International Journal of Pharmaceutics 486 (2015) 217–225
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
A slow-release system of bacterial cellulose gel and nanoparticles for hydrophobic active ingredients Yukari Numata a,b, * , Leticia Mazzarino a,c , Redouane Borsali a a
Centre de Recherches sur les Macromoléculés Végétales (CERMAV, UPR-CNRS 5301), Université Grenoble Alpes, BP 53, F-38041 Grenoble Cedex 9, France Department of Materials Chemistry, Asahikawa National College of Technology, Asahikawa 071-8142, Japan c Departamento de Ciências Farmacêuticas, Universidade Federal de Santa Catarina, Florianópolis 88040-900, Brazil b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 January 2015 Received in revised form 6 March 2015 Accepted 30 March 2015 Available online 31 March 2015
A combination of bacterial cellulose (BC) gel and amphiphilic block copolymer nanoparticles was investigated as a drug delivery system (DDS) for hydrophobic active ingredients. Poly(ethylene oxide)-bpoly(caprolactone) (PEO-b-PCL) and retinol were used as the block copolymer and hydrophobic active ingredient, respectively. The BC gel was capable of incorporating copolymer nanoparticles and releasing them in an acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. The percentage of released copolymer reached a maximum value of approximately 60% after 6 h and remained constant after 24 h. The percentage of retinol released from the copolymer-containing BC gel reached a maximum value at 4 h. These results show that the combination of BC gel and nanoparticles is a slow-release system that may be useful in the cosmetic and biomedical fields for skin treatment and preparation. ã 2015 Elsevier B.V. All rights reserved.
Chemical compounds studied in this article: Retinol (PubChem CID: 445354) Keywords: Bacterial cellulose gel Nanoparticles Block copolymer Biocompatibility Drug delivery system Retinol
1. Introduction Bacterial cellulose (BC) gel is a unique hydrogel consisting of greater than 99% (w/w) water by weight. BC gels have a threedimensional network structure of ultrafine fiber made from pure cellulose (Brown, 1996; Iguchi et al., 2000; Ross et al., 1991) and have various useful properties, such as softness, translucence, and good biocompatibility and water retention capacity (Klemm et al., 2005; Miyamoto et al., 1989; Okiyama et al., 1993). BC gels have been extensively investigated as potential soft materials for use in biomedical fields (Fu et al., 2013; Shah et al., 2013). In skin tissue repair, the advantages of BC, such as biocompatibility, conformability, elasticity, transparency, the ability to maintain a moist environment in the wound, and the ability to absorb exudate during the inflammatory phase, confer great potential for application in wound healing systems. In cosmetic applications, BC gels offer good biocompatibility, conformability, elasticity, and transparency, as well as the ability to maintain a moist
* Corresponding author. Present address: General Education, Faculty of Commerce, Otaru University of Commerce, 3-5-21, Midori, Otaru 047-8501, Japan. Tel.: +81 134 27 5412; fax: +81 134 27 5412. E-mail address: [email protected] (Y. Numata). http://dx.doi.org/10.1016/j.ijpharm.2015.03.068 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
environment for active ingredients. BC gels are commercially available as a cosmetic facial mask in Japan. BC facial masks adhere well to the face and retain moisture better than normal cellulose sheets, because BC consists of very thin nano-fibers. However, the incorporation of hydrophobic active ingredients into BC gels remains a challenging task. New methods to effectively incorporate hydrophobic molecules would allow their application in BC gel instead of in oily cosmetic products such as facial creams; therefore, the development of such methods will be useful in the cosmetics market and in biomedical applications. Amphiphilic block copolymer systems are used to incorporate hydrophobic molecules, and these copolymers have the ability to self-assemble into well-organized structures, such as nanoparticles, micelles, and vesicles, in aqueous solutions. Micelles can have a diameter of 10–200 nm, but most have a diameter ranging from 10 to 100 nm. Polymers typically have molecular weights ranging from several hundred to several tens of thousands. Particle size depends on sample preparation conditions and the type of copolymer, as well as its components, molecular weight, and organic solvent composition (Aliabadi et al., 2007). Micelles can be produced by methods such as dialysis (Allen et al., 1999) and cosolvent evaporation (Aliabadi et al., 2005b). Nanoparticles produced from biocompatible and biodegradable copolymers have shown great potential as carriers for active ingredients in cosmetic
218
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
and pharmaceutical applications. Poly(ethylene oxide)-b-poly (caprolactone) (PEO-b-PCL) is composed of biocompatible and biodegradable synthetic polymers and is one of the most extensively studied copolymer systems for nanoparticle formation (Wei et al., 2009). The PEO-b-PCL copolymer forms nano-sized micelles, in which PEO is the hydrophilic shell-forming block and PCL is the hydrophobic core-forming block. Many researchers have reported encapsulation of hydrophobic active ingredients into PEO-b-PCL nanoparticles (Aliabadi et al., 2005a; Kim et al., 1998; Mahmud and Lavasanifar, 2005; Shuai et al., 2004). The ultimate objective of such studies was the development of a drug delivery system (DDS). PEO (poly(ethylene glycol) (PEG)), the hydrophilic shellforming block of PEO-b-PCL nanoparticles, has high affinity for cellulose (Numata et al., 2009). Therefore, we hypothesized that PEO-b-PCL nanoparticles can be introduced into BC gel, which can then be used to encapsulate hydrophobic molecules. If the BC gel can release PEO-b-PCL nanoparticles and the encapsulated hydrophobic molecules, then the system can be utilized in topical compound preparations in the cosmetic and the biomedical fields. In this study, we designed a DDS by combining BC gels with a nanoparticle-encapsulated hydrophobic active ingredient. We used PEO-b-PCL and retinol as the block copolymer and hydrophobic active ingredient, respectively. Retinol, vitamin A alcohol, is extensively used in the pharmaceutical and cosmetic industries and is a well-known anti-aging ingredient. The PEO-bPCL nanoparticle-containing BC gel was investigated for the presence of BC containing nanoparticles by Fourier transform infrared (FT-IR) spectra and field-emission gum scanning electron microscopy (FEG-SEM). The released nanoparticles indicated that the PEO-b-PCL nanoparticle-containing BC gel can be utilized as a DDS, and the retinol release indicated that packaged retinol can be released from the BC gel; however, the DDS must be improved to support practical use. Our results show that the DDS described herein may be useful in topical preparations in the cosmetic and biomedical fields. 2. Material and methods 2.1. Materials Three types of PEO-b-PCL block copolymers were purchased from Polymer Source, Inc. (Quebec, Canada). The molecular weights (Mw) of PEO and PCL in the three PEO-b-PCL block copolymers were 5000 and 5000, 5000 and 10,000, and 2000 and 11,500, respectively. The Mw/number average molecular weight (Mn) ratios of the three types of PEO-b-PCL copolymers were 1.20, 1.26, and 1.20. Retinol was purchased from Sigma–Aldrich Co. LLC (St. Louis, MO, USA). 2.2. Preparation of PEO-b-PCL nanoparticles and retinol-loaded PEOb-PCL nanoparticles PEO-b-PCL nanoparticles were prepared by co-solvent evaporation (Aliabadi et al., 2005b). PEO-b-PCL (10 mg) dissolved in acetone (0.5 mL) was added in a drop-wise manner (5.08 mL/h) to Milli-Q water (1 mL) with stirring. The remaining acetone was removed by evaporation at room temperature under vacuum. Retinol-loaded PEO-b-PCL nanoparticles were prepared by cosolvent evaporation as described above except for dissolving retinol and PEO-b-PCL in acetone. PEO-b-PCL (10 mg) and retinol (3, 6, or 15 mg) dissolved in acetone (0.5 mL) were added in a dropwise manner (5.08 mL/h) to Milli-Q water (1 mL) with stirring. The remaining acetone was removed by evaporation at room temperature under vacuum. The obtained nanoparticle suspension was
centrifuged at 5673 g for 5 min to remove the retinol precipitate from the outside of the nanoparticles. 2.3. Preparation of BC gel containing PEO-b-PCL nanoparticles BC gels were biosynthesized using Gluconacetobacter xylinus ATCC 12733 in Hestrin and Schramm medium for 3 weeks under static conditions (Hestrin and Schramm, 1954). Subsequently, the gels were purified using running water for 2 days, deproteinized in 0.5% (w/w) NaOH solution, neutralized in 0.5% (w/w) acetic acid solution, and washed with distilled water. BC gels containing PEO-b-PCL nanoparticles were prepared by diffusion of nanoparticles into the gels. BC gels (sample size = 2 cm 2 cm, thickness = 2 mm) were immersed in the PEO-b-PCL nanoparticle suspension (10 mg/mL) for 1 day at room temperature. BC gels containing retinol-loaded PEO-b-PCL nanoparticles were prepared by immersion of the gels in the PEO-b-PCL nanoparticle suspension (10 mg/mL, initial retinol/copolymer weight ratio of 0.15) for 1 day at 4 C. 2.4. Characterization of PEO-b-PCL nanoparticles 2.4.1. Dynamic light scattering The hydrodynamic radius of the obtained nanoparticles in aqueous media and acetic acid-sodium acetate buffer (pH 5.2) was measured using dynamic light scattering (DLS) at 25 C. The hydrodynamic diameter was calculated from the Stokes–Einstein relation. The instrument was an ALV/CGS-8FS/ N069 apparatus (ALV GmbH, Langen, Germany) equipped with an ALV/LSE-5004 multiple tau digital correlator with a 125 ns initial sampling time (ALV GmbH) and a 35 mW red helium–neon linearly polarized laser operating at a wavelength of 632.8 nm (JDS Uniphase Corporation, Milpitas, CA, USA). The concentration of the block copolymer was 1 mg/mL, and it was diluted using distilled water or acetic acid-sodium acetate buffer (pH 5.2). 2.4.2. Transmission electron microscopy The morphology of the PEO-b-PCL nanoparticles was observed under a Philips CM200 microscope (Royal Philips, Amsterdam, Netherlands) operated at 120 kV. PEO-b-PCL nanoparticles in aqueous media (4 mL) or acetic acid–sodium acetate buffer (pH 5.2) (4 mL), with a block copolymer concentration of 1 mg/mL, were dropped on a glow-discharge carbon-coated copper grid. Next, 2% (w/v) uranyl acetate negative stain (4 mL) was added and allowed to dry completely. 2.4.3. Retinol encapsulation The amount of retinol in the PEO-b-PCL nanoparticles was determined using a Perkin-Elmer Lambda 10 UV/Vis spectrophotometer (PerkinElmer, Inc., Waltham, MA, USA) at 325 nm. The retinol-loaded PEO-b-PCL nanoparticle suspension was dissolved in acetonitrile to extract the retinol from the nanoparticles. The retinol encapsulation efficiency was calculated using the following equation: Loaded retinolðmgÞ Encapsulation ð%Þ ¼ Initial retinolðmgÞ
2.4.4. Stability of encapsulated retinol The retinol-loaded PEO(5000)-b-PCL(5000) nanoparticle suspension was diluted so that the concentrations of block copolymer and retinol were 5 mg/mL and 1.5 mg/mL, respectively. Next, the acetic acid–sodium acetate buffer (pH 5.2) (19 mL) and the
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
nanoparticle suspension (1 mL) were mixed and kept at 37 C. Periodically, the retinol-loaded PEO-b-PCL nanoparticles were sampled from the suspension and dissolved in acetonitrile. The amount of retinol was determined by measuring the absorbance of the solution at 325 nm. 2.5. Characterization of BC gel containing PEO-b-PCL nanoparticles 2.5.1. Fourier transform infrared (FT-IR) spectroscopy FT-IR spectroscopy was used to confirm the introduction of the PEO-b-PCL nanoparticles into the BC gels. The instrument was a Perkin-Elmer spectrum RXI FTIR spectrometer (PerkinElmer, Inc., Waltham, MA, USA). The samples of BC gel and the BC gel containing PEO(10,000)-b-PCL(5000) nanoparticles were freeze-dried to prepare them for FT-IR spectroscopy. For each measurement, the sample was pressed into a pellet with KBr powder and 64 spectra were accumulated at a resolution of 4 cm1. 2.5.2. High-resolution scanning electron microscopy The BC fiber and the BC gel containing PEO-b-PCL nanoparticles were analyzed using a Zeiss Ultra 55 field-emission gum scanning electron microscope (FEG-SEM) (Zeiss, Germany) in the secondary electron imaging mode. The samples were freeze-dried and deposited on copper stubs with silver adhesion. The samples were sputtered with 6 nm of Au/Pd.
Introduced copolymerðmg=mLÞ ¼
2.6.1. Retinol release from nanoparticles The amount of retinol released from the nanoparticles was calculated on the basis of the retinol residue in the nanoparticles. The concentration of retinol residue was determined from the amount of retinol in acetic acid–sodium acetate buffer (pH 5.2) as described above. The nanoparticle suspension (1 mL; 10 mg/L copolymer concentration, 0.500 initial retinol/copolymer weight ratio) was added to acetic acid–sodium acetate buffer (19 mL, pH 5.2), and the solution was kept at 37 C. Periodically, the retinolloaded PEO(5000)-b-PCL(5000) nanoparticles were sampled from the suspension. To calculate the amount of retinol in the nanoparticles, the solution was centrifuged at 5673 g for 5 min and dissolved in acetonitrile. The amount of retinol was determined by measuring the UV absorbance of the solution at 325 nm. The amount of retinol released from the nanoparticles (NP) was calculated from the concentration of retinol using the following equation: Table 1 Characteristics of PEO-b-PCL micelles with copolymers of different molecular weights. Nanoparticle size SD (nm)
PEO(5000)-b-PCL(5000) in water PEO(5000)-b-PCL(10,000) in water PEO(2000)-b-PCL(11,500) in water PEO(5000)-b-PCL(5000) in buffer PEO(5000)-b-PCL(10,000) in buffer PEO(2000)-b-PCL(11,500) in buffer
39.7 71.8 29.4 38.0 73.8 32.0
Data are presented as mean SD (n = 3).
1.6 1.5 1.0 0.45 2.0 1.4
Polydispersity index SD 0.245 0.095 0.141 0.019 0.215 0.11 0.125 0.0078 0.116 0.011 0.170 0.071
Retinol released from NPð%Þ ¼
Initial loaded retinolðmg=mLÞ Retinol residue in NPðmg=mLÞ Initia l loaded retinolðmg=mLÞ
where the initial loaded retinol (mg/mL) was calculated from the encapsulated retinol, and the retinol residue in NP (mg/mL) was calculated from the detected value as described above. 2.6.2. PEO-b-PCL nanoparticle release from BC gels BC gels containing PEO-b-PCL nanoparticles were placed into acetic acid–sodium acetate buffer (pH 5.2) (2 mL) at 37 C. The amount of released PEO-b-PCL was measured using an analytical method for nonionic surfactants (Boyer et al., 1977), which has been used to measure cobalt thiocyanate active substance (CTAS) in biodegradation and environmental studies. Cobalt thiocyanate solution was prepared by dissolving Co(NO3)6H2O (30 g) and NH4SCN (200 g) in 1 L of distilled water. The buffer containing the released PEO-b-PCL (200 mL) dissolved in methylene chloride (CH2Cl2, 5 mL) was added to the cobalt thiocyanate solution (5 mL) and shaken for 5 min. After a few minutes, the solution separated into an aqueous layer and an organic layer containing PEO-b-PCL. The amount of PEO-b-PCL was determined by measuring the absorbance of the organic layer using a Perkin-Elmer Lambda 10 UV/Vis spectrophotometer (PerkinElmer, Inc., Waltham, MA, USA) at 626 nm. The introduced PEO-b-PCL copolymer concentration was calculated from the concentration of nanoparticles in the suspension using the following equation:
Copolymer introduced BC gel weightðmgÞ Initial BC gel weightðmgÞ Initial BC gel volumeðmLÞ
2.6. In vitro release
Copolymer
219
where the initial BC gel volume (mL) was regarded as equal to the initial BC gel weight (g), because the BC fiber weight was negligible in comparison with the water in the gel. The copolymer introduced BC gel weight is the weight of the BC gel after immersion in the nanoparticle suspension. The released copolymer (%) was calculated from the following equation: Detected copolymerðmg=mLÞ Released copolymerð%Þ ¼ Introduced copolymerðmg=mLÞ
2.6.3. Retinol release from BC gels BC gels containing retinol-loaded PEO-b-PCL nanoparticles were prepared as described in Section 2.3, except that the gel weight was measured before the gel was immersed in the nanoparticle suspension. After the gel was removed from the nanoparticle suspension, the weight was measured again, and the gel was immersed in acetic acid–sodium acetate buffer (pH 5.2) (2 mL) and kept at 37 C. The amount of retinol released from the BC gel was determined by measuring the absorbance of the solution at 325 nm as described in Section 2.4, and calculated using the following equation: Retinol released from BCð%Þ ¼
Detected retinolðmg=mLÞ Introduced retinol in BC gelðmg=mLÞ
where the introduced retinol in BC gel (mg/mL) was calculated from the retinol concentration of the nanoparticle suspension residue after the BC gel was removed, because the retinol concentration in the BC gel was assumed to be equal to the retinol concentration of the nanoparticle suspension residue after the BC gel was removed.
220
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
3. Results and discussion 3.1. Characterization of PEO-b-PCL nanoparticles Block copolymer PEO-b-PCL nanoparticles were prepared using a co-solvent evaporation method (Aliabadi et al., 2005b). The hydrodynamic diameters and polydispersity indices of nanoparticles prepared using different molecular weights of PEO-b-PCL determined by DLS are shown in Table 1. The mean diameters of PEO(5000)-b-PCL(5000), PEO(5000)-b-PCL(10,000), and PEO (2000)-b-PCL(11,500) nanoparticles were approximately 39.7, 71.8, and 29.4 nm, and their polydispersity indices were 0.245, 0.141, and 0.215. A previous study using light scattering showed that the diameters of PEO(5000)-b-PCL(5000) and PEO(5000)-bPCL(13,000) nanoparticles prepared by a co-solvent evaporation method were 87.5 28.2 nm and 78.7 11.8 nm, respectively, and their polydispersity indices were 0.198 and 0.111, respectively (Aliabadi et al., 2005b). The sizes and polydispersity indices of the nanoparticles reported here are smaller than those reported in previous studies. Preparation conditions can affect nanoparticle size, and three important factors influencing the results of the cosolvent evaporation methods are as follows: the type of organic cosolvent, the ratio of the volume of the organic phase to the volume of the aqueous phase during the assembly process, and the order of addition of the two phases (Aliabadi et al., 2007). In addition, the stirring time of the mixture after the drop-wise addition of the components influences the size of the nanoparticles (Aliabadi et al., 2007, 2005b). Our nanoparticle preparation conditions were different from those reported in previous studies; we used smaller drops and a longer stirring time. In the co-solvent evaporation method, copolymers dissolved in an organic solution assemble and form micelles when the organic solution is mixed with the aqueous solution, and the copolymers are impossible to dissolve in the organic–aqueous mixture solution. The size of the PEO(5000)-bPCL(5000) nanoparticles reported in this study was significantly different from that of previous reports, and this small size may explain the finding that PEO(5000)-b-PCL(5000) dissolved well in acetone and was easily affected by decreased solubility when the acetone solution was mixed with the aqueous solution. The PEO(2000)-b-PCL(11,500) nanoparticles were smaller in size than the other copolymers. PEO(5000)-b-PCL(5000) and PEO (5000)-b-PCL(10,000) nanoparticles were selected for subsequent studies. The nanoparticles were applied to a DDS for human skin. The physicochemical characteristics of the nanoparticles in the buffer are important for the effectiveness of the DDS. The hydrodynamic diameter of the nanoparticles was measured by DLS in acetic acid– sodium acetate buffer with a pH of 5.2, the same pH as human skin. These results indicated that the nanoparticles maintained their size in the acetic acid–sodium acetate buffer (Table 1). In addition, the copolymer nanoparticles were analyzed using transmission electron microscopy (TEM). The TEM images of PEO(5000)-b-PCL (5000) and PEO(5000)-b-PCL(10,000) nanoparticles in aqueous solution and acetic acid–sodium acetate buffer are shown in Fig. 1. A good correlation was observed between the TEM images and DLS results. 3.2. Preparation and characterization of retinol-loaded PEO-b-PCL nanoparticles Retinol was selected as a test compound because it is a typical active hydrophobic molecule. Retinol is used as an anti-aging ingredient in cosmetic products. Retinol-loaded nanoparticles were prepared in the same way as PEO-b-PCL nanoparticles, except that both PEO-b-PCL and retinol were dissolved in acetone. The physicochemical characteristics of retinol-loaded PEO-b-PCL
Fig. 1. TEM images of PEO-b-PCL nanoparticles; (a) PEO(5000)-b-PCL(5000) in aqueous solution, (b) PEO(5000)-b-PCL(10,000) in aqueous solution, (c) PEO(5000)b-PCL(5000) in acetic acid–sodium acetate buffer solution (pH 5.2), and (d) PEO (5000)-b-PCL(10,000) in acetic acid–sodium acetate buffer solution (pH 5.2).
nanoparticles are shown in Table 2. The hydrodynamic diameter of the retinol-loaded PEO(5000)-b-PCL(5000) nanoparticles was the same as that of the unloaded nanoparticles. The retinol-loaded PEO(5000)-b-PCL(10,000) copolymers formed larger particles than those formed by the retinol-loaded PEO(5000)-b-PCL(5000), similar to the unloaded nanoparticles. In addition, retinol-loaded PEO(5000)-b-PCL(10,000) nanoparticle (initial retinol/copolymer weight ratio 0.15) size showed a similar trend to retinol-loaded PEO(5000)-b-PCL(5000) (initial retinol/copolymer weight ratio 0.15 and 0.30) size. However, the retinol-loaded PEO(5000)-b-PCL (10,000) nanoparticles (initial retinol/copolymer weight ratio 0.75) were 16% larger than the unloaded nanoparticles. The encapsulation efficiency of retinol in the PEO(5000)-b-PCL (5000) nanoparticles was 83.9% (initial retinol/copolymer weight ratio 0.15), 84.1% (initial retinol/copolymer weight ratio 0.30), and 66.9% (initial retinol/copolymer weight ratio 0.75). This result suggests that the PEO(5000)-b-PCL(5000) nanoparticles (initial retinol/copolymer weight ratio 0.15) have sufficient space to contain more retinol. The encapsulation efficiencies of some hydrophobic substrates into PEO(5000)-b-PCL(5000) nanoparticles have been reported, including curcumin (3.05%, initial curcumin/copolymer weight ratio 0.10) (Ma et al., 2008), doxorubicin (DOX; 4.03%; initial DOX/copolymer weight ratio 0.20) (Shuai et al., 2004), hydrophobic fluorescent probe DiI (80%; initial DiI/copolymer weight ratio 0.0010) (Mahmud and Lavasanifar, 2005), and cyclosporine A (CsA; 52.2%; initial CsA/copolymer weight ratio 0.10) (Aliabadi et al., 2005b). Retinol is more suitable than other hydrophobic substrates for loading into PEO(5000)-bPCL(5000) nanoparticles. In general, drug hydrophobicity plays a decisive role in the drug-loading process. Retinol is highly hydrophobic and interacts strongly with PCL. For retinol-loaded PEO(5000)-b-PCL(10,000) nanoparticles, the encapsulation efficiencies were 80.5% (initial retinol/copolymer weight ratio 0.15) and 81.5% (initial retinol/copolymer weight ratio 0.75). In previous studies, the encapsulation efficiency for paclitaxel increased significantly with increases in PCL length (Aliabadi et al., 2005b; Mahmud and Lavasanifar, 2005; Shuai et al., 2004), whereas a
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
221
Table 2 Physicochemical characteristics of retinol-loaded PEO-b-PLC nanoparticles in aqueous solution. Copolymer
Initial retinol (mg)
Initial retinol /copolymer weight ratio Encapsulation SD Encapsulation SD Nanoparticle size SD Polydispersity (mg/mg) (mg) (%) (nm) index SD
PEO(5000)-b-PCL (5000)
3 6 15 3 15
0.15 0.3 0.75 0.15 0.75
PEO(5000)-b-PCL (10,000)
2.52 5.05 10.0 2.42 12.2
0.16 0.38 0.39 0.12 0.23
83.9 84.1 66.9 80.5 81.5
5.2 6.3 2.6 4.1 1.5
38.4 38.3 39.2 70.2 83.3
0.85 0.79 0.20 2.5 1.2
0.112 0.097 0.098 0.0944 0.121
0.001 0.0034 0.022 0.023 0.014
Data are presented as mean SD (n = 3).
lesser effect was observed in this study, due to the low initial retinol/copolymer weight ratio (0.15). When a high initial retinol/ copolymer weight ratio (0.75) was used, PCL length influenced retinol encapsulation. These results suggest that PEO-b-PCL can encapsulate a large amount of retinol. Retinol-loaded nanoparticles in pH 5.2 acetic acid–sodium acetate buffer were analyzed by DLS. The retinol-loaded nanoparticles maintained their size in the acetic acid–sodium acetate buffer, similar to retinol-non-loaded nanoparticles, as indicated by measurements of hydrodynamic diameter. TEM images of retinolloaded PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticles (initial retinol/copolymer weight ratio 0.15) in acetic acid–sodium acetate buffer are shown in Fig. 2. The shape of the nanoparticles was not affected by drug loading. 3.3. Stability of retinol encapsulated in PEO-b-PCL nanoparticles Retinol is very unstable and easily oxidized (Szuts and Harosi, 1991; Tsunoda and Takabayashi, 1995). In a previous study, the stability of the unstable substrate curcumin increased when it was encapsulated into PEO-b-PCL nanoparticles (Mahmud and Lavasanifar, 2005). Therefore, the stability of retinol was also expected to increase when it was encapsulated into PEO-b-PCL nanoparticles. The stability of free retinol and retinol encapsulated in PEO(5000)-b-PCL(5000) nanoparticles was evaluated in acetic acid–sodium acetate buffer (pH 5.2) at 37 C, and the results are shown in Fig. 3. The concentration of free retinol in the buffer solution was decreased by 70% after 45 min, and free retinol in the buffer solution was almost undetectable after 1 day. However, the retinol-encapsulated nanoparticles retained more than 50% of the loaded retinol after 8 h, and approximately 10% of the loaded retinol remained after 1 week, indicating that encapsulation into PEO-b-PCL nanoparticles can increase the stability of hydrophobic active molecules. In a previous study on the stability of retinolloaded nanoliposomes (Ko and Lee, 2010), the authors placed the retinol-loaded nanoliposome suspension in a glass vial with oxygen at 37 C, and approximately 50% of the retinol remained after 2 days. In contrast, in this study, approximately 25% of the
Fig. 2. TEM images of retinol-loaded PEO-b-PCL nanoparticles in acetic acid– sodium acetate buffer solution (pH 5.2); (a) PEO(5000)-b-PCL(5000), (b) PEO (5000)-b-PCL(10,000). The initial retinol/copolymer weight ratio (mg/mg) was 0.15.
retinol remained after 2 days, and this difference was likely caused by the retinol incorporation method used. Nanoliposomes incorporate retinol into lipid bilayers, whereas PEO-b-PCL nanoparticles incorporate retinol inside of the micelle, from which it is more likely to be released. 3.4. Preparation of PEO-b-PCL nanoparticle-containing BC gel Nanoparticle-containing BC gel was prepared using a diffusion method. The BC gel was placed in the nanoparticle suspension to absorb the nanoparticles. The FT-IR spectra of the gels were measured to establish that the PEO-b-PCL copolymers were incorporated into the BC gel. Fig. 4 shows the FT-IR spectra of the freeze-dried BC gel, the freeze-dried nanoparticle-containing BC gel, and the PEO-b-PCL powder. A C¼O bond indicative of PCL was observed at 1725 cm1 in the spectra of the PEO-b-PCL powder and the nanoparticle-containing BC gel, but this bond was not observed in the BC gel spectrum. These results indicate that the PEO-b-PCL copolymers were incorporated into the BC gel. To establish the relative position between PEO-b-PCL and BC, the nanoparticle-containing BC gel was analyzed by FEG-SEM. The FEG-SEM images of BC gel prepared with or without nanoparticle are shown in Fig. 5. Very fine cellulose fibers were observed in the pure BC gel (Fig. 5(a)). However, aggregates of PEO-b-PCL were observed on the surface of the PEO(5000)-b-PCL(10,000) nanoparticle-containing BC gel (Fig. 5(b) and (c)). The nanoparticles gathered to the cellulose fibers on the surface of the gel and became stuck to them. FEG-SEM images were obtained of the cross-sections of the BC gels containing PEO(5000)-b-PCL(5000) (Fig. 5(d) and (e)) and PEO(5000)-b-PCL(10,000) (Fig. 5(f) and (g))
Fig. 3. Effect of encapsulation into PEO(5000)-b-PCL(5000) nanoparticles in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. Filled gray circle: retinolloaded nanoparticles (NP(retinol)) suspension; filled black diamond: retinol alone in solution (free retinol). The results are shown as the mean SD.
222
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
C=O
T%
BC BC+NP
PEO-b-PCL 2000
1800
1600
wavenumber
1400
1200
(cm-1)
Fig. 4. FT-IR spectra of the BC gel, PEO(5000)-b-PCL(10,000) nanoparticlecontaining BC gel, and PEO(5000)-b-PCL(10,000) powder.
nanoparticles. Some large particles were also observed inside the samples. Nanoparticles of similar size to those observed in the TEM images were not observed in the SEM cross-section images, which showed 770 nm (Fig. 5(d)) and 80 nm (Fig. 5(e)) particles in the PEO (5000)-b-PCL(5000) nanoparticle-containing BC gel, as well as 290 nm (Fig. 5(f)) and 20 nm (Fig. 5(g)) particles in the PEO(5000)b-PCL(10,000) nanoparticle-containing BC gel. In comparison with TEM, it is difficult to observe small size nanoparticles using FEGSEM. However, the results showed that the PEO-b-PCL copolymers were introduced into the BC gel as well as onto its surface. To determine the amount of nanoparticles incorporated into the BC gel, PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticle residue suspensions were subjected to measurement using the CTAS method after the BC gel was removed (Boyer et al., 1977). The incorporation ratios for PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticles were 99.7% and 93.6%, respectively. The ratio was calculated based on the weight of the nanoparticle-containing gel and the residue nanoparticle suspension after nanoparticle diffusion. There were more PEO(5000)-bPCL(5000) nanoparticles than PEO(5000)-b-PCL(10,000) nanoparticles in the BC gel. The PEO(5000)-b-PCL(5000) nanoparticles
Fig. 5. FEG-SEM images of the BC gel and nanoparticle-containing BC gel. (a) BC cross-section; (b) and (c) PEO(5000)-b-PCL(10,000) nanoparticle-containing near the BC surface; (d) and (e) PEO(5000)-b-PCL(5000) nanoparticle-containing BC cross-section; and (f) and (g) PEO(5000)-b-PCL(10,000) nanoparticle-containing BC cross-section.
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
223
were smaller than the PEO(5000)-b-PCL(10,000) nanoparticles, allowing them to more easily pass between cellulose fibers. 3.5. In vitro release studies 3.5.1. Copolymer release from BC gel The amount of copolymer (formed and unformed nanoparticle ratio) released from the BC gel in acetic acid–sodium acetate buffer (pH 5.2) at 37 C was determined at defined time intervals using the CTAS method (Boyer et al., 1977). The results show that the combination of BC gel and nanoparticles can be used as a slowrelease system (Fig. 6). PEO(5000)-b-PCL(5000) nanoparticles were released more rapidly than PEO(5000)-b-PCL(10,000) nanoparticles at early time points. The released copolymer ratio reached a maximum of approximately 60% after 6 h and remained constant after 24 h. Although the small particles were more easily released, the total amount of released nanoparticles was not dependent on nanoparticle size. The nanoparticle release ratios did not reach 100%, suggesting that a portion of the nanoparticles remained inside the BC gel and on its surface. The FEG-SEM images suggested that some PEO-b-PCL nanoparticles stuck to the cellulose fibers, and also showed that large PEO-b-PCL particles existed in the BC gel. PEO (PEG) has high affinity for BC fibers (Numata et al., 2009); therefore, PEO-b-PCLs are difficult to remove from the gel completely. The release of copolymers after 24 h was analyzed using DLS. The results showed that nanoparticle size did not change significantly after nanoparticle release from the BC gel (data not shown). Fig. 7 shows TEM images of released copolymers after 24 h in acetic acid–sodium acetate buffer (pH 5.2). The TEM images corroborated the DLS results: the copolymers remained in the formed nanoparticles and the size did not change. These results indicate that BC gel is a good carrier for PEO-b-PCL nanoparticles. 3.5.2. Retinol release from PEO-b-PCL nanoparticles To determine the amount of retinol released from PEO(5000)-bPCL(5000) nanoparticles in acetic acid–sodium acetate buffer (pH 5.2) at 37 C, the nanoparticle-containing buffer was centrifuged to remove the retinol present outside the nanoparticles. The released retinol ratio was calculated after measuring the absorption of the
Fig. 7. TEM images of nanoparticles released from the BC gel in acetic acid–sodium acetate buffer solution (pH 5.2). (a) PEO (5000)-b-PCL (5000); (b) PEO (5000)-b-PCL (10,000).
retinol-containing nanoparticles (Fig. 8). The results show that 55% of retinol encapsulated in the nanoparticles was released after 8 h. The residue retinol ratio at the same time point was 65% (Fig. 3), indicating that 10% of the retinol in the nanoparticle suspension was located outside of the nanoparticles and precipitated in suspension. After 24 h, the residue retinol ratio in the suspension was 35.9% (Fig. 3), and ratio of retinol released from the nanoparticles was 66.4% (Fig. 8), indicating that 33.6% of the retinol remained in the nanoparticles. Consequently, the retinol in suspension was calculated to be 2.3%, indicating that the retinol present in the suspension was almost entirely encapsulated in nanoparticles. Retinol released outside the nanoparticles was almost undetectable because the retinol was unstable in the buffer. 3.5.3. Encapsulated retinol release from BC gel To determine the quantities of nanoparticle-encapsulated retinol and free retinol released from the BC gel, we performed a diffusion experiment with PEO(5000)-b-PCL(5000) and PEO (5000)-b-PCL(10,000) retinol-loaded nanoparticles. The BC gel was placed in the retinol-loaded nanoparticle suspension for 1 day at 4 C to prevent retinol deterioration. The PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) retinol-loaded nanoparticle suspensions retained 78.6% and 71.9% of the initial retinol amount, respectively, after 1 day at 4 C. In the assay of the effect of encapsulation into nanoparticles in buffer (pH 5.2) at 37 C, only
100
80
PEO(5000)-b-PCL(5000)
0.8 80
Released retinol [%]
Released copolymer [%]
1001
60 PEO(5000)-b-PCL(10,000)
40
20
0.6 60
0.4 40
0.2 20
0
0
5
10
15
20
25
Time [h] Fig. 6. Released copolymer ratio from the BC gel in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C; open gray circle: PEO(5000)-b-PCL(5000); open black circle: PEO(5000)-b-PCL(10,000). The results are shown as the mean SD.
00 0
20
40
60
80
100
[h] Time [hr] Fig. 8. Ratio of retinol released from PEO(5000)-b-PCL(5000) nanoparticles in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. The results are shown as the mean SD.
224
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
4. Conclusion
10 PEO(5000)-b-PCL(10,000)
Released retinol [%]
8
6 PEO(5000)-b-PCL(5000)
4
2
0 0
5
10
15
20
25
Time [h] Fig. 9. Retinol with nanoparticles released from BC gel in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C; gray square: PEO(5000)-b-PCL(5000); black triangle: PEO(5000)-b-PCL(10,000). The results are shown as the mean SD.
35.9% (PEO(5000)-b-PCL(5000), Fig. 3) of the loaded retinol remained after 1 day. These results show that retinol deterioration was prevented by encapsulation. Next, the nanoparticle-encapsulated retinol containing gel was placed in acetic acid–sodium acetate buffer (pH 5.2) at 37 C and the absorption was measured at defined time intervals (Fig. 9). We assumed that the BC gel contained the same concentration of retinol-loaded nanoparticles as the nanoparticle suspension after the BC gel was removed. The ratio of released retinol from the nanoparticles was calculated on the basis of the initial retinol concentration of BC gel. The retinol ratio reached its maximum of 7.7% ((PEO(5000)-b-PCL(5000)) at 4 h and 8.7% (PEO(5000)-b-PCL(10,000)) at 4 h, and started to decrease after another hour. The ratio of released copolymer reached a maximum at 6 h, similar to the released copolymer ratio from the BC gel (Fig. 6), suggesting that the detected retinol was contained in the nanoparticles. However, the detected amount of released retinol was smaller than the amount of retinol released from the nanoparticles (Fig. 8). Most of the retinol released from the nanoparticles precipitated and remained in the BC gel. Once retinol was released from the nanoparticles, it was precipitated in the aqueous solution. Therefore, the precipitated retinol was trapped by the cellulose fibers. The detected amount of released retinol was consequently small in comparison with the amount of retinol released from the nanoparticles. However, the rate of retinol release from the BC gel decreases after 24 h, indicating that the released of retinol into the buffer solution decreases slowly, which is consistent with the effect of encapsulation into nanoparticles (Fig. 3). The system described herein requires some improvement. First, the quantities of introduced retinol should be considered. In previous studies, the encapsulation ratios were smaller than that used in this study, whereas the same drug/copolymer weight ratio was used (Aliabadi et al., 2005b; Ma et al., 2008). A large loaded drug releases easily because the drug is likely to be distributed from the nanoparticle core. The quantity of retinol release and the release rate of loaded-retinol are difficult to control. Second, if an antioxidant substance is added to the nanoparticle suspension or loaded into the nanoparticle with retinol, the system can release larger quantities of retinol.
We have developed a DDS based on diffusion of retinol-loaded PEO-b-PCL nanoparticles from BC gel. The nanoparticles increased the stability of the encapsulated retinol. The retinol-loaded nanoparticles released retinol slowly. However, most of the retinol released from the nanoparticles was precipitated and remained in the BC gel, and this problem requires further study to overcome. The system described herein can contribute considerably to cosmetic and medical applications, such as skin treatment and skin tissue repair. PEO-b-PCL nanoparticles can also encapsulate other hydrophobic molecules (Aliabadi et al., 2005b; Ghoroghchian et al., 2006; Ma et al., 2008; Mahmud and Lavasanifar, 2005; Shuai et al., 2004). Stable, active ingredients can be delivered with this system, which has the potential to increase the quantities of drugs released from BC gel. PEO-b-PCL can also form composite nanoparticles encapsulating hydrophobic organic and inorganic compounds, and these nanoparticles can serve as multimodal carriers for drugs and imaging agents (Gindy et al., 2008). Additionally, PEO-b-PCL forms micelles, and also forms vesicles with a hydrophilic membrane surface, which are capable of encapsulating hydrophilic molecules (Ding and Liu, 1997; Kukula et al., 2002; Nardin et al., 2000). Therefore, the system described herein has potential uses in various skin preparations. Acknowledgments The authors thank Amandine Durand-Terrasson for her help with the TEM and FEG-SEM experiments. The authors acknowledge support from CNRS, PolyNat Carnot Institute, and LabEx ARCANE (ANR-11-LABX-0003-01), France. Yukari Numata was funded by the Institute of National Colleges of Technology, Japan. References Aliabadi, H.M., Brocks, D.R., Lavasanifar, A., 2005a. Polymeric micelles for the solubilization and derivery of cyclosporine A: pharmacokinetics and biodistribution. Biomaterials 26, 7251–7259. Aliabadi, H.M., Mahmud, A., Sharifabadi, A.D., Lavasanifar, A., 2005b. Micelles of methoxy poly(ethylene oxide)-b-poly(e-caprolactone) as vehicles for the solubilization and controlled delivery of cyclosporine A. J. Control. Release 104, 301–311. Aliabadi, H.M., Elhasi, S., Mahmud, A., Gulamhusein, R., Mahdipoor, P., Lavasanifar, A., 2007. Encapsulation of hydrophobic drugs in polymeric micelles through cosolvent evaporation: the effect of solvent composition on micellar properties and drug loading. Int. J. Pharm. 329, 158–165. Allen, C., Maysinger, D., Eisenberg, A., 1999. Nano-engineering block copolymer aggregates for drug delivery. Colloids Surf. B 16, 3–27. Boyer, S.L., Guin, K.F., Kelley, R.M., Mausner, M.L., Robinson, H.F., Schmitt, T.M., Stahl, C.R., Setzkorn, E.A., 1977. Analytical method for nonionic surfactants in laboratory biodegradation and environmental studies. Environ. Sci. Technol. 11, 1167–1171. Brown, J.M.B., 1996. The biosynthsis of cellulose. J. Macromol. Sci. Part A Pure Appl. Chem. 33, 1345–1373. Ding, J., Liu, G., 1997. Polyisoprene-block-poly(2-cinnamoylethyl methacrylate) vesicles and their aggregates. Macromolecules 30, 655–657. Fu, L., Zhang, J., Yang, G., 2013. Present status and applications of bacterial cellulosebased materials for skin tissue repair. Carbohydr. Polym. 92, 1432–1442. Ghoroghchian, P.P., Li, G., Levine, D.H., Davis, K.P., Bates, F.S., Hammer, D.A., Therien, M.J., 2006. Bioresorbable vesicles formed through spontaneous self-assembly of amphiphilic poly(ethylene oxide)-block-polycaprolactone. Macromolecules 39, 1673–1675. Gindy, M.E., Panagiotopoulos, A.Z., Prud’homme, R.K., 2008. Composite block copolymer stabilized nanoparticles: simultaneous encapsulation of organic active and inorganic nanoparticles. Langmuir 24, 83–90. Hestrin, S., Schramm, M., 1954. Syntheie of cellulose by Acetobacter xylinum. Biochem. J. 58, 345–352. Iguchi, M., Yamanaka, S., Budhiono, A., 2000. Review bacterial cellulose – a masterpiece of nature’s arts. J. Mater. Sci. 35, 261–270. Kim, S.Y., Shin, I.G., Lee, Y.M., Cho, C.S., Sung, Y.K., 1998. Methoxy poly(ethylene glycol) and e-caprolactone amphiphilic block copolymeric micelle containing indomethacin. II. Micelle formation and drug release behaviours. J. Control. Release 51, 13–22.
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225 Klemm, D., Heublein, B., Fink, H.P., Bohn, A., 2005. Cellulose: fascinating biopolymer and sustainble raw material. Angew. Chem. Int. Ed. 44, 2–37. Ko, S., Lee, S.C., 2010. Effect of nanoliposomes on the stabilization of incorporated retinol. Afr. J. Biotechnol. 9, 6158–6161. Kukula, H., Schlaad, H., Antonietti, M., Forster, S., 2002. The formation of polymer vesicles or peptosomes by polybutadiene-block-poly(L-glutamate)s in dilute aqueous solution. J. Am. Chem. Soc. 124, 1658–1663. Ma, Z., Haddadi, A., Molavi, O., Lavasanifar, A., Lai, R., Samuel, J., 2008. Micelles of poly(ethylene oxide)-b-poly(e-caprolactone) as vehicles for the solubilization, stabilization, and controlled delivery of curcumin. J. Biomed. Mater. Res. Part A 86A, 300–310. Mahmud, A., Lavasanifar, A., 2005. The effect of block copolymer structure on the internalization of polymeric micelles by human breast cancer cells. Colloids Surf. B 45, 82–89. Miyamoto, T., Takahashi, S., Ito, H., Inagaki, H., Noishiki, Y., 1989. Tissue biocompatibility of cellulose and its derivatives. J. Biomed. Mater. Res. 23, 125– 133. Nardin, C., Hirt, T., Leukel, J., Meier, W., 2000. Polymerized ABA triblock copolymer vesicles. Langmuir 16, 1035–1041.
225
Numata, Y., Muromoto, K., Furukawa, H., Gong, J.P., Tajima, K., Munekata, M., 2009. Nonvolatile and shape-memoried bacterial cellulose gels swollen by poly (ethylene glycol). Polym. J. 41, 524–525. Okiyama, A., Motoki, M., Yamanaka, S., 1993. Bacterial cellulose III. Development of a new form of cellulose. Food Hydrocolloids 6, 493–501. Ross, P., Mayer, R., Benziman, M., 1991. Cellulose biosynthesis and function in bacteria. Microbiol. Rev. 55, 35–58. Shah, N., UI-Islam, M., Khattak, W.A., Park, J.K., 2013. Overview of bacterial cellulose composites: a multipurpose advanced. Carbohydr. Polym. 98, 1585–1598. Shuai, X., Ai, H., Nasongkla, N., Kim, S., Gao, J., 2004. Micellar carriers based on block copolymers of poly(e-caprolactone) and poly(ethylene glycol) for doxorubicin delivery. J. Control. Release 98, 415–426. Szuts, E.Z., Harosi, F.I., 1991. Solubility of retinoids in water. Arch. Biochem. Biophys. 287, 297–304. Tsunoda, T., Takabayashi, K., 1995. Stability of all-trans-retinol in cream. J. Cosmet. Chem. 46, 191–198. Wei, X.W., Gong, C.Y., Gou, M.L., Fu, S.Z., Guo, Q.F., Shi, S., Luo, F., Guo, G., Qiu, L.Y., Qian, Z.Y., 2009. Biodegradable poly(e-caprolactone)-poly(ethylene glycol) copolymers as drug delivery system. Int. J. Pharm. 381, 1–18.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
A slow-release system of bacterial cellulose gel and nanoparticles for hydrophobic active ingredients Yukari Numata a,b, * , Leticia Mazzarino a,c , Redouane Borsali a a
Centre de Recherches sur les Macromoléculés Végétales (CERMAV, UPR-CNRS 5301), Université Grenoble Alpes, BP 53, F-38041 Grenoble Cedex 9, France Department of Materials Chemistry, Asahikawa National College of Technology, Asahikawa 071-8142, Japan c Departamento de Ciências Farmacêuticas, Universidade Federal de Santa Catarina, Florianópolis 88040-900, Brazil b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 9 January 2015 Received in revised form 6 March 2015 Accepted 30 March 2015 Available online 31 March 2015
A combination of bacterial cellulose (BC) gel and amphiphilic block copolymer nanoparticles was investigated as a drug delivery system (DDS) for hydrophobic active ingredients. Poly(ethylene oxide)-bpoly(caprolactone) (PEO-b-PCL) and retinol were used as the block copolymer and hydrophobic active ingredient, respectively. The BC gel was capable of incorporating copolymer nanoparticles and releasing them in an acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. The percentage of released copolymer reached a maximum value of approximately 60% after 6 h and remained constant after 24 h. The percentage of retinol released from the copolymer-containing BC gel reached a maximum value at 4 h. These results show that the combination of BC gel and nanoparticles is a slow-release system that may be useful in the cosmetic and biomedical fields for skin treatment and preparation. ã 2015 Elsevier B.V. All rights reserved.
Chemical compounds studied in this article: Retinol (PubChem CID: 445354) Keywords: Bacterial cellulose gel Nanoparticles Block copolymer Biocompatibility Drug delivery system Retinol
1. Introduction Bacterial cellulose (BC) gel is a unique hydrogel consisting of greater than 99% (w/w) water by weight. BC gels have a threedimensional network structure of ultrafine fiber made from pure cellulose (Brown, 1996; Iguchi et al., 2000; Ross et al., 1991) and have various useful properties, such as softness, translucence, and good biocompatibility and water retention capacity (Klemm et al., 2005; Miyamoto et al., 1989; Okiyama et al., 1993). BC gels have been extensively investigated as potential soft materials for use in biomedical fields (Fu et al., 2013; Shah et al., 2013). In skin tissue repair, the advantages of BC, such as biocompatibility, conformability, elasticity, transparency, the ability to maintain a moist environment in the wound, and the ability to absorb exudate during the inflammatory phase, confer great potential for application in wound healing systems. In cosmetic applications, BC gels offer good biocompatibility, conformability, elasticity, and transparency, as well as the ability to maintain a moist
* Corresponding author. Present address: General Education, Faculty of Commerce, Otaru University of Commerce, 3-5-21, Midori, Otaru 047-8501, Japan. Tel.: +81 134 27 5412; fax: +81 134 27 5412. E-mail address: [email protected] (Y. Numata). http://dx.doi.org/10.1016/j.ijpharm.2015.03.068 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
environment for active ingredients. BC gels are commercially available as a cosmetic facial mask in Japan. BC facial masks adhere well to the face and retain moisture better than normal cellulose sheets, because BC consists of very thin nano-fibers. However, the incorporation of hydrophobic active ingredients into BC gels remains a challenging task. New methods to effectively incorporate hydrophobic molecules would allow their application in BC gel instead of in oily cosmetic products such as facial creams; therefore, the development of such methods will be useful in the cosmetics market and in biomedical applications. Amphiphilic block copolymer systems are used to incorporate hydrophobic molecules, and these copolymers have the ability to self-assemble into well-organized structures, such as nanoparticles, micelles, and vesicles, in aqueous solutions. Micelles can have a diameter of 10–200 nm, but most have a diameter ranging from 10 to 100 nm. Polymers typically have molecular weights ranging from several hundred to several tens of thousands. Particle size depends on sample preparation conditions and the type of copolymer, as well as its components, molecular weight, and organic solvent composition (Aliabadi et al., 2007). Micelles can be produced by methods such as dialysis (Allen et al., 1999) and cosolvent evaporation (Aliabadi et al., 2005b). Nanoparticles produced from biocompatible and biodegradable copolymers have shown great potential as carriers for active ingredients in cosmetic
218
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
and pharmaceutical applications. Poly(ethylene oxide)-b-poly (caprolactone) (PEO-b-PCL) is composed of biocompatible and biodegradable synthetic polymers and is one of the most extensively studied copolymer systems for nanoparticle formation (Wei et al., 2009). The PEO-b-PCL copolymer forms nano-sized micelles, in which PEO is the hydrophilic shell-forming block and PCL is the hydrophobic core-forming block. Many researchers have reported encapsulation of hydrophobic active ingredients into PEO-b-PCL nanoparticles (Aliabadi et al., 2005a; Kim et al., 1998; Mahmud and Lavasanifar, 2005; Shuai et al., 2004). The ultimate objective of such studies was the development of a drug delivery system (DDS). PEO (poly(ethylene glycol) (PEG)), the hydrophilic shellforming block of PEO-b-PCL nanoparticles, has high affinity for cellulose (Numata et al., 2009). Therefore, we hypothesized that PEO-b-PCL nanoparticles can be introduced into BC gel, which can then be used to encapsulate hydrophobic molecules. If the BC gel can release PEO-b-PCL nanoparticles and the encapsulated hydrophobic molecules, then the system can be utilized in topical compound preparations in the cosmetic and the biomedical fields. In this study, we designed a DDS by combining BC gels with a nanoparticle-encapsulated hydrophobic active ingredient. We used PEO-b-PCL and retinol as the block copolymer and hydrophobic active ingredient, respectively. Retinol, vitamin A alcohol, is extensively used in the pharmaceutical and cosmetic industries and is a well-known anti-aging ingredient. The PEO-bPCL nanoparticle-containing BC gel was investigated for the presence of BC containing nanoparticles by Fourier transform infrared (FT-IR) spectra and field-emission gum scanning electron microscopy (FEG-SEM). The released nanoparticles indicated that the PEO-b-PCL nanoparticle-containing BC gel can be utilized as a DDS, and the retinol release indicated that packaged retinol can be released from the BC gel; however, the DDS must be improved to support practical use. Our results show that the DDS described herein may be useful in topical preparations in the cosmetic and biomedical fields. 2. Material and methods 2.1. Materials Three types of PEO-b-PCL block copolymers were purchased from Polymer Source, Inc. (Quebec, Canada). The molecular weights (Mw) of PEO and PCL in the three PEO-b-PCL block copolymers were 5000 and 5000, 5000 and 10,000, and 2000 and 11,500, respectively. The Mw/number average molecular weight (Mn) ratios of the three types of PEO-b-PCL copolymers were 1.20, 1.26, and 1.20. Retinol was purchased from Sigma–Aldrich Co. LLC (St. Louis, MO, USA). 2.2. Preparation of PEO-b-PCL nanoparticles and retinol-loaded PEOb-PCL nanoparticles PEO-b-PCL nanoparticles were prepared by co-solvent evaporation (Aliabadi et al., 2005b). PEO-b-PCL (10 mg) dissolved in acetone (0.5 mL) was added in a drop-wise manner (5.08 mL/h) to Milli-Q water (1 mL) with stirring. The remaining acetone was removed by evaporation at room temperature under vacuum. Retinol-loaded PEO-b-PCL nanoparticles were prepared by cosolvent evaporation as described above except for dissolving retinol and PEO-b-PCL in acetone. PEO-b-PCL (10 mg) and retinol (3, 6, or 15 mg) dissolved in acetone (0.5 mL) were added in a dropwise manner (5.08 mL/h) to Milli-Q water (1 mL) with stirring. The remaining acetone was removed by evaporation at room temperature under vacuum. The obtained nanoparticle suspension was
centrifuged at 5673 g for 5 min to remove the retinol precipitate from the outside of the nanoparticles. 2.3. Preparation of BC gel containing PEO-b-PCL nanoparticles BC gels were biosynthesized using Gluconacetobacter xylinus ATCC 12733 in Hestrin and Schramm medium for 3 weeks under static conditions (Hestrin and Schramm, 1954). Subsequently, the gels were purified using running water for 2 days, deproteinized in 0.5% (w/w) NaOH solution, neutralized in 0.5% (w/w) acetic acid solution, and washed with distilled water. BC gels containing PEO-b-PCL nanoparticles were prepared by diffusion of nanoparticles into the gels. BC gels (sample size = 2 cm 2 cm, thickness = 2 mm) were immersed in the PEO-b-PCL nanoparticle suspension (10 mg/mL) for 1 day at room temperature. BC gels containing retinol-loaded PEO-b-PCL nanoparticles were prepared by immersion of the gels in the PEO-b-PCL nanoparticle suspension (10 mg/mL, initial retinol/copolymer weight ratio of 0.15) for 1 day at 4 C. 2.4. Characterization of PEO-b-PCL nanoparticles 2.4.1. Dynamic light scattering The hydrodynamic radius of the obtained nanoparticles in aqueous media and acetic acid-sodium acetate buffer (pH 5.2) was measured using dynamic light scattering (DLS) at 25 C. The hydrodynamic diameter was calculated from the Stokes–Einstein relation. The instrument was an ALV/CGS-8FS/ N069 apparatus (ALV GmbH, Langen, Germany) equipped with an ALV/LSE-5004 multiple tau digital correlator with a 125 ns initial sampling time (ALV GmbH) and a 35 mW red helium–neon linearly polarized laser operating at a wavelength of 632.8 nm (JDS Uniphase Corporation, Milpitas, CA, USA). The concentration of the block copolymer was 1 mg/mL, and it was diluted using distilled water or acetic acid-sodium acetate buffer (pH 5.2). 2.4.2. Transmission electron microscopy The morphology of the PEO-b-PCL nanoparticles was observed under a Philips CM200 microscope (Royal Philips, Amsterdam, Netherlands) operated at 120 kV. PEO-b-PCL nanoparticles in aqueous media (4 mL) or acetic acid–sodium acetate buffer (pH 5.2) (4 mL), with a block copolymer concentration of 1 mg/mL, were dropped on a glow-discharge carbon-coated copper grid. Next, 2% (w/v) uranyl acetate negative stain (4 mL) was added and allowed to dry completely. 2.4.3. Retinol encapsulation The amount of retinol in the PEO-b-PCL nanoparticles was determined using a Perkin-Elmer Lambda 10 UV/Vis spectrophotometer (PerkinElmer, Inc., Waltham, MA, USA) at 325 nm. The retinol-loaded PEO-b-PCL nanoparticle suspension was dissolved in acetonitrile to extract the retinol from the nanoparticles. The retinol encapsulation efficiency was calculated using the following equation: Loaded retinolðmgÞ Encapsulation ð%Þ ¼ Initial retinolðmgÞ
2.4.4. Stability of encapsulated retinol The retinol-loaded PEO(5000)-b-PCL(5000) nanoparticle suspension was diluted so that the concentrations of block copolymer and retinol were 5 mg/mL and 1.5 mg/mL, respectively. Next, the acetic acid–sodium acetate buffer (pH 5.2) (19 mL) and the
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
nanoparticle suspension (1 mL) were mixed and kept at 37 C. Periodically, the retinol-loaded PEO-b-PCL nanoparticles were sampled from the suspension and dissolved in acetonitrile. The amount of retinol was determined by measuring the absorbance of the solution at 325 nm. 2.5. Characterization of BC gel containing PEO-b-PCL nanoparticles 2.5.1. Fourier transform infrared (FT-IR) spectroscopy FT-IR spectroscopy was used to confirm the introduction of the PEO-b-PCL nanoparticles into the BC gels. The instrument was a Perkin-Elmer spectrum RXI FTIR spectrometer (PerkinElmer, Inc., Waltham, MA, USA). The samples of BC gel and the BC gel containing PEO(10,000)-b-PCL(5000) nanoparticles were freeze-dried to prepare them for FT-IR spectroscopy. For each measurement, the sample was pressed into a pellet with KBr powder and 64 spectra were accumulated at a resolution of 4 cm1. 2.5.2. High-resolution scanning electron microscopy The BC fiber and the BC gel containing PEO-b-PCL nanoparticles were analyzed using a Zeiss Ultra 55 field-emission gum scanning electron microscope (FEG-SEM) (Zeiss, Germany) in the secondary electron imaging mode. The samples were freeze-dried and deposited on copper stubs with silver adhesion. The samples were sputtered with 6 nm of Au/Pd.
Introduced copolymerðmg=mLÞ ¼
2.6.1. Retinol release from nanoparticles The amount of retinol released from the nanoparticles was calculated on the basis of the retinol residue in the nanoparticles. The concentration of retinol residue was determined from the amount of retinol in acetic acid–sodium acetate buffer (pH 5.2) as described above. The nanoparticle suspension (1 mL; 10 mg/L copolymer concentration, 0.500 initial retinol/copolymer weight ratio) was added to acetic acid–sodium acetate buffer (19 mL, pH 5.2), and the solution was kept at 37 C. Periodically, the retinolloaded PEO(5000)-b-PCL(5000) nanoparticles were sampled from the suspension. To calculate the amount of retinol in the nanoparticles, the solution was centrifuged at 5673 g for 5 min and dissolved in acetonitrile. The amount of retinol was determined by measuring the UV absorbance of the solution at 325 nm. The amount of retinol released from the nanoparticles (NP) was calculated from the concentration of retinol using the following equation: Table 1 Characteristics of PEO-b-PCL micelles with copolymers of different molecular weights. Nanoparticle size SD (nm)
PEO(5000)-b-PCL(5000) in water PEO(5000)-b-PCL(10,000) in water PEO(2000)-b-PCL(11,500) in water PEO(5000)-b-PCL(5000) in buffer PEO(5000)-b-PCL(10,000) in buffer PEO(2000)-b-PCL(11,500) in buffer
39.7 71.8 29.4 38.0 73.8 32.0
Data are presented as mean SD (n = 3).
1.6 1.5 1.0 0.45 2.0 1.4
Polydispersity index SD 0.245 0.095 0.141 0.019 0.215 0.11 0.125 0.0078 0.116 0.011 0.170 0.071
Retinol released from NPð%Þ ¼
Initial loaded retinolðmg=mLÞ Retinol residue in NPðmg=mLÞ Initia l loaded retinolðmg=mLÞ
where the initial loaded retinol (mg/mL) was calculated from the encapsulated retinol, and the retinol residue in NP (mg/mL) was calculated from the detected value as described above. 2.6.2. PEO-b-PCL nanoparticle release from BC gels BC gels containing PEO-b-PCL nanoparticles were placed into acetic acid–sodium acetate buffer (pH 5.2) (2 mL) at 37 C. The amount of released PEO-b-PCL was measured using an analytical method for nonionic surfactants (Boyer et al., 1977), which has been used to measure cobalt thiocyanate active substance (CTAS) in biodegradation and environmental studies. Cobalt thiocyanate solution was prepared by dissolving Co(NO3)6H2O (30 g) and NH4SCN (200 g) in 1 L of distilled water. The buffer containing the released PEO-b-PCL (200 mL) dissolved in methylene chloride (CH2Cl2, 5 mL) was added to the cobalt thiocyanate solution (5 mL) and shaken for 5 min. After a few minutes, the solution separated into an aqueous layer and an organic layer containing PEO-b-PCL. The amount of PEO-b-PCL was determined by measuring the absorbance of the organic layer using a Perkin-Elmer Lambda 10 UV/Vis spectrophotometer (PerkinElmer, Inc., Waltham, MA, USA) at 626 nm. The introduced PEO-b-PCL copolymer concentration was calculated from the concentration of nanoparticles in the suspension using the following equation:
Copolymer introduced BC gel weightðmgÞ Initial BC gel weightðmgÞ Initial BC gel volumeðmLÞ
2.6. In vitro release
Copolymer
219
where the initial BC gel volume (mL) was regarded as equal to the initial BC gel weight (g), because the BC fiber weight was negligible in comparison with the water in the gel. The copolymer introduced BC gel weight is the weight of the BC gel after immersion in the nanoparticle suspension. The released copolymer (%) was calculated from the following equation: Detected copolymerðmg=mLÞ Released copolymerð%Þ ¼ Introduced copolymerðmg=mLÞ
2.6.3. Retinol release from BC gels BC gels containing retinol-loaded PEO-b-PCL nanoparticles were prepared as described in Section 2.3, except that the gel weight was measured before the gel was immersed in the nanoparticle suspension. After the gel was removed from the nanoparticle suspension, the weight was measured again, and the gel was immersed in acetic acid–sodium acetate buffer (pH 5.2) (2 mL) and kept at 37 C. The amount of retinol released from the BC gel was determined by measuring the absorbance of the solution at 325 nm as described in Section 2.4, and calculated using the following equation: Retinol released from BCð%Þ ¼
Detected retinolðmg=mLÞ Introduced retinol in BC gelðmg=mLÞ
where the introduced retinol in BC gel (mg/mL) was calculated from the retinol concentration of the nanoparticle suspension residue after the BC gel was removed, because the retinol concentration in the BC gel was assumed to be equal to the retinol concentration of the nanoparticle suspension residue after the BC gel was removed.
220
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
3. Results and discussion 3.1. Characterization of PEO-b-PCL nanoparticles Block copolymer PEO-b-PCL nanoparticles were prepared using a co-solvent evaporation method (Aliabadi et al., 2005b). The hydrodynamic diameters and polydispersity indices of nanoparticles prepared using different molecular weights of PEO-b-PCL determined by DLS are shown in Table 1. The mean diameters of PEO(5000)-b-PCL(5000), PEO(5000)-b-PCL(10,000), and PEO (2000)-b-PCL(11,500) nanoparticles were approximately 39.7, 71.8, and 29.4 nm, and their polydispersity indices were 0.245, 0.141, and 0.215. A previous study using light scattering showed that the diameters of PEO(5000)-b-PCL(5000) and PEO(5000)-bPCL(13,000) nanoparticles prepared by a co-solvent evaporation method were 87.5 28.2 nm and 78.7 11.8 nm, respectively, and their polydispersity indices were 0.198 and 0.111, respectively (Aliabadi et al., 2005b). The sizes and polydispersity indices of the nanoparticles reported here are smaller than those reported in previous studies. Preparation conditions can affect nanoparticle size, and three important factors influencing the results of the cosolvent evaporation methods are as follows: the type of organic cosolvent, the ratio of the volume of the organic phase to the volume of the aqueous phase during the assembly process, and the order of addition of the two phases (Aliabadi et al., 2007). In addition, the stirring time of the mixture after the drop-wise addition of the components influences the size of the nanoparticles (Aliabadi et al., 2007, 2005b). Our nanoparticle preparation conditions were different from those reported in previous studies; we used smaller drops and a longer stirring time. In the co-solvent evaporation method, copolymers dissolved in an organic solution assemble and form micelles when the organic solution is mixed with the aqueous solution, and the copolymers are impossible to dissolve in the organic–aqueous mixture solution. The size of the PEO(5000)-bPCL(5000) nanoparticles reported in this study was significantly different from that of previous reports, and this small size may explain the finding that PEO(5000)-b-PCL(5000) dissolved well in acetone and was easily affected by decreased solubility when the acetone solution was mixed with the aqueous solution. The PEO(2000)-b-PCL(11,500) nanoparticles were smaller in size than the other copolymers. PEO(5000)-b-PCL(5000) and PEO (5000)-b-PCL(10,000) nanoparticles were selected for subsequent studies. The nanoparticles were applied to a DDS for human skin. The physicochemical characteristics of the nanoparticles in the buffer are important for the effectiveness of the DDS. The hydrodynamic diameter of the nanoparticles was measured by DLS in acetic acid– sodium acetate buffer with a pH of 5.2, the same pH as human skin. These results indicated that the nanoparticles maintained their size in the acetic acid–sodium acetate buffer (Table 1). In addition, the copolymer nanoparticles were analyzed using transmission electron microscopy (TEM). The TEM images of PEO(5000)-b-PCL (5000) and PEO(5000)-b-PCL(10,000) nanoparticles in aqueous solution and acetic acid–sodium acetate buffer are shown in Fig. 1. A good correlation was observed between the TEM images and DLS results. 3.2. Preparation and characterization of retinol-loaded PEO-b-PCL nanoparticles Retinol was selected as a test compound because it is a typical active hydrophobic molecule. Retinol is used as an anti-aging ingredient in cosmetic products. Retinol-loaded nanoparticles were prepared in the same way as PEO-b-PCL nanoparticles, except that both PEO-b-PCL and retinol were dissolved in acetone. The physicochemical characteristics of retinol-loaded PEO-b-PCL
Fig. 1. TEM images of PEO-b-PCL nanoparticles; (a) PEO(5000)-b-PCL(5000) in aqueous solution, (b) PEO(5000)-b-PCL(10,000) in aqueous solution, (c) PEO(5000)b-PCL(5000) in acetic acid–sodium acetate buffer solution (pH 5.2), and (d) PEO (5000)-b-PCL(10,000) in acetic acid–sodium acetate buffer solution (pH 5.2).
nanoparticles are shown in Table 2. The hydrodynamic diameter of the retinol-loaded PEO(5000)-b-PCL(5000) nanoparticles was the same as that of the unloaded nanoparticles. The retinol-loaded PEO(5000)-b-PCL(10,000) copolymers formed larger particles than those formed by the retinol-loaded PEO(5000)-b-PCL(5000), similar to the unloaded nanoparticles. In addition, retinol-loaded PEO(5000)-b-PCL(10,000) nanoparticle (initial retinol/copolymer weight ratio 0.15) size showed a similar trend to retinol-loaded PEO(5000)-b-PCL(5000) (initial retinol/copolymer weight ratio 0.15 and 0.30) size. However, the retinol-loaded PEO(5000)-b-PCL (10,000) nanoparticles (initial retinol/copolymer weight ratio 0.75) were 16% larger than the unloaded nanoparticles. The encapsulation efficiency of retinol in the PEO(5000)-b-PCL (5000) nanoparticles was 83.9% (initial retinol/copolymer weight ratio 0.15), 84.1% (initial retinol/copolymer weight ratio 0.30), and 66.9% (initial retinol/copolymer weight ratio 0.75). This result suggests that the PEO(5000)-b-PCL(5000) nanoparticles (initial retinol/copolymer weight ratio 0.15) have sufficient space to contain more retinol. The encapsulation efficiencies of some hydrophobic substrates into PEO(5000)-b-PCL(5000) nanoparticles have been reported, including curcumin (3.05%, initial curcumin/copolymer weight ratio 0.10) (Ma et al., 2008), doxorubicin (DOX; 4.03%; initial DOX/copolymer weight ratio 0.20) (Shuai et al., 2004), hydrophobic fluorescent probe DiI (80%; initial DiI/copolymer weight ratio 0.0010) (Mahmud and Lavasanifar, 2005), and cyclosporine A (CsA; 52.2%; initial CsA/copolymer weight ratio 0.10) (Aliabadi et al., 2005b). Retinol is more suitable than other hydrophobic substrates for loading into PEO(5000)-bPCL(5000) nanoparticles. In general, drug hydrophobicity plays a decisive role in the drug-loading process. Retinol is highly hydrophobic and interacts strongly with PCL. For retinol-loaded PEO(5000)-b-PCL(10,000) nanoparticles, the encapsulation efficiencies were 80.5% (initial retinol/copolymer weight ratio 0.15) and 81.5% (initial retinol/copolymer weight ratio 0.75). In previous studies, the encapsulation efficiency for paclitaxel increased significantly with increases in PCL length (Aliabadi et al., 2005b; Mahmud and Lavasanifar, 2005; Shuai et al., 2004), whereas a
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
221
Table 2 Physicochemical characteristics of retinol-loaded PEO-b-PLC nanoparticles in aqueous solution. Copolymer
Initial retinol (mg)
Initial retinol /copolymer weight ratio Encapsulation SD Encapsulation SD Nanoparticle size SD Polydispersity (mg/mg) (mg) (%) (nm) index SD
PEO(5000)-b-PCL (5000)
3 6 15 3 15
0.15 0.3 0.75 0.15 0.75
PEO(5000)-b-PCL (10,000)
2.52 5.05 10.0 2.42 12.2
0.16 0.38 0.39 0.12 0.23
83.9 84.1 66.9 80.5 81.5
5.2 6.3 2.6 4.1 1.5
38.4 38.3 39.2 70.2 83.3
0.85 0.79 0.20 2.5 1.2
0.112 0.097 0.098 0.0944 0.121
0.001 0.0034 0.022 0.023 0.014
Data are presented as mean SD (n = 3).
lesser effect was observed in this study, due to the low initial retinol/copolymer weight ratio (0.15). When a high initial retinol/ copolymer weight ratio (0.75) was used, PCL length influenced retinol encapsulation. These results suggest that PEO-b-PCL can encapsulate a large amount of retinol. Retinol-loaded nanoparticles in pH 5.2 acetic acid–sodium acetate buffer were analyzed by DLS. The retinol-loaded nanoparticles maintained their size in the acetic acid–sodium acetate buffer, similar to retinol-non-loaded nanoparticles, as indicated by measurements of hydrodynamic diameter. TEM images of retinolloaded PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticles (initial retinol/copolymer weight ratio 0.15) in acetic acid–sodium acetate buffer are shown in Fig. 2. The shape of the nanoparticles was not affected by drug loading. 3.3. Stability of retinol encapsulated in PEO-b-PCL nanoparticles Retinol is very unstable and easily oxidized (Szuts and Harosi, 1991; Tsunoda and Takabayashi, 1995). In a previous study, the stability of the unstable substrate curcumin increased when it was encapsulated into PEO-b-PCL nanoparticles (Mahmud and Lavasanifar, 2005). Therefore, the stability of retinol was also expected to increase when it was encapsulated into PEO-b-PCL nanoparticles. The stability of free retinol and retinol encapsulated in PEO(5000)-b-PCL(5000) nanoparticles was evaluated in acetic acid–sodium acetate buffer (pH 5.2) at 37 C, and the results are shown in Fig. 3. The concentration of free retinol in the buffer solution was decreased by 70% after 45 min, and free retinol in the buffer solution was almost undetectable after 1 day. However, the retinol-encapsulated nanoparticles retained more than 50% of the loaded retinol after 8 h, and approximately 10% of the loaded retinol remained after 1 week, indicating that encapsulation into PEO-b-PCL nanoparticles can increase the stability of hydrophobic active molecules. In a previous study on the stability of retinolloaded nanoliposomes (Ko and Lee, 2010), the authors placed the retinol-loaded nanoliposome suspension in a glass vial with oxygen at 37 C, and approximately 50% of the retinol remained after 2 days. In contrast, in this study, approximately 25% of the
Fig. 2. TEM images of retinol-loaded PEO-b-PCL nanoparticles in acetic acid– sodium acetate buffer solution (pH 5.2); (a) PEO(5000)-b-PCL(5000), (b) PEO (5000)-b-PCL(10,000). The initial retinol/copolymer weight ratio (mg/mg) was 0.15.
retinol remained after 2 days, and this difference was likely caused by the retinol incorporation method used. Nanoliposomes incorporate retinol into lipid bilayers, whereas PEO-b-PCL nanoparticles incorporate retinol inside of the micelle, from which it is more likely to be released. 3.4. Preparation of PEO-b-PCL nanoparticle-containing BC gel Nanoparticle-containing BC gel was prepared using a diffusion method. The BC gel was placed in the nanoparticle suspension to absorb the nanoparticles. The FT-IR spectra of the gels were measured to establish that the PEO-b-PCL copolymers were incorporated into the BC gel. Fig. 4 shows the FT-IR spectra of the freeze-dried BC gel, the freeze-dried nanoparticle-containing BC gel, and the PEO-b-PCL powder. A C¼O bond indicative of PCL was observed at 1725 cm1 in the spectra of the PEO-b-PCL powder and the nanoparticle-containing BC gel, but this bond was not observed in the BC gel spectrum. These results indicate that the PEO-b-PCL copolymers were incorporated into the BC gel. To establish the relative position between PEO-b-PCL and BC, the nanoparticle-containing BC gel was analyzed by FEG-SEM. The FEG-SEM images of BC gel prepared with or without nanoparticle are shown in Fig. 5. Very fine cellulose fibers were observed in the pure BC gel (Fig. 5(a)). However, aggregates of PEO-b-PCL were observed on the surface of the PEO(5000)-b-PCL(10,000) nanoparticle-containing BC gel (Fig. 5(b) and (c)). The nanoparticles gathered to the cellulose fibers on the surface of the gel and became stuck to them. FEG-SEM images were obtained of the cross-sections of the BC gels containing PEO(5000)-b-PCL(5000) (Fig. 5(d) and (e)) and PEO(5000)-b-PCL(10,000) (Fig. 5(f) and (g))
Fig. 3. Effect of encapsulation into PEO(5000)-b-PCL(5000) nanoparticles in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. Filled gray circle: retinolloaded nanoparticles (NP(retinol)) suspension; filled black diamond: retinol alone in solution (free retinol). The results are shown as the mean SD.
222
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
C=O
T%
BC BC+NP
PEO-b-PCL 2000
1800
1600
wavenumber
1400
1200
(cm-1)
Fig. 4. FT-IR spectra of the BC gel, PEO(5000)-b-PCL(10,000) nanoparticlecontaining BC gel, and PEO(5000)-b-PCL(10,000) powder.
nanoparticles. Some large particles were also observed inside the samples. Nanoparticles of similar size to those observed in the TEM images were not observed in the SEM cross-section images, which showed 770 nm (Fig. 5(d)) and 80 nm (Fig. 5(e)) particles in the PEO (5000)-b-PCL(5000) nanoparticle-containing BC gel, as well as 290 nm (Fig. 5(f)) and 20 nm (Fig. 5(g)) particles in the PEO(5000)b-PCL(10,000) nanoparticle-containing BC gel. In comparison with TEM, it is difficult to observe small size nanoparticles using FEGSEM. However, the results showed that the PEO-b-PCL copolymers were introduced into the BC gel as well as onto its surface. To determine the amount of nanoparticles incorporated into the BC gel, PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticle residue suspensions were subjected to measurement using the CTAS method after the BC gel was removed (Boyer et al., 1977). The incorporation ratios for PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) nanoparticles were 99.7% and 93.6%, respectively. The ratio was calculated based on the weight of the nanoparticle-containing gel and the residue nanoparticle suspension after nanoparticle diffusion. There were more PEO(5000)-bPCL(5000) nanoparticles than PEO(5000)-b-PCL(10,000) nanoparticles in the BC gel. The PEO(5000)-b-PCL(5000) nanoparticles
Fig. 5. FEG-SEM images of the BC gel and nanoparticle-containing BC gel. (a) BC cross-section; (b) and (c) PEO(5000)-b-PCL(10,000) nanoparticle-containing near the BC surface; (d) and (e) PEO(5000)-b-PCL(5000) nanoparticle-containing BC cross-section; and (f) and (g) PEO(5000)-b-PCL(10,000) nanoparticle-containing BC cross-section.
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
223
were smaller than the PEO(5000)-b-PCL(10,000) nanoparticles, allowing them to more easily pass between cellulose fibers. 3.5. In vitro release studies 3.5.1. Copolymer release from BC gel The amount of copolymer (formed and unformed nanoparticle ratio) released from the BC gel in acetic acid–sodium acetate buffer (pH 5.2) at 37 C was determined at defined time intervals using the CTAS method (Boyer et al., 1977). The results show that the combination of BC gel and nanoparticles can be used as a slowrelease system (Fig. 6). PEO(5000)-b-PCL(5000) nanoparticles were released more rapidly than PEO(5000)-b-PCL(10,000) nanoparticles at early time points. The released copolymer ratio reached a maximum of approximately 60% after 6 h and remained constant after 24 h. Although the small particles were more easily released, the total amount of released nanoparticles was not dependent on nanoparticle size. The nanoparticle release ratios did not reach 100%, suggesting that a portion of the nanoparticles remained inside the BC gel and on its surface. The FEG-SEM images suggested that some PEO-b-PCL nanoparticles stuck to the cellulose fibers, and also showed that large PEO-b-PCL particles existed in the BC gel. PEO (PEG) has high affinity for BC fibers (Numata et al., 2009); therefore, PEO-b-PCLs are difficult to remove from the gel completely. The release of copolymers after 24 h was analyzed using DLS. The results showed that nanoparticle size did not change significantly after nanoparticle release from the BC gel (data not shown). Fig. 7 shows TEM images of released copolymers after 24 h in acetic acid–sodium acetate buffer (pH 5.2). The TEM images corroborated the DLS results: the copolymers remained in the formed nanoparticles and the size did not change. These results indicate that BC gel is a good carrier for PEO-b-PCL nanoparticles. 3.5.2. Retinol release from PEO-b-PCL nanoparticles To determine the amount of retinol released from PEO(5000)-bPCL(5000) nanoparticles in acetic acid–sodium acetate buffer (pH 5.2) at 37 C, the nanoparticle-containing buffer was centrifuged to remove the retinol present outside the nanoparticles. The released retinol ratio was calculated after measuring the absorption of the
Fig. 7. TEM images of nanoparticles released from the BC gel in acetic acid–sodium acetate buffer solution (pH 5.2). (a) PEO (5000)-b-PCL (5000); (b) PEO (5000)-b-PCL (10,000).
retinol-containing nanoparticles (Fig. 8). The results show that 55% of retinol encapsulated in the nanoparticles was released after 8 h. The residue retinol ratio at the same time point was 65% (Fig. 3), indicating that 10% of the retinol in the nanoparticle suspension was located outside of the nanoparticles and precipitated in suspension. After 24 h, the residue retinol ratio in the suspension was 35.9% (Fig. 3), and ratio of retinol released from the nanoparticles was 66.4% (Fig. 8), indicating that 33.6% of the retinol remained in the nanoparticles. Consequently, the retinol in suspension was calculated to be 2.3%, indicating that the retinol present in the suspension was almost entirely encapsulated in nanoparticles. Retinol released outside the nanoparticles was almost undetectable because the retinol was unstable in the buffer. 3.5.3. Encapsulated retinol release from BC gel To determine the quantities of nanoparticle-encapsulated retinol and free retinol released from the BC gel, we performed a diffusion experiment with PEO(5000)-b-PCL(5000) and PEO (5000)-b-PCL(10,000) retinol-loaded nanoparticles. The BC gel was placed in the retinol-loaded nanoparticle suspension for 1 day at 4 C to prevent retinol deterioration. The PEO(5000)-b-PCL(5000) and PEO(5000)-b-PCL(10,000) retinol-loaded nanoparticle suspensions retained 78.6% and 71.9% of the initial retinol amount, respectively, after 1 day at 4 C. In the assay of the effect of encapsulation into nanoparticles in buffer (pH 5.2) at 37 C, only
100
80
PEO(5000)-b-PCL(5000)
0.8 80
Released retinol [%]
Released copolymer [%]
1001
60 PEO(5000)-b-PCL(10,000)
40
20
0.6 60
0.4 40
0.2 20
0
0
5
10
15
20
25
Time [h] Fig. 6. Released copolymer ratio from the BC gel in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C; open gray circle: PEO(5000)-b-PCL(5000); open black circle: PEO(5000)-b-PCL(10,000). The results are shown as the mean SD.
00 0
20
40
60
80
100
[h] Time [hr] Fig. 8. Ratio of retinol released from PEO(5000)-b-PCL(5000) nanoparticles in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C. The results are shown as the mean SD.
224
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225
4. Conclusion
10 PEO(5000)-b-PCL(10,000)
Released retinol [%]
8
6 PEO(5000)-b-PCL(5000)
4
2
0 0
5
10
15
20
25
Time [h] Fig. 9. Retinol with nanoparticles released from BC gel in acetic acid–sodium acetate buffer solution (pH 5.2) at 37 C; gray square: PEO(5000)-b-PCL(5000); black triangle: PEO(5000)-b-PCL(10,000). The results are shown as the mean SD.
35.9% (PEO(5000)-b-PCL(5000), Fig. 3) of the loaded retinol remained after 1 day. These results show that retinol deterioration was prevented by encapsulation. Next, the nanoparticle-encapsulated retinol containing gel was placed in acetic acid–sodium acetate buffer (pH 5.2) at 37 C and the absorption was measured at defined time intervals (Fig. 9). We assumed that the BC gel contained the same concentration of retinol-loaded nanoparticles as the nanoparticle suspension after the BC gel was removed. The ratio of released retinol from the nanoparticles was calculated on the basis of the initial retinol concentration of BC gel. The retinol ratio reached its maximum of 7.7% ((PEO(5000)-b-PCL(5000)) at 4 h and 8.7% (PEO(5000)-b-PCL(10,000)) at 4 h, and started to decrease after another hour. The ratio of released copolymer reached a maximum at 6 h, similar to the released copolymer ratio from the BC gel (Fig. 6), suggesting that the detected retinol was contained in the nanoparticles. However, the detected amount of released retinol was smaller than the amount of retinol released from the nanoparticles (Fig. 8). Most of the retinol released from the nanoparticles precipitated and remained in the BC gel. Once retinol was released from the nanoparticles, it was precipitated in the aqueous solution. Therefore, the precipitated retinol was trapped by the cellulose fibers. The detected amount of released retinol was consequently small in comparison with the amount of retinol released from the nanoparticles. However, the rate of retinol release from the BC gel decreases after 24 h, indicating that the released of retinol into the buffer solution decreases slowly, which is consistent with the effect of encapsulation into nanoparticles (Fig. 3). The system described herein requires some improvement. First, the quantities of introduced retinol should be considered. In previous studies, the encapsulation ratios were smaller than that used in this study, whereas the same drug/copolymer weight ratio was used (Aliabadi et al., 2005b; Ma et al., 2008). A large loaded drug releases easily because the drug is likely to be distributed from the nanoparticle core. The quantity of retinol release and the release rate of loaded-retinol are difficult to control. Second, if an antioxidant substance is added to the nanoparticle suspension or loaded into the nanoparticle with retinol, the system can release larger quantities of retinol.
We have developed a DDS based on diffusion of retinol-loaded PEO-b-PCL nanoparticles from BC gel. The nanoparticles increased the stability of the encapsulated retinol. The retinol-loaded nanoparticles released retinol slowly. However, most of the retinol released from the nanoparticles was precipitated and remained in the BC gel, and this problem requires further study to overcome. The system described herein can contribute considerably to cosmetic and medical applications, such as skin treatment and skin tissue repair. PEO-b-PCL nanoparticles can also encapsulate other hydrophobic molecules (Aliabadi et al., 2005b; Ghoroghchian et al., 2006; Ma et al., 2008; Mahmud and Lavasanifar, 2005; Shuai et al., 2004). Stable, active ingredients can be delivered with this system, which has the potential to increase the quantities of drugs released from BC gel. PEO-b-PCL can also form composite nanoparticles encapsulating hydrophobic organic and inorganic compounds, and these nanoparticles can serve as multimodal carriers for drugs and imaging agents (Gindy et al., 2008). Additionally, PEO-b-PCL forms micelles, and also forms vesicles with a hydrophilic membrane surface, which are capable of encapsulating hydrophilic molecules (Ding and Liu, 1997; Kukula et al., 2002; Nardin et al., 2000). Therefore, the system described herein has potential uses in various skin preparations. Acknowledgments The authors thank Amandine Durand-Terrasson for her help with the TEM and FEG-SEM experiments. The authors acknowledge support from CNRS, PolyNat Carnot Institute, and LabEx ARCANE (ANR-11-LABX-0003-01), France. Yukari Numata was funded by the Institute of National Colleges of Technology, Japan. References Aliabadi, H.M., Brocks, D.R., Lavasanifar, A., 2005a. Polymeric micelles for the solubilization and derivery of cyclosporine A: pharmacokinetics and biodistribution. Biomaterials 26, 7251–7259. Aliabadi, H.M., Mahmud, A., Sharifabadi, A.D., Lavasanifar, A., 2005b. Micelles of methoxy poly(ethylene oxide)-b-poly(e-caprolactone) as vehicles for the solubilization and controlled delivery of cyclosporine A. J. Control. Release 104, 301–311. Aliabadi, H.M., Elhasi, S., Mahmud, A., Gulamhusein, R., Mahdipoor, P., Lavasanifar, A., 2007. Encapsulation of hydrophobic drugs in polymeric micelles through cosolvent evaporation: the effect of solvent composition on micellar properties and drug loading. Int. J. Pharm. 329, 158–165. Allen, C., Maysinger, D., Eisenberg, A., 1999. Nano-engineering block copolymer aggregates for drug delivery. Colloids Surf. B 16, 3–27. Boyer, S.L., Guin, K.F., Kelley, R.M., Mausner, M.L., Robinson, H.F., Schmitt, T.M., Stahl, C.R., Setzkorn, E.A., 1977. Analytical method for nonionic surfactants in laboratory biodegradation and environmental studies. Environ. Sci. Technol. 11, 1167–1171. Brown, J.M.B., 1996. The biosynthsis of cellulose. J. Macromol. Sci. Part A Pure Appl. Chem. 33, 1345–1373. Ding, J., Liu, G., 1997. Polyisoprene-block-poly(2-cinnamoylethyl methacrylate) vesicles and their aggregates. Macromolecules 30, 655–657. Fu, L., Zhang, J., Yang, G., 2013. Present status and applications of bacterial cellulosebased materials for skin tissue repair. Carbohydr. Polym. 92, 1432–1442. Ghoroghchian, P.P., Li, G., Levine, D.H., Davis, K.P., Bates, F.S., Hammer, D.A., Therien, M.J., 2006. Bioresorbable vesicles formed through spontaneous self-assembly of amphiphilic poly(ethylene oxide)-block-polycaprolactone. Macromolecules 39, 1673–1675. Gindy, M.E., Panagiotopoulos, A.Z., Prud’homme, R.K., 2008. Composite block copolymer stabilized nanoparticles: simultaneous encapsulation of organic active and inorganic nanoparticles. Langmuir 24, 83–90. Hestrin, S., Schramm, M., 1954. Syntheie of cellulose by Acetobacter xylinum. Biochem. J. 58, 345–352. Iguchi, M., Yamanaka, S., Budhiono, A., 2000. Review bacterial cellulose – a masterpiece of nature’s arts. J. Mater. Sci. 35, 261–270. Kim, S.Y., Shin, I.G., Lee, Y.M., Cho, C.S., Sung, Y.K., 1998. Methoxy poly(ethylene glycol) and e-caprolactone amphiphilic block copolymeric micelle containing indomethacin. II. Micelle formation and drug release behaviours. J. Control. Release 51, 13–22.
Y. Numata et al. / International Journal of Pharmaceutics 486 (2015) 217–225 Klemm, D., Heublein, B., Fink, H.P., Bohn, A., 2005. Cellulose: fascinating biopolymer and sustainble raw material. Angew. Chem. Int. Ed. 44, 2–37. Ko, S., Lee, S.C., 2010. Effect of nanoliposomes on the stabilization of incorporated retinol. Afr. J. Biotechnol. 9, 6158–6161. Kukula, H., Schlaad, H., Antonietti, M., Forster, S., 2002. The formation of polymer vesicles or peptosomes by polybutadiene-block-poly(L-glutamate)s in dilute aqueous solution. J. Am. Chem. Soc. 124, 1658–1663. Ma, Z., Haddadi, A., Molavi, O., Lavasanifar, A., Lai, R., Samuel, J., 2008. Micelles of poly(ethylene oxide)-b-poly(e-caprolactone) as vehicles for the solubilization, stabilization, and controlled delivery of curcumin. J. Biomed. Mater. Res. Part A 86A, 300–310. Mahmud, A., Lavasanifar, A., 2005. The effect of block copolymer structure on the internalization of polymeric micelles by human breast cancer cells. Colloids Surf. B 45, 82–89. Miyamoto, T., Takahashi, S., Ito, H., Inagaki, H., Noishiki, Y., 1989. Tissue biocompatibility of cellulose and its derivatives. J. Biomed. Mater. Res. 23, 125– 133. Nardin, C., Hirt, T., Leukel, J., Meier, W., 2000. Polymerized ABA triblock copolymer vesicles. Langmuir 16, 1035–1041.
225
Numata, Y., Muromoto, K., Furukawa, H., Gong, J.P., Tajima, K., Munekata, M., 2009. Nonvolatile and shape-memoried bacterial cellulose gels swollen by poly (ethylene glycol). Polym. J. 41, 524–525. Okiyama, A., Motoki, M., Yamanaka, S., 1993. Bacterial cellulose III. Development of a new form of cellulose. Food Hydrocolloids 6, 493–501. Ross, P., Mayer, R., Benziman, M., 1991. Cellulose biosynthesis and function in bacteria. Microbiol. Rev. 55, 35–58. Shah, N., UI-Islam, M., Khattak, W.A., Park, J.K., 2013. Overview of bacterial cellulose composites: a multipurpose advanced. Carbohydr. Polym. 98, 1585–1598. Shuai, X., Ai, H., Nasongkla, N., Kim, S., Gao, J., 2004. Micellar carriers based on block copolymers of poly(e-caprolactone) and poly(ethylene glycol) for doxorubicin delivery. J. Control. Release 98, 415–426. Szuts, E.Z., Harosi, F.I., 1991. Solubility of retinoids in water. Arch. Biochem. Biophys. 287, 297–304. Tsunoda, T., Takabayashi, K., 1995. Stability of all-trans-retinol in cream. J. Cosmet. Chem. 46, 191–198. Wei, X.W., Gong, C.Y., Gou, M.L., Fu, S.Z., Guo, Q.F., Shi, S., Luo, F., Guo, G., Qiu, L.Y., Qian, Z.Y., 2009. Biodegradable poly(e-caprolactone)-poly(ethylene glycol) copolymers as drug delivery system. Int. J. Pharm. 381, 1–18.
