A Novel Microtubule Inhibitor Overcomes Multidrug ...Translational Science A Novel Microtubule Inhibitor Overcomes Multidrug Resistance in Tumors Nannan Ning1,Yamei Yu2, Min Wu3, Ruihong

Translational Science

A Novel Microtubule Inhibitor OvercomesMultidrug Resistance in TumorsNannan Ning1, Yamei Yu2, Min Wu3, Ruihong Zhang3, Ting Zhang4,Changjun Zhu5, Lei Huang6, Cai-Hong Yun7, Cyril H. Benes8, Jianming Zhang3,8,Xianming Deng4, Qiang Chen2, and Ruibao Ren3,9

Abstract

Microtubule inhibitors as chemotherapeutic drugs arewidely used for cancer treatment. However, the develop-ment of multidrug resistance (MDR) in cancer is a majorchallenge for microtubule inhibitors in their clinical imple-mentation. From a high-throughput drug screen using cellstransformed by oncogenic RAS, we identify a lead heteroarylamide compound that blocks cell proliferation. Analysis ofthe structure-activity relationship indicated that this seriesof scaffolds (exemplified by MP-HJ-1b) represents a potentinhibitor of tumor cell growth. MP-HJ-1b showed activitiesagainst a panel of more than 1,000 human cancer cell lineswith a wide variety of tissue origins. This compound depo-lymerized microtubules and affected spindle formation. Italso induced the spike-like conformation of microtubulesin vitro and in vivo, which is different from typical microtu-

bule modulators. Structural analysis revealed that this seriesof compounds bound the colchicine pocket at the intra-dimer interface, although mostly not overlapping withcolchicine binding. MP-HJ-1b displayed favorable pharma-cological properties for overcoming tumor MDR, both invitro and in vivo. Taken together, our data reveal a novelscaffold represented by MP-HJ-1b that can be developed as acancer therapeutic against tumors with MDR.

Significance: Paclitaxel is a widely used chemotherapeuticdrug in patients with multiple types of cancer. However,resistance to paclitaxel is a challenge. This study describes anovel class of microtubule inhibitors with the ability tocircumvent multidrug resistance across multiple tumor celllines. Cancer Res; 78(20); 5949–57. �2018 AACR.

IntroductionMicrotubule is considered as one of the most important targets

of anticancer drugs due to its important role in cell mitosis (1–3).Microtubule inhibitors, which inhibit cell growth by binding onmicrotubules and further influencing microtubules' function,have been used with great success in the treatment of variety oftumors (4, 5). Microtubule inhibitors are classified into differentcategories according to their binding sites: paclitaxel binding site,vinca alkaloid binding site, and colchicine binding site (6–8).Paclitaxel-binding site inhibitors disrupt mitosis by reinforcingmicrotubule and preventing the depolymerization of microtu-bule (9). Although inhibitors binding two other sites can depo-lymerize microtubule and block cell cycle in mitosis (10–13).These inhibitors, especially paclitaxel and vincristine, are widelyused as cancer chemotherapeutics. Unfortunately, more than halfcancer patients eventually would develop multidrug resistance(MDR) to chemotherapeutics (14, 15).

The most commonly characterized mechanism of MDR ismediated by the overexpression of ATP-binding cassette proteins(16–18). ABC proteins, as a kind of cell membrane transportproteins, reduce drug accumulation in cancer cells by pumpingdiverse compounds out. The key members of ABC transporterfamily include MDR1/P-gP/ABCB1 (MDR gene 1, P-glycopro-tein), MRPs/ABCCs (MDR-related proteins), and BCRP/ABCG2(breast cancer resistance protein; refs. 19, 20). With overexpres-sion of these transport pumps, the efflux of drug increased, whichresults in drug resistance. As MDR becomes an inevitable obstaclefor the success of chemotherapy, overcoming MDR has been ahigh priority for both clinical and investigation oncologists. Thestrategy of targeting ABC proteins has been virtually abandoned

1State Key Laboratory for Medical Genomics, Shanghai Institute of Hematology,Collaborative Innovation Center of Hematology, Collaborative InnovationCenter of System Biology, Ruijin Hospital, School of Life Sciences and Biotech-nology and School of Medicine, Shanghai Jiao Tong University, Shanghai, China.2State Key Laboratory of Biotherapy and Cancer Center, West China Hospital,Collaborative Innovation Center for Biotherapy, Sichuan University, Chengdu,China. 3State Key Laboratory for Medical Genomics, Shanghai Institute ofHematology, Collaborative Innovation Center of Hematology, CollaborativeInnovation Center of System Biology, Ruijin Hospital, Shanghai Jiao TongUniversity School ofMedicine, Shanghai, China. 4StateKey LaboratoryofCellularStress Biology, Innovation Center for Cell Signaling Network, School of LifeSciences, Xiamen University, Xiamen, Fujian, China. 5Tianjin Key Laboratory ofAnimal and Plant Resistance, College of Life Sciences, Tianjin Normal University,Tianjin, China. 6Key Laboratory of Cell Differentiation and Apoptosis of TheChinese Ministry of Education, Department of Pathophysiology, Shanghai JiaoTong University School of Medicine, Shanghai, China. 7Institute of SystemsBiomedicine, Department of Biophysics, School of Basic Medical Sciences,Peking University Health Science Center, Beijing, China. 8Massachusetts GeneralHospital, Charlestown, Massachusetts. 9Department of Biology, BrandeisUniversity, Waltham, Massachusetts.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Authors: Ruibao Ren, State Key Laboratory for Medical Geno-mics, Shanghai Institute of Hematology, Collaborative Innovation Center ofHematology, Collaborative InnovationCenter of SystemBiology, Ruijin Hospital,Shanghai Jiao Tong University School of Medicine, Shanghai, China, 197 Ruijin ErRoad, Shanghai, 200025 China. Phone: 86-21-64370045-610625; Fax: 781-736-2405; E-mail: [email protected]; Qiang Chen, Sichuan University,[email protected]; and Xianming Deng, Xiamen University,[email protected]

doi: 10.1158/0008-5472.CAN-18-0455

�2018 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 5949

on February 8, 2020. © 2018 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 22, 2018; DOI: 10.1158/0008-5472.CAN-18-0455

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-18-0455&domain=pdf&date_stamp=2018-9-29

http://cancerres.aacrjournals.org/

by the pharmaceutical industry because of the limitations of theavailable target-specific inhibitors, and the difficulty in designingclinical trials (21). Therefore, to seek alternative ways to circum-vent cancer drug resistance is urgently needed.

In this study, we describe a small-moleculeMP-HJ-1b as a novelmicrotubule inhibitor that binds on colchicine site. Mechanisti-cally, MP-HJ-1b restrains microtubule polymerization andpromotes microtubule depolymerization. The pharmaceuticaleffectiveness of MP-HJ-1b shows that it can overcome MDRin vitro and in vivo.

Materials and MethodsCell culture and proliferation assay

The HeLaR cell line was provided by Dr. Lei Huang (ShanghaiJiao Tong University School of Medicine, Shanghai, China) andcultured as described previously (22). The K562 doxorubicin-resistant cell line, A2780 paclitaxel-resistant cell line and A2780cell line were purchased from the Cell Center of Institutes ofBiomedical Sciences (Fudan University, China). The MDR andexpression of ABC transporters in these cell lines were tested anddescribed in the text. The RL95-2 and SK-NEP-1 cell lines werepurchased from the Center for Type Culture Collection (ChineseAcademy of Sciences, China). The RL95-2 cells were grownin DMEM: F12 Medium (Gibco, catalog #11330032) with 10%fetal bovine serum (Gibco, catalog #10099-141) and 0.005 mg/mL insulin. The SK-NEP-1 cells were grown in McCoy's 5aMedium (Gibco, catalog #16600082) with 10% FBS. The NB4,SEM, MOLM-13, NOMO-1, and SHI-1 cell lines were purchasedfrom Deutsche Sammlung von Mikroorganismen und Zellkultu-ren. The SEM and SHI-1 cells were grown in IMDM (Invitrogen,catalog #12440-053)with 10%FBS,whereasNB4,MOLM-13 andNOMO-1 cells were grown in RPMI-1640 Medium (Gibco,catalog #C22400500BT) with 10% FBS. The HeLa, MES-SA,A549, HCC827, H1975, MCF-7, HT-29, H9, MV-4-11, HL-60,KU812, K562, RS4;11 and SUP-B15 cell lines were purchasedfrom the ATCC. All cell lines were authenticated using shorttandem repeat profiling analysis according to the AmericanNational Standard ANS-0002-2011. The cell passages were lim-ited to 10 generations for all experiments in the study.Mycoplasmatesting was performed usingMycoFluorMycoplasmaDetection Kit(Invitrogen, catalog #M7006) every 3 weeks.

Cellular IC50 valuewasmeasured byCellTiter-Glo LuminescentCell Viability Assay (Promega, catalog #G7572). Cells were seed-ed in 96-well plates (5,000 cells in 100 mL complete medium perwell) with different doses of compounds (MP-HJ-1b, paclitaxel orvincristine) in triplicates. After 48 hours incubation, CellTiter-Glowas added into each well. After 30 minutes incubation at roomtemperature, the luminescent signal was measured, collected andanalyzed with EnVision Multimode Plate Reader.

Cell-cycle analysisA total of 5� 105 cells were cultured in a 10-cm dish. After 24

hours treatment with MP-HJ-1b or dimethyl sulfoxide(DMSO), cells were harvested and washed in PBS. 70% ethanolwas used to fix cells at �20�C overnight. Cells were washed twotimes in PBS. RNase was added and cells were incubated for 30minutes at 37�C. Propidium iodide (PI) labeling was per-formed for 10 minutes at room temperature prior to DNAcontent analysis by flow cytometry. Cell-cycle analysis wasperformed using FlowJo software.

Cell morphology analysisAfter 12 hours treatment of MP-HJ-1b or DMSO, cells were

harvested and washed with PBS for three times, and then fixedwith 3% paraformaldehyde (containing 0.2% sucrose) for 30minutes. Cells were blocked with 10% goat serum containing0.4% TritonX-100 for 1 hour at room temperature. Three hourstreatment of primary antibodies [anti-tubulin antibody (Abcam,catalog #ab6161), anti-crest antibody (Cortex Biochem, humanCREST serum), anti-CEP192 antibody (Santa Cruz Biotechnolo-gy, catalog #sc-84785) or anti-pericentrin antibody (Abcam,catalog #ab4448)] was conducted at room temperature. Afterwashing with PBS, the secondary antibodies [anti-mouse IgG(CST, catalog #4408), anti-rabbit IgG (CST, catalog #4413)] wereapplied for 1 hour at room temperature. Cells were washed withPBS and labeled with DAPI for 10 minutes at room temperature.Finally, anti-fade polyvinylpyrrolidone mounting medium wasused to seal coverslips.

Tubulin polymerization assayTubulin was re-dissolved in the cold G-PEM buffer

(80 mmol/L PIPES PH6.9, 2 mmol/L MgCl2, 0.5 mmol/LEGTA, 1 mmol/L GTP) and frozen in liquid nitrogen forstorage. Thawed-tubulin was centrifuged at 14,000 � g for10 minutes at 4�C before experiments. G-PEM, paclitaxel, orMP-HJ-1b was added into tubulin homogeneous solution.Absorbance at 340 nm was continuously measured for 1 hourwith Envision plate reader (PE).

Microtubule formation in vitroHomogeneous tubulin solution was prepared as mentioned

previously. MP-HJ-1b, paclitaxel or vincristine was added into thesolution and incubated for 20 minutes at room temperature.Tubulin samples were dropped onto the copper grids. The coppergrids were dried after staining by 2% uranyl acetate-lead citrate.Samples were observed by transmission electronmicroscope (FEITecnai G2 Spirit TEM).

Western blotThe whole-cell proteins were obtained by using mammalian

protein extraction reagent (M-PER) and centrifuged at 12,000 r.p.m. at 4�C. Proteins weremixed with loading buffer and boiled for10 minutes. Denatured proteins were separated in 8% sodiumdodecyl sulfate-polyacrylamide gel (SDS-PAAG) and transferredto the polyvinylidene fluoride (PVDF) membrane. Membraneswere incubated with antibodies [MDR1/ABCB1 rabbit antibody(CST, catalog #13342), MRP1/ABCC1 rabbit antibody(CST, catalog #14685), anti-MRP2 antibody (Abcam, catalog#ab3373), MRP3/ABCC3 rabbit antibody (CST, catalog#39909) or ABCG2 antibody (CST, catalog #4477)] at 4�C over-night. HRP-conjugated secondary antibodies were used and thespecific blots were detected by ECL reagent.

Mouse modelFemale nudemicewere fed in the Shanghai JiaotongUniversity

School of Medicine Experimental Animal Center with mimicnormal diurnal. Nude mice (4–6-weeks-old) were selected andhypodermically inoculated with 1 � 107 HeLa cells. Four weekslater, nude mice were grouped randomly. MP-HJ-1b, vincristine,or corn oil was administrated by intraperitoneal injection, respec-tively. HeLaR cells were inoculated into nude mice similarly.When tumor was large enough, the tumor tissue was surgically

Ning et al.

Cancer Res; 78(20) October 15, 2018 Cancer Research5950




extracted and transplanted to newnudemice. This process neededtwice. These tumor-bearing nude mice were stochasticallygrouped for drug treatment after tumor formed in 2 weeks.

All the animal experiments were guided and approved by TheAnimal Care and Welfare Committee of Shanghai Jiaotong Uni-versity School of Medicine.

Protein expression and purificationThe complex of two tubulins, one stathmin-like domain of RB3

(RB3-SLD) and one tubulin tyrosine ligase (TTL; the T2R–TTLcomplex) was produced as described with slight modifications(7, 23). RB3-SLDwasoverexpressed inEscherichia coliBL21 (DE3),purified sequentially by anion-exchange chromatography (QFF;GE Healthcare) and gel filtration (Superdex 75; GE-Healthcare).The purified protein was concentrated to 10 mg/mL and stored at�80�C until use. TTL was overexpressed in E. coli BL21 (DE3)and purified by nickel-affinity chromatography, followed by gelfiltration (Superdex 200; GE-Healthcare). Purified TTL in Bis-Trispropane (pH6.5), 200 mmol/L NaCl, 2.5 mmol/L MgCl2, 5mmol/L b-mercaptoethanol and 1% glycerol was concentratedto 20 mg/mL and stored at �80�C until use. Porcine braintubulin (Cytoskeleton, catalog #T-238P) was supplied at10mg/mL (buffer: 80 mmol/L Pipes, pH 6.9, 2.0 mmol/L MgCl2,0.5mmol/L EGTA and 1mmol/L GTP) and stored at�80�C untiluse. The T2R–TTL complex was prepared by mixing tubulin,RB3-SLD and TTL in a 2: 1.3: 1.2 molar ratio, and then1 mmol/L b, g-methyleneadenosine 50-triphosphate disodiumsalt, 5 mmol/L tyrosine and 10 mmol/L DTT were added and thecomplex was concentrated to 20 mg/mL at 4�C.

Crystallization and crystals soakingThe T2R-TTL crystals were obtained at 20�C in a buffer con-

sisting of 6% PEG4000, 8% glycerol, 0.1 mol/L MES (pH 6.7),30 mmol/L CaCl2 and 30 mmol/L MgCl2. Seeding method wasused to obtain single crystals. Rod-like crystals appeared after2 days and grew to maximum dimensions within 1 week. Forcrystal soaking, 0.1 mL of MP-HJ-1c (dissolved in DMSO at10mmol/L concentration) was added to a 2 mL crystal-containingdrop for 18 hours at 20�C.

Data collection and structure determinationThe reservoir solution supplemented with 20% (v/v) glycerol

was used as the cryoprotectant. The crystals were transferred intothe cryoprotectant for a few seconds, and then mounted in nylonloops and flash-cooled in liquid nitrogen. Diffraction data werecollected on beamline BL19U1 of National Facility for ProteinScience Shanghai (NFPS) at Shanghai Synchrotron RadiationFacility (Shanghai, China). Data were processed using HKL3000(24). The structures were determined by molecular replacementmethod using the T2R-TTL structure (PDB ID: 4I55) as a search

model. The refinement was performed using COOT (25) andPHENIX (26). Themodel qualitywas checkedwithMOLPROBITY(27). Atomic coordinates and structure factors of tubulin–MP-HJ-1c complex have been deposited in the Protein Data Bankunder accession code 5YZ3.

Statistical analysisAll data are expressed as the mean � standard error (SEM) of

three independent experiments unless stated otherwise. The datawere analyzedwith Student t tests. P values <0.05were consideredsignificance.

ResultsIdentification of MP-HJ-1b

In an effort to conquer the seemingly impregnable fortressof oncogenic RAS, we constructed a phenotypic screening tosearch for compounds that are able to inhibit the proliferationof NRASG12D-transformed cells (28). In brief, by comparingIL-3–dependent BaF3 cell proliferation with that of mutantRAS-driven BaF3 cell, this cell-based screening system couldprovide an unbiased search for cytotoxic compounds that pref-erentially target RAS signaling. We screened an in-house com-pound library comprisedmore than 30,000 chemical entities. Thecompound library was made of small molecules with drug-likeproperties, which were generated by combinatorial chemistryspecific designed for high-throughput screening (29). One

Figure 1.

Compound structures. A, Chemicalstructure of MP-HJ-1a. B, Chemicalstructure of MP-HJ-1b.

Table 1. Antiproliferative IC50 values of MP-HJ-1b against cancer cell lines

Cells Tumor types Ras mutant IC50(mmol/L)

HeLa Cervical cancer — 0.024MES-SA Uterine sarcoma — 0.045RL95-2 Endometrial cancer HRAS-Q61H 0.017A549 Lung cancer KRAS-G12S 0.040HCC827 Lung cancer — 0.068H1975 Lung cancer — 0.151A2780 Ovarian cancer — 0.043SK-NEP-1 Nephroblastoma — 0.026MCF-7 Breast cancer — 0.013HT-29 Colon cancer — 0.104H9 Lymphoma NRAS-Q61K 0.122NB4 Acute myelogenous leukemia KRAS-A18D 0.011SEM Acute myelogenous leukemia — 0.075MV-4-11 Acute myelogenous leukemia — 0.111MOLM-13 Acute myelogenous leukemia — 0.089NOMO-1 Acute myelogenous leukemia KRAS-G13D 0.067SHI-1 Acute myelogenous leukemia KRAS-Q61H 0.057HL-60 Acute myelogenous leukemia NRAS-Q61L 0.090KU812 Chronic myeloid leukemia — 0.055K562 Chronic myeloid leukemia — 0.059RS4;11 Acute lymphoblastic leukemia — 0.037SUP-B15 Acute lymphoblastic leukemia — 0.075

Microtubule Inhibitor Overcoming Multidrug Resistance

www.aacrjournals.org Cancer Res; 78(20) October 15, 2018 5951




compoundMP-HJ-1a (Fig. 1A) was identified as the top lead for itsinhibitory effects on the proliferation of NRASG12D-driven BaF3cells.On thebasis of the core structureofMP-HJ-1a,weput intensivemedicinal chemistry efforts to generate a series of analogues. Com-pound MP-HJ-1b (Fig. 1B) was eventually chosen as the leadingcompound for following pharmacological characterization.

To investigate whether MP-HJ-1b is specific effective to RAS-mutant cells, 22 human cell lines, from different types of tumors,were initially tested withMP-HJ-1b and the viability was detectedby CellTiter-Glo Luminescent Cell Viability Assay (Table 1). Toour surprise, MP-HJ-1b showed strong anti-proliferation effectson all tumor cell lines. These data suggested that MP-HJ-1b hasan interesting property of general cytotoxicity instead of RAS-mutant-specific killing. In addition, the anti-proliferative effect of

MP-HJ-1b was also observed with elevated IC50 values on othernon-oncogenic cells (Supplementary Table S1; SupplementaryFig. S1). This result has been further verified overmore than 1,000tumor cell lines (Supplementary Excel file), which lead us tohypothesize that MP-HJ-1b might have interrupted essentialcomponents of cancer cells growth and survival.

MP-HJ-1b blocks mitosis by inhibiting microtubulepolymerization

To understand the deleterious effect of MP-HJ-1b on cellsviability, we examined the influence of MP-HJ-1b on differentcellular events by monitoring cell cycle with flow cytometer.Typically, MP-HJ-1b treated HeLa cells showed cell cycle arrestat G2–M phase (Fig. 2A). The mitotic blocking function of

Figure 2.

MP-HJ-1b arrests cell cycle in prometaphase by inhibiting microtubules formation. A, PI-marked cell-cycle analysis by flow cytometry after compoundtreatment (DMSOor 200nmol/LMP-HJ-1b).B,CellularmorphologyofHeLa cells inmitosis (red, centromere; green,a-tubulin; blue, chromosome).C–E,Microtubuleswere degraded and became spots but not centrosomes andMTOC (centrosome: CEP192, Pericentrin; MTOC: g-tubulin). F andG,CompoundMP-HJ-1b and vincristinesuppressed tubulin polymerization in dose-dependent manners.

Ning et al.





MP-HJ-1b was further verified from other cell lines (Supplemen-tary Fig. S2). Consistently,MP-HJ-1bpromoted cell apoptosis andincreased cell death (Supplementary Fig. S3).

To clarify the specific period of cell-cycle blockage, weperformed immunofluorescence-based imaging analysis toMP-HJ-1b–treated cells. As shown in Fig. 2B, the metaphase celldisplays a bipolar spindle apparatus and chromosomes lined upalong the equatorial plate. After MP-HJ-1b treatment, the normalspindle was disappeared. Interestingly, several tubulin-spots,scattered chromosomes and centromeres were present in the cell(Fig. 2B; Supplementary Fig. S4). This phenomenon suggests thatthe cell was arrested in pro-metaphase according to the disap-pearance of nuclear membrane and characterization of chromo-somes and centromeres.

On the basis of this observation, we hypothesized thatMP-HJ-1b might have interfered cell cytoskeleton system. Wewere specifically interested in exploring which composition ofthese tubulin-spots was abrupted by MP-HJ-1b, because all threecommon subtypes of tubulin: a-tubulin, b-tubulin and g-tubulincan constitutemicrotubule, spindle, centrosome andmicrotubuleorganizing center (MTOC) in mitosis, respectively. The existenceof two centrosomes in each cell after the compound treatmentindicated that the spike-like spots were not centrosomes (Fig. 2Cand D). MTOC, as the beginning of microtubule, is centered onthe g-tubulin andmediates tubulin nucleation phenomenon andreplaces centrosome in cells without centriole. Next, we labeledg-tubulin and a-tubulin at the same time (Fig. 2E). As the resultsshowed in Fig. 2E, the spike-like spots are not MTOC. These datasuggest thatMP-HJ-1b blocksmitosis by influencingmicrotubule.

To investigate whether compound MP-HJ-1b affects thetubulin polymerization, a cell-free in vitro tubulin polymerizationassay was performed. Dissolved tubulin polymerizes gently in

vitro at 37�C with supply of GTP. This process was acceleratedwhen paclitaxel (a microtubule stabilizing agent) presented(Fig. 2F and G). On the contrast, vincristine inhibited polymer-ization reaction and depolymerized microtubule dramatically(Fig. 2F). Interestingly, MP-HJ-1b showed the same depolymer-ization effect as that of vincristine in a dose-dependent manner(Fig. 2G). These data lead us to hypothesize that MP-HJ-1bsuppresses microtubule polymerization via direct binding to thetubulin heterodimer.

MP-HJ-1b alters microtubule in a distinct mechanismTo further examine the biophysical impact of MP-HJ-1b on

microtubule, we compared it with vinca alkaloids and paclitaxel,in which both are the most widely used microtubule-relatedantitumor drugs. Paclitaxel is a microtubule stabilizing agent andpromotes microtubule polymerization. Although vinca alkaloidsinhibit microtubule polymerization and promote microtubuledepolymerization. Both MP-HJ-1b and vincristine blocked mito-sis and arrested cells in pro-metaphase (Fig. 3A). Disorderedchromosomes and dissembled nuclear envelope surroundedscattered chromosomes were observed in cell treated with bothcompounds. A close examination revealed that MP-HJ-1binduced a spike-like structure in the compounds treated cells.

We further verified the inhibitory effect of MP-HJ-1b on thepolymerization of microtubules in vitro by transmission electronmicroscope (TEM). The a-tubulin and b-tubulin built a hollowtubular structure at room temperature in the present of guanosinetriphosphate (GTP; Fig. 3B), and paclitaxel could stabilize thisstructure (Fig. 3C). In contrast, there was no visible microtubulestructure when vincristine was added at the beginning of thereaction (Fig. 3D). Although these globular proteins, a-tubulinand b-tubulin, were aggregated smooth and irregular pellets when

Figure 3.

Alteration of microtubules in cells and in vitro caused by tubulin inhibitors. A, Compound MP-HJ-1b (200 nmol/L), vincristine (50 nmol/L), and colchicines(250 nmol/L) depolymerize microtubules in HeLa cells. B–E, Appearance of tubulin formation was monitored by TEM. 1 mmol/L paclitaxel, 8 mmol/L vincristine,or 20 mmol/L compound MP-HJ-1b.






MP-HJ-1b added (Fig. 3E). To our knowledge, this kind of tubulinconformation has not been reported. Compound MP-HJ-1binduced formation of a new conformation of microtubulesalthough the efficiency was lower than vincristine (Fig. 2Fand G). These results suggested that MP-HJ-1b uses a differentmechanism to block microtubule's normal function from that ofvincristine.

Crystal structure of the tubulin–MP-HJ-1c complexTo understand the molecular mechanism of MP-HJ-1b on

inhibition of microtubule polymerization, we performed a struc-tural analysis. CompoundMP-HJ-1c, an analog ofMP-HJ-1bwiththe comparable pharmacological efficacy (Supplementary Fig. S5)and better solubility was used. The crystal structure of a proteincomplex composed of a/b-tubulin, the stathmin-like proteinRB3 and tubulin tyrosine ligase (T2R–TTL), complexed withMP-HJ-1c, was solved at 2.55 Å resolution (Fig. 4A). Detailsof the data collection and refinement statistics are summarizedin Supplementary Table S2.

The high-resolution and clear density map enabled us todetermine the position and orientation of the small-moleculeinhibitor unambiguously (Fig. 4B), thus revealing the detailedinteractions between MP-HJ-1c and tubulin (Fig. 4C). As seen inthe crystal structure, MP-HJ-1c bound to the colchicine-bindingsite, which is a big pocket surrounded by a super b-sheet andtwo a-helices, and is capped by two loops (Fig. 4C). The agentMP-HJ-1c made hydrogen bonds with the side chains of bN165,bE198 and bY200, and the main-chain oxygen atom of bV236(Fig. 4C). MP-HJ-1c also made extensive hydrophobic interac-tions with b-tubulin (Fig. 4C). A structure-based pharmacophoremodel for CBSIs has been proposed recently (30), which consistsof three hydrophobic centers (I, II, and III) and two hydrogenbond centers (IV and V). MP-HJ-1c occupied two hydrophobiccenters I and II, and one hydrogen bond center IV (Fig. 4C).

Comparedwith colchicine,MP-HJ-1c was locatedmuch deeperin the b subunit and made no interaction with the a subunit(Fig. 4D). Very little overlap with colchicine was observed. Com-parison between tubulin–MP-HJ-1c and tubulin–colchicine com-plex structures showed that the bindingof the different colchicine-site ligands did not affect the global conformation of tubulin norof the T2R complex. The root mean square deviation (RMSD) for2,146 Ca atoms between tubulin–MP-HJ-1c and tubulin–colchi-cine complexes is 0.33 Å. The major conformational changesconcern the aT5 loop of the colchicine domain (Fig. 4D).

MP-HJ-1b overcomes MDR in vitro and in vivoAs MP-HJ-1b could efficiently block cell migration and colony

formation (Supplementary Fig. S6; Supplementary Fig. S7), wenext investigated whether MP-HJ-1b is able to overcome MDR incancer, as MDR has posed a major challenge for cancer chemo-therapeutic agents in clinical treatment. Three pairs of cell lines,including HeLa (cervical cancer) and HeLaR (paclitaxel resistanceHeLa), A2780 (ovarian cancer), andA2780R (paclitaxel resistanceA2780), K562 (chronic myelogenous leukemia, CML), andK562R (doxorubicin resistance K562) were used for furtherexperiments. The drug efflux is regarded as the main mechanismofMDR.We first checked the expression ofMDR1/ABCB1,MRP1/ABCC1, MRP2/ABCC2 and MRP3/ABCC3 in those cells (Fig. 5A;Supplementary Fig. S8). We then evaluated the IC50 values ofpaclitaxel, vincristine, and MP-HJ-1b on these three pairs of cells.The IC50 values of each compound on different cell lines were

shown on Fig. 5B; Supplementary Table S3. MP-HJ-1b clearlyshowed a distinct pharmacological efficacy from these two com-pounds by sufficiently suppressing MDR in different type oftumor cell lines.

As MP-HJ-1b showed promising pharmacological propertieswith half-life of more than 9 hours (Supplementary Fig. S9), wefurther performed in vivo experiments to examine the relevantpharmacological antitumor efficacyofMP-HJ-1busing vincristineas a reference. We built three subcutaneous xenograft models bytransplantedHeLa,HeLaRor K562 cells into nudemice.When thetumor volume reached approximately 100 mm3, the nude micewere randomly assigned to three groups that were intraperitonealadministered 25 mg/kg MP-HJ-1b, 0.2 mg/kg vincristine, or cornoil, respectively. For tumors derived from HeLa and K562 cells,both compounds inhibited tumor growth and showed excellentanticancer activity (Fig. 5C, Supplementary Tables S4–S5; Sup-plementary Figs. S10–S11). Consistent with cellular proliferation

Figure 4.

Crystal structure of MP-HJ-1c complexed with tubulin. A, Overall structure ofMP-HJ-1c–tubulin complex. RB3-SLD, green; TTL, yellow; a-tubulin, black;b-tubulin, gray; GTP, red; GDP, orange; MP-HJ-1c, cyan. GTP, GDP, and MP-HJ-1care shown as spheres. B, The chemical structure of an analog derived from MP-HJ scaffold, MP-HJ-1c, and its electron densities. The Fo-Fc omit map is gray andcontoured at 3s.C, Interaction between MP-HJ-1c and tubulin. Sticks, MP-HJ-1c.Residues that make interactions with MP-HJ-1c are shown as sticks and arelabeled. The hydrophobic centers are indicated by green semitransparent cycles(small, center I; big, center II), and the hydrogen bond center is indicated by ayellow semitransparent cycle (center V). D, Comparison of the tubulin-bindingmodes between MP-HJ-1c and colchicine. The complex structures oftubulin–MP-HJ-1c (cyan) and tubulin-colchicine (PDB ID: 4O2B, yellow) aresuperimposed. MP-HJ-1c and colchicine are shown as sticks.

Ning et al.





results, the tumor growth was also inhibited by MP-HJ-1b in theHeLaR xenograft model but not by vincristine (Fig. 5C; Supple-mentary Table S4). In addition, there were no body weight lostand observed side effects during the drug treatment (Fig. 5D;Supplementary Fig. S10), suggesting the tolerability ofMP-HJ-1b.The results show that MP-HJ-1b overcomes MDR both in vivo andin vitro.

DiscussionTargeting microtubule remains the first line ammunition for

many clinicians, even with more selective approaches available

for cancer treatment. Because of the essential role of microtubuleplays in cancer cells and existence of several difficulty-to-treatmalignancies, such as pancreatic cancer, microtubule inhibitors,paclitaxel and vincristine are commonly used in the treatment ofvarieties cancers, including breast carcinoma, ovarian cancer,acute leukemia, malignant lymphoma, lung cancer (31–33).Specifically, the results of recent clinical studies indicated thatmicrotubule inhibitors nab-paclitaxel combining with gemcita-bine prolongs survival of pancreatic ductal adenocarcinomapatients. Authors have proposed that nab-paclitaxel may play arole for targeting KRAS by disrupting its intracellular trafficking(34). Even though the primary observation indicated that

Figure 5.

In vitro and in vivo efficacy of MP-HJ-1b in both wild-type and MDR cell lines. A, MDR-related proteins expression increased in three resistant cell lines. B, IC50

values of paclitaxel, vincristine, and compound MP-HJ-1b on three pairs cell lines (HeLaR, paclitaxel resistance HeLa; A2780R, paclitaxel resistance A2780;K562R, doxorubicin resistance K562). C, Compound MP-HJ-1b and vincristine suppress tumor volume in HeLa xenograft or HeLaR xenograft models. D, The weightof each group of nude mice during treatment.






MP-HJ-1b might not be a RAS-mutant–specific compound, thepotential role of MP-HJ-1b in RAS protein trafficking is worth toexplore in the future.

Second, the development of MDR in cancer is an unavoidablechallenge for microtubule inhibitors for their clinical implemen-tation. ABC transporters play the crucial role in the developmentof MDR. Even though MDR1/ABCB1 is considered as one of themost common and critical pumps for drug efflux and reducingintracellular drug accumulation, it is extremely difficult to targetABC transporters in overriding cancer MDR. Because the com-plexity of drug pumps expression pattern and functional redun-dancy greatly limit the utility of those ABC inhibitors in cancertherapy (35–37). Thus to develop a mechanistic different micro-tubule inhibitors could be clinically beneficiary to circumventcancer drug resistance. Our data reveal that MP-HJ-1b binds tomicrotubule colchicine-binding site, which has no drug beingapproved to target this site so far. It is interesting to observe thatcolchicine-binding sitemicrotubule inhibitors can act as vascular-targeting agents. Therefore, besides used as antimitotic agents,colchicine-binding site compounds can rapidly depolymerizemicrotubules of newly formed vasculature to shut down theblood supply to tumors (38, 39). The potent in vivo pharmaco-logical efficacy of MP-HJ-1b need to be further explored on itsinhibition on vascular-targeting function.

As we know that currently most microtubule targeted com-pounds have been discovered from natural products. It is difficultand expensive to make and optimize those natural products forclinic use. Obviously, it can save some extra financial and envi-ronmental burden for cancer patients as well as pharmaceuticalindustry if there were available small-molecule microtubule inhi-bitors in the market. Thus, to develop small-molecule inhibitors,such as MP-HJ-1b would be urgent needed.

It is still unclear that different tubulin targeting agents, such astaxanes and vincas show separate antitumor effects. However, thistissue-specific inhibitory mechanisms provides a rationale todevelop novel approaches aim to improve on existing com-pounds by profiling tumor sensitivities, that could reduce sideeffects such as peripheral neuropathy toxicity (1). With moreavailable compounds targeting the numerous other componentsof the tubulin–microtubule complex certainly would be animportant aspect of fully exploiting its anticancer potential inthat the drug synergistic effect can be achieved by combining twoor more targets in the same system.

Nevertheless, the feedback from bedside clearly indicated anurgent need of search new strategy of targeting microtubule andovercoming MDR in human cancer treatment. The series ofsmall molecules represented by MP-HJ-1b showed promisingpharmacological efficacy on overriding MDR both in vitro andin vivo. The molecular mechanism of MP-HJ-1b's inhibitoryfunction is different from previous reported anti-MDR drugs bybinding to the colchicine-binding site of tubulin, consequentlydepolymerizesmicrotubules and affects formation of the spindle.Our data reveal a novel scaffold that could be further developed ascancer therapeutics, especially for MDR cancers.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: N. Ning, M. Wu, C.-H. Yun, X. Deng, Q. Chen, R. RenDevelopment of methodology:N. Ning, M.Wu, C. Zhu, C.-H. Yun, C.H. BenesAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): N. Ning, R. Zhang, C.H. BenesAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):N. Ning, Y. Yu, T. Zhang, J. Zhang, X. Deng, Q. Chen,R. RenWriting, review, and/or revision of the manuscript: N. Ning, C.H. Benes,J. Zhang, X. Deng, Q. Chen, R. RenAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): N. Ning, Y. Yu, M. Wu, R. Zhang, L. Huang,X. Deng, R. RenStudy supervision: X. Deng, R. Ren

AcknowledgmentsThis work was supported by the National Key Research and Development

Program (2016YFC0902800 to R. Ren), the Key Project of Natural ScienceFoundation of China (81230055 to R. Ren), the Samuel Waxman CancerResearch Foundation CoPI Program, the National Natural Science Foundationof China (No. U1405223 to X. Deng), the Fundamental Research Funds for theCentral Universities of China (No. 20720160064 to X. Deng), and the KeyProject of Natural Science Foundation of China (81672722 to Q. Chen).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 10, 2018; revised July 12, 2018; accepted August 16, 2018;published first August 22, 2018.

References1. JordanMA,Wilson L.Microtubules as a target for anticancer drugs. Nat Rev

Cancer 2004;4:253–65.2. Schmidt M, Bastians H. Mitotic drug targets and the development of novel

anti-mitotic anticancer drugs. Drug Resist Updat 2007;10:162–81.3. Berges R, Tchoghandjian A, Honore S, Esteve MA, Figarella-Branger D,

Bachmann F, et al. The Novel Tubulin-Binding Checkpoint ActivatorBAL101553 Inhibits EB1-dependentmigration and invasion andpromotesdifferentiation of glioblastoma stem-like cells. Mol Cancer Ther2016;15:2740–9.

4. Downing KH, Nogales E. Tubulin structure: insights into microtubuleproperties and functions. Curr Opin Struct Biol 1998;8:785–91.

5. Seve P, Reiman T, Dumontet C. The role of betaIII tubulin in predictingchemoresistance in non-small cell lung cancer. Lung Cancer 2010;67:136–43.

6. Kupchan SM, Komoda Y, Court WA, Thomas GJ, Smith RM, Karim A, et al.Maytansine, a novel antileukemic ansa macrolide from Maytenus ovatus.J Am Chem Soc 1972;94:1354–6.

7. Prota AE, Bargsten K, Zurwerra D, Field JJ, Diaz JF, Altmann KH, et al.Molecular mechanism of action of microtubule-stabilizing anticanceragents. Science 2013;339:587–90.

8. Kavallaris M. Microtubules and resistance to tubulin-binding agents. NatRev Cancer 2010;10:194–204.

9. Perez EA.Microtubule inhibitors: differentiating tubulin-inhibiting agentsbased on mechanisms of action, clinical activity, and resistance. MolCancer Ther 2009;8:2086–95.

10. Kruczynski A, Barret JM, Etievant C, Colpaert F, Fahy J, Hill BT. Antimitoticand tubulin-interacting properties of vinflunine, a novel fluorinated Vincaalkaloid. Biochem Pharmacol 1998;55:635–48.

11. Toso RJ, Jordan MA, Farrell KW, Matsumoto B, Wilson L. Kinetic stabili-zation of microtubule dynamic instability in vitro by vinblastine.Biochemistry 1993;32:1285–93.

12. Ravelli RB, Gigant B, Curmi PA, Jourdain I, Lachkar S, Sobel A, et al. Insightinto tubulin regulation froma complexwith colchicine and a stathmin-likedomain. Nature 2004;428:198–202.


Ning et al.




13. Pommier Y, Marchand C. Interfacial inhibitors: targeting macromolecularcomplexes. Nat Rev Drug Discov 2011;11:25–36.

14. Bray F, Jemal A, Grey N, Ferlay J, Forman D. Global cancer transitionsaccording to the Human Development Index (2008–2030): a population-based study. Lancet Oncol 2012;13:790–801.

15. TredanO,Galmarini CM, Patel K, Tannock IF.Drug resistance and the solidtumor microenvironment. J Natl Cancer Inst 2007;99:1441–54.

16. Fojo AT, Menefee M. Microtubule targeting agents: basic mechanisms ofmultidrug resistance (MDR). Semin Oncol 2005;32:S3–8.

17. Rees DC, Johnson E, Lewinson O. ABC transporters: the power to change.Nat Rev Mol Cell Biol 2009;10:218–27.

18. Locher KP. Structure andmechanismof ABC transporters. CurrOpin StructBiol 2004;14:426–31.

19. Rumjanek VM, Trindade GS, Wagner-Souza K, de-Oliveira MC, Marques-Santos LF, Maia RC, et al. Multidrug resistance in tumour cells: character-ization of the multidrug resistant cell line K562-Lucena 1. An Acad BrasCienc 2001;73:57–69.

20. Annilo T, Chen ZQ, Shulenin S, Dean M. Evolutionary analysis of a clusterof ATP-binding cassette (ABC) genes.Mammalian Genome 2003;14:7–20.

21. GottesmanMM, Lavi O,Hall MD, Gillet JP. Toward a better understandingof the complexity of cancer drug resistance. Annu Rev Pharmacol Toxicol2016;56:85–102.

22. Jin W, Liao X, Lv Y, Pang Z, Wang Y, Li Q, et al. MUC1 induces acquiredchemoresistance by upregulating ABCB1 in EGFR-dependentmanner. CellDeath Dis 2017;8:e2980.

23. Charbaut E, Curmi PA, Ozon S, Lachkar S, Redeker V, Sobel A. Stathminfamily proteins display specific molecular and tubulin binding properties.J Biol Chem 2001;276:16146–54.

24. Minor W, Cymborowski M, Otwinowski Z, Chruszcz M. HKL-3000: theintegration of data reduction and structure solution–from diffractionimages to an initial model in minutes. Acta Crystallogr D Biol Crystallogr2006;62:859–66.

25. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics.Acta Crystallogr D, Biol Crystallogr 2004;60:2126–32.

26. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW, Echols N, et al.PHENIX: a comprehensive Python-based system for macromolecularstructure solution. Acta Crystallogr D Biol Crystallogr 2010;66:213–21.

27. Chen VB, Arendall WB 3rd, Headd JJ, Keedy DA, Immormino RM, KapralGJ, et al. MolProbity: all-atom structure validation for macromolecularcrystallography. Acta Crystallogr D Biol Crystallogr 2010;66:12–21.

28. WuM,Huang J, Zhang J, BenesC, JiaoB, RenR.N-ArachidonoylDopamineInhibits NRAS neoplastic transformation by suppressing its plasma mem-brane translocation. Mol Cancer Ther 2017;16:57–67.

29. Jiang J, Gui F, He Z, Li L, Li Y, Li S, et al. Targeting BRK-positive breastcancers with small-molecule kinase inhibitors. Cancer Res 2017;77:175–86.

30. Wang Y, ZhangH,Gigant B, Yu Y,WuY,ChenX, et al. Structures of a diverseset of colchicine binding site inhibitors in complex with tubulin provide arationale for drug discovery. FEBS J 2016;283:102–11.

31. Ojima I, Chen J, Sun L, Borella CP, Wang T, Miller ML, et al. Design,synthesis, and biological evaluation of new-generation taxoids. J MedChem 2008;51:3203–21.

32. Mekhail TM, Markman M. Paclitaxel in cancer therapy. Expert OpinPharmacother 2002;3:755–66.

33. Dumontet C, Jordan MA. Microtubule-binding agents: a dynamic field ofcancer therapeutics. Nat Rev Drug Discov 2010;9:790–803.

34. Basseville A, Bates S, Fojo T. Pancreatic cancer: Targeting KRAS andthe vitamin D receptor via microtubules. Nat Rev Clin oncol2015;12:442–4.

35. Binkhathlan Z, Lavasanifar A. P-glycoprotein inhibition as a therapeuticapproach for overcomingmultidrug resistance in cancer: current status andfuture perspectives. Curr Cancer Drug Targets 2013;13:326–46.

36. Krishna R, de Jong G, Mayer LD. Pulsed exposure of SDZ PSC 833 tomultidrug resistant P388/ADR and MCF7/ADR cells in the absence ofanticancer drugs can fully restore sensitivity to doxorubicin. Anticancer Res1997;17:3329–34.

37. Robert J, Jarry C. Multidrug resistance reversal agents. J Med Chem2003;46:4805–17.

38. Davis PD, Dougherty GJ, Blakey DC, Galbraith SM, Tozer GM, Holder AL,et al. ZD6126: a novel vascular-targeting agent that causes selectivedestruction of tumor vasculature. Cancer Res 2002;62:7247–53.

39. Hamel E, Lin CM, Plowman J, WangHK, Lee KH, Paull KD. Antitumor 2,3-dihydro-2-(aryl)-4(1H)-quinazolinone derivatives. Interactions withtubulin. Biochem Pharmacol 1996;51:53–9.






2018;78:5949-5957. Published OnlineFirst August 22, 2018.Cancer Res Nannan Ning, Yamei Yu, Min Wu, et al. TumorsA Novel Microtubule Inhibitor Overcomes Multidrug Resistance in

Updated version

10.1158/0008-5472.CAN-18-0455doi:

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/78/20/5949.full#ref-list-1

This article cites 39 articles, 7 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/78/20/5949To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-18-0455

http://cancerres.aacrjournals.org/content/78/20/5949.full#ref-list-1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/78/20/5949


Documents

A Novel Microtubule Inhibitor Overcomes Multidrug ...Translational Science A Novel Microtubule Inhibitor Overcomes Multidrug Resistance in Tumors Nannan Ning1,Yamei Yu2, Min Wu3, Ruihong