 μM) ofFA-free counterpart. This demonstrates that the combination of the en-hanced cellular uptake via FA receptor-mediated endocytosis withpH-triggered intracellular drug release can significantly improve drugefficacy. Similar results have been reported elsewhere [11,40,41].
The anticancer effects of the developed SPION/DOX-loaded NCPs inmagneto-thermo-chemotherapy were further evaluated. The viabilityof HeLa cells treated with drug-free FA-conjugated magnetic NCPs at37 °C for 4 h, followed by the 20-min HFMF treatment and additional20-h incubation, was appreciably reduced (Fig. 6b), demonstratingthat the magneto-thermotherapy of magnetic NCPs is a promisingtreatment for temperature-sensitive cancer cells. By integratingchemo- and thermotherapy into a single platform, the tailored-madeFA-conjugated SPION/DOX-loaded NCPs exhibit superior capability ofinhibiting cell proliferation compared to free DOX (Fig. 6b). The acid/magnet-triggered rapid drug liberation in the intracellular environmentor close vicinity of cancer cells can further induce apoptosis of the sur-viving cells from the heating treatment [42]. Themultimodal anticancerefficacy of SPION/DOX-loaded NCPs in terms of the tumor growth inhi-bition was preliminarily evaluated in Tramp-C1 tumor models (see the
Fig. 6. (a) Cell viability of HeLa cells incubated with free DOX, FA-conjugated and FA-freeSPION/DOX-loaded NCPs with various DOX doses at 37 °C for 24 h. (b) Cell viability ofHeLa cells incubated with free DOX, FA-conjugated SPION/DOX-loaded NCPs (DOX con-centration = 10 μΜ), and FA-conjugated drug-free magnetic NCPs with a concentrationof 80 μg/mL corresponding to that of the above payloads-loaded NCPs at 37 °C for 4 h,followed by the HFMF treatment of 20 min and additional incubation of 20 h.
0.2 0.4 0.6 0.8 1.00
40
80
120
160
200
240
280
r2 = 183.4 mM-1 s-1Resovist
r2 = 255 mM-1 s-1
FA-conjugated SPION/DOX-loaded NCPs
1 / T
2 (S
-1)
Fe concentration (mM)
a
b
Fig. 7. (a) Iron concentration-dependent T2 relaxation rate (1/T2, s−1) of the FA-conjugatedSPION/DOX-loaded NCPs and Resovist in aqueous solutions. (b) T2-weightedMR phantomimages of the FA-conjugated SPION/DOX-loaded NCPs at 37 °C.
Fig. 8. (a) T2-weighed MR images of HeLa cells incubated with FA-conjugated SPION/DOX-loaded NCPs at 37 °C for 4 h. (b) T2-weighed MR images of HeLa cells incubatedwith FA-conjugated and FA-free SPION/DOX-loaded NCPs (Fe concentration =0.36 mM) at 37 °C for 1 and 4 h.
287W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
Supplementary Data for detail description). As shown in Fig. S10, thetumor growth of mice receiving the single treatment of SPION/DOX-loaded NCPs combined with HFMF at day 0 was significantlysuppressed within 14 days as compared to either only PBS as controlor free DOX. Although rather preliminary, these NCPs show great prom-ise in advanced hybrid anticancer modality.
3.7. MR imaging
The application of the SPION/DOX-loaded NCPs as a MRI contrastagent for cancer diagnosis was evaluated. Through the linearleast-square fitting of 1/T2 (s−1) versus iron concentration, the transver-sal relaxivity (r2) reflecting the capability of the contrast agent to alter T2relaxation was accordingly obtained. As shown in Fig. 7a, the r2 value(255.2 Fe mM−1 S−1) of the FA-conjugated SPION/DOX-loaded NCPswas about 40% higher than that (183.4 Fe mM−1 S−1) of Resovist, acommercially available T2 contrast agent. The enhancement in MRI T2sensitivity can be ascribed to the synergistic effect of clustered SPIONsconfined within the inner aqueous compartment of polymersomes, asreported in the literature [11,16]. It is worthy to emphasize that theSPION/DOX-loadedNCPs in either aqueous solution or intracellular envi-ronment maintained sound MRI contrast as shown in Figs. 7b and 8a.Moreover, after being internalized by HeLa cells, the FA-conjugatedSPION/DOX-loaded NCPs displayed considerably enhanced contrast inthe T2-weighted MR images compared to the FA-free counterpart(Fig. 8b). Higher intracellular iron concentrations were also detected byICP-MS in HeLa cells incubated with the FA-conjugated SPION/
DOX-loaded NCPs (Table S1). In agreement with the above flowcytometry and CLSM data (Fig. 5), these results suggest that effectivetumor-targetedMRI can be achievedwith the FA-conjugated theranosticNCPs via FA receptor-mediated endocytosis.
4. Conclusion
In this work, multifunctional tumor-targeting NCPs carrying bothcitric acid-coated SPIONs and DOXwere developed for improved cancertheranosis. Through the self-assembly of lipid-containing copolymer,poly(AAc-co-DSA), in aqueous suspension of citric acid-coated SPIONsand the subsequent DOX encapsulation, the SPION/DOX-loaded
288 W.-H. Chiang et al. / Journal of Controlled Release 168 (2013) 280–288
NANOMEDICIN
E
polymersomes were attained. Upon sequential deposition of chitosanand poly(γ-PGA-co-γ-GAOSu)-g-PEG-FA by electrostatic interactionsand in situ covalent crosslinking, the NCPs exhibited enhanced colloidalstability in vitro and in vivo, pH-tunable drug release and activetumor-targeting ability. The FA-conjugated SPION/DOX-loaded NCPsshow superior capability of inhibiting cell proliferation by the combinedintracellular pH-triggered rapid drug liberation and magnetic hyper-thermia as compared to free DOX species. By virtue of the SPION cluster-ing effect in the inner aqueous chambers of polymersomes, theFA-conjugated SPION/DOX-loaded NCPs have an appreciably higher r2relaxivity value than Resovist. The high magnetic relaxivity of thetumor-targeting NCPs coupled with enhanced cellular uptake greatlypromotes the MRI contrast of target cancer cells. These results demon-strate the great potential of the FA-conjugated SPION/DOX-loadedNCPs as an advanced theranostic nanodevice to improve the anticancerefficacy.
Acknowledgments
This work is supported by the National Science Council(NSC99-2627-M007-009 and NSC99-2221-E007-006-MY3) and the Na-tional Tsing Hua University (101N2716E1 and 101N2053E1), Taiwan.We thank 7 T animal MRI Core Lab of the Neurobiology and CognitiveScience Center for technical and facility supports, and InstrumentationCenter for MRI experiments at National Taiwan University.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jconrel.2013.03.029.
[2] W. Wang, D. Cheng, F. Gong, X. Miao, X. Shuai, Design of multifunctional micellefor tumor-targeted intracellular drug release and fluorescence imaging, Adv.Mater. 24 (2012) 115–120.
[3] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.F. Chin, A.D. Sherry,D.A. Boothman, J. Gao, Multifunctional polymeric micelles as cancer-targeted,MRI-ultrasensitive drug delivery systems, Nano Lett. 6 (2006) 2427–2430.
[4] M. Liong, J. Lu, M. Kovochich, T. Xia, S.G. Ruehm, A.E. Nel, F. Tamanoi, J.I. Zink,Multifunctional inorganic nanoparticles for imaging, targeting, and drug delivery,ACS Nano 2 (2008) 889–896.
[5] Y. Xiao, H. Hong, A. Javadi, J.W. Engle, W. Xu, Y. Yang, Y. Zhang, T.E. Barnhart, W. Cai,S. Gong, Multifunctional unimolecular micelles for cancer-targeted drug deliveryand positron emission tomography imaging, Biomaterials 33 (2012) 3071–3082.
[6] E.K. Lim, Y.M. Huh, J. Yang, K. Lee, J.S. Suh, S. Haam, pH-Triggered drug-releasingmagnetic nanoparticles for cancer therapy guided by molecular imaging by MRI,Adv. Mater. 23 (2011) 2436–2442.
[7] M.K. Yu, Y.Y. Jeong, J. Park, S. Park, J.W. Kim, J.J. Min, K. Kim, S. Jon, Drug-loadedsuperparamagnetic iron oxide nanoparticles for combined cancer imaging andtherapy in vivo, Angew. Chem. Int. Ed. 47 (2008) 5362–5365.
[8] L. Zhang, H. Xue, Z. Cao, A. Keefe, J.Wang, S. Jiang,Multifunctional anddegradable zwit-terionic nanogels for targeted delivery, enhanced MR imaging, reduction-sensitivedrug release, and renal clearance, Biomaterials 32 (2011) 4604–4608.
[9] J. Hu, Y. Qian, X. Wang, T. Liu, S. Liu, Drug-loaded and superparamagnetic ironoxide nanoparticle surface-embedded amphiphilic block copolymer micelles forintegrated chemotherapeutic drug delivery and MR imaging, Langmuir 28(2012) 2073–2082.
[10] H. Cabral, N. Nishiyama, K. Kataoka, Supramolecular nanodevices: from designvalidation to theranostic nanomedicine, Acc. Chem. Res. 44 (2011) 999–1008.
[11] X. Yang, J.J. Grailer, I.J. Rowland, A. Javadi, S.A. Hurley, V.Z. Matson, D.A. Steeber, S.Gong, Multifunctional stable and pH-responsive polymer vesicles formed byheterofunctional triblock copolymer for targeted anticancer drug delivery andultrasensitive MR imaging, ACS Nano 4 (2010) 6805–6817.
[12] S.M. Lee, Y. Song, B.J. Hong, K.W. MacRenaris, D.J. Mastarone, T.V. O'Halloran, T.J.Meade, S.T. Nguyen, Modular polymer-caged nanobins as a theranostic platformwith enhanced magnetic resonance relaxivity and pH-responsive drug release,Angew. Chem. Int. Ed. 122 (2010) 10156–10160.
[13] E. Amstad, J. Kohlbrecher, E. Muller, T. Schweizer, M. Textor, E. Reimhult, Trig-gered release from liposomes through magnetic actuation of iron oxide nanopar-ticle containing membranes, Nano Lett. 11 (2011) 1664–1670.
[14] M.S. Martina, C. Wilhelm, S. Lesieur, The effect of magnetic targeting on the up-take of magnetic-fluid-loaded liposomes by human prostatic adenocarcinomacells, Biomaterials 29 (2008) 4137–4145.
[15] N. Kamaly, T. Kalber, M. Thanou, J.D. Bell, A.D. Miller, Folate receptor targeted bi-modal liposomes for tumor magnetic resonance imaging, Bioconjug. Chem. 20(2009) 648–655.
[16] Q. Sun, D. Cheng, X. Yu, Z. Zhang, J. Dai, H. Li, B. Liang, X. Shuai, A pH-sensitivepolymeric nanovesicle based on biodegradable poly(ethylene glycol)-b-poly(2-(diisopropylamino)ethyl aspartate) as a MRI-visible drug delivery sys-tem, J. Mater. Chem. 21 (2011) 15316–15326.
[17] C. Sanson, O. Diou, J. Thevenot, E. Ibarboure, A. Soum, A. Brulet, S. Miraux, E.Thiaudiere, S. Tan, A. Brisson, V. Dupuis, O. Sandre, S. Lecommandoux, Doxorubicinloaded magnetic polymersomes: theranostic nanocarriers for MR imaging andmagneto-chemotherapy, ACS Nano 5 (2011) 1122–1140.
[18] Z. Cheng, A. Tsourkas, Paramagnetic porous polymersomes, Langmuir 24 (2008)8169–8173.
[19] K. Shiraishi, K. Kawano, Y. Maitani, M. Yokoyama, Polyion complexmicelle MRI con-trast agents from poly(ethylene glycol)-b-poly(L-lysine) block copolymers havingGd-DOTA; preparations and their control of T1-relaxivities and blood circulationcharacteristics, J. Control. Release 148 (2010) 160–167.
[20] P. Chandrasekharan, D. Maity, C.X. Yong, K.H. Chuang, J. Ding, S.S. Feng,Vitamin E (D-alpha-tocopheryl-co-poly(ethylene glycol) 1000 succinate) micelles-superparamagnetic iron oxide nanoparticles for enhanced thermotherapy andMRI, Biomaterials 32 (2011) 5663–5672.
[21] X. Li, H. Li, G. Liu, Z. Deng, S. Wu, P. Li, Z. Xu, H. Xu, P.K. Chu, Magnetite-loadedfluorine-containing polymeric micelles for magnetic resonance imaging anddrug delivery, Biomaterials 33 (2012) 3013–3024.
[22] J. Dai, S. Lin, D. Cheng, S. Zou, X. Shuai, Interlayer-crosslinked micelles with par-tially hydrated core showing reduction and pH dual sensitivity for pinpointed in-tracellular drug release, Angew. Chem. Int. Ed. 50 (2011) 9404–9408.
[23] H. Xu, F. Meng, Z. Zhong, Reversibly crosslinked temperature-responsivenano-sized polymersomes: synthesis and triggered drug release, J. Mater. Chem.19 (2009) 4183–4190.
[24] S. Ning, Q. Huang, X. Sun, C. Li, Y. Zhang, J. Li, Y.N. Liu, Carboxymethyl dextran-coatedliposomes: toward a robust drug delivery platform, Soft Matter 7 (2011) 9394–9401.
[29] S. Trabulo, A.M. Cardoso, T. Santos-Ferreira, A.L. Cardoso, S. Simoes, M.C. Pedrosode Lima, Survivin silencing as a promising strategy to enhance the sensitivity ofcancer cells to chemotherapeutic agents, Mol. Pharm. 8 (2011) 1120–1131.
[30] H.J. Lee, Y. Bae, Cross-linked nanoassemblies from poly(ethylene glycol)-poly(aspartate) block copolymers as stable supramolecular templates for partic-ulate drug delivery, Biomacromolecules 12 (2011) 2686–2696.
[31] Y. Tian, L. Bromberg, S.N. Lin, T.A. Hatton, K.C. Tam, Complexation and release ofdoxorubicin from its complexes with pluronic P85-b-poly(acrylic acid) block co-polymers, J. Control. Release 121 (2007) 137–145.
[32] W. Chen, F. Meng, R. Cheng, Z. Zhong, pH-sensitive degradable polymersomes fortriggered release of anticancer drugs: a comparative study with micelles, J. Con-trol. Release 142 (2010) 40–46.
[33] C. Duan, J. Gao, D. Zhang, L. Jia, Y. Liu, D. Zheng, G. Liu, X. Tian, F. Wang, Q. Zhang,Galactose-decorated pH-responsive nanogels for hepatoma-targeted delivery oforidonin, Biomacromolecules 12 (2011) 4335–4343.
[34] Q. Tian, C.N. Zhang, X.H.Wang,W.Wang,W. Huang, R.T. Cha, C.H.Wang, Z. Yuan, M.Liu, H.Y. Wan, H. Tang, Glycyrrhetinic acid-modified chitosan/poly(ethylene glycol)nanoparticles for liver-targeted delivery, Biomaterials 31 (2010) 4748–4756.
[35] N.C. Bigall, A. Curcio, M.P. Leal, A. Falqui, D. Palumberi, R.D. Corato, E. Albanesi, R.Cingolani, T. Pellegrino, Magnetic nanocarriers with tunable pH dependence forcontrolled loading and release of cationic and anionic payloads, Adv. Mater. 23(2011) 5645–5650.
[36] J. Ge, E. Neofytou, T.J. Cahill, R.E. Beygui, R.N. Zare, Drug release fromelectric-field-responsive nanoparticles, ACS Nano 6 (2012) 227–233.
[37] P. Pradhan, J. Giri, F. Rieken, C. Koch, O. Mykhaylyk, M. Doblinger, R. Banerjee, D.Bahadur, C. Plank, Targeted temperature sensitive magnetic liposomes forthermo-chemotherapy, J. Control. Release 142 (2010) 108–121.
[38] E. Amstad, J. Kohlbrecher, E. Muller, T. Schweizer, M. Textor, E. Reimhult, Triggeredrelease from liposomes through magnetic actuation of iron oxide nanoparticlescontaining membranes, Nano Lett. 11 (2011) 1664–1670.
[39] Y. Chen, A. Bose, G.D. Bothun, Controlled release from bilayer-decoratedmagnetoliposomes via electromagnetic heating, ACS Nano 4 (2010) 3215–3221.
[40] M. Guo, C. Que, C. Wang, X. Liu, H. Yan, K. Liu, Multifunctional superparamagneticnanocarriers with folate-mediated and pH-responsive targeting properties foranticancer drug delivery, Biomaterials 32 (2011) 185–194.
[41] M. Prabaharan, J.J. Grailer, S. Pilla, D.A. Steeber, S. Gong, Folate-conjugated amphi-philic hyperbranched block copolymers based on boltorn H40, poly(L-lactide)and poly(ethylene glycol) for tumor-targeted drug delivery, Biomaterials 30(2009) 3009–3019.
[42] S. Brule, M. Levy, C. Wilhelm, D. Letourneur, F. Gazeau, C. Menager, C. Le Visage,Doxorubicin release triggered by alginate embedded magnetic nanoheaters: acombined therapy, Adv. Mater. 23 (2011) 787–790.
