Zerumbone, a cyclic sesquiterpene, exerts antimitotic activity in HeLa cells through tubulin binding and exhibits synergistic activity with vinblastine and paclitaxel

Abstract Objectives The aim of this study was to elucidate the antimitotic mechanism of zerumbone and to investigate its effect on the HeLa cells in combination with other mitotic blockers. Materials and methods HeLa cells and fluorescence microscopy were used to analyse the effect of zerumbone on cancer cell lines. Cellular internalization of zerumbone was investigated using FITC‐labelled zerumbone. The interaction of zerumbone with tubulin was characterized using fluorescence spectroscopy. The Chou and Talalay equation was used to calculate the combination index. Results Zerumbone selectively inhibited the proliferation of HeLa cells with an IC50 of 14.2 ± 0.5 μmol/L through enhanced cellular uptake compared to the normal cell line L929. It induced a strong mitotic block with cells exhibiting bipolar spindles at the IC50 and monopolar spindles at 30 μmol/L. Docking analysis indicated that tubulin is the principal target of zerumbone. In vitro studies indicated that it bound to goat brain tubulin with a Kd of 4 μmol/L and disrupted the assembly of tubulin into microtubules. Zerumbone and colchicine had partially overlapping binding site on tubulin. Zerumbone synergistically enhanced the anti‐proliferative activity of vinblastine and paclitaxel through augmented mitotic block. Conclusion Our data suggest that disruption of microtubule assembly dynamics is one of the mechanisms of the anti‐cancer activity of zerumbone and it can be used in combination therapy targeting cell division.

kinases such as the Aurora family of proteins; the Polo-like kinase; and the mitotic checkpoint proteins. 5,7 Microtubules, the key components of the cytoskeleton, are composed of alpha and beta tubulin heterodimer. They are highly dynamic polymers that undergo polymerization and depolymerization in a short span of time and play essential role in the maintenance of cell shape, intracellular trafficking, cell motility and cell signalling apart from cell division and mitosis. 8 Tubulin has two nucleotide (GTP) binding sites and three well-characterized drug binding sites such as the colchicine binding site, the paclitaxel binding site and the Vinca alkaloid binding site. The GTP binding site is located at the N-terminal region of the α and the β subunits, and the colchicine binding site is present at the interface of the α-β subunit. 8,9 The paclitaxel binding site is located at the β-tubulin, and the Vinca alkaloids binding site is located in the N-terminal region of the β-tubulin subunit close to the GTP binding site. 9 The clinically successful antitubulin agents such as the paclitaxel and the vinblastine are obtained from plants. Natural product research is gaining a huge attention because many of the phytochemicals exhibit excellent chemopreventive and chemotherapeutic potential in addition to their selectivity against cancer cells and low cost of production. 10 Natural products such as genistein, apigenin, quercetin, curcumin, berberine, limonene, coumarin, indirubin, brassinin, indole-3-carbinol, lycopene and resveratrol are in clinical/ preclinical trials either alone or in combination therapy for the treatment of cancer. [11][12][13] In the present study, we have investigated the anti-proliferative mechanism of the natural product zerumbone isolated from the plant Zingiber zerumbet belonging to the ginger family of flowering plants (Zingiberaceae). Zerumbone is a sesquiterpene and is reported to exhibit anti-cancer potential and other pharmacological activities such as anti-inflammatory, antibacterial, antimalarial and antioxidant properties. [14][15][16][17] Zerumbone was found to be effective in preventing tumour angiogenesis by inhibiting the VEGF expression and NF-κB activity. 18 It was reported to induce apoptosis in various cancer cell lines by modulating the FAS and TRAIL signalling pathways, through enhanced expression of TNF and modulating Bax/Bcl-2 ratio . [19][20][21] Recently, zerumbone was reported to block the cell cycle at mitosis 17,22 and induce apoptosis in cancer cells through inhibition of microtubule assembly. 23 However, the mechanism behind the mitotic block was not clearly established, and hence, we have performed this study to elaborate its anti-cancer mechanism through biophysical, biochemical and cell culture studies.
Cell culture studies showed an excellent observation that zerumbone exhibited selective toxicity against HeLa cells through enhanced internalization of the compound and inhibited their migration. The anti-proliferative effect of zerumbone in HeLa cells correlated well with its ability to inhibit the cell cycle at mitosis through tubulin binding. Zerumbone bound to tubulin at the colchicine binding site and inhibited the polymerization of tubulin into microtubules. Zerumbone exhibited excellent synergistic antimitotic and anti-proliferative activity in HeLa cells when combined with clinically established drugs such as paclitaxel and vinblastine. Together, the results suggest that the anti-proliferative effects of zerumbone could be partly through its inhibitory effects on tubulin and induction of mitotic block, and combination of zerumbone and other anti-cancer drugs might provide a therapeutic advantage in controlling the growth of cancer cells.

| Materials
Paclitaxel, vinblastine sulphate, podophyllotoxin, colchicine, 5,5′-  Calicut. Zerumbone was extracted and isolated from the rhizomes of Zingiber zerumbet as described earlier. 24 Briefly, 1 kg of fresh rhizomes was washed under running tap water and cut into slices. The slices were then shade-dried at 37°C for 5 days. The dried samples were then soaked in methanol for 3 days, and the methanolic extract was concentrated by using the rotary evaporator (Heidolph Instruments, GmbH & CO. KG, Schwabach, Germany). The extract was then fractionated by silica gel (mesh size 200) column chromatography using organic solution mixture of hexane: ethyl acetate (8:2; v/v). Zerumbone thus obtained was further purified by crystallization. The purity of zerumbone was analysed and confirmed using the Shimadzu liquid chromatography-mass spectrometry (LC-MS) and Bruker Avance III Nuclear Magnetic Resonance spectroscopy (NMR) using the standard procedure.

| Fluorescent labelling of zerumbone
Zerumbone does not have any characteristic fluorescence; hence, we labelled it with fluorescein isothiocyanate (FITC) by conjugating zerumbone oxime with fluoresceinthiocarbamyl ethylenediamine (EDF) to characterize the binding site of zerumbone on tubulin.
Zerumbone oxime was synthesized as described earlier. 25 In brief, zerumbone (0.3 g) was dissolved in 10 mL of ethanol containing 0.9 g of hydroxylamine hydrochloride and 1.8 g potassium carbonate. The mixture was stirred for 5 hours at room temperature. The reaction mixture was then filtered, and the residue was washed with methanol.
The filtrate was concentrated under reduced pressure and was then mixed with dichloromethane (10 mL). The organic layer was collected and washed with water. The resultant mixture was concentrated and dried to get crystalline zerumbone oxime, which was subjected to FTIR analysis. Fluoresceinthiocarbamyl ethylenediamine (EDF) was synthesized as described earlier. 26 Zerumbone oxime (20 mg),

| Calculating the percentage of apoptotic cell death using AO staining
HeLa cells (0.5 × 10 5 cells/mL) grown on poly-l-lysine-coated glass coverslips (12 mm) in 24-well tissue culture plates were treated with either 0.1% DMSO or different concentrations of zerumbone (10,20 and 30 μmol/L) for 24 hours. The live cells were immediately viewed under an inverted Nikon ECLIPSE Ti (Tokyo, Japan) fluorescent microscope after adding AO (2 μg/mL), and the images were captured using the CoolSNAP digital camera.

| Cell migration assay
HeLa cells (1 × 10 6 cells/mL) were grown in minimum essential medium supplemented with 10% FBS in 35 mm cell culture dishes. At 90% confluence, a wound was made using a sterile micropipette tip. 30,31 The floating cells were removed immediately after wounding, and the media were changed with fresh one containing different concentrations of zerumbone (0, 5, 10 and 15 μmol/L). Cells were observed at 24, 48 and 72 hours of intervals, and the bright-field images of the wound closure were recorded using the Nikon ECLIPSE Ti inverted microscope. Percentage wound healing was calculated by using the formula: anti-mouse IgG conjugated to Alexa Fluor 568. The DNA was stained with Hoechst 33342 to visualize the DNA. Gamma tubulin staining was performed using rabbit monoclonal anti-gamma tubulin antibody at 1:1000 dilutions as described earlier. 27,32,33 Immunofluorescence images were acquired using the CoolSNAP digital camera and were processed by using ImageJ (NIH, USA).

| Molecular docking study
The interaction of zerumbone with tubulin dimer and other cell divi- Then, they were energy-minimized until the average RMSD of the non-

| Purification of tubulin
Goat brain tubulin was isolated by two cycles of polymerization and depolymerization in the presence of glutamate as described earlier. 34, 35 Bradford assay was used to estimate the tubulin concentration using bovine serum albumin as the standard. 36 The protein was stored in aliquots at −80°C until further use. All the experiments with tubulin were performed in PEM buffer (25 mmol/L PIPES, 1 mmol/L EGTA, 3 mmol/L MgCl 2 , pH 6.8).

| Spectral measurements
All the absorbance measurements were carried out in Systronics  where L f is the concentration of free zerumbone. 30,39 The experiment was repeated three times.

| Sedimentation assay
The in vitro microtubule sedimentation assay was performed to detect the effect of zerumbone on the polymerization of tubulin.
Different concentrations of zerumbone were incubated with tubulin (12 μmol/L) in PEM buffer containing 0.8 mol/L glutamate and 1 mmol/L GTP at 37°C for 1 hour. The reaction mixture was then subjected to centrifugation at 50 000 × g for 1 hour. The supernatant and pellet were collected separately, and the protein concentration in the supernatant was measured using Bradford assay. 30

| Light scattering assay
The effect of zerumbone on the assembly of microtubule was also analysed by monitoring the kinetics of tubulin polymerization.
Different concentrations of zerumbone were added to 12 μmol/L tubulin in the polymerization buffer containing 25 mmol/L PIPES, 1 mmol/L EGTA, 3 mmol/L MgCl2 and 0.8 mol/L glutamate. The assembly reaction was initiated by adding 1 mmol/L GTP and incubated at 37°C. 38 The polymerization of tubulin was monitored by light scattering at 550 nm for 15 minutes using JASCO FP-8300 spectrofluorometer (Tokyo, Japan) connected with circulating water bath maintained at 37°C.

| Binding site competition assay
Colchicine has a very weak fluorescence in aqueous buffers but exhibits a strong fluorescence after binding to tubulin. 40

| Determination of combination index
HeLa cells were incubated with zerumbone or vinblastine or pacli- where D m is the median dose, f a is the fraction affected, and f u is the fraction unaffected (f u = 1 − f a ). The median dose (D m ) was calculated as described earlier. 42 A CI < 1 indicates synergism, CI = 1 shows additivity, and CI > 1 specifies antagonism. HeLa cells grown on coverslips in 24-well tissue culture plate were treated with zerumbone in combination with paclitaxel or vinblastine and processed to visualize microtubules and DNA.

| Isolation and characterization
The methanolic extract of 1 kg of fresh Zingiber zerumbet extract

| Internalization of fluorozerumbone by HeLa and L929 cells
The fluorozerumbone internalized by HeLa and L929 cells treated with 40 µmol/L fluorozerumbone for 4 hours was extracted using methanol and quantified based on the absorption spectra of the standard fluorozerumbone. We found that the uptake of fluorozerumbone by HeLa cells was 26 nmol/cell and that by L929 cells was 14.8 nmol/cell. The methanolic extracts were further subjected to fluorometric analysis by exciting them at 494 nm as explained in Materials and methods. As shown in Figure 2C, the fluorescence intensity of the extract obtained from HeLa cells was 16% higher than that of the extract obtained from L929 cells. These results indicate that the cellular uptake of fluorozerumbone is much higher in tumour cells.

| Zerumbone-induced apoptosis in HeLa cells
Acridine orange staining is a common method used to detect apoptotic cell death. After 24 hours, the control cells remained viable and healthy and the zerumbone-treated cells displayed brightly stained hypercondensed nucleus and membrane blebbing, which indicated the characteristic of apoptosis ( Figure 3A).

| Zerumbone inhibited the migration of HeLa cells in a concentration-dependent manner
Wound healing assay was used to check the migration of HeLa cells upon treatment with different concentrations of zerumbone.
As shown in Figure 4A, zerumbone effectively inhibited the  Figure 4B). The results suggest that zerumbone can effectively prevent the migration of cancer cells in a concentration-and time-dependent manner even at concentrations lower than the IC 50 . to be 3%, 7% and 27%, respectively, and the MI, which is the ratio of the total number of the mitotic cells to the total cells, was found to be 15%, 20% and 34%, respectively. Under similar conditions, the MI of the control cells was 3.5% ( Figure 5D).

| Probing the possible targets of zerumbone through computational docking analysis
Since zerumbone treatment produced cells with mitotic abnormalities, we investigated the interaction of the zerumbone with cell division proteins such as tubulin, Eg5, Aurora A, Plk1,   Figure 6B-F. Based on the Glide docking score (Table 1) and MM-GBSA scoring, it is possible to speculate that Eg5 and Aurora kinase A could also be the potential target for zerumbone in addition to tubulin.

| Binding of zerumbone to tubulin
Results from the cell culture studies and docking analysis indicated that tubulin could be one of the primary targets for Zerumbone.
Hence, binding of zerumbone on tubulin was analysed using spectrofluorometer by measuring the intrinsic tryptophan fluorescence of tubulin. Zerumbone quenched the intrinsic fluorescence of tubulin in a concentration manner ( Figure 7A).

| Zerumbone inhibited the polymerization of tubulin in vitro
The effect of zerumbone on tubulin assembly was analysed by using the sedimentation assay and the light scattering   Figure 9C).

| Zerumbone inhibited the proliferation of HeLa cells synergistically in combination with vinblastine and paclitaxel
Vinblastine and paclitaxel are FDA-approved clinically used drugs for the treatment of various types of tumours. 48  Zerumbone synergistically increased the MI in combination with vinblastine and paclitaxel. As shown in Figure 11A, vinblastine when used alone induced a mitotic block of 9.4%; however, when combined with zerumbone 5, 10 and 12 µmol/L, the mitotic block was increased to 20%, 25% and 31%, respectively. Similarly, vinblastine 1.2 nmol/L when used alone induced a mitotic block of 19%, and when combined with zerumbone 5, 10 and 12 µmol/L, the mitotic block was found to be increased 28%, 38% and 46%, respectively. Zerumbone induced a significant hike in the mitotic cells when combined with paclitaxel, similar to its synergistic activity with vinblastine. When 5 nmol/L paclitaxel was combined with 10 and 12 µmol/L zerumbone, the MI was found to be 32% and 37%, respectively ( Figure 11B), and when 10 nmol/L paclitaxel was combined with 10 and 12 µmol/L zerumbone, the MI was increased to 54% and 60%, respectively, while paclitaxel alone at 5 and 10 nmol/L induced 15% and 31% mitotic block ( Figure 11B). In addition to the enhanced mitotic arrest, the combined addition of two drugs induced drastic mitotic abnormalities in the organization of the mitotic spindle and alignment of chromosomes ( Figures   11C,D). The significance of plants as a major source of anti-cancer agents can be understood from the fact that most of the presently used chemotherapeutic agents are derived from natural sources in one way or the other. 51 In this study, we have found that the potential mechanism behind the anti-cancer activity of zerumbone is through its inhibitory activity on tubulin polymerization and mitotic arrest.

| D ISCUSS I ON
Results from LC-MS and NMR were in conformity with the molecular weight (218.34) and the structure of the compound. The data were in agreement with the earlier published reports. 25 Our  be used as a relatively less toxic, safe and effective chemotherapeutic agent. Curcumin, the dihydroxy polyphenol from Curcuma longa, was also reported to induce selective toxicity in cancer cells due to preferential uptake by the cancer cells compared to normal cells. 29 AO staining was used to analyse whether the cytotoxic effect of zerumbone was due to necrotic cell death or apoptosis.
Apoptotic cells will appear brightly stained, hypercondensed and often fragmented chromatin in spherical or irregular shapes under fluorescent microscope. 52 It was clearly evident that by the end of one cell cycle, most of the zerumbone-treated cells underwent apoptosis since their characteristics were similar to those of the apoptotic cells. 52 Molecules preventing metastasis are highly valuable in cancer chemotherapy as they can prevent the cancer cells spreading to other tissues. Results from the cell migration assays indicate that zerumbone strongly inhibited the migration of cancer cells at 5 and 10 μmol/L, which are much lower concentrations than its IC 50 . This result is in agreement with the previous report suggesting the anti-metastatic property of zerumbone. 53 Microtubules play a very important role in cell migration, 54 and most of the potent tubulin-targeted drugs inhibit the migration of the cell at concentrations lower than their IC 50 . 55,56 Since zerumbone showed excellent mitotic block and inhibited the migration of cancer cells, we analysed the effect of zerumbone on interphase and mitotic microtubules using immunofluorescence microscopy. In our study, we observed that zerumbone at IC 50  The mechanism of action of zerumbone was similar to that of the other clinically used chemotherapeutic drugs such as vinblastine and paclitaxel, which bind to tubulin and induce mitotic block. In addition to tubulin, zerumbone might have another target involved in centrosome separation that gets inhibited only at higher concentrations.
Our finding that internalization of zerumbone is higher in cancer cells leading to it preferential killing and that it is highly effective in preventing the migration of cancer cells will be more valuable in F I G U R E 11 Zerumbone potentiated the mitotic block in HeLa cells in combination with vinblastine and paclitaxel. Zerumbone increased the number of mitotic cells in HeLa cells in combination with vinblastine (A) and paclitaxel (B). All the experiments were performed three times. The data represents mean ± SD. Effect of zerumbone on the spindle microtubule and chromosome alignment in combination with vinblastine (C) and paclitaxel (D). HeLa cells treated with indicated concentrations of zerumbone, vinblastine and paclitaxel were fixed and processed for immunofluorescence microscopy as described in Section 2. The scale bar represents 10 μm cancer therapy either alone or in combination with other established chemotherapeutic drugs.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflict of interests.