An Anti-Wnt5a Antibody Suppresses Metastasis of Gastric ... · Cancer Cells In Vivo by Inhibiting Receptor-Mediated Endocytosis Hideaki Hanaki1,2, Hideki Yamamoto 1, Hiroshi Sakane

Therapeutic Discovery

An Anti-Wnt5a Antibody Suppresses Metastasis of GastricCancer Cells In Vivo by Inhibiting Receptor-MediatedEndocytosis

Hideaki Hanaki1,2, Hideki Yamamoto1, Hiroshi Sakane1, Shinji Matsumoto1, Hideki Ohdan2,Akira Sato1, and Akira Kikuchi1

AbstractWnt5a is a representative ligand that activates the b-catenin–independent pathway inWnt signaling. It was

reported that the expression of Wnt5a in human gastric cancer is associated with aggressiveness and poor

prognosis and that knockdown ofWnt5a reduces the ability of gastric cancer cells to metastasize in nudemice.

Therefore, Wnt5a and its signaling pathway might be important targets for the therapy of gastric cancer. The

aim of this studywas to examinewhether an anti-Wnt5a antibody affectsmetastasis of gastric cancer cells. One

anti-Wnt5a polyclonal antibody (pAb5a-5) inhibited migration and invasion activities in vitro of gastric cancer

cells with a high expression level ofWnt5a. Previously, it was shown thatWnt5a induces the internalization of

receptors, which is required for Wnt5a-dependent activation of Rac1. pAb5a-5 inhibited Wnt5a-dependent

internalization of receptors, thereby suppressed Wnt5a-dependent activation of Rac1. Laminin g2 is one of

target genes of Wnt5a signaling and Rac1 was involved in its expression. pAb5a-5 also inhibited Wnt5a-

dependent expression of laminin g2. In an experimental liver metastasis assay, gastric cancer cells were

introduced into the spleens of nudemice. Laminin g2 was required for liver metastatic ability of gastric cancer

cells in vivo. Furthermore, intraperitoneal injection of pAb5a-5 inhibited the metastatic ability of gastric cancer

cells. These results suggest that an anti-Wnt5a antibody was capable of suppressing Wnt5a-dependent

internalization of receptors, resulting in the prevention of metastasis of gastric cancer cells by inhibiting the

activation of Rac1 and the expression of laminin g2. Mol Cancer Ther; 11(2); 298–307. �2011 AACR.

Introduction

The Wnt protein family plays roles in embryogenesisand carcinogenesis (1). Wnt activates at least 2 intracel-lular signaling pathways, the b-catenin–dependent andb-catenin–independent pathways (2–4). Abnormal acti-vation of the b-catenin–dependent pathway because ofgenetic alteration in b-catenin, adenomatous polyposiscoli (APC), or Axin can be associated with human cancer(5). Recent evidence has clarified that the b-catenin–inde-pendent pathway is also involved in tumorigenesis (6, 7).Wnt5a is a representative of the Wnt protein family thatactivates the b-catenin–independent pathway, which pri-

marily modulates cell movement and polarity (2, 6, 7). Bybinding to receptors, Frizzled (Fz) and receptor tyrosinekinase–like orphan receptor 2 (Ror2), Wnt5a induces theinternalization of the receptors (8, 9), which is necessaryfor theWnt5a-dependent activation of small GTP-bindingprotein Rac1 that plays important roles in the migrationand invasion activities of cancer cells (10).

It has been shown that expression ofWnt5a is correlatedwith the aggressiveness of melanoma, breast cancer, lungcancer, gastric cancer, and prostate cancer (2, 6, 11–16),suggesting thatWnt5ahasoncogenicproperties. Inexperi-ments in vivowithnudemice,metastasis fromthespleen tothe liver was suppressed significantly in Wnt5a-knock-downcells (15), stronglysuggesting thatWnt5a is involvedin gastric cancer invasion ormetastasis and thatWnt5a is agood molecular target for gastric cancer therapy.

In addition to the activation of Rac1, several possiblemechanisms bywhichWnt5a regulates cell migration andinvasion have been proposed.Wnt5a controls cell polarityand directional migration inmelanoma cells by recruitingactin andmyosin IIB to melanoma cell adhesionmolecule(17). Wnt5a signaling has also been shown to cooperatewith the integrin signalingpathway to regulate cellmigra-tion and adhesion (18). In this model, Wnt5a, Fz2, andintegrin b1 colocalize to the leading edge of polarizedmigrating cells, and this signal is transmitted to focal

Authors'Affiliations: 1Department ofMolecular Biology andBiochemistry,Graduate School of Medicine, Osaka University, Yamadaoka, Suita; and2Department of Surgery, Graduate School of Biomedical Sciences, Hir-oshima University, Kasumi, Minami-ku, Hiroshima, Japan

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Akira Kikuchi, Department of Molecular Biologyand Biochemistry, Graduate School of Medicine, Osaka University, 2-2Yamadaoka, Suita 565-0871, Japan. Phone: 81-6-6879-3410; Fax: 81-6-6879-3419; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-11-0682

�2011 American Association for Cancer Research.

MolecularCancer

Therapeutics

Mol Cancer Ther; 11(2) February 2012298

on July 21, 2021. © 2012 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst November 18, 2011; DOI: 10.1158/1535-7163.MCT-11-0682

http://mct.aacrjournals.org/

adhesionkinase (FAK)andpaxillin through thebindingofDvl andAPC, respectively (18).Moreover,Wnt5a inducesthe expression of laminin g2 andmatrixmetalloproteinase1 in gastric and prostate cancer cells (15, 16), which areknown tobe involved in invasionandmetastasis of cancer.Thus, Wnt5a has different roles in cell migration andinvasion in a cell or tissue context–dependent manner.Wnt1 and Wnt2, which activate the b-catenin–depen-

dent pathway, have been shown to be overexpressed inhepatoma, malignant melanoma, colorectal cancer, non–small cell lung cancer, breast cancer, and sarcoma (19–22).Blockade ofWnt signaling using anti-Wnt1 and anti-Wnt2antibodies induced apoptosis and inhibit proliferationprobably by inhibiting the b-catenin–dependent pathwayin these cancer cells. Although how these antibodiesinhibit Wnt signaling is unclear, interference with theb-catenin–dependent pathway would provide a potentand selective therapeutic strategy for cancers that over-express Wnt1 or Wnt2. However, it has been never clar-ifiedwhether an anti-Wnt5a antibodyhas antitumor effectthrough suppression of the b-catenin–independent path-way. Here we show that an anti-Wnt5a antibody inhibitsmetastasis of gastric cancer cells in vivoby suppressing theWnt5a-dependent internalization of receptors.

Materials and Methods

Cell lines and cell cultureGastric cancer cell lines, KKLS, MKN-1, MKN-45, and

TMK-1 cells, were provided by Dr. W. Yasui (HiroshimaUniversity, Hiroshima, Japan) inAugust 2006 (no authen-tication was done by the authors). These gastric cancercells were grown in RPMI-1640 supplemented with 10%FBS and were maintained at 37�C in humidified air with5%CO2.More information is described in SupplementaryData.

Cell migration and invasion assaysTo measure cell migration and invasion activities,

Transwell assays were done using a modified Boydenchamber (tissue culture treated, 6.5 mm in diameter,10-mm thick, and 8-mm pores; Transwell) and a Matri-gel-coated modified Boyden chamber (BD Biosciences),respectively, as described previously (14, 15). The detailsare described in Supplementary Data.

Animals and implantation of tumor cellsSix-week-old male BALB/cAnNCrj-nu mice (Charles

River Laboratory Japan, Inc.) were anesthetized with acombination of medetomidine (0.3 mg/kg body weight),midazolam (4 mg/kg), and butorphanol (5 mg/kg), andthen KKLS (2.5 � 105), MKN-45 (2.5 � 106), or TMK-1(1.0 � 106) cells in 50 mL Hanks balanced salt solution(HBSS) were injected into the spleen through a 27-gaugeneedle (day 0). pAb5a-5, anti-glutathione-S-transferase(GST) antibody (10 mg/g body weight), or PBS was inject-ed into the intraperitoneal cavity twiceweekly for 5weeks(at days �2, 1, 5, 8, 12, 15, 19, 22, 26, and 29). Five weeks

later, under deep anesthesia with pentobarbital, hepatec-tomy was carried out. Then, formalin-fixed samples weresectioned and the number of metastatic nodule wascounted macroscopically. The protocols used for all ani-mal experiments in this study were approved by theAnimal Research Committee of Hiroshima Universityand Osaka University, Japan.

Receptor internalization assayInternalization of FLAG-Fz2 by Wnt5a and Wnt3a was

examined as described (8) and its details are described inSupplementary Data. When the internalization of Ror2 orlow-density lipoprotein-like receptor protein 6 (LRP6) atendogenous levels was examined, KKLS cells were pre-incubated with 25 mg/mL pAb5a-5 or anti-GST antibodyand then stimulatedwith 50 ng/mLWnt5a orWnt3a. Thecells were incubated with 0.5 mg/mL sulfo-NHS-LC-biotin (Pierce) at 4�C for 30minutes (23, 24). After quench-ing of excess biotinwith 50mmol/LNH4Cl, the cells werelysed in 0.2 mL of TNE buffer [25 mmol/L Tris-HCl(pH 7.5), 150mmol/LNaCl, and 5mmol/L EDTA-NaOH(pH 8.5) containing 0.4% sodium deoxycholate, 1%Triton X-100, 2 mg/mL leupeptin, 2 mg/mL aprotinin, and1 mmol/L phenylmethylsulfonyl fluoride). The lysateswere precipitated with the NeutrAvidin Agarose Resin(Pierce) for 2 hours at 4�C, and the precipitates wereprobed with anti-Ror2 or anti-LRP6 antibody.

Statistical analysisThe incidence of metastasis was compared using the

Mann–Whitney U test. A P-value less than 0.05 wasconsidered statistically significant. Other experimentswere carried out at least 3 times and the results wereexpressed as means � SE.

Results

Generation of anti-Wnt5a polyclonal antibodiesUsing different synthetic peptides that corresponded to

amino acid residues 87–103, 165–181, and 275–290 ofhumanWnt5a,wegenerated rabbit polyclonal anti-Wnt5aantibodies (Supplementary Fig. S1A). Among them, ananti-Wnt5a antibody, named pAb5a-1, was used forimmunohistochemical analyses in human gastric andprostate cancer samples (14–16). However, this antibodydid not affect cell migration of gastric cancer cells in vitro(data not shown). pAb5a-2 recognized both Wnt5a andWnt5b, whereas pAb5a-5 reacted with Wnt5a only (Sup-plementary Fig. S1B). These anti-Wnt5a antibodiesdidnotreact with Wnt3a (Supplementary Fig. S1B).

Among gastric cancer cell lines Wnt5a mRNA levelswere relatively higher in MKN-1, TMK-1, and KKLS cellsthan in MKN-45 cells (Supplementary Fig. S2A; ref. 14).Consistent with mRNA levels, Wnt5a protein wasdetected in lysates of MKN-1, TMK-1, and KKLS cells atendogenous levels but not in that of MKN-45 cells (Sup-plementary Fig. S2B). pAb5a-5 detected endogenousWnt5a, which was secreted from MKN-1 and KKLS cells

Wnt5a Antibody and Metastasis

www.aacrjournals.org Mol Cancer Ther; 11(2) February 2012 299




and attached to the extracellular matrix (ECM) fraction(Supplementary Fig. S2C).

An anti-Wnt5a antibody suppresses gastric cancercell migration and invasion in vitro

Knockdown of Wnt5a suppressed migration and inva-sion activities of MKN-1, KKLS, and TMK-1 cells in vitro(14, 15) and themetastatic ability ofKKLSandTMK-1 cells

in vivo (15). Therefore,migration andmetastatic abilities ofthese gastric cancer cells depend onWnt5a expression. Totest whether anti-Wnt5a antibodies affect these functionsof gastric cancer cells, cell migration activity was mea-sured using a Boyden chamber.WhenKKLS,MKN-1, andTMK-1 cells were treated with pAb5a-5, their ability tomigrate was decreased, but pAb5a-2 did not affect it(Fig. 1A–C). The invasive potential of the cells was

Figure 1. Anti-Wnt5a antibodyinhibits migration, invasion, andadhesion of gastric cancer cells.A–C, inhibition of migration (i) andinvasion (ii) of KKLS (A), MKN-1 (B),or TMK-1 (C) cells by anti-Wnt5aantibody. i, cells were placed in aTranswell chamber for themigration assay in the presence of25 mg/mL anti-Wnt5a (5a-2 or 5a-5)or anti-GST antibody in the upperchamber. ii, cells treated with anti-Wnt5a or anti-GST antibody for 24hours were applied to a Matrigel-coated modified Boyden chamberfor the invasion assay. Relativemigration or invasion activity wasexpressed as a percentage ofmigrated or invading cell numbersof anti-GST antibody–treated cells.Results are shown as means � SEof 3 independent experiments. �, P< 0.05. D, migration (i) and invasion(ii) activities of MKN-45/controlcells or MKN-45 cells stablyexpressing Wnt5a (MKN-45/Wnt5a) were examined in thepresence of pAb5a-5 or anti-GSTantibody. �, P < 0.05. E, KKLS (i),MKN-1 (ii), and MKN-45 (iii) cellswere subjected to the adhesionassay in the presence of pAb5a-5or anti-GST antibody. �, P < 0.05;��, P < 0.01. F, KKLS (i) and MKN-1(ii) cells were cultured in thepresence of pAb5a-5 (closedcircles) or anti-GST antibody (opencircles).

Hanaki et al.

Mol Cancer Ther; 11(2) February 2012 Molecular Cancer Therapeutics300




measured using a Matrigel-coated modified Boydenchamber. pAb5a-5 but not pAb5a-2 inhibited invasionactivity of these 3 gastric cancer cells in vitro (Fig. 1A–C). The inhibitory effect of pAb5a-5 onKKLS cellswas in adose-dependent manner (Supplementary Fig. S3).To determine the specificity of pAb5a-5, MKN-45 cells,

which show a low expression level of Wnt5a, were used.Unlike the effects on KKLS, MKN-1, and TMK-1 cells,pAb5a-5 did not affectmigration and invasion activities ofMKN-45 cells (Fig. 1D). Expression of Wnt5a in MKN-45cells enhanced migration and invasion activities, andpAb5a-5 inhibited the activities of Wnt5a-expressingMKN-45 cells (Fig. 1D). Wnt5a was also shown to beinvolved in cell-to-substrate adhesion and to enhanceadhesion-dependent activation of FAK (14, 18). pAb5a-5 inhibited adhesion activity in KKLS and MKN-1 cells(Fig. 1E). pAb5a-5 did not affect adhesion activity ofMKN-45 cells but inhibited that of MKN-45 cells expres-sing Wnt5a (Fig. 1E). Furthermore, pAb5a-5 also sup-pressed adhesion-dependent activation of FAK in KKLSandWnt5a-expressingMKN-45 cells (Supplementary Fig.S4). Taken together, pAb5a-5 inhibited cell migration,invasion, and adhesion activities of gastric cancer cellsexpressing Wnt5a. However, pAb5a-5 did not affect thegrowth of KKLS and MKN-1 cells in vitro (Fig. 1F), whichwas consistent with our previous findings that knock-down of Wnt5a did not affect proliferation of gastriccancer cells in vivo and in vitro (14, 15).

pAb5a-5 inhibits Wnt5a-dependent activation ofRac1 and expression of laminin g2Wnt5a activated Rac1 (8, 14), which plays important

roles in cell migration and adhesion (10). Consistent withinhibitory effects of pAb5a-5 on migration and invasion,the antibody inhibited Wnt5a-dependent activation ofRac1 in KKLS cells (Fig. 2A). Furthermore, the inhibitoryeffect of pAb5a-5 on invasion activity of KKLS cells wasrestored by the expression of a constitutive active form ofRac1 (Rac1V12; Supplementary Fig. S5), suggesting thatthe antibody affects Wnt5a signaling upstream of Rac1.It has been reported that Wnt5a induces the expression

of laminin g2 (15), which is suggested to be involved ininvasion activity of gastric cancer cells (25). To examinewhether laminin g2 is necessary for metastatic ability ofgastric cancer cells in vivo, wild-typeTMK-1 or laminin g2-knockdown TMK-1 cells were inoculated into the spleenof nudemice (Fig. 2B and Supplementary Fig. S6). Knock-down of laminin g2 indeed reducedmetastatic nodules inthe liver (Fig. 2B).To examine whether Rac1 is involved in Wnt5a-depen-

dent expression of laminin g2 gene (LAMC2), LAMC2-Luc,a reporter gene containing the LAMC2 promoter and aluciferase gene, was transfected intoMKN-1, TMK-1, andMKN-45 cells. In these gastric cancer cells, Rac1V12increased luciferase activity and a dominant negativeform of Rac1 (Rac1N17) suppressed Wnt5a-dependentluciferase activity (Fig. 2C), suggesting that Wnt5ainduces the expression of laminin g2 through the activa-

tion of Rac1. Wnt5a increased protein levels of laminin g2in MKN-45 and MKN-1 cells, although basal expressionlevels were different (Fig. 2D). pAb5a-5 suppressed pro-tein and mRNA levels of laminin g2 in MKN-1 cells andWnt5a-treated MKN-45 cells (Fig. 2D). Immunocyto-chemical analyses confirmed that pAb5a-5 inhibitsWnt5a-dependent expression of laminin g2 in MKN-1cells (Fig. 2E and Supplementary Fig. S7). These resultssuggest that pAb5a-5 has a potential ability to suppressmigration and invasion activities of Wnt5a-expressingcells by inhibiting the activation of Rac1 and the expres-sion of laminin g2.

pAb5a-5 inhibits Wnt5a-dependent receptorinternalization

How pAb5a-5 inhibits Wnt5a signaling was examined.Wnt5a bound to the extracellular domain of its receptors,Fz2 and Ror2, in vitro (8, 26). Secreted frizzled-relatedprotein 2 (sFRP2) is a secretedprotein that is able to bind toWnt ligands, thereby suppressing their binding to recep-tors (27). pAb5a-5 did not inhibit the interaction betweenWnt5a and Fz2 under conditions where sFRP2 inhibitedtheir binding (Fig. 3A). Wnt5a has been shown to inducephosphorylation of Ror2 in HeLaS3, HEK293, andNIH3T3 cells (28). Consistent with these results, Wnt5ainduced a mobility shift of Ror2 in KKLS cells on SDS-PAGE, which indicated the phosphorylation of Ror2 (Fig.3B). pAb5a-5didnot affect themobility shift ofRor2underconditions where sFRP2 reduced it (Fig. 3B). Takentogether, these results suggest that pAb5a-5 does notaffect the binding of Wnt5a to its receptors, Fz2 and Ror2.

Wnt5a has been shown to induce the internalization ofFLAG-Fz2 through clathrin-mediated endocytic routeand that this process is necessary for Wnt5a-dependentRac1 activation in HeLaS3 and HEK293 cells (8). FLAG-Fz2 was internalized in KKLS cells in response to Wnt5a(Fig. 3C). Treatment of cells with pAb5a-5 increasednumbers of cells expressing FLAG-Fz2 on the cell surfaceanddecreased numbers of cells having intracellular punc-tuate structures of FLAG-Fz2, suggesting that pAb5a-5inhibits Wnt5a-dependent internalization of FLAG-Fz2(Fig. 3C). Endogenous Ror2 was also found to be inter-nalized byWnt5a in HeLaS3 cells in a clathrin-dependentmanner (9). Wnt5a induced the internalization of endog-enous Ror2 in KKLS cells in time- and dose-dependentmanners, and pAb5a-5 suppressed it (Fig. 3D). Wnt3ainduced the internalization of Fz2 and LRP6, and theirinternalization is required for Wnt3a-dependent activa-tion of the b-catenin–dependent pathway (8, 23, 24).Unlike the effects on Wnt5a-dependent internalization ofFz2 and Ror2, pAb5a-5 did not affect Wnt3a-dependentinternalization of Fz2 and LRP6 (Fig. 3E).

Furthermore, knockdown of clathrin suppressedWnt5a-dependent increase in LAMC2 mRNA of TMK-1and MKN-45 cells (Fig. 3F), suggesting that receptor-mediated endocytosis througha clathrin-dependent routeis necessary for the expression of laminin g2. Takentogether, these results suggested that pAb5a-5 inhibits






Wnt5a-dependent internalization of its receptors, therebysuppressing Wnt5a-dependent activation of Rac1 andexpression of laminin g2.

pAb5a-5 recognizedWnt5a released fromKKLScellsin 3D culture

Next we tested whether pAb5a-5 recognizes Wnt5asecreted from cells. To observeWnt5a secreted from cells,KKLS cells were embedded into 3D-Matrigel. KKLS cellsformed amorphous aggregates in 3D culture, and endog-

enous integrin b1 was enriched in the cell surface (Fig. 4).However, endogenous Wnt5a was hard to detect usingpAb5a-5 in 3D culture conditions (Fig. 4A). When Wnt5awas expressed inKKLS cells transiently, itwas recognizedby pAb5a-5 in the area surrounding cells (Fig. 4B). Thisdetection was specific, because pAb5a-2 and anti-GSTantibody did not recognize Wnt5a produced from thecells (Fig. 4C and D). These results suggest that secretedWnt5a is localized to the immediate vicinity of KKLS cellsprobably because of the interactionwithMatrigel proteins

A

B

C

D

E

Figure 2. pAb5a-5 inhibitsactivation of Rac1 and expressionof laminin g2 in gastric cancer cells.A, KKLS cells were stimulated withthe indicated concentrations ofWnt5a for 1 hour in the presence orabsence of 25 mg/mLpAb5a-5. Thecells were incubated with GSTfusion Cdc42/Rac-interactingbindingdomain (CRIB) immobilizedonglutathione sepharose. The totallysates (total Rac) and precipitates(active Rac) were probed with anti-Rac1 antibody. Band intensitieswere quantified using NIH imagesoftware. Values in control cellswithoutWnt5a stimulation were setto 100%. Results are shown asmeans � SE of 3 independentexperiments. B, TMK-1/control(n ¼ 4) and TMK-1/laminin g2-knockdown (KD; n ¼ 5) cells wereinjected into the spleens of nudemice. Photographs representingliver metastases (i) and numbers ofmetastatic nodules per mouse (ii)are shown. Scale bars, 5 mm.�, P < 0.05. C, LAMC2-Luc wastransfected into MKN-1 (i), TMK-1(ii), and MKN-45 (iii) cells withpEXV/Rac1V12 or pEXV/Rac1N17in the presence or absence of 150ng/mL Wnt5a. Relative laminin g2promoter activity was expressedas fold increases compared withthe luciferase induction in controlcells. D, MKN-1 (i) or MKN-45 (ii)cells were treated with 150 ng/mLWnt5a for 8 (left) or 4 (right) hours inthe presence or absence of pAb5a-5 or anti-GST antibody. Left, ECMfractions from cells were probedwith anti-laminin g2 antibody. Thelysates of these cells were probedwith anti-b-actin antibody. Right,LAMC2 mRNA level wasquantified. Relative mRNA levelswere expressed as percentages ofLAMC2 mRNA levels in anti-GSTantibody–treated cells. �, P < 0.05.E, MKN-1 cells were treated with150 ng/mLWnt5a for 8 hours in thepresence of pAb5a-5 or anti-GSTantibody and then stained withanti-laminin g2 antibody. Scalebars, 20 mm.

Hanaki et al.





Figure 3. pAb5a-5 inhibits receptor-mediated endocytosis. A, FLAG-Fz2was incubatedwith 200 ng/mLWnt5a for 2 hours in the presence of 25mg/mLpAb5a-5 or anti-GST antibody or 480 ng/mL sFRP2, and then the complexes were precipitated. The precipitates were probed with pAb5a-5. Results arerepresentative of 3 independent experiments. B, KKLScellswere stimulatedwith 25 ng/mLWnt5a for 1 hour in thepresenceof pAb5a-5, anti-GSTantibody, orsFRP2, and then the lysates were probed with anti-Ror2 antibody. C, KKLS cells expressing FLAG-Fz2 were treated with 50 ng/mL Wnt5a for 30 minutesin the presence of pAb5a-5 or anti-GST antibody. Left, confocal images; right, quantification of internalized FLAG-Fz2. Internalization of FLAG-Fz2 wasestimated as described in Supplementary Data. Scale bars, 20 mm. D, i, KKLS cells were treated with 50 ng/mL Wnt5a for the indicated periods in thepresence of pAb5a-5 or anti-GST antibody. After cell surface biotinylation, the lysates were precipitated with the NeutrAvidin Agarose Resin. Left, theprecipitates (cell surfaceRor2) and lysates (total Ror2) were probedwith anti-Ror2 antibody. Right, the amounts of cell surface Ror2were quantified usingNIHimage software. Values at zero time in the presence of anti-GST antibody were set to 100%. ii, KKLS cells were treated with the indicated concentrations ofWnt5a for 15 minutes in the presence of pAb5a-5 or anti-GST antibody. Values at no Wnt5a in the presence of anti-GST antibody were set to 100%.






or heparan sulfate proteoglycans (HSPG; ref. 29). There-fore, it was expected that pAb5a-5 is able to react with anative form of Wnt5a, which is secreted from cells andaffects Wnt5a-producing cells in vivo.

pAb5a-5 inhibits metastasis in vivoFinally, the effects of pAb5a-5 on metastatic ability in

vivo of gastric cancer cells were examined using a mousemodel. KKLS or MKN-45 cells were inoculated into thespleen of nude mice. Injection of pAb5a-5 significantlyinhibited the numbers of liver metastatic foci comparedwith that of PBS or anti-GST antibody (Fig. 5A). However,

in vivo metastasis of MKN-45 cells was not affected bytreatment with pAb5a-5 (Fig. 5B). No obvious toxicity,including weight loss, during the treatment and no his-tologic damage of liver or spleen were observed in any ofthe control mice administrated with pAb5a-5 (data notshown).

A polyclonal antibody contains multiple antibodiesagainst different epitopes on the antigen even though asynthetic peptide was used as an antigen. Therefore, theneutralizing activity could vary among 5 different rabbit-generated pAb5a-5 antibodies. All of the antibodies sup-pressed migration and invasion activities of KLLS cells invitro, although the degree of inhibition varies (Fig. 5C).Antibodies inhibiting migration well had a tendency tosuppress invasion strongly. Furthermore, all pAb5a-5from 5 different rabbits suppressed metastatic ability ofKKLS cells in vivo (Fig. 5D). Antibodies reducing invasionin vitrowell had a tendency to suppressmetastasis strong-ly in vivo (Fig. 5D). Taken together, these results suggestedthat the effects of pAb5a-5 on gastric cancer cell migrationand invasion activities in vitro reflect those of the antibodyto suppress metastatic ability in vivo.

Discussion

The molecular mechanisms by which pAb5a-5inhibits invasion and metastasis

In this study, we have, for the first time, shown that ananti-Wnt5a antibody (pAb5a-5) inhibits invasion andmetastasis of gastric cancer cells. The inhibitory activityof pAb5a-5 was specific for gastric cancer cells expressingWnt5a, because it did not affect migration, invasion,adhesion, and metastasis of MKN-45 cells, which expressWnt5a little. Wnt5a was shown to be involved in theactivation of Rac1 and the expression of laminin g2 (14,15). Wnt5a induced the internalization of Fz2 and Ror2through a clathrin-dependent manner and the internali-zation was necessary for Wnt5a-dependent activation ofRac1 (8) and expression of laminin g2 (this study). Ourpresent results revealed that pAb5a-5 suppresses Wnt5a-dependent receptor internalization, although pAb5a-5does not inhibit the binding of Wnt5a to its receptors,thereby inhibiting the Wnt5a-dependent activation ofRac1 and expression of laminin g2. Because an anti-FLAGantibody was internalized into cells with FLAG-Fz2in response to Wnt5a (8) and pAb5a-5 did not affectWnt3a-dependent internalization of Fz2 and LRP6, theeffect of pAb5a-5 to inhibit the internalization of Wnt5areceptors could be specific. Taken together, it is likely thatpAb5a-5 suppresses invasion and metastasis of gastric

E, i, KKLS cells expressing FLAG-Fz2 were treated with 50 ng/mLWnt3a for 15 minutes in the presence of pAb5a-5 or anti-GST antibody. ii, KKLS cells weretreated with the indicated concentrations of Wnt3a for 30 minutes in the presence of pAb5a-5 or anti-GST antibody. After cell surface biotinylation, thelysates were precipitated with the NeutrAvidin Agarose Resin. The amounts of cell surface LRP6 were quantified. F, MKN-45 and TMK-1 cells,which were treated with control or clathrin short interfering RNA, were stimulated with or without 150 ng/mL Wnt5a for 4 hours, and then LAMC2 mRNAlevels were quantified. Relative mRNA levels were expressed as fold increases compared with LAMC2 mRNA levels in control cells without Wnt5astimulation. �, P < 0.05. IP, immunoprecipitation; KD, knockdown.

Figure 4. pAb5a-5 recognizes Wnt5a secreted from KKLS cells. KKLScells transfectedwithout (A) or with (B–D) pPGK2-neo/Wnt5awere grownin theMatrigel for 36hours. After the cellswere incubatedwith anti-Wnt5a(A andB,pAb5a-5;C, pAb5a-2) or anti-GST (D; green) andanti-integrinb1antibodies (A–D; red), the cellswere viewed using a confocalmicroscope.Scale bars, 40 mm.

Hanaki et al.





cancer cells by inhibiting Wnt5a-dependent receptorinternalization.Currently we do not know the mechanism by which

pAb5a-5 blocks Wnt5a-dependent internalization ofreceptors. The interaction of Wnt5a with its receptorsand other factors including HSPG and secreted pro-teins is highly complex. One possibility is that pAb5a-5may inhibit the interaction of Wnt5a and componentsother than receptors. For example, a transmembrane-type HSPG, syndecan (SDC), has been reported to regu-late Wnt5a signaling. Expression of SDC1 and SDC4correlates with Wnt5a expression in melanoma cell linesand knockdown of SDC1 and SDC4 decreases cell inva-sion activity, which is restored by Wnt5a (30). SDC4 hasalso been shown to be required for Wnt5a-dependentinternalization of Fz7 in Xenopus embryos (31). In addi-tion, another type of HSPG, glypican-4, enhancedWnt5a-dependent activation of Rac1 (9). Therefore, it

is intriguing to examine effects of pAb5a-5 on the form-ation of a complex between Wnt5a and HSPG.

Possible effects of pAb5a-5 on other functions ofWnt5a

It has been suggested thatWnt5ahas tumor suppressiveactions, becauseWnt5a heterozygous mice develop B-celllymphoma (32). Furthermore, Wnt5a reduced the prolif-eration, migration, and invasiveness in thyroid tumorcells and inhibited the migration activity of colorectalcancer cells (6). The mechanism of the tumor suppressiveeffects ofWnt5amight be because of its inhibitory activityfor the b-catenin pathway. It has been shown that Wnt5aantagonizes the b-catenin pathway by inhibiting either of3 points; b-catenin–dependent transcriptional activation,the stabilization of b-catenin, or the binding of Wnt thatactivates the b-catenin pathway to Fz (7, 8, 33). In the firstcase, Ror2 functions as a receptor (33), and this inhibitory

Figure 5. pAb5a-5 inhibitsmetastases in vivo. A, KKLS cells (2.5� 105 cells) in 50 mL HBSS (�) wereinjected into the spleen of nudemice.i, PBS, pAb5a-5, or anti-GSTantibody (10 mg/g) was injected intointraperitoneal cavity twice weeklyfor 5 weeks. Results are shown asmeans � SE of 7 to 14 experiments.�, P < 0.05. ii, photographsrepresenting liver metastases (left,anti-GST antibody treatment; right,pAb5a-5 treatment). Scale bars, 5mm.B,MKN-45 cells (2.5�106 cells)in 50 mL HBSS (�) were injected intothe spleen of nude mice. i, thenumber of metastatic liver tumorswas counted. Results are averagesof 2 experiments. ii, photographsrepresenting liver metastases. C andD, relative migration (C, i), invasion(C, ii), or metastatic (D) activity ofKKLS cells treated with pAb5a-5generated by 5 different rabbits(#1–#5) was expressed aspercentage of KKLS cells treatedwith anti-GST antibody. In migrationand invasion experiments in vitro,results are shownasmeans�SEof 3independent experiments for eachantibody. In metastatic experimentsin vivo, results are shown asaverages of 2 or 3 independentexperiments for each antibodyexcept for #4 and #5. ND means nometastatic nodule in the liver. PBStreatment, n¼ 14; anti-GST antibodytreatment, n¼ 7; pAb5a-5 treatment,n ¼ 10 [#1 (n ¼ 2); #2 (n ¼ 3);#3 (n ¼ 3); #4 (n ¼ 1); #5 (n ¼ 1)].

A

B

C D






mechanism may require the internalization of Ror2.Therefore, if Wnt5a suppresses tumorigenesis throughthis mechanism, pAb5a-5may exacerbate cancer progres-sion. The third inhibitory mechanism of Wnt5a doesnot require the internalization of receptors but competewith other Wnt ligand for the binding to receptors (8).Therefore, pAb5a-5 is likely to show no effects on tumorswhen Wnt5a exerts this inhibitory action. At present itis not known whether receptor-mediated endocytosis isinvolved in the second inhibitory mechanism.

It has also been reported that Wnt5a activates theb-catenin pathway when Fz5 or Fz4 and LRP5 are over-expressed (7, 33), suggesting that Wnt5a promotes cellgrowth in cancers with Fzs and LRP5 overexpression.Whether the internalization of these receptors in responseto Wnt5a is required for the activation of the b-cateninpathway is not known.However, it has been reported thatWnt3a induces the internalization of LRP6 and Fz2 or Fz5througha caveolin-mediated route and the internalizationis necessary for the activation of the b-catenin pathway(8, 23, 34). Therefore, it is possible that by inhibitingreceptor internalization pAb5a-5 suppresses cell growthin certain types of cancers in which Fzs and LRP5 areoverexpressed and Wnt5a stimulates cell proliferation.

Our results showed that anti-Wnt5a polyclonal anti-bodies generated from different rabbits suppress migra-tion and invasion activities of gastric cancer cells in vitro tovarious extents. However, the inhibitory effects onmetas-tasis in vivo were almost parallel with those on invasion

in vitro. Therefore, in vitro invasion assays are useful forscreening to find anti-Wnt5a antibodies capable of sup-pressing metastatic ability of cancer cells in vivo. It will beimportant to obtain anti-Wnt5a monoclonal antibody forfurther investigation how the antibody suppressesWnt5asignaling and inhibit metastases of cancer cells for theclinical management of cancers with overexpression ofWnt5a.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

Theauthors thankDr.HidekiYamamoto (HiroshimaUniversity) for histechnical assistance andDrs.W.Yasui,K.Kaibuchi, T.Akiyama, S. Takada,J. Olsen, and A. Hall for donating cells and plasmids.

Grant Support

This study was supported by Grants-in-Aid for Scientific Research andfor Scientific Research on Priority Areas from the Ministry of Education,Science, and Culture of Japan (2008–2011) and a grant from KobayashiFoundation for Cancer Research (2010) and Nagase Foundation (2011) toA. Kikuchi.

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 September 2, 2011; revised October 26, 2011; acceptedNovember 15, 2011; published OnlineFirst November 18, 2011.

References1. Logan CY, Nusse R. The Wnt signaling pathway in development and

disease. Annu Rev Cell Dev Biol 2004;20:781–810.2. Veeman MT, Axelrod JD, Moon RT. A second canon. Functions and

mechanisms of b-catenin-independent Wnt signaling. Dev Cell 2003;5:367–77.

3. Kikuchi A, Yamamoto H, Sato A. Selective activation mechanisms ofWnt signaling pathways. Trends Cell Biol 2009;19:119–29.

4. MacDonald BT, Tamai K, He X.Wnt/b-catenin signaling: components,mechanisms, and diseases. Dev Cell 2009;17:9–26.

5. Polakis P. The many ways of Wnt in cancer. Curr Opin Genet Dev2007;17:45–51.

6. Kikuchi A, Yamamoto H. Tumor formation due to abnormalities in theb-catenin-independent pathway ofWnt signaling. Cancer Sci 2008;99:202–8.

7. Kikuchi A, Yamamoto H, Sato A, Matsumoto S. Wnt5a: its signalling,functions and implication in diseases. Acta Physiol (Oxf) 2011 Mar 29.doi: 10.1111/j.1748-1716.2011.02294.x.

8. Sato A, Yamamoto H, Sakane H, Koyama H, Kikuchi A. Wnt5aregulates distinct signalling pathways by binding to Frizzled2. EMBOJ 2010;29:41–54.

9. SakaneH, YamamotoH,MatsumotoS, Sato A, Kikuchi A. Localizationof glypican-4 in different membrane microdomains is involved in theregulation of Wnt signaling. J Cell Sci 2012. In press.

10. Yamazaki D, Kurisu S, Takenawa T. Regulation of cancer cell motilitythrough actin reorganization. Cancer Sci 2005;96:379–86.

11. Weeraratna AT, Jiang Y, Hostetter G, Rosenblatt K, Duray P, BittnerM,et al. Wnt5a signaling directly affects cell motility and invasion ofmetastatic melanoma. Cancer Cell 2002;1:279–88.

12. Huang CL, Liu D, Nakano J, Ishikawa S, Kontani K, Yokomise H, et al.Wnt5a expression is associated with the tumor proliferation and the

stromal vascular endothelial growth factor—an expression in non–small cell lung cancer. J Clin Oncol 2005;23:8765–73.

13. Pukrop T, Klemm F, Hagemann T, Gradl D, Schulz M, Siemes S, et al.Wnt 5a signaling is critical for macrophage-induced invasion of breastcancer cell lines. Proc Natl Acad Sci U S A 2006;103:5454–9.

14. KurayoshiM,OueN,YamamotoH,KishidaM, InoueA,Asahara T, et al.Expression of Wnt-5a is correlated with aggressiveness of gastriccancer by stimulating cell migration and invasion. Cancer Res2006;66:10439–48.

15. Yamamoto H, Kitadai Y, Yamamoto H, OueN, OhdanH, YasuiW, et al.Laminin g2 mediates Wnt5a-induced invasion of gastric cancer cells.Gastroenterology 2009;137:242–52.

16. Yamamoto H, Oue N, Sato A, Hasegawa Y, Matsubara A, YamamotoH, et al. Wnt5a signaling is involved in the aggressiveness of prostatecancer and expression of metalloproteinase. Oncogene 2010;29:2036–46.

17. Witze ES, Litman ES, Argast GM, Moon RT, Ahn NG. Wnt5a control ofcell polarity and directional movement by polarized redistribution ofadhesion receptors. Science 2008;320:365–9.

18. MatsumotoS, FumotoK,Okamoto T, Kaibuchi K, Kikuchi A. Binding ofAPC and dishevelled mediates Wnt5a-regulated focal adhesiondynamics in migrating cells. EMBO J 2010;29:1192–204.

19. WeiW,ChuaMS,Grepper S, SoSK.BlockadeofWnt-1 signaling leadsto anti-tumor effects in hepatocellular carcinoma cells. Mol Cancer2009;8:76.

20. You L, He B, Xu Z, Uematsu K, Mazieres J, Fujii N, et al. An anti-Wnt-2monoclonal antibody induces apoptosis in malignant melanoma cellsand inhibits tumor growth. Cancer Res 2004;64:5385–9.

21. He B, Reguart N, You L, Mazieres J, Xu Z, Lee AY, et al. Blockadeof Wnt-1 signaling induces apoptosis in human colorectal cancer

Hanaki et al.





cells containing downstream mutations. Oncogene 2005;24:3054–8.

22. He B, You L, Uematsu K, Xu Z, Lee AY, Matsangou M, et al. Amonoclonal antibody against Wnt-1 induces apoptosis in humancancer cells. Neoplasia 2004;6:7–14.

23. Yamamoto H, Komekado H, Kikuchi A. Caveolin is necessary for Wnt-3a-dependent internalization of LRP6 and accumulation of b-catenin.Dev Cell 2006;11:213–23.

24. Sakane H, Yamamoto H, Kikuchi A. LRP6 is internalized by Dkk1 tosuppress its phosphorylation in the lipid raft and is recycled for reuse. JCell Sci 2010;123:360–8.

25. Miyazaki K. Laminin-5 (laminin-332): unique biological activity and rolein tumor growth and invasion. Cancer Sci 2006;97:91–8.

26. Oishi I, Suzuki H, Onishi N, Takada R, Kani S, Ohkawara B, et al. Thereceptor tyrosine kinase Ror2 is involved in non-canonical Wnt5a/JNKsignalling pathway. Genes Cells 2003;8:645–54.

27. Kawano Y, Kypta R. Secreted antagonists of the Wnt signallingpathway. J Cell Sci 2003;116:2627–34.

28. Yamamoto H, Yoo SK, Nishita M, Kikuchi A, Minami Y. Wnt5amodulates glycogen synthase kinase 3 to induce phosphorylation

of receptor tyrosine kinase Ror2. Genes Cells 2007;12:1215–23.

29. Mythreye K, Blobe GC. Proteoglycan signaling co-receptors: roles incell adhesion, migration and invasion. Cell Signal 2009;21:1548–58.

30. O'Connell MP, Fiori JL, Kershner EK, Frank BP, Indig FE, Taub DD,et al. Heparan sulfate proteoglycan modulation of Wnt5A signal trans-duction in metastatic melanoma cells. J Biol Chem 2009;284:28704–12.

31. Ohkawara B, Glinka A, Niehrs C. Rspo3 binds syndecan 4 and inducesWnt/PCP signaling via clathrin-mediated endocytosis to promotemorphogenesis. Dev Cell 2011;20:303–14.

32. Liang H, Chen Q, Coles AH, Anderson SJ, Pihan G, Bradley A, et al.Wnt5a inhibits B cell proliferation and functions as a tumor suppressorin hematopoietic tissue. Cancer Cell 2003;4:349–60.

33. Mikels AJ, Nusse R. Purified Wnt5a protein activates or inhibitsb-catenin-TCF signaling depending on receptor context. PLoS Biol2006;4:570–82.

34. Yamamoto H, Sakane H, Yamamoto H, Michiue T, Kikuchi A. Wnt3aandDkk1 regulate distinct internalization pathways of LRP6 to tune theactivation of b-catenin signaling. Dev Cell 2008;15:37–48.






2012;11:298-307. Published OnlineFirst November 18, 2011.Mol Cancer Ther Hideaki Hanaki, Hideki Yamamoto, Hiroshi Sakane, et al.

by Inhibiting Receptor-Mediated EndocytosisIn VivoCells An Anti-Wnt5a Antibody Suppresses Metastasis of Gastric Cancer

Updated version

10.1158/1535-7163.MCT-11-0682doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2011/11/17/1535-7163.MCT-11-0682.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/11/2/298.full#ref-list-1

This article cites 32 articles, 8 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/11/2/298.full#related-urls

This article has been cited by 7 HighWire-hosted articles. Access the articles 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://mct.aacrjournals.org/content/11/2/298To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-11-0682

http://mct.aacrjournals.org/content/suppl/2011/11/17/1535-7163.MCT-11-0682.DC1

http://mct.aacrjournals.org/content/11/2/298.full#ref-list-1

http://mct.aacrjournals.org/content/11/2/298.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/11/2/298


Documents

An Anti-Wnt5a Antibody Suppresses Metastasis of Gastric ... · Cancer Cells In Vivo by Inhibiting Receptor-Mediated Endocytosis Hideaki Hanaki1,2, Hideki Yamamoto 1, Hiroshi Sakane