ICANCERRESEARCH57, 3208-3213. August 1, 19971

ABSTRACT

E7010 (N-[2-[(4-hydroxyphenyl)amino]-3-pyridinyll-4-methoxybenzenesulfonamide), an orally active sulfonamide antitumor agent that iscurrently in a Phase I clinical trial, showed rather consistent growthinhibitory activities against a panel of 26 human tumor cell lines(IC@ 0.06-0.8 @ag/ml),in contrastto vlncristine(VCR;ICso 0.00020.04 @.tg/ml), 5-fluorouracil (IC@ = 0.2—30 gag/mI), Adriamycin(IC50= 0.002-0.7 gag/ml),mitomycinC (ICso 0.0073 @Lg/Ifll),1-@3-D-arobinofuronoxylcytosine (ICse 0.005 to >30 gLg/ml), camptothecin

(ICso 0.®2@0.4@ andcisplatin(ICso 0.520 @LgJ1fll).It causeda dose-dependent increase in the percentage of mitotic cells in parallelwith a decrease in cell proliferation, like VCR. It also showed a dosedependent inhibition of tubulin polymerization, which correlated wellwith the cell growth-Inhibitory activity. ‘4C-labeledE7010 bound to purifled tubulin, and this binding was Inhibited by colchicine but not by

VCR. However,its bindingpropertieswere differentfrom those of colchicine, as well as those of VCR. E7010 was active against two kinds ofVCR-resistant P388 cell lines, one of which showed multidrug resistancedue to the overexpressionof P-glycoproteln(resistantto Taxol),and theotherdidnotshowmultidrugresistance(sensitiveto Taxol).Furthermore,four E7010-resistant P388 cell lines showed no cross-resistance to VCR, adifferent pattern of resistance to podophyllotoxin, and collateral sensitivity to Taxol. Therefore, E7010 is a novel tubulin-binding agent that has awider antitumor spectrum than VCR and has different properties from

those of VCR or Taxol.

INTRODUCTION

As part of the search for clinically effective antitumor drugs to treatrefractory solid tumors, we have screened a number of sulfonamides,because this class of compounds is well known to have a variety ofbiological activities, such as antibacterial, insulin-releasing, carbonicanhydrase-inhibitory, and antithyroid activities (1). We found thatE7010 showed a wide therapeutic range of antitumor activity in thecolon 38 carcinoma model (2). It inhibited the growth of colon 38cells inoculated s.c. by 60—99%when administered at doses of 25—100 mg/kg daily for 8 days (3). It was also active against mousetumors (s.c. inoculated M5076 fibrosarcoma, s.c. inoculated Lewislung carcinoma, and i.p. inoculated P388 leukemia), a rat tumor (s.c.inoculated SST-2 mammary carcinoma), and human xenografts (LC376, LC-6, and LX- 1 lung cancer; H-8l , H-! 11, SC-2, and SC-6gastric cancer; H-143, COLO32ODM, and WiDr colon cancer; and

H-31 and MX-l breast cancer) upon oral administration (3). Moreover, E70l0 was effective against VCR2-resistant, CDDP-resistant,and 5-Ri-resistant P388 cells in vivo (3).

The present study was designed to elucidate the mechanism ofaction of E7010, and we report here that E70l0 is a novel antimitotic

Received 3/4/97; accepted 6/2/97.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

I To whom requests for reprints should be addressed. Phone: 81-298-47-5749; Fax:

81-298-47-2037.2 The abbreviations used are: VCR, vincristine; 5-FU, S-fluorouracil; ADM, Adria

mycin; MMC, mitomycin C; Ara-C, l-@-o-arabinofuranoxylcytosine; CPT, camptothecin;CDDP,cisplatin;CLC, colchicine;POD, podophyllotoxin;NOC,nocodazole;MTX,methotrexate; VP16, etoposide; MNNG, N-methyl-N'-nitro-N-mtrosoguanidine; MU,3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; MDR, multidrug-resistant;CQS, chioroquinoxaline sulfonamide.

and tubulin-binding agent with different properties from those ofTaxol, VCR, or CLC.

MATERIALS AND METHODS

Drugs. E7010 and 14C-labeledE70l0 (specific activity, 56 mCi/mmol;phenyl-U-RING-'4C; Fig. 3B) were synthesized at Tsukuba Research Laboratories, Eisai Co. Ltd. Reference compounds were obtained from the following sources: VCR, vinblastine, CDDP, CPT, VP16, and melphalan were

obtained from Sigma (St. Louis, MO); 5-FU was obtained from Tokyo Kasei(Tokyo, Japan); ADM, MMC, CLC, POD, and MTX were obtained fromWako Junyaku (Tokyo, Japan); and Ara-C (Cylocide) was obtained fromNippon Sinyaku (Kyoto, Japan). Taxol was obtained from the Drug Synthesis

and Chemstry Branch, Division of Cancer Treatment, National Cancer Institate. Trisnordolastatin was kindly provided by Dr. Shioiri (Nagoya CityUniversity, Nagoya, Japan). [3H]VCR was purchased from Amersham, and[3HICLC was purchased from DuPont New England Nuclear.

Cell Lines. P388 murine leukemia and MX-l were supplied by CancerChemotherapy Center, Japan Foundation for Cancer Research (Tokyo, Japan).VCR-resistant P388 cell line (P388/VCR) with overexpression of P-glycopro

1cm(4)waskindlysuppliedbyDr.M.Inaba(CancerChemotherapyCenter,Japan Foundation for Cancer Research). H460, H520, H596, HCT-!16,SW480, SW620, COLO2O5,DLD-!, HCF-l5, HT-29, MiApaca-Il, and HL-60were obtained from the American Type Culture Collection (Rockville, MD).PC1, PC9, and MKN28 were obtained from the Immuno Biology Laboratory(Gunma, Japan). A549, WiDr, LS!74T, COLO32ODM, ZR-75-l, CCRFCEM, and K562 were obtained from Dainippon Pharmaceutical Co., Ltd.(Osaka, Japan). PC1O,MKN45, and KATO-3 were obtained from the JapaneseCancer Research Resources Bank (Tokyo, Japan). All cell lines were mainmined in RPMI 1640 containing 10% heat-inactivated fetal bovine serum,

penicillin (100 units/nil), streptomycin (100 @g/ml),2-mercaptoethanol (50@LM),and sodium pyruvate (1 mM).

MTT Assay. Exponentially growing cells were plated into 96-well tissueculture plates and precultured for 1 day. Various concentrations of test compounds were added, and the cells were incubated for an additional 3 days. Theantiproliferative activity was measured by MU assay, and the IC50 valueswere determined.

Mitotic Index. P388 cells were treated with the indicated concentrations ofE70l0 or VCR for 12 h. Viable cell numbers were determined by the trypanblue dye exclusion method, and cells were treated with 0.075 M KC1, fixed(3:1, methanol:acetic acid), and stained with 0.1% crystal violet. The mitotic

index was determined by counting at least 500 cells.Microtubule Polymerization Assay. Microtubule protein from bovine

brain was prepared by two cycles of assembly and disassembly in assemblybuffer [0.1 M morpholinoethanesulfonate, 1 mM EGTA, and 0.5 mM MgCl2(pH 6.8); Ref. 5]. Protein concentration was determined using the Bio-Radprotein assay kit. The reaction mixture contained 2 mg/mI mcrotubule proteinand the half-log step concentrations (0.89, 2.7, 8.9, and 27 ,.@M)of E70l0 inassembly buffer containing 1 mM GTP. Microtubule polymerization wasinitiated by warming the sample from 0°Cto 37°Cin a 1-mi cuvette, andturbidity (@A350) was recorded during the indicated time intervals with aspectrophotometer (Fig. 3A). In the study of the relationship between antiproliferative activity and microtubule polymerization-inhibitory activity (Fig. 3B),the wells of 96-well plates contained 1 mg/ml microtubule protein and log stepconcentrations of E70l0 derivatives. The plates were incubated at 37°C,andthe turbidity was recorded with a mcroplate reader (ThermoMax; MolecularDevices) for 20 mn kinetically. The turbidity at 20 mm (the polymerization inthe control reached a plateau level) was plotted on a linear scale versus theconcentration of the test compound on a log scale, and the concentrationcausing a 50% inhibition of polymerization was determined.

3208

Mechanism of Action of E7010, an Orally Active Sulfonamide Antitumor Agent:

Inhibition of Mitosis by Binding to the Colchicine Site of Tubulin

Kentaro Yoshimatsu,' AtSUmi Yamaguchi, Hiroshi Yoshino, Nozomu Koyanagi, and Kyosuke Kitoh

Tsukuba Research Laboratories, Eisai Co. Ltd., 1-3 Tokodai 5-chome, Tsukuba-shi, Ibaraki 300-26, Japan

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MECHANISM OF ACFION OF E70l0

E7010-resistant clones 0.5r-D, l.Or-H, and 4.Or-M were cloned, respectively.E7010-resistant cell line 0.6r-8 was isolated by a four-step selection. MNNGtreatment was used at each step, and the dose of E7010 at each step was 0.1,0.2, 0.3, and 0.6 @g/ml.After the four-step selection, 0.6r-8 was cloned.

The Effect of E7010 on Cell Cycle Progression. Antitumoragents are known to arrest cell cycle progression in S phase, 02, or Mphase (10, 11). Therefore, the effect of E7010 on the cell cycleprogression of P388 cells was examined by flow cytometry, andE70l0 was found to arrest P388 cells in the G2-M phase. Next, weexamined whether E7010 increases the percentage of mitotic cells,because agents that arrest cells in M phase, such as VCR, increase themitotic index, but agents that cause G2 arrest, such as VP16, decreaseit. E7010 caused a dose-dependent increase in the mitotic index, likeVCR (Fig. 2), and the rise in mitotic index was well correlated withthe decrease in cell proliferation. We also observed an increase in themitotic index of s.c. implanted M5076 tumor at 18—24h after the oraladministration of E7010 (data not shown).

The Effects of E7010 and Its Derivatives on MicrotubulePolymerization. The results described above and the immunofluorescence observation that E70l0 caused the disappearance of cytoplasmic microtubules in colon 38 cells (data not shown) led us toexamine E7010 for effects on microtubule polymerization. E70l0caused dose-dependent inhibition of microtubule polymerization (Figure 3A). Furthermore, a linear relationship between the antiprolifera

.@)

‘3

0

RESULTS

The Effects of E7010 and Antitumor Drugs on the Proliferationof Human Tumor Cell Lines. The antiproliferative activities ofE7010 and various clinically used antitumor drugs were tested againsta panel of 26 human tumor cell lines (7 lung cancers, 10 coloncancers, 3 stomach cancers, 1 pancreatic cancer, 2 breast cancers, and3 leukemias) by MU assay. E70l0 showed a dose-dependent and

broad-spectrum antiproliferative activity (Fig. 1). The IC50 values ofE7010 on various human tumors were rather similar (0.06—0.8 p.g/ml), in contrast to those of VCR (0.0002—0.04 @.tg/ml),5-FU (0.2—30j@g/ml), ADM (0.002-0.7 @tg/ml),MMC (0.007-3 @g/ml),Ara-C(0.005->30 @.Lg/ml),CPT (0.002-0.4 @@g/ml),and CDDP (0.5-20

@@g/ml).

j

Fig. 1. The effects of E70l0 and antitumor drugs on the proliferation of 26 humantumor cell lines. Cells were treated with various concentrations of the compounds for 3days.The antiproliferativeactivitywas determinedby MU assay,and IC50for eachcompoundwascalculated.A:•,E7010;0, VCR;i@,5-Ri; D, ADM.B:0, MMC; @,Ara-C; 0, CPT; X, CDDP. The cell lines used were non-small cell lung cancers PC1, PC9,PC1O, A549, H520, H460, and H596; colon cancers WiDr, LS174T, COLO32ODM,HCT-l 16, SW620, C0L0205, DLD-l, HCF-15, SW480, and HT29; stomach cancersMKN28, MKN45, and KATO3; pancreatic cancer MiApaca-Il; breast cancers MX-l and

@-75-l;and leukemias CCRF-CEM, 1(562, and HL-60.

Binding Assay. Tubulin was purified from microtubule protein on a phosphocellulose column (6). The binding assay was performed using the DEAEcellulose filter method of Borisy (7). The reaction mixture (0.2 ml) containedtubulin and ‘4C-labeledE70l0 (2 aiM;0.0224 pCi/reaction), [3H]VCR (0.25,.tM; 0.05 MCi/reaction), or [3H]CLC (1 ,.LM;0.1 pCi/reaction) in assembly

buffer. The reaction was stopped by rapid cooling to 4°C,and then the mixturewas diluted with 5 ml ofO.lX assembly buffer, filtered through a stack of threeWhatman DE81 ion exchange paper discs using a BRANDEL cell harvester(Biomedical Research and Development Laboratories, MD), and washed with10 ml of 0.lX assembly buffer.

Isolation ofNon-MDR VCR-reslstant P388. VCR-resistant cell lines thatdid not show the MDR phenotype were isolated from MNNG (Aldrich)-treatedP388 cells by selection with both increasing doses of VCR (0.5, 1.0, 2.0, and4.0 ng/ml) and 10 @Mverapamil to avoid the development of the mdrphenotype (8, 9). When the relative resistance of VCR-resistant cells to VCRreached about 10 times, the VCR-resistant clone P3881V4—4 was cloned inVCR and verapamil-containing medium using the limiting dilution method.The relative resistance was determined in the absence of verapamil by MUassay.

Isolation of E7010-resistant P388. E70l0-resistant cell lines 0.5r-D,1.Or-H,and 4.Or-M were isolated by MNNG-treated P388 cells by selectionwith increasing doses of E70l0 (0.1, 0.2, 0.3, 0.4, 0.5, 0.7, 0.8, 0.9, 1.0, 1.2,2.0, 3.0, and 4.0 p@g/ml).When the relative resistance of E7010-resistant cellsto E70l0 reached about 5 (13 weeks), 10 (18 weeks), and 40 times (21 weeks),

3209

8

1@

.01 .1 1 10

60

50

40

30xS.u.E 20U

0

@ 10

0.001

Concentration (@sg/mI)

Fig. 2. The effects of E70l0 and VCR on proliferation and the mitotic index. P388 cellswere treated with the indicated concentrations of E70l0 (•, ) or VCR (0, []) for 12 h.The number of viable cells (, El) and the mitotic index (•,0) were determined asdescribed in “Materialsand Methods.―Values represent the mean ±SE (bars) of threeindependent experiments.

@iO@OO@ @-@(OOU)@

@@ .1.

c@@

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

J//,,,,,,,—,—.--..,—------'—--———--.@ Control0.89 sM

2.7 @tM

8.9 sM

27 pM

5 10 15 20 25Time (mm)

MECHANISM OF ACTION OF E7010

Inhibition of ‘4C-labeledE7010, [3H]VCR, or [3H]CLC Bindingto Tubulin. To determine the binding site of E7010, inhibition of thebinding of ‘4C-labeledE70l0, [3H]VCR, or [3H]CLC to tubulin byeach unlabeled compound was examined. The binding of E70l0 totubulin was inhibited by CLC and enhanced by VCR (Fig. 5A). Thebinding of CLC was inhibited by E7010 and was also enhanced byVCR (Fig. SB). The enhancement of the binding by VCR was causedby stabilization of the CLC-binding activity of tubulin (12, 16). Thebinding of VCR was not inhibited by E7010 or CLC (Fig. SC).Lineweaver-Burk analysis demonstrated that E7010 is a competitiveinhibitor of the binding of CLC and binds at the CLC site on tubulin(Fig. SD). The K1value for E70l0 of [3H]CLC was 3.3 @M(Fig. SD),and the K1 values for CLC, POD, and NOC, reported to bind at theCLC site, were 2.7, 0.94, and 2.3 @M,respectively (data not shown).

Cross-Resistance in VCR-resistant P388 Cells. Antimitoticdrugs that are widely used in cancer chemotherapy, such as VCR,

vinblastine, and Taxol, are good substrates for transport by P-glycoprotein (classical MDR phenotype; Refs. 17—19).Moreover, othertypes of resistance to Vinca alkaloids have been reported (20—22),andwe isolated VCR-resistant cells (P388/V4—4) that did not show theclassical MDR phenotype. Therefore, the effects of VCR, Taxol,ADM, CLC, and E70!0 on two MDR P388 cells (P388IVCR andP388/V0.256) and P3881V4—4cells were examined. The results aresummarized in Table 1. P388/VCR and P388N0.256 cells, whichoverexpress P-glycoprotein, were significantly resistant to VCR,Taxol, CLC, and ADM but were sensitive to E7010. P3881V4—4cells,

0.3

0.2

0.1

0.0

A

i

B

0

@-U,00C.2a).0@

C

4

3

0

R=0.92

14

R2=HR3=OH A

2 63 4 5InhibitIon of tubulin polymerization

log [1C50(nM)]

Fig. 3. The effect of E70l0 on microtubule polymerization (A) and the correlationbetween antiproliferative activities and inhibitory activities against microtubule polymerization in E7010 derivatives (B). A, the reaction mixture contained 2 mg/mI microtubule protein and the indicated concentration of E70l0 in assembly buffer. Microtubulepolymerization was recorded as described in “Materialsand Methods.―B, in this study, theeffects of E7010 derivatives on microtubule polymerization were determined using 1mg/mI microtubule protein. Each point, the antiproliferative activity and antipolymerization activity of an E7010 derivative. 1, R' CH3O, R2 = OH, and R3 = H; 2,R' = CH3O,R2= H,andR3= H;3, R' = CH1O,R2= H,andR3= F;4, R' CH3O,R2 = CH@,and R3 = OH; 5, R' = C2H5, R2 = H, and R3 = H; 6, R' = CH3O, R2 = H,and R3 = Cl; 7, R' CH1O, R2 = H, and R3 = OH; 8, R' CH3, R2 = H, and R3 = H;9, R' = C2H50, R2 = H, and R3 = H; 10, R' = CH3O, R2 = H, and R3 = C2H5O; 11,R@= CH1O,R2= H,andR3= phenyl;12,R' = CH1O,R2= H,andR3= COOC2H5;13, R' = CH3O, R2 = H, and R3 = CH3; 14, R' = CH3O, R2 = H, and R3 = CH3O;15, R' = CH3O, R2 = H, and R3 = SCH3; 16, R' = CH3CO, R2 = H, and R3 = H; 17,R' = CN,R2= H,andR3= H; 18,R' = NO2,R2= H,andR3= H; 19:R' = CH3O,R2 CH3O,andR3= CH3O.

tive activity of a compound and its activity to inhibit microtubulepolymerization was found for various E70l0 derivatives that had been

synthesized during the development of E70l0 (Fig. 3B). These resultsstrongly suggest that the inhibition of microtubule polymerization isresponsible for the antiproliferative action of E7010.

The Binding of ‘4C-labeledE7010 to Tubulin. Compounds thatarrest the cells in mitosis and inhibit tubulin polymerization areknown to bind to tubulin at the Vinca site or the CLC site (12—15).The binding of E7010 to tubulin was examined and compared withthe binding of VCR and CLC, which are typical tubulin-bindingagents (Fig. 4). The binding of E70l0 to tubulin was rapid and didnot show temperature dependence. On the other hand, the bindingof VCR was slow at 4°C,whereas CLC could not bind to tubulinat 4°Cas reported previously (12), and its binding was slow evenat 37°C.

0E0.

•0C

0

B

0E0.

•0C

0

0 30 60 90 120Time (mm)

Fig. 4. The binding of ‘4C-labeledE70l0, [3H]VCR, and [3H]CLC to tubulin at 4°Cand 37°C.Purified tubulin (0.45 mg/ml) was incubated with ‘4C-labeledE70l0 (•),[3H]VCR (0), or [3H]CLC (@) at 4°Cor 37°Cfor the time indicated. The binding totubulin was determined by the DEAE-cellulose filter method as described in “Materialsand Methods―and expressed as the mean for duplicate samples in a single experiment,which was repeated with similar results.

3210

l4@

R@H@Q

@1fl2

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MECHANISMOF ACI1ON OF E70l0

01@•1C0U

•0C

0.000p..Lua

0I-C0U

0

•0C

0.0

0>I

C')

Fig. 5. A comparison of the effects of E7010, CLC, and VCR on thebinding of ‘4C-labeledEbb, [3HICLC, and [3H]VCR to tubulin.Microtubule protein (0.4 mg/mI) was incubated with ‘4C-labeledE70l0(A),[3HICLC(B andD) or [3H]VCR(C)andthe indicatedconcentrations of unlabeled compounds for 2 h at 37°C.A, B, and C: •,E70l0;& CLC;0, VCR.In the Lineweaver-Burkanalysis(D),thefollowingconcentrations of [3H]CLC were used: 0.56, 0.83, 1.25, 2, 4, and 8 @sM.The binding was determinedby the DEAE-cellulose filter methodasdescribed in “Materialsand Methods.―Points, the mean for duplicatesamples in a single experiment, which was repeated with similar results.

@10@@ io6 io@ i@-@ io@ io6 io@ io@

Concentration (M) Concentration (M)

1o6 io@ io@Concentration (M)

-1 0 11/Coichicine (1/pM)

2

which were isolated by exposure to VCR in the presence of verapamil,showed cross-resistance to vinblastine, vindesine, and trisnordolastatin (a dolastatin analogue; Ref. 23) with resistance factors of 4.7, 9.4,and 8.7, respectively, but they had a collateral sensitivity to Taxol.P388/V4—4 cells were sensitive to E7010 and slightly resistant to

CLC.Cross-Resistance in E7010-resistant P388 Cell Lines. For the

clinical application of a new drug, understanding the mechanisms bywhich resistance is developed is important. Therefore, we isolatedfour E7010-resistant cell lines and examined the sensitivity of E70l0-resistant cell lines to clinically used antitumor drugs (VP16, 5-FU,CDDP, ADM, MTX, MMC, Ara-C, and melphalan; Fig. 6A) andantimitotic agents (VCR, Taxol, CLC, NOC, and POD; Fig. 6B).None of E7010-resistant cell lines (0.6r-8, 0.5r-D, l.Or-H, and 4.Or-M)showed cross-resistance to any of the antitumor drugs of which thetarget is not tubulin. On clinically used antimitotic drugs, E70l0-

resistant cell lines exhibited no significant resistance to VCR but hadincreased sensitivity to Taxol. Among antimitotic agents that bind tothe CLC site, NOC showed a similar pattern of resistance across thespectrum of E70l0-resistant cell lines (resistance factor, 4.Or

M 1.Or-H > 0.5r-D 0.6r-8). On the other hand, POD exhibiteda different pattern of resistance (resistance factor, O.5r-D and l.OrH > 0.6r-8 and 4.Or-M). The pattern of resistance to CLC was similarto that of POD, although the resistance to CLC was minimal.

DISCUSSION

E70l0 was found by screening a number of sulfonamides withwidely different chemical structures (2) and shows a broad spectrumof activity against mouse and human tumors (3). In a Phase I study ofE70l0 using a single-dose schedule, some responses were observed(24), and a 5-day continuous administration study is in progress. Wehave demonstrated here that E7010 is an antimitotic agent, and thetarget of its antiproliferative action is tubulin. CQS was reported tohave antiproliferative activity in vitro (25, 26), but the mechanism ofits action has not yet been elucidated. Therefore, we examined theeffect of CQS on cell cycle progression and tubulin polymerization.CQS did not arrest P388 cells in the G2-M phase and did not affecttubulin polymerization. Therefore, E70l0 seems to be a unique sulfonamide antitumor agent.

Most of the tumor cell lines tested were almost equally sensitive toE7010 (IC50 = 0.06—0.8 @sg/m1).This is consistent with the idea thatthe target of E7010 is tubulin, because microtubules in the mitoticspindle are important for cell proliferation of all cells. In contrast, awider range of IC50 values (0.002—0.04 p.g/ml) was observed withVCR. This might be due to the differential drug accumulation of VCRamong these tumor cells, which express P-glycoprotein to variousextents. Indeed, HCT-15 and DLD-l, which were sensitive to E70l0(IC50 values of 0.08 and 0.14 p@g/ml,respectively) but very insensitive

3211

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Table 1 The effects of VCR, Taxol, ADM, CLC, and E7010 on three VCR-resistant P388 celllinesP388

cells and VCR-resistant P388 cell lines were treated with various concentrations of the compounds for 3 days. Antiproliferative activity was determined as indicated in

“Materialsand Methods,―and IC50 for each compound was calculated. Resistance factors were calculated as IC50 for the resistant cell line: IC50 for the sensitive cellline.Resistance

factorCell

line VCR Taxol E7010 CLCADMP388/VCR

11.0 ±2.4'@ 6.3 ±0.9 1.2 ±0.2 — —P388N@,56 82.8 ±7.7 16.9 ±0.3 1.5 ±0.3 64.5 29.9 ±2.9P388/V4-4 8.4 ±0.7 0.18 ±0.05 1.3 ±0.2 2.7 ±0.2 0.58 ±0.08a

Values representthe mean ±SE.

MECHANISM OF ACtiON OF E7010

to VCR, were reported to express P-glycoprotein and to show differential sensitivity to non-MDR versus MDR drugs (27). These observations are consistent with the finding that P388IVCR and P388/V0.256 were resistant to VCR but sensitive to E7010. The patterns ofantiproliferative activity against tumor cell lines except HCT-l5 and

DLD-l were similar for VCR and E7010, although the ranges of IC50values were different.

To examine the mechanism of action of E70l0, we first examinedthe effect on cell cycle progression, because this is related to themechanism of action (10, 11). Antimetabolites, such as MTX andAra-C, cause cells to accumulate in the S phase. Topoisomeraseinhibitors, VP16 and CPT, ADM, and MMC inhibit cell cycle pro

gression in the 02 phase. Vinca alkaloids and Taxol arrest cells in theM phase. By flow cytometric analysis, we found that E70l0 accumulated P388 cells in G2-M phase. E70l0 caused dose-dependent increases in the mitotic index in parallel with an inhibition of cellproliferation, like VCR.

Because the antimitotic effects of antimitotic agents may be mediated by disruption (VCR) or stabilization (Taxol) of microtubules inmitotic spindles, the effects of E70l0 on microtubule structure incolon 38 cells were examined, and E7010 was shown to cause thedisappearance of cytoplasmic microtubules and mitotic spindles (datanot shown). Therefore, the effects of E7010 and E70l0 derivatives onmicrotubule assembly in a cell-free system were examined and cornpared with the growth-inhibitory activities. These experiments clearlydemonstrated that the growth-inhibitory activity of E70l0 is causedby the inhibition of rnicrotubule assembly. Next, we examinedwhether E70l0 can bind to tubulin and compared the binding properties of E70l0 with those of VCR and CLC, although the structure ofE70l0 is quite different from those of the known tubulin binders,

Fig. 6. Effects of antitumor agents and antimitotic agents on E7OIOresistant P388 cell lines. P388 cells and E7OIO-resistant cell lines(0.6r-8, 0.5r-D, I Or-H, and 4.Or-M) were treated with various concentrations of the compounds for 3 days. Antiproliferative activity wasdetermined as indicated in “Materialsand Methods,―and IC,0 for eachcompound was calculated. Resistant factors were calculated as IC50 forthe resistant cell line:IC@0 for the sensitive cell line and shown graphically as the mean value ±SE (bars; n = 3—10).

which bind to the CLC site or Vinca site on tubulin, such as CLC,NOC, Vinca alkaloids, rhizoxin, halichondrin, and so on. We foundthat E7010 binds to purified tubulin and that the binding site is theCLC site, not the Vinca site, although the binding properties of E7010to tubulin were quite different from those of CLC. Therefore, E7010is quite distinct in its action from clinically used Vinca alkaloids orTaxol. The preliminary structure-activity study (Fig. 3B) reveals that

the addition of methoxyl groups to the lower phenyl ring decreases theactivity (compounds 2, 14, and 19), which is intriguing in relation tothe structural factors of other CLC site ligands such as CLC, andsuggests that the ligand recognition on tubulin may be differentbetween E7010 and CLC.

The antimitotic drugs, Vinca alkaloids and Taxol, are key drugs incancer chemotherapy, but drug resistance mediated by P-glycoproteinlimits their usefulness against MDR tumors that have acquired resistance in the course of chemotherapy or naturally (17—19).In this studyand in the previous one (3), we found that MDR cells selected byexposure to VCR did not show cross-resistance to E7010. In the invitro tumor panel study, E7010 seemed to be effective against naturally resistant MDR cell lines such as HCT15, DLD-1, andOVCAR-4. There was also no cross-resistance with other MDR celllines (KB 8—5,KB C-!, and KB C-2; data not shown). Because othertypes of drug resistance to VinCa alkaloids have been reported (20—22), we selected VCR-resistant P388/V4—4 cells, which did not showthe classical MDR phenotype, by exposure to VCR and verapamil.Interestingly, P388/V4—4 cells showed resistance to Vinca alkaloids(VCR, vinblastine, and vindesine) and trisnordolastatin, an analogueof dolastatin 10 that binds to the Vinca site on tubulin (13) but hadcollateral sensitivity to Taxol. Surprisingly, P3881V4—4cells were notcross-resistant to E70l0. Furthermore, to clarify the resistance to

III

d AU

III 4.Or-M

J.4 U 1.Or-H

0 O.5r-D

0 O.6r-8

@ t I I I I I t I I I t I I@ I I I I@ I I

p//,

L-PAM

Ara-C

MMC

NOC

POD

CLC

Taxoi

VCR

E7010

MDC

ADM

CDDP

5-FU

VP16

E7010

0.1 1 10Resistance factor

100 0.1 1 10 100Resistance factor

3212

llllllllllllllllllP

1nhT@.I

@-I El 4.Or-M

U 1.Or-H

0 O.5r-D

@l 0 O.6r-8

:Ji

l1lll1IIIllllllIllIll@'i

W//@

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

MECHANISM OF ACTION OF E7010

E7010, four E70l0-resistant P388 cell lines were selected by exposureto E7010. E7010-resistant P388 cell lines were not cross-resistant toVCR, but collateral sensitivity to Taxol was observed. The pattern ofthe cross-resistance to tubulin binders that bind at the CLC bindingsite was very interesting. The pattern of the cross-resistance to NOCacross the spectrum of four E70l0-resistant P388 cell lines wassimilar to that of E70l0, but the pattern of the cross-resistance to PODwas quite different. The pattern ofcross-resistance to CLC was similarto that of POD, although the resistance to CLC was minimum. Thesedifferences in the resistance pattern seem to be consistent with thedifferences of the binding properties (binding site, temperature dependency, rate of binding, and reversibility of binding) among tubulinbinders. These suggest the involvement of alterations of the target,tubulin itself or the polymerization state of tubulin, that specificallyaffect the binding of the E70l0-type agent to tubulin in E70l0-resistant cell lines, because the accumulation of 14C-labeled E70l0was similar in sensitive P388 cells and in E70l0-resistant cell lines(data not shown). CDDP-, 5-FU-, VP16-, and CPT-resistant cell linesalso lacked cross-resistance to E70l0 (data not shown).

Few CLC site binding agents are active in in vivo solid tumor models,although many compounds have been reported to bind to the CLC site ontubulin, such as CLC analogues (28), steganacin (29), POD analogues(30),combrestatinanalogues(31),benzimidazolederivatives(32),1,2-hydropyrido[3,4-b]pyrazine derivatives (33), 2-styrylquinazolin-4(3H)-one derivatives (34), 2,3@ihydro-2-(aiyl)4(lfI)-quinazolinone derivatives (35), and curacin A (36). We have shown here that E70l0 is aunique antimitotic agent with a broad spectrum ofantitumor activity. Thisdrug is considered to be a candidate for clinical use, either alone or incombination chemotherapy.

REFERENCES

I. Maren, T. H. Relations between structure and biological activity of sulfonamides.Annu. Rev. Pharmacol. Toxicol., 16: 309—327,1976.

2. Yoshino, H., Ueda, N., Niijima, J., Sugumi, J., Kotake, Y., Okada, T., Koyanagi, N.,Asada, M., Yoshimatsu, K., and Kitoh, K. Novel sulfonamides as potential, systemically active antitumor agents. J. Med. Chem., 35: 2496—2497, 1992.

3. Koyanagi, N., Nagasu, T., Fujita, F., Watanabe, T., Tsukahara, K., Funahashi, Y.,Fujita, M., Taguchi, T., Yoshino, H., and Kitoh, K. In viva tumor growth inhibitionproduced by a novel sulfonamide, E7O1O.against rodent and human tumors. CancerRes.,54: 1702—1706,1994.

4. Shinoda, H., Inaba, M., and Tsuruo, T. In viva circumvention of vincristine resistancein mice with P388 leukemia using a novel compound, AHC-52. Cancer Res., 49:1722—1726,1989.

5. Shelanski, M. L., Gaskin, F., and Cantor, C. R. Microtubule assembly in the absenceof added nucleotides. Proc. Natl. Aced. Sci. USA, 70: 765—768,1973.

6. Williams, R. C., Jr., and Detrich, H. M., 111.Separation of tubulin from microtubuleassociated proteins on phosphocellulose: accompanying alterations in concentrationsof buffer components. Biochemistry, 18: 2499—2503, 1979.

7. Borisy, 0. 0. A rapid method for quantitative determination of microtubule proteinusing DEAE-cellulose filters. Anal. Biochem.. 50: 373—385,1972.

8. Chen, Y-N., Mickley, L. A., Schwartz, A. M., Acton, E. M., Hwang, J., and Fojo.A. T. Characterizationof Adriamycin-resistanthumanbreastcancercells whichdisplay overexpression of a novel resistance-related membrane protein. J. Biol.Chem., 265: 10073—10080,1990.

9. Futscher,B. W., Foley,N. E., Gleason-Guzman,M. C., Meltzer,P. S., Sullivan,D. M.. and Dalton, W. S. Verapamil suppresses the emergence of P-glycoproteinmediated multidrug resistance. lot. J. Cancer, 66: 520—525, 1996.

10. Charcosset, J-Y. Effects of antineoplastic agents on the cell cycle progression. Biol.Cell, 58: 136—138,1986.

I I. Mitomo, Y., and Takamoto, S. Study of the effect of antitumor agents on the cellcycle traverse by flow cytometry. Jpn. J. Cancer Chemother., 15: 1019—1025,1988.

12. Cortese, F., Bhattacharyya, B., and Wolff, J. Podophyllotoxin as a probe for thecolchicine binding site of tubulin. J. Biol. Chem., 252: 1134—1140, 1977.

13. Bai, R., Pettit, G. R., and Hamel, E. Binding of dolastatin 10 to tubulin at a distinctsite for peptide antimitotic agents near the exchangeable nucleotide and Vincaalkaloid sites. J. Biol. Chem., 265: 17141—17149, 1990.

14. Bai, R., Paull, K. D., Herald, C. L., Malspeis, L., Pettit, G. R., and Hamel, E.Halichondrin B and homohalichondrin B. marine natural products binding in theVinca domain of tubulin. J. Biol. Chem., 266: 15882—15889,1991.

15. Iwasaki, S. Antimitotic agents: chemistry and recognition of tubulin molecule. Med.Res. Rev., 13: 183—198,1993.

16. Wheeler, G. P., Bowdon, B. J., Temple, C., Jr., Adamson, D. J., and Webster,J. Biologicaleffectsand structureactivityrelationshipsof l.2-dihydropyndol3,4-blpyrazines. Cancer Rca., 43: 3567—3575,1983.

17. Beck, W. T. The cell biology of multiple drug resistance. Biochem. Pharmacol., 36:2879—2887, 1987.

18. Moscow, J. A., and Cowan, K. H. Multidrug resistance. J. NatI. Cancer Inst., 80:14—20,1988.

19. Endicott, J. A., and Ling, V. The biochemistry of P-glycoprotein-mediated multidrugresistance. Annu. Rev. Biochem., 58: 137—171,1989.

20. Pain, J., Sirotnak, F. M., Barrveco, J. R., Yang, C. H., and Biedler, J. L. Alteredmolecular properties of tubulin in a multidrug resistant of Chinese hamstercells selected for resistance to Vinca alkaloids. J. Cell. Physiol., 136: 341—347,1988.

21. Ohta, S., Nishio, K., Kubo, S., Nishio, M., Ohmori, T., Takahashi, T., and Saijo, N.Characterization of a vindesine-resistant human small cell lung cancer cell line. Br. J.Cancer,68: 74—79,1993.

22. Debal, V., Allam, N., Morjani, H., Millot, J. M., Braguer, D., Breillout, F., andManfait, M. Characterization of a navelbine-resistant bladder carcinoma cell linecross-resistant to taxoids. Br. J. Cancer, 70: 1118—1125, I994.

23. Shioiri, T., Hayashi, K., and Hamada, Y. Stereoselective synthesis of dolastatin 10andits congeners.Tetrahedron,49: 1913—1924,1993.

24. Tsukagoshi, S., Niitani, H., Taguchi, T., Koyanagi, N., and Kitoh, K. A noveltubulin-interacting agent, E7010: its antitumor efficacies and the current clinicalstatus. Ann. Oncol., 55: 126 (A499), 1994.

25. Shoemaker, R. H. New approach to antitumor drug screening: the human colonyforming assay. Cancer Treat. Rep.. 70: 9—12,1986.

26. Branda, R. F., McCormack, J. J., and Perlmutter, C. A. Cellular pharmacology ofchloroquinoxaline sulfonamide and related compound in murine B 16 melanoma cells.Biochem. Pharmacol., 37: 4557—4564, 1988.

27. Wu. L., Smythe, A. M., Stinson, S. F., Mullendore, L. A., Monks, A., Schudiero,D. A., Paull, K. D., Koutsoukos,A. D., Rubinslein,L. V., Boyd, M. R., andShoemaker, R. H. Multiresistant pbenotype of disease-oriented panels of humantumor cell lines used for anticancer drug screening. Cancer Res., 52: 3029—3034,1992.

28. Dumont, R., Brossi, A., Chingnell. C. F., Quinn, F. R., and Suffness, M. A novelsynthesis of colchicine and analogues from thiocolchicine and congeners: reevaluation of colchicine as a potential antitumor agent. J. Med. Chem., 30: 732—735,1987.

29. Wang, R. W-J., Rebhun, L. I., and Kupchan. S. M. Antimitotic and antitubulinactivity of the tumor inhibitor steganacin. Cancer Res., 37: 3071—3079, 1977.

30. Batra, J. K., Kang, G. J., Jurd, L., and Hamel, E. Methylenedioxy-benzopyran analogsof podophyllotoxin, a new synthetic class of antimitotic agents that inhibit tubulinpolymerization. Biochem. Pttarmacol., 37: 2595—2602,1988.

31. Lin, C. M., Singh, S. B., Chu, P. S., Dempcy, R. 0., Schmidt, J. M., Pettit, G. R., andHamel, E. Interactions of tubulin with potent natural and synthetic analogs of theantimitotic agent combrestatin: a structure-activity study. Mol. Pharmacol.. 34: 200—208, 1988.

32. Hoebeke, J., Nijen, G. V., and Brabander, M. D. interaction of oncodazole (R17934),a new antitumoral drug, with rat brain tubulin. Biochem. Biophys. Res. Commun.. 69:319—324,1976.

33. Bowdon, B. J., Waud, W. R., Wheeler, G. P., Ham, R., Dansby, L., and Temple, C.,Jr. Comparison of l.2-dihydmpyrido[3,4-b)pyrazines (l-deaza-7.8-dihydropteridines)with several other inhibitors of mitosis. Cancer Res., 47: 1621—1626,1987.

34. Lin, C. M., Kang, G. J., Roach, M. C., Jiang, J. B., Hesson, D. P., Luduena, R. F.,and Hamel, E. Investigation of the mechanism of the interaction of tubulin withderivatives of 2-styrylquinazolin-4(3H)-one. Mol. Pharmacol., 40: 827—832,1991.

35. Hamel, E., Lin, C. M., Plowman, J., Wang, H. K., Lee, K. H., and Paull, K. D.Antitumor 2,3-dihydro-2-(aryl)-4( lH)-quinazolinone derivatives: interactions withtubulin. Biochem. Phannacol., 51: 53—59,1996.

36. Blokhin, A. V., Yoo, H-D., Geralds, R. S., Nagle, D. G., Gerwick, W. H., and Hamel,E. Characterization of the interaction of the marine cyanobacteria natural productcuracin A with the colchicine site of tubulin and initial structure-activity studies withanalogues. Mol. Pharmacol., 48: 523—531, I995.

3213

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1997;57:3208-3213. Cancer Res   Kentaro Yoshimatsu, Atsumi Yamaguchi, Hiroshi Yoshino, et al.   Colchicine Site of TubulinAntitumor Agent: Inhibition of Mitosis by Binding to the Mechanism of Action of E7010, an Orally Active Sulfonamide

  Updated version

  http://cancerres.aacrjournals.org/content/57/15/3208

Access the most recent version of this article at:

   

   

   

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

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/57/15/3208To request permission to re-use all or part of this article, use this link

on April 8, 2018. © 1997 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from