Pluronic F127–chondroitin sulfate micelles prepared through a facile method for passive and active tumor targeting

Abstract

Tumor-specific drug delivery is still a challenge in cancer therapy. Passive tumor targeting strategies, such as the enhanced permeability and retention (EPR) effect, cause nanocarriers to accumulate in tumors. However, this strategy can not provide specific tumor targeting. In this study, Pluronic F127 (PF127), a block copolymer which can inhibit drug efflux transporters in cancer therapy, was modified to form tumor-specific micelles with a natural polysaccharide, chondroitin sulfate (ChS), which imparts the site-specific property. A facile and efficient method based on Schiff base reaction was developed to facilitate both basic and clinical research. A series of PF127–ChS micelles with different ratios of PF127 and ChS were fabricated and evaluated in terms of size, morphology, drug loading efficiency and drug release behavior. Spherical micelles with a mean diameter of 155–241 nm were obtained. Their critical micelle concentration (CMC) was significantly reduced in contrast to PF127 micelles and their stability was enhanced. Doxorubicin (DOX) was loaded into the hydrophobic core of PF127 or adsorbed by ChS through electrostatic interactions with the negative charges of chondroitin sulfate. In vitro DOX release studies showed that DOX release from the micelles was enhanced at acidic pH values compared to physiological pH. A cytotoxicity assay (MTT) determined that the micelles possess significantly lower toxicity. Confocal microscopy and flow cytometry analysis indicated that DOX loaded micelles could efficiently release DOX inside cells by specific cellular uptake. These outcomes revealed that PF127–ChS micelles could be exploited as carriers for anti-tumor drugs for site-specific therapy of solid tumors.

---

Introduction

During the last decade, pharmaceutically acceptable micelles have received much enthusiasm among researchers for clinical cancer chemotherapy as they could overcome the obstacles of conventional administration and improve therapeutic outcomes.^1–3 Micelles possess various advantages in their applications for drug delivery. Their intrinsic core–shell structures allow them to load poorly water-soluble anticancer drugs in the hydrophobic core to improve bioavailability. They can also serve as drug reservoirs to allow sustained release to prevent or decrease tissue damage upon accidental extravasations. Moreover, micelles with a nano-size could get to the lesion place (tumor site) by passive targeting due to an enhanced permeability and retention (EPR) effect. These special advantages make micelles a hot research topic as drug nanocarriers and many researchers have successfully prepared a variety of micelles for cancer treatment.^4–6 However, passive tumor targeting based on the EPR effect can not provide specific tumor targeting. Therefore, tumor-specific ligands, such as folic acid, nucleic acids, peptides and antibodies, are introduced to micelles, which can interact with receptors overexpressed in cancer cells and enhance their recognition and internalization by the target tissues.

Even though introducing tumor-specific ligands to micelles results in better anticancer activity, only a few of them successfully tested in clinical trials have shown promise. The reason for this lies in the toxicity, instability, inefficiency, complexity of the synthesis, and other factors related to scaling. Park et al. prepared folate-receptor targeted micelles based on block copolymers of poly(D,L-lactic-co-glycolic acid) (PLGA) and poly(ethylene glycol) (PEG) using carbodiimide chemistry for delivery of doxorubicin (DOX).⁷ In vitro and in vivo studies showed that folate conjugated micelles exhibited a greater extent of cellular uptake against KB cells over-expressing folate receptors and regressed tumor volume significantly. However, excess folate would negatively affect the aqueous stability of the micelles, resulting in formation of insoluble aggregates.⁸ Monoclonal antibodies are the most well-known and efficient targeting ligands and about 30 of them have been approved for clinical use.⁹ They have been widely employed in many active targeting schemes and in vitro and in vivo studies exhibited excellent active tumor targeting. However, some drawbacks such as the large size, low stability, high cost and potential immunogenicity impede their applications.¹⁰ In addition, antibodies are always conjugated in micelles via chemical modification, which could decrease the affinity toward receptors.¹¹

In this study, Pluronic F127 (PF127), a block copolymer, was modified to form micelles with a natural polysaccharide, chondroitin sulfate (ChS), to achieve both passive and active targeting effects. A facile and efficient method based on Schiff base reaction was developed to facilitate both basic and clinical research. The use of PF127 as a micelle for a drug delivery vehicle to treat cancer is a rapidly developing area for cancer chemotherapy.^12,13 PF127 has a special structure consisting of hydrophilic poly(ethylene oxide) (PEO) and hydrophobic poly(propylene oxide) (PPO) segments, therefore it could self-assemble into micelles with its hydrophobic PPO inner core and hydrophilic PEO outer shell above the critical micelle concentration (CMC).¹⁴ The PF127 based micellar formulation of DOX, SP1049C, comprising mixed micelles of PF127 and L61, was the first micelle drug to advance to clinical stage¹⁵ and has completed a phase II clinical trial in advanced esophageal cancer patients.¹⁶ However, the use of PF127 micelles in vivo has been hampered by the following factors: the small size of the PF127 (20–30 nm) may allow premature renal excretion of the carrier and penetration through the tight junctions of healthy endothelia; and the micelles disassociate at low concentrations (below CMC) and are no longer able to keep the incorporated hydrophobic drugs. The main shortcomings of an unstable structure and easy excretion limit the bioavailability of PF127 micelles in vivo.^17,18

Due to these obstacles, we prepared PF127 micelles with a natural polysaccharide, ChS, in this study. ChS is an important structural component found in bone, cartilage and connective tissue, composing of alternating units of β-1,4-linked glucuronic acid and β-1,3-nacetylgalactosamine.^19,20 ChS has been widely used as a potential biomaterial in pharmaceutical studies with many promising properties such as good hydrophilicity, biocompatibility, biodegradability and anti-inflammatory. In addition, ChS as one of the glycosaminoglycans (GAGs) of all vertebrates binds endogenous proteins with many functional properties of growth factors, adhesion molecules and enzymes, and has been shown that it has an anti-atherogenic effect. More importantly, ChS could target tumor cells and facilitate cellular uptake by binding to CD44 receptors over expressed on various tumor cells.^21,22 In this study, ChS was introduced into the micelles to improve their stabilization and impart the site-specific property.

The formation of PF127–ChS micelles is based on the Schiff base reaction between aldehydes and amine groups. Briefly, the terminal hydroxyl groups of PF127 were oxidized (PF127–CHO) to react with the amine groups on the modified chondroitin sulfate (ChS–ADH). Both of the modifications of ChS and PF127 were acquired through facile methods. The Schiff base reaction takes place just by stirring the mixture solution (PF127–CHO and ChS–ADH) under mild conditions in PBS (pH 7.4) without using toxic chemical agents and water is the only byproduct, which facilitates both basic and clinical research for cancer therapy.^23,24 In addition, it is known that both tumor extracellular space (pH ∼ 6.8) and subcellular organelles (endosomes and lysosomes) (pH ∼ 5.5) are significantly more acidic than normal physiological environments (pH 7.4), while Schiff base bonds are hydrolyzed faster at low pH.²⁵ Therefore, PF127–ChS micelles fabricated through Schiff base interactions are well suited for delivery of anticancer drugs.

Materials and methods

Materials

Chondroitin sulfate (ChS, MW 20–50 K) and Pluronic F127 were provided from Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. (Shanghai, China) and used as received. Dess–Martin periodinane was purchased from Aladdin (Shanghai, China). Adipic dihydrazide (ADH), 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS) and other chemicals were used as received.

Preparation of PF127–ChS micelles

The terminal alcohols on PF127 were converted to aldehydes (PF127–CHO) via an oxidation reaction of Dess–Martin periodinane.²⁶ Briefly, PF127 (5.16 g, 0.410 mmol) and Dess–Martin periodinane (0.6 g, 1.415 mmol) were placed in a dry round-bottom flask with anhydrous methylene chloride (400 mL). The reaction mixture was agitated and refluxed at 40 °C for 30 h. Later, most of the solvent was evaporated under vacuum and the product was precipitated in excess hexane, filtered. The extent of oxidation of PF127–CHO, which is defined as the number of oxidized PF127–CHO residues per 100 original PF127 feeds, was estimated using proton nuclear magnetic resonance (¹H-NMR) analysis.

Adipic acid dihydrazide was used in the process of functionalization of ChS with hydrazine groups.²⁷ Briefly, ChS (0.450 g) was dissolved in deionized water at a concentration of 2.5 mg mL⁻¹. EDC (0.1772 g, 1.4392 mmol) and NHS (0.1656 g, 1.4392 mmol) were added to activate the carboxyl groups of ChS and the pH of the solution was adjusted to pH ≈ 5.5. After stirring for 0.5 h, ADH (7.5 mmol) was added to the mixture and the pH was adjusted to 6.0. The reaction was left to continue for 24 h. Finally, the polymer solution was dialyzed against distilled water for 3 days and lyophilized. The obtained ChS–ADH was confirmed by ¹H-NMR analysis. The degree of substitution of ChS–ADH, which is defined as the number of hydrazine groups per 100 sugar residues of ChS, was determined by elemental analysis.

A series of PF127–ChS micelles with different ratios of PF127–CHO and ChS–ADH were prepared by Schiff base reactions. In brief, 45.5 mg PF127–CHO and 4.5 mg ChS–ADH were dissolved in 10 mL phosphate buffer solution (PBS, pH 7.4), and stirred continuously for 24 h at 500 rpm to obtain micelles with a functional group molar ratio of 1 : 1 (named as PF127₁–ChS₁). After centrifugation for 20 min at 15 000 rpm to remove unreacted material, the micelles were dissolved in PBS again. Likewise, other micelles with different ratios (1 : 0.5, 1 : 2, and 1 : 5, named as PF127₁–ChS_0.5, PF127₁–ChS₂, and PF127₁–ChS₅, respectively) were prepared by varying the amount of ChS–ADH. The structure of PF127₁–ChS₅ was determined by ¹H-NMR analysis.

Characterization of PF127–ChS micelles

IR absorption spectra of ChS, PF127–CHO and PF127₁–ChS₅ were obtained on a Fourier transform infrared (FTIR) spectrometer (Nicolet 670 FTIR, USA) over the region of 4000 to 400 cm⁻¹.

Transmission electron microscopy (TEM) images were visualized on a transmission electron microscope (JEM-1200EX/S, Hitachi, Japan) operated at an accelerating voltage of 200 kV. Micelle solution (1 mg mL⁻¹) was deposited onto a carbon coated copper grid and dried at room temperature for 1 day. A drop of 2% phosphotungstic acid was then used to stain the grid before examination with the electron microscope.

Hydrodynamic diameter and size distribution of the micelles were measured by dynamic light scattering (DLS) using a laser light scattering spectrometer (ALV/SP-125). Scattered light was collected with a laser of wavelength 514.5 nm at a 90° scattering angle for a duration of 10 min. All data were averages of five tests. The concentration of all of the micelle solutions was 1 mg mL⁻¹ and the temperature was maintained at 25 °C. Before measurement, the solutions were filtered through nylon filters (0.45 mm pore size).

Zeta potentials of the micelles dispersed in deionized water were determined with a potentiostat (Malvern ZEN3600, England).

The critical micelle concentration (CMC) of PF127–ChS micelles in PBS (pH 7.4) was determined using fluorescence probe technique spectroscopy (Lamda 35, Perkin-Elmer Co., USA), using pyrene as a fluorescence probe. In a typical test, pyrene in diethyl ether (6.0 × 10⁻⁶ M) was prepared and 1 mL of the solution was added to a series of 10 mL colorimetric cylinders. The diethyl ether was evaporated by placing at 30 °C for 12 h to form a thin film at the bottom of the colorimetric cylinder. Then, 10 mL of the micelle solutions (in PBS, pH 7.4) with different concentrations (0.004–3 mg mL⁻¹) were added to the colorimetric cylinders. The mixture solution was equilibrated at room temperature for 24 h before measurement. Finally, the excitation spectra of the solutions were recorded from 300 to 400 nm with an emission wavelength of 490 nm and excitation and emission bandwidths set at 15 nm. The CMC value was taken from the crosspoint when extrapolating the intensity ratio I₃₈₄/I₃₇₃.

Disassembly of micelles under acidic environments

In order to investigate the hydrolysis of PF127–ChS micelles in acidic conditions, the diameter and size distribution changes of the micelles were observed by DLS. The micelle solution (1 mg mL⁻¹) was first adjusted to pH 7.4, pH 6.5 and pH 5.5 using hydrochloric acid solution. Then the solution was stirred for 12 h at room temperature. All data were averaged over five measurements.

Preparation of DOX loaded micelles

To prepare DOX loaded micelles, 10 mg of DOX·HCl was first dissolved in 2 mL DMSO containing 25 μL of triethylamine, and stirred for 5 h. Then the free DOX solution was added into the PF127–ChS solution (5 mg mL⁻¹) and stirred vigorously for 24 h. The mixed solution was dialyzed against NaOH aqueous solution (pH 7.8–8.0) for 30 h to remove DMSO, triethylamine and non-encapsulated DOX. Finally, an appropriate amount of PBS was added to the DOX loaded micelle solution to achieve a final concentration of 1 mg mL⁻¹. The final solution was used directly in subsequent tests. All the drug loaded samples were prepared just prior to testing. The loading amount of DOX in the micelles was determined using a UV/vis spectrophotometer (Lambda 35, Perkin-Elmer, America) at 490 nm. 500 μL of the pre-made DOX-loaded micelle solution was introduced into 8 mL of DMSO under a vigorous vortex. The concentration of DOX was calculated according to a standard curve generated from DMSO with DOX. The drug loading (DL) and the drug encapsulation efficiency (EE) were calculated using the following formulas:

DL (wt%) = (mass of DOX in micelles/mass of micelles) × 100%

EE (wt%) = (mass of DOX in micelles/mass of feeding DOX) × 100%

Drug release study in vitro

The in vitro release behavior of the DOX loaded micelles was tested under different pHs. The DOX loaded micelle solution (1 mg mL⁻¹) in a dialysis bag (molecular weight cut-off 3500) was submerged into 30 mL of PBS with pH values of 7.4 and 5.5, and then incubated at 37 °C under oscillation at 100 rpm. At given time intervals, 5 mL of the release medium was taken out to measure the DOX concentration in the dialysate and 5 mL of fresh buffer solution was added to maintain a sink condition. UV/vis absorbance was recorded at 490 nm for DOX.

Cytotoxicity test

The cytotoxicity of the PF127–ChS micelles was evaluated using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay with human lung cancer A549 cells. 200 μL of culture medium containing human lung cancer A549 cells was added into a 96-well plate at a density of 10 000 cells per well and cultured for 12 h. The micelle solutions, which had been sterilized just before use, were injected into the wells to a concentration of 5, 10, 20, 50 and 100 μg mL⁻¹. After culturing for 48 h in a humidified atmosphere with 5% CO₂ at 37 °C, the medium was removed carefully and replaced by MTT reagent for another 4 h. Then the cells were dissolved by 200 μL of DMSO. Cell viability was determined by the absorbance values at 490 nm of the cells which was measured with a microplate reader (Tecan, Mannedorf, Switzerland).

The toxicity of free DOX and DOX loaded PF127₁–ChS₅ micelles on A549 cells was evaluated using the MTT method as described before. The IC₅₀ values of free DOX and DOX loaded PF127₁–ChS₅ micelles were calculated using SPSS software.

Cell uptake study

To examine the cellular uptake of free DOX and DOX loaded micelles, confocal laser scanning microscopy (CLSM) was employed using human lung cancer A549 cells. First, the human lung cancer A549 cells were added into a 6-well plate at a density of 1 × 10⁴ cells per well and cultured for 12 h. Then DOX loaded PF127₁–ChS₅ and free DOX were added, and the cells were cultured for 0.5 h, 1 h and 4 h. The concentration of DOX in each well was 10 μg mL⁻¹. Finally, the location of intracellular fluorescence was validated using a CLSM imaging system (LYMPUSFV-1000) at the excitation wavelength of 490 nm.

Flow cytometry analysis was employed to quantitatively examine the cellular uptake of free DOX and DOX loaded micelles, and the model cells were human lung cancer A549 cells. First, 5 mL of the human lung cancer A549 cells were added into 6-well plates at a density of 1 × 10⁴ cells per well and cultured for 48 h. Then DOX loaded PF127₁–ChS₅ and free DOX, which had been sterilized just before use, were added, the concentration of DOX in each well was 10 μg mL⁻¹. It was cultivated for 0.5 h, 1 h and 4 h in a humidified atmosphere with 5% CO₂ at 37 °C. Then we harvested the cells. Finally, the positive ratio of cells marked by DOX was validated using flow cytometry (Accuri C6, Becton, Dickinson and Company, American) at the excitation wavelength of 490 nm.

Statistical analysis

Statistical analysis of the data was performed using the one-way analysis of variance (ANOVA), assuming a confidence level of 95% (p < 0.05) for statistical significance.

Results and discussion

Synthesis and characterization of PF127–ChS micelles

Recent progress in biomaterials has given new insights into target-specific anticancer drug delivery. A variety of carriers have been fabricated.^28–30 However, complex synthesis processes and the possible toxicity of the agents used were found to be some critical challenges for their application. In this study, we have embarked on the design and synthesis of PF127 based micelles using a facile and fast method for target-specific drug delivery to tumors. Schiff base reaction was employed to achieve such a goal, as shown in Scheme 1. PF127–CHO was first synthesized using Dess–Martin periodinane, making some of the terminal alcohols on PF127 oxidize to aldehydes. Then ChS, possessing high affinity toward CD44 receptors, was modified under mild conditions with covalent attachment of ADH to the carboxylic acid groups, resulting in pendant hydrazine amino functionalities in ChS–ADH. The PF127–ChS micelles were synthesized via Schiff base reaction between the aldehydes of PF127–CHO and the amine groups of ChS–ADH. The synthesis method is facile and straightforward, without using an initiator, crosslinker or other catalyst. Therefore, the micelles should facilitate both basic and clinical research for tumor targeting treatment. Fig. 1A shows the ¹H NMR spectrum of PF127–CHO. The peak at 9.7 ppm was attributed to the hydrogen in the aldehydes (–CHO), demonstrating that the primary alcohols of PF127 were converted to aldehydes. From the peak areas of 9.7 ppm and 1.18 ppm (OCH₂OH(CH₃)O of PF127–CHO) in the PF127–CHO spectrum, we can estimate that about 95% of the proportion of alcohols in PF127 were converted to aldehydes. Fig. 1B shows the ¹H NMR spectrum of ChS–ADH, in which the typical resonance peaks of both ChS and ADH were clearly exhibited, demonstrating the successful preparation of ChS–ADH. The peaks between 3.23 ppm and 4.65 ppm can be assigned to the protons in the sugar rings of ChS, and the peaks at 1.74 and 2.60 ppm refer to the methylene protons of ADH residues. The spectrum of the PF127₁–ChS₅ micelles (Fig. 1C) presents both the resonance peaks from PF127–CHO and ChS–ADH, demonstrating the occurrence of the Schiff base reaction.

Scheme 1 Schematic illustration of DOX loaded in PF127–ChS micelles and their tumor targeting and intracellular drug release action.

Fig. 1 ¹H NMR spectra of PF127–CHO (A), ChS–ADH (B) and PF127₁–ChS₅ micelles (C).

FTIR spectra of PF127–CHO, ChS–ADH and PF127₁–ChS₅ are given in Fig. 2. There is a strong absorbance peak at 1729 cm⁻¹ in the PF127–CHO spectrum (Fig. 2a), which is attributed to C=O stretching of the aldehydes, indicating that the primary alcohols have been successfully oxidized to aldehydes. In the spectrum of ChS–ADH (Fig. 2b), some characteristic peaks can be ascribed at 1625 cm⁻¹ for amide bands, and 1700 cm⁻¹ for secondary amines (–NHNH–), suggesting the substitution of ChS. In the spectrum of PF127–ChS (Fig. 2c), the peak at 1729 cm⁻¹ almost disappears while the characteristic peak at 1660 cm⁻¹ (–CN–) appears, which suggests that the aldehydes groups in the PF127–CHO have reacted with the amino groups in the ChS by the Schiff base reaction.

Fig. 2 FTIR spectra of PF127–CHO (a), ChS–ADH (b) and PF127₁–ChS₅ micelles (c).

The size of the micelles modulated their cellular uptake. TEM and DLS experiments revealed that compact PF127–ChS micelles formed with an average size less than 250 nm. Fig. 3 shows the TEM micrographs of the PF127–ChS micelles. As shown in the figure, the particle sizes of PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅ are about 48 nm, 50 nm, 190 nm and 230 nm, respectively. Aggregation occurred for PF127₁–ChS_0.5 and PF127₁–ChS₁, which may be due to their small size. However, a well dispersed spherical morphology was observed for PF127₁–ChS₂ and PF127₁–ChS₅, indicating that the introduction of ChS improved the stabilization of the micelles. Fig. 4 shows the average hydrodynamic diameters of the micelles, which was tested in PBS at a concentration of 1 mg mL⁻¹. The results were 155 nm, 163 nm, 176 nm and 241 nm for PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅, respectively. The size of the micelles is much bigger than that of PF127 micelles (20–30 nm), indicating that the introduction of ChS to the system makes bigger micelles. The difference in the TEM and DLS results is attributed to the different test status and the hydrodynamic diameter was recorded in DLS. Micelles with a diameter smaller than 250 nm could passively accumulate in tumor sites via the EPR effect, which provides the possibility for the targeted delivery of drugs to specific tumor sites.¹⁰ The surface charge of the micelles also affects their cellular uptake. Positively charged carriers could interact with proteoglycans with negative charges in the cell membrane due to strong electrostatic interactions.³¹ The zeta potential of the PF127–ChS micelles in PBS are −13.4, −24.9, −26.2 and −30.4 for PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅, respectively. The negative potential provides micelles with long term stability and the relatively higher zeta potential provides better suspension stability.

Fig. 3 TEM images of PF127–ChS micelles: PF127₁–ChS_0.5 (a), PF127₁–ChS₁ (b), PF127₁–ChS₂ (c) and PF127₁–ChS₅ (d).

Fig. 4 Size distributions of PF127–ChS micelles determined by DLS: PF127₁–ChS_0.5 (a), PF127₁–ChS₁ (b), PF127₁–ChS₂ (c) and PF127₁–ChS₅ (d).

The fluorescence emission spectrum of pyrene was used to measure the CMC of the micelles and to demonstrate the presence of a hydrophobic environment in the stabilized micelles. The fluorescence emission spectrum of pyrene is highly dependent upon the hydrophobicity of the local environment. Fig. 5 shows the change in the intensity ratio of I₃₈₄/I₃₇₃ in pyrene excitation spectra. There is no obvious change in the intensity ratio at lower concentrations, but an increase emerges above the CMC value, which indicates the incorporation of pyrene in the hydrophobic core of the micelles. The CMC of PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅ is about 0.018 mg mL⁻¹, 0.0068 mg mL⁻¹, 0.0019 mg mL⁻¹ and 0.0018 mg mL⁻¹, respectively. The CMC of the PF127–ChS micelles is much smaller than that of PF127 (about 0.035 mg mL⁻¹), confirming the micelles hold better stabilities, which is because the micelles were stabilized by conjugating with ChS. The low CMC values implied a strong tendency of the polymers to form micelles in water, which would be beneficial for the stabilization of micelle circulation in physiological conditions.³²

Fig. 5 Intensity ratios (I₃₈₄/I₃₇₃) from pyrene excitation spectra as a function of PF127–ChS micelles in PBS buffer: PF127₁–ChS_0.5 (a), PF127₁–ChS₁ (b), PF127₁–ChS₂ (c) and PF127₁–ChS₅ (d).

Disassembly of micelles under acidic environments

The pH sensitivity of the micelles was studied by DLS measurements with changes in the external pH (from 7.4 to 5.5). With the decrease of the media pH value from 7.4 to 5.5, the hydrodynamic diameter of the micelles changed, as shown in Fig. 6. At the slightly basic pH (pH 7.4), the micelles have a hydrodynamic diameter of about 240 nm with a normal symmetrical distribution. In acidic environments (pH 6.5 and pH 5.5), the size of the micelles underwent a significant increase to about 410 nm for pH 6.5 and about 650 nm for pH 5.5, and a heterogeneous non-symmetrical distribution appeared. This phenomenon indicates that the micelles have been damaged under acidic pH conditions by breaking of the Schiff base bonds. Then the degradation products of the micelles aggregate, making the DLS results show a higher value. Such pH-induced damage and degradation of the micelles may ensure the ability of PF127–ChS micelles as an ideal intracellular drug-delivery system for tumor cells.³³

Fig. 6 Size distributions of PF127₁–ChS₅ micelles in PBS of different pH values (pH 7.4, 6.5 and 5.5) for 12 h.

Drug loading and in vitro release of DOX

To investigate the pH stimuli-responsive release behavior of PF127–ChS micelles, DOX, a wide-spectrum antitumor drug, was used as a model anticancer drug and loaded into the micelles using the solvent replacement approach. Free DOX is poorly water-soluble, limiting its development in clinical settings. In this study, DOX was encapsulated into the hydrophobic core of the micelles. In addition, electrostatic interactions between ChS and DOX also contribute to the DOX loading in the micelles. The pK_a constant of ChS is about 4.4–4.5 by pH titration³⁴ and the pK_a of DOX is about 8.6.³⁵ The electrostatic interactions governing the complex formation between ChS and DOX occur at a pH range that is within their pK_a values. Since DOX loading was performed at pH 8.0, the DOX molecules could form electrostatic interactions with the unreacted carboxyl groups on ChS. Consequently, such micelles hold a high loading capacity and encapsulation efficiency attributed to the hydrophobic effect and ionic bonding. The EE and DL of DOX is about 14.3% and 38.7% for PF127₁–ChS_0.5, 15.0% and 40.9% for PF127₁–ChS₁, 15.4% and 42.2% for PF127₁–ChS₂, and 15.6% and 42.8% for PF127₁–ChS₅, respectively. There is little difference in the drug loading efficiency, indicating that the hydrophobic DOX molecules are mainly accommodated in the hydrophobic PPO space of the PF127 molecules, and the electrostatic interactions between ChS and DOX had little effect on drug loading.

The pH dependent release performances of the PF127–ChS micelles were assessed at pH 5.5 and 7.4 at 37 °C throughout the experiment. With the same concentration of micelles (1 mg mL⁻¹), we examined the release behaviors of DOX from different formulations and the release profiles are shown in Fig. 7. In 57 h, about 43.8%, 30.7%, 29.8% and 32.8% of DOX was released from PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅ at pH 7.4, respectively. About 69.0%, 53.2%, 50.4% and 77.6% of DOX was released from PF127₁–ChS_0.5, PF127₁–ChS₁, PF127₁–ChS₂ and PF127₁–ChS₅ at pH 5.5, respectively. The samples, except for PF127₁–ChS_0.5, all showed a sustained and relatively slow release behavior of DOX at pH 7.4, while a relatively fast release behavior at pH 5.5. This may be because there is not an adequate amount of ChS in the PF127₁–ChS_0.5 micelles to retard DOX release. PF127₁–ChS₅ showed better release behavior, which is due to more cleavage of imine bonds at pH 5.5. It is known that a normal physiological pH is 7.4, and the pH of endocytic compartments, such as endosomes and lysosomes, are 5.5. Therefore, PF127–ChS micelles show the desired characteristic for drug release in cancer cells.

Fig. 7 DOX release behaviors from DOX loaded PF127–ChS micelles at 37 °C as a function of pH (pH 7.4 and 5.5): PF127₁–ChS_0.5 (a), PF127₁–ChS₁ (b), PF127₁–ChS₂ (c) and PF127₁–ChS₅ (d).

Cytotoxicity analysis

The safety of the synthesized micelles was determined through cell viability to ensure they do not harm cells during the internalization process. The results are shown in Fig. 8a. Human lung cancer A549 cells cultured without micelles were used as a blank value to normalize. The micelles were relatively non-toxic up to a concentration of 100 μg mL⁻¹ as the cell viability is more than 90% (about 91–106%). This is in relation to the biocompatibility of PF127 and ChS. In addition, it is reported that the hydrophilic polymer on the surface of the micelles can prevent polymer aggregation and nonspecific adhesion to the cell surface, resulting in lower cytotoxicity.³⁶ These results suggested that PF127–ChS micelles had low toxicity and may do less harm to normal cells. Therefore, this kind of micelles show considerable prospects in clinical applications.

Fig. 8 Cell viability measured by MTT assay for (a) PF127₁–ChS₅ at different concentrations and (b) DOX and DOX loaded PF127₁–ChS₅ micelles after 48 h of incubation with human lung cancer A549 cells.

The cytotoxicity of DOX and DOX loaded PF127₁–ChS₅ micelles on human lung cancer A549 cells is shown in Fig. 8b. DOX loaded PF127₁–ChS₅ micelles showed a similar anticancer activity to DOX, indicating that the DOX loaded micelles were efficiently taken up by the A549 cells. However, the IC₅₀ values of free DOX and the DOX loaded PF127₁–ChS₅ micelles is 37.5 and 55.3 μg mL⁻¹, respectively. The lower anticancer activity compared with free DOX may be caused by the time-consuming DOX release from micelles.

Confocal microscopy and flow cytometry analysis of cellular uptake

To evaluate the cellular uptake and sequential drug distribution of DOX and DOX loaded PF127–ChS, confocal laser scanning microscope (CLSM) analysis was performed with human lung cancer A549 cells. The results in Fig. 9 indicate that the both free DOX and DOX loaded PF127–ChS can be internalized into cells, shown by the red fluorescence in the cytoplasm. As small molecule uptake is faster than endocytosis, free DOX could be easily internalized into the cells, correspondingly the cells show stronger red fluorescence than those with the DOX loaded micelles. After 1 h of treatment of the cells with DOX loaded micelles, faint red fluorescence was observed in the cell cytoplasm, then, strong red fluorescence appeared in the nucleus of the treated cells after 4 h, indicating more DOX escaped from the DOX loaded micelles and was efficiently released inside cells.

Fig. 9 CLSM images of human lung cancer A549 cells incubated with DOX and DOX loaded PF127₁–ChS₅ micelles (10 μg mL⁻¹ of DOX) for 0.5 h, 1 h and 4 h.

The cellular uptake of DOX and DOX loaded PF127–ChS micelles by A549 cells was further evaluated by flow cytometry, as shown in Fig. 10. The fluorescence intensity augmented with the increase of incubation time. The cellular uptake of DOX loaded micelles was less than that of free DOX, which is in agreement with the foregoing MTT and CLSM results. The results indicated that PF127–ChS is efficient for delivering DOX into the cytoplasm as a drug carrier, however, it would take time to transport DOX into the nuclei and perform its action. In addition, with the incubation with DOX loaded PF127–ChS micelles, we observed a reduction in the number of cells, indicating that with more DOX entering into the cells, cells were gradually killed. This result is attributed to the affinity of ChS to CD44 receptors. CD44 is a transmembrane glycoprotein with extracellular, membrane and cytoplasmic domains, which can bind a variety of ECM molecules and contains attachment sites for ChS.³¹ Liu et al. synthesized ChS-g-poly(ε-caprolactone) micelles and encapsulated an anticancer drug, camptothecin, for tumor targeting delivery. To confirm the cellular uptake of the micelles occurred through CD44 receptor-mediated endocytosis, cells were pretreated with the CD44 antibody to block the CD44 receptor on the cell membrane. The flow cytometric diagrams showed the inhibition of intercellular uptake of micelles when the CD44 receptor was blocked.³⁷ These results showed that it is feasible to introduce ChS to PF127 micelles to achieve both passive and active targeting effects in this study.

Fig. 10 Flow cytometry images (A) of human lung cancer A549 cells incubated with DOX and DOX loaded PF127₁–ChS₅ micelles (10 μg mL⁻¹ of DOX) for 0.5 h, 1 h and 4 h. Quantitative analysis of flow cytometry images (B).

Conclusions

Micelles based on Pluronic F127 (PF127) and chondroitin sulfate (ChS) with both passive and active targeting properties were fabricated to specifically deliver DOX to tumors. A series of PF127–ChS micelles with different ratios of PF127 and ChS were formed via a facile and efficient method. The micelles have a mean diameter below 250 nm, narrow size distribution and negative zeta potential. Their critical micelle concentration (CMC) was significantly reduced in contrast to PF127 micelles and their stability was enhanced. DOX release from the micelles was higher at acidic pH values compared to physiological pH. The cytotoxicity assay determined that the micelles possess significantly lower toxicity. Confocal microscopy and flow cytometry analysis indicated that the micelles could efficiently release DOX inside cells by specific cellular uptake. These studies indicate that PF127–ChS micelles have potential application in triggered intracellular drug release.

Acknowledgements

The authors gratefully acknowledge the financial support of the National Natural Science Foundation of China (grant no. 51541304, 51273086, 51503091), Special Doctorial Program Fund from the Ministry of Education of China (grant no. 20130211110017), and the Fundamental Research Funds for the Central Universities (grant no. lzujbky-2015-26).

Notes and references

1. S. Biswas, P. Kumari, P. M. Lakhani and B. Ghosh, Eur. J. Pharm. Sci., 2016, 83, 184–202.
2. C. Deng, Y. Jiang, R. Cheng, F. Meng and Z. Zhong, Nano Today, 2012, 7, 467–480.
3. Y. Xiao, H. Hong, A. Javadi, J. W. Engle, W. Xu, Y. Yang, Y. Zhang, T. E. Barnhart, W. Cai and S. Gong, Biomaterials, 2012, 33, 3071–3082.
4. M. Moussa, S. N. Goldberg, G. Kumar, R. R. Sawant, T. Levchenko, V. P. Torchilin and M. Ahmed, PLoS One, 2014, 9, e102727.
5. H. Wei, R.-X. Zhuo and X.-Z. Zhang, Prog. Polym. Sci., 2013, 38, 503–535.
6. R. Gupta, J. Shea, C. Scafe, A. Shurlygina and N. Rapoport, J. Controlled Release, 2015, 212, 70–77.
7. H. S. Yoo and T. G. Park, J. Controlled Release, 2004, 96, 273–283.
8. W. Cao, J. Zhou, A. Mann, Y. Wang and L. Zhu, Biomacromolecules, 2011, 12, 2697–2707.
9. N. Kamaly, Z. Xiao, P. M. Valencia, A. F. Radovic-Moreno and O. C. Farokhzad, Chem. Soc. Rev., 2012, 41, 2971–3010.
10. Y. Zhong, F. Meng, C. Deng and Z. Zhong, Biomacromolecules, 2014, 15, 1955–1969.
11. A. S. Manjappa, K. R. Chaudhari, M. P. Venkataraju, P. Dantuluri, B. Nanda, C. Sidda, K. K. Sawant and R. S. Ramachandra Murthy, J. Controlled Release, 2011, 150, 2–22.
12. D. A. Chiappetta and A. Sosnik, Eur. J. Pharm. Biopharm., 2007, 66, 303–317.
13. W. Zhang, Y. Shi, Y. Chen, J. Ye, X. Sha and X. Fang, Biomaterials, 2011, 32, 2894–2906.
14. D. Attwood, J. Collett and C. Tait, Int. J. Pharm., 1985, 26, 25–33.
15. S. Danson, D. Ferry, V. Alakhov, J. Margison, D. Kerr, D. Jowle, M. Brampton, G. Halbert and M. Ranson, Br. J. Cancer, 2004, 90, 2085–2091.
16. J. W. Valle, A. Armstrong, C. Newman, V. Alakhov, G. Pietrzynski, J. Brewer, S. Campbell, P. Corrie, E. K. Rowinsky and M. Ranson, Invest. New Drugs, 2010, 29, 1029–1037.
17. K. H. Bae, S. H. Choi, S. Y. Park, Y. Lee and T. G. Park, Langmuir, 2006, 22, 6380–6384.
18. J. D. Pruitt, G. Husseini, N. Rapoport and W. G. Pitt, Macromolecules, 2000, 33, 9306–9309.
19. D. Uebelhart, J.-M. T. Eugene, P. D. Delmas, A. Chantraine and E. Vignon, Osteoarthritis Cartilage, 1998, 6, 39–46.
20. A. J. Bollet, J. R. Handy and B. C. Sturgill, J. Clin. Invest., 1963, 42, 853.
21. F. Ronca, L. Palmieri, P. Panicucci and G. Ronca, OsteoarthritisCartilage, 1998, 6, 14–21.
22. Y.-L. Chen, H.-P. Lee, H.-Y. Chan, L.-Y. Sung, H.-C. Chen and Y.-C. Hu, Biomaterials, 2007, 28, 2294–2305.
23. H. Tan, C. R. Chu, K. A. Payne and K. G. Marra, Biomaterials, 2009, 30, 2499–2506.
24. S. Lü, C. Gao, X. Xu, X. Bai, H. Duan, N. Gao, C. Feng, Y. Xiong and M. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 13029–13037.
25. X. Guo, C. Shi, J. Wang, S. Di and S. Zhou, Biomaterials, 2013, 34, 4544–4554.
26. N. Gao, S. Lü, C. Gao, X. Wang, X. Xu, X. Bai, C. Feng and M. Liu, Chem. Eng. J., 2016, 287, 20–29.
27. S. Lü, B. Li, B. Ni, Z. Sun, M. Liu and Q. Wang, Soft Matter, 2011, 7, 10763–10772.
28. X.-H. Peng, Y. Wang, D. Huang, Y. Wang, H. J. Shin, Z. Chen, M. B. Spewak, H. Mao, X. Wang, Y. Wang, Z. Chen, S. Nie and D. M. Shin, ACS Nano, 2011, 5, 9480–9493.
29. C. Duan, J. Gao, D. Zhang, L. Jia, Y. Liu, D. Zheng, G. Liu, X. Tian, F. Wang and Q. Zhang, Biomacromolecules, 2011, 12, 4335–4343.
30. Y. Patil, T. Sadhukha, L. Ma and J. Panyam, J. Controlled Release, 2009, 136, 21–29.
31. A. Pathak, P. Kumar, K. Chuttani, S. Jain, A. K. Mishra, S. P. Vyas and K. C. Gupta, ACS Nano, 2009, 3, 1493–1505.
32. L. Y. Qiu, R. J. Wang, C. Zheng, Y. Jin and L. Q. Jin, Nanomedicine, 2010, 5, 193–208.
33. J. W. Wojtkowiak, D. Verduzco, K. J. Schramm and R. J. Gillies, Mol. Pharmaceutics, 2011, 8, 2032–2038.
34. A. Nakajima and K. Shinoda, J. Colloid Interface Sci., 1976, 55, 126–132.
35. L. Mayer, M. Bally and P. Cullis, Biochim. Biophys. Acta, Biomembr., 1986, 857, 123–126.
36. M. Ogris, S. Brunner, S. Schüller, R. Kircheis and E. Wagner, Gene Ther., 1999, 6, 595–605.
37. Y.-S. Liu, C.-C. Chiu, H.-Y. Chen, S.-H. Chen and L.-F. Wang, Mol. Pharm., 2014, 11, 1164–1175.

---