Enhancing the solubility and masking the bitter taste of propiverine using crystalline complex formation

http://informahealthcare.com/ddi ISSN: 0363-9045 (print), 1520-5762 (electronic) Drug Dev Ind Pharm, 2014; 40(8): 1084–1091 ! 2014 Informa Healthcare USA, Inc. DOI: 10.3109/03639045.2013.807280 RESEARCH ARTICLE Enhancing the solubility and masking the bitter taste of propiverine using crystalline complex formation Tetsuo Ogata1,2, Daisuke Tanaka3, and Tetsuya Ozeki2 1Formulation Research Laboratory, Tokushima Research Center, Taiho Pharmaceutical Co., Ltd., 224-2 Hiraishi-ebisuno, Kawauchi-cho, Tokushima, Japan, 2Graduate School of Pharmaceutical Sciences, Nagoya City University, 3-1 Tanabe-dori, Mizuho-ku, Nagoya, Aichi, Japan, and 3Chemistry Research Laboratory, Tsukuba Research Center, Taiho Pharmaceutical Co., Ltd., 3 Ohkubo, Tsukuba, Ibaraki, Japan Abstract Context: Patient compliance can be reduced when bitter-tasting compounds, such as propiverine hydrochloride, are administered orally. Propiverine hydrochloride is an example of a drug with a bitter taste, used for the treatment of overactive bladders. Objective: This study tested whether propiverine free base palatability and aqueous solubility could be improved by crystalline complex formation. Materials and methods: We used 42 compounds, and found 9 new propiverine crystalline complexes. The properties and solubility of these complexes were studied using a range of techniques. A taste perception study was carried out using a taste sensor to evaluate the taste masking ability of the crystalline complex formation. Results: The melting points of the crystalline complexes were higher than that of propiverine. The dissolution rates of the crystalline complexes in aqueous buffer solution (pH 6.8) and in purified water were much faster than that of propiverine. Propiverine salicylic acid crystalline complex had substantially less bitterness than propiverine hydrochloride, which was extremely bitter. Discussion: The present findings indicated that crystalline complex formation provided an effective approach to enhancing propiverine solubility, and to masking its bitter taste. Conclusion: Crystalline complex formation represents a useful and valuable technique for the preparation of orally disintegrating tablets and improving patient compliance, even for substances with bitter tastes. Keywords Crystalline complex, dissolution test, propiverine, taste masking, taste sensor History Received 2 February 2013 Revised 13 May 2013 Accepted 14 May 2013 Published online 21 June 2013 Introduction Taste is one of the most important parameters governing patient compliance, as unpleasant tasting medicines are generally diffi- cult to take. Within the range of unpleasant tastes, bitterness is disliked most. Therefore, masking the bitterness of the active pharmaceutical ingredients (API) in an oral formulation is an important consideration when making the drug product. Bitterness can be masked using a number of methods, including inclusion complexation1–3, ion-exchange resins4–6 and salt preparation7,8. The latter approach suggests that complex formation could be an effective technique for masking bitterness; however, most previous studies have focused on salt preparation as an approach to improve solubility of poorly soluble com- pounds8–10. In addition, recent work has indicated that solubility can be improved by cocrystallization, in addition to salt forma- tion, use of amorphous drug forms, cosolvency, and crystal polymorphism11–25. In this study, propiverine was used as a model drug. Propiverine hydrochloride, a medication used for the treatment of overactive bladders, is a basic salt which is very bitter and highly water-soluble (pKa¼ 8.57)26. Its free base (propiverine) has poor solubility in water27 but also results in a bitter-tasting solution. To date, no research has been conducted on crystalline complexes of propiverine, and no information is available regarding their solubility or taste. This study explored the preparation of propiverine crystalline complexes using 42 compounds. By adding a compound solution to propiverine solution, we successfully created nine new propiverine crystalline complexes. These were confirmed by 1H- NMR, differential thermal analysis, and powder X-ray diffraction analysis, and shown to have improved aqueous solubility and taste. Methods and materials Materials Propiverine and propiverine hydrochloride were obtained from Taiho Pharmaceutical Co., Ltd. (Tokyo, Japan) and Fukujyu Pharmaceutical Co., Ltd. (Toyama, Japan), respectively. Benzotriazole, glutaric acid, erythorbic acid, saccharin, sorbic Address for correspondence: Testuo Ogata, Formulation Research Laboratory, Tokushima Research Center, Taiho Pharmaceutical Co., Ltd., 224-2 Hiraishi-ebisuno, Kawauchi-cho, Tokushima, Japan. Tel: +81-88-665-3571. Fax: +81-88-665-7225. E-mail: tetsu-ogata@taiho. co.jp D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. acid, succinic acid, and DL-tartaric acid (special grade), acetanil- ide, and urea (analytical grade) were purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). Acetonitrile (HPLC grade), adipic acid, L-(þ)-ascorbic acid, L-aspartic acid, benzoic acid, dibutylhydroxytoluene, L-glutamic acid, DL-malic acid, oxalic acid, D-(�)-fructose, fumaric acid, glycine, hippuric acid, maleic acid, malonic acid, (�)-mandelic acid, D-(�)-mannitol, myo- inositol, nicotineamide, nicotinic acid, palmitic acid, phthalic acid, salicylic acid, p-toluic acid, p-tosic acid monohydrate, vaniline, xylitol, potassium dihydrogenphosphate, acetone, tetra- hydrofuran (THF) (special grade), aspartame, citric acid, L- cysteine, isophthalic acid, propyl gallate, terephthalic acid (analytical grade) and 1-octanesulfonic acid sodium salt (ion pair chromatography-grade) were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). D-(þ)-glucose was purchased from Lancaster Synthesis (Lancashire, United Kingdom). Taste standard solutions were purchased from Intelligent Sensor Technology, Inc. (INSENT, Kanagawa, Japan). 1H-NMR spectra 1H-NMR spectra were recorded using a JEOL JNM-EX270 spectrometer. Chemical shifts were given as � values relative to tetramethylsilane (TMS) as internal standard. Preparation of propiverine crystalline complexes Figure 1 shows the structural formula of propiverine. The propiverine molecule is comprised of a piperidine nitrogen atom (Figure 1, small circle) and an aromatic rings (Figure 1, large circles). The nitrogen atom and/or the �–� interaction between the aromatic ring and the coformer are involved in the generation of crystalline complexes. Crystalline complexes were prepared by mixing one of the compounds listed in Table 1 with propiverine. For 33 of the 42 compounds, mixing with propiverine failed to generate a crystalline complex. Nine crystalline complexes were created successfully, and their preparation methods are presented individually below. Succinic acid crystalline complex (Form I): Propiverine (4.73 g, 12.9 mmol) and succinic acid (0.770 g, 6.52 mmol) were dissolved in acetone (60 mL). The solution was evaporated under reduced pressure to produce succinic acid crystalline complex (Form I). Succinic acid crystalline complex (Form II): Propiverine (5.30 g, 14.4 mmol) and succinic acid (1.70 g, 14.4 mmol) were dissolved in acetone (28 mL). The solution was stirred at room temperature and the precipitate was filtered to isolate the succinic acid crystalline complex (Form II). Fumaric acid crystalline complex: Propiverine (7.60 g, 20.7 mmol) and fumaric acid (2.40 g, 20.7 mmol) were dissolved in THF (15 and 60 mL, respectively). The 2 solutions were then mixed for 4 h at 5 �C and the precipitate was filtered to isolate the fumaric acid crystalline complex. Malonic acid crystalline complex: Propiverine (6.23 g, 17.0 mmol) and malonic acid (1.77 g, 17.0 mmol) were dissolved in acetone (8 mL) at 50 �C and stirred at room temperature. The precipitate was filtered to isolate the malonic acid crystalline complex. DL-tartaric acid crystalline complex: Propiverine (4.26 g, 11.6 mmol) and DL-tartaric acid (1.74 g, 11.6 mmol) were dissolved in acetone (72 mL) at 60 �C and stirred at room temperature. The precipitate was filtered to isolate DL-tartaric acid crystalline complex. Benzoic acid crystalline complex: Propiverine (7.51 g, 20.4 mmol) and benzoic acid (2.50 g, 20.4 mmol) were dissolved in acetone (30 mL) at 40 �C and stirred below 5 �C for 2 h. The precipitate was filtered to isolate the benzoic acid crystalline complex. p-Toluic acid crystalline complex: Propiverine (5.47 g, 14.9 mmol) and p-toluic acid (2.03 g, 14.9 mmol) were dissolved in THF (15 mL), mixed with water (225 mL), and then stirred for 6 h at room temperature. The precipitate was filtered to isolate the p-toluic acid crystalline complex. Salicylic acid crystalline complex: Propiverine (5.09 g, 13.8 mmol) and salicylic acid (1.91 g, 13.8 mmol) were dissolved in acetone (56 and 14 mL, respectively), mixed, and further diluted with water (210 mL). The solution was then stirred for 4.5 h at room temperature. The precipitate was filtered to isolate the salicylic acid crystalline complex. Phthalic acid crystalline complex: Propiverine (4.82 g, 13.1 mmol) and phthalic acid (2.18 g, 13.1 mmol) were dissolved in THF (17.5 mL). The solution was then mixed with water (210 mL) and stirred for 4 h at room temperature. The precipitate was filtered to isolate the phthalic acid crystalline complex. Powder X-ray diffraction (PXRD) measurements PXRD patterns were obtained using a Rigaku X-ray diffraction system (model RINT2100 Ultimaþ/PC, Rigaku Corporation, Tokyo, Japan) with copper as the target (CuKa), a voltage of 40 kV and a current of 40 mA. The scanning speed was 2�/min at 5–40� angles. Table 1. Compounds used to prepare crystalline complexes. Group A Group B adipic acid, L-(þ)-ascorbic acid, L-aspartic acid, benzoic acid, citric acid, L-cysteine, dibutylhydroxytoluene, D-(�)-fructose, fumaric acid, L- glutamic acid, glutaric acid, glycine, hippuric acid, isophthalic acid, erythorbic acid, maleic acid, DL-malic acid, malonic acid, (�)-man- delic acid, D-(�)-mannitol, myo-inositol, nicotinic acid, oxalic acid, palmitic acid, phthalic acid, salicylic acid, sorbic acid, succinic acid, DL-tartaric acid, terephthalic acid, p-toluic acid, p-tosic acid mono- hydrate, urea, vaniline, xylitol acetanilide, aspartame, benzoic acid, benzotriazole, dibutylhydroxyto- luene, D-(þ)-glucose, hippuric acid, isophthalic acid, (�)-mandelic acid, nicotinamide, nicotinic acid, phthalic acid, propyl gallate, saccharin, salicylic acid, terephthalic acid, p-toluic acid, p-tosic acid monohydrate, vaniline Group A, Compound that interacts with the nitrogen atom in propiverine. Group B, Compound that uses a �-� interaction with the propiverine aromatic ring. Figure 1. Structural formula of propiverine. Small circle, Nitrogen atom in propiverine molecule. Large circle, Propiverine molecule’s aromatic ring. DOI: 10.3109/03639045.2013.807280 Solubility and taste of propiverine crystalline complexes 1085 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. Differential thermal analysis (DTA) Samples (9–10 mg) were placed in an open aluminum pan and analyzed using a differential thermal balance apparatus (model TG8120; Rigaku Corporation, Japan) at a scan rate of 5.0 �C/min under a dry N2 gas flow (100 mL/min) at a temperature range of 50–250 �C (except fumaric acid: 320 �C). Dissolution studies Dissolution analyses of propiverine, its crystalline complexes (the equivalent of 10 mg of propiverine hydrochloride), and propiver- ine hydrochloride, were performed according to the method described in the Japanese Pharmacopoeia (JP XVI), using apparatus 2 with the paddle rotating at 50 rpm in 900 mL of medium at 37� 0.5 �C. The tests used 900 mL of phosphate buffer (pH 6.8) or water and were performed in duplicate. Samples (10 mL) were taken at predetermined time intervals and filtered through a membrane filter (pore size 0.45 mm; Millipore). The filtrate was diluted with an equal volume of 1 mM hydrochloric acid. A 15 -mL aliquot of this diluted filtrate was then assayed using a Shimadzu LC-2010C HPLC system with an Inertsil Ph-3 column (50 mm� 4.6 mm i.d., 5mm; GL Science, Inc., Tokyo, Japan) kept at 40 �C. The mobile phase was 25 mM potassium dihydrogenphosphate with 10 mM 1-octanesulfonic acid sodium salt (pH 3.2)/acetonitrile (13:7), at a flow rate of 1.2 mL/min. Propiverine was detected at a wavelength of 210 nm. Taste sensor measurements A taste-responding system (SA402B; INSENT, Kanagawa, Japan) was used to measure the electrical potential of the sample solutions. The electrode set was attached to a mechanically controlled robotic arm. The detection sensor consisted of eight electrodes, composed of lipid/polymer membranes, described in Table 2. Each lipid was mixed in a test tube containing a plasticizer, dissolved in THF, and dried on a glass plate at 30 �C to form a thin, transparent, 200-mm-thick film. The electrodes consisted of silver wires plated with Ag/AgCl, with an internal cavity filled with 3.33 M KCl solution. The difference between the electric potentials of the working electrode and the reference electrode was measured using a high-input impedance amplifier, connected to a computer30,31. Fresh 30 mM KCl solution containing 0.3 mM tartaric acid (corresponding to saliva) was used as the reference solution and rinsing solution for the electrodes after every measurement. The electrode was dipped first into the reference solution (Vr) and then into the sample solution (Vs). The relative sensor output was calculated as the difference between the potentials of the sample and the reference solutions (Vs–Vr). When the electrode was returned to the reference solution, the new potential of the reference solution was defined as Vr0 . The difference (Vr0–Vr) between the potentials of the reference solution before and after sample measurement was defined as the change in membrane potential due to adsorption (CPA), which also corresponded to the aftertaste. The measuring intervals were set at 20 s, and the electrodes were rinsed after each measurement. Sample solutions were prepared using the following methods. The drug samples (corresponding to 0.05 mmol) were placed in 10 mL injection cylinders. A standard solution (30 mM KCl with 0.3 mM tartaric acid) was poured into the injection cylinder (4 mL, 37 �C), and the injection cylinder was reversed 10 times every 3 s. The extract was filtered using a membrane filter (pore size 0.45 mm; Nihon Millipore K.K., Japan). Sample solutions were added to 2 mL of the filtrate and the final volume was adjusted to 250 mL. The types of information provided by the sensor are presented in Table 3 and the taste standard solutions employed in the study are presented in Table 4. Results and discussion 1H-NMR spectra The 1H-NMR spectrum data of 9 crystalline complexes are enumerated as follows: Succinic acid crystalline complex (Form I): �: 0.86 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.47–1.57 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.72–1.73 (2H, m, CHCH2CH2N), 2.11 (3H, s, NCH3), 2.17–2.29 (4H, m, CHCH2CH2N), 2.39 (2H, s, succinic acid CH2), 3.14 (2H, t, J¼ 8.4 Hz, CH3CH2CH2O), 4.83–4.85 (1H, m, piperidine CH), 7.31–7.37 (10H, m, ArH). Succinic acid crystalline complex (Form II): �: 0.87 (3H, t, J¼ 7.4 Hz, CH3CH2CH2O), 1.50–1.58 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.73–1.75 (2H, m, CHCH2CH2N), 2.14 (3H, s, NCH3), 2.17–2.37 (4H, m, CHCH2CH2N), 2.40 (4H, s, succinic acid CH2), 3.14 (2H, t, J¼ 6.4 Hz, CH3CH2CH2O), 4.79–4.91 (1H, m, piperidine CH), 7.26–7.46 (10H, m, ArH). Fumaric acid crystalline complex: �: 0.87 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.50–1.60 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.73–1.93 (2H, m, CHCH2CH2N), 2.23 (3H, s, Table 2. Lipid and plasticizers used for the sensor membranes28. Sensor Lipid Plasticizer AAE Trioctylmethyl ammonium chloride Dioctyl phenylphosphonate Phosphoric acid di(2-ethylhexyl) ester AC0 Palmitic acid Dioctyl phenylphosphonate AE1 Tetradodecyl ammonium bromide Dioctyl phenylphosphonate AN0 Phosphoric acid di-n-decyl ester Dioctyl phenylphosphonate BT0 Phosphoric acid di-n-decyl ester Bis(1-butylpentyl) adipate Tributyl O-acetylcitrate CA0 Phosphoric acid di(2-ethylhexyl) ester Dioctyl phenylphosphonate Oleic acid Trioctylmethyl ammonium chloride C00 Tetradodecyl ammonium bromide 2-Nitrophenyl octyl ether CT0 Tetradodecyl ammonium bromide Dioctyl phenylphosphonate 1-Hexadecanol Table 3. Taste information converted from taste sensor outputs28,29. Sensor Taste information from relative value Taste information from CPA* value AAE Umami Richness AC0 (none) Aftertaste from basic bitterness AE1 Astringency Aftertaste from astringency AN0 (none) Aftertaste from basic bitterness BT0 (none) Aftertaste from HCl salt CA0 Sourness (none) C00 Acidic bitterness Aftertaste from acidic bitterness CT0 Saltiness (none) *CPA, The change in membrane potential due to adsorption. Table 4. Taste standard solutions. Taste Composition Saltiness 300 mM KClþ 0.3 mM tartaric acid Sourness 30 mM KClþ 3 mM tartaric acid Umami 30 mM KClþ 0.3 mM tartaric acidþ 10 mM sodium hydrogen glutamate Basic bitterness 30 mM KClþ 0.3 mM tartaric acidþ 0.1 mM quinine hydrochloride Acidic bitterness 30 mM KClþ 0.3 mM tartaric acidþ iso-a acid Astringency 30 mM KClþ 0.3 mM tartaric acidþ 0.05% tannic acid 1086 T. Ogata et al. Drug Dev Ind Pharm, 2014; 40(8): 1084–1091 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. Figure 2. Powder X-ray diffraction spectra of the compounds indicated above each trace. Figure 3. Powder X-ray diffraction spectra of propiverine and propiverine crystalline com- plexes indicated above each trace. DOI: 10.3109/03639045.2013.807280 Solubility and taste of propiverine crystalline complexes 1087 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. NCH3), 2.50–2.51 (4H, m, CHCH2CH2N), 3.14 (2H, t, J¼ 6.5 Hz, CH3CH2CH2O), 4.85–4.97 (1H, m, piperidine CH), 6.56 (2H, s, fumaric acid CH), 7.32–7.38 (10H, m, ArH). Malonic acid crystalline complex: �: 0.87 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.55–1.57 (2H, m, CH3CH2CH2O), 1.61–1.73 (2H, m, CHCH2CH2N), 1.84–1.96 (2H, m, CHCH2CH2N), 2.49 (3H, s, NCH3), 2.64 (2H, s, malonic acid CH2), 2.85–2.88 (4H, m, CHCH2CH2N), 3.14 (2H, t, J¼ 6.5 Hz, CH3CH2CH2O), 4.92– 5.04 (1H, m, piperidine CH),7.33–7.39 (10H, m, ArH). DL-tartaric acid crystalline complex: �: 0.87 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.50–1.57 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.79–1.91 (2H, m, CHCH2CH2N), 2.32 (3H, s, NCH3), 2.49–2.52 (4H, m, CHCH2CH2N), 2.59 (2H, s, OH), 3.15 (2H, t, J¼ 6.5 Hz, CH3CH2CH2O), 4.13 (2H, s, DL-tartaric acid CH), 4.86–4.98 (1H, m, piperidine CH), 7.32–7.38 (10H, m, ArH). Benzoic acid crystalline complex: �: 0.86 (3H, t, J¼ 7.0 Hz, CH3CH2CH2O), 1.50–1.58 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.72–1.74 (2H, m, CHCH2CH2N), 2.10 (3H, s, NCH3), 2.13–2.33 (4H, m, CHCH2CH2N), 3.14 (2H, t, J¼ 6.5 Hz, CH3CH2CH2O), 4.83–4.86 (1H, m, piperidine CH), 7.31–7.37 (10H, m, ArH), 7.46–7.64 (3H, m, ArH), 7.94 (2H, t, J¼ 7.0 Hz, ArH). p-Toluic acid crystalline complex: �: 0.87 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.50–1.58 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.67–1.79 (2H, m, CHCH2CH2N), 2.10 (3H, s, NCH3), 2.11–2.31 (4H, m, CHCH2CH2N), 2.37 (3H, s, ArCH3), 3.14 (2H, t, J¼ 6.2 Hz, CH3CH2CH2O), 4.78–4.90 (1H, m, piperidine CH), 7.28–7.38 (12H, m, ArH), 7.83 (2H, t, J¼ 8.1 Hz, ArH). Salicylic acid crystalline complex: �: 0.86 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.47–1.72 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.88–2.00 (2H, m, CHCH2CH2N), 2.53 (3H, s, NCH3), 2.75 (1H, s, OH), 2.90–3.02 (4H, m, CHCH2CH2N), 3.14 (2H, t, J¼ 6.4 Hz, CH3CH2CH2O), 4.94–5.06 (1H, m, piperidine CH), 6.64–6.71 (2H, m, ArH),7.19–7.71 (12H, m, ArH). Phthalic acid crystalline complex: �: 0.87 (3H, t, J¼ 7.3 Hz, CH3CH2CH2O), 1.50–1.71 (4H, m, CHCH2CH2N and CH3CH2CH2O), 1.81–2.07 (2H, m, CHCH2CH2N), 2.61 (3H, s, NCH3), 2.62–2.88 (2H, m, CHCH2CH2N), 2.98–3.04 (2H, m, CHCH2CH2N), 3.15 (2H, t, J¼ 6.5 Hz, CH3CH2CH2O), 4.95– 5.07 (1H, m, piperidine CH), 7.38–7.53 (12H, m, ArH), 8.08–8.11 (2H, m, ArH). In the above-mentioned result, because the existence of the proton that originated in the propiverine and these compounds (organic acids) had been confirmed by each complex, it was suggested that these complexes that had obtained it in the present study be composed of these two elements. PXRD Figure 2 shows the PXRD spectra of propiverine, succinic acid and propiverine succinic acid crystalline complex; and Figure 3 shows the PXRD spectra of all the other propiverine crystalline complexes generated by the present study. In Figure 2, the diffraction peak pattern of the propiverine succinic acid crystalline complex was significantly different, compared to the individual source materials (propiverine and succinic acid). In Figure 3, the characteristic peak intensities of propiverine (2�¼ 6.8�, 12.4�, 18.0� and 27.3�) were weaker in all the other propiverine crystalline complexes, con- firming changes to the crystal structure of propiverine. The existence of the diffraction peaks showed that these were crystal- line complexes. DTA Figure 4 shows the DTA curves for propiverine, phthalic acid, and the propiverine phthalic acid crystalline complex, whereas Figure 5 shows the DTA curves for all other the propiverine crystalline complexes. The endothermic melting peak for propiverine phthalic acid crystalline complex differed from the source materials (propiverine and phthalic acid). Similarly, all other propiverine crystalline complexes, except propiverine phthalic acid crystalline complex, showed a different endothermic melting peak than propiverine or the source material organic acid (measured values: succinic acid: 189 �C; fumaric acid: 256 �C; malonic acid: 137 �C; DL-tartaric acid: 208 �C; benzoic acid: 123 �C; p-toluic acid: 180 �C and salicylic acid: 159 �C). The endothermic melting peaks of propiverine crystalline complexes ranged from 30 �C to 90 �C higher than the melting point of propiverine, which was approximately 65 �C. It was confirmed that these compounds were new compounds because it observed a Figure 5. Differential thermal analysis curves for propiverine and the propiverine crystalline complexes indicated below each trace. Figure 4. Differential thermal analysis curves for the compounds indicated above each trace. 1088 T. Ogata et al. Drug Dev Ind Pharm, 2014; 40(8): 1084–1091 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. single endothermic melting peak at the temperature different from the source materials other than the propiverine and the organic acids that were the source material. Dissolution studies We examined the effects of these changes in crystal structure and melting points on the dissolution ability of propiverine. Figure 6 shows the dissolution percentages for propiverine hydrochloride, propiverine and the propiverine crystalline complexes in buffer (pH 6.8) and water. The 5-min dissolution percentage revealed that crystalline complexes, other than p-toluic acid, salicylic acid, and phthalic acid, behaved similarly to propiverine hydrochloride. At the 15-min time point, all crystalline complexes showed a similar dissolution percentage to propiverine hydrochloride. Propiverine is a basic compound, showing poor dissolution in both water and a buffer solution at pH 6.8. However, these results demonstrated that converting propiverine to a crystalline complex considerably improved its dissolution in a buffer solution at pH 6.8 (similar to intraoral pH) and water, to the point where propiverine solubility was comparable to that of propiverine hydrochloride. Taste sensor measurement We investigated whether the dissolved propiverine crystalline complexes were bitter tasting, in a similar manner to propiverine hydrochloride. Figure 7 (upper panel) shows the sensor outputs of Figure 6. Dissolution of propiverine hydrochloride, propiverine and the propiverine crystalline complexes indicated, expressed as the percentage of starting material in solution 5 (open bars) and 15 (shaded bars) min after dissolution tests were started. Each data point represents the mean of duplicate analyses. DOI: 10.3109/03639045.2013.807280 Solubility and taste of propiverine crystalline complexes 1089 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. the probes detecting umami, saltiness, sourness, acidic bitterness and astringency of 0.1 mM propiverine solution, 0.1 mM propiver- ine hydrochloride solution, and 0.1 mM propiverine crystalline complex solutions. Figure 7 (lower panel) presents the sensor probe readings for the indicated aftertastes and richness. In these graphs, if the output value of the sensor is large, it is shown that the taste that relates to the sensor is strong. Moreover, if the output value of the sensor is small, it is shown that the taste that relates to the sensor is weak. Propiverine hydrochloride, propiverine and propiverine crystalline complexes produced minimal sensor outputs for AAE, CT0, CA0, AE1, CPA (AAE), CPA (C00) and CPA (AE1), compared to the relevant standard taste solution (Table 4). Therefore, propiverine crystalline complexes were thought that umami, saltiness, sourness, astrin- gency, richness, aftertaste from acidic bitterness and aftertaste from astringency were extremely weak. Of the nine propiverine crystalline complexes generated, the C00 (acidic bitterness) sensor output was strongest in the propiverine phthalic acid crystalline complex, suggesting that it had a strongly bitter and acidic taste. Crystalline complexes with benzoic acid, p-toluic acid, salicylic acid, and phthalic acid showed significantly smaller sensor output values for CPA (BT0), CPA (AN0), and CPA (AC0) than propiverine hydrochloride, suggesting that these compounds were less bitter than propiverine hydrochloride. However, the 5- min dissolution percentages of these crystalline complexes were low, as shown in Figure 6, probably due to the increased melting points of the crystalline complexes. This lower solubility may have contributed to the reduced bitterness sensor output. In other words, it is thought that bitterness sensor output becomes small when an initial dissolution rate of crystalline complex is low. Of these four crystalline complexes, the sensor output value for the salicylic acid crystalline complex was similar to that of propiverine. We have previously reported that propiverine was less bitter than propiverine hydrochloride, based on both taste sensor measurements and sensory evaluation17. The present study showed that the propiverine salicylic acid crystalline complex was substantially less bitter than propiverine hydro- chloride. Furthermore, our results showed that the propiverine salicylic acid crystalline complex possessed high dissolution ability. Conclusions We generated nine propiverine crystalline complexes by mixing an organic acid solution with a propiverine solution. We confirmed the identities of these crystalline complexes through 1H-NMR, DTA and PXRD spectra. Whilst propiverine dissolves poorly in buffer (pH 6.8) or water, we have shown that the crystalline complexes possessed a greatly improved capacity for dissolution. The present study also demonstrated that use of the taste sensor was a valid approach to screening taste of many compounds. This taste analyses showed that propiverine salicylic acid crystalline complexes were less bitter than propiverine hydrochloride. Figure 7. Taste sensor outputs of propiverine hydrochloride, propiverine, propiverine crystalline complexes and standard taste solutions. In the upper panel, AAE¼ umami, CT0¼ saltiness, CA0¼ sourness, C00¼ acidic bitterness and AE1¼ astringency. In the lower panel, CPA (AAE)¼ richness, CPA (C00)¼ aftertaste from acidic bitterness, CPA (AE1)¼ aftertaste from astringency, CPA (BT0)¼ aftertaste from HCl salt, CPA (AN0)¼ aftertaste from basic bitterness and CPA (AC0)¼ aftertaste from basic bitterness. The standard taste solutions used for each taste are listed in Table 4. 1090 T. Ogata et al. Drug Dev Ind Pharm, 2014; 40(8): 1084–1091 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y. We succeeded in both improving aqueous solubility and masking the bitterness of propiverine. At present, we cannot confirm whether the crystalline complexes obtained in this way are salt or cocrystal. Childs et al.32 have described a method for prediction of whether a product was salt or cocrystal, using DpKa value (pKa of base – pKa of acid). In addition, salts and cocrystals are multicomponent crystals that can be distinguished by the location of the proton between an acid and a base. Proton transfer can be analyzed using several experimental techniques, including single-crystal X-ray crystallography, Raman spectroscopy, infra- red spectroscopy and solid-state NMR spectroscopy. Future studies will use these techniques to clarify whether crystalline complexes obtained as described here are salt or cocrystal. Declaration of interest The authors report no declarations of interest. References 1. Patel AR, Vavia PR. Preparation and evaluation of taste masked famotidine formulation using drug/b-cyclodextrin/polymer ternary complexation approach. AAPS PharmSciTech 2008;9:544–50. 2. Vandelli MA, Salvioli G, Mucci A, et al. 2-hydroxypropyl- b-cyclodextrin complexation with ursodeoxycholic acid. Int J Pharm 1995;118:77–83. 3. Szejtli J, Szente L. Elimination of bitter, disgusting tastes of drugs and food by cyclodextrins. Eur J Pharm Biopharm 2005;61:115–25. 4. Agarwal R, Mittal R, Singh A. Studies of ion-exchange resin complex of chloroquine phosphate. Drug Dev Ind Pharm 2000;26: 773–6. 5. Anand V, Kandarapu R, Garg S. Ion-exchange resins: carrying drug delivery forward. Drug Discov Today 2001;6:905–14. 6. Lu MF, Borodkin S, Woodward L, et al. A polymer carrier system for taste masking of macrolide antibiotics. Pharm Res 1991;8: 706–12. 7. Menegon RF, Blau L, Janzantti NS, et al. A nonstaining and tasteless hydrophobic salt of chlorhexidine. J Pharm Sci 2011;100:3130–8. 8. Bastin RJ, Bowker MJ, Slater BJ. Salt selection and optimization procedures for pharmaceutical new chemical entities. Org Process Res Dev 2000;4:427–35. 9. Serajuddin ATM. Salt formation to improve drug solubility. Adv Drug Deliv Rev 2007;59:603–16. 10. Li S, Woug SM, Sethia S, et al. Investigation of solubility and dissolution of a free base and two different salt forms as a function of pH. Pharm Res 2005;22:628–35. 11. Babu NJ, Nangia A. Solubility advantage of amorphous drugs and pharmaceutical cocrystals. Cryst Growth Des 2011;11:2662–79. 12. Nehm SJ, Rodrı́guez-Spong B, Rodrı́guez-Hornedo N. Phase solubility diagrams of cocrystals are explained by solubility product and solution complexation. Cryst Growth Des 2006;6:592–600. 13. Chiarella RA, Davey RJ, Peterson ML. Making co-crystals — the utility of ternary phase diagrams. Cryst Growth Des 2007;7:1223–6. 14. Stanton MK, Bak A. Physicochemical properties of pharmaceutical co-crystals: a case study of ten AMG 517 co-crystals. Cryst Growth Des 2008;8:3856–62. 15. Reddy LS, Bethune SJ, Kampf JW, Rodrı́guez-Hornedo N. Cocrystals and salts of Gabapentin: pH dependent cocrystal stability and solubility. Cryst Growth Des 2009;9:378–85. 16. Good DJ, Rodrı́guez-Hornedo N. Solubility advantage of pharma- ceutical cocrystals. Cryst Growth Des 2009;9:2252–64. 17. Bethune SJ, Huang N, Jayasankar A, Rodrı́guez-Hornedo N. Understanding and predicting the effect of cocrystal components and pH on cocrystal solubility. Cryst Growth Des 2009;9:3976–88. 18. Good DJ, Rodrı́guez-Hornedo N. Cocrystal eutectic constants and predication of solubility behavior. Cryst.Growth Des 2010;10: 1028–32. 19. Huang N, Rodrı́guez-Hornedo N. Effect of micellar solubilization on cocrystal solubility and stability. Cryst Growth Des 2010;10: 2050–3. 20. Hickey MB, Peterson ML, Scoppettuolo LA, et al. Performance comparison of a co-crystal of carbamazepine with marketed product. Eur J Pharm Biopharm 2007;67:112–19. 21. Shiraki K, Takata N, Takano R, et al. Dissolution improvement and the mechanism of the improvement from cocrystallization of poorly water-soluble compounds. Pharm Res 2008;25:2581–92. 22. Childs SL, Chyall LJ, Dunlap JT, et al. Crystal engineering approach to forming cocrystals of amine hydrochlorides with organic acids. Molecular complexes of fluoxetine hydrochloride with benzoic, succinic, and fumaric acids. J Am Chem Soc 2004;126:13335–42. 23. Bak A, Gore A, Yanez E, et al. The co-crystal approach to improve the exposure of a water-insoluble compound: AMG 517 sorbic acid co-crystal characterization and pharmacokinetics. J Pharm Sci 2008; 97:3942–56. 24. McNamara DP, Childs SL, Giordano J, et al. Use of a glutaric acid cocrystal to improve oral bioavailability of a low solubility API. Pharm Res 2006;23:1888–97. 25. Morissette SL, Almarsson Ö, Peterson ML, et al. High-throughput crystallization: polymorphs, salts, co-crystals and solvates of pharmaceutical solids. Adv Drug Deliv Rev 2004;56:275–300. 26. Ushio T, Yamamoto S, Sudoh Y, et al. Physico-chemical properties and stability of propiverine hydrochloride. Pharm Regul Sci 1991; 22:41–51. 27. Ogata T, Koide A, Kinoshita M, Ozeki T. Taste masking of propiverine hydrochloride by conversion its free base. Chem Pharm Bull 2012;60:976–84. 28. Kobayashi Y, Habara M, Ikezaki H, et al. Advanced taste sensors based on artificial lipids with global selectivity to basic taste qualities and high correlation to sensory scores. Sensors 2010;10: 3411–43. 29. Okamoto M, Sunada H, Nakano M, Nishiyama R. Bitterness evaluation of orally disintegrating famotidine tablets using a taste sensor. Asian J Pharm Sci 2009;4:1–7. 30. Harada T, Uchida T, Yoshida M, et al. A new method for evaluating the bitterness of medicines in development using a taste sensor and a disintegration testing apparatus. Chem Pharm Bull 2010;58: 1009–14. 31. Tanigake A, Miyanaga Y, Nakamura T, et al. The bitterness intensity of Clarithromycin evaluated by a taste sensor. Chem Pharm Bull 2003;51:1241–5. 32. Childs SL, Stahly GP, Park A. The salt-cocrystal continuum: the influence of crystal structure on ionization state. Mol Pharm 2007;4: 323–38. DOI: 10.3109/03639045.2013.807280 Solubility and taste of propiverine crystalline complexes 1091 D ru g D ev el op m en t a nd I nd us tr ia l P ha rm ac y D ow nl oa de d fr om in fo rm ah ea lth ca re .c om b y U ni ve rs ity o f M el bo ur ne o n 09 /1 3/ 14 Fo r pe rs on al u se o nl y.