Stabilization of Polyion Complex Nanoparticles Composed of Poly

Dec 17, 2009 - ... Coatings, Inks, and Related Products · Dyes, Organic Pigments, ... Articles ASAP · Current Issue · Submission & Review .... pH/Suga...
0 downloads 0 Views 968KB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Stabilization of Polyion Complex Nanoparticles Composed of Poly(amino acid) Using Hydrophobic Interactions Takami Akagi,†,‡ Kazuki Watanabe,† Hyungjin Kim,†,‡ and Mitsuru Akashi*,†,‡ †

Department of Applied Chemistry, Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita 565-0871, Japan and ‡Japan Science and Technology Agency (JST), Core Research for Evolutional Science and Technology (CREST), Saitama, Japan Received August 4, 2009. Revised Manuscript Received November 25, 2009 We report the design and preparation of polyion complex (PIC) nanoparticles composed of anionic hydrophobically modified and cationic poly(amino acid) and the effect of hydrophobic interactions on the stability of these PIC nanoparticles under physiological conditions. We selected poly(γ-glutamic acid) (γ-PGA) as the biodegradable anionic polymer and poly(ε-lysine) (ε-PL) as the cationic polymer. Amphiphilic graft copolymers consisting of γ-PGA and L-phenylalanine (L-Phe) as the hydrophobic side chain were synthesized by grafting L-Phe to γ-PGA. The PIC nanoparticles were prepared by mixing γ-PGA-graft-L-Phe (γ-PGA-Phe) with ε-PL in phosphate buffered saline (PBS). The formation and stability of the PIC nanoparticles were investigated by dynamic light scattering (DLS) measurements. Monomodal anionic PIC nanoparticles were obtained using nonstoichiometric mixing ratios. When unmodified γ-PGA was mixed with ε-PL in PBS, the formation of PIC nanoparticles was observed. However, within a few hours after the preparation, the PIC nanoparticles dissolved in the PBS. In contrast, γ-PGA-Phe/ε-PL nanoparticles showed high stability for a prolonged period of time in PBS and over a wide range of pH values. The stability and size of the PIC nanoparticles depended on the γ-PGA-Phe/ε-PL mixing ratio and the hydrophobicity of the γ-PGA. The improved stability of the PIC nanoparticles was attributed to the formation of hydrophobic domains in the core of the nanoparticles. The fabrication of PIC nanoparticles using hydrophobic interactions was very useful for the stabilization of the nanoparticles. These results will provide a novel concept in the design of carrier systems composed of PIC. It is expected that the γ-PGA-Phe/ε-PL nanoparticles will have great potential as multifunctional carriers for pharmaceutical and biomedical applications, such as drug and vaccine delivery systems.

Introduction Polymeric nanoparticles have attracted increasing interest as drug delivery carriers as well as for peptides, proteins, and DNA.1-3 Polymer complexes associated with two or more complementary polymers are widely used in potential applications in the form of particles, hydrogels, films, and membranes. In general, electrostatic forces, hydrophobic interactions, hydrogen bonds, van der Waals forces, or combinations of these interactions are available as the driving forces for the formation of the polymer complexes.4-8 In particular, a polyion complex (PIC) can be easily formed when oppositely charged polyelectrolytes are mixed in aqueous solution and interact via electrostatic (Coulombic) interactions. Nanoscaled structural materials (e.g., nanoparticles, micelles, nanogels, and hollow nanospheres) composed of PIC are prepared by tuning the preparation conditions, such as the charge ratio of the anionic-to-cationic polymers, temperature, concentration, and type of polyelectrolyte.9 *Corresponding author: Tel þ81-6-6879-7356, Fax þ81-6-6879-7359, e-mail [email protected].

(1) Torchilin, V. P. Adv. Drug Delivery Rev. 2006, 58, 1532–1555. (2) Vasir, J. K.; Labhasetwar, V. Adv. Drug Delivery Rev. 2007, 59, 718–728. (3) Mundargi, R. C.; Babu, V. R.; Rangaswamy, V.; Patel, P.; Aminabhavi, T. M. J. Controlled Release 2008, 125, 193–209. (4) Kakizawa, Y.; Kataoka, K. Adv. Drug Delivery Rev. 2002, 54, 203–222. (5) Zhang, L.; Eisenberg, A. Science 1995, 268, 1728–1731. (6) Dou, H.; Jiang, M.; Peng, H.; Chen, D.; Hong, Y. Angew. Chem., Int. Ed. 2003, 42, 1516–1519. (7) Kang, N.; Perron, M. E.; Prud’homme, R. E.; Zhang, Y.; Gaucher, G.; Leroux, J. C. Nano Lett. 2005, 5, 315–319. (8) M€uller, M.; Reihs, T.; Ouyang, W. Langmuir 2005, 21, 465–469. (9) Hartig, S. M.; Greene, R. R.; DasGupta, J.; Carlesso, G.; Dikov, M. M.; Prokop, A.; Davidson, J. M. Pharm. Res. 2007, 24, 2353–2369.

2406 DOI: 10.1021/la902868g

However, the stability and characteristics of prepared PIC are influenced by various factors involving their chemical compositions and their surrounding environment. In particular, for PIC micelles or nanoparticles, the ionic strength of the solution is a key parameter for stability because of the shielding effect of the ionic species on the electrostatic interactions.10 Therefore, destabilization of PIC under physiological conditions limit their applications as a drug carrier. Several groups have reported the stabilization of PIC micelles or nanoparticles by chemical cross-linking of the core or shell.11-14 The structure of the cross-linked micelles was fixed while their dissociation was permanently suppressed. In many cases, however, these micelles have a disadvantage in terms of the biocompatibility and biodegradability of the material for medical applications. Moreover, chemical reactions in the cores or shells are undesirable because they may induce a change in the properties of the micelles and/or encapsulated drugs by undesirable intermicellar cross-linking. To overcome this problem, Kataoka et al. prepared PIC micelles composed of poly(R,βaspartic acid) and poly(ethylene glycol)-block-poly(L-lysine) (PEG-b-pLys) containing thiol groups. The PIC micelles with (10) Jaturanpinyo, M.; Harada, A.; Yuan, X.; Kataoka, K. Bioconjugate Chem. 2004, 15, 344–348. (11) Chen, Q.; Hu, Y.; Chen, Y.; Jiang, X.; Yang, Y. Macromol. Biosci. 2005, 5, 993–1000. (12) Hu, Y.; Chen, Y.; Chen, Q.; Zhang, L.; Jiang, X.; Yang, C. Polymer 2005, 46, 12703–12710. (13) Hu, F. Q.; Wu, X. L.; Du, Y. Z.; You, J.; Yuan, H. Eur. J. Pharm. Biopharm. 2008, 69, 117–125. (14) Zhang, J.; Zhou, Y.; Zhu, Z.; Ge, Z.; Liu, S. Macromolecules 2008, 41, 1444–1454. (15) Kakizawa, Y.; Harada, A.; Kataoka, K. J. Am. Chem. Soc. 1999, 121, 11247–11248.

Published on Web 12/17/2009

Langmuir 2010, 26(4), 2406–2413

Akagi et al.

cores cross-linked by the disulfide bonds showed remarkable stability against high salt concentrations.15 Sung et al. reported the stabilization of multi-ion-cross-linked nanoparticles composed of chitosan and poly(γ-glutamic acid) blended with tripolyphosphate (TPP) and MgSO4.16 After PIC formation, the crosslinking of PIC is caused by the cross-linkable groups inside the PIC. Thus, the PIC structure is locked in a reversible fashion and efficiently prevents any dissociation under physiological conditions. On the other hand, nanoparticles fabricated by the self-assembly of amphiphilic block copolymers or hydrophobically modified polymers have been explored as drug carrier systems. In general, these amphiphilic copolymers consisting of hydrophilic and hydrophobic segments are capable of forming polymeric structures in aqueous solutions via hydrophobic interactions.17,18 These self-assembled nanoparticles are composed of an inner core of hydrophobic moieties and an outer shell of hydrophilic groups. The hydrophobic core acts as a drug incorporation site, especially for hydrophobic drugs. Amphiphilic block copolymers such as PEG-b-poly(lactic acid) or PEG-b-poly(ε-caprolactone) are very attractive for drug delivery applications.19,20 In a previous study, we prepared nanoparticles composed of hydrophobically modified poly(γ-glutamic acid) (γ-PGA).21-23 These nanoparticles showed great potential as vaccine delivery carriers.24-26 Like electrostatic interactions, the fabrication of nanoparticles by hydrophobic interactions has attracted considerable attention in drug delivery systems. However, few studies on nanoparticle preparation using both electrostatic and hydrophobic interactions as the driving forces of particle formation and their physiochemical properties have been reported. Muller et al. prepared polyelectrolyte complex (PEC) dispersions by mixing poly(diallyldimethylammonium chloride) (PDADMAC) with poly(maleic acid-co-R-methylstyrene) (PMA-MS). The PEC dispersions were stabilized by intraparticle hydrophobic interactions by phenyl residue of PMA-MS and electrostatic attraction between PMA-MS and PDADMAC and by interparticle electrostatic repulsion, respectively.27,28 Others, stabilization of DNAcontaining PEC by hydrophobic interactions29 or preparation of micelles composed of oppositely charged diblock copolymers using both electrostatic and hydrophobic interactions as a driving force for micellization,30,31 have been reported. It is expected that (16) Lin, Y. H.; Sonaje, K.; Lin, K. M.; Juang, J. H.; Mi, F. L.; Yang, H. W.; Sung, H. W. J. Controlled Release 2008, 132, 141–149. (17) Gaucher, G.; Dufresne, M. H.; Sant, V. P.; Kang, N.; Maysinger, D.; Leroux, J. C. J. Controlled Release 2005, 109, 169–188. (18) Letchford, K.; Burt, H. Eur. J. Pharm. Biopharm. 2007, 65, 259–269. (19) Hagan, S. A.; Coombes, A. G. A.; Garnett, M. C.; Dunn, S. E.; Davies, M. C.; Illum, L.; Davis, S. S. Langmuir 1996, 12, 2153–2161. (20) Allen, C.; Yu, Y.; Maysinger, D.; Eisenberg, A. Bioconjugate Chem. 1998, 9, 564–572. (21) Matsusaki, M.; Hiwatari, K.; Higashi, M.; Kaneko, T.; Akashi, M. Chem. Lett. 2004, 33, 398–399. (22) Akagi, T.; Kaneko, T.; Kida, T.; Akashi, M. J. Controlled Release 2005, 108, 226–236. (23) Akagi, T.; Higashi, M.; Kaneko, T.; Kida, T.; Akashi, M. Biomacromolecules 2006, 7, 297–303. (24) Uto, T.; Wang, X.; Sato, K.; Haraguchi, M.; Akagi, T.; Akashi, M.; Baba, M. J. Immunol. 2007, 178, 2979–2986. (25) Akagi, T.; Wang, X.; Uto, T.; Baba, M.; Akashi, M. Biomaterials 2007, 28, 3427–3436. (26) Wang, X.; Uto, T.; Akagi, T.; Akashi, M.; Baba, M. J. Virol. 2007, 81, 10009–10016. (27) Reihs, T.; M€uller, M.; Lunkwitz, K. J. Colloid Interface Sci. 2004, 271, 69– 79. (28) M€uller, M.; Kessler, B.; Richter, S. Langmuir 2005, 21, 7044–7051. (29) Izumrudov, V. A.; Zhiryakova, M. V. Macromol. Chem. Phys. 1999, 200, 2533–2540. (30) Voets, I. K.; de Keizer, A.; Cohen Stuart, M. A.; Justynska, J.; Schlaad, H. Macromolecules 2007, 40, 2158–2164. (31) Voets, I. K.; de Keizer, A.; Leermakers, F. A. M.; Debuigne, A.; Jer^ome, R.; Detrembleur, C.; Cohen Stuart, M. A. Eur. Polym. J. 2009, 45, 2913–2925.

Langmuir 2010, 26(4), 2406–2413

Article

Figure 1. Chemical structures of γ-PGA, γ-PGA-Phe, and ε-PL.

the hydrophobically modified polyelectrolyte will act to stabilize not only the PIC nanoparticles by hydrophobic interactions but also cell membrane-disruptive carriers. It has been reported that the hydrophobicity of the polymer affects the destabilization of lipid membranes.32-34 The hydrophobic moieties and lipid membranes may interact via hydrophobic interactions and hydrogen bonds and cause membrane expansion or disruption.35 This hydrophobically driven membrane-disruptive process is important for the molecular design of nanoparticles as drug delivery systems. In the present study, we focused on a novel approach for the stabilization of PIC nanoparticles by hydrophobic interactions. We selected poly(γ-glutamic acid) (γ-PGA) as the biodegradable anionic polymer and poly(ε-lysine) (ε-PL) as the cationic polymer (Figure 1). γ-PGA is a naturally occurring water-soluble biodegradable, edible, and nontoxic polyamide that is synthesized by certain strains of Bacillus.36 It is a high-molecular-weight polypeptide composed of γ-linked glutamic acid units, and its R-carboxylate side chains can be chemically modified. In this study, γ-PGA was hydrohobically modified by a hydrophobic amino acid, L-phenylalanine (L-Phe). ε-PL is produced by a Streptomyces albulus strain and has been used as a food additive due to its antimicrobial activities.37 ε-PL is water-soluble and biodegradable and has a molecular weight of ∼5000. Here, we report the formation of PIC nanoparticles composed of γ-PGAgraft-L-Phe (γ-PGA-Phe) and ε-PL and the effects of hydrophobic interactions on their stability under physiological conditions.

Experimental Section Materials. γ-PGA (Mw=3.8  105, D-Glu/L-Glu=60/40, pKa =2.3) and ε-PL (Mw=4700, pKa=9.0) were kindly donated from Meiji Seika Kaisha, Ltd. (Tokyo, Japan) and Chisso Corporation (Tokyo, Japan). L-Phenylalanine ethyl ester (L-Phe) was purchased from Sigma (St. Louis, MO). 1-Ethyl-3-(3-(dimethylamino)propyl)carbodiimide (water-soluble carbodiimide, WSC), Coomassie Brilliant Blue (CBB) G-250, and dimethyl sulfoxide (DMSO) were purchased from Wako Pure Chemical Industries (Osaka, Japan). Synthesis of Hydrophobically Modified γ-PGA. γ-PGAgraft-L-Phe (γ-PGA-Phe) was synthesized as previously described.21-23 Briefly, γ-PGA (4.7 unit mmol) was hydrophobically modified by L-Phe (4.7 mmol) in the presence of WSC (1.2, 2.4, or 3.5 mmol) in 50 mM NaHCO3 aqueous solution. The purified γ-PGA-Phe was then characterized by 1H NMR (32) Murthy, N.; Robichaud, J. R.; Tirrell, D. A.; Stayton, P. S.; Hoffman, A. S. J. Controlled Release 1999, 61, 137–143. (33) Jones, R. A.; Cheung, C. Y.; Black, F. E.; Zia, J. K.; Stayton, P. S.; Hoffman, A. S.; Wilson, M. R. Biochem. J. 2003, 372, 65–75. (34) Chen, R.; Khormaee, S.; Eccleston, M. E.; Slater, N. K. Biomaterials 2009, 30, 1954–1961. (35) Seki, K.; Tirrell, D. A. Macromolecules 1984, 17, 1692–1698. (36) Shih, I. L.; Van, Y. T. Bioresour. Technol. 2001, 79, 207–225. (37) Saimura, M.; Takehara, M.; Mizukami, S.; Kataoka, K.; Hirohara, H. Biotechnol. Lett. 2008, 30, 377–385.

DOI: 10.1021/la902868g

2407

Article spectroscopy. The degree of grafting of L-Phe was determined from the integral intensity ratio of the methylene peaks of γ-PGA to the phenyl group peaks of L-Phe. γ-PGA-Phe with 16, 28, and 49 L-Phe groups per 100 glutamic acid units of γ-PGA (γ-PGAPhe-16, γ-PGA-Phe-28, and γ-PGA-Phe-49) were prepared. Preparation of PIC Nanoparticles. PIC nanoparticles composed of γ-PGA-Phe and ε-PL were prepared by a simple mixing method. First, γ-PGA-Phe (10 mg/mL) and ε-PL (5 mg/mL) were dissolved in phosphate buffered saline (PBS, pH 7.4). To prepare the PIC nanoparticles, γ-PGA-Phe (10 mg/mL in PBS) was added to each concentration of ε-PL solution (0, 0.63, 1.25, 2.0, 2.5, and 5 mg/mL in PBS) at the same volume to yield a translucent solution. The fleshly prepared mixed solutions were used for the characterization of PIC nanoparticles. For measurements of the particle yield and composition, the mixed solutions were then centrifuged at 14000g for 15 min. The mixed solution was separated three phase: supernatant, coacervate, and precipitate.27 After the supernatant removal, the coacervate and precipitate phases were resuspended in pure water and were then freeze-dried for further studies. Characterization of PIC Nanoparticles. The particle size distribution of a mixed solution of γ-PGA-Phe and ε-PL in PBS was measured without filtering by a dynamic light scattering (DLS) method using a Zetasizer Nano ZS (Malvern Instruments, UK). The surface charge of the PIC nanoparticles in PBS was determined by zeta potential measurements using a Zetasizer Nano ZS. The unit ratio of γ-PGA and ε-PL contained within the collected PIC nanoparticles was measured by elemental analysis. The morphology of the PIC nanoparticles was observed by scanning electron microscopy (SEM) (JEOL JSM-6701F) at 30 kV. A drop of the nanoparticle suspension was placed on a glass surface, which was fixed on metallic supports with carbon tape. After drying, the samples were coated with osmic acid. Stability of PIC Nanoparticles. The stability of the PIC nanoparticles in PBS was examined by changing the particle size with respect to time. γ-PGA-Phe (10 mg/mL) and ε-PL (1.25, 2.0, and 2.5 mg/mL) dissolved in PBS were mixed at the same volume, and then the mixed solution was placed in a microtube at 4 °C. At different time intervals, the size of the PIC nanoparticles was measured by DLS. To evaluate the effects of pH on the stability of the PIC nanoparticles, γ-PGA-Phe (10 mg/mL) and ε-PL (2 mg/ mL) dissolved in pure water were mixed at the same volume, and a 1/10 volume of 0.1 M buffer (pH 3-5; citrate, pH 6-8; phosphate, pH 9-11; carbonate) was then added into the mixed solution. The measurement of particle size was carried out in 10 mM buffer at various pH values. To evaluate the effects of hydrogen bonds on the formation of PIC nanoparticles, γ-PGA-Phe (10 mg/mL) and ε-PL (2 mg/mL) dissolved in PBS were mixed at the same volume. After mixing, urea as a hydrogen bond inhibitor was added into the mixed solution at a final concentration of 0.5 or 5 M, and then the changes in particle size were measured.

Detection of the Hydrophobic Domain in PIC Nanoparticles. The hydrophobic domains in the PIC nanoparticles were detected with Coomassie Brilliant Blue (CBB) G-250 dye, according to the protocol previously described by Duval-Terrie et al.38 Binding experiments were performed by varying the nanoparticle concentration at a constant CBB. A mixed solution of equal volumes of γ-PGA-Phe (10 mg/mL) and ε-PL (2 mg/mL) solution was diluted with PBS. To 1 mL of these solutions, 1 mL of 0.1 mM CBB was added. The mixtures were then incubated at room temperature for 10 min. The absorption of CBB was monitored between 400 and 800 nm with a UV/vis spectrophotometer. The PIC nanoparticles show slightly absorption at wavelength from 400 to 800 nm. The absorption spectra of CBB were analyzed by subtraction of the absorption of PIC nanoparticles alone from the absorption of PIC nanoparticle mixed with CBB. (38) Duval-Terrie, C.; Huguet, J.; Muller, G. Colloids Surf., A 2003, 220, 105– 115.

2408 DOI: 10.1021/la902868g

Akagi et al. Table 1. Synthesis of γ-PGA-graft-L-Phe Copolymers sample

γ-PGA (unit mmol)

L-Phe (mmol)

WSC (mmol)

yield (%)

grafting degree (%)a

γ-PGA4.7 4.7 1.2 51 16 Phe-16 γ-PGA4.7 4.7 2.4 46 28 Phe-28 γ-PGA4.7 4.7 3.5 54 49 Phe-49 a The degree of grafting of L-Phe was measured by 1H NMR.

Results and Discussion Formation of PIC Nanoparticles by γ-PGA-Phe and ε-PL Complex. To study the influence of hydrophobic interactions on the formation and stability of PIC nanoparticles, hydrohobically modified γ-PGA (γ-PGA-Phe) was synthesized by the conjugation of L-Phe as the hydrophobic group (Figure 1). The degree of L-Phe grafting was controlled by altering the amount of WSC. In this experiment, γ-PGA-Phe with 16, 28, and 49 L-Phe groups per 100 glutamic acid units of γ-PGA (γ-PGA-Phe-16, γ-PGA-Phe-28, and γ-PGA-Phe-49) was prepared (Table 1). The graft copolymers were soluble in PBS, whereas γ-PGA-Phe with an over 50% grafting degree was not dissolved in PBS. PIC nanoparticles composed of γ-PGA-Phe and ε-PL were prepared by a simple mixing method. To prepare PIC nanoparticles, the γ-PGA-Phe (10 mg/mL) dissolved in PBS was added to the ε-PL (2 mg/mL) solution at the same volume. In this case, the PIC nanoparticles were prepared by the addition of excess amounts of γ-PGA-Phe to ε-PL. It has been reported that complexation between polyelectrolytes having significantly different molecular weights, weak ionic groups, or nonstoichiometric mixing ratios leads to the formation of water-insoluble aggregates.39-41 The interaction of oppositely charged polyelectrolytes can lead to soluble complexes or precipitates. Intermediates between these two states are coacervates. Coacervation is a process during which a homogeneous solution of charged polyelectrolytes undergoes liquid-liquid phase separation, giving rise to a polymer-rich dense phase. The phenomenon can be divided into simple and complex coacervation. In simple polyelectrolyte coacervation addition of salt or alcohol normally promotes coacervation. In complex coacervation two oppositely charged polyelectrolytes or colloidal species can undergo coacervation through associative interactions.42,43 Muller et al. reported on purification of nonstoichometric polyelectrolyte complex nanoparticles by consecutive centrifugation. Centrifugation of mixed polyelectrolyte solution leads to three phases: supernatant, coacervate, and precipitate.27,28 In this study, fleshly prepared mixed solutions were used for the characterization of PIC nanoparticles because there was no difference on the size of nanoparticles between with or without centrifugal wash. Therefore, it seems that particle size measured by DLS is derived from precipitate and coacervate in mixed solution. Table 2 summarizes the properties of PIC nanoparticles composed of γ-PGA-Phe with different hydrophobicities and ε-PL. The formation and size of the PIC nanoparticles depended on the hydrophobicity of γ-PGA. When γ-PGA-Phe was mixed with ε-PL, the mixing solutions became turbid, whereas a solution (39) Guillemet, D. Macromolecules 1997, 30, 7810–7815. (40) Schatz, C.; Domard, A.; Viton, C.; Pichot, C.; Delair, T. Biomacromolecules 2004, 5, 1882–1892. (41) Hartig, S. M.; Carlesso, G.; Davidson, J. M.; Prokop, A. Biomacromolecules 2007, 8, 265–272. (42) Piculell, L.; Lindman, B. Adv. Colloid Interface Sci. 1992, 41, 149–179. (43) Wang, Y.; Kimura, K.; Huang, Q.; Dubin, P.; Jaeger, W. Macromolecules 1999, 32, 7128–7134.

Langmuir 2010, 26(4), 2406–2413

Akagi et al.

Article Table 2. Characterization of PIC Nanoparticles

sample γ-PGA γ-PGA-Phe-16 γ-PGA-Phe-28 γ-PGA-Phe-49

γ-PGA:ε-PL unit feed ratioa 83:17 80:20 78:22 75:25

diameter (nm)b [PDI]c

zeta potential (mV)b

yield of NPs (%)d

15.6 ( 0.7 [0.38 ( 0.04] 805 ( 5.1 [0.20 ( 0.03] 337 ( 45 [0.10 ( 0.03] 255 ( 58 [0.20 ( 0.04]

-20.2 -30.8 -26.3 -24.5

-e 29 56 68

γ-PGA:ε-PL unit ratio of NPsa 75:25 84:16 79:21

a The unit ratio was measured by elemental analysis. b The size and zeta potential of the nanoparticles were measured in PBS by DLS and laser Doppler velocimetry using a Zetasizer nano ZS. c PDI represents polydispersity index. d The PIC complexes from the mixed solution were collected by a centrifugation method. e Not detected.

Table 3. Effects of the Mixing Ratio of γ-PGA-Phe and ε-PL on the Size of γ-PGA-Phe/ε-PL Nanoparticles ε-PL (mg/mL) sizeb (nm)

0

0.63

1.25

2.0

2.5

5.0

size 14.5 ( 1.9 10.9 ( 0.3 10.7 ( 0.7 15.6 ( 0.7 535 ( 19 931 ( 172 PDI 0.39 ( 0.12 0.37 ( 0.07 0.40 ( 0.20 0.38 ( 0.04 0.50 ( 0.12 0.25 ( 0.04 γ-PGA-Phe-16 size 16.9 ( 3.1 12.3 ( 1.0 13.9 ( 2.7 805 ( 5.1 1667 ( 272 -c PDI 0.39 ( 0.06 0.48 ( 0.09 0.41 ( 0.08 0.20 ( 0.03 0.20 ( 0.07 γ-PGA-Phe-28 size 12.5 ( 0.4 225 ( 13 231 ( 13 337 ( 45 362 ( 9.2 PDI 0.20 ( 0.01 0.15 ( 0.05 0.11 ( 0.03 0.10 ( 0.03 0.14 ( 0.05 γ-PGA-Phe-49 size 16.6 ( 1.3 134 ( 8.0 199 ( 6.1 255 ( 58 PDI 0.24 ( 0.01 0.20 ( 0.01 0.25 ( 0.01 0.20 ( 0.04 a γ-PGA or γ-PGA-Phe (10 mg/mL) was mixed with indicated concentration of ε-PL solution at the same volume in PBS. b The size and polydispersity index (PDI) of nanoparticles were measured by DLS. c Large-sized aggregations were observed. γ-PGAa

Figure 2. Size distribution (a) and photographs (b) of PIC nanoparticles composed of γ-PGA-Phe with different hydrophobicities and ε-PL. γ-PGA-Phe (10 mg/mL) and ε-PL (2 mg/mL) dissolved in PBS were mixed to final concentrations of 5 and 1 mg/mL. The size of the nanoparticles was measured in PBS by DLS.

of unmodified γ-PGA mixed with ε-PL remained transparent (Figure 2b). The formation of insoluble PIC nanoparticles was observed during the mixing of γ-PGA-Phe and ε-PL. The size of the PIC nanoparticles decreased with an increasing grafting degree of L-Phe (Table 2 and Figure 2a). It is suggested that the increased grafting degree enhanced the hydrophobic interactions between the L-Phe groups attached to the γ-PGA backbone, resulting in an increased packing of the γ-PGA-Phe. The PIC nanoparticles from the mixed solution were collected by a centrifugation method. The γ-PGA/ε-PL unit ratio of the collected PIC nanoparticles was the same as the feed ratio. The yield of the PIC nanoparticles was found to be dependent on the grafting degree of L-Phe. These results suggest that a high grafting degree of γ-PGA-Phe could easily form PIC complexes, and a large number of γ-PGA-Phe were associated with the formation of PIC nanoparticles. The obtained PIC nanoparticles showed a highly negative zeta potential in PBS, indicating the presence of excess γ-PGA-Phe at their surfaces. This negative charge of the nanoparticle surface was due to the carboxyl groups of γ-PGA. Langmuir 2010, 26(4), 2406–2413

Influence of the Mixing Ratio of γ-PGA-Phe and ε-PL. To investigate the effect of the mixing ratio of γ-PGA-Phe and ε-PL on the formation of PIC nanoparticles, the preparation of nanoparticles was performed by the addition of γ-PGA-Phe solution (10 mg/mL in PBS) to each concentration of ε-PL solution (0, 0.63, 1.25, 2.0, 2.5, and 5 mg/mL in PBS) at the same volume, and then the particle size of the mixed solution was measured. The results of the particle size measurements are summarized in Table 3. As the concentration of the added ε-PL increased, the mixing solutions became more turbid, and the size of the PIC nanoparticles increased according to the increase in aggregated polymers. The size of the PIC nanoparticles depended on the concentration ratio in mixing solution of γ-PGA-Phe and ε-PL. When the concentration of ε-PL increased from 0.63 to 2.5 mg/mL, the size of the γ-PGA-Phe-28/ε-PL nanoparticles increased from 225 to 362 nm and showed a monodispersed size distribution (Figure 3). When γ-PGA-Phe-28 (10 mg/mL) was mixed with 0.63, 1.25, 2.0, or 2.5 mg/mL of ε-PL at the same volume, the charge ratio, R=[COOH]/[NH2], was 8.2, 4.1, 2.6, DOI: 10.1021/la902868g

2409

Article

Figure 3. Influence of the mixing ratio on the size of γ-PGA-Phe28/ε-PL nanoparticles. γ-PGA-Phe-28 (10 mg/mL) and ε-PL (0-2.5 mg/mL) at the same volume were mixed in PBS, and then the size of the nanoparticles was measured by DLS.

or 2.1, respectively. The yield of the PIC nanoparticles increased with increasing the concentration of ε-PL solution. In the case of the concentration of ε-PL was 1.25, 2.0, and 2.5 mg/mL, the yields of γ-PGA-Phe-28/ε-PL nanoparticles were 42.1, 56.3, and 61.0%, respectively. These results suggest that the increased size of γ-PGA-Phe/ε-PL nanoparticles by changing the ε-PL concentration are attributed to the differences of the association number of γ-PGA-Phe/ε-PL pre one particle formation. When γ-PGA-Phe was mixed with 5 mg/mL of ε-PL, large-sized aggregations were detected. In contrast, if unmodified γ-PGA was mixed with 2.5 or 5 mg/mL of ε-PL, the formation of PIC nanoparticles was observed. However, within a few hours after the preparation, the PIC nanoparticles prepared from γ-PGA dissolved in PBS, and the solution became clear, indicating that the unmodified γ-PGA cannot form stable PICs in PBS. In this study, ε-PL with a Mw of 4700 was used as a cationic polymer. The Mw of the polyelectrolyte is a major parameter affecting the formation of PIC. In general, the high Mw of the polymer leads to the formation of large-sized particles.44-46 It is suggested that the chain length of ε-PL was not sufficient to form stable γ-PGA/ ε-PL nanoparticles. Figure 4 shows SEM images of the PIC nanoparticles. The γ-PGA-Phe-16/ε-PL and γ-PGA-Phe-28/ε-PL nanoparticles were spherical in shape, and a white circle surrounding particles was observed. The samples for SEM observation are contained PIC, free γ-PGA-Phe, and free ε-PL because all of polymers are not involved in the particle formation. The particular SEM images may be attributed to the presence of free polymers which are not involved in the formation of PIC nanoparticles. The morphology of the PIC nanoparticles purified by centrifugal wash was also observed. The γ-PGA-Phe/ε-PL nanoparticles prepared by centrifugal wash were spherical in shape with a relatively smooth and rigid structure. However, a large amount of aggregations was also found (data not shown). The particles size obtained from the SEM image was different from the DLS data. The size of γ-PGA-Phe-16/ε-PL and γ-PGA-Phe-28/ε-PL nanoparticles obtained from the SEM image was 458 ( 106 and 183 ( 34 nm (n=30). The size measured by the DLS method (γ-PGA-Phe-16: 805 ( 5 nm; γ-PGA-Phe-28: 337 ( 45 nm) was higher than that of (44) Schatz, C.; Lucas, J. M.; Viton, C.; Domard, A.; Pichot, C.; Delair, T. Langmuir 2004, 20, 7766–7768. (45) Luo, K.; Yin, J.; Song, Z.; Cui, L.; Cao, B.; Chen, X. Biomacromolecules 2008, 9, 2653–2661. (46) Yusa, S.; Yokoyama, Y.; Morishima, Y. Macromolecules 2009, 42, 376– 383.

2410 DOI: 10.1021/la902868g

Akagi et al.

the SEM method. This difference can be attributed to the difference between the dried and hydrated state. In the case of the SEM method, SEM image exhibits the size at the dried state of the sample, whereas the DLS method involves the measurement of size in the hydrated state. Therefore, in the hydrated state, the nanoparticles will have a higher particle size due to solvent effect. In the case of γ-PGA-Phe-49/ε-PL nanoparticles, the aggregation of the nanoparticles was observed. The γ-PGA-Phe-49/ε-PL nanoparticles showed a monodispersed size distribution of about 250 nm in PBS. This difference is attributed to the aggregation of the nanoparticles by the drying process. In the case of the highly hydrophobized γ-PGA (γ-PGA-Phe-49), some hydrophobic groups (Phe) may be located near the nanoparticle surfaces. Therefore, the increased hydrophobicity of γ-PGA induces formation of aggregation by the drying process. Stability of PIC Nanoparticles on the Incubation Time and pH. The influence of the incubation time on the size of the PIC nanoparticles is shown in Figure 5a. Following the just-completed preparation of γ-PGA-Phe-16/ε-PL nanoparticles, the size of the nanoparticles was about 800 nm. After a 1-day incubation in PBS, however, the nanoparticles collapsed and the solution became clear. The size of the γ-PGA-Phe-28/ε-PL nanoparticles gradually increased when the incubation time was extended. The nanoparticles showed time-dependent aggregation behavior. In contrast, the size change of the γ-PGA-Phe-49/ε-PL nanoparticles as a function of time was smaller than that of the γ-PGA-Phe-16 and -28/ε-PL nanoparticles. As a result of a more compact complex structure, the aggregation of the γ-PGA-Phe-49/ε-PL nanoparticles in PBS was inhibited. These results indicate that the loss of particle stability is obviously slower for the γ-PGAPhe-49/ ε-PL nanoparticles than for the γ-PGA-Phe-28/ε-PL nanoparticles. Figure 5b shows the effect of the γ-PGA-Phe-28/ε-PL ratio on the stability of the PIC nanoparticles. When the concentration of ε-PL was increased from 1.25 to 2.5 mg/mL, the size of the nanoparticles increased gradually with increasing incubation time. When the ε-PL concentration is high, the nanoparticles would aggregate with each other, resulting in an increase of the particle size over time. This result might be attributed to the electrostatic repulsions of the PIC nanoparticles; electrostatic repulsion between the nanoparticles with the same negative charge prevents the aggregation of these nanoparticles. In the case of a high ε-PL concentration, the electrostatic repulsion between the nanoparticles is reduced due to a screening of the electric charge of the γ-PGA-Phe. Therefore, with increasing ε-PL content, the stability of the nanoparticles decreased in PBS. These results suggest that the stability of the PIC nanoparticles is affected by the hydrophobicity of the γ-PGA-Phe and the γ-PGA-Phe/ε-PL ratio. On the other hand, if γ-PGA conjugated with tyrosine (γ-PGA-Tyr) as a hydrophobic unit was used to prepare PIC nanoparticles, γ-PGA-Tyr/ε-PL could not form stable PIC nanoparticles in PBS. This result indicates that the hydrophobicity of grafting units is also important for the formation of stable PIC nanoparticles. The stability and characteristics of the prepared PIC nanoparticles are influenced by the ionic strength and pH of the solution. It has been reported that PIC nanoparticles composed of chitosan (CS) and γ-PGA are stable in water, but not in PBS (pH 7.4) due to deprotonation of the CS amino groups.47 Moreover, Kunioka et al. reported that hydrogels prepared by γ-irradiation from γ-PGA and ε-PL showed pH-sensitive swelling behaviors. This 50/50 wt % γ-PGA/ε-PL hydrogel swelled at a low (6.0) but deswelled at pH values from 4.0 to 6.0.48 To observe the effects of the pH on the size of the γ-PGA-Phe-28/ ε-PL nanoparticles, DLS measurements were carried out in buffers at various pH values. The PIC nanoparticles did not change their particle size over a pH range from 3 to 9 (data not shown). The nanoparticles showed high stability against a wide range of pH values. (48) Kunioka, M.; Choi, H. J. J. Environ. Polym. Degrad. 1996, 4, 123–129.

Langmuir 2010, 26(4), 2406–2413

Effects of Hydrogen Bonds on the Stability and Formation of Nanoparticles. It is well know that polyelectrolytes such as polypeptides interact with each other not only through electrostatic interactions but also through hydrogen bonds.49,50 In this study, we selected anionic and cationic poly(amino acid) to (49) Haynie, D. T.; Zhang, L.; Rudra, J. S.; Zhao, W.; Zhong, Y.; Palath, N. Biomacromolecules 2005, 6, 2895–2913. (50) Zhou, J.; Wang, B.; Tong, W.; Maltseva, E.; Zhang, G.; Krastev, R.; Gao, C.; M€ohwald, H.; Shen, J. Colloids Surf., B 2008, 62, 250–257.

DOI: 10.1021/la902868g

2411

Article

Akagi et al.

Figure 6. Changes in the size of the PIC nanoparticles as a function of urea treatment. (a) γ-PGA-Phe of different hydrophobicities (10 mg/ mL) and ε-PL (2 mg/mL) dissolved in PBS were mixed at the same volume, and urea as a hydrogen bond inhibitor was then added into the mixed solution to a final concentration of 0.5 M. At different time intervals, the size of the nanoparticles was measured by DLS. (b) γ-PGAPhe-49 (10 mg/mL) and ε-PL (2 mg/mL) dissolved in PBS were mixed at the same volume, and urea was then added to the mixed solution to a final concentration of 0.5 or 5 M. The size of the nanoparticles was measured soon after urea addition.

prepare the PIC nanoparticles. Therefore, it can be expected that the hydrogen-bond interactions between the poly(amino acid) would influence PIC formation and stability. The effect of hydrogen bonds on the stability and formation of the PIC nanoparticles was studied using urea, which is able to form stronger hydrogen bonds with the peptide groups. After adding urea (0.5 or 5 M) to the γ-PGA-Phe/ε-PL nanoparticles, the size of the nanoparticles was measured by DLS, and the results are shown in Figure 6. It can be observed that the size of the PIC nanoparticles was drastically changed by the addition of urea. After 0.5 M urea treatment, the size of the γ-PGA-Phe-16/ε-PL nanoparticles increased rapidly, and then the nanoparticles collapsed and the solution became transparent. In contrast, the aggregation or swelling of the nanoparticles by an inhibition of the hydrogen bonds was suppressed with an increase in polymer hydrophobicity. Increased polymer hydrophobicity resulted in increased stability against the urea treatment. In addition, the effect of hydrogen bonds was evaluated by adding urea to ε-PL solution. Urea was added to ε-PL solution, and then the mixing solution (ε-PL 2 mg/mL, urea 10 M) was added to γ-PGA-Phe (10 mg/mL) at the same volume. The size of the PIC nanoparticles premixed with urea was measured by DLS. As a result, γ-PGAPhe-16/ε-PL could not form nanoparticles in the presence of 5 M urea. In the case of γ-PGA-Phe-28 and -49, the particle size without urea was 331 and 263 nm. In contrast, the particle size increased to 2440 and 568 nm by the preaddition of urea, respectively. These results clearly showed that the intraand/or intermolecular hydrogen bonds in the γ-PGA-Phe/ε-PL contributed to the stability and formation of the PIC nanoparticles. Detection of the Hydrophobic Domains into the PIC Nanoparticles. The PIC nanoparticles formed between γ-PGA-Phe and ε-PL had a drastic improvement in their stability. It is thought that the hydrophobic interactions of γ-PGA-Phe are important for the fabrication of stable PIC nanoparticles. If this stabilization is caused by the hydrophobic interactions, then the hydrophobic domain in the particle core should be formed. Hydrophobic domains in the PIC nanoparticles were detected with Coomassie Brilliant Blue (CBB) G-250 dye. CBB is highly soluble in water but also possesses a marked hydrophobic character, mainly due to the presence of six aromatic rings. This molecule therefore participates in strong hydrophobic interactions. 2412 DOI: 10.1021/la902868g

Figure 7. Changes in the wavelength of the maximum absorption of CBB in PBS upon adding PIC nanoparticles. The mixed solution of γ-PGA-Phe (5 mg/mL) and ε-PL (1 mg/mL) was diluted with PBS, and CBB was then added to the mixed solution to a final concentration of 0.05 mM. The absorption of CBB was monitored between 400 and 800 nm with a UV/vis spectrophotometer.

Moreover, the wavelength shift of the maximum absorption was observed in nonpolar media.33 Thus, in this experiment, CBB was used as a probe of its microenvironment polarity. Figure 7 shows the changes in the wavelength of the maximum absorption of CBB as a function of the PIC nanoparticle concentration. As expected, the maximum absorption peak of CBB was shifted toward the higher wavelengths when the γ-PGA-Phe/ε-PL nanoparticle concentration was increased. An increase in the peak wavelength shift was dependent on the hydrophobicity of γ-PGA. The maximum absorption change in the case of γ-PGA-Phe/ε-PL nanoparticles could be attributed to the presence of hydrophobic domains. For the mixture of γ-PGA/ε-PL, the dose-dependent change in the absorption spectra was not observed. Likewise, for unmodified γ-PGA, γ-PGA-Phe-16, and -28 (polymer alone), absorbance remained constant (data not shown). However, in the case of γ-PGA-Phe-49 alone, the CBB peak shift was observed in a dose-dependent manner. When γ-PGA-Phe-49 concentration was 300 μg/mL, the peak shift width (Δλ) was 15 nm. These results suggest that γ-PGA-Phe-49 itself has some hydrophobic domains due to the intra- or intermolecular associations of L-Phe attached to Langmuir 2010, 26(4), 2406–2413

Akagi et al.

the γ-PGA. However, the CBB peak shift of γ-PGA-Phe-49/ε-PL nanoparticles was higher than that of γ-PGA-Phe-49 alone at the same concentration. This is attributed to the fact that an increase in the peak wavelength shift is dependent on increase of hydrophobic domains in the core of nanoparticles associated with PIC formation. The formation of hydrophobic domains by the intra- or intermolecular associations of γ-PGA-Phe is important for the stability of the PIC nanoparticles. In this study, it was shown that the stability of the PIC nanoparticles depended on the hydrophobicity of γ-PGA. The increased stability of the PIC nanoparticles was ascribed to the formation of hydrophobic domains in the core of the nanoparticles. It is considered that the PIC nanoparticles consisting of more hydrophobized γ-PGA have a larger number or size of hydrophobic domains in the particles. Therefore, the γ-PGAPhe-49/ε-PL nanoparticle was more stable compared to nanoparticles composed of γ-PGA, γ-PGA-Phe-16, or -28. However, the possibility exists that γ-PGA-Phe-49 has a different conformation in solution compared to unmodified γ-PGA, γ-PGA-Phe-16, or -28. It can be speculated that the more hydrophobized γ-PGA, the more the conformation tends toward a compact coil rather than an expanded one. Thus, γ-PGA-Phe/ε-PL nanoparticles can be formed from crosslinking of compact coiled γ-PGA-Phe by ε-PL. The compact coiled γ-PGA-Phe might be efficiently and strongly crosslinked by ε-PL and consequently increased stability. There is a possibility that the particle formation process depends on the conformation of the polymers in solution. Further studies are in progress to obtain detailed information on the formation and structure of the PIC nanoparticles.

Langmuir 2010, 26(4), 2406–2413

Article

Conclusions PIC nanoparticles are promising carriers for DDS. However, their utility is limited by their sensitivity to the environment, such as ionic strength or the pH of the solution. The present study focused on the preparation of PIC nanoparticles composed of biodegradable γ-PGA-Phe and ε-PL and their stability under physiological conditions. We prepared γ-PGA-Phe/ε-PL nanoparticles by a simple mixing method without using any organic solvents and surfactants. The formation of PIC nanoparticles with added γ-PGA-Phe of different hydrophobicities in ε-PL solution was studied as a function of the mixing ratio. The formation of PIC nanoparticles depended on the γ-PGA-Phe/εPL ratio and the hydrophobicity of γ-PGA. The stability of the PIC nanoparticles was significantly improved by the hydrophobic modification of γ-PGA. The γ-PGA-Phe/ε-PL nanoparticles were stable under physiological conditions, exhibiting no aggregation, precipitation, or dissociation for a prolonged period of time. Moreover, the stability of the PIC nanoparticles against the inhibition caused by hydrogen bonds increased with increasing polymer hydrophobicity. This study revealed that the stabilization of the PIC nanoparticles could be attributed to the formation of the hydrophobic domain in the core of the nanoparticles. Our results suggest that the fabrication of PIC nanoparticles using hydrophobic interactions is very useful for the stabilization of nanoparticles. These results provide a novel concept for particle design for DDS. Acknowledgment. This work was supported by CREST from the Japan Science and Technology Agency (JST).

DOI: 10.1021/la902868g

2413