Combretastatin-Inspired Heterocycles as Antitubulin Anticancer Agents.

Combretastatin (CA-4) and its analogues are undergoing several clinical trials for treating different types of tumors. In this work, the antiproliferative activity of a series of 2-aminoimidazole-carbonyl analogs of clinically relevant combretastatins A-4 (CA-4) and A-1 was evaluated using a cell-based assay. Among the compounds tested, C-13 and C-21 displayed strong antiproliferative activities against HeLa cells. C-13 inhibited the proliferation of lung carcinoma (A549) cells more potently than combretastatin A-4. C-13 also retarded the migration of A549 cells. Interestingly, C-13 displayed much stronger antiproliferative effects against breast carcinoma and skin melanoma cells compared to noncancerous breast epithelial and skin fibroblast cells. C-13 strongly disassembled cellular microtubules, perturbed the localization of EB1 protein, inhibited mitosis in cultured cells, and bound to tubulin at the colchicine site and inhibited the polymerization of reconstituted microtubules in vitro. C-13 treatment increased the level of reactive oxygen species and induced apoptosis via poly(ADP-ribose) polymerase-cleavage in HeLa cells. The results revealed the importance of the 2-aminoimidazole-carbonyl motif as a double bond replacement in combretastatin and indicated a pharmacodynamically interesting pattern of H-bond acceptors/donors and requisite syn-templated aryls.

Introduction

Several natural products and their derivatives such as paclitaxel, docetaxel, vinblastine, vincristine, vinorelbine, vindesine, and ixabepilone are highly successful as microtubule-targeting anticancer agents.[1−7] These compounds act by interfering with the microtubule dynamics upon binding to tubulin.[2−4] In addition, several natural products such as combretastatins, epothilones, dolastatins, and 2-methoxyestradiol are undergoing clinical trials for cancer chemotherapy.[2,3,8] Among these natural products, combretastatin A-4 (CA-4), isolated from the Cape Bush willow tree, Combretum caffrum(9) and its several derivatives have shown strong anticancer potential in clinical trials.[2,3,8,10] In addition to its strong antiproliferative activity, CA-4 is also known to display anti-angiogenic[11,12] and anti-vascular activity.[13−16] Currently, combretastatin A4 phosphate (CA-4P), a phosphate prodrug of CA-4,[17−19] has completed phase II clinical trials for the therapy of advanced anaplastic thyroid cancer (ClinicalTrials.gov identifier: NCT00060242), pathologic myopia (ClinicalTrials.gov identifier: NCT01423149), and polypoidal choroidal vasculopathy (ClinicalTrials.gov identifier: NCT01023295). In addition, CA-4P is undergoing phase III clinical trials for anaplastic thyroid cancer in combination with carboplatin/paclitaxel (ClinicalTrials.gov identifier: NCT00507429). Oxi4503, a synthetic phosphorylated prodrug of combretastatin A-1,[20−22] has completed phase I clinical trials for hepatic tumor (ClinicalTrials.gov identifier: NCT00960557) and advanced solid tumors (ClinicalTrials.gov identifier: NCT00977210) and is undergoing phase I/II trials on acute myelogenous leukemia and myelodysplastic syndromes in combination with cytarabine (ClinicalTrials.gov identifier: NCT02576301). A benzophenone analogue of CA-4, CKD-516,[23−25] and a serine amino acid derivative of CA-4, AVE8062[26−28] have completed phase I clinical trials on solid tumors (ClinicalTrials.gov identifier: NCT01560325 and NCT00968916). Analogues of combretastatin thus may have a great potential in cancer chemotherapy.

Although CA-4 is a promising clinical candidate, a number of issues are associated with it. It has a short biological half-life[29] and undergoes cis–trans isomerization in heat, light, and protic media, forming the inactive trans form from the active cis form.[30,31] To inhibit metabolic degradation and to retain the cis configuration of the combretastatins, replacement of the olefinic motif by heterocyclic[32−40] or carbocyclic[41−43] rings and bridging functional groups[44−50] has become a valuable approach.[8] Although a number of significant contributions have been made in the field of rigid combretastatin analogs, only limited examples of this approach using an imidazole core have been documented.[44,51,52]

Incorporation of a relevant scaffold in generating a new molecular skeleton as the potential target-specific bioactive agent is a crucial aspect in drug discovery. 2-Aminoimidazole is a valuable heterocyclic motif that is common in the marine-sponge alkaloids[53,54] and in therapeutic/bioactive agents.[29,37,55,56] It can provide an interesting hydrogen bond donor/acceptor pattern, while interacting with a target. With varied poly-substitutions/functionalities, the 2-aminoimidazoles exhibit diverse physicochemical properties.[57,58] Recently, we found a new series of 4,5-diaryl-2-aminoimidazole analogues as potent tubulin assembly inhibitors[44] and explored the distinctive pharmacophoric features of 2-aminoimidazole as a bridging motif. The incorporation of an additional linker between the bridging alkene motif and the aryl group to provide the flexibility is significant.[59] To synthesize CA-4-inspired new molecular entities, a series of 4-aryl-5-aroylimidazole-2-amines has been considered (Figure ). We imagined that the substitution of the olefinic bond of CA-4 with a 2-aminoimidazole bridging motif would provide hydrogen bond donor/acceptor points by C2-amino and ring N/NH functionalities for tubulin binding (Figure ). Further, it can induce drug-like properties and increase water solubility. It is worth mentioning that the carbonyl oxygen of phenstatin generates essential interactions with the colchicine binding site of tubulin.[46] The carbonyl oxygen in new agents (aroylindole and diarylketone–chalcone) was found to impede the assembly of tubulin and enhance the antiproliferative ability.[40,60−62] Therefore, the assembly of 2-aminoimidazole-carbonyl as a bridging motif, trimethoxyphenyl as a ring, and varied relevant aryls and heteroaryls as the other ring can be an important approach for discovering potential tubulin inhibitors (Figure ).

Figure 1

Design of novel 4-aryl-5-aroyl-imidazole-2-amines.[40,44,59,61,62]

Interestingly, several of the tested 2-aminoimidazole-carbonyl analogues of CA-4 displayed pronounced antiproliferative activities against HeLa and A549 cells. Among these, C-13 was found to be significantly more potent than CA-4 against lung carcinoma cells. Most importantly, C-13 displayed 15.7 and 4 folds higher antiproliferative activity toward breast cancer and skin melanoma cells compared to their noncancerous counterparts, respectively. Further, C-13 displayed more specific cytotoxicity toward breast cancer cells as compared to noncancerous breast epithelial cells than its parent compound CA-4. C-13 strongly depolymerized both interphase and mitotic microtubules, perturbed chromosome organization, halted mitotic progression in cultured cells, and induced apoptotic cell death. The importance of the molecular skeleton in the structural modulation of combretastatin-inspired clinical agents has been explored, and the evidence presented in the study indicated that C-13 has a strong anticancer potential.

Results and Discussion

Chemistry

2-Aminoimidazoles are routinely synthesized via the condensation of an α-haloketone with an acetylated guanidine, or condensation of an α-aminoketone with cyanamide[63−65] or by functionalization of imidazole scaffold via protection, C2-amination, introduction of substituents, and deprotection.[66] These classical reactions as key steps in synthesis are not feasible to prepare our designed 4-aryl-5-aroyl-2-aminoimidazole series of compounds with relevant substitution diversity. We have developed recently a method[67] to access polysubstituted aminoimidazoles. The method circumvents the necessity of electronic modulation of guanidine. The reaction pathway involves aza-Michael addition, SN2, and a unique redox-neutral process. Significant advantages of the developed methodology are the convenient procedure, step economy, excellent substrate scope, introduction of distinctive substitutions/functionalities into the 2-aminoimidazole core, and use of easily accessible materials. Keeping in view the potential structural features of 4-aryl-5-aroylimidazole-2-amines for exhibiting tubulin polymerization inhibitory activity, we synthesized a series of investigational compounds in our previous work[67] and two new compounds (8 and 10) with relevant substitutions (compounds 1–21, Figure ). Compounds 8 and 10 that possess hydroxyl functionality were obtained from compounds 7 and 9, respectively, by Pd–C-catalyzed hydrogenative debenzylation. The trimethoxyphenyl motif was considered in ring A. Several relevantly substituted aryls and heteroaryls were considered in ring B. Aromatics contain methoxy, hydroxyl, methoxy-hydroxyl, dimethoxy, trimethoxy, fluoro, chloro, and cyano groups. Heteroaromatics are quinoline, pyridine, and thiophene groups/motifs. It is worth to mention that the quinolinyl motif as ring B in our previously explored combretastatin analog[44] showed its pharmacophoric importance in displaying potent tubulin polymerization inhibitory property. Switch in substituted aryl rings (A vs B) was also considered in the synthesis (4 and 16). Also, to validate the significance of 3,4,5-trimethoxyphenyl moiety in ring A, few compounds without trimethoxy substitutions were prepared (17–20).

Figure 2

Structures of investigated combretastatin inspired 2-aminoimidazoles (1–7, 9, 11–21)[67] and 8 and 10: all the compounds except compounds 8 and 10 were prepared by vicinal Csp2–H and Csp2–Br guanidination reaction and were reported previously.[67]

All the products were identified by 1H and 13C NMR and IR spectroscopic techniques and confirmed by high-resolution mass spectrometry (HRMS). Further, single-crystal X-ray crystallographic analysis of the synthesized compound[67] indicated that the skeleton possesses syn-locking of two aryl rings across the 2-aminoimidazole-carbonyl motif and a favorable range of dihedral angle of two aryls as known in the literature,[59,68] indicating skeleton’s flexibility to have an important three-dimensional structural feature required in binding with tubulin at the colchicine site.

Biological Studies

Screening Assay

The combretastatin analogues (1–21) were initially screened using HeLa cells at 25 nM concentration by sulforhodamine B assay.[44,69] The compounds inhibited HeLa cell proliferation with varying ability as shown in Figure . Of the compounds evaluated, 7 compounds showed more than 50% inhibition of HeLa cell proliferation at 25 nM concentration (Figure ).

Figure 3

Screening of combretastatin analogues in HeLa cells. The inhibition of cell proliferation was plotted for each of the analogue at 25 nM concentration in HeLa cells. Three sets of experiments were carried out. Error bars represent standard deviation.

To find out the most potent compound against lung carcinoma, the half-maximal inhibitory concentration (IC50) of these active compounds was deduced in A549 cells in comparison with CA-4 (Table , Figure S5A). C-13 and CA-4 inhibited the proliferation of A549 cells with an IC50 of 48 ± 10 and 112 ± 7 nM, respectively, indicating that C-13 was more potent than CA-4 against A549 cells. Therefore, the antiproliferative activity of C-13 was further characterized.

Table 1

IC50 of the 7 Active Combretastatin Analogues in the A549 Cell Linea

compound no.	IC50 (nM)
12	307 ± 19
13	48 ± 10
14	725 ± 59
16	183 ± 12
17	247 ± 43
20	314 ± 66
21	147 ± 44
CA-4	112 ± 7

Error represents standard deviation. Data is an average of at least three sets.

C-13 Showed Potent Antiproliferative Activity against Various Cancer Cell Lines

C-13 exerted potent cytotoxic effects on different types of tumor cells in culture including a multidrug-resistant breast cancer, EMT6/AR1 cell line (Table , Figure S5B–G). For example, C-13 inhibited HeLa cell proliferation with an IC50 of 36 ± 3 nM. Interestingly, C-13 inhibited the proliferation of MCF-7, MCF 10A, B16F10, and L929 cells with an IC50 of 121 ± 10, 1891 ± 55, 416 ± 20, and 1655 ± 111 nM, respectively (Table , Figure S5B–F). The finding indicated that C-13 exerted 15.7 and 4 times stronger antiproliferative effects on MCF-7 breast cancer cells and B16F10 skin melanoma cells than noncancerous MCF 10A breast epithelial cells and L929 skin fibroblast cells, respectively. CA-4 inhibited the proliferation of MCF-7 and MCF 10A cells with an IC50 of 18 ± 3 and 144 ± 31 nM, respectively (Figure S6). The antiproliferative activity of C-21 was also tested in B16F10 and L929 cells. C-21 showed an IC50 of 456 ± 78 and 1812 ± 56 nM in B16F10 and L929 cells, respectively.

Table 2

IC50 of C-13 in Different Cell Linesa

cell lines	IC50 (nM)
HeLa	36 ± 3
MCF-7	121 ± 10
MCF 10A	1891 ± 55
B16F10	416 ± 20
L929	1655 ± 11
EMT6/AR1	2432 ± 47

Error represents standard deviation. Data is an average of at least three sets.

C-13 Depolymerized Microtubules in HeLa and A549 Cells

Combretastatin derivatives are well-known microtubule depolymerizing agents. C-13 also induced microtubule depolymerization in HeLa and A549 cells (Figure ). At a concentration of 100 nM, C-13 induced a significant depolymerization of microtubules and 200 nM C-13 induced a strong depolymerization of interphase microtubules in HeLa and A549 cells (Figure ). Further, C-13 strongly depolymerized spindle microtubules in HeLa and A549 cells and perturbed the bipolar spindle formation, causing misalignment of chromosomes at the metaphase plate (Figure ). Using γ-tubulin staining, 3, 24, and 58% of the mitotic spindles was found to be multipolar in the absence and presence of 100 and 200 nM C-13. Similar microtubule depolymerizing effects were observed in HeLa and A549 cells with C-21 (Figures S7 and S8).

Figure 4

C-13 depolymerized microtubules in HeLa and A549 cells. HeLa and A549 cells were incubated in the absence and in the presence of 100 and 200 nM C-13 for 24 h and subjected to immunostaining. Grayscale images are shown. The scale bar is 10 μm.

Figure 5

Effect of C-13 on mitotic spindles of HeLa and A549 cells. HeLa and A549 cells were incubated in the absence and presence of 100 and 200 nM C-13 for 24 h. Cells were fixed and stained for α-tubulin (green) and DNA (blue). The scale bar is 10 μm.

HeLa cells were incubated with 200 nM C-13 for 1 and 2 h. This short exposure of C-13 also strongly disassembled both interphase and spindle microtubules and affected chromosome congression in the mitotic HeLa cells (Figure S9A,B). To check the reversibility of the effect of the compound, HeLa cells were treated with 200 nM C-13 for 1 h followed by washing the compound. The percentage of dead cells in C-13-treated cells was found to be two times higher than the vehicle-treated cells, suggesting that even a brief exposure of C-13 can induce cell killing (Figure S10).

To confirm that C-13 induced microtubule depolymerization in HeLa cells, the ratio of polymeric to soluble tubulin in HeLa cells was determined using western blotting (Figure A). The ratio of polymeric to soluble tubulin in HeLa cells was determined to be 2.1 ± 0.4, 1.6 ± 0.1, and 1.1 ± 0.1 in the absence and presence of 120 and 240 nM C-13 (Figure B). The ratio of polymeric to soluble tubulin of cells treated with 25 nM vinblastine was found to be 1.4 ± 0.9. The decrease in the polymeric tubulin and the increase in the soluble fraction of tubulin suggested that C-13 depolymerizes microtubules in HeLa cells.

Figure 6

C-13 reduced the polymeric/soluble tubulin fraction in HeLa cells as determined by western blot. (A) HeLa cells were treated either without or with 120 and 240 nM C-13 for 24 h. Vinblastine (25 nM) was used as a positive control. The polymeric and soluble fractions of tubulin were isolated and an equal amount of protein was loaded onto sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Immunoblotting was carried out using a monoclonal antibody against α-tubulin. The experiment was carried out three times. One of the blots is shown. (B) Ratio of polymer/soluble tubulin fraction was quantified using ImageJ software, and statistical significance was determined using student’s t-test. (ns: p > 0.05; *: p < 0.05). The error bar signifies standard deviation.

Microtubule-targeting agents generally perturb microtubule dynamics at a lower concentration than it is required to visibly depolymerize microtubules.[3,70,71] EB1, a plus-tip binding protein,[72] binds to the growing end of dynamic microtubules; therefore, a change in the localization of EB1 may provide an idea about the perturbation of microtubule dynamics. Thus, we examined the effect of C-13 on the localization of EB1 in GFP-EB1-expressing HeLa cells. In control HeLa cells, EB1 comets were distinctly observed at the tips of microtubules (Figure ). However, at 35 and 70 nM C-13, the localization of EB1 was perturbed, and fewer, diffused comets of EB1 were observed (Figure ), suggesting that the delocalization of EB1 was in response to perturbation of the microtubule architecture in HeLa cells. The finding indicated that C-13 could perturb microtubule dynamics.

Figure 7

C-13 affected the localization of EB1. GFP-EB1-expressing HeLa cells were treated with either the vehicle or 35 and 70 nM C-13, and live-cell imaging was carried out. The scale bar is shown in the figure.

C-13 Induced Mitotic Block in HeLa Cells

Because chromosome movement during mitosis is dependent on microtubules and an improper alignment of the chromosomes can induce mitotic block, we examined whether C-13 can block cells at mitosis or not. The effect of C-13 on the progression of HeLa cells was first examined by flow cytometry. The percentage of HeLa cells in the G2/M phase was determined to be 26, 70, and 78% in the absence and presence of 75 and 200 nM C-13, indicating that C-13 treatment prevents the progression of HeLa cells at the G2/M phase (Figure A). HeLa cells treated with 20 nM CA-4 showed 73% of cells in the G2/M phase (Table S1).

Figure 8

C-13 increased the mitotic index in HeLa cells. (A) HeLa cells were incubated in the absence (a) and presence of 75 nM C-13 (b), 200 nM C-13 (c), and 20 nM CA-4 (d) for 12 h, and cell cycle analysis was performed using flow cytometry by staining the DNA in cells with propidium iodide (PI). The experiment was performed twice. (B) HeLa cells were incubated in the absence and presence of 75 and 200 nM C-13 for 12 h. Cells were stained with antibody against phospho-histone H3 (green), and Hoechst 33258 was used for staining the DNA (blue). The experiment was performed thrice. The scale bar is 10 μm.

C-13 treatment was found to halt the progression of HeLa cells at the G2/M phase; therefore, we next determined the effect of the compound on the mitotic index. The effect of C-13 on the mitotic index in HeLa cells was first determined based on the morphology of DNA, stained using Hoechst 33258 dye (Table ). The mitotic index of the vehicle-treated control was found to be 3 ± 0.6, whereas in the presence of 75 and 200 nM C-13, the mitotic index increased to 12 ± 2.3 and 23 ± 2.2, indicating that C-13 treatment increased the mitotic index of HeLa cells. Under similar conditions, HeLa cells treated with 20 nM CA-4 showed a mitotic index of 40 ± 1.2.

Table 3

Mitotic Index Determined by Hoechst and Phospho-Histone Staining in HeLa Cellsa

	mitotic index
	Hoechst staining	phospho-histone staining
control	3 ± 0.6	3 ± 0.2
75 nM C-13	12 ± 2.3	13 ± 0.3
200 nM C-13	23 ± 2.2	23 ± 0.7
20 nM CA-4	40 ± 1.2	38 ± 0.3

Error represents standard deviation. Data are an average of at least three sets, and 500 cells were counted in each case.

In mitosis, histone H3 at serine 10 gets phosphorylated and acts as a marker for mitosis.[73]C-13 increased the number of phospho-histone-positive cells with the increasing concentration compared to the vehicle-treated control (Figure B, Table ). The percentages of mitotic cells were found to be 3 ± 0.2, 13 ± 0.3, and 23 ± 0.7 in the presence of 0, 75, and 200 nM C-13, respectively. The results together suggested that C-13 induces a mitotic block in HeLa cells. Under similar experimental conditions, the percentage of mitotic cells in HeLa cells treated with 20 nM CA-4 was found to be 38 ± 0.3.

Characterization of C-13 Binding to Tubulin

The binding interaction of C-13 to tubulin was investigated using the tryptophan fluorescence of tubulin. C-13 decreased the fluorescence intensity of tubulin, indicating that the compound interacts with tubulin (Figure A). The changes in the fluorescence intensities were fitted into a binding equation to determine a dissociation constant (Kd) for the interaction of tubulin with C-13 (Figure B). The Kd was calculated to be 1.3 ± 0.4 μM. Similarly, a Kd for the interaction of tubulin with C-21 was found to be 2.0 ± 1.4 μM (Figure S11A,B).

Figure 9

C-13 decreased the intrinsic tryptophan fluorescence of tubulin. (A) Tubulin (2 μM) was incubated without (box solid) and with different concentrations 0.5 (circle solid), 1 (triangle up solid), 2 (triangle down solid), 5 (triangle left-pointing solid), 10 (triangle right-pointing solid), 15 (diamond solid), 20 (pentagon up solid), 25 (pentagon down solid), 30 (star solid), and 40 (circle with all but the upper left quadrant black) μM of C-13 in PIPES buffer (25 mM, pH 6.8) for 30 min at 25 °C. Fluorescence emission spectra were monitored using 295 nm as excitation wavelength. (B) Changes in fluorescence intensity at 335 nm upon binding of C-13 to tubulin were plotted. The experiment was done five times. One of the five sets is shown.

Determination of the C-13 Binding Site on Tubulin

Combretastatins are reported to interact with tubulin at the colchicine site.[74,75] Molecular docking analysis was performed to find the putative binding site of C-13 on the tubulin dimer, as reported recently[76] (Figure S12). We first performed docking of N-deacetyl-N-(2-mercaptoacetyl)colchicine (DAMA-colchicine) on tubulin (PDB ID: 1SA0)[77] as well as the docking of CA-4 on the tubulin dimer (PDB ID: 5LYJ).[75] The root mean square deviation (RMSD) values between the docked and the crystal structures of the DAMA-colchicine and CA-4 were found to be 1.2 and 1.6 Å, respectively, indicating acceptable docking (Figure S12).

Using a similar procedure, C-13 was docked on tubulin. On overlapping the docked conformations of colchicine, CA-4 and C-13, we observed that C-13, similar to colchicine and CA-4, binds to tubulin at the colchicine-binding site (Figure S13). Further analysis of the interacting residues lying within 4 Å of the docked conformation of C-13 suggested that the binding pocket is mainly composed of hydrophobic amino acids (Table S2). Several of the interacting residues present in the binding pocket of C-13 were also found to be present in the binding pocket of colchicine and CA-4 (Table S2), suggesting that C-13, colchicine, and CA-4 share a common binding site.

In addition to the hydrophobic interactions, the docked conformation of C-13 also showed possible hydrogen bonding interactions of C-13 with tubulin residues (Table S3). The distance between the thiol group of Cys241B and the aryl-methoxy group of C-13 was measured to be 3.09 Å, indicating a hydrogen bond between them. Another probable hydrogen bonding interaction was observed between the amide hydrogen of Ile318B and aryl-methoxy of C-13 as the distance between them was 3.07 Å. Further, we measured the distances between carbonyl oxygen of the amide bond of Asn251B and hydrogen atoms of the amine of the imidazole moiety of C-13, with distances of 2.54 and 2.77 Å indicating either of the two possible hydrogen bonding interactions. This suggests that the hydrogen bonding interactions in addition to the hydrophobic interactions may stabilize C-13 in the binding pocket. A similar analysis was carried out for colchicine and CA-4 (Table S3).

The binding energies for colchicine, CA-4, and C-13 were calculated using AutoDock 4.2 software and were found to be −9.02, −7.87, and −8.44 kcal/mol, respectively, indicating that C-13 binds to tubulin dimer more strongly than CA-4 (Table S3).

Biochemically, we performed a competitive binding assay with colchicine to confirm the putative binding site of tubulin. We found out that C-13 decreased the fluorescence intensity of the tubulin–colchicine complex, implying that C-13 inhibits the binding of colchicine to tubulin (Figure A). A Ki value of 1.1 ± 0.13 μM was estimated for the inhibitor (Figure B). Similarly, C-21 was also found to decrease the fluorescence intensity of the tubulin–colchicine complex with a Ki value of 11.7 ± 7.3 μM (Figure S14A,B). In addition, the previously reported[44] compound 12, a combretastatin analogue, is reported to bind at the colchicine-binding site on tubulin and the compound fluoresces strongly upon binding to tubulin. C-13 diminished the fluorescence intensity of the compound 12–tubulin complex, indicating that C-13 binds to the combretastatin-binding site on tubulin (Figure C,D). Thus, the biochemical competitive binding assays and the docking analysis strongly suggested that C-13 binds to tubulin at the colchicine-binding site.

Figure 10

C-13 binds to the colchicine binding site. (A) Tubulin (5 μM) was incubated without (box solid) and with different concentrations 2 (circle solid), 5 (triangle up solid), 10 (triangle down solid), 20 (triangle left-pointing solid), 30 (triangle right-pointing solid), 50 (diamond solid), 60 (pentagon up solid), and 70 (hexagon solid) μM of C-13 in PIPES buffer (25 mM, pH 6.8) for 15 min at 37 °C. Colchicine (10 μM) was then added into the reaction mixture and incubated at 37 °C for 45 min. Fluorescence spectra were monitored using 340 nm as the excitation wavelength. (B) Fluorescence intensity (at 434 nm) change is plotted against different logarithmic concentrations of C-13. The experiment was carried out four times. One of the four sets is shown. (C) Tubulin (2 μM) was incubated without (box solid) and with different concentrations 2 (circle solid), 5 (triangle up solid), 10 (triangle down solid), 15 (triangle left-pointing solid), 20 (triangle right-pointing solid), 30 (diamond solid), 50 (pentagon up solid), and 60 (hexagon solid) μM of C-13 in PIPES buffer (25 mM, pH 6.8) for 20 min at 37 °C. Subsequently, a previously reported compound 12 (5 μM) was added into the reaction milieu and incubated for 10 min at room temperature. Fluorescence spectra were measured using 350 nm as the excitation wavelength. (D) Percentage inhibition of compound 12 binding against the concentration of C-13 was plotted. Three sets of experiments were carried out. One of the three sets is shown.

C-13 Inhibited Microtubule-Associated Protein-Rich Tubulin and Taxol-Induced Tubulin Polymerization

C-13 disrupted microtubule organization in cells; therefore, the influence of C-13 on microtubule-associated protein (MAP)-rich tubulin polymerization was analyzed. C-13 inhibited the polymerization of MAP-rich tubulin with an IC50 of 63 ± 27 μM (Figure A,B). Also, C-13 inhibited taxol-induced polymerization of purified tubulin (Figure C,D). For example, 30, 60, 90, and 150 μM C-13 inhibited tubulin polymerization by 33 ± 9, 40 ± 8, 55 ± 8, and 65 ± 18% respectively.

Figure 11

C-13 inhibited MAP-rich tubulin and taxol-induced tubulin polymerization. (A) MAP-rich tubulin (2 mg/mL) was polymerized in the presence of 1 mM GTP without (box solid) and with different concentrations 5 (circle solid), 10 (triangle up solid), 20 (triangle down solid), 50 (triangle left-pointing solid), 100 (triangle right-pointing solid), and 200 (diamond solid) μM of C-13 in PEM buffer. Assembly kinetics was monitored at 350 nm (37 °C). (B) Percentage inhibition of polymerization against the C-13 concentration was plotted. The experiment was carried out four times. One of the four sets is shown. (C) Tubulin (18 μM) was incubated without (box solid) and with different concentrations: 30 (circle solid), 60 (triangle up solid), 100 (triangle down solid), and 150 (triangle left-pointing solid) μM of C-13 in PEM buffer on ice for 10 min. Subsequently, 10 μM taxol was added to the reaction mixture followed by 1 mM guanosine 5′-triphosphate (GTP). Kinetics of the tubulin assembly was monitored by taking absorbance at 350 nm (37 °C). (D) Percentage inhibition of polymerization against the C-13 concentration was plotted. Three sets of experiments were performed. One of the three sets is shown.

C-13 Induced Apoptosis in HeLa Cells

C-13 treatment increased the number of dead cells (Figure A). For example, the percentages of PI-stained positive cells in the absence and presence of 75 and 200 nM C-13 were 6, 42, and 74%. Under similar conditions, the percentage of PI-stained positive cells in HeLa cells treated with 20 nM CA-4 was 87%. Cleavage of poly(ADP-ribose) polymerase (PARP), which is mainly involved in DNA repair, acts as a marker for apoptosis.[78] On western blot analysis, a single band of PARP (116 kDa) was observed in vehicle-treated HeLa cells, whereas two bands (116 and 89 kDa) were observed in HeLa cells treated with 75 and 200 nM C-13, confirming the cleavage of PARP (Figure B).

Figure 12

C-13 induced apoptosis and increased the production of ROS in HeLa cells. (A) HeLa cells were incubated without and with 75 and 200 nM C-13 for 24 h. PI staining was carried out. Two independent sets of experiments were performed. The scale bar is 10 μm. (B) HeLa cells were treated without and with 75 and 200 nM C-13 for 24 h. C-13 treatment produced PARP cleavage. β-Actin was used as a loading control. The experiment was carried out three times. One of the blots is shown. (C) HeLa cells were treated without and with 75 and 200 nM C-13, 20 nM CA-4, and 100 μM H2O2 and incubated for 6 h. Cells were incubated with DCFDA (25 μM) dye for 1 h at 37 °C in dark. Fluorescence intensity per cell at 525 nm was calculated, and statistical significance was deduced using student’s t-test. Three sets of experiments were performed independently. (*: p < 0.05; **: p < 0.01; ***: p < 0.001). The error bar represents standard deviation.

C-13 Increased the Intracellular Reactive Oxygen Species in HeLa Cells

The detection of the intracellular reactive oxygen species (ROS) concentration was carried out using 2′,7′-dichloro fluorescein diacetate (DCFDA) dye.[79]C-13 treatment increased the intracellular ROS concentration in HeLa cells (Figure C). The fluorescence intensity of DCFDA per cell in the absence and presence of 75 and 200 nM C-13 was determined to be 2.4 ± 0.1, 3.0 ± 0.1, and 3.9 ± 0.2 (Figure C). The fluorescence intensity of DCFDA per cell was determined to be 3.6 ± 0.4 and 4.5 ± 0.2 when HeLa cells were treated with 20 nM CA-4 and 100 μM H2O2, respectively. A brief exposure (2 h) of C-13 produced a significant depolymerization of microtubules in HeLa cells, while it did not stimulate ROS. For example, the intensity of microtubules reduced significantly by 33 ± 8% (p < 0.0001) upon 2 h of incubation with 200 nM C-13 (Figure S9A), while the fluorescence intensity of ROS was increased by only 2.8 ± 1%. The results indicated that the depolymerization of microtubules preceded ROS production.

C-13 Retarded A549 Cell Migration

Combretastatins are known to be vascular-targeting agents, so we checked the effect of C-13 on the migration of A549 cells using a wound-closure assay (Figure ). The wound was completely healed in 10 h in the vehicle-treated control cells. However, after 10 h, the wound was only partly healed in C-13-treated cells (Figure S15A), and 100 and 200 nM C-13 inhibited the migration of A549 cells by 50 ± 14 and 58 ± 13%, respectively (Figure ). C-13 also affected the rate of migration of A549 cells (Figure S15B). The rate of migration in vehicle-treated cells was found to be 3 ± 0.7 μm/h, while in 100 and 200 nM C-13-treated cells, the rate of migration was found to be 1.6 ± 0.3 and 1.1 ± 0.1 μm/h, respectively. A549 cells treated with 100 nM CA-4 inhibited the migration by 77 ± 5% and retarded the rate to 0.4 ± 0.2 μm/h.

Figure 13

C-13 inhibited the migration of A549 cells. The wound made by a scratch was allowed to heal without or in the presence of 100 and 200 nM C-13. Three sets of experiments were carried out. The scale bar is 10 μm.

Conclusions

In conclusion, twenty-one combretastatin-inspired analogues were investigated for their anticancer activity using cultured cells. The structure modulation of combretastatins involved the incorporation of the 2-aminoimidazole-carbonyl group as a double bond replacement motif that possesses a pharmacodynamically interesting pattern of H-bond acceptors/donors as well as provides the requisite syn-orientation of aryls, relevant heteroaryls, or functionalized aryls in ring B, and the switch of two aryls. Here, we report the highly potent tubulin-targeting antiproliferative agent C-13 that possesses thiophene as ring B. It is the most active among the compounds investigated and shows strong differential activity against cancerous and noncancerous cells in culture. The evidence presented in the study suggests that C-13 inhibits the proliferation of cells by targeting microtubules similar to CA-4 and colchicine. The present study thus will aid in the rational structural modulation of clinical agents that target the colchicine-binding site in tubulin and may provide guidance in the synthesis of novel tubulin-targeting anticancer agents with improved efficacy.

Experimental Section

General Remarks

C-13 was screened for PAINS using several biochemical assays. The results suggested that C-13 is not a PAINS molecule. First, C-13 was found to inhibit the binding of colchicine to tubulin in a concentration-dependent manner with a Ki of 1.1 ± 0.13 μM. To determine the specificity of C-13 binding, we checked whether C-13 could bind to proteins such as bovine serum albumin (BSA), lysozyme, trypsin, and Streptococcus pneumoniae FtsZ. C-13 was found not to bind to any of these proteins.

To examine whether C-13 affects the activity of any unrelated targets, we determined the effect of C-13 on polymerization of S. pneumoniae FtsZ, a bacterial homolog of tubulin, and also on the enzymatic activity of alkaline phosphatase. C-13 did not affect the polymerization of S. pneumoniae FtsZ. Further, C-13 did not inhibit the enzymatic activity of alkaline phosphatase. These experiments together strongly suggest that the anti-tubulin activity shown by C-13 is not an artifact. In addition, C-13 satisfies Lipinski’s rule of five for drug-likeness.[80]

General Considerations

Organic substrates, reagents, and solvents were utilized as obtained from commercial suppliers without their additional purification. The progress of the organic reactions was monitored by thin layer chromatography (TLC) (Merck, silica gel 60 F254). The acquisition of the NMR spectra was done on a Bruker Avance DPX 400 MHz spectrometer using solvent CDCl3, DMSO-d6, or CD3OD and tetramethylsilane as an internal standard. J values are calculated in Hz. HRMS (ESI) spectrometric data were obtained in a Bruker-maXis mass instrument. IR spectra of samples as thin films (neat) were recorded on a Nicolet FT-IR Impact 410 instrument. High-performance liquid chromatography (HPLC) (Shimadzu LC-6AD system) was used to analyze the purity of the synthesized compounds. The HPLC method involves the Phenomenex RP-C18 column (250 × 4.60 mm), particle size 5 μm, water–acetonitrile as an eluting solvent system, and a flow rate of 1 mL/min. The purity of all the compounds was found to be >95%.

General Experimental Procedure for the Synthesis of 4-Aryl-5-aroyl-1H-imidazole-2-amines (1–7, 9, 11–21)

The compounds were prepared following our previously reported method.[67] A sealed tube was made oven-dried. It was set up with a rubber septum and a magnetic bar under a nitrogen atmosphere (flow of nitrogen gas). In the sealed tube, α-bromochalcone (E/Z-mixture, 0.5 mmol), guanidine·HCl (1.5 mmol, 143 mg, 3 equiv), potassium carbonate (1.75 mmol, 242 mg, 3.5 equiv), and manganese dioxide (0.75 mmol, 65 mg, 1.5 equiv) were subsequently taken. Then, the solvent dioxane (anhyd., 8 mL) was added, and the tube with chemicals and nitrogen gas was sealed. The mixture of components was then heated with stirring at 100 °C. The progress of the transformation was checked by TLC. After maximum conversion (16 h), the resultant mixture was cooled down to room temperature, diluted with EtOAc–MeOH (1:1, 60 mL), and filtered through celite. The rotary evaporator-vacuum pump was used to evaporate the solvent(s) from the solution. The crude mass was subjected to column chromatographic purification–isolation on silica gel (100–200 mesh), which was partially acid-neutralized by passing a small volume (1–4 mL) of triethylamine base. MeOH–EtOAc (5:95) was used as the eluting solvent. 4-Aryl and 5-aroyl substituted 1H-imidazole-2-amines (compounds 1–7, 9, 11–21) were obtained. The molecular constitutions of these investigated compounds were identified by spectroscopic data (1H and 13C NMR, IR) and confirmed by HRMS. The samples were found to be >95% pure by HPLC analysis.

The spectroscopic data and the scanned spectra of the compounds 1–7, 9, 11–21 were published in our previously reported article.[67]

General Procedure for Debenzylation of Compounds 7 and 9 toward the Synthesis of Compounds 8 and 10, Respectively

Compound 7 or compound 9 (0.5 mmol) was taken in a round bottom flask, and degassed anhydrous MeOH (3.5 mL) was added under nitrogen. Palladium on carbon (10%) was added. The mixture was stirred under hydrogen atmosphere. The progress of the reaction was monitored by TLC. After completion of the reaction, the mixture was diluted with EtOAc–MeOH (1:1, 60 mL). The resultant mixture was filtered through celite. The solvent was evaporated under reduced pressure (rotary evaporator). The column chromatographic purification of crude mass was performed on silica gel (100–200 mesh) partially deacidified by passing triethylamine (1–4 mL). MeOH–EtOAc (5:95) as eluting solvent was used. It provided 4-aryl-5-aroyl-1H-imidazole-2-amines (compound 8 or compound 10).

Characterization Data of Compounds 8 and 10

4-(3-Hydroxy-4-methoxyphenyl)-5-(3,4,5-trimethoxybenzoyl)-1H-imidazol-2-amine (8)

Yellow solid; 160 mg, 80%; mp > 200 °C; 1H NMR (400 MHz, DMSO-d6): δ 10.87 (s, NH), 8.78 (s, 1H), 6.77 (s, 1H), 6.71 (s, 2H), 6.60 (d, J = 8.2 Hz, 1H), 6.56 (d, J = 7.2 Hz, 1H), 5.95 (s, NH2), 3.67 (s, 3H), 3.62 (s, 3H), 3.56 (s, 6H) ppm; 13C{1H} NMR (100 MHz, DMSO-d6): δ 182.5, 153.2, 152.5, 149.3, 147.7, 146.1, 140.3, 134.4, 128.4, 121.8, 121.2, 116.9, 111.5, 106.8, 60.4, 56.1, 55.9, 55.4 ppm; IR (KBr) νmax: 3344, 2924, 1579, 1223, 1123 cm–1; HRMS (ESI) m/z: calcd for C20H22N3O6 [M + H]+, 400.1508; found, 400.1500.

4-(4-Hydroxyphenyl)-5-(3,4,5-trimethoxybenzoyl)-1H-imidazol-2-amine (10)

Yellow solid; 166 mg, 90%; mp > 200 °C; 1H NMR (400 MHz, DMSO-d6): δ 10.88 (s, NH), 9.43 (s, 1H), 7.05 (d, J = 7.8 Hz, 2H), 6.74 (s, 2H), 6.49 (d, J = 8.3 Hz, 2H), 5.94 (s, NH2), 3.62 (s, 3H), 3.55 (s, 6H) ppm; 13C{1H} NMR (100 MHz, DMSO-d6): δ 182.6, 157.4, 153.1, 152.6, 140.3, 134.4, 130.9, 126.0, 114.6, 106.9, 60.5, 55.9 ppm; IR (KBr) νmax: 3399, 1647, 1275, 1023, 995 cm–1; HRMS (ESI) m/z: calcd for C19H20N3O5 [M + H]+, 370.1403; found, 370.1400.

Cell Culture

HeLa, MCF-7, MCF 10A, A549, L929, and B16F10 were bought from National Centre for Cell Sciences (NCCS), Pune, India. The drug-resistant EMT6/AR1 cell line was bought from Sigma, St. Louis, MO, USA. HeLa and MCF-7 were maintained using Eagle’s minimal essential medium. L929 and B16F10 cells were grown in Dulbecco’s modified Eagle’s medium (DMEM), whereas A549 cells were cultured in F-12K nutrient medium, supplemented with 10% (v/v) fetal bovine serum and 1% (v/v) antibiotic–antimycotic solution as described earlier.[81] The EMT6/AR1 cell line was maintained as described previously.[44] MCF 10A was grown in DMEM and F12K in a 1:1 ratio supplemented with 10% (v/v) fetal bovine serum, 1% (v/v) antibiotic–antimycotic solution, epidermal growth factor (20 ng/mL), hydrocortisone (0.5 μg/mL), and insulin (10 μg/mL).[82] All the cells were grown in optimized conditions in a humidified incubator in 5% CO2 at 37 °C (Sanyo, Tokyo, Japan).

Screening of Combretastatin Analogues

A cell-based screening assay was performed to check the antiproliferative potential of the 21 synthesized compounds. The stocks of the combretastatin analogues were prepared in 100% dimethyl sulfoxide (DMSO) (cell culture grade). The compounds (1–21) were screened in HeLa cells using sulforhodamine B assay.[69] Briefly, HeLa cells (10 000 cells/well) were incubated for 24 h in 96-well cell culture plates in a humidified CO2 incubator. Then, the cells were incubated with either vehicle (0.1% DMSO) or 25 nM of each of the compounds for 24 h. After the incubation, the cells were fixed using 50% trichloroacetic acid for 1 h at 4 °C, washed, completely dried, and subsequently incubated with sulforhodamine B (0.4% in 1% acetic acid) for 1 h, followed by washing with 1% glacial acetic acid. When the plates dried completely, tris chloride (10 mM, pH 10.0) was added to the wells and incubated for 30 min at room temperature. Absorbance was measured at 520 nm using Spectramax M2e, and the inhibition of cell proliferation was determined.

Determination of IC50

A549 cells (10 000 cells/well) were incubated for 24 h in a 96-well cell culture plate. Then, the cells were incubated with different concentrations of the 7 active compounds for 24 h and further processed for sulforhodamine B assay.[69]

The IC50 values were calculated using GraphPad software version 6.0 (GraphPad Software, CA, USA) by fitting the values in an equation from nonlinear regression[81]where Y is the response, X is the logarithmic concentration of the compound, Bottom is the minimum response, Top is the maximum response, and IC50 is the concentration of the compound that gives a response mid-way between Top and Bottom.

The IC50 values for both C-13 and C-21 in L929 and B16F10 and C-13 in HeLa, MCF-7, MCF 10A, and EMT6/AR1 cells were obtained similarly by incubating the cells for one cell cycle. All IC50 values were determined three times independently for each of the cell lines.

Immunofluorescence Assay

Immunofluorescence assay was performed as described previously.[83] Briefly, HeLa or A549 cells (2.5 × 104 cells/well) were seeded onto glass coverslips and incubated for 24 h. HeLa cells and A549 cells were treated either with the vehicle (0.1% DMSO) or with 100 and 200 nM of C-13. Immunostaining was performed by staining the cells with α-tubulin antibody (1:400), γ-tubulin (1: 300), and phospho-histone H3 (serine 10) antibody (1:400) and diluting in 2% BSA in phosphate-buffered saline (PBS) for 3 h at room temperature or overnight at 4 °C. Fluorescein isothiocyanate (FITC)-conjugated IgG secondary antibody (1:400 dilution in 2% BSA in PBS) was later added to the wells and incubated for 1 h at room temperature.[44] DNA was stained with Hoechst 33258 (10 μg/mL). The percentage of the multipolar spindle in the absence and presence of C-13 was determined by scoring 300 mitotic cells in each of the experimental conditions. The images were captured using an Eclipse TE 2000U microscope (Nikon, Tokyo, Japan) at 60× magnification and processed with Image-Pro Plus software (Media Cybernetics, Silver Spring, MD). Similarly, immunostaining was performed by treating HeLa cells with 35 and 70 nM C-21 and A549 cells with 300 and 600 nM C-21 as stated above.

Drug Retention Experiment

HeLa cells were treated with the vehicle (0.1% DMSO) and 200 nM C-13 and incubated for 1 h in a CO2 incubator at 37 °C. Later, the cells were washed thrice with Dulbecco’s phosphate-buffered saline, and fresh media was added to the flasks. After 24 h, the cells were processed for live/dead assay by flow cytometry using 5 μL of PI (50 μg/mL) and analyzed on a BD FACSAria instrument (BD, San Jose, CA, USA).

Determination of the Amount of Soluble and Polymeric Tubulin in Cells after C-13 Treatment

The effect of C-13 on the polymerized amount of tubulin in HeLa cells was determined by western blot as described earlier.[83] HeLa cells (in T-25 flasks) were incubated in the absence and presence of 120 and 240 nM C-13 for 24 h. After 24 h, the cell pellet was collected and added to the pellet PEM buffer with 25% glycerol and 0.5% Triton X-100 without disturbing the pellet. It was then incubated for 2 min at 37 °C, and the supernatant was removed gently from the top of the cell pellet. The supernatant represents the soluble fraction of tubulin. From the remaining cell pellet, lysates were prepared[84] by incubating the cells with lysis buffer [tris(hydroxymethyl)aminomethane (Tris) 20 mM, NaCl 200 mM, Triton X-100—0.1%, dithiothreitol (DTT) 1 mM at pH 7.2] for 1 h at 4 °C followed by centrifugation and collection of supernatants representing the polymeric fraction of tubulin. The protein concentrations for both the fractions of tubulin were measured by Bradford’s assay.[85] Protein (20 μg) from both the soluble and polymeric fraction was taken and subjected to SDS-PAGE. The protein band was transferred onto the poly(vinylidene difluoride) (PVDF) membrane via electro-blotting. Immunoblotting was carried out using the α-tubulin monoclonal antibody, and the intensity of the bands was analyzed using ImageJ software.

Effect of C-13 on EB1 Localization

The effect of C-13 on the localization of EB1 was studied in live HeLa cells. HeLa cells were transfected with GFP-EB1 plasmid[86] using Lipofectamine 3000 following manufacturer’s instruction and incubated for 6 h. Fresh media was added after 6 h and cells were allowed to grow for one cell cycle. Cells were then treated either with vehicle (0.1% DMSO) or with 35 and 70 nM C-13 and incubated for 24 h. Later, images were captured with a 60× oil objective using a laser-scanning microscope (Olympus FluoView 500).

Cell Cycle Analysis

HeLa cells were incubated with either the vehicle (0.1% DMSO) or 75 and 200 nM C-13 for 12 h. Subsequently, the cells were trypsinized, washed, and fixed with chilled ethanol (70%) in PBS. The cells were then incubated with 50 μg/mL PI and 1 μg/mL RNase for 30 min. Analysis of the cell cycle was performed using the BD FACSAria instrument (BD, San Jose, CA, USA), and the data were evaluated with ModFit LT version 5.0 (Verity Software House).[71]

Determination of the Dissociation Constant (Kd) for Binding of C-13 to Tubulin

Tubulin was isolated from the goat brain using two cycles of polymerization and depolymerization,[87] and the concentration of the purified tubulin was determined using Bradford’s assay.[85] Tubulin (2 μM) was incubated without or with various concentrations (0–40 μM) of C-13 in 25 mM PIPES buffer, pH 6.8 for 30 min at 25 °C. The tryptophan fluorescence was assessed in the absence and presence of C-13 using an excitation wavelength of 295 nm in a cuvette of 0.3 cm path length using a spectrofluorometer (JASCO FP-6500, Tokyo, Japan). Inner filter effect correction was performed for the measured fluorescence intensities using the formula[88]where Fcorrected is the corrected fluorescence intensity, Fobserved is the observed fluorescence intensity, Aexcitation is the absorbance of the compound at the excitation wavelength (295 nm), and Aemission is the absorbance of the compound at the emission wavelength (335 nm).

The dissociation constant (Kd) was calculated using GraphPad software version 6.0 (GraphPad Software, CA, USA) using the following equationwhere ΔF is the difference in the fluorescence intensity on binding with the compound, ΔFmax is the maximum difference in the fluorescence intensity when the compound saturates the binding site of tubulin, P0 is the concentration of tubulin, and L0 is the concentration of the compound. A similar experiment was carried out with C-21.

Determination of the Binding Site of C-13 on Tubulin

Tubulin (5 μM) was mixed with several concentrations (0–70 μM) of C-13 in PIPES buffer (25 mM, pH 6.8) for 15 min at 37 °C. Colchicine (10 μM) was then added to the reaction milieu and incubated at 37 °C for 45 min. Fluorescence spectra (410–500 nm) were recorded using an excitation wavelength of 340 nm in a cuvette of path length 0.3 cm using a spectrofluorometer (JASCO FP-6500, Tokyo, Japan). The experiment was done four times. The inhibition constant (Ki) was calculated using GraphPad software version 6.0 (GraphPad Software, CA, USA) by fitting the difference in fluorescence intensity using the following equation[89]where Ki is an inhibition constant, EC50 is the concentration of C-13 at which the fluorescence intensity was reduced to half, L is the concentration of C-13, and Kd is the dissociation constant of the binding of colchicine to tubulin.[44] A similar experiment was carried out with C-21.

Competition Assay with Compound 12

Compound 12 was reported to bind at the colchicine-binding site on tubulin.[44] Tubulin (2 μM) was incubated either without or with different concentrations (2–60 μM) of C-13 in 25 mM PIPES buffer, pH 6.8 for 20 min at 37 °C. Later, compound 12 (5 μM) was added into the reaction milieu and incubated for 10 min at room temperature. Fluorescence spectra (410–500 nm) were monitored using 350 nm as the excitation wavelength in a cuvette of path length 0.3 cm using a spectrofluorometer (JASCO FP-6500, Tokyo, Japan). The fluorescence intensities were corrected for the inner filter effect in the presence of C-13 and the inhibition of the binding of compound 12 to tubulin by different concentrations of C-13 was determined.

Molecular Docking Analysis of C-13

To find the putative binding site of C-13 on the tubulin dimer, molecular docking was performed using AutoDock 4.2[90] using the protocol as described earlier.[76] The three-dimensional atomic co-ordinates for C-13 were obtained using the PRODRG server.[91] Briefly, global docking for colchicine (PDB: 1SA0),[77] CA-4 (PDB: 5LYJ)[75] and C-13 was performed individually on the tubulin dimer using a grid box that covered the entire surface of the tubulin dimer. Because the maximum number of conformations for all these molecules was found to be at the interface of the tubulin dimer, local docking was performed at the tubulin interface only.

A grid box of 56 × 60 × 58 Å with a spacing of 0.375 Å was used to cover the tubulin dimer interface to perform local docking, and 50 independent docking jobs were carried out, each of 100 runs to obtain 5000 conformations. An RMSD cut off of 4 Å was used for clustering of the conformations and analysis of binding energies using AutoDock 4.2. The interaction with tubulin was analyzed using UCSF Chimera version 1.11.[92]

Effect of C-13 on MAP-Rich Tubulin Polymerization

MAP-rich tubulin was purified as described earlier.[84] MAP-rich tubulin (2 mg/mL) was incubated in the absence and presence of different concentrations (5, 10, 20, 50, 100, and 200 μM) of C-13 in PEM (25 mM PIPES pH 6.8, 3 mM MgCl2, 1 mM EGTA) buffer for 10 min on ice. Subsequently, GTP (1 mM) was added to the reaction mixture, and the kinetics of microtubule assembly was assessed at 37 °C by light scattering (at 350 nm) using a spectrofluorometer.

Effect of C-13 on Taxol-Induced Tubulin Polymerization

Tubulin (18 μM) was incubated without and with different concentrations (30, 60, 100, and 150 μM) of C-13 in PEM buffer on ice for 10 min. Subsequently, taxol and GTP were added to the reaction mixture to a final concentration of 10 μM taxol and 1 mM GTP. The kinetics of tubulin assembly was assessed by taking the absorbance of the reaction mixture at 350 nm, 37 °C using Spectramax M2e.

PI Staining

HeLa cells (2.5 × 104 cells/well) were seeded onto glass coverslips and incubated for 24 h. The cells were then treated either with vehicle (0.1% DMSO) or with 75 and 200 nM C-13 and 20 nM CA-4 for 24 h. The plates were centrifuged at 2500 rpm for 15 min and washed with PBS. Live/dead cell staining with an FITC-Annexin V apoptosis detection kit (BD Biosciences, San Jose, CA, USA) was carried out.[71] Cells were incubated with 5 μL of PI solution (50 μg/mL) and incubated for 15 min at room temperature in dark. Images were captured using an Eclipse TE 2000U microscope (Nikon, Tokyo, Japan) at 60× magnification and processed with Image-Pro Plus software (Media Cybernetics, Silver Spring, MD).

Confirmation of Apoptosis by PARP Cleavage

HeLa cells were treated with 75 and 200 nM C-13 and 20 nM CA-4 in T-25 flasks for 24 h. After 24 h, cells were scrapped and lysed with lysis buffer (Tris 20 mM, NaCl 200 mM, Triton X-100, 0.1%, DTT 1 mM at pH 7.2). The protein concentration was determined by Bradford’s assay, and 100–200 μg of the proteins was loaded onto the SDS-PAGE gel. The protein band was transferred onto the PVDF membrane via electroblotting. Immunoblotting was carried out using the anti-PARP monoclonal antibody and anti-β-actin antibody (loading control).[81]

Effect of C-13 on Generation of ROS

Production of intracellular ROS was quantified using DCFDA dye. HeLa cells (2 × 105 cells/mL) were seeded in 24-well cell culture plates and treated either with vehicle (0.1% DMSO) or with C-13 (75 and 200 nM) and incubated for 6 h. CA-4 (20 nM) was used for comparison, while H2O2 was used as a positive control. In a separate assay, HeLa cells were treated without and with 200 nM C-13 for 2 h and the amount of ROS generated was estimated after 2 h of C-13 treatment. The cells were collected by centrifugation at 2500 rpm for 10 min and washed twice with PBS. Counting of cells was carried out using trypan blue dye. Cells were then incubated with DCFDA dye (25 μM) in dark at 37 °C for 1 h. Fluorescence spectra (510–600 nm) were monitored using an excitation wavelength of 488 nm.[93] Fluorescence intensity per cell at 525 nm was calculated, and statistical significance was deduced using student’s t-test.

Scratch Wound Healing Assay

A549 cells were seeded on coverslips (1 × 105 cells or up to 90% confluency). After the attachment of the cells, a wound was created on the coverslip by a sterile 10 μL micropipette tip.[44] The coverslips were then treated with 0, 100, and 200 nM of C-13, and differential interference contrast images were taken at different time intervals (0, 6, and 10 h or till the wound healed) to see the effect of C-13 on wound closure. The percentage of wound healed was calculated using Image-Pro Plus software (Media Cybernetics, Silver Spring, MD).