Natural product derivative Bis(4-fluorobenzyl)trisulfide...

3318

Published Online First on December 8, 2009 as 10.1158/1535-7163.MCT-09-0548Published OnlineFirst December 8, 2009; DOI: 10.1158/1535-7163.MCT-09-0548

Natural product derivative Bis(4-fluorobenzyl)trisulfideinhibits tumor growth by modification of β-tubulin atCys 12 and suppression of microtubule dynamics

Wanhong Xu,1 Biao Xi,1,3 Jieying Wu,1 Haoyun An,3

Jenny Zhu,3 Yama Abassi,3 Stuart C. Feinstein,4

Michelle Gaylord,4 Baoqin Geng,2 Huifang Yan,5

Weimin Fan,6 Meihua Sui,6 Xiaobo Wang,3

and Xiao Xu1,3

1Hangzhou High Throughput Drug Screening Center, ACEA Bioand 2Department of Pharmacology, School of Medicine, ZhejiangUniversity, Hangzhou, Zhejiang, P.R. China; 3Department ofBiology, ACEA Biosciences, San Diego, California;4Neuroscience Institute, University of California Santa Barbara,Santa Barbara, California; 5Department of Pharmacology,Shanghai Institutes of Pharmaceutical Industry, Shanghai,P.R. China; and 6Department of Pathology, South CarolinaMedical University, Charleston, South Carolina

AbstractBis(4-fluorobenzyl)trisulfide (BFBTS) is a synthetic mole-cule derived from a bioactive natural product, dibenzyltri-sulfide, found in a subtropical shrub, Petiveria allieacea.BFBTS has potent anticancer activities to a broad spec-trum of tumor cell lines with IC50 values from high nano-molar to low micromolar and showed equal anticancerpotency between tumor cell lines overexpressing multi-drug-resistant gene, MDR1 (MCF7/adr line and KBv200line), and their parental MCF7 line and KB lines. BFBTS in-hibited microtubule polymerization dynamics in MCF7cells, at a low nanomolar concentration of 54 nmol/L,while disrupting microtubule filaments in cells at low mi-cromolar concentration of 1 μmol/L. Tumor cells treatedwith BFBTS were arrested at G2-M phase, conceivably re-sulting from BFBTS-mediated antimicrotubule activities.Mass spectrometry studies revealed that BFBTS bound and

eceived 6/22/09; revised 10/23/09; accepted 10/25/09; publishednlineFirst 12/8/09.

rant Support: Hangzhou Municipal Grant for new drug discovery andevelopment and Zhejiang Technology and Science Department Grant forew drug discovery and development.

he costs of publication of this article were defrayed in part by theayment of page charges. This article must therefore be hereby markeddvertisement in accordance with 18 U.S.C. Section 1734 solely todicate this fact.

ote: Supplementary material for this article is available at Molecularancer Therapeutics Online (http://mct.aacrjournals.org/).

. Xu and B. Xi contributed equally to this work.

equests for reprints: Xiao Xu, ACEA Biosciences, Inc., 6779 Mesa Ridgeoad, Suite 100, San Diego, CA 92121. Phone: 858-724-0928, ext.006; Fax: 858-724-0927. E-mail: [email protected]

opyright © 2009 American Association for Cancer Research.

oi:10.1158/1535-7163.MCT-09-0548

RO

Gdn

Tpain

NC

W

RR3

C

d

on May 16, 2018.mct.aacrjournals.org Downloaded from

modified β-tubulin at residue Cys12, forming β-tubulin-SS-fluorobenzyl. The binding site differs from known antimicro-tubule agents, suggesting that BFBTS functions as a novelantimicrotubule agent. BFBTS at a dose of 25 mg/kg inhib-ited tumor growth with relative tumor growth rates of19.91%, 18.5%, and 23.42% in A549 lung cancer, Bcap-37 breast cancer, and SKOV3 ovarian cancer xenografts,respectively. Notably, BFBTS was more potent againstMDR1-overexpressing MCF7/adr breast cancer xenograftswith a relative tumor growth rate of 12.3% than paclitaxelwith a rate of 43.0%. BFBTS displays a novel antimicro-tubule agent with potentials for cancer therapeutics. [MolCancer Ther 2009;8(12):3318–30]

IntroductionMicrotubules are dynamic polymers of α-tubulin and β-tu-bulin heterodimers, arranged in the form of slender fila-ments whose polymerization dynamics are tightlyregulated both spatially and temporally. Microtubules areinvolved in a variety of cellular processes, making theman important target for anticancer drugs (1). In interphasecells, microtubules exchange their tubulin with solubletubulin in the cytoplasmic pool with half times of ∼3minutes to several hours (2, 3). With the onset of mitosis,the interphase microtubule network disassembles and isreplaced by a population of highly dynamic microtubules,forming mitotic spindle, driving the intricate movement ofchromosomes and separation of sister chromatids. Mitoticspindle microtubules with a short tubulin half-life of ∼15seconds are 20 to 50 times more dynamic than interphasemicrotubule (3–5). Therefore, the dynamics of mitoticspindle microtubules are especially sensitive to modulationby microtubule-binding proteins and to microtubule-activedrugs (6). Chemically diversified group of microtubule-targeted drugs can alter microtubule polymerizationand dynamics in a wide variety of ways (1) and are classi-fied based on their distinct tubulin-binding domains, in-cluding Vinca domain, colchicine domain, taxane domain,and other domains distinct from the above three domains(1, 7)Bis(4-fluorobenzyl)trisulfide (BFBTS) is a synthetic ana-

logue of an organosulfur compound, dibenzyltrisulfide(DBTS; ref. 8). DBTS is a bioactive natural compound foundin a subtropical shrub, Petiveria Allieacea, which had beenused as a folk medicine for antitumor and antirheumatic ef-fects in Central and South America since the Aztecs (9). Pri-mary bioactive compounds in the essential oil of the roots ofthis shrub include benzyldehyde, dibenzydisulfide, DBTS,cis- and trans-stilbene, and 2[(phenylmethyl)dithio]-ethanol.

Mol Cancer Ther 2009;8(12). December 2009

© 2009 American Association for Cancer Research.

http://mct.aacrjournals.org/

Molecular Cancer Therapeutics 3319

Published OnlineFirst December 8, 2009; DOI: 10.1158/1535-7163.MCT-09-0548

DBTS is the principle bioactive compound in the lipophilicextract of plant roots (10, 11) and its reported biological ac-tivities include enhancement of phagocytosis in treated mice(12), in vitro antifungal action mainly mediated by itsderivative, methyl benzyl sulphonic anhydric (13), inhibi-tion of neurite outgrowth and proliferation in neuroblasto-ma cell line, SH-SY5Y (9, 11, 14), and induction ofmonocyte-like differentiation of promyelocytic acute leuke-mia cell line, HL-60 (15). The molecular mechanisms under-lying the biological activities of DBTS remain largelyunknown, although it was reported that DBTS induceddisassembly of microtubules and decreased the total expres-sion of tubulin in SH-SY5Y cells, indicating that the antipro-liferative activity of DBTS may result from its potentialantimitotic effect (9).In our previous studies, we reported that BFBTS was

more cytotoxic than DBTS against nine human tumor celllines (8). To further explore the anticancer potency and themechanism of action, we compared BFBTS to paclitaxel andother antimitotic agents, and found that BFBTS is cytotoxicto a broad range of cancer cell lines including drug-resistantcell lines that overexpress the drug-resistant gene, MDR1.The anticancer activities may result from covalent modifica-tion of β-tubulin at a unique site of Cys 12, culminating inmicrotubule inhibition and leading to suppression of micro-tubule dynamics. BFBTS inhibited tumor growth in fourhuman cancer xenografts including one MDR1-expressingbreast cancer xenografts.

Materials and MethodsChemicals

BFBTS was synthesized using the method reported previ-ously (8). Other chemicals used in the study were purchasedfrom Sigma. For all in vitro experiments, BFBTS and pacli-taxel were dissolved in DMSO, and for all in vivo animalexperiments, BFBTS and paclitaxel were dissolved incremaphor/ethanol (w:v 4:6). BFBTS-injectable solution incremaphor/ethanol (w:v 4:6) were prepared using the pro-cess reported previously (16, 17).Cell Lines and Reagents

MCF7, MCF7/adr, KB, and KBv200 cell lines were fromDr. Fan's laboratory in South Carolina Medical University,Charleston, SC. Bcap-37 cell line was from Dr. Yan's labora-tory in Shanghai Institute of Pharmaceutical Industry,Shanghai, P.R. China. MCF7 cells stably expressing greenfluorescent protein (GFP) -α-tubulin were from Dr. Feinsteinof University of California Santa Barbara, Santa Barbara,CA (18). All other cell lines used were purchased fromAmerican Type Culture Collection. Cell culture reagentswere purchased from Hyclone. All cancer cells were culti-vated in 10% fetal bovine serum DMEM media at 37°C,5% CO2.Cell Proliferation, Viability, and Apoptosis Analysis

Label-free real-time detection technology developed byACEA Biosciences (19, 20) was used for continuously mon-itoring cell proliferation and viability after BFBTS addition.



The system xCELLigence RTCA system (Roche AppliedSciences) was operated based on the instruction in the usermanual. In brief, cells were seeded into wells of 96-well mi-crotiter E-plate (Roch Applied Sciences) at cell concentrationof 5,000 cells per well, cultured in CO2 incubator at 37°C, 5%CO2, and continuously monitored on xCELLigence SP sys-tem. At about 20 h after cell seeding, the cells in exponentialgrowth phase were treated with different concentrations ofBFBTS. The cell response to BFBTS treatment was continu-ously monitored by xCELLigence SP system for a period of48 h.Cell apoptosis and death detection was measured by Cell

Death Detection ELISA kit (Roche Applied Sciences),according to assay protocol for the kit. Briefly, cells wereseeded into the 96-well plate. Then cells were treated for48 h with BFBTS at 1 μmol/L, a concentration close to theIC50 value of cell proliferation. Cells were harvested andlysed. Twenty microliters of lysate were removed and trans-ferred to streptavidin-coated microplate and then incubatedwith anti–histone-biotin and anti–DNA-POD antibodies for2 h followed by adding 2,2′-Azinobis-3-ethylbenzthiazoline-sulphonic-acid substrate for color development. The platewasmeasured at an absorbance of 405 and 490 nm in BeckmanMultimode DTX880 plate reader.Phosphorylated Histone H3 Staining, Mitotic Index,

and Flow Cytometry

Cultured A549, MCF7, and MCF7/adr cells were treatedwith 2 μmol/L BFBTS or 30 nmol/L paclitaxel for 24 h (thisconcentration are close to compound IC90 value in cell pro-liferation), fixed with ice-cold methanol for 3 min, permea-bilized with PBS/0.25% TX-100 for 10 min, followed byblocking in PBS/1% bovine serum albumin/0.1% TX-100for 30 min at room temperature. The cells were stained withanti–phosphorylated histone H3 (Ser-10) polyclonal anti-body (Upstate) followed by Rhodamine-labeled anti–rabbitIgG (Chemicon). Mitotic index was determined by calculat-ing the ratio of the number of phosphorylated histoneH3–positive cells versus total 4′,6-diamidino-2-phenylindole–positive cells. For flow cytometry (fluorescence-activated cellsorting) analysis, A549 cell were cultured in Petri dishes untilreaching ∼40% confluence. Cells were treated with BFBTS at2 μmol/L for 12 or 24 h, were harvested, and then followedby fixation in 100% ethanol and fluorescence-activated cellsorting analysis.Tubulin Polymerization Assay

Tubulin and the fluorescence-based tubulin polymeriza-tion assay kit, and the tubulin polymerization assay wasdone based on the product instruction (Cytoskeleton).Briefly, 2 mg/mL of tubulin were mixed with BFBTS (atdifferent final concentrations: 30, 300, or 3 μmol/L), pacli-taxel (3 μmol/L), vincristine (1.5 μmol/L), or DMSO at fi-nal concentration of 0.1%, which served as the solventcontrol. The polymerization reaction was done at 37°Cin 20% glycerol, 80 mmol/L PIPES, 2.0 mmol/L MgCl2,0.5 mmol/L EGTA (pH 6.9), and 1 mmol/L GTP, andwas monitored every minute for 30 min in a temperature-controlled plate reader (BMS) at excitation and emission of360 and 420 nm, respectively.



BFBTS Inhibits Microtubule Dynamics3320



on May 16, 2018. © 2009 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

l

-

.

-

r-,

.

r-

Figure 1. Inhibition of tumor celgrowth and induction of apoptosisby BFBTS. A, chemical structuresof DBTS (A) and BFBTS (B). B, tumor cell growth inhibition dynamicsof 10 tumor cell lines by BFBTSLive cell response to the exposureof BFBTS was monitored in realtime on xCELLigence system. Theconcentrations of BFBTS used fothe treatment of PC3, A549, SHSY5Y, Bcap-37, Panc-1, MV522NCI-69AR, SKOV3, HT29, andMKN45 lines were indicated. 0.1%DMSO serves as solvent controlEach curve represented an averageof normalized CI from continuousmeasurement of duplicate wells ove48 h after compound addition (arrow). The representative curvesshown were from three repeatedexperiments.


Figure 1 Continued. C, inductionof apoptosis by BFBTS in A549,HT29, and SH-SY5Y cell lines at theconcentration 1 μmol/L, close to IC50

values. The mean apoptosis value ±SD for BFBTS-treated A549, HT29,and SHSY5Y cells were 1.37 ±0.23, 0.52 ± 0.05, and 3.54 ±0.65, respectively, and 0.15 ±0.0.001, 0.10 ± 0.01, and 0.05 ±0.0.65, respectively, for DMSO-trea-ted A549, HT29, and SHSY5Ycells. The data shown was the rep-resentative data of two repeated ex-periments. *, P < 0.5; **, P <0.01. D, inhibitory effect of BFBTSand paclitaxel on MDR1–highly ex-pressing cell lines, MCF7/adr andKBv200, in comparison with the pa-rental cell lines, MCF7 and KB. IC50

values at 24 and 48 h were indicat-ed. Arrows, the compound addition.The representative data shown werethe average of three triplicate samplewells. Concentrations of BFBTSused for all four lines were indicatedand 0.1% DMSO serves as solventcontrol.



Mass Spectrometry

Purified tubulin was incubated in the presence or ab-sence of BFBTS at 37°C for 30 min. BFBTS-treated tubulinwas digested with trypsin and the peptides were analyzed



by liquid chromatography tandem mass spectrometryusing reverse phase chromatography generated in Biomo-lecular Mass Spectrometry Facility, Department of Chemis-try and Biochemistry, University of California, San Diego,





California, using a pressure bomb. The microcapillary waspacked with Zorbax SB-C18 reversed phase material(Agilent). The peptides were eluted over a 60-min periodwith a linear gradient 5% to 80% of solvent B going fromsolvent A (5% acetonitrile, 0.2% formic acid) to solvent B(95% acetonitrile, 0.2% formic acid), with a flow rate of0.15/min. The tryptic peptide samples were separatedand analyzed with a LTQ ion trap mass spectrometer(ThermoFinnigan). Tandem mass spectrometry data ob-tained were analyzed by using SEQUEST program (ThermoScientific).Microtubule Staining and Microtubule Dynamics

Analysis

For microtubule staining assay, MCF7 cells were firstseeded in the 1 cm2 cover glass slide and then placed in tis-sue culture dishes for cultivation. Cells were then treatedwith BFBTS, paclitaxel, and vinblastine at the concentrationclose to their IC50 value in cytotoxicity assay. DMSO (0.1%)was served as solvent control. After 4 h of treatment,cells were fixed, followed by immunostaining with fluores-cence-conjugated anti-tubuline antibody. The intracellularmicrotubule networks were visualized under a fluorescentmicroscope.The microtubule dynamics analysis was done based on

the method described previously (18). Briefly, MCF7 cellsstably expressing GFP-α-tubulin were seeded onto thecover glass slides that had been first coated with poly-L-lysine (0.4 mg/mL; Sigma) for 2 h, and then with lami-nin and fibronectin (Sigma) at concentration of 10 and20 μg/mL, respectively, for addition 2 h at 37°C. Inaddition, cells were incubated with BFBTS at 54, 72, and108 nmol/L for 6 h, an incubation duration that allowedattainment of an equilibrium drug concentration in cells.Microtubules were observed using a Nikon Eclipse E800fluorescence microscope with a 1.4 numerical aperature×100 objective lens. Thirty images of each cell were ac-quired at 4-s intervals using a Hamamatsu Orca II digitalcamera driven by Metamorph software (Universal Imag-ing, Media).Xenograft Studies

For A549, non–small cell lung carcinoma, Bcap-37 breastcancer, and SKOV3 ovarian cancer xengograft models, mice(nu/nu) purchased from the Animal Institutes, ChineseAcademy of Science were implanted s.c. in the right flankwith the 4 × 106 cells suspended in 200 μL media. To preparetumor cells for the implantation, tumor cells harvested inculture flasks were first implanted in nu/nu mice. Afterthe tumor formation, tumor cell suspensions were preparedfrom the tumor tissue, and reimplanted in nu/nu mice foradditional two times before they are used for in vivo experi-ments. The tumor formation duration and variation weremuch more controllable using the passaged tumor cells thanthe tumor cells directly from culture flasks. The implantedtumors were allowed to grow for 7 d and then the tumor-bearing mice with the tumor sizes around 30 mm3 wererandomly separated into different test groups (n = 6 foreach group). The experiment in MCF7/adr breast cancerxenograft model was carried out in Dr. Fang's laboratory


in South Carolina Medical University. Mice (female nu/nu)were purchased from Charles River Laboratories. In theMCF7/adr model, the mice did not receive the treatmentof additional estrogen. The tumor cell implantation proto-col was same as above and for the MCF7/adr experiment.Four mice (n = 4) were used for each test group and con-trol group. BFBTS was administrated i.v. with dose regi-mens of 6.25, 12.5, and 25 mg/kg, respectively. The doseregimens were determined based on the maximum toler-ant dose of BFBTS, which was 100 mg/kg (data notshown). Tumors and mouse body weights were measuredevery 5 d for efficacy and toxicity. Tumor growth wasmonitored using external measurements with a caliperand tumor volumes were calculated using the formula(width2 × length)/2, where the width represents the smal-ler tumor diameter. For the MCF7/adr xneografts, thetumors were dissected and weighted after termination ofthe experiment. The relative tumor volume (RTV) wasused for evaluating in vivo efficacy of BFBTS, calculatedaccording to relative tumor volume = Vt/Vo, where Vo isthe tumor volume before BFBTS treatment. Vt is the tumorvolume at a given experiment time point. The relative tu-mor growth rate (T/C %) is TRTV (treated group)/CRTV

(control group) × 100%. Statistical ANOVA was used forstatistic analysis.

ResultsInhibition of Cancer Cell Proliferation and Induction of

Apoptosis by BFBTS

BFBTS (Fig. 1A, compound b) was derivatized by attach-ing fluoro-atoms at the para-position of the two benzenerings of DBTS (Fig. 1A, compound a). This modification re-sulted in improved in vitro cytotoxicity activity comparedwith DBTS in at least nine tumor cell lines of different ori-gins (8). In these cytotoxicity assays, we also found BFBTSto be the most potent compound among 16 different DBTSderivatives tested (8). In this study, we further explore theantitumor activity of BFBTS by monitoring cell response toBFBTS in additional 10 tumor cell lines, using a real-timecell analysis system (xCELLigence). Unlike the conventional

Table 1. IC50 values of BFBTS in 10 deferent types of tumor celllines after 48 h of treatment

Cell line (tumor type)

Mol Cancer Ther 20

© 2009 American Association for C

IC50 value (μmol/L) ± SD

PC3 (prostate cancer)
0.360 ± 0.117 Bcap-37 (breast cancer) 0.394 ± 0.321 NCI-69AR (SCLC) 0.415 ± 0.067 SKOV3 (ovarian cancer) 0.441 ± 0.012 SH-SY5Y (neuroblastoma) 0.652 ± 0.077 A549 (NSCLC) 0.768 ± 0.259 MV522 (NSCLC) 1.295 ± 0.279 HT29 (colon cancer) 1.678 ± 0.070 Panc-1 (pancreatic cancer) 2.350 ± 1.202 MNK45 (gastric cancer) 113.630 ± 16.9
Abbreviations: SCLC, small cell lung carcinoma; NSCLC, non–small cell lungcarcinoma.

09;8(12). December 2009

ancer Research.




end point cell viability assays, the real-time cell detectionsystem continuously monitors live cells in response to theexposure of BFBTS and provides kinetic cell response infor-mation from the same cell population. Dose- and time-dependent cell responses to BFBTS were clearly shown in10 tumor cell lines (Fig. 1B). The IC50 values derived fromnormalized cell index (CI) at 48 hours after BFBTS treatmentin 10 tumor cell lines are listed in Table 1, and ranged fromhigh nanomolar concentration (∼400 nmol/L for Bcap-37line) to high micromolar concentration (∼113 μmol/L forMNK line). The normalized CI is an arbitrary unit derivedfrom electrical impedance measurement (19–22). The corre-



lation between CI and cell viability is well documented inthe literature (23–26). Furthermore, we also tested cell deathand apoptosis in three selected tumor lines, A549, HT29,and SH-SY5Y, treated with 1 μmol/L BFBTS, the concen-tration approximating the IC50 value for cell proliferationand cytotoxicity (Table 1). BFBTS induced significant apo-ptosis in all three tumor lines compared with untreated cells(Fig. 1C).One major mechanism of acquired drug resistance is the

overexpression of efflux pumps, i.e., the p-gp170/MDR(21, 27–31), and many clinically used drugs are good sub-strate of the drug efflux pump (21, 27). To test the effect of

Figure 2. Induction of mitotic arrest by BFBTS. A, A549 cell growth inhibition patterns induced by BFBTS, paclitaxel, vinblastine, etopside, indirubin-3′-oxime, and latruculin. The dynamic cell growth inhibition patterns were represented as normalized CIs (Y axis) over time of compound exposure (X axis).The concentration for each compound used in the experiment was around the IC50 value obtained from the RTCA assay in the xCELLigence system. Eachcurve represents an average of triplicate sample wells under a given concentration as indicated. B, increase in phosphorylated histone H3–positive nuclearstaining in 2 μmol/L BFBTS-treated and 30 nmol/L paclitaxel-treated A549, MCF7, and MCF7/adr cells. Red, the specific phosphorylated histone H3; blue,the background nuclear DNA staining. Bar, 20 μm. C, the mitotic indexes in BFBTS- or paclitaxel-treated cells, and control. **, P < 0.01 in comparisonwith the control group. Cells in 10 randomly selected fields and total 1,000 cells were counted to obtain the mitotic indexes. D, A549 cell cycle arrest atG2-M phase by 2 μmol/L BFBTS. Control, cells treated with DMSO; 12 h, cells treated with BFBTS for 12 h; and 24 h, cells treated with BFBTS for 24 h.**, P < 0.01.





MDR expression on the sensitivity of tumor cells to BFBTS,we used MCF7/adr and human epidermoid carcinomacell line KBv200, where the overexpression of MDR1 is in-duced by exposure to low dose of vinblastine or daunoru-bincin (28–31). The MCF7/adr and KBv200 as well as theirparental lines (MCF7 and KB) were treated with differentconcentrations of BFBTS and paclitaxel, and the cell re-sponses were continuously monitored in real-time for 48hours after treatment. As shown in Fig. 1D, the MDR1-over-expressing cell lines and their parental cell lines were equal-ly sensitive to BFBTS, with IC50 values (48 hours after


treatment) of 1.31 μmol/L in KB, 1.85 μmol/L in KBV200,1.09 μmol/L in MCF7, and 3.19 μmol/L in MCF7/adr. Incontrast, the cytotoxic effect of paclitaxel in the resistant celllines was much lower than those in their parent lines, withIC50 values (48 hours after treatment) of >50 nmol/L inKBv200 compared with 1.63 nmol/L in KB, and >400nmol/L in MCF7/adr compared with 6.54 nmol/L inMCF7. The equal inhibitory effect of BFBTS in the MDR1–highly expressing cell lines and their parent cell lines strong-ly suggests that BFBTS may not be a substrate for the MDR1transporters.

Figure 3. Inhibition of microtubule polymerization dynamics and specific modification of β-tubulin at Cys 12 by BFBTS. A, inhibition of microtubuleorganization in MCF7 cells by BFBTS, paclitaxel, and vinblastine. The drug concentrations were indicated. DMSO served as a solvent control. B, in-hibition of tubulin polymerization in cell-free system by BFBTS, paclitaxel, and vincristine at different concentrations as indicated. Tubulin polymerizationkinetics of different compounds and DMSO control were recorded and represented by relative fluorescent unit (RFU) in the wavelength range between360 and 430 nm over time (in minutes). C, life history plots of the microtubule dynamics from DMSO-treated or 54 nmol/L BFBTS-treated MCF7 cells.The curves represent the changes in length (distance) of individual microtubules over time. The bar graph shows a representative data of two separateexperiments. The quantitative measurement (columns, mean; bars, SD) of the dynamicity of BFBTS-treated MCF7 cells at 54, 72, and 108 nmol/L are4.62 ± 2.15, 3.65 ± 0.9, and 2.34 ± 0.64 in comparison with 10.13 ± 2.56 in DMSO-treated cells. *, P < 0.05; **, P < 0.01 in comparison withDMSO treatment.






BFBTS-Mediated Mitotic Arrest and Microtubule

Depolymerization

In our previous study, we identified and reported aunique cellular response pattern for cells treated with anti-mitotic agents (32, 33). The kinetic curves of cells treatedwith antimitotic agents declined between 5 to 14 hoursafter compound addition, and then steadily increasedafterwards (32, 33). Interestingly, 7 of 10 tumor cell lines(PC3, A549, SH-SY5Y, Bcap-37, Panc-1, MV522, andHT29) treated with BFBTS also displayed such typical ki-netic pattern (Fig. 1C). This finding allowed us to hypo-thesize that BFBTS may also function as an antimitoticagent. To further test whether BFBTS induces antimitoticpattern, we treated A549 cells with BFBTS in conjunctionwith two well-known antimitotic agents, paclitaxel andvinorelbine. As a control, we treated A549 cells with com-pounds that have different mechanisms of action andreveal distinguishable kinetic patterns in live cells, in-cluding DNA-damaging agent, etopside, CDK inhibitor,indirubin-3′-oxime, and actin disruption agent, latruculin.Cells were continuously monitored for 48 hours in real-time after the treatment. A549 cells treated with BFBTSgenerated a mitotic arrest pattern identical to that obtainedfor paclitaxel and vinorelbine (Fig. 2A), but distinct fromthose of etopside, indirubin-3′-oxime, and latruculin-treated cells (Fig. 2A), indicating that BFBTS may functionas an antimitotic agent inducing mitotic arrest.To further test whether BFBTS induces mitotic arrest, we

treated A549, MCF7, and MCF7/adr cells with 2 μmol/LBFBTS, 30 nmol/L paclitaxel, or 0.1% DMSO that served

Figure 4. Mass spectrometry anal-ysis of the BFBTS-treated tubulin asdisplayed by the main fragmentationseries (carboxy, y and amino, b, se-ries) for peak from the retention timeof 35.51 min at m/z 1,137.18. Theb ions (b4–b15) and y ions (y4–y15)were observed to have an increase of172 kDa.



as negative control. The treated cells were then stained withan antibody against phosphohistone histone H3, a markerfor cells undergoing mitosis and mitotic arrest (34, 35). Asshown in Fig. 2B, increased number of phosphorylated his-tone H3–positive cells were visualized in BFBTS or paclitaxel-treated A549 cells compared with the DMSO-treated cells.In BFBTS-treated MCF7 and MCF7/adr cells, phasphory-lated histone H3–positive cells were increased, and in con-trast, such increase can only be seen in paclitaxel-treatedMCF7 cells but not in paclitaxel-treated MCF7/adr cellswhere MDR1 gene was highly expressed. The extent of mi-totic arrest was quantified by calculating the mitotic index,which was 48.4 ± 4.49%, 44.92 ± 8.42%, and 53.63 ± 5.98%for BFBTS-treated A549, MCF7, and MCF7/adr, respective-ly, and 42.35 ± 9.53%, 49.35 ± 12.50%, and 6.44 ± 1.96% forpaclitaxel-treated A549, MCF7, and MCF7/adr cells, respec-tively (Fig. 2C). Consistent with the cytotoxicity assayshown in Fig. 1D, the BFBTS showed equal mitotic arrestpotency in the MCF7 cells and MDR1-expressing MCF7/adr cells. Furthermore, using fluorescence-activated cellsorting analysis, we determined that the accumulation ofcells with 4N DNA content increased after 12 and 24 hoursof BFBTS treatment (Fig. 2D), strongly indicating thatBFBTS induced cell cycle arrest at G2-M phase.Next, we questioned whether BFBTS-induced mitotic ar-

rest resulted from its effect on microtubule polymerization.To address this question, BFBTS-treated MCF7 cells werestained with FITC-conjugated anti-tubulin antibody. Asshown in Fig. 3A, MCF7 cells treated with 1 μmol/L BFBTSfor 4 hours display a significant reduction in polymerized





microtubules and microtubule networks, which were simi-lar to MCF7 cells treated with the microtubule inhibitor,vinblastine, but different from MCF7 cells treated with mi-crotubule stabilizer, paclitaxel. This observation suggeststhat BFBTS may function as a microtubule inhibitor, possi-bly by preventing polymerization of tubulin into microtu-bules or by facilitating the disruption of microtubularnetworks. In cell-free tubulin assembly assay system usingpurified bovine tubulin, we found that BFBTS at concentra-tion of 3 μmol/L indeed completely inhibited the tubulinassembly into microtubule processes, whereas paclitaxelpromoted the assembly (Fig. 3B). Collectively, the resultsstrongly support the notion that BFBTS induces mitoticarrest in the tumor cells by directly inhibiting microtubulepolymerization.Inhibition of Microtubule Dynamics by BFBTS

Both in cell-free assay system and in cultured cells, micro-tubules are dynamic polymers that grow and shorten by theaddition and loss of tubulin subunits at the microtubuleends (18, 36). In contrast to treatment with relatively highconcentrations where microtubule-targeted compoundscan be distinguished based on their ability to either promotepolymerization of tubulin or induce microtubule disassem-bly, at low concentrations, they share a common underlyingmechanism, which is suppression of microtubule dynamics(6, 36). The suppression of microtubule dynamics, whichprevents chromosome alignment at the metaphase plate, re-sults in a sustained block at or before the metaphase-ana-phase transition leading to mitotic arrest followed by celldeath or apoptosis (18). Previous work has also supportedthe notion that the concentration of compounds that affectsmicrotubule dynamics, but not microtubule polymerizationin live cells, is similar to the concentration that inhibits pro-liferation (6, 7). To test the effect of BFBTS on microtubuledynamics, we used MCF7 cells that had been stably trans-fected with GFP-α-tubulin. At the flat lamellar edge of theinterphase cells, GFP-labeled microtubules are readily visi-ble and can be followed for several minutes by time-lapsefluorescent microscopy (data not shown). The position ofeach microtubule end with time was then determined andgraphed to generate a “life history plot” (Fig. 3C), from


which parameters of microtubule dynamics were deter-mined. The parameters used here for testing the effect ofBFBTS on microtubule dynamics include growth rate, short-ening rate, attenuation, time- and length-based catastrophefrequencies, time- and length-based rescue frequencies, anddynamicity (Table 2). The dynamic curves in Fig. 3C, eachrepresenting a single microtubule, showed the frequentminute changes in length of the extreme ends of microtu-bules. In the BFBTS-treated cells, low nanomolar BFBTS(54 nmol/L) significantly stabilized the microtubules com-pared with the DMSO-treated cells, showing very smallchanges in microtubule length over time (Fig. 3C). The dy-namicity, which represents the changes in length over time(μm/minute), was also significantly inhibited in the cellstreated with BFBTS in a dose-dependent manner (Fig. 3),and dynamicity was inhibited by approximately 54%,64%, and 76% at the concentrations of 54, 78, and 108nmol/L of BFBTS, respectively.To further confirm inhibitory effect of BFBTS on micro-

tubule dynamics, we analyzed two microtubule dynamicparameters, the frequencies of catastrophe and rescue,which are important parameters in determining microtu-bule function (18, 37). The catastrophe frequency is the fre-quency with which the microtubules switch from eitherpause or growth to shortening. The rescue frequency isthe frequency with which the microtubules switch fromshortening to either growth or pause. In BFBTS-treatedcells, both the time- and length-based rescue frequenciessignificantly decreased whereas both the time- andlength-based catastrophe frequencies significantly in-creased compared with DMSO-treated cells that servedas the control (Table 2), indicating that BFBTS suppressesthe microtubule dynamics by destabilizing the microtubulepolymerization.Specific Modification of β-tubulin at Cys 12 by BFBTS

A previous study reported a sulfhydryl group of cy-steines in β-tubulin that could be oxidatively modified bythe garlic-derived organosulfur compound diallyl trisulfidewhose trisulfide group plays a key role in this modification(38). BFBTS contains the same trisulfide group, although theside group of BFBTS is a fluorobenzene ring. Therefore,

Table 2. Inhibition of microtubule dynamics in MCF7 cells by BFBTS

DMSO
BFBTS (54 nmol/L)
© 2009 Ame

BFBTS (72 nmol/L)

Mol Cancer Ther 2009;8(

rican Association for Cance

BFBTS (108 nmol/L)

% Time spent
Growing 58.92% 32.97% 23.80% 25.14% Shortening 29.51% 32.85% 27.56% 17.59% Attenuated 11.57% 34.18% 48.64% 57.27% Frequency of Catatrophe (events/min) 2.967 2.580 1.572 1.389 Catatrophe (events/μm) 0.480 0.850 1.010 0.950 Rescue (events/min) 6.976 5.134 4.131 5.858 Rescue (events/μm) 0.403 0.940 0.881 1.138 Dynamicity 10.130 4.620 3.650 2.430
12). December 2009

r Research.




BFBTS may also bind sulfhydryl group of cysteines in >β-tubulin. To test this hypothesis, we used mass spectro-metry to analyze the possible interaction between BFBTSand tubulin. To identify how BFBTS directly interacts withtubulin, we digested the BFBTS-treated tubulin with tryp-sin and analyzed the digests with liquid chromatographytandem mass spectrometry. Peptide mass mapping ofBFBTS-treated tubulin identified >50% of β-tubulin. Bycomparing with the trypsinized tubulin fragmental massspectrum of DMSO control samples, the BFBTS-treatedβ-tubulin peptides showed an increase of 172 kDa in themass, which corresponded to the expected fragment ofBFBTS, containing two sulfur atoms forming β-tubulin-ss-fluorobenzyl. The cysteine residue modification byBFBTS was confirmed by identifying the presence of thepeptide of 3EIVHIQAGQCGNQIGAKFW21 ([M+2H]2+of m/z 1137.18; Fig. 4). Thus, in the β-tubulin of BFBTS-treated samples, the residue Cys 12 was identified to bemodified by BFBTS with increased mass of 172 kDa. Nomodification could be detected in the other cysteine resi-dues either in β-tubulin or in α-tubulin (data not shown),



suggesting that the cysteine modification is limited to theindicated peptide region.Inhibition of Tumor Growth in Human Tumor

Xenografts by BFBTS

Three human cancer lines, A549, Bcap-37, and SKOV3,were selected for testing BFBTS antitumor activity in xeno-grafts models, using three dose treatment cohorts (6.25, 12.5,and 25 mg/kg) once per day (q.d.), i.v. × 10 days. Mice trea-ted with paclitaxel at 10 mg/kg, i.v., q.d. for 7 consecutivedays served as a positive control. Compared with solventcontrol group, BFBTS in three dose treatment cohorts inhib-ited tumor growth in A549, Bacp-37, and SKOV3 xeno-grafts, showing significantly low relative tumor growthrates (Figs. 1B, 5A, B, and C). The relative tumor growthrates (T/C %) in BFBTS-treated A549 xenografts were19.91%, 25.36%, and 31.24% for 25, 12.5, and 6.25 mg/kgdoses, respectively, and the relative tumor growth ratewas 26.92% when administrating paclitaxel. In Bcap-37 xe-nografts, the relative tumor growth rates (T/C%) were18.57%, 29.23%, and 40.40% when given at doses of 25,12.5, and 6.25 mg/kg, respectively, and 23.14% of relative

Figure 5. Inhibition of tumor growth in human tumor xenografts by BFBTS. Mice were injected s.c. with A549 cells, Bcap-37 cells, SKOV3 cells, andMCF7/adr cells. For mice bearing A549 (A), Bcap-37 (B), and SKOV3 (C) tumors, the mice were treated with BFBTS and paclitaxel 7 d after tumorinoculation. The BFBTS doses and treatment schedule are 6.25, 12.5, or 25 mg/kg i.v. q.d, × 10 d. The paclitaxel dose and treatment schedule of10 mg/kg i.v. q.d. × 7 d served as positive control. Cremaphor/ethanol i.v. q.d. × 10 d was used as solvent control. **, P < 0.01 in comparison withsolvent control group. D, after 7 d after tumor inoculation, the mice bearing MCF7/adr tumors were treated with BFBTS every other day at 50 mg/kg i.v. ×5 times (10 d), or paclitaxel every other day at 20 mg/kg i.v. × 5 times (10 d), or cremaphor/ethanol (solvent control) i.v. every other day × 5 times (10 d).The tumor growth was indicated by tumor volume (mm3).





tumor growth rate was seen in the Bcap-37 xenografts trea-ted with paclitaxel. Similar relative tumor growth rates werealso seen in SKOV3 xenografts with 23.47%, 34.46%, and46.77% when administrated with BFBTS at doses of 25,12.5, and 6.25 mg/kg, whereas paclitaxel administrationshowed 18.29% relative tumor growth rate. In cultured cells,MDR1-overexpressing MCF7/adr cell line was sensitive toBFBTS but resistant to paclitaxel (Figs. 1D, 2B and C). Tocompare the in vivo antitumor efficacy of BFBTS with pacli-taxel in MCF7/adr cells, we used the MCF7/adr xenograftmodel. The mice bearing the MCF7/adr tumors were trea-ted with either BFBTS at dose of 50 mg/kg i.v., once every 2days × 5 times, or paclitaxel at maximum tolerant dose of 20mg/kg i.v., once every 2 days × 5 times. The concentrationsand dose schedule used in the MCF7/adr xenograft modelwere determined by a pretest in the xenograft model(data not shown). As shown in Fig. 5D, SupplementFig. S1, and Supplement Table S1, BFBTS was much morepotent in MCF7/adr xenografts with relative tumor growthrate (T/C %) of 12.3% than paclitaxel with high relativetumor growth rate (T/C %) of 42.98%. The tumor weightsin the BFBTS-treated mice were also significantly less thanthose in either solvent or paclitaxel treated mice (Supple-ment Fig. S1).

DiscussionBFBTS is shown to be one of the growing classes of natu-rally derived microtubule inhibitors with potent antitumoractivities. Microtubules are among the most successful tar-gets for anticancer therapy, and antimicrotubule agents areknown to directly bind to microtubules and/or soluble tu-bulin altering highly dynamic microtubule polymerizationprocess (39). The microtubule binding sites of known anti-microtubule agents are diverse and can be classified intoVinca domain, colchicine domain, Taxane domain, and oth-er microtubule binding sites (1, 39). Using mass spectrumassay, we found that the sulfhydryl group of Cys12 of β-tubulin was covalently modified and formed a disulfidebridge with BFBTS, resulting in β-tubulin-SS-fluorobenzyl.Notably, among 12 cysteine residues in α-tubulin and 8 cys-teine residues in β-tubulin, BFBTS only targeted β-tubulinat Cys12. Interestingly, Cys12 is located in Vinca domainnear the exchangeable GTP-binding site (40), and it wasreported that C12S mutation of β-tubulin significantlyaltered tubulin function in Saccharomyces cerevisiae andwas lethal, indicating the important role of Cys 12 inβ-tubulin (41). The covalent modification of Cys 12 in β-tubulin by BFBTS leads to blockage of sulfhydryl group,which is situated in a polar environment to maintain nor-mal tubulin function. Modification by BFBTS is apparentlyvery selective, because no other BFBTS-modified cysteineresidues in either α-tubulin or β-tubulin were detected (da-ta not shown). Recently, an antitumor agent, imidazolyl di-sulfide IV-2, and garlic-derived organosulfur compound,diallyl trisulfide, have been reported to have antimicrotu-bule activity by modifying cysteine residues in β-tubulin


(37, 42). Interestingly, the trisulfide analogue, diallyl trisul-fide, modified the tubulin at Cys 12 and at Cys 354 (37).Together with our findings on BFBTS, we suggest that or-ganosulfur compounds may share a common mechanismfor microtubule inhibition through modification of sulfhy-dryl groups of cysteine residues in tubulin, and such mod-ification mechanism differs from other microtubuleinhibitors. Therefore, this group of compounds may forma new class of antimicrotubule agents.Compared with clinically used microtubule inhibitors,

taxanes and Vinca alkaloids, BFBTS is a moderately potentmicrotubule inhibitor with a IC50 value at high nanomolarrange in the most sensitive cell lines tested in vitro. Interest-ingly, in tumor xenografts, BFBTS displayed potent tumorgrowth inhibition activity in A549, Bcap-37, and SKOV3 xe-nografts with the dose range between 6.25 and 25 mg/kg,similar to paclitaxel, of which the in vitro IC50 values were11.95, 3.48, and 4.55 nmol/L in A549, Bcap-37, and SKOV3lines, respectively (data not shown). Although in vitro anti-tumor activities of a given compound is not necessary di-rectly correlated to in vivo potency of tumor growthinhibition in xenografts, the strong potent antitumor growthactivities of BFBTS in A549, Bcap-37, and SKOV3 xenograftsmay indicate its unique in vivo characteristics, such as tumortissue penetration and/or unique compound metabolismprofiles, which deserve further study. The potent inhibitoryactivity of BFBTS in xenografts may conceivably result fromits potent suppression of microtubule dynamics. Indeed,BFBTS suppresses microtubule dynamics in live cells atthe concentration of 54 nmol/L (Fig. 3C), which was almost19-fold lower than the dose needed for disrupting micro-tubule organization (Fig. 3A). This clearly indicates that amuch lower BFBTS concentration is needed to alter cancercell microtubule dynamics, which are extremely importantin the processes of cancer cell division and mitosis (29, 32).Interestingly, by real-time monitoring of live MCF7 cells,cell proliferation inhibition can be clearly detected in re-sponse to the exposure of 190 nmol/L BFBTS (Fig. 1D).This result is consistent with previous findings that po-tent suppression of microtubule dynamic by microtubuleinhibitors was more directly correlated to cell prolifera-tion inhibition (6, 7).In addition, antitumor activities of microtubule inhibitors

may also result from their effect on angiogenesis. Inhibitionof angiogenesis by disrupting microtubules has also beenreported for microtubule inhibitors such as taxanes, vincris-tine, noscapine, and a new antimicrotubule agent, 2ME2(34, 36). Microtubule inhibition by these agents down-regulates hypoxia-inducible factor-1at the posttranscript-ional level and inhibits hypoxia-inducible factor-1–inducedtranscriptional activation of vascular endothelial growthfactor expression, which plays a key role in cancer angio-genesis (34, 36). It has been reported that inhibition ofangiogenesis by another structurally similar natural productdiallyl trisulfide resulted from inactivation of Akt anddownregulation of vascular endothelial growth factorand vascular endothelial growth factor R2 (37, 38). Inthis study, we found that BFBTS inhibited the endothelial






tube formation (data not shown). Therefore, it is plausiblethat some aspects of BFBTS antitumor activity may bedue to disruption of angiogenesis, which warrants furtherinvestigation.In summary, the natural active product derivative

BFBTS was shown to have broad antitumor therapeuticpotentials, which most likely resulted from its antimitoticeffect. BFBTS directly inhibited microtubule polymeriza-tion by covalently binding β-tubulin Cys 12, leading tothe disruption of the microtubule stability and inhibitionof microtubule dynamics. BFBTS inhibited tumor cellgrowth in A549, Bcap-37, and SKOV3 human tumor xeno-grafts; furthermore, both in vitro cell culture and inMCF7/adr xenografts, the tumor inhibitory effect was in-dependent of the MDR1 expression, suggesting thatBFBTS would have clinical benefit of overcoming MDR1-mediated drug resistance. Dissection of underlying mech-anism of BFBTS-tubulin interaction and modification andtumor cell apoptosis is warranted and will help to furtherexplore BFBTS anticancer therapeutic potentials.

Disclosure of Potential Conflicts of Interest

X. Xu and X. Wang: employees, shareholders, and patent holders,ACEA Biosciences, Inc. No other potential conflicts of interest weredisclosed.

References

1. Jordan MA, Wilson L. Microtubules as a target for anticancer drugs.Nat Rev Cancer 2004;4:253–65.

2. Pepperkok R, Bre MH, Davoust J, Kreis TE. Microtubules are stabilizedin confluent epithelial cells but not in fibroblasts. J Cell Biol 1990;111:3003–12.

3. Saxton WM, Stemple DL, Leslie RJ, Salmon ED, Zavortink M, McIntoshJR. Tubulin dynamics in cultured mammalian cells. J Cell Biol 1984;99:2175–86.

4. Zhai Y, Kronebusch PJ, Simon PM, Borisy GG. Microtubule dynamicsat the G2/M transition: abrupt breakdown of cytoplasmic microtubulesat nuclear envelope breakdown and implications for spindle morphogene-sis. J Cell Biol 1996;135:201–14.

5. Rusan NM, Fagerstrom CJ, Yvon AM, Wadsworth P. Cell cycle-dependent changes in microtubule dynamics in living cells expres-sing green fluorescent protein-α tubulin. Mol Biol Cell 2001;12:971–80.

6. Yvon AM, Wadsworth P, Jordan MA. Taxol suppresses dynamics of in-dividual microtubules in living human tumor cells. Mol Biol Cell 1999;10:947–59.

7. Hadfield JA, Ducki S, Hirst N, McGown AT. Tubulin and microtubulesas targets for anticancer drugs. In: Meijer L, Jezequel A, Rocberge M, edi-tors. Progress in Cell Cycle Research. 2003, p. 309–25.

8. An H, Zhu J, Wang X, Xu X. Synthesis and anti-tumor evaluation ofnew trisulfide derivatives. Bioorg Med Chem Lett 2006;16:4826–9.

9. Rosner H, Williams LA, Jung A, Kraus W. Disassembly of microtubulesand inhibition of neurite outgrowth, neuroblastoma cell proliferation, andMAP kinase tyrosine dephosphorylation by dibenzyl trisulphide. BiochimBiophys Acta 2001;1540:166–77.

10. Kim S, Kubec R, Musah RA. Antibacterial and antifungal activity ofsulfur-containing compounds from Petiveria alliacea L. J Ethnopharmacol2006;104:188–92.

11. Williams LA, Rosner H, Moller W, Conrad J, Nkurunziza JP, Kraus W.In vitro anti-proliferation/cytotoxic activity of sixty natural products on thehuman SH-SY5Y neuroblastoma cells with specific reference to dibenzyltrisulphide. West Indian Med J 2004;53:208–19.



12. Williams LA, Vasquez E, Klaiber I, Kraus W, Rosner H. A sulfonicanhydride derivative from dibenzyl trisulphide with agro-chemical activ-ities. Chemosphere 2003;51:701–6.

13. Williams L, The TL, Gardner M, et al. Immunomodulatory activities ofPetiveria alliacea L. Phytother Res 1997;11:251–3.

14. Williams L, Rosner H, Moller W, Kraus W. Anti-proliferation/cytotoxicaction of dibenzyl trisulphide, a secondary metabolite of Petiveria alliacea.Jam J Sci 2004;15:54–60.

15. Mata-Greenwood E, Ito A, Westenburg H, et al. Discovery of novelinducers of cellular differentiation using HL-60 promyelocytic cells. Anti-cancer Res 2001;21:1763–70.

16. Bao Y, Mo X, Xu X, et al. Stability studies of anticancer agent bis(4-fluorobenzyl)-trisulfide and synthesis of related substances. J PharmBiomed Anal 2008;48:664–71.

17. Bao Y, He Y, Xu X, et al. Quantitative determination of anti-tumoragent bis(4-fluorobenzyl)trisulfide, fluorapacin and its pharmaceuticalpreparation by high-performance liquid chromatography. J Pharm BiomedAnal 2008;46:206–10.

18. Kamath K, Jordan MA. Suppression of microtubule dynamics byepothilone B is associated with mitotic arrest. Cancer Res 2003;63:6026–31.

19. Xi B, Yu N, Wang X, Xu X, Abassi YA. The application of cell-based label-free technology in drug discovery. Biotechnol J 2008;3:484–95.

20. Xing JZ, Zhu L, Jackson JA, et al. Dynamic monitoring of cytotoxicityon microelectronic sensors. Chem Res Toxicol 2005;18:154–61.

21. Sangrajrang S, Fellous A. Taxol resistance. Chemotherapy 2000;46:327–34.

22. Solly K, Wang X, Xu X, Strulovici B, Zheng W. Application of real-timecell electronic sensing (RT-CES) technology to cell-based assays. AssayDrug Dev Technol 2004;2:363–72.

23. Mosse YP, Laudenslager M, Longo L, et al. Identification of ALK asa major familial neuroblastoma predisposition gene. Nature 2008;455:930–5.

24. Martin-Manso G, Galli S, Ridnour LA, Tsokos M, Wink DA, RobertsDD. Thrombospondin 1 promotes tumor macrophage recruitment and en-hances tumor cell cytotoxicity of differentiated U937 cells. Cancer Res2008;68:7090–9.

25. Tucker LA, Zhang Q, Sheppard GS, et al. Ectopic expression ofmethionine aminopeptidase-2 causes cell transformation and stimulatesproliferation. Oncogene 2008;27:3967–76.

26. Zhang J, Ren H, Yuan P, et al. Down-regulation of hepatoma-derivedgrowth factor inhibits anchorage-independent growth and invasion of non-small cell lung cancer cells. Cancer Res 2006;66:18–23.

27. Hosono T, Fukao T, Ogihara J, et al. Diallyl trisulfide suppressesthe proliferation and induces apoptosis of human colon cancer cellsthrough oxidative modification of β-tubulin. J Biol Chem 2005;280:41487–93.

28. Azarenko O, Okouneva T, Singletary KW, Jordan MA, Wilson L.Suppression of microtubule dynamic instability and turnover inMCF7 breast cancer cells by sulforaphane. Carcinogenesis 2008;29:2360–8.

29. Jordan MA, Kamath K. How do microtubule-targeted drugs work? Anoverview. Curr Cancer Drug Targets 2007;7:730–42.

30. Benderra Z, Trussardi A, Morjani H, et al. Regulation of cellular gluta-thione modulates nuclear accumulation of daunorubicin in human MCF7cells overexpressing multidrug resistance associated protein. Eur J Cancer2000;36:428–34.

31. Zhu X, Sui M, Fan W. In vitro and in vivo characterizations of tetran-drine on the reversal of P-glycoprotein-mediated drug resistance to pacli-taxel. Anticancer Res 2005;25:1953–62.

32. Jordan MA, Horwitz SB, Lobert S, Correia JJ. Exploring the mechan-isms of action of the novel microtubule inhibitor vinflunine. Semin Oncol2008;35:S6–12.

33. Abassi YA, Xi B, Zhang W, et al. Kinetic cell-based morphologicalscreening: prediction of mechanism of compound action and off-targeteffects. Chem Biol 2009;16:712–23.

34. Newcomb EW, Lukyanov Y, Schnee T, et al. Noscapine inhibitshypoxia-mediated HIF-1α expression andangiogenesis in vitro: a novelfunction for an old drug. Int J Oncol 2006;28:1121–30.





35. Hendzel MJ, Wei Y, Mancini MA, et al. Mitosis-specific phosphory-lation of histone H3 initiates primarily within pericentromeric hetero-chromatin during G2 and spreads in an ordered fashion coincidentwith mitotic chromosome condensation. Chromosoma 1997;106:348–60.

36. Mabjeesh NJ, Escuin D, LaVallee TM, et al. 2ME2 inhibits tumorgrowth and angiogenesis by disrupting microtubules and dysregulatingHIF. Cancer Cell 2003;3:363–75.

37. Xiao D, Li M, Herman-Antosiewicz A, et al. Diallyl trisulfide inhibitsangiogenic features of human umbilical vein endothelial cells by causingAkt inactivation and down-regulation of VEGF and VEGF-R2. Nutr Cancer2006;55:94–107.

38. Shankar S, Chen Q, Ganapathy S, Singh KP, Srivastava RK. Diallyltrisulfide increases the effectiveness of TRAIL and inhibits prostate cancer


growth in an orthotopic model: molecular mechanisms. Mol Cancer Ther2008;7:2328–38.

39. Jackson JR, Patrick DR, Dar MM, Huang PS. Targeted anti-mitotictherapies: can we improve on tubulin agents? Nat Rev Cancer 2007;7:107–17.

40. Gupta S, Bhattacharyya B. Antimicrotubular drugs binding to vincadomain of tubulin. Mol Cell Biochem 2003;253:41–7.

41. Gupta ML, Jr., Bode CJ, Dougherty CA, Marquez RT, Himes RH.Mutagenesis of β-tubulin cysteine residues in Saccharomyces cerevisiae:mutation of cysteine 354 results in cold-stable microtubules. Cell MotilCytoskeleton 2001;49:67–77.

42. Huber K, Patel P, Zhang L, et al. 2-[(1-methylpropyl)dithio]-1H-imidaz-ole inhibits tubulin polymerization through cysteine oxidation. Mol CancerTher 2008;7:143–51.




Published OnlineFirst December 8, 2009.Mol Cancer Ther Wanhong Xu, Biao Xi, Jieying Wu, et al. and suppression of microtubule dynamics

-tubulin at Cys 12βinhibits tumor growth by modification of Natural product derivative Bis(4-fluorobenzyl)trisulfide

Updated version

10.1158/1535-7163.MCT-09-0548doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2009/12/14/1535-7163.MCT-09-0548.DC1

Access the most recent supplemental material at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://mct.aacrjournals.org/content/early/2009/12/04/1535-7163.MCT-09-0548To request permission to re-use all or part of this article, use this link

on May 16, 2018. © 2009 American Association for Cancer Research. mct.aacrjournals.org Downloaded from


http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-09-0548

http://mct.aacrjournals.org/content/suppl/2009/12/14/1535-7163.MCT-09-0548.DC1

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/early/2009/12/04/1535-7163.MCT-09-0548


Natural product derivative Bis(4-fluorobenzyl)trisulfide...

Documents

Transcript of Natural product derivative Bis(4-fluorobenzyl)trisulfide...