Biocompatible reduction-responsive polypeptide micelles as nanocarriers for enhanced chemotherapy efficacy in vitro

Abstract

To develop biocompatible reduction-responsive micellar systems for efficient intracellular drug delivery, disulfide-linked block copolymers of methoxyl poly(ethylene glycol) (mPEG) and poly(ε-benzyloxycarbonyl-L-lysine) (PZLL) were synthesized through ring-opening polymerization of ε-benzyloxycarbonyl-L-lysine N-carboxyanhydride with amino group terminated disulfide functionalized mPEG as macroinitiator. The copolymers self-assembled into micelles in phosphate buffered saline (PBS) at pH 7.4 through direct dissolution and dialysis methods. The micelles were revealed to have excellent hemocompatibilities, and cell and tissue compatibilities, which rendered them with potential for drug delivery applications. Doxorubicin (DOX), an anthracycline antitumor drug, was loaded into the micelles through nanoprecipitation with a drug loading efficiency of about 30 wt%. The in vitro DOX release from all DOX-loaded micelles was accelerated in PBS with 10.0 mM GSH, mimicking intracellular reductive conditions. DOX-loaded micelles showed higher cellular proliferation inhibition towards glutathione monoester pretreated HeLa (a human cervical cell line) and HepG2 cells (a human hepatoma cell line) as compared to unpretreated or buthionine sulfoximine pretreated cells. The above results indicated that the biocompatible reduction-responsive micelles have vast potential in targeted intracellular delivery of antitumor drugs to achieve enhanced efficacy in malignancy therapy.

---

1. Introduction

Presently, malignancy (i.e. cancer) is one of the most serious diseases around the world, which exhibits high morbidity and mortality.^1,2 In spite of the considerable development of various antitumor drugs, such as doxorubicin (DOX), paclitaxel, and chlorambucil, clinical outcomes have been disappointing due to their life-threatening side effects, including leukaemia, cardiotoxicity, etc.^3–5 To eliminate or reduce the severe side effects, many kinds of nanocarriers, such as micelles,^6,7 vesicles,^8,9 liposomes¹⁰ and nanogels,^2,11,12 have been developed as vehicles for delivery of antitumor drugs. Among all of the aforementioned nanovehicles, polymeric micelles self-assembled from amphiphilic block copolymers have emerged as one of the most promising nanocarriers for various antitumor drugs. Micelles are usually associated with several merits as polymeric drug carriers, such as improved drug solubility in water, prolonged circulation time, enhanced accumulation in tumor sites, decreased side effects, and elevated drug bioavailability and efficacy.^5,13

In recent years, stimuli-responsive micelles, which show sharp intelligent response to intracellular microenvironmental stimuli, such as pH,^14,15 redox,^16,17 temperature¹⁸ and enzymes,¹⁹ have attracted considerable attention. Just to reach the targeted tumor tissues, the smart nanocarriers can rapidly enter cells through endocytosis and quickly release the payload triggered by intracellular stimuli, resulting in aggressive activity toward tumor cells and maximal chemotherapeutic efficacy with fewer side effects.²⁰ In particular, reduction-responsive micelles containing disulfide bonds have been considered to be the most promising nanovehicles due to the difference in redox potential between the mildly oxidizing extracellular milieu and the reducing intracellular microenvironment. The disulfide bonds can be reductively degraded in the presence of glutathione (GSH), a thiol-containing tripeptide. The intracellular concentration of GSH is ∼0.5 to 10.0 mM and is significantly higher than that outside cells (∼2.0 to 20.0 μM).¹ The dramatic variation of GSH concentration provides an opportunity for designing intracellular specific drug delivery systems. Therefore, such smart micelles containing GSH-cleavable disulfide bonds may hold great potential for efficient intracellular delivery of antitumor drugs.

Very recently, several reports have appeared concerning reduction-responsive micelles formed from amphiphilic disulfide-linked block copolymers. Zhong's group prepared reduction-responsive micelles based on disulfide-linked poly(ethylene glycol) (PEG) or dextran and poly(ε-caprolactone) (PCL) block copolymers and applied them for rapid intracellular release of DOX.^13,21 Wang et al. developed a micellar system from disulfide-linked block copolymers of PCL and poly(ethyl ethylene phosphate) aimed at intracellular DOX release triggered by GSH in tumor cells.²² Li and co-workers exploited two kinds of reduction-responsive micelles originating from disulfide-linked PEG-polypeptides (i.e. poly(ε-benzyloxycarbonyl-L-lysine) (PZLL) and poly(rac-leucine)) to achieve intracellular delivery of DOX.^23,24

In this work, a type of biocompatible reduction-responsive micelle based on disulfide-linked methoxyl poly(ethylene glycol) (mPEG) and PZLL was developed for efficient intracellular delivery of antitumor drugs. The block copolymers were synthesized through ring-opening polymerization (ROP) of ε-benzyloxycarbonyl-L-lysine N-carboxyanhydride (ZLL NCA) with amino group terminated disulfide functionalized mPEG (mPEG-S₂-NH₂) as macroinitiator. As compared to previous work, the synthesis pathway for the block copolymers was more convenient.^23,24 The micelles of copolymers in phosphate buffered saline (PBS) at pH 7.4 were prepared through direct dissolution and dialysis methods. The micelles were revealed to have excellent hemocompatibilities, and cell and tissue compatibilities. DOX was loaded into the micelles through nanoprecipitation, and the in vitro DOX release was accelerated in intracellular reductive conditions. The DOX-loaded micelles showed higher cellular proliferation inhibition towards glutathione monoester (GSH-OEt) pretreated HeLa (a human cervical cell line) and HepG2 (a human hepatoma cell line) cells as compared to unpretreated or buthionine sulfoximine (BSO) pretreated cells. These reducible micelles are highly promising for the targeted intracellular delivery of antitumor drugs.

2. Experimental section

2.1. Materials

ZLL was purchased from GL Biochem Co., Ltd. ZLL NCA was synthesized as described in our previous work with slight modification.^25,26 mPEG (M_n = 5000) was purchased from Sigma-Aldrich without further purification. GSH, GSH-OEt, buthionine sulfoximine (BSO) and 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) were purchased from Sigma-Aldrich. Doxorubicin hydrochloride (DOX·HCl) was purchased from Zhejiang Hisun Pharmaceutical Co., Ltd. N,N-Dimethylformamide (DMF) was dried over CaH₂ at 25 °C for 48 h before vacuum distillation. All other reagents and solvents were purchased from Sinopharm Chemical Reagent Co., Ltd. and used as obtained.

2.2. Sample syntheses

The three-step procedure employed for the synthesis of methoxy poly(ethylene glycol)-S-S-poly(ε-benzyloxycarbonyl-L-lysine) (mPEG-S₂-PZLL) was shown in Scheme 1.

Scheme 1 Synthesis pathway for mPEG-S₂-PZLL.

2.2.1. Synthesis of mPEG-S₂-NH₂ macroinitiator. mPEG-S₂-NH₂ was synthesized through a ring-opening reaction of succinic anhydride (SA) and a subsequent condensation reaction. Briefly, mPEG (10.0 g, 2.0 mmol) and SA (0.24 g, 2.4 mmol) were dissolved in 100.0 mL of dichloromethane (DCM)–pyridine (9 : 1, v/v) mixed solvent, and the reaction proceeded at 25 °C for 24 h. Then, the solution was precipitated into excess diethyl ether. The obtained product was further washed twice with diethyl ether and dried under vacuum at 25 °C for 24 h, yielding carboxyl group terminated mPEG (mPEG–COOH, yield: 95.3%). Subsequently, mPEG–COOH (5.0 g, 0.98 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (EDC·HCl, 0.94 g, 4.9 mmol) and N-hydroxysuccinimide (NHS, 0.23 g, 1.96 mmol) were dissolved into 50.0 mL of dimethyl sulfoxide (DMSO) and were kept stirring at 25 °C for 12 h. Then, the above reaction mixture was added into a solution of cysteamine hydrochloride (4.42 g, 19.61 mmol) in 10.0 mL of water, dropwise. After the completion of the addition, the reaction proceeded at 25 °C for an additional 48 h. The solution was extracted with 90.0 mL of DCM 3 times (30.0 mL × 3). The DCM phase was dried, concentrated and then precipitated into excess diethyl ether. The final mPEG-S₂-NH₂ was further washed twice with diethyl ether and dried under vacuum at 25 °C for 24 h (yield: 92.7%).

2.2.2. Syntheses of disulfide-linked mPEG and PZLL (mPEG-S₂-PZLL) block copolymers. As shown in Scheme 1, mPEG-S₂-PZLL block copolymers were synthesized through ROP of ZLL NCA, initiated by mPEG-S₂-NH₂ in DMF. The different reactant feeding molar ratios were listed in Table 1. The reaction was conducted at 25 °C for 72 h. Then, the reaction solution was precipitated into an excess diethyl ether–ethanol (2 : 1, v/v) mixture. The obtained product was further washed twice with diethyl ether and dried under vacuum at 25 °C for 24 h (yield: 88.7%).

Table 1 Characterization of mPEG-S₂-PZLL block copolymers

Entry	Feeding molar ratio of ZLL NCA/mPEG-S2-NH2	DP of PZLL^a	M n (g mol^−1)	M n (g mol^−1)	PDI^b
a Determined by ^1H NMR. b Determined by GPC.
mPEG113-S2-PZLL18	20	18	10 000	11 900	1.09
mPEG113-S2-PZLL35	40	35	14 400	16 300	1.14

---

2.3. Preparation of micelles

Micelles were prepared by direct dissolution and dialysis methods.

2.3.1. Direct dissolution method. Typically, the micelles were prepared by directly dissolving 5.0 mg of mPEG-S₂-PZLL in 5.0 mL of PBS, marked as micelle₁.

2.3.2. Dialysis method. Briefly, 5.0 mg of mPEG-S₂-PZLL was dissolved in 1.0 mL of DMF, and the solution was allowed to stir at 25 °C for 2 h. Then, to this solution, 2.0 mL of PBS was added at a rate of 1.0 mL min⁻¹ under gentle stirring. After standing for 3 h at 25 °C, DMF was removed by dialysis using a dialysis bag (MWCO 3500 Da) against PBS for 24 h and the volume was set to the mark in a 5.0 mL volumetric flask to obtain the micelle solution, marked as micelle₂.

2.4. Characterization

¹H NMR spectra were recorded on a Bruker AV 400 NMR spectrometer in chloroform-d (CDCl₃) or trifluoroacetic acid-d (TFA-d). FT IR spectra were recorded on a Bio-Rad Win-IR instrument using the potassium bromide method. Number-, weight-average molecular weights (M_n, M_w) and molecular weight distributions (polydispersity index, PDI = M_w/M_n) were determined by gel permeation chromatography (GPC) using a series of linear TSKgel Super columns (AW3000 and AW5000) and a Waters 515 HPLC pump, with an OPTILAB DSP Interferometric Refractometer (Wyatt Technology) as the detector. The eluent was DMF containing 0.01 M lithium bromide at a flow rate of 1.0 mL min⁻¹ at 40 °C. Monodispersed polystyrene standards purchased from Waters Co., Ltd. with a molecular weight range of 1790–200 000 were used to generate the calibration curve. Transmission electron microscopy (TEM) measurements were performed on a JEOL JEM-1011 transmission electron microscope with an accelerating voltage of 100 kV. A drop of the micelle solution (0.05 g L⁻¹) was deposited onto a 230 mesh copper grid coated with carbon and allowed to dry in air at 25 °C before measurements. Dynamic laser scattering (DLS) measurements were performed on a WyattQELS instrument with a vertically polarized He–Ne laser (DAWN EOS, Wyatt Technology). The scattering angle was fixed at 90°.

2.5. Critical micellization concentration (CMC) measurements

CMC was measured by fluorescence spectroscopy using pyrene as probe on a PTI Fluorescence Master System with the software Felix 4.1.0. Typically, a predetermined amount of pyrene in acetone was added into a series of ampoules, and the acetone was removed first by gently flowing N₂ and then by vacuum. A predetermined volume of mPEG-S₂-PZLL solution and PBS were added into the ampoules, consecutively, to get solutions with different mPEG-S₂-PZLL concentrations ranging from 6.10 × 10⁻⁵ to 1.0 g L⁻¹, whereas the concentration of pyrene in each flask was fixed at 6.0 × 10⁻⁷ mol L⁻¹, slightly lower than the saturation solubility of pyrene in water. These solutions were shaken vigorously and then allowed to equilibrate at 25 °C for at least 24 h. The excitation spectra of pyrene with various mPEG-S₂-PZLL concentrations were measured at the detection emission wavelength (λ_em = 390 nm). The CMC value was obtained from the intersection of the tangent to the horizontal line of I₃₄₀/I₃₃₅ with relative constant value and the diagonal line with rapidly increased I₃₄₀/I₃₃₅ ratio.

2.6. Protein adsorption tests

Bovine serum albumin (BSA) was used as a model protein to determine the protein adsorption of mPEG-S₂-PZLL micelles. The micelles were incubated with a solution of BSA in PBS at pH 7.4, with the final concentrations of micelles at 0.125, 0.25 or 0.5 g L⁻¹, and BSA at 0.5 g L⁻¹, respectively. After incubation at 37 °C for a determined time, 1.0 mL of each sample was withdrawn after vortexing to ensure homogeneity, and then centrifuged at 16 000 g for 15 min to precipitate the protein-adsorbed micelles. The BSA concentration of the supernatant was determined using UV-Vis spectroscopy (UV-2401PC, Shimadzu) by measuring the maximal absorbance at 280 nm. Then, the amount of BSA adsorbed on the micelle was calculated against a standard calibration curve of BSA using a previously reported method.¹¹ mPEG₁₁₃ and PEI25k were used as negative and positive controls, respectively.

2.7. Hemolysis activity examination

The hemolysis properties of the mPEG-S₂-PZLL micelles were examined by spectrophotometry. Rabbit blood stabilized with dipotassium ethylene diamine tetraacetate was obtained from the Experimental Animal Center of Jilin University. To obtain red blood cells (RBCs), 5.0 mL of a rabbit blood sample was added to 10.0 mL of physiological saline (PS), and then the RBCs were isolated from serum by centrifugation at 1200 rpm for 8 min. The RBCs were further washed 6 times with PS (40.0 mL × 6). Following the last wash, the remaining RBCs were dispersed in 30.0 mL of PS. 0.4 mL of micelles at different concentrations were separately mixed with 0.4 mL of RBCs suspended in PS. The mixtures were then incubated at 37 °C in a thermostatted water bath for 2 h. Then, the RBCs were centrifuged at 3000 rpm for 10 min and 100.0 μL of the supernatant of each sample was transferred into a 96-well plate. Free hemoglobin in the supernatant was measured with a Bio-Rad 680 microplate reader at 540 nm. PS and triton X-100 (10.0 g L⁻¹) were used as negative and positive controls, respectively. All hemolysis experiments were carried out in triplicate. The hemolysis ratio (HR) of the RBCs was calculated using eqn (1):

In eqn (1), A_sample, A_{negitive control}, and A_{positive control} were denoted as the absorbances of the sample, negative and positive controls, respectively.

2.8. Tissue response evaluations

Five-week-old female BALB/c nude mice were obtained and housed from the Animal Holding Unit of Northeast Normal University. The animal procedures were approved by the School of Life Sciences Animal Care and Use Committee of Northeast Normal University. Micelles were dissolved in PBS (1.0 g L⁻¹, 10.0 μL) and injected into the tibialis anterior muscles in the mice. The muscles that received the injections were isolated one week later, fixed in 4% (w/v) paraformaldehyde, washed and embedded in paraffin. Tissue sections were cut into 8.0 μm thick slices, which were placed onto gelatin-coated slides and stained with hematoxylin and eosin (H&E) for histological examinations. The muscles without injection and receiving an intramuscular injection of 10.0 μL of PBS and PEI25k in PBS (1.0 g L⁻¹, 10.0 μL) were used as controls.

2.9. Drug loading and release

DOX was used as a model drug for drug loading. DOX-loaded mPEG-S₂-PZLL micelles were prepared by a nanoprecipitation technique (Scheme 2). Typically, mPEG-S₂-PZLL (20.0 mg), DOX·HCl (4.0 mg) and triethylamine (0.72 mg) were mixed in 4.0 mL of DMF. The solution was allowed to stand at 25 °C for 2 h. Then, 2.0 mL of deionized water was added dropwise to the solution under stirring. The mixture was stirred at 25 °C for 6 h and the organic solvent was removed by dialysis against deionized water for 24 h. The dialysis medium was refreshed 5 times and the whole procedure was performed in the dark. Then, the solution was filtered and lyophilized. For determination of drug loading content (DLC) and drug loading efficiency (DLE), the drug loaded micelle was dissolved in DMF and analyzed with fluorescence measurements using a standard curve method (λ_ex = 480 nm). The DLC and DLE of the drug loaded micelles were calculated according to eqn (2) and (3), respectively:

Scheme 2 Schematic illustration of DOX loading and targeted intracellular release.

The release profiles of the DOX-loaded micelles were investigated in PBS (pH 7.4) without or with 10.0 mM GSH. The weighed, freeze-dried DOX-loaded micelle was suspended in 3.0 mL of release medium and transferred into a dialysis bag (MWCO 3500 Da). The release experiment was initiated by placing the end-sealed dialysis bag into 50.0 mL of release medium at 37 °C with continuous shaking at 70 rpm. At predetermined intervals, 2.0 mL of external release medium was taken out and an equal volume of fresh release medium was replenished. The amount of released DOX was determined using fluorescence measurements (λ_ex = 480 nm). The release experiments were conducted in triplicate.

2.10. Intracellular drug release

The cellular uptake and intracellular release behaviors of the DOX-loaded micelles were observed by confocal laser scanning microscopy (CLSM) and flow cytometric analyses toward HeLa or HepG2 cells.

2.10.1. CLSM. The cells were seeded in 6-well plates at 2.0 × 10⁵ cells per well in 2.0 mL of complete Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum, supplemented with 50.0 IU mL⁻¹ penicillin and 50.0 IU mL⁻¹ streptomycin, and cultured for 24 h, and then treated with 0.5 mM BSO for 12 h or treated with 10.0 mM GSH-OEt for 2 h. Cells were washed by PBS and incubated at 37 °C for an additional 2 h with DOX-loaded micelles at a final DOX concentration of 10.0 mg L⁻¹ in complete DMEM. Cells without BSO and GSH-OEt pretreatment were used as control. Then, the culture medium was removed and cells were washed with PBS thrice. Thereafter, the cells were fixed with 4% (w/v) paraformaldehyde for 30 min at 25 °C, and the cells were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, blue) for the cell nuclei and Alexa Fluor® 488 phalloidin (Alexa 488, green) for the cytoskeleton, according to the standard protocols provided by the suppliers. CLSM images of the cells were obtained through confocal microscopy (Olympus FluoView 1000).

2.10.2. Flow cytometric analyses. The cells were seeded in 6-well plates at 2.0 × 10⁵ cells per well in 2.0 mL of complete DMEM, and cultured for 24 h, and then treated with 0.5 mM BSO for 12 h or treated with 10.0 mM GSH-OEt for 2 h. The cells were washed by PBS and incubated at 37 °C for an additional 2 h with the DOX-loaded micelles at a final DOX concentration of 10.0 mg L⁻¹ in complete DMEM. Cells without BSO and GSH-OEt pretreatment were used as control. Thereafter, the culture medium was removed and the cells were washed with PBS thrice and treated with trypsin. Then, 1.0 mL of PBS was added to each culture well, and the solutions were centrifuged for 4 min at 3000 rpm. After removal of supernatants, the cells were resuspended in 0.3 mL of PBS. Data for 1.0 × 10⁴ gated events were collected, and analysis was performed by flow cytometry (Beckman, California, USA).

2.11. Cell viability assays

The relative cytotoxicities of the micelles against HeLa and HepG2 cells were evaluated in vitro by a standard MTT assay. The cells were seeded in 96-well plates at 7000 cells per well in 100.0 μL of complete DMEM and incubated at 37 °C in a 5% CO₂ atmosphere for 24 h, followed by removing the culture medium and adding micelle solutions at different concentrations (0–0.5 g L⁻¹). The cells were subjected to MTT assay after being incubated for another 72 h. The absorbance of the solutions were measured on a Bio-Rad 680 microplate reader at 490 nm. Cell viability (%) was calculated based on eqn (4):

In eqn (4), A_sample and A_control represented the absorbances of the sample and control wells, respectively.

The cytotoxicities of the DOX-loaded micelles against HeLa and HepG2 cells were also evaluated in vitro by MTT assay. Similarly, cells were seeded into 96-well plates at 10 000 cells per well in 200.0 μL of complete DMEM and further incubated for 24 h. Then, the cells were treated with 0.5 mM BSO for 12 h or 10.0 mM GSH-OEt for 2 h. Cells without pretreatment were used as control. After washing the cells with PBS, the DOX-loaded micelles were added at different DOX concentrations (0–10.0 mg L⁻¹ DOX). The cells were subjected to MTT assay after being incubated for another 24, 48 or 72 h. The absorbance of the solutions was measured on a Bio-Rad 680 microplate reader at 490 nm. Cell viability (%) was calculated based on eqn (4).

3. Results and discussion

3.1. Synthesis and characterization of disulfide-linked block copolymers

Synthesis and polymerization of α-amino acid N-carboxyanhydrides were first reported by Hermann Leuchs in 1906.²⁷ Since then, these cyclic and highly reactive amino acid derivatives have been used for the formation of polypeptides by ROP with a primary amine initiator in a controlled manner.^28,29 In this work, the disulfide-linked mPEG-S₂-PZLL block copolymers were synthesized through ROP of ZLL NCA, initiated by the terminated primary amine group in the mPEG-S₂-NH₂ chain (Scheme 1). First, mPEG-S₂-NH₂ was synthesized through the ring-opening reaction of SA and a subsequent condensation reaction. The characteristic resonances at 3.29 and 3.18 ppm are attributed to the methylene protons neighboring the disulfide bond (–C(O)NHCH₂CH₂–S₂–CH₂CH₂NH₂, 2H and 2H) in the ¹H NMR spectrum (Fig. 1Ab), and the typical amide I and II bands at 1662 (υ_C=O) and 1546 cm⁻¹ (υ_C(O)–NH) observed from FT IR spectroscopy (Fig. 1Cb) confirmed the successful synthesis of mPEG-S₂-NH₂. The ZLL NCA reacted with mPEG-S₂-NH₂ to obtain the mPEG-S₂-PZLL block copolymer, whose chemical structure was also characterized by both ¹H NMR and FT IR. The ¹H NMR spectra verified the successful synthesis of mPEG-S₂-PZLL, and all signals were well assigned, as shown in Fig. 1B. The degree of polymerization (DP, Table 1) of PZLL in the resultant mPEG-S₂-PZLL was calculated by comparing the integrated area of the peak at 5.06 or 3.08 ppm, attributed to the methylene proton of the side group of the ZLL unit, with the signal at 3.65 ppm of the methylene proton assigned to mPEG. The results of FT IR (Fig. 1Cc and Cd) also demonstrated the generation of a PZLL block, based on the increased strength of absorption at 1654 (υ_C=O) and 1539 cm⁻¹ (υ_C(O)–NH), attributed to the amide bond in the backbone of PZLL. The successful synthesis of mPEG-S₂-PZLL was further confirmed by GPC. The PDI values of the copolymers were around 1.09–1.14. The relatively low molecular weight distribution may be attributed to the living character of the ROP of the NCA monomer, initiated by primary amine. It should be noted that the M_n measured by GPC was relatively higher than that measured by ¹H NMR, due to the structural differences between the resultant copolymer and the monodispersed polystyrene standards that were used to generate the calibration curve in GPC analyses.

Fig. 1 ¹H NMR spectra of mPEG₁₁₃-COOH (a) and mPEG₁₁₃-S₂-NH₂ (b) (A, in CDCl₃), and mPEG₁₁₃-S₂-PZLL₁₈ (a) and mPEG₁₁₃-S₂-PZLL₃₅ (b) (B, in TFA-d); FT IR spectra of mPEG₁₁₃-COOH (a), mPEG₁₁₃-S₂-NH₂ (b), mPEG₁₁₃-S₂-PZLL₁₈ (c) and mPEG₁₁₃-S₂-PZLL₃₅ (d) (C).

3.2. Self-assembly behaviors of block copolymers

Amphiphilic block copolymers with both hydrophilic and hydrophobic segments are a kind of interesting material which can self-assemble into various morphologies, such as vesicles and micelles, in aqueous solution.^30,31 In this study, the micelles of amphiphilic mPEG-S₂-PZLL block copolymers were prepared by direct dissolution (micelle₁) and dialysis (micelle₂) methods. The morphologies of the micelles were studied by TEM measurements. These revealed that both mPEG₁₁₃-S₂-PZLL₁₈ micelle₁ and micelle₂ had a spherical morphology with the respective average radii around 130 and 103 nm (Fig. 2A and B). In contrast, the hydrodynamic radii (R_h) measured by DLS were 181.0 ± 15.7 and 125.0 ± 5.9 nm, respectively (Fig. 2C and Table 2). The smaller values of R_h from TEM observations as compared to those determined by DLS may be attributed to the dehydration of the micelles in the process of TEM sample preparation. The shrinkage of the mPEG shell may also contribute to this difference. The micelle size and dispersity depended on the preparation approach. Micelle₂ prepared by the dialysis method exhibited a smaller radius and narrower dispersity than micelle₁ fabricated by the direct dissolution approach. This result was attributed to the random stage of copolymer dissolution in a co-solvent of hydrophilic and hydrophobic moieties (i.e. DMF) in the preparation process of micelle₂, which provided a sufficient degree of freedom to the macromolecular chain motion. The R_h value of the micelle could also be affected by the composition of the block copolymer. Longer hydrophobic PZLL segments resulted in a smaller R_h value, because of the increase in hydrophobic interactions between the PZLL chains. To determine the micellization process and investigate the effect of the copolymer composition on the formation and properties of the micelles, the widely reported pyrene-probe-based fluorescence technique was used.^6,32,33Fig. 3A shows the fluorescence excitation spectra of pyrene in mPEG₁₁₃-S₂-PZLL₁₈ aqueous solution at different concentrations. A red shift of the maximal absorption band from 335 to 340 nm was observed when the concentration of the copolymer was increased from 2.44 × 10⁻⁴ to 1.0 g L⁻¹. This red shift resulted from the transfer of pyrene molecules from a water environment to the hydrophobic micellar core, indicating the formation of micelles.

Fig. 2 Typical TEM micrographs of mPEG₁₁₃-S₂-PZLL₁₈ micelle₁ (A) and micelle₂ (B), and hydrodynamic radii (R_h) of mPEG₁₁₃-S₂-PZLL₁₈ micelle₁ (a) and micelle₂ (b) (C), and mPEG₁₁₃-S₂-PZLL₃₅ micelle₁ (a) and micelle₂ (b) (D).

Table 2 Properties of mPEG-S₂-PZLL block copolymers

Entry	CMC^a (mg L^−1)	CMC^b (mg L^−1)	R h (nm)	R h (nm)	DLC (wt%)	DLE (wt%)
a Micelle1. b Micelle2. c Micelle1. d Micelle2.
mPEG113-S2-PZLL18	18.74 ± 1.28	9.63 ± 1.13	181.0 ± 15.7	125.0 ± 5.9	5.35 ± 0.93	28.24 ± 1.25
mPEG113-S2-PZLL35	13.91 ± 1.07	3.68 ± 0.68	116.0 ± 10.2	60.8 ± 5.0	6.51 ± 0.86	34.83 ± 1.13

---

Fig. 3 Excitation spectra of pyrene in directly dissolved aqueous solutions of mPEG₁₁₃-S₂-PZLL₁₈ at different concentrations (a: 1.0, b: 1.25 × 10⁻¹, c: 1.56 × 10⁻², d: 1.95 × 10⁻³, e: 2.44 × 10⁻⁴ g L⁻¹, λ_em = 390 nm) (A), and the intensity ratio (I₃₄₀/I₃₃₅) as a function of the log of the concentration of mPEG₁₁₃-S₂-PZLL₁₈ (B) and mPEG₁₁₃-S₂-PZLL₃₅ (C) (a: micelle₁, b: micelle₂).

CMC is one of the key parameters to describe the physical properties of micelles. At the CMC, the hydrophobic portion starts to associate to reduce the interaction with water molecules, leading to micelle formation. From the plots of the pyrene fluorescence intensity ratio of I₃₄₀/I₃₃₅ (Fig. 3B and C) versus the log of copolymer concentration, the CMC values were obtained, taken as the intersection of the tangents to the horizontal line of intensity ratio with relatively constant value and the diagonal line with rapidly increased intensity ratio (Table 2). Both the micelle preparation approach and the copolymer composition affected the CMC value. Micelle₁ has a higher apparent CMC value than micelle₂, attributed to the non-micellar aggregation and incomplete diffusion of copolymeric chains. The CMC value of mPEG₁₁₃-S₂-PZLL₃₅ was lower than that of mPEG₁₁₃-S₂-PZLL₁₈, induced by the longer PZLL segment, which played an important role in the stability of the micelle. With the lower dispersity and CMC value of micelle₂, the dialysis method was more appropriate to prepare the micelles. In addition, the low CMC values below 10 mg L⁻¹ of the copolymers provided the possibility of retaining the stability of the micelles in very dilute aqueous milieu (e.g. blood stream), which is one of the important features for drug delivery.

3.3. Hemocompatabilities of micelles

Excellent hemocompatibilities of nanocarriers are one of the requisites because nanocarriers are designed to be administrated via intravenous injection for most drug delivery applications. Therefore, the interaction of the micelles and blood components should be effectively minimized and the abilities for resistance to nonspecific protein adsorption and elimination of hemolysis were requested.

In this work, the mPEG shell of the micelles can prevent recognition by the reticuloendothelial system and resist nonspecific protein adsorption (i.e. nonfouling properties) in complicated in vivo circumstances, which results in prolonged periods of circulation.^34,35 Herein, the interaction of the mPEG-S₂-PZLL micelle₂s with the model protein (i.e. BSA) was addressed to examine the potential of the micelle₂s for in vivo applications. As shown in Fig. 4, for all concentrations (0.125, 0.25 and 0.5 g L⁻¹) and incubation times (2, 4 and 6 h), the micelle₂s and mPEG₁₁₃ showed slight BSA adsorption, while PEI25k interacted strongly with BSA. It is believed that such a characteristic is the result of the formation of a strong hydration layer around the surface of the micelle because of the existence of the hydrophilic mPEG shell. In addition, the HR of the micelle was assessed by a hemolysis assay. An enhanced HR results in a higher level of broken RBCs. The RBCs were exposed to the micelles at different concentrations for 2 h. As shown in Fig. 5, the mPEG-S₂-PZLL micelle₂s showed negligible hemolysis destruction to RBCs, even at a very high concentration of 0.5 g L⁻¹. These results indicated that the mPEG-S₂-PZLL micelle₂s were hemocompatible, allowing potential application as drug delivery vehicles.

Fig. 4 BSA adsorption on the mPEG₁₁₃-S₂-PZLL₁₈ micelle₂ (a), mPEG₁₁₃-S₂-PZLL₃₅ micelle₂ (b), mPEG₁₁₃ (c) and PEI25k (d) after incubation at 37 °C with different concentrations and periods of time: 2 (A), 4 (B) and 6 h (C).

Fig. 5 Percentage of RBC hemolysis incubated with mPEG₁₁₃-S₂-PZLL₁₈ micelle₂ (a) and mPEG₁₁₃-S₂-PZLL₃₅ micelle₂ (b). PS (−) and triton X-100 (10.0 g L⁻¹) (+) were used as negative and positive controls, respectively. Data were represent as mean ± standard deviation (n = 3).

3.4. Cell and tissue compatibilities of micelles

Evaluation of the cell and tissue compatibilities of micelles is necessary for drug delivery applications. The in vitro cytotoxicities of the micelles to HeLa and HepG2 cells were evaluated by a MTT assay. Fig. 6 shows the viability of cells treated with micelles at different concentrations for 72 h, which was compared with those treated with PEI25k. More than 90% of the cells cultured with the micelles remained viable, even at a concentration of 0.5 g L⁻¹, whereas there were few live cells left after treatment with PEI25k at the same concentration. This suggested that the micelles had low cytotoxicities and could be safely used as biocompatible carriers for delivering bioactive substances.

Fig. 6 Cell viabilities of HeLa (A) and HepG2 (B) cells incubated with mPEG₁₁₃-S₂-PZLL₁₈ micelle₂ (a) and mPEG₁₁₃-S₂-PZLL₃₅ micelle₂ (b) with PEI25k (c) as control for 72 h. Data were presented as mean ± standard deviation (n = 6).

To further demonstrate the biocompatibilities of the micelles in vivo, the acute inflammatory response toward micelles was evaluated in the muscle in female BALB/c nude mouse. As shown in Fig. 7, severe necrosis was observed in the muscle sample after 7 days following PEI25k injection. However, histological analysis revealed no obvious inflammatory reactions at the injection sites with the same dose of the micelles, which were comparable to muscle samples without injection and after receiving PBS injection, indicating the good biocompatibilities of the micelles in vivo.

Fig. 7 Tissue responses after intramuscular injections of mPEG₁₁₃-S₂-PZLL₁₈ micelle₂ (A) and mPEG₁₁₃-S₂-PZLL₃₅ micelle₂ (B) as compared with that without (C), and with PBS (D) and PEI25k (E) injections. Tissue sections were analyzed by H&E staining after 7 days following injection.

3.5. DOX loading and reduction-responsive release

DOX is one of the most effective anthracycline antitumor drugs against a wide range of malignancies, which interacts with DNA through insertion and then inhibits the biosyntheses of bioactive macromolecules.^1,13 In the current study, to better understand the potential applications of the obtained micelles in drug delivery, DOX was selected as a model drug to determine the loading and release properties.

DOX was loaded into the micelles by a nanoprecipitation technique (Scheme 2). Comparison of the drug-loading abilities of the micelles indicated that the higher DLC and DLE were attributed to longer hydrophobic PZLL blocks (Table 2). The DLC values for mPEG₁₁₃-S₂-PZLL₁₈ and mPEG₁₁₃-S₂-PZLL₃₅ were 5.35 ± 0.93 and 6.51 ± 0.86 wt%, depending on the compositions of the block copolymers, while the DLE values were 28.24 ± 1.25 and 34.83 ± 1.13 wt%, respectively.

The drug release behaviors of the DOX-loaded micelles with different DP of PZLL were investigated in PBS without and with 10.0 mM GSH at pH 7.4. The cumulative release percentages of DOX loaded in micelles versus time were plotted in Fig. 8. In the absence of GSH, less than 50 wt% loaded DOX was released from the micelles over the test duration (60 h). However, the DOX release rates were accelerated by the presence of GSH, and over 90% of loaded DOX was released with 10.0 mM GSH, analogous to the intracellular reductive microenvironment, in 60 h. The fast DOX release from the micelles under the reductive conditions was most likely due to the degradation of the micelles during cleavage of the disulfide bonds (Scheme 2). The length of PZLL was found to be relevant to the drug release kinetics, and the release rates of DOX from micelles were of the order of mPEG₁₁₃-S₂-PZLL₁₈ micelle (with 10.0 mM GSH) > mPEG₁₁₃-S₂-PZLL₃₅ (with 10.0 mM GSH) > mPEG₁₁₃-S₂-PZLL₁₈ micelle (without GSH) > mPEG₁₁₃-S₂-PZLL₃₅ (without GSH). The increase in the DP of PZLL led to the decrease of the DOX release rate. Therefore, it was possible to modulate the release rate of DOX from the micelles by adjusting the reductive environment and polypeptide content. These release profiles are not only beneficial for minimizing drug loss in circulation, but also for selective accumulation in tumor tissue by the enhanced permeability and retention effect, which may improve the overall therapeutic efficacy in vivo.

Fig. 8 DOX release from DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles (a, b) and mPEG₁₁₃-S₂-PZLL₃₅ micelles (c, d) at pH 7.4 without GSH (a, c) and with 10.0 mM GSH (b, d) in PBS at 37 °C. Data were presented as mean ± standard deviation (n = 3).

3.6. Intracellular DOX release and cellular proliferation inhibition

To demonstrate the feasibility of the micelles for intracellular drug delivery in malignancy therapy, the cellular uptake and intracellular release behaviors of the DOX-loaded micelles were evaluated by CLSM and flow cytometry toward HeLa or HepG2 cells. HeLa or HepG2 cells were first pretreated with 0.5 mM BSO for 12 h, or with 10.0 mM GSH-OEt for 2 h, and then incubated with typical DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ or mPEG₁₁₃-S₂-PZLL₃₅ micelles for 2 h (10.0 mg L⁻¹ DOX). HeLa or HepG2 cells without pretreatment were used as controls. It is documented that BSO is an inhibitor for the intracellular synthesis of GSH, and GSH-OEt can obviously improve the intracellular GSH concentration.¹ As expected, the strongest intracellular DOX fluorescence was observed in the GSH-OEt pretreated cells after incubation with the DOX-loaded micelles (Fig. 9). In contrast, the weakest DOX fluorescence was shown in the cells preincubated with BSO. It was noted that stronger intracellular fluorescence was observed in the cells incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelles due to the larger total amount of intracellular released DOX, which was associated with the difference in the amount of released DOX between those without and with GSH (Fig. 8). The triggered intracellular drug release was further confirmed by flow cytometric analyses. As shown in Fig. 10, the flow cytometric histogram of the GSH-OEt pretreated cells incubated with DOX-loaded micelles shifts clearly in the direction of high fluorescence intensity as compared with that of the control, and the weakest fluorescence intensity was observed in the cells incubated with BSO. The free DOX was reported to have stronger fluorescence as compared with the DOX loaded in nanoparticles at the same absolute DOX concentration due to the self-quenching effects of DOX.^21,36 Thus, the enhanced fluorescence intensity in the GSH-OEt pretreated HeLa or HepG2 cells may be the result of the high endocytosis efficiency of the DOX-loaded micelles and the improved intracellular release of DOX, caused by the reduction-responsive degradation of the micelles.

Fig. 9 Representative CLSM images of HeLa cells incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles (A, B and C) and mPEG₁₁₃-S₂-PZLL₃₅ micelles (D, E and F), and HepG2 cells incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles (G, H and I) for 2 h: (A, D and G) cells without pretreatment; (B, E and H) cells pretreated with 0.5 mM BSO; (C, F and I) cells pretreated with 10.0 mM GSH-OEt. The cell nuclei and cytoskeleton of the cells were stained with DAPI (blue) and Alexa 488 (green), respectively.

Fig. 10 Flow cytometric profiles of HeLa cells incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles for 2 h: (a) cells without pretreatment; (b) cells pretreated with 0.5 mM BSO; (c) cells pretreated with 10.0 mM GSH-OEt.

The in vitro cellular proliferation inhibition of DOX-loaded mPEG-S₂-PZLL micelles against HeLa and HepG2 cells was also estimated by a MTT assay. Cells were pretreated with 0.5 mM BSO or 10.0 mM GSH-OEt for 12 h and 2 h, separately, and then incubated with the DOX-loaded micelles for 24, 48 and 72 h. The cells without pretreatment were used as controls. According to the previous literature, 0.5 mM BSO and 10.0 mM GSH-OEt did not show any cytotoxicities toward HeLa and HepG2 cells.^1,37 As shown in Fig. 11, in contrast to the control cells, the HeLa and HepG2 cells pretreated with BSO and GSH-OEt exhibited the lowest and highest inhibition efficiency, respectively. The results revealed that the faster DOX release from the DOX-loaded micelles was triggered by higher intracellular GSH concentration, which enhanced the inhibition of the cellular proliferation. In addition, the DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelles exhibited higher cellular proliferation inhibiting ability than the DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles in all the test conditions and durations. This was due to the slower extracellular DOX release and almost identical intracellular DOX release from the DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelles (Fig. 8). The half maximal inhibitory concentration (IC₅₀) values of the DOX-loaded micelles were listed in Table 3. In contrast to the control cells, the cells pretreated with BSO and GSH-OEt exhibited the highest and lowest IC₅₀ values, respectively. With the same test conditions and duration, all the cells cultured with DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelles showed lower IC₅₀ values in contrast with the DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles. It quantificationally confirmed the differences in cellular proliferation inhibitions of unpretreated and pretreated cells with various DOX-loaded mPEG-S₂-PZLL micelles and revealed the adjustable reduction-responsive ability of the micelles in cells.

Fig. 11 Proliferation inhibition towards HeLa (A, B, C, G, H and I) and HepG2 cells (D, E, F, J, K and L) incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₁₈ micelles (A, B, C, D, E and F) and mPEG₁₁₃-S₂-PZLL₃₅ micelles (G, H, I, J, K and L) with various DOX concentrations for 24 (A, D, J and G), 48 (B, E, H and K) and 72 h (C, F, I and L). The cells were unpretreated (a), pretreated with 0.5 mM BSO (b) and 10.0 mM GSH-OEt (c), respectively. Data were presented as mean ± standard deviation (n = 6).

Table 3 IC₅₀ values of DOX-loaded micelles after incubation with unpretreated and pretreated HeLa and HepG2 cells for 24, 48 and 72 h

Entry	Pretreatment	IC50^a (mg L^−1)		IC50^b (mg L^−1)		IC50^c (mg L^−1)
		HeLa	HepG2^d	HeLa	HepG2	HeLa	HepG2
a 24 h. b 48 h. c 72 h. d Beyond the test range.
mPEG113-S2-PZLL18	Unpretreated	8.05	—	1.01	6.62	0.55	2.34
	0.5 mM BSO for 12 h	8.65	—	1.15	8.98	0.99	4.72
	10.0 mM GSH-OEt for 2 h	2.75	—	0.46	5.19	0.30	1.48
mPEG113-S2-PZLL35	Unpretreated	2.75	—	0.67	3.26	0.86	1.70
	0.5 mM BSO for 12 h	3.32	—	1.08	3.49	0.94	2.69
	10.0 mM GSH-OEt for 2 h	1.31	—	0.46	3.04	0.24	1.25

---

4. Conclusions

Two disulfide-linked mPEG-S₂-PZLL block copolymers were synthesized through ROP of ZLL NCA, initiated by mPEG-S₂-NH₂. The block copolymers formed into micelles in PBS at pH 7.4 by direct dissolution and dialysis methods. The micelles prepare by the dialysis method have a lower dispersity and CMC value than those prepared by the direct dissolution approach. The hemocompatibilities, and cell and tissue compatibilities of the micelles were revealed to be excellent, which rendered them with potential for delivering bioactive substances. DOX was loaded into the micelles, and the DOX release was accelerated in the presence of 10.0 mM GSH or from micelles with lower PZLL content. The DOX-loaded micelles exhibited faster DOX release in HeLa and HepG2 cells, which were pretreated with 10.0 mM GSH-OEt or incubated with DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelles. Moreover, a higher cellular proliferation inhibition efficacy was achieved toward GSH-OEt pretreated or DOX-loaded mPEG₁₁₃-S₂-PZLL₃₅ micelle cultured HeLa and HepG2 cells. These above features, including biocompatibilities, stability in storage conditions and intracellular stimuli triggered drug release, bestowed the mPEG-S₂-PZLL micelles with unique advantages as vehicles for intelligent intracellular drug delivery in clinical chemotherapy.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (Projects 51233004, 51273196, 51273037, 51203153, 51273080, 51003103, 50973108, 20904053 and 21174142).

Notes and references

1. F. Shi, J. Ding, C. Xiao, X. Zhuang, C. He, L. Chen and X. Chen, J. Mater. Chem., 2012, 22, 14168–14179.
2. J. Ding, F. Shi, C. Xiao, L. Lin, L. Chen, C. He, X. Zhuang and X. Chen, Polym. Chem., 2011, 2, 2857–2864.
3. A. Bhaw-Luximon, L. M. Meeram, Y. Jugdawa, W. Helbert and D. Jhurry, Polym. Chem., 2011, 2, 77–79.
4. J. Liu, W. Huang, Y. Pang, X. Zhu, Y. Zhou and D. Yan, Biomaterials, 2010, 31, 5643–5651.
5. J. Cheng, J. X. Ding, Y. C. Wang and J. Wang, Polymer, 2008, 49, 4784–4790.
6. W. Wang, J. X. Ding, C. S. Xiao, Z. H. Tang, D. Li, J. Chen, X. L. Zhuang and X. S. Chen, Biomacromolecules, 2011, 12, 2466–2474.
7. X. Zhao, Z. Poon, A. C. Engler, D. K. Bonner and P. T. Hammond, Biomacromolecules, 2012, 13, 1315–1322.
8. J. Ding, C. Xiao, C. He, M. Li, D. Li, X. Zhuang and X. Chen, Nanotechnology, 2011, 22, 494012.
9. X. Yang, J. J. Grailer, I. J. Rowland, A. Javadi, S. A. Hurley, V. Z. Matson, D. A. Steeber and S. Gong, ACS Nano, 2010, 4, 6805–6817.
10. S. H. Jung, S. H. Jung, H. Seong, S. H. Cho, K.-S. Jeong and B. C. Shin, Int. J. Pharm., 2009, 382, 254–261.
11. Y. Y. Yuan, J. Z. Du, W. J. Song, F. Wang, X. Z. Yang, M.-H. Xiong and J. Wang, J. Mater. Chem., 2012, 22, 9322.
12. J. Ding, C. Xiao, L. Yan, Z. Tang, X. Zhuang, X. Chen and X. Jing, J. Controlled Release, 2011, 152, E11–E13.
13. H. Sun, B. Guo, R. Cheng, F. Meng, H. Liu and Z. Zhong, Biomaterials, 2009, 30, 6358–6366.
14. C. Y. Zhang, Y. Q. Yang, T. X. Huang, B. Zhao, X. D. Guo, J. F. Wang and L. J. Zhang, Biomaterials, 2012, 33, 6273–6283.
15. V. N. Rao, S. R. Mane, A. Kishore, J. Das Sarma and R. Shunmugam, Biomacromolecules, 2012, 13, 221–230.
16. N. Ma, H. Xu, L. An, J. Li, Z. Sun and X. Zhang, Langmuir, 2011, 27, 5874–5878.
17. X. J. Cai, H. Q. Dong, W. J. Xia, H. Y. Wen, X. Q. Li, J. H. Yu, Y. Y. Li and D. L. Shi, J. Mater. Chem., 2011, 21, 14639.
18. X. Jiang, L. Li, J. Liu, W. E. Hennink and R. Zhuo, Macromol. Biosci., 2012, 12, 703–711.
19. K. Y. Choi, H. Y. Yoon, J.-H. Kim, S. M. Bae, R.-W. Park, Y. M. Kang, I. S. Kim, I. C. Kwon, K. Choi, S. Y. Jeong, K. Kim and J. H. Park, ACS Nano, 2011, 5, 8591–8599.
20. S. R. MacEwan, D. J. Callahan and A. Chilkoti, Nanomedicine, 2010, 5, 793–806.
21. H. Sun, B. Guo, X. Li, R. Cheng, F. Meng, H. Liu and Z. Zhong, Biomacromolecules, 2010, 11, 848–854.
22. L. Y. Tang, Y. C. Wang, Y. Li, J. Z. Du and J. Wang, Bioconjugate Chem., 2009, 20, 1095–1099.
23. T. B. Ren, W. J. Xia, H. Q. Dong and Y. Y. Li, Polymer, 2011, 52, 3580–3586.
24. H. Y. Wen, H. Q. Dong, W. J. Xie, Y. Y. Li, K. Wang, G. M. Pauletti and D. L. Shi, Chem. Commun., 2011, 47, 3550–3552.
25. J. Ding, C. Xiao, Z. Tang, X. Zhuang and X. Chen, Macromol. Biosci., 2011, 11, 192–198.
26. J. Ding, C. Xiao, L. Zhao, Y. Cheng, L. Ma, Z. Tang, X. Zhuang and X. Chen, J. Polym. Sci., Part A: Polym. Chem., 2011, 49, 2665–2676.
27. H. Leuchs, Ber. Dtsch. Chem. Ges, 1906, 39, 857–861.
28. T. J. Deming, Prog. Polym. Sci., 2007, 32, 858–875.
29. T. Thambi, H. Y. Yoon, K. Kim, I. C. Kwon, C. K. Yoo and J. H. Park, Bioconjugate Chem., 2011, 22, 1924–1931.
30. A. Lalatsa, A. G. Schaetzlein, M. Mazza, L. Thi Bich Hang and I. F. Uchegbu, J. Controlled Release, 2012, 161, 523–536.
31. T. Smart, H. Lomas, M. Massignani, M. V. Flores-Merino, L. R. Perez and G. Battaglia, Nano Today, 2008, 3, 38–46.
32. J. Rodriguezhernandez, F. Checot, Y. Gnanou and S. Lecommandoux, Prog. Polym. Sci., 2005, 30, 691–724.
33. J. Ding, X. Zhuang, C. Xiao, Y. Cheng, L. Zhao, C. He, Z. Tang and X. Chen, J. Mater. Chem., 2011, 21, 11383–11391.
34. K. Osada, H. Cabral, Y. Mochida, S. Lee, K. Nagata, T. Matsuura, M. Yamamoto, Y. Anraku, A. Kishimura, N. Nishiyama and K. Kataoka, J. Am. Chem. Soc., 2012, 134, 13172–13175.
35. G. Pasut and F. M. Veronese, J. Controlled Release, 2012, 161, 461–472.
36. Y. Bae, S. Fukushima, A. Harada and K. Kataoka, Angew. Chem., Int. Ed., 2003, 42, 4640–4643.
37. J. Y. Liu, Y. Pang, W. Huang, X. H. Huang, L. L. Meng, X. Y. Zhu, Y. F. Zhou and D. Y. Yan, Biomacromolecules, 2011, 12, 1567–1577.

---