Properties of tropolone/γ-cyclodextrin complexes prepared using different methods

World Journal of Pharmaceutical Sciences ISSN (Print): 2321-3310; ISSN (Online): 2321-3086 Available online at: http://www.wjpsonline.org/ Original Article Properties of tropolone/γ-cyclodextrin complexes prepared using different methods Rina Suzuki, Yutaka Inoue*, Isamu Murata and Ikuo Kanamoto Laboratory of Drug Safety Management, Faculty of Pharmacy and Pharmaceutical Sciences, Josai University; 1-1 Keyakidai, Sakado-shi, Saitama, 3500295, Japan Received: 20-06-2017 / Revised Accepted: 05-08-2017 / Published: 02-09-2017 ABSTRACT The aim of the present study is to evaluate the properties of the inclusion complex of tropolone (TPN)/ -cyclodextrin ( CD) prepared by cogrinding method and coprecipitation method. The physical properties of the preparation were evaluated by differential scanning calorimetry, powder X-ray diffraction, infrared absorption spectra, and 1H-1H NOESY NMR spectrum. Intermolecular interactions in the solid state were confirmed to be in the molar ratios TPN/ CD = β/1 and TPN/ CD = 4/1 in the cogrinding method and molar ratio TPN/ CD = β/1 in the coprecipitation method. In addition, in GM (TPN/ CD = 4/1), two molecules of TPN were encapsulated in the molecular space formed between CD. Therefore, it was suggested that different inclusion structures of TPN/ CD complexes were formed using different preparation methods. Keywords: Tropolone, Cyclodextrin, Ground mixture, Copreciptate, Molecular interaction Address for Correspondence: Dr. Yutaka Inoue, Laboratory of Drug Safety Management, Faculty of Pharmacy and Pharmaceutical Sciences, Josai University; 1-1 Keyakidai, Sakado-shi, Saitama, 3500295, Japan; E-mail: yinoue@josai.ac.jp How to Cite this Article: Rina Suzuki, Yutaka Inoue, Isamu Murata and Ikuo Kanamoto. Properties of tropolone/ -cyclodextrin complexes prepared using different methods. World J Pharm Sci 2017; 5(9): 250261. This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercialShareAlike 4.0 International License, which allows adapt, share and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms. © 2017 World J Pharm Sci Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 used. For example, in the case of actarit, an antirheumatic drug, it has been reported that inclusion complexes of different molar ratios are obtained using cogrinding and freeze-drying methods, and the inclusion structures formed by both the methods are different [13]. Higashi et al. reported that salicylic acid is encapsulated in the outer molecular space formed not only within the cavity of CD, but also between CD [14]. This specific inclusion mode can be applied for the development of new medicines in the future. Thus, elucidating the mechanism of formation of a new inclusion complex via encapsulation of guest molecules in the molecular space formed between CDs is useful in the development of future products. Authors previously reported that the TPN derivative HT forms inclusion complexes with α-, -, and CDs [15, 16]. INTRODUCTION Tropolone (TPN, 2-hydroxy-2,4,6cycloheptatriene-1-one) is a non-benzenoid aromatic compound with a planar seven-member ring structure, and it is an isomer of benzoic acid. Notably, TPN has been reported to form intra- and intermolecular hydrogen bonds and exhibit specific properties. It possesses pharmacological and biochemical effects such as antibacterial [1], antiinflammatory [2], antioxidant [3], and antitumor [4] activities. Owing to these properties, TPN derivatives colchicine and hinokitiol (HT) have been applied and studied in various fields as pharmaceuticals and quasi-drugs [5, 6]. In addition, the TPN skeleton can be induced to various compounds, and it would be possible to develop a wide range of structures as a new pharmacophore in the pharmaceutical field in the future. The purpose of this study is to evaluate the mechanism of TPN/ CD inclusion complex formation using the space where TPN is formed between CDs, using different preparation methods. Cyclodextrin (CD) is a cyclic polysaccharide in which D-glucopyranose is cyclically bound by a α1, 4 bond. It is classified as α-cyclodextrin (αCD), -cyclodextrin ( CD), and -cyclodextrin ( CD) according to the number of glucopyranose units, and these CDs are widely used as host molecules in the formation of inclusion complex. CDs are hydrophilic near the edge and on the outside of the ring, while the interior cavity shows hydrophobic properties. It is known that a host is capable of including a guest to form an inclusion complex by hydrophobic interactions in aqueous solution [7]. The methods for preparing inclusion complexes include cogrinding [8], coprecipitation [9], freeze drying [10], spray drying [11], and sealed heating [12] methods. There are reports that different inclusion structures are formed depending on the preparation method, even when the same CD is MATERIALS AND METHODS Materials: TPN used as a bulk powder was purchased from Sigma-ALDRICH Ltd. (Fig.1). CD was donated by Cyclo Chem Co. Ltd (Tokyo, Japan) and used after storage at 40°C at a relative humidity of 82% for 7 days [17]. The moisture content of CDs was confirmed by coulometric titration using Karl Fischer moisture meter (CA-06, Mitsubishi Chemical Co., Ltd). All other reagents were special grade reagents manufactured by Wako Pure Chemical Industries, Ltd. Fig. 1 Chemical Structures of (a) Tropolone (TPN) and (b) -cyclodextrin ( CD) Preparation of ground mixture molar ratios of respectively, and mixer to prepare a physical mixture (PM). The ground mixture (GM) was prepared from the PM. For each PM, 1 g of material was charged in an alumina cell and cogrinding was conducted for 30 physical mixture (PM) and (GM): HT/ CD was weighed at 5/1, 4/1, 3/1, 2/1, and 1/1, mixed for 1 min using a vortex 251 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 min using a vibration rod mill (TI-500ET, CMT Co., Ltd) to obtain GM. 1 H- nuclear magnetic resonance (NMR) measurement: NMR spectra were obtained using a Varian NMR System 400 MHz (manufactured by Agilent Technologies). Dimethyl sulfoxide-d6 (DMSO-d6) was used as a solvent, and the measurement was carried out with a pulse-width of 90°, delay time of 6.4 μs, scan time of 3.723 s, and 128 integration steps at 26°C. Preparation of humidification: GM was conditioned and recrystallized by storing in a desiccator at 40°C and 82% relative humidity in the presence of a saturated aqueous solution of potassium chloride. Preparation of coprecipitate (CP): To 5 mL of TPN aqueous solution (0.10 mol/L), 5 mL of CD aqueous solution (0.13 mol/L) was added in portions, stirred for 6 h at room temperature, and allowed to stand at room temperature for 24 hours. Then, the precipitate was separated by filtration and dried in a desiccator, under vacuum for 24 h at room temperature. 1 H-1H nuclear overhauser effect difference spectroscopy (NOESY) NMR measurement: 1H1 H NOESY spectra were recorded using the Varian NMR System 400 MHz (manufactured by Agilent Technologies). Using D2O as a solvent, the measurement was carried out with a pulse-width of 90°, relaxation time 500 ms, scan time 0.500 s, and cumulative frequency of 256 integration steps at 25°C. Physicochemical characterization Differential scanning calorimetry (DSC): The thermal behavior of samples was recorded using a differential scanning calorimeter (Thermo plus Evo, Rigaku). Approximately 2 mg of a sample was filled in a sealed aluminum pan, and the measurement was conducted under N2 gas flow rate of 60 mL/min and heating rate of 5°C /min. RESULTS AND DISCUSSION DSC analysis: It has been reported that inclusion complexes are formed by intermolecular interactions, resulting in changes in their thermal behavior [19]. Therefore, thermal behavior of TPN/ CD complexes was examined by DSC measurement (Fig. 2). The endothermic peak derived from the melting point of TPN was confirmed to be around 57°C in TPN crystals and ground TPN alone (Fig. 2a, 2b). For PM (TPN/ CD=β/1), PM (TPN/ CD=4/1), and GM (TPN/ CD=5/1), an endothermic peak derived from the melting of TPN crystal was observed at around 57°C (Fig. 2e-f, 2k). However, for GM (TPN/ CD=1/1), GM (TPN/ CD=β/1), GM (TPN/ CD=γ/1), GM (TPN/ CD=4/1), and CP (TPN/ CD), no endothermic peak derived from the melting of TPN crystal was observed (Fig. 2g-j, 2l). In a previous study, it has been reported that changes in thermal behavior indicate the inclusion complexes of guest drug and CD in solid dispersion or the formation of inclusion complexes with different properties [20]. From the results of DSC measurement, it was inferred that intermolecular interaction is formed between TPN and CD. The low temperature shift of the endothermic peak derived from the melting of TPN in GM and the decrease in caloric value were attributed to the mechanochemical effect caused by the mechanical energy developed in the grinding method. Thermogravimetry (TG): The thermal behavior of the samples was recorded using a differential scanning calorimeter (Thermo plus Evo, Rigaku). Approximately 10 mg of a sample was placed in an aluminum pan, and the measurement was conducted under N2 gas flow rate of 200 mL/min and heating rate of 5°C /min. Powder X-ray diffraction (PXRD): PXRD was performed on an X-ray diffractometer (MiniFlex II, Rigaku), and the diffraction intensity was measured with a NaI scintillation counter. Cu line (30 kV, 15 mA) with a scan speed of 4°/min over the βθ ranges of 3-35° were used to carry out X-ray diffraction measurement. The powder sample was filled in a glass plate so that the sample plane became flat and measured. Fourier transform infrared (FT-IR) spectroscopy: FT-IR absorption spectroscopy of the samples was performed using the KBr tablet method and recorded using a spectrometer (FT/IR410, JASCO). The number of integration steps was 32, resolution was 4 cm-1, and measurements were recorded in the wavenumber range of 4000-400 cm1 . For preparing tablets, potassium bromide (KBr) was added to the sample at a weight ratio of 1/10 (sample/KBr), mixed, and compressed by a manual press. Background correction was performed using KBr only tablet. PXRD analysis: PXRD measurement was carried out to investigate the crystalline state of TPN/ CD in the cogrinding method and coprecipitation method. 252 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 Fig. 2 DSC curves of TPN/ CD systems (a) TPN, (b) TPN ground, (c) CD, (d) CD ground, (e) PM (TPN/ CD=2/1), (f) PM (TPN/ CD=4/1), (g) GM (TPN/ CD=1/1), (h) GM (TPN/ CD=2/1), (i) GM (TPN/ CD=3/1), (j) GM (TPN/ CD=4/1), (k) GM (TPN/ CD=5/1), (l) CP (TPN/ CD) Characteristic peaks of TPN were observed at βθ = 14.4°, 25.3° in TPN crystal and ground TPN alone (Fig. γa, γb). The peak of CD alone was observed at βθ = 14.4°, 18.γ° (Fig. γc). In PM (TPN/ CD = β/1) and PM (TPN/ CD = 4/1), diffraction peaks derived from TPN crystals were observed around βθ = 14.5°, β4.7°, and diffraction peaks derived from CD were observed near βθ = 1β.β°, 18.8° (Fig. 3e, 3f). However, in the case of GM (TPN/ CD = 1/1), GM (TPN/ CD = β/1), and GM (TPN/ CD = γ/1), diffraction peaks derived from the TPN crystal and CD showed a halo pattern (Fig. 3g-i). A previous study reported that crystallization occurs when an amorphous sample is stored in a humidity-controlled environment [24]. Therefore, PXRD measurement was performed after GM (TPN/ CD = β/1) and GM (TPN/ CD = 4/1), which showed a halo pattern in cogrinding, were stored under humidity-controlled conditions. Diffraction peaks derived from TPN and CD were not found in the diffraction pattern even after crystallization. The diffraction peaks of humidity-conditioned GM (TPN/ CD = β/1) and humidity-conditioned GM (TPN/ CD = 4/1) were approximately βθ = 7.5°, 12.0°, and 16.7° (Fig. 3l, 3m). When inclusion complexes are formed by tetragonal columnar type CD, characteristic diffraction peaks are known to be observed around βθ = 7.4°, 1β.1°, and 16.5° [24]. In the diffraction pattern of humidityconditioned GM (TPN/ CD = β/1), humidityconditioned GM (TPN/ CD = 4/1), and CP, diffraction peaks similar to the tetragonal columnar type structure of CD (βθ = 7.γ°, 1β.0° and 16.5°) were observed. From the results, it was speculated that a complex of TPN and CD with a tetragonal columnar type structure was formed. It is known that inclusion complexes formed by cogrinding method become amorphous [21]. During cogrinding with CD, the regularity of the crystal lattice in the TPN crystal structure is disturbed and crystallinity declines; thus, there is a possibility of the complex becoming amorphous or the mechanochemical reaction progresses to form inclusion complexes. It was presumed that the complex might become amorphous in the process of changing to a different crystalline structure from the TPN crystal [22, 23]. 253 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 Fig. 3 PXRD patterns of TPN/ CD systems (a) TPN, (b) TPN ground, (c) CD, (d) CD ground, (e) PM (TPN/ CD=2/1), (f) PM (TPN/ CD=4/1), (g) GM (TPN/ CD=1/1), (h) GM (TPN/ CD=2/1), (i) GM (TPN/ CD=3/1), (j) GM (TPN/ CD=4/1), (k) GM (TPN/ CD=5/1), (l) GM (TPN/ CD=2/1) after storage at 40ºC and 82% for 7days, (m) GM (TPN/ CD=4/1) after storage at 40ºC and 82% for 7days, (n) CP (TPN/ CD) ■:TPN, ○: CD, △: tetragonal columnar form 260ºC was confirmed. This weight reduction corresponded to 98.3% of TPN contained in the sample. In CP, a weight loss of about 14.7% was confirmed from around 153-260ºC. According to the report by Daniel, the weight loss of drugs observed above 110ºC is considered to be due to the formation of complexes [25]. It is considered that the respective weight losses observed from around 157 ºC in GM and CP were due to TPN/ CD complex formation. Similar weight loss was observed in GM (TPN/ CD = β/1) and CP (TPN/ CD), which suggested the presence of TPN with the same molecular state. However, in GM (TPN/ CD = 4/1), two-stage weight loss was confirmed at 100-178ºC and 178-260°C, indicating that GM (TPN/ CD = β/1) and TPN had different molecular states. TG analysis: TG measurement was carried out to examine the weight change of TPN molecules involved in complex formation (Fig. 4). In TPN alone, around 99% weight loss from the TPN crystal was confirmed from around 60ºC. In addition, weight loss of 11.5% in GM (TPN/ CD = β/1), 6.0% in GM (TPN/ CD = 4/1), and 5.4% in CP (TPN/ CD) was observed between γ0-100ºC. These weight losses were presumed to be from the degree of decrease in the TG curve, which was suggested to be due to the evaporation of crystal water or absorbed water of CD. For GM (TPN/ CD = 2/1), a weight loss of approximately 11.2% was confirmed from around 157-260ºC. This weight reduction corresponded to 91.3% of TPN contained in the sample. For GM (TPN/ CD = 4/1), a weight loss of approximately 8.0% from around 100178ºC and approximately 9.7% from around 178- 254 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 Fig. 4 TG curves of TPN/ CD systems (a) TPN, (b) GM (TPN/ CD=2/1), (c) GM (TPN/ CD=4/1), (d) CP (TPN/ CD) 1 H-NMR analysis: From the results of DSC, PXRD, and TG measurements, intermolecular interactions in both GM and CP can be inferred. 1 H-NMR spectrum measurement was carried out to investigate the inclusion molar ratio of CP [26]. The results of the 1H-NMR spectrum measurement of TPN, CD, GM (TPN/ CD = 4/1), and CP are shown in Fig. 5. In TPN, a signal derived from the seven-membered ring hydrogen was observed around 7.0-7.5 ppm. In CD, signals derived from hydrogen and hydroxyl groups of the glucose unit were observed. In CP, signals from TPN and CD were confirmed respectively. Since the number of protons of the signal derived from the seven- membered ring hydrogen of TPN observed around 7.0-7.41 ppm was 1.32, it was shown that the number of protons per hydrogen atom of TPN in CP was 0.264. In addition, since the number of protons of the signal derived from hydrogen number 1 in the glucose unit of CD was 1, it was shown that the number of protons per hydrogen atom of CD was 0.1β5. From this result, it’s can suggest that when the inclusion molar ratio of CP was calculated using the formula, the molar ratio of inclusion complex formation between CD and TPN was TPN/ CD = β/1 when TPN/ CD = β.15/1 [27]. 255 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 E D A C A-D B Tropolone (a) 2, 4 H 4 6 5HO O H HO 3, 6, 5 OH 2 OH 1 3 H H Glucopyranose (b) 2, 4 (c) (1.00) (2.46) A-D 1 3, 6, 5 2, 4 (d) 9 10 8 (1.00) (1.32) A-D 7 6 1 5 (ppm) 3, 6, 5 4 3 2 1 0 1 Fig.5 H-NMR (DMSO-d6) spectra of TPN/ CD systems (a) TPN, (b) CD, (c) GM (TPN/ CD=4/1), (d) CP (TPN/ CD) measurement was performed to examine the molecular state of the inclusion complex (Fig. 6). FT-IR analysis: From the results of DSC, TG, and PXRD measurements, intermolecular interaction between TPN and CD was inferred. FT-IR spectroscopy is a useful analytical method to confirm the formation of inclusion complexes in solid state [28]. Therefore, FT-IR spectrum 256 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 (a) 2998 3198 1613 1545 (b) 1647 % Transmission 2929 3387 (c) 1613 2929 1540 (d) 3387 2929 1613 1545 (e) 3387 2925 (f) 3375 2924 (g) 3365 2925 3361 3800 3300 1543 1612 1543 1611 1643 1610 1700 1500 2800 -1 Wavenumber (cm ) 1300 1100 900 Fig. 6 FT-IR spectra of TPN/ CD systems (a) TPN, (b) CD, (c) PM (TPN/ CD=2/1), (d) PM (TPN/ CD=4/1), (e) GM (TPN/ CD=2/1), (f) GM (TPN/ CD=4/1), (g) CP (TPN/ CD) and CP, the absorption peak near 3198 cm-1 derived from the hydroxyl group (OH stretching vibration) disappeared. It has been reported that TPN forms a dimer through intermolecular hydrogen bonding in its crystal structure [29]. The peak shift observed in this study was presumed to be due to the cleavage of intermolecular hydrogen bond forming the TPN dimer and formation of new intermolecular interactions between TPN and CD. In general, CDs incorporate water molecules when guest molecules are not clathrated. Moreover, the water molecules and guest molecules in the CD cavity exchange with each other leading to a stable energy state during the formation of inclusion complexes [30]. The peaks derived from crystal water present inside the CD ring that were confirmed around 1647 cm-1 were lost in GM (TPN/ CD = β/1), GM (TPN/ CD = 4/1), and CP. Thus, it was inferred that intermolecular interaction with the guest molecules was due to the dehydration of water of crystallization in CD [31]. In the TPN crystal, an absorption peak in the vicinity of 1613 cm-1 derived from the carbonyl group (C=O stretching vibration) in the TPN molecular structure and an absorption peak near 3198 cm-1 derived from the hydroxyl group (O-H stretching vibration) were observed using FT-IR (Fig. 6a). In CD alone, a broad absorption peak derived from hydroxyl group (O-H stretching vibration) was confirmed between 3800-3100 cm-1 centered on 3387 cm-1 (Fig. 6b). In PM (TPN/ CD = β/1) and PM (TPN/ CD = 4/1), the absorption peaks that were derived from carbonyl and hydroxyl groups in the TPN molecular structure were similar to that in TPN crystal (Fig. 6c, 6d). However, the absorption peak in the vicinity of 1613 cm-1 derived from the TPN carbonyl group (C=O stretching vibration) is 1612 cm-1 in GM (TPN/ CD = β/1). It was observed that for the absorption peak in GM (TPN/ CD = 4/1), the wave number shifted to 1611 cm-1, and the absorption peak in CP was 1610 cm-1 (Fig. 6e-g). In addition, in GM (TPN/ CD = β/1), GM (TPN/ CD = 4/1), 257 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 (TPN/ CD = 4/1), it was inferred that the sevenmembered ring of two molecules of TPN is in the cavity and located near the narrower edge of CD. Furthermore, since a cross peak between H-2 and H-4 located outside CD was confirmed, it is speculated that the remaining two molecules of TPN were present in the molecular space formed by CDs. In CP, it was inferred that the sevenmembered ring of TPN is located near the narrow edges of CD. Since cross peaks were confirmed between H-A, H-B, H-D, and H-E of TPN and H-5 and H-6 of CD as well as between H-C of TPN and H-6 of CD, it is assumed that the carbonyl and hydroxyl groups are located on the narrower edge side of CD. Higashi et al. reported similar results using the sealed heating method for salicylic acid and CD [β4]. In other words, TPN has a planar structure similar to salicylic acid, and it is not only included into CD, but also interacts with CD in the void space between CD molecules. This may be because of the strong interaction between the void space and TPN structure. From these results, can suggest that TPN and CD form inclusion complexes of different structures due to differences in preparation methods. 1 H-1H NOESY NMR measurement: 1H - 1H NOESY NMR measurement was performed to evaluate the molecular state in aqueous solution [32]. In GM (TPN/ CD = β/1), a cross peak was observed between H-A, H-B, H-D, and H-E peaks derived from the seven-membered ring of TPN and H-3, H-5, and H-6 peaks of CD (Fig. 7-a). In GM (TPN/ CD = 4/1), a cross peak was observed between the H-A, H-B, H-D, and H-E peaks derived from the seven-membered ring of TPN and the H-3, H-5, and H-6 peaks located inside CD. In addition, a cross peak was observed between H-2 and H-4 located outside CD (Fig. 7-b). In CP, a cross peak was observed between H-A, H-B, H-D and H-E peaks derived from the seven-membered ring of TPN and H-5 and H-6 peaks located inside CD (Fig. 7-c). From the results of 1H-1H NOESY NMR measurement, GM (TPN/ CD = β/1) was inferred to be located near the wide edge of the sevenmembered ring of TPN is a CD. Furthermore, from the cross peaks between H-6 of CD and H-A and H-E of TPN, it was suggested that the carbonyl and hydroxyl groups of TPN were located in the narrower edge of CD. However, in GM 1 1 Fig. 7-a H- H NOESY NMR spectrum of GM (TPN/ CD=2/1) in D2O 258 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 1 1 1 1 Fig. 7-b H- H NOESY NMR spectrum of GM (TPN/ CD=4/1) in D2O Fig. 7-c H- H NOESY NMR spectrum of CP (TPN/ CD) in D2O 259 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 Diagram1 Structural view of TPN/ CD complex (a) GM (TPN/ CD=2/1), (b-1) GM (TPN/ CD=4/1) side view, (b-2) GM (TPN/ CD=4/1) top view, (c) CP (TPN/ CD=2/1) an interesting new discovery. As with the salicylic acid system, it became possible to form a novel ternary complex. In the future, further elucidation of the encapsulation mechanism of the drug in the molecular space formed by these specific CDs would broaden the use of CD as drug carriers in pharmaceutical development. CONCLUSIONS In this study, revealed the formation of TPN/ CD inclusion complex using cogrinding and coprecipitation methods. Owing to the differences in preparation methods, inclusion complex with different structures were formed. The molar ratio of the inclusion complex formed by the cogrinding method was TPN/ CD = β/1 and TPN/ CD = 4/1 and that by coprecipitation method was TPN/ CD = 2/1. In addition, in GM (TPN/ CD = 4/1), two molecules of TPN were encapsulated in the molecular space formed between CD. Encapsulation of drugs in the molecular space between CDs in the cogrinding method has become ACKNOWLEDGMENT The authors are grateful to Cyclo Chem Co., Ltd for the provision of CD. Conflict of Interests: The authors declare no conflict of interests regarding the publication of this paper. 260 Inoue et al., World J Pharm Sci 2017; 5(9): 250-261 REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Trust TJ. Antibacterial Activity of Tropolone. Antimicrob. Agents Chemother 1975; 7(5): 500-506. Ye J et al. Anti-inflammatory effects of hinokitiol on human corneal epithelial cells: an in vitro study. Eye (Lond) 2015; 29(7): 964-971. Kadoma Y et al. Kinetic radical-scavenging activity of colchicine and tropolone. In Vivo 2007; 21(3): 481-486. Lee YS et al. Hinokitiol inhibits cell growth through induction of S-phase arrest and apoptosis in human colon cancer cells and suppresses tumor growth in a mouse xenograft experiment. J Nat Prod 2013; 76(12): 2195-2202. Jiménez C et al. Exploring the size adaptability of the B ring binding zone of the colchicine site of tubulin with paranitrogen substituted isocombretastatins. Eur J Med Chem 2015; 11: 210-222. Inamori Y et al. Cytotoxic effect of hinokitiol and tropolone on the growth of mammalian cells and on blastogenesis of mouse splenic T cells. Biol Pharm Bull 1993; 16(5): 521-523. Brewster M.E et al. Cyclodextrins as pharmaceutical solubilizers. Adv. Drug. Deliv. Rev 2007; 59: 645-666. Ezawa T et al. Changes in the Physicochemical Properties of Piperine/ -Cyclodextrin due to the Formation of Inclusion Complexes. Int J Med Chem 2016 (in press). Abd El-Gawad AE et al. Improvement of the Ocular Bioavailability of Econazole Nitrate upon Complexation with Cyclodextrins. AAPS PharmSciTech 2017; 18(5): 1795-1809. Van den Hoven JM et al. Cyclodextrin as membrane protectant in spray-drying and freeze-drying of PEGylated liposomes. Int J Pharm 2012; 438: 209-216. Cabral-Marques H et al. Optimisation of spray-drying process variables for dry powder inhalation (DPI) formulations of corticosteroid/cyclodextrin inclusion complexes. Eur J Pharm Biopharm 2009; 73: 121-129. Corti G et al. Physical-chemical characterization of binary systems of metformin hydrochloride with triacetyl-betacyclodextrin. J Pharm Biomed Anal 2007; 45(3): 480-486. Inoue Y et al. Evaluation of actarit/ -cyclodextrin complex prepared by different methods. J Incl Phenom Macrocycl Chem 2015; 81: 161-168. Higashi K et al. Structural evaluation of crystalline ternary -cyclodextrin complex. J Pharm Sci 2011; 100: 325-333. Suzuki R et al. Effect of -cyclodextrin derivative complexation on the physicochemical properties and antimicrobial activity of hinokitiol. J Incl Phenom Macrocycl Chem 2015; 83: 177-186. Suzuki R et al. Molecular interactions of the inclusion complexes of hinokitiol and various cyclodextrins. AAPS PharmSciTech 2017 (in press). Specogna E et al. Dehydration, dissolution, and melting of cyclodextrin crystals. J. Phys. Chem. B 2015; 119: 14331442. Xiao CF et al. Investigation of inclusion complex of epothilone A with cyclodextrins. Carbohydr Polym 2014; 102: 297-305. Giordano F et al. Thermal analysis of cyclodextrins and their inclusion compounds. Thermochim. Acta 2001; 380: 123-151. Inoue Y et al. Ternary inclusion complex formation and stabilization of limaprost, a prostaglandin E1 derivative, in the presence of α- and -cyclodextrins in the solid state. Int. J. Pharm 2016; 509: 338-347. Aigner Z et al. DSC, X-ray and FTIR studies of a gemfibrozil/dimethyl- -cyclodextrin inclusion complex produced by co-grinding. J Pharm Biomed Anal 2012; 57: 62-67. Iwata M et al. Effectiveness of mechanochemical treatment with cyclodextrins on increasing solubility of glimepiride. Pharmazie 2009; 64(6): 390-394. Nakai Y et al. Properties of crystal water of a-, -, and -cyclodextrin. Chem Pharm Bull 1986; 34: 2178-2182. Higashi K et al. Salicylic acid/gamma-cyclodextrin 2:1 and 4:1 complex formation by sealed-heating method. J Pharm Sci 2010; 99: 4192-4200. Daniel I et al. Water content of flavonoid/cyclodextrin nanoparticles: Relationship with the structural descriptors of biologically active compounds. Food Chemistry 2012; 132: 1651-1659. Zhao R et al. NMR studies on puerarin and its interaction with beta-cyclodextrin. J. Biol. Phys 2011; 37(4): 387-400. Ogawa N et al. Solid-state characterization of sertraline base- -cyclodextrin inclusion complex. J. Pharm. Biomed. Anal 2015; 107: 265-272. Fernandes CM et al. Physicochemical characterization and in vitro dissolution behavior of nicardipine-cyclodextrins inclusion compounds. Eur. J. Pharm. Sci 2002; 15: 79-88. Mitsuzuka A et al. Infrared spectroscopy of OH stretching vibrations of hydrogen-bonded tropolone-(H2O)n (n51-3) and tropolone-(CH3OH)n (n51 and 2) clusters. Journal of chemical physics 1996; 105(7): 2618-2627. Mohamad S et al. Conventional study on novel dicationic ionic liquid inclusion with -cyclodextrin. Int. J. Mol. Sci 2011; 12: 6329-6345. Tárkányi G et al. Structure and stability of warfarin-sodium inclusion complexes formed with permethylated monoamino- -cyclodextrin. J. Pharm. Biomed. Anal 2013; 72: 292-298. Yao Y et al. Development of a myricetin/hydroxypropyl- -cyclodextrin inclusion complex: preparation, characterization, and evaluation. Carbohydr Polym 2014; 110: 329-227. 261