INTRODUCTION

In recent years, chemoselectivity in 1,3-dipolar cycloadditions of nitrile imines to multifunctional dipolarophiles possessing two or more different dipolarophilic sites has attracted the interest of several research groups [1]. For example, the cycloaddition of nitrile imines to compounds I and II (Fig. 1) which contain exocyclic C=C and endocyclic C=N double bonds was found to occur selectively to the exocyclic C=C double bond to give the corresponding spiro pyrazole derivative [2, 3]. Contrary to the foregoing reports, it was found that the cycloaddition of nitrile imines to III and IV (Fig. 1) occurred selectively to the endocyclic C=N double bond to yield fused 1,2,4-triazoles [4, 5]. Both spiro pyrazoles and fused 1,2,4-triazoles have been reported to exhibit a wide range of biological activities [6, 7].

Fig. 1.
figure 1

Structures of compounds I–IV.

Full size image

Quantum chemical calculations play an important role in elucidation and confirmation of the actual structure of heterocyclic compounds and stereoselectivity of the reactions. In this article, we deeply investigated the reasons behind preferences of one mechanism over the other by density functional theory (DFT) calculations.

Guided by the above-mentioned findings and continuing our studies on nitrile imine precursors [8–16], we thought it interesting to explore the site selectivity in the cycloaddition of nitrile imines to 11-aryl-4-(arylmethylidene)-1,2,3,4,11,11a-hexahydrodibenzo[b,e][1,4]thiazepines 3 (Scheme 1) which possess two dipolarophilic sites, namely the exocyclic C=C and endocyclic C=N double bonds. In addition, the cycloaddition products seemed promising for studying their in vitro antitumor activity.

Scheme
scheme 1

1.

Full size image

RESULTS AND DISCUSSION

The required bifunctional dipolarophiles 3 were prepared by refluxing an equimolar mixture of 2,6-bis(arylmethylidene)cyclohexan-1-ones 1a and 1b with 2-aminobenzenethiol (2) in ethanol in the presence of acetic acid. On the basis of spectroscopic and elemental analyses data, the isolated compounds proved to have structure 3 while structure 4 was discarded. For example, the 1H NMR spectra of 3, apart from signals of the aromatic protons and the three CH2 groups, showed one doublet at δ 5.05 ppm (1H, J = 12.6 Hz, 11-H) and a multiplet signal near δ 3.00 ppm (1H, 11a-H). The large coupling constant J = 12.6 Hz indicated E configuration of the exocyclic C=C double bond. Table 1 contains the total energies (Etot), HOMO and LUMO energies (EHOMO, ELUMO), and dipole moments (μ) of all possible products and reactants in all possible pathways of the proposed mechanisms. For the reaction between 1 and 2 to form either 3 or 4, the total energy of 3 is less than that of 4 by 6.28 kcal/mol. The dipole moment and energy gap of 3 are comparable to those of initial compound 1. Molecule 4 has the highest dipole moment among 1–4. Compound 2 has a higher band gap than compounds 1, 3, and 4 which are characterized by comparable values. Higher band gap usually indicates lower reactivity which leads to the conclusion that the reaction is mainly driven by the high reactivity of 1.

Table 1. Table 1. The total energy Etot, energies of HOMO, LUMO, energy gaps ΔE, and dipole moments μ of structures 1–16
Full size table

Compounds 3a and 3b were reacted with C,N-diaryl nitrile imines 6 generated in situ from the corresponding hydrazonoyl halides 5 by the action of triethylamine in benzene under reflux. In each case, only one pure product was formed (according to the data of TLC monitoring). These findings indicated that the described reaction is chemo- and regioselective. Theoretically, the addition of nitrile imines 6 to compounds 3 could lead to the formation of four isomeric cycloadducts 7–10 (Scheme 2). The isolated compounds were assigned structure 7 on the basis of spectroscopic data (see Experimental). In addition to signals of aromatic protons, three CH2 groups, 11-H, and 11a-H, the 1H NMR spectra of 7a–7i showed a singlet of 4′-H (dihydropyrazole ring) in the region δ 5.4–5.6 ppm. The position of the latter signal is in agreement with published data for structurally related dihydropyrazoles, whereas proton in the 5-position of dihydropyrazoles usually resonates at a lower field (δ > 5.6 ppm)[17, 18].

Scheme
scheme 2

2.

Full size image

The formation of compounds 7 rather than 8–10 was also confirmed by theoretical calculations. The total energies of 7 and 8 are lower than those of 9 and 10 by at least 27 kcal/mol. The energy of the reaction calculated by subtracting the sum of the energies of the reactants (compounds 3 and 6) from the energy of the product (compound 7 or 8) was 32.86 and 34.76 kcal× mol–1 for the formation of 7 and 8, respectively. Thus, the difference is only ~5.7%. This small energy difference cannot be regarded as strong evidence in favor of one or another pathway. The molecular orbital plot was utilized as the best way for explanation the postulated mechanism. We expected that the reaction proceed through the interaction of HOMO of one reactant with LUMO of the other as usually interpreted in cycloaddition reactions. An obvious pattern by which route the organic reaction will proceed is obtained by contrasting the signs of the molecular orbital (MO) lobes at the reaction sites. Similar signs of the lobes indicate constructive interference (bonding overlap), while different sings indicate destructive interference (anti-bonding overlap). Figures 2–5 show the LUMO of 6 and HOMO of 3 in different orientations to each other to form compound 7–10. It is clear that Figs. 2 and 4 correspond to the constructive reaction, whereas it is destructive in Figs. 3 and 5. Thus, the formation of 7 and 9 is possible, and the formation of 8 and 10 is not. Joint consideration of the total energies and molecular orbital interactions led us to conclude that the only possible pathway of the reaction between 3 and 6 is to produce compound 7.

Fig. 2.
figure 2

Molecular orbital plot for the LUMO of compound 6 (above) and HOMO of 3 (below) in the orientation to form compound 7.

Full size image
Fig. 3.
figure 3

Molecular orbital plot for the LUMO of compound 6 (above) and HOMO of 3 (below) in the orientation to form compound 8.

Full size image
Fig. 4.
figure 4

Molecular orbital plot for the LUMO of compound 6 (above) and HOMO of 3 (below) in the orientation to form compound 9.

Full size image
Fig. 5.
figure 5

Molecular orbital plot for the LUMO of compound 6 (above) and HOMO of 3 (below) in the orientation to form compound 10.

Full size image

The reactions of 3a and 3b with C-ethoxycarbonyl-N-aryl nitrile imines 12 generated in situ from the corresponding hydrazonoyl chlorides 11 in refluxing benzene in the presence of triethylamine also yielded only one product in each case (compounds 13a–13e; Scheme 3). The structure of the isolated compounds was determined on the basis of their IR, 1H NMR, and mass spectra and elemental analyses (see Experimental). The other possible isomeric structures, 14–16 (Scheme 3) were thus discarded. For example, the 1H NMR spectra of the products lacked dihydropyrazole 4-H proton signal found in the spectra of 7. These findings indicated that the products obtained in the reaction of 3 with 12 resulted from 1,3-dipolar cycloaddition of the latter to the endocyclic C=N double bond rather than to the exocyclic C=C bond.

Scheme
scheme 3

3.

Full size image

A similar treatment can be applied for the reaction of 3 with 12 to yield one of compounds 13–16. As followed from the total energy calculations, compounds 13 and 14 are more stable than 15 and 16 by more than 30 kcal/mol. The molecular orbital simulation showed that the LUMO of 12 has no π* character, so that the first π*-molecular orbital is LUMO+1. Figures 6–9 show the LUMO+1 of 12 and HOMO of 3 in the orientations required to form structures 13–16. The reaction is constructive in the cases corresponding to Figs. 6 and 8 and is destructive in the cases depicted in Figs. 7 and 9. These MO plots indicate the possibility of formation of 13 and 15. Likewise, a combination of the total energy and molecular orbital considerations suggested only one possibility, i.e., the formation of only compound 13.

Fig. 6.
figure 6

Molecular orbital plot for the LUMO+1 of compound 12 (above) and HOMO of 3 (below) in the orientation to form compound 13.

Full size image
Fig. 7.
figure 7

Molecular orbital plot for the LUMO+1 of compound 12 (above) and HOMO of 3 (below) in the orientation to form compound 14.

Full size image
Fig. 8.
figure 8

Molecular orbital plot for the LUMO+1 of compound 12 (above) and HOMO of 3 (below) in the orientation to form compound 15.

Full size image
Fig. 9.
figure 9

Molecular orbital plot for the LUMO+1 of compound 12 (above) and HOMO of 3 (below) in the orientation to form compound 16.

Full size image

To explore the significance of the newly synthesized compounds, the in vitro antitumor activity of dibenzothiazepine derivatives 3, 7, and 13 was tested at the Regional Center for Mycology and Biotechnology (Al-Azhar University, Cairo, Egypt) against MCF-7 breast cancer cell line; doxorubicin (IC50 = 0.426 µg· mL–1) was used as reference drug. The results are collected in Table 2. Compounds 3b, 13a, 13b, and 13c showed a promising pharmacological activity (IC50 = 2.7, 2.9, 6.2, and 13.5 µg/mL, respectively), while the activity of the other tested compounds was only moderate. It is important that the conversion of compound 3a (IC50 = 24.8 µg/mL) into triazole derivatives 13a–13c enhanced the antitumor activity.

Table 2. Cytotoxicity of compounds 3a, 3b, 7a–7c, 7e, 7g, 7i, and 13a–13c against MCF-7 cancer cell line
Full size table

Table 3 shows the results of theoretical toxicity predictions for dibenzothiazepine series 3, 7, and 13 using the Osiris program; it was found that the toxicity of all these compounds is lower than that of doxorubicin (DOX). It should also be noted that dibenzothiazepine 7d can be used as antibiotic with some pharmacomodulation (DS = 1.85). It follows from the data in Table 3 that the tested structures are non-mutagenic. With regard to irritant and reproductive effects, all dibenzothiazepine compounds 3, 7, and 13 are at low risk compared to the standard medicine used. The hydrophilicity character of each derivative was expressed in terms of cLogP value. It has been shown that absorption or permeability is significantly affected by hydrophilicity (cLogP value). When the cLogP value is greater than five, the permeability or absorption decreases. On this basis, most dibenzothiazepine derivatives have cLogP values outside accepted range, but another critical parameter must be considered. This is related to the geometric structure of the pharmacophore site (Fig. 10) which is flexible in all benzothiazepine derivatives. The absorption, distribution properties, and bio-efficiency proved to be dependent on the geometrical parameters and solubility in water.

Table 3. Osiris calculations of toxicity risksa of dibenzothiazepine derivatives 3, 7, and 13
Full size table
Fig. 10.
figure 10

Identification of pharmacophore sites on the basis of POM theory [19–34].

Full size image

Further, drug-likeness (DL) of 3, 7, and 13 is not in the comparable zone with the standard drug used (Table 3). We have estimated DS (the overall drug score) for dibenzothiazepine derivatives 3, 7, and 13 and compared it with that of the standard drug (doxorubicin). The drug score is an integral parameter comprising drug-likeness, cLogS, cLogP, toxicity risks, and molecular weight in one handy value that may be applied to judge the overall ability of tested compound to qualify for a drug. The newly synthesized compounds showed low to moderate DS values in comparison to DOX (Tables 3, 4).

Table 4. Molinspiration calculations of compounds 3, 7, and 13
Full size table

We succeeded in synthesizing a new series of spiro-1,2,4-triazole derivatives containing a hexahydrodibenzo[b,e][1,4]thiazepine moiety, starting from commercially available compounds. The antitumor activity of these molecules was evaluated against breast cancer cell line (MCF-7) using doxorubicin as reference. Compound 13b showed higher cytotoxicity than all other tested compounds; however, its biological activity remains modest (IC50 = 2.90 µg/mL) in comparison to doxorubicin (IC50 = 0.426 µg/mL). The results of POM analysis of the relative cytotoxicity of these new derivatives in comparison to doxorubicin have been reported. It seems that compounds of the 7a–7i series possess an important antiparazite/antifungal and antiviral N,N-pharmacophore site which deserves a separate antiviral/antiparasite screening. Our previous experience with similar spiro systems suggests that a subtle change in the pharmacophore can give rise to antitubercular, antitumor, antitrypanosomal, and/or anti-HIV activities.

EXPERIMENTAL

Melting points were determined on a Stuart SMP3 melting point apparatus using 0.5-mm (o.d.) glass capillaries. The IR spectra (4000 to 200 cm–1) were recorded on a Perkin Elmer 1430 spectrophotometer from samples prepared as KBr discs. The NMR spectra were acquired on a Bruker Avance 400 instrument at 400 MHz for 1H in DMSO-d6 solutions, using the residual solvent signal as reference. The mass spectra (electron impact, 70 eV) were obtained on a Finnigan-MAT 8222 instrument at the Microanalytical Center (Cairo University). Elemental analyzes were carried out on a Vario-LIII Elementar CHNS analyzer (Germany). Initial compounds 1a, and 1b and hydrazonoyl chlorides 5 and 11 were prepared as reported previously [35–37].

Compounds 3a and 3b (general procedure). A mixture of 1.7 g (5 mmol) of 2,6-bis(4-chlorobenzylidene)cyclohexan-1-one (1a) or 1.6 g (5 mmol) of 2,6-bis(4-fluorobenzylidene)cyclohexan-1-one (1b), 0.65 g (5 mmol) of 2-aminobenzenethiol (2), and 3 mL of acetic acid in 30 mL of ethanol was refluxed for 5 h. The solvent was evaporated under reduced pressure, and the product was filtered off and recrystallized from ethanol.

4-(4-Chlorobenzylidene)-11-(4-chlorophenyl)-1,2,3,4,11,11a-hexahydrodibenzo[b,e][1,4]thiazepine (3a). Yield 1.6 g (70%), yellow crystals, mp 180–182°C. IR spectrum, ν, cm–1: 3055, 2927, 2859, 1553, 1483, 1443, 1093. 1H NMR spectrum, δ, ppm: 1.11–1.80 m and 2.53–3.17 m (7H, CH2, CH), 5.05 d (J = 12.6 Hz , 1H, CH), 7.15 – 7.58 m ( 13H, Ar-H, =CH); 13C NMR spectrum, δC, ppm: 23.12, 27.0, 27.9, 39.1, 57.21, 119.6, 125.8, 126.9, 127.5, 128.2, 129.0, 130.1, 130.4, 131.5 131.9, 133.2, 134.4, 135.3, 136.3, 150.2, 152.1, 165.8. Mass spectrum, m/z (Irel, %): 452 (1) [M + 2]+, 451 (6) [M + 1]+, 450 (6) [M]+, 353 (35), 338 (100), 323 (24), 311 (18), 308 (19), 126 (24), 90 (19), 76 (24), 63 (28), 50 (24) Found, %: C 69.21; H 4.56; N 3.02. C26H21Cl2NS. Calculated, %: C 69.33; H 4.70; N 3.11. M 450.42.

4-(4-Fluorobenzylidene)-11-(4-fluorophenyl)-1,2,3,4,11,11a-hexahydrodibenzo[b,e][1,4]thiazepine (3b). Yield 1.7 g (80%), yellow fluorescing crystals, mp 174–176°C. IR spectrum, ν, cm–1: 3052, 2935, 2841, 1567, 1482, 1440, 1093. 1H NMR spectrum, δ, ppm: 1.06–1.85 m and 3.00–3.14 m (7H, CH2, CH), 5.05 d (1H, J = 12.6 Hz, CH), 7.10–7.31 m and 7.45–7.61 m (12H, Harom, =CH), 7.38 s (1H, =CH). 13C NMR spectrum, δC, ppm: 22.70, 27.10, 29.37, 37.12, 57.54, 120.27, 123.93, 125.90, 128.27, 128.36, 129.28, 131.54, 134.21, 135.36, 135.94, 136.65, 137.10, 140.54, 147.15, 149.33, 150.11, 163.12. Mass spectrum, m/z (Irel, %): 419 (4) [M + 2]+, 418 (10) [M + 1]+, 417 (32) [M]+, 308 (10), 280 (5), 270 (12), 268 (100), 254 (12), 162 (5), 160 (5), 149 (6), 146 (10), 136 (8), 133 (28), 109 (44), 107 (6), 83 (7), 69 (7). Found, %: C 74.57; H 5.16; N 3.19. C26H21F2NS. Calculated, %: C 74.79; H 5.07; N 3.35. M 417.51.

Compounds 7a–7i and 13a–13e (general procedure). Triethylamine (0.28 mL, 2 mmol), was added to a mixture of equimolar amounts of compound 3a (0.9 g, 2 mmol) or 3b (0.84 g, 2 mmol) and the corresponding hydrazonoyl chloride 5 (in the synthesis of 7a–7i) or 11 (in the synthesis of 13a–13e) (2 mmol) in dry benzene (30 mL), and the mixture was refluxed for 20 h. The solvent was evaporated, the liquid residue was triturated with methanol, and the solid product was collected by filtration under vacuum and crystallized from dioxane.

4′,11-Bis(4-chlorophenyl)-2′,5′-diphenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7a). Yield 0.64 g (50%), greenish brown solid, mp 220–222°C. IR spectrum, ν, cm–1: 3048, 2928, 2866, 1625, 1589, 1487, 1451, 1371, 1285, 1091. 1H NMR spectrum, δ, ppm: 1.06–1.70 m and 3.38–4.40 m (7H, CH2, CH), 5.45 d (1H, J = 12.6 Hz, C–H), 5.50 s (1H, 4′-H), 6.58–7.71 m (22H, Harom). 13C NMR spectrum, δC, ppm: 21.0, 26.27, 28.2, 36.7, 57.6, 80.2, 117.8, 118.2, 121.1, 122.0, 124.3, 125.9, 127.1, 127.6, 128.7, 128.9, 129.1, 129.3, 130.2, 130.4, 131.8, 133.5, 134.1, 134.5, 136.4, 137.5, 140.7, 145.4, 152.6, 154.7, 160.2. Mass spectrum, m/z (Irel, %): 646 (11) [M + 2]+, 645 (23) [M + 1]+, 644 (23) [M]+, 642 (25), 370 (18), 359 (35), 357 (100), 344 (25), 342 (20), 321 (18), 286 (24), 272 (16), 268 (22), 204 (19), 162 (48), 136 (16), 128 (25), 125 (40), 118 (15), 115 (15), 104 (15), 96 (23), 91 (17), 77 (50), 65 (17). Found, %: C 72.48; H 4.62; N 6.41%. C39H31Cl2N3S. Calculated, %: C 72.66; H 4.85; N 6.52. M 644.66.

4′,5′,11-Tris(4-chlorophenyl)-2′-phenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7b). Yield 0.68 g (50%), brown solid, mp 210–212°C. IR spectrum, ν, cm–1: 3053, 2928, 2860, 1588, 1485, 1286, 1257, 1093. 1H NMR spectrum, δ, ppm: 1.11–2.36 m and 2.99– 3.16 m (7H, CH2, CH), 5.06 d (1H, J = 12.6 Hz, CH), 5.47 s (1H, 4′-H), 6.60–8.27 m (21H, Harom). Mass spectrum, m/z (Irel, %): 680 (10) [M + 2]+, 679 (13) [M + 1]+, 678 (24) [M]+, 676 (24), 575 (25), 573 (50), 571 (47), 393 (45), 391 (77), 379 (19), 377 (32), 340 (15), 338 (23), 302 (17), 286 (40), 284 (52), 274 (17), 272 (33), 204 (19), 162 (39), 149 (19), 138 (16), 136 (29), 127 (17), 125 (33), 115 (31), 108 (17), 102 (17), 91 (37), 89 (21), 77 (100), 64 (23), 51 (19). Found, %: C 69.05; H 4.34; N 6.36. C39H30Cl3N3S. Calculated, %: C 68.98; H 4.45; N 6.19. M 679.10.

4′,11-Bis(4-chlorophenyl)-2′-(4-nitrophenyl)-5′-phenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7c). Yield 0.87 g (63%), red crystals, mp 226–228°C (from EtOH). IR spectrum, ν, cm–1: 2923, 2855, 1588, 1489, 1455, 1296, 1102. 1H NMR spectrum, δ, ppm: 0.77–1.78 m and 2.99–3.32 m (7H, CH2, CH), 5.04 d (1H, J = 12.6 Hz , CH), 5.66 s (1H, 4′-H), 6.57–8.19 m (21H, Harom). Mass spectrum, m/z (Irel, %): 690 (2) [M + 1]+, 689 (2) [M]+, 540 (75), 537 (100), 536 (34), 380 (20), 372 (26), 284 (26), 272 (19), 204 (16), 151 (19), 149 (16), 136 (25), 129 (23), 127 (25), 125 (37), 115 (48), 109 (16), 105 (16), 103 (21), 91 (19), 89 (35), 77 (46), 64 (22), 62 (21), 51 (21). Found, %: C 68.12; H 4.49; N 8.26. C39H30Cl2N4O2S. Calculated, %: C 67.92; H 4.38; N 8.12. M 689.65.

4′,11-Bis(4-chlorophenyl)-2′-(4-nitrophenyl)-5′-(thiophen-2-yl)-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7d). Yield 0.79 g (57%), red crystals, mp 248–250°C (from EtOH). IR spectrum, ν, cm–1: 3061, 2923, 2858, 1586, 1490, 1377, 1294. 1H NMR spectrum, δ, ppm: 1.03–1.90 m and 2.51–3.10 m ( 7H, CH2, CH), 5.35 d (1H, J = 12.6 Hz, CH), 5.58 s (1H, 4′-H), 6.46–8.19 m (19H, Harom). Mass spectrum, m/z (Irel, %): 697 (5) [M + 2]+, 696 (3) [M + 1]+, 695 (20) [M]+, 663 (23), 545 (60), 544 (50), 543 (53), 542 (83), 272 (33), 139 (30), 126 (23), 115 (37), 103 (47), 85 (37), 84 (33), 77 (43). Found, %: C 63.95; H 4.31; N 7.98. C37H28Cl2N4O2S2. Calculated, %: C 63.88; H 4.06; N 8.05. M 695.68.

4′,11-Bis(4-fluorophenyl)-2′,5′-diphenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7e). Yield 0.59 g (48%), white crystals, mp 214–216°C (from EtOH). IR spectrum, ν, cm–1: 3049, 2924, 2864, 1595, 1499, 1452, 1373, 1289. 1H NMR spectrum, δ, ppm: 1.12–2.40 m and 3.20–3.25 m (7H, CH2, CH), 5.45 d (1H, J = 12.6 Hz, CH), 5.52 s (1H, 4′-H), 6.59 d (4H, J = 8 Hz, Harom), 6.89–7.41 m (14H, Harom), 7.69 d (4H, J = 8 Hz, Harom). Mass spectrum, m/z (Irel, %): 613 (2) [M + 2]+, 612 (9) [M + 1]+, 611 (17) [M]+, 506 (100), 504 (24), 356 (14), 341 (71), 327 (31), 306 (16), 270 (17), 268 (34), 267 (18), 256 (13), 251 (16), 224 (13), 162 (11), 147 (10), 136 (16), 133 (21), 115 (13), 109 (51), 107 (14), 104 (16), 91 (24), 77 (69), 65 (13), 51 (16). Found, %: C 76.72; H 5.35; N 6.59. C39H31F2N3S. Calculated, %: C 76.57; H 5.11; N 6.87. M 611.75.

5′-(4-Chlorophenyl)-4′,11-bis(4-fluorophenyl)-2′-phenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7f). Yield 0.52 g (40%), yellow crystals, mp 60–62°C (from EtOH). IR spectrum, ν, cm–1: 3058, 2921, 2852, 1596, 1550, 1503, 1404, 1369, 1305. 1H NMR spectrum, δ, ppm: 0.74–3.15 m (7H, CH2, CH), 5.50 d (1H, J = 13 Hz, CH), 5.40 s (1H, 4′-H), 6.93–7.65 m (21H, Harom). Mass spectrum, m/z (Irel, %): 647 (8) [M + 2]+, 646 (11) [M + 1]+, 645 (21) [M]+, 643 (17), 377 (37), 375 (86), 361 (16), 270 (25), 268 (100), 256 (17), 162 (25), 133 (18), 109 (39), 107 (16), 77 (38). Found, %: C 72.58; H 4.39; N 6.58. C39H30ClF2N3S. Calculated, %: C 72.49; H 4.68; N 6.50. M 646.19.

4′,11-Bis(4-fluorophenyl)-2′-(4-nitrophenyl)-5′-phenyl-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7g). Yield 0.83 g (63%), orange crystals, mp 254–256°C. IR spectrum, ν, cm–1: 3063, 2933, 1590, 1500, 1388, 1306. 1H NMR spectrum, δ, ppm: 1.14–2.43 m and 2.75– 3.27 m (7H, CH2, CH), 5.46 d (1H, J = 13 Hz, CH), 5.67 s (1H, 4′-H), 6.50–8.32 m (21H, Harom). Mass spectrum, m/z (Irel, %): 656 (13), [M]+, 506 (100), 504 (18), 472 (21), 396 (22), 394 (13), 372 (23), 364 (28), 256 (30), 327 (13), 325 (12), 312 (10), 270 (21), 268 (43), 257 (14), 256 (23), 253 (12), 250 (10), 236 (12), 222 (15), 207 (10), 197 (13), 185 (10), 183 (16), 177 (10), 168 (10), 165 (10), 151 (10), 146 (14), 135 (25), 133 (25), 121 (16), 110 (15), 108 (33), 105 (14), 103 (15), 101 (13), 95 (13), 91 (14), 89 (22), 82 (12), 78 (12), 76 (23), 74 (10), 65 (22), 63 (16), 51 (15). Found, %: C 71.52; H 4.84; N 8.24. C39H30F2N4O2S. Calculated, %: C 71.32; H 4.60; N 8.53. M 656.74.

4′,11-Bis(4-fluorophenyl)-2′-(4-nitrophenyl)-5′-(thiophen-2-yl)-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7h. Yield 0.68 g (51%), orange crystals, mp 218–220°C. IR spectrum, ν, cm–1: 3444, 2934, 2869, 2589, 1502, 1448, 1377, 1294. 1H NMR spectrum, δ, ppm: 1.12–2.80 m (7H, CH2, CH), 4.91 s (1H, 4′-H), 5.38 d (1H, J = 13 Hz, CH), 6.47–8.28 m (19H, Harom). Mass spectrum, m/z (Irel, %): 664 (12) [M + 2]+, 663 (4) [M + 1]+, 662 (39) [M]+, 567 (21), 472 (15), 440 (30), 364 (23), 309 (11), 241 (15), 122 (5), 95 (63), 83 (16), 76 (34). Found, %: C 67.25; H 4.39; N 8.18. C37H28F2N4O2S2. Calculated, %: C 67.05; H 4.26; N 8.45. M 662.77.

4′,11-Bis(4-fluorophenyl)-5′-(furan-2-yl)-2′-(4-nitrophenyl)-2,2′,3,4′,11,11a-hexahydro-1H-spiro[dibenzo[b,e][1,4]thiazepine-4,3′-pyrazole] (7i). Yield 0.84 g (65%), red crystals, mp 250–252°C (from EtOH). IR spectrum, ν, cm–1: 3070, 2923, 2858, 1590, 1502, 1453, 1371, 1323, 1302. 1H NMR spectrum, δ, ppm: 1.03–3.44 m (7H, CH2, CH), 5.34 d (1H, J = 13 Hz, CH), 5.49 s (1H, 4′-H), 6.51–8.19 m (19H, Harom). Mass spectrum, m/z (Irel, %): 646 (10) [M]+, 495 (100), 462 (17), 346 (20), 268 (26), 256 (21), 147 (15), 139 (15), 136 (13), 133 (20), 121 (14), 115 (22), 109 (44), 93 (17), 77 (17), 75 (16), 67 (15), 65 (22), 63 (19), 58 (15), 53 (13), 51 (24). Found, %: C 68.97; H 4.15; N 8.54. C37H28F2N4O3S. Calculated, %: C 68.72; H 4.36; N 8.66. M 646.71.

Ethyl 4-(4-chlorobenzylidene)-8-(4-chlorophenyl)-3-(4-methylphenyl)-4,5,6,7,7a,8-hexahydro-3H-dibenzo[b,e][1,2,4]triazolo[4,3-d][1,4]thiazepine-1-carboxylate (13a). Yield 0.68 g (52%), yellow solid, mp 130–132°C. IR spectrum, ν, cm–1: 3045, 2925, 2857, 1700 (C=O), 1575, 1480, 1376. 1H NMR spectrum, δ, ppm: 0.83 t (3H, J = 7 Hz, CH3), 1.09–1.80 m and 2.80–3.15 m (6H, CH2), 2.39 s (3H, CH3), 3.88 q (2H, J = 6 Hz, OCH2), 4.70 m (1H, CH), 5.06 d (1H, J = 12 Hz, CH), 7.19–7.63 m (17H, Harom, =CH). 13C NMR spectrum, δC, ppm: 14.1, 19.4, 23.5, 28.1, 30.3, 40.7, 54.8, 62.2, 109.3, 115.8, 119.4, 120.6, 122.0, 124.3, 125.5, 126.6, 127.5, 128.2, 129.1, 129.9, 130.7, 131.6, 132.0, 133.2, 133.7, 137.6, 141.7, 142.5, 146.2, 155.4, 159.3. Mass spectrum, m/z (Irel, %): 655 (27) [M + 1]+, 654 (33) [M]+, 630 (26), 601 (31), 597 (26), 592 (31), 590 (27), 554 (26), 532 (39), 523 (33), 521 (31), 513 (29), 509 (31), 507 (35), 506 (33), 502 (34), 499 (27), 489 (32), 486 (30), 460 (100), 451 (28), 443 (29), 404 (38), 397 (36), 386 (37), 377 (32), 367 (31), 365 (63), 363 (34), 336 (36), 332 (70), 298 (32), 284 (30), 256 (32), 245 (43), 240 (32), 201 (38), 194 (27), 179 (32), 130 (35), 100 (33), 80 (27), 72 (28). Found, %: C 67.75; H 5.14; N 6.53. C37H33Cl2N3O2S. Calculated, %: C 67.88; H 5.08; N 6.42. M 654.65.

Ethyl 4-(4-chlorobenzylidene)-3,8-bis(4-chlorophenyl)-4,5,6,7,7a,8-hexahydro-3H-dibenzo[b,e][1,2,4]triazolo[4,3-d][1,4]thiazepine-1-carboxylate (13b). Yield 0.65 g (48%), pale yellow solid, mp 142–144°C. IR spectrum, ν, cm–1: 3045, 2924, 2856, 1702 (C=O), 1582, 1514, 1482, 1445, 1375. 1H NMR spectrum, δ, ppm: 0.82 t (3H, J = 7 Hz, CH3), 1.10–1.72 m and 2.78–3.18 m (6H, CH2), 3.90 q (2H, J = 7 Hz, OCH2), 4.70 d (1H, J = Hz, CH), 5.05 d (1H, J = 12.3 Hz, CH), 7.14–7.62 m (17H, Harom, =CH). Mass spectrum, m/z (Irel, %): 675 (15) [M]+, 674 (19), 597 (18), 565 (18), 493 (18), 486 (19), 462 (44), 460 (52), 451 (24), 449 (30), 404 (23), 396 (24), 332 (21), 323 (22), 296 (25), 288 (27), 286 (42), 284 (100), 274 (25), 271 (22), 248 (26), 235 (25), 224 (22), 192 (22), 181 (21), 163 (27), 153 (28), 149 (33), 136 (22), 129 (25), 127 (49), 125 (73), 121 (29), 117 (27), 115 (79), 113 (23), 111 (21), 109 (43), 105 (22), 102 (25), 99 (21), 94 (32), 91 (25), 89 (44), 80 (95), 77 (38), 72 (25), 69 (41), 64 (92). Found, %: C 64.14; H 4.16; N 6.29. C36H30Cl3N3O2S. Calculated, %: C 64.05; H 4.48; N 6.22. M 675.07.

Ethyl 4-(4-chlorobenzylidene)-8-(4-chlorophenyl)-3-(4-nitrophenyl)-4,5,6,7,7a,8-hexahydro-3H-dibenzo[b,e][1,2,4]triazolo[4,3-d][1,4]thiazepine-1-carboxylate (13c). Yield 0.96 g (70%), orange solid, mp 180–182°C. IR spectrum, ν, cm–1: 3055, 2925, 2853, 1699 (C=O), 1593, 1513, 1480, 1444, 1377. 1H NMR spectrum, δ, ppm: 0.81 t (3H, J = 7 Hz, CH3), 1.14–1.95 m and 2.67–3.20 m (6H, CH2), 3.92 q (2H, J = 7 Hz, OCH2), 4.70 m (1H, CH), 5.05 d (1H, J = 11.4 Hz, CH), 7.17–7.63 m (17H, Harom, =CH). Mass spectrum, m/z (Irel, %): 686 (17) [M + 1]+, 685 (25) [M]+, 661 (21), 598 (28), 574 (24), 553 (21), 505 (22), 498 (24), 494 (22), 481 (20), 475 (25), 455 (22), 406 (26), 393 (21), 367 (23), 363 (21), 354 (26), 349 (26), 342 (22), 338 (32), 326 (30), 312 (25), 300 (21), 288 (21), 282 (24), 263 (22), 247 (22), 224 (21), 218 (20), 213 (23), 210 (24), 199 (22), 194 (33), 180 (22), 178 (23), 161 (25), 152 (22), 145 (21), 141 (24), 126 (31), 120 (23), 115 (38), 113 (23), 111 (28), 93 (34), 90 (28), 86 (93), 80 (100), 76 (38). Found, %: C 63.12; H 4.51; N 8.02. C36H30Cl2N4O4S. Calculated, %: C 63.06; H 4.41; N 8.17. M 685.62.

Ethyl 4-(4-fluorobenzylidene)-8-(4-fluorophenyl)-3-(4-methylphenyl)-4,5,6,7,7a,8-hexahydro-3H-dibenzo[b,e][1,2,4]triazolo[4,3-d][1,4]thiazepine-1-carboxylate (13d). Yield 0.84 g (68%), yellow solid, mp 194–196°C. IR spectrum, ν, cm–1: 3045, 2929, 2859, 1700 (C=O), 1596, 1508, 1472, 1448, 1377. 1H NMR spectrum, δ, ppm: 0.82 t (3H, J = 7 Hz, CH3), 1.07–2.34 m and 2.68–3.20 m (6H, CH2), 2.36 s (3H, CH3), 3.91 q (2H, J = 7 Hz, OCH2), 4.65 m (1H, CH), 5.03 d (1H, J = 11.7 Hz, CH), 7.03–7.61 m (17H, Harom, =CH). Mass spectrum, m/z (Irel, %): 622 (52) [M + 1]+, 621 (51) [M]+, 617 (45), 580 (81), 563 (45), 553 (48), 538 (54), 519 (61), 502 (68), 477 (42), 473 (45), 463 (50), 457 (46), 450 (45), 441 (65), 432 (46), 421 (51), 414 (45), 410 (63), 393 (53), 386 (45), 382 (47), 373 (62), 371 (48), 364 (100), 355 (57), 341 (53), 334 (88), 330 (63), 323 (63), 310 (86), 293 (62), 290 (47), 246 (64), 240 (75), 216 (51), 187 (48), 167 (42), 152 (42), 97 (64), 81 (55), 79 (58). Found, %: C 71.41; H 5.39; N 6.69. C37H33F2N3O2S. Calculated, %: C 71.48; H 5.35; N 6.76. M 621.74.

Ethyl 4-(4-fluorobenzylidene)-8-(4-fluorophenyl)-3-(4-nitrophenyl)-4,5,6,7,7a,8-hexahydro-3H-dibenzo[b,e][1,2,4]triazolo[4,3-d][1,4]thiazepine-1-carboxylate (13e). Yield 0.91 g (70%), brown solid, mp 178–180°C. IR spectrum, ν, cm–1: 3038, 2922, 2852, 1696 (C=O), 1598, 1505, 1474, 1441, 1373. 1H NMR spectrum, δ, ppm: 0.82 t (3H, J = 7 Hz, CH3), 1.15–1.90 m and 2.77–3.08 m (6H, CH2), 3.89 q (2H, J = 7 Hz, OCH2), 4.69 m (1H, CH), 5.04 d (1H, J = 11.4 Hz, CH), 7.13–7.63 m (17H, Harom, =CH). Mass spectrum, m/z (Irel, %): 655 (54) [M + 3]+, 654 (47) [M + 2]+, 653 (57) [M + 1]+, 652 (19) [M]+, 642 (51), 639 (45), 626 (55), 617 (56), 598 (45), 594 (70), 591 (43), 572 (46), 570 (57), 567 (46), 556 (56), 554 (56), 547 (54), 543 (43), 530 (62), 528 (46), 515 (47), 510 (50), 508 (41), 499 (66), 460 (65), 455 (58), 449 956), 442 (56), 400 (55), 397 (64), 382 (64), 360 (55), 337 (58), 330 (77), 319 (54), 306 (79), 300 (61), 263 (56), 239 (58), 229 (56), 218 (65), 201 (64), 199 (66), 190 (68), 176 (60), 174 (54), 170 (68), 166 (62), 154 (52), 148 (76), 145 (75), 132 (77), 106 (100), 102 (58). Found, %: C 66.19; H 4.59; N 8.52. C36H30F2N4O4S. Calculated, %: C 66.24; H 4.63; N 8.58. M 652.71.

Quantum chemical calculations. Hybrid density functional theory B3LYP method [38–43] was utilized to calculate the molecular geometry of all newly synthesized dibenzothiazepine derivatives. All calculations were performed using Gaussian 09W software package [44]. The calculated structures were visualized using GaussView version 5.0.9 [45].

Anticancer activity. Eleven dibenzothiazepine derivatives 3a, 3b, 7a–7c, 7e, 7g, 7i, and 13a–13c were tested against MCF-7 human breast cancer cell line by the known MTT [3-(4,5-dimethyl-1,3-thiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay, and the results were compared with the data for doxorubicin used as reference drug. The IC50 values were determined from the dose–response curve plotted on the basis of the MTT assay data. The cytotoxicity was expressed as mean IC50 value for three independent runs (Table 2). The procedure was identical to that published by Vijayan et al. [46] using Crystal Violet dye (1%). The cells were siphoned into a 96-well plate to a concentration of 1×104 cells per well (100 μL). The microplates were incubated at 37°C for 48 h in a humidified incubator (5% CO2). Three wells were used for each concentration value. Control wells contained only DMSO without any test compound. After incubation of cells for 24 h at 37°C, different concentrations of test samples (50, 25, 12.5, 6.25, 3.125, and 1.56 μg) were added, the incubation was continued for 48 h, and a solution of Crystal Violet was added to each well for at least 30 min. Excess dye was removed using tap water, 30% acetic acid (30%) was then added to each well, the content of the wells was thoroughly mixed, and the absorbance at λ 490 nm was measured using a microplate reader. All results were corrected for background absorption detected in the wells without adding a stain. All experiments were conducted in triplicate. The results are collected in Table 2.