Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Protocols
Iontophoretic transdermal delivery using chitosan-coated PLGA nanoparticles for positively charged drugs Issei Takeuchi a,b,c , Tomoyoshi Takeshita a , Takaaki Suzuki a , Kimiko Makino a,b,c,∗ a b c
Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan Center for Drug Delivery Research, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan Center for Physical Pharmaceutics, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan
a r t i c l e
i n f o
Article history: Received 12 July 2017 Received in revised form 7 September 2017 Accepted 3 October 2017 Keywords: Donepezil hydrochloride Transdermal administration Iontophoresis Coumarin Rhodamine poly(dl-lactide-co-glycolide) PLGA Positively charged nanoparticles Hydrophilic drug
a b s t r a c t Recently, poly(dl-lactide-co-glycolide) (PLGA) nanoparticles prepared using a combination of an antisolvent diffusion method with preferential solvation was shown to be beneficial for the iontophoretic transdermal delivery of therapeutic agents. Also, this preparation method can contain a hydrophilic drug. However, since PLGA nanoparticles were negatively charged, it was difficult to apply iontophoresis for positively charged hydrophilic drugs. In this study, we prepared positively charged PLGA nanoparticles containing donepezil hydrochloride (DP). DP was used as a positively charged hydrophilic drug model. The PLGA nanoparticles were coated with chitosan hydroxypropyltrimonium chloride. The average particle diameter of the nanoparticles was 117.7 ± 60.6 nm and the surface charge number density changed from negative to positive. Ex vivo skin accumulation study was carried out using abdominal rat skin and a Franz-type diffusion cell with/without iontophoresis. When iontophoresis was applied, the DP concentration in the rat skin of chitosan-coated PLGA nanoparticles was 2.2 times higher than that of non-coated PLGA nanoparticles. This indicated that chitosan-coated PLGA nanoparticles were suitable for iontophoresis. To investigate the transdermal delivery route of the nanoparticles, we prepared chitosancoated PLGA nanoparticles containing DP, coumarin-6, and rhodamine 6G. Coumarin-6 and rhodamine 6G were used as a trace marker of the PLGA nanoparticles and positively charged hydrophilic drug model, respectively. From the results of ex vivo accumulation test of this fluorescent nanoparticles, it was suggested that positively charged hydrophilic drugs reached the hair follicles as a nanoparticle, and then they were released from the nanoparticles. © 2017 Published by Elsevier B.V.
1. Introduction The skin has been studied as a site of administration for systemic and local delivery of therapeutic agents. Transdermal delivery can avoid the effect of first-pass hepatic metabolism, and deliver therapeutic agents for a long period of time. The skin is composed of the epidermis (thickness of 50–100 m), dermis (thickness of 3–5 mm) and subcutaneous tissue (thickness of 1–2 mm) [1]. The stratum corneum, the outermost layer of the skin (epidermis), is composed of keratin-filled dead cells (corneocytes) embedded in a complex intercellular lipid mixture, particularly rich in ceramides, cholesterol, fatty acids, organized in bilayer arrays. The stratum corneum functions as a barrier against intrusion such as microorganisms
∗ Corresponding author at: Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan. E-mail address: [email protected] (K. Makino). https://doi.org/10.1016/j.colsurfb.2017.10.011 0927-7765/© 2017 Published by Elsevier B.V.
(bacteria and viruses), radiation and transpiration of water [2,3]. Simultaneously, the permeability of drugs into the skin is also poor. In order to therapeutic agents to penetrate the skin, drugs must be able to pass through this barrier. Various penetration enhancing approaches of drugs through the skin have been studied. Occlusive dressing technique (ODT) is the facilitation of percutaneous absorption by hydrating stratum corneum [4]. Penetration enhancers, chemicals that interact with skin constituents, is one long–standing approach to promoting drug flux [5,6]. Physical modifications such as iontophoresis which promotes the penetration of hydrophilic and charged molecules across skin by electric energy [7–10], electroporation which produces small pore on the surface of stratum corneum by adding pulse voltage [11], microneedles [12], and needle-free injectors [13], and combinations of physical and chemical enhancement technique have also been studied [14]. Moreover, nano-sized systems such as liposomes, micelles, nanoparticles, and dendrimers have attracted attention as drug carriers for transdermal drug deliv-
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
ery system [15]. Recently, we have reported a combination of an antisolvent diffusion method [16] with preferential solvation phenomenon [17,18] for nanoparticles including a hydrophobic drug. The nanoparticles having high surface charge number density were prepared using poly(dl-lactide-co-glycolide) (PLGA), widely used biocompatible and biodegradable copolymer [19–21], and they were beneficial for the iontophoretic transdermal delivery of therapeutic agents [22]. It was suggested that the combined use of this nanoparticle preparation method and iontophoresis was actually useful for the treatment of osteoporosis [23]. In drug delivery using iontophoresis, electrorepulsion and electroosmosis play an important role [24]. In the drug-loaded nanoparticles, to fully utilize these effects, it is necessary to adjust the positive or negative of the surface charge of the carrier nanoparticles with the drug. In the previous study, we reported that positively charged hydrophilic drug containing nanoparticles prepared using a combination of an antisolvent diffusion method with preferential solvation were useful for transdermal administration [25]. However, since the PLGA particle surface was negatively charged, iontophoresis could not be applied. Salt formation is the most common and effective method of increasing the solubility and dissolution rate of acidic and basic drugs [26]. Examples of positively charged drugs include drugs in the form of hydrochloride, sulfate or sodium salts. The main aim of the present study was to apply a transdermal delivery system using PLGA nanoparticles and iontophoresis to positively charged drugs for efficient transdermal delivery. We prepared Donepezil hydrochloride (DP)-loaded PLGA nanoparticles and to obtain positively charged PLGA nanoparticles, the surface of the nanoparticles was coated with chitosan (CS). DP was used as a positively charged hydrophilic drug model. We determined physicochemical characteristics and skin accumulation of DP of the chitosan-coated PLGA nanoparticles. Moreover, the transdermal delivery routes of negatively charged PLGA nanoparticles and positively charged CS-coated PLGA nanoparticles were observed. Coumarin-6 and rhodamine 6G were used as a trace marker of the PLGA nanoparticles and positively charged hydrophilic drug model, respectively [27]. 2. Materials and methods 2.1. Materials Poly(dl-lactide-co-glycolide) (PLGA, Mw: 10,000, monomer composition of lactic acid/glycolic acid = 75/25), sucrose, l(+)-arginine (purity ≥ 98%), N-methyl-2-pyrrolidone (NMP, C5 H9 NO, purity > 98%), potassium dihydrogen phosphate (KH2 PO4 , purity > 99.5%), ethyl p-hydroxybenzoate (C9 H10 O3 ), and rhodamine 6G (C28 H31 ClN2 O3 ) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Donepezil hydrochloride (DP, C24 H29 NO3 ·HCl, purity > 98%) was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Coumarin-6 (C20 H18 N2 O2 S, purity > 98%) was purchased from Sigma–Aldrich (St. Louis, MO, USA). Acetonitrile (CH3 CN, JP, USP/NF, EP) was purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Isoflurane for the animal was purchased from Mylan Inc. (Pittsburgh, PA, USA). Chitosan hydroxypropyltrimonium chloride (CS, Mw: 100,000–200,000) was kindly donated by Katakura & Co-op Agri Corp. (Tokyo, Japan). Other chemicals were of the highest reagent grade commercially available. 2.2. Preparation of DP-loaded PLGA nanoparticles coated with CS DP-loaded PLGA nanoparticles (bare nanoparticles) were prepared using a combination of an antisolvent diffusion method with preferential solvation [22,23]. Briefly, 96 mg of PLGA and 4 mg of DP
521
were dissolved in 2 mL of NMP. The solution was injected into 20 mL of 0.3% (w/v) l-arginine aqueous solution. Bare nanoparticles were immediately precipitated by Marangoni effect [28]. The suspension was dialyzed for 24 h in a dialysis tube (UC36-32-100, molecular weight cut off: 14,000, EIDIA Co., Ltd., Tokyo, Japan) to remove excess NMP and unloaded-DP. After dialysis, the suspension was mixed with equivalent amounts of 0.02% (w/v) chitosan solution, and DP-loaded CS-coated PLGA nanoparticles (CS-coated nanoparticles) were obtained [29,30]. Then, 500 mg of sucrose was added to the suspension as a cryoprotectant. After freezing at −30◦ C, this suspension was lyophilized by usage of a freeze dryer (FDU-1200, Tokyo Rikakikai Co., Ltd., Tokyo, Japan). Surface properties of the bare and CS-coated nanoparticles were observed using a scanning electron microscope (SEM, JSM-6060LA, JEOL Ltd., Akishima, Japan). The mean volume diameters and polydispersity index values of these nanoparticles were measured using a particle size analyzer (ELSZ-2, Otsuka Electronics Co., Ltd., Osaka, Japan). Samples were dispersed in purified water and measured at 25 ◦ C. Also, the electrophoretic mobility of the nanoparticles was measured in physiological saline with nine different ionic strengths, at skin surface temperature (32◦ C) [31]. DP content in the nanoparticles was measured using high-performance liquid chromatography (HPLC, SIL-20A prominence, SPD-20A prominence, LC-20AD prominence, CTO-10ASvp, DGU-20A3 prominence, Shimadzu) at 271 nm with a polymer-coated ODS column (CAPCELL PAK MF Ph-1, size: 4.6 mm × 150 mm, Shiseido Co. Ltd., Tokyo, Japan). The mobile phase consisted of a solution of phosphate buffer solution (pH 2.6) and acetonitrile with a volume ratio of 85:15. Ten milligrams of samples were dissolved in 5 mL of an acetonitrile-phosphate buffer solution with a volume ratio of 1:1. HPLC measurements were carried out at 40 ◦ C (flow rate: 1.0 mL/min), and 30 L of sample solution were applied. All HPLC measurements were carried out under the same conditions. To confirm the state of DP in the CS-coated nanoparticles during skin accumulation study, the release rate of DP from the nanoparticles was investigated by using dialysis membrane method [32]. The CS-coated nanoparticles were redispersed in 50 mM of NaCl solution to a concentration of 0.1% (w/v) DP. The suspension was placed in a dialysis tube (UC20-32-100, molecular weight cut off: 14,000, EIDIA Co., Ltd.) and added in 95 mL of 50 mM NaCl solution. The sample suspensions were shaken at 30 rpm at 32◦ C. After 0.5, 1, 2, 6, and 12 h, the samples were taken and the amount of released DP was quantified using HPLC. We compared this result to the release rate of 0.1% (w/v) DP solution, prepared using 50 mM of NaCl solution, from the dialysis tube (UC20-32-100). 2.3. Ex vivo skin accumulation study of CS-coated nanoparticles Male Sprague-Dawley rats aging 6 weeks were purchased from Japan SLC Inc. (Tokyo Japan). All animal care was conducted under the Guidelines for Animal Experimentation of Tokyo University of Science, which is based on the Guidelines for Animal Experimentation of the Japanese Association for Laboratory Animal Science. Rats were anesthetized by intraperitoneal administration of a combination anesthetic, which was prepared with 0.3 mg/kg of medetomidine, 4.0 mg/kg of midazolam, and 5.0 mg/kg of butorphanol, under isoflurane anesthesia [33]. After shaved their abdominal hair, animals were killed by exsanguination. Then, abdominal skin was excised and mounted on a Franz-type diffusion cell [23]. Bare and CS-coated nanoparticles were dispersed in 50 mM NaCl solution, and 0.1% (w/v) DP contained suspension was prepared. Then, 3 mL of the suspension and physiological saline (30 mL, pH 7.4, I = 0.154 M) were used as donor medium and receptor medium, respectively. Also, 0.1% (w/v) DP solution was used for donor medium for comparison. The receptor medium was kept at 32◦ C and stirred with a magnetic stirrer. Two hours after the start
522
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
of the permeation test, 1 mL of the receptor medium was collected and 1 mL of fresh physiological saline was added to receptor cell. When iontophoresis was applied, Ag/AgCl electrodes were used. Iontophoresis was performed at a constant current of 0.25 mA/cm2 for 2 h. After the permeability study using a Franz-type diffusion cell, the skin was carefully washed to remove nanoparticles or free DP accumulated on the surface of the skin. To measure DP amount in rat skin, the sample skins were lyophilized and broken into fractions. Then, 0.1 mL of acetonitrile solution containing 0.3% (w/v) of ethyl p-hydroxybenzoate as an internal standard substance, and 4.9 mL of a solution of acetonitrile and phosphate buffer solution (pH 6.0) with a volume ratio of 2:1 were added to collected samples. After shaking for 60 min, the sample solutions were centrifuged (Model 2410, Kubota Corp., Tokyo, Japan) at 4000 rpm for 10 min to remove protein. The supernatants were collected and passed through 0.45 m filter and the amounts of DP in the samples were measured using HPLC.
2.4. Preparation of fluorescent PLGA nanoparticles To investigate the transdermal delivery route of CS-coated nanoparticles, we prepared PLGA nanoparticles containing DP, coumarin-6, and rhodamine 6G (CS-coated fluorescent nanoparticles). Bare fluorescent nanoparticles were also prepared for comparison. Ninety-six milligrams of PLGA, 3 mg of DP, 500 g of coumarin-6, and 500 g of rhodamine 6G were dissolved in 2 mL of NMP. Nanoparticle preparation was carried out in the same manner described in Section 2.2. Contents of coumarin-6 and rhodamine 6G in the nanoparticles were measured using HPLC. To measure coumarin-6 and rhodamine 6G, a solution of acetonitrile and purified water with a volume ratio of 45:55 and a solution of acetonitrile and purified water with a volume ratio of 3:7 were used as mobile phases, respectively. To investigate the retention of coumarin-6 in PLGA nanoparticles and the release behavior of rhodamine 6G from PLGA nanoparticles, we carried out release test of those substances from fluorescent nanoparticles. The nanoparticles were redispersed in 50 mM of NaCl solution to a concentration of 0.1% (w/v) DP. The suspension was placed in a dialysis tube (UC2032-100) and added in 95 mL of 50 mM NaCl solution. The sample suspensions were shaken at 30 rpm at 32◦ C. After 0.5, 1, 2, 4, 6, 8, 10, 12, 24, 72, and 120 h, the samples were taken and the amount of released DP was quantified using HPLC.
Fig. 1. Particle size distribution of donepezil hydrochloride-loaded PLGA nanoparticles (bare nanoparticles) and donepezil hydrochloride-loaded chitosan hydroxypropyltrimonium chloride-coated PLGA nanoparticles (CS-coated nanoparticles).
3. Results and discussion 3.1. Characterization of DP-loaded bare and CS-coated nanoparticles As shown in Table 1, bare nanoparticles and CS-coated nanoparticles with mean volume diameters of 80.2 ± 30.1 and 117.7 ± 60.6 nm, respectively, were prepared. Their particle size distributions were shown in Fig. 1. DP contents of bare and CS-coated nanoparticles were 3.14 ± 0.05 and 2.87 ± 0.06%, respectively. The entrapment efficiency of DP in bare nanoparticles was 78.5% and yield of CS-coated nanoparticles was 79.9%. Fig. 2 displays SEM images of the nanoparticles, showing spherical dispersed particles. Their electrophoretic mobility were also measured and the results were shown in Fig. 3. The data were analyzed by using Ohshima’s electrokinetic theory for soft particles [34]. At a colloidal particle covered with a layer of polyelectrolyte chains moving in a liquid containing a symmetrical electrolyte of valency v and bulk concentration (number density) n (m−3 ), ionized groups of valency z are uniformly distributed at a number density of N (m−3 ). Therefore, it is expressed as r 0 0 /m + DON / d zeN f( ) + 2 a 1/m + 1/
= with 2.5. Evaluation of transdermal delivery route of fluorescent nanoparticles Ex vivo skin permeation experiments of fluorescent nanoparticles were carried out in the same manner described in Section 2.3. The cross sections of the skin after 2 h permeability studies were used as samples. The samples were fixed using 4% (w/v) paraformaldehyde phosphate buffer solution which con® tained 10% of sucrose and embedded in Jung tissue freezing medium (Leica Instruments, Nussloch, Germany) in liquid nitrogen. Then, they were cut into 20 m sections using a cryostat (CM3050S, Leica Instruments) at −20◦ C. The samples were mounted on a sliding glass and were observed under a fluorescence microscope (coumarin-6 Ex/Em = 488/519 nm, rhodamine 6G Ex/Em = 518/543 nm, Fluoview FV1000, Olympus Corp., Tokyo, Japan).
f
d a
2 = 3
DON =
0 =
(1)
1+
1
3
(2)
2 1 + d/a
kT zN 2 zN ln[ + {( ) + 1} ve 2vn 2vn
zN kT zN 2 (ln[ + {( ) + 1} ve 2vn 2vn
1/2
(3)
],
1/2
]+
2vn zN 2 [1 − {( ) + 1} zN 2vn
1/2
]), (4)
1/2
= ␥/
,
m = [1 + (
zN 2 ) ] 2vn
(5) 1/4
,
(6)
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
523
Table 1 Physicochemical properties of donepezil hydrochloride (DP)-loaded bare nanoparticles and chitosan (CS)-coated nanoparticles (mean ± S.D., n = 3).
Particle diameter (nm) Polydispersity index DP content in nanoparticles% (w/w) Electrophoretic mobility at I = 50 mM Particle surface charge number density (mM) Particle softness parameter (nm)
Bare nanoparticles
CS-coated nanoparticles
80.2 ± 30.1 0.15 3.14 ± 0.04 −68.1 ± 4.50 −72.4 2.21
117.7 ± 60.6 0.18 2.87 ± 0.06 23.1 ± 0.31 14.3 2.85
Fig. 2. Scanning electron microscopy images of donepezil hydrochloride-loaded PLGA nanoparticles taken at an accelerating voltage of 10–15 kV (magnification: 20,000×). (a) Bare nanoparticles. (b) Chitosan hydroxypropyltrimonium chloridecoated nanoparticles.
=(
2ne2 v2 ) r 0 kT
1/2
,
(7)
where a is the particle diameter, d is the thickness of the ionpenetrable surface layer, is the viscosity, is the frictional coefficient of the surface layer, εr is the relative permittivity of the solution, ε0 is the permittivity of a vacuum, DON is the Donnan potential of the surface layer, 0 is the potential at the boundary between the surface layer (it means the surface potential of the polyelectrolyte-coated particle) and the surrounding solution and k is the Boltzmann constant. The parameter is the DebyeH`u` ckel parameter, and m shows the Debye-H`u` ckel parameter in the polyelectrolyte layer. We have determined the density of fixed-charges zN and the reciprocal of (: the degree of friction exerted on the liquid flow in the polyelectrolyte layer of the parti-
Fig. 3. (a) Electrophoretic mobility of donepezil hydrochloride-loaded PLGA nanoparticles. Solid line are theoretical results calculated with zN = −7.24 × 10−2 M and 1/ = 2.21 nm (mean ± S.D., n = 3). (b) Electrophoretic mobility of donepezil hydrochloride-loaded chitosan hydroxypropyltrimonium chloride-coated PLGA nanoparticles. Solid line are theoretical results calculated with zN = 1.43 × 10−2 M and 1/ = 2.85 nm (mean ± S.D., n = 3).
cles). 1/ has the dimension of length and shows the permeability of the polyelectrolyte layer. Therefore, it can be considered as the parameter of “softness” of particle surface [34,35]. In the previous report, the curve-fitting procedure was used to determine zN and 1/ [10,25,36,37]. As shown in Fig. 3, we have applied it to our measurement result of electrophoretic mobility of bare and CS-coated nanoparticles. From the result of the procedures (shown as a solid
524
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
Fig. 5. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study with iontophoresis using bare fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
nanoparticles were more rigid than CS-coated nanoparticles, since bare nanoparticles did not have the CS layer at the surface of the nanoparticles. 3.2. Ex vivo skin accumulation study of CS-coated nanoparticles
Fig. 4. (a) Cumulative release rate of donepezil hydrochloride (DP) from DP-loaded PLGA nanoparticles and DP-loaded chitosan-coated PLGA nanoparticles (CS-coated nanoparticles, mean ± S.D., n = 3, * p < 0.05, ** p < 0.01, t-test). (b) DP amount in the rat skin when applied with passive diffusion or iontophoresis (mean ± S.D., n = 4, *p < 0.05, **p < 0.01, Tukey’s test).
line in Fig. 3), zN = −7.24 × 10−2 M and 1/ = 2.21 nm were obtained in the bare nanoparticles. Also, zN = 1.43 × 10−2 M and 1/ = 2.85 nm were obtained in the CS-coated nanoparticles. By coating CS on the surface of bare nanoparticles, the value of the surface charge number density of nanoparticles changed from negative to positive. PLGA molecule has negative charges at neutral pH because of their ionized terminal carboxyl groups [38]. This result indicated that the particle surface was fully covered with CS and the negative charge derived from PLGA was shielded. Also, the value of softness parameter of CS-coated nanoparticles was about 1.3 times higher than that of bare nanoparticles. It was assumed that bare
The cumulative release rate of DP from CS-coated nanoparticles in 50 mM of NaCl solution at 32◦ C is shown in Fig. 4a. Compared to 0.1% (w/v) DP solution, CS-coated nanoparticles significantly decreased release of DP in 1–12 h. The release rates of CS-coated nanoparticles and DP solution after 2 h from the start of the test were 46.1 ± 3.06 and 81.6 ± 2.18%, respectively. Therefore, it was considered that half of the DP was retained in the nanoparticles 2 h after the start of the skin accumulation study. Also, since this result was similar to that of bare nanoparticles, CS-coated nanoparticles and bare nanoparticles showed no difference in drug release behavior. Fig. 4b shows results of ex vivo skin accumulation study of bare and CS-coated nanoparticles. Without iontophoresis, the value of DP accumulation in the skin of CS-coated nanoparticles was 5.3 times higher than that of DP solution. This result assumed that hydrophobicity of DP was enhanced by particulation using PLGA, and the hydrophilic drug was partitioned into the stratum corneum. When iontophoresis was applied to CS-coated nanoparticles, the DP concentration in the rat skin was 1.9 times higher than that of CS-coated nanoparticles without iontophoresis. In addition,
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
Fig. 6. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study without iontophoresis using CS-coated fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
when iontophoresis was applied, the value of DP accumulation in the skin increased to 2.2 times by CS modification. These significant difference indicated that CS-coated nanoparticles were suitable for transdermal delivery of positively charged drug using iontophoresis. 3.3. Transdermal delivery route of fluorescent nanoparticles in rat skin Contents of coumarin-6 and rhodamine 6G in the nanoparticles measured using HPLC were 0.4 and 0.5%, respectively. The cumulative release rate of coumarin-6 from the nanoparticles was 0.67% at 8 h and 2.63% at 120 h. In the previous study, the release rate of coumarin-6 from 150 nm PLGA nanoparticles was less than 1% of the content in the particles at 2–8 h [27]. Thus, we considered that coumarin-6 was successfully retained in nanoparticles. In contrast, rhodamine 6G was rapidly released from the nanoparticles and reached 13.6% at 2 h, 60.71% at 12 h, and 89.04% at 120 h. From these results, we confirmed that coumarin-6 had the capability of trace marker of CS-coated PLGA nanoparticles, and rhodamine 6 G could be used as a hydrophilic drug model. In Figs. 5–7, the cross sections of the rat skin after 2 h accumulation study were shown. As shown in Fig. 5, bare fluorescent nanoparticles were observed in deep follicles, however, the fluorescence of rhodamine 6G was weak, and it was not observed in the deep follicles. The released rates of coumarin-6 and rhodamine 6 G from the nanoparticles
525
Fig. 7. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study with iontophoresis using CS-coated fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
after 2 h were 0.24% and 13.6%, respectively. From these results, it was suggested that diffusion of released rhodamine 6G was prevented by iontophoresis. Fig. 7 shows the results of CS-coated fluorescent nanoparticles without iontophoresis, coumarin-6 and rhodamine 6G were detected in stratum corneum and hair follicle after 2 h accumulation study. Since the intercellular spaces in the stratum corneum are 70 nm, we considered that a part of CS-coated fluorescent nanoparticles penetrated the intercellular spaces of the stratum corneum [39]. As shown in Fig. 7, when iontophoresis was applied to nanoparticles, coumarin-6 and rhodamine 6G were observed in deep follicles compared to passive diffusion. The fluorescence of rhodamine 6G was observed in the deeper part of the hair follicles compared to coumarin-6. These results indicated that rhodamine 6 G, which contained in CS-coated nanoparticles, reached the hair follicles, and then it was released from the nanoparticles. 4. Conclusions DP-loaded PLGA nanoparticles coated with CS having positive surface charge number density were prepared by using a combination of an antisolvent diffusion method with preferential solvation. This nanoparticle showed high skin accumulation of DP in passive diffusion due to improved hydrophobicity. In addition, when iontophoresis was applied to CS-coated nanoparticles, the DP accu-
526
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
mulation was 2.2 times higher than that of bare nanoparticles. In the study of the transdermal delivery route using the fluorescent nanoparticles, we confirmed that transdermal delivery of a positively charged substance from PLGA nanoparticles was improved by coating the nanoparticle surface with CS. From these results, it was suggested that CS-coated nanoparticles can efficiently deliver positively charged agents to the deep hair follicles when iontophoresis is applied. This nanoparticle preparation technique may be applicable to minoxidil sulfate preparation for treating alopecia [40] and terbinafine hydrochloride preparation for tinea disease [41]. Acknowledgement This work was supported by MEXT-Supported Program for the Strategic Research Foundation at Private Universities 2010–2014, (Grant Number: S1001019). References [1] M.R. Prausniz, R. Langer, Nat. Biotechol. 26 (2008) 1261–1268. [2] J.A. Bouwstra, P.L. Honeywell-Nguyen, Adv. Drug Deliv. Rev. 54 (2002) S41–S55. [3] M. Sugino, H. Todo, K. Sugibayashi, Yakugaku Zasshi. 129 (2009) 1453–1458. [4] C.R. Behl, G.L. Flynn, T. Kurihara, N. Harper, W. Smith, W.I. Higuchi, N.F.H. Ho, C.L. Pierson, J. Invest. Dermatol. 75 (1980) 346–352. [5] A.C. Williams, B.W. Barry, Adv. Drug Deliv. Rev. 64 (2012) 128–137. [6] T. Ghafourian, P. Zandasrar, H. Hamishekar, A. Nokhodchi, J. Controlled Release 99 (2004) 113–125. [7] Y.N. Kalia, A. Naik, J. Garrison, R.H. Guy, Adv. Drug Deliv. Rev. 56 (2004) 619–658. [8] K. Tomoda, H. Terashima, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 88 (2011) 706–710. [9] K. Tomoda, H. Terashima, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 92 (2012) 50–54. [10] I. Takeuchi, K. Fukuda, S. Kobayashi, K. Makino, Biomed. Mater. Eng. 27 (2016) 475–483. [11] K. Mori, T. Hasegawa, S. Sato, K. Sugibayashi, J. Controlled Release 90 (2003) 171–179. [12] H.A.E. Benson, S. Namjoshi, J. Pharm. Sci. 97 (2008) 3591–3610. [13] N. Inoue, H. Todo, D. Iidaka, Y. Tokudome, F. Hashimoto, T. Kishino, K. Sugibayashi, Int. J. Pharm. 391 (2010) 65–72. [14] F. Bounoure, M. Lahiani Skiba, M. Besnard, P. Arnaud, E. Mallet, M. Skiba, J. Dermatol. Sci. 52 (2008) 170–177. [15] G. Cevc, U. Vierl, J. Controlled Release 141 (2010) 277–299. [16] H. Fessi, F. Puisieux, J.P. Devissaguet, N. Ammoury, S. Benita, Int. J. Pharm. 55 (1989) R1–R4. [17] J.Y. Xiong, X.Y. Liu, P.D. Sawant, S.B. Chen, T.S. Chung, K.P. Pramoda, J. Chem. Phys. 121 (2004) 12626–12631.
[18] J.Y. Xiong, X.Y. Liu, S.B. Chen, T.S. Chung, J. Phys. Chem. 109 (2005) 13877–13882. [19] I. Takeuchi, K. Tomoda, A. Hamano, M. Makino, Colloids Suf. A: Physicochem. Eng. Asp. 520 (2017) 771–778. [20] I. Takeuchi, S. Yamaguchi, S. Goto, K. Makino, Drug release behavior of hydrophobic drug-loaded poly (lactide-co-glycolide) nanoparticles: effects of glass transition temperature, Colloids Surf. A: Physicochem. Eng. Asp. 529 (2017) 328–333. [21] I. Takeuchi, Y. Tetsuka, T. Nii, M. Shinogase, K. Makino, Inhalable nanocomposite particles using amino acids with improved drug content and humidity resistance, Colloids Surf. A: Physicochem. Eng. Asp. 529 (2017) 387–393. [22] K. Tomoda, N. Yabuki, H. Terada, K. Makino, Colloids Surf. A: Physicochem. Eng. Asp. 457 (2014) 88–93. [23] I. Takeuchi, S. Kobayashi, Y. Hida, K. Makino, Colloids Surf. B 155 (2017) 35–40. [24] R.H. Guy, Y.N. Kalia, M.B. Delgado-Charro, V. Merino, A. López, D. Marro, J. Controlled Release 64 (2000) 129–132. [25] I. Takeuchi, K. Tomoda, M. Koji, K. Makino, Hydrophilic drug-loaded PLGA nanoparticles for transdermal delivery, Colloid Polym. Sci. 295 (2017) 911–983. [26] A.T.M. Serajuddin, Adv. Drug Deliv. Rev. 59 (2007) 603–616. [27] K. Tomoda, A. Watanabe, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 97 (2012) 84–89. [28] H. Fessi, F. Puisieux, J. Ph. Decissaguet, N. Ammoury, S. Benita, Int. J. Pharm. 55 (1989) R1–R4. [29] H. Yamamoto, Y. Kuno, S. Sugimoto, H. Takeuchi, Y. Kawashima, J. Controlled Release 102 (2005) 373–381. [30] H. Tsujimoto, Y. Tsukada, K. Hara, Y. Kawashima, J. Soc. Powder Technol. Japan 44 (2007) 453–458. [31] D.M. DiPasquale, M.J. Buono, F.W. Kolkhorst, Jpn. J. Physiol. 53 (2003) 427–430. [32] G. Mittal, D.K. Sahana, V. Bhardwaj, M.N.V. Ravi Kumar, J. Controlled Release 119 (2007) 77–85. [33] S. Kawai, Y. Takagi, S. Kaneko, T. Kurosawa, Exp. Anim. 60 (2011) 481–487. [34] H. Ohshima, J. Colloid Interface Sci. 163 (1994) 474–483. [35] K. Makino, T. Taki, M. Ogura, S. Handa, M. Nakajima, T. Kondo, H. Ohshima, Biophys. Chem. 47 (1993) 261–265. [36] K. Makino, M. Ikekita, T. Kondo, S. Tanuma, H. Ohshima, Coll. Pol. Sci. 272 (1994) 487–492. [37] K. Makino, F. Fukai, S. Hirata, H. Ohshima, Colloids Surf. B: Biointerfaces 7 (1996) 235–238. [38] S.K. Sahoo, J. Panyam, S. Prabha, V. Labhasetwar, J. Controlled Release 82 (2002) 105–114. [39] H. Tsujimoto, K. Hara, C.C. Huang, T. Yokoyama, H. Yamamoto, H. Takeuchi, Y. Kawashima, K. Akagi, N. Miwa, J. Soc. Powder Technol. Japan 41 (2004) 867–875. [40] G.M. Gelfuso, T. Gratieri, M.B. Delgado-Charro, R.H. Guy, R.F.V. Lopez, J. Pharm. Sci. 102 (2013) 1488–1494. [41] A. Sumitha, M. Geetha, G.H. Shashikala, H.S. Somashekar, K.S. Chandan, Int. J. Basic Clin. Pharmacol. 4 (2015) 474–478.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Protocols
Iontophoretic transdermal delivery using chitosan-coated PLGA nanoparticles for positively charged drugs Issei Takeuchi a,b,c , Tomoyoshi Takeshita a , Takaaki Suzuki a , Kimiko Makino a,b,c,∗ a b c
Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan Center for Drug Delivery Research, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan Center for Physical Pharmaceutics, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan
a r t i c l e
i n f o
Article history: Received 12 July 2017 Received in revised form 7 September 2017 Accepted 3 October 2017 Keywords: Donepezil hydrochloride Transdermal administration Iontophoresis Coumarin Rhodamine poly(dl-lactide-co-glycolide) PLGA Positively charged nanoparticles Hydrophilic drug
a b s t r a c t Recently, poly(dl-lactide-co-glycolide) (PLGA) nanoparticles prepared using a combination of an antisolvent diffusion method with preferential solvation was shown to be beneficial for the iontophoretic transdermal delivery of therapeutic agents. Also, this preparation method can contain a hydrophilic drug. However, since PLGA nanoparticles were negatively charged, it was difficult to apply iontophoresis for positively charged hydrophilic drugs. In this study, we prepared positively charged PLGA nanoparticles containing donepezil hydrochloride (DP). DP was used as a positively charged hydrophilic drug model. The PLGA nanoparticles were coated with chitosan hydroxypropyltrimonium chloride. The average particle diameter of the nanoparticles was 117.7 ± 60.6 nm and the surface charge number density changed from negative to positive. Ex vivo skin accumulation study was carried out using abdominal rat skin and a Franz-type diffusion cell with/without iontophoresis. When iontophoresis was applied, the DP concentration in the rat skin of chitosan-coated PLGA nanoparticles was 2.2 times higher than that of non-coated PLGA nanoparticles. This indicated that chitosan-coated PLGA nanoparticles were suitable for iontophoresis. To investigate the transdermal delivery route of the nanoparticles, we prepared chitosancoated PLGA nanoparticles containing DP, coumarin-6, and rhodamine 6G. Coumarin-6 and rhodamine 6G were used as a trace marker of the PLGA nanoparticles and positively charged hydrophilic drug model, respectively. From the results of ex vivo accumulation test of this fluorescent nanoparticles, it was suggested that positively charged hydrophilic drugs reached the hair follicles as a nanoparticle, and then they were released from the nanoparticles. © 2017 Published by Elsevier B.V.
1. Introduction The skin has been studied as a site of administration for systemic and local delivery of therapeutic agents. Transdermal delivery can avoid the effect of first-pass hepatic metabolism, and deliver therapeutic agents for a long period of time. The skin is composed of the epidermis (thickness of 50–100 m), dermis (thickness of 3–5 mm) and subcutaneous tissue (thickness of 1–2 mm) [1]. The stratum corneum, the outermost layer of the skin (epidermis), is composed of keratin-filled dead cells (corneocytes) embedded in a complex intercellular lipid mixture, particularly rich in ceramides, cholesterol, fatty acids, organized in bilayer arrays. The stratum corneum functions as a barrier against intrusion such as microorganisms
∗ Corresponding author at: Faculty of Pharmaceutical Sciences, Tokyo University of Science, 2641, Yamazaki, Noda, Chiba 278-8510, Japan. E-mail address: [email protected] (K. Makino). https://doi.org/10.1016/j.colsurfb.2017.10.011 0927-7765/© 2017 Published by Elsevier B.V.
(bacteria and viruses), radiation and transpiration of water [2,3]. Simultaneously, the permeability of drugs into the skin is also poor. In order to therapeutic agents to penetrate the skin, drugs must be able to pass through this barrier. Various penetration enhancing approaches of drugs through the skin have been studied. Occlusive dressing technique (ODT) is the facilitation of percutaneous absorption by hydrating stratum corneum [4]. Penetration enhancers, chemicals that interact with skin constituents, is one long–standing approach to promoting drug flux [5,6]. Physical modifications such as iontophoresis which promotes the penetration of hydrophilic and charged molecules across skin by electric energy [7–10], electroporation which produces small pore on the surface of stratum corneum by adding pulse voltage [11], microneedles [12], and needle-free injectors [13], and combinations of physical and chemical enhancement technique have also been studied [14]. Moreover, nano-sized systems such as liposomes, micelles, nanoparticles, and dendrimers have attracted attention as drug carriers for transdermal drug deliv-
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
ery system [15]. Recently, we have reported a combination of an antisolvent diffusion method [16] with preferential solvation phenomenon [17,18] for nanoparticles including a hydrophobic drug. The nanoparticles having high surface charge number density were prepared using poly(dl-lactide-co-glycolide) (PLGA), widely used biocompatible and biodegradable copolymer [19–21], and they were beneficial for the iontophoretic transdermal delivery of therapeutic agents [22]. It was suggested that the combined use of this nanoparticle preparation method and iontophoresis was actually useful for the treatment of osteoporosis [23]. In drug delivery using iontophoresis, electrorepulsion and electroosmosis play an important role [24]. In the drug-loaded nanoparticles, to fully utilize these effects, it is necessary to adjust the positive or negative of the surface charge of the carrier nanoparticles with the drug. In the previous study, we reported that positively charged hydrophilic drug containing nanoparticles prepared using a combination of an antisolvent diffusion method with preferential solvation were useful for transdermal administration [25]. However, since the PLGA particle surface was negatively charged, iontophoresis could not be applied. Salt formation is the most common and effective method of increasing the solubility and dissolution rate of acidic and basic drugs [26]. Examples of positively charged drugs include drugs in the form of hydrochloride, sulfate or sodium salts. The main aim of the present study was to apply a transdermal delivery system using PLGA nanoparticles and iontophoresis to positively charged drugs for efficient transdermal delivery. We prepared Donepezil hydrochloride (DP)-loaded PLGA nanoparticles and to obtain positively charged PLGA nanoparticles, the surface of the nanoparticles was coated with chitosan (CS). DP was used as a positively charged hydrophilic drug model. We determined physicochemical characteristics and skin accumulation of DP of the chitosan-coated PLGA nanoparticles. Moreover, the transdermal delivery routes of negatively charged PLGA nanoparticles and positively charged CS-coated PLGA nanoparticles were observed. Coumarin-6 and rhodamine 6G were used as a trace marker of the PLGA nanoparticles and positively charged hydrophilic drug model, respectively [27]. 2. Materials and methods 2.1. Materials Poly(dl-lactide-co-glycolide) (PLGA, Mw: 10,000, monomer composition of lactic acid/glycolic acid = 75/25), sucrose, l(+)-arginine (purity ≥ 98%), N-methyl-2-pyrrolidone (NMP, C5 H9 NO, purity > 98%), potassium dihydrogen phosphate (KH2 PO4 , purity > 99.5%), ethyl p-hydroxybenzoate (C9 H10 O3 ), and rhodamine 6G (C28 H31 ClN2 O3 ) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Donepezil hydrochloride (DP, C24 H29 NO3 ·HCl, purity > 98%) was purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Coumarin-6 (C20 H18 N2 O2 S, purity > 98%) was purchased from Sigma–Aldrich (St. Louis, MO, USA). Acetonitrile (CH3 CN, JP, USP/NF, EP) was purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Isoflurane for the animal was purchased from Mylan Inc. (Pittsburgh, PA, USA). Chitosan hydroxypropyltrimonium chloride (CS, Mw: 100,000–200,000) was kindly donated by Katakura & Co-op Agri Corp. (Tokyo, Japan). Other chemicals were of the highest reagent grade commercially available. 2.2. Preparation of DP-loaded PLGA nanoparticles coated with CS DP-loaded PLGA nanoparticles (bare nanoparticles) were prepared using a combination of an antisolvent diffusion method with preferential solvation [22,23]. Briefly, 96 mg of PLGA and 4 mg of DP
521
were dissolved in 2 mL of NMP. The solution was injected into 20 mL of 0.3% (w/v) l-arginine aqueous solution. Bare nanoparticles were immediately precipitated by Marangoni effect [28]. The suspension was dialyzed for 24 h in a dialysis tube (UC36-32-100, molecular weight cut off: 14,000, EIDIA Co., Ltd., Tokyo, Japan) to remove excess NMP and unloaded-DP. After dialysis, the suspension was mixed with equivalent amounts of 0.02% (w/v) chitosan solution, and DP-loaded CS-coated PLGA nanoparticles (CS-coated nanoparticles) were obtained [29,30]. Then, 500 mg of sucrose was added to the suspension as a cryoprotectant. After freezing at −30◦ C, this suspension was lyophilized by usage of a freeze dryer (FDU-1200, Tokyo Rikakikai Co., Ltd., Tokyo, Japan). Surface properties of the bare and CS-coated nanoparticles were observed using a scanning electron microscope (SEM, JSM-6060LA, JEOL Ltd., Akishima, Japan). The mean volume diameters and polydispersity index values of these nanoparticles were measured using a particle size analyzer (ELSZ-2, Otsuka Electronics Co., Ltd., Osaka, Japan). Samples were dispersed in purified water and measured at 25 ◦ C. Also, the electrophoretic mobility of the nanoparticles was measured in physiological saline with nine different ionic strengths, at skin surface temperature (32◦ C) [31]. DP content in the nanoparticles was measured using high-performance liquid chromatography (HPLC, SIL-20A prominence, SPD-20A prominence, LC-20AD prominence, CTO-10ASvp, DGU-20A3 prominence, Shimadzu) at 271 nm with a polymer-coated ODS column (CAPCELL PAK MF Ph-1, size: 4.6 mm × 150 mm, Shiseido Co. Ltd., Tokyo, Japan). The mobile phase consisted of a solution of phosphate buffer solution (pH 2.6) and acetonitrile with a volume ratio of 85:15. Ten milligrams of samples were dissolved in 5 mL of an acetonitrile-phosphate buffer solution with a volume ratio of 1:1. HPLC measurements were carried out at 40 ◦ C (flow rate: 1.0 mL/min), and 30 L of sample solution were applied. All HPLC measurements were carried out under the same conditions. To confirm the state of DP in the CS-coated nanoparticles during skin accumulation study, the release rate of DP from the nanoparticles was investigated by using dialysis membrane method [32]. The CS-coated nanoparticles were redispersed in 50 mM of NaCl solution to a concentration of 0.1% (w/v) DP. The suspension was placed in a dialysis tube (UC20-32-100, molecular weight cut off: 14,000, EIDIA Co., Ltd.) and added in 95 mL of 50 mM NaCl solution. The sample suspensions were shaken at 30 rpm at 32◦ C. After 0.5, 1, 2, 6, and 12 h, the samples were taken and the amount of released DP was quantified using HPLC. We compared this result to the release rate of 0.1% (w/v) DP solution, prepared using 50 mM of NaCl solution, from the dialysis tube (UC20-32-100). 2.3. Ex vivo skin accumulation study of CS-coated nanoparticles Male Sprague-Dawley rats aging 6 weeks were purchased from Japan SLC Inc. (Tokyo Japan). All animal care was conducted under the Guidelines for Animal Experimentation of Tokyo University of Science, which is based on the Guidelines for Animal Experimentation of the Japanese Association for Laboratory Animal Science. Rats were anesthetized by intraperitoneal administration of a combination anesthetic, which was prepared with 0.3 mg/kg of medetomidine, 4.0 mg/kg of midazolam, and 5.0 mg/kg of butorphanol, under isoflurane anesthesia [33]. After shaved their abdominal hair, animals were killed by exsanguination. Then, abdominal skin was excised and mounted on a Franz-type diffusion cell [23]. Bare and CS-coated nanoparticles were dispersed in 50 mM NaCl solution, and 0.1% (w/v) DP contained suspension was prepared. Then, 3 mL of the suspension and physiological saline (30 mL, pH 7.4, I = 0.154 M) were used as donor medium and receptor medium, respectively. Also, 0.1% (w/v) DP solution was used for donor medium for comparison. The receptor medium was kept at 32◦ C and stirred with a magnetic stirrer. Two hours after the start
522
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
of the permeation test, 1 mL of the receptor medium was collected and 1 mL of fresh physiological saline was added to receptor cell. When iontophoresis was applied, Ag/AgCl electrodes were used. Iontophoresis was performed at a constant current of 0.25 mA/cm2 for 2 h. After the permeability study using a Franz-type diffusion cell, the skin was carefully washed to remove nanoparticles or free DP accumulated on the surface of the skin. To measure DP amount in rat skin, the sample skins were lyophilized and broken into fractions. Then, 0.1 mL of acetonitrile solution containing 0.3% (w/v) of ethyl p-hydroxybenzoate as an internal standard substance, and 4.9 mL of a solution of acetonitrile and phosphate buffer solution (pH 6.0) with a volume ratio of 2:1 were added to collected samples. After shaking for 60 min, the sample solutions were centrifuged (Model 2410, Kubota Corp., Tokyo, Japan) at 4000 rpm for 10 min to remove protein. The supernatants were collected and passed through 0.45 m filter and the amounts of DP in the samples were measured using HPLC.
2.4. Preparation of fluorescent PLGA nanoparticles To investigate the transdermal delivery route of CS-coated nanoparticles, we prepared PLGA nanoparticles containing DP, coumarin-6, and rhodamine 6G (CS-coated fluorescent nanoparticles). Bare fluorescent nanoparticles were also prepared for comparison. Ninety-six milligrams of PLGA, 3 mg of DP, 500 g of coumarin-6, and 500 g of rhodamine 6G were dissolved in 2 mL of NMP. Nanoparticle preparation was carried out in the same manner described in Section 2.2. Contents of coumarin-6 and rhodamine 6G in the nanoparticles were measured using HPLC. To measure coumarin-6 and rhodamine 6G, a solution of acetonitrile and purified water with a volume ratio of 45:55 and a solution of acetonitrile and purified water with a volume ratio of 3:7 were used as mobile phases, respectively. To investigate the retention of coumarin-6 in PLGA nanoparticles and the release behavior of rhodamine 6G from PLGA nanoparticles, we carried out release test of those substances from fluorescent nanoparticles. The nanoparticles were redispersed in 50 mM of NaCl solution to a concentration of 0.1% (w/v) DP. The suspension was placed in a dialysis tube (UC2032-100) and added in 95 mL of 50 mM NaCl solution. The sample suspensions were shaken at 30 rpm at 32◦ C. After 0.5, 1, 2, 4, 6, 8, 10, 12, 24, 72, and 120 h, the samples were taken and the amount of released DP was quantified using HPLC.
Fig. 1. Particle size distribution of donepezil hydrochloride-loaded PLGA nanoparticles (bare nanoparticles) and donepezil hydrochloride-loaded chitosan hydroxypropyltrimonium chloride-coated PLGA nanoparticles (CS-coated nanoparticles).
3. Results and discussion 3.1. Characterization of DP-loaded bare and CS-coated nanoparticles As shown in Table 1, bare nanoparticles and CS-coated nanoparticles with mean volume diameters of 80.2 ± 30.1 and 117.7 ± 60.6 nm, respectively, were prepared. Their particle size distributions were shown in Fig. 1. DP contents of bare and CS-coated nanoparticles were 3.14 ± 0.05 and 2.87 ± 0.06%, respectively. The entrapment efficiency of DP in bare nanoparticles was 78.5% and yield of CS-coated nanoparticles was 79.9%. Fig. 2 displays SEM images of the nanoparticles, showing spherical dispersed particles. Their electrophoretic mobility were also measured and the results were shown in Fig. 3. The data were analyzed by using Ohshima’s electrokinetic theory for soft particles [34]. At a colloidal particle covered with a layer of polyelectrolyte chains moving in a liquid containing a symmetrical electrolyte of valency v and bulk concentration (number density) n (m−3 ), ionized groups of valency z are uniformly distributed at a number density of N (m−3 ). Therefore, it is expressed as r 0 0 /m + DON / d zeN f( ) + 2 a 1/m + 1/
= with 2.5. Evaluation of transdermal delivery route of fluorescent nanoparticles Ex vivo skin permeation experiments of fluorescent nanoparticles were carried out in the same manner described in Section 2.3. The cross sections of the skin after 2 h permeability studies were used as samples. The samples were fixed using 4% (w/v) paraformaldehyde phosphate buffer solution which con® tained 10% of sucrose and embedded in Jung tissue freezing medium (Leica Instruments, Nussloch, Germany) in liquid nitrogen. Then, they were cut into 20 m sections using a cryostat (CM3050S, Leica Instruments) at −20◦ C. The samples were mounted on a sliding glass and were observed under a fluorescence microscope (coumarin-6 Ex/Em = 488/519 nm, rhodamine 6G Ex/Em = 518/543 nm, Fluoview FV1000, Olympus Corp., Tokyo, Japan).
f
d a
2 = 3
DON =
0 =
(1)
1+
1
3
(2)
2 1 + d/a
kT zN 2 zN ln[ + {( ) + 1} ve 2vn 2vn
zN kT zN 2 (ln[ + {( ) + 1} ve 2vn 2vn
1/2
(3)
],
1/2
]+
2vn zN 2 [1 − {( ) + 1} zN 2vn
1/2
]), (4)
1/2
= ␥/
,
m = [1 + (
zN 2 ) ] 2vn
(5) 1/4
,
(6)
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
523
Table 1 Physicochemical properties of donepezil hydrochloride (DP)-loaded bare nanoparticles and chitosan (CS)-coated nanoparticles (mean ± S.D., n = 3).
Particle diameter (nm) Polydispersity index DP content in nanoparticles% (w/w) Electrophoretic mobility at I = 50 mM Particle surface charge number density (mM) Particle softness parameter (nm)
Bare nanoparticles
CS-coated nanoparticles
80.2 ± 30.1 0.15 3.14 ± 0.04 −68.1 ± 4.50 −72.4 2.21
117.7 ± 60.6 0.18 2.87 ± 0.06 23.1 ± 0.31 14.3 2.85
Fig. 2. Scanning electron microscopy images of donepezil hydrochloride-loaded PLGA nanoparticles taken at an accelerating voltage of 10–15 kV (magnification: 20,000×). (a) Bare nanoparticles. (b) Chitosan hydroxypropyltrimonium chloridecoated nanoparticles.
=(
2ne2 v2 ) r 0 kT
1/2
,
(7)
where a is the particle diameter, d is the thickness of the ionpenetrable surface layer, is the viscosity, is the frictional coefficient of the surface layer, εr is the relative permittivity of the solution, ε0 is the permittivity of a vacuum, DON is the Donnan potential of the surface layer, 0 is the potential at the boundary between the surface layer (it means the surface potential of the polyelectrolyte-coated particle) and the surrounding solution and k is the Boltzmann constant. The parameter is the DebyeH`u` ckel parameter, and m shows the Debye-H`u` ckel parameter in the polyelectrolyte layer. We have determined the density of fixed-charges zN and the reciprocal of (: the degree of friction exerted on the liquid flow in the polyelectrolyte layer of the parti-
Fig. 3. (a) Electrophoretic mobility of donepezil hydrochloride-loaded PLGA nanoparticles. Solid line are theoretical results calculated with zN = −7.24 × 10−2 M and 1/ = 2.21 nm (mean ± S.D., n = 3). (b) Electrophoretic mobility of donepezil hydrochloride-loaded chitosan hydroxypropyltrimonium chloride-coated PLGA nanoparticles. Solid line are theoretical results calculated with zN = 1.43 × 10−2 M and 1/ = 2.85 nm (mean ± S.D., n = 3).
cles). 1/ has the dimension of length and shows the permeability of the polyelectrolyte layer. Therefore, it can be considered as the parameter of “softness” of particle surface [34,35]. In the previous report, the curve-fitting procedure was used to determine zN and 1/ [10,25,36,37]. As shown in Fig. 3, we have applied it to our measurement result of electrophoretic mobility of bare and CS-coated nanoparticles. From the result of the procedures (shown as a solid
524
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
Fig. 5. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study with iontophoresis using bare fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
nanoparticles were more rigid than CS-coated nanoparticles, since bare nanoparticles did not have the CS layer at the surface of the nanoparticles. 3.2. Ex vivo skin accumulation study of CS-coated nanoparticles
Fig. 4. (a) Cumulative release rate of donepezil hydrochloride (DP) from DP-loaded PLGA nanoparticles and DP-loaded chitosan-coated PLGA nanoparticles (CS-coated nanoparticles, mean ± S.D., n = 3, * p < 0.05, ** p < 0.01, t-test). (b) DP amount in the rat skin when applied with passive diffusion or iontophoresis (mean ± S.D., n = 4, *p < 0.05, **p < 0.01, Tukey’s test).
line in Fig. 3), zN = −7.24 × 10−2 M and 1/ = 2.21 nm were obtained in the bare nanoparticles. Also, zN = 1.43 × 10−2 M and 1/ = 2.85 nm were obtained in the CS-coated nanoparticles. By coating CS on the surface of bare nanoparticles, the value of the surface charge number density of nanoparticles changed from negative to positive. PLGA molecule has negative charges at neutral pH because of their ionized terminal carboxyl groups [38]. This result indicated that the particle surface was fully covered with CS and the negative charge derived from PLGA was shielded. Also, the value of softness parameter of CS-coated nanoparticles was about 1.3 times higher than that of bare nanoparticles. It was assumed that bare
The cumulative release rate of DP from CS-coated nanoparticles in 50 mM of NaCl solution at 32◦ C is shown in Fig. 4a. Compared to 0.1% (w/v) DP solution, CS-coated nanoparticles significantly decreased release of DP in 1–12 h. The release rates of CS-coated nanoparticles and DP solution after 2 h from the start of the test were 46.1 ± 3.06 and 81.6 ± 2.18%, respectively. Therefore, it was considered that half of the DP was retained in the nanoparticles 2 h after the start of the skin accumulation study. Also, since this result was similar to that of bare nanoparticles, CS-coated nanoparticles and bare nanoparticles showed no difference in drug release behavior. Fig. 4b shows results of ex vivo skin accumulation study of bare and CS-coated nanoparticles. Without iontophoresis, the value of DP accumulation in the skin of CS-coated nanoparticles was 5.3 times higher than that of DP solution. This result assumed that hydrophobicity of DP was enhanced by particulation using PLGA, and the hydrophilic drug was partitioned into the stratum corneum. When iontophoresis was applied to CS-coated nanoparticles, the DP concentration in the rat skin was 1.9 times higher than that of CS-coated nanoparticles without iontophoresis. In addition,
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
Fig. 6. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study without iontophoresis using CS-coated fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
when iontophoresis was applied, the value of DP accumulation in the skin increased to 2.2 times by CS modification. These significant difference indicated that CS-coated nanoparticles were suitable for transdermal delivery of positively charged drug using iontophoresis. 3.3. Transdermal delivery route of fluorescent nanoparticles in rat skin Contents of coumarin-6 and rhodamine 6G in the nanoparticles measured using HPLC were 0.4 and 0.5%, respectively. The cumulative release rate of coumarin-6 from the nanoparticles was 0.67% at 8 h and 2.63% at 120 h. In the previous study, the release rate of coumarin-6 from 150 nm PLGA nanoparticles was less than 1% of the content in the particles at 2–8 h [27]. Thus, we considered that coumarin-6 was successfully retained in nanoparticles. In contrast, rhodamine 6G was rapidly released from the nanoparticles and reached 13.6% at 2 h, 60.71% at 12 h, and 89.04% at 120 h. From these results, we confirmed that coumarin-6 had the capability of trace marker of CS-coated PLGA nanoparticles, and rhodamine 6 G could be used as a hydrophilic drug model. In Figs. 5–7, the cross sections of the rat skin after 2 h accumulation study were shown. As shown in Fig. 5, bare fluorescent nanoparticles were observed in deep follicles, however, the fluorescence of rhodamine 6G was weak, and it was not observed in the deep follicles. The released rates of coumarin-6 and rhodamine 6 G from the nanoparticles
525
Fig. 7. Fluorescence microscope images of cross sections of the rat skin after 2 h accumulation study with iontophoresis using CS-coated fluorescent nanoparticles. (a) Fluorescence of coumarin-6. (b) Fluorescence of rhodamine 6G.
after 2 h were 0.24% and 13.6%, respectively. From these results, it was suggested that diffusion of released rhodamine 6G was prevented by iontophoresis. Fig. 7 shows the results of CS-coated fluorescent nanoparticles without iontophoresis, coumarin-6 and rhodamine 6G were detected in stratum corneum and hair follicle after 2 h accumulation study. Since the intercellular spaces in the stratum corneum are 70 nm, we considered that a part of CS-coated fluorescent nanoparticles penetrated the intercellular spaces of the stratum corneum [39]. As shown in Fig. 7, when iontophoresis was applied to nanoparticles, coumarin-6 and rhodamine 6G were observed in deep follicles compared to passive diffusion. The fluorescence of rhodamine 6G was observed in the deeper part of the hair follicles compared to coumarin-6. These results indicated that rhodamine 6 G, which contained in CS-coated nanoparticles, reached the hair follicles, and then it was released from the nanoparticles. 4. Conclusions DP-loaded PLGA nanoparticles coated with CS having positive surface charge number density were prepared by using a combination of an antisolvent diffusion method with preferential solvation. This nanoparticle showed high skin accumulation of DP in passive diffusion due to improved hydrophobicity. In addition, when iontophoresis was applied to CS-coated nanoparticles, the DP accu-
526
I. Takeuchi et al. / Colloids and Surfaces B: Biointerfaces 160 (2017) 520–526
mulation was 2.2 times higher than that of bare nanoparticles. In the study of the transdermal delivery route using the fluorescent nanoparticles, we confirmed that transdermal delivery of a positively charged substance from PLGA nanoparticles was improved by coating the nanoparticle surface with CS. From these results, it was suggested that CS-coated nanoparticles can efficiently deliver positively charged agents to the deep hair follicles when iontophoresis is applied. This nanoparticle preparation technique may be applicable to minoxidil sulfate preparation for treating alopecia [40] and terbinafine hydrochloride preparation for tinea disease [41]. Acknowledgement This work was supported by MEXT-Supported Program for the Strategic Research Foundation at Private Universities 2010–2014, (Grant Number: S1001019). References [1] M.R. Prausniz, R. Langer, Nat. Biotechol. 26 (2008) 1261–1268. [2] J.A. Bouwstra, P.L. Honeywell-Nguyen, Adv. Drug Deliv. Rev. 54 (2002) S41–S55. [3] M. Sugino, H. Todo, K. Sugibayashi, Yakugaku Zasshi. 129 (2009) 1453–1458. [4] C.R. Behl, G.L. Flynn, T. Kurihara, N. Harper, W. Smith, W.I. Higuchi, N.F.H. Ho, C.L. Pierson, J. Invest. Dermatol. 75 (1980) 346–352. [5] A.C. Williams, B.W. Barry, Adv. Drug Deliv. Rev. 64 (2012) 128–137. [6] T. Ghafourian, P. Zandasrar, H. Hamishekar, A. Nokhodchi, J. Controlled Release 99 (2004) 113–125. [7] Y.N. Kalia, A. Naik, J. Garrison, R.H. Guy, Adv. Drug Deliv. Rev. 56 (2004) 619–658. [8] K. Tomoda, H. Terashima, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 88 (2011) 706–710. [9] K. Tomoda, H. Terashima, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 92 (2012) 50–54. [10] I. Takeuchi, K. Fukuda, S. Kobayashi, K. Makino, Biomed. Mater. Eng. 27 (2016) 475–483. [11] K. Mori, T. Hasegawa, S. Sato, K. Sugibayashi, J. Controlled Release 90 (2003) 171–179. [12] H.A.E. Benson, S. Namjoshi, J. Pharm. Sci. 97 (2008) 3591–3610. [13] N. Inoue, H. Todo, D. Iidaka, Y. Tokudome, F. Hashimoto, T. Kishino, K. Sugibayashi, Int. J. Pharm. 391 (2010) 65–72. [14] F. Bounoure, M. Lahiani Skiba, M. Besnard, P. Arnaud, E. Mallet, M. Skiba, J. Dermatol. Sci. 52 (2008) 170–177. [15] G. Cevc, U. Vierl, J. Controlled Release 141 (2010) 277–299. [16] H. Fessi, F. Puisieux, J.P. Devissaguet, N. Ammoury, S. Benita, Int. J. Pharm. 55 (1989) R1–R4. [17] J.Y. Xiong, X.Y. Liu, P.D. Sawant, S.B. Chen, T.S. Chung, K.P. Pramoda, J. Chem. Phys. 121 (2004) 12626–12631.
[18] J.Y. Xiong, X.Y. Liu, S.B. Chen, T.S. Chung, J. Phys. Chem. 109 (2005) 13877–13882. [19] I. Takeuchi, K. Tomoda, A. Hamano, M. Makino, Colloids Suf. A: Physicochem. Eng. Asp. 520 (2017) 771–778. [20] I. Takeuchi, S. Yamaguchi, S. Goto, K. Makino, Drug release behavior of hydrophobic drug-loaded poly (lactide-co-glycolide) nanoparticles: effects of glass transition temperature, Colloids Surf. A: Physicochem. Eng. Asp. 529 (2017) 328–333. [21] I. Takeuchi, Y. Tetsuka, T. Nii, M. Shinogase, K. Makino, Inhalable nanocomposite particles using amino acids with improved drug content and humidity resistance, Colloids Surf. A: Physicochem. Eng. Asp. 529 (2017) 387–393. [22] K. Tomoda, N. Yabuki, H. Terada, K. Makino, Colloids Surf. A: Physicochem. Eng. Asp. 457 (2014) 88–93. [23] I. Takeuchi, S. Kobayashi, Y. Hida, K. Makino, Colloids Surf. B 155 (2017) 35–40. [24] R.H. Guy, Y.N. Kalia, M.B. Delgado-Charro, V. Merino, A. López, D. Marro, J. Controlled Release 64 (2000) 129–132. [25] I. Takeuchi, K. Tomoda, M. Koji, K. Makino, Hydrophilic drug-loaded PLGA nanoparticles for transdermal delivery, Colloid Polym. Sci. 295 (2017) 911–983. [26] A.T.M. Serajuddin, Adv. Drug Deliv. Rev. 59 (2007) 603–616. [27] K. Tomoda, A. Watanabe, K. Suzuki, T. Inagi, H. Terada, K. Makino, Colloids Surf. B: Biointerfaces 97 (2012) 84–89. [28] H. Fessi, F. Puisieux, J. Ph. Decissaguet, N. Ammoury, S. Benita, Int. J. Pharm. 55 (1989) R1–R4. [29] H. Yamamoto, Y. Kuno, S. Sugimoto, H. Takeuchi, Y. Kawashima, J. Controlled Release 102 (2005) 373–381. [30] H. Tsujimoto, Y. Tsukada, K. Hara, Y. Kawashima, J. Soc. Powder Technol. Japan 44 (2007) 453–458. [31] D.M. DiPasquale, M.J. Buono, F.W. Kolkhorst, Jpn. J. Physiol. 53 (2003) 427–430. [32] G. Mittal, D.K. Sahana, V. Bhardwaj, M.N.V. Ravi Kumar, J. Controlled Release 119 (2007) 77–85. [33] S. Kawai, Y. Takagi, S. Kaneko, T. Kurosawa, Exp. Anim. 60 (2011) 481–487. [34] H. Ohshima, J. Colloid Interface Sci. 163 (1994) 474–483. [35] K. Makino, T. Taki, M. Ogura, S. Handa, M. Nakajima, T. Kondo, H. Ohshima, Biophys. Chem. 47 (1993) 261–265. [36] K. Makino, M. Ikekita, T. Kondo, S. Tanuma, H. Ohshima, Coll. Pol. Sci. 272 (1994) 487–492. [37] K. Makino, F. Fukai, S. Hirata, H. Ohshima, Colloids Surf. B: Biointerfaces 7 (1996) 235–238. [38] S.K. Sahoo, J. Panyam, S. Prabha, V. Labhasetwar, J. Controlled Release 82 (2002) 105–114. [39] H. Tsujimoto, K. Hara, C.C. Huang, T. Yokoyama, H. Yamamoto, H. Takeuchi, Y. Kawashima, K. Akagi, N. Miwa, J. Soc. Powder Technol. Japan 41 (2004) 867–875. [40] G.M. Gelfuso, T. Gratieri, M.B. Delgado-Charro, R.H. Guy, R.F.V. Lopez, J. Pharm. Sci. 102 (2013) 1488–1494. [41] A. Sumitha, M. Geetha, G.H. Shashikala, H.S. Somashekar, K.S. Chandan, Int. J. Basic Clin. Pharmacol. 4 (2015) 474–478.
