DPCP (Lot No. LEH3622) was purchased from Fujifilm Wako Pure Chemical Corporation (Tokyo, Japan) (Figure 1). β-CD was supplied by Cyclo Chem Co., Ltd. (Kobe, Japan) and stored at 40°C with 82% relative humidity for 7 days. All other reagents used, including those for NMR analysis, were obtained from Fujifilm Wako Pure Chemical Corporation.

Physical mixtures (PMs) were prepared by weighing the components at a molar ratio of 1/1 and mixing them using a vortex mixer for 1 min. To prepare 3D ground mixtures (3DGMs), 500 mg of the PM (DPCP/β-CD, DPCP/HPβCD) was weighed and placed in a grinding cell along with 96 g of Φ5 mm alumina balls. Then, the mixtures were processed using a 3D ball mill (3D-80, Nagao system Inc., Kawasaki, Japan) for 30 min.

A solubility phase diagram was constructed following the method described by Higuchi et al. [23]. Excess DPCP (10 mg) was added to aqueous solutions (10 mL) of varying β-CD (0–50 mM) and HPβCD (0–10 mM) concentrations. The mixtures were shaken (25 ± 0.5°C) at 100 rpm for 24 h to obtain a suspension. The suspension was filtered through a 0.20-μm membrane filter (Advantec^®, Tokyo, Japan), and the filtrate was quantified by high-performance liquid chromatography (HPLC).

The stability constant (Ks) was calculated from the slope of the solubility phase diagram and the solubility (S ₀ ) of DPCP in the absence of β-CD or HPβCD using the following equation (Eq. 1). K S = slope / S 0 ⋅ 1 ‐ slope (1)

The complexation efficiency (CE) was calculated using the following equation (Eq. 2). CE = S 0 ⋅ K S = slope / 1 ‐ slope (2)

Samples were quantified using HPLC (LC-20ADvp, Shimadzu Corp., Kyoto, Japan). The detection wavelength was set at 293 nm. An Inertsil ODS-3 column (4.6 × 150 mm, Φ5 μm; GL Sciences, Tokyo, Japan) was utilized, with a sample injection volume of 10 μL and a column temperature of 40°C. The mobile phase consisted of acetonitrile and water in a 6:4 ratio. The retention time for DPCP was set at 7 min.

Diffraction intensities were measured using a NaI scintillation counter on a Miniflex II powder X-ray diffractometer (Rigaku Corporation, Tokyo, Japan). Cu-rays (30 kV, 15 mA) were used as the X-ray source. The X-ray diffraction measurements were conducted at a scan speed of 4°/min over a scan range of 2θ = 5–40°. The powder samples were placed flat on a glass plate for measurement.

Measurements were performed using the attenuated total reflection (ATR) method with a JASCO FT/IR-4600 instrument (JASCO Corporation, Tokyo, Japan). The measurement conditions were as follows: integration frequency, 16; resolution, 4 cm⁻¹; and measurement range, 400 cm⁻¹–4,000 cm⁻¹.

A field-emission SEM (JSM-IT800, Schottky Field Emission Scanning Electron Microscope, JOEL Ltd., Tokyo, Japan) was used. Each sample was coated with a thin layer of gold for 60 s to enhance conductivity and observed under a pressurized voltage of 1 kV.

The dissolution test was conducted using a dissolution tester (NTR-593, Toyama Sangyo, Osaka, Japan) in accordance with the JP18 revised paddle method. The dissolution medium consisted of 300 mL of distilled water maintained at 37 ± 0.5°C, with stirring at 50 rpm. DPCP (50 mg) was placed in a paddle. Samples of 10 mL were collected at 5, 10, 15, 60, and 120 min and, then, filtered through a 0.20-μm membrane filter (DISMIC 25AS, Advantec^®). To maintain a constant volume of dissolved solution, distilled water (10 mL) was added at the same temperature and volume after each sample collection. The obtained filtrates were diluted with a mixture of acetonitrile and distilled water (6/4), and subsequently analyzed using HPLC.

An AVANCE NEO 600 NMR system (Bruker, Billerica, MA, United States) was used. Deuterium oxide (D₂O) was employed as the solvent, and the measurement conditions were as follows: resonance frequency, 600.13 MHz; measurement temperature, 25°C; integration frequency, 32; pulse width, 90°; and waiting time, 3 s (ROE) and 4 s (NOE).

The spleen was removed from a male C57BL/6 mouse (Tokyo Laboratory Animals Science Co., Ltd., Tokyo, Japan), minced with scissors, and filtered through a 55-μm mesh. Thereafter, the cells were hemolyzed using a red blood cell lysis buffer (Pluriselect-USA Inc., CA, United States) and resuspended in RPMI 1640 medium (Sigma-Aldrich Co., St. Louis, MO, United States) by centrifugation. Isolated splenocytes were confirmed to have >98% viability using trypan blue exclusion assay. Splenocytes (5.0 × 10⁶ cells/mL) were plated and cultured for 48 h in RPMI 1640 medium containing 5% fetal calf serum (GE Healthcare Lifesciences, Inc., IL, United States), 0.01 μg/mL lipopolysaccharide (LPS) (Sigma-Aldrich Co.), and various reagents. The reagents added to splenocytes were CPDP, HPβCD/CPDP, and β-CD/CPDP.

The IFN-γ concentration in the culture medium (100 μL) was quantified using an ELISA kit (Proteintech Inc., Rosemont, IL, United States), following the manufacturer’s instructions. After treating the culture medium with various reagents for 48 h in primary cultures of splenocytes, the samples were collected and centrifuged at 500 × g for 5 min to remove any particulate matter. Subsequently, the supernatant (100 μL) was used for the quantification of IFN-γ concentration.

NIH3T3 (NIH3T3/14-1) mouse fibroblast cells were obtained from Riken BioResource Center (Ibaragi, Japan) and maintained in Dulbecco’s modified eagle medium (DMEM) supplemented with 10% fetal bovine serum and buffered with 44 mM NaHCO₃ at 37°C in a 5% CO₂ incubator [24].

For cell injury experiments, a cell suspension was prepared in the aforementioned medium to a concentration of 1.2 × 10³ viable cells/mL. Then, 2 mL of the suspension was seeded onto a 35-mm-φ six-well plate (2.5×10² viable cells/cm²). The cells were cultured in a CO₂ incubator at 37°C with 5% CO₂ for 96 h. To quantify PGE₂ concentration in the culture medium, cells were prepared similarly at a concentration of 1.0 × 10² viable cells/mL and, then, seeded in 2-mL portions (2.5 × 10² viable cells/cm²) and cultured for 96 h under the same conditions.

After adding 2 mL of phosphate-buffered saline (PBS) (−) to the NIH-3T3 cells, the dish was irradiated with UV rays from the bottom for 10 s using an 8-m W/cm², 254-nm UV lamp (15 W × 6). Immediately after UV irradiation, PBS (−) with 2.5 mM glucose and 0.9 mM CaCl₂ was added. The solution was removed by suction and replaced with the conditioning culture medium (PBS (−) with 2.5 mM glucose and 0.9 mM CaCl₂) to which each drug had been added. The cells were subsequently cultured in a CO₂ incubator (37°C, 5% CO₂) for 4 h [25].

After adding each drug (to the UV-irradiated NIH-3T3 cells and culturing for 4 h, 50 μL of the conditioning culture medium was sampled from each dish. The concentration of PGE₂ in the culture medium was measured using an EIA kit (Arbor Assays, Ann Arbor, MI, United States). The PGE₂ levels were determined colorimetrically at 450 nm. The control groups consisted of cells subjected to UV irradiation without the addition of any drugs, as well as cells without UV irradiation and without drug treatment [26].

The results are presented as means ± standard error of the mean from 3-4 separate experiments for pharmacological measurements. Statistical comparisons between the control and treatment groups were performed using Student's t-test, with significance set at P < 0.05. The statistical analysis was conducted using Statcel-the Useful Addin Forms on Excel (fourth edition; OMS Publishing, Tokyo, Japan).

This study was approved by the Institutional Animal Care and Use Committee of the Josai University (No. JU 22037). All procedures involving animals were conducted in accordance with the regulations outlined in the Josai University Guidelines for the Care and Use of Laboratory Animals.

Solubility phase diagrams were constructed to analyze the solubility changes of DPCP/β-CD and DPCP/HPβCD inclusion complexes in solution, which allowed for the calculation of stability constants and prediction of inclusion molar ratios (Figure 2). According to the classification of solubility phase diagrams, they can be categorized into type A (soluble complexes) and type B (insoluble complexes). Type A includes AL (linear diagram), Ap (positive deviation from linearity), and AN (negative deviation from linearity) types, while type B includes BS (the complex has some but limited solubility) and BI (the complex is insoluble) types [27].

DPCP/β-CD, the solubility phase diagram, exhibited a linear AL-type relationship, proportional to the increase in β-CD concentration. The Ks was calculated to be 200.4 M⁻¹, with a complexation efficiency (CE) of 1.8 × 10⁻². Similarly, the DPCP/HPβCD solubility phase diagram also showed an AL-type linear relationship, indicating an increase in DPCP solubility proportional to the HPβCD concentration. The Ks for DPCP/HPβCD was calculated to be 233.5 M⁻¹, with a CE of 2.3 × 10⁻².

The results from the solubility phase diagrams indicate that both DPCP/β-CD and DPCP/HPβCD form linear AL-type complexes as per the Higuchi et al. classification, suggesting a 1/1 inclusion molar ratio in solution. The higher stability constant for DPCP/HPβCD compared to DPCP/β-CD suggests that DPCP has a greater tendency to form a complex with HPβCD than with β-CD.

The researchers conducted PXRD measurements to examine changes in the crystalline state of solid dispersions of DPCP/β-CD and DPCP/HPβCD that were prepared using 3D milling (Figure 3). Characteristic diffraction peaks for DPCP were observed at 2θ = 6.7°, 19.4°, and 30.0°, while β-CD exhibited characteristic peaks at 2θ = 12.3° and 18.6°. HPβCD displayed a halo pattern without any characteristic diffraction peaks.

For the DPCP/β-CD solid dispersion prepared by physical mixing (PM), both DPCP and β-CD peaks were observed. However, in the 3DGM of DPCP/β-CD, the peaks from DPCP and β-CD disappeared, and a halo pattern was observed. Similarly, in the DPCP/HPβCD solid dispersion PM, the DPCP-derived peak was present, but in the 3DGM of DPCP/HPβCD, both the DPCP- and β-CD-derived peaks disappeared, resulting in a halo pattern.

Previous studies have reported that co-milling can disrupt the crystalline structure, leading to the formation of amorphous solids that exhibit a halo pattern [28]. It has also been suggested that mechanical energy applied during milling and friction can contribute to the formation of amorphous inclusion complexes [29]. Based on the PXRD results, the characteristic diffraction peaks of DPCP, β-CD, and HPβCD were not observed, indicating the formation of inclusion complexes with a presumed molar ratio of 1/1 in the 3DGM of DPCP/β-CD and DPCP/HPβCD.

The PXRD results suggested that DPCP/β-CD and DPCP/HPβCD in the solid state form inclusion complexes with a DPCP/CD molar ratio of 1:1. To further confirm the intermolecular interactions within these complexes, FT-IR absorption spectroscopy was conducted (Figure 4). For DPCP, the FT-IR spectrum exhibited characteristic peaks at 688 cm⁻¹ for the -CH group derived from the aromatic ring, 1,606 cm⁻¹ for C=C, and 1844 cm⁻¹ for the ketone group.

In the physical mixtures (PMs) of DPCP/β-CD and DPCP/HPβCD, the original peaks derived from DPCP and β-CD were observed (Figure 4A). However, in the 3DGM of DPCP/β-CD, shifts were observed in the DPCP-derived peaks: the -CH group peak shifted to 689 cm⁻¹, the C=C peak to 1,612 cm⁻¹, and the ketone group peak to 1850 cm⁻¹. Similarly, in the 3DGM of DPCP/HPβCD, the -CH group peak shifted to 689 cm⁻¹, the C=C peak to 1,616 cm⁻¹, and the ketone group peak to 1849 cm⁻¹ (Figure 4B).

Additionally, in the 3DGM samples, the peaks around 3,348 cm⁻¹ for β-CD and 3,376 cm⁻¹ for HPβCD, which correspond to the -OH groups, shifted to 3,351 cm⁻¹ and 3,390 cm⁻¹, respectively. The shifts in these peaks, which are attributed to DPCP and β-CD alone, suggest the formation of intermolecular interactions between the C=O and C=C groups of DPCP, the aromatic ring -CH, and the -OH groups of β-CD. These interactions confirm the inclusion complex formation in the solid state through mixed grinding.

In the solid state, 3DGMs of DPCP/β-CD and DPCP/HPβCD, both with a 1:1 M ratio, were suggested to form inclusion complexes. PXRD measurements indicated changes in the crystal state of these complexes. To evaluate the microstructure and surface morphology of the prepared inclusion complexes, SEM measurements were performed. SEM, with its high resolution, can observe fine structures on the nanometer scale, providing detailed insights into the microstructure and surface characteristics of the materials (Figure 5).

The SEM images revealed that the crystal surface of intact DPCP appeared smooth and columnar, with particle sizes around 100 μm (Figure 5A). Intact β-CD displayed a smooth and angular crystal surface with particle sizes around 150 μm (Figure 5B). In the PM of DPCP/β-CD (1/1), no significant changes in the particle surfaces of DPCP and β-CD were observed (Figure 5C).

Generally, when β-CD forms inclusion complexes, cubic crystals are reported to appear. In the 3DGM of DPCP/β-CD (1/1), the particle size was reduced to around 100 μm due to milling, and the particles exhibited a cubic shape with fine surface attachments, suggesting the formation of an inclusion complex (Figure 5D).

HPβCD intact exhibited a smooth and spherical crystal surface with particle sizes around 100 μm (Figure 5E). In the PM of DPCP/HPβCD (1/1), the particle surfaces of DPCP and HPβCD remained unchanged (Figure 5F). Typically, when HPβCD forms inclusion complexes, crystals with a folded structure are observed [30]. In the 3DGM of DPCP/HPβCD (1/1), the particle size was further reduced to approximately 30 μm, and a plate-like folded structure with fine particles on the surface was observed, indicating the formation of an inclusion complex (Figure 5G).

These SEM observations suggest that the particle shapes and surface morphologies changed due to the formation of inclusion complexes through the 3DGM process, confirming the successful inclusion of DPCP with β-CD and HPβCD.

Solid-state characterization confirmed the formation of inclusion complexes in 3DGMs of DPCP/β-CD (1/1) and DPCP/HPβCD (1/1). To further investigate the change in solubility of DPCP with each inclusion complex, dissolution tests were conducted using samples of intact DPCP, 3DGM (DPCP/β-CD = 1/1), and 3DGM (DPCP/HPβCD = 1/1) (Figure 6).

Initially, the dissolution concentration of intact DPCP at 15 min was measured at 5.70 μg/mL. In contrast, both 3DGM (DPCP/β-CD = 1/1) and 3DGM (DPCP/HPβCD = 1/1) exhibited faster dissolution of DPCP compared to DPCP alone, with concentrations measured at 17.5 μg/mL and 58.4 μg/mL, respectively, from study initiation. This significant increase in the solubility of DPCP in the presence of inclusion complexes can be attributed to two main factors: the formation of inclusion complexes between DPCP and β-CD or HPβCD, and the increase in specific surface area due to the smaller particle size achieved through milling.

Notably, the solubility of DPCP was higher in 3DGM (DPCP/HPβCD = 1/1) compared to that in 3DGM (DPCP/β-CD = 1/1). This suggests that there may be differences in the inclusion mode or interaction strength between DPCP/β-CD and DPCP/HPβCD complexes, in addition to the findings from the physical properties evaluation in the solid state.

NMR spectroscopy was employed to elucidate the inclusion modes of DPCP with β-CD and HPβCD in the prepared 3DGM at a 1:1 mole ratio. In the case of 3DGM (DPCP/β-CD = 1/1), cross-peaks were identified between the -CH (A; 7.94 ppm) of DPCP and the -OH groups of β-CD (H-6; 3.71 ppm, H-3; 3.82 ppm), as well as between -CH (B; 7.55 ppm) of DPCP and -OH (H-6; 3.71 ppm) of β-CD (Figure 7A) [31–33]. These findings indicate that DPCP is encapsulated by β-CD from the wide edge (H-3) to the narrow edge (H-6), encompassing the benzene ring of DPCP.

Conversely, in 3DGM (DPCP/HPβCD = 1/1), cross-peaks were observed between -CH of DPCP (A; 7.95 ppm) and -OH of HPβCD (H-6; 3.72 ppm), as well as between -CH (B; 7.55 ppm) of DPCP and -OH (H-6; 3.72 ppm) of HPβCD (Figure 7B). These results from ROESY and NOESY NMR measurements suggest that HPβCD encapsulates DPCP such that the CD surrounds the benzene ring of DPCP from the narrow edge (H-6) to the wide edge (H-3).

The distinct cross-peak patterns observed in the NMR spectra of DPCP/β-CD and DPCP/HPβCD complexes indicate different inclusion modes between DPCP and the CDs in solution. Specifically, DPCP/β-CD predominantly interacts with the -OH groups of β-CD, while DPCP/HPβCD interacts in a manner where HPβCD surrounds the benzene ring of DPCP.

These findings further support the hypothesis that the 3DGM preparation method affects the inclusion mode and interaction strength of DPCP with β-CD and HPβCD, influencing their solubility enhancement properties observed in dissolution tests.

In this study, 3DGM samples indicated that DPCP/β-CD and DPCP/HPβCD form inclusion complexes at a molar ratio of 1/1, albeit with distinct inclusion modes. DPCP is recognized for its ability to reduce elevated serum IFN-γ levels in patients with AA [34]. Thus, we investigated the anti-inflammatory effects of DPCP inclusion complexes (DPCP/β-CD and DPCP/HPβCD) using LPS-induced IFN-γ as a model to assess their efficacy (Figure 8).

The results demonstrated that DPCP alone significantly decreased LPS-induced IFN-γ levels at a concentration of 10⁻⁷ M compared to the control group. Conversely, β-CD and HPβCD alone did not show any effect on LPS-induced IFN-γ levels. Furthermore, both DPCP/β-CD and DPCP/HPβCD inclusion complexes significantly attenuated LPS-induced IFN-γ at a DPCP concentration of 10⁻⁷ M compared to the control group. Particularly, HPβCD/DPCP exhibited a more pronounced reduction in LPS-induced IFN-γ at lower concentrations (10⁻⁸ M), indicating a stronger inhibitory effect than that observed with DPCP alone or DPCP/β-CD. Additionally, the HPβCD group without encapsulated DPCP did not influence the LPS-induced IFN-γ levels.

These findings underscore that DPCP/β-CD and DPCP/HPβCD exert differential effects on reducing LPS-induced IFN-γ levels. The enhanced efficacy of HPβCD/DPCP suggests potential advantages in therapeutic applications aimed at mitigating conditions associated with elevated IFN-γ, such as AA.

We investigated the protective effects of each drug on UV-irradiated NIH-3T3 cells subjected to UV damage (Figure 9). Following 4 h of UV irradiation, the number of viable NIH-3T3 cells decreased to approximately 70% compared to that of the untreated group (no UV irradiation, no drug added). In each drug-treated group, the decrease in cell viability was significantly attenuated in a dose-dependent manner.

The group supplemented with DPCP showed a dose-dependent suppression of UV-induced cell damage, with recovery to 85% of the non-UV-irradiated group observed at 10⁻⁶ M DPCP. Similarly, the β-CD/DPCP group, where DPCP was encapsulated in β-CD, exhibited a comparable protective effect against cell damage. In contrast, the HPβCD/DPCP group showed significant inhibition of cell damage starting from 10⁻⁸ M HPβCD/DPCP, with recovery to 97% of the non-UV-irradiated group at 10⁻⁷ M HPβCD/DPCP.

Furthermore, the release of PGE₂ into the culture medium of UV-irradiated NIH-3T3 cells after 4 h was evaluated (Figure 10). PGE₂ is a known chemical mediator involved in inflammation following UV radiation exposure. The addition of DPCP and its inclusion complexes led to a significant increase in PGE₂ release compared to the untreated group. However, DPCP and its complexes also showed a dose-dependent suppression of PGE₂ release induced by UV radiation. At 10^−-6 M DPCP, PGE₂ release was reduced to 50% of the UV-irradiated control, while β-CD inclusion and HPβCD groups exhibited inhibitory effects at lower concentrations. Notably, the HPβCD group at 10⁻⁷ M suppressed PGE₂ release to 25% of the UV-irradiated control, indicating a potent anti-inflammatory effect.

These results suggest that the addition of DPCP and its complexes protects UV-damaged NIH-3T3 cells, preserves cell viability, and reduces PGE₂ release, which is indicative of anti-inflammatory activity. The enhanced efficacy observed with HPβCD/DPCP compared to β-CD/DPCP may be attributed to differences in the inclusion modes of these complexes, highlighting the importance of inclusion complex formation in modulating biological activity.

The ex vivo cell studies aimed to validate whether inclusion complexes with β-CD derivatives could preserve the anti-inflammatory effects of DPCP. Results from these studies confirmed that both β-CD and HPβCD inclusion complexes maintained the anti-inflammatory properties of DPCP. Particularly noteworthy was the finding that the DPCP/HPβCD complex exhibited anti-inflammatory effects at lower concentrations compared to the DPCP/β-CD complex, which represents a novel discovery. This difference in inclusion modes, elucidated through NOESY NMR measurements, suggests a potential pharmacological advantage where DPCP demonstrates enhanced anti-inflammatory efficacy.

However, the safety and pharmacokinetic profiles of DPCP/β-CD inclusion complexes, which typically require comprehensive in vivo studies, remain a limitation of this research and should be addressed in future investigations.

The study unequivocally demonstrated that DPCP/β-CD complexes enhance solubility and retain anti-inflammatory properties. A clinical challenge highlighted is the current use of acetone, an organic solvent, in preparing DPCP for AA treatment. Based on these findings, future efforts will focus on scaling up the preparation method using 3D ball milling and exploring alternative formulation methods. Addressing the stability of DPCP/β-CD complexes in both solid-state and solution phases remains a critical challenge that requires further investigation.

As this study focused primarily on improving the solubility of DPCP, future research directions should include investigating skin permeability for potential clinical applications, thereby addressing the broader implications of these findings.