2015 Volume 63 Issue 2 Pages 59-67
Lutein is a macular pigment that contributes to maintaining eye health. The development of lutein-laden nanocarriers for ocular delivery would have the advantages of user friendliness and cost-effectiveness. Nano-scaled vehicles such as cyclodextrin (CD) and nanoemulsion could overcome the barriers caused by the scleral structure. This study focused on the development of hybrid nanocarriers containing nanoemulsion and CD for scleral lutein accumulation. In the presence of the nanoemulsion, CD forms such as βCD and hydroxyethyl (HE) βCD increased the partition of lutein into the porcine sclera. A combination of nanoemulsion and 2% HEβCD enhanced lutein accumulation to 119±6 µg g−1 h−1, which was 9.2-fold higher than that with lutein suspension alone. We explored the dose effect of CD in nanoemulsion on scleral lutein and found that the scleral accumulation of lutein was enhanced by increasing the CD content. The novel nanoemulsion had 95% drug-loading efficiency and low cytotoxicity in retinal cells. The CD-modified nanoemulsion not only improved the stability and entrapment efficacy of lutein in the aqueous system but also enhanced scleral lutein accumulation. An increase in the partition coefficient of lutein in porcine sclera when using the CD-modified nanoemulsion was also confirmed.
Lutein, a macular pigment, performs significant roles in maintaining the health of eyes. This carotenoid can reduce the risk of macular degeneration and scavenge harmful reactive oxygen species.1) Epidemiologic research shows lutein levels in eye tissues is inversely associated with ocular diseases such as cataracts and macular degeneration.2) Lutein also prevents ocular inflammation diseases including uveitis, choroidal neovascularization, retinal ischemia and retinitis.3) Since lutein cannot be synthesized in the human body, now dietary lutein is the main source to supply the macular pigment.4) The bioavailability of supplemental lutein in the retina depends on the administration routes and the delivery vehicles.5,6) However, the studies related to lutein ocular delivery are limited. How to improve the efficiency of topical lutein delivery to the macula is a challenge since the eye is isolated by multiple layers of biological barriers. Lutein has to overcome obstacles including the corneal tight junctions, the mucosal surface, and the blood–retinal barriers in order to reach the macula. Conventional eye drops and ointments have the drawback of poor ocular bioavailability, whereas systemic administration suffers from the first-pass metabolism.7)
The use of nano lipid vehicles has led to the solution of solubility-related problems for poorly soluble drugs, like curcumin, granisetron and so on.8,9) Oil in water nanoemulsions have been recognized as promising vehicles for lipophilic drugs since their biocompatibility, drug solubilization and biodegradability.10) Lipophilic drugs have been encapsulated in regular emulsions for ocular delivery such as flurbiprofen,11) indomethacin,12) baicalin and cyclosporine.13) Nanoemulsions have smaller size ranging about 50–200 nm and have transparent or translucent appearance as compared with the regular emulsion. They are kinetically stable and physically stable on the long term with no apparent flocculation or coalescence. The benefits of nanoemulsions as drug carriers include high payload, easy processing, fast sterilization and low production costs. Nanoemulsions have be proposed as ocular eye drops in virtue of their distinct advantages including sustained release of the drug applied to the eyes and high penetration in the deeper layers of the ocular structure.14)
Cyclodextrins (CDs) have been adapted as drug carriers in eye drops since they can stabilize drugs, reduce ocular irritation, enhance drug solubility and increase permeability.15) The following mechanisms of CD–drug complexes account for these advantages. CDs pack the lipophilic drugs in their cave and deliver the cargo through the aqueous tear fluid to the eye surface with the help of their hydrophilic outer surface.16) CDs also function as permeation enhancers by increasing the drug retention on the surface of the corneal epithelium.17) CDs can decrease drug irritation by limiting free drug concentration to a nonirritating level.18) There has been growing interest in the development of CD as delivery vehicles for lipophilic drugs.19,20) For example, the addition of γCD in lipid emulsions enhances the transdermal delivery of steroidal drugs such as progesterone21) and fludrocortisone.22) However, the studies combining CD and nanoemulsions for ophthalmic formulations are rare according to our literature survey. Ocular drug delivery is one of the most challenging fields because of the critical and specific environment of the eyes. These conventional dosage forms suffer from the problems of poor ocular bioavailability, caused by various anatomical and pathophysiological barriers prevailing in the eye. Recent studies demonstrate the applications of nanoparticulate systems like nanoparticles, liposomes, niosomes and cyclodextrins in the field of ocular drug delivery.23) Lutien exerts antioxidant activities in lipid phases by quenching singlet oxygen or free radicals, but their bioavailability through dietary supplementation is limited. Lutein is also a potent antioxidant and is found to enhance immune function, suppress mammary tumor growth and enhance lymphocyte proliferation.24) The scarce water solubility (1.3×10−9 M) of lutein represent a drawback for bioavailability. To circumvent these unfavorable characteristics, Jambhekar et al. report that lutein can be encapsulated in butyrate modified β-cyclodextrins to assemble nanocarriers with a diameter of 320 nm and a zeta potential of −8 mV.25) Pfitzner et al. also adapt methyl-β-cyclodextrin to enhance the water solubility of lutein, zeaxanthin for the medium supplementation in cell culture.26) However, their studies are limited to the lutein and CD complex. To the best of our knowledge, no study has been reported on the combining effects of cyclodextrin and nanoemulsion on scleral accumulation of lutein. With the relatively large area of sclera, transscleral delivery was suggested to be a promising method to deliver drugs to the posterior segment of the eye. Several approaches, including transscleral iontophoresis and fibrin sealants, have been reviewed for transscleral drug delivery.27) Although the above results are promising but further understanding on partition of drug laden vehicles is necessary to optimize transscleral delivery. Additionally, data about scleral lutein partition are limited, and the influences of lutein to sclera accumulation on the transport behavior of lutein have not been systematically studied. Therefore, our objectives were to (a) characterize the vehicles including the morphology, encapsulation, size distribution, zeta potential, (b) examine the partition coefficients and accumulation of lutein in the porcine sclera, (c) analyze the penetration depth of lutein, and (d) determine their cytotoxicity in retinal ganglion cells. The results in this study could help researchers to develop ocular delivery systems and to evaluate of the complex interplay between the lutein, the carrier composition and the sclera.
The lutein (20% content in corn oil) is kindly provided by DKSH Co. (Taipei, Taiwan). Fresh porcine eyes are kindly donated from Ya-Shin Pork Industry (Taoyuan, Taiwan). Transcutol HP is obtained from Gattefossé Co. (Lyon, France). Cyclodextrins were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). All other chemicals and reagents were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.) unless otherwise stated.
Preparation of Lutein-Laden NanoemulsionsLutein-laden vehicles (Table 1) were manufactured by a hot sonication method. Briefly, lutein was first dissolved in diethylene glycol monoethyl ether (Transcutol). Lutein in Transcutol (0.2 g in 19 g) was mixed with corn oil (1.8 g) and Span 20 (3.0 g) at 80°C in a dry bath. Pluronic F68 (2.0 g) and CD (2.0 g) was dispersed in the water (72 g). Finally, the mixture solution was sonicated for 1.5 min using a probe-type sonicator. Samples were stored in the dark for further characterization. Bare NE (no CD added) and hydroxyethyl (HE)βCD (2%) suspension were also made by the same procedure. The aqueous lutein suspension was prepared by sonicating the mixture of lutein (0.2%), corn oil (0.8%), transcutol (19%) and water (80%). Composition of the developed vehicles such as NE and NE+HEβCD is shown in Table 1.
| Composition (%) | Vehicles | |||
|---|---|---|---|---|
| NE | NE+HEβCD | HEβCD suspension | Aqueous lutein suspension | |
| Lutein/Transcutol | 0.2/19 | 0.2/19 | 0.2/19 | 0.2/19 |
| Corn oil | 1.8 | 1.8 | 0.8 | 0.8 |
| Pluronic F68 | 2 | 2 | — | — |
| Span 20 | 3 | 3 | — | — |
| HEβCD | — | 2 | 2 | — |
| Water | 74 | 72 | 78 | 80 |
| Characteristics | ||||
| Mean (nm) | 97.7±2.3 | 91.7±0.8 | 1132.4±183.7 | 1114.0±74.9 |
| Zeta potential (mV) | −33.2±0.6 | −33.1±0.1 | −26.8±0.7 | −21.7±1.5 |
| Entrapment efficiency (%) | 85.7±3.5 | 95.1±1.4 | 26.83±0.67 | 12.3±0.19 |
| Partition coefficient (4 h, 10−2) | 15.83±0.38 | 23.81±1.20 | 12.57±1.14 | 2.58±0.66 |
a) The lutein content in all vehicles is 2000 µg/g. The results represents average ±S.D. (n=3).
The size and surface property of NE were characterized by transmission electron microscopy (TEM) and dynamic light scattering. A drop of diluted sample was dispersed onto a 100-mesh copper grid, and then the excess drop was removed with a filter paper. The sample containing copper grid (CF200-Cu, Electron Microscopy Science, Washington, DC, U.S.A.) dried for 2 h at 55°C prior to microstructure analysis. The morphology of the various NE was observed by TEM (JEOL JEM 2000 EXII, Tokyo, Japan). NE samples were diluted 1 : 25 with Milli-Q water and dried on carbon film (CF200-Cu, Electron Microscopy Science, Washington, PA, U.S.A.) for 12 h. After stained with a 1% phosphotungstic acid (Merck, Darmstadt, Germany) for 30 s and vacuum dried in the incubator, NE samples were then analyzed by TEM. TEM technique is a well-documented method for microstructure analysis of lipid nanoparticles. In the TEM measurement, an electron bean may melt the lipid nanoparticles and cause uncertainty in the size distribution. The size distribution of lipid nanoparticles are often analyzed by using TEM methods as indicated in the several references.28,29) TEM still remains a reasonable technique to measure the microstructure of lipid nanoparticles. The average particle size and zeta potential in different formulations were characterized by using Zetasizer Nano ZS 90 (Malvern, Worcestershire, U.K.) at a fixed angle of 90° and a temperature of 25°C. Zeta potential is the surface charge of particles, which is an indicator of the long-term stability. Since the lutein has the strong absorbance, the measurement of size distribution and zeta potential of lutein-laden vehicles needs dilution by using water in order to obtain the correct analysis. Two dilution ratios including 100 and 1000-fold have been evaluated in this study and no significant changes in the size distribution were found. Therefore, lutein laden vehicles are diluted with water to a 100-fold dilution before the analysis of size distribution and zeta potential.
Lutein Accumulation and Partition Coefficient in VitroTo investigate possible drug penetration or retention in the scleral tissues, the following experiments were performed to measure partition coefficient and accumulation rate of lutein in the sclera. The tendency of lutein into scleral tissue was estimated by measuring its partition coefficient between porcine sclera and the formulation at 32°C, the body surface temperature. Lutein dissolved in the vehicle for the partition experiments is fixed at a concentration of 2000 µg g−1. The lutein-loaded vehicle solution (2 mL) was mixed with the porcine sclera (18–22 mg per piece) in a 10 mL vial. The vehicle and the ocular tissues were incubated at the shaker with 200 rpm agitation for 4 h. Three pieces of sclera were randomly withdrawn from the vial, rinsed with phosphate buffer saline, wiped with paper and weighed. The ocular tissue was then homogenized and the lutein accumulated in tissue was extracted by using 2 mL of the extraction buffer (10% tetrahydrofuran and 90% methanol). The homogenized solution was centrifuged at 10000 rpm, filtered by a nylon filter (0.22 µm pore size) and analyzed by HPLC to determine total lutein content (Wl, µg) in the extraction buffer. Each experiment was replicated at least four times. The lutein concentration [Cl] in the sclera after the incubation period was calculated as previously described with some modification30):
 |
Finally, the partition coefficient (K) was calculated using the equation:
 |
The prepared nanoemulsions were separated from the free lutein using Sephadex G-50 (GE Healthcare) resin and a plastic column for the measurement of entrapment efficiency. The lutein-loaded vehicle of 0.1 mL was separated by using 2 mL of resin. The part of the outflow with opalescence and metered volume to 6 mL were collected and measured by HPLC. The entrapment efficiency (EE) of lutein in the nanoemulsions was calculated according to the following equations31):
 |
Porcine eye-balls are freshly collected and stored in ice during transport. The sclera is dissected from porcine eye within 24 h after slaughter, wrapped in wetted tissue paper and stored at −80°C in a polypropylene bag and used within one month.17) Ocular tissue was fixed in phosphate buffered saline (PBS) solution containing 10% formalin and cut vertically, dehydrated using ethanol, embedded in paraffin, and stained with hematoxylin and eosin (H&E) staining. These samples were then observed under light microscope (Olympus BX51, Tokyo, Japan) using 10 magnification. For penetration analysis by confocal laser scanning microscopy, the ocular tissue was removed from the lutein vehicles, rinsed with PBS and then the surface of the sclera was wiped gently. The ocular tissue was directly sandwiched between a glass slide and a cover slip, and examined confocal microscopy without additional tissue processing. Leica TCS SP2 confocal laser scanning microscope (Leica, Heerbrugg, Switzerland) was used in this study. Lutein fluorescence emitted at 515 nm was recorded when excited at a wavelength of 488 nm by means of an argon laser. The sample was scanned from the tissue surface (0 µm) to a depth of 100 µm at a 15 µm interval.
Cell Culture and Cytotoxicity of NanoemulsionsRetinal ganglion cells (RGC) were kindly supplied by the Department of Ophthalmology at Chang Gung Memorial Hospital (Taoyuan, Taiwan). RGC cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) in a humidified atmosphere of 5% CO2 at 37°C. Cells were subcultured using trypsin–ethylenediaminetetraacetic (EDTA) when reaching 80% confluence. For cytotoxicity experiments, RGC cells were seeded on 48 well plates at a density of 5×104 cells/well and allowed to grow for 48 h. NEs were diluted to a series of concentration using DMEM medium. Cells were fixed by 0.2 mL of 4% formaldehyde, the nucleus stained by Hoechst 33342, then analyzed by INCell Analyzer 1000 (GE Healthcare, Piscataway, NJ, U.S.A.). The fluorescence at 455 nm in the nucleus stained by Hoechst 33342 was recorded after excitation at 350 nm. The cell viability is defined as follows: (the number of nuclei in treated cells/the number of nuclei in controlled cells)×100%.
Statistical AnalysisStatistical analysis of differences between different vehicles is performed by using the ANOVA and the Bonferroni post hoc test. The tests were performed using statistical software (SPSS v. 13.0; SPSS Inc., Chicago, IL, U.S.A.), and data were expressed as the mean±S.D. A 0.05 level of probability was taken as the level of significance.
In order to evaluate the scleral lutein accumulation by CD and NE, the retention ability of lutein in the sclera was examined in vitro. The accumulation rate and partition coefficient of drug in targeted tissue are important indicators for the evaluation of drug retention ability. We investigated the effects of combination of various 0.5% CDs with NEs on lutein accumulation in the scleral tissue during 4 h incubation. As shown in Table 2, the formulations were ranked in terms of scleral lutein accumulation as NE+αCD<NE+methylβCD<NE<NE+βCD<NE+HEβCD. The promoting effect of the CDs modified NE for lutein accumulation in the sclera was confirmed. Additionally HEβCD in NE showed the maximal elevation in lutein accumulation. The partition coefficient of lutein in the sclera using four kinds of CD containing vehicles was ranked in the following order: αCD< methylβCD<βCD<HEβCD. The partition coefficient was measured by assuming homogenous sclera and the passive diffusion model between the vehicles and the porcine sclera. The trend for the partition coefficient of lutein in the porcine sclera was similar to that of lutein accumulation. The partition coefficient of NE+HEβCD was significantly higher (p<0.05) than that of NE alone, whereas no significant difference (p>0.05) existed between the two vehicles, NE+αCD and NE+methyl-βCD. The average lutein accumulation rate in sclera for NE+0.5%HEβCD and NE was 97.4±4.3 and 79.2±1.9 µg g−1 h−1, respectively. NE combined with 0.5% HEβCD could elevate 1.2-flod lutein accumulation as compared with that of NE alone. Besides, the addition of CD in the NE at the concentration of 0.5% did not affect the stability of formulation. CD-based formulations have been applied to deliver various drugs into the eye.33) Among the tested CDs, HEβCD most effectively enhance the lutein accumulation in the porcine sclera. We also investigated whether the lutein accumulation was affected by the amount of HEβCD in NE carriers. The dose effect of HEβCD in NE on the scleral lutein accumulation is shown in Table 3. The sclera could accommodate lutein after it immersed in the vehicles during 4 h incubation. The concentration of HEβCD in the NE up to 2% could enhance the lutein accumulation in the porcine sclera. However, more than 3% HEβCD in NEs could not further increase the lutein accumulation. The instability of NE+3% HEβCD might account for the reducing accumulation since the phase separation was observed in this vehicle. Among the tested HEβCD concentration, 2% HEβCD in NE carrier could significantly enhance the lutein accumulation in the porcine sclera. The average lutein accumulation rate in sclera for NE+HEβCD was 119.0±6.0 µg g−1 h−1, indicating 1.5-flod lutein accumulation as compared with that of NE alone (Tables 2, 3). The dose effect of lutein (1000–8000 µg g−1) in the NE+HEβCD vehicle on scleral lutein accumulation is depicted in Fig. 1. The amount of lutein permeated into the pig sclera significantly elevated when lutein in the vehicles increased (p<0.05). The increase of the lutein concentration in the vehicles could improve the driving force needed for the diffusion of lutein into the sclera. Fick’s first law has postulated that diffusion flux goes from regions of high concentration to regions of low concentration with a magnitude proportional to the concentration gradient.34) As shown in Table 3, the partition coefficient and the accumulation rate of NE+HEβCD (8000 µg g−1 lutein) was significantly higher (p<0.05) than those of 2000 µg g−1 lutein. Similarly, the lutein accumulation rate in sclera for 8000 and 1000 µg g−1 lutein in NE+HEβCD was 205.1±8.0 and 45.9±3.3 µg g−1 h−1, respectively. The increase of lutein concentration also increased the lutein accumulation in the porcine sclera as indicated in Table 3. NE+HEβCD loaded with 8000 µg g−1 lutein could elevate 4.5-flod lutein accumulation as compared with that of NE loaded with 1000 µg g−1 lutein. In contrast, the partition coefficient increased with increasing lutein concentration to 2000 µg/mL and then decreased with further increase to 4000 or 8000 µg/mL (Fig. 1). The vehicles loaded 4000 or 8000 µg g−1 lutein were not stable after 3 weeks storage. The lutein payload in the vehicles was maintained at 2000 µg g−1 in the following sections. Since the sclera is a homogeneous tissue, several research groups adapt the similar procedure to measure the partition coefficient.30,35) The measurement of the partition coefficient assumes homogeneity in a tissue, which is actually heterogeneous. As with many experiments in biological tissues or organs, this problem can never be fully resolved. However, it remains reasonable to measure the partition coefficient by using this techniques reported in the aforementioned literature.
| Vehicles | Accumulation rate (µg g−1 h−1) | Partition coefficient (×10−2) | Enhancement (%) |
|---|---|---|---|
| NE | 79.15±1.88 | 15.83±0.38 | 100.00 |
| NE+0.5% α-CD | 59.56±7.99* | 11.91±1.60* | 75.25 |
| NE+0.5% β-CD | 80.49±10.58 | 16.10±2.12 | 101.70 |
| NE+0.5% M-β-CD | 60.16±4.36* | 12.03±0.87* | 76.00 |
| NE+0.5% HEβCD | 97.42±4.28* | 19.48±0.86* | 123.09 |
a) The CD and lutein content in the tested vehicles is 0.5% and 2000 µg g−1, respectively. The results represents average ±S.D. (n=3). Sclera accumulation rate of NE is set as the comparison basis (100%). * p<0.05 as compared with the values of the accumulation rate and partition coefficient in control NE (79.15±1.88 µg g−1 h−1, 15.83±0.38).
| Formulation | Lutein (µg g−1) | Accumulation rate (µg g−1 h−1) | Partition coefficient (×10−2) | Enhancement (%) |
|---|---|---|---|---|
| Aqueous solution | 2000 | 10.57±1.59* | 2.58±0.32* | 13.36 |
| NE+1% HEβCD | 2000 | 93.89±9.40 | 18.78±1.88* | 118.62 |
| NE+3% HEβCD | 2000 | 106.97±5.31* | 21.39±1.06* | 135.15 |
| NE+2% HEβCD | 1000 | 45.92±3.29* | 18.37±1.31* | 58.87 |
| NE+2% HEβCD | 2000 | 119.03±6.01* | 23.81±1.20* | 150.39 |
| NE+2% HEβCD | 4000 | 154.87±8.39* | 15.49±0.84 | 195.67 |
| NE+2% HEβCD | 8000 | 205.11±8.03* | 10.26±0.40* | 259.14 |
a) The results represents average ±S.D. (n=3). Sclera accumulation rate and partition coefficient of NE (Table 1) is set as the comparison basis (100%). * p<0.05 as compared with the values of the accumulation rate and partition coefficient in control NE (79.15±1.88 µg g−1 h−1, 15.83±0.38).
The error bar stands for standard deviation (n=3).
In order to understand the mechanism of the CD’s enhancing effect, the characterization of the four vehicles was performed. The compositions of the four vehicles including NE, NE+HEβCD, HEβCD suspension and aqueous lutein suspension are shown in Table 1. Corn oil was used as a lipid core for lutein payload in the NE. Our NE system is stabilized by using the two surfactants Pluronic F68 with a hydrophile–lipophile balance (HLB) of 24 and Span 20 with an HLB of 8.6. A combination of hydrophilic and lipophilic emulsifiers has been proved to maintain a stable dispersion by imparting more rigidity and strength to the binary-surfactant film.36) In addition, surfactants can reduce the colloidal size by reducing the surface tension and fluidizing the interfacial droplet film. The HLB of surfactant blends (F68/Span 20) is 14.8 by calculating the weight percentages of surfactants and their HLB values. The droplet sizes in the surfactant-free suspensions, i.e., HEβCD suspension and aqueous lutein suspension, were significantly larger than those of the NE and NE+HEβCD. Phase separation in these suspensions was quickly observed after the several-hours storage. The lutein entrapment efficiency in the two suspensions was significantly lower than that of the NE or NE+HeβCD (p<0.05) as demonstrated in Table 1. The lutein partition coefficients for the four vehicles were ranked as aqueous lutein suspension<HEβCD suspension<NE <NE+HEβCD. The detailed mechanism is discussed in the next section. The vehicle effect on scleral lutein accumulation is demonstrated in Fig. 2. The lutein-loaded vehicles including NE, NE+HEβCD, HEβCD suspension and aqueous lutein suspension had significant difference in scleral lutein accumulation. When lutein suspended in the water, it would form oil droplets and have little accumulation in the porcine sclera. In contrast, the lutein embedded in the HEβCD vehicle could enhance the lutein accumulation in the sclera. The partition coefficient of lutein in vehicle of NE+HEβCD was 1.9-fold higher than that of HEβCD suspension and 1.5-fold higher than that of NE. This result clearly indicated the combination of NE and HEβCD on scleral lutein accumulation. As demonstrated in Table 1, the zeta potentials of these vehicles were around −21 to −33 mV. The addition of HEβCD into NE would change the zeta potential on the surface of NE. The size distribution of the NE was smaller than that of NE+HEβCD. TEM photographs provided some information for the size change. CD (small solid, arrow in Fig. 3B) around the NEs was observed in the vehicle of NE+HEβCD as shown in the TEM. The adhesion of CDs on the surface of NEs resulted in the increase of particle size. The morphology of lutein-loaded NEs and NE+HEβCD could be clearly observed using TEM. In the TEM image, the mean size of the NEs and NE+HEβCD was 97 and 91 nm, respectively, which was consistent with the size measured by photon correlation spectroscopy. The droplets of NE and NE+HEβCD prepared in this study had spherical shape as indicated in the TEM photographs (Figs. 3A, B). The entrapment efficiency for NE and NE+HEβCD was 85 and 95%, respectively. Formulation stability during storage is an important indicator in formulation development.37) The effect of storage duration on the size and zeta potential of lutein-loaded vehicles was studied at the room temperature for four weeks. As indicated in Fig. 4, NE could maintain their initial size and zeta potential during 4 weeks period. NE with 2% HEβCD could maintain their initial size for 3 weeks. Significant fluctuation of size distribution was observed for the aqueous and HEβCD suspensions during the storage. The size of the CD suspension and aqueous lutein suspension was around 1100 nm which was significantly large than that of NE vehicles. Absence of surfactants in these suspension accounted for the sedimentation and large aggregation. The zeta potentials for the NE and NE+HEβCD were around −33 mV which could maintain the particles in suspension by repelling each other. In contrast, the zeta potential of aqueous lutein suspension and CD suspension was −21 and −26 mV, respectively. Electrostatic force was insufficient to maintain the droplets in the suspended state and cannot avoid their aggregation. No lutein precipitation in the vehicles of NE and NE+HEβCD when stored at a temperature at 25°C for 21 d. In CD suspension or aqueous lutein suspension, lutein sediment and phase separation was quickly observed. The lutein accumulation in the porcine sclera is evidenced by using the confocal laser scanning microscopy (CLSM) in the next section.
The lutein content in all vehicles is 2000 µg g−1. The error bar stands for standard deviation (n=3).
Scale bars in (A) and (B) represent 100 and 50 nm, respectively. The size distributions of two NEs were 100 and 110 nm, respectively. The arrows in Fig. 3B indicate the HeβCD.
The error bar stands for standard deviation (n=3).
CLSM provides the localization and permeation pathway of fluorescent compounds in the porcine sclera without embedding procedures. Lutein has a light emission of 515 nm after being excited by 488-nm laser light. This fluorescent characteristic of lutein in porcine sclera for NE, NE+HEβCD, HEβCD suspension and aqueous lutein suspension was indicated in Fig. 5. The lutein located at various sclera depths (from 0 to 106 µm) were examined using CLSM. A depth of 0 µm represents the outside layer, whereas a depth of 106 µm represents the inside layer. Lutein in NE+HEβCD could penetrate the most depth of 93 µm into the sclera and could have the most fluorescence as compared with other vehicles. The fluorescence intensities of lutein in the sclera were ranked in the order of NE+HEβCD>NE>HEβCD suspension=aqueous lutein suspension. These confocal results are consistent with scleral lutein accumulation using 4 vehicles in Fig. 2. Lutein in NE and NE+HEβCD had the fluorescence intensity of 139 and 81 arbitrary unit (AU) respectively at the top layer. The vehicle of NE+HEβCD exhibited the largest lutein accumulation which provided a driving force for deeper lutein penetration. Very slight lutein penetration from the HEβCD suspension and the aqueous lutein suspension was observed at 16 µm depth. In contrast, lutein could penetrate to the depth of 74 µm in the sclera with the help of NE vehicle. The HEβCD or aqueous lutein suspension only had the fluorescence intensity of 20 AU at the outermost epithelium.
Lutein fluorescence emitted at 515 nm was recorded when excited at a wavelength of 488 nm by means of an argon laser. The sample was scanned from the tissue surface (0 µm) to a depth of 352 µm at a 29.3 µm interval.
The histological examination of the porcine sclera by the NE+HEβCD was analyzed using HE staining in order to understand the impact of vehicles on the regular tissue. Porcine sclera was treated with NE+HEβCD and control (PBS) sclera for 24 h before HE staining. Since the sclera is composed of type I collagen and glycosaminoglycan,38) the sclera was stained as red in the Fig. 6. Micrographs of control and vehicle-treated sclera demonstrated normal histology as shown in Figs. 6a–e. The sclera of NE or HeβCD containing NE was similar to that of PBS treated sclera. Some loose space in the sclera was found which might contribute to the enhanced accumulation of lutein delivery using the NE+HEβCD vehicle.
Scale bar represents 100 µm. Porcine sclera after 24-h treatment was fixed, HE stained and observed using light microscopy.
The bare vehicles and drug loaded vehicles containing 2000 µg g−1 lutein were used for the cytotoxicity experiments for RGC cells. The dilution range of vehicles is 0.2–2.5% using DMEM as the diluent. Similar cell viability was observed for the bare and lutein-loaded vehicles as indicated in Fig. 7. Lutein payload would not increase the toxicity as compared with the bare vehicles. Their viability value for RGC cells at the dose of 1.25% of bare NE+HEβCD and NE was 51.2 and 36.7%, respectively. Cytotoxic effects were alleviated when NE+HEβCD in the drug-free or lutein loaded conditions. It has been noted that the toxicity of drug was decreased when CD incorporated in the ocular formulation. In a previous study, remarkable cytotoxic effects on macrophage cells were found at 0.1% addition of bare nanoemulsion.39) The nanoemulsions formulated with the stearic acid were cytotoxic at the 1% concentration for mouse J774 macrophages, 3T3 fibroblasts and human HaCaT keratinocytes.40) Very limited information is available for the toxicity effects of nanoemulsion on ocular cell lines. Our results of the cytotoxicity test on RGC cells indicated the safety of CD modified lipid vehicles.
The error bar stands for standard deviation (n=3).
CD-based vehicles are already applied to deliver ophthalmic drugs.33,41) However, the studies combining CD and nanoemulsions for ophthalmic lutein formulation are limited. The addition of HEβCD in NE was found to enhance the lutein encapsulation and the scleral lutein partition (Table 2). Effect on scleral lutein accumulation and partition by using NE+HEβCD (Fig. 2, Table 2) and the fluorescence data of scleral lutein intensities (Fig. 5) indicate that the amount of scleral lutein accumulation using the NE+HEβCD was significantly higher than that of NE or HeβCD alone. Also the amount of scleral lutein accumulation using the NE was significant higher than that of HeβCD suspension (Fig. 2). Several mechanisms contributing to the increased scleral delivery by NE include the enhancement of lutein solubility, the large surface of nanoemulsions and the occlusive effect. The first reason was that lutein can readily be dissolved in NE but not in water. Hence the higher drug loading in the NEs increased the concentration gradient towards the sclera. Secondly, the NEs had more scleral lutein accumulation because the small particle size ensures close contact and better adhesion on the sclera to deliver the lutein. However, the transport of intact droplets into the scleral is difficult since the barrier effects of the scleral epithelium.42) Thirdly, the enhanced scleral lutein by the vehicle NE might be due to the surfactants in the formulation. Surfactants, which can loosen or fluidize the lipid bilayers on the scleral epithelium, can act as permeation enhancers.43)
Cyclodextrins (CDs) are a group of cyclic oligosaccharides capable of forming inclusion complexes with hydrophobic drugs. The advantages of cyclodextrins in the ocular delivery include enhanced solubility, enhanced permeation, enhanced stability, decreased irritation and controlled release.23) In ophthalmic preparations, co-administration of CDs has been reported to increase corneal penetration, ocular absorption and the efficacy of poorly water-soluble drugs. CDs are also effective in decreasing drug irritation as they mask the irritating effects of the drugs by forming inclusion complexes. The modified CDs result from the introduction of other functional groups at the 2-, 3- and 6-hydroxyl groups of the glucose residues. The solubility can be improved through two mechanisms as by breaking the 2- and 3-OH hydrogen bonds and by preventing crystallization due to creation of a new material which is made up of many isomeric components and gives rise to an amorphous product.44) Hydroxyalkylation on βCD moieties can not only convert the parent βCD into its amorphous and non-crystallizable derivatives but also can reduce hemolytic and renal toxicities of parent CDs.45) Lipid nanoparticles modified by CDs have been reported to modulate the biodistribution and to enhance the bio-availability of administered drugs.46) Paclitaxel loaded solid lipid nanoparticles coated with HPβCD can enhance cellular accumulation of the drug into p-glycoprotein expressing cells.47) Nanostructured lipid carrier modified by cysteine–polyethylene glycol stearate has the mucoadhesive properties on the surface of rabbit eyes.48) HPβCD combined with the lipid nanoparticles can improve the lutein loading efficiency and the partition of lutein into the porcine cornea.49) These reports suggest CDs modified lipid vehicles as a promising tool for ocular drug delivery. The proposed mechanism of HEβCD modified NE in the lutein accumulation includes the following possibility. HEβCD could provide extraordinary space to accommodate lutein. Lutein would firstly release from the vehicles, then would be partitioned into or absorbed by the sclera. Lutein loaded in HEβCD was more easily released as compared to that incorporated in the fat matrix of NEs. Additionally, CD could extract cholesterol, phospholipids and proteins from the sclera and can reduce the barrier effect of the scleral epithelium.18)
Every CD has its own characteristics including water solubility, lipophilic internal cavity and ability to form guest-host inclusion complexes. The guest-host inclusion complexes render the hydrophobic compounds more soluble. However, the cavity of cyclodextrin determines the molecules that will be able to form inclusion complexes. The diameters of the hydrophobic cavity in α- and β-cyclodextrin are 4.7–13.7 Å and 7.8–15.3 Å, respectively.50,51) The size of lutein molecule was determined to be 1.58 Å (diameter)×31.29 Å (length) in a rod shape by using Jmol software. Based on the aforementioned information, the lutein may be too loose to completely accommodate within α-, β-cyclodextrin or HEβCD cavity. Lutein solubilization by using CDs may contribute to the ability of forming non-inclusion based complexes and aggregates. Another possibility is that HEβCD can stabilize emulsion systems by complexation of fatty acid residues of the oil phase. The concept of using CDs as emulsifying agents in multiphase systems was first introduced by Shimada.52) Very recently Klang et al. have tested the feasibility of using CDs as additional stabilizing agents in lecithin-based nanoemulsions for transdermal delivery.21,22) These reports have shown that CDs may have a stabilizing effect on nanoemulsions including stabilizing zeta potential values. It is most likely that the CDs are incorporated into the interfacial film around the oil droplets. The mixed interfacial films are more flexible and thus more suitable to form stable droplets.21,22) The proposed mechanism for HEβCD that enhancing the scleral accumulation of lutien was due to the enhanced stability on the nanoemulsions. Nevertheless further investigation shall be performed to confirm the speculation.
The localization and permeation pathway of lutein in the sclera was confirmed by using the CLSM technique. Our CLSM results (Fig. 5) showed the penetration profiles of lutein within the sclera tissue were vehicle dependent. The confocal lutein intensity indicated that the lutein accumulation by NE+HEβCD in the sclera was much higher than that of NE alone. Since the lutein partition into the sclera is the first step for scleral accumulation, the improved scleral accumulation in the presence of CD may be due to increased partitioning of lutein-loaded NE in the sclera. The increased lutein accumulation was observed again when HEβCD combined with NE as the ocular delivery vehicles by the CLSM results. In fact, the scleral lutein accumulation is controlled by the factors like diffusivity coefficient, path length and partition coefficient in the scleral tissue. Type of cyclodextrins and lutein concentration in the NEs impacted on the partition coefficient and sequentially affect the lutein accumulation (Table 2).
In conclusion, we systematically characterize the nanoemulsions, examine their partition coefficients in the porcine sclera, analyze the penetration depth of lutein, and determine their cytotoxicity. Among the tested enhancers, HEβCD most effectively enhance the lutein accumulation in the porcine sclera. A combination of nanoemulsion and 2% HEβCD could further enhance the lutein partition up to 23.8±1.2 (µg g−1 h−1) which was 9.2-fold higher than that of the lutein suspension. The modified nanoemulsions are potent to modulate the distribution of the lutein in the sclera as proven by CLSM. From the confocal images, this improvement may be due to the increased partitioning of lutein to the scleral surface by CD modified nanoemulsion. Our results of the cytotoxicity test on RGC cells indicated the safety of CD modified lipid vehicles. However, in vivo lutein accumulation in the macula by using this hybrid nanoemulsion merits further investigation.
The project was kindly supported by Chang Gung Memorial Hospital (CMRPD 2A0102, 1D0091), Taiwan. The authors appreciate the financial support from the Ministry of Science and Technology, ROC (MOST 102–2221-E-182–075, 103–2221-E-182–010).
The authors declare no conflict of interest.
