Construction of novel pH-sensitive hybrid micelles for enhanced extracellular stability and rapid intracellular drug release

Abstract

It is still a major challenge for successful cancer chemotherapy to design drug delivery systems with desirable extracellular stability and intracellular drug release. Here, novel pH-sensitive hybrid micelles were constructed by introducing positively charged arginine (Arg) into negatively charged micelles formed from a ternary graft copolymer (mPAL) comprising poly(acrylic acid) (PAA) as the backbone and methoxy poly(ethylene-glycol) (mPEG) and poly(lactide) (PLA) as the grafts to balance the extracellular stability and intracellular drug release. Paclitaxel (PTX) loaded hybrid micelles (PTX@mPAL/Arg) exhibited a desirable particle size (62.8 ± 2.9 nm), and a high entrapment efficiency (93.3 ± 2.9%). AFM and TEM images demonstrated the smooth surface and distinct spherical shape of the micelles. PTX@mPAL/Arg displayed high stability against dilution and serum, impeded drug release at physical conditions, and accelerated PTX release in mildly acidic medium. Blank micelles possessed satisfactory compatibility and would be safe for biomedical applications. PTX@mPAL/Arg micelles demonstrated much higher antitumor activity with low IC₅₀ values of 0.67 and 0.20 μg mL⁻¹ for A549 and HepG2 cells following 48 h incubation, respectively. A cellular internalization experiment indicated that the micelles could deliver and release cargo into the cytoplasm of HepG2 cells. Pharmacokinetic study in rats proved that PTX loaded micelles enhanced the AUC of PTX and prolonged circulation time in comparison to Taxol. These intelligent micelles with improved stability and drug release could be further studied as a promising delivery carrier for anticancer agents to improve therapeutic efficacy and to minimize adverse effects.

---

1. Introduction

Cancer has been a primary threat to human health and life expectancy around the world; tremendous amount of resource and effort has been invested to develop innovative and effective therapeutic strategies. Chemotherapy is the first-line therapy of choice for cancer patients; paclitaxel (PTX) has been one of the most important chemotherapeutic agents for cancers including breast cancer, lung cancer, ovarian cancer, and cervical cancer.^1,2 PTX possesses powerful antineoplastic effects via cellular microtubules hyperstabilization, capable of inhibiting cellular replication and inducing cellular apoptosis.^3,4 However, PTX is slightly soluble in water and the use of large amount of Cremophor EL in commercially viable PTX pharmaceuticals has reportedly been associated with numerous adverse effects such as allergy, hypersensitivity and neurotoxicity.^5,6 As such innovative preparation methods are needed both to improve PTX solubility and to minimize PTX toxicity.

Biodegradable polymeric micelles have become an emerging and promising platform to delivery anticancer agent to targets for cancer chemotherapy in recent years. The capability and versatility of polymeric micelles have been well demonstrated in literatures, ranging from increasing water solubility, to prolonging retention of drug preparations in circulation, from improving pharmacokinetic and pharmacodynamic profiles of the administered drug, in particular targeted accumulation of drug in tumor sites, to reducing overall side effects of the chemotherapy.^7–11 In addition, PEGylation of polymeric micelles would prolong the retention of preparation in vivo, improving the distribution in tumor tissues.^12,13 However, the polymeric nanocarriers are often plagued by the inadequate extracellular stability and insufficient intracellular drug release, which hampers the efficacy of cancer chemotherapy and causes increased side effects.^14,15

Stimuli-responsive nanocarriers are playing an increasingly crucial role in drug delivery system, which present superior extracellular stability and activated intracellular drug release. The environmental stimuli involve temperature, the presence of reactive oxygen species, difference in glutathione (GSH) concentrations, enzymatic expression, pH gradient variation, etc.^16–19 The pH values of blood and tumor tissues are 7.4 and 6.8 respectively, but pH value of intracellular endo/lysosome ranges from 4.0 to 6.0.^20,21 Polyacrylic acid (PAA), a biocompatible material, is frequently used as a pH-responsive carrier, which would accept protons at low pH and release protons at high pH.²² Some studied have demonstrated the pH-controlled release of cationic drugs, such as doxorubicin.^23,24 As a negative charged copolymer, PAA is not very suitable for the delivery of anion or non-ionic antitumor drugs by intravenous injection.

Beyond the simple assembly of copolymers into micelles, the shell or/and core cross-linked polymeric micelles is a research target of great significance, which could prevent the premature drug release and to increase intracellular drug release. In this study, novel pH-sensitive hybrid micelles based on ternary graft copolymer (mPAL) were designed and constructed using arginine (Arg) as stabilizer to realize physical cross-linking. Methoxy poly(ethylene-glycol) (mPEG) and poly(lactide) (PLA) were grafted onto the backbone of poly(acrylic acid) (PAA) to fabricate amphipathic copolymer to enhance the entrapment efficiency and the stability of micelles. As described in Scheme 1, the pH-sensitive hybrid PTX loaded mPAL/Arg micelles (PTX@mPAL/Arg) were fabricated by a film dispersion-ultrasonic method. Micelles with a desirable size range would be stable in circulation and accumulate at tumor tissues by EPR effect after intravenous injection. Once in tumor cells, the hybrid micelles would rapidly release the loaded drug under the stimuli of low pH, and then the drug would transport to the target, thus inhibiting the growth of tumor. The ultimate objective of the study was to develop a feasible, simple and functional nanocarrier system based on graft copolymer for anticancer compounds to achieve superior extracellular stability and activated intracellular drug release. Both the in vitro particulate characteristics and in vivo pharmacokinetic properties of PTX@mPAL/Arg micelles were investigated.

Scheme 1 (A) Schematic illustration of fabrication and self-assembly process of PTX@mPAL/Arg micelles. (B) Experimental rationale of drug accumulation and intracellular trafficking pathway; drug delivery includes steps of intravenous injection, EPR effect, (a) cellular internalization, (b) endo/lysosomal escape, (c) pH-triggered drug release and transport and (d) apoptosis induction.

2. Materials and methods

2.1. Materials

PTX was purchased from Xi'an Sanjiang Bio-Engineering Co., Ltd. (Xi'an, Shanxi, China). Methyl poly(ethylene glycol) (mPEG, MW, 2000) was purchased from Sigma-Aldrich (St. Louis, MO, USA). D,L-Lactide was purchased from Jinan Daigang Bioengineering Co. Ltd. (Jinan, Shandong, China), and recrystallized in ethyl acetate twice before use. Poly(acrylic acid) (PAA, MW, 1200) was purchased from Shanghai Reagent Factory (Shanghai, China). 1-(3-Dimethylaminopropyl)-3-ethyl-carbodiimide (EDC) and N-hydroxysuccinimide (NHS) were purchased from Shanghai Covalent Chemicals Co., Ltd. (Shanghai, China). Pyrene (>99%) was purchased from Fluka Company (St. Louis, MO, USA). Arginine was purchased from Aladdin Reagent Database Inc. (Shanghai, China). All reagents were of analytical grade and used as received.

2.2. Synthesis of ternary graft copolymer mPAL

The designed grafted polymer was composed of a hydrophilic main chain of PAA, a partial hydrophilic branched-chain of mPEG and a partial hydrophobic branched-chain of PLA. Therefore, the synthesis of the graft polymer was divided into two steps: mPEG was conjugated to activated carboxyl group of the main chain to form the hydrophilic section; PLA was conjugated to the residual carboxyl group of the main chain via ring-opening polymerization to form the final copolymer (Scheme 2).

Scheme 2 Synthesis process of the ternary graft copolymer mPAL.

2.2.1. Synthesis of PAA-g-mPEG. In the synthesis of PAA-g-mPEG, the carboxylate groups of PAA were first activated by NHS and EDC. Briefly, 1.2 g of PAA dissolved in 20 mL of DMSO reacted with 0.9 g of NHS and 1.5 g of EDC under nitrogen flow at room temperature for 12 h; the activated PAA then reacted with 16.7 g of mPEG dissolved in 10 mL of DMSO for another 8 h, followed by dilution in 45 mL acetone and centrifugation. The supernatant was dialyzed (SpectraPor 6, MW cutoff 8000 Da) against deionized water for 24 h and lyophilized.

2.2.2. Synthesis of ternary graft copolymer. mPAL was synthesized by ring-opening polymerization reaction. In brief, 1.2 g of D,L-lactide was added to a flame-dried and nitrogen-purged glass flask, followed by adding 2.8 g of PAA-g-mPEG and 0.3% (mol mol⁻¹) Sn(Oct)₂ toluene solution. The reaction vessel was immersed in a thermostatic oil bath at 125 °C, under agitation for 24 h. The reaction product was then precipitated in 200 mL of cool ether, filtered and dried at 40 °C under vacuum for 48 h. A series of graft copolymers with variable PLA blocks were synthesized by changing feed ratio of PAA-g-mPEG to D,L-lactide.

2.3. Characterization of ternary graft polymer mPAL

The structure of mPAL was characterized by Fourier transform infrared (FT-IR) spectrometry (Nicolet Impact 410 Spectrometer, Nicolet Analytical Instruments, Madison, WI, USA) and ¹H nuclear magnetic resonance (¹H NMR) spectrometry (Bruker Avance Spectrometer AV-500, Bruker, Karlsruhe, Germany).

Critical micelle concentration (CMC) of mPAL was measured by fluorescence spectroscopy, using pyrene as a hydrophobic fluorescence probe.^25,26 The copolymers were diluted in distilled water with concentrations ranged from 1 × 10⁻⁷ to 1 g L⁻¹. The final pyrene concentration was 6 × 10⁻⁷ M. Then, the samples were incubated at 65–70 °C for 2 h and kept overnight at room temperature. Fluorescence spectra were recorded using a Shimadzu RF-5301 PC fluorescence spectrophotometer (Kyoto, Japan). The emission wavelength was set at 392 nm, and the excitation spectrum of pyrene was recorded from 270 nm to 370 nm. Peak height ratio of the fluorescent intensities at 338 nm and 333 nm (I₃₃₈/I₃₃₃) against the logarithm of micelle concentrations was plotted, and CMC calculated. The polymers were further dispersed in arginine solution (1.5 mg mL⁻¹) to study the effects of arginine on CMC values of mPAL.

2.4. Preparation of PTX loaded micelles

PTX@mPLA/Arg micelles were prepared by a film dispersion-ultrasonic method with modifications.²⁷ Briefly, mPLA and PTX were first dissolved in 10 mL mixture of dichloromethane and methanol (7 : 1, v/v), evaporated in rotary evaporator at 45 °C for 30 minutes to form a thin film. 10 mL Arg solution (1.5 mg mL⁻¹) was then added to the vessel to hydrate the film to prepare PTX@mPLA/Arg hybrid micelles. Amphiphilic mPAL copolymer would self-assemble into micelles and encapsulate PTX into hydrophobic cores and cationic Arg would further stabilize the micelles through ionic interaction of Arg with the copolymers. The preparation was sonicated at 100 W with a probe-type ultrasonicator for 5 minutes in an ice bath, filtered with a microporous membrane (0.22 μm) to remove free PTX, and then lyophilized. PTX@mPLA micelles were prepared by the similar method except that Arg solution was replaced with the same volume of distilled water.

2.5. Characterization of PTX loaded micelles

2.5.1. Drug loading capability. After filtration by 0.22 μm microporous membrane, PTX micelle solution was diluted with 20 times of methanol and vortexed for 3 min for demulsification. Then, the solution was centrifuged at 8000 rpm for 10 minutes and the drug concentration in the supernatant was quantified by high performance liquid chromatography (HPLC, Shimadzu LC-10AD system, Kyoto, Japan) coupled with a Diamonsil C18 column (250 mm × 4.6 mm, 5 μm particle size, Dikma Technology Company, China). The mobile phase was composed of methanol and water (70 : 30, v/v), flow rate at 1.0 mL min⁻¹. PTX was separated at 30 °C and detected at 227 nm. The calibration curve was prepared in the range of 0.2–20 μg mL⁻¹ (r = 0.9998). Drug loading (DL) and entrapment efficiency (EE) were calculated as following:

2.5.2. Morphological observation. The morphology and size distribution of PTX@mPLA and PTX@mPLA/Arg micelles were visualized by atomic force microscopy (AFM, SPA 3800N, SEIKO, Japan) and transmission electron microscopy (TEM, Hitachi H-7650, Hitachi, Japan). For AFM observation, samples were dissolved in distilled water and dripped onto freshly cleaved mica plate, followed by air-drying at room temperature and mounted on microscope scanner.^28,29 Explorer atomic force microscope was a tapping mode, using high resonant frequency (F₀ = 129 kHz), pyramidal cantilevers with silicon probes having force constants of 20 N m⁻¹. Scan speed was set at 2 Hz. For TEM observation, samples were negatively stained with 2% phosphotungstic acid prior to the observation.

2.5.3. Particle size, size distribution and zeta potential. Particle size, size distribution and zeta potential of polymeric micelles were determined using a Malvern Zetasizer 3000 system (Malvern Instruments Ltd., Malvern, UK). Experimental temperature was kept at 25 °C during measurement with a scattering angle of 90°. Samples were diluted properly and equilibrated to the defined temperature for 30 minutes before measurement.

2.5.4. In vitro PTX release from micelles. PTX release from PTX@mPLA and PTX@mPLA/Arg micelles was studied in PBS buffer solution at three pH conditions (pH 5.0, pH 6.5, pH 7.4) using a dialysis method. Briefly, 1 mL of PTX loaded micelles (0.5 mg PTX) was placed in a dialysis bag (MWCO 3500 Da); the dialysis bag was then immersed in 150 mL of PBS buffer containing 0.1% (w/v) Tween 80, stirred at 37 °C and 100 rpm. At predetermined intervals, samples (1.0 mL) were collected for HPLC analysis as described in Section 2.5.1 and the medium was replenished. The particle sizes of the two micelles in pH 5.0, pH 6.5 and pH 7.4 were also measured by a Malvern Zetasizer.

2.5.5. Stability of PTX loaded micelles. To evaluate the storage stability of PTX@mPLA and PTX@mPLA/Arg solution, micelles were dispersed in distilled water with a corresponding PTX concentration of 0.5 mg mL⁻¹ and kept at 25 °C for 96 h. Changes in particle size and drug loading of the micelles were monitored and compared at 0, 24, 48, 72, and 96 h by dynamic light scattering (DLS) and HPLC, respectively.

The stability of micelles in fetal bovine serum (FBS) solution was evaluated by measuring the size changes. PTX@mPLA and PTX@mPLA/Arg micelles were dispersed in phosphate buffered saline (PBS) containing 10% FBS and kept at 37 °C shaking incubator with a rate of 100 rpm. To monitor the interactions of the particles with proteins in the presence of FBS, the particle size was measured by DLS at predetermined time intervals (0, 4, 8, 12, 24 h).

2.6. Biocompatibility of the copolymer

Hemolysis testing was used to assess hemocompatibility of mPAL and mPAL/Arg blank micelles in vitro.³⁰ 2.5 mL of 2% rabbit red blood suspension was mixed with samples of mPAL, mPAL/Arg, Tween 80 and Cremophor EL to achieve the systematically varied concentrations of 0.2–2 mg mL⁻¹, and incubated at 37 °C for 1 h. Distilled water and saline were used as positive and negative controls, respectively. A Shimadzu UV-2450 spectrophotometer was used to determine hemoglobin release at 416 nm after collecting the supernatant. Three replicates were utilized for each testing sample.

Cytotoxicity of mPAL and mPAL/Arg blank micelles was evaluated by an MTT assay for A549 and HepG2 cell lines (American Type Culture Collection, Manassas, VA, USA). Cells were cultured in Dulbecco's modified eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37 °C under a humid atmosphere of 5% CO₂. Cells were incubated with blank micelles and Taxol vehicle Cremophor EL/ethanol (50 : 50, v/v) at corresponding PTX concentrations ranging from 0.001 to 20 μg mL⁻¹ for 48 h. Cellular viability was calculated by measuring the absorbance of each well at 570 nm by a microplate reader (Powerwave X, BioTek Instruments, Inc., Winooski, VT, USA).

2.7. In vitro antitumor efficacy

A549 and HepG2 cell lines were also used to evaluate antitumor efficacy of PTX@mPLA and PTX@mPLA/Arg micelles. Antitumor efficacy of PTX loaded micelles was determined by an MTT assay via detection of the mitochondrial activity. Cells seeded in 96-well plates were incubated with Taxol and PTX loaded micelles at PTX concentrations ranged from 0.001 to 20 μg mL⁻¹ for 48 h. Afterwards, 20 μL of freshly prepared MTT (5 mg mL⁻¹) solution was added to each well and incubation took place for another 4 h. Then, 150 μL of DMSO was added to dissolve formazan crystals. The absorbance was measured at 570 nm using a microplate reader (Powerwave X, BioTek Instruments, Inc., Winooski, VT, USA). Negative control groups used cells without treatment, and all results were normalized to untreated group.

2.8. Cellular internalization experiment

The cellular uptake of mPLA and mPLA/Arg micelles was tracked using the confocal laser scanning microscopy (CLSM). Fluorescent labeled micelles were fabricated by changing PTX into hydrophobic coumarin (C6). HepG2 cells were seeded into glass bottom culture dishes and cultured for 24 h at 37 °C and 5% CO₂. Then the cells were treated with C6 labeled micelles at the concentration of 100 ng mL⁻¹. After incubation for 2 and 4 h, cells were washed with PBS for three times and fixed with 4% paraformaldehyde. The cell nuclei were stained with DAPI. The cellular uptake of mPLA and mPLA/Arg micelles was visualized by a CLSM (Carl Zeiss LSM 700, Germany).

2.9. In vivo pharmacokinetics

Male Sprague-Dawley (SD) rats (220 ± 20 g) were obtained from Center for New Drug Screening, China Pharmaceutical University; the study animals were housed in pathogen-free facility and provided with food and water ad libitum, acclimated for one week at 25 °C, 55% RH, regular day/night cycle before the experiment commenced. The animal study was carried out in accordance to the guidelines of the Health Guide for the Care and Use of Laboratory Animal, and the animal use protocol was approved by the ethics committee of China Pharmaceutical University.

Eighteen rats were randomly divided into three study groups. Taxol, PTX@mPLA and PTX@mPLA/Arg micelles were injected through the tail vain at a dose of 5 mg kg⁻¹ of PTX. 0.3 mL of blood was collected from the orbital plexus of rats at 5, 10, 15, 20, 30, 45 min, and 1.0, 2.0, 4.0, 6.0, 8.0, 10.0 and 12.0 h after single i.v. administration of each formulation. The samples were centrifuged at 3000 rpm for 10 minutes and plasma was separated for drug analysis.

A liquid–liquid extraction method was designed to extract the individual drugs and internal standard from plasma samples. Briefly, 100 μL plasma was mixed with 20 μL internal standard (diazepam, 4 μg mL⁻¹) and vortexed for 30 s. 2 mL diethyl ether was added to the sample, vortexed for 2 minutes and centrifuged at 3000 rpm for 10 minutes to extract PTX. The organic layer was then transferred to a clean tube and dried under nitrogen flow at 37 °C. The extraction residue was reconstituted in 100 μL methanol and centrifuged at 10 000 rpm for another 10 minutes. 20 μL supernatant was injected into the HPLC system for PTX quantification. The extraction recovery of PTX from plasma was greater than 90%.

Chromatographic conditions for the analysis of PTX in biological samples were the same as those of in vitro drug samples, with the mobile phase slightly adjusted to methanol/water at 68/32 (v/v). The standard curve was constructed in the range of 0.025–5 μg mL⁻¹ PTX (r = 0.9999); the limit of quantitation (LOQ) was 10.0 ng mL⁻¹ PTX in plasma. No interference was observed in detecting PTX and the internal standard.

2.10. Data analysis

Experiments were conducted at least 3 replicates, and data were expressed as mean ± standard deviation (SD). Pharmacokinetic parameters were calculated using Kinetica 5.0 Pharmacokinetics Software, in compartmental and non-compartmental models. GraphPad Prism Software 5.0 was used to perform statistical analysis of the data. Groups were compared and analyzed using Student's t-test or one-way ANOVA with Tukey's Post Hoc test comparison and statistical significance was set at P < 0.05.

3. Results and discussion

One of the differences between tumor vasculatures and normal vasculatures is known as the enhanced permeability and retention (EPR), which can provide feasibility for selective delivery of therapeutic agents to solid tumors.³¹ Polymeric micelles have shown potentials of significant benefits in antitumor therapy. They are capable of improving pharmacokinetic properties, facilitating drug delivery across various biobarriers, accumulating in tumor tissues via EPR effect, and increasing tissue specificity by surface modification with targeting moieties.³² Furthermore, smart micelle drug delivery platforms that release payloads in response to an intrinsic biological signal also permit favorable drug loading capacity, stability and circulation time.^33,34 In this study, novel smart hybrid micelles were constructed on the basis of pH-sensitive ternary graft copolymer mPAL using cationic arginine (Arg) as stabilizer to balance the extracellular stability and intracellular drug release. Subsequently the profiles of smart micelles as drug delivery system were evaluated.

3.1. Synthesis and characterization of mPAL

A new amphiphilic grafted copolymer with mPEG and PAA as hydrophilic moieties and PLA as hydrophobic moieties was synthesized, as shown in Scheme 2. mPEG and PLA were conjugated to PAA by esterification and ring-opening polymerization, respectively. The structure of the graft polymer was characterized by FT-IR and ¹H NMR spectra.

As shown in Fig. 1A, a broad –OH stretch absorption band between 3500 and 3200 cm⁻¹ and the peaks at 2948 cm⁻¹ and 2878 cm⁻¹ represented the –CH stretch. The strong peak at 1756 cm⁻¹ represented the stretch absorption of C=O, which indicated that acrylic acid was not fully reacted. 1270–1180 cm⁻¹ band and 1180–1050 cm⁻¹ band represented –C=O, –C–O stretch of ester and –C–O stretch of ether, respectively.

Fig. 1 Characterization of the copolymers. (A) FT-IR spectra of lactide (LA), mPAL, PAA-g-mPEG, mPEG, and PAA. ¹H NMR spectra of (B) PAA-g-mPEG and (C) PAL in CDCl₃.

¹H NMR spectra of PAA-g-mPEG and mPAL were shown in Fig. 1B and C. Peaks were assigned as the following: 3.32 ppm (–CH₃, mPEG group), 3.54 ppm (–CH₂, mPEG group); signals at 2.30 ppm and 1.76 ppm were –CH₂ of PAA segment; quartet peak at 4.98 ppm and duplex peak at 1.50 ppm were attributed to consecutive –CH and –CH₃ of PLA segment, respectively. The substitution degree of mPEG in mPAL was 18.4%, depending on the ratio of integration values of methyl protons in mPEG to that of methylene protons in PAA. The molecular weights of mPAL with PAA-g-mPEG/D,L-lactide feed ratio of 6 : 4, 7 : 3, and 8 : 2 were 19 200, 16 600 and 12 900, respectively, measured by Ubisch viscometer method.

Critical micelle concentration (CMC) plays an important role in maintaining stability of various drug delivery systems. In this study CMC was measured by fluorescence technique using pyrene as a probe, of which the spectra contained a vibrational band exhibiting high sensitivity to the polarity of pyrene environment.³⁵ A substantial increase of the intensity ratio was observed when the concentration was higher than the CMC value, indicating the formation of micelles. Therefore, CMC was calculated as the interception of two straight lines, i.e., one being the fitted line at low nanoaggregate concentrations and the other being the fitted line on the rapid rising part of the curve. As listed in Table 1, CMC values of mPAL with various feed ratios were in the range of 2.69–14.3 mg L⁻¹, which was significantly lower than that of low molar mass surfactant.³⁶ With the addition of Arg, the CMC values of all mPAL copolymers decreased slightly, due to the additional stabilizing forces of ionic interaction between Arg and PAA in the copolymers. Lower CMC values would be essential in maintaining thermodynamic stability, improving circulation time and tumor targeting of the micelles.^37,38 The higher hydrophobic segment ratio was, the lower the CMC value became; hydrophobicity of the copolymer played a pivotal role in lowering the CMC. However, the ratio of hydrophobic segment should also be adjusted to achieve desirable solubility of the copolymer. Based on low CMC value, it was postulated that micelles formed from mPAL and Arg would possess good stability in aqueous medium and retain sufficient stability even after intravenous administration and dilution within the blood circulation, which was beneficial to delivering anticancer molecules to tumor sites. The sizes and size distributions of nanoassemblies were measured using DLS. With the incorporation of Arg, the sizes of all micelles increased slightly, but the size distributions changed little, indicating the interaction between mPAL and Arg. The size enlargement of micelles with Arg may be attributed to the adsorption of arginine on the surface of mPAL, in which the anionic PAA is hydrophilic and formed the outer shell of the micelles.

Table 1 Characteristics of the copolymers with various feed ratio (mean ± SD, n = 3)

mPAL	Without Arg			With Arg
	Size^b (nm)	PDI^b	CMC^c (mg L^−1)	Size^b (nm)	PDI^b	CMC^c (mg L^−1)
a Feed ratio of PAA-g-mPEG to D,L-lactide.b Determined using a Malvern Zetasizer 3000 system.c Determined using pyrene as the fluorescence probe.
6 : 4^a	57.6 ± 3.3	0.14 ± 0.02	2.69	58.8 ± 3.4	0.14 ± 0.04	2.38
7 : 3^a	79.6 ± 4.8	0.21 ± 0.03	5.62	81.3 ± 2.9	0.22 ± 0.02	5.29
8 : 2^a	126.1 ± 6.2	0.18 ± 0.01	14.3	128.6 ± 5.7	0.19 ± 0.06	12.7

---

3.2. Preparation and characterization of PTX loaded micelles

It was possible to enhance PTX aqueous solubility, to prolong its circulation time in body and to facilitate specific drug targeting by micelle formulation. A film dispersion-ultrasonic method was used to prepare PTX loaded micelles in this study. Of amphipathic nature, the graft copolymer mPAL self-assembled into micellar structures in aqueous medium. Arg, a stabilizer and cross-linker, was introduced to the micelles to enhance the stability of micelles. The prepared PTX loaded micelles were lyophilized with 3% mannitol, indicating viability of formulating as an injectable preparation. Various parameters of PTX loaded micelles based on mPAL with different substitution degrees are listed in Table 2.

Table 2 Characteristics of PTX loaded micelles (mean ± SD, n = 3)

Micelles	Size^b (nm)	PDI^b	Zeta-potential^b (mV)	DL (%)	EE (%)
a Feed ratio of PAA-g-mPEG to D,L-lactide.b Determined using a Malvern Zetasizer 3000 system.
mPAL (6 : 4)^a/Arg	62.8 ± 2.9	0.16 ± 0.02	−18.7 ± 2.9	8.5 ± 0.2	93.3 ± 2.9
mPAL (7 : 3)^a/Arg	82.3 ± 5.6	0.23 ± 0.04	−20.3 ± 3.4	7.9 ± 0.4	85.4 ± 4.2
mPAL (8 : 2)^a/Arg	129.4 ± 4.7	0.19 ± 0.06	−23.8 ± 4.4	6.7 ± 0.3	72.6 ± 3.4
mPAL (6 : 4)^a	63.2 ± 4.6	0.15 ± 0.03	−26.2 ± 3.1	9.3 ± 0.3	87.5 ± 3.3

---

Drug loading (DL) and encapsulation efficiency (EE) of micelles were also measured. As shown in Table 2, the hybrid micelles formed from mPAL and Arg could efficiently entrap PTX in the hydrophobic core of micelles with high encapsulation efficiency (>72.6%) and the drug loading content was ranged from 6.7% to 8.5%. The results demonstrated that the drug loading content of mPLA/Arg micelles was adjustable, which is an important characteristic for drug delivery systems. Interestingly, when substitution of PLA was increased, particle size of PTX loaded micelles would decrease, entrapment efficacy and drug loading would increase. The introduction of Arg showed no influence on the encapsulation of PTX. Subsequently, micelle formulation with the highest drug loading, smallest particle size and proper poly index, zeta potential and encapsulation efficiency was selected for further testing.

TEM and AFM techniques were utilized to directly visualize the size and morphology of PTX loaded micelles. Both PTX@mPAL/Arg (Fig. 2A and B) and PTX@mPAL (Fig. S1†) micelles were spherical in shape with smooth surface and uniform distribution. The average particle size of micelles was approximately 50 nm, which was smaller than the hydrodynamic diameters measured by DLS analysis. This difference might be attributed to collapse of micelles during drying process; nevertheless uniform particle size distribution was beneficial to physical stability of the drug delivery system.³⁹ Particle size is also critical in maintaining desirable circulation, biodistribution and cellular uptake of the polymeric micelles upon systemic administration. When micelles ranged in diameter of 20–100 nm, it was possible to avoid filtration by the kidney, minimize specific sequestration by sinusoids in the spleen and fenestra in the liver, which was reportedly approximately 150–200 nm in size.³¹ Particle size and polydispersity index (PDI) of the selected PTX@mPAL/Arg micelles were 62.8 ± 2.9 nm and 0.16 ± 0.02 determined by DLS (Fig. 2C), which would be suitable for spontaneous penetration of the micelles into leaky vasculatures of tumor interstitium via EPR effect.⁴⁰

Fig. 2 Physico-chemical characterization of micelles. (A) Transmission electron microscopy (TEM) image, (B) atomic force microscope (AFM) image, and (C) size distribution of PTX@mPAL/Arg micelles. (D) PTX release profiles from drug loaded micelles under different pH conditions with 0.1% (w/v) Tween 80 (mean ± SD, n = 3). (E) Stability of micelles in terms of particle size and drug loading stored at 25 °C (mean ± SD, n = 3, *P < 0.05, significant difference from the characteristics of micelles at 0 h).

Surface charge of the micelles is another important parameter that could influence in vivo behaviors such as adsorption to plasma proteins, recognition by macrophages, and elimination from the circulation. Literatures have reported that nano-drug delivery system with too negative charge was difficult to be internalized, while the ones with too positive charge had strong adsorptive interactions with the cell membranes.^41,42 Zeta potential of the selected mPAL micelles was −18.7 ± 2.9 mV, which would reduce phagocytic uptake, thereby elongating circulation time in blood.⁴³

3.3. Drug release and stability of PTX loaded micelles

Specific drug targeting and release at tumor site is one of the key elements for anticancer delivery systems; premature drug leaking from polymeric micelles within systemic circulation may often compromise chemotherapy efficacy.⁴⁴ Hence, it is critical to design a nano-sized preparation that is capable of self-triggered drug release at tumor location. Unlike normal tissues, tumor microenvironment is generally acidic (pH 4.0–7.0) due to the accumulation of acidic metabolites caused by anaerobic glycolysis under hypoxia.^21,45 A sustained and pH-dependent drug release at slightly acidic condition would certainly enhance chance for drug delivery to tumor site. As shown in Fig. 2D, both the micelles displayed a typical two-phase release with rapid drug release during the initial 12 h and slow release at the later time. There was no significant initial burst at pH 7.4 for the two micelles, indicating the high drug entrapment efficiency and few desorption from the surface. The release mechanisms of PTX loaded micelles for the faster stage and the slower stage were diffusion through pores, and degradation of the polymeric matrix, respectively. Drug release from PTX@mPAL and PTX@mPAL/Arg micelles was tested at pH 5.0, 6.5 and 7.4 in this study. Cumulative drug release of PTX@mPAL micelles was slightly faster at pH 7.4 than that at pH 5.0, which was agree with the particle size changes of PTX@mPAL according to pH 7.4 and 5.0 (Fig. S2†). At lower pH, the protonation of carboxyl groups in PAA segments caused the shrinkage of PAA chains, which tightly wrapped inner cores and prevented the encapsulated drug release. Therefore, particles size and drug release of PTX@mPAL micelles were reduced. It was interesting that drug release from PTX@mPAL/Arg micelles were significantly accelerated with decreasing pH in PBS buffer. For example, 79.93%, 65.17% and 40.44% drugs were released from PTX@mPAL/Arg micelles in 36 h at pH 5.0, 6.5 and 7.4, respectively. The average particle sizes of PTX@mPAL/Arg micelles according to pH 5.0, 6.5 and 7.4 were 79.7, 68.9 and 62.9 nm. Notably, the almost complete drug release was observed at pH 5.0 wherein 94.24% of PTX was released from mPAL/Arg micelles within 72 h. The pH sensitive release behaviour of hybrid micelles was related closely to the particle size changes, which was the intuitive performance of the electrostatic interaction of cationic Arg and carboxyl groups in PAA segments as well as the hydrogen bond formation between guanidinium in Arg and carboxylate oxygen in PTX. This selective drug release characteristics could potentiate the micelles as an antitumor delivery vehicle that facilitated drug release to cancer cells and reduced toxicity to normal tissues at the same time.

Stability of PTX loaded micelles upon reconstitution in distilled water was tested by checking changes in particle size and drug loading for 96 h at 25 °C (Fig. 2E). No significant changes were observed in particle size and drug leakage within 96 h for PTX@mPAL/Arg micelles; however, average particle size of PTX@mPAL micelles increased from 63.2 to 88.3 nm after 96 h while drug loading decreased to 64.6%. The results indicated that the presentation of Arg in micelles enhanced the stability of micelles, which was attributed to the strong electrostatic interaction between positively charged Arg moleculars and negatively charged carboxyl groups in PAA segments. Meanwhile, the stability of micelles under physiological conditions was also evaluated using PBS solution containing 10% FBS (Fig. S3†). Hydrodynamic sizes of both PTX@mPAL and PTX@mPAL/Arg micelles did not change much, suggesting the prevention of non-specific protein adsorption and aggregation by PEG modification on the surface of micelles.

3.4. Biocompatibility of copolymer

Interactions between nanocarrier ingredients and blood components may trigger pathophysiological reaction or alter nanocarrier behaviors in vivo, including changes in pharmacokinetics, biodistribution and therapeutic efficacy.⁴⁶ It is therefore important to evaluate blood biocompatibility of the copolymer for safe intravenous administration of the micelles. Tween 80 is a hydrophilic surfactant with high capability for cell membrane destruction, which is often used as positive control in hemolytic test. As shown in Fig. 3A, hemolysis induced by mPAL and mPAL/Arg was significantly lower than that by Tween 80, especially at high concentrations, which met the criteria of <5% in the biological function test for biomaterials according to the Standard Practice for Assessment of Haemolytic Properties of Materials (ASTM F756-2008).⁴⁷ The possible reasons for decreased hemolysis of mPAL and mPAL/Arg may be the surface modification of PEG, a biocompatible copolymer, which could reduce the interaction with red blood cells. Both mPAL and mPAL/Arg exhibited no significant difference with Cremophor EL at any testing concentration. The results suggested that mPAL and mPAL/Arg possessed satisfactory compatibility in blood, which might be further formulated as injectable micelle carriers.

Fig. 3 (A) Hemolysis of Tween 80, Cremophor EL, blank micelles of mPAL and mPAL/Arg at different concentrations. Cytotoxicity of blank micelles and Taxol vehicles for (B) A549 cells and (C) HepG2 cells at corresponding PTX concentrations after incubation for 48 h (mean ± SD, n = 3, *P < 0.05 compared to Tween 80 or Taxol vehicle).

Cytotoxicity of blank micelles and Taxol vehicle was also tested in A549 and HepG2 cell lines by MTT assay (Fig. 3B). Blank micelles of mPAL and mPAL/Arg were practically non-toxic with cell viabilities above 87.5% at corresponding PTX concentrations ranged from 0.001 to 20 μg mL⁻¹ for 48 h. The results confirmed that these polymeric micelles would be safe for biomedical applications. On the other hand, when testing concentrations were above 1 μg mL⁻¹, Taxol vehicle displayed significant cytotoxicity, which was partly attributed to cytostatic action by Cremophor EL. Comparatively blank mPAL and mPAL/Arg micelles would be desirable for PTX as a drug delivery carrier.

3.5. In vitro antitumor efficacy

The cytotoxicity of Taxol and PTX loaded micelles was tested in A549 and HepG2 cells using MTT assays. As shown in Fig. 4, cell viability of A549 and HepG2 negatively correlated with the concentrations of drugs. PTX@mPAL and PTX@mPAL/Arg micelles displayed inferior antitumor activity at higher concentrations for A549 cells to that for HepG2 cells. The MTT assay of PTX formulations yielded comparable 50% inhibitory concentration (IC₅₀) values for MCF-7 and A549 cells, which were calculated by GraphPad Prism Software 5.0 and shown in Table S1.† Due to cytotoxicity from Taxol vehicles (1.15 ± 0.28 μg mL⁻¹), IC₅₀ values of PTX@mPAL (1.76 ± 0.42 μg mL⁻¹) and PTX@mPAL/Arg (0.67 ± 0.13 μg mL⁻¹) micelles for A549 cells was significantly higher than that of Taxol (0.19 ± 0.08 μg mL⁻¹). It should be noted that both PTX@mPAL and PTX@mPAL/Arg micelles exhibited low IC₅₀ values of 0.33 ± 0.12 and 0.20 ± 0.04 μg mL⁻¹ for HepG2 cells, which approached that of Taxol (0.23 ± 0.07 μg mL⁻¹). Interestingly, PTX@mPAL/Arg micelles displayed lowed IC₅₀ values than PTX@mPAL micelles for both A549 and HepG2 cells, which indicated that Arg hybridization had advantageous effects on the intracellular drug release and antitumor effects. However, the differences of tumor cell inhibition between PTX@mPAL and PTX@mPAL/Arg were not so big as that of in vitro drug release. This could be attributed to the functions of hydrolytic enzymes, especially esterase in tumor cells. The graft copolymer might also affect the microenvironment of cell membrane and alter the structure and function of proteins, thus reducing P-gp mediated drug efflux.⁴⁸ The antitumor efficacy of PTX@mPAL/Arg micelles might be further enhanced by decoration with targeting ligands such as antibody, folic acid or aptamer, which could improve cellular uptake of micelles. It was confirmed that PTX@mPAL/Arg micelles possessed superior extracellular stability could efficiently deliver and release drugs after endocytosis by cancer cells.

Fig. 4 In vitro antitumor efficacy of PTX loaded micelles and Taxol at various PTX concentrations after incubation for 48 h. (A) A549 cells; (B) HepG2 cells (mean ± SD, n = 3, *P < 0.05 compared to Taxol).

3.6. Cellular internalization

A cellular uptake assay was performed to ascertain the endocytosis and intracellular drug release of mPAL and mPAL/Arg micelles in HepG2 cells with CLSM. As shown in Fig. 5, some fluorescence of C6 could be observed in cytoplasm after incubating C6 labeled micelles with tumor cells for 2 h. The C6 fluorescence in the cytosol became brighter as the extension of incubation time to 4 h, which was crucial for antimitotic agents such as PTX.³³ Small size and lower zeta potential of hybrid micelles might help facilitate the endocytosis of PTX@mPAL/Arg micelles, leading to enhanced drug uptake. It was noted that the average fluorescence intensity for mPAL/Arg micelles at 2 h was 2.6 times that for mPAL micelles after the same incubation time (Fig. S4†), which was most likely due to the improved cellular uptake and pH stimulated fast drug release from mPAL/Arg micelles. These results have demonstrated that the hybrid micelles could mediate more efficient cellular uptake and intracellular anticancer drug release.

Fig. 5 Representative confocal laser scanning microscopy (CLSM) images of HepG2 cells incubated with coumarin (C6) labeled mPAL and mPAL/Arg micelles for 2 and 4 h. For each panel, the images from left to right showed green fluorescence from C6 labeled micelles, cell nuclei stained blue by DAPI, and overlays of both images. The scale bars correspond to 20 μm in all the images.

3.7. In vivo pharmacokinetics

Drug concentration–time profiles following a single intravenous dosing of Taxol, PTX@mPAL and PTX@mPAL/Arg micelles in rats are displayed in Fig. 6. PTX plasma levels of all the three preparations exhibited biphasic decline which could be characterized as a rapid distribution phase (<0.5 h) followed by an elimination phase. PTX@mPAL and PTX@mPAL/Arg micelles significantly prolonged drug retention time in blood circulation, even though initial plasma PTX concentration from the micelles was lower than that of Taxol upon administration. PTX concentration from the micelles was still detectable 12 h after administration, while that from Taxol was below the quantification limit 6 h after administration.

Fig. 6 Average drug plasma concentration–time profiles of Taxol and PTX loaded micelles in rats after intravenous administration of 5 mg kg⁻¹ PTX (mean ± SD, n = 6).

Table 3 summarizes pharmacokinetic parameters of PTX in the three formulations, calculated using two compartmental or non-compartmental methods; there was no difference between the two models. In comparison, plasma elimination half-life (t_1/2) values of PTX@mPAL and PTX@mPAL/Arg micelles were significantly increased by 1.63 and 3.72-fold, respectively (P < 0.05). The area under the concentration–time curve (AUC_0→∞) of PTX@mPAL/Arg micelles was 11.23 ± 0.77 μg mL⁻¹ h⁻¹, which was the highest in PTX formulations. The rate constants of distribution phase (α) and elimination phase (β) in plasma determined the AUC of drugs. Compared to that of Taxol, t_1/2α and t_1/2β values of PTX@mPAL and PTX@mPAL/Arg micelles were much higher, indicating the slower disappearance of drugs in both distribution and elimination stage. Mean residence time (MRT_0→∞) of PTX@mPAL/Arg micelles was 6.18-fold higher than that of Taxol and 2.26-fold higher than that of PTX@mPAL micelles. Total clearance (CL) rate for micelles was much lower than that of Taxol. The results indicated that PTX loaded micelles produced a well-sustained drug release in rats, which might be attributed to the rigid core–shell micellar architecture, plus the introduction of PEG to micelles resulting in low protein adsorption and long body circulation. Numerous studies have confirmed that intravenous administration of nanoscaled drug systems could significantly alter pharmacokinetic properties of drug substances.^49–51 Nanocarriers with PEG modification could prolong circulation time, prevent rapid drug elimination, and increase EPR effect, which would further potentiate passive drug targeting to non-mononuclear phagocyte system (MPS) disease regions such as tumor tissues.⁵² The significant differences in t_1/2, MRT_0→∞ and CL between PTX@mPAL and PTX@mPAL/Arg micelles (P < 0.01) indicated that the introduction of positively charged Arg did influence the in vivo circulation properties of the micelles. PTX@mPAL/Arg micelles with more thermodynamic stability were less susceptible to disintegrate after blood dilution upon administration, which could slow down the decrease of drug concentration in the elimination phase.

Table 3 Pharmacokinetic parameters of PTX after single intravenous administration of Taxol and PTX loaded micelles in rats (mean ± SD, n = 6)^c

Pharmacokinetic parameters	Taxol	PTX@mPAL	PTX@mPAL/Arg
a Pharmacokinetic parameters calculated by a two-compartment model.b Pharmacokinetic parameters calculated by noncompartment model.c *P < 0.05, **P < 0.01 compared to Taxol. ^#P < 0.05, ^##P < 0.01 compared to PTX@mPAL.
α^a (h^−1)	9.35 ± 0.52	6.29 ± 0.43**	7.17 ± 0.46**
β^a (h^−1)	0.35 ± 0.03	0.22 ± 0.02**	0.14 ± 0.01**^#
t1/2α^a (h)	0.07 ± 0.01	0.11 ± 0.01**	0.10 ± 0.01**
t1/2β^a (h)	1.97 ± 0.21	3.09 ± 0.29**	5.11 ± 0.34**^#
t1/2^b (h)	1.77 ± 0.43	2.89 ± 0.52*	6.58 ± 0.59**^##
CL^b (mL h^−1 kg^−1)	1374.87 ± 82.53	927.27 ± 72.73**	445.19 ± 69.37**^##
AUC0→12^b (μg mL^−1 h^−1)	3.47 ± 0.41	5.10 ± 0.42**	8.31 ± 0.61**^##
AUC0→∞^b (μg mL^−1 h^−1)	3.64 ± 0.44	5.39 ± 0.48**	11.23 ± 0.77**^##
MRT0→∞^b (h)	1.41 ± 0.22	3.86 ± 0.25**	8.72 ± 0.43**^##

---

4. Conclusion

It has been demonstrated in this paper that novel pH-sensitive hybrid micelles (PTX@mPAL/Arg) based on ternary biocompatible graft copolymer mPAL and cross-linker arginine (Arg) showed enormous potential in solving the dilemma of superior extracellular stability and rapid intracellular drug release. The intelligent micelles displayed high drug loading capacity with spherical morphology and a particle size around 60 nm. It was worth noting that PTX@mPAL/Arg micelles exhibited excellent stability against dilution and serum with inhibited premature drug release. Significantly, the hybrid micelles promoted fast and maximum drug release at artificial tumor microenvironment relative to physiological conditions, and demonstrated significant cytotoxicity to cancer cells attributing to enhanced PTX solubility and efficient cellular internalization. In addition, PTX@mPAL/Arg micelles prolonged in vivo circulation time and produced satisfactory pharmacokinetic parameters in rats, which was desirable for enhancing therapeutic efficacy of PTX. Further studies are ongoing to research the interaction mechanisms of the three components. These results indicated that the pH-sensitive hybrid micelles could be a promising nanocarrier for efficient anticancer drug delivery.

Acknowledgements

This study was financially supported by Grants from the National Natural Science Foundation of China (No. 81373363), Key New Drug Innovation Project from the Ministry of Science and Technology of the People's Republic of China (No. 2009ZX09310-004), the National Major Scientific and Technological Special Project for “Significant New Drugs Development” during the Twelfth Five-year Plan Period (No. 2015ZX09501001), and the Fundamental Research Funds for the Central Universities (No. PT2014YX0085). We thank the Cellular and Molecular Biology Center of China Pharmaceutical University for assistance with confocal microscopy work and we are grateful to Xiaonan Ma for her technical help.

References

1. S. S. Razi, S. Rehmani, X. Li, K. Park, G. S. Schwartz, M. J. Latif and F. Y. Bhora, J. Surg. Res., 2015, 194, 622–630.
2. Y. Liu, Y. Ng, M. R. Toh and G. N. Chiu, J. Controlled Release, 2015, 220, 438–446.
3. V. Das, J. Štěpánková, M. Hajdúch and J. H. Miller, Biochim. Biophys. Acta, 2015, 1855, 172–182.
4. L. M. Zasadil, K. A. Andersen, D. Yeum, G. B. Rocque, L. G. Wilke, A. J. Tevaarwerk, R. T. Raines, M. E. Burkard and B. A. Weaver, Sci. Transl. Med., 2014, 6, 229ra43.
5. R. P. Symonds, C. Gourley, S. Davidson, K. Carty, E. Mccartney, D. Rai, S. Banerjee, D. Jackson, R. Lord, M. McCormack, E. Hudson, N. Reed, M. Flubacher, P. Jankowska, M. Powell, C. Dive, C. M. L. West and J. Paul, Lancet Oncol., 2015, 16, 1515–1524.
6. A. Sparreboom, O. van Tellingen, W. J. Nooijen and J. H. Beijnen, Cancer Res., 1996, 56, 2112–2115.
7. J. Nicolas, S. Mura, D. Brambilla, N. Mackiewicz and P. Couvreur, Chem. Soc. Rev., 2013, 42, 1147–1235.
8. T. M. Allen and P. R. Cullis, Science, 2004, 303, 1818–1822.
9. J. Zhang, K. Chen, Y. Ding, X. Xin, W. Li, M. Zhang, H. Hu, M. Qiao, X. Zhao and D. Chen, RSC Adv., 2016, 6, 23693–23701.
10. M. E. Davis, Z. Chen and D. M. Shin, Nat. Rev. Drug Discovery, 2008, 7, 771–782.
11. H. Maeda, H. Nakamura and J. Fang, Adv. Drug Delivery Rev., 2013, 65, 71–79.
12. C. A. Simpson, A. C. Agrawal, A. Balinski, K. M. Harkness and D. E. Cliffel, ACS Nano, 2011, 5, 3577–3584.
13. S. Yin, J. Huai, C. Xi, Y. Yang, X. Zhang, Y. Gan, G. Wang, X. Gu and J. Li, Acta Biomater., 2015, 26, 274–285.
14. W. Chen, F. Meng, R. Cheng, C. Deng, J. Feijen and Z. Zhong, J. Controlled Release, 2015, 210, 125–133.
15. S. C. Owen, D. P. Y. Chan and M. S. Shoichet, Nano Today, 2012, 7, 53–65.
16. Y. Zhao, C. Shi, X. Yang, B. Shen, Y. Sun, Y. Chen, X. Xu, H. Sun, K. Yu, B. Yang and Q. Lin, ACS Nano, 2016, 10, 5856–5863.
17. K. M. Poole, C. E. Nelson, R. V. Joshi, J. R. Martin, M. K. Gupta, S. C. Haws, T. E. Kavanaugh, M. C. Skala and C. L. Duvall, Biomaterials, 2015, 41, 166–175.
18. M. Karimi, A. Ghasemi, P. Sahandi-Zangabad, R. Rahighi, S. M. Moosav-Basri, H. Mirshekari, M. Amiri, Z. Shafaei-Pishabad, A. Aslani, M. Bozorgomid, D. Ghosh, A. Beyzavi, A. Vaseghi, A. R. Aref, L. Haghani, S. Bahrami and M. R. Hamblin, Chem. Soc. Rev., 2016, 45, 1457–1501.
19. E. J. Kwon, J. H. Lo and S. N. Bhatia, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 14460–14466.
20. H. Deng, Y. Zhang, X. Wang, J. Zhang, Y. Cao, J. Liu, J. Liu, L. Deng and A. Dong, Acta Biomater., 2015, 11, 126–136.
21. R. Mo, Q. Sun, J. Xue, N. Li, W. Li, C. Zhang and Q. Ping, Adv. Mater., 2012, 24, 3659–3665.
22. Y. Tian, L. Bromberg, S. N. Lin, T. A. Hatton and K. C. Tam, J. Controlled Release, 2007, 121, 137–145.
23. E. S. Choo, B. Yu and J. Xue, J. Colloid Interface Sci., 2011, 358, 462–470.
24. Y. Tian, P. Ravi, L. Bromberg, T. A. Hatton and K. C. Tam, Langmuir, 2007, 23, 2638–2646.
25. H. Gao, Y. Tan, Q. Guan, T. Cai, G. Liang and Q. Wu, RSC Adv., 2015, 5, 49376–49384.
26. A. T. Wang, D. S. Liang, Y. J. Liu and X. R. Qi, Biomaterials, 2015, 53, 160–172.
27. H. Zhang, L. Zhao, L. Chu, X. Han and G. Zhai, J. Colloid Interface Sci., 2014, 434, 40–47.
28. J. O. Kim, A. V. Kabanov and T. K. Bronich, J. Controlled Release, 2009, 138, 197–204.
29. I. Montasser, H. Fessi and A. W. Coleman, Eur. J. Pharm. Biopharm., 2002, 54, 281–284.
30. R. Raveendran, G. S. Bhuvaneshwar and C. P. Sharma, Carbohydr. Polym., 2016, 137, 497–507.
31. X. Duan and Y. Li, Small, 2013, 9, 1521–1532.
32. K. Kettler, K. Veltman, D. van-de-Meent, A. van-Wezel and A. J. Hendriks, Environ. Toxicol. Chem., 2014, 33, 481–492.
33. T. Yin, L. Wang, L. Yin, J. Zhou and M. Huo, Biomaterials, 2015, 61, 10–25.
34. Y. Li, K. Xiao, W. Zhu, W. Deng and K. S. Lam, Adv. Drug Delivery Rev., 2014, 66, 58–73.
35. L. Zhang, R. Guo, M. Yang, X. Jiang and B. Liu, Adv. Mater., 2007, 19, 2988–2992.
36. K. N. Prasad, T. T. Luong, A. T. Paris, F. C. Vaution, M. Seiller and F. Puisieux, J. Colloid Interface Sci., 1979, 69, 225–232.
37. S. Su, Y. Tian, Y. Li, Y. Ding, T. Ji, M. Wu, Y. Wu and G. Nie, ACS Nano, 2015, 9, 1367–1378.
38. C. T. Nguyen, T. H. Tran, M. Amiji, X. Lu and R. M. Kasi, Nanomedicine, 2015, 11, 2071–2082.
39. Y. Zhao, J. Li, H. Yu, G. Wang and W. Liu, Int. J. Pharm., 2012, 430, 282–291.
40. C. Shi, X. Guo, Q. Qu, Z. Tang, Y. Wang and S. Zhou, Biomaterials, 2014, 35, 8711–8722.
41. S. Chen, S. Hu, E. F. Smith, P. Ruenraroengsak, A. J. Thorley, R. Menzel, A. E. Goode, M. P. Ryan, T. D. Tetley, A. E. Porter and M. S. P. Shaffer, Biomaterials, 2014, 35, 4729–4738.
42. K. Saha, A. Bajaj, B. Duncan and V. M. Rotello, Small, 2011, 7, 1903–1918.
43. F. Yamashita and M. Hashida, Adv. Drug Delivery Rev., 2013, 65, 139–147.
44. R. R. Arvizo, O. R. Miranda, D. F. Moyano, C. A. Walden, K. Giri, R. Bhattacharya, J. D. Robertson, V. M. Rotello, J. M. Reid and P. Mukherjee, PLoS One, 2011, 6, e24374.
45. V. P. Torchilin, Nat. Rev. Drug Discovery, 2014, 13, 813–827.
46. W. Hu, L. Qiu, L. Cheng, Q. Hu, Y. Liu, Z. Hu, D. Chen and L. Cheng, Acta Biomater., 2016, 36, 241–253.
47. L. Zhang, P. Zhang, Q. Zhao, Y. Zhang, L. Cao and Y. Luan, J. Colloid Interface Sci., 2016, 464, 126–136.
48. H. Zou, Z. Wang and M. Feng, J. Controlled Release, 2015, 214, 121–133.
49. L. Xiao, X. Xiong, X. Sun, Y. Zhu, H. Yang, H. Chen, L. Gan, H. Xu and X. Yang, Biomaterials, 2011, 32, 5148–5157.
50. G. Tripodo, G. Pasut, A. Trapani, A. Mero, F. M. Lasorsa, T. Chlapanidas, G. Trapani and D. Mandracchia, Biomacromolecules, 2015, 16, 550–557.
51. G. M. Garcia, L. T. Oliveira, R. Pitta-Ida, C. de-Lima-Mdo, J. M. Vilela, M. S. Andrade, D. S. Abdalla and V. C. Mosqueira, J. Controlled Release, 2015, 209, 207–218.
52. A. Baumann, D. Tuerck, S. Prabhu, L. Dickmann and J. Sims, Drug Discovery Today, 2014, 19, 1623–1631.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1, S2 and Table S1. See DOI: 10.1039/c6ra23050d

---