2.2. Preparation of solid-state CBD-CD inclusion complexed polymeric micelles (CBD-β-CDPM)

CBD 7.9 g (25 mmol) was dissolved in 78.0 g of the warm ethanol at 50 °C and stirred with a magnetic stirrer (C-MAG HS 7, IKA® Works Co. Ltd, Bangkok, Thailand.) until a clear solution was obtained that resulted in an alcoholic CBD solution. Poloxamer 407, 9.9 g (0.79 mmol) and β-CD 28.4 g (25 mmol) were dispersed in separate vessels in 400 mL and 476 mL, respectively, of water at 70 °C by slow addition and stirring with a high shear mixer (IKA RW 20, Co. Ltd, Bangkok, Thailand) at 1000 rpm for at least 60 min or until a clear solution was formed. The aqueous β-CD solution and aqueous poloxamer 407 solution were then combined to create a homogenous mixture. The aqueous mixture of β-CD and poloxamer 407 was then mixed with CBD ethanolic solution and the mixture was stirred at a low shear rate (500 rpm) for at least 30 min or until a uniform dispersion was created. As a consequence, 1000 g of the combination were produced.

The mixture solution was freeze-dried to produce a water-soluble CBD-β-CDPM using the Martin Christ Model Delta 2–24 LSC plus freeze-dryer (Osterode am Harz, Germany) using the following conditions: pre-freeze temperature at −40 °C for 24 h; primary drying at −10 °C for 24 h at a chamber pressure of 0.633 mbar; and secondary drying at 10 °C for 4 h and 20 °C for 20 h. The schematic diagram of CBD-β-CDPM preparation is illustrated in Supplementary Fig. S1, and the ingredients of the CBD-β-CDPM are listed in Table 1 .

2.3. Preparation of the nasal spray solution containing CBD-β-CD complexed polymeric micelles (CBD-β-CDPM-NS)

All ingredients for the nasal spray formulation containing (CBD-β-CDPM) are provided in Table 2 . For each 100 mL of nasal solution, 20 mL of sterile water for injection and 2 g of ethanol were combined. Then, 0.3 g of the dried CBD-β-CDPM powder (equivalent to CBD 50 mg) was gradually added during constant stirring. After all excipients were dissolved, sodium citrate, benzalkonium chloride, and citric acid monohydrate were added. Glycerin was then added in the last step. HEC was disseminated in sterile water for injection and then homogeneously combined with the CBD-β-CDPM solution to obtain the nasal spray solution The resultant nasal spray formulation was kept at controlled room temperature for subsequent analysis. As a control, a formulation without CBD-β-CDPM was prepared. Also, CBD-β-CD was prepared similarly to the above method without adding poloxamer 407 to compare the interaction or characteristics of CBD and β-CD without surfactant interference.

2.4. Assay of CBD in the CBD-β-CDPM-NS formulation

High-performance liquid chromatography (HPLC) was used to analyze the CBD content in the CBD-β-CDPM powder and CBD-β-CDPM-NS formulation. The HPLC system included the CBM-20A system controller (Shimadzu Corporation, Tokyo, Japan), LC-30AD pump, SIL-30AC autosampler, SPD-M20A PDA detector, and CTO-20AC column oven. A Shim-pack GIS C18 (150 mm × 4.6 mm, 3 μm) from Shimadzu (Shimadzu Corporation, Tokyo, Japan) at 15 °C was used to perform the separations. A degassed mixture of 70% acetonitrile and 30% ultrapure water made up the mobile phase. The sample or standard CBD substance was dissolved in the mobile phase. The injection volume and flow rate were 10 μL and 1 mL/min, respectively. A spectrophotometer was used to monitor the separation at 207 nm. LC Solution software (Shimadzu Corporation, Tokyo, Japan) was used to process the acquired data. The validation parameters included precision, accuracy, specificity, limit of quantitation, limit of detection, linearity, and robustness. There was no matrix interference of either CBD-β-CDPM powder or CBD-β-CDPM-NS in the chromatographic analysis.

2.5. Characterization of dried CBD-β-CDPM powder

The moisture content of the dried CBD-β-CDPM powder was analyzed using thermogravimetric analysis (TGA). Samples were accurately weighed and subjected to heating at a constant rate of 10 °C/min from 30 to 150 °C under a nitrogen purge using a thermogravimetric analyzer (Model TGA 7 PerkinElmer, Inc., USA).

All samples including CBD, β-CD, poloxamer 407, CBD-β-CD, CBD-β-CDPM powder, and the physical mixing of CBD, β-CD, and poloxamer 407 were subjected to differential scanning calorimetry (DSC). DSC analysis was carried out on a differential scanning calorimeter (DSC 800, Perkin Elmer Inc., USA). Each sample (5–10 mg) was heated at a rate of 10 °C/min from 30 to 300 °C under a flowing argon atmosphere (flow rate: 20 mL/min) in an aluminum pan.

All samples including CBD, β-CD, poloxamer 407, CBD-β-CD, and CBD-β-CDPM powder were determined by Fourier transform infrared (FT-IR) spectroscopy. The sample powder (1 mg) was mixed with KBr in a 1:100 ratio following by compression to obtain 2 mm transparent disc. FT-IR spectra between 4000 and 400 cm⁻¹ were determined after an accumulation of 16 scans using a Spectrum One (Perkin Elmer, Inc., USA).

2.6. Characterization of CBD-β-CDPM-NS

CBD-β-CDPM was dispersed in a purified water at a concentration of 0.3% w/v which was equal to the concentration of CBD-β-CDPM in the CBD-β-CDPM-NS formulation. The particle size distribution and zeta potential were then determined by a Zetasizer (Nano ZS ZEN3600, Malvern Instruments Ltd., Worcestershire, UK). CBD-β-CDPM-NS was also analyzed without dilution. The sample measurements were performed at 25 °C. The average of ten measurements at an angle of 90° was used to calculate the particle size and polydispersity index of the investigated formulations.

The viscosity of CBD-β-CDPM-NS was analyzed using a Modular Advanced Rheometer System (HAAKE MARS 60; Thermo Fisher Scientific, Bremen, Germany). The rheometer used a 60 mm diameter parallel plate at a gap of 0.5 mm with a Peltier temperature control system. After loading a 3 g sample onto the bottom plate, the upper plate was moved to the designed gap. The flow experiments were conducted at a constant 25 °C with a shear rate of 1 to 1000 s⁻¹. A frequency sweep was conducted in a range of 0.1 Hz to 100 kHz. Each batch of tests was performed in triplicate.

The surface tension of CBD-β-CDPM-NS formulation was measured by the Du Noüy ring method. The Du Noüy ring is an automated operation to determine the surface tension. The tensiometer used in this experiment was a model DY-300 (Kyowa Interface Science Co., Ltd., Tokyo, Japan). Approximately 200 mL of CBD-β-CDPM-NS was placed in a glass surface tension cup and loaded on the equipment. The ring was cleaned after each measurement to ensure the residue was completely removed. All measurements were performed in triplicate.

The osmolality of CBD-β-CDPM-NS was determined by an automatic cryoscopic osmometer (Osmomat® 030, Gonotec, ELITechGroup Inc., USA). All measurements were performed in triplicate.

2.7. Stability study

To determine the stability under actual storage conditions in Thailand, which is in a hot and humid zone, 20 mL of the CBD-β-CDPM-NS formulation was placed in a polyethylene nasal spray bottle and kept for six months at 30 ± 0.5 °C and at a relative humidity of 75 ± 5% (ASEAN Guideline on Stability of Drug Products, 2018). The content of CBD in CBD-β-CDPM-NS and physical properties of the formulation including appearance, pH, particle size, and zeta potential were recorded.

2.8. In vitro drug release studies

In vitro dissolution of the nasal formulation in a nasal cavity was developed according to the previous study (Inoure et al., 2022). The CBD dissolution of CBD-β-CDPM-NS was evaluated. Phosphate-buffered saline (PBS) pH 6.4 was used as an artificial nasal fluid (Washington et al., 2000). The temperature of the medium was kept at 35 °C. A quantity of 500 μL of CBD-β-CDPM-NS (equivalent to 50 μg × 5 doses of CBD) was loaded into the simulated chamber with 10 mL dissolution medium. After dosing, a 100 μL aliquot of medium was collected from the bottom of the chamber at intervals of 0, 1, 2, 3, 4, 5, 7, 10, and 15 min and replaced with 100 μL fresh medium. The amount of CBD dissolved was analyzed by HPLC as described in Section 2.4.

2.9. Cell culture conditions for cell toxicity studies

The RAW 264.7 mouse monocyte/macrophage cell line (ATCC TIB-71, USA) was grown in Dulbecco’s Modified Eagle Medium (DMEM, Gibco®, USA) containing 10% fetal bovine serum (FBS, Gibco®, USA) and 100 U/mL penicillin/streptomycin (Gibco®, USA). The media was replenished every two days while the cells were cultured at 37 °C in a 5% CO₂ incubator. The cells were collected by gentle rocking. A new cell suspension was then prepared by adding fresh culture media, which was then used for further incubation.

The RPMI 2650 human nasal septum epithelial cell line (ATCC: CCL-30, Rockville, MD, USA) was cultured in Eagle’s Minimum Essential Medium (EMEM, Gibco®, USA) supplemented with 10% fetal bovine serum (FBS, Gibco®, USA), 100 U/mL penicillin/streptomycin (Gibco®, USA). The cells were incubated at 37 °C in a 5% CO₂ incubator and the media were changed every two days. They were harvested by gentle rocking, followed by the addition of fresh culture medium to create a new cell suspension for further incubation.

2.10. Cell proliferation and viability assay

A quantity of 100 μL of RAW 264.7 cells or RPMI 2650 at a concentration of 1 × 10⁵ cell/mL was seeded in a 96-well plate with complete medium. After 24 h of incubation, the culture plates were treated with CBD-β-CDPM-NS at CBD concentrations equivalent to 3.12–50 µg/mL in fresh media. Untreated cells were used as a negative control. After the cells were exposed to the test samples for 24 h, cell viability was measured using a solution (5 mg/mL) of 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT). Briefly, the cells were treated with 80 µL of fresh media and the MTT solution (20 µL), which was then incubated at 37 °C under 5% CO₂ for 4 h. The media containing MTT were then removed and dimethyl sulfoxide (100 µL) was added. Absorbance was recorded using a microplate reader (Biohit 830, Biohit®, Helsinki, Finland) at 570 nm. The percentage of cell proliferation was calculated and compared to the negative control.

2.11. In vitro CBD-β-CDPM-NS efficacy to reduce S-RBD-induced pro-inflammatory cytokine production

This experiment was performed according to previous studies ( Srichana et al., 2022 ). The coding sequence for the S-RBD of SARS-CoV-2 (S-RBD, residue 317–539) was optimized for expression in mammalian cells from the sequence obtained from the reference strain (Wuhan-Hu-1) and cloned into pSecTag using restriction enzyme cloning (BsiWI and XhoI) in-frame with the Igκ leader sequence at the N-terminus and the Myc-His tag at the C-terminus. Recombinant tagged S-RBD protein was produced according to a previously published protocol (Amanat et al., 2020). Briefly, human embryonic kidney 293 T cells were maintained in OptiMEM with 10% fetal bovine serum supplement and transfected with pSecTag-SRBD using FuGENE transfection reagent according to the manufacturer’s protocol. The cell media were then changed to OptiMEM with no supplement at 6 h post-transfection. At 72 h post-transfection, the supplement was conditioned with binding buffer (4X) (1.2 mM NaCl, 200 mM NaH₂PO₄, 40 mM imidazole). The conditioned supernatant was then mixed with Ni-NTA agarose resin (Qiagen) for 1 h. The unbound proteins were washed off with a wash buffer (300 mM NaCl, 50 mM NaH₂PO₄, 20 mM imidazole). Recombinant S-RBD was then eluted with an elution buffer (300 mM NaCl, 50 mM NaH₂PO₄, 250 mM imidazole). The eluted fractions were pooled, concentrated and buffer-exchanged into PBS using a 10 molecular weight cut-off centrifugal filter unit. The protein concentration was measured using the Bradford method (Bradford, 1976). Recombinant S-RBD was stored at −80 °C in small aliquots prior to use.

RAW 264.7 cells (100 μL) at a cell density of 10⁵ cells/mL in complete media were seeded into each well of a 96-well plate. The cells were allowed to grow until 70–80% confluence for 2 h. Following incubation, the cells were treated with S-RBD and CBD-β-CDPM-NS in different orders as follows:

Treatment mode: Cells were stimulated with S-RBD protein (10 μg/mL) and incubated at 37 °C under 5% CO₂ for 24 h before treatment with CBD-β-CDPM-NS containing CBD equivalent to 50 μg/mL and incubated at 37 °C under 5% CO₂ for 24 h.

Co-administration mode: Cells were stimulated with S-RBD protein (10 μg/mL) and concurrently treated with CBD-β-CDPM-NS containing CBD equivalent to 50 μg/mL for 24 h at 37 °C in 5% CO₂.

After 24 h of incubation, the cell supernatant of each experimental mode was assayed for pro-inflammatory cytokine levels that included TNF-α, IL-1β, or IL-6, which were generated in the cell supernatants obtained from the experiment explained earlier. Measurements used a rat TNF-α, IL-1β, or IL-6 Quantikine enzyme-linked immunosorbent assay kit (R&D Systems, Inc., Minneapolis, MN, USA). Using the 96-well plates included in the ELISA kit, 50 μL of TNF-α, IL-1β, or IL-6 diluent was added to each well. The experimental cell supernatant in the amount of 50 μL was added and incubated for 2 h at room temperature. Each well was washed (5X) with a buffer solution followed by adding the conjugate solution (100 μL) of either TNF-α, IL-1β, or IL-6 to each well and incubated for another 2 h. Each well was then washed with a buffer (5X) and 100 μL of substrate solution was added. The plates were incubated for 30 min at room temperature followed by adding a stop solution (100 μL). The absorbance of a consequent reaction was measured at 450 nm with a microplate reader (Biohit 830, Biohit®, Helsinki, Finland). The absorbance results were quantified using TNF-α, IL-1β, and IL-6 standard curves.

2.12. Ex vivo studies on the activity of CBD-β-CDPM-NS on porcine nasal epithelia mucosa

Pigs (22–23 weeks old, 80 kg) were dissected by the veterinary staff of Charoen Pokphand (CP) Foods Public Company, Ltd, Thailand. To maintain the tissue integrity, the nasal mucosa tissue was immediately removed from the pig snout. The mucosa explants were prepared using the procedure described by Tulinski et al. (2013), with slight modifications. In brief, isolated nasal mucosa explants were placed on sterile gauze soaked in DMEM supplemented with 1 mg/mL streptomycin and 1000 U/mL penicillin. The air–liquid interface was kept open and the cilia membrane exposed to the air. The process of explant preparation is depicted in Fig. 1 .

Porcine nasal mucosa explants were divided into pieces approximately 1 cm² and placed into a well of a 24-well plate filled with DMEM to the half-height of the explants. The ability of CBD-β-CDPM-NS to suppress the cytokine provoked by the S-RBD protein was examined in two experimental studies: prevention and treatment. The surface of a porcine nasal mucosa explant was exposed to S-RBD protein 50 µL (10 µg/mL), which was then incubated for 6 h in a CO₂ incubator to activate the immune system and generate pro-inflammatory cytokines. In the prevention study, nasal mucosa explants were treated with 25 μL of CBD-β-CDPM-NS for 30 min before S-RBD protein challenge. Following 6 h of S-RBD incubation in the treatment study, 25 μL of CBD-β-CDPM-NS was applied to the explant surface and incubated for 30 min. After the incubation time ended, the tissue was diced into small fractions suspended in culture medium (DMEM) and transferred to a microcentrifuge tube, placed in an ice bath, and then subjected to a 10 min sonication process. The generated cytokines (TNF-α, IL-1β, or IL-6) were detected in the supernatant after centrifuging the tissue suspension at 5000 rpm for 5 min. The Quantikine® ELISA Kit (R&D Systems, Inc., Minneapolis, MN, USA) was used to measure the porcine TNF-α, IL-1β, or IL-6 cytokine levels following the manufacturer’s instructions.

2.13. Ex vivo nasal epithelium permeation studies using Franz diffusion cells

Porcine nasal epithelium explants were cut into 2 × 2 cm² pieces and CBD transport was determined through the membrane. The explants were mounted between the donor and receptor chamber of the Franz diffusion cells (Hanson Technology, Chatsworth, CA USA) with an effective diffusion area of 1.76 cm² and cell volume of 8 mL filled with freshly prepared PBS pH 6.4. After an equilibrium time of 15 min at 35 °C ± 0.5 °C, 500 μL of CBD-β-CDPM-NS (equivalent to 250 μg CBD) were loaded into the donor compartment through direct contact with nasal epithelium explant membrane. The diffusion cells were maintained in a circulating water bath at 100 rpm. Samples of 500 μL were collected at each time point of 0, 5, 10, 15, 20, 25, 30, 45, and 60 min and fresh PBS pH 6.4 was used to replace the equal amounts of sampling. The 500 μL collected samples were diluted with 300 μL of mobile phase. The diluted samples were filtered with a 0.22 μm nylon membrane filter before the CBD analysis using HPLC described in Section 2.4. Following the experiment, the leftover porcine nasal mucosa membranes were placed in centrifuge tubes with an additional 800 μL of mobile phase. The samples were digested with a cell digestion machine (IKA, Bangkok, Thailand) and filtered through a 0.22 μm nylon membrane. The supernatant was then subjected to HPLC analysis to determine the CBD content retained in the membrane. The experiment was done in quadruplet. From the in vitro membrane permeation data, steady-state flux (J_ss) (μg/cm²/min) and the cumulative amount of CBD through the mucosa membranes (μg/cm²) were determined.

3.2. Development of CBD-β-CDPM

CBD complexation with CD was developed to improve water solubility. β-CD has undergone extensive safety testing and has been approved for use in pharmaceutical products (Loftsson and Brewster, 2010, Lv et al., 2019, Haimhoffer et al., 2019). Also, β-CD is safe for human ingestion and application (Pereira et al., 2021, Poulson et al., 2022; European Medicines Agency, 2014). Furthermore, β-CD was chosen in this study due to the similarity of its internal cavity size of 6.0–6.4 Å (Poulson, et al., 2022) and the size of the CBD molecule. From the literature, the crystal structure of CBD was reported to be monoclinic with dimensions of a = 10.617 Å, b = 10.649 Å, and c = 17.266 Å and β = 95.30(4)° (Jones et al., 1977), and another study reported similar dimensions of a = 10.4395 Å, b = 10.8739 Å, and c = 16.7853 Å and β = 95.448(1)° (Mayr et al., 2017). We confirmed the results of those studies from our CBD crystallography as shown in the supplementary data files. These findings indicated that a portion of the CBD molecule could be inserted into the β-CD cavity of 6.0–6.4 Å. In addition, another group reported that the β-CD cavity was 7.8 Å (Păduraru et al., 2022). β-CD has hydrophilic properties with a solubility in 25 °C water of 18 mg/mL (Chatjigakis et al., 1992). Furthermore, as the temperature is increased, β-CD becomes more soluble (Poulson et al., 2022, Yong et al., 2008). For this reason, the manufacturing process in this study was performed at 70 °C. After preparation, CBD-β-CDPM was lyophilized to produce a dry powder. This technique provided the amorphous formulation of CBD as reported in a previous study (Koch, et al., 2020). This current method is similar to the method by Koch et al., (2020) except CBD was dissolved in ethanol before adding it to the CD solution, which resulted in a higher amount of CBD loaded into the CD molecules.

The CBD encapsulated within the cavity of β-CD to form a 1:2 host–guest inclusion complex is described in the literature (Lv et al., 2019). Another consistent study prepared CBD as a β-CD inclusion complex. The molar stoichiometry of the CBD and β-CD complex was 1:2 by the precipitation method in which only 7 mg of β-CD was required to complex 1 mg of CBD (Mannila et al., 2007). However, Li et al. (2022) reported CBD was inclusion complexed with hydroxypropyl-β-CD at a ratio of 1:1.

In this study, CBD was formulated with β-CD in a 1:1 ratio and poloxamer to improve the aqueous solubility and bioavailability of the poorly soluble CBD. The CBD-β-CDPM solution was freeze-dried into powder and reconstituted as an active ingredient in a nasal spray formulation. The developed water-soluble CBD-β-CDPM is white to an off-white amorphous powder at room temperature after the lyophilization process, and the powder is free flowing with a high bulk volume. The yield of water-soluble CBD-β-CDPM was 90.9% (42.0 g) of the total weight of the solid ingredients (46.2 g) in the formulation. The TGA analysis showed moisture content of the water-soluble CBD-β-CDPM was 4.8 ± 0.3% w/w. After reconstituting the CBD-β-CDPM with water for injection, a clear colorless solution that easily dissolved was obtained. The solubility of CBD-β-CD was 0.4 μg/mL, which is 4 times higher than pure CBD. When CBD-β-CD was formulated with poloxamer 407, its solubility at 25 °C was 2.5 mg/mL, which is equal to 427.5 μg/mL of CBD (theoretically, 46.2 g of CBD-β-CDPM powder contains 7.9 g of CBD). Thus, the aqueous solubility of CBD-β-CDPM was 4275 times higher than the solubility of pure CBD at 0.1 μg/mL (Amanat et al., 2020), which is beneficial for developing aqueous nasal spray solutions.

3.3. Characterization of CBD-β-CDPM

A previously published thermal analysis reported that poloxamer 407 had the lowest melting point of 59.49 °C (Emam et al., 2021) followed by CBD at 67.5 °C (Stinchcomb et al., 2004), whereas β-CD melted at 267–269 °C (Thurein et al., 2018). However, the broad endothermic profile of β-CD showed a peak at around 100–150 °C, which corresponds to the loss of crystalized water molecules from the β-CD cavity. Several other researchers reported results such as a centered peak at 145 °C (Marangoci et al., 2019), a range of 34–155 °C (peak ∼ 125 °C) (Júnior et al., 2019), and a peak at 113.2 °C (David et al., 2019). In this study, the thermograms of the raw materials of pure compound poloxamer 407, CBD, and β-CD had melting peaks at 53.9, 64.2, and 122.9 °C, respectively. The thermogram of the physical mixture showed an absence of the characteristic melting peaks of poloxamer 407 and β-CD (Supplementary Fig. S2). The melting point of CBD shifted to a lower melting point temperature of 60 °C. This indicated that the physical mixture formed a complex with β-CD during the heating phase that resulted in a much lower melting point than the original melting point of pure β-CD caused by eutectic phenomena. Also, the melting point of CBD shifted to a lower value in the mixture of the three components with an absence of the poloxamer 407 peak. This indicated that the three molecules could interact even in the physical mixture during the heating stage, and that the lowest melting components of poloxamer 407 and CBD could act as a cosolvent system while forming a complex structure of those three components.

When the CBD complexed with β-CD (CBD-β-CD), it was found that part of the CBD structure was possibly inserted into the CD cavity. The small CBD peak at 60 °C was found to have a much smaller peak area together with an amorphous halo from 100 to 120 °C that was possibly caused from free CBD that did not form the inclusion complex with β-CD (Fig. S2).

The DSC thermogram showed that the CBD-β-CDPM is an amorphous solid, while the DSC thermogram did not show the response peak of any compounds. Hence the mixture was completely amorphous. It can be postulated that the mixture is readily soluble in water without free CBD remaining (Fig. S2). When CBD and poloxamer 407 were prepared as a mixture, there was no crystallinity observed in the thermogram, which indicated only an amorphous phase in the mixture.

The FT-IR spectra of CBD, poloxamer 407, β-CD, CBD-β-CD and CBD-β-CDPM are shown in Supplementary Fig. S3. The potential interactions between CBD and β-CD were investigated by FT-IR analysis by comparing the respective absorption spectra, considering specific functional groups of the CBD, poloxamer 407 and β-CD alone, and complexes of CBD-β-CD. The potential interactions of poloxamer 407 in the complexes system were also investigated. If CBD forms the inclusion complexes with β-CD, the characteristic peaks of CBD probably shift, decrease, or disappear (Li et al., 2021). The characteristic bands at 2925–2856 cm⁻¹ (C–H stretching) and at ∼ 3408 cm⁻¹ (O–H stretching) were found in CBD and β-CD alone and also in the complex of CBC-β-CD and CBD-β-CDPM. The CBD spectrum showed characteristic bands at 1623–1583 cm⁻¹ (cyclic alkene C = C stretching) that were not found in the β-CD and poloxamer 407. Only C–O stretching of aliphatic ether was found in β-CD at the wave number 1158 cm⁻¹ and also in the poloxamer 407 that usually occurs in the range 1260–1000 cm⁻¹ (Pavia et al., 2001). The characteristic IR bands of β-CD and CBD were found in the complex of CBD-β-CD and peak intensity decreased compared to raw compounds, which confirmed that molecular interaction between CBD and β-CD occurred.

From the FT-IR results, CBD, β-CD and poloxamer did not chemically interact with one another as all fingerprints of each component appeared in the spectra. Possible structures of CBD-β-CDPM could be a mixture of: (A) the CBD inclusion complex with β-CD at a ratio of 1:1; (B) CBD without complex encapsulation in the core of poloxamer; (C) CBD-β-CD complex may encapsulate in the core of the poloxamer but might be difficult to occur due to hydrophilicity molecules; and (D) CBD-β-CD complex inserted between hydrophilic parts of poloxamer micelles (Supplementary Fig. S4).

3.4. Development and stability of CBD-β-CDPM-NS

CBD-β-CDPM was used as an active ingredient along with a buffering agent, preservative, and ethanol as a permeation enhancer (Bhise et al., 2008) to formulate a nasal spray solution. The formulation for CBD-β-CDPM nasal spray (CBD-β-CDPM-NS) was a clear solution.

HEC (0.05 %w/w) was used as a mucoadhesive polymer in this experiment due to its adhesiveness to the nasal epithelial cells (Cho et al., 2021). In addition to the good mucoadhesive properties, Hansen et al. reported that HEC enhanced the transport of drug molecules across the nasal epithelium (Hansen et al., 2015). A preliminary study of HEC concentration at varying concentrations of 0.05, 0.1, 0.15, 0.2, and 0.25% w/w had no effect on surface tension but had a significant impact on viscosity. The viscosity of HEC ranged from 6.2 cP at 0.05% (w/v) to 82.1 cP at 0.25% (w/v) (Supplementary Fig. S5). On the basis of these results, the lowest concentration of HEC with the lowest viscosity was chosen for the formulation for easy release from the container.

The average size of the CBD-β-CDPM before freeze-drying was 101.4 ± 6.9 d.nm and its polydispersity index was 0.249 ± 0.013 (n = 6) (Table 3 ). The literature reported that polymeric micelles of pure poloxamer 407 showed a particle size under 100 nm (Bahman et al., 2019). The particle size of CBD-β-CDPM was larger than empty poloxamer micelles, which was likely due to the insertion of CBD or CBD-β-CD in the core of the micelles. The apparent zeta potential of the CBD-β-CDPM before freeze-drying was −8.9 ± 0.4 mV. This result indicated the micelles were negatively charged, which was similar to the results of −13.57 mV in a previous report on polymeric micelles of gambogic acid (Saxena and Hussain, 2012). Both the polypropylene oxide and polyethylene oxide segments in poloxamer 407 are nonionic; therefore, the addition of CBD and β-CD affected the surface charge of the polymeric micelles.

After storage of CBD-β-CDPM for six months, reconstitution of the powder with sterile water for injection resulted in a particle size of CBD-β-CDPM of 102.5 ± 7.3 d.nm and a zeta potential of −10.5 ± 0.2 mV. The size of the CBD-β-CDPM was reduced with a lower negative charge. This was possibly due to stabilization of the size and charge by equilibrium moisture during storage.

The particle size of CBD-β-CDPM-NS initially was 111.9 ± 0.7 d.nm and the zeta potential was 0.8 ± 0.1 mV. The particle size of CBD-β-CDPM-NS after storage at room temperature for six months increased to 121.4 ± 1.1 d.nm with a small change in the surface charge. The CBD-β-CD complex polymer micelle enlargement indicated some degree of instability although no precipitation or aggregation occurred, and the surface charge increased to nearly zero. This result was possibly caused by the buffering agent and preservative in the nasal spray formulation that affected the counter ions of the colloidal system. In addition, the environmental change of the colloidal system may lead to slightly larger sizes and structures of the micelles (Perumal et al., 2022).

The osmolarity of this formulation was 300 ± 0.8 mOsm/kg in the range of 285–310, which indicated an isotonic solution suitable for nasal preparation (USP, Ehrick et al., 2013, Barros et al., 2022). The viscosity of the nasal spray formulation was 12.04 ± 2.64 cP at the shear rate of 50 s⁻¹. The rheogram of the CBD-β-CDPM-NS showed a shear thinning system, which was typical of pseudoplastic behavior (Supplementary Fig. S6). As the shear rate increased, the viscosity decreased. The local pH value inside the nasal cavity directly influences the rate and extent of drug absorption. It has been suggested that the optimum pH value of nasal spray should range from 4.5 to 6.5 (Dhakar et al., 2011) or 5.5 to 6.5, which is close to the physiological pH (Barros et al., 2022). The pH value of the nasal spray formulation in this current study was 6.02 ± 0.02, which is suitable for nasal epithelial cells. The surface tension was 34.5 mN/m, which was lower than sterile water (72.8 mN/m at 20 °C) (Sinko, 2011). This result indicated that poloxamer 407 reduced the surface tension and formed complex polymeric micelles in the bulk solution.

The assay results of CBD are shown in Table 3. The CBD content in the CBD-β-CDPM was 102.1 ± 0.5% of the theoretical value (found 8.06 g in the theoretical CBD content of 7.9 g in a total solid content of 46.2 g). When CBD-β-CDPM powder was formulated as a nasal spray solution (CBD-β-CDPM-NS), the assay content was 99.8 ± 1.3% of the theoretical content.

The chemical and physical properties, which included CBD content, appearance, pH, osmolality, particle size, and zeta potential, of the CBD-β-CDPM powder and CBD-β-CDPM-NS did not significantly change (p-value < 0.5). These results confirmed that the product was stable in storage up to six months at room temperature.

3.5. CBD release profile from CBD-β-CDPM-NS and permeation through nasal epithelium

The in vitro CBD release from polymeric micelles of CBD-β-CDPM-NS in simulated nasal cavity conditions through the dialysis membrane is shown in Fig. 2 . CBD release from CBD-β-CDPM-NS was 104 ± 0.26%, while CBD release from CBD-β-CDPM was 91.94 ± 0.15% in 15 min. The CBD in the nasal spray (CBD-β-CDPM-NS) was 100% released within the first minute because the CBD molecules were complexed with β-CD in the form of micelle and colloidal systems that were easily released from the complexed state. In contrast to the CBD-β-CDPM powder, CBD release was slower than CBD-β-CDPM-NS.

The CBD permeation rate and amount of CBD per area of the membrane from CBD-β-CDPM-NS through the porcine nasal mucosa are shown in Fig. 3 (A) and (B). The steady-state flux (J_ss) of CBD over excised porcine nasal mucosa was 4.8 μg/cm²/min, and the rapid absorption of CBD from CBD-β-CDPM-NS increased after CBD-β-CDPM-NS was administered to the diffusion cells. After 15 min the CBD permeated the receptor cells at a lower permeation rate. The results showed that the permeation rate of CBD was enhanced by increased aqueous solubility of CBD (Ghezzi et al., 2021). However, CBD remained in the nasal epithelium and in the donor phase. The CBD retained in the nasal epithelial explants and the permeation across the mucosa correlated with the ex vivo CBD-β-CDPM-NS as described in Section 3.6.

3.6. Cell viability and in vitro efficacy of CBD-β-CDPM-NS

The cytotoxicity profiles of the CBD-β-CDPM, CBD-β-CDPM-NS, and nasal spray blank formulation against the human nasal septum epithelial cell line (RPMI 2650) and mouse monocyte/macrophage cell line (RAW264.7) are presented in Fig. 4 A and Fig. 4B, respectively. CBD-β-CDPM and CBD-β-CDPM-NS with CBD concentrations equivalent to 3.125–50 μg/mL were used to challenge the cell line. The results showed that all concentrations of CBD-β-CDPM, CBD-β-CDPM-NS, and the nasal spray blank formulation were safe for both RPMI 2650 and RAW264.7 cells with nearly 100% viability.

The SARS-CoV-2 virus’s recombinant S-RBD is a viral protein that binds to the host cell’s ACE2 receptor and enters the cell, causing pathological changes in lung injury and inflammatory responses that lead to excessive production of pro-inflammatory cytokines, inflammation, and tissue injury (Montazersaheb et al., 2022). These immune reactions might indicate the onset of a cytokine storm with severe clinicopathological consequences and severe COVID-19 disease (Mahmudpour et al., 2020). Stimulation of RAW264.7 macrophage-like cells by S-RBD as a COVID-19 surrogate was employed to determine the efficacy of CBD-β-CDPM-NS formulation to inhibit excessive pro-inflammatory cytokines, such as TNF-α, IL-1β, and IL-6. Stimulation resulted in the release of extraordinarily high levels of TNF-α, IL-1β, and IL-6, which were 539.18 ± 5.22 pg/mL (Fig. 5 A), 273.38 ± 4.61 pg/mL (Fig. 5B), and 1666.36 ± 5.96 pg/mL (Fig. 5C), respectively. The CBD-β-CDPM-NS formulation, on the other hand, did not trigger RAW264.7 cells to release any pro-inflammatory cytokines because the levels detected were very low and equivalent to the levels detected in the untreated control group (3rd column in Fig. 5 A–C). This indicated that the CBD-β-CDPM-NS formulation did not cause injury to the cells.

The capacity of the CBD-β-CDPM-NS formulation to suppress the production of pro-inflammatory cytokines as a result of S-RBD stimulation was evaluated when CBD-β-CDPM-NS formulation was applied after S-RBD stimulation (treatment mode) or concurrently with S-RBD stimulation (co-administration mode). Upon comparison with the S-RBD stimulation control group, all experiment modes demonstrated the capacity of the CBD-β-CDPM-NS formulation to significantly lower the levels of TNF-α, IL-1β, and IL-6. However, it is apparent from comparing the treatment mode and co-administration mode that the co-administration mode demonstrated significantly greater efficacy than the treatment mode. All measured cytokine levels generated when S-RBD and CBD-β-CDPM-NS formulation were administered together were extremely low and similar to the untreated control levels as depicted in Fig. 5, A–C. These findings are in agreement with our previous reports (Srichana et al., 2022), which showed that co-administration of a CBD formulation and allergens had the greatest impact on reducing the level of macrophage-produced cytokines. CBD binding to the CB₂ receptor on macrophages/monocyte resulted in a decrease in cytokine production in RAW264.7 cells exposed to S-RBD (Perwee, 2008). Co-administration of the CBD-β-CDPM-NS formulation and S-RBD would also function through additional mechanisms in addition to the CBD-CB₂ interaction to lessen the cytokine release induced by S-RBD. This was suggested to be the result of competing interactions between S-RBD and CBD with macrophages/monocytes; however, CBD interactions with macrophages/monocytes did not lead the immune cells to release response cytokines like S-RBD did. In consequence of this, macrophage/monocyte production of cytokines decreased when CBD was used instead of S-RBD in the interaction between macrophages/monocytes. According to the findings, the combined effects of CBD-CB₂ interaction and CBD competitive contact with macrophages resulted in a very low cytokine level that was identical to the untreated control.

This research was supported by the National Science, Research and Innovation Fund and Prince of Songkla University (PHA 64050875). The authors would like to thank the Drug Delivery System Excellence Center, Faculty of Pharmaceutical Sciences, Prince of Songkla University for providing the facilities and the National Biotechnology Center (BIOTEC), NSTDA, Ministry of Higher Education Science Research and Innovation, Thailand for providing the spike protein of SARS-CoV-2, as well as Ms. Titpawan Nakpheng for technical support. The authors would like to thank the veterinary staff of Charoen Pokphand, Foods Public Company, Ltd, Thailand for nasal tissue preparation.