Controlled Release of Ionic Drugs from Complex Micelles with

Mar 19, 2012 - Jianzu Wang , Yiqing Song , Pingchuan Sun , Yingli An , Zhenkun Zhang ... Aoting Qu , Huiru Yang , Zhenkun Zhang , Yingli An , and Linq...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Biomac

Controlled Release of Ionic Drugs from Complex Micelles with Charged Channels Xiaojun Liu, Rujiang Ma,* Junyang Shen, Yanshuang Xu, Yingli An, and Linqi Shi* Key Laboratory of Functional Polymer Materials, Ministry of Education, and Institute of Polymer Chemistry, Nankai University, Tianjin 300071, China S Supporting Information *

ABSTRACT: Oral administration of ionic drugs generally encounters with significant fluctuation in plasma concentration due to the large variation of pH value in the gastrointestinal tract and the pH-dependent solubility of ionic drugs. Polymeric complex micelles with charged channels on the surface provided us with an effective way to reduce the difference in the drug release rate upon change in pH value. The complex micelles were prepared by self-assembly of PCL-b-PAsp and PCL-b-PNIPAM in water at room temperature with PCL as the core and PAsp/PNIPAM as the mixed shell. With an increase in temperature, PNIPAM collapsed and enclosed the PCL core, while PAsp penetrated through the PNIPAM shell, leading to the formation of negatively charged PAsp channels on the micelle surface. Release behavior of ionic drugs from the complex micelles was remarkably different from that of usual core− shell micelles where diffusion and solubility of drugs played a key role. Specifically, it was mainly dependent on the conformation of the PAsp chains and the electrostatic interaction between PAsp and drugs, which could partially counteract the influence of pH-dependent diffusion and solubility of drugs. As a result, the variation of drug release rate with pH value was suppressed, which was favorable for acquiring relatively steady plasma drug concentration.

1. INTRODUCTION Over the past several decades, as a kind of drug carrier, polymeric micelles self-assembled from block copolymers have received enormous interest due to their capability to increase the solubility of insoluble drugs as well as their specific core− shell structure and good stability under physiological conditions.1−4 Typically, polymeric micelle-based drug carriers consist of a biodegradable polyester core (e.g., poly(lactide) or poly(ε-caprolactone)) and a hydrophilic poly(ethylene glycol) shell that is biocompatible and nontoxic to the human body.5 Normally, for drugs physically encapsulated in the core of sufficiently stable polymeric micelles, release is controlled by the rate of drug diffusion in the micellar core.6 Lots of drugs for treatment of some diseases are ionic, such as propranolol,7 diltiazem,8 and captopril,9 for cardiovascular and cerebrovascular diseases, which have seriously threatened human health. For ionic drugs (refers to ionic drugs whose ionicity is reflected by an amine or carboxyl group in this paper), their solubility in aqueous solution usually varies remarkably with pH values. When polymeric micelles as drug carriers are applied to an oral system, they pass through the complicated environment of the human body. For instance, it is highly acidic in the stomach while nearly neutral in the intestine. Moreover, the pH value of the intestine is not constant.10 Due to the pH-dependent solubility, the release rates of ionic drugs from polymeric micelles are not constant and will change with surrounding environment of the carriers.11 Such a pHdependent release behavior is unfavorable for treatment of some © 2012 American Chemical Society

diseases, which require steady plasma drug concentration. Drug carriers with relatively steady drug release rate are desired. However, pH-dependent release behavior of ionic drugs existed in micelles with simple core−shell structure. Yang et al.12 prepared micelles based on poly(ε-caprolactone)-b-poly(ethylene glycol)-b-poly(εcaprolactone) (PCL-b-PEG-b-PCL) triblock copolymers. Doxorubicin (DOX) was loaded in the PCL core as a model drug. The results showed that release rate of DOX from micelles at pH 7 was remarkably slower than that at pH 5.5. To advance functions of polymeric micelles, micelles with core−shell-corona three-layered structure have been developed and demonstrated superior functions compared to their simple core−shell counterparts. A key advantage of core−shell-corona micelles is that channels can be created in the shell,13 which could be applied in controlled drug release and other aspects. For example, Wu et al.14 prepared complex micelles by selfassembly of poly(lactide)-block-poly(ethylene glycol) (PLA-bPEG) and poly(lactide)-block-poly(N-isopropylacrylamide) (PLA-b-PNIPAM) with a PLA core and a mixed PEG/PNIPAM shell at room temperature. By increasing the temperature, the PNIPAM chains collapsed onto the PLA core, while the PEG chains remained soluble and stretched through the collapsed PNIPAM layer, leading to the formation of PEG channels. Received: December 23, 2011 Revised: March 16, 2012 Published: March 19, 2012 1307

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

Scheme 1. Illustration of Formation of Complex Micelles with PAsp and PEG Channels

added. After freeze-degas-thaw cycles, polymerization was performed at 110 °C for 12 h. Then the mixture was diluted with dichloromethane and then precipitated into excess diethyl ether. The precipitate was dried under vacuum. 2.2.2. Synthesis of PCL-b-PNIPAM. Synthesis of PCL-b-PINPAM was illustrated in Scheme 2B. PCL-Br macroinitiator was synthesized by ROP of ε-caprolactone using 2-hydroxyethyl 2-bromoisobutyrate (HEBIB) as the initiator and Sn(Oct)2 as the catalyst in toluene solution. Briefly, HEBIB (0.21 g, 1 mmol), ε-caprolactone (6.8 g, 60 mmol), and Sn(Oct)2 (0.48 g, 1.2 mmol) were added into the reaction flask and then 13 mL of toluene was added. After freeze-degas-thaw cycles, polymerization was conducted at 110 °C for 12 h. Then the mixture was diluted with dichloromethane and then precipitated into excess diethyl ether. PCL-b-PNIPAM was synthesized by atom transfer radical polymerization (ATRP) using PCL-Br as the initiator and CuCl/Me6Tren as the catalyst in N,N-dimethylformamide (DMF) solution. Briefly, PCLBr (2.0 g), CuCl (0.045 g), Me6TREN (0.11 g), and NIPAM (3.6 g) were added into the reaction flask and then 8 mL of DMF was added. After freeze-degas-thaw cycles, polymerization was performed at 45 °C for 24 h. The mixture was purified by passing through an Al2O3 column to remove the copper catalyst followed by precipitation in excess diethyl ether. 2.2.3. Synthesis of PCL-b-PAsp. PCL-b-PAsp was synthesized according to ref 19, illustrated in Scheme 2C. Synthesis of PCL-NH2 Macroinitiator. PCL-NHBoc was synthesized by ROP of ε-caprolactone using N-(tert-butoxycarbonyl)ethanolamine as the initiator and Sn(Oct)2 as the catalyst in toluene solution. Briefly, N-(tert-butoxycarbonyl)ethanolamine (0.16 g, 1 mmol), ε-caprolactone (6.8 g, 60 mmol), and Sn(Oct)2 (0.48 g, 1.2 mmol) were added into the reaction flask and then 13 mL of toluene was added. After freeze-degas-thaw cycles, polymerization was conducted at 110 °C for 12 h. Then the mixture was diluted with dichloromethane and precipitated into excess diethyl ether. The removal of the Boc group of PCL-NHBoc was performed by treatment in the mixing solvent of trifluoroacetic acid and dichloromethane (1/1, V/V) at room temperature for 12 h, and then the mixture was precipitated into excess diethyl ether. The precipitate was then treated in the mixing solvent of triethylamine (TEA) and dichloromethane (1/1, V/V) at room temperature for 12 h, and then the mixture was precipitated into excess diethyl ether to obtain the macroinitiator PCL-NH2. Synthesis of PCL-b-PAsp Diblock Copolymer. PCL-b-PAsp(OBzl) was synthesized by ROP of Asp(OBzl)-NCA using PCL-NH2 as the initiator in dichloromethane solution. Briefly, PCL-NH2 (1.0 g) and

The collapsed PNIPAM layer could not only restrain burst drug release, but also control degradation of the PLA core. Complex micelles with PEG channels possessed excellent functions such as suppressing burst drug release and controlling degradation of the polyester core. However, aforementioned pH-dependent release behavior would still exist. Herein, we show the first example of complex micelles with electronegative PAsp channels prepared by self-assembly of poly(ε-caprolactone)block-poly(aspartic acid) (PCL-b-PAsp) and poly(ε-caprolactone)block-poly(N-isopropylacrylamide) (PCL-b-PNIPAM) in aqueous solution (Scheme 1). PAsp could ionize and its ionization degree and conformation could vary with pH values in aqueous solution. Because of the variation of conformation of PAsp chains and the interaction between charges of PAsp and ionic drugs at different pH values, it was predicted that release behavior of ionic drugs from complex micelles with charged channels would be remarkably different from these with uncharged PEG channels. Our goal is to reduce pH dependent release behavior of ionic drugs through the interaction between charges and the variation of configuration of PAsp chains at different pH values.

2. EXPERIMENTAL SECTION 2.1. Materials. ε-Caprolactone from Alfa was distilled under reduced pressure before use. Methoxy poly(ethylene glycol) (CH3O-PEG113-OH) (Mn = 5000 and the polydispersity index (PDI) = 1.05) was purchased from Aldrich and dried in a vacuum. 2-Hydroxyethyl 2-bromoisobutyrate (HEBIB) was synthesized according to ref 15. N-Carboxy-β-benzylL-aspartate anhydride (Asp(OBzl)-NCA) was synthesized according to ref 16. CuCl was purchased from Aldrich and purified according to ref 17. Tris[2-(dimethylamino)-ethyl]amine (Me6TREN) was synthesized according to ref 18. N-Isopropylacrylamide (NIPAM), from Aldrich, was purified by recrystallization from hexane and dried in a vacuum. N-(tertButoxycarbonyl)ethanolamine (Alfa), stannous octoate (Sn(Oct)2; Alfa), trifluoroacetic acid (Alfa), and trifluoromethanesulfonic acid (Alfa) were used without further purification. All solvents were redistilled before use. 2.2.1. Synthesis of the Diblock Copolymers. Synthesis of PCL-b-PEG. Synthesis of PCL-b-PEG was illustrated in Scheme 2A. PCL-b-PEG was synthesized by ring-opening polymerization (ROP) of ε-caprolactone using PEG-OH as the initiator and Sn(Oct)2 as the catalyst in toluene solution. Briefly, PEG-OH (2.0 g, 0.4 mmol), ε-caprolactone (3.65 g, 32 mmol), and Sn(Oct)2 (0.19 g, 0.48 mmol) were added into the reaction flask and then 10 mL of toluene was 1308

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

added dropwise into the polymer solution under vigorous stirring until opalescence appeared, indicating the formation of micelles. The solution was stirred overnight and then dialyzed against water to remove DMF. Micelles with different composition ratios of block copolymers were prepared by the same way. 2.4. Drug Release Experiments. Drug-loaded micelles were prepared by adding distilled water dropwise into the polymer and drug (ibuprofen or doxorubicin (the free base, not the salt with HCl)) solution in DMF under vigorous stirring until opalescence appeared, indicating the formation of micelles. The solution was stirred overnight and then dialyzed against water to remove DMF. DOX and ibuprofen both have very low solubility in neutral water so they could be loaded in the hydrophobic PCL core through hydrophobic interaction during the formation of the micelles in neutral water. Finally, the micelle solution was filtered through a 0.45 μm Millipore filter to remove unencapsulated drug particles. The drug release experiment was performed in acetate buffer solution (pH 5.7, 0.05 M) and in phosphate buffer solution (pH 8.0, 0.05 M), respectively. Briefly, 4 mL of the micelle solution was transferred into a dialysis bag (molecular weight cut off: 12−14 KD), then the bag was immersed in 16 mL of buffer solution at 37 °C. Periodically, 4 mL of the solution outside the dialysis bag was taken out for UV−vis measurements. The volume of solution was kept constant by adding 4 mL of original buffer solution after each sampling. The amount of drug was measured using a UV−vis spectrophotometer at 266 nm for ibuprofen and at 583 nm for DOX, respectively. 2.5. Characterizations. 1H nuclear magnetic resonance (1H NMR) spectra were recorded on a Varian UNITY-plus 400 M NMR spectrometer at room temperature with CDCl3 and DMSO-d6 as solvents and TMS as a reference. Relative molecular weights and molecular weight distributions of block copolymers were measured by gel permeation chromatography (GPC) at 35 °C with a Waters 1525 chromatograph equipped with a Waters 2414 refractive index detector. GPC measurements were carried out using THF and DMF as eluents with a flow rate of 1.0 mL/ min, respectively. Polystyrene standards were used for calibration. Dynamic light scattering (DLS) and static light scattering (SLS) experiments were performed on a laser light scattering spectrometer (BI200SM) equipped with a digital correlator (BI-9000AT) at 532 nm at given temperatures. All the samples were obtained by filtering through a 0.45 μm Millipore filter into a clean scintillation vial. Transmission electron microscopy (TEM) measurements were conducted using a Philips T20ST electron microscope at an acceleration voltage of 200 kV. To prepare the TEM samples, the sample solution was dropped onto a carbon-coated copper grid and dried slowly in air.

Scheme 2. Synthesis of PEG-b-PCL (A), PCL-b-PNIPAM (B) and PCL-b-PAsp (C) diblock copolymers

3. RESULTS AND DISCUSSION 3.1. Synthesis of Three Block Copolymers. The block copolymer PCL-b-PAsp was synthesized by sequential ROP of ε-caprolactone and Asp (OBzl)-NCA. N-(tert-Butoxycarbonyl)ethanolamine was first used to initiate the ROP of ε-caprolactone to obtain the polymer PCL-NHBoc. After removal of the Boc group and desalinization, the macroinitiator PCL-NH2 was obtained and used to initiate the ROP of Asp(OBzl)-NCA to get the polymer PCL-b-PAsp(OBzl). After removal of the benzyl group, the block copolymer PCL-b-PAsp was obtained. Figure 1A,B shows 1 H NMR spectra of PCL-NHBoc and PCL-b-PAsp(OBzl) in CDCl3. Figure 1C shows 1H NMR spectra of PCL-b-PAsp in DMSO-d6. Based on the relative intensities, the block copolymer was denoted as PCL45-b-PAsp60. The polydispersity index of PCLb-PAsp measured by GPC using DMF as the eluent was 1.16 (see the Supporting Information). 1 H NMR spectra of PCL-b-PEG, PCL-Br, and PCL-b-PNIPAM in CDCl3 are shown in the Supporting Information. Based on the relative intensities, the two block copolymers were denoted as PCL62-b-PEG113 and PCL45-b-PNIPAM55, respectively. The polydispersity index of PCL-b-PEG and PCL-b-PNIPAM measured by GPC using THF as the eluent was 1.23 and 1.34, respectively.

Asp(OBzl)-NCA (3.6 g) were added into the reaction flask, and then 15 mL of dichloromethane was added. After freeze-degas-thaw cycles, polymerization was performed at 30 °C for 24 h. Then the mixture was diluted with dichloromethane and then precipitated into excess diethyl ether to obtain the polymer PCL-b-PAsp (OBzl). Then PCL-b-PAsp (OBzl) was treated with a mixture of trifluoroacetic/trifluoromethanesulfonic acid/anisole to remove the benzyl group. The mixture was then gently stirred at 0 °C for 2 h and then precipitated into excess diethyl ether to obtain the polymer PCL-b-PAsp. 2.3. Preparation of the Polymeric Micelles. The block copolymers PCL-b-PEG, PCL-b-PNIPAM and PCL-b-PAsp were first dissolved in DMF to make the polymer solution with a concentration of 1.0 g/L, respectively. Subsequently, a given amount of distilled water was 1309

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

complex micelles, respectively. All of the micellar solutions have the same polymer concentration of 0.1 g/L. Figure 2 shows the

Figure 2. Hydrodynamic diameter distribution f(Dh) of (A) PCL-bPEG/PCL-b-PNIPAM complex micelles, (B) PCL-b-PAsp/PCL-bPNIPAM complex micelles, (C) PCL-b-PEG individual micelles, and (D) PCL-b-PAsp individual micelles measured at the scattering angle of 90° at 25 °C.

hydrodynamic diameter distributions of the four kinds of blank micelles (drugs not loaded) measured by DLS at the scattering angle of 90° at 25 °C. Obviously, each of the micelles has a narrow diameter distribution and the average hydrodynamic diameters (Dh) of these micelles are 109.2, 120.7, 143.9, and 149.0 nm, respectively. TEM images (see the Supporting Information) of them supported the DLS results. To prove the thermoresponsive behavior of the complex micelles and the formation of channels in the complex micelles, the Rh (Rh = 0.5 Dh) and the radii of gyration (Rg) of the four kinds of micelles were measured by DLS and SLS at 25 and 37 °C. The results are listed in Table 1. The Rg/Rh value can Table 1. DLS and SLS Data for the Four Kinds of Micelles Measured at 25 and 37 °C PCL-b-PEG micelles, 25 °C PCL-b-PEG micelles, 37 °C PCL-b-PEG/PCL-b-PNIPAM micelles, 25 °C PCL-b-PEG/PCL-b-PNIPAM micelles, 37 °C PCL-b-PAsp micelles, 25 °C PCL-b-PAsp micelles, 37 °C PCL-b-PAsp/PCL-b-PNIPAM micelles, 25 °C PCL-b-PAsp/PCL-b-PNIPAM micelles, 37 °C

Rh (nm)

Rg (nm)

Rg/Rh

82.8 81.2 67.3 64.5 73.6 72.5 62.7 60.3

63.2 63.9 60.6 49.2 51.3 52.9 52.9 43.8

0.76 0.79 0.90 0.76 0.70 0.73 0.84 0.73

reflect the morphology of nanoparticles in solution.20 It can be seen that for individual micelles, the Rh and Rg values remain nearly invariable after increase of temperature, indicating that the micelle structure changes little. But for the complex micelles, the Rg and Rg/Rh values decreases remarkably at higher temperature, indicating that the structure of the complex micelles became more compact. Our group did detailed characterization and analysis on the formation of channels in complex micelles.13,21 Similarly, the results above reveal the collapse of the PNIPAM chains which form a hydrophobic shell around the PCL core. The PEG or PAsp chains are still soluble at higher temperature and they can penetrate through the PNIPAM shell forming hydrophilic or charged channels. 1H NMR spectra of the PCL-b-PAsp/PCL-b-PNIPAM

Figure 1. 1H NMR spectra of PCL-NHBoc (A), PCL-b-PAsp(OBzl) (B) in CDCl3, and PCL-b-PAsp (C) in DMSO-d6.

3.2. Formation of the Micelles. In this paper, except where mentioned otherwise, four kinds of micelles were employed, which were PCL-b-PEG individual micelles, PCL-b-PAsp individual micelles, PCL-b-PEG/PCL-b-PNIPAM (1:1, mol/mol) complex micelles and PCL-b-PAsp/PCL-b-PNIPAM (1:1, mol/mol) 1310

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

Table 2. Hydrodynamic Diameter (Dh), Loading Efficiency, and Loading Content of the Drug-Loaded Micelles

PCL-b-PEG micelles PCL-b-PAsp micelles PCL-b-PEG/ PCL-bPNIPAM micelles PCL-b-PAsp/ PCL-bPNIPAM micelles

Dh of blank micelles (nm)

Dh of DOX loaded micelles (nm)

Dh of ibuprofen loaded micelles (nm)

loading efficiencya/loading contentb of DOX (%)

loading efficiency/loading content of ibuprofen (%)

143.9 149.0 120.7

151.6 160.2 132.4

153.2 158.7 135.3

12.8/8.8 11.2/7.7 11.6/8.0

15.9/10.7 14.7/9.9 14.3/9.7

109.2

120.3

118.6

12.3/8.4

15.1/10.2

a

The loading efficiency was defined as the ratio of the weight of loaded drug to the initially added drug. bThe loading content was defined as the ratio of the weight of the loaded drug to the weight of drug and copolymers in drug-loaded micelle solution.

Figure 3. Release profiles of (A) ibuprofen and (B) DOX from PCL-b-PEG micelles and PCL-b-PEG/PCL-b-PNIPAM complex micelles at pH 5.7 and 8.0 at 37 °C, respectively.

Figure 4. Release profiles of (A) ibuprofen and (B) DOX from PCL-b-PEG micelles, PCL-b-PAsp micelles, PCL-b-PEG/PCL-b-PNIPAM complex micelles, and PCL-b-PAsp/PCL-b-PNIPAM complex micelles at pH 8.0 at 37 °C, respectively.

complex micelles above and below LCST of PNIPAM (see the Supporting Information) also prove the thermoresponsive behavior of the complex micelles. It can be seen that at 37 °C PNIPAM completely loses the mobility, while PAsp remains the mobility. 3.3. Release of DOX and Ibuprofen from the Micelles. Table 2 shows the hydrodynamic diameter (Dh), loading efficiency, and loading content of the drug-loaded micelles. It can be seen that the Dh of the drug-loaded micelles are slightly larger than that of corresponding blank micelles. The drug release experiment was performed in acetate buffer solution (pH 5.7, 0.05 M) and in phosphate buffer solution (pH 8.0, 0.05 M), respectively. Figure 3 shows the release profiles of ibuprofen and DOX from PCL-b-PEG micelles and PCL-b-PEG/PCL-b-PNIPAM complex micelles at pH 5.7 and 8.0 at 37 °C, respectively. It could be seen that for the same micelles the release rate of ibuprofen at pH 8.0 was remarkably faster than that at pH 5.7 but the release rate of DOX at pH 8.0 was markedly slower than that at pH 5.7. This is because ibuprofen is a kind of anionic drug and its solubility in alkaline water is relatively higher than that in acidic water while DOX is a kind of cationic drug and its solubility in alkaline water is relatively lower than that in acidic water (solubility of ibuprofen

at pH 5.7, 7.0, and 8.0 are 0.005, 0.093, and 0.930 mg/mL, respectively; solubility of DOX at pH 5.7, 7.0, and 8.0 are 0.940, 0.047, and 0.005 mg/mL, respectively). These results show that the release behavior of ionic drugs from complex micelles with PEG channels is markedly pH-sensitive. In addition, the release rate of ibuprofen and DOX from individual micelles was faster than that of the corresponding complex micelles. This is because the PNIPAM chains could collapse onto the PCL core at 37 °C, which could hinder the release of drugs from the micelles. Moreover, the release rate of ibuprofen and DOX from PCL-b-PEG/PCL-b-PNIPAM and PCL-b-PAsp/ PCL-b-PNIPAM complex micelles at 25 and 37 °C (see the Supporting Information) showed that the drug release rate at 25 °C was faster than that at 37 °C. This is because PNIPAM is soluble at 25 °C, while insoluble, and could collapse onto the PCL core at 37 °C. The collapsed PNIPAM layer could decrease the drug release rate. Figure 4 shows the release profiles of ibuprofen and DOX (pK of PAsp, DOX, and ibuprofen are 4.4, 8.6, and 4.6) from the four kinds of micelles mentioned above at pH 8.0 at 37 °C. It could be seen in Figure 4A that the release rate of ibuprofen from the micelles with PAsp shell or channels was slower than 1311

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

that with PEG shell or channels. This is because the PAsp chains and the ionized ibuprofen are both electronegative in aqueous solution, then they could repel each other and the mutual repulsion between them could decrease the release rate of ibuprofen (illustrated in Scheme 3A). Conversely, the release

accelerate the migration of DOX from inner core to the outside through long-range electrostatic interactions. The diffusion of DOX from hydrophobic inner core to its surface was a slow process, which might be the controlling step of the entire drug release process. When DOX came out of the core, they would combine with PAsp through electrostatic interactions, but this combination was not very strong. At the same time, the PAsp chains were stretched and incompact with a large number of water molecules surrounding them, which is favorable for the diffusion of DOX from the surface of micellar core to the solvent through ion exchange. Additionally, rapid diffusion of DOX to the solution can also be ascribed to the gradient of DOX concentration between the surface of the micellar core and the solvent. Overall, the release rate of DOX from the micelles with PAsp channels was faster than that with PEG channels. Figure 5 shows the release profiles of ibuprofen and DOX from PCL-b-PAsp/PCL-b-PNIPAM complex micelles with different ratios of PCL-b-PAsp to PCL-b-PNIPAM at pH 8.0 at 37 °C, respectively. Obviously, the drug release rate increased with decreasing ratio of PCL-b-PNIPAM to PCL-b-PAsp. This is because the density of PAsp channels increased with decreasing the content of PNIPAM. As a result, the drug release rate rose. Here, the drug release rate was controlled by the density of channels. Figure 3 shows that the release behavior of ionic drugs from micelles with PEG chains or channels is markedly pH-sensitive. Actually, the obvious difference of drug release rate under different pH values might be unfavorable for treatment of some diseases since relatively steady drug release is desired. Figure 6 shows the release profiles of ibuprofen and DOX at pH 5.7 and 8.0 at 37 °C from PCL-b-PEG individual micelles and PCL-bPAsp/PCL-b-PNIPAM complex micelles, respectively. It could be seen that compared with micelles bearing PEG chains (Figure 3), the difference between the release rate of ibuprofen or DOX from micelles bearing PAsp channels at pH 8.0 and that of pH 5.7 is relatively smaller. This means that the release rate of ibuprofen or DOX from micelles with electronegative PAsp channels did not increase remarkably, while their solubility in water enhanced obviously. For the release of ibuprofen through PAsp channels, interactions between ibuprofen and partially charged PAsp were weak because ibuprofen was almost neutral at lower pH value. So the release of ibuprofen was mainly controlled by diffusion. At higher pH value, ibuprofen became negatively charged and its solubility increased dramatically. But the release rate did not increase markedly due to the electrostatic repulsion between like-charged ibuprofen and PAsp (shown in Scheme 3A). In the case of DOX, although its solubility was higher at lower pH value, PAsp was in

Scheme 3. (A) Illustration of Release of Ibuprofen from PCL-b-PAsp/PCL-b-PNIPAM Complex Micelles at 37 °C; (B) Illustration of Contraction of PAsp Chains of PCL-bPAsp/PCL-b-PNIPAM Complex Micelles Due to the Decrease of pH Values

rate of DOX from the micelles with PAsp shell or channels was faster than that with PEG shell or channels as shown in Figure 4B. This may be ascribed to two aspects. On one hand, at pH 8.0, the PAsp chains possessed high dissociation extent so they were highly hydrophilic and in the stretched state. Thus, the size of PAsp channels was relatively larger, leading to more free space for drugs to pass through. On the other hand, electrostatic interaction between PAsp and DOX may pull DOX from inner core to outside, which could accelerate the release of DOX. To prove this hypothesis, zeta potential of the PCL-b-PAsp/PCL-b-PNIPAM micelles before and after loading DOX were determined as −25.85 and −25.71 mV at pH 7.4, respectively. Zeta potential varied little after loading DOX, indicating that the great majority of DOX was loaded in the inner core of micelles. PAsp could

Figure 5. Release profiles of (A) ibuprofen and (B) DOX at pH 8.0 at 37 °C from PCL-b-PAsp/PCL-b-PNIPAM complex micelles with different ratios of PCL-b-PAsp to PCL-b-PNIPAM (a) 6/4, mol/mol, (b) 5/5, mol/mol, (c) 4/6, mol/mol, respectively. 1312

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

Figure 6. Release profiles of (A) ibuprofen and (B) DOX from PCL-b-PEG micelles and PCL-b-PAsp/PCL-b-PNIPAM complex micelles at pH 5.7 and 8.0 at 37 °C, respectively.

■

the coil state (illustrated in Scheme 3B) and the size of PAsp channels would become smaller, which decreased the effective space in the channels so as to slow down the release rate of DOX. Thus, rapid release was suppressed. At higher pH value, its solubility was lower, but PAsp became more negatively charged and expanded into linear state. Thus, the size of PAsp channels would become larger and the effective space in the channels increased, which could partially counteract the influence of decreasing solubility of DOX. Based on the overall effect of the factors above, the pH sensitivity of release behavior of ionic drugs from complex micelles was reduced effectively to a certain extent, which is favorable for obtaining a relatively steady release profile upon variation of pH value. Moreover, it is shown from Figures 3−6 that the release rate of DOX is remarkably slower than that of ibuprofen. The remarkable difference of drug release rate of DOX and ibuprofen is ascribed to the chemical structure of them. DOX is composed of a very large hydrophobic group and an amidocyanogen which could ionize while ibuprofen is composed of a remarkably smaller hydrophobic group and a carboxyl group which could ionize. Thus, diffusion rate of ibuprofen is much faster than DOX. As a result, the release of DOX is remarkably slower than the release of ibuprofen.

ASSOCIATED CONTENT

S Supporting Information *

Figure S-1: 1H NMR spectra of PCL-b-PEG (A), PCL-Br (B), and PCL-b-PNIPAM (C) in CDCl3. Figure S-2: TEM images of (A) PCL-b-PEG/PCL-b-PNIPAM complex micelles, (B) PCLb-PAsp/PCL-b-PNIPAM complex micelles, (C) PCL-b-PEG individual micelles, and (D) PCL-b-PAsp individual micelles. Figure S-3: Release profiles of (A) ibuprofen and (B) DOX from PCL-b-PEG/PCL-b-PNIPAM complex micelles and PCLb-PAsp/PCL-b-PNIPAM complex micelles at 25 and 37 °C, respectively. Figure S-4: 1H NMR spectra of the PCL-b-PAsp/ PCL-b-PNIPAM complex micelles at 25 and 37 °C. Figure S-5: GPC trace of PCL-b-PAsp in DMF at room temperature. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the National Natural Science Foundation of China (No. 91127045, 50830103, and 20904025) and the National Basic Research Program of China (973 Program, No. 2011CB932500) for financial support.

4. CONCLUSIONS Compared with usual core−shell micelles, complex micelles with charged PAsp channels could control the drug release not only through diffusion but also through the interaction between charges and the variation of conformation of the PAsp chains at different pH values. Specifically, for anionic drugs, at lower pH values, they were almost uncharged and their release was mainly controlled by diffusion. At higher pH values, their solubility was higher, but the release rate increased little due to mutual repulsion between the PAsp chains and drugs. In the case of cationic drugs, at lower pH values, although their solubility was higher, fast release was restrained since the PAsp chains would contract, leading to less free space for drugs to pass through. At higher pH values, their solubility was lower, but the release rate decreased little because PAsp chains were more negatively charged and in the stretched state so that PAsp channels would supply more effective space. Overall, the release rate of ionic drugs from complex micelles with negatively charged PAsp channels did not vary remarkably with the pH values to some extent. Thus, relatively steady drug release under different pH values was achieved. Complex micelles with charged channels may find application in drug delivery for treatment of some diseases which require relatively steady drug release.

■

REFERENCES

(1) Rösler, A.; Vandermeulen, G. W. M.; Klok, H. Adv. Drug Delivery Rev. 2001, 53, 95−108. (2) Adams, M. L.; Lavasanifar, A.; Kwon, G. S. J. Pharm. Sci. 2003, 92, 1343−1355. (3) Allen, T. M.; Cullis, P. R. Science 2004, 303, 1818−1822. (4) Jones, M.; Leroux, J. Eur. J. Pharm. Biopharm. 1999, 48, 101−111. (5) Otsuka, H.; Nagasaki, Y.; Kataoka, K. Adv. Drug Delivery Rev. 2003, 55, 403−419. (6) Lavasanifar, A.; Samuel, J.; Kwon, G. S. Adv. Drug Delivery Rev. 2002, 54, 169−190. (7) Sui, J. J.; Zhang, J. H.; Ching, C. B.; Chen, W. N. Mol. Pharm. 2008, 5, 885−890. (8) Carosati, E.; Budriesi, R.; Ioan, P.; Ugenti, M. P.; Frosini, M.; Fusi, F.; Corda, G.; Cosimelli, B.; Spinelli, D.; Chiarini, A.; Cruciani, G. J. Med. Chem. 2008, 51, 5552−5565. (9) Mukherjee, R.; McCaddon, A.; Smith, C. A.; Brasch, N. E. Inorg. Chem. 2009, 48, 9526−9534. (10) Sonaje, K.; Lin, K.; Wang, J.; Mi, F.; Chen, C.; Juang, J.; Sung, H. Adv. Funct. Mater. 2010, 20, 3695−3700. (11) La, S. B.; Okano, T.; Kataoka, K. J. Pharm. Sci. 1996, 85, 85−90.

1313

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314

Biomacromolecules

Article

(12) Yang, Y.; Hua, C.; Dong, C. M. Biomacromolecules 2009, 10, 2310−2318. (13) Li, G. Y.; Shi, L. Q.; Ma, R. J.; An, Y. L.; Huang, N. Angew. Chem., Int. Ed. 2006, 45, 4959−4962. (14) Wu, C. L.; Ma, R. J.; He, H.; Zhao, L. Z.; Gao, H. H.; An, Y. L.; Shi, L. Q. Macromol. Biosci. 2009, 9, 1185−1193. (15) Jayachandran, K. N.; Takacs-Cox, A.; Brooks, D. E. Macromolecules 2002, 35, 4247−4257. (16) Yuan, M. L.; Wang, Y. H.; Li, X. H.; Xiong, C. D.; Deng, X. M. Macromolecules 2000, 33, 1613−1617. (17) Xia, J. H.; Zhang, X.; Matyjaszewski, K. Macromolecules 1999, 32, 3531−3533. (18) Ciampolini, M.; Nardi, N. Inorg. Chem. 1966, 5, 41−44. (19) Arimura, H.; Ohya, Y.; Ouchi, T. Macromol. Rapid Commun. 2004, 25, 743−747. (20) Tu, Y. F.; Wan, X. H.; Zhang, D.; Zhou, Q. F.; Wu, C. J. Am. Chem. Soc. 2000, 122, 10201−10205. (21) Ma, R. J.; Wang, B. L.; Xu, Y. L.; An, Y. L.; Zhang, W. Q.; Li, G. Y.; Shi, L. Q. Macromol. Rapid Commun. 2007, 28, 1062−1069.

1314

dx.doi.org/10.1021/bm2018382 | Biomacromolecules 2012, 13, 1307−1314