Synthesis, Computational Analysis, and Antiproliferative Activity of Novel Benzimidazole Acrylonitriles as Tubulin Polymerization Inhibitors: Part 2

We used classical linear and microwave-assisted synthesis methods to prepare novel N-substituted, benzimidazole-derived acrylonitriles with antiproliferative activity against several cancer cells in vitro. The most potent systems showed pronounced activity against all tested hematological cancer cell lines, with favorable selectivity towards normal cells. The selection of lead compounds was also tested in vitro for tubulin polymerization inhibition as a possible mechanism of biological action. A combination of docking and molecular dynamics simulations confirmed the suitability of the employed organic skeleton for the design of antitumor drugs and demonstrated that their biological activity relies on binding to the colchicine binding site in tubulin. In addition, it also underlined that higher tubulin affinities are linked with (i) bulkier alkyl and aryl moieties on the benzimidazole nitrogen and (ii) electron-donating substituents on the phenyl group that allow deeper entrance into the hydrophobic pocket within the tubulin’s β-subunit, consisting of Leu255, Leu248, Met259, Ala354, and Ile378 residues.


Introduction
Microtubules, being key dynamic structural components in cells, have attracted considerable attention from medicinal chemists as targets for anticancer drug discovery [1][2][3]. These protein biopolymers, formed through the polymerization of heterodimers of αand β-tubulins, play an important role in cellular shape organization, cell division, mitosis, and intracellular movement. Potent microtubule-targeting drugs such as paclitaxel, vinblastine, colchicine, and vincristine are structurally complex natural products that are widely used in anticancer therapy [4]. These products alter the dynamics of tubulin, such as the polymerization and depolymerization [5], by binding to specific sites on the tubulin heterodimers [6], of which the most important are those for paclitaxel, vinblastine, and colchicine; thus, within the binding to the tubulin heterodimers, inhibitors could suppress tubulin dynamic instability and interfere with microtubule functions, including the mitotic spindle formation.
Inhibitors that bind to the vinblastine and colchicine binding sites are known as inhibitors of tubulin polymerization, while inhibitors interacting with the paclitaxel binding site are known as microtubule-stabilizing compounds [7].
During the past decades, there has been an increase in the development of novel tubulin polymerization inhibitors, while versatile classes of organic derivatives, both natural and synthetic, have been extensively studied [8][9][10][11][12]. Nowadays, considerable interest is focused on the development, design, and biological activity of novel heteroaromatic systems, whereby nitrogen heterocycles have become the essential structural motifs in medicinal and pharmaceutical chemistry, being very important in drug discovery. Benzimidazoles can easily interact with essential biomolecules of living systems due to the structural similarity of their nuclei with naturally occurring purines. Benzimidazoles are, therefore, very prominent scaffolds for the development of novel derivatives with a wide range of diverse biological activities. Considering all of the biological activities displayed by benzimidazoles, one of the most important is their antitumor activity, which is exerted by acting on numerous biological targets (Figure 1). For example, they can interact with DNA and RNA as intercalators or minor groove binders [13]; they can inhibit topoisomerases I and II [14]; and they can act as antiangiogenic agents [15], androgen receptor antagonists [16], and inhibitors of kinases [17,18]. Some benzimidazoles were also recognized as tubulin polymerization inhibitors, mostly binding to the colchicine binding site [19][20][21][22]. Nocodazole ( Figure 1a) is a well-known anticancer agent, which significantly inhibits tubulin polymerization at low nanomolar concentrations [23]. Additionally, benzimidazole-5-carboxylate derivatives causes mitotic catastrophes by specifically targeting the microtubule system [24]. Several studies have reported on the biological activity of 2-aryl-1,2,4-oxadiazolo-benzimidazole derivatives ( Figure 1b) with different mechanisms of biological action, such as binding to the colchicine binding site [25]. A series of benzimidazole-2-urea derivatives (Figure 1c) has been described as novel β-tubulin inhibitors that might bind in a new binding site different from the three well-known ones [26]. Novel 2-aryl-benzimidazole derivatives of dehydroabietic acid have been reported as tubulin polymerization inhibitors, which significantly disrupt the intracellular microtubule network by binding to the colchicine site of tubulin [26]. inhibitors of tubulin polymerization, while inhibitors interacting with the paclitaxel binding site are known as microtubule-stabilizing compounds [7]. During the past decades, there has been an increase in the development of novel tubulin polymerization inhibitors, while versatile classes of organic derivatives, both natural and synthetic, have been extensively studied [8][9][10][11][12]. Nowadays, considerable interest is focused on the development, design, and biological activity of novel heteroaromatic systems, whereby nitrogen heterocycles have become the essential structural motifs in medicinal and pharmaceutical chemistry, being very important in drug discovery. Benzimidazoles can easily interact with essential biomolecules of living systems due to the structural similarity of their nuclei with naturally occurring purines. Benzimidazoles are, therefore, very prominent scaffolds for the development of novel derivatives with a wide range of diverse biological activities. Considering all of the biological activities displayed by benzimidazoles, one of the most important is their antitumor activity, which is exerted by acting on numerous biological targets ( Figure 1). For example, they can interact with DNA and RNA as intercalators or minor groove binders [13]; they can inhibit topoisomerases I and II [14]; and they can act as antiangiogenic agents [15], androgen receptor antagonists [16], and inhibitors of kinases [17,18]. Some benzimidazoles were also recognized as tubulin polymerization inhibitors, mostly binding to the colchicine binding site [19][20][21][22]. Nocodazole ( Figure 1a) is a well-known anticancer agent, which significantly inhibits tubulin polymerization at low nanomolar concentrations [23]. Additionally, benzimidazole-5-carboxylate derivatives causes mitotic catastrophes by specifically targeting the microtubule system [24]. Several studies have reported on the biological activity of 2-aryl-1,2,4-oxadiazolo-benzimidazole derivatives ( Figure 1b) with different mechanisms of biological action, such as binding to the colchicine binding site [25]. A series of benzimidazole-2-urea derivatives ( Figure  1c) has been described as novel β-tubulin inhibitors that might bind in a new binding site different from the three well-known ones [26]. Novel 2-aryl-benzimidazole derivatives of dehydroabietic acid have been reported as tubulin polymerization inhibitors, which significantly disrupt the intracellular microtubule network by binding to the colchicine site of tubulin [26]. Recently, as a continuation of our previous efforts aimed at the design and discovery of novel benzimidazoles with promising antitumor activities, we prepared novel N-substituted, benzimidazole-derived acrylonitriles as potential tubulin polymerization inhibitors. N,N-dimethylamino-substituted acrylonitriles I and II ( Figure 2) with submicromolar inhibitory concentrations (IC50 0.2-0.6 μM) were chosen as lead compounds, while their interference with the tubulin activity was confirmed by in vitro studies of the tubulin polymerization inhibition and the computational analysis [27].  Encouraged by our findings and the fact that some of the tested compounds showed strong and selective antiproliferative activity, we further optimized the presented structure by designing and synthesizing novel N-substituted benzimidazole acrylonitriles. Here, we present the synthesis, biological activity, and tubulin polymerization inhibition of the most active compounds and demonstrate their binding within the colchicine site of tubulin via computational docking analysis and molecular dynamics simulations.

Chemistry
The synthesis of novel N-substituted, benzimidazole-derived acrylonitriles 32-71 is illustrated in Scheme 1, starting from the ortho-chloronitrobenzenes 1-2.  Encouraged by our findings and the fact that some of the tested compounds showed strong and selective antiproliferative activity, we further optimized the presented structure by designing and synthesizing novel N-substituted benzimidazole acrylonitriles. Here, we present the synthesis, biological activity, and tubulin polymerization inhibition of the most active compounds and demonstrate their binding within the colchicine site of tubulin via computational docking analysis and molecular dynamics simulations.

Chemistry
The synthesis of novel N-substituted, benzimidazole-derived acrylonitriles 32-71 is illustrated in Scheme 1, starting from the ortho-chloronitrobenzenes 1-2.  Encouraged by our findings and the fact that some of the tested compounds showed strong and selective antiproliferative activity, we further optimized the presented structure by designing and synthesizing novel N-substituted benzimidazole acrylonitriles. Here, we present the synthesis, biological activity, and tubulin polymerization inhibition of the most active compounds and demonstrate their binding within the colchicine site of tubulin via computational docking analysis and molecular dynamics simulations.

Chemistry
The synthesis of novel N-substituted, benzimidazole-derived acrylonitriles 32-71 is illustrated in Scheme 1, starting from the ortho-chloronitrobenzenes 1-2. By using uncatalyzed microwave-assisted amination in acetonitrile with an excess of desired amine, N-substituted precursors 3-8 bearing i-butyl, methyl, phenyl, and nhexyl substituents were obtained in good reaction yields. Within the reduction of nitrosubstituted compounds 3-8 with SnCl 2 ·2H 2 O in MeOH followed by cyclocondensation with 2-cyanoacetamide at high temperatures, corresponding N-substituted 2-(cyanomethyl)benzimidazoles 17-25 as main precursors were obtained in moderate yields [27]. Targeted acrylonitrile derivatives 32-71 were synthesized in the condensation with the chosen unsubstituted and methoxy-, N,N-dimethyl-, and N,N-diethyl-substituted aromatic aldehydes 26-31 in absolute ethanol by using a few drops of piperidine as a weak base. All acrylonitriles were obtained in moderate to high reaction yields, while some of them were obtained as mixtures of Eand Z-isomers (38, 41, 45, 61, 64, 67, and 68), which could not be efficiently separated by column chromatography.
The structural determination of newly prepared acrylonitriles was performed using 1 H and 13 C NMR spectroscopies and elemental analysis. The structural characterization was performed based on the chemical shifts in both spectra and H-H coupling constant values in the 1 H spectra. The successful nucleophilic substitution was confirmed within the signals in the aliphatic part of both 1 H and 13 C NMR spectra of compounds 3-8. The structure of amino-substituted derivatives 9-16 was confirmed via the observation of signals related to the protons of amino groups placed in the range of 5.25-4.44 ppm in the 1 H NMR spectra. A singlet of acrylonitrile protons in the range of 8.54-7.20 ppm confirmed the formation of acrylonitrile derivatives.

Antiproliferative Activity In Vitro
All newly prepared compounds were tested for their antiproliferative activity in vitro. The results are displayed in Table 1 as IC 50 values (50% inhibitory concentrations) and are compared to known antiproliferative agents combretastatin A4 (CA4) and docetaxel.
Thirteen derivatives did not show any activity against the tested cell lines. Some of the tested derivatives exhibited strong and selective antiproliferative activity but were less active in comparison to the used standard drugs. Among the most active compounds, the N,N-diethylamino-substituted derivative with the i-butyl substituent placed at the nitrogen atom of benzimidazole core 64 did not show any significant selectivity towards the tested cancer cell lines and was the most potent system elucidated here. Compound 50 substituted with cyano and phenyl rings at the benzimidazole core bearing two methoxy groups, 68 substituted with cyano and i-butyl substituents at the benzimidazole core bearing a N,N-diethylamino group, and 69 substituted with cyano and methyl substituents at the benzimidazole core bearing a N,N-diethylamino group showed selectivity against the hematological cancer cell lines (acute lymphoblastic leukemia (DND-41), acute myeloid leukemia (HL-60), chronic myeloid leukemia (K-562), and non-Hodgkin lymphoma (Z-138). Among other derivatives with moderate activities, compound 48 substituted with cyano and i-butyl at the benzimidazole core bearing two methoxy groups showed some selectivity against lung carcinoma (NCI-H460) and colorectal carcinoma (HCT-116).
In general, comparing the unsubstituted and cyano-substituted derivatives bearing the same substituents at the nitrogen atom of the benzimidazole core and at the phenyl ring, it was observed that some cyano-substituted derivatives showed slightly improved antiproliferative activity, while for some others the presence of the -CN moiety reduced the activity.   The latter also holds for the most potent compound 64, whose cyano derivative 68 is generally less active. Among the tested systems, i-butyl-substituted derivatives showed improved activity relative to methyl, phenyl or n-hexyl-substituted analogues. According to the obtained results, it is obvious that the most significant impact on the antiproliferative activity enhancement relates to the introduction of the N,N-diethylamino group at the para-position of the phenyl ring. In comparison to the previously published results [27], it can be concluded that the introduction of the N,N-diethylamino group instead of the N,Ndimethylamino group decreased the antiproliferative activity against Capan-1, HCT-116, NCI-H460, DND-41, and HL-60 cancer cells from a submicromolar to micromolar range of inhibitory concentrations.
In order to establish whether the observed antiproliferative activity is selective towards cancer cells, normal peripheral blood mononuclear cells (PBMCs) from two healthy donors were treated with compounds 50, 64, 68, and 69 ( Figure 3). The compounds with the best antiproliferative activity in hematological cancer cells (compound 50, 68, and 69) also induced apoptosis in normal cells, although this was limited to the highest concentration tested (20 µM) for compounds 68 and 69, resulting in a favorable selectivity window. Derivative 64, which was inhibitory against all tested cancer cell lines, had no effect on the viability of normal PBMC and is, thus, a selective anticancer compound.
Pharmaceuticals 2021, 14, x FOR PEER REVIEW 6 of 26 cancer cell lines, had no effect on the viability of normal PBMC and is, thus, a selective anticancer compound.

In Vitro Inhibition of the Tubulin Polymerization
The ability of derivatives 50, 64, 68, and 69 to inhibit the polymerization of tubulin was confirmed in vitro in a purified protein system. All tested compounds showed effective activity in a dose-dependent manner ( Figure 4).

In Vitro Inhibition of the Tubulin Polymerization
The ability of derivatives 50, 64, 68, and 69 to inhibit the polymerization of tubulin was confirmed in vitro in a purified protein system. All tested compounds showed effective activity in a dose-dependent manner (Figure 4).
At the highest concentrations tested (30 and 10 µM), these four derivatives were all able to inhibit tubulin polymerization. Even when lowering the dose to 3 µM, compounds 50 (to a lesser extent) and 68 still showed some inhibitory activity in the in vitro assay. and PI followed by live cell monitoring. The percentages of live (white), dead (red), and apoptotic (blue) cells after 72 h are shown (means ± standard error bars).

In Vitro Inhibition of the Tubulin Polymerization
The ability of derivatives 50, 64, 68, and 69 to inhibit the polymerization of tubulin was confirmed in vitro in a purified protein system. All tested compounds showed effective activity in a dose-dependent manner ( Figure 4). Docetaxel and combretastatin A4 (CA4) (1 μM) were used as controls for tubulin-stabilizing and tubulin-destabilizing agents, respectively, while DMSO was used as a negative control. The effects on tubulin assembly were monitored in a Tecan Spark multimode plate reader at one minute intervals for one hour at 37 °C. Each condition was tested in duplicate. The level of polymerization was measured by an increase in fluorescence emission intensity at λem = 435 nm. Purified porcine neuronal tubulin and GTP were mixed in a 96-well plate. Docetaxel and combretastatin A4 (CA4) (1 µM) were used as controls for tubulin-stabilizing and tubulin-destabilizing agents, respectively, while DMSO was used as a negative control. The effects on tubulin assembly were monitored in a Tecan Spark multimode plate reader at one minute intervals for one hour at 37 • C. Each condition was tested in duplicate. The level of polymerization was measured by an increase in fluorescence emission intensity at λ em = 435 nm.

Computational Analysis
Computational analysis was performed to interpret the observed biological properties and to identify structural and electronic features responsible for the highest activity of 64, which should aid in the subsequent design of even more effective ligands based on the utilized organic skeleton. To do so, we considered a representative set of compounds, including 50, 63, 64, 66, 68, and 69, together with two model systems m1 and m2, as well as colchicine, which was taken as a typical ligand for the colchicine binding site in tubulin ( Figure 5), in line with our previous results [27]. At the highest concentrations tested (30 and 10 μM), these four derivatives were all able to inhibit tubulin polymerization. Even when lowering the dose to 3 μM, compounds 50 (to a lesser extent) and 68 still showed some inhibitory activity in the in vitro assay.

Computational Analysis
Computational analysis was performed to interpret the observed biological properties and to identify structural and electronic features responsible for the highest activity of 64, which should aid in the subsequent design of even more effective ligands based on the utilized organic skeleton. To do so, we considered a representative set of compounds, including 50, 63, 64, 66, 68, and 69, together with two model systems m1 and m2, as well as colchicine, which was taken as a typical ligand for the colchicine binding site in tubulin ( Figure 5), in line with our previous results [27]. The systems investigated here consisted of variously substituted benzimidazole and phenyl fragments, linked together with an acrylonitrile unit, which can formally exist as either E-or Z-isomers. Our DFT analysis showed that all systems were between 0.5 and 4.0 kcal mol -1 more stable as Z-isomers (Table 2 and Figure S120), where the cyano group and the benzimidazole moiety reside on the same side of the C=C double bond. Although these values neglected any kinetic aspects of the isomer formation, the obtained thermodynamic data were found to be in good quantitative agreement with the experiments, which generally predict the predominance of Z-isomers. The differences in the stability among isomers were higher for the systems with bulkier N-substituents due to larger unfavorable steric interactions with the attached phenyl group occurring in their E-analogues. Interestingly, the only exception was provided by the most active system, 64, which was 1.2 kcal mol -1 more stable as an E-analogue. This confirms its experimentally observed preference for the E-isomer at a 2:1 ratio, yet such the small The systems investigated here consisted of variously substituted benzimidazole and phenyl fragments, linked together with an acrylonitrile unit, which can formally exist as either Eor Z-isomers. Our DFT analysis showed that all systems were between 0.5 and 4.0 kcal mol −1 more stable as Z-isomers (Table 2 and Figure S120), where the cyano group and the benzimidazole moiety reside on the same side of the C=C double bond. Although these values neglected any kinetic aspects of the isomer formation, the obtained thermodynamic data were found to be in good quantitative agreement with the experiments, which generally predict the predominance of Z-isomers. The differences in the stability among isomers were higher for the systems with bulkier N-substituents due to larger unfavorable steric interactions with the attached phenyl group occurring in their E-analogues. Interestingly, the only exception was provided by the most active system, 64, which was 1.2 kcal mol −1 more stable as an E-analogue. This confirms its experimentally observed preference for the E-isomer at a 2:1 ratio, yet such the small energy difference observed here, and in other evaluated systems, allows for the presence of the other isomer as well. With this in mind, we will focus our discussion on the most stable Z-isomers of all systems, while both isomers will be considered for the most active compound 64. Table 2. Relative stability levels of the isomers, obtained using the DFT (SMD)/M06-2X/6-31+G(d) model, and the binding free energies (∆G bind,CALC ), obtained through docking simulations, for the studied ligands and colchicine (all values in kcal mol −1 ). a The experimental value for colchicine is taken from [28]. The docking procedure produced the binding free energies shown in Table 2. The highest affinity was obtained for colchicine, ∆G bind = −9.3 kcal mol −1 , closely matching the value of ∆G bind = −9.0 kcal mol −1 reported by Silva-García and co-workers obtained using the AutoDock docking software [28]. Moreover, both of these values were in excellent agreement with the experimental value of ∆G bind,EXP = −8.3 kcal mol −1 calculated from the colchicine binding constant K bind,EXP = 6.3 × 10 5 L mol −1 measured by Wilson and Meza [29]. In addition, the predicted colchicine binding position very closely matched that in the crystal structure, suggesting that the docking procedure correctly positioned it within the colchicine binding site ( Figure S121). Such an agreement in terms of both the binding energy and the position of the ligand leads us to conclude that these results lend firm credence to the employed computational methodology and support the reliability of the other results as well.

System
In certain cases, the values correspond to allosteric positions on tubulin ( Figure S122), which is why we also analyzed the most favorable orthosteric poses within the colchicine binding site, since the latter are responsible for a potential tubulin polymerization inhibition. The lowest binding affinity was displayed by the least substituted m1, ∆G bind = −8.0 kcal mol −1 , corresponding to the allosteric binding, which suggests a lack of antitumor activity. This already suggests that the substitution of the used organic framework is likely crucial for the binding and that specific protein-ligand interactions govern the activity. Since m1 is a reference system, let us note that a binding pose within the colchicine binding site comes with a further lower affinity, ∆G bind = −7.6 kcal mol −1 , being the least exergonic site here. In that case ( Figure S123), m1 is oriented so that its phenyl unit is immersed into the β-subunit close to Cys241, but without any significant interaction with it. In contrast, the importance of the benzimidazole unit is seen in favorable N-H···π interactions with Lys352, yet this is outperformed by the unfavorable steric contacts between N-i-butyl, Lys254, and Asn258, which decrease the binding and disfavor such orthosteric positions for m1.
The addition of electron donors on the phenyl unit, such as the p-OMe group in m2, improves the binding. This allows deeper penetration of the β-subunit ( Figure S123) facilitated by the S-H···O(Me) hydrogen bonding with Cys241, which is absent in m1. This maintains favorable contacts with Lys352, while offering reduced steric interactions with Lys254 and Asn258, with the latter being allowed to engage in N-H···N hydrogen bonds with the benzimidazole moiety. All of this contributes around 0.4 kcal mol −1 to the binding within the colchicine binding site, yet still promotes the allosteric binding as being the most favorable at ∆G bind = −8.4 kcal mol −1 . Replacing the p-OMe group with p-NEt 2 immediately offers the most potent system 64. With its better hydrogen-bond-accepting properties, 64 forms a stronger S-H···N(Et 2 ) interaction with Cys241 ( Figure S123), which is evident in the reduced S···N distance of 0.4 Å from that observed for the matching S···O distance in m2. This allows 64 to rotate and avoid the unfavorable steric contacts with Lys254 and Asn258, enabling both to bind the benzimidazole fragment-the former through the N-H···π interactions, while the latter through the N-H···N hydrogen bonds. The attached cyano group in 64 accepts hydrogen bonding from Lys352, further promoting the binding.
All of this positions 64 within the colchicine binding site as the most favorable binding location, linked with the most exergonic binding energy of ∆G bind = −8.7 kcal mol −1 . This confirms its high activity and promotes the tubulin polymerization inhibition as its likely biological mechanism of action.
The presence of the bulky N-i-butyl group is also significant in this activity. In general, this allows the investigated ligands to better position themselves within the hydrophobic interior of the β-subunit. If this is replaced by a smaller N-Me group as in 63, the system is reverted back to the allosteric binding as being most favorable, confirming its reduced activity, while its potential binding within the colchicine binding is also reduced to ∆G bind = −8.0 kcal mol −1 ( Figure S124). Along these lines, the introduction of the aromatic N-phenyl unit in 66 improves the binding within the colchicine binding site to ∆G bind = −8.6 kcal mol −1 , mostly through favorable N-H···π interactions with this substituent, while positive contributions from Lys254 remain limited, yet this binding pose is also dominated by the allosteric binding that is 0.2 kcal mol −1 higher, making 66 a non-active compound.
Lastly, the presence of the electron-withdrawing cyano group on the benzimidazole core generally leads to reduced activities in the investigated cases. As illustrative examples, both 68 and 69 are associated with lower affinities than 64, regardless of having either N-i-butyl or N-methyl groups attached to the benzimidazole unit. In 68, this results in promoting the allosteric binding and allowing for only a moderate orthosteric binding at ∆G bind = −8.1 kcal mol −1 , while in 69 the effect is smaller, although seen in reduced orthosteric binding at ∆G bind = −8.3 kcal mol −1 due to a notable departure from the βsubunit interior ( Figure S124). In both cases, the reduced affinity likely comes as a result of a depleted electron density within the benzimidazole unit, which makes it less susceptible for the N-H···π interactions with either Lys254 or Lys352 residues and favors ligand departure from the colchicine binding site.
The results presented so far confirm 64 as the most potent ligand and reveal its position within the colchicine binding site as the most favorable binding location. Knowing that it was isolated as a mixture of both isomers, we decided to further support its prevalence for the E-isomer and its likely biological activity through a series of MD simulations considering both isomers. It turned out that when a less stable Z-isomer is placed within the colchicine binding site, it leaves this location after 130 ns of the simulation time, only to remain allosterically bound for the rest of the simulations ( Figure S125). The MM-GBSA analysis reveals that during the first part of the simulation, while orthosterically bound, its binding free energy is 2.1 kcal mol −1 lower than during the second part, when it is positioned outside the colchicine binding site, thereby providing the driving force for the departure ( Figure S126). On the other hand, its E-isomer remains within the colchicine binding site throughout the MD simulations ( Figure S127), while the decomposition of the obtained binding energy into contributions from individual residues demonstrates interesting trends ( Figure 6). This confirms the hydrophobic nature of the β-subunit interior and the orthosteric binding site, as Leu255 and Leu248 dominate the binding, being solely responsible for over 40% of the binding energy. This is followed by the mentioned Lys254 and Lys352 residues, which establish hydrogen bonds mainly with the cyano group and the unsaturated benzimidazole nitrogen atom, respectively, with the former occasionally being supported through Asn258 as well ( Figure S128). electron-donating substituents on the phenyl group that allow deeper entrance into the hydrophobic pocket within the β-subunit predominantly consisting of Leu255, Leu248, Met259, Ala354, and Ile378 residues.

General Methods
All chemicals and solvents used for the synthesis were obtained from the commercial suppliers Aldrich and Acros. Melting points were determined on an SMP11 Bibby and Büchi 535 apparatus and were uncorrected. NMR spectra were taken in DMSO-d6 solutions with TMS as an internal standard. The 1 H and 13 C NMR spectra were recorded on a Varian Bruker Advance III HD 400 MHz/54 mm Ascend instrument. Chemical shifts are given in ppm (δ) relative to TMS. All prepared compounds were checked by TLC with Merck silica gel 60F-254 glass plates. Microwave-assisted synthesis was performed in a Milestone start S microwave oven using quartz cuvettes under a pressure of 40 bar. Elemental analyses for C, H, and N were performed on a Perkin-Elmer 2400 elemental analyzer. Where analyses are indicated only as symbols of elements, the analytical results obtained were within 0.4% of the theoretical value. NMR spectra of In concluding this section, we can emphasize that docking simulations confirmed 64 as the most potent ligand studied here, while MD simulations support E-isomer as its biologically active form. The investigated ligands compete between orthosteric binding into the colchicine binding site responsible for the observed antitumor activities and other allosteric positions, where the latter prevails in several cases, leading to compounds that are inactive against tubulin polymerization; however, the obtained insights regarding the most potent systems suggest that higher tubulin affinities are associated with (i) bulkier alkyl and aryl moieties on the benzimidazole nitrogen and (ii) electron-donating substituents on the phenyl group that allow deeper entrance into the hydrophobic pocket within the β-subunit predominantly consisting of Leu255, Leu248, Met259, Ala354, and Ile378 residues.

General Methods
All chemicals and solvents used for the synthesis were obtained from the commercial suppliers Aldrich and Acros. Melting points were determined on an SMP11 Bibby and Büchi 535 apparatus and were uncorrected. NMR spectra were taken in DMSO-d 6 solutions with TMS as an internal standard. The 1 H and 13 C NMR spectra were recorded on a Varian Bruker Advance III HD 400 MHz/54 mm Ascend instrument. Chemical shifts are given in ppm (δ) relative to TMS. All prepared compounds were checked by TLC with Merck silica gel 60F-254 glass plates. Microwave-assisted synthesis was performed in a Milestone start S microwave oven using quartz cuvettes under a pressure of 40 bar. Elemental analyses for C, H, and N were performed on a Perkin-Elmer 2400 elemental analyzer. Where analyses are indicated only as symbols of elements, the analytical results obtained were within 0.4% of the theoretical value. NMR spectra of newly prepared compounds are presented in Supporting Information (Figures S1-S119).

General Method for Preparation of Compounds 3-8
Compounds 3-8 were prepared using microwave irradiation at optimized reaction times at 170 • C, with a power level of 800 W and 40 bar pressure from 1 or 2 in acetonitrile (10 mL), with an excess of the corresponding amine. After cooling, the resulting product was purified by column chromatography on SiO 2 using dichloromethane/methanol at 200:1 as the eluent. The synthesis of the previously published derivatives 3-6 is outlined in the Supporting Materials.

General Method for Preparation of Compounds 9-11
Derivatives 3, 7, and 8 and a solution of SnCl 2 ·2H 2 O in MeOH and concentrated HCl were refluxed for 0.5 h. After cooling, the reaction mixture was evaporated under vacuum conditions and dissolved in water (20 mL). The resulting solution was treated with 20% NaOH to pH = 14. The resulting precipitate was filtered off, washed with hot ethanol, then filtered again. The filtrate was evaporated at a reduced pressure and extracted with ethyl acetate. The organic layer was dried over anhydrous MgSO4 and concentrated at reduced pressure. The synthesis of the previously published derivative 9 is outlined in Supporting Materials.

General Method for Preparation of Compounds 14-16
Benzonitrile derivatives 4-6 and a solution of SnCl 2 ·2H 2 O in MeOH and concentrated HCl were refluxed for 0.5 h. After cooling, the reaction mixture was evaporated under vacuum conditions and dissolved in water (20 mL). The resulting solution was treated with 20% NaOH to pH = 14. The resulting precipitate was filtered off, washed with hot ethanol, then filtered again. The filtrate was evaporated at a reduced pressure, a small amount of water was added, then the product was filtered again. The synthesis of the previously published derivatives 14-16 is outlined in the Supporting Materials.

General Method for Preparation of Compounds 18-21
A mixture of the corresponding substituted 1,2-phenylenediamines 9, 10, 12, 13, and 2cyanoacetamide was heated in an oil bath for 35-50 min at 185 • C. After cooling, the resulting product was purified by column chromatography on SiO 2 using dichloromethane/methanol at 200:1 as the eluent. The synthesis of the previously published derivatives 18-20 is outlined in the Supporting Materials.