2202
Biomacromolecules 2008, 9, 2202–2206
Locally Controlled Release of Basic Fibroblast Growth Factor from Multilayered Capsules Yuki Itoh,†,‡ Michiya Matsusaki,† Toshiyuki Kida,† and Mitsuru Akashi*,†,‡,§ Department of Applied Chemistry, Graduate School of Engineering, Osaka University, 2-1 Yamada-oka, Suita, Osaka 565-0871, Japan, Center of Excellence Program for 21st Century, Osaka University, Japan, and Core Research for Evolutional Science and Technology, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan Received March 29, 2008; Revised Manuscript Received May 8, 2008
Biodegradable multilayered capsules encapsulating basic fibroblast growth factor (bFGF) were developed as a cytokine release carrier for drug delivery systems. The multilayered hollow capsules were fabricated via the layer-by-layer (LbL) assembly of chitosan (CT) and dextran sulfate (Dex). The bFGF was encapsulated into the CT/Dex multilayered capsules by controlling the membrane permeability, and the local and sustained release of bFGF from the capsules was examined. At pH < 8.0, the capsule membrane tightened, and FITC-dextran (Mw ) 4000) could not enter the capsules. However, FITC-dextran (Mw ) 250000) easily entered the capsules at pH > 8.0, which can be attributed to the electrostatic repulsion of Dex caused by the deprotonation of the amine group in CT. After treatment with acetic acid buffer (pH 5.6), FITC-dextran or bFGF was successfully encapsulated into the capsules. The amount of encapsulated bFGF was approximately 34 µg/1 mg of capsule. Initially, about 30% of the encapsulated bFGF was released in serum-free medium within a few hours, however, the release was sustained over 70 h. When the bFGF encapsulating capsules were added to cell culture medium (serum-free), the mouse L929 fibroblast cells proliferated well for 2 weeks as compared to cultures, where bFGF was added to the medium or where bFGF and empty hollow capsules were added separately. The proliferation is due to the local and sustained release of bFGF from the adsorbent capsule to the cell surface.
Introduction Tissue engineering is the medical treatment whereby the regeneration of tissues and organs that have degenerated into dysfunction or dysergia can be achieved by grafting host cells into the damaged tissues and organs.1–3 To enhance tissue regeneration, angiogenesis is necessary because nutrition is supplied to the tissues through blood vessels. However, the native angiogenic process is very slow, and the resulting capillary density is so low that sufficient nutrients cannot be provided to the damaged tissues and organs.4 Therefore, the applications of angiogenic cytokines, including basic fibroblast growth factor (bFGF) and vascular endothelial growth factor (VEGF), which control cell proliferation and differentiation and accelerate tissue regeneration, is important for well-fashioned tissue regeneration.5–8 However, the simple administration of dissolved cytokines into the damaged site cannot achieve tissue regeneration because of the diffusion and denaturation of these cytokines. The wrong dosage of cytokines, such as the administration of large amounts of cytokines, leads to side effects such as carcinomas and tumor growth. It is necessary to control the released amount, the rate, and the delivery site of the cytokines from the drug carriers for effective tissue regeneration. Self-assembling polymeric particles, micelles, and capsules have been actively pursued to create novel drug delivery systems due to their unique functionalities, such as biodegradability or stimuli responsiveness.9–11 In particular, multilayered hollow capsules prepared by the layer-by-layer (LbL) assembly12 of * To whom correspondence should be addressed. Tel.: +81-6-6879-7356. Fax: +81-6-6879-7359. E-mail:
[email protected]. † Graduate School of Engineering, Osaka University. ‡ Center of Excellence Program for 21st Century, Osaka University. § Japan Science and Technology Agency.
oppositely charged polyelectrolytes have attracted a great deal of attention as novel drug carriers because their diameters, membrane thickness, and permeability can be easily controlled.13–15 LbL assembly is a powerful technology to fabricate nanometersized multilayers composed of macromolecules such as synthetic and biodegradable polymers, proteins, and polysaccharides.16 The membrane permeability can be controlled by changing the physicochemical properties of the membrane constituent polymers, which respond to stimuli such as pH or ionic strength.17–20 Many studies on drug delivery systems using multilayered hollow capsules have been reported.21–27 For example, Sukhorukov and co-workers reported that encapsulated peroxidase was released from dextran sulfate (Dex)/protamine capsules by a change in pH.21 Caruso and co-workers demonstrated that encapsulated catalase was released from poly(L-lysine)/poly(Lglutamic acid) capsules by pH and ion strength changes.22 In a previous study, we reported the preparation of biodegradable multilayered hollow nanocapsules via the LbL assembly of chitosan (CT) and Dex on nanosized silica particles and the subsequent removal of silica particles with hydrofluoric acid.28 A novel sustained release system in which the encapsulated proteins are released over a long period of time from multilayered capsules by the enzymatic degradation of the capsule membranes was previously demonstrated in development.26,27 However, to the best of our knowledge, there have been no reports on the controlled release of cytokines from biodegradable multilayered hollow capsules and the bioactivity of released cytokines in cell cultured medium. These points are a drawback for the application of multilayered capsules as cytokine release carriers. Therefore, we propose the controlled release of cytokine, the bioactivity of released cytokine, and the cell study with multilayered capsules.
10.1021/bm800321w CCC: $40.75 2008 American Chemical Society Published on Web 06/27/2008
Release of Basic Fibroblast Growth Factor
Figure 1. Schematic illustration of the preparation of CT/Dex multilayered hollow capsules via the layer-by-layer assembly of CT and Dex onto silica particles and the encapsulation of bFGF into the capsules by manipulating the pH.
In this study, we designed biodegradable multilayered capsules encapsulating bFGF prepared via the LbL assembly of CT and Dex. The permeability of the capsule membranes was controlled by manipulating the environmental pH (Figure 1). The controlled release from the multilayered capsules and the bioactivity of the growth factor on mouse fibroblast cells (L929) were evaluated. The bFGF, which was released from the multilayered capsules, effectively acted on the proliferation of L929 cells, possibly due to a stabilization of bFGF and the locally sustained release of bFGF. This study is the first report of an in vitro evaluation of the controlled release of cytokine from multilayered capsules.
Experimental Section Materials. CT (Mw ) 6.5 × 105), Dex (Mw ) 5.0 × 105), sodium chloride (NaCl), hydrofluoric acid (HF), and basic fibroblast growth factor (bFGF, human recombinant expressed in E. coli) were purchased from Wako Pure Chemicals Inc. and were used without further purification. Potassium chloride (KCl), sodium hydrogen carbonate (NaHCO3), potassium dihydrogenphosphate (KH2PO4), and disodium hydrogenphosphate (Na2HPO4) were of guaranteed reagent grade and were also purchased from Wako Pure Chemical Inc. Fluorescein isothiocyanate-labeled dextran (Mw: 4000 (FD4), 20000 (FD20), and 250000 (FD250)) were obtained from Sigma. Acetic acid, sodium acetate (CH3COONa), and Tris(hydroxymethyl)aminomethane were purchased from Wako Pure Chemical Inc. WST-1 and 1-methoxy PMS were purchased from Dojindo. All of the aqueous solutions were freshly prepared using ultrapure water (milliQ-Plus system, Millipore, 18.2 MΩ cm). Preparation of Biodegradable Multilayered Capsules Encapsulating bFGF. The silica particles (3 µm) were immersed into an aqueous CT solution (1.0 mg mL-1, 0.5 M NaCl) containing 25% formic acid for 15 min under gentle shaking at ambient temperature. The silica particles were then rinsed three times in ultrapure water. Next, the silica particles were immersed into an aqueous Dex solution (1.0 mg mL-1, 0.5 M NaCl) for 15 min, and the same procedure was repeated seven times. The silica particles were then dissolved in HF (1.0 M) to yield the multilayered hollow capsules. The obtained capsules were immersed into a bFGF solution (0.5 mg mL-1, pH 8.0, 0.05 M tris buffer) to encapsulate the bFGF into the capsules through the capsule membrane. The encapsulation of bFGF into the capsules was confirmed by the confocal microscopic observation of Texas-red-labeled bFGF (TR-bFGF). Permeability Control of the Multilayered Hollow Capsules. The obtained multilayered hollow capsules were added into buffer solutions (pH 5.6, 0.05 M acetic acid buffer, pH 6.8 and 8.0, 0.05 M tris buffer) containing dissolving FITC-dextran (Mw: 4000 (FD4), 20000 (FD20),
Biomacromolecules, Vol. 9, No. 8, 2008
2203
and 250000 (FD250)). After 1 h, the capsules were observed by confocal microscopy. After the addition of the capsules to the pH 8.0 buffer solutions in which the FDs were dissolved, the obtained capsules were added into a pH 5.6 buffer solution to encapsulate the FDs. Cell Proliferation With or Without Capsules. Multilayered hollow capsules prepared via the LbL assembly of FITC-CT and Dex were added into cell culture medium, and L929 cells (1.0-105 cells) were then incubated for 24 h. After 1 h, the shapes of the L929 cells, in which the nuclei and Actin were stained by DAPI and TRITCphalloidin, respectively, were observed by confocal microscopy. After 1 and 2 days, the numbers of L929 cells were counted with a hemacytometer. Release of bFGF from Multilayered Capsules. To encapsulate the bFGF in multilayered hollow capsules, the multilayered hollow capsules were added into a TR-bFGF solution (0.5 mg mL-1, 0.05 M Tris buffer solution) at pH 8.0 for 12 h. The obtained capsules were then immersed into a 0.05 M acetic acid buffer (pH 5.6), and the multilayered capsules encapsulating TR-bFGF were observed by confocal microscopy. The amount of TR-bFGF encapsulated was determined quantitatively by the fluorescent spectra of the TR-bFGF. Multilayered capsules encapsulating TR-bFGF were added into serum-free medium without phenol red at 37 °C. After the indicated times in incubation, the amount of TR-bFGF released was observed by the fluorescent spectra of the collected supernatant. Cell Proliferation With or Without Capsules Encapsulating bFGF in Serum-Free Medium. A serum-free suspension of L929 cells was added into 24-well multiplates at a density of 0.5-105 cells/well. After 24 h of incubation, the following solutions were added: bFGFencapsulating multilayered capsules (10 µg, amount of bFGF encapsulated: 0.34 µg), empty multilayered capsules (10 µg) in a bFGF solution (amount of bFGF: 0.34 µg), empty multilayered capsules (10 µg), or dissolved bFGF (amount of bFGF: 0.34 µg). After the prescribed times, the bioactivity of bFGF released from the capsules was evaluated by counting the cell numbers using the WST-1 reagent.
Results and Discussion Permeability Control of Hollow Capsules. The permeability of CT/Dex multilayered hollow capsules in several buffer solutions (pH 5.6, 6.8, and 8.0) was evaluated using FITCdextran (Mw: 4000 (FD4), 20000 (FD20), and 250000 (FD250)) by confocal microscopic observation. Figure 2a shows fluorescent images of the dissolving FDs after the addition of the capsules to the buffer solutions. The permeability of the capsules was dramatically changed between pH 6.8 and 8.0. At pH < 6.8, the fluorescence assigned to FD was not observed in any capsules, suggesting impermeability of the capsules to FD with Mw ) 4000-250000. On the other hand, the fluorescence of FDs was observed in all capsules at over pH 8.0, indicating penetration of the tracker into the capsules. This is likely due to the large pores generated in the multilayers under basic conditions due to the electrostatic repulsion of Dex caused by deprotonation of the amine groups in CT.19,29,30 The reversibility of this permeability event was observed. The capsules were added into a pH 8.0 buffer solution containing each dissolved FDs. These capsules were added into a pH 5.6 buffer solution without FDs. The fluorescent images showed the successful encapsulation of all FDs by controlling permeability of the capsule membrane (Figure 2a). These results suggest that bioactive macromolecules such as proteins can be easily encapsulated into CT/Dex multilayered capsules by pH control. Multilayers prepared via LbL assembly were constructed by electrostatic interactions between the polycations and the polyanions. Therefore, the pH-responsive permeability change could be attributed to an electric charge change in the polyelectrolytes comprising the multilayer.19,29,30 The amine groups
2204
Biomacromolecules, Vol. 9, No. 8, 2008
Itoh et al.
Figure 3. (a) Schematic illustration outlining the procedure of the cytophilic test on multilayered capsules with CT and Dex surfaces. (b) Confocal microscopic images of multilayered capsules with CT and Dex surfaces, which were constructed from FITC-CT and Dex, and L929 cells with their nuclei and Actin stained by DAPI and TRITCphalloidin, respectively, in medium and serum-free medium.
Figure 2. (a) Capsule membrane permeability control by manipulating the pH. Confocal microscopic images of multilayered capsules in buffer solutions (pH 5.6, 0.05 M acetic acid buffer; pH 6.8 and 8.0, 0.05 M Tris buffer) containing dissolved FITC-dextran (Mw: 4000 (FD4), 20000 (FD20), and 250000 (FD250)). (b) Schematic illustration of multilayer permeability control by manipulating the pH.
Figure 4. Cell proliferation profiles of L929 cells cultured with hollow capsules with a CT surface (open circle) or a Dex surface (closed circle). These results are presented as means ( SD (n ) 3).
in the CT molecules were cationic under acidic conditions (pH < 6.4), and the cationic charges decreased with increased pH. In basic conditions at pH > 6.4, the permeability increased because the electrostatic interactions between CT and Dex decreased and the electrostatic repulsions increased between the excess sulfate groups in Dex due to the reducing cationic charges in CT. In contrast, stable electrostatic interactions were formed between CT and Dex under acidic conditions (Figure 2b). Cell Proliferation With or Without Capsules. The cytotoxicity of the capsules was evaluated. Mouse L929 fibroblasts were cultured in culture medium for 24 h, and then capsules with FITC-CT or Dex surfaces were added, and the cells and capsules were observed after 1 h of incubation by confocal microscopy. Figure 3 shows a schematic illustration of this cell culture test and fluorescent images of L929 cells and the capsules with or without serum. The L929 cells were still alive and showed a good extended shape under all conditions. The capsules adsorbed onto the surface of the cells independent of the composition of the outermost surface of the capsules. These results indicate that these capsules may be able to deliver encapsulated drugs to the cell surface directly. To quantitatively evaluate the cytotoxicity of the capsules, the proliferation profile of L929 cells during incubation with multilayered hollow capsules with CT or Dex surfaces was examined for 3 days (Figure 4). After a preculture of the L929 cells on tissue culture polystyrene (TCPS) dishes at a density of 1.0-105 cells/dish for 24 h, the multilayered hollow capsules with CT or Dex surfaces were added into the medium, and the living cell numbers were counted with a hemacytometer. The proliferation
profiles were almost the same under all conditions, and no significant differences were observed. These results suggest that the multilayered hollow capsules with CT and Dex surfaces did not affect cell proliferation. Nevertheless, they absorbed onto the cell surfaces (Figure 3b). Therefore, these multilayered capsules are usage as a direct drug delivery carrier to the cell surface. Release of bFGF from Multilayered Capsules. The release of bFGF from the Dex surface multilayered hollow capsules was evaluated to determine if these capsules are suitable as cytokine release carriers. The multilayered hollow capsules (diameter, 3 µm; membrane thickness calculated from quartz crystal microbalance (QCM) analysis, 69 nm) were immersed into a Texas-red labeled bFGF (TR-bFGF) solution (0.5 mg mL-1, 0.05 M Tris buffer pH 8.0) for 12 h at 4 °C. After centrifugation, the obtained capsules were washed with a pH 5.6 acetic acid buffer to encapsulate bFGF. The encapsulation of bFGF was confirmed by confocal microscopic observation, and the amount of encapsulated bFGF was determined quantitatively by measuring the fluorescence intensity. Figure 5a shows a confocal microscopic image of the capsules after the encapsulation step, and the encapsulation of TR-bFGF was clearly confirmed. The amount of bFGF in the capsules was ascertained to be 34 µg per 1 mg of capsules. The multilayered capsules encapsulating TR-bFGF were then incubated in serumfree medium at 37 °C for the prescribed times. The amount of released bFGF from the capsules was determined quantitatively from the fluorescence intensity. The release profile of bFGF is shown in Figure 5b. The initial release of bFGF was observed for 6 h, and about 50% of the bFGF was released after 12 h of
Release of Basic Fibroblast Growth Factor
Figure 5. (a) Confocal microscopic image of multilayered capsules encapsulating TR-bFGF and (b) release profile of encapsulated bFGF from the multilayered capsules in serum-free medium.
Biomacromolecules, Vol. 9, No. 8, 2008
2205
the capsules encapsulating bFGF could be adsorbed onto the surface of the cell membrane as shown in Figure 3b and might release bFGF directly and continuously to the cell membrane. Accordingly, the L929 cells proliferated for 2 weeks continuously. Furthermore, encapsulation would protect the protein, bFGF, and thus unstable bFGF could maintain its activity even after 2 weeks. These results indicate that the encapsulated bFGF can be directly released to the bFGF receptor on the L929 cell surface for a long time without diffusion of the bFGF into the medium or denaturation. These multilayered capsules are expected to have great potential for tissue engineering applications and other controlled drug release systems.
Conclusions
Figure 6. Proliferation of L929 cells cultured in serum-free medium in the presence of hollow capsules (open triangle), dissolved bFGF (open square), hollow capsules in a bFGF solution (open circle), and bFGF-encapsulating capsules (closed circle). After 7 days of incubation, the cell culture medium was replaced with fresh medium. These results are presented as means ( SD (n ) 3).
incubation. After that, the release profile became saturated as 75% of the bFGF was released within 72 h. This initial release of bFGF can be attributed to the drastic permeability change of the capsule membrane caused by the pH change from pH 5.6 to 7.6. Therefore, because bFGF could be entrapped in the capsule by electrostatic interaction, controlled release of bFGF appears possible. The multilayered capsules prepared via the LbL assembly have great potential as a controlled release carrier for bFGF or other cytokines. Cell Proliferation With or Without Capsules Encapsulating bFGF in Serum-Free Medium. To confirm the bioactivity of bFGF released from the capsules, L929 cell proliferation with or without the capsules encapsulating bFGF was evaluated, because bFGF is known as an unstable cytokine.1,2,7,8 Multilayered capsules encapsulating bFGF were prepared via LbL assembly, plus the permeability control of the capsule membranes. Mouse L929 fibroblast cells were then cultured with or without these capsules (Figure 6). L929 cells with or without (control) empty hollow capsules, which means that the multilayered capsules encapsulated no bFGF, did not proliferate for 14 days because of the serumfree conditions. When bFGF solution or the empty hollow capsules mixed with bFGF were added into the cell cultured medium, the cells grew for 4 days and then the cell growth then stopped for 2 weeks. On the other hand, the number of cells cultured with capsules encapsulating bFGF increased continuously for 2 weeks. In the case of the bFGF solution and the empty capsules mixed with bFGF, the cytokine may have diffused or denatured, and therefore, it seems that cell proliferation was not observed after 4 days of incubation. In contrast,
Biodegradable multilayered capsules encapsulating bFGF as a cytokine release carrier for tissue engineering were evaluated. Cell proliferation with or without capsules encapsulating bFGF in serum-free medium was employed to evaluate the activity of the released bFGF. L929 cells cultured with encapsulated bFGF proliferated in serum-free medium better than without the capsules. The capsules were harmless to L929 cells but allowed a sustained release of bFGF. These cell proliferation results suggest that the bFGF was directly and continuously released from the capsules to the fibroblast cells without denaturation or diffusion of the bFGF. These preliminary results indicated that multilayered capsules can be useful as a novel drug carrier that can encapsulate various growth factors or proteins and release them locally and persistently on a cell surface. Acknowledgment. This work was supported by Core Research for Evolutional Science and Technology (CREST) from the Japan Science and Technology Agency (JST) and Center of Excellence (COE) Program for 21st Century in Osaka University. We thank Suzuki Yushi Industrial Co., Ltd. for their donation of the silica particles.
References and Notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18)
Langer, R.; Vacanti, J. Science 1993, 260, 920. Langer, R. Nat. Med. 1996, 2, 742. Bourque, M. Int. J. DeV. Biol. 1993, 37, 573. Perets, A.; Baruch, Y.; Weisbuch, F.; Shoshany, G.; Neufeld, G.; Cohen, S. J. Biomed. Mater. Res. 2003, 65A, 489. Richardson, T.; Peters, M.; Ennett, A.; Mooney, D. Nat. Biotechnol. 2001, 19, 1029. Holland, T.; Tabata, Y.; Mikos, A. J. Controlled Release 2005, 101, 111. Matsusaki, M.; Serizawa, T.; Kishida, A.; Akashi, M. Biomacromolecules 2005, 6, 400. Matsusaki, M.; Akashi, M. Biomacromolecules 2005, 6, 3351. Bae, Y.; Fukushima, S.; Harada, A.; Kataoka, K. Angew. Chem., Int. Ed. 2003, 42, 4640. Matsusaki, M.; Hiwatari, K.; Higashi, M.; Kaneko, T.; Akashi, M. Chem. Lett. 2004, 33, 398. Akagi, T.; Ueno, M.; Hiraishi, K.; Baba, M.; Akashi, M. J. Controlled Release 2005, 109, 49. Decher, G. Science 1997, 227, 1232. Donath, E.; Sukhorukov, G. B.; Caruso, F.; Davis, S. A.; Mo¨hwald, H. Angew. Chem., Int. Ed. 1998, 37, 2201. Caruso, F.; Caruso, R. A.; Mo¨hwald, H. Science 1998, 282, 1111. Voigt, A.; Lichtenfeld, H.; Sukhorukov, G. B.; Zastrow, H.; Donath, E.; Ba¨umler, H.; Mo¨wald, H. Ind. Eng. Chem. Res. 1999, 38, 4037. Haynie, T. D.; Zhang, L.; Rudra, S. J.; Zhao, W.; Zhong, Y.; Palath, N. Biomacromolecules 2005, 6, 2895. Lvov, Y.; Antipov, A. A.; Mamedov, A.; Mo¨hwald, H.; Sukhorukov, G. B. Nano Lett. 2001, 1, 125. Sukhorukov, G. B.; Antipov, A. A.; Voigt, A.; Donath, E.; Mo¨hwald, H. Macromol. Rapid Commun. 2001, 22, 44.
2206
Biomacromolecules, Vol. 9, No. 8, 2008
(19) Antipov, A. A.; Sukhorukov, G. B.; Leporatti, S.; Radtchenko, I. L.; Donath, E.; Mo¨hwald, H. Colloids Surf., A 2002, 198, 535. (20) Ibarz, G.; Dahne, L.; Donath, E.; Mohwald, H. AdV. Mater. 2001, 13, 1324. (21) Balabushevich, N. G.; Tiourina, O. P.; Volodkin, D. V.; Larionova, N. I.; Sukhorukov, G. B. Biomacromolecules 2003, 4, 1191. (22) Yu, A.; Wang, Y.; Barlow, E.; Caruso, F. AdV. Mater. 2005, 17, 1737. (23) Antipov, A. A.; Sukhorukov, G. B.; Mo¨hwald, H. Langmuir 2003, 19, 2444. (24) Park, M. K.; Deng, S.; Advincula, R. C. Langmuir 2005, 21, 5272. (25) Antipov, A. A.; Sukhorukov, G. B.; Donath, E.; Mo¨hwald, H. J. Phys. Chem. B 2001, 105, 2281.
Itoh et al. (26) Itoh, Y.; Matsusaki, M.; Kida, T.; Akashi, M. Biomacromolecules 2006, 7, 2715. (27) Itoh, Y.; Matsusaki, M.; Kida, T.; Akashi, M. Chem. Lett. 2008, 37, 238. (28) Itoh, Y.; Matsusaki, M.; Kida, T.; Akashi, M. Chem. Lett. 2004, 33, 1552. (29) Antipov, A. A.; Sukhorukov, G. B. AdV. Colloid Interface Sci. 2004, 111, 49. (30) De Geest, B. G.; Sanders, N. N.; Sukhorukov, G. B.; Demeester, J.; De Smedt, S. C. Chem. Soc. ReV. 2007, 36, 636.
BM800321W
