Systematic Studies on Anti-Cancer Evaluation of Stilbene and Dibenzo[b,f]oxepine Derivatives

Cancer is one of the most common causes of human death worldwide; thus, numerous therapies, including chemotherapy, have been and are being continuously developed. In cancer cells, an aberrant mitotic spindle—a microtubule-based structure necessary for the equal splitting of genetic material between daughter cells—leads to genetic instability, one of the hallmarks of cancer. Thus, the building block of microtubules, tubulin, which is a heterodimer formed from α- and β-tubulin proteins, is a useful target in anti-cancer research. The surface of tubulin forms several pockets, i.e., sites that can bind factors that affect microtubules’ stability. Colchicine pockets accommodate agents that induce microtubule depolymerization and, in contrast to factors that bind to other tubulin pockets, overcome multi-drug resistance. Therefore, colchicine-pocket-binding agents are of interest as anti-cancer drugs. Among the various colchicine-site-binding compounds, stilbenoids and their derivatives have been extensively studied. Herein, we report systematic studies on the antiproliferative activity of selected stilbenes and oxepine derivatives against two cancer cell lines—HCT116 and MCF-7—and two normal cell lines—HEK293 and HDF-A. The results of molecular modeling, antiproliferative activity, and immunofluorescence analyses revealed that compounds 1a, 1c, 1d, 1i, 2i, 2j, and 3h were the most cytotoxic and acted by interacting with tubulin heterodimers, leading to the disruption of the microtubular cytoskeleton.


Introduction
Cancer, the leading cause of human death worldwide, remains difficult to treat, despite a huge effort to develop novel therapies [1][2][3][4][5][6]. Chemotherapy, a traditional and commonly used anti-cancer treatment, targets specific proteins and, as a consequence, cellular structures or processes [7]. Among the most common targets of chemotherapy are proteins involved in cell cycle progression [8]. In cancer cells, the mitotic spindle-a structure that is indispensable for cell division-frequently shows abnormalities, as it causes the improper division of genetic material between daughter cells and, as a consequence, genomic instability [8]. The scaffold of the mitotic spindle is formed by microtubules (MTs), which are cylindrical, hollow biopolymers that are continuously polymerized and depolymerized. MTs are composed of tubulin heterodimers consisting of two~55 kDa proteins, αand β-tubulin. In interphase cells, MTs form a network involved in different cellular processes, including the establishment and maintenance of cell shape, intracellular transport, organelle distribution, and cell motility [9]. Thus, in both interphase and dividing cells, tubulin is among the most useful targets in chemotherapy [7].
The surface of tubulin contains so-called "pockets" or "sites" that can accommodate small molecules, which are called microtubule-targeting agents (MTAs). Intercalation of MTAs with the pockets affects MTs' polymerization/depolymerization, leading to changes On the other hand, CSBAs that include stilbenoids are poorly soluble in water, and they cause strong side effects [19]. For example, CA4P/fosbretabulin, which is generally well tolerated, causes acute but transient hypertension [25]. Therefore, efforts are being made to develop new stilbenoid derivatives in the search for more potent anti-cancer agents with improved features [26][27][28][29][30][31]. On the other hand, CSBAs that include stilbenoids are poorly soluble in water, and they cause strong side effects [19]. For example, CA4P/fosbretabulin, which is generally well tolerated, causes acute but transient hypertension [25]. Therefore, efforts are being made to develop new stilbenoid derivatives in the search for more potent anti-cancer agents with improved features [26][27][28][29][30][31].
In our previous study [39,40], we synthesized and evaluated the cytotoxicity of five stilbenes (1a-1e) and six dibenzo[b,f ]oxepines (2a-2e, 2h) (the latter are frequently used as scaffolds in medicinal chemistry, they have a (Z)-stilbene motif in their skeleton, and their aromatic rings are connected by oxygen). We showed that one substituted dibenzo[b,f ]oxepine and two stilbene derivatives, while operating through tubulin binding, acted more selec-tively toward cancer cell lines (HeLa and U87) than toward normal cell lines. In search of novel compounds with higher cytotoxicity and selectivity toward cancer cells, we studied previously obtained compounds and synthesized new derivatives of hydroxy, methoxy, nitro, and amine groups with 4 stilbenes (1f-1i) and 13 dibenzo[b,f ]oxepines (2f, 2g, 2i, 2j, 3a-3j). For the new compounds, we determined their structure by using NMR and theoretical calculations. For all 28 compounds, we evaluated cytotoxicity for the HCT116 and MCF-7 cell lines. We collected these compounds to conduct a systematic study of stilbenes with methoxy and nitro groups and of dibenzo[b,f ]oxepines with nitro, methoxy, and amino groups. Our results indicated that the addition of a methoxy group could maintain or abolish the cytotoxicity of stilbene derivatives in a position-dependent manner, while, with some exceptions, the restriction of the double bonds that connected phenyl rings to create a (Z)-stilbene motif with an oxygen bridge decreased the compounds' activity. Moreover, we demonstrated that three of the most active agents were selective toward cancer cell lines and that they targeted tubulin, thus disrupting the microtubular cytoskeleton.

Chemistry and NMR Spectra
We synthesized a series of compounds 3a-h, 3j in three synthetic steps. The synthetic routes are summarized in Scheme 1. Various stilbene moieties 1a-i were obtained by starting from the condensation of suitable 2-hydroxyaldehyde and 2,4-dinitrotoluene with pyrrolidine as a catalyst to obtain the substituted stilbene. The next step was the reaction between a derivative of 2,4-dinitrostilbene 1a-i and sodium azide. In the reactions, the corresponding nitrodibenzo[b,f ]oxepine 2a-i appeared. Nitrodibenzo[b,f ]oxepine bearing a hydroxyl substituent 2i was protected in the reaction with acetic anhydride and catalytic amounts of concentrated sulfuric acid, which yielded an acetoxy derivative of dibenzo[b,f ]oxepine 2j. Nitrodibenzo[b,f ]oxepines 2a-f, 2h, 2j were converted into amino derivatives 3a-f, 3h, 3j in the presence of Zn in acetic acid. Nitrodibenzo[b,f ]oxepine with two nitro groups 2g was selectively reduced to 3g by using dicobalt octocarbonyl in water. methoxy groups, can enhance the biological activity of natural stilbenoids [32][33][34][35][36][37][38].
In our previous study [39,40], we synthesized and evaluated the cytotoxicity of five stilbenes (1a-1e) and six dibenzo[b,f]oxepines (2a-2e, 2h) (the latter are frequently used as scaffolds in medicinal chemistry, they have a (Z)-stilbene motif in their skeleton, and their aromatic rings are connected by oxygen). We showed that one substituted dibenzo[b,f]oxepine and two stilbene derivatives, while operating through tubulin binding, acted more selectively toward cancer cell lines (HeLa and U87) than toward normal cell lines. In search of novel compounds with higher cytotoxicity and selectivity toward cancer cells, we studied previously obtained compounds and synthesized new derivatives of hydroxy, methoxy, nitro, and amine groups with 4 stilbenes (1f-1i) and 13 dibenzo[b,f]oxepines (2f, 2g, 2i, 2j, 3a-3j). For the new compounds, we determined their structure by using NMR and theoretical calculations. For all 28 compounds, we evaluated cytotoxicity for the HCT116 and MCF-7 cell lines. We collected these compounds to conduct a systematic study of stilbenes with methoxy and nitro groups and of dibenzo[b,f]oxepines with nitro, methoxy, and amino groups. Our results indicated that the addition of a methoxy group could maintain or abolish the cytotoxicity of stilbene derivatives in a position-dependent manner, while, with some exceptions, the restriction of the double bonds that connected phenyl rings to create a (Z)-stilbene motif with an oxygen bridge decreased the compounds' activity. Moreover, we demonstrated that three of the most active agents were selective toward cancer cell lines and that they targeted tubulin, thus disrupting the microtubular cytoskeleton.

Chemistry and NMR Spectra
We synthesized a series of compounds 3a-h, 3j in three synthetic steps. The synthetic routes are summarized in Scheme 1. Various stilbene moieties 1a-i were obtained by starting from the condensation of suitable 2-hydroxyaldehyde and 2,4-dinitrotoluene with pyrrolidine as a catalyst to obtain the substituted stilbene. The next step was the reaction between a derivative of 2,4-dinitrostilbene 1a-i and sodium azide. In the reactions, the corresponding nitrodibenzo[b,f]oxepine 2a-i appeared. Nitrodibenzo[b,f]oxepine bearing a hydroxyl substituent 2i was protected in the reaction with acetic anhydride and catalytic amounts of concentrated sulfuric acid, which yielded an acetoxy derivative of dibenzo[b,f]oxepine 2j. Nitrodibenzo[b,f]oxepines 2a-f, 2h, 2j were converted into amino derivatives 3a-f, 3h, 3j in the presence of Zn in acetic acid. Nitrodibenzo[b,f]oxepine with two nitro groups 2g was selectively reduced to 3g by using dicobalt octocarbonyl in water. In order to determine the structures of the reaction products of the derivatives of stilbenes 1a-i and dibazo[b,f ]oxepines 2a-j, 3a-h, 3j in solution, the 1 H and 13 C NMR spectra of all of the products were measured (see Figures S8-S61 in the Supplementary Materials). The coupling constants ( 1 H-1 H) were measured directly by using the resolution-enhanced 1D spectra and were confirmed, when necessary, through homo-decoupling. The coupling constants ( 1 H-1 H) for olefin protons α and β in compounds 1a-i were about 16 Hz (E configuration) and changed to about 11 Hz for molecules 2a-j, 3a-h, 3j (Z configuration). In summary, we developed a method with mild conditions and operational simplicity for

Anti-Cancer Potential of Stilbenes and Oxepines
To verify if the synthesized compounds showed cytotoxic effects in cancer cells, we determined the viability of compound-treated HTC116 cells by performing an MTT colorimetric assay ( Figure 2) [41]. Besides compound 1a, in which only the hydroxy group was present at position 2, the most active stilbenes were 1c, 1d, 1i; e.g., compounds that were also at position 4 substituted with methoxy or hydroxy groups in 1c, 1i or in the compound 1d containing a methoxy group at position 5. Generally, we observed that the stilbene derivatives were more cytotoxic than the dibenzo[b,f ]oxepine derivatives and that the substitution of -NO 2 with -NH 2 decreased oxepine activity, but often improved its solubility in organic solvents (such as dimethyl sulfoxide). The opposite tendency was observed in the 2h, 3h series, but these were the only derivatives in which the phenyl ring was substituted by the naphthalene ring. In two series of derivatives, 1b, 2b, 3b and 1f, 2f, 3f, cell viability was only minimally affected. These agents had a methoxy group in the ortho position or two methoxy groups in the meta and para positions. Therefore, these substitutions seemed to abolish the anti-cancer activity of stilbene and oxepine derivatives. stilbenes 1a-i and dibazo[b,f]oxepines 2a-j, 3a-h, 3j in solution, the H and C NMR spec-tra of all of the products were measured (see Figures S8-S61 in the Supplementary Materials). The coupling constants ( 1 H-1 H) were measured directly by using the resolution-enhanced 1D spectra and were confirmed, when necessary, through homo-decoupling. The coupling constants ( 1 H-1 H) for olefin protons α and β in compounds 1a-i were about 16 Hz (E configuration) and changed to about 11 Hz for molecules 2a-j, 3a-h, 3j (Z configuration). In summary, we developed a method with mild conditions and operational simplicity for the versatile synthesis of derivatives of stilbenes or dibenzo[b,f]oxepines 1a-i, 2a-2j, 3a-h, 3j from substituted benzaldehydes and 2,4-dinitrotoluene.

Anti-Cancer Potential of Stilbenes and Oxepines
To verify if the synthesized compounds showed cytotoxic effects in cancer cells, we determined the viability of compound-treated HTC116 cells by performing an MTT colorimetric assay ( Figure 2) [41]. Besides compound 1a, in which only the hydroxy group was present at position 2, the most active stilbenes were 1c, 1d, 1i; e.g., compounds that were also at position 4 substituted with methoxy or hydroxy groups in 1c, 1i or in the compound 1d containing a methoxy group at position 5. Generally, we observed that the stilbene derivatives were more cytotoxic than the dibenzo[b,f]oxepine derivatives and that the substitution of -NO2 with -NH2 decreased oxepine activity, but often improved its solubility in organic solvents (such as dimethyl sulfoxide). The opposite tendency was observed in the 2h, 3h series, but these were the only derivatives in which the phenyl ring was substituted by the naphthalene ring. In two series of derivatives, 1b, 2b, 3b and 1f, 2f, 3f, cell viability was only minimally affected. These agents had a methoxy group in the ortho position or two methoxy groups in the meta and para positions. Therefore, these substitutions seemed to abolish the anti-cancer activity of stilbene and oxepine derivatives. Interestingly, a previous report on the anti-cancer activity of compound 2h in U87 (glioblastoma) and HeLa (cervical cancer) cell lines [39] showed that it was significantly Figure 2. The cytotoxicity effect of tested compounds 1a-g, 1i, 2a-2j, 3a-3h, 3j at 100 µM on cancer HCT116 cells based on MTT assay after 48h of treatment.
Interestingly, a previous report on the anti-cancer activity of compound 2h in U87 (glioblastoma) and HeLa (cervical cancer) cell lines [39] showed that it was significantly more active than in the HTC116 (colon carcinoma) cells used in this study. On the other hand, 2h also showed severe activity in normal cell lines [39].
To determine the selectivity of the chosen stilbenes and dibenzo[b,f ]oxepine derivatives 1a, 1c, 1d, 1i, 2i, 2j, 3h toward cancer cells, we estimated the cell viability in a series of increasing concentrations of these compounds by using two cancer cell lines ( human colon carcinoma (HCT116) and human breast adenocarcinoma (MCF-7)) and two normal cell lines (human embryonic kidney (HEK293) and human dermal fibroblasts (HDF-A)) ( Figure 3). The tested compounds inhibited cell growth dose-dependently, and generally, a stronger effect was observed for HCT166 cells than for MCF-7 cells and for HEK293 cells than for HDF-A cells.
To determine the selectivity of the chosen stilbenes and dibenzo[b,f]oxepine derivatives 1a, 1c, 1d, 1i, 2i, 2j, 3h toward cancer cells, we estimated the cell viability in a series of increasing concentrations of these compounds by using two cancer cell lines ( human colon carcinoma (HCT116) and human breast adenocarcinoma (MCF-7)) and two normal cell lines (human embryonic kidney (HEK293) and human dermal fibroblasts (HDF-A)) ( Figure 3). The tested compounds inhibited cell growth dose-dependently, and generally, a stronger effect was observed for HCT166 cells than for MCF-7 cells and for HEK293 cells than for HDF-A cells. Next, based on the data presented above, we calculated the half-maximal inhibitory concentration (IC50) and selectivity index (SI). As observed here ( Figure 2) and in previous studies [39], in most cases, stilbenes exhibited stronger cytotoxicity than dibenzo[b,f]oxepines did ( Figure 3), but, with the exception of compound 1d, they showed no selectivity toward cancer cells (Table 1). The highest selectivity indexes were obtained for stilbene 1d and dibenzo[b,f]oxepine derivatives 2i, 2j, which varied between 1.4 and 2.4 (Table 1). Next, based on the data presented above, we calculated the half-maximal inhibitory concentration (IC 50 ) and selectivity index (SI). As observed here ( Figure 2) and in previous studies [39], in most cases, stilbenes exhibited stronger cytotoxicity than dibenzo[b,f ]oxepines did ( Figure 3), but, with the exception of compound 1d, they showed no selectivity toward cancer cells (Table 1). The highest selectivity indexes were obtained for stilbene 1d and dibenzo[b,f ]oxepine derivatives 2i, 2j, which varied between 1.4 and 2.4 (Table 1). ; "*"-The predicted IC 50 (above the maximum concentration tested-100 µM); "ND"-the IC 50 could not be calculated. The IC 50 value of stilbene 1d obtained for HTC116 cells (18 µM) was the lowest obtained in cancer cells for all tested compounds, while the selectivity index was among the highest reached for HTC116 cells with respect to both types of control cells. Therefore, we decided to visualize the microtubular network within the control and 1d-treated HCT116 cells. Until now, the arrangement of microtubules during treatment with dibenzo[b,f ]oxepine derivatives has not been analyzed, but several results exist for stilbenes [42][43][44]. Immunostaining of α-tubulin showed that the control cells contained a well-developed microtubular cytoskeleton with dense microtubules that were arranged along the long axis and on the cells' borders in elongated cells (the arrow and arrowhead, respectively, in Figure 4a).  ; "*"-The predicted IC50 (above the maximum concentration tested-100 µM); "ND"-the IC50 could not be calculated.

The Structure of Microtubular Cytoskeleton in HT116 Cells Treated with Compound (1d)
The IC50 value of stilbene 1d obtained for HTC116 cells (18 µM) was the lowest obtained in cancer cells for all tested compounds, while the selectivity index was among the highest reached for HTC116 cells with respect to both types of control cells. Therefore, we decided to visualize the microtubular network within the control and 1d-treated HCT116 cells. Until now, the arrangement of microtubules during treatment with dibenzo[b,f]oxepine derivatives has not been analyzed, but several results exist for stilbenes [42][43][44]. Immunostaining of α-tubulin showed that the control cells contained a well-developed microtubular cytoskeleton with dense microtubules that were arranged along the long axis and on the cells' borders in elongated cells (the arrow and arrowhead, respectively, in Figure 4a). In contrast, the HCT116 cells treated with 60 µm of compound 1d were rounded, and the density of the microtubular network was decreased (Figure 4b). Moreover, in many In contrast, the HCT116 cells treated with 60 µm of compound 1d were rounded, and the density of the microtubular network was decreased (Figure 4b). Moreover, in many cells, anti-α-tubulin stained an amorphous material, which was probably depolymerized tubulin (the arrow in Figure 4b). In contrast to the control cells, in the 1d-treated cells, the point of the organization of MTs, which was presumably a centrosome, was well defined, with several MTs (probably the most stable) arising from it (the arrowhead in Figure 4b). This observation confirmed that compound 1d affected the microtubular cytoskeleton, which, in turn, could lead to the cells' death.

Computational Analysis and Molecular Docking Simulations
The combination of experimental and computational methodologies has been of considerable importance in the discovery and further improvement of new candidates for potent anti-cancer compounds [45,46]. Molecular docking methods rely on the exploration of the ligand conformation within the active site of a protein or macromolecule until the minimum energy is achieved. Notably, this technique allows the computation of the binding free energy by including processes that are responsible for molecular recognition [47,48]. As mentioned earlier, in the Z derivatives of the stilbenes combretastatin CA1 and CA4 phosphate (CA1P and CA4P, respectively) and their disodium salts (OXi4503 and fosbretabulin, respectively), as well as dibenzo[b,f ]oxepines, which have a (Z)-stilbene in their skeleton, the motif docks at the colchicine-binding site [9,49]. Based on the available literature and our cytotoxicity results, we expected that selected stilbene and dibenzo[b,f ]oxepine derivatives could be potent tubulin inhibitors that bind to colchicine pockets [39,[50][51][52][53]. Therefore, the interactions between the most active compounds 1a, 1c, 1d, 1i, 2i, 2j, 3h and tubulin heterodimer (crystal structure from PDB: 1SA0) were analyzed with molecu-lar modeling. The optimal structures of the compounds with the best biological activity were calculated by using the DFT B3LYP/6-311G 6-311++g (2d,p) method (and with a polarizable continuum model (PCM), Gaussian 03W) [54,55] (Figures S1-S7, Tables S1-S7). Molecular docking was performed by simulating the incorporation of compounds 1a, 1c,  1d, 1i, 2i, 2j, 3h into the colchicine-binding site in tubulin ( Figure 5, Table S8). The docking protocol was validated by re-docking the DAMA-colchicine that was extracted from PDB crystal structure. In the crystal structure, the tropolone ring of colchicine is stabilized by van der Waals contacts with Val181, Ser178, and Val315. The carbonyl group is stabilized by a hydrogen bond with Val181. The ring with trimethoxy substituents is buried in the aliphatic part confined by Lys352, Asn350, Leu378, Ala316, Leu255, Lys254, Ala250, and Leu242. The methoxy group at position 3 participated in a hydrogen bond with the -SH group of Cys241 [56]. The algorithm that was applied correctly reproduced the binding mode of the native ligand. The binding poses of all tested compounds exhibited a high affinity for the target protein. The estimated binding free energy varied from −7.7 to −9.3 kcal/mol, in contrast to the −8.9 kcal/mol of redocked DAMA-colchicine (for a detailed list of the estimated binding free energies and the predicted interactions, see the supporting information). Analysis of the binding model of the most active compounds and tubulin allowed the discovery of several interactions with the protein residues in the colchicine-binding site (Table S8) . Compounds 1a, 1c, 1d, 1i, 2i, 2j, 3h were stabilized in their binding poses by hydrophobic interactions with β-tubulin residues (Leu242, Leu248, Ala250, Lys254, Leu255, Lys352, Ile378, Val318). Of the active compounds, four ligands (1d, 1i, 2i, 3h) were stabilized in their binding poses by hydrogen bond interactions with the tubulin heterodimer ( Figure 5 and Figure S3). However, no hydrogen bonds with Cys241 were predicted, though these are considered to be the most crucial for high potency. This may partially justify the high micromolar cytotoxicity of the tested compounds [57,58].

Conclusions
In this study, we report the synthesis and cytotoxic effects of nine stilbenoids 1a-i and nineteen dibenzo [b,f] oxepine derivatives 2a-i, 3a-h, 3j. The compounds were obtained by substituting a methoxy group in various positions of the 2-hydroxy-2′,4′-dinitrostilbene skeleton or nitro, hydroxy, acetoxy, or benzo groups of stilbene or dibenzo[b,f]oxepine. For the most promising compounds, we performed molecular docking, anti-cancer activity, measurements, and MT visualization through immunofluorescence. Our results showed that the stilbene derivatives were more active than the dibenzo[b,f]oxepine derivatives. The most active compounds were 1a, 1c, 1d, 1i, 2i, 2j, and  3h.
It is worth noting that for both stilbenes and the nitro or amine dibenzo[b,f]oxepines, the molecular skeletons were the same, and the compounds differed in their substituents

Conclusions
In this study, we report the synthesis and cytotoxic effects of nine stilbenoids 1a-i and nineteen dibenzo [b,f ] oxepine derivatives 2a-i, 3a-h, 3j. The compounds were obtained by substituting a methoxy group in various positions of the 2-hydroxy-2 ,4 -dinitrostilbene skeleton or nitro, hydroxy, acetoxy, or benzo groups of stilbene or dibenzo[b,f ]oxepine. For the most promising compounds, we performed molecular docking, anti-cancer activity, measurements, and MT visualization through immunofluorescence. Our results showed that the stilbene derivatives were more active than the dibenzo[b,f ]oxepine derivatives. The most active compounds were 1a, 1c, 1d, 1i, 2i, 2j, and 3h.
It is worth noting that for both stilbenes and the nitro or amine dibenzo[b,f ]oxepines, the molecular skeletons were the same, and the compounds differed in their substituents and the positions that they took. Some regularities could be observed: The introduction of electron-donating (EDG; enriched electron density in the aromatic ring) hydroxy, methoxy, or acetyloxy groups into the ring in the para or ortho position in 1c, 1i, 1a and in the meta position in 1d had a positive effect on the anticancer activity of both stilbene and nitro dibenzo[b,f ]oxepine 2i, 2j. The last set of molecules consisted of amine dibezo[b,f ]oxepines 3a-3h, 3j. In this case, only compound 3h showed an effect, and there were little data for investigating the effect of the electron density of the substituent on anticancer activity.
The selectivity toward cancer cells (HCT116 and MCF-7) was investigated in comparison with the selectivity toward normal cells (HEK293 and HDF-A) for stilbenes 1a, 1c, 1d, 1i and dibenzo[b,f ]oxepines 2i, 2j, 3h; this indicated that the three most promising anti-cancer agents were compounds 1d, 2i, and 2j. The molecular docking and visualization of the microtubular cytoskeleton in the stilbene and/or dibenzo[b,f ]oxepine derivatives confirmed that the cytotoxicity was related to their tubulin-targeting abilities and, as a consequence, depolymerization of cytoplasmic MTs. 1a-1i (Figure 6) To a stirred mixture of 2,4-dinitrotoluene (12.00 mmol), appropriate aldehyde (12.00 mmol), and toluene (20 mL) under argon, dry pyrrolidine (0.9 mL, 0.782 g, 11.00 mmol) was added. After 24 h of heating at 100 • C, the solvent was distilled off on a rotary evaporator. Then, ethyl acetate (150 mL) was added to the residue, and the resulting mixture was washed with 0.5 M hydrochloric acid (2 × 40 mL) and water to neutral pH. Next, the mixture was dried by the addition of anhydrous MgSO 4 and the solvent was removed in vacuo. The crude product was purified by crystallization from methanol or ethanol with a small amount of the activated charcoal (in order to absorb colored impurities). To a stirred mixture of 2,4-dinitrotoluene (12.00 mmol), appropriate aldehyde (12.00 mmol), and toluene (20 mL) under argon, dry pyrrolidine (0.9 mL, 0.782 g, 11.00 mmol) was added. After 24 h of heating at 100 °C, the solvent was distilled off on a rotary evaporator. Then, ethyl acetate (150 mL) was added to the residue, and the resulting mixture was washed with 0.5 M hydrochloric acid (2 × 40 mL) and water to neutral pH. Next, the mixture was dried by the addition of anhydrous MgSO4 and the solvent was removed in vacuo. The crude product was purified by crystallization from methanol or ethanol with a small amount of the activated charcoal (in order to absorb colored impurities). To the solution of appropriate stilbene (5.00 mol) in DMF (15 mL/mmol) NaN3 (9.50 mmol) was sequentially added and the reaction flask was fitted with a condenser. The mixture was stirred at 120 °C overnight and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel with DCM as the mobile phase. To the solution of appropriate stilbene (5.00 mol) in DMF (15 mL/mmol) NaN 3 (9.50 mmol) was sequentially added and the reaction flask was fitted with a condenser. The mixture was stirred at 120 • C overnight and concentrated in vacuo. The residue was purified by flash column chromatography on silica gel with DCM as the mobile phase. The compound 2i (1.00 mmol) was dissolved in excess of acetic anhydride. A few drops of concentrated sulfuric acid were added to the obtained solution and the mixture was stirred at room temperature overnight. Next, the reaction was quenched by dropwise addition of a concentrated aqueous solution of sodium bicarbonate (1 mL) and evaporated under reduced pressure. The residue was re-dissolved in ethyl acetate and transferred to a separatory funnel, washed two times with water and with brine. After solvent evaporation, the residue was purified by flash column chromatography on silica gel with a 10% mixture of ethyl acetate in hexane as a mobile phase. Compound 3a-3f and 3j (Figure 8) To the solution of corresponding nitrodibenzo[b,f]oxepine (1.00 mmol) in acetic acid (20.0 mL/mmol) activated zinc dust was added (10 mmol) and the resulting slurry was allowed to stir overnight at room temperature. Next, the mixture was filtrated through a pad of celite and a solvent was evaporated under reduced pressure. The residue was subjected to column chromatography on silica gel with hexane: ethyl acetate 1:1 mixture (v:v) as the eluent. The compound 2i (1.00 mmol) was dissolved in excess of acetic anhydride. A few drops of concentrated sulfuric acid were added to the obtained solution and the mixture was stirred at room temperature overnight. Next, the reaction was quenched by dropwise addition of a concentrated aqueous solution of sodium bicarbonate (1 mL) and evaporated under reduced pressure. The residue was re-dissolved in ethyl acetate and transferred to a separatory funnel, washed two times with water and with brine. After solvent evaporation, the residue was purified by flash column chromatography on silica gel with a 10% mixture of ethyl acetate in hexane as a mobile phase. Compound 3a-3f and 3j (Figure 8) To the solution of corresponding nitrodibenzo[b,f ]oxepine (1.00 mmol) in acetic acid (20.0 mL/mmol) activated zinc dust was added (10 mmol) and the resulting slurry was allowed to stir overnight at room temperature. Next, the mixture was filtrated through a pad of celite and a solvent was evaporated under reduced pressure. The residue was subjected to column chromatography on silica gel with hexane: ethyl acetate 1:1 mixture (v:v) as the eluent. Dicobalt octacarbonyl (1.00 g, 3.00 mmol) was placed in a 50 mL three-necked flask equipped with a reflux condenser. The vessel was purged with argon, and (1.76 mmol) of 2.7-dinitrodibenzo[b,f ]oxepine 2g and 20 mL of dimethoxyethane were added and the solution was stirred under an argon atmosphere. After the precipitate had dissolved, 2 mL of water was added and it was heated at 100 • C for 5 h. Next, the mixture was cooled down to room temperature and concentrated on a rotary evaporator. The resulting precipitate was dissolved in chloroform and then purified by column chromatography using chloroform as the eluent. mixture of ethyl acetate in hexane as a mobile phase. Compound 3a-3f and 3j (Figure 8) To the solution of corresponding nitrodibenzo[b,f]oxepine (1.00 mmol) in acetic acid (20.0 mL/mmol) activated zinc dust was added (10 mmol) and the resulting slurry was allowed to stir overnight at room temperature. Next, the mixture was filtrated through a pad of celite and a solvent was evaporated under reduced pressure. The residue was subjected to column chromatography on silica gel with hexane: ethyl acetate 1:1 mixture (v:v) as the eluent. and the maps were center located (39.82, 53.24, −8.21) in the binding site of the protein. A grid spacing of 0.375 Å (approximately one-fourth of the length of a carbon-carbon covalent bond) was used for the calculation of the energetic map. All computations were performed on an Intel ® Core TM and 7-4702MQ 3.2 GHz processor running Ubuntu 18.04 Workstation Linux distribution. PyMOL software (www.pymol.org/ (accessed on 1 January 2021)) was used to analyze the docking results [62]. The Protein-Ligand Interaction Profiler (PLIP) was used to predict protein-docked ligand interactions (the default threshold was used for detection steps: 4.0 Å max. distance of carbon atoms for a hydrophobic interaction; 4.1 Å max. distance between acceptor and donor in hydrogen bonds) [63].