This article was downloaded by: [University of California, San Francisco]On: 04 December 2014, At: 10:44Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Cancer Biology & TherapyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/kcbt20

Neuropilin-1 interacts with integrin β1 and modulatespancreatic cancer cell growth, survival and invasionMitsuharu Fukasawa, Akira Matsushita & Murray KorcPublished online: 01 Aug 2007.

To cite this article: Mitsuharu Fukasawa, Akira Matsushita & Murray Korc (2007) Neuropilin-1 interacts with integrin β1 andmodulates pancreatic cancer cell growth, survival and invasion, Cancer Biology & Therapy, 6:8, 1184-1191, DOI: 10.4161/cbt.6.8.4363

To link to this article: http://dx.doi.org/10.4161/cbt.6.8.4363

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

e1 CancerBiology&Therapy 2007;Vol.6Issue8

[CancerBiology&Therapy6:8,e1-e8,EPUBAheadofPrint:http://www.landesbioscience.com/journals/cbt/abstract.php?id=4363;August2007];©2007LandesBioscience

Mitsuharu Fukasawa†

Akira Matsushita†

Murray Korc*

Departments of Medicine, Pharmacology and Toxicology, and the Norris Cotton Comprehensive Cancer Center; Dartmouth Medical School; Hanover, New Hampshire and Dartmouth Hitchcock Medical Center; Lebanon, New Hampshire USA

†These authors contributed equally to this work.

*Correspondence to: Murray Korc; Department of Medicine; Dartmouth Hitchcock Medical Center ;One Medical Center Drive; Lebanon, New Hampshire 03756 USA; Tel.: 603.650.7936; Fax: 603.650.6122; Email: murray.korc@dartmouth.edu

Original manuscript submitted: 03/01/07Revised manuscript submitted: 04/29/07Manuscript accepted: 04/29/07

This manuscript has been published online, prior to printing for Cancer Biology & Therapy, Volume 6, Issue 8. Definitive page numbers have not been assigned. The current citation is: Cancer Biol Ther 2007; 6(8):http://www.landesbioscience.com/journals/cc/abstract.php?id=4363Once the issue is complete and page numbers have been assigned, the citation will change accordingly.

Key words

neuropilin-1, antisense, adhesion, apoptosis,integrin,pancreaticcancer,Bcl-xL,STAT5

AcKnowledgeMents

This work was supported in part byUnited States Public Health Service GrantCA-102687awardedbytheNationalCancerInstitutetoM.K.

Research Paper

Neuropilin-1 Interacts with Integrin b1 and Modulates Pancreatic Cancer Cell Growth, Survival and Invasion

AbstrActNeuropilin‑1 (Np‑1) is a coreceptor for vascular endothelial growth factor‑A (VEGF‑A),

and both are expressed at high levels in pancreatic ductal adenocarcinomas (PDACs). While VEGF‑A has been implicated in tumor angiogenesis, the role of Np‑1 in PDAC is less clearly defined. Accordingly, PANC‑1 pancreatic cancer cells, which express relatively high levels of Np‑1, were transfected with the Np‑1 antisense cDNA. By comparison with sham transfected cells, Np‑1 antisense expressing clones (Np‑1AS) exhibited decreased anchorage independent growth, adhesion and invasiveness, and prolonged doubling times. Np‑1AS were also more sensitive to the pro‑apoptotic actions of ActD, as evidenced by PARP cleavage, caspase 9 activation, and annexin V staining. ActD decreased Bcl‑xL and STAT5 levels in the antisense expressing cells, but not in sham‑transfected cells, and did not alter STAT3, Bcl‑2, phospho‑AKT, AKT, Bad, Bax or Bak levels. Immunoprecipitation followed by immunoblotting revealed that Np‑1 associated with integrin b1, and integrin b1 blockade attenuated adhesion. However, Np‑AS expressing clones exhibited enhanced tyrosine phosphorylated focal adhesion kinase. Thus, Np‑1 confers a growth and survival advantage to PANC‑1 cells, and interacts with integrin b1 to coordinate signaling events that promote cell adherence and invasiveness.

IntroductIonPancreaticductaladenocarcinoma(PDAC)isadeadlymalignancy1thatischaracter-

izedbyahighrateofmutationsintheK-rasoncogene,thep53tumorsuppressorgenes,andthep16cellcycleregulatinggene,andbyanabundanceoftransmembranetyrosinekinase receptorsand their ligands.2-3These include theepidermalgrowth factor (EGF)receptor,erb-B2,erb-B3, the insulin-likegrowth factor-1 (IGF-1) receptor,and ligandssuchastransforminggrowthfactoralpha(TGFa),IGF-1andvascularendothelialgrowthfactor-A (VEGF-A).4 In addition, the cancer cells in PDAC often exhibit constitutiveactivation of NFkB,5 AKT6 and STAT3.7Together, these alterations combine to givepancreaticcancercellsadistinctgrowthandsurvivaladvantage.

Neuropilin-1 (Np-1), originally identified as a mediator of chemorepulsive guidancefor axons in the developing nervous system,8 is also expressed at high levels in thecancercellsinPDAC.9-10Np-1isatransmembraneproteinthatactsasacoreceptorforVEGF-A11 and semaphorins.12 Its extracellular region of consists of two complementbinding-like domains (a1 and a2), two coagulation factor V/VIII homology domains(b1andb2),andameprinA5,orcdomain.13Itsintracellulardomainconsistsofashortcytoplasmictailofabout40aminoacidswithac-terminalSEAdomain.13TheSEAdomainisconservedfromXenopustohumanandhasbeenshowntointeractwithaproteinthathasbeenvariablycalledNpinteractingprotein(NIP)orGIPCforRGS-GAIP-interactingprotein.14GIPCcontainsacentralPSD-95/Dlg/ZO-1(PDZ)domainandaC-terminalacylcarrierproteindomain.15

Np-1hasbeenimplicatedinpromotingcellsurvival,andhasbeenrecentlyshowntohaveacoreceptorfunctionwithrespecttotheinteractionsofHBGFswiththeircognatehighaffinityreceptors.16Thepurposeofthepresentstudywastoassesstheroleofendog-enousNp-1 inpancreatic cancer cells, in relation to cell growthand survival.Wenowreport thatdownregulationofNp-1decreases theproliferation, chemoresistance, adhe-sionandinvasionofPANC-1humanpancreaticcancercells,underscoringthepotentialimportanceofNp-1asatherapeutictargetinthisdisease.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

www.landesbioscience.com CancerBiology&Therapy e2

Neuropilin-1InteractswithIntegrinb1

MAterIAl And MethodsMaterials.The followingmaterialswerepurchased: fetal bovine

serum(FBS)fromOmegaScientific(Tarzana,CA);DMEM,trypsinsolution, and penicillin, streptomycin and amphotericin B solu-tionfromMediatech,Inc.(Herndon,VA);LipofectAMINEPLUSReagent kit from Invitrogen (Carlsbad, CA); Immobilon P fromMillipore Intertech (Bedford, MA); Immobilized anti-HA, SuperSignal Substrate System detection system from Pierce (Rockford,IL); Bio-Max light film from Eastman Kodak (Rochester, NY);horseradish-peroxidase conjugated ant-goat and anti-mouse anti-bodies fromBio-RadLaboratories, Inc. (Hercules,CA); anti-Np-1(A12)mousemonoclonal,anti-ERK2,anti-HA(Y-11)rabbitpoly-clonalantibodies,normalmouseIgGandproteinA/GplusagarosefromSantaCruzBiotechnology,Inc.(SantaCruz,CA);anti-STAT3,anti-AKT, anti-phospho AKT, anti-Bad, anti-Bak, anti-Bax,anti-Bcl-2,anti-Bcl-xL,anti-EGFRandanti-STAT5antibodiesfromCell SignalingTechnology (Beverly, MA); anti-Poly (ADP-Ribose)Polymerase(PARP)andanti-integrinb1mousemonoclonalantibodyfromBDTransductionLaboratories(SanDiego,CA);anti-integrinb1(6S6)monoclonalantibodiesfromChemiconInternational,Inc.(Temecula, CA); anti-PARP monoclonal antibody from BIOMOLInternational, L.P. (PlymouthMeeting,PA); anti-FAKmonoclonalantibodyfromUpstateBiotechnology(LakePlacid,NY);anti-phos-pho-FAK (Tyr397) rabbit polyclonal antibody from BioSourceInternational(Camarillo,CA).HumanrecombinantVEGF-AwasagiftfromGenentech(SouthSanFrancisco,CA).AllotherchemicalsandreagentswerepurchasedfromSigmaChemicalCorp.(St.Louis,MO). PANC-1 human pancreatic cancer cells were obtained fromthe American Type Culture Collection (ATCC, Rockville, MD).Anti-NIPantibodyisagiftfromDr.MichaelSimonsatDartmouthMedicalSchool.

Cell culture.PANC-1cellsweregrowninDMEMwith8%fetalbovine serum, 100 units/ml penicillin, 100 mg/ml streptomycin,250ng/mlamphotericinB.Cellsweremaintainedat37˚Cin95%airand5%CO2.

Construction of vectors. The Np-1 antisense cDNA wassubcloned into pLNCX2 vector. Authenticity was confirmed bysequencing. PANC-1 cells were transfected with the vector usingLipofectAMINE PLUS and stably transfected clones were selectedwithgeneticin (G418) inDMEM.Individualcloneswere isolated,then Np-1 antisense expression was Northern blot analysis andexpression of the 923 amino acid Np-1 protein was determinedbyWesternblottingwith anti-Np-1 (A-12) antibody.The forwardPrimer (5'- GAATTC ACC ATG GAG AGG GGG CTG CCG-3')andreverseprimer(5'-GCGGCCGCCCGTACAGCACGACCCCACAGACAG-3')wereusedtoamplifyacDNAencodinga truncated, 883 amino acid segment of Np-1 that was devoid oftheintracellulardomainandthatwenamedDNp-1.ThefulllengthNp-1andDNp-1cDNAweresubclonedintothepMHvectorwhichexpressesaC-terminalHAtag.PANC-1cellsweretransientlytrans-fectedwiththevectorusingserum-freemediumandLipofectAMINEPLUS.ExpressionoftheHA-taggedNp-1proteinwasconfirmedbyimmunoblottingwithananti-HA(Y-11)antibody.

Northern blot analysis. Total RNA was extracted, size frac-tioned, transferred onto nylon membranes, and UV cross-linked,as reported previously.17 Blots were hybridized overnight with the[33P]dCTP-labeled differential display Np-1 cDNA probe, washedextensively, and thenexposedat -80˚C toKodakBiomaxMS filmwith intensifying screens. A 190 bp 7S cDNA probe was used toconfirmequalRNAloadingandtransfer.17

Immunoblotting and immunoprecitation. For immunoblot-tingof plasmamembrane, cellswere cultured in8%FBS,washedin phosphate-buffered saline (PBS) and lysed in buffer containing20 mM HEPES pH 7.4, 1.5 mM MgCl2, 1 mM EGTA, andproteaseinhibitorsconsistingof2mMbenzamidine,1mg/mleachofaprotinin, leupeptin,andpepstatinand,and1mMphenylmethyl-sulfonyl fluoride (BALPP).The lysateswere centrifuged at 1500 xg for 10 min at 4˚C and the membrane pellet was solubilized for30minat4˚Cinbuffercontaining20mMHEPESpH7.4,1%Cholamidopropyl[dimethylammonio]-1-propanesulfonate (CHAPS),1 mM EDTA, 10% Glycerol, and BALLP. Insoluble material wasremoved by centrifugation at 14,000 x g at 4˚C for 20 min. Forimmunoblotting of total cell proteins, cells were lysed in buffercontaining 20 mM Tris/HCl pH 7.5, 0.1 M NaCl, 5 mM NaF,1%TritonX-100,1mMNa3VO4,10mMNa4P2O7,andBALLP.ProteinconcentrationwasdeterminedusingtheBCAproteinassaykit(Pierce,Rockford,IL).Followingfractionationona7.5%poly-acrylamidegel,proteinsweretransferredtoImmobilonPmembranesandblotted asdescribedpreviously.18Toperform immunoblottingof membrane proteins, proteins were incubated overnight at 4˚Cwith the indicated primary antibodies and for 60 min with thecorresponding horseradish-conjugated secondary antibodies. Afterwashing,boundantibodywasvisualizedusingenhancedchemilumi-nescence.Allexperimentswereperformedatleasttwicewithsimilarresults.Forproteinimmunoprecipitationstudies,celllysates(1mg)wereincubatedovernightat4˚Cwithanti-HAagaroseoranti-inte-grinb1,anti-FAKantibodiesfollowedbya2hourincubationwithproteinA/Gplusagarose.Thebeadswerethenwashedwith0.05%Tween20inTris-bufferedsaline(TBS),andtheproteinssubjectedtoSDS-PAGEgelsforimmunoblotting.

Cell proliferation assay.Toassessdoublingtime,cellswereseededin 6-well plates at a density of 3.0 x 104 cells/well and incubatedfortheindicatedtimes.Cellnumbersweredeterminedbycountingcellswithahemocytometer.Cellproliferationwasalsodeterminedbythe3-(4,5-methylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide(MTT)dye-reduction assay,whichmeasures the conversionof theMTT tetrazolium salt into MTT formazan.19 We have previouslyshownthat results fromthisassaycorrelatewellwithcell countingand[3H] thymidine incorporationdata.20-21Cellswereplatedatadensityof10,000cells/wellin96-wellplatesandincubatedat37˚Covernightpriortotreatment.Absorbancewasmeasuredat570nmwithamicrotiterplatereader.

Soft agar assay.Anchorage-independentgrowthwas assessedbyadouble-layer soft agar assay, as previously reported.22Briefly, 3 x103cellsweresuspendedincompletemediumcontaining0.3%agarand seeded in6-wellplatesonto abase layerof completemediumcontaining0.5%agar.Completemediumcontaining0.3%agarwasadded every three days for ten days before colony counting. Eachassay was performed in triplicate, and experiments were repeatedthreetimes.

Adhesion assay. Cells were suspended in serum-free mediumcontaining0.1%BSAwereseededatadensityof1x105cells/plateon nonadhesive Falcon 24-well culture plates (BD BiosciencesLabware,Bedford,MA)thatwerecoatedwithfibronectin, lamininorcollagenIV.Cellswerethenincubatedfor6hat37˚C,andwashedwith HBSS. The remaining adherent cells were trypsinized andcountedwith ahemocytometer. Inhibition studieswereperformedbypre-incubating thecells for60minat37˚Cwith5mg/mlanti-integrinb1antibody(6S6)orwithnormalmouseIgG,usingFalcon96-well nontissue culture-treated plates (BD Biosciences Labware)

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

Neuropilin-1InteractswithIntegrinb1

e3 CancerBiology&Therapy 2007;Vol.6Issue8

thatwerecoatedwithfibronectin,lamininorcollagenIV.Cellswerethen incubated for3h at 37˚C.AfterwashingwithPBS, adherentcellswere fixed in50ml of96%ethanol for10min, stainedwith50 ml of 0.1% crystal violet, rinsed with water and dried at roomtemperature.Stainedcellsweresolubilizedwith50mlof0.2%TritonX-100 and absorbance was measured at 570 nm with a microtiterplatereader.

Invasion assay. Invasion studies were performed as reportedpreviouslywithsomemodification.23-24Briefly,2.5x104cellsweresuspendedin500mlofserum-freemediumsupplementedwith0.1%BSA and placed onto the upper compartment of Matrigel-coatedtranswell chambers (8 mm pore size, BioCoat Matrigel InvasionChambers,Falcon24-wellcultureplates(BDBiosciencesLabware).Thelowercompartmentwerefilledwithmediumcontaining0.5%FBSintheabsenceorpresenceof1nMVEGF-A.After16h,cellsontheuppersurfaceofthefilterwerecarefullyremovedwithacottonswaband themembraneswere fixed inmethanolandstainedwithcrystal violet.The cells that had migrated through the membraneto the lower surface of the filter were counted in nine differentfieldsusingalightmicroscope(magnificationx200).Eachassaywasperformedinduplicate,andexperimentswererepeatedthreetimes.

Apoptosis assays. Apoptosis was assessed by monitoring PARPcleavage,measuringcaspaseactivity,andperformingFACSanalysis.To monitor PARP cleavage, 1 x 106 cells were seeded on 10 cmplates and incubated overnight, then treated with Actinomycin D(ActD) for 24 hr. Cells lysates were subjected to immunoblottingwith anti-PARP antibody (1:1000 dilution). Caspase-9 activitywas measured using a colorimetric assay kit from R&D Systems.BrieflyaftertreatmentwithActDfor24h,cellswerelysed,centri-fuged and assayed for caspase-9 activity using DEVD-pNA andLEHD-pNApeptidesubstrates.Opticaldensitywasmeasuredwithaspectrophotometeratawavelengthof405nm.ToperformFACSanalysis,cellswereincubatedintheabsenceorpresenceof500ng/mlActD for 24 h, harvested by trypsinization, stained with AnnexinV-PE and 7-Amino-actinomycin D (7-AAD) (BD Pharmingen).Cellswere thenanalyzedby flowcytometryusingaFACScan(BDBiosciences).25

Statistics. Student t test was used for statistical analysis of theexperiments.p<0.05wastakenasthelevelofsignificance.

resultsEffects of Np‑1 antisense on Np‑1 levels on the growth of

PANC‑1 cells.PANC-1cells express relativelyhigh levelsofNp-1butlowlevelsofNp-2,9andarethereforeidealforassessingtheroleofendogenousNp-1inpancreaticcancercells.Accordingly,PANC-1cellswere stably transfectedwithacDNAencodingthe full lengthNp-1intheantisenseorientation.ClonestransfectedwiththeNp-1antisensecDNAexhibitedtheantisensetranscriptbyNorthernblotanalysis(Fig.1A).Immunoblottingofcellmembranelysatesusingamonoclonal anti-Np-1 antibody revealed the presence of relativelyhigh levels of Np-1 in membranes from sham transfected cells,whereasmembranesfromthetwoclonesexhibitedmarkedlyreducedlevelsofNp-1(Fig.1B).

To assess the consequences of decreased Np-1 expression ondoublingtimes,cellgrowthwasmonitoredforsixdays(Fig.2A).Thecalculateddoublingtimeforshamtransfectedcellswas20.0hours.Bycontrast,doublingtimesfortheantisensetransfectedcloneswere30.6(p<0.002)and29.5hours(p<0.003),respectively.Moreover,by comparison with sham transfected cells, both Np-1 antisenseexpressing clones exhibited significantly decreased growth in softagar(Fig.2B).Thus,colonynumbersweredecreasedby35%(p<0.0002)and58%(p<0.002)withclones1and2,respectively.

Effects of Np‑1 suppression on actinomycin‑D‑mediated actions.Actinomycin-D(ActD)inhibitsgrowthandinducesapop-tosis in PANC-1 cells.17 Therefore, we next compared the effectsof ActD on growth and apoptosis in sham transfected and Np-1antisense expressing cells. Incubation with ActD for 48 h resultedindose-dependentdecrease incellgrowth. In theshamtransfectedcells, a maximal decrease of 73% occurred at a concentration of1000ng/mlActD(Fig.3).Bycontrast,bothNp-1antisenseclonesexhibitedasignificantlygreatergrowthinhibitoryeffectof86%and90%with the sameconcentrationofActD(Fig.3).ActD-inducedPARPcleavageandcaspase-9generationwerealsomorepronounced

Figure 1. Effects of anti‑sense Np‑1 on Np‑1 mRNA and protein expression. (A) Northern blot. Total RNA (20 mg/lane) from sham‑transfected (sham) and Np‑1 anti‑sense clones (C1, C2) was hybridized with a 32P‑labeled Np‑1 sense riboprobe. 7S cDNA was used as a loading and transfer control. (B) Np‑1 protein expression. Sham and Np‑1 anti‑sense clones were used to prepare membrane preparations. Membrane lysates (30 mg/lane) were subjected to immunoblot analysis using an anti‑Np‑1 antibody (A‑12). Western blotting with an anti‑EGF receptor (EGFR) antibody served as a loading control.

Figure 2. Effects of anti‑sense Np‑1 expression on cell growth. (A) Doubling time. Cells were seeded in 6‑well plates and incubated for the indicated times. Results are shown as means ± SE of triplicate determinations from three separate experiments. *p < 0.05 and **p < 0.01 when compared with sham. (B) Soft agar assay. Cells were seeded in 6‑well plates with 0.3% agar solution, and colonies were counted 10 days later. Data are the means ± SE of triplicate determinations from three separate experiments. *p < 0.0002 and **p < 0.002 when compared with sham transfected cells.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

www.landesbioscience.com CancerBiology&Therapy e4

Neuropilin-1InteractswithIntegrinb1

in both clones (Fig. 4). Moreover, following a 24 h of incubationwithActD,thepercentageofapoptoticcellsasdeterminedbyflowcytometry was markedly increased in both Np-1 antisense clonesbycomparisonwithshamtransfectedcells (Fig.5).Thus, thehighendogenouslevelsofNp-1inPANC-1cellscontributetoenhancedresistancetoapoptosis.

Effects of actinomycin D on apoptosis‑regulating proteins. InNp-1 antisense-transfected cells, ActD caused a marked decreasein the levels of Bcl-xL, without altering Bcl-2 levels (Fig. 6A). Bycontrast,ActDdidnot alterBcl-xL levels in sham-transfectedcells(Fig.6A).Moreover,ActDfailedtoalterphospho-AKT,AKT,Bad,Bax,orBaklevelsineithershamtransfectedorNp-1antisensetrans-fectedcells (Fig.6B).Thus,amongseveral importantproteins thatparticipate in the modulation of apoptosis susceptibility, only thelevelsofBcl-xLweredecreasedbyActDandthiseffectwasobservedonlyintheNp-1antisensetransfectedcells.

STAT5 is known to upregulate Bcl-xL expression, and bothSTAT5andSTAT3havebeenimplicatedintheregulationofapop-tosis.26Therefore,wenextsoughttodeterminewhetherthelevelsofthesetranscriptionfactorswerealteredinourcells.ActDdecreasedSTAT5levels intheNp-1antisenseexpressingcells,butnotinthesham-transfectedcells(Fig.7).Bycontrast,ActDdidnotalterSTAT3levelsineithershamorNp-1antisensetransfectedcells(Fig.7).

Figure 3. Effects of ActD on cell growth. Sham cells and Np‑1 anti‑sense clones were plated at a density of 10,000 cells/well (96‑well plates) and incubated overnight at 37˚C. Cells were then incubated for 48 h with the indicated concentrations of ActD. MTT assays were performed as described in “Materials and Methods.” Results are expressed as the percent change from respective controls, and are the means ± SE of six determinations from three independent experiments. *p < 0.04, †p < 0.03, and ††p < 0.02 when compared with the corresponding values in sham transfected cells.

Figure 4. Effects of ActD on markers of apoptosis. (A) PARP cleavage. Sham and Np‑1 anti‑sense clones were incubated for 24 hr with the indicated con‑centrations of ActD, and cell lysates were subjected to Western blot analysis with an antibody that recognizes PARP. The membrane was reprobed for ERK2 to assess protein loading. (B) Caspase‑9 activity. Sham and Np‑1 anti‑sense clones were incubated for 24 hr with the indicated concentrations of ActD, and caspase‑9 activities were measured in cell lysates. Data expressed as percent change above control, are the means ± SE of triplicate determi‑nations from three individual experiments. *p < 0.05; **p < 0.01 when compared with the corresponding values in sham transfected cells.

Figure 5. FACS analysis. Sham and Np‑1 anti‑sense cells were incubated for 24 h with 500 ng/ml of ActD. Cells were then stained with Annexin‑V and 7‑AAD, and FACS analysis was carried out. The cells at the upper right corner are either dead or undergoing apoptosis. The cells at the lower left corner are undergoing apoptosis. Results shown are representative of two experiments.

Figure 6. (A) Effects of ActD on Bcl‑2 and Bcl‑xL protein levels. Cells were incubated for 24 h in the absence or presence of the indicated concentra‑tions of ActD. Cell lysates (30 mg/lane) were then subjected to immunoblot analysis using anti‑Bcl‑2, and anti‑Bcl‑xL antibodies. The membrane was probed for ERK2 to assess protein loading. (B) Effects of ActD on AKT activa‑tion and AKT, Bad, Bax, Bak protein levels. Cells were incubated for 24 h in the absence or presence of the indicated concentrations of ActD. Cell lysates (30 mg/lane) were subjected to immunoblot analysis using anti‑phosphoAKT, anti‑AKT, anti‑Bad, anti‑Bax and anti‑Bak antibodies. The membrane was probed for ERK2 to assess protein loading.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

Neuropilin-1InteractswithIntegrinb1

e5 CancerBiology&Therapy 2007;Vol.6Issue8

Interactions of integrin b1 with Np‑1 in PANC‑1 Cells. TheobservationthatthelevelsofNp-1,acellsurfaceproteinthatisdevoidofintrinsicsignalingcapabilities,canparticipateinthemodulationof cell proliferation and apoptosis, raised thepossibility thatNp-1maybeabletointeractwithintegrins,alargefamilyofglycoproteins,consisting of heterodimers of a and b subunits.27This possibilitywas strengthened by previous reports that integrin b1 participatesin the regulation of STAT5 activation,28 and by our finding thatsuppression of Np-1 led to ActD-mediated downregulation ofSTAT5.Accordingly,wenextsoughttodeterminewhetherintegrinb1associatedwithNp-1.Tothisend,PANC-1cellsweretransientlytransfectedwithHA-taggedNp-1 cDNA. ImmunoprecipitationofNp-1withananti-HAantibodyfollowedbyimmunoblottingwith

ananti-integrinb1antibody revealeda strongbandcorrespondingtointegrinb1(Fig.8),indicatingthatNp-1formsacomplexwithintegrin b1. Next, PANC-1 cells were transiently transfected withan HA-tagged cDNA encoding a truncated Np-1 (DNp-1) that isdevoidoftheintracellulardomain.ImmunoprecipitationofDNp-1with an anti-HA antibody followed by immunoblotting with an

anti-integrinb1antibodyagainrevealedastrongbandcorrespondingtointegrinb1(Fig.8A),indicatingthatintegrin b1 associates with either the extracellular ortransmembrane domains of Np-1, but not with itsintracellulardomain.

TheC-terminaldomainofNp-1isknowntointeractwithadockingproteinthathasbeenvariablynamedasNp-1interactingprotein(NIP),GLUT1CBP(GIPC1),andsynectin.29ImmunoprecipitationofNp-1withan

Figure 7. Effects of ActD on STAT3 and STAT5 protein levels. Cells were incu‑bated for 24 h in the absence or presence of the indicated concentrations of ActD. Cell lysates (30 mg/lane) were subjected to immunoblot analysis using anti‑STAT3 and anti‑STAT5 antibodies. The membrane was probed for ERK2 to assess protein loading.

Figure 8. Association of integrin b1 with Np‑1. (A) Immunoprecipitation of full length Np‑1. The HA‑tagged full length Np‑1 was transiently expressed into PANC‑1 cells and lysates were immunoprecipitated with anti‑HA agarose. Immunoblotting was performed next using the anti‑integrin b1 anti‑body. (B) Immunoprecipitation of DNp‑1. The cDNA encoding HA‑tagged DNp‑1 was transiently expressed into PANC‑1 cells and lysates were immu‑noprecipitated with anti‑HA agarose, followed by immunoblotting with the anti‑NIP antibody.

Figure 9. Effects of anti‑sense Np‑1 anti integrin‑b1 antibod‑ies on cell adhesion. A‑C, Effects of Np‑1 suppression. Sham transfected PANC‑1 cells and clones 1 (C1) and 2 (C2) expressin the Np‑1 antisense cDNA were seeded on plates coated with fibronectin (A), laminin (B), and collagen IV (C), and incubated for 6h at 37˚C. After washing, the adherent cells were counted. Data are the means ± SE of triplicate determinations from three separate experiments. *p < 0.01, **p < 0.002, †p < 0.001, ††p < 0.002, #p < 0.003, and #p < 0.02 when compared with the corresponding values in sham transfected cells. D‑F. Effects of anti‑integrin b1 blocking antibody on cell adhesion. Cells were incubated for 60 min at 37˚C in serum‑free medium containing 0.1% BSA in the absence or presence of 5 mg/ml anti‑integrin b1 antibody (6S6) or normal mouse IgG. Cells were then seeded in 96‑well non tissue culture‑treated plates coated with fibronectin (D), laminin (E), and collagen IV (F). Following incubation for 3h at 37˚C cells were washed, and adherent cells were fixed in ethanol and stained with crystal violet. Stained cells were solu‑bilized with Triton X‑100 and absorbance was measured with a plate reader. Data are the means ± SE of six determinations from three independent experiments. ***p < 0.05, †††p < 0.02, and ## p < 0.02 when compared with values observed with normal mouse IgG.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

www.landesbioscience.com CancerBiology&Therapy e6

Neuropilin-1InteractswithIntegrinb1

anti-HA antibody followed by immunoblotting with an anti-NIPantibody revealed the expected association of Np-1 with NIP(Fig. 8A). By contrast, immunoprecipitation of DNp-1 followedby immunoblotting with an anti-NIP antibody did not reveal anyproteinbands(Fig.8B).

Effect of Np‑1 suppression on cell adhesion, invasion and FAK phosphorylation.Integrinsmediatetheadhesionofcellstocompo-nents of the extracellular matrix, and this effect activates a varietyof signaling pathways that are important for the regulation of celladhesionandinvasion.30InviewofourfindingthatNp-1interactswith integrin b1, we next sought to compare the effects of Np-1suppressionand integrinb1blockadeon theadhesionofPANC-1cells on three different substrates. By comparison with shamtransfected cells, both Np-1 antisense expressing clones exhibitedsignificantly decreased cell adhesion to fibronectin, laminin, andcollagen IV (Fig. 9). Similarly, by comparison with cells treatedwithanon-specificcontrolIgG,cellstreatedwithanantibodythatdisrupts integrin b1 function exhibited significantly decreased celladhesiontofibronectin,laminin,andcollagenIV(Fig.9).

Invivo,cancercelladhesionisaprocessthatinvolvescancercellsinteracting with adjacent cancer cells as well as with componentsof the extracellular matrix.31 In general, cancer cells that are lessadherent are more likely to be invasive, yet successful tumor celladhesionisanimportantfeatureofthemetastaticprocess.Itwasofinterest,therefore,todeterminewhethersuppressionofNp-1expres-sion,whichresultedinattenuatedcelladhesion,wouldleadtoalteredinvasivecapacityofthecancercells.Accordingly,invasionassayswereperformed next. By comparison with sham transfected cells, bothNp-1antisenseexpressingclonesexhibitedasignificantlydecreasedability to invade across a matrigel membrane (Fig. 10). Moreover,VEGF did not enhance the invasiveness of sham transfected cells,anddidnotrestoretheinvasivecapacityofNp-1antisenseexpressingclones(Fig.10).

FAK is a cytoplasmic tyrosine kinase that colocalizes with inte-grins within focal contacts in adherent cells and participates inintegrin-mediated signal transductionpathways.32FAK isactivatedfollowing tyrosine phosphorylation, which can lead to the activa-tionofsrc,MAPKandPI3-kinasepathways.32GiventheabilityofNp-1tointeractwithintegrinb1,wenextsoughttodeterminetheconsequences of Np-1 suppression on FAK phosphorylation. FAK

levelswereslightlyincreasedinNp-1antisenseexpressingclonesbycomparisonwithsham-transfectedcells(Fig.11).Moreimportantly,therewasadramaticupregulationoftyrosinephosphorylationintheNp-1antisenseexpressingclones,thatwasalsoreadilyevidentwithrespecttophosphorylationontyrosineresidue397(Fig.11),asiteknowntoreflectFAKactivation.

dIscussIonNp-1functionsasacoreceptor forbothSema3AandVEGF-A,

facilitating their interactions with their respective high affinityreceptors.13Invivostudieswithtransgenicmicerevealedthatover-expressionofNp-1resultedinembryonic lethalityduetoexcessivecapillaries and blood vessel formation, cardiac malformation, andectopic sprouting and defasciculation of nerve fibers.33 However,Np-1knock-outmicealsodiedinuteroandexhibitedvarioustypesofvasculardefectsincludingtranspositionofthegreatvesselsandthedisorganized and insufficient development of vascular networks intheyolksac.34TheseresultsindicatethatappropriatelevelsofNp-1arerequiredfornormalembryonicvasculogenesisandangiogenesis.

Np-1isoverexpressedinthecancercellsinPDAC,9andsuppres-sionofNp-1expression inPANC-1cellshasbeenassociatedwithenhanced chemosensitivity to gemcitabine,35 raising the possibilitythat Np-1 may contribute to chemoresistance in PDAC. In thepresent study, down regulation of Np-1 increased ActD- inducedpro-apoptoticactions,asevidencedbyenhancedPARPcleavageandcaspase-9 activity, an by Annexin-V staining. ActD also decreasedBcl-xL,andSTAT5levelsintheantisenseexpressingcells,butnotintheshamtransfectedcells.However,ActDdidnotalterBcl-2,phos-pho-AKT,AKT,Bad,Bax,Bak,orSTAT3levels.TheseobservationssuggestthatBcl-xLandSTAT5maycontributetotheanti-apoptoticeffectsofhighlevelsofNp-1.

Np-1 suppression in PANC-1 cells has been associated withenhanced phosphorylation of mitogen activated protein kinase(MAPK) and with upregulation of phospho-Akt,35,36 even thoughtheactivationofMAKPandAktgenerallyleadstoenhancedmito-genesisandsurvival.However,inthepresentstudy,Np-1antisenseexpressing clones did not exhibit increased phospho-AKT levels(Fig. 6B). Moreover, we did not find that these clones exhibitedincreasedactivationofeitherMAPKor junkinase (JNK) levels,asdeterminedusingthecorrespondingphospho-specificantibodies(datanotshown).Instead,suppressionofNp-1levelsresultedinprolonged

Figure 11. Effects of anti‑sense Np‑1 on tyrosine phosphorylation of FAK. Cell lysates were prepared from sham‑transfected and Np‑1 anti‑sense expressing clones that were grown in 10 cm culture plates in the presence of 10% serum and allowed to become 80% confluent. Immunoblotting was performed using anti‑phospho FAK (Tyr397) or anti‑phosphotyrosine antibodies. To assess FAK protein levels, cell lysates (30 mg/lane) were also subjected to immunoblot‑ting using anti‑FAK antibody. Membranes were reprobed for ERK2 to assess equivalent loading of lanes.

Figure 10. Invasion assay. Sham‑transfected and Np‑1 anti‑sense expressing clones were seeded on Matrigel in the upper cell culture inserts of a 24‑well invasion chamber. The lower compartment was filled with the medium containing 0.5% FBS, in the absence or presence of 1 nM VEGF. After 16 h, the membranes were fixed in methanol and stained with crystal violet. The cells that had migrated through the membrane to the lower surface of the filter were counted in nine different fields using a light microscope (magnification ×200). Data are the means ± SE of triplicate determinations from three separate experiments. *p < 0.03 and †p < 0.02 when compared with values for the corresponding sham transfected cells.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

Neuropilin-1InteractswithIntegrinb1

e7 CancerBiology&Therapy 2007;Vol.6Issue8

doubling times and decreased anchorage independent growth,indicating that Np-1 may confer a growth advantage to PANC-1cells and may enhance their tumorigenic potential. Inasmuch asSTAT5activationleadstotranscriptionalregulationofseveralgenes,includingthoseinvolvedintheregulationofcellsurvivalandprolif-eration,37itispossiblethatNp-1allowsforenhancedproliferation,inpart,byupregulatingSTAT5.

Integrinsarelargefamilyofcell-surfacereceptorswhichmediatetheeffectsoftheextracellularmatrix(ECM)andbindECMcompo-nents,organizethecytoskeleton,andactivateintracellularsignalingpathway.31 In thepresent studywedetermined thatNp-1 forms acomplexwith integrinb1,andthat this interactiondoesnotoccurthroughtheintracellulardomainofNp-1.SeverallinesofevidencesuggestthatNp-1interactionswithintegrinb1couldbebiologicallymeaningfulinPDAC.First,attenuatedexpressionofNp-1resultedindecreasedadhesionofPANC-1cells tofibronectin, lamininandcollagen type IV, and this effect was mimicked by an integrin b1antibody that interrupts integrin b1 function. Second, attenuatedexpression of Np-1 resulted in decreased Bcl-xL expression in thepresence of ActD, and integrin b1 is known to participate in theregulation of STAT5 activation,29 whereas STAT5 is known toinduce the expression of Bcl-xL.38,39 In this context, low levels ofNp-1 would interfere with STAT5 activation, thereby preventingSTAT5-mediated induction of Bcl-xL.Third, a2b1, a5b1, a6b1integrins are the predominant adhesion receptors for fibronectin,laminin,andcollageninpancreaticcancercells,40underscoringtheimportanceofintegrinb1inthisprocess.Fourth,a2,a3,a5,a6,and aV as well as b1, b3, and b4 are overexpressed in PDAC,41raising the possibility that the concomitant abundance of Np-1and integrin b1 in this malignancy may contribute to enhancedstromal-cancercellinteractionsthatleadtotheaberrantactivationofgrowthandsurvivalpathwaysinthecancercells.

To date, all of the biological actions of Np-1 in endothelialcells or in cancer cells havebeen attributed to its coreceptor func-tions in relation to ligands such as VEGF and Semaphorins.42-45Moreover, the SEA domain of Np-1 interacts with GIPC, a PDZdomain protein that associates with at least 25 binding partners,many of which are either adhesion molecules or transmembranereceptors.42Itispossible,therefore,thatNp-1maysignalindirectlythrough such mechanisms.To our knowledge, however, Np-1 hasnotpreviouslybeenshowntointeractwithintegrins,andtheinter-actionofb1integrinwithNp-1occurredevenintheabsenceoftheintracellular domain of Np-1. Moreover, VEGF did not alter theinvasivenessofPANC-1cells,irrespectiveofthelevelsofNp-1,yetthesecells expressVEGFreceptors.46Thus, in spiteof theabsenceofdirectevidencethatendogenousNp1interactedwithb1integrinsin PANC1 cells, Np-1 exerted its effects in a ligand independentmanner,anditsattenuatedexpressionresultedindecreasedinvasivepropertiesofPANC-1cells.Itislikely,therefore,thatthedecreasedexpressionand/oractivationofintegrinb1mayexplaintheattenu-atedinvasivenessofPANC-1cellswhenNp-1levelsaresuppressed,pointingtoanovelmechanismofactionforNp-1.Insupportofthisconclusion,previousstudiesthatrevealedthatintegrinb1isrequiredfor the invasive characteristics of lymphomas47 and squamous cellcarcinomas.48

In cancer cells, enhanced cellular invasion is usually associatedwithacceleratedmigration,decreasedadhesion,and increasedFAKactivation, whereas inhibition of FAK activity is associated withattenuated migration and invasion.32 By contrast, in the presentstudy, FAK phosphorylation on tyrosyl residue 397 and overall

tyrosine phosphorylation were markedly increased in the clonesthat expressed low Np-1 levels, whereas both invasion and adhe-sion were greatly attenuated.Taken together with our integrin b1data, these observations raise the possibility that Np-1 functionstocoordinate integrin-mediatedadhesionandmigrationwithFAKphosphorylationandactivation,andthatintheabsenceofNp-1thisintegration is lost.Thus, in addition topromoting tumorprogres-sionbyenhancingtumorangiogenesisandvascularremodeling,49-50Np-1mayconferagrowth,survivalandinvasiveadvantagetocancercells by promoting more efficient interactions with integrin b1.Conceivably, integrin interaactions with Np-1 may also be occur-ring in ednothelial cells.Therefore, targeting Np-1 in PDAC mayenhancetheeffectivenessofothertherapeuticmodalitiesbyattenu-atingangiogenesis,bysuppressingcellularinvasionandbyenhancingcancerchemosensitivity.

References 1. JemalA,MurrayT,WardE,SamuelsA,TiwariRC,GhafoorA,FeuerEJ,ThunMJ.Cancer

statistics,2005.CACancerJClin2006;56:106-30. 2. Kern SE. Molecular genetic alterations in ductal pancreatic adenocarcinomas. Med Clin

NorthAm2000;84:691-5. 3. Korc M. Role of growth factors in pancreatic cancer. Surg Oncol Clin N Am 1998;

7:25-41. 4. KorcM.Pathwaysforaberrantangiogenesisinpancreaticcancer.MolCancer2003;2:1-8. 5. WangW,AbbruzzeseJL,EvansDB,LarryL,ClearyKR,ChiaoPJ.Thenuclearfactor-kappa

BRelAtranscriptionfactorisconstitutivelyactivatedinhumanpancreaticadenocarcinomacells.ClinCancerRes1999;5:119-27.

6. Cheng JQ, Ruggeri B, Klein WM, Sonoda G, Altomare DA, Watson DK, Testa JR.AmplificationofAKT2inhumanpancreaticcellsandinhibitionofAKT2expressionandtumorigenicitybyantisenseRNA.ProcNatlAcadSciUSA1996;93:3636-41.

7. ScholzA,HeinzeS,DetjenKM,PetersM,WelzelM,HauffP,SchirnerM,WiedenmannB, Rosewicz S. Activated signal transducer and activator of transcription 3 (STAT3)supports the malignant phenotype of human pancreatic cancer. Gastroenterology 2003;125:891-905.

8. HeZ,Tessier-LavigneM.Neuropilinisareceptorfortheaxonalchemorepellentsemapho-rinIII.Cell1997;90:739-51.

9. FukahiK,FukasawaM,NeufeldG,ItakuraJ,KorcM.Aberrantexpressionofneuropilin-1and-2inhumanpancreaticcancercells.ClinCancerRes2004;10:581-90.

10. ParikhAA,LiuWB,FanF,StoeltzingO,ReinmuthN,BrunsCJ,BucanaCD,EvansDB,EllisLM.Expressionandregulationofthenovelvascularendothelialgrowthfactorreceptorneuropilin-1 by epidermal growth factor in human pancreatic carcinoma. Cancer 2003;98:720-9.

11. SokerS,TakashimaS,MiaoHQ,NeufeldG,KlagsbrunM.Neuropilin-1isexpressedbyendothelialandtumorcellsasanisoform-specificreceptorforvascularendothelialgrowthfactor.Cell1998;92:735-45.

12. Kolodkin AL, Levengood DV, Rowe EG,TaiYT, Giger RJ, Ginty DD. Neuropilin is asemaphorinIIIreceptor.Cell1997;90:753-62.

13. Neufeld G, Cohen T, Shraga N, Lange T, Kessler O, Herzog Y. The Neuropilins:MultifunctionalsemaphorinandVEGFreceptorsthatmodulateaxonguidanceandangio-genesis.TrendsCardiovascMed2002;12:13-9.

14. DeVriesL,LouX,ZhaoG,ZhengB,FarguharMG.GIPC,aPDZdomaincontainingprotein,interactsspecificallywiththeCterminusofRGS-GAIP.ProcNatlAcadSciUSA1998;95:12340-5.

15. Cai H, Reed RR. Cloning and characterization of neuropilin-1-interacting protein: APSD-95/Dlg/ZO-1domain-containingproteinthatinteractswiththecytoplasmicdomainofneuropilin-1.JNeurosci1999;19:6519-27.

16. WestDC,ReesCG,DuchesneL,PateySJ,TerryCJ,TurnbullJE,DeleheddeM,HeegaardCW,AllainF,VanpouilleC,RonD,FernigDG.Interactionsofmultipleheparinbindinggrowthfactorswithneuropilin-1andpotentiationoftheactivityoffibroblastgrowthfac-tor-2.JBiolChem2005;280:13457-64.

17. KleeffJ,KornmannM,SawhneyH,KorcM.ActinomycinDinducesapoptosisandinhibitsgrowthofpancreaticcancercells.IntJCancer2000;86:399-407.

18. KornmannM,DanenbergKD,ArberN,BegerHG,DanenbergPV,KorcM.InhibitionofcyclinD1expressioninhumanpancreaticcancercellsisassociatedwithincreasedchemo-sensitivityanddecreasedexpressionofmultiplechemoresistancegenes.CancerRes1999;59:3505-11.

19. MosmannT.Rapidcolorimetricassayforcellulargrowthandsurvival:Applicationtopro-liferationandcytotoxicityassays.JImmunolMethods1983;65:55-63.

20. Raitano AB, Korc M.Tumor necrosis factor upregulates gamma-interferon binding in ahumancarcinomacellline.JBiolChem1990;265:10466-72.

21. BaldwinRL,KorcM.Growthinhibitionofhumanpancreaticcarcinomacellsbytransform-inggrowthfactorbeta-1.GrowthFactors1993;8:23-34.

22. KornmannM,ArberN,KorcM. Inhibitionof basal andmitogen-stimulatedpancreaticcancer cell growth by cyclin D1 antisense is associated with loss of tumorigenicity andpotentiationofcytotoxicitytocisplatinum.JClinInvest1998;101:344-52.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014

©2007 L

ANDES BIOSCI

ENCE.

DO NOT DIST

RIBUTE.

www.landesbioscience.com CancerBiology&Therapy e8

Neuropilin-1InteractswithIntegrinb1

23. AlbiniA,IwamotoY,KleinmanHK,MartinGR,AaronsonSA,KozlowskiJM,McEwanRN.Arapidinvitroassayforquantitatingtheinvasivepotentialoftumorcells.CancerRes1987;47:3239-45.

24. Rowland-Goldsmith MA, Maruyama H, Kusama T, Ralli S, Korc M. Soluble type IItransforminggrowth factor-b (TGF-b) receptor inhibitsTGF-b signaling inCOLO-357pancreatic cancer cell in vitro and attenuates tumor formation. Clin Cancer Res 2001;7:2931-40.

25. HeraultO,ColombatP,DomenechJ,DegenneM,BremondJL,SensebeL,BernardMC,BinetC.Arapidsingle-laserflowcytometricmethodfordiscriminationofearlyapoptoticcellsinaheterogenouscellpopulation.BrJHaematol1999;104:530-7.

26. YuH,JoveR.TheSTATsofcancer-Newmoleculartargetscomeofage.NatRevCancer2004;4:97-105.

27. HynesRO.Integrins:Bidirectional,allostericsignalingmachines.Cell2002;110:673-87. 28. Brizzi MF, Defilippi P, Rosso A, Venturino M, Garbarino G, Miyajima A, Silengo L,

TaroneG,PegoraroL. Integrin-mediatedadhesionofendothelialcells inducesJAK2andSTAT5A activation: Role in the control of c-fos gene expression. Mol Biol Cell 1999;10:3463-3471.

29. NaccacheSN,HassonT,HorowitzA.BindingofinternalizedreceptorstothePDZdomainofGIPC/synectinrecruitsmyosinVItoendocyticvesicles.ProcNatlAcadSciUSA2006;103:15735-40.

30. GuoW,GiancottiFG.Integrinsignallingduringtumourprogression.NatRevMolCellBiol2004;5:816-26.

31. LiottaLA,SteegPS,Stetler-StevensonWG.Cancermetastasisandangiogenesis:Animbal-anceofpositiveandnegativeregulation.Cell1991;64:327-36.

32. ParsonsJT.Focaladhesionkinase:Thefirsttenyears.JCellSci2003;116:1409-16. 33. KitsukawaT,ShimonoA,KawakamiA,KondohH,FujisawaH.Overexpressionofmem-

braneprotein,neuropilin,inchimericmicecausesanomaliesinthecardiovascularsystem,nervoussystemandlimbs.Development1995;121:4309-18.

34. KawasakiT,KitsukawaT,BekkuY,MatsudaY,SanboM,YajiT,FujisawaH.Arequirementforneuropilin-1inembryonicvesselformation.Development1999;126:4895-902.

35. WeyJS,GrayMJ,FanF,BelchevaA,McCartyMF,StoeltzingO,SomcioR,LiuW,EvansDB,KlagsbrunM,GallickGE,EllisLM.Overexpressionofneuropilin-1promotesconsti-tutiveMAPKsignallingandchemoresistanceinpancreaticcancercells.BrJCancer2005;93:233-41.

36. GrayMJ,WeyJS,BelchevaA,McCartyMF,TrevinoJG,EvansDB,EllisLM,GallickGE.Neuropilin-1suppressestumorigenicpropertiesinahumanpancreaticadenocarcinomacelllinelackingneuropilin-1coreceptors.CancerRes2005;65:3664-70.

37. PaukkuK,SilvennoinenO.STATsascriticalmediatorsofsignaltransductionandtranscrip-tion:LessonslearnedfromSTAT5.CytokineGrowthFactorRev2004;15:435-55.

38. SocolovskyM,FallonAE,WangS,BrugnaraC,LodishHF.FetalanemiaandapoptosisofredcellprogenitorsinStat5a-/-5b-/-mice:AdirectroleforStat5inBcl-X(L)induction.Cell1999;98:181-91.

39. LordJD,McIntoshBC,GreenbergPD,NelsonBH.TheIL-2receptorpromoteslympho-cyteproliferationandinductionofthec-myc,bcl-2,andbcl-xgenesthroughthetrans-acti-vationdomainofStat5.JImmunol2000;164:2533-41.

40. AraoS,MatsumotoA,OtsukiM.Beta1 integrinsplay an essential role in adhesionandinvasionofpancreaticcarcinomacells.Pancreas2000;20:129-37.

41. LohrM,TrautmannB,GottlerM,PetersS,ZaunerI,MaierA,KloppelG,LiebeS,KreuserED.Expressionandfunctionofreceptorsforextracellularmatrixproteinsinhumanductaladenocarcinomasofthepancreas.Pancreas1996;12:248-59.

42. Guttmann-Raviv N, Kessler O, Shraga-Heled N, Lange T, Herzog Y, Neufeld G. TheNeuropilins and their role in tumorigenesis and tumor progression. Cancer Lett 2006;231:1-11.

43. Bielenberg DR, Pettaway CA, Takashima S, Klagsbrun M. Neuropilins in neoplasms:Expression,regulation,andfunction.ExpCellRes2006;312:584-93.

44. OsadaR,HoriuchiA,KikuchiN,OhiraS,OtaM,KatsuyamaY,Konishi I.Expressionof semaphorins,vascularendothelialgrowth factor,andtheircommonreceptorneuropil-insandalleic lossof semaphorin locus inepithelialovarianneoplasms: Increased ratioofvascular endothelial growth factor to semaphorin is a poor prognostic factor in ovariancarcinomas.HumPathol2006;37:1414-25.

45. Nguyen QD, Rodrigues S, Rodrigue CM, Rivat C, Grijelmo C, Bruyneel E, Emami S,AttoubS,GespachC. Inhibitionof vascular endothelial growth factor (VEGF)-165 andsemaphorin 3A-mediated cellular invasion and tumor growth by the VEGF signalinginhibitor ZD4190 in human colon cancer cells and xenografts. Mol CancerTher 2006;5:2070-7.

46. Itakura J, Ishiwata T, Shen B, Kornmann M, Korc M. Concomitant overexpression ofvascularendothelialgrowthfactoranditsreceptorsinpancreaticcancer.IntJCancer2000;85:27-34.

47. StroekenPJ,vanRijthovenEA,vanderValkMA,RoosE.Targeteddisruptionofthebeta1integringeneinalymphomacelllinegreatlyreducesmetastaticcapacity.CancerRes1998;58:1569-77.

48. BrockbankEC,BridgesJ,MarshallCJ,SahaiE.Integrinbeta1isrequiredfortheinvasivebehaviourbutnotproliferationofsquamouscellcarcinomacellsinvivo.BrJCancer2005;92102-12.

49. MiaoHQ,LeeP,LinH,SokerS,KlagsbrunM.Neuropilin-1 expressionby tumor cellspromotestumorangiogenesisandprogression.FASEBJ2000;14:2532-9.

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a, S

an F

ranc

isco

] at

10:

44 0

4 D

ecem

ber

2014