Synthesis of novel [3,1]-benzothiazepine and [3,1]-benzoxazepine derivatives with antitumoral activity

Abstract

A new method for the synthesis of [3,1]-benzothiazepines and [3,1]-benzoxazepines from the reaction of C-allylanilines and isothiocyanates or isocyanates without the need for the isolation of any intermediate is described. The compounds were obtained in good to moderate yields and some exhibited cytotoxic activity against tumor cell lines.

---

Benzothiazepines and benzoxazepines constitute important building blocks in pharmaceutical research while they can be found in many drugs and preclinical leads.¹

The synthesis of different isomeric forms of these compounds by varying the position of the heteroatoms in their structures together with their unique pharmacological properties make the development of new methods for the synthesis of medium-sized heterocycles containing N, S and O atoms a subject of great interest. Thus, several isomers of benzothiazepine² and benzoxazepine³ rings were already described and some of them exhibited a wide variety of pharmacological effects.⁴

Herein, we wish to describe a new method for the synthesis of [3,1]-benzothiazepines and [3,1]-benzoxazepines for further evaluation of their antitumoral activities. The general method for the synthesis of [3,1]-benzoxazepines is based on the photochemical isomerization of quinoline N-oxides.⁵ However, it requires a significant number of steps for preparation of the starting materials used in the cyclization event, and sometimes proceed with a lack of regioselectivity.⁶ In the case of [3,1]-benzothiazepines derivatives, to our knowledge there are even few reports in the literature describing the synthesis of this class of compounds.⁷

We initially focused on experiments to find a route which gave the required starting materials. Thus, allylation of commercially available anilines 1a–d was performed following a literature procedure⁸ to yield the corresponding N-allylanilines 2a–d in good yields (Scheme 1).

Scheme 1 Synthesis of N-allylanilines 2a–d.

Subsequent aza-Claisen rearrangement⁹ at 114–140 °C without the use of any solvent gave the corresponding C-allylanilines 3a–d in good yields (Scheme 2).

Scheme 2 Synthesis of C-allylanilines 3a–d.

C-Allylanilines, 3 were then reacted with commercially available isothiocyanates, in dichloromethane to give in situ the corresponding thioureas. Subsequent addition of iodine¹⁰ to the reaction mixture gave the corresponding [3,1]-benzothiazepines 4 through a 7-exo-trig mode cyclization in good yields and without the isolation of any intermediate compound (Table 1).

Table 1 Synthesis of [3,1]-benzothiazepines 4 through iodocyclization

Entry	3	Ar	Time (h)	[3,1]-Benzothiazepine		Yield^a (%)
a Isolated yield.b A complex mixture of products was obtained.
1	3a	C6H5	72		4a	—^b
2	3b	4-Cl-C6H4	99		4b	74
3	3b	3,4-Me2-C6H3	72		4c	37
4	3b	C10H7	120		4d	63
5	3c	4-Cl-C6H4	72		4e	69
6	3c	C10H7	96		4f	76
7	3d	C6H5	72		4g	52
8	3d	4-Cl-C6H4	120		4h	77
9	3d	4-F-C6H4	48		4i	70
10	3d	4-MeO-C6H4	120		4j	64
11	3d	3,4-Me2-C6H3	120		4k	74

---

The reaction yield does not seem to be strongly influenced by the electronic effects on the groups attached to the aromatic ring of both C-allylanilines 3 or isothiocyanates since the [3,1]-benzothiazepines bearing neutral, electron-withdrawing and electron-donating groups in the aromatic rings were compatible with the reaction conditions. The only exception occurred when 3a was used as precursor, where a complex mixture of products was observed. Further attempts to purify 4a from this mixture were unsuccessful (Table 1, entry 1).

Next, we extended the method for the synthesis of the [3,1]-benzoxazepines derivatives, 5a–c. In this case, the reaction solvent was replaced by tetrahydrofuran due to the low solubility of intermediates N,N′-diarylureas in dichloromethane.

Thus, reaction of C-allylanilines, 3a–b with phenylisocyanate led to the formation in situ of the corresponding N,N′-diarylureas after one hour. Subsequent cyclisation using iodine gave the corresponding [3,1]-benzoxazepines 5a–c in moderate yields (Table 2). The reaction using isocyanates gave lower yields (38–56%) when compared to the reaction using thiocyanates. This result could be explained by the lower nucleophilic character of the oxygen atom from the carbonyl group of the synthesized ureas when compared to the sulfur atom present in the thioureas.

Table 2 Synthesis of [3,1]-benzoxazepines 5 through iodocyclization

Entry	3	[3,1]-Benzoxazepine, 5		Yield (%)
1	3a		5a	47
2	3b		5b	56
3	3c		5c	38

---

The cyclisation of ureas or thioureas could lead to two isomers being the C=N bond endo or exo (Fig. 1). For example, for compound 5a, if the signal attributed to the proton bonded to nitrogen – Hb – presents an HMBC correlation with the carbon nuclei Cb from ring B, the obtained compound should have an endo C=N bond (Fig. 1, structure 5a). However, if the hydrogen atom bonded to nitrogen –Ha – presents an HMBC correlation with the carbon nuclei Ca from ring A, the obtained compound should have an exo C=N bond (Fig. 1, structure 5a′).

Fig. 1 HMBC correlation between the endo and exo compounds.

From the spectrum obtained for 5a, ring A shows four different signals in the ¹H NMR spectrum at δ_H 7.74 (d, 8.1 Hz, 1H), 7.23 (d, 7.2 Hz, 1H), 7.14 (t, 8.1 Hz, 1H) and 6.93 (td, 7.5, 1.0 Hz, 1H). Conversely, ring B shows only three signals at δ_H 7.52 (dd, 8.7, 0.9 Hz, 2H), 7.34–7.26 (m, 2H) and 7.07–7.00 (m, 1H). HMBC spectrum shown a correlation between the signal at δ_H 8.85, corresponding to N–H and the signals at δ_C 152.0 and 120.6. The signal at δ_C 152.0 was attributed to C=N. In addition, the signal at δ_C 120.6 showed an HMQC correlation with the signal at δ_H 7.52, attributed to the proton at the ortho position in ring B. In this way, the signal at δ_C 120.6 was attributed to the carbon at the ortho position in the ring B, confirming the structure of the obtained compound as 5a.

The in vitro anticancer activity of benzoxathiepin,¹¹ benzodioxepin¹² and benzoxazepines¹³ on MCF-7 cells has been previously described. Further studies of 4,1-benzoxazepines by microarray technology showed that the main molecular targets of some of these compounds are pro-apoptotic genes with protein kinase activity such as GP132, ERN1 or RAC1, which prevent the metastatic progression. These facts prompt us to submit the synthesized compounds to the MTT assay¹⁴ for the evaluation of their cytotoxic effects on tumor cells.

The synthesized compounds were then screened at 25 μg mL⁻¹ against HL-60 (human pro-myelocytic leukemia), NCI-H292 (human lung carcinoma) and HT-29 (human colon carcinoma) tumor cells lineages. The samples with growth inhibition over 90% were used to determine the IC₅₀ values (concentration that causes 50% growth inhibition). Analogues 4h and 4i exhibited good cytotoxicities and their IC₅₀ values (μg mL⁻¹) were determined. The results are described in Table 3.

Table 3 IC₅₀ and standard error values (μg mL⁻¹) of two different experiments for compounds 4h and 4i

Compound	HL-60	NCI-H292	HT-29
a Doxorubicin (DOX) was the positive control. *p < 0.05 compared compound 4i to 4h by t test.
4h	8.2 ± 0.2	15.1 ± 1.6	14.8 ± 1.1
4i	2.1 ± 0.9*	12.4 ± 3.1	7.7 ± 0.8*
DOX^a	0.02 ± 0.005	0.1 ± 0.05	0.4 ± 0.05

---

From Table 3 it can be seen that compound 4h displayed moderate cytotoxic activity against all tested cancer cell lines, being more active for HL-60 cells. Compound 4i exhibited better antiproliferative activity against HL-60 and HT-29 cell lines,¹⁵ being also more active for HL-60 with an IC₅₀ = 2.1 μg mL⁻¹ demonstrating the potential of the newly synthesized compounds as antitumoral agents.

Conclusions

In conclusion, the method demonstrates to be useful for the synthesis of [3,1]-benzothiazepines and [3,1]-benzoxazepines derivatives without the need of the isolation of any intermediate.

Compounds 4h and 4i exhibit good antitumoral activity for HL-60 cells and are promising intermediates for the synthesis of an array of more potent target structures while the iodine atom can be used as an additional point of diversity.

Acknowledgements

The authors gratefully acknowledge CNPq (484778/2011-0), FACEPE (PRONEX APQ-0859-1.06/08), CAPES and INCT-INAMI for financial support. The authors are also thankful to Central Analítica (IQ-UNICAMP) for HRMS. P.H.M is thankful to CNPq for his fellowship.

Notes and references

1. (a) J. B. Bariwal, K. D. Upadhyay, A. V. Manvar, J. C. Trivedi, J. S. Singh, K. S. Jain and A. K. Shah, Eur. J. Med. Chem., 2008, 43, 2279; (b) X. Wu, F. Liu and Y. Zhou, J. Heterocycl. Chem., 2011, 48, 368; (c) R. Budriesi, B. Cosimelli, P. Ioan, M. P. Ugenti, E. Carosati, M. Frosini, F. Fusi, R. Spisani, S. Saponara, G. Cruciani, E. Novellino, D. Spinelli and A. Chiarini, J. Med. Chem., 2009, 52, 2352; (d) C. W. Lindsley, ACS Chem. Neurosci., 2010, 1, 343; (e) C. W. Yin, S. K. Wo and Z. Zuo, J. Pharm. Biomed. Anal., 2012, 58, 83; (f) S. Fukamachi, H. Konishi and K. Kobayashi, Heterocycles, 2011, 83, 883.
2. Benzothiazepines synthesis: [1,5] (a) Z. Q. Dong, F. M. Liu and Z. L. Yuan, Mol. Diversity, 2011, 15, 963; (b) D. B. Yadav, B. Dharmendra, G. L. Morgans, B. A. Aderibigbe, L. G. Madeley, M. A. Fernandes, J. P. Michael, C. B. de Koning and W. A. L. van Otterlo, Tetrahedron, 2011, 67, 2991; (c) C. B. W. Phippen and C. S. P. McErlean, Tetrahedron Lett., 2011, 52, 1490; (d) E. Ayral, P. Gloanec, G. Berge, G. de Nanteuil, P. Mennecier, A. Rupin, T. J. Verbeuren, P. Fulcrand, J. Martinez and J. F. Hernandez, Bioorg. Med. Chem. Lett., 2009, 19, 1386 ; [1,4]; (e) C. Spitz, V. Reboul and P. Metzner, Tetrahedron Lett., 2011, 52, 6321; (f) L. Fodor, P. Csomós, A. Csampai and P. Sohar, Synthesis, 2010, 2943; (g) M. Incerti, D. Acquotti, P. Sandor and P. Vicini, Tetrahedron, 2009, 65, 7487 ; [1,3]; (h) G. Campiani, S. Butini, C. Fattorusso, B. Catalanotti, S. Gemma, V. Nacci, E. Morelli, A. Cagnotto, I. Mereghetti, T. Mennini, M. Carli, P. Minetti, M. A. Di Cesare, D. Mastroianni, N. Scafetta, B. Galletti, M. A. Stasi, M. Castorina, L. Pacifici, M. Vertechy, S. di Serio, O. Ghirardi, O. Tinti and P. Carminati, J. Med. Chem., 2004, 47, 143; (i) G. Campiani, V. Nacci, S. Bechelli, S. M. Ciani, A. Garofalo, I. Fiorini, H. Wikstrom, P. de Boer, Y. Liao, P. G. Tepper, A. Cagnotto and T. Mennini, J. Med. Chem., 1998, 41, 3763 ; [1,2]; (j) R. A. Abramovitch, R. A. Mavukel, B. Mavukel and J. R. Stowers, J. Chem. Soc., Chem. Commun., 1983, 520; (k) Y. Tamura, S. M. Bayomi, C. Mukai, M. Ikeda and M. Murase, Tetrahedron Lett., 1980, 21, 533.
3. Benzoxapines synthesis: [1,5]: (a) J. F. Liegeois, M. Deville, S. Dilly, C. Lamy, F. Mangin, M. Resimont and F. I. Tarazi, J. Med. Chem., 2012, 55, 1572; (b) Y. Uto, Y. Ueno, Y. Kiyotsuka, Y. Miyazawa, H. Kurata, T. Ogata, T. Takagi, S. Wakimoto and J. Ohsumi, Eur. J. Med. Chem., 2011, 46, 1892; (c) A. Levai, Heterocycles, 2008, 75, 2155 ; [1,4]; (d) M. Ghandi, T. Momeni, M. T. Nazeri, N. Zarezadeh and M. Kubicki, Tetrahedron Lett., 2013, 54, 2983; (e) H. Kwiecien, M. Smist and A. Wrzesniewska, Curr. Org. Synth., 2012, 9, 828 ; [1,3]; (f) N. Kakusawa, K. Sakamoto, J. Kurita and T. Tsuchiya, Heterocycles, 1996, 43, 2091; (g) A. R. Katrizky, O. Rubio, R. Awartani, N. Latif, N. Mishriky and F. M. Assad, Heterocycles, 1984, 22, 1155; (h) A. Albini, G. F. Bettinetti and G. Minoli, Tetrahedron Lett., 1979, 20, 3761 ; [1,2]; (i) T. Kametani and H. Nemoto, Chem. Pharm. Bull., 1971, 19, 1325.
4. (a) K. Lombardo, M. A. Stasi and F. Borsini, Eur. J. Pharmacol., 2009, 621, 53; (b) K. Lombardo, M. A. Stasi and F. Borsini, Eur. J. Pharmacol., 2009, 621, 53; (c) R. Coccurello, A. Caprioli, R. Conti, O. Ghirardi, F. Borsini, P. Carminati and A. Moles, J. Med. Chem., 2004, 47, 143; (d) R. Coccurello, A. Caprioli, R. Conti, O. Ghirardi, F. Borsini, P. Carminati and A. Moles, J. Pharmacol. Exp. Ther., 2008, 326, 905; (e) G. Campiani, S. Butini, S. Gemma, V. Nacci, C. Fattorusso, B. Catalanotti, G. Giorgi, A. Cagnotto, M. Goegan, T. Mennini, P. Minetti, M. A. Di Cesare, D. Mastroianni, N. Scafetta, B. Galletti, M. A. Stasi, M. Castorina, L. Pacifici, O. Ghirardi, O. Tinti and P. Carminati, J. Med. Chem., 2002, 45, 344.
5. (a) A. Albini, G. F. Bettinetti and G. Minoli, Org. Synth., 1990, 7, 23; (b) C. Kaneko, H. Fujii, K. Hashiba, Y. Karasawa, M. Wakai, R. Hayashi and M. Somei, Chem. Pharm. Bull., 1982, 30, 74; (c) A. Albini, G. F. Bettinetti and G. Minoli, Tetrahedron Lett., 1979, 20, 98; (d) C. Kaneko, A. Yamamoto and M. Hashiba, Chem. Pharm. Bull., 1979, 27, 946; (e) N. Hata, Chem. Lett., 1978, 7, 1359; (f) C. Kaneko and R. Kitamura, Heterocycles, 1977, 6, 111; (g) R. Kitamura, H. Fujii, K. Hashiba, M. Somei and C. Kaneko, Tetrahedron Lett., 1977, 18, 2911; (h) S. Yamada, M. Ishikawa and C. Kaneko, Chem. Pharm. Bull., 1975, 23, 2818; (i) C. Kaneko, S. Hayashi and Y. Kobayashi, Chem. Pharm. Bull., 1974, 22, 2147; (j) A. Kubo, S.-I. Sakai, S. Yamada, I. Yokoe and C. Kaneko, Chem. Pharm. Bull., 1968, 16, 1533; (k) C. Kaneko, I. Yokoe and M. Ishikawa, Tetrahedron Lett., 1967, 8, 5237; (l) C. Kaneko and I. Yokoe, Tetrahedron Lett., 1967, 8, 5355; (m) O. Buchardt, Tetrahedron Lett., 1966, 7, 6221.
6. (a) W. Y. Yang, S. A. Marrone, N. Minors, D. A. R. Zorio and I. V. Alabugin, Beilstein J. Org. Chem., 2011, 7, 813; (b) T. Besson, G. Guillaumet, C. Lamazzi and C. W. Rees, Synlett, 1997, 704; (c) P. Dalidowicz and J. S. Swenton, J. Org. Chem., 1993, 58, 4802; (d) G. Capozzi, R. Ottana and G. Romeo, Heterocycles, 1987, 26, 39; (e) J. P. Le Roux, P. L. Desbene and J. C. Cherton, J. Heterocycl. Chem., 1981, 18, 847; (f) Y. Ito, K. Kobayashi and T. Saegusa, Tetrahedron Lett., 1978, 19, 2087; (g) C. Kaneko, H. Fujii, S. Kawai and M. Somei, Chem. Lett., 1978, 1277; (h) C. Kaneko, S. Kawai and M. Somei, Chem. Lett., 1978, 1281; (i) Y. Kobayashi, I. Kumadaki, Y. Hirose and Y. Hanzawa, J. Org. Chem., 1974, 39, 1836.
7. W. Fathalla, J. Marek and P. Pazdera, Heterocycl. Commun., 2002, 8, 79.
8. I. Gonzalez, I. Bellas, A. Souto, R. Rodríguez and J. Cruces, Tetrahedron Lett., 2008, 49, 2002.
9. L. M. Acosta, A. Palma and A. Bahsas, Tetrahedron, 2010, 66, 8392.
10. (a) M. Jereb, D. Vrazic and M. Zupan, Tetrahedron, 2011, 67, 1355; (b) K. Gilmore and I. V. Alabugin, Chem. Rev., 2011, 111, 6513; (c) B. Godoi, R. F. Schumacher and G. Zeni, Chem. Rev., 2011, 111, 2937.
11. M. C. Nunez, F. Rodriguez-Serrano, J. A. Marchal, O. Caba, A. Aranega, M. A. Gallo, A. Espinosa and J. M. Campos, Tetrahedron, 2007, 63, 183.
12. A. Conejo-Garcia, M. C. Nunez, J. A. Marchal, F. Rodriguez-Serrano, A. Aranega, M. A. Gallo, A. Espinosa and J. M. Campos, Eur. J. Med. Chem., 2008, 43, 1742.
13. (a) K. Samanta, B. Chakravarti, J. K. Mishra, S. K. D. Dwivedi, L. V. Nayak, P. Choudhry, H. K. Bid, R. Konwar, N. Chattopadhyay and G. Panda, Bioorg. Med. Chem. Lett., 2010, 20, 283; (b) M. Diaz-Gavilan, F. Rodriguez-Serrano, J. A. Gomez-Vidal, J. A. Marchal, A. Aranega, M. A. Gallo, A. Espinosa and J. M. Campos, Tetrahedron, 2004, 60, 11547.
14. J. M. Edmondson, L. S. Armstrong and A. O. Martinez, J. Tissue Cult. Methods, 1988, 11, 15.
15. In order to understand if the IC₅₀ values are statistically significant, the t test, comparing 4i to 4h was performed. The results showed that the IC₅₀ values for 4i and 4h are different only for HL-60 and HT-29.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, spectra and spectroscopic data for all compounds. See DOI: 10.1039/c3ra44937h

---