lable at ScienceDirect

Biomaterials 34 (2013) 9602e9614

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Reversing multidrug resistance by intracellular delivery of Pluronic�

P85 unimers

Wei Hong a, Dawei Chen a,b, Xiaojun Zhang a, Jianfu Zeng a, Haiyang Hu a, Xiuli Zhao a,Mingxi Qiao a,*

a School of Pharmacy, Shenyang Pharmaceutical University, Wenhua Road 103, P.O. Box 42, Shenyang, Liaoning Province 110016, PR Chinab School of Pharmacy, Medical College of Soochow University, Suzhou, Jiangsu Province 215123, PR China

a r t i c l e i n f o

Article history:Received 19 June 2013Accepted 10 August 2013Available online 7 September 2013

Keywords:Pluronic� P85 unimersMultidrug resistanceIntracellular deliveryTriggered releasepH sensitive

* Corresponding author. Tel.: þ86 24 23986308; faE-mail address: qiaomingxi@163.com (M. Qiao).

0142-9612/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2013.08.032

a b s t r a c t

Pluronics have been demonstrated as excellent multidrug resistance (MDR) reversal agent in the form ofunimers rather than micelles. However, the effective intracellular delivery of Pluronic� unimers to MDRcancer cells still remains a big challenge. To address this issue, a mixed micellar system based mainly onthe pH-sensitive copolymer of poly (L-histidine)-poly (D,L-lactide)-polyethyleneglycol-poly (D,L-lactide)-poly (L-histidine) (PHis-PLA-PEG-PLA-PHis) and Pluronic� F127, some of which was conjugated withfolate, was constructed to intracellularly deliver the unimers of Pluronic� P85 to MDR cells. The folate-mediated endosomal pH-sensitive mixed micelles (pHendoSM-P85/f) were prepared by a thin-film hy-dration method, by which Pluronic� P85 unimers and doxorubicin (DOX) were incoporated into themixed micelles. The incorporation of Pluronic� P85 unimers was investigated by the surface tension test.The results indicated that the Pluronic� P85 unimers probably first inserted into the binary mixed mi-celles and then formed a triple-component mixed micelles with Pluronic� F127 and PHis-PLA-PEG-PLA-PHis as the loading content increased. Further analyzed with flow cytometry, confocal laser scanningmicroscopy (CLSM) and MTT assay, the micelles with inserted Pluronic� P85 unimers demonstratedmuch more cellular uptake and higher cytotoxicity against MDR cells than the triple-component mixedmicelles and plain Pluronic� micelles. The enhanced MDR reversal effect was attributed to the successfulintracellular delivery of Pluronic� P85 unimers to the MDR cells, which was confirmed by the subcellularcolocalization of Pluronic� P85 unimers with mitochondria, the decreased ATP energy and mitochondrialmembrane potential (MP) in the MCF-7/ADR cells. The pHendoSM-P85/f/DOX also demonstrated moredramatic antitumor efficiency and remarkable reduction of ATP energy in the MDR cells in tumors thanthe control formulations. The intracellular delivery of Pluronic� P85 unimers to the MDR cells based onthe targeted and endosomal pH triggerd release mixed micelles has been demonstrated as a promisingapproach to reverse MDR.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Nowadays, chemotherapy is still the primary option for cancertherapy. However, the efficacy of chemotherapy treatment has beenseverely compromised by the development of multidrug resistance(MDR) [1,2]. Even though much effort has been made to overcomethe MDR in cancer, limited successes have been achieved withsome therapeutic regimens and antineoplastic agents [3e8]. Themajor reason for that is the complex mechanisms through whichthe drug resistance exhibited [9,10]. Mechanisms of MDR have been

x: þ86 24 23986306.

All rights reserved.

identified to date including reduced drug accumulation due to asuperfamily of ATP binding cassette (ABC) proteins, such as the P-glycoprotein and the MDR-associated proteins (MRPs); drugdetoxification caused by the glutathione (GSH)/glutathione-S-transferase (GST) system; altered targets involving topoisomeraseII; alteration in drug-induced apoptosis owing to the change of Bcl-2 pathway; and the sequestration of drugs within cytoplasmicvesicles. Several independent mechanisms of MDR may act simul-taneously and/or in concert, which further increases the difficultyin MDR reversal.

Recently, amphiphilic block copolymers of Pluronics, such asPluronic� P85, have been identified to be the most promising MDRreversal agent due to their reversal effect on several distinct drugresistance mechanisms. It has been demonstrated that Pluronics

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9603

are able to 1) block drug efflux transporters, such as P-glycoprotein(P-gp) [11], multidrug resistance proteins (MRPs) [12,13], andbreast cancer resistance protein (BCRP) [14,15]; 2) incorporate intomembranes changing its microviscosity [16]; 3) induce a remark-able reduction of ATP levels in MDR cells [17]; 4) inhibit theglutathione (GSH)/glutathione (GST) detoxification system [18e22]; 5) induce release of cytochrome C and increase of reactiveoxygen species (ROS) levels in the cytoplasm [23]; 6) enhance pro-apoptotic signaling and decrease anti-apoptotic defense in MDRcells [24]; 7) abolish drug sequestration in acidic vesicles [25,26].Remarkably, these reversal effects are most apparent at thecopolymer concentrations below their critical micellization con-centration CMC [27], suggesting that the Pluronics unimers, i.e.single block copolymer molecules, are responsible for the MDRreversal [28,29]. This was attributed to the hydrophobic PPO chainsof the Pluronic� unimers immersed into the biomembrane hydro-phobic areas, resulting in the alternations of the membrane struc-ture and decreasing its microviscosity (“membrane fluidization”)[30,31]. Many attempts have been made to deliver Pluronicsreversal agents to MDR cells using micellar delivery systems due totheir ability to self-assemble into micelles [32e34]. For example,Wang and Zhang et al. developed floate-mediated Pluronic� P105micelles and Pluronic� F127/P123 mixed micelles to overcomeMDR, respectively. The authors claimed that the combination offolate-mediated active targeting and Pluronic� MDR reversal effectcould enhance the paclitaxel accumulation in both the resistantbreast cancer MCF-7/ADR cells and resistant ovarian cancer SKOV-3/PTX cells, leading to enhanced cytotoxicity against MDR cells.However, the major limitations of Pluronic copolymers were thelowmicellization and solubilization capacity to hydrophobic drugs,as well as the low stability of the self-assembled micelles upondilution in the blood-stream due to their relatively high CMC values[35]. Therefore, two similar mixed micelles consisting of mPEG-PLAcopolymers and Pluronic� L61 were later constructed because ofthe better micellar stability and higher solubilization to hydro-phobic drugs of the mPEG-PLA copolymers than Pluronics [36,37].Both results showed that the mixed micelles could significantlyenhance the cytotoxicity and cellular accumulation of the anti-cancer drugs, as compared to the plain mPEG-PLA micelles. How-ever, the major concern with the previously developed micellardelivery systems is that whether the Pluronics unimers can be trulytransported to the MDR cells. The plain Pluronics micelles wereexpected to lose the Pluronic MDR reversal agent in the blood cir-culation due to the poormicellar stability. While themixedmicellardelivery systems based on mPEG-PLA were designed to be stableenough to withstand the severe dilution in the blood circulation inorder to carry the Pluronics unimers to the tumor tissue via passiveor active targeting [17,38]. However, it is hardly expected that the

Fig. 1. Schematic diagram of the DOX-loaded folate-mediated endosomal

stabilized micelles were able to effectively release the Pluronicsunimers into the cytosol after being taken up into the endosomes orlysosomes of the MDR cells, which created another barrier on theway to the target due to the enhanced sequestration and degra-dation of MDR cells. The moderate MDR reversal effect of thepreviously developed micelles was probably attributed to theincreased drug accumulation in MDR cells resulting from the activetargeting and the micelles bypassing the efflux mediated by MDR-related proteins rather than the Pluronic unimers [39].

The objective of the present research was to evaluate the MDRreversal effect by intracellular delivery of Pluronic� P85 unimersto MDR cells. To achieve this, an endosomal pH-sensitive mixedmicellar delivery system with a targeting ligand of folate (pHen-

doSM-P85/f) based on the pH-sensitive copolymer of PHis-PLA-PEG-PLA-PHis and Pluronic� F127 were constructed (Fig. 1). Asmall proportional Pluronic� F127 was conjugated with folate toimpart the active targeting of the mixed micelles. The relativelylonger hydrophilic PEO block (4500 Da) of Pluronic� F127 ensureda prolonged circulation of the micelles and the exposure of thefolate ligand at the periphery of the pHendoSM-P85/f for folatereceptor-medicated endocytosis [40]. The pH-sensitive copolymerof PHis-PLA-PEG-PLA-PHis was responsible for dissociating themicellar structure in early or late endosomes and subsequentlydisrupting endosomal membrane due to the protonation of PHisblocks, releasing the Pluronic� P85 unimers into the cytosol forexerting the MDR reversal effect (Fig. 2). In this study, the intra-cellular delivery of Pluronic� P85 unimers and the MDR reversaleffect of pHendoSM-P85/f were evaluated at the cellular and ani-mal levels by performing in vitro cytotoxicity, intracellular accu-mulation, sub-cellular distribution, intracellular ATP level andmitochondrial membrane potential in MCF-7/ADR cells, in vivobiodistribution and antitumor effect in MCF-7/ADR tumor-xeno-grafted nude mice.

2. Materials and methods

2.1. Materials

Doxorubicin (DOX) was purchased from Beijing HuaFeng United Technology Co.Ltd. (Beijing, China). Pluronic� F127, Pluronic� P105 and Pluronic� P85 was kindlysupplied by BASF Ltd. (Shanghai, China). PEG (Mn: 2000 g/mole), N, N0- Carbon-yldiimidazole (CDI), 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl-tetrazolium bro-mide (MTT), fluorescein isothiocyanate (FITC) and Hoechst 33258 were purchasedfrom Sigma (St Louis, MO, USA). RPMI 1640without folic acid and fetal bovine serum(FBS) were purchased from Gibco BRL (Gaithersberg, MD, USA). ATP assay kit,mitochondrial membrane potential assay kit with JC-1, MitoTracker Red and Lyso-tracker Green were purchased from Beyotime� Biotechnology Co. Ltd (Nantong,China). Purified deionized water was prepared by the Milli-Q plus system (MilliporeCo., Billerica, MA, USA). All the other reagents and chemicals were of analytical orchromatographic grade and were purchased from Concord Technology (Tianjin,China).

pH-sensitive mixed micellar delivery system (pHendoSM-P85/f/DOX).

Fig. 2. Schematic diagram of the approach to reverse MDR by intracellular delivery of Pluronic� P85 unimers.

W. Hong et al. / Biomaterials 34 (2013) 9602e96149604

The human breast cancer cell line MCF-7 and the multidrug resistant humanbreast cancer cell line MCF-7/ADR were purchased from American Type Cell Culture(ATCC, USA). Culture plates and dishes were purchased from Corning Inc. (NY, USA).MCF-7 cells were cultured in RPMI 1640 medium without folic acid, supplementedwith 10% FBS, 100 IU/mL penicillin and 100 mg/mL streptomycin sulfate. MCF-7/ADRcells were cultured in the same medium with the addition of 1000 ng/mL ofDoxorubicin. All the cells were cultured at 37 �C in a humidifier with 5% CO2 at-mosphere. All the experiments were performed on the cells in the logarithmic phaseof growth.

Female BALB/c nude mice (20 � 2 g), supplied by Department of ExperimentalAnimals, Shenyang Pharmaceutical University (Shenyang, China), were acclimatedat 25 �C and 55% of humidity under natural light/dark conditions. All animal ex-periments were carried out in accordance with guidelines evaluated and approvedby the ethics committee of Shenyang Pharmaceutical University.

Tested formulations: DOX solution; pHendoSM-P85/f: blank folate-mediatedendosomal pH sensitive mixed micelles composed of PHis-PLA-PEG-PLA-PHis,Pluronic� F127 and folate-Pluronic� F127 loaded with Pluronic� P85; pHendoSM-P85/f/DOX: folate-mediated endosomal pH sensitive mixed micelles composed ofPHis-PLA-PEG-PLA-PHis, Pluronic� F127 and folate-Pluronic� F127 loaded withPluronic� P85 and DOX; pHendoSM/f: blank folate-mediated endosomal pH sensitivemixed micelles composed of PHis-PLA-PEG-PLA-PHis, Pluronic� F127 and folate-Pluronic� F127 without Pluronic� P85; pHendoSM/f/DOX: folate-mediated endo-somal pH sensitive mixed micelles composed of PHis-PLA-PEG-PLA-PHis, Pluronic�

F127 and folate-Pluronic� F127 loaded with DOX; pHIM-P85/f: blank folate-mediated Pluronic� P105 micelles loaded with Pluronic� P85; pHIM-P85/f/DOX:folate-mediated Pluronic� P105 micelles loaded with Pluronic� P85 and DOX.

2.2. Synthesis and characterization of PHis-PLA-PEG-PLA-PHis and folate-Pluronic�

F127

The copolymers used in this workwere home-made. The details of synthesis andcharacterization of PHis-PLA-PEG-PLA-PHis and folate-Pluronic� F127 were shownin the Supplementary data.

2.3. Preparation and characterization of DOX-loaded mixed micelles

The pHendoSM-P85/f/DOX was prepared via thin-film hydration method. First,doxorubicin (DOX) hydrochloride (20 mg) was stirred overnight with triethylamine(molar ratio, 1/3) in acetonitrile (20 mL) to obtain the DOX base. Then 20 mg of DOXbase and 180 mg of copolymer mixtures consisting of PHis-PLA-PEG-PLA-PHis,Pluronic� F127, folate-Pluronic� F127 and Pluronic� P85 were co-dissolved in20 mL of acetonitrile and sonicated for 30 min. The solvent was removed by rotaryevaporation at 40 �C to obtain a thin film. Residual acetonitrile remaining in the filmwas further evaporated overnight at room temperature under vacuum. The resultantthin film was hydrated with 10 mL of PBS (pH 7.4) for 30 min to obtain a micellarsolution. The micellar solution was filtrated through a 0.22 mm film to remove theunincorporated DOX aggregates. The pHendoSM/f/DOX was prepared as described

above without adding Pluronic� P85. For the preparation of pHIM-P85/f/DOX,identical operation was conducted except that the equivalent weight ratio of PHis-PLA-PEG-PLA-PHis was replaced by Pluronic� P105.

The pHendoSM-P85/f/DOX and pHIM-P85/f/DOX were characterized on a Zeta-sizer Naso ZS (Malvern, UK) and transmission electron microscopy (JEM-1230,Japan) operating at an acceleration voltage of 80 kV without staining. Drug loadingcoefficient (DL%) and entrapment efficiency (EE%) were calculated by the followingequations, respectively.

DL% ¼ Weight of the drug in micellesWeight of the feeding copolymers and drug

� 100%

EE% ¼ Weight of the drug in micellesWeight of the feeding drug

� 100%

2.4. Dynamic light scattering (DLS)

Dynamic light scattering (DLS) was used to measure the mean hydrodynamicdiameter and particle size distribution of themixedmicelles at various weight ratiosof PHis-PLA-PEG-PLA-PHis/Pluronic� F127 and their counterparts. All the mea-surements were carried out on a Zetasizer Naso ZS (Malvern, UK) at 25 �C afterequilibration for 5 min. The micellar solutions were filtered though a 0.45 mmdisposablemembrane filter prior tomeasurement. All the values were the average ofat least three independent samples.

2.5. Surface tension measurements

The surface tensionwas measured using Auto Surface Tensionmeter (A201, USA)based on Wilhelmy Plate method. The copolymer mixtures of PHis-PLA-PEG-PLA-PHis and Pluronic� F127 (80/20, w/w) at the concentration of 10 mg/mL were dis-solved in water under gentle stirring at 25 �C. The mixed micellar solution wasallowed to equilibrate for 24 h at room temperature. Then various amounts ofPluronic� P85 were added into the mixed micellar solution to obtain differentloading content from 1�10�5 wt% to 10 wt%. The surface tension of the solutionwasmeasured after equilibration for 24 h.

2.6. pH-dependent drug release from the micelles

Aliquots of DOX-loaded micelles (3 mL) containing 3 mg of DOX base weretransferred into the dialysis bags (molecular weight cut-off size 3500 Da) and dia-lyzed against 80 mL of phosphate buffers with different pH (pH 7.4, pH 6.8, pH 6.5,pH 6.0, pH 5.5, pH 5.0, pH 4.5 and pH 4.0) at 37 �C under sink conditions. Periodically,0.2 mL of aliquots were withdrawn and replaced with the equal volume of freshmedium. The DOX concentration was measured by multifunctional microplate

Fig. 3. Cumulative release of DOX from the mixed micelles composed of Pluronic�

F127/PHis-PLA-PEG-PLA-PHis (20/80, w/w) as a function of pH at 37 �C. The concen-tration of the copolymer mixture was 10 mg/mL (mean � SD, n ¼ 6).

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9605

reader (Tecan, Austria) at 490 nm and the cumulative DOX release was plotted as afunction of time.

2.7. In vitro cytotoxicity

The in vitro cytotoxicity of DOX solution and DOX-loaded micelles were assessedby a standard thiazolyl blue tetrazolium bromide (MTT) assay. Briefly, MCF-7/ADRcells were seeded in 96-well plates at the density of 5 � 103 cells per well andincubated overnight. The growth medium was then replaced with fresh mediumcontaining an indicated concentration of the tested formulations (pHendoSM-P85/f/DOX, pHendoSM/f/DOX, pHIM-P85/f/DOX and DOX solution), respectively. Controlwells were treated with equivalent volume of DOX-free medium. After incubationfor 48 h, thewells were rinsed with PBS, and then replacedwithMTT solution (5mg/mL). The plates were further incubated for 4 h at 37 �C, allowing the viable cells toreduce the yellow MTT into purple formazan crystals. At last, the medium wasremoved completely and 150 mL of dimethyl sulphoxide (DMSO) was added to eachwell to dissolve purple formazan crystals. The absorbance was measured with at570 nm using a multifuctional microplate reader (Tecan, Austria). The IC50 valueswere calculated using nonlinear regression analysis and the MDR reversal effect wasassessed by quantifying the IC50 values of the tested formulations.

2.8. Flow cytometry

MCF-7/ADR cells were seeded at the density of 1�105 cells/well in 6-well platesand incubated for 24 h to allow cell attachment. After 24 h, the medium wasreplaced with cell culture medium containing pHendoSM/f/DOX with differentweight ratios of Pluronic� P85. After 4 h incubation, the cells were washed threetimes with PBS. The cells were then harvested by trypsinization, centrifuged at1000 rpm for 5min, re-suspended in 500 mL of PBSmedium and analyzed using a BDFACS Caliber flow cytometer (FACSCAN, Becton Dickinson, San Jose, CA, USA).

2.9. Confocal laser scanning microscope (CLSM)

MCF-7/ADR cells were seeded on cover-slide system at the density of 2.5 � 104

cells/well in a humidifier with 5% CO2 atmosphere for 24 h at 37 �C and followed byincubation with the tested formulations (pHendoSM-P85/f/DOX, pHendoSM/f/DOX,pHIM-P85/f/DOX and DOX solution). After incubation for pre-determined time in-tervals, the cells were washed three times with cold PBS and stained with 100 nM

MitoTracker-Red (10 min) and 10 mM Hoechst 33258 (10 min) to visualize mito-chondria and nuclei, respectively. Then the cells were fixed with 4% para-formaldehyde for 30 min. The microscope images were captured using a confocallaser scanning microscope (CLSM, Olympus FV1000-IX81, Japan).

2.10. Mitochondrial membrane potential assay

The change in mitochondrial membrane potential was assessed using thelipophilic cationic membrane potential-sensitive dye JC-1 (5,50 ,6,60etetrachloro-1,10,3,30- tetraethyllenzimidazolycarbocyanine iodide) [41]. Briefly, the confluentMCF-7/ADR cells were treated with the different tested formulations (Pluronic� P85unimers, pHendoSM-P85/f, pHendoSM/f and pHIM-P85/f) for 2 h, and then washedthree times with cold PBS. The trypsinized cells (5 � 105) were suspended in 500 mLof diluted JC-1 staining solution for 20 min. After staining, the cells were rinsed withphysiological saline twice. Subsequently, the cells were suspended in 500 mL of JC-1staining buffer and immediately measured by a multifunctional microplate reader(Tecan, Austria) at lex (488 nm)/lem (590 nm) for red fluorescence or lex (488 nm)/lem (530 nm) for green fluorescence according to the manufacturer’s instructions.The obtained values were then expressed as average JC-1 red/green (R/G) signalintensity ratio (n ¼ 6).

2.11. ATP content assay

For intracellular ATP assay, the confluent MCF-7/ADR cells were treated withPluronic� P85 unimers and blank micelles (pHIM-P85/f, pHendoSM/f and pHendoSM-P85/f) for 2 h. Then the cells were washed twice with ice-cold PBS and solubilized incell lysates followed by immediate centrifugation (12,000�g) at 4 �C for 10 min. Thesupernatant was collected for ATP quantification using the kit based on luciferin/luciferase assay. Light emission was measured with Ultra-Weak LuminescenceAnalyzer (model BPCL, China). Raw data were converted to ATP concentrations ac-cording to the standard calibration curve. ATP contents were normalized by proteincontent in each sample determined by a BCA kit. The blank medium was used ascontrol.

2.12. DIR fluorescence real-time tumor imaging

MCF-7/ADR cells were induced to female BALB/c nude mice by subcutaneousinjection of 1 � 107 cells suspending in cell culture medium. When the tumorreached approximately 150e200 mm3, 0.2 mL of DIR-loaded pHIM-P85/f (pHIM-P85/f/DIR), pHendoSM/f (pHendoSM/f/DIR) and pHendoSM-P85/f (pHendoSM-P85/f/DIR) was intravenously injected through a tail vein, respectively. The time-

dependent biodistribution in MCF-7/ADR tumor-bearing nude mice was imagedby the Kodak In Vivo Imaging System FX PRO (Carestream Health, Inc., USA) at thetime points of 0.5 h, 1 h, 2 h, 4 h, 8 h, 12 h, 24 h, 36 h, 48 h post injection. The miceunder anesthetic state via inhalation of Gerolan Sol were automatically moved intothe imaging chamber for scanning. After 48 h, the mice were sacrificed and themajor organs and tumors were harvested. Each organ and tumor was rinsed withphysiological saline for three times followed by the capture of fluorescent images.

2.13. Anti-tumor efficiency

MCF-7/ADR cells were introduced to female BALB/c mice as described above.When the tumors grew to approximately 50e100 mm3 in volume, the administra-tion was started and this day was designed as day 0. At day 0, mice were randomlydivided into 5 groups. Theywere intravenously injectedwith 0.2mL of DOX solution,pHIM-P85/f/DOX, pHendoSM/f/DOX and pHendoSM-P85/f/DOX through the tail veinat the dose of 10 mg/kg every 3 days for 15 days, respectively. The physiologicalsaline was used as control. At the end of the experiment, all mice were sacrificed andthe tumors were harvested and weighed. For the luciferin/luciferase assay, eachtumor sample was supplemented with PBS and homogenized, followed by imme-diate centrifugation (12,000 g) at 4 �C for 10 min. The ATP content was determinedas described in Section 2.11 with physiological saline as control. Throughout theexperiment, the body weights of the mice and the volumes of the tumors weremeasured every other day. The antitumor activity was assessed with the tumorvolume (V), which was calculated by the following equation: V (mm3) ¼ (LW2)/2,where length (L) was the longest diameter and width (W) was the shortest diameterperpendicular to length.

2.14. Statistical analysis

All experiments were performed at least three times. Quantitative data arepresented as the mean � S.D. Statistical comparisons were determined by theanalysis of variance (ANOVA) among �3 groups or Student’s t-test between 2groups. P-values<0.05 and P-values<0.01 were considered statistically significant.

3. Results and discussion

3.1. Composition of the folate-mediated endosomal pH sensitivemicellar delivery system (pHendoSM-P85/f)

In order to intracellularly deliver Pluronic� P85 unimers to theMDR cells, the endosomal pH (pHw5.5) sensitive micellar deliverysystem conjugated with folate (pHendoSM-P85/f) was developedusing PHis-PLA-PEG-PLA-PHis and Pluronic� F127. The pH sensitivedelivery systems are required to have a proper triggering pH (thepH at which a remarkable accelerated drug release occur, pHt),depending on their intended use due to the different pH valuesbetween tumor extracellular interstitium (pHexw6.5) and tumor

Fig. 4. The change of pHt of the mixed micelles as a function of Pluronic� F127/PHis-PLA-PEG-PLA-PHis weight ratio (mean � SD, n ¼ 6).

Fig. 5. TEM images of the pHendoSM-P85/f/DOX (a) and pHIM-P85/f/DOX (b).

W. Hong et al. / Biomaterials 34 (2013) 9602e96149606

intracellular endosome (pHendow5.5). The pHt of the mixed mi-celles were investigated at different weight ratios of Pluronic�

F127/PHis-PLA-PEG-PLA-PHis by testing the in vitro pH-dependentDOX release. The typical pH-dependent drug release profile fromthe mixed micelles at the weight ratio of 20/80 was shown in Fig. 3.At pH 7.4, the mixed micelles exhibited a sustained drug releasepattern. When the pH decreased from 7.4 to 6.0, the mixed micellesshowed an unchanged drug release profile. However, an acceler-ated drug release occurred when the pH decreased to 5.5. Thiscould be attributed to the mixed micelles underwent a remarkablestructure transition due to the protonation of PHis blocks at pH 5.5.The corresponding pH was referred as pHt for this system. Similartransitions were observed for the other mixed micelles withPluronic� F127/PHis-PLA-PEG-PLA-PHis weight ratio ranging from5/95 to 35/65. The pHt was found to decrease as the Pluronic� F127weight ratio in the mixed micelles increased (Fig. 4). This wasattributed to the additional hydrophobic force arising from theinsertion of hydrophobic PPO block of Pluronic� F127 into the innercore of the micelles, which demanded more imidazole groups inthe PHis blocks to protonate at much lower pH to disrupt themicellar structure. When the Pluronic� F127 weight ratio reached40/60, the drug release behavior did not change from pH 7.4 to 4.0,suggesting the micelles probably lost pH sensitivity within thestudied pH range. The mixed micelles composed of Pluronic� F127/PHis-PLA-PEG-PLA-PHis at the weight ratio of 20/80 were found todisplay a desirable pHt (w5.5), which was specifically responsive tothe early endosomal pH.

To further enhance the delivery efficiency of the mixed micelles,a proportional of Pluronic� F127 was conjugated with folate toimpart the possible active targeting. Folate is an attractive targetingligand due to its high binding affinity for the folate receptors(kdw10�10 M) and it is able to be efficiently internalized into the

Table 1The physicochemical characterization of DOX-loaded polymeric mixed micelles (n ¼ 3).

Formulations Particle size (nm) x potential (mv)

pHendoSM-P85/f/DOX 117.0 �18.9 �6.09 � 0.18pHendoSM/f/DOX 102.3 � 12.1 �6.11 � 0.09pHIM-P85/f/DOX 108.7 � 13.6 �5.68 � 0.11

cancer cells through the receptor-mediated endocytosis [42]. In thepreliminary experiments, cellular uptake study was performed toinvestigate the optimum content of folate-conjugated Pluronic�

F127, which was set as 20% based on the weight of Pluronic� F127(data not shown).

3.2. The characterization of the mixed micelles

The physical characterizations of the mixed micelles weresummarized in Table 1. The mean diameters of the different mi-celles were all close to 100 nm, with good polydispersity indexesless than 0.1. As presented in Fig. 5, both TEM images of pHendoSM-P85/f/DOX and pHIM-P85/f/DOX showed spherical and homoge-neous morphology with the particle size correlated well with thoseobtained by DLS. The bright region of the micellar image wasprobably attributed to the outer shell of the mixed micelles formed

PDI DL% EE%

0.089 � 0.005 9.03 � 0.12 90.3 � 1.220.067 � 0.007 9.15 � 0.16 91.5 � 1.560.091 � 0.004 8.97 � 0.12 89.7 � 1.17

Fig. 7. The surface tension variation of the Pluronic� F127/PHis-PLA-PEG-PLA-PHismixed micelles with different loading content of Pluronic� P85. The weight ratio ofPluronic� F127/PHis-PLA-PEG-PLA-PHis was 20/80 (w/w) and the concentration ofcopolymer mixture was 10 mg/mL (mean � SD, n ¼ 6).

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9607

by the hydrophilic blocks of the copolymers, while the dark regioncorresponded to the inner core of the mixedmicelles formed by thehydrophobic blocks [43]. All the mixed micelles exhibited thesimilar slightly negative surface charge at pH 7.4, as characterizedby the low zeta potential. The nearly net surface charge of themicelles was able to prevent the recognition of opsonin, leading to aprolonged in vivo circulation and subsequent passive or activetargeting [44].

3.3. Co-micellization of PHis-PLA-PEG-PLA-PHis and PEO-PPO-PEO

When two different copolymers are applied to form the mixedmicelles, the co-micellization of them need to be characterized toexclude the possibility of forming the micelle mixture composed ofPluronic� F127 and PHis-PLA-PEG-PLA-PHis mono-copolymer mi-celles [45]. The average particle size of the binary mixed micellesmeasured by DLS was around 80 nmwhen the weight ratio of PHis-PLA-PEG-PLA-PHis/Pluronic� F127 was 80/20 (Fig. 6). The particlesize of the binary mixed micelles fell into the window of PHis-PLA-PEG-PLA-PHis mono-micelles (100 nm) and Pluronic F127 mono-micelles (40 nm). The median particle size and the relatively nar-row size distribution polydispersity (PDI < 0.1) of the mixed mi-celles indicates the co-micellization of the two copolymers [45]. Itcould be also noted that the binary mixed micelles could be formedat various weight ratio of PHis-PLA-PEG-PLA-PHis/Pluronic F127from 20/80 to 80/20, indicating the good miscibility of the twocopolymers.

3.4. The incorporation of Pluronic� P85 unimers into thepHendoSM/f

The incorporation of Pluronic� P85 unimers into the pHen-

doSM/f was confirmed by the surface tension measurement due toits surface activity. The effect of the addition of Pluronic� P85 intothe mixed micellar solution was shown in Fig. 7. The surfacetension of the mixed micellar solution containing the lowestloading content of Pluronic� P85 unimers (1 � 10�5 wt%) wasmeasured to be 38.70 mN/m, which was much lower than that ofthe mixed micellar solution (55.43 mN/m). The surface tensionkept decreasing with the loading content of Pluronic� P85increasing to “a”. The decrease in surface tension was due to thePluronic� P85 unimers distributing to the air/water interface. Thisindicates that Pluronic� P85 unimers are free in the mixed

Fig. 6. The hydrodynamic diameter (dh) and polydispersity index (PDI) of mixed mi-celles as a function of PHis-PLA-PEG-PLA-PHis/Pluronic� F127 weight ratio(mean � SD, n ¼ 6).

micellar solution within this loading content. It could be observedthat the surface tension started to increase with the loadingcontent of Pluronic� P85 increasing from “a” to “b”, within whichwas below the CMC value of Pluronic� P85 (6.5 � 10�5 mg/mL).This was probably because the Pluronic� P85 unimers insertedinto the binary mixed micelles due to the physical interactionswith Pluronic� F127 or PHis-PLA-PEG-PLA-PHis copolymer, lead-ing to a reduction of Pluronic� P85 unimers at the air/waterinterface, and subsequent an increase in surface tension. Thesurface tension kept increasing until “c”, due to the continuousinsertion of the Pluronic� P85 unimers into the binary mixedmicelles. After the loading content passed “c”, the surface tensionbegan to decrease, which was probably because of the replace-ment of the copolymers of Pluronic� F127 or PHis-PLA-PEG-PLA-PHis by the Pluronic� P85 unimers and the subsequent distrib-uting to the air/water interface. Further increasing the loadingcontent of Pluronic� P85 unimers from “d” to “e”, the surfacetension increased again. This could be due to the replacedcopolymer of Pluronic� F127 and PHis-PLA-PEG-PLA-PHis formedtriple-component mixed micelles with Pluronic� P85 unimers.When the loading content of Pluronic� P85 was above “e”, thesurface tension started to decrease again. This was because themixed micelles were saturated by the Pluronic� P85 unimers andthe excessive Pluronic� P85 unimers distributed to the air/waterinterface again, leading to a decrease in the surface tension.Further increasing the loading content of Pluronic� P85, the sur-face tension remained constant due to the formation of plainPluronic� P85 micelles in the mixed micellar solution. Similarphenomenon was also observed by the previous research onPluronic� F127 and another surfactant [46]. The above resultsindicate that the existence form of Pluronic� P85 unimers in thebinary mixed micelles varied with its loading content. When theloading content of Pluronic� P85 was between “a” (5 � 10�5 wt%)to “c” (2.5 � 10�3 wt%), Pluronic� P85 unimers probably insertedinto the binary mixed micelles. While the loading content ofPluronic� P85 was between “c” (2.5 � 10�3 wt%) and “e”(5 � 10�2 wt%), Pluronic� P85 unimers formed the triple-component mixed micelles with PHis-PLA-PEG-PLA-PHis,Pluronic� F127. Therefore, the loading content of Pluronic� P85in the binary mixed micelles between “a” to “e” was selected forthe following studies.

W. Hong et al. / Biomaterials 34 (2013) 9602e96149608

3.5. The effect of the loading content of Pluronic� P85 on the MDRreversal

The effect of loading content of Pluronic� P85 unimers inpHendoSM-P85/f on the cytotoxicity and cellular uptake was eval-uated using MTT method and flow cytometry, respectively (Fig. 8aand b). Prior to the test, the resistance characteristics of MCF-7/ADRwas confirmed by determination of the resistance index (ratio ofIC50 of DOX solution against MCF-7/ADR cells to that against MCF-7cells). The resistance index of MCF-7/ADR was 102, indicating thatMCF-7/ADR cells in the current study possessed considerably highdrug resistance to DOX compared to the parental cell lines. The IC50value decreased with the increasing of Pluronic� P85 loadingcontent until the loading content reached 1 � 10�3 wt%. When theloading content was higher than 1 � 10�3 wt%, the cytotoxicity ofDOX against MCF-7/ADR cells tended to dramatically decrease andthen leveled off. This was in accordance with previous results[17,27]. Similar results were obtained in the celluar uptake studyusing flow cytometry. The maximum celluar uptake occurred whenthe loading content of Pluronic� P85 was 1 � 10�3 wt%. Furtherincreasing the loading content of Pluronic� P85, the cellular uptakefirst decreased and then leveled off (Fig. 8b).

Fig. 8. In vitro cytotoxicity of the DOX-loaded mixed micelles with different loadingcontent of Pluronic� P85 against MCF-7/ADR cells (a) and flow cytometry analysis ofthe DOX-loaded mixed micelles with different loading content of Pluronic� P85uptaken by MCF-7/ADR cells (b) (mean � SD, n ¼ 6).

It could be noted that the loading content of Pluronic� P85 in themixed micelles had a remarkable effect on the cytotoxicity andcellular uptake of DOX. The optimum loading content of Pluronic�

P85 was 1�10�3 wt%, which fell into the range between “a” and “c”.As indicated by the surface tension test, the existence form ofPluronic� P85 unimers in the mixed micelles was different with thevariation of the loading content of Pluronic� P85. Within the rangebetween “a” and “c”, Pluronic� P85 unimers randomly inserted intothe mixed micelles. This could result in much easier release ofPluronic� P85 unimers into the cytosol when the mixed micellesbegan to dissociate in the early or late endosomes, leading to highercytotoxicity and cellular uptake of DOX. When the loading contentfurther increased to the range of “d” and “e”, the triple-componentmixed micelles were formed by Pluronic� P85, Pluronic� F127 andPHis-PLA-PEG-PLA-PHis. The more ordered micellar structure andstronger interaction with other copolymers probably impeded thetriggered release of Pluronic� P85 unimers from the mixed micellesin the acidic endosomes. These results also indicate that the previ-ously developedmixedmicellar delivery systems based on Pluronic�

MDR reversal agent and other type of polymers could not efficientlydeliver the Pluronic� unimers to the MDR cells. The highly orderedstable drug carriers could generate the physical barrier effect asso-ciated with sequestration in acidic organelles in MDR cells, resultingin poor intracellular delivery of MDR reversal agent [47].

Because of the minor amount of Pluronic� P85 barely affect thepH triggered drug release profiles, the mixed micelles pHendoSM-P85/f consisting of 20 wt% Pluronic� F127 (containing 20 wt% folateconjugated Pluronic� F127 based on the weight of Pluronic� F127),80 wt% PHis-PLA-PEG-PLA-PHis and 1 � 10�3 wt% Pluronic� P85were prepared for further studies.

3.6. In vitro cytotoxicity and cellular uptake of the DOX-loadedmicelles

To evaluate the in vitro cytotoxicity of the pHendoSM-P85/f/DOX,the micelles were tested against MCF-7/ADR cell lines andcompared to the control formulations (DOX solution, pHIM-P85/f/DOX and pHendoSM/f/DOX). The cytotoxicity of pHendoSM-P85/f/DOX and the control formulations against MCF-7/ADR was plottedin Fig. 9. The IC50 values were 4.83 � 0.41 mg/mL for pHendoSM-P85/f/DOX, 59.6 � 5.14 mg/mL for pHendoSM/f/DOX, 88.2 � 4.79 mg/mLfor pHIM-P85/f/DOX and 168.8 � 3.67 mg/mL for DOX solution,respectively. The pHendoSM-P85/f/DOX showed clear dose-dependent cytotoxicity and the lowest IC50, as compared to thecontrol formulations. It could be noted that the IC50 of pHendoSM-P85/f/DOX was more than 10 times lower than that of pHendoSM/f/DOX and 22 times lower than that of pHIM-P85/f/DOX. The effect ofthe different micelles (pHendoSM-P85/f/DOX, pHendoSM/f/DOX andpHIM-P85/f/DOX) on the uptake of DOX was evaluated using CLSM.As illustrated in Fig. 10, the pHendoSM-P85/f/DOX showed time-dependent DOX accumulation and broad distribution of DOX inthe MCF-7/ADR cells during the incubation. The DOX fluorescenceintensity increased from 0.5 h to 1 h, and reached the highest in-tensity at 2 h. The relatively strong fluorescence lasted for morethan 8 h. Compared to pHendoSM-P85/f/DOX, even though thepHendoSM/f/DOX also revealed a broad distribution of DOX in thecytosol as well as in nuclear compartments, but the fluorescenceintense became much weaker after 4 h of incubation and almostdisappeared after 8 h incubation. The pHIM-P85/f/DOX could not beinternalized as much as the other two micelles as observed by theminimal fluorescence intensity in the cells. Moreover, the pocketsof DOX intensity and a localized DOX intensity distribution alongwith a few dot-shapes of fluorescence inside the cells wereobserved, indicating the exocytosis of pHIM-P85/f out of the MDRcells by the endosome buffering.

Fig. 9. In vitro cytotoxicity of DOX solution, pHIM-P85/f/DOX, pHendoSM/f/DOX andpHendoSM-P85/f/DOX against MCF-7/ADR cells after 24 h and 48 h incubation(mean � SD, n ¼ 6). **P < 0.01: significantly different from the DOX solution,##P < 0.01: significantly different from the pHIM-P85/f/DOX, DDP < 0.01: signifi-cantly different from the pHendoSM/f/DOX.

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9609

The significant difference between pHendoSM-P85/f/DOX andpHendoSM/f/DOX in reversing MDR is obviously caused by thePluronic� P85 unimers incorporated in the mixed micelles. Thestrong and prolonged fluorescence of DOX in CLSM indicates thatthe Pluronic� P85 unimers can be released to the cytosol from thepHendoSM-P85/f/DOX and prevents the exclusion of DOX out of the

Fig. 10. The confocal microscope images of MCF-7/ADR cells incubated with DOX solution, p8 h at 37 �C.

MDR cells by suppressing the P-gp pumps. For pHendoSM/f/DOX,after DOX escaping from the endosomes, some of the released drugmight be sequestered in the cytosolic vesicles again or excluded outof the MDR cells by the P-gp pumps, leading to the decreasd fluo-rescence intensity as the incubation proceeded. It could be notedthat the pHIM-P85/f showed much less reversal effect compared tothe other two endosomal triggered release micelles (pHendoSM/f/DOX and pHendoSM-P85/f/DOX). This indicates that the effectiveintracellular delivery of Pluronic MDR agent to the MDR cellscannot be achieved with the plain Pluronic mixed micelles.Although pHIM-P85/f/DOX could be internalized into the tumorcells, the lack of pH-sensitivity and endosomal escape led to therecycling of the endosomes back to the cell surface or degradationin the lysosomes [48,49]. Due to the protonation of pH sensitivePHis block and its “proton sponge” effect [50], both the pHendoSM-P85/f/DOX and pHendoSM/f/DOX showed endosomal escape andrelease of their payload into the cytosol of the MDR cells. In addi-tion, all the micelles showed MDR reversal effect to some extent incomparison to the DOX solution, which might be due to thedifferent uptake mechanism between the micelles and drug solu-tion. This was in accordance with previous reports that active tar-geting showed limited MDR reversal effect [51,52].

3.7. Intracellular delivery of Pluronic� P85 unimers by thepHendoSM-P85/f

To further investigate the intracellular delivery of Pluronic� P85unimers, the subcellular distribution of Pluronic� P85 unimers andits effect on mitochondria were studied by the CLSM observation,mitochondrial membrane potential (MP) and ATP assay, respectively.

HIM-P85/f/DOX, pHendoSM/f/DOX and pHendoSM-P85/f/DOX for 0.5 h, 1 h, 2 h, 4 h and

Fig. 12. The effects of the blank mixed micelles (pHendoSM-P85/f, pHendoSM/f andpHIM-P85/f) and Pluronic� P85 unimers on mitochondrial membrane potential (a) andintracellular ATP level (b) of MCF-7/ADR cells (mean � SD, n ¼ 6). **P < 0.01: signif-icantly different from the blank medium, ##P < 0.01: significantly different from thepHIM-P85/f, þþ P < 0.01: significantly different from the pHendoSM/f, e P > 0.05: notsignificantly different from the Pluronic� P85 unimers.

W. Hong et al. / Biomaterials 34 (2013) 9602e96149610

The subcellular distribution of Pluronic� P85 unimers in MCF-7/ADRcells was investigated by triple labeling with the nucleus selectivedye (Hoechst 33258, blue), the fluorescent micelles (FITC-micelles,green), and the dye selective for mitochondria (MitoTracker, red).The fluorescence microphotographs of the MCF-7/ADR cells incu-bated with blank pHendoSM-P85/f and pHIM-P85/f for 2 h werepresented in Fig. 11. The colocalization of green and red fluorescencewas observed after incubation with pHendoSM-P85/f, indicating thatPluronic� P85 unimers were released from the micelles and reachedthe mitochondria. Compared to pHendoSM-P85/f, pHIM-P85/f dis-played much weaker green fluorescence, suggesting less Pluronic�

P85 unimers existed in the cytosol. The difference between the twomicelles was attributed to the pH-triggered release of Pluronic� P85unimers and endosomal escape property of pHendoSM-P85/f.

The effect of Pluronic� P85 unimers on the mitochondrialmembrane potential (MP) and ATP level of MCF-7/ADR cells wereassessed with different formulations (pHendoSM-P85/f, pHendoSM/fand pHIM-P85/f). The MP was determined using JC-1, which un-derwent reversible transformation from a monomer (greenflorescence) into an aggregate form (red florescence) when itbound with high MP. Mitochondrial depolarization (non-functionalmitochondria) was indicated by a decrease in the ratio of red/greenfluorescence intensity. As shown in Fig. 12a, pHendoSM-P85/fsignificantly decreased the average JC-1 red/green fluorescenceintensity ratio in the MCF-7/ADR cells (R/G ¼ 0.19 � 0.023,P < 0.05), as compared to pHIM-P85/f, which showed moderateeffect on the reduction of the fluorescence intensity ratio (R/G ¼ 0.38 � 0.012). The pHendoSM/f showed no effect on the MP (R/G ¼ 0.45 � 0.027) due to the absence of Pluronic� P85 unimers.

The intracellular ATP level was determined using luciferin/luciferase assay [53]. The cells treated with blank medium wereused as control and the ATP level was normalized as 100%. Asshown in Fig. 12b, when MCF-7/ADR cells were treated withpHendoSM-P85/f, the ATP level dropped to 40% of the normal level,which was not significantly different from the ATP level reduced bythe Pluronic� P85 unimers (38% of the normal level, P > 0.05).Compared to pHendoSM-P85/f and Pluronic� P85 unimers, pHIM-P85/f slightly dropped the ATP level to 86% of the normal level.The ATP level of the MDR cells treated with the pHendoSM/f wasfound to increase to about 140% due to the energy dependentendocytosis [54].

It could be noted that the pHendoSM-P85/f showed comparablereduction in both MP and ATP level with Pluronic� P85 unimers.This indicates that the endosomal pH triggered release and escapeproperties of the pHendoSM-P85/f could effectively deliver thePluronic� P85 unimers into the cytosol, dysfunctionating themitochondria of the MCF-7/ADR cells. Central to mitochondrialfunction is an electrochemical proton gradient across the

Fig. 11. The confocal microscope images of intracellular localization of FITC-Pluronic� P85 upHIM-P85/f for 2 h at 37 �C.

mitochondrial inner membrane, which establishes an electricalmembrane potential (MP). The decreased MP caused by Pluronic�

P85 unimers could stop the ATP synthesis and change the mito-chondrial membrane permeability, ultimately leading to cell

nimers in MCF-7/ADR cells. The cells were incubated with blank pHendoSM-P85/f and

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9611

apoptosis. The ATP depletion phenomenon is obviously extremelyimportant for MDR reversal since various drug resistance mecha-nism such as drug transport and detoxification systems, requiresconsumption of energy to sustain their function in MDR cells.

The pHIM-P85/f showed much less reduction in the MP and ATPlevel in MCF-7/ADR cells than the pHendoSM-P85/f. This suggeststhat effectively intracellular delivery of Pluronic� P85 unimerscould not be achieved with the plain Pluronic� mixed micelles. Asindicated in by the CLSM observation, the pHIM-P85/f wereencapsulated in the endosomes after endocytosis, compromisingthe efficient release of Pluronic� P85 unimers to the cytosol of theMDR cells. The moderate reduction of MP and ATP content by thepHIM-P85/f was probably due to some Pluronic unimers capable ofincorporating into endosomal membrane followed by subsequenttranslocation into the cytosol [23].

Fig. 13. The in vivo non-invasive images of time-dependent whole body imaging of MCF-7/ADIR-pHIM-P85/f/DIR (a). The ex vivo optical images of tumors and organs of MCF-7/ADR tumDIR and pHIM-P85/f/DIR (b) (n ¼ 3).

3.8. Tumor targeting and anti-tumor efficacy of different micelles

To monitor the time-dependent biodistribution of the differentmicelles (pHendoSM-P85/f/DIR, pHendoSM/f/DIR and pHIM-P85/f/DIR), the mice bearing MCF-7/ADR tumors were intravenouslyinjected with the micelles loaded with 0.1 wt% DIR (a near-infraredfluorescent dye) and observed using an optical image system. Asshown in Fig. 13, the pHendoSM-P85/f/DIR showed effectively tumortargeting and relative long residence at the tumor tissue. At 12 hpost injection, the maximum fluorescence was observed and thefluorescence could last for more than 48 h at the tumor tissue. ThepHendoSM/f/DIR showed almost the same time-dependent bio-distribution due to the similar composition of the micelles exceptthe absence of Pluronic� P85 unimers. The pHIM-P85/f/DIR showedworse tumor targeting property and shorter circulating time than

DR tumor-bearing mice after i.v. injection of pHendoSM-P85/f/DIR, pHendoSM/f/DIR andor-bearing mice sacrificed at 48h after i.v. injection of pHendoSM-P85/f/DIR, pHendoSM/f/

Fig. 14. The tumor volume changes of MCF-7/ADR tumor-bearing mice post i.v. injection of saline, DOX solution, pHIM-P85/f/DOX, pHendoSM/f/DOX and pHendoSM-P85/f/DOX (a).The weights of excised tumors fromMCF-7/ADR tumor-bearing mice at the time of sacrifice (b). The images of excised tumor tissue fromMCF-7/ADR tumor-bearing mice at the timeof sacrifice (c). The effects of thee mixed micelles (pHendoSM-P85/f/DOX, pHendoSM/f/DOX and pHIM-P85/f/DOX) and Pluronic� P85 unimers on the ATP level of isolated MCF-7/ADRtumors (d). Variation of body weights of MCF-7/ADR timor-bearing mice post i.v. injection of saline, DOX solution, pHIM-P85/f/DOX, pHendoSM/f/DOX and pHendoSM-P85/f/DOX (e).**P < 0.01: significantly different from the DOX solution, ##P < 0.01: significantly different from the Pluronic� P85 unimers, þþ P < 0.01: significantly different from the pHIM-P85/f/DOX, P < 0.01: significantly different from the pHendoSM/f/DOX, e P > 0.05: not significantly different from the physiological saline.

W. Hong et al. / Biomaterials 34 (2013) 9602e9614 9613

the pHendoSM-P85/f/DIR due to the poor dilution stability ofPluronic� micelles arising from the high CMC value [35]. The majororgans (heart, liver, spleen, lung, kidney and brain) and tumor tis-sues were isolated and the ex vivo images were further studied. ThepHendoSM-P85/f/DIR showed much intenser fluorescence in tumortissue than pHIM-P85/f/DIR. The results indicate that pHendoSM-P85/f/DIR possess much better dilution stability than pHIM-P85/f/DIR.

To evaluate the in vivo antitumor efficacy, the pHendoSM-P85/f/DOX and control formulations (pHIM-P85/f/DOX pHendoSM/f/DOX,DOX solution and saline) were intravenously injected to theMCF-7/ADR xenografted nude mice, respectively. Neither complete tumorgrowth regression nor toxicity-induced death was observed. Asshown in Fig. 14 that the pHendoSM-P85/f/DOX demonstrated thesignificant difference in inhibition of tumor growth as compared topHendoSM/f/DOX (P < 0.01), pHIM-P85/f/DOX (P < 0.005) and DOXsolution (P < 0.001) (Fig. 14a). The weight of tumors treated withthe pHendoSM-P85/f/DOX was approximately 1.5 times and 2.0times smaller than those treated with pHendoSM/f/DOX and pHIM-P85/f/DOX, respectively (Fig. 14b). The pHendoSM-P85/f/DOXresulted in significantly decreased the ATP levels in the MCF-7/ADRisolated tumors (P< 0.005) after the treatments as compared to thepHIM-P85/f/DOX, which correlated well with the in vitro results.These results indicate more efficient intracellular delivery ofPluronic� P85 unimers and better MDR reversal effect could beachieved by the pHendoSM-P85/f/DOXwith endosomal pH triggeredrelease and endosome escape property. The pHIM-P85/f/DOX evendisplayed significantly less antitumor efficiency than pHendoSM/f/DOX (289.4 � 13.5 mg vs 196.9 � 11.9 mg, P < 0.05). This could bedue to the worse tumor targeting ability, as indicated in the in vivobiodistribution experiment, which led to less accumulation of DOXand Pluronic� P85 unimers. In addition, the pHIM-P85/f/DOX takenup by the MDR cells might be sequestrated in acidic endosomes orlysosomes, further compromising the anti-tumor efficiency [47].The in vivo anti-tumor results were consistent with the MDRreversal effects observed in the in vitro cytotoxicity and cellularuptake tests. Fig. 14e showed the body weight change of thedifferent mice groups. Body weight loss is one of the indicators forevaluation of formulation toxicity. Mice treated with DOX solutionexhibited 19% decrease of weight with bad physical condition,indicating the severe systemic toxicity of DOX solution. Comparedto DOX solution, mice treated with all the micelles did not showobvious weight loss. Moreover, mice treated pHendoSM-P85/f/DOXshowed no significant weight loss in comparison to those of pHIM-P85/f/DOX (P > 0.05), indicating the comparable safety with thePluronic micelles.

4. Conclusion

In summary, an approach of intracellular delivery of thePluronic� P85 unimers to reverse MDR was evaluated by con-structing a folate mediated endosomal pH-triggered release mixedmicellar delivery system based on PHis-PLA-PEG-PLA-PHis andPluronic� F127. The incorporation of Pluronic� P85 unimers in themixedmicelles was found first inserted into themixedmicelles andthen formed triple-component mixed micelles with PHis-PLA-PEG-PLA-PHis and Pluronic� F127, as its loading content increased. ThepHendoSM-P85/f with inserted Pluronic� P85 unimers showedmoreefficient intracellular delivery of Pluronic� P85 unimers to thecytosol of the MDR cells than the triple-component mixed micellesand other control formulations. In addition, much more efficientintracellular delivery of Pluronic� P85 unimers was also confirmedby the higher antitumor efficiency against MCF-7/ADR tumors andlower cellular ATP levels in the isolated tumors. Therefore, theintracellular delivery of Pluronic� P85 unimers has been

demonstrated as an effective and a promising approach to reverseMDR. The study also indicates that how to achieve the unimerdelivery of Pluronic MDR agent with drug delivery systems need tobe considered when they were used for MDR reversal.

Acknowledgment

The authors are grateful to the financial support from the Na-tional Natural Science Foundation of China (81273448), NaturalScience Foundation of Liaoning Province (201102211) and Programfor Liaoning Excellent Talents in University (LJQ2011106).

Appendix A. Supplementary data

Supplementary data related to this article can be found online athttp://dx.doi.org/10.1016/j.biomaterials.2013.08.032.

References

[1] Naito S, Yokomizo A, Koga H. Mechanisms of drug resistance in chemotherapyfor urogenital carcinoma. Int J Urol 1999;6:427e39.

[2] Kroger N, Achterrath W, Hegewisch-Becker S, Mross K, Zander AR. Currentoptions in treatment of anthracycline-resistant breast cancer. Cancer TreatRev 1999;25:279e91.

[3] Fields A, Hochster H, Runowicz C, Speyer J, Goldberg G, Cohen C, et al. PSC833:initial clinical results in refractory ovarian cancer patients. Curr Opin Oncol1998;10(Suppl. 1):S21.

[4] Wigler PW. PSC833, cyclosporin A, and dexniguldipine effects on cellularcalcein retention and inhibition of the multidrug resistance pump in humanleukemic lymphoblasts. Biochem Biophys Res Commun 1999;257:410e3.

[5] Cruz F, Wolf A. Effects of the novel cyclosporine derivative PSC833 on glucosemetabolism in rat primary cultures of neuronal and glial cells. BiochemPharmacol 2001;62:129e39.

[6] Baekelandt M, Lehne G, Trope CG, Szanto I, Pfeiffer P, Gustavssson B, et al.Phase I/II trial of the multidrug-resistance modulator valspodar combinedwith cisplatin and doxorubicin in refractory ovarian cancer. J Clin Oncol2001;19:2983e93.

[7] Chico I, Kang MH, Bergan R, Abraham J, Bakke S, Meadows B, et al. Phase Istudy of infusional paclitaxel in combination with the P-glycoprotein antag-onist PSC 833. J Clin Oncol 2001;19:832e42.

[8] Verschraegen CF, Gupta E, Loyer E, Kavanagh JJ, Kudelka AP, Freedman RS,et al. A phase II clinical and pharmacological study of oral 9-nitrocamptothecin in patients with refractory epithelial ovarian, tubal orperitoneal cancer. Anticancer Drugs 1999;10:375e83.

[9] Gottesman MM, Fojo T, Bates SE. Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer 2002;2:48e58.

[10] Zgurskaya HI, Nikaido H. Multidrug resistance mechanisms: drug efflux acrosstwo membranes. Mol Microbiol 2000;37:219e25.

[11] Venne A, Li S, Mandeville R, Kabanov A, Alakhov V. Hypersensitizing effect ofpluronic L61 on cytotoxic activity, transport, and subcellular distribution ofdoxorubicin in multiple drug-resistant cells. Cancer Res 1996;56:3626e9.

[12] Evers R, Kool M, Smith AJ, van Deemter L, de Haas M, Borst P. Inhibitory effectof the reversal agents V-104, GF120918 and Pluronic L61 on MDR1 Pgp-,MRP1- and MRP2-mediated transport. Br J Cancer 2000;83:366e74.

[13] Miller DW, Batrakova EV, Kabanov AV. Inhibition of multidrug resistance-associated protein (MRP) functional activity with pluronic block co-polymers. Pharm Res 1999;16:396e401.

[14] Yamagata T, Kusuhara H, Morishita M, Takayama K, Benameur H, Sugiyama Y.Effect of excipients on breast cancer resistance protein substrate uptake ac-tivity. J Control Release 2007;124:1e5.

[15] Yamagata T, Kusuhara H, Morishita M, Takayama K, Benameur H, Sugiyama Y.Improvement of the oral drug absorption of topotecan through the inhibitionof intestinal xenobiotic efflux transporter, breast cancer resistance protein, byexcipients. Drug Metab Dispos 2007;35:1142e8.

[16] Dudeja PK, Anderson KM, Harris JS, Buckingham L, Coon JS. Reversal ofmultidrug resistance phenotype by surfactants: relationship to membranelipid fluidity. Arch Biochem Biophys 1995;319:309e15.

[17] Kabanov AV, Batrakova EV, Alakhov VY. Pluronic block copolymers for over-coming drug resistance in cancer. Adv Drug Deliv Rev 2002;54:759e79.

[18] Van Aubel RA, Koenderink JB, Peters JG, Van Os CH, Russel FG. Mechanismsand interaction of vinblastine and reduced glutathione transport in mem-brane vesicles by the rabbit multidrug resistance protein Mrp2 expressed ininsect cells. Mol Pharmacol 1999;56:714e9.

[19] Bagrij T, Klokouzas A, Hladky SB, Barrand MA. Influences of glutathione onanionic substrate efflux in tumour cells expressing the multidrug resistance-associated protein, MRP1. Biochem Pharmacol 2001;62:199e206.

[20] Harbottle A, Daly AK, Atherton K, Campbell FC. Role of glutathione S-trans-ferase P1, P-glycoprotein and multidrug resistance-associated protein 1 inacquired doxorubicin resistance. Int J Cancer 2001;92:777e83.

W. Hong et al. / Biomaterials 34 (2013) 9602e96149614

[21] Priebe W, Krawczyk M, Kuo MT, Yamane Y, Savaraj N, Ishikawa T. Doxoru-bicin- and daunorubicin-glutathione conjugates, but not unconjugated drugs,competitively inhibit leukotriene C4 transport mediated by MRP/GS-X pump.Biochem Biophys Res Commun 1998;247:859e63.

[22] Evers R, de Haas M, Sparidans R, Beijnen J, Wielinga PR, Lankelma J, et al.Vinblastine and sulfinpyrazone export by the multidrug resistance proteinMRP2 is associated with glutathione export. Br J Cancer 2000;83:375e83.

[23] Alakhova DY, Rapoport NY, Batrakova EV, Timoshin AA, Li S, Nicholls D, et al.Differential metabolic responses to pluronic in MDR and non-MDR cells: anovel pathway for chemosensitization of drug resistant cancers. J ControlRelease 2010;142:89e100.

[24] Minko T, Batrakova EV, Li S, Li Y, Pakunlu RI, Alakhov VY, et al. Pluronic blockcopolymers alter apoptotic signal transduction of doxorubicin in drug-resistant cancer cells. J Control Release 2005;105:269e78.

[25] Altan N, Chen Y, Schindler M, Simon SM. Defective acidification in humanbreast tumor cells and implications for chemotherapy. J Exp Med 1998;187:1583e98.

[26] van Veen HW, Margolles A, Putman M, Sakamoto K, Konings WN. Multidrugresistance in lactic acid bacteria: molecular mechanisms and clinical rele-vance. Antonie Van Leeuwenhoek 1999;76:347e52.

[27] Batrakova E, Lee S, Li S, Venne A, Alakhov V, Kabanov A. Fundamental re-lationships between the composition of pluronic block copolymers and theirhypersensitization effect in MDR cancer cells. Pharm Res 1999;16:1373e9.

[28] Batrakova EV, Han HY, Miller DW, Kabanov AV. Effects of pluronic P85unimers and micelles on drug permeability in polarized BBMEC and Caco-2cells. Pharm Res 1998;15:1525e32.

[29] Miller DW, Batrakova EV, Waltner TO, Alakhov V, Kabanov AV. Interactionsof pluronic block copolymers with brain microvessel endothelial cells: evi-dence of two potential pathways for drug absorption. Bioconjug Chem1997;8:649e57.

[30] Batrakova EV, Li S, Vinogradov SV, Alakhov VY, Miller DW, Kabanov AV.Mechanism of pluronic effect on P-glycoprotein efflux system in blood-brainbarrier: contributions of energy depletion and membrane fluidization.J Pharmacol Exp Ther 2001;299:483e93.

[31] Batrakova EV, Li S, Elmquist WF, Miller DW, Alakhov VY, Kabanov AV.Mechanism of sensitization of MDR cancer cells by Pluronic block copolymers:selective energy depletion. Br J Cancer 2001;85:1987e97.

[32] Zhang W, Shi Y, Chen Y, Ye J, Sha X, Fang X. Multifunctional Pluronic P123/F127 mixed polymeric micelles loaded with paclitaxel for the treatment ofmultidrug resistant tumors. Biomaterials 2011;32:2894e906.

[33] Wang Y, Yu L, Han L, Sha X, Fang X. Difunctional Pluronic copolymer micellesfor paclitaxel delivery: synergistic effect of folate-mediated targeting andPluronic-mediated overcoming multidrug resistance in tumor cell lines. Int JPharm 2007;337:63e73.

[34] Wang YLY, Zhang L, Fang X. Pharmacokinetics and biodistribution ofpaclitaxel-loaded pluronic P105 polymeric micelles. Arch Pharm Res 2008;31:530e8.

[35] Wang Y, Hao J, Li Y, Zhang Z, Sha X, Han L, et al. Poly(caprolactone)-modifiedPluronic P105 micelles for reversal of paclitaxcel-resistance in SKOV-3 tu-mors. Biomaterials 2012;33:4741e51.

[36] Mu CF, Balakrishnan P, Cui FD, Yin YM, Lee YB, Choi HG, et al. The effects ofmixed MPEG-PLA/Pluronic copolymer micelles on the bioavailability andmultidrug resistance of docetaxel. Biomaterials 2010;31:2371e9.

[37] Li X, Li P, Zhang Y, Zhou Y, Chen X, Huang Y, et al. Novel mixed polymericmicelles for enhancing delivery of anticancer drug and overcoming

multidrug resistance in tumor cell lines simultaneously. Pharm Res2010;27:1498e511.

[38] Burt HM, Zhang X, Toleikis P, Embree L, Hunter WL. Development of co-polymers of poly (D, L-lactide) and methoxy polyethylene glycol as micellarcarriers of paclitaxel. Colloids Surf B Biointerfaces 1999;16:161e71.

[39] Zhang W, Shi Y, Chen Y, Yu S, Hao J, Luo J, et al. Enhanced antitumor efficacyby paclitaxel-loaded pluronic P123/F127 mixed micelles against non-smallcell lung cancer based on passive tumor targeting and modulation of drugresistance. Eur J Pharm Biopharm 2010;75:341e53.

[40] Gaucher G, Dufresne MH, Sant VP, Kang N, Maysinger D, Leroux JC. Blockcopolymer micelles: preparation, characterization and application in drugdelivery. J Control Release 2005;109:169e88.

[41] Simon L, Szilagyi G, Bori Z, Telek G, Magyar K, Nagy Z. Low dose (-)deprenyl iscytoprotective: it maintains mitochondrial membrane potential and elimi-nates oxygen radicals. Life Sci 2005;78:225e31.

[42] Chen H, Ahn R, Van den Bossche J, Thompson DH, O’Halloran TV. Folate-mediated intracellular drug delivery increases the anticancer efficacy ofnanoparticulate formulation of arsenic trioxide. Mol Cancer Ther 2009;8:1955e63.

[43] Lo CL, Huang CK, Lin KM, Hsiue GH. Mixed micelles formed from graft anddiblock copolymers for application in intracellular drug delivery. Biomaterials2007;28:1225e35.

[44] Liu D, Liu F, Song YK. Recognition and clearance of liposomes containingphosphatidylserine are mediated by serum opsonin. Biochim Biophys Acta1995;1235:140e6.

[45] Yin H, Lee ES, Kim D, Lee KH, Oh KT, Bae YH. Physicochemical characteristicsof pH-sensitive poly(L-Histidine)-b-poly(ethylene glycol)/poly(L-Lactide)-b-poly(ethylene glycol) mixed micelles. J Control Release 2008 Mar 3;126(2):130e8.

[46] Zhang Y, Lam YM. Study of mixed micelles and interaction parameters forpolymeric nonionic and normal surfactants. J Nanosci Nanotechnol 2006;6:3877e81.

[47] Belhoussine R, Morjani H, Millot JM, Sharonov S, Manfait M. Confocal scanningmicrospectrofluorometry reveals specific anthracyline accumulation in cyto-plasmic organelles of multidrug-resistant cancer cells. J Histochem Cytochem1998;46:1369e76.

[48] Belting M, Sandgren S, Wittrup A. Nuclear delivery of macromolecules: bar-riers and carriers. Adv Drug Deliv Rev 2005;57:505e27.

[49] Couvreur P. Drug vectorization or how to modulate tissular and cellular dis-tribution of biologically active compounds. Ann Pharm Fr 2001;59:232e8.

[50] Oh KT, Lee ES, Kim D, Bae YH. L-histidine-based pH-sensitive anticancer drugcarrier micelle: reconstitution and brief evaluation of its systemic toxicity. IntJ Pharm 2008;358:177e83.

[51] Aouali N, El Btaouri H, Dumontet C, Eddabra L, Malagarie-Cazenave S,Madoulet C, et al. Accumulation of lactosylceramide and overexpression of aPSC833-resistant P-glycoprotein in multidrug-resistant human sarcoma cells.Oncol Rep 2011;25:1161e7.

[52] Nagao K, Maeda M, Manucat NB, Ueda K. Cyclosporine A and PSC833 inhibitABCA1 function via direct binding. Biochim Biophys Acta 2013;1831:398e406.

[53] Garewal HS, Ahmann FR, Schifman RB, Celniker A. ATP assay: ability todistinguish cytostatic from cytocidal anticancer drug effects. J Nati Cancer Inst1986;77:1039e45.

[54] Allen C, Yu Y, Eisenberg A, Maysinger D. Cellular internalization of PCL(20)-b-PEO(44) block copolymer micelles. Biochim Biophys Acta 1999;1421:32e8.