Cannabidiol-Loaded Lipid-Stabilized Nanoparticles Alleviate Psoriasis Severity in Mice: A New Approach for Improved Topical Drug Delivery

2.2. In Vitro Skin Permeability of Semi-Solid LSN Formulation Development

The skin permeability of CBD from the LSN semi-solid formulations was analyzed using fresh rat skin mounted on a Franz diffusion cell system. No significant differences were observed between the skin permeation and retention of the various LSNs. However, since SALA-stabilized nanoparticles showed a relatively high CBD release in our previous study [18], they were used in skin penetration experiments as a representative LSN system. The LSN dispersion was thickened with fumed silica (CAB-O-SIL^® 530), MAS (Veegum HV), and Poloxamer 407 thermogel, while LSNs dispersed in water served as a non-gelled control. Volumes of 0.2 mL (about 0.2 g) from the various LSN formulations, each containing CBD at a concentration of 500 µg/g, were applied on the excised skin, providing 100 µg dose of CBD over 1.77 cm² of skin surface area. (i.e., 100 µg CBD/0.2 g quantity and 56.5 µg/cm² skin surface area). The unstirred volume of these LSN products (about 0.2 mL) was relatively high for typical dermal application, and only a tiny portion of CBD in the SLNs came in direct contact with the skin. Typically, semi-solid formulations are applied to the skin in much lower volumes per cm², constituting about 1 µL/cm² [28,29] and accompanied by rubbing. Thus, considerably higher availability of LSNs would be expected in clinical use. Considering the hindrance caused by the gel structures, the total permeation from the liquid aqueous dispersion was expected to be the highest.

As seen in Figure 2, there was a quantitative penetration of CBD into rat skin. The extent of CBD penetration depended on the type of gelling agent used for formulating the LSNs. Skin penetration of CBD from the liquid dispersion of the LSNs was relatively high, as expected (100% ± 28%, with actual values of 2.0 ± 0.8 µg/cm²), compared to the penetration of CBD from the poloxamer thermogel, which was the lowest (15.4% ± 14.6%, with actual values of 0.4 ± 0.3 μg/cm²; p < 0.001, ANOVA test). The relatively low skin penetration of LSNs from the poloxamer gel can be explained by the micelle density formed by poloxamers. Liu et al. [30] found that the micellar face-centered cubic lattice length is 29.5 nm for 15–45% w/w poloxamer 407 in an aqueous solution. Thus, with such small distances between the poloxamer micelles, the mobility of the 200 nm diameter LSNs through the Poloxamer gel network should have been significantly obstructed at rest (without external mechanical shear, i.e., rubbing). MAS particles are extremely thin, negatively charged plates approximately 2 nm thick and about 200–500 nm long, forming a three-dimensional network after water dispersion [31]. CAB-O-SIL^® 530 fumed silica has an average aggregate size of 200–300 nm, though the distance between the aggregates may be even greater. According to localization and hopping theory [32,33], the permeation of nanoparticles through an entangled polymer network depends on the confinement ‘parameter C’ described as the ratio of effective particle diameter to the effective tube diameter, which roughly corresponds to the distance between the polymer or mesh crosslinks. Particles with the confinement parameter C < 1 permeate relatively free, whereas the movement in the mesh of particles with 1 < C < 3 is due to hopping, given a sufficient activation energy. According to this theory, the diffusion coefficient in the mesh quickly drops proportionally to the exponent of (-C²). Thus, CAB-O-SIL^® 530 gel with the largest distances between the gel crosslinks would have the highest diffusion coefficient compared to the other tested gels, explaining the differences observed in Figure 2. It was also noted that the skin penetration of CBD was dependent on the silica concentration. Compared to 5% w/w gel, penetration extent decreased to 63.5 ± 17.5% for 4% w/w gel and further to 33.5% ± 7.4% for 3% w/w gel. This result is counter-intuitive to the expectation that reducing the gelling agent’s content would increase the molecules’ or NPs’ permeation through its polymeric network. It has been shown by Binder et al. [34] that an increase in the concentration of hydroxypropyl methylcellulose (HPMC) and hydroxyethyl cellulose (HEC) reduced the permeation of sulphadiazine sodium through the skin, which was explained by increased viscosity and entanglement of these cellulosic gels. In contrast to these polymers, the higher concentration of CAB-O-SIL^® 530 silica gel increased skin permeation, probably due to the hydrophobicity of the silica particles. Additional silica particles tend to form aggregates, building a mesh with thicker strands-fibers and larger openings. These results established CAB-O-SIL^® 530 5% gel as a more suitable semi-solid vehicle for LSNs and NPs in general.

2.3. In-Vivo Skin Penetration and Retention

Further assessment of the skin penetrability of LSNs from the topical silica gel vehicle was performed in an in-vivo study using SALA-stabilized NPs loaded with a fluorescent marker—coumarin 6 (C6) [35]. The LSN dispersion was combined with CAB-O-SIL^® 530 to obtain a 5% silica gel with 7.5 × 10¹¹ NPs/g. Then, the gel was applied for 2 h onto the abdomen of an anesthetized rat, which had previously been shaved and depilated as described in the experimental section. As seen in Figure 3, C6 penetrated to about 100 µm skin depth, roughly corresponding to or slightly exceeding the thickness of the epidermis layer of rat skin [36,37].

3.2. Size and Microscopic Analysis

3.3. Determination of CBD in NP Dispersion

To quantify CBD (within nano-sized particles), to about 25 μL aliquots from each particle sample that was carefully weighed, 975 μL MeOH was added and stirred. After at least 10 min, the samples were further diluted 1:10, and then the liquid was injected into an HPLC system (Shimadzu VP series, Shimadzu Corp., Tokyo, Japan), equipped with a prepacked column (ReproSil-Pur 300 ODS-3, 5 µm, 250 mm 4.6 mm, Dr. Maisch, Ammerbuch, Germany), which was constantly maintained at 30 °C. The samples were chromatographed using a mobile phase of acetonitrile-35 mM acetic acid (75:25) at a 1 mL/min flow rate. A calibration curve, peak area measured at 208 nm versus CBD concentration, was constructed by running standard drug solutions in MeOH for each series of chromatographed samples.

3.4. In-Vitro Skin Penetration Study

3.5. In-Vivo Skin Penetration Study and Image Analysis

The in-vivo penetration study was performed on Sprague–Dawley rats. On the day prior to the experiment, the rats were anesthetized, and their abdominal hair was carefully clipped and depilated (Veet cream, Reckitt Benckiser, Chartres, France). On the day of the experiment, the rats were anesthetized, and Coumarin 6 (TCI, Tokyo, Japan) loaded NPs formulated in a 5% silica gel were applied at about 50 mg/cm². After two hours, the rats were euthanized by aspiration of CO₂. The abdominal skin was washed with plenty of water and removed from the carcasses. The animal treatments were performed in accordance with protocol authorization number: IL-30-06-2020(C). SALA stabilized, Coumarin 6 (C6) loaded NPs were prepared similarly to CBD loaded NPs (see Section 3.1), with initial 5 × 10⁻⁴% w/w C6 content in the ethanol solution. The C6-loaded NP dispersion in 5% silica gel was prepared as described in Section 3.4.2 with a final concentration of about 0.002%w/w C6. The excised skin was snap-frozen in liquid nitrogen and sectioned with a cryotome using 100µm thickness for further confocal microscopy analysis (Spinning disc confocal microscope, 3i, Denver, CO, USA). The micrographs were collected with 1.5 μm depth steps. Further, they were processed with Fiji software (version 2.9.0/1.54f) [61] using the Z-project function.

3.6. Imiquimod-Induced Psoriasis in Mice

The authors are grateful for the professional assistance and technical support of the staff at Ilze Katz Institute for Nanoscale Science & Technology. The authors want to especially express their appreciation to Ilya Eydelman for his support and helpful assistance.