Terpene Composited Lipid Nanoparticles for Enhanced Dermal Delivery of All-trans-Retinoic Acids

Abstract

In the present study, terpene composited lipid nanoparticles and lipid nanoparticles were developed and evaluated for dermal delivery of all-trans-retinoic acids (ATRA). Terpene composited lipid nanoparticles and lipid nanoparticles were investigated for size, size distribution, zeta potential, entrapment efficiency, photostability, and cytotoxicity. In vitro skin permeation of ATRA lipid formulations were also evaluated. To explore the ability of lipid nanocarriers to target the skin, the distribution of rhodamine B base in the skin was investigated using confocal laser scanning microscopy (CLSM). The results indicated that the physicochemical characteristics of terpene composited lipid nanoparticles influenced skin permeability. All lipid nanocarriers significantly protected ATRA from photodegradation and were non-toxic to normal human foreskin fibroblast cells in vitro. Solid lipid nanoparticles containing 10% limonene (10% L-SLN) had the highest ATRA skin permeability. Terpene composited SLN and nanostructured lipid carriers (NLC) showed higher epidermal permeation of rhodamine B across the skin based on CLSM image analysis. Our study suggests that terpene composited SLN and NLC can be potentially used as dermal drug delivery carriers for ATRA.

Transdermal drug delivery has been chosen as a feasible alternative route of drug delivery due to its various advantages over conventional oral and intravenous routes such as reduction of drug metabolism via first pass effect, minimization of pain, and possible controlled drug release.^1,2) However, the effectiveness of transdermal drug delivery depends on the capability of drugs to penetrate across the skin in sufficient amounts to reach therapeutic levels.³⁾ The stratum corneum is an important barrier of the skin for drug absorption.^4,5) To facilitate drug delivery through the skin, penetration enhancers, which ideally cause a temporary reversible reduction in the barrier function of the stratum corneum, are extensively used to increase percutaneous absorption.⁶⁾

Terpenes are a series of naturally occurring compounds consisting of isoprene (C₅H₈) units. They have been used in transdermal research since 1960s as skin permeation enhancers. They are reported to be a very safe and are an effective class of penetration enhancers that has been classified by the Food and Drug Administration (FDA) as generally regarded as safe (GRAS).⁷⁾ Limonene is a hydrocarbon lipophilic terpene obtained from the lemon peel of citrus lemon.⁸⁾ Previous studies have demonstrated that permeability enhancement by limonene can occur through multiple possible mechanisms, which may have contributed to the enhanced permeability of ketoprofen.⁹⁾ 1,8-Cineole, a terpene, has also been used to promote percutaneous absorption of several lipophilic drugs through hairless mouse skin^10,11) and was recently reported to have an enhancing effect on percutaneous Zidovudine (AZT) absorption across rat skin.¹²⁾ The mechanism of percutaneous permeation enhancement involves increasing the solubility of drugs in skin lipids, disruption of lipid/protein organization, and/or extraction of skin micro-constituents that are responsible for maintenance of barrier status. Hence, terpenes appear to offer great promise for use in transdermal formulations.¹³⁾

Solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC), which are the first and the second generation of lipid nanoparticles, respectively, are improved compared to nanoemulsions (NE). Lipid nanoparticles (SLN, NLC and NE) were chosen as the transdermal dosage form because of their promising novel dosage form and suitability for efficient delivery of active ingredients through the skin. The large surface area of these lipid nanoparticle systems allows rapid penetration of active molecules.^14,15) The lipid matrices of SLN are composed of solid lipids only, whereas NLC are composed of both solid and liquid lipids. In contrast with NE, both NLC and SLN are in the solid state at room and body temperatures. Moreover, the degradation of ascorbyl palmitate loaded in SLN and NLC showed lower degradation than NE.¹⁶⁾ A number of drugs can be used as a model drug for transdermal drug delivery; however, liphophilic drugs typically pose stability problems and thus, lipophilic drugs are good model drugs, which represent poor water solubility and/or photo instable drugs.

When applied topically, all-trans-retinoic acids (ATRA) have demonstrated efficacy in keratinization disorders and in the treatment of other cutaneous lesions.¹⁷⁾ Nevertheless, ATRA is sensitive to decomposition by light or high temperature and possesses poor aqueous solubility.¹⁸⁾ Accordingly, a desirable drug delivery system, such as lipid nanoparticles (SLN, NLC and NE), should be used to solve these problems. Moreover, little knowledge is available regarding the incorporation of terpenes in various lipid nanoparticles and the comparison for their effectiveness to other lipid nanoparticles as skin delivery carriers.

The objective of this study was to enhance the skin permeability of ATRA using lipid nanoparticles incorporated with terpenes as a permeation enhancer and to compare their effectiveness with other lipid nanoparticles as skin delivery carriers. Particle size, size distribution, zeta potential, entrapment efficiency, photo-stability, cytotoxicity, and in vitro skin permeability of these formulations were evaluated.

MATERIALS AND METHODS

Materials

D-Limonene (L), 1,8-cineole (C), rhodamine B base, ATRA and butylated hydroxytoluene (BHT) were purchased from Sigma (St. Louis, MO, U.S.A.). Cetyl palmitate (CP) was purchased from SABO SpA (Levate, Italy). Oleic acid (O) was purchased from Fluka Chemie AG (Seelze, Germany). Medium chain triglycerides (M) was obtained from Uniqema Asia Pacific (Kuala Lumpur, Malaysia). Polysorbate-80 (T80) was purchased from the NOF Corporation (Osaka, Japan). Diethylene glycol monoethyl ether (transcutol P) and polysorbate-20 (T20) was purchased from Ajax Finechem, (Auckland, New Zealand). Normal human foreskin fibroblast (NHF) cells were obtained from the American-Type Culture Collection (ATCC) (Rockville, MD, U.S.A.). Dimethyl sulfoxide (DMSO) was obtained from BDH Laboratories, UK. Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), trypsin–ethylenediaminetetraacetic acid (EDTA), and penicillin–streptomycin were purchased from Gibco BRL (Rockville, MD, U.S.A.). All other reagents and solvents were commercially available and were of analytical grade.

Preparation of ATRA Loaded Lipid Nanoparticles

Lipid nanocarriers loaded with ATRA were prepared using a de-novo emulsification method. Lipid nanocarriers were reduced to the nanometer size range using an ultrasonicator (Sonics & Materials, CT, U.S.A.). The oil phase was composed of ATRA (0.3%), lipid matrix core (SLN: CP, NLC: CP and O and NE: O and M), L, C, transcutol P, and BHT. The aqueous phase was composed of T80, T20, and distilled water (Table 1). The oil and aqueous phases were heated to 65°C. The aqueous phase was then added into the oil phase and subsequently stirred at 1400 RPM for 5 min using a magnetic stirrer. Finally, the mixture was sonicated for 15 min using a probe type ultrasonicator (Sonics & Materials, CT, U.S.A.). The pH of the formulations was adjusted to 7±0.1 with 1 N NaOH and then filtered through a 0.45 µm membrane filter to remove precipitates. The formulations were kept in light-protected and sealed containers at 4°C in the refrigerator.

Table 1. Compositions of ATRA-Loaded Lipid Nanoparticle Formulations

Composition (%)	Formulation
	NLC	5% L-NLC	10% C-NLC	10% L-NLC	10% L-SLN	10% L-NE
All-trans-retinoic acid	0.3	0.3	0.3	0.3	0.3	0.3
Medium chain triglycerides						20
Oleic acids	10	10	10	10		10
Cetyl palmitate	20	20	20	20	30
1,8-Cineole			10
D-Limonene		5		10	10	10
Transcutol P	1.2	1.2	1.2	1.2	1.2	1.2
Butylated hydroxytoluene	0.002	0.002	0.002	0.002	0.002	0.002
Tween 20	2	2	2	2	2	2
Tween 80	8	8	8	8	8	8
Water qs.	100	100	100	100	100	100

The ATRA-suspension was prepared by incorporating 0.3% of ATRA in water followed by stirring for over 24 h to ensure constant thermodynamic activity through the course of the skin permeation experiment.

Vesicles Size, Size Distribution, Zeta Potential and Morphology Evaluation

Droplet size, polydispersity index (PDI), and zeta potential of ATRA-loaded lipid nanoparticles were determined at 25°C by photon correlation spectroscopy (PCS) using a Zetasizer Nano ZS (Brookhaven, U.S.A.). Medium viscosity and refractive index values of 1.054 cP and 1.33, respectively, were used for the aqueous phase with a sample refractive index of 1.460 and an absorption value of 0.00 (similar to the particles named Intralipid® within the Malvern software). Prior to the measurement, 1 µL of lipid nanoparticles was diluted with 1 mL distilled water, which was filtered through a 0.45 µm membrane filter.

Transmission electron microscopy (TEM, JEM-1230, JEOL, Tokyo, Japan) was utilized to analyze particle shape, size, and occurrence of lipid nanoparticle aggregation. Prior to TEM observation, the samples were prepared by placing a drop of the lipid nanoparticles, which were diluted 50-fold with double-distilled water, onto a formvar-coated copper grid. Excess sample volume was removed using a sheet of filter paper. After drying, the sample was visualized under the TEM at 300 k and 800 k magnification. Microscopy was performed at 80 kV.

Determination of ATRA Incorporation Efficiency

Due to the poor water solubility of ATRA, unentrapped ATRA was present in a yellow crystal. After reducing the particle size by probe type ultrasonication, lipid nanoparticle formulations were then filtered through a 0.45-µm membrane filter to remove precipitated ATRA. To investigate the concentration of ATRA in lipid nanoparticles, 0.1 g of lipid nanoparticles were disrupted with 1.5 mL of isopropyl alcohol (IPA). The lipid nanoparticle/IPA solution was centrifuged at 14000 for 15 min. The supernatant was appropriately diluted with IPA. The amount of ATRA incorporated in the lipid nanoparticle formulation was directly determined by high-performance liquid chromatography (HPLC). The percent yield or efficiency of ATRA incorporated into the lipid nanoparticles and the ATRA content were calculated using the following equations:

  

(1)

  

(2)

where C_LN is the concentration of ATRA in the lipid nanoparticles, C_INT is the initial concentration of ATRA, A_LN is the amount of ATRA in the lipid nanoparticles, and A_L is the amount of lipid in the lipid nanoparticles.

Wide-Angle X-Ray Scattering (WAXS) Investigation

To investigate the crystallinity of ATRA, WAXS was performed using a Siemens D8 Advance X-ray generator (Bruker AG, Thailand) with a copper anode (CuKα radiation, 40 kV, 25 mA, k=0.15418 nm). The scattering intensities were detected by a proportional detector (Goniometer PW18120). Measurements were analyzed at 2θ from 5.0° to 50.0° Bragg’s equation was used to transform the data from scattering angle to the spacing of lipid chains.

In Vitro Skin Permeation Study

In vitro skin permeability was determined using the shed skin of the snake, Naja kaouthia. The shed snake skin was used as a model of the human stratum corneum layer because of its similarity to human skin with regard to lipid content and permeability.¹⁹⁾ The shed snake skin was used to compare the skin permeability of ATRA in lipid nanoparticle formulations. Skin permeation experiments were performed using a Franz diffusion cell with a water jacket connected to water bath at 32°C. Each compartment had a volume of 6.5 mL and an effective diffusion area of 2.20 cm². The donor compartment was filled with 1 mL of each formulation. Ethanol and acetate phosphate buffer (pH 3.4) in the ratio of 1 : 1 was selected as the acceptor medium to maintain a sink condition over the experimental duration. This receptor solution does not disturb skin permeability.²⁰⁾ The mixture was stirred with a magnetic Teflon stirrer at 500 rpm. A sample (0.5 mL) of the acceptor medium was removed at 2, 4, 8, 12 and 24 h and replaced with the same volume of medium to maintain a constant volume. The concentration of ATRA in samples was assayed by HPLC. The cumulative amount of ATRA was plotted against time and the pseudo-steady state flux was determined from the slope using linear regression analysis. All experiments were performed for a minimum of three times.

Confocal Laser Scanning Microscopy (CLSM) Measurement

CLSM was employed to elucidate the skin penetration pathway of both whole skin and skin cross sections using the co-localization technique. Porcine skin was utilized as a model membrane in the CLSM study because it is reported that the skin permeability of both hydrophilic and lipophilic permeants across porcine skin is similar to human skin. Abdominal neonatal porcine skin from animals that died during being born (intrapartum stillbirths) was used as a model membrane for this study. Newborn pigs (ca. 1 kg) were obtained from a local pig farm in Nakhon Pathom Province, Thailand. The subcutaneous fat was carefully removed using medical scissors and surgical blades. The dorsal skin thickness was approximately 0.6–0.7 mm. The skin samples were frozen at −20°C until use. Samples were thawed at room temperature using phosphate buffered saline (PBS) prior to experiments. The difference between the red color of the fluorescent compound Rhodamine B base and the blue color of the skin auto-fluorescence was used to visualize drug distribution on the treated porcine skin surface. The tissue was immediately investigated by CLSM after removing the tissue from the refrigerator. To obtain x–y plane or top-view images of the skin, a cover slip of 22×50 mm (MENZEL-GLÄSER®, Braunschweig, Germany) was used, and a reservoir was made using silicone glue spread around the rim of the cover slip. A piece of tissue was placed on the constructed cover slip with the stratum corneum facing up toward the microscope condenser, and then, an adequate amount of methyl salicylate was added to the constructed cover slip to serve as immersion oil. Visualization was performed using the 20× objective lens system of an inverted Zeiss LSM 510 META microscope (Carl Zeiss, Jena, Germany) equipped with a He–Ne 1 laser (excitation wavelength=543 nm, emission wavelength=580 nm) and a diode laser for rhodamine B base and skin autofluorescence, respectively (excitation wavelength=358 nm, emission wavelength=461 nm). To compare skin penetration depths and fluorescence intensity of entrapped rhodamine in the particles at nonfollicular region, skin regions were scanned using the 20× objective lens to obtain x–z plane images at which the laser could scan through the tissue. The fluorescence intensity was determined at the middle horizontal line of each image using Zeiss LSM 5 operating software. The mean fluorescence intensity of each image was plotted against different skin depths.

Stability and Photostability Study

ATRA-loaded lipid nanoparticles with limonene formulations were stored in light-protected, sealed containers at 4, 25 and 40°C for 30 d. Both chemical and physical stabilities of ATRA were evaluated. Particle size and zeta potential were determined on days 0, 7, 14 and 30, and at 4, 25 and 40°C. Chemical stability was determined by measuring the amount of ATRA remaining on days 7, 14 and 30 after initial preparation. The amount of ATRA at these time points was quantified by HPLC and the percentage of ATRA remaining was compared to the amount on day 0.

To investigate the ability for lipid nanocarriers to protect photodegradation over the time, lipid nanoparticles (1 mL) were kept in a tightly closed glass bottle and were irradiated by a UVA lamp up to 6 h, which was placed 15 cm above the samples. Samples were taken from each lipid nanocarrier at given time intervals (0.5, 1, 2, 4, 6 h) followed by dilution with isopropyl alcohol. The amount of intact ATRA was analyzed by HPLC. In comparison, an isopropyl alcohol solution of ATRA in the same concentration of lipid nanoparticles was also investigated using the same procedures performed on lipid nanoparticles.

Cytotoxicity Evaluation

The cytotoxicity of the nanocarriers was evaluated based on a procedure adapted from the ISO10993-5 standard test method (indirect contact). Lipid nanocarriers (with and without 0.3% ATRA) were diluted with serum-free medium (SFM; containing DMEM, 1% (v/v) l glutamine, 1% (v/v) lactalbumin, 1% (w/v) antibiotic and antimycotic formulation) to obtain varying concentrations (0.1, 1, 10, 100, 1000 and 10000 µg/mL). NHF cells were plated in 100 µL of DMEM supplemented with 10% FBS and were distributed at a density of 10000 cells/well in 96-well plates. Cells were grown under a humidified atmosphere (5% CO₂, 95% air, 37°C) until confluent (typically 48 h after plating). The cell medium was removed, and the tested extraction media with varying concentrations of lipid nanoparticles were then re-incubated for 24 h. After treatment, the tested extraction solutions were removed. Finally, the cells were incubated with 100 µL of an 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)-containing medium (1 mg/mL) for 4 h. The medium was removed, the cells were rinsed with PBS (pH 7.4), and the formazan crystals formed in the living cells were dissolved in 100 µL DMSO per well. Cell viability (%) was calculated based on the absorbance at 550 nm, which was determined using a microplate reader. The viability of the non-treated control cells was arbitrarily defined as 100%.

HPLC Analysis of ATRA

ATRA concentration was determined using an HPLC (Agilent Technology, U.S.A.). A C₁₈ reversed-phase column (Symmetry®, VertiSep™, Vertical, Thailand) with dimensions of 5 µm, 4.6×150 mm was used. The mobile phase was a mixture of acetonitrile–water–glacial acetic acid with the volume ratio of 84.5 : 15.0 : 0.5. A UV detector was set at 353 nm for ATRA detection. The injection volume was 20 µL and the flow rate was 1.5 mL/min at ambient temperature. Quantitative determination of the amount of ATRA in each sample was obtained from the calibration curve, which had good linearity (r²=0.99) in the range of 0.02–100 µg/mL.

Data Analysis

All experiment measurements were collected in triplicate. Values are expressed as the mean±standard deviation (S.D.). Statistical significance was evaluated using a one-way ANOVA followed by a least-significant difference post-hoc test. The significance level was set at a p value <0.05.

RESULTS AND DISCUSSION

Physicochemical Characteristics of ATRA Loaded Lipid Nanoparticles

In our previous study, ATRA-loaded SLN (30% CP) and NLC (10% CP and 20% O) consisting of 30% lipid as the oil phase and 1.2% lecithin as an emulsifier combined with 8% T80 as a co-emulsifier showed good chemical and physical stability at 4°C.²¹⁾ Therefore, these formulations were selected for this study. Transcutol P, T20 and T80 were used as emulsifier, and terpene oil (L or C) was used as the skin penetration enhancer. The compositions of ATRA-loaded lipid nanoparticles with terpene are shown in Table 1.

Particle size and zeta potential of NLC, 5% L-NLC, 10% C-NLC, 10% L-NLC, 10% L-SLN and 10% L-NE formulations are shown in Table 2. All lipid nanoparticles containing 0.3% ATRA with and without terpene formulations were in the nanometer range as the average size was less than 200 nm (139 to 160 nm) with a narrow PDI (0.12–0.15) and exhibited a negative zeta potential (−28 to −46 mV). Based on TEM imaging (Fig. 1), lipid nanoparticles with 10% terpene had nanometer-size spherical shapes; however, they were not perfectly round particles.

Table 2. Particle Size, Zeta Potential, Percent Yield, and ATRA Content of NLC, 5% L-NLC, 10% C-NLC, 10% L-NLC, 10% L-SLN and 10% L-NE Formulations (n=3)

Formulation	Size±S.D. (nm)	Zeta potential±S.D.	Yield (%)	ATRA content (mg/g)
NLC	160.87±0.75	−38.66±2.14	72.61±1.54	6.98±0.45
5% L-NLC	136.56±0.89	−39.05±3.14	81.58±3.25	7.84±0.22
10% C-NLC	138.23±0.71	−30.65±2.24	84.28±5.61	8.10±0.44
10% L-NLC	139.74±0.72	−39.93±3.13	84.50±4.64	8.12±0.35
10% L-SLN	146.28±0.93	−28.37±3.09	85.18±3.57	8.13±0.25
10% L-NE	144.68±0.69	−45.47±3.56	88.81±5.62	8.76±0.45

Fig. 1. TEM Images (a) 30000× of 10% L-SLN, (b) 30000× of 10% L-NLC, (c) 30000× of 10% L-NE, (d) 80000× of 10% L-SLN, (e) 80000× of 10% L-NLC and (f) 80000× of 10% L-NE

Different amounts of liquid lipids in lipid nanoparticles with terpene (10% L-SLN, 15% in L-NLC, 20% in L-NLC and C-NLC and 40% in L-NE) did not influence particle size. These results are consistent with Wang et al.²²⁾ and Charoenputtakhun et al.²³⁾ Under optimized production conditions (40% amplitude, 15 min for ultrasonication), small lipid nanoparticles comprised of different chemical liquid lipids with a relatively uniform size distribution were obtained. The zeta potential of the 10% L-NE formulations (−45.47 mV) was higher than that of other formulations due to the negative charged MCT at their carboxylic groups. The ZP values of NLC with and without terpene were higher than for SLN formulations. This might be due to the accumulation of oleic acids on the surface of NLC formulations. In fact, the melting point of cetyl palmitate is higher than that of oleic acids (ca. 50°C versus ca. (−13°C). Consequently, when preparing NLC, the lipid core of the particles, composed of oleic acids dissolved in the melted cetyl palmitate, starts recrystallizing before being entrapped within the solid lipid matrix. Subsequently, excess oleic acids are accumulated in the outer shell of the nanoparticles, similar to the drug-enriched shell model proposed as a drug encapsulation model for SLN.²⁴⁾ Moreover, higher negative zeta potential values for formulations containing oleic acids may be caused by the acid groups of fatty acids, which migrated to and concentrated at the lipid–water interface of NLC droplets.^22,23) A similar result has been proposed for coenzyme Q10-loaded NLC composed of cetyl palmitate and MCT as lipid matrix.²⁴⁾

Major determinants of drug loading capacity were the types and amounts of lipid. At 0.3% ATRA, the percentage yield in all the formulations was greater than 80%. Thus, the lipid nanoparticles had a good capacity for ATRA loading. These results were consistent with result of Lim et al. ⁽²⁵⁾ that ATRA could be efficiently incorporated into SLN. The percentage of ATRA in all formulations except NLC was more than 80, implying that limonene increases ATRA solubility and consequently, increases drug loading capacity. 10% L-NE had the highest percent yield because the solubility of drug in oil (liquid lipid) was higher than in solid lipid. The amount of oil loaded in the lipid matrix did not change the morphology of the particles. Moreover, almost all of the particles were uniform shape and did not adhere to each other.

Ten percent L-NE had the highest percent yield in drug loading capacity. There was no significant difference in ATRA loading between 10% L-SLN and 10% L-NLC (Table 2). Crystallinity as determined by X-ray diffraction (XRD) (Fig. 2) showed that the ATRA powder was highly crystalline as evident from sharp peaks of the XRD pattern. No ATRA peak was found in the XRD pattern of lipid nanocarriers. XRD patterns of 10% L-SLN did not exhibit a sharp peak, indicating at least partial loss of crystalline nature. The CP of the 10% L-SLN was arranged in a lamellar lattice structure. The XRD patterns broadened when the quantity of liquid lipids in the formulation were increased. As observed in previous studies, the XRD peaks of 10% L-SLN resulted from the crystallinity of cetylpalmitate. The decrease in crystallinity of 10% L-NLC and 10% L-NE may provide more space for ATRA accommodation.

Fig. 2. XRD Pattern of ATRA Powder, 10% L-SLN, 10% L-NLC and 10% L-NE

Skin Permeation and Skin Distribution Study

Figures 3(a) and (b) show the permeation profile of ATRA-loaded NLC formulations with different amounts of limonene (0, 5, 10%) and different types of terpene (10% limonene and 10% cineole), respectively. The addition of terpene oils (limonene or cineole) to the formulation increased ATRA skin permeation. ATRA permeability was significantly higher when formulated in 10% L-NLC compared to 5% L-NLC, and similarly, 10% L-NLC was significantly higher than that of 10% C-NLC. Our results are in accordance with the results reported by Subongkot et al.,²⁶⁾ which demonstrated that the skin permeation flux of sodium fluorescein was higher than that of cineole when encapsulated in liposomes containing limonene. Enhancement in skin permeation by limonene may be due to increased ATRA solubility within the skin or due to increased partitioning of the drug into the stratum corneum membrane. It is also possible that penetration enhancers may increase drug permeability by causing reorganization of lipid domains or by changing confirmations within the keratinized protein component due to extraction of lipids from the stratum corneum.^26,27) The flux of ATRA formulated in 10% limonene composited lipid nanoparticles was in the order of 10% L-SLN >10% L-NLC >10% L-NE>suspension (Fig. 4). The amount of solid lipids (CP) influenced ATRA skin permeability. Teerachaideekul et al.²⁴⁾ reported that an increase in medium chain triglyceride content in Nile red loaded NLC led to a decrease in the occlusion factor, which is related to the amount and depth of Nile red skin penetration. The occlusion factor is dependent on the sample volume, particle size, crystallinity, lipid concentration, and type of colloidal system.²⁸⁾ Therefore, it is expected that 10% L-SLN would lead to a higher ATRA flux than 10% L-NLC and 10% L-NE. Stecova et al. also reported that the skin permeability of cyproterone acetate loaded in SLN was higher than that in NLC.²⁹⁾ No intact SLN and NLC penetrated into the skin,³¹⁾ therefore ATRA might be released from the nanoparticle the drug with or without permeation enhancers permeate through the skin. After application of the lipid nanoparticles onto the skin, film formation occurs, which subsequently leads to an occlusive effect. This leads to reduced water loss, resulting in increased skin hydration, and consequently, enhances the penetration of the drug into the skin.³⁰⁾ Therefore, it is anticipated that the formulation containing a greater amount of solid lipid would lead to greater skin penetration.

Fig. 3. The in Vitro Skin Permeation of ATRA from (a) ATRA-Loaded NLC: (□) 0% L-NLC, (▲) 5% L-NLC, (■) 10% L-NLC and (X)10% C-NLC (b) ATRA-Loaded Lipid Nanoparticles: (●) 10% L-SLN, (■) 10% L-NLC, (◆) 10% L-NE and (×) ATRA Suspension

Fig. 4. The Flux at Steady State of ATRA-Loaded Suspension NLC, 5% L-NLC, 10% C-NLC 10% L-NLC, 10% L-SLN and 10% L-NE

Each value represents the mean±S.D. (n=3–6).

To make correlative conclusions regarding the skin targeting of lipid nanoparticles, distributions of rhodamine b base in porcine skin was investigated using CLSM. It is believed that rhodamine b base (a lipophilic compound) appears in the inner phase of lipid nanoparticles (the lipid phase). Figure 5 illustrates the results obtained from CLSM following dermal delivery of rhodamine B base via the skin for 8 h from rhodamine B base in M (control) (5a), rhodamine B base loaded 10% L-SLN (5b), rhodamine B base loaded 10% L-NLC (5c), and rhodamine B base loaded 10% L-NE (5d). The skin thickness was optically scanned at ca. 10 µm increments from the surface of the skin (left to right, top to bottom). Ten percent L-SLN showed the highest intensity of fluorescence, followed by 10% L-NLC, 10% L-NE and dye in M, respectively (Fig. 6). The fluorescence intensity in the horizontal section at 30 to 75 µm of 10% L-SLN and 10% L-NLC was very strong and gradually decreased with increasing depth.

Fig. 5. Confocal Laser Scanning Microscopic (CLSM) Micrographs of Pig Dorsal Skin after the in Vitro Topical Administration of Rhodamine B Base via the Skin for 8 h from Rhodamine B Base in MCT (Control) (a) Rhodamine B Base Loading 10% L-SLN (b) Rhodamine B Base Loading 10% L-NLC (c) Rhodamine B Base Loading 10% L-NE (d)

The skin specimen was viewed by CLSM at ca. 10 µm increments through the Z-axis from skin surface.

Fig. 6. Intensity from CLSM Micrographs of Pig Dorsal Skin after the in Vitro Topical Administration of Rhodamine B Base via the Skin for 8 Hours from Rhodamine B Base in MCT (Control) (×), Rhodamine B Base Loading 10% L-SLN (●), Rhodamine B Base Loading 10% L-NLC (■) and Rhodamine B Base Loading 10% L-NE (◆)

Rhodamine B base was used as a model lipophilic drug for studying skin distribution by CLSM because it has low solubility and is a fluorescent molecule. The skin distribution results for rhodamine B base loaded terpene composited lipid nanoparticles revealed higher epidermal permeability as indicated by the higher intensity of fluorescence at the epidermis compared to that at the dermis. The results from CLSM were in agreement with the skin permeation results. It was observed that even after 8 h, the film formation and creamy texture on the skin surface was maintained for 10% L-SLN and 10% L-NLC. In contrast, no film formation or creamy texture was formed on the skin surface after applying 10% L-NE. After applying 10% L-SLN on the skin, a monolayer film formed, which causes water loss and ultimately, lipid modification leading to drug expulsion from 10% L-SLN. Afterwards, occlusion is achieved with 10% L-SLN. In addition, the occlusive factor depends on matrix crystallinity. In skin permeation, rhodamine B base was released from nanoparticle and then permeated into the skin. As a result, 10% L-SLN showed the highest dye distribution and intensity in the epidermis when compared to other formulations. These results are in accordance with study results by Teeranachaideekul et al., which showed that lipid formulations with higher amounts of oil influenced skin permeability and accumulation.^23,24) Therefore, it is anticipated that adding 10% limonene into SLN can maintain the occlusive property and 10% L-SLN can maximize skin permeability compared to 10% L-NLC and 10% L-NE.

Stability and Photostability Study

The average particle size and zeta potential of ATRA loaded lipid nanoparticles with limonene at days 0 and 90 stored at 4, 25 and 40°C, respectively, are shown in Table 3. Ten percent L-SLN formulations were found to be more stable than 10% L-NLC and 10% L-NE formulations. Sedimentation was not observed for any of the formulations after fresh preparation. After storage at 4°C for 90 d, no sedimentation was observed, and the size, size distribution, and zeta potential at 4°C for 90 d were not significantly different from the initial preparation. After storage at 25°C for 90 d, no sedimentation was observed for any formulation; however the size, size distribution, and zeta potential were slightly different from the initial preparation. After storage at 40°C for 90 d, the size of all lipid nanoparticles was significantly larger than the initial preparation. Thus SLN can improve the photo-stability and stability of ATRA.²⁵⁾

Table 3. Particle Size and Zeta Potential of ATRA Loaded Lipid Nanoparticles with Limonene at Days 0 and 90 Stored at 4, 25 and 40°C (n=3)

		10% L-SLN		10% L-NLC		10% L-NE
		Size±S.D. (nm)	Zeta±S.D. (mV)	Size±S.D. (nm)	Zeta±S.D. (mV)	Size±S.D. (nm)	Zeta±S.D. (mV)
0 d		146.28±0.93	−39.37±3.09	139.70±0.02	−39.93±3.13	144.68±1.19	−45.47±3.56
90 d	4°C	164.43±3.15	−25.53±1.22	149.33±1.45	−28.63±2.24	176.95±1.67	−27.90±2.19
	25°C	156.10±1.38	−20.56±1.89	134.60±1.07	−21.07±1.65	144.93±0.87	−20.80±1.63
	40°C	195.95±0.89	−29.00±2.27	364.23±3.44	−25.70±2.01	6937.0±926	−23.23±1.82

The percentage of ATRA remaining in ATRA loaded lipid nanoparticles with limonene at 4, 25 and 40°C at 7, 14 and 30 d compared to day 0 are represented in Fig. 7. The percentage of ATRA remaining in all the lipid nanoparticle formulations decreased when the temperature increased. ATRA loaded lipid nanoparticles with limonene formulations were the most chemically stable at 4°C after 30 d (Fig. 7). The stability of 10% L-SLN at 30 d, for all temperatures was not significantly different between 10% L-NLC and 10% L-NE. The percentage ATRA remaining after 30 d at 4°C, 25°C and 40°C was higher than 80% for 10% L-SLN and 10% L-NLC. ATRA stability in lipid nanoparticles may be due to the slow transition of lipids in the nanoformulation or the steric effect of T80. During long-term storage, triglycerides undergo degradation into fatty acids and mono- and diglycerides, which could compete with formulation surfactants for positioning on the surface. Fatty acids and monoglycerides can form mixed micelles, which may enhance hydrophobic drug partitioning out of the nanoparticle.^31,32)

Fig. 7. The Percentage of ATRA Remaining in 0.3% ATRA in ■ 10% L-SLN, 10% L-NLC and □ 10% L-NE ATRA Loaded Lipid Nanoparticle with Limonene at 4, 25 and 40°C at 7, 14 and 30 d Compared to Day 0

As the temperature increased, ATRA degradation also increased. The effect of temperature on the reaction rate can be explained by the Arrhenius equation.³³⁾ ATRA loaded lipid nanoparticles with limonene formulations were the most chemically stable at 4°C after 30 d (Fig. 7). Consequently, degradation of these formulations could easily be triggered by a high temperature. The percentage of ATRA remaining in SLN formulations at all temperatures after 30 d post production was higher compared to NLC formulations. The chemical instability of SLN was likely caused by lipid modification during storage, which may have caused the drug to expel from the lipid matrix.^34–36) When limonene is incorporated into SLN, limonene may decrease the formation of lipid modifications within the lipid matrix.

Since ATRA degrades when exposed to light, it is important to develop formulations that are photo-protective. Incorporation of ATRA in liposomes, niosomes, lipid emulsions, and NLC have been found to protect ATRA against photodegradation. Photodegradation of ATRA follows first order kinetics. The amount of ATRA versus time was fitted to the equation y=exp(−kt), where is an equivalent to the total concentration of ATRA at t=0. The photodegradation rate constant (k) and the corresponding half-life (t_1/2=0.693/k) of 0.3% ATRA in isopropyl alcohol (IPA), 10% L-SLN, 10% L-NLC, and 10% L-NE at 25±0.5°C under the UV light for 6 h are summarized in Table 4. The photodegradation of ATRA in isopropyl solution was the fastest, whereas the lipid nanoparticles significantly protected ATRA from photodegradation. After 2 h of UV light exposure, approximately 72% of the initial concentration of ATRA existed in the isopropanol solution, whereas ATRA-loaded lipid nanoparticles exhibited much slower photodegradation. Lipid nanoparticles protected ATRA from photodegradation after 6 h of UV exposure with 64.2%, 64.5% and 67.02% of the initial concentration of ATRA in 10% L-SLN, 10% L-NLC and 10% L-NE, respectively (Fig. 8). Our results showed that the entrapment of ATRA into lipid nanocarriers strongly protected ATRA from photodegradation. Lim et al. also reported that the degradation of ATRA was greatly retarded by incorporation in SLN.²⁵⁾

Table 4. Photodegradation Rate and Half Life of ATRA in Isopropyl Alcohol and Various Lipid Nanoparticle Formulations at 25±0.5°C in the Presence of UVA Light for 6 h (n=3)

Formulation	Photodegradation rate (×10−2/h)	Half life (h)
ATRA in isopropyl alcohol	17.23±1.12	4.02±0.26
ATRA 10% L-SLN	5.76±0.25	12.16±0.52
ATRA 10% L-NLC	5.72±0.32	12.11±0.68
ATRA 10% L-NE	5.43±0.42	12.76±0.97

Fig. 8. The Photodegradation Profile of (×) 0.3% ATRA in Isopropyl Alcohol (IPA), (●) 10% L-SLN, (■) 10% L-NLC and (◆) 10% L-NE at 25±0.5°C in the Presence of UVA Light for 6 h

Cytotoxicity Evaluation

Cell culture studies were conducted to evaluate fibroblast cell viability of both lipid nanoparticle with limonene and lipid nanoparticle with limonene loaded with 0.3% ATRA. The effect of lipid nanoparticles with and without ATRA on mitochondrial dehydrogenase activity is shown as an IC₅₀ value, where a high IC₅₀ value indicates low toxicity. The IC₅₀ values of lipid nanoparticles with and without ATRA are shown in Table 5. ATRA powder had the lowest IC₅₀ value. The IC₅₀ value of ATRA loaded lipid nanoparticles was approximately 4- to 57-fold higher than ATRA powder, whereas the IC₅₀ value of ATRA loaded lipid nanoparticles with 10% limonene was slightly higher than that of ATRA loaded lipid nanoparticles. In comparison, the cytotoxicity among lipid nanoparticles with and without limonene was in the order of 10% L-NLC >10% L-NE >10% L-SLN. However, their cytotoxicity level was low because the IC₅₀ value was higher than 1 mg/mL. The result from the cytotoxicity study implies that lipid nanocarriers decrease ATRA toxicity to fibroblast cells. Gokce et al. reported that fibroblast cells remained completely viable after exposure to resveratrol concentrations of 100 µM incorporated into SLN, and they concluded that the SLN formulation itself was not toxic.³⁷⁾

Table 5. IC₅₀ Value for ATRA Powder, 10% Limonene SLN, 10% Limonene NLC and 10% Limonene NE (n=3)

Formulation	IC50±S.D. (µg/mL)
ATRA powder	350±25
10% L-SLN	20000±1500
10% L-NLC	1250±100
10% L-NE	2000±150
ATRA 10% L-SLN	10000±1000
ATRA 10% L-NLC	1500±200
ATRA 10% L-NE	2500±200

CONCLUSION

Terpene composited lipid nanoparticles for dermal drug delivery of ATRA were successfully formulated. The use of terpenes in lipid nanoparticle formulations significantly improved ATRA permeability through the skin compared to lipid nanoparticles without terpene and ATRA suspension. All terpene composited lipid nanoparticles exhibited good ATRA loading capacity, protected ATRA from photodegradation, increased the skin permeability of ATRA, and exhibited a low toxicity to fibroblast cells. Our study suggests that 10% L-SLN and 10% L-NLC are the best potential carriers for dermal drug delivery of ATRA.

Acknowledgment

The authors would like to acknowledge The Thailand Research Funds through the Royal Golden Jubilee Ph.D. Program (Grant No. PHD/0141/2552) and the Basic Research Grant (Grant No. BRG 5680016) for financial support. We wish to thank the Division of Medical Molecular Biology, Department of Research and Development, Faculty of Medicine Siriraj Hospital, Mahidol University for providing the confocal laser scanning microscope used in this research.

REFERENCES

• 1) Subedi RK, Oh SY, Chun MK, Choi HK. Recent advances in transdermal drug delivery. Arch. Pharm. Res., 33, 339–351 (2010).
• 2) Vijaya R, Ruckmani K. In vitro and in vivo characterization of the transdermal delivery of sertraline hydrochloride films. Daru, 19, 424–432 (2011).
• 3) Pathan IB, Setty CM. Chemical penetration enhancers for transdermal drug delivery systems. Trop. J. Pharm. Res., 8, 173–180 (2009).
• 4) Bouwstra JA, Honeywell-Nguyen PL, Gooris GS, Ponec M. Structure of the skin barrier and its modulation by vesicular formulations. Prog. Lipid Res., 42, 1–36 (2003).
• 5) Morgan CJ, Renwick AG, Friedmann PS. The role of stratum corneum and dermal microvascular perfusion in penetration and tissue levels of water-soluble drugs investigated by microdialysis. Br. J. Dermatol., 148, 434–443 (2003).
• 6) Bouwstra JA, Honeywell-Nguyen PL, Gooris GS, Ponec M. Structure of the skin barrier and its modulation by vesicular formulations. Prog. Lipid Res., 42, 1–36 (2003).
• 7) Vikas S, Seema S, Gurpreet S, Rana AC, Baibhav J. Penetration enhancers: a novel strategy for enhancing transdermal drug delivery. Int. Res. J. Pharm., 2, 32–36 (2011).
• 8) Di Vaio C, Graziani G, Gaspari A, Scaglione G, Nocerino S, Ritieni A. Essential oils content and antioxidant properties of peel ethanol extract in 18 lemon cultivars. Sci. Hortic. (Amsterdam), 126, 50–55 (2010).
• 9) Sakeena MH, Elrashid SM, Muthanna FA, Ghassan ZA, Kanakal MM, Laila L, Munavvar AS, Azmin MN. Effect of limonene on permeation enhancement of ketoprofen in palm oil esters nanoemulsion. J. Oleo Sci., 59, 395–400 (2010).
• 10) Lim PF, Liu XY, Kang L, Ho PC, Chan YW, Chan SY. Limonene GP1/PG organogel as a vehicle in transdermal delivery of haloperidol. Int. J. Pharm., 311, 157–164 (2006).
• 11) Ahad A, Aqil M, Kohli K, Sultana Y, Mujeeb M, Ali A. Role of novel terpenes in transcutaneous permeation of valsartan: effectiveness and mechanism of action. Drug Dev. Ind. Pharm., 37, 583–596 (2011).
• 12) Narishetty STK, Panchagnula R. Transdermal delivery of zidovudine: effect of terpenes and their mechanism of action. J. Control. Release, 95, 367–379 (2004).
• 13) Sapra B, Jain S, Tiwary AK. Percutaneous permeation enhancement by terpene: mechanistic view. AAPS J., 10, 120–132 (2008).
• 14) Teeranachaideekul V, Souto EB, Junyaprasert VB, Müller RH. Cetyl palmitate-based NLC for topical delivery of Coenzyme Q(10)- development, physicochemical characterization and in vitro release studies. Eur. J. Pharm. Biopharm., 67, 141–148 (2007).
• 15) Mukherjee S, Ray S, Thakur RS. Solid lipid nanoparticles: a modern formulation approach in drug delivery system. Indian J. Pharm. Sci., 71, 349–358 (2009).
• 16) Uner M, Wissing SA, Yener G, Müller RH. Solid lipid nanoparticles (SLN) and nanostructured lipid carriers (NLC) for application of ascorbyl palmitate. Pharmazie, 60, 577–582 (2005).
• 17) Baldwin HE, Nighland M, Kendall C, Mays DA, Grossman R, Newburger J. 40 years of topical tretinoin use in review. J. Drugs Dermatol., 12, 638–642 (2013).
• 18) Sen T, Tarimci N. The effect of cyclodextrins on the in vitro characteristics of semisolid formulations of all-trans-retinoic acid. Turk. J. Pharm. Sci., 10, 95–108 (2013).
• 19) Chinsriwongkul A, Opanasopit P, Ngawhirunpat T, Chareansriwilaiwat N, Sila-On W, Ruktanonchai U. Physicochemical properties of lipid emulsions formulated with high-load all-trans-retinoic acid. PDA J. Pharm. Sci. Technol., 61, 461–471 (2007).
• 20) Ngawhirunpat T, Opanasopit P, Rojanarata T, Panomsuk S, Chanchome L. Evaluation of simultaneous permeation and metabolism of methyl nicotinate in human, snake, and shed snake skin. Pharm. Dev. Technol., 13, 75–83 (2008).
• 21) Chern WH, Thakker KD. Development and validation of in vitro release tests for semisolid dosage forms—Case study. Dissolution Technol., 10–15 (2003).
• 22) Wang JJ, Liu KS, Sung KC, Tsai CY, Fang JY. Lipid nanoparticles with different oil/fatty ester ratios as carriers of buprenorphine and its prodrugs for injection. Eur. J. Pharm. Sci., 38, 138–146 (2009).
• 23) Charoenputtakhun P, Opanasopit P, Rojanarata T, Ngawhirunpat T. All-trans-retinoic acid-loaded lipid nanoparticles as a transdermal drug delivery carrier. Pharm. Dev. Technol., 19, 164–172 (2014).
• 24) Teeranachaideekul V, Boonme P, Souto EB, Muller RH, Junyaprasert VB. Influence of oil content on physicochemical properties and skin distribution of Nile red-loaded NLC. J. Control. Release, 128, 134–141 (2008).
• 25) Lim SJ, Lee MK, Kim CK. Altered chemical and biological activities of all-trans-retinoic acid incorporated in solid lipid nanoparticle powders. J. Control. Release, 100, 53–61 (2004).
• 26) Subongkot T, Opanasopit P, Rojanarata T, Ngawhirunpat T. Effect of limonene and 1,8-cineole on the skin penetration of fluorescein sodium deformable liposomes. Adv. Mat. Res., 506, 449–452 (2012).
• 27) William AC, Barry BW. Penetration enhancer. Adv. Drug Deliv., 56, 603–618 (2004).
• 28) Wissing S, Muller RH. The influence of the crystallinity of lipid nanoparticles on their occlusive properties. Int. J. Pharm., 242, 377–379 (2002).
• 29) Stecová J, Mehnert W, Blaschke T, Kleuser B, Sivaramakrishnan R, Zouboulis CC, Seltmann H, Korting HC, Kramer KD, Korting MS. Cyproterone acetate loading to lipid nanoparticles for topical acne treatment: particle characterization and skin uptake. Pharm. Res., 24, 991–1000 (2007).
• 30) Pardeike J, Hommoss A, Müller RH. Lipid nanoparticles (SLN, NLC) in cosmetic and pharmaceutical dermal products. Int. J. Pharm., 366, 170–184 (2009).
• 31) Schneider M, Stracke F, Hansen S, Schaefer UF. Nanoparticles and their interactions with the dermal barrier. Dermatoendocrinol, 1, 197–206 (2009).
• 32) Waraho T, McClements DJ, Decker EA. Impact of free fatty acid concentration and structure on lipid oxidation in oil-in-water emulsions. Food Chem., 129, 854–859 (2011).
• 33) Thatipamula R, Palem C, Gannu R, Mudragada S, Yamsani M. Formulation and in vitro characterization of domperidone loaded solid lipid nanoparticles and nanostructured lipid carriers. Daru, 19, 23–32 (2011).
• 34) Teeranachaideekul V, Muller RH, Junyaprasert VB. Encapsulation of ascorbyl palmitate in nanostructured lipid carriers (NLC)—Effects of formulation parameters on physicochemical stability. Int. J. Pharm., 340, 198–206 (2007).
• 35) Heurtault B, Saulnier P, Pech B, Proust JE, Benoit JP. Physico-chemical stability of colloidal lipid particles. Biomaterials, 24, 4283–4300 (2003).
• 36) Pardeshi C, Rajput P, Belgamwar V, Tekade A, Patil G, Chaudhary K, Sonje A. Solid lipid based nanocarriers: An overview. Acta Pharm., 62, 433–472 (2012).
• 37) Gokce EH, Korkmaz E, Dellera E, Sandri G, Bonferoni MC, Ozer O. Resveratrol-loaded solid lipid nanoparticles versus nanostructured lipid carriers: evaluation of antioxidant potential for dermal applications. Int. J. Nanomedicine, 7, 1841–1850 (2012).