A Promising Drug Controlled-Release System Based on Diacetylene

Oct 23, 2009 - Caixin Guo, Shaoqin Liu, Chang Jiang, Wenyuan Li and Zhifei Dai*. Nanobiotechnology Division, Bio-X Center, State Key Laboratory of Urb...
0 downloads 0 Views 2MB Size
pubs.acs.org/Langmuir © 2009 American Chemical Society

A Promising Drug Controlled-Release System Based on Diacetylene/ Phospholipid Polymerized Vesicles Caixin Guo, Shaoqin Liu, Chang Jiang, Wenyuan Li, and Zhifei Dai* Nanobiotechnology Division, Bio-X Center, State Key Laboratory of Urban Water Resources and Environment, School of Sciences, Harbin Institute of Technology, Harbin, 150001, China

Hiroe Fritz and Xiaoyi Wu* Department of Aerospace and Mechanical Engineering, University of Arizona, Tucson, Arizona 85721 Received April 9, 2009. Revised Manuscript Received October 1, 2009 A novel polymerized vesicular carrier loaded with paclitaxel was developed by introducing the ultraviolet (UV) crosslinkable 10,12-pentacosadiynoic acid (PCDA) into bilayered phospholipid vesicles with the purpose of improving the physicochemical stability as well as the controlled-release property of liposomes. Dynamic light scattering (DLS) and transmission electron microscopy (TEM) results revealed the enhanced stability of PCDA-polymerized vesicles against Triton X-100. In particular, alteration in PCDA/phospholipids ratios and UV-irradiation time can modulate the cumulative paclitaxel released. For instance, vesicles composed of phospholipids only released 98.0 ( 2.1% of paclitaxel within 24 h. Over the same time period, 72.0 ( 5.8%, 43.9 ( 6.5%, and 20.1 ( 5.4% of paclitaxel was released from polymerized PCDA/phospholipid vesicles at molar ratios of 1:3, 1:1, and 3:1, respectively. Likewise, by increasing the UV-irradiation time from 20 to 40 min, the cumulative release of paclitaxel from polymerized PCDA/phospholipid vesicles at molar ratio of 1:1 decreased from 90.5 ( 3.7% to 37.6 ( 2.3% over a time period of experimental observation of 24 h. The influences of vesicle composition (i.e., PCDA/phospholipids ratio) and UV-irradiation time on the release rates of paclitaxel were further examined by finite element (FE) analyzed using Abaqus. Our results demonstrate that novel polymerized vesicles capable of regulating the release of anticancer drugs such as paclitaxel have been developed.

1. Introduction Vesicles composed of phospholipids (i.e., liposome) have been extensively investigated as biomembranes, self-assembly colloids as well as delivery vehicles for various chemotherapy drugs and diagnostic and cosmetic agents. The liposome encapsulation can significantly improve the circulation longevity and site accumulation of encapsulated agents.1-4 Additionally, liposomal drug formulations offer the advantage of reducing drug toxicities that are otherwise observed at concentrations similar to, or lower than, those required for maximum therapeutic activity. However, their applications for long-term, sustained release have been plagued by their insufficient morphological stability, susceptibility to acids, bases, or surfactants, and the resulting poor sustainedand controlled-release efficiency. In response to these challenges, various kinds of polymerized vesicles, either surfactant- or lipid-type, have been introduced to overcome the instability problem.5 Once polymerized, the vesicles could maintain their shapes for an extended period of time, less susceptible to surfactant-induced disruption, and decrease the *Corresponding author. Tel/Fax: 86-451-86402692. E-mail: zhifei.dai@ hit.edu.cn, [email protected]. (1) Gabizon, A.; Papahadjopoulos, D. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 6949. (2) Mayer, L. D.; Tai, L. C.; Ko, D. S.; Masin, D.; Ginsberg, R. S.; Cullis, P. R.; Bally, M. B. Cancer Res. 1989, 49, 5922. (3) Mayer, L. D.; Masin, D.; Nayar, R.; Boman, N. L.; Bally, M. B. Br. J. Cancer 1995, 71, 482. (4) Forssen, E. A.; Male-Brune, R.; Adler-Moore, J. P.; Lee, M. J.; Schmidt, P. G.; Krasieva, T. B.; Shimizu, S.; Tromberg, B. J. Cancer Res. 1996, 56, 2066. (5) Fendler, J. H. Science 1984, 223, 888. (6) Eaton, P. E.; Jobe, P. G.; Nyi, K. J. Am. Chem. Soc. 1980, 102, 6638. (7) Sisson, T. M.; Lamparski, H. G.; Ko1lchens, S.; Elayadi, A.; O’Brien, D. F. Macromolecules 1996, 29, 8321. (8) Sadownik, A.; Stefely, J.; Regen, S. L. J. Am. Chem. Soc. 1986, 108, 7789.

13114 DOI: 10.1021/la9034112

vesicle-vesicle and vesicle-cell fusion.6-9 Nevertheless, polymerized liposomes that have been previously reported as delivery vehicles were all composed of unsaturated phospholipids.10-14 Although the biocompatibility of phospholipids and their passive targeting to the receptors render them desirable for drug delivery, most unsaturated phospholipids are not commercially available and often require sophisticated organic syntheses. Also, phospholipids containing ester-linked hydrocarbon chains are also susceptible to acid and base hydrolysis. In addition, it is relatively difficult to attach ligands to the surface of phospholipids for sitespecifically active targeting. Polydiacetylene vesicles have been widely used in chem/biosensors owing to their blue to red color change in response to a variety of external stimuli, such as pH,15 organic solvent,16 temperature,17 mechanical stress,18 and molecular recognition.19,20 Because of its nontoxicity, polydiacetylene is also employed in the development of ingestible formulations21 and gene carriers.22 (9) Bonte, F.; Hsu, M. J.; Papp, A.; Wu, K.; Regen, S. L.; Juliano, R. L. Biochim. Biophys. Acta 1987, 900, 1. (10) Okada, J.; Cohen, S.; Langer, R. Pharm. Res. 1995, 12, 576. (11) Chen, H. M.; Torchilin, V.; Langer, R. Pharm. Res. 1996, 13, 1378. (12) Chung, Y. C.; Jeong, J. M.; Hwang, J. H. Bull. Korean Chem. Soc. 1998, 19, 780. (13) Jeong, J. M.; Chung, Y. C.; Hwang, J. H. J. Biotechnol. 2002, 94, 255. (14) Chen, H. M.; Torchilin, V.; Langer, R. J. Controlled Release 1996, 42, 263. (15) Cheng, Q.; Stevens, R. C. Langmuir 1998, 14, 1974. (16) Chance, R. R. Macromolecules 1980, 13, 396. (17) Chance, R. R.; Patel, G. N.; Witt, J. D. J. Chem. Phys. 1979, 71, 206. (18) Nallicheri, R. A.; Rubner, M. F. Macromolecules 1991, 24, 517. (19) Charych, D. H.; Nagy, J. O.; Spevak, W.; Bednarski, M. D. Science 1993, 261, 585. (20) Guo, C. X.; Boullanger, P.; Liu, T.; Jiang, L. J. Phys. Chem. B 2005, 109, 18765. (21) Khan, N.; Wyres, C. A. Multi-colour printing, U.S. Patent 20080286483. (22) Yu, G. S.; Choi, H.; Bae, Y. M.; Kim, J.; Kim, J. M.; Choi, J. S. J. Nanosci. Nanotechnol. 2008, 8, 1.

Published on Web 10/23/2009

Langmuir 2009, 25(22), 13114–13119

Guo et al.

Figure 1. Structure of 10,12-pentacosadiynoic acid used for the fabrication of polymerized vesicles.

However, to our best knowledge, the use of polydiacetylene vesicles alone or blended with liposome as anticancer drug carriers have not been exploited. The general idea followed in this work was the development of paclitaxel-loaded polydiacetylene-phospholipid vesicular formulation. We incorporated a polymerizable molecule, 10,12-pentacosadiynoic acid (PCDA) (Figure 1), into liposomes composed of phospholipids. By UV irradiation, the polymerization of PCDA monomers occurred between the diacetylene groups on their hydrophobic moieties, resulting in the formation of polydiacetylene-linkage. Compared with the conventional liposomal formulations, polydiacetylene-phospholipid vesicles take advantages: (1) photoinduced polymerization results in higher stability; (2) the conjugated backbone of polydiacetylene can be distorted upon exposure to pH, temperature, and biorecognition, which can be used for stimuli-responsive drug controlled-release systems; (3) by altering the PCDA proportion, we can regulate the degree of polymerization, hence the release rate of the entrapped drugs; (4) the carboxyl group in the diacetylene molecule is chemically active, so it provides an opportunity to attach various targeting ligands to the surface of the vesicles, allowing selective targeting of the vesicles to desired sites. The focus of this study is the stability of polymerized vesicles against surfactant as well as the effect of vesicle composition and the degree of polymerization on the cumulative release of paclitaxel.

2. Materials and Methods 2.1. Materials. PCDA of 98% in purity was purchased from Lancaster Co. and further purified by dissolving the compound in chloroform and removing polymerized monomers before use. 1,2Distearoyl-sn-glycero-3-phosphocholine (DSPC) was obtained from Avanti Polar Lipids (Alablaster, AL). Paclitaxel was purchased from Shanghai Jinhe Bio-Technology Co., Ltd. (Shanghai, China). Tween 80 was obtained from Tianjin Institute of Fine Chemicals (Tianjin, China). Triton X-100 was purchased from Sigma. Dialysis membrane was obtained from Spectrum Laboratories Inc. (MWCO: 12-14000, California, U.S.), and used for release studies. In all the experiments, organic solvents used were reagent or high-performance liquid chromatography (HPLC) grade, and doubly distilled water was used.

2.2. Preparation of PCDA/DSPC Vesicles and Determination of Entrapment Efficiency and Drug Loading Capacity of Vesicles. Vesicles composed of DSPC and of PCDA/ DSPC at different molar ratios were prepared by the thin-film hydration method. Briefly, PCDA and DSPC at various molar ratios were codissolved with paclitaxel in chloroform at a total lipid concentration of 5 mmol/L. The molar ratio of drug to lipid maintains at 1:30 in all the experiments. After the chloroform was rotoevaporated to dryness, the resulted thin film was resuspended in distilled water to achieve a lipid concentration of 2.5 mmol/L. The suspension was then heated to 70 C and sonicated in an ultrasonicator bath for about 15 min. Consequently, a semitransparent vesicle solution was obtained. Unentrapped paclitaxel was removed from the vesicle suspensions by centrifuging at 2000 rpm for 10 min. The supernatant of PCDA/DSPC vesicles was irradiated by UV light at a wavelength of 254 nm to polymerize the vesicles, while DSPC vesicles without any treatment were used for a control study. To characterize the ability to load drugs in unpolymerized and polymerized vesicles, two important physicochemical parameters were introduced. First, the entrapment effiLangmuir 2009, 25(22), 13114–13119

Article ciency (EE) of vesicles was defined as the ratio of the amount of drug encapsulated in the vesicles to the total amount of drug used in the vesicular dispersion. Second, the drug loading (DL) capacity was determined by the ratio of the amount of encapsulated drug to the total amount of vesicles. The two parameters were calculated using the followamount of drug encapsulated in the vesicles ing equations: EE% ¼ total amount of drug in vesicular dispersion  100 of drug encapsulated in the vesicles  100. DL% ¼ amount total amount of freeze-dried vesicles To determine the amount of paclitaxel loaded in vesicles, unpolymerized vesicles were lysed with 10 times the volume of ethanol and then analyzed using HPLC (HPLC, Waters Co., U.S.A.). A mixed solvent of 70% methanol and 30% water at a flow rate of 1 mL/min was used for HPLC analysis. 2.3. Transmission Electron Microscope (TEM). The vesicle morphology was characterized using a Philip TECNAIG2 TEM equipped with a CCD camera. Vesicle solutions were deposited onto a carbon-coated copper grid and negatively stained with phosphotungstic acid. An accelerating voltage of 120 KV was applied in a TEM analysis. 2.4. Dynamic Light Scattering (DLS). The mean size distributions of the polymerized and unpolymerized vesicles were determined by a 90Plus/BI-MAS instrument (Brookhaven Instruments Co., U.S.A.). Each experiment was repeated 3 times in order to acquire an average. The average diameter given by this instrument is number weighed. Multimodal size distribution (MSD) using the non-negatively constrained least squares (NNLS) algorithm was used for the data analysis. 2.5. Stability of the Polymerized Vesicles. The stability of the polymerized vesicles was analyzed by stepwise addition of Triton X-100 to 1 mL vesicle solution at room temperature. This method has been widely used to evaluate the stability of liposomes. The average sizes of vesicles after the addition of Triton X-100 were measured using DLS. The morphology and size of vesicles were also characterized by TEM after addition of a given amount of Triton X-100, in order to further evaluate the vesicle stability. 2.6. In Vitro Drug Release Studies. The release of paclitaxel from conventional and polymerized vesicles was examined using the dialysis method at room temperature and was compared with free paclitaxel dissolved in ethanol. Before the in vitro release experiments, the concentrations of paclictaxel were all adjusted to 35 μg/mL. An aliquot of each vesicle solution (1 mL) was placed in a dialysis membrane and was tightly sealed. The dialysis bag was immersed in 50 mL of medium that was 1/10 PBS containing 0.1% (v/v) of Tween 80 at pH of 7.4.23 A magnetic stirrer was used to slowly stir the release medium. Samples of 1 mL were taken out from the release medium and replaced with the same amount of fresh medium at given time intervals. The release was analyzed over a time period of 24 h and the amount of released paclitaxel was determined using HPLC. 2.7. Finite Element Analysis of Drug Release. The diffusion-driven release of paclitaxel can be described by the Fick’s second law, ∂C/∂t = Dr2C, where C is the local concentration of paclitaxel, D is the diffusion coefficient of paclitaxel in a vesicle, and r2 is the Laplace operator. Moreover, the local paclitaxel flux on the surface of the vesicle is determined by the difference of the paclitaxel concentrations across the surface: qm =hm(Cs - C¥), in which qm is the paclitaxel flux on the surface, Cs is the paclitaxel concentration on the surface, C¥ is the paclitaxel concentration in the release medium, and hm is the local mass transfer coefficient, also known as the convection coefficient. Analytical solutions have been obtained for drug release from simple matrix systems, such as a plane film and a spherical matrix with an infinite mass transfer coefficient (hm = ¥) in the boundary layer.24 In contrast,

(23) Yang, T.; Cui, F. D.; Choi, M. K.; Cho, J. W.; Chung, S. J.; Shimb, C. K.; Kimb, D. D. Int. J. Pharm. 2007, 338, 317. (24) Siepmann, J.; Ainaoui, A.; Vergnaud, J. M.; Bodmeier, R. J. Pharm. Sci. 1998, 87, 827.

DOI: 10.1021/la9034112

13115

Article

Guo et al.

finite element (FE) analysis can be used to investigate drug release from complex matrix systems (e.g., complicated geometry and material compositions, and various boundary conditions).25,26 In this study, FE analysis of the release of paclitaxel from the conventional and polymerized vesicles was conducted using Abaqus 6.7 student edition (Simulia, USA). Specifically, heat diffusion elements were used to simulate the drug release from the presumed spherical vesicles. There are two reasons to use heat diffusion elements instead of mass diffusion elements for this study. First, heat and mass diffusion problems are mathematically identical after a parameter conversion. Second, Abaqus includes heat convection on a surface in heat transfer analyses, but no mass convection for simulating mass diffusion problems. By comparing the FE results with the analytical solutions available in literature, Zhao recently verified the FE procedures using Abaqus heat diffusion elements in the simulations of drug release from planar, cylindrical, and spherical matrices.27 Herein, we used 492 axisymmetric, triangular heat diffusion elements to analyze the release of paclitaxel. In all the FE simulations, the paclitaxel concentration in the release medium was assumed to be zero, that is, C¥ = 0.

Figure 2. Average mean diameter for unpolymerized (9) and polymerized (b) vesicles as a function of [Triton X-100]/[lipid].

3. Results and Discussion 3.1. Stability of the Polymerized Vesicles. Ease of solubilization by surfactant micelles can be usefully employed to characterize the stability of lipid bilayer vesicles in aqueous suspension. Among various surfactants, the nonionic surfactant Triton X-100 has been selected given its properties as a good agent for membrane solubilization.28,29 In this article, the stabilities of polymerized and unpolymerized vesicles were characterized by Triton X-100-induced solubilization in aqueous suspension. Revealed by a DLS analysis, the addition of Triton X-100 into unpolymerized vesicles resulted in an initial increase in vesicle size followed by a rapid decrease with increased amount of Triton X-100, indicating significant disruption of vesicles (Figure 2). In contrast, the polymerized vesicles with a PCDA/DSPC molar ratio of 1:1 have no decrease in size up to the addition of 48 equiv of Triton X-100 compared to the initial size, suggesting the stability of vesicles to surfactant is greatly enhanced by the UV irradiation-initiated polymerization. It is likely that more Triton X-100 will be needed to disrupt polymerized vesicles with higher PCDA/DSPC molar ratios. Presented in Figure 3 is a proposed structural transformation mechanism of unpolymerized and polymerized vesicles induced by Triton X-100. Upon the addition of Triton X-100, surfactant amphiphiles are first incorporated into vesicles bilayer. When the surfactant concentration exceeds a critical level, the bilayered structure of unpolymerized vesicles exists no longer. Instead, formed will be the mixed micelles of lipid and surfactant. However, polymerized PCDA/DSPC vesicles only undergo the conformational changes of PCDA backbones and thus a blue-to-red color transition up to 48 equiv of Triton-100, because the polymerized PCDA segregated DSPC molecules and minimized the surfactant-induced extraction of the lipid molecules. The color transition of PCDA vesicles induced by Triton X-100 has been reported by Prof. Jiang,30 which demonstrated that Triton X-100 micelles directly attacked the mixed lipid vesicles. The adsorbed micelle could break at the vesicle surface, and part of released (25) Wu, X. Y.; Zhou, Y. J. Controlled Release 1998, 51, 57. (26) Zhou, Y.; Wu, X. Y. J. Controlled Release 1997, 49, 277. (27) Zhao, K. M. Finite element simulation of drug release using Abaqus. M.S. Thesis. University of Arizona, Tucson, AZ, 2008. (28) Lopez, O.; de la Maza, A.; Coderch, L.; Lopez -Iglesiasb, C.; Wehrli, E.; Parra, J. L. FEBS Lett. 1998, 426, 314. (29) Lopez, O.; Cocera, M.; Pons, R.; Azemar, N.; Lopez-Iglesias, C.; Wehrli, E.; Parra, J. L.; de la Maza, A. Langmuir 1999, 15, 4678. (30) Su, Y. L.; Li, J. R.; Jiang, L. Colloids Surf., A 2005, 25, 257.

13116 DOI: 10.1021/la9034112

Figure 3. Schematic representation of structural transformations of unpolymerized (upper) and polymerized vesicles (lower) upon addition of Triton X-100. The blue color (lower) stands for polymerized PCDA vesicles. The blue to red color transition is due to the conformational changes of PCDA backbones induced by Triton X-100.

monomers could insert into the mixed lipid vesicles. The insertion of surfactant molecules would impart stress to the adjacent PCDA network and induce the change of the polymer conformation. In addition, for PCDA/DSPC polymerized vesicles, the most probable region for the insertion of Triton X-100 will be the DSPC domain. When the amount of Triton X-100 increased to a critical value, DSPC probably disintegrated to form micelles with Triton X-100, which might change the size of vesicles but with the vesicular structure remained. Therefore, the decreased sizes of the polymerized vesicle were seen in Figure 2 when 40-fold Triton X-100 was added. A TEM was used to further analyze the surfactant-induced morphological change of unpolymerized and polymerized vesicles. The as-prepared polymerized vesicles with PCDA/DSPC molar ratio of 1:1 possess a spherical architecture with diameter ranging from 210 to 380 nm (the size and morphology of unpolymerized and polymerized vesicles are similar), as shown in Figure 4a. When 14 equiv of Triton X-100 were added to unpolymerized vesicles, aggregates of 10 nm in size were obtained, indicating that the unpolymerized vesicles have been disrupted and converted into micelles (Figure 4b). In contrast, the polymerized vesicles exhibited remarkable resistance to morphological changes induced by surfactants. For instance, the polymerized vesicles retained their vesicular structure upon the addition of 140 equiv of Triton X-100, although their mean size increased from 210-380 nm to 600-1000 nm (Figure 4c). Consistent with the proposed structural change mechanism shown in Figure 3, the Langmuir 2009, 25(22), 13114–13119

Guo et al.

Article

Figure 5. In vitro paclitaxel release from free (9), DSPC lipoFigure 4. TEM image of polymerized vesicles (PCDA/DSPC molar ratio of 1:1, 30-min irradiation) before (a) and after the addition of 14 (b) and 140 (c) equiv of Triton X-100: (b) unpolymerized vesicles; (c) polymerized vesicles.

TEM analysis confirmed the enhanced stability of PCDA/DSPC vesicles against Triton-100 by UV irradiation-induced polymerization of PCDA. 3.2. Effects of UV Irradiation on the Absorption Spectrum of Paclitaxel. Because UV irradiation was utilized to polymerize PCDA/DSPC vesicles, its influence on paclitaxel needs to be examined. After being irradiated by UV light for various given times (e.g., 0, 20, 30, and 40 min), paclitaxel in ethanol solutions was analyzed using UV-visible spectroscopy (data not shown). Results showed that UV irradiation had little effect on the UV absorbance spectra of paclitaxel. Therefore, it is feasible to determine the amount of paclitaxel encapsulated in vesicles by HPLC equipped with a UV detector. 3.3. In Vitro Release of Paclitaxel. EE and DL, two physicochemical parameters characterizing the drug-loading capacity of vesicles, were evaluated for both unpolymerized and polymerized vesicles. In particular, the EE was evaluated to be 82.1 ( 10.2%, 93.3 ( 2.2%, 84.1 ( 9.3% and 75.3 ( 9.5% for the vesicles at PCDA/DSPC molar ratios of 0:1, 1:3, 1:1, and 3:1, respectively. It was seen that incorporation of PCDA into conventional phospholipids bilayer had no significantly effect on the entrapment efficiency of paclitaxel. The EE of all four types of vesicles was over 75.0% and the PCDA/DSPC molar ratio of 1:3 showed the optimal entrapment efficiency. The DL was 2.7 ( 0.3%, 3.8 ( 0.1%, 3.7 ( 0.3%, and 3.3 ( 0.2% for the vesicles at PCDA/DSPC molar ratios of 0:1, 1:3, 1:1, and 3:1, respectively. Clearly, vesicles with PCDA displayed better drug loading capacity, if compared with DSPC liposomes. It is likely that with the incorporation of PCDA reduced vesicles’ weight, the drug loading capacity of polymerized vesicles per unit mass was greatly improved. As opposed to sustained-release system, the controlled-release system can achieve a site-selective, controlled-release system pattern, such that the therapeutic efficacy was improved. At present, the controlled-release system generally contains two patterns. One is that the delivery carriers can release the drug automatically and selectively under the triggering of the physical and chemical conditions of a target, including pH, ionic strength, and temperature, etc. The other is that the drug release can be modulated artificially by the alteration in the structure of drug carriers in vitro.10 Herein, we examined how the vesicle composition (i.e., the PCDA/DSPC molar ratio) and polymerization time (i.e., UV irradiation time) influenced the controlled release of paclitaxel from polymerized PCDA/DSPC vesicles. Langmuir 2009, 25(22), 13114–13119

somes (b), and PCDA/DSPC polymerized vesicles at molar ratios of 1:3 (2), 1:1 (1), and 3:1 (() into PBS (pH 7.4) release medium containing 0.1% (v:v) of Tween 80 at room temperature.

Surfactants or organic solvents are often used to maintain sink conditions in the dissolution test of water-insoluble drugs from tablets or drug carrier.31-34 In this study, 0.1% (v:v) of Tween 80 was added to the release medium (PBS, pH 7.4) to increase the solubility of paclitaxel. In Figure 5, the PCDA/DSPC molar ratio varied while the UV-irradiation time was kept constant at 30 min. As a control study, nearly complete release (e.g., 98.0 ( 2.1%) was observed for DSPC liposomes within 24 h. Still, if there was no drug carrier, paclitaxel would be completely released within 8 h. In contrast, the incorporation of PCDA significantly slowed down the release of paclitaxel, and a biphasic release pattern which was characterized by an initial faster release followed by a sustainedrelease phase was exhibited. Furthermore, with the increase of the PCDA proportion, the sustained-release character became more and more obvious. For example, 72.0 ( 5.8%, 43.9 ( 6.5% and 20.1 ( 5.4% of paclitaxel was released within 24 h of dialysis at room temperature corresponding to PCDA/DSPC molar ratios of 1:3, 1:1 and 3:1, respectively. As the proportion of PCDA increases, the degree of polymerization (i.e., the number of conjugated double and triple bond units), or in other word, the length of conjugated backbone in PCDA molecules, also increases correspondingly. The conjugated backbone, which linearly links PCDA molecules into a covalently linked polymeric structure, blocks the channels of drugloaded polymerized vesicles. Thus, it takes longer time for drugs to be released from vesicles with the high degree of polymerization than from that with low one. The above results indicate that a promising drug controlled-release system has been constructed successfully by photopolymerization of PCDA. In Figure 6, the UV-irradiation time was varied in the range of 20-40 min while the molar ratio of PCDA/DSPC was kept constant at 1:1. Obviously, the increased irradiation time resulted in a lower cumulative amount released. The released drug was evaluated to be 90.5 ( 3.7% within 24 h with vesicles being irradiated for 20 min while this value decreased to 43.9 ( 6.5% and 37.6 ( 2.3% in the case of 30 and 40 min, respectively. As we know, the longer the irradiation time is, the higher the degree of polymerization is. However, the irradiation time can not be excessive, because the excessive UV irradiation may impair paclitaxel (31) Shah, V. P.; Konecny, J. J.; Everett, R. L.; McCullough, B.; Noorizadeh, C. A.; Skelly, J. P. Pharm. Res. 1989, 6, 612. (32) Takahashi, M.; Mochizuki, M.; Itoh, T.; Ohta, M. Chem. Pharm. Bull. 1994, 42, 333. (33) Maggi, L.; Torre, M. L.; Giunchedi, P.; Conte, U. Int. J. Pharm. 1996, 135, 73. (34) Savolainen, P. S.; J€arvinen, T.; Taipale, H.; Urtti, A. Int. J. Pharm. 1997, 159, 27.

DOI: 10.1021/la9034112

13117

Article

Guo et al. Table 1. Mass Diffusion and Convection Coefficients Used in the FE Analysis of the Influence of Vesicle Compositions on the Controlled Release of Paclitaxel vesicle-free

0:1a

k (nm2/s) 0.025 0.0015 h (nm/s) 0.075 0.0034 a The PCDA/DSPC molar ratio.

1:3a

1:1a

3:1a

0.0012 0.0041

0.001 0.03

0.0008 0.049

Figure 6. Effect of UV-irradiation time on in vitro release of paclitaxel from PCDA/DSPC vesicles at a molar ratio of 1:1: (9) 20 min; (2) 30 min; (b) 40 min.

Figure 8. The FE model fits in vitro release of paclitaxel from polymerized vesicles with a PCDA/DSPC molar ratio of 1:1 and with UV-irradiation time of 20 (b); 30 (2); and 40 (9) min, respectively. Solid lines represent the FE simulation results.

Figure 7. The FE model fits in vitro release of paclitaxel from different vesicles: vesicle-free (9) and vesicles with PCDA/DSPC molar ratios of 0:1 (b), 1:3 (2), 1:1 (1), and 3:1 ((), respectively. Solid lines represent the FE simulation results.

which should be protected from light. Moreover, prolonged exposure of a polymerized PCDA vesicle to UV irradiation can change its configuration,35 which probably induces the leakage of the entrapped drugs. Nevertheless, these negative aspects were not observed in the present experiments. 3.4. FE Analysis of Drug Release. FE simulations were pursued to gain new insights into the influences of polymerization parameters, including PCDA/DSPC molar ratio and UV-irradiation time, on the release of paclitaxel from unpolymerized and polymerized vesicles. Figure 7 shows a comparison of the FE simulation results with the in vitro release of paclitaxel from vesicles with different PCDA/DSPC molar ratios. Materials parameters used in the FE analysis were provided in Table 1. Notably, the FE model well captured the rapid release of paclitaxel under vesiclefree conditions and from DSPC vesicles. It is likely that the diffusion-driven mechanism is responsible for the rapid release of paclitaxel in both cases. The FE simulation also reasonably captured the in vitro release of paclitaxel from polymerized vesicles with a PCDA/DSPC ratio of 1:3. Unlike the nearly complete depletion of paclitaxel from DSPC vesicles within 24 h, the cumulative release of paclitaxel from polymerized vesicles with a PCDA/DSPC molar ratio of 1:3 remained at around 72.0%. Less satisfactory FE results were obtained in simulating the release of paclitaxel from polymerized vesicles with high PCDA/ (35) Huo, Q.; Russell, K. C.; Leblanc, R. M. Langmuir 1999, 15, 3972.

13118 DOI: 10.1021/la9034112

DSPC molar ratios (i.e., 1:1 and 3:1). It is worth mentioning that a cumulative release of 43.9 ( 6.5% and 20.1 ( 5.4% of paclitaxel within 24 h was experimentally observed in these cases. Clearly, these highly polymerized PCDA/DSPC vesicles were capable of retaining a large portion of paclitaxel, and thus sustained release was achieved over the time period of experimental observation. In the diffusion-driven release model used in the FE simulations, lower diffusion and convection coefficients can result in slower release of paclitaxel. However, a FE model with very low diffusion and convection coefficients will predict a linear release of paclitaxel as a function of time at low cumulative release. It seems that the diffusion-driven release model cannot fully capture the twophase release behavior of paclitaxel from highly polymerized PCDA/DSPC vesicles: a rapid drug release phase in the first 4 h and a sustained release phase after 12 h. A missing mechanism in the current FE model might be the physical binding of drug molecules with polymerized vesicles. Indeed, the Mooney group has exploited the reversible physical binding as a mechanism for the controlled, sustained release of growth factor from synthetic extracellular matrices.36 Moreover, mechanical strains that may break the physical bonds between drugs and carriers have been used to regulate the release rates of growth factors. This physical binding mechanism that is likely responsible for the sustained release of paclitaxel from highly polymerized PCDA/DSPC vesicles can be included in the development of a new FE model in the future. The comparison of the FE simulation results with in vitro release measurements of paclitaxel from PCDA/DSPC vesicles at a molar ratio of 1:1 but with different exposure time to UV irradiation was shown in Figure 8. Listed in Table 2 were the material parameters used in the FE analysis. Notably, the FE simulations of the release of paclitaxel from PCDA/DSPC (36) Lee, K. Y.; Peters, M.C.; Anderson, K. W.; Mooney, D. L. Nature 2000, 408, 998.

Langmuir 2009, 25(22), 13114–13119

Guo et al.

Article

Table 2. Mass Diffusion and Convection Coefficients Used in the FE Analysis of the Effects of UV Irradiation Time on the Controlled Release of Paclitaxel UV irradiation time (min) k (nm2/s) h (nm/s)

20

30

40

0.0041 0.0089

0.001 0.03

0.0007 0.01

vesicles exposed to UV irradiation for 20 or 40 min agreed with the experimental measurements very well. Less satisfactory is the FE simulation of the paclitaxel release from PCDA/DSPC vesicles irradiated for 30 min, which has been discussed in the previous section. Interestingly, the diffusion coefficients, k, decreased with the increased exposure of PCDA/DSPC vesicles to UV-irradiation (Table 2). Likely, the enhanced UV irradiation resulted in higher cross-linking densities of PCDA/DSPC vesicles, lowering the diffusion of paclitaxel. A similar decrease in k was observed when the PCDA/DSPC molar ratios increased (Table 1). In contrast, the convection coefficient, h, displayed a more complicated dependence on the PCDA/DSPC molar ratio and the UV-irradiation time (Tables 1 and 2). This is because the convection coefficient depends not only on the cross-linking density but also the surface properties of PCDA/DSPC vesicles.

4. Conclusion In summary, we developed a novel drug delivery system through inclusion of UV-cross-linkable PCDA molecules into

Langmuir 2009, 25(22), 13114–13119

bilayered phospholipid vesicles followed by UV-irradiation induced polymerization in situ. Results from preliminary studies demonstrated that the polymerized vesicles possessed improved stability against surfactant-induced disruption of vesicles, and that alteration in vesicle composition and UV-irradiation time modulated the controlled-release of anticancer drug such as paclitaxel. Unpolymerized vesicles ruptured upon exposure to Triton X-100, while polymerized ones kept the spherical structure. It is believed that the conjugated backbone in PCDA molecules blocks the channels of drug-loaded polymerized vesicles. As the proportion of PCDA and/or the irradiation time increased, the length of conjugated backbone in PCDA molecules increased correspondingly, hence the drug cumulative release decreased. In addition, polydiacetylene vesicles can undergo configuration changes upon exposure to external stimuli, such as pH and temperature, which may make them promising in pHor temperature-responsive controlled drug delivery. This research is in progress. Acknowledgment. This research is financially supported by the National High Technology Research and Development Program of China (No. 2007AA03Z316), the Cultivation Fund of the Key Scientific and Technical Innovation Project, Ministry of Education of China (No. 707021) HIT, NSRIF, 2008, 10 (No. AUWQ18400004) and State Key Lab of Urban Water Resource and Environment (2009TS01).

DOI: 10.1021/la9034112

13119