Synthesis of new benzothiazepine derivatives of monocarbonyl curmumin analogs

95 HNUE JOURNAL OF SCIENCE DOI: 10.18173/2354-1059.2018-0076 Natural Sciences 2018, Volume 63, Issue 11, pp. 95-102 This paper is available online at SYNTHESIS OF NEW BENZOTHIAZEPINE DERIVATIVES OF MONOCARBONYL CURMUMIN ANALOGS Duong Quoc Hoan 1 and Nguyen Thi Thanh Xuan 2 1 Faculty of Chemistry, Hanoi National University of Education 2 Hoang Long High school Abstract. Monocarbonyl curcumin analogs 1 and 6 reacted with o-aminothiophenol gave benzothiazepines 4 and 7, respectively. The benzothiazepine cyclyzation reaction was carried out with microwave and conventional methods. Silica gel in DMF was the best choice for microwave method and gave benzothiazepine 4 in 50 % in 10 minutes. The conventional method in presence of acetic acid gave the target product 4 in 48% for 3 hours. Meanwhile, yield of compound 7 was 15% at the same condition. Retro-Aldol condensation was observed in these cases. Structures of compounds 4 and 7 were elucidated by using IR, 1D NMR, 2D NMR, and MS methods. Biological test showed that compound 4 was active against KB cancer cell line at IC50 = 47.6 g/L but it was inactive against bacteria. Keywords: Benzothiazepine, curcumin, monocarbonyl curcumin analog, KB cancer cell line. 1. Introduction In previous papers [1, 2], we published some results of modification of curcumin and curcumin analogs at both monoketone and hydroxyl groups (Figure 1, analogs 1 and 2). Unfortunately, all modifications did not give any good luck about their bioactive improvement. Figure. 1. Modifications of curcumin, curcumin analog and benzothiazepine drug Received October 29, 2018. Revised November 9, 2018. Accepted November 16, 2018. Contact Duong Quoc Hoan, e-mail: hoandq@hnue.edu.vn Duong Quoc Hoan and Nguyen Thi Thanh Xuan 96 Recently, benzothiazepine 3 derivatives have been improved that they had anti-inflammatory activity [3, 4], antifungal etc. Benzothiazepines have firmly stood as a “drug prejudice core” due to its presence in a wide range of bioactive compounds like antimicrobial, antifungal, Ca +2 antagonist, CNS depressant, antiplatelet aggregation, anti-HIV, calmodulin antagonist and bradykinin receptor antagonist [5]. The therapeutic journey of benzothiazepines can be traced back by entry of the anti-depressant “Thiazesim” into the pharmaceutical market followed by Diltiazem, Clentiazem, and Siratiazem being the cardiovascular drugs of this family [6]. Further optimization of substituents around the benzothiazepine nucleus resulted in many drugs like quetiapine fumarate and thiazesim for treating CNS disorders, 2164U90 as bile acid active transport system inhibitor and JMV1645 as bradikynin receptor antagonist [7]. This seven membered ring system is also widely acclaimed as eminent cardiotherapeutic and psychotherapeutic scaffold [8] as well as cholinesterase inhibitors [9]. Therefore, in this work, a mixture of two pairs of benzothiazepine enantiomers was obtained from reaction of a monocarbonyl curcumin analog 1 or 6 with o-aminothiphenol. The bioactivities of benzothiazepine 4 were also checked. 2. Content 2.1. Experimental section 2.1.1. General Solvents and other chemicals were purchased from Sigma-Aldrich, Merck, Aladdin and used as received, unless indicated. The 1 H NMR and 13 C NMR spectra were recorded on the Bruker Avance 500 NMR spectrometer in CDCl3. Chemical-shift data for each signal was reported in ppm. IR spectra were recorded on the Mattson 4020 GALAXY Series FT-IR. Mass spectra were obtained from Mass Spectrometry Facility of Vietnam Academy of Science and Technology on LC-MSD-Trap-SL spectrometer. The domestic microwave oven (Sanyo EM - S1065, 800W Microwave Power, made in Thailand 2005) was used. 2.1.2. Synthesis procedure (E)-4-(4-(4-hydroxy-3-methoxybenzylidene)-1,2,3,4,11,11a-hexahydrodibenzo[b,e][1,4] thiazepin-11-yl)-2-methoxyphenol (4) * Microwave method With acetic acid: A mixture of o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol), monocarbonyl curcumin analog 1 (366 mg, 1 mmol) and acetic acid (3 drops) were thoroughly mixed in a glass beaker. Then, the beaker containing the mixture partially was immersed in a sand bath in a domestic microwave oven and irradiated for 3 minutes at the medium power level (400 W). The progress of the reaction was monitored after every 1 min of irradiation by TLC with petroleum n-hexane/ethyl acetate (1:1 v/v mixture) as eluent. When reaction finished, it was diluted with ethyl acetate (20 mL). The resulting solution was extracted with ethyl acetate. The organic layer was washed with water, brined and dried over with saturated Na2SO4. Products were separated with a flash column chromatography. Without acetic acid: To a solution of o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol) and monocarbonyl curcumin analog 1 (366 mg, 1 mmol) in dichloromethane (DCM, 5 mL) was thoroughly mixed in a flask containing 2 g of silica gel. Then the solvent was evaporated in vacou to get a dry mixture. The resulting mixture was irradiated for 10 minutes at medium power level (400W). The progress of the reaction was monitored after every 1 min of irradiation by TLC with n-hexane/ethyl acetate (1:1 v/v mixture) as eluent. The benzothiazepine product 4 was purified with a flash column chromatography. Synthesis of new benzothiazepine derivatives of monocarbonyl curmumin analogs 97 Following the above procedure using: o-aminothiophenol (1 mmol, 0.1 mL, d = 1.17g/mL, 125 g/mol) and monocarbonyl curcumin analog 1 (366 mg, 1 mmol), dimethylformamide (DMF, 3 mL) and 0.5 g of silica gel. Following the above procedure using: o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol) and monocarbonyl curcumin analog 1 (366 mg, 1 mmol), dimethylformamide (DMF, 3 mL) and glacial acetic acid (2 drops). * Conventional method A solution of monocarbonyl curcumin analog 1 (366 mg, 1 mmol) and o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol) in dichloromethane (5 mL) was mixed with silica gel (2 g) in a 50mL flask. The solvent was removed by evaporation under reduced pressure and stirred at 80 ºC for 3 hr under nitrogen atmosphere. To a solution of monocarbonyl curcumin analog 1 (366 mg, 1 mmol) and o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol) in 10 mL ethanol and 2–3 drops of glacial acetic acid were added. The reaction mixture was refluxed by heating for 3 hours, and the progress of the reaction was monitored by TLC. After completion of the reaction, the resulting solution was cooled and transferred into crushed ice. The product was extracted with ethyl acetate (10 ml x3).The organic phase was washed with water, brined and dried over with saturated Na2SO4. The residue was purified with a flash column chromatography. IR (KBr, cm -1 ): 3419 (br), 3120, 2933, 2858, 1593, 1509, 1458, 1441, 1267, 1154. EI-MS: C28H27NO4S: 473.59 g/mol found for [M-H]: 472.1 au; [M+H]: 474.1au (E)-4-(3-(4-hydroxy-3-methoxybenzylidene)-2,3,10,10a-tetrahydro-1H-benzo[b]cyclopentano [e] [1,4]thiazepin-10-yl)-2-methoxyphenol (7) To a solution of o-aminothiophenol (1 mmol, 0.1 mL, d =1.17g/mL, 125 g/mol) and monocarbonyl curcumin analog 6 (350 mg, 1 mmol) in dimethylformamide (DMF, 3 mL) was thoroughly mixed in beaker containing 0.5 g of silica gel. The resulting mixture was irradiated for 10 minutes at medium power level (400 W). The progress of the reaction was monitored after every 1 min of irradiation by TLC with n-hexane/ethyl acetate (1:1 v/v mixture) as eluent. The benzothiazepine product 7 was purified with a flash column chromatography as a brown foam (69 mg, 15%, 459 g/mol). IR (KBr, cm -1 ): 3419 (br), 3100, 2932, 2858, 1598, 1509, 1458, 1442, 1265, 1150. 1 H NMR (500 MHz, CDCl3):  3.20 (m, 1H, H2), 7.51 (d, J = 7.5 Hz, 1H, H2”), 4.69 (d, J = 12.5Hz, 1H, H3), 7.40 (s, 1H, H3’), 7.15 (m, 1H, H3”), 7.47 (m, 1H, H4”), 6.67 (s, 1H, H5), 6.98 (s, 1H, H5’), 7.65 (d, J = 7.0 Hz, 1H, H5”), 6.80 (d, J = 8.0 Hz, 1H, H8), 6.92 (d, J = 8.0 Hz, 1H, H8’), 6.88 (d, J = 8.0 Hz, 1H, H9), 7.00 (d, J = 7.5Hz, 1H, H9’), 3.89(s, 3H, H10), 3.78(s, 3H, H10’), 3.00 (m, 1H,Hx1), 2.75 (m, 1H, Hx2), 1.40 (m, 1H, Hy1), 1.60 (m, 1H, Hy2) ; 13C NMR (125 MHz, CDCl3):  124.2 (C1), 137.5 (C1”), 46.2 (C2), 138.4 (C2’), 126.0 (C2”), 62.9 (C3), 130.0 (C3’), 125.2 (C3”), 135.5 (C4), 129.0 (C4’), 130.1(C4”), 108.2 (C5), 113.8 (C5’), 135.0 (C5”), 146.1 (C6), 146.3 (C6’), 152.0 (C6”), 145.4 (C7), 146.1 (C7’), 114.3 (C8), 114.0 (C8’), 119.2(C9), 124.1 (C9’), 55.9 (C10), 55.8 (C10’), 33.2 (Cx), 27.0 (Cx); EI-MS: C27H25NO4S: 459.56 g/mol found for [M-H]: 458.2 au; [M+H]: 460.4au. 2.1.3. Bioactivity test Bioactivity tests were followed by the Broth dilution method [11]. The thiazepine product 4 was selected for bacterial test including Gram (+) (Staphylococcus aureus, Bacillus subtilis, Lactobacillus fermentum) and Gram (-) (Salmonella enteric, Escherichia coli, Pseudomonas Duong Quoc Hoan and Nguyen Thi Thanh Xuan 98 aeruginosa) bacteria and fungal test (Candida albican) and for anticancer test with KB cancer cell line. All tests were carried out in the Laboratory of Applied Biochemistry of Institute of Vietnam Academy of Science and Technology. 2.2. Results and discussion 2.2.1. Synthesis Synthesis of monocarbonyl curcumin analog was following our paper [2]. Thus, the synthesis of benzothiazepine 4 from monocarbonyl curcumin analog 1 and o-aminothiophenol was screened under some methods (Scheme 1): In the microwave-assisted method, it was carried out in five conditions: in acetic acid; without acetic acid; in presence of silica gel; in silica gel and limited amount of DMF; and HOAc catalyst in DMF: conventional method: it was heated with silica gel without solvent; refluxed in ethanol and presence of acetic acid. Results were shown in Table 1. Although the mixture was irradiated in acetic acid as a protocol shown by Kumar et al. [3], unfortunately, retro-aldol condensation of the chalcone was happened. Then the vanillin was cyclized with o-aminothiophenol to yield benzo[d]thiazole 5 (Scheme 2). Similarly, the second try was without acetic acid. The reaction was burned and got a mess. Kodomari et al. published a protocol for synthesis of benzothiazepine derivatives showing that silica gel was the best catalyst for the benzothiazepine cyclization without solvent. Hence, this condition was tried in both microwave method and conventional method. Unluckily, the burn and mess were obtained giving rise to a lot of spots on TLC. Recently, in our group, DMF was found the best solvent for condensating and forming Schiff bases so the next entries were for comparison of silica gel and HOAc in DMF (Table 1). Surprisingly, silica gel in DMF (entry 4, in 50 % yield) was better than HOAc in DMF (entry 5, in 25% yield). Besides the product, it was observed the retro-aldol condensation reaction in the case of HOAc. In the conventional methods, entry 6 did not give any product; the mixture became brown black and showed a mess on TLC. Interestingly, entry 7 gave quite good yield about 48%, but it took 3 hours that was much longer than microwave method (entry 4). H3CO HO OCH3 OH O Retro-Aldol condesation H3CO HO O 2 O H2N HS H3CO HO N S H3CO HO OCH3 OH O ketone linker SN H3CO HO OCH3 OH 10' 1 2 3 4 79 10 1'' 2'' 6'' 3' x' x y 4' 6' 9' Hetrocyclic linker Conditionsa. b. 5 1 4 Scheme 1. Reaction forming thiazepine derivate 4 Scheme 2. Retro-aldol condensation 1 and forming benzo[d]thiazole Synthesis of new benzothiazepine derivatives of monocarbonyl curmumin analogs 99 Table 1. Optimization of 1,5-thiazepine cyclization Items Microwave method Conventional method Entry 1 2 3 4 5 6 7 Conditi on HOAc [3] without HOAc Silica gel Silica gel in DMF in DMF and some drops of HOAc SiO2[10] HOAc, Ethanol [4] Yield 0% 0% 0% 50% 25% 0% 48% Time 3 min. 10 min. 10 min. 10 min. 10 min. 3 h. 3 h. Remark Retro-Aldol condensation of chalcone and get benzo[d]thia zole of vanillin No reaction Clean reaction , got a mixture of two starting material s and product Clean reaction, got a mixture of two starting materials and product Retro- Aldol condensatio n of chalcone and get benzo[d]thi azole of vanillin Decompose d and mess Clean reaction, got a mixture of two starting materials and product As the acceptable procedure was in hand, the ketone 6 was treated with o-aminothiophenol under the microwave irradiation. Unfortunately, from ketone 6, the product of the retro-aldol reaction was more dominate than benzothiazepine 7 (Scheme 3). Scheme 3. Reaction of ketone 6 with o-aminothiophenol 2.2.2. Structure determination Spectral analysis of compound 7 was referred by the results of compound 4 because it wasn’t taken 2D NMR. The results were shown in the experimental part. Here is the detail of compound 4’s structural determination. In order to determine the product structure 4, it was recorded IR, 1D NMR, 2D NMR and MS spectra. IR spectrum indicated the vibrations of O-H bond at 3410 cm -1 ; C-H bonds in ranges of 3120  2858 cm-1 and stretching vibration of >C=CC=N- bonds of aromatic and benzothiazepine rings. Especially, there was no vibration of >C=O of monocarbonyl compound. On the MS spectra, the pseudo molecular peaks agreed with molecular formula C28H27NO4S: 473.59 g/mol found for [M-H]: 472.1 au; [M+H]: 474.1au (Figure 2). Figure 2 also showed HSQC and HMBC correlations of benzothiazepnine product 4. The product has 28 carbons; 27 hydrogens; one nitrogen; four oxygens and one sulfur. Moreover, this was a mixture of two pair of diastereomers making NMR analysis difficult. NMR data were shown in the table 2. 1 H NMR of benzothiazepine product 4 indicated two protons in the strong field at  3.20 ppm (m, 1H) and at  4.68 ppm (d, J = 12.5 Hz, 1H). H3 was confirmed at  4.68 ppm due to a doublet Duong Quoc Hoan and Nguyen Thi Thanh Xuan 100 peak; and H2 was at  3.20 ppm as well. These peaks were not only indicated for the cyclyzation happened but also for NMR analysis because of the left and right aromatic rings were quite similar. H3 had a cross peak with C3 at  63.1 ppm on HSQC spectrum and 6 cross peaks on HMBC spectrum (small picture on the HMBC, Figure 2) indicating C5 at 108.4ppm; C9 at 119.3 ppm; C1 at 123.5 ppm; C1” at 135.5 ppm; Cx at 26.8 and C2 at 45.4 ppm. These data helped to identify three parts: C2 and Cx for the left; C1, C5 and C9 for the right and C1” for the top. For example, H5 was at  6.66 (s,1H); H9 was at  6.88 (d, J =8.0 Hz, 1H) since they had a cross peak with C5 on HSQC spectrum. C7 was at  145.3ppm and C8 was at  114.2 ppm due to cross peaks with H9. C6 was at  146.1 ppm because it had a cross peak with H8. H10 was at  3.90 ppm (s, 3H) due to a cross peak with C6, consequently. C10 was at  55.9 ppm, C4 was at  136.0 ppm as a result of interaction with H8. A B C Figure 2. HSQC (A), HMBC (B) and a part of MS spectra (C) of compound 4 Table 2. 1 H NMR and 13 C NMR data [ (ppm), J (Hz)] 1 H NMR (δH ppm, J Hz) HMBC HnxCm 13 C NMR (δC ppm) HMBC CnxCm HSQC HnxCn - - - C1 123.5 H3, Hxa - - - - C1” 135.5 H3 H2 3.20 (m, 1H) - C2 45.4 H3 H2xC2 - - - C2’ 134.9 - - H2’’ 7.52 (d, J7.5, 1H) - C2” 125.9 H5’’ H3 4.68 (d, J12.5, 1H) C5, C9,C1, C1’’, C2, Cx C3 63.1 Hx H3xC3 Synthesis of new benzothiazepine derivatives of monocarbonyl curmumin analogs 101 H3’ 7.41 (s, 1H) - C3’ 130.0 - H3’xC3’ H3’’ 7.12 (m, 1H) C6’’ C3” 125.0 H5’’ - - C4 136.0 H8 - - - C4’ 129.1 H8’ - H4’’ 7.47 (m, 1H) C2’’ C4” 130.0 H2’’ H5 6.66 (s,1H) C7, C9 C5 108.4 H3,H9 H5xC5 H5’ 6.98 (s,1H) C9’,C7’ C5’ 113.3 H9’ H5’xC5’ H5” 7.65 (d, J7.0, 1H) C2’’ C5” 134.9 - H5” xC5” - - - C6 146.1 H10, H8 - - - - C6’ 146.7 H10’, H8’ C6” 151.5 H3’’ - - C7 145.3 H5, H9 - C7’ 146.1 H5’, H9’ H8 6.79 (d, J8.0, 1H) C4, C6 C8, 114.2 - H8xC8 H8’ 6.93 (d, J8.0, 1H) C4’, C6’ C8’ 114.2 - H9 6.88 (d, J8.0, 1H) C5, C7 C9 119.3 H3, H5 H9 x C9 H9’ 7.03 (d, J7.5, 1H) C5’, C7’ C9’ 124.0 H5' H10 3.90 (s, 3H) C6 C10 55.9 - H10 x C10 H10’ 3.78 (s, 3H) C6’ C10’ 55.8 - Hy 1.70 (m, 2H) Cy 20.5 Hxa Hy x Cy Hxe 1.37 (d, J11.0, 1H) Cx’ - - Hxe xCx Hxa 1.70 (m, 1H) Cx’ Cx 26.8 H3 Hxa x Cx Cx’ 29.8 Hxa,e Hx’e 3.20 (m, 1H) Cx - - - Hx’e x Cx Hx’a 2.61 (t, J7.5, 1H) Cx - - - Hx’a x Cx H (OH) 5.70 (br, 2H) - - - - - 2.2.3. Bioactivity test results The benzothiazepine product 4 was not active against Gram (+) (Staphylococcus aureus, Bacillus subtilis, Lactobacillus fermentum) and Gram (-) (Salmonella enteric, Escherichia coli, Pseudomonas aeruginosa) bacteria and fungus (Candida albican). Encouragingly, it performed a weak anticancer activity towards KB cancer cell line at IC50 47.6 g/mL. 3. Conclusions The benzothiazepine derivative 4 was synthesized with microwave method and conventional method in about 50 % yield. The microwave method gave many advantages such as taking short time, using less solvent... Meanwhile benzothiazepine derivative 7 was obtained in 15% since the retro-aldol condensation happened. The benzothiazepine derivative’s structures were assigned with IR, NMR and MS spectral analysis. The benzothiazepine derivative 4 was against on KB cell line at IC50 = 47.6 g/mL but it was inactive on bacterial test. Duong Quoc Hoan and Nguyen Thi Thanh Xuan 102 REFERENCES [1] Duong Quoc Hoan, Nguyen Thi Thanh Xuan, Truong Minh Luong, Tran Thi Trang, 2016. Characterization and bioactivity evaluation of two new acetohydrazides synthesized from curcumin and monocarbonyl curcumin analog. J. Sci. HNUE, Chemical and Biological Sci., 61(9), 11-20. [2] Duong Quoc Hoan, Dam Thi Uyen, Pham Thi Yen, Nguyen Hien, 2015. Synthesis and structure of some phenoxy acetic acid derivatives from curcumin and monocarbonyl curcumin analogs. Vietnam J. Chem., 53 (6e1,2), 348-353. [3] Kumar, R. R., Perumal, S., 2007. A facile synthesis and highly atom economic 1,3-dipolar cycloaddition of hexahydropyrido[3,4-c][1,5]benzothiazepines with nitrile oxide: stereoselective formation of hexahydro- [1,2,4]oxadiazolo[5,4-d]pyrido[3,4- c][1,5]benzothiazepines. Tetrahedron, 63,7850–7857. [4] Mhaske, G. R., Bajod, S. S., Ambhore, D. M., Shelke, S. N., 2014. Synthesis and Evaluation of Novel 1, 5-Benzothiazepine Derivatives as Anti- Inflammatory Agents. International Journal of Innovative Research in Science, Engineering and Technology, Vol. 3, Issue 6, 13208-13215. [5] a) Kawakita, S., Kinoshita, M., Ishikawa, H., Kagoshima, T., Katori, R., Ishikawa, K., Hirota, Y., 1991. Efficacy and safety of clentiazem in patients with essential hypertension: results of an early pilot test. Clin. Cardiol., 14, 53-60. b). Geyer, H. M., Watzman, N., Buckley, J. P., 1970. Effects of a tranquilizer and two antidepressants on learned and unlearned behaviors. J. Pharm.Sci., 59, 964. [6] Bagal, S. K., Brown, A. D., Cox, P. J., Omoto, K., Owen, R. M., Pryde, D. C., Sidders, B., Skerratt, S. E., Stevens, E. B., Storer, R. I., Swain, N. A., 2013. Ion channels as therapeutic targets: a drug discovery perspective. J. Med. Chem., 56, 593-624. [7] Ye, N., Neumeyer, J. L., Baldessarini, R. J., Zhen, X., Zhang, A., 2013. Update 1 of: Recent progress in development of dopamine receptor subtype-selective agents: potential therapeutics for neurological and psychiatric disorders. Chem. Rev., 113, 123-178. [8] Bariwal, J. B., Upadhyay, K. D., Manvar, A. T., Trivedi, J. C., Singh, J. S., Jain, K. S., Shah, A. K, 2008. 1,5-Benzothiazepine, a versatile pharmacophore: a review. Eur. J. Med. Chem., 43, 2279. [9] Farzana Latif Ansari, Saima Kalsoom, Zaheer-ul-Haq, Zahra Ali, Farukh Jabeen, 2012. In silico studies on 2,3-dihydro-1,5-benzothiazepines as cholinesterase inhibitors. Med. Chem. Res. 21, 2329–2339. [10] Kodomari, M., Tomohiro Noguchi, and Tadashi Aoyama, 2004. Solvent-Free Synthesis of 1,5- Benzothiazepines and Benzodiazepines on Inorganic Supports. Synthetic Communications, 34(10), 1783-1790. [11] Ericsson, J. M.; Sherris, J. C., 1971. Antibiotic sensitivity testing: report of an international collaborative study. Acta Pathol Microbiol Scand, 217, 1-90.