Stimuli-responsive terpolymer mPEG-b-PDMAPMA-b-PAH mediated co-delivery of adriamycin and siRNA to enhance anticancer efficacy

Abstract

Combination of chemotherapy and small interfering RNA (siRNA)-based therapy has emerged as a promising approach for cancer treatment with the synergistic effect. In this study, a novel terpolymer methoxy poly(ethylene glycol)-block-poly(N,N-dimethylaminopropyl methacrylamide)-block-poly(acrylhydrazine) (mPEG-b-PDMAPMA-b-PAH) with a disulfide linkage between mPEG and PDMAPMA blocks was developed for the intracellular targeted co-delivery of adriamycin and P-gp siRNA. The terpolymer was synthesized by sequential reversible addition–fragmentation chain transfer (RAFT) polymerization of N,N-dimethylaminopropyl methacrylamide and N-tert-butoxycarbonyl-N′-acryl hydrazine (Boc-acrylhydrazine) in the presence of a PEGylated macro-RAFT agent, followed by Boc-deprotection. The terpolymer could chemically conjugate adriamycin via an acid-cleavable hydrazone bond and simultaneously condense the negatively charged siRNA through electrostatic interactions at an N/P ratio of 2. The resultant adriamycin-conjugated nanoparticles/siRNA complexes (ADR-NPs/siRNA complexes) showed a spherical morphology and had an average diameter of 186 nm. The release profiles of the two payloads from the ADR-NPs/siRNA complexes exhibited a pH/reduction dual-dependent sustained release behavior. The ADR-NPs/siRNA complexes could simultaneously deliver adriamycin and siRNA efficiently into MCF-7/ADR cells and significantly inhibit their growth in a synergistic fashion. All the results indicated that the terpolymer mPEG-b-PDMAPMA-b-PAH could serve as a potential vehicle for the combination of chemotherapy and gene therapy.

---

1. Introduction

Chemotherapy is one of the most effective approaches to treating cancers in the clinic, however there remain some challenges for the applications of anticancer drugs, such as non-solubility in water, individual sensitivity and non-specific toxicity.¹ Moreover, overexpression of drug efflux transporters such as P-glycoprotein (P-gp) which may pump the anticancer drugs out of cancer cells enhances drug resistance in chemotherapy.² Recently, small interference RNA (siRNA), which can disrupt cellular pathways by knocking down genes, has been regarded as a promising strategy to sensitize tumor cells to chemotherapy.³ Therefore, the co-delivery of an anticancer drug and siRNA to achieve the synergistic/combined effect of gene therapy and chemotherapy as an emerging strategy has drawn increasing attention in the treatment of cancers.⁴ So far, in order to achieve such co-delivery purpose, several promising systems have been developed,⁵ such as polymeric,⁶ liposomal⁷ and silica-based cationic nanoparticles.⁸ For example, Meng et al. utilized mesoporous silica nanoparticles to deliver adriamycin and P-gp siRNA to drug-resistant KB-V1 cells, which resulted in significantly enhanced cell killing of adriamycin in a synergistic fashion.⁹ Chen et al. used liposomes in co-formulation of c-Myc siRNA and adriamycin to enhance the uptake of adriamycin, which obviously improved tumor growth inhibition.¹⁰ These studies clearly verified that the co-delivery system could effectively silence the expression of efflux transporter and substantially enhance the anticancer activity of adriamycin. In comparison with liposomal and silica-based cationic nanoparticles, polymer-based non-viral vectors have great advantages with respect to safety, convenient large-scale production, and physiological stability.¹¹

In recent years, various polymeric cationic nanoparticles have been developed to co-deliver an anticancer drug and siRNA.¹² However, the co-loading of hydrophobic drug molecules and negatively charged siRNA in the same nanoparticle is still a challenge. Anticancer drugs are usually encapsulated in nanoparticles through hydrophobic–hydrophobic interactions with biodegradable polymers such as poly(ε-caprolactone),¹³ and poly(lactide-co-glycolide).¹⁴ These polymers exhibit gradual degradation kinetics inside the body, leading to sustained drug release over a period of days to weeks and reduced drug efficacy. Recently, pH-triggered controlled release co-delivery systems based on hydrazone linkages have received a lot of attentions.¹⁵ Hydrazone linkage is relatively stable at neutral pH simulating the environment of the blood (pH 7.4), but can rapidly hydrolyze in the mildly acidic environments, such as endosomes (pH 5–6) and lysosomes (pH 4–5).¹⁶ Up to now, a variety of synthetic and natural cationic polymers have been investigated as siRNA carriers in combined therapy, including poly(ethylene imine),¹⁷ poly(L-lysine),¹⁸ chitosan,¹⁹ and polyamidoamine dendrimers.²⁰ These cationic polymers can form stable siRNA–polymer complexes via electrostatic interactions to protect siRNA from enzymatic degradation and facilitate endocytosis and endosomal disruption by the proton sponge effect. Recently, polycations containing primary, secondary and/or tertiary amine, which are synthesized via reversible addition–fragmentation chain transfer (RAFT) polymerization, have received many attentions because of its facile and precise control over polymer molecular weights and narrow molecular weight distributions.^21,22 On the other side, a cationic polymer poly N,N-dimethylaminopropyl methacrylamide (PDMAPMA) has been regard as a potential siRNA vehicle in gene delivery.^23,24 However, the application of polycations is often hindered by the relatively high cytotoxicity originating from the positive charge. To decrease the toxicity and improve the stability of polycationic carriers, PEG has extensively been used as a shielding layer to modify the polymeric vehicles. However, such PEGylation would hinder siRNA complexation and the proton sponge effect, limiting its application.²⁵ Takae et al. synthesized a PEG-detachable copolymer based on disulfide linkages as a gene vector, and the copolymer-derived micelles were sensitive to an intracellular reducing environment and showed high cellular uptakes and gene transfection efficiency in gene delivery.²⁶ To the best of our knowledge, very limited studies have been reported on the stimuli-responsive PEG-detachable copolymers as nanovehicles for the co-delivery of siRNA and an anticancer drug.

In this study, we developed a novel cationic terpolymer methoxy poly(ethylene glycol)-block-poly(N,N-dimethylaminopropyl methacrylamide)-block-poly(acrylhydrazine) (mPEG-b-PDMAPMA-b-PAH) for the purpose of co-delivering adriamycin and P-gp siRNA. The terpolymer was synthesized by sequential RAFT polymerization of N,N-dimethylaminopropyl methacrylamide and N-tert-butoxycarbonyl-N′-acryl hydrazine (Boc-acrylhydrazine) in the presence of a PEGylated macro-RAFT agent containing a disulfide bond, followed by Boc-deprotection. The terpolymer could chemically conjugate adriamycin on the PAH block via hydrazone bonds and simultaneously complex negatively charged siRNA with the cationic PDMAPMA block through electrostatic interactions, resulting in the formation of nanoparticle. The properties of the vehicle, such as structure, morphology, particle size, siRNA-binding ability and in vitro drug release behavior were evaluated in detail. Furthermore, drug uptake and cytotoxicity were evaluated through confocal laser scanning microscopy (CLSM), flow cytometry and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assays in MCF-7/ADR breast cancer cells.

2. Materials and methods

2.1 Materials

Pyrene, 3,3′-dithiodipropionic acid, anhydrous hydrazine, methoxy poly(ethylene glycol) (mPEG, molecular weight: 2 kDa), N,N′-dicyclohexylcarbodiimide (DCC), 4-(N,N-dimethylamino)pyridine (DMAP), 2-(4-amidinophenyl)-6-indolecarbamidine dihydrochloride (DAPI), paraformaldehyde and MTT were purchased from Sigma-Aldrich (Shanghai, China). Acryloyl chloride, tert-butyl carbazate and N,N-dimethylaminopropyl methacrylamide (DMAPMA) were purchased from Aladdin Chemistry Co., Ltd. (Shanghai, China), and DMAPMA was distilled from calcium hydride under reduced pressure prior to use. Adriamycin hydrochloride was purchased from Hisun Pharmaceutical Co., Ltd. (Zhejiang, China). All other reagents and solvents were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Anhydrous organic solvents were dried by refluxing over sodium wire or calcium hydride and distilled before use.

Targeting human P-gp siRNA (sense: 5′-GAA ACC AAC UGU CAG UGU AdTdT-3′, anti-sense: 5′-UAC ACU GAC AGU UGG UUU CdTdT-3′) and fluorescein-tagged siRNA (FAM–siRNA) were supplied by Shanghai GenePharma Co., Ltd. (Shanghai, China). Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), 0.25% (w/v) trypsin-0.03% (w/v) EDTA solution and penicillin–streptomycin were purchased from Invitrogen. Annexin V-FITC apoptosis detection kit from 7-Sea Biotech Co., Ltd. (Shanghai, China) was used according to its protocol.

RAFT agent S-1-dodecyl-S′-(α,α′-dimethyl-α′′-acetic acid)trithiocarbonate (DDACT) was synthesized as a yellow crystalline solid (yield: 35.6%, m.p.: 60–61 °C) according to the literature procedure.^{27 1}H NMR (CDCl₃, Bruker 400 MHz, δ_ppm): δ 0.87 (t, 3H), 1.25–1.38 (m, 18H), 1.68–1.73 (m, 8H), 3.28 (t, 2H). Bis(β-isocyanatoethyl) disulfide (BIED) was synthesized following published procedure.²⁸

2.2 Synthesis of terpolymer mPEG-b-PDMAPMA-b-PAH

mPEG-b-PDMAPMA-b-PAH terpolymer was synthesized via sequential RAFT polymerization of DMAPMA and Boc-acrylhydrazine followed by Boc-deprotection. Boc-acrylhydrazine was synthesized as follows. Tert-butyl carbazate (6.61 g, 50 mmol) was dissolved in 50 mL of distilled water, acryloyl chloride solution (4.06 mL, 50 mmol in 10 mL of dichloromethane (CH₂Cl₂)) was added dropwise. The mixture was stirred for 16 h at room temperature. The solution was extracted by CH₂Cl₂ and the organic phase was separated, dried and precipitated in cold diethyl ether. The crude product was recrystallized in toluene and dried in vacuum, obtaining Boc-acrylhydrazine (yield: 90.4%, m.p.: 154–155 °C). ¹H NMR (in CDCl₃): δ 1.49 (t, –CH₃), 5.76 (m, CH₂=CH–), 6.20 (m, CH₂=CH–), 6.40 (m, CH₂=CH–).

The terpolymer was synthesized in three steps. Firstly, mPEG–DDACT macro-RAFT agent was synthesized by conjugating the carboxyl group of DDACT to the primary amino end group of cystamine-modified mPEG (mPEG–Cys). mPEG–Cys was synthesized by reacting PEG with BIED using dibutyltin dilaurate as the catalyst followed by hydrolysis. Briefly, pre-dried mPEG (4 g, 2.0 mmol) and BIED (2.04 g, 10 mmol) were dissolved in 30 mL of toluene, and then dibutyltin dilaurate (0.05 g) was added to the solution. The mixture solution was stirred at 85 °C for 48 h and precipitated in an excess of n-hexane. The intermediate product was hydrolysed in water at 60 °C for 6 h and lyophilized, obtaining mPEG–Cys. Then, mPEG–Cys (2 g, 1 mmol), DDACT (1.82 g, 5 mmol), DCC (2.06 g, 10 mmol) and DMAP (0.122 g, 1 mmol) were dissolved in 50 mL of CH₂Cl₂. The reaction was performed for 48 h and the solution was precipitated in cold diethyl ether. mPEG–DDACT macro-RAFT agent was obtained as a pale yellow solid (2.09 g, yield: 92.6%). Secondly, terpolymer mPEG-b-PDMAPMA-b-P(Boc-AH) was synthesized via sequential RAFT polymerization of DMAPMA and Boc-acrylhydrazine using azodiisobutyronitrile (AIBN) as the initiator. Briefly, mPEG–DDACT macro-RAFT agent (1.13 g, 0.5 mmol) and DMAPMA (2.554 g, 15 mmol) were dissolved in 12 mL of 1,4-dioxane in a 100 mL round-bottomed flask sealed by rubber septa under a nitrogen atmosphere, and the dissolved oxygen was dislodged through three freeze–pump–thaw cycles. An AIBN solution (8.21 mg, 0.05 mmol in 0.5 mL of 1,4-dioxane) was injected into the flask using degassed syringe and stirred in a preheated oil bath at 60 °C. After the polymerization, the solvent was removed under vacuum, and the crude product was re-dissolved in 10 mL of CH₂Cl₂ and isolated by forming a precipitate in an 8-fold excess of cold diethyl ether. The precipitation process was repeated three times and the copolymer mPEG-b-PDMAPMA–DDACT was dried overnight under vacuum. Yield: 3.38 g (91.8%). Terpolymer mPEG-b-PDMAPMA-b-P(Boc-AH) was synthesized by the same RAFT process. Yield: 4.07 g (94.1%). Thirdly, to remove the protecting Boc groups, mPEG-b-PDMAPMA-b-PAH (2 g) was dissolved in a 1 : 1 mixture of TFA and CH₂Cl₂ and stirred for 6 h at room temperature. After the solvent was removed under reduced pressure, the remaining solid was dialyzed (MWCO 3.5 kDa) against 0.25% ammonia solution for 48 h followed by lyophilisation, obtaining a white product mPEG-b-PDMAPMA-b-PAH. Yield: 1.81 g (90.8%). The nominal degree of polymerization of each block was 45, 30, and 50 respectively.

¹H NMR spectra were obtained on a Bruker spectrometer (400 MHz) at 25 °C using DMSO-d₆ or CDCl₃ as solvents and tetramethylsilane as the internal reference. Gel permeation chromatography (GPC) measurements were performed using a Waters 1515 GPC instrument.

2.3 Preparation and characterization of drug loaded nanoparticles

Adriamycin was chemically conjugated onto the PAH block of terpolymer mPEG-b-PDMAPMA-b-PAH via a hydrazone linkage. In a typical process, mPEG-b-PDMAPMA-b-PAH (40 mg) was dissolved in 4 mL of dimethyl sulfoxide (DMSO), and an adriamycin hydrochloride solution (10 mg in 1 mL of DMSO) was added. After 24 h of reaction in the dark at room temperature, the system was dialyzed against phosphate buffer saline (PBS) for 2 days using a dialysis bag (MWCO 3.5 kDa). After lyophilization, the adriamycin-conjugated (ADR-conjugated) terpolymer was obtained. For the determination of drug loading content in the drug-loaded terpolymer, the amount of adriamycin in dialysis solution was measured by a fluorescence spectrometer (LS55, Perkin Elmer, USA) with excitation at 480 nm and emission at 588 nm according to an adriamycin calibration curve.

High performance liquid chromatography (HPLC) analysis was performed on a Shimadzu LC-10AVP HPLC system to determine the form of adriamycin molecules existing in the drug-loaded terpolymer nanoparticles (ADR-NPs). The applied column was Shim-pack VP-ODS (150 × 4.6 mm, 5 μm). The mobile phase consisted of CH₃OH and H₂O (4.5 : 5.5, v/v) with a flow rate of 0.8 mL min⁻¹. UV detection wavelength was 490 nm. In addition, Fourier transform infrared (FTIR) spectroscopy was also used to characterize the chemical conjugation of adriamycin onto the terpolymer on an IR Prestige-21 FTIR spectrometer (Shimadzu, Japan) using the KBr disc method.

The critical micelle concentration (CMC) of the ADR-conjugated terpolymer was measured using pyrene fluorescent probe method on the fluorescence spectrometer. The concentration of ADR-conjugated terpolymer varied from 2.5 × 10⁻⁵ to 0.05 mg mL⁻¹ and the concentration of pyrene was fixed at 6 × 10⁻⁷mol L⁻¹. The excitation wavelength was adjusted to 339 nm, and the emission fluorescence intensities at 373 and 384 nm were recorded. The CMC was estimated as the cross-point of a decrease in the intensity ratio of the peaks at 373 and 384 nm plotted versus the logarithm of ADR-conjugated terpolymer concentrations.

The ADR-NPs were prepared by the probe sonication method and their morphology, size, and zeta potential were characterized. Briefly, a predetermined amount (2 mg) of adriamycin-loaded terpolymer was suspended in 5 mL of pure water and sonicated for 10 min in ice bath using a probe-type ultrasonicator (JY 92-2D; Ningbo Scientz Biotechnology Co., Ltd., Ningbo, China). The morphology of nanoparticles was observed on a transmission electron microscope (TEM, JEM-2000CX, JEOL Ltd. Inc., Japan). The particle size and zeta potential of nanoparticles in pure water were measured by a dynamic light scattering (DLS) zetasizer instrument (Nano ZS-90, Malvern Instruments Ltd, UK).

2.4 Preparation of ADR-NPs/siRNA complexes

P-gp siRNA was complexed with the PDMAPMA block of the terpolymer through electrostatic interactions. The siRNA binding ability of ADR-NPs was studied by agarose gel retardation assay. In brief, a predetermined amount (0.2 μg) of siRNA was mixed with various amounts (0–6.4 μg) of adriamycin-loaded terpolymer in aqueous solutions to obtain carrier/siRNA (N/P) ratios of 0–32. 10 μL of the complexes was mixed with 2 μL of 6× loading buffer and electrophoresed by a 1% (w/v) agarose gel containing ethidium bromide (0.25 μg mL⁻¹ of the gel) at 80 V for 30 min in TAE buffer solution (40 mM Tris–HCl, 1 v/v% acetic acid, and 1 mM EDTA). The bands of RNA in gel were detected by UV illuminator (254 nm), and the photographs were captured in a Bio-Rad imaging system (Hercules, CA). These results were used to calculate the threshold N/P ratio for subsequent experiments. The threshold is defined as the lowest N/P ratio value that retards siRNA migration in the gel. To determine the influence of adriamycin loading, the gel electrophoresis assay of adriamycin-free terpolymer was also performed. The size distributions and zeta potentials of the ADR-NPs/siRNA complexes at different N/P ratios were measured in triplicate using DLS. The morphology of the complexes was observed by TEM.

2.5 Release of adriamycin and siRNA from the ADR-NPs/siRNA complexes

In vitro release of the payloads from the ADR-NPs/siRNA complexes was studied as follows: lyophilized samples (5 mg each) were re-suspended in 5 mL of PBS and then transferred into different dialysis bags with a molecular weight cut-off of 3.5 kDa for adriamycin and 50 kDa for FAM–siRNA. The dialysis bags were immersed in 47 mL of buffered solution (with or without 10 mM glutathione (GSH) at pH 5.0 and pH 7.4). Release study was performed for 144 h at 37 °C under horizontal shaking (180 rpm). At predetermined time intervals, 2 mL of release medium was withdrawn for fluorescence measurement and an equivalent volume of fresh release medium was added. The fluorescence intensities of adriamycin and FAM–siRNA were recorded at excitation/emission wavelengths of 480/590 nm and 488/520 nm, respectively. In the assessment of drug release behavior, the cumulative amount of released payload was calculated using a standard calibration curve (0.01–2 μg mL⁻¹ for adriamycin and 1–200 nM for FAM–siRNA), and the percentages of drug released were plotted against time.

2.6 Cell culture

MCF-7/ADR (multidrug resistant human breast cancer cell line) cells were obtained from the Medical Center of Xi'an Jiaotong University (China), and maintained in 25 cm² cell culture flasks in which the cells were passaged at 70–80% confluence every 2–3 days. The MCF-7/ADR cells were cultured in DMEM medium containing 10% FBS, 100 U mL⁻¹ penicillin and 100 μg mL⁻¹ streptomycin in a humidified incubator at 37 °C with 5% CO₂.

2.7 CLSM observation

For cellular uptake studies by laser confocal microscopy, 5 × 10⁴ MCF-7/ADR cells were seeded into a CLSM dish and incubated overnight at 37 °C. The ADR-NPs/siRNA complexes (equivalent concentrations: 4 μg mL⁻¹ adriamycin and 200 nM FAM–siRNA) were added to the CLSM dishes. After incubation for different time intervals (4 or 24 h), the medium was discarded, and the cells were rinsed with PBS three times, fixed with 4% paraformaldehyde for 20 min and stained with DAPI solution for 5 min. The samples were observed using a confocal laser scanning microscope (Olympus Fluoview FV1000, Japan). The fluorescence signals of the adriamycin and FAM–siRNA were excited at 488 nm and measured at emission ranges of 580–610 nm and 500–530 nm, respectively.

2.8 Flow cytometry analysis

Dual-color flow cytometry was used to characterize the simultaneous cellular uptake of adriamycin and FAM–siRNA by MCF-7/ADR cells. MCF-7/ADR cells were seeded in a 6-well plate at 2.5 × 10⁵ per well and incubated overnight. The cells were treated with media containing the terpolymer mPEG-b-PDMAPMA-b-PAH, free adriamycin, ADR-NPs, NPs/siRNA complexes or ADR-NPs/siRNA complexes (equivalent concentrations: 4 μg mL⁻¹ adriamycin and 200 nM FAM–siRNA) for 3 h at 37 °C. Cells were collected, washed with cold PBS and re-suspended in 500 μL of PBS. Samples were examined on flow cytometer (BD Biosciences, USA) equipped with a 488 nm argon laser for excitation. The fluorescence intensities of adriamycin and FAM–siRNA were measured at FL1 band-pass emission (581 ± 21 nm) for the red adriamycin or FL2 band-pass emission (530 ± 20 nm) for the green FAM–siRNA. FlowJo software was used to analyze the data.

2.9 In vitro apoptosis detection

Detection of apoptotic cells was performed by flow cytometry measurement using Annexin V-FITC apoptosis detection kit. The terpolymer, free adriamycin, ADR-NPs or ADR-NPs/siRNA complexes (equivalent concentrations: 2 μg mL⁻¹ adriamycin and 100 nM siRNA) were added to wells and MCF-7/ADR cells were cultured for 48 h. The cells were harvested and stained using Annexin V-FITC apoptosis detection kit. Data analysis was performed using FlowJo software.

2.10 In vitro cytotoxicity measurements

The cytotoxicity induced by ADR-conjugated terpolymer in the presence or absence of P-gp siRNA was assayed using the MTT assay. MCF-7/ADR cells were seeded into 96-well plates at a density of 4000 cells per well and then incubated overnight at 37 °C. After the culture medium in each well had been replaced with 200 μL of formulation solution (equivalent terpolymer concentrations: 0.1, 1.0, 2.0, 5.0, 10.0 and 20.0 μg mL⁻¹), the cells were then incubated for 48 h. Afterwards, 150 μL of the medium from each well was replaced with same volume of fresh medium. 20 μL of MTT solution (5 mg mL⁻¹) was added. The cells were incubated for a further 4 h in the dark, and the medium was replaced with 150 μL of DMSO to dissolve the dark-blue formazan crystals. After gentle agitation for 10 min, the absorbance at 570 nm of each well was recorded on a microplate spectrophotometer (Bio-Rad 60, USA). Cell viability was calculated according to the following equation:

2.11 Statistical analysis

Data are presented as the mean values ± standard deviation of triplicate measurements for each experiment. The differences between the mean values were analyzed using Student's t-test. The results were considered statistically significant at p < 0.05.

3. Results and discussion

3.1 Synthesis of a terpolymer mPEG-b-PDMAPMA-b-PAH

To accomplish combination therapy of cancer, a novel terpolymer mPEG-b-PDMAPMA-b-PAH was developed to simultaneously deliver adriamycin and P-gp siRNA into breast cancer cells to achieve a synergistic effect. As described in Fig. 1, mPEG-b-PDMAPMA-b-PAH was synthesized by sequential RAFT polymerization of DMAPMA and Boc-acrylhydrazine followed by Boc-deprotection. In detail, a macro-RAFT agent mPEG–DDACT was firstly synthesized by conjugating the carboxyl group of DDACT to the primary amino end group of mPEG–Cys. After this, mPEG-b-PDMAPMA-b-P(Boc-AH) was synthesized via sequential RAFT polymerization of DMAPMA and Boc-acrylhydrazine in the presence of mPEG–DDACT macro-RAFT agent. Finally, the protecting Boc groups of mPEG-b-PDMAPMA-b-P(Boc-AH) were deprotected using acidolysis, resulting in a terpolymer mPEG-b-PDMAPMA-b-PAH. The terpolymer could chemically conjugate adriamycin onto the PAH block via a hydrazone bond and simultaneously complex siRNA with the PDMAPMA block through electrostatic interactions.

Fig. 1 Synthesis route of terpolymer mPEG-b-PDMAPMA-b-PAH.

The as-synthesized mPEG–DDACT, mPEG-b-PDMAPMA, mPEG-b-PDMAPMA-b-P(Boc-AH) and mPEG-b-PDMAPMA-b-PAH were characterized using ¹H NMR, and the results are showed in Fig. 2. As indicated in the ¹H NMR spectrum in Fig. 2A, the successful synthesis of the mPEG–DDACT macro-RAFT agent was confirmed by both the appearance of chemical shifts at 0.89 (g), 1.26 (g) and 1.7 ppm (f) corresponding to the protons of DDACT²⁹ and the characteristic signal at 3.67 ppm (b) ascribed to the methylene protons of mPEG. As shown by the ¹H NMR spectrum in Fig. 2B, the chemical shifts at 0.98 (c), 1.94 (d), 2.28–2.4 (e and g) and 3.2 ppm (f) were ascribed to the DMAPMA blocks,³⁰ and the peak at 3.67 ppm (b) was due to the methylene protons of mPEG. The number-average molecular weight (M_n) of the diblock copolymer mPEG-b-PDMAPMA determined by GPC was 8460 g mol⁻¹, which was consistent with its theoretical value (7100 g mol⁻¹) (Table 1). These results confirmed the synthesis of mPEG-b-PDMAPMA. In the ¹H NMR spectrum of mPEG-b-PDMAPMA-b-P(Boc-AH) in Fig. 2C, in addition to the characteristic peaks of the mPEG and PDMAPMA blocks, some new chemical shifts at 1.49 (j), 1.79 (h) and 3.74 ppm (i) attributed to the P(Boc-AH) blocks were also observed, indicating the successful polymerization of P(Boc-AH). At the same time, the M_n (19 370) of the terpolymer mPEG-b-PDMAPMA-b-P(Boc-AH) was much higher than that of the diblock copolymer mPEG-b-PDMAPMA (Table 1). These results suggested the successful synthesis of the terpolymer mPEG-b-PDMAPMA-b-P(Boc-AH). As showed in Fig. 2D, the absence of the peak at 1.49 ppm due to the protons of Boc groups revealed the complete deprotection of the Boc groups. An obvious decrease in the M_n after deprotection (Table 1) provided an experimental verification of the Boc-deprotection. These results demonstrated the successful synthesis of the terpolymer mPEG-b-PDMAPMA-b-PAH.

Fig. 2 ¹H NMR spectra of (A) mPEG–DDACT in CDCl₃, (B) mPEG-b-PDMAPMA in CDCl₃, (C) mPEG-b-PDMAPMA-b-P(Boc-AH) in CDCl₃ and (D) mPEG-b-PDMAPMA-b-PAH in DMSO-d₆.

Table 1 GPC data of the copolymers synthesized by RAFT polymerization

Copolymer	Mn,theoretical (g mol^−1)	Mn,GPC (g mol^−1)	PDIGPC
mPEG-b-PDMAPMA	7100	8460	1.02
mPEG-b-PDMAPMA-b-P(Boc-AH)	16 400	19 370	1.11
mPEG-b-PDMAPMA-b-PAH	11 400	13 650	1.05

---

3.2 Preparation and characterization of ADR-NPs

The terpolymer mPEG-b-PDMAPMA-b-PAH could chemically conjugate adriamycin via a hydrazone linkage to obtain an amphiphilic copolymer. The formation of ADR-conjugated terpolymer was verified by comparing the FTIR spectra of mPEG-b-PDMAPMA-b-PAH terpolymer before and after conjugation with adriamycin (Fig. 3A). The appearance of a characteristic absorption peak of adriamycin at 1070 cm⁻¹ suggested that adriamycin molecules were chemically conjugated onto the PAH blocks.^31,32 To verify this result, HPLC analysis was employed to reveal the form of adriamycin existing in the ADR-conjugated terpolymer by separating and determining the amount of adriamycin unconjugated onto the terpolymer. The results of HPLC analysis (Fig. 3B) showed that the ADR-conjugated terpolymer eluted at 7.2 min and the free adriamycin eluted at 10.5 min. The content of free adriamycin absorbed in the ADR-conjugated terpolymer was determined to be less than 1%. These results confirmed the successful chemical conjugation of adriamycin onto the terpolymer mPEG-b-PDMAPMA-b-PAH via hydrazone bonds. The drug loading content and loading efficiency of adriamycin calculated from fluorescence measurement were 14.9% and 69.8%, respectively.

Fig. 3 (A) FTIR spectra of (a) mPEG-b-PDMAPMA-b-PAH and (b) ADR-conjugated mPEG-b-PDMAPMA-b-PAH. (B) HPLC trace of ADR-conjugated mPEG-b-PDMAPMA-b-PAH.

The CMC of the ADR-conjugated terpolymer was determined using a pyrene fluorescence method. As showed in Fig. 4A, the CMC of ADR-NPs was calculated to be approximately 1.2 mg L⁻¹ which is significantly lower than that of PEG-based amphiphilic linear polymers reported in the literature.³³ This result suggested that the adriamycin-loaded terpolymer could keep stable cationic micelle structure at relatively low concentrations, which is desirable for the systemic drug delivery. The particle size and zeta potential of ADR-NPs were characterized by TEM and DLS measurements. As showed in Fig. 4B, the ADR-NPs exhibited a spherical morphology and had an average diameter of 202 ± 6.2 nm. The difference in the average size of the ADR-NPs between the DLS and TEM results was caused by the volumetric shrinkage of the ADR-NPs during the TEM sample preparation, which was also observed in the literature.³⁴ The zeta potential of the ADR-NPs was +36.3 ± 1.6 mV. Such a positive zeta potential is suitable for ADR-NPs to condense the negatively charged siRNA to form ADR-NPs/siRNA complexes.

Fig. 4 (A) CMC of ADR-conjugated mPEG-b-PDMAPMA-b-P(AH) determined by fluorescence spectroscopy using pyrene as a fluorescent probe. (B) Particle size distribution and morphology of ADR-NPs as determined by DLS and TEM.

3.3 Formation and characterization ADR-NPs/siRNA complexes

To estimate the ability of cationic ADR-NPs to condense siRNA, gel retardation assay was carried out at different N/P ratios, using adriamycin-free terpolymer as a control. As showed in Fig. 5A and B, the gel retardation assay demonstrated that siRNA could be effectively condensed by the terpolymer or ADR-NPs at low N/P ratios of 1 and 2, respectively, indicating the full neutralization of negatively charged siRNA. Notably, the conjugation of adriamycin onto the terpolymer did not fundamentally affect the siRNA loading capacity, because the complexation of siRNA with the ADR-NPs was mainly ascribed to electrostatic interactions while the adriamycin molecules were loaded in a site-isolated state.

Fig. 5 Electrophoretic mobility of free siRNA and siRNA complexed with (A) terpolymer and (B) ADR-NPs at different N/P ratios. (C) Sizes and zeta potentials of ADR-NPs/siRNA complexes at various N/P ratios. (D) Particle size distribution and morphology of ADR-NPs/siRNA complexes.

Generally, the zeta potential and size of the nanocarriers are known as crucial parameters that may determine the cellular uptake level of nanocarriers for drug delivery. Small particle size is in favor of easy entry of nanocarriers into cancer cells by endocytosis via the enhanced permeability and retention effect.³⁵ The size and zeta potential of the ADR-NPs/siRNA complexes were analyzed using DLS over a wide range of N/P ratios. As showed in Fig. 5C, the zeta potential of the ADR-NPs/siRNA complexes significantly increased from +21.3 to +35.8 mV as the N/P ratio increased from 2 to 16. This result is apparently due to the electrostatic neutralization of the positively charged NPs by the negatively charged siRNA. The particle size increased from 106 to 194 nm with the increasing N/P ratio. The increase in the complex size might be due to the formation of a more compact structure because the siRNA binding made the complexes be more compressed. It has been reported that 200 nm (particle size) is a rough upper limit for cellular uptake by phagocytosis,³⁶ and the nanoparticles bigger than this size might be excluded from cellular internalization altogether. Therefore, the nanoparticles with a less size and a higher positive zeta potential would more facilitate their cellular internalization. In this study, an N/P ratio of 4 for the ADR-NPs/siRNA complexes was chosen for subsequent in vitro cellular uptake and cytotoxicity experiments. As showed in Fig. 5D, the ADR-NPs/siRNA complexes exhibited a uniformly spherical morphology and had an average particle size of 144 ± 5.4 nm at a fixed N/P ratio of 4. The ADR-NPs/siRNA complexes would be internalized via endocytosis more easily than the ADR-NPs due to their smaller size.

3.4 Release of adriamycin and siRNA from ADR-NPs/siRNA complexes

To validate the stimuli-responsive release manner of payloads from the terpolymeric vehicles, the release experiments were carried out in four different solutions (pH 5.0 and 7.4, containing or not 10 mM GSH) which reflected the mildly acidic condition and reductive potential of the intracellular environment (pH 4–6, 2–10 mM GSH). As showed in Fig. 6, the release profiles revealed that the release of adriamycin and siRNA from the ADR-NPs/siRNA complexes exhibited pH-dependence. The release rates of the payloads increased with decreasing pH in both the presence and absence of GSH, namely the release rates of adriamycin and siRNA at pH 5.0 were higher than those at pH 7.4. This may be due to both the acidic hydrolysis of the hydrazone linkage and the protonation of PDMAPMA in an acidic environment. This phenomenon was also observed in other systems in which adriamycin was conjugated via a hydrazone linkage.^37,38 Furthermore, the co-delivery system also exhibited a reduction-responsive release behavior, namely adriamycin and siRNA were released more rapidly in the presence of GSH than in the absence of GSH. For instance, the percentages (46% and 29%) of adriamycin and siRNA released within the first 12 h under reductive conditions (pH 5.0, GSH 10 mM) were much higher than those (32% and 12%) at pH 5.0. The reduction-responsive release behavior could be ascribed to the GSH-triggered PEG detachment, which was also found in other PEGylated vehicles based on disulfide linkages.³⁹ These results suggested that the adriamycin/siRNA co-loaded terpolymeric vehicle would be stable under the physiological conditions, but achieve rapid payload release in the environment inside the cancer cells.

Fig. 6 In vitro accumulative release profiles of adriamycin and FAM–siRNA from ADR-NPs/siRNA complexes in the (A) absence or (B) presence of 10 mM GSH at pH 5.0 and 7.4.

3.5 Confocal microscopy study

To demonstrate the co-delivery of adriamycin and siRNA, MCF-7/ADR cells were incubated with the ADR-NPs/siRNA complexes and observed by confocal laser scanning microscopy. As showed in Fig. 7A, the red fluorescence originated from adriamycin and the green fluorescence came from FAM–siRNA. The cell nuclei were stained blue with DAPI. After 4 h of incubation, both red and green fluorescence was observed in MCF-7/ADR cells, suggesting that the ADR-NPs/siRNA complexes could be internalized by the MCF-7/ADR cells. Importantly, the red fluorescence started to separate from the green fluorescence, indicating that the payloads had been released from the ADR-NPs/siRNA complexes. After 24 h of incubation, both the red and green fluorescence in the cells became stronger. Notably, the strong fluorescence from adriamycin appeared in the nuclei and the green fluorescence became more diffusely distributed, indicating that the adriamycin and siRNA were released from the ADR-NPs/siRNA complexes in the intracellular environment. It is well known that the accumulation of adriamycin in the nuclei is essential for its cell-killing activity. In contrast, as showed in Fig. 7B, no green fluorescence was detected in the MCF-7/ADR cells treated with a mixture of free adriamycin and naked siRNA while red fluorescence of adriamycin was detected. It is because the naked negative FAM–siRNA can't cross the negative cell membrane to enter cancer cells.³

Fig. 7 Confocal microscope images of MCF-7/ADR cells incubated with (A) ADR-NPs/siRNA complexes and (B) mixture of free adriamycin and naked siRNA for 4 and 24 h. Scale bars = 30 μm.

3.6 Cellular uptake studies using flow cytometry

Two-color flow cytometry was also used to reveal the ability of the ADR-NPs/siRNA complexes to simultaneously deliver adriamycin and siRNA into MCF-7/ADR cells. As showed in Fig. 8, the cells incubated with the ADR-NPs and terpolymer/siRNA complexes emitted red and green fluorescence, respectively. A portion (8.81%) of the MCF-7/ADR cells treated with the ADR-NPs/siRNA complexes was both green and red fluorescence positive, demonstrating that the ADR-NPs/siRNA complexes could co-deliver ADR and FAM–siRNA into the MCF-7/ADR cell.

Fig. 8 Cellular uptake of different formations in MCF-7/ADR cells after 3 h of incubation, as determined by flow cytometry. (A) Terpolymer group, (B) free adriamycin group, (C) ADR-NPs group, (D) terpolymer/siRNA complexes group, (E) ADR-NPs/siRNA complexes group, (dose: 4 μg mL⁻¹ adriamycin and 200 nM FAM–siRNA).

3.7 Annexin V-FITC apoptosis assay

To investigate the cytotoxicity of adriamycin in different transportation modalities and the synergistic effect of adriamycin and P-gp siRNA, we used Annexin V-FITC apoptosis detection kit to detect the apoptosis of MCF-7/ADR cells by flow cytometry. In these formulations, the equivalent concentrations of adriamycin and P-gp siRNA were 2 μg mL⁻¹ and 100 nM, respectively. The flow cytometry data (Fig. 9) showed a rank order in efficacy of different formulations on apoptosis as follows: blank, free adriamycin, ADR-NPs and ADR-NPs/siRNA complexes. As expected, the percentage of cell growth inhibition of the ADR-NPs/siRNA complexes was higher than those of free adriamycin and ARD-NPs. The ADR-NPs/siRNA complexes group exhibited a lower percentage of live cells (43.1%) than ADR-NPs group (59.0%) and free adriamycin group (54.2%). The co-delivery of adriamycin and P-gp siRNA allowed adriamycin to bypass the P-gp efflux pump and increased the intracellular drug concentration, resulting in enhanced cell death and apoptosis.⁴⁰ Furthermore, free adriamycin group exhibited a slightly higher cytotoxicity than ADR-NPs group. The phenomenon can be explained by the fact that free adriamycin is a hydrophilic and can be readily transported into cells by passive diffusion.⁴¹ In contrast, the gradual release of adriamycin from the ADR-NPs decreased its cytotoxicity. Although ADR-conjugated NPs emitted red fluorescence, their fluorescence was not detected at a low concentration (2 μg mL⁻¹). These results demonstrated that the ADR-NPs/siRNA complexes could effectively co-deliver adriamycin and P-gp siRNA into the MCF-7/ADR cells and enhance the therapy efficacy via a synergistic effect.

Fig. 9 Apoptosis and cell death of MCF-7/ADR cells treated with different formulations. (A) Control group, (B) free adriamycin group, (C) ADR-NPs group, (D) ADR-NPs/siRNA complexes group (equivalent concentrations: 2 μg mL⁻¹ adriamycin and 100 nM siRNA).

3.8 In vitro cytotoxicity

It is well known that the clinical application of chemotherapeutic agents is often hindered by their intrinsic or acquired multidrug resistance (MDR) in cancer cells. This resistance increases drug efflux, resulting in a decreased concentration of cytotoxic agents and reduced apoptotic activity in cancer cells. P-gp, the product of the MDR1 gene, is able to decrease intracellular drug accumulation through the active transport driven by adenosine triphosphate. To evaluate the synergistic effect of co-delivered adriamycin and P-gp siRNA, the in vitro cytotoxicity of the terpolymer, NPs/siRNA complexes, ADR-NPs and ADR-NPs/siRNA complexes (ADR-loading content 14.9%, N/P ratio = 4) was evaluated in MCF-7/ADR cells by the MTT assay after incubation for 48 h. As showed in Fig. 10, both ADR-NPs and ADR-NPs/siRNA complexes inhibited cell growth in a dose-dependent manner, and the ADR-NPs/siRNA complexes were more cytotoxic than the ADR-NPs. For the ADR-NPs/siRNA complexes and ADR-NPs, the concentration at which 50% of MCF-7/ADR cells were killed (IC₅₀) was determined to be 3 and 10 μg mL⁻¹, respectively. The IC₅₀ of the ADR-NPs/siRNA complexes was about 3.3 times lower than that of the ADR-NPs. The results showed that the simultaneous delivery of adriamycin and P-gp siRNA into tumor cells could effectively enhance the therapy efficiency of adriamycin via a synergistic effect in comparison to adriamycin delivery alone. The result is consistent with that previously reported.⁴² Under the same condition, no significant cytotoxicity against the MCF-7/ADR cells was observed for the terpolymer mPEG-b-PDMAPMA-b-PAH and NPs/siRNA complexes, reflecting that the terpolymer showed good biocompatibility because of the PEG shielding.⁴³ In a word, all the results demonstrated that the co-delivery of adriamycin and siRNA based on the multifunctional polymeric vehicle would be a promising combination therapeutic strategy for enhanced anti-tumor therapy.

Fig. 10 In vitro cytotoxicity of different formulations with different concentrations against MCF-7/ADR cells obtained by the MTT assay. Cell viabilities of MCF-7/ADR cells treated with the terpolymer, NPs/siRNA complexes, ADR-NPs and ADR-NPs/siRNA complexes for 48 h.

4. Conclusions

In this study, a novel multifunctional terpolymer mPEG-b-PDMAPMA-b-PAH was designed and synthesized in order to co-deliver adriamycin and P-gp siRNA into breast cancer cells. The terpolymer could conjugate adriamycin via a hydrazone linkage and simultaneously complex P-gp siRNA through electrostatic interactions. The resultant ADR-NPs/siRNA complexes had a narrow size distribution and showed a spherical morphology in aqueous solutions. The in vitro release kinetics of adriamycin and P-gp siRNA from the ADR-NPs/siRNA complexes exhibited a pH and reduction dual-dependent fashion. The terpolymer could efficiently co-deliver adriamycin and P-gp siRNA into the MCF-7/ADR breast cancer cells. In comparison with the adriamycin-loaded terpolymer, the co-delivery system exhibited an enhanced cytotoxicity via a synergistic effect. Therefore, the terpolymeric vehicle would be a promising combination therapeutic strategy for enhanced anti-tumor therapy.

Acknowledgements

The authors gratefully acknowledge financial supports from the National Natural Science Foundation of China (50603020 and 50773062), the Project of Natural Science Foundation of Shaanxi Province, China (2013K09-27) and the Fundamental Research Funds for the Central Universities (XJJ2014124 and XJJ2012146).

Notes and references

1. F. Li, H. T. Zhang, C. H. Gu, L. Fan, Y. B. Qiao, Y. C. Tao, C. Cheng, H. Wu and J. Yi, Polym. Int., 2013, 62, 165–171.
2. Y. B. Patil, S. K. Swaminathan, T. Sadhukha, L. Ma and J. Panyam, Biomaterials, 2010, 31, 358–365.
3. K. A. Whitehead, R. Langer and D. G. Anderson, Nat. Rev. Drug Discovery, 2009, 8, 129–138.
4. M. Saad, O. B. Garbuzenko and T. Minko, Nanomedicine, 2008, 3, 761–776.
5. M. Creixell and N. A. Peppas, Nano Today, 2012, 7, 367–379.
6. P. F. Liu, H. Yu, Y. Sun, M. J. Zhu and Y. R. Duan, Biomaterials, 2012, 33, 4403–4412.
7. Y. H. Yu, E. Kim, D. E. Park, G. Shim, S. Lee, Y. B. Kim, C. W. Kim and Y. K. Oh, Eur. J. Pharm. Biopharm., 2012, 80, 268–273.
8. A. M. Chen, M. Zhang, D. G. Wei, D. Stueber, O. Taratula, T. Minko and H. X. He, Small, 2009, 5, 2673–2677.
9. H. Meng, M. Liong, T. Xia, Z. X. Li, Z. X. Ji, J. I. Zink and A. E. Nel, ACS Nano, 2010, 4, 4539–4550.
10. Y. C. Chen, S. R. Bathula, J. Li and L. Huang, J. Biol. Chem., 2010, 285, 22639–22650.
11. C. F. Zheng, M. B. Zheng, P. Gong, J. Z. Deng, H. Q. Yi, P. F. Zhang, Y. J. Zhang, P. Liu, Y. F. Ma and L. T. Cai, Biomaterials, 2013, 34, 3431–3438.
12. H. Wang, Y. Zhao, H. Wang, G. J. Nie and K. H. Nan, Curr. Drug Metab., 2012, 13, 1087–1096.
13. T. M. Sun, J. Z. Du, Y. D. Yao, C. Q. Mao, S. Dou, S. Y. Huang, P. Z. Zhang, K. W. Leong, E. W. Song and J. Wang, ACS Nano, 2011, 5, 1483–1494.
14. H. J. Wang, P. Q. Zhao, W. Y. Su, S. Wang, Z. Y. Liao, R. F. Niu and J. Chang, Biomaterials, 2010, 31, 8741–8748.
15. D. W. Dong, B. Xiang, W. Gao, Z. Z. Yang, J. Q. Li and X. R. Qi, Biomaterials, 2013, 34, 4849–4859.
16. X. W. Ding, Y. Liu, J. H. Li, Z. Luo, Y. Hu, B. L. Zhang, J. J. Liu, J. Zhou and K. Y. Cai, ACS Appl. Mater. Interfaces, 2014, 6, 7395–7407.
17. H. Y. Huang, W. T. Kuo, M. J. Chou and Y. Y. Huang, J. Biomed. Mater. Res., Part A, 2011, 97, 330–338.
18. D. Ma, H. B. Zhang, Y. Y. Chen, J. T. Lin and L. M. Zhang, J. Colloid Interface Sci., 2013, 405, 305–311.
19. W. Wei, P. P. Lv, X. M. Chen, Z. G. Yue, Q. Fu, S. Y. Liu, H. Yue and G. H. Ma, Biomaterials, 2013, 34, 3912–3923.
20. M. Han, Q. Lv, X. J. Tang, Y. L. Hu, D. H. Xu, F. Z. Li, W. Q. Liang and J. Q. Gao, J. Controlled Release, 2012, 163, 136–144.
21. C. H. Zhu, S. Y. Jung, G. Y. Si, R. Cheng, F. H. Meng, X. L. Zhu, T. G. Park and Z. Y. Zhong, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 2869–2877.
22. M. Ahmed and R. Narain, Prog. Polym. Sci., 2013, 38, 767–790.
23. J. L. Zhu, H. Cheng, Y. Jin, S. X. Cheng, X. Z. Zhang and R. X. Zhuo, J. Mater. Chem., 2008, 18, 4433–4441.
24. S. Kirkland-York, Y. L. Zhang, A. E. Smith, A. W. York, F. Q. Huang and C. L. McCormick, Biomacromolecules, 2010, 11, 1052–1059.
25. N. Cao, D. Cheng, S. Y. Zou, H. Ai, J. M. Gao and X. T. Shuai, Biomaterials, 2011, 32, 2222–2232.
26. S. Takae, K. Miyata, M. Oba, T. Ishii, N. Nishiyama, K. Itaka, Y. Yamasaki, H. Koyama and K. Kataoka, J. Am. Chem. Soc., 2008, 130, 6001–6009.
27. Y. A. Vasilieva, D. B. Thomas, C. W. Scales and C. L. McCormick, Macromolecules, 2004, 37, 2728–2737.
28. Y. Teramura, Y. Kaneda and H. Iwata, Biomaterials, 2007, 28, 4818–4825.
29. J. Rieger, C. Grazon, B. Charleux, D. Alaimo and C. Jerome, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 2373–2390.
30. A. Das, S. Ghosh and A. R. Ray, Polymer, 2011, 52, 3800–3810.
31. S. Kayal and R. V. Ramanujan, Mater. Sci. Eng., C, 2010, 30, 484–490.
32. A. Lanz-Landazuri, A. Martinez de Ilarduya, M. Garcia-Alvarez and S. Munoz-Guerra, React. Funct. Polym., 2014, 81, 45–53.
33. H. L. Sun, B. N. Guo, R. Cheng, F. H. Meng, H. Y. Liu and Z. Y. Zhong, Biomaterials, 2009, 30, 6358–6366.
34. L. P. Qiu, Z. Li, M. X. Qiao, M. M. Long, M. Y. Wang, X. J. Zhang, C. M. Tian and D. W. Chen, Acta Biomater., 2014, 10, 2024–2035.
35. H. Maeda, J. Wu, T. Sawa, Y. Matsumura and K. Hori, J. Controlled Release, 2000, 65, 271–284.
36. V. P. Torchilin, Pharm. Res., 2007, 24, 1–16.
37. M. Prabaharan, J. J. Grailer, S. Pilla, D. A. Steeber and S. Q. Gong, Biomaterials, 2009, 30, 5757–5766.
38. S. H. Liu, Y. B. Guo, R. Q. Huang, J. F. Li, S. X. Huang, Y. Y. Kuang, L. Han and C. Jiang, Biomaterials, 2012, 33, 4907–4916.
39. J. Q. Zhao, J. J. Liu, S. X. Xu, J. H. Zhou, S. C. Han, L. D. Deng, J. H. Zhang, J. F. Liu, A. M. Meng and A. J. Dong, ACS Appl. Mater. Interfaces, 2013, 5, 13216–13226.
40. J. M. Li, Y. Y. Wang, M. X. Zhao, C. P. Tan, Y. Q. Li, X. Y. Le, L. N. Ji and Z. W. Mao, Biomaterials, 2012, 33, 2780–2790.
41. C. X. Liu, F. X. Liu, L. X. Feng, M. Li, J. Zhang and N. Zhang, Biomaterials, 2013, 34, 2547–2564.
42. X. B. Xiong and A. Lavasanifar, ACS Nano, 2011, 5, 5202–5213.
43. M. L. Patil, M. Zhang and T. Minko, ACS Nano, 2011, 5, 1877–1887.

---