Characterization of Ion Channels Involved in the Proliferative Response of Femoral Artery Smooth Muscle Cells Pilar Cidad, Alejandro Moreno-Domínguez, Laura Novensá, Mercé Roqué, Leire Barquín, Magda Heras, M. Teresa Pérez-García and José R. López-López Arterioscler Thromb Vasc Biol 2010, 30:1203-1211: originally published online March 18, 2010 doi: 10.1161/ATVBAHA.110.205187 Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association. 7272 Greenville Avenue, Dallas, TX 72514 Copyright © 2010 American Heart Association. All rights reserved. Print ISSN: 1079-5642. Online ISSN: 1524-4636 The online version of this article, along with updated information and services, is located on the World Wide Web at: http://atvb.ahajournals.org/content/30/6/1203 Data Supplement (unedited) at: http://atvb.ahajournals.org/content/suppl/2010/03/17/ATVBAHA.110.205187.DC1.html Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online at http://atvb.ahajournals.org//subscriptions/ Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, a division of Wolters Kluwer Health, 351 West Camden Street, Baltimore, MD 21202-2436. Phone: 410-528-4050. Fax: 410-528-8550. E-mail: [email protected] Reprints: Information about reprints can be found online at http://www.lww.com/reprints Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 Cell Biology/Signaling Characterization of Ion Channels Involved in the Proliferative Response of Femoral Artery Smooth Muscle Cells Pilar Cidad, Alejandro Moreno-Domínguez, Laura Novensa´, Merce´ Roque´, Leire Barquín, Magda Heras, M. Teresa Pe´rez-García, Jose´ R. Lo´pez-Lo´pez Objective—Vascularsmoothmusclecells(VSMCs)contributesignificantlytoocclusivevasculardiseasesbyvirtueoftheir abilitytoswitchtoanoncontractile,migratory,andproliferatingphenotype.Althoughtheparticipationofionchannels in this phenotypic modulation (PM) has been described previously, changes in their expression are poorly defined becauseoftheirlargemoleculardiversity.Weobtainedaglobalportraitofionchannelexpressionincontractileversus proliferating mouse femoral artery VSMCs, and explored the functional contribution to the PM of the most relevant changes that we observed. Methods and Results—High-throughput real-time polymerase chain reaction of 87 ion channel genes was performed in 2 experimental paradigms: an in vivo model of endoluminal lesion and an in vitro model of cultured VSMCs obtainedfromexplants.mRNAexpressionchangesshowedagoodcorrelationbetweenthe2proliferativemodels, with only 2 genes, Kv1.3 and Kv(cid:1)2, increasing their expression on proliferation. The functional characterization demonstrates that Kv1.3 currents increased in proliferating VSMC and that their selective blockade inhibits migration and proliferation. Conclusion—These findings establish the involvement of Kv1.3 channels in the PM of VSMCs, providing a new therapeutical target for the treatment of intimal hyperplasia. (Arterioscler Thromb Vasc Biol. 2010;30:1203-1211.) Key Words: gene expression (cid:1) ion channels (cid:1) restenosis (cid:1) vascular biology (cid:1) vascular muscle (cid:1) Kv1.3 channels (cid:1) vascular remodeling Vascular smooth muscle cells (VSMCs) are differenti- programs to readjust cellular activity to mechanical and atedcellsthatregulatevesseldiameteranddetermine hormonal factors.1–3 tissue perfusion. However, they can exhibit a variety of See accompanying article on page 1073 functionally dissimilar phenotypes. In response to local The switch in ion transport mechanisms associated with cues, VSMCs experience a phenotypic modulation (PM), PM is getting increasing amounts of attention. Coordinate with profound and reversible changes leading to prolifer- changes in ion channels are an integral component of ation, migration, and secretion of extracellular matrix VSMCplasticity,astheycanredirectbiochemicalactivity components.1 This plasticity is essential for injury repair, toward new functional responses.4,5 Moreover, both con- but it also contributes to the development and progression tractileandproliferativesignalsrequirespecificchangesin ofvasculardiseaseinresponsetoabnormalenvironmental intracellular [Ca2(cid:1)] and membrane potential that are de- signals. It is becoming evident that contractile and prolif- termined by the ion channels expressed in VSMCs. Re- erative phenotypes represent extreme cases of a spectrum modeling of several ion channels has shown to be func- of phenotypes that may coexist as the result of a develop- tionally important for the PM of VSMCs in several mentally regulated genetic program constantly modulated preparations.5–10 These data contribute to our understand- by environmental cues. This explains both a relatively ing of VSMC modulation, but also, importantly, they can stable expression of certain transcriptional programs in providenewtargetsforthetreatmentofvasculardisorders. different VSMCs and a marked plasticity of these cells, Todate,studiesofionchanneldistributionwithinspecific including the ability to respond with different genetic vascular beds and the modifications of ion channels on Receivedon:April3,2009;finalversionacceptedon:March10,2010. FromDepartamentodeBioquímicayBiologíaMolecularyFisiologíaeInstitutodeBiologíayGene´ticaMolecular,UniversidaddeValladolid yConsejosuperiordeInvestigacionesCientificas,Valladolid,Spain(P.C.,A.M.-D.,M.T.P.-G.,J.R.L.-L.);ServiciodeCardiología.InstitutClinic delTo`rax,HospitalClinic,InstitutodeInvestigacionesBiomedicalsAugustPiiSunyer,UniversidaddeBarcelona,Barcelona,Spain(L.N.,M.R., L.B.,M.H.). CorrespondencetoJose´ Ramo´nLo´pez-Lo´pez,DepartamentodeBioquímicayBiologíaMolecular,UniversidaddeValladolid,EdificioInstitutode BiologíayGene´ticaMolecular,c/SanzyFore´ss/n,47003Valladolid,[email protected] ©2010AmericanHeartAssociation,Inc. ArteriosclerThrombVascBiolisavailableathttp://atvb.ahajournals.org DOI:10.1161/ATVBAHA.110.205187 Downloaded from http://atvb.ahajou1r2n0a3ls.org/ by guest on January 26, 2012 1204 Arterioscler Thromb Vasc Biol June 2010 remodeling have been limited to a small number of tionsofthepatch-clamptechniquewereusedforfunctionalstudies candidate subunits believed to be of importance. In addi- inisolatedVSMCs8,12 tion,thecomplexityoftheircharacterizationintheinvivo models has led most investigators to extrapolate from the Migration and Proliferation Studies data obtained from cultured VSMCs as a proliferative Migration of cultured VSMCs was studied by a scratch assay and phenotype, because VSMCs in intimal hyperplasic lesions proliferation was determined by 5-bromo-2(cid:3)-deoxyuridine (BrdU) resemble dedifferentiated myofibroblasts or cultured incorporation. VSMCs. However, the emerging picture regarding the phenotypic regulation of VSMCs shows a large degree of Results diversity, reflecting both intrinsic variability of VSMCs Expression Profile of Ion Channel Genes in among vascular beds and the specific responses to the Femoral VSMCs different proliferative signals present in cultured cells and The transcriptional expression levels of 87 ion channel in the in vivo lesions. subunits in VSMCs from mouse femoral arteries were Here, we used a quantitative approach to obtain a global investigatedbyreal-timepolymerasechainreaction.These portraitofionchannelgeneexpressionincontractileVSMCs channelgenesinclude(cid:2),(cid:1),and(cid:3)subunitsofK(cid:1)channels from mouse femoral arteries and their changes on PM in 2 and (cid:2)subunits of voltage-dependent Ca2(cid:1) channels, Cl(cid:4) proliferative models: an in vivo model of neointimal channels, and Trp channels. The expression levels of hyperplasia induced by endoluminal lesion11 and an in markers of VSMCs (calponin, Cnn1) and endothelium vitro model using cultured VSMCs from arterial explants. Two genes (Kv1.3 and Kv(cid:1)2) showed a concordant up- (endothelial nitric oxide synthase, Nos3, and von Wille- brand factor), were also explored, as well as other endog- regulationinbothmodels,andtheexpressionandfunction enous controls (Gapdh, B2m, Hprt-1, and Klf5). Expres- ofKv1.3proteinswasexplored.Electrophysiologicalstud- sion of 26 channel genes was undetectable after 40 cycles ies in cultured VSMCs and VSMCs from injured arteries of amplification under all conditions. showanincreasedfunctionalexpressionofKv1.3currents. Therelativeabundanceofthe54channelgenesfoundin Pharmacological or genetic Kv1.3 blockade inhibits cul- the control preparation (mRNA from intact femoral arter- tured VSMC migration and proliferation, and this inhibi- ies, with endothelium, C ) is shown in Figure 1. Genes tion cannot be mimicked by VSMC depolarization. Fi- E(cid:1) are grouped by families, and their expression levels are nally, the analysis of other vascular beds confirms that upregulationofKv1.3associateswithPM,suggestingthat normalized to the endogenous control ribosomal protein they can represent a good therapeutical target by which to 18S.Wedetectedexpressionof1Cl(cid:4)channel(Clcn3)and control VSMC proliferation. several inward rectifier and 2-pore domain K(cid:1) channels. We found expression of all members of the Ca2(cid:1)- dependentK(cid:1)channels(K )family,withtheexceptionof Ca Materials and Methods themaxiK(BK )(cid:1)subunitBK(cid:1)2;ofthem,theBK(cid:2)and Ca AnexpandedMaterialsandMethodssectionisgiveninthesupple- BK(cid:1)1subunitswereamongthemostabundantlyexpressed mentalmaterial,availableonlineathttp://atvb.ahajournals.org. channel genes. Within the voltage-dependent K(cid:1) channels (Kv channels), we found expression of several accessory Animals subunits, as well as pore-forming (cid:2)subunits of members Bloodpressurenormal(BPN)mice(TheJacksonLaboratory)were of the Kv1 to Kv4 and Kv11 subfamilies. Voltage- maintainedwithinbredcrossingintheanimalfacilitiesoftheschools dependent Ca2(cid:1) channels are represented by Cav1 and ofmedicineoftheuniversitiesofValladolidandBarcelona.Unilat- Cav3 subfamily members. Finally, we detected expression eral transluminal injury of femoral arteries was performed by passageofa0.25-mm-diameterangioplastyguidewire,aspreviously ofalargenumberofTrpchannelsoftheTrpc,Trpv,Trpm, described.11 andTrppsubfamilies,withTrpp2beingthemostabundant transcript. VSMC Isolation Changes in the expression profile of ion channels Micewerekilledbydecapitationafterisofluoraneanesthesia.Fem- induced by the switch to a proliferative phenotype were oralarteriesweredissectedandcleanedofconnectiveandendothe- studiedinVSMCsobtainedfromexplantsofendothelium- lialtissues,andtheywerestoredforRNAextractionoruseddirectly denuded femoral arteries and kept in culture (in vitro toobtainfreshdispersedVSMCsorculturedVSMCsaspreviously model) and in endoluminal lesion-induced intimal hyper- described.8,12 plasia (in vivo model), in which expression changes were studied at 3 different times after lesion (48 hours, 1 week, RNA Expression Profile and 4 weeks). Each model had its own control: endo- TaqMan low-density arrays (Applied Biosystems) were used to thelium-free arteries (C ) for the in vitro model and simultaneously quantify the expression of 96 genes per sample.12 E(cid:4) Each sample derived from (cid:2)500 ng of mRNA from contractile or endothelium intact arteries (CE(cid:1)) for the in vivo model. proliferatingVSMCs. Differencesbetweenthese2controlsaresummarizedintheinset inFigure1,wheregeneswhoseexpressionwasmodifiedinthe Protein Expression and Function CE(cid:4)samples(relativetoCE(cid:1))areshown. Selective antibodies were used for protein detection with immuno- Changes in proliferation were analyzed by 2 methods. logic methods.8,11 The whole-cell and perforated-patch configura- First, we studied, for each individual gene, the differences Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 Cidad et al Ion Channels and VSMC Proliferation 1205 Figure1.Relativeabundanceoftheionchannel genesstudiedincompletefemoralarteries(C ) E(cid:1) expressedas2(cid:4)(cid:5)Ct,where(cid:5)Ct(cid:6)Ct (cid:4)Ct . channel 18s Eachbaristhemean(cid:7)SEMof10determinations obtainedin5duplicateassays.Genesaregrouped byfamilies(seeonlinesupplementaldataforthe listofchannelgenes).Theinsetshowsfold changesinexpressioninendothelium-freearteries (C )expressedaslog(2(cid:4)(cid:5)(cid:5)Ct),where E(cid:4) (cid:5)(cid:5)Ct(cid:6)(cid:5)Ct(C )(cid:4)(cid:5)Ct(C ).Inallfigures,*P(cid:8)0.05, E(cid:4) E(cid:1) **P(cid:8)0.01,and***P(cid:8)0.001. betweencontrolandproliferationinthe2modelsbyusing sificationsinthemodelofendoluminallesion(Figure2A). the 2(cid:4)(cid:5)(cid:5)Ct relative quantification method13 (Supplemental Groups of genes with similarity in the pattern of expres- Figure III). Second, a 2-way hierarchical clustering anal- sion (indicated by the correlation coefficient) are high- ysis of genes and different experimental conditions14 was lighted in Figure 2. Within downregulated genes, we performed and visualized with the Treeview software identified clusters of genes with late decrease expression (Figure 2). This analysis categorized genes according to (Figure 2, box A), early decrease followed by a partial theirresponsestoPMintoupregulatedanddownregulated recovery (box B), or time-independent decrease (box C). genes and also demonstrated time-course–related subclas- Similarly, PM-upregulated genes can show a late increase Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 1206 Arterioscler Thromb Vasc Biol June 2010 Figure2.A,Two-wayhierarchicalagglomerativeclusteringappliedto59genes(horizontally)andto5contractileVSMCssamples (C1toC5),2samplesat48hoursafterendoluminallesion,2samplesafter1week,and3samplesafter4weeks(vertically).The inputdatawerethe(cid:5)Ctvaluesforthegenes.Colorpatchrepresentstheexpressionlevelforeachgeneandsample.Expression levelswerenormalizedwithineachsample,withascalerangingfrombrightgreen(lowest)tobrightred(highest).Missingvalues areshownasgraypatches.Thelengthofthetreebranchesisproportionaltothecorrelationofthegeneexpressionpattern,and someoftheseclustersandcorrelationcoefficientsareindicated.B,SamerepresentationasinA,butcomparingthecontractile VSMCs(labeledhereastissuesT1toT3)withtheculturedVSMCs(C1toC3).C,Diagramshowinggenesexhibitingsignificant expressionchanges(upregulation,redfontordownregulation,greenfont)inendoluminallesionsamplesandinculturedVSMCs, whenanalyzedwiththeT-REXsetoftoolsoftheGEPASsuite(seeSupplementalMethods).Thelightgrayintersectioncontains thegenesshowingthesamesignificantchangesinbothmodels. (boxD),anearlyincreasewithpartialrecovery(boxE),or obtained a significant variation in the expression of 22 a sustained increase (box F). Following these criteria, the genes in the in vivo model and 29 genes in the in vitro hierarchical clustering of the genes whose expression was model (Figure 2C). The endoluminal lesion induced a modified in cultured VSMC (Figure 2B) also identified 2 significant increase in 3 genes (ClCa1, Kv1.3, and Kv(cid:1)2), groups of genes with decreased (boxes A through C) or andasignificantdecreasein19genes.InculturedVSMCs, increased (boxes D through F) expression with 7 genes were overexpressed and 22 exhibited significant proliferation. decrease. The overlapping region in Figure 2C shows the To discern whether these expression patterns were genes with significant changes in their expression profile simply arbitrary structures or rather reflect biologically commontobothproliferativemodels(18genes).Ofthem, significantassociations,weperformedastatisticalanalysis wefocusedonthosewithincreasedexpression(Kv1.3and of the data with the GEPAS suite (http://gepas.bioinfo. Kv(cid:1)2), as they are more amenable to represent therapeu- cipf.es/), with false discovery rate (FDR) correction.15We tical targets. We studied the presence, distribution, and Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 Cidad et al Ion Channels and VSMC Proliferation 1207 Figure3.A,FamilyofcurrenttracesobtainedinfreshlydissociatedandculturedVSMCswiththeindicatedpulseprotocol.Cur- rentdensityofthetotaloutwardK(cid:1)currentwascalculatedindepolarizingpulsesto(cid:1)40mV.Kvcurrentdensitywasobtained afterBK blockwith500nmol/Lpaxilline.Mean(cid:7)SEMof14to18cells.B,TheupperplotshowsthefractionofthetotalKvcur- Ca rentrepresentedbyKv1.xcurrents(correolide-sensitive)andKv1.3currents(margatoxin-orPAP1-sensitive)incontractile(tissue) versusculturedVSMCs.Theamplitudeofthedepolarizationinducedbytheseblockersinperforated-patchrecordingsisrepre- sentedinthelowerplot.Mean(cid:7)SEMof11to14cells.C,PharmacologicalcharacterizationofoutwardK(cid:1)currentsincontractile andculturedVSMCs.Shownarecurrent/voltagerelationshipsobtainedin2cellsincontrolconditionsandaftersequentialappli- cationofpaxilline(500nmol/L),margatoxin(10nmol/L),correolide(5(cid:4)mol/L),andTEA(20mmol/L).TheKv1.3componentisrep- resentedbytheshadedareas.Thetracesshowcurrentat(cid:1)80mVinbothcellswitheachblocker,andpiechartsrepresentthe proportionofKv1-Kv4currentsinbothpreparations. functional contribution of Kv1.3 in both proliferative larger in cultured VSMCs (from 44.12(cid:7)1.6% in tissue to models. 68.69(cid:7)6.5% in culture, Figure 3B). The increased func- tional contribution of Kv1.3 channels can account for this Functional Expression of Kv Channels in change, as revealed by selective Kv1.3 blockers (marga- Femoral VSMCs toxin and PAP-117,18). Current sensitive to 10 nmol/L Outward K(cid:1) currents were studied with the whole-cell margatoxinrepresented58.15(cid:7)4.54%oftheKvcurrentin patch clamp technique in freshly dissociated VSMCs and culturedVSMCsversus10.66(cid:7)2.03%infreshlydispersed in cultured VSMCs from femoral arteries. Figure 3A VSMCs. Similar results were obtained with 10 nmol/L shows representative traces obtained in each condition. PAP-1.Thefunctionalcontributionofthesechannelstoset Both total outward K(cid:1) current density, and the Kv com- restingV wasexploredwiththesameblockers,providing M ponent of this current (ie, the 500 nmol/L paxilline-resis- parallel results (Figure 3B, bottom). Figure 3C depicts tant current16) was significantly larger at all voltages in examplesofthepharmacologicaldissectionoftheoutward VSMCs in the contractile phenotype, but there were no K(cid:1) currents. In the presence of paxilline (500 nmol/L), significant differences in the proportion of the two com- application of margatoxin or PAP-1 (10 nmol/L each) ponents. However, the Kv1 fraction of the current (the allowed quantification of the Kv1.3 component. After 5 (cid:4)mol/L correolide-sensitive current) was significantly Kv1.3 blockade, correolide (5 (cid:4)mol/L) was used to selec- Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 1208 Arterioscler Thromb Vasc Biol June 2010 Figure5.A,Representativeexampleofthechangesinpeak currentamplitudeat(cid:1)40mVinaVSMCisolatedfroman injuredarterywiththeindicateddrugs.B,Labelingwithanti- SM22(green)demonstratesthevascularsmoothmusclelin- eageofthesecells.C,Absolutemagnitudeofthe margatoxin-sensitivecomponent(I )incontrol(contractile) Kv1.3 andinculturedandinjuredVSMCs.Mean(cid:7)SEMof11to15 cells. In addition to the increased proportion of Kv1 currents, cultured VSMCs also showed a significant decrease in the Kv2(from36.52(cid:7)2.97%to21.34(cid:7)3.71%)andKv3(from 16.4(cid:7)2.74% to 5.56(cid:7)0.97%) components. The Kv3 com- ponent was estimated as the fraction of the currents Figure4.A,ImmunocytochemicalidentificationofKv1.3and blockedby100(cid:4)mol/LTEAaftercorreolide,andtheKv2 Kv1.5proteininfreshlydispersedandculturedfemoral component was calculated as the difference between VSMCs.Anti-Kv1.3oranti-Kv1.5labeling(green)iscombined with4(cid:3),6-diamidino-2-phenylindolestaining(blue)B,Paraffin 20 mmol/L and 100 (cid:4)mol/L TEA-sensitive currents. sectionsofcontrolandinjuredfemoralarterieswerelabeled withtheindicatedantibodiesandcounterstainedwithhema- Kv1.3 Protein Expression in Femoral VSMCs toxylin.Arrowsindicateinternalelasticlamina(IEL).C,Sum- marydatarepresentthepercentageoflabeledareaoverthe Changes in Kv1.3 protein expression were explored by totalvesselarea.Mean(cid:7)SEMof10to25sectionsfrom6to immunocytochemistry in freshly dispersed and cultured 13animals. VSMCs and by immunohistochemical labeling in sections of control and injured femoral arteries. We have also tively block the remaining Kv1 currents. Subsequent studied Kv1.5 protein expression as one example of application of tetraethylammonium chloride (TEA) channeldownregulatedonPM.Kv1.3labelingincreasedin (20 mmol/L) identified the amplitude of the Kv2(cid:1)Kv3 cultured VSMCs when compared with freshly dispersed component. The minimal residual current, which was cells, whereas Kv1.5 labeling disappeared (Figure 4A). insensitive to correolide and TEA, could reflect Kv4 Figure 4B shows similar results in sections of femoral current component. The molecular composition of Kv arteries: Kv1.3 staining was weak in control sections, and currents in each preparation is illustrated in the pie charts. increased after lesion (mainly in the intimal layer), Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 Cidad et al Ion Channels and VSMC Proliferation 1209 Figure6.A,EffectofKv1.3blockersonfemoralVSMCmigrationcalculatedasthepercentageofinvadedareainscratchassays. Imagesshowarepresentativeexperimentwith10nmol/LPAP-1.B,Dose-responseeffectofPAP-1ontheproliferationrateof culturedfemoralVSMCsmeasuredwithaBrdUincorporationassay.n(cid:6)3to8experimentsineachgroup;SFindicatesserum- freemedium.C,TwodifferentKv1.3shortinterferingRNAs(siRNAs)wereabletoreducemRNAexpression(normalizedtoGapdh) andBrdUincorporationintransfectedVSMCscomparedwithmock-transfectedorsiRNAC–transfectedVSMCs(n(cid:6)4).D, EffectsofKchannelblockersorhigh[K(cid:1)] onrestingV (current-clampexperiments)andonproliferationinhibition(BrdUrdincor- e M poration).Mean(cid:7)SEMof8to20determinations.TheinsetshowstheantiproliferativeeffectofMgTx(10nmol/L)incontrol mediumandin20mmol/LK(cid:1)-containingmedium.n(cid:6)7experiments. whereaslabelingwithanti-Kv1.5antibodywasdetectedin liferation rate, determined by the number of BrdU(cid:1) cells, the muscular layer of almost half of the cells in control was significantly decreased by Kv1.3 channels blockade arteries, decreasing on injury. Functional characterization (Figure 6B through 6D). The effect of PAP-1 inhibiting of Kv1.3 currents from VSMCs obtained from injured proliferation was dose dependent (Figure 6B) and could arteries (Figure 5A) showed an augmented expression of be mimicked by selective knockdown of Kv1.3 currents the Kv1 component (85.2(cid:7)4.5% of the Kv current) due to with short interfering RNA (Figure 6C). PAP-1 and mar- an increase of the MgTx-sensitive component gatoxindepolarizedVSMCs,butothermaneuversthatalso (61.5(cid:7)5.2%), very similar to the case with cultured depolarized cultured VSMCs, such as increasing [K(cid:1)] or e VSMCs (see Figure 3B). Moreover, increased Kv1.3 blockade of Kv2 or BK channels, did not affect prolif- Ca current was not simply reflecting downregulation of other eration (Figure 6D). Moreover, proliferation inhibition by Kv channels, as Kv1.3 current density increased in both 10 nmol/L margatoxin was attenuated but still significant proliferation models (Figure 5C). in the presence of 20 mmol/L K(cid:1)e (Figure 6D, inset). Effects of Kv1.3 Blockade on VSMC Migration Discussion and Proliferation The upregulated functional expression of Kv1.3 could This work provides a comprehensive study of the expres- reflect a link between the channel and the establishment sion pattern of ion channels in VSMCs, using a quantita- and maintenance of the proliferative phenotype. To ex- tive high-throughput technique, with the goal of elucidat- plorethis,westudiedtheeffectofKv1.3currentblockade ing PM-associated changes. We defined the relative on the ability of cultured VSMCs to migrate and prolifer- expression of the ion channel genes studied and compared ate. Migration was determined in confluent femoral this profile among different vascular beds, observing a VSMCs by a scratch migration assay. After 24 hours in continuous pattern that correlates with the size of the serum-free medium alone (control) or with 10 nmol/L arteries (see Supplemental Figure IV), in agreement with PAP-1 or 10 nmol/L margatoxin, the invaded area was previousdatafocusedonindividualchannels.19–23Charac- significantly larger in control cells than in cells treated terization of the changes in expression pattern associated with the Kv1.3 blockers (Figure 6A). Similarly, the pro- with proliferation in in vivo and in vitro models of PM Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 1210 Arterioscler Thromb Vasc Biol June 2010 showed a good correlation between both situations. Eigh- larization, proliferation was not affected if depolarization teen of 22 genes that changed in the in vivo model were was induced with other channel blockers or with high conservedinculturedVSMCs.Thelargernumberofgenes [K(cid:1)] . e that changed in cultured VSMCs can be attributed to their However, Kv1.3 inhibition could restrain hyperpolariz- more homogeneous nature (a uniform population of syn- ing signals required for proliferation. This hypothesis is thetic VSMCs), whereas endoluminal lesion samples are consistentwithourresultsshowinganattenuatedresponse partially contaminated with endothelial cells, contractile to MgTx in high [K(cid:1)] . e VSMCs, and connective tissue. These contaminants could We have previously reported a similar association be- also explain some of the changes present only in this tween Kv3.4 channel upregulation and proliferation in model,asitismostlikelythecaseofSK3channels(small human uterine artery VSMCs,8 although in this prepara- conductance K channels), whose expression is restricted tion, high [K(cid:1)] depolarization mimics proliferation inhi- Ca e to endothelial cells.9,24 Although specific changes induced bition by Kv3.4 blockade.30 Interestingly, the Kv1.3 gene by the culture conditions cannot be excluded, our analysis is also upregulated in these cells,8 and Kv1.3 blockers indicates that ion channel expression profile in cultured inhibit proliferation (Supplemental Figure VIB and VIC), VSMCs reproduces reasonably well the changes in prolif- suggesting that the association between Kv1.3 upregula- erating lesions in vivo. This observation is important tion and proliferation may be present in different vascular beds.Infact,theionchannelexpressionprofileofcultured consideringthetechnicallimitationsoftheinvivomodels mesenteric VSMCs shows a remarkable similarity to that for functional studies. of femoral arteries (Supplemental Figure V), with Kv1.3 Only some of the channels highlighted here have been being the predominant Kv1 channel gene and Kv1.3 previously reported as relevant to VSMC proliferation. blockade inhibiting proliferation. Downregulation of several Kv1 genes (Kv1.5, Kv1.2, and Kv1.3 channels have been reported to associate with Kv2.1) has been reported in other preparations.8,25 Also, proliferation in T cells,17 endothelial cells,31 microglia, decreased expression of Cav1.2 mRNA is consistent with macrophages, oligodendrocyte progenitors, and carcinoma the reported reduction in the functional expression of L-typeCa2(cid:1)channelsinsyntheticVSMCs,5,26althoughwe cells,32,33 but this is the first description of their role in VSMCproliferationandmigration.Inthisscenario,Kv1.3 didnotdetecttheconcomitantupregulationofgenesofthe blockade could represent a new therapeutic approach to Trpc and Cav3.x families described in these studies. prevent unwanted remodeling. Similarly, decreases of the BK channel genes (BK(cid:2)and BK(cid:1)1 subunits) have been described,5,8,9,27 but we did not Acknowledgments find the associated increase in IK1 mRNA that was We thank Esperanza Alonso for excellent technical assistance and previously reported.5,9,10 Nevertheless, a contribution to theMerckCompanyforthegiftofcorreolide. PM in our preparation of these channels (Trpc, Cav3.x, or IK1) via posttranscriptional modulation cannot be ex- Sources of Funding cluded. Overall, vascular bed variations may account for This work was supported by Ministerio de Sanidad, Instituto de some of the recounted discrepancies, and it is tempting to Salud Carlos III grants R006/009 (Red Heracles), FS041139-0 (M.R.), and PI041044 (J.R.L.-L.); Ministerio de Educacio´n y speculate that not only the expression profile of ion CienciagrantsBFU2004-05551(M.T.P.-G.)andBFU2007-61524 channels but also the PM-induced changes could be (J.R.L.-L.); and Junta de Castilla y Leon grant GR242. Dr vascular bed specific. In this context, the finding of Moreno-Domínguez is a fellow of the Spanish Ministerio de conserved changes across different preparations is rele- Educacion y Ciencia. vant, as it may reflect obligatory associations of certain Disclosures channels with vascular remodeling, representing novel None. therapeutic opportunities. With this idea, we explored the functional expression References and the contribution to VSMC proliferation of the genes 1. OwensGK,KumarMS,WamhoffBR.Molecularregulationofvascular upregulated by the PM. We focused on Kv1.3 channels smoothmusclecelldifferentiationindevelopmentanddisease.Physiol because the tools for determining the contribution of the Rev.2004;84:767–801. modulatory subunit Kv(cid:1)2 are more limited. Furthermore, 2. ShanahanCM,WeissbergPL.Smoothmusclecellheterogeneity:patterns ofgeneexpressioninvascularsmoothmusclecellsinvitroandinvivo. the concomitant increased expression of Kv1.3 and Kv(cid:1)2 ArteriosclerThrombVascBiol.1998;18:333–338. may reflect the fact that they form heteromultimers, 3. WamhoffBR,BowlesDK,OwensGK.Excitation-transcriptioncoupling because Kv(cid:1)2 preferentially associates with Kv1.x chan- inarterialsmoothmuscle.CircRes.2006;98:868–878. 4. Neylon CB. Potassium channels and vascular proliferation. Vasc nels,28 and both association and upregulation of Kv1.3 Pharmacol.2002;38:35–41. and Kv(cid:1)2 have been described on mitogen-stimulated 5. Beech DJ. Ion channel switching and activation in smooth-muscle activation of T lymphocytes.17,29 Functional expression of cells of occlusive vascular diseases. Biochem Soc Trans. 2007;035: 890–894. Kv1.3 proteins increased in both proliferative models, and 6. GollaschM,HaaseH,RiedC,LindschauC,MoranoI,LuftFC,HallerH. selective blockade of Kv1.3 currents decreased cultured L-typecalciumchannelexpressiondependsonthedifferentiatedstateof VSMC migration and proliferation, demonstrating a link vascularsmoothmusclecells.FASEBJ.1998;12:593–601. 7. KumarB,DrejaK,ShahSS,CheongA,XuSZ,SukumarP,NaylorJ, between Kv1.3 channels functional expression and PM. ForteA,CipollaroM,McHughD,KingstonPA,HeagertyAM,Munsch Finally, although Kv1.3 inhibition leads to VSMC depo- CM, Bergdahl A, Hultgardh-Nilsson A, Gomez MF, Porter KE, Hell- Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012 Cidad et al Ion Channels and VSMC Proliferation 1211 strand P, Beech DJ. Upregulated TRPC1 channel in vascular injury in 19. Jackson WF. Ion Channels and Vascular Tone. Hypertension. 2000; vivo and its role in human neointimal hyperplasia. Circ Res. 2006;98: 35:173. 557–563. 20. MichelakisED,ReeveHL,HuangJM,TolarovaS,NelsonDP,WeirEK, 8. Miguel-VeladoE,Moreno-DominguezA,ColinasO,CidadP,HerasM, ArcherSL.Potassiumchanneldiversityinvascularsmoothmusclecells. Perez-Garcia MT, Lo´pez-Lo´pez JR. Contribution of Kv channels to CanJPhysiolPharmacol.1997;75:889–897. phenotypicremodelingofhumanuterinearterysmoothmusclecells.Circ 21. SmithPD,BrettSE,LuykenaarKD,SandowSL,MarrelliSP,Vigmond Res.2005;97:1280–1287. EJ, Welsh DG. KIR channels function as electrical amplifiers in rat 9. KohlerR,WulffH,EichlerI,KneifelM,NeumannD,KnorrA,GrgicI, vascularsmoothmuscle.JPhysiol.2008;586:1147–1160. Kampfe D, Si H, Wibawa J, Real R, Borner K, Brakemeier S, 22. TeramotoN.PhysiologicalrolesofATP-sensitiveK(cid:1)channelsinsmooth OrzechowskiHD,ReuschHP,PaulM,ChandyKG,HoyerJ.Blockadeof muscle.JPhysiol.2006;572:617–624. theintermediate-conductancecalcium-activatedpotassiumchannelasa 23. Fountain SJ, Cheong A, Flemming R, Mair L, Sivaprasadarao A, new therapeutic strategy for restenosis. Circulation. 2003;108: BeechDJ.Functionalup-regulationofKCNAgenefamilyexpression 1119–1125. in murine mesenteric resistance artery smooth muscle. J Physiol. 10. TharpDL,BowlesDK.Theintermediate-conductanceCa2(cid:1)-activated 2004;556:29–42. K(cid:1)channel(KCa3.1)invasculardisease.CardiovascHematolAgents 24. Jackson WF. Potassium channels in the peripheral microcirculation. MedChem.2009;7:1–11. Microcirculation.2005;12:113–127. 11. RoqueM,FallonJT,BadimonJJ,ZhangWX,TaubmanMB,ReisED. 25. MandegarM,FungYC,HuangW,RemillardCV,RubinLJ,YuanJX. Mouse model of femoral artery denudation injury associated with the Cellularandmolecularmechanismsofpulmonaryvascularremodeling: rapid accumulation of adhesion molecules on the luminal surface and role in the development of pulmonary hypertension. Microvasc Res. recruitment of neutrophils. Arterioscler Thromb Vasc Biol. 2000;20: 2004;68:75–103. 335–342. 26. HouseSJ,PotierM,BisaillonJ,SingerHA,TrebakM.Thenon-excitable 12. Moreno-Dominguez A, Cidad P, Miguel-Velado E, Lopez-Lopez JR, smooth muscle: calcium signaling and phenotypic switching during Perez-GarciaMT.DenovoexpressionofKv6.3contributestochangesin vasculardisease.PflugersArch.2008;456:769–785. vascular smooth muscle cell excitability in a hypertensive mice strain. 27. LongX,TharpDL,GeorgerMA,SlivanoOJ,LeeMY,WamhoffBR, JPhysiol.2009;587:625–640. BowlesDK,MianoJM.Thesmoothmusclecell-restrictedKCNMB1ion 13. LivakKJ,SchmittgenTD.Analysisofrelativegeneexpressiondatausing channelsubunitisadirecttranscriptionaltargetofserumresponsefactor real-time quantitative PCR and the 2(cid:4)(cid:5)(cid:5)Ct method. Methods. 2001;25: andmyocardin.JBiolChem.2009;284:33671–33682. 402–408. 28. RhodesKJ,StrassleBW,MonaghanMM,Bekele-ArcuriZ,MatosMF, 14. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and TrimmerJS.AssociationandcolocalizationoftheKvbeta1andKvbeta2 displayofgenome-wideexpressionpatterns.ProcNatlAcadSciUSA. beta-subunitswithKv1alpha-subunitsinmammalianbrainK(cid:1)channel 1998;95:14863–14868. complexes.JNeurosci.1997;17:8246–8258. 15. GusnantoA,CalzaS,PawitanY.Identificationofdifferentiallyexpressed 29. McCormackT,McCormackK,NadalMS,VieiraE,OzaitaA,RudyB. genesandfalsediscoveryrateinmicroarraystudies.CurrOpinLipidol. Theeffectsofshakerbeta-subunitsonthehumanlymphocyteK(cid:1)channel 2007;18:187–193. Kv1.3.JBiolChem.1999;274:20123–20126. 16. LiG,CheungDW.EffectsofpaxillineonK(cid:1)channelsinratmesenteric 30. Miguel-Velado E, Perez-Carretero FD, Colinas O, Cidad P, Heras M, arterialcells.EurJPharmacol.1999;372:103–107. Lopez-LopezJR,Perez-GarciaMT.Cell-cycledependentexpressionof 17. BeetonC,WulffH,StandiferNE,AzamP,MullenKM,PenningtonMW, Kv3.4channelsmodulatesproliferationofhumanuterinearterysmooth Kolski-AndreacoA,WeiE,GrinoA,CountsDR,WangPH,LeeHealey musclecells.CardiovascRes.2010cvq011. CJ, Andrews S, Sankaranarayanan A, Homerick D, Roeck WW, 31. Erdogan A, Schaefer CA, Schaefer M, Luedders DW, Stockhausen F, TehranzadehJ,StanhopeKL,ZiminP,HavelPJ,GriffeyS,KnausHG, AbdallahY,SchaeferC,MostAK,TillmannsH,PiperHM,Kuhlmann NepomGT,GutmanGA,CalabresiPA,ChandyKG.Kv1.3channelsare CR.MargatoxininhibitsVEGF-inducedhyperpolarization,proliferation atherapeutictargetforTcell-mediatedautoimmunediseases.ProcNatl and nitric oxide production of human endothelial cells. J Vasc Res. AcadSciUSA.2006;103:17414–17419. 2005;42:368–376. 18. SchmitzA,SankaranarayananA,AzamP,Schmidt-LassenK,Homerick 32. PardoLA.Voltage-gatedpotassiumchannelsincellproliferation.Phys- D, Hansel W, Wulff H. Design of PAP-1, a selective small molecule iology.2004;19:285–292. Kv1.3blocker,forthesuppressionofeffectormemoryTcellsinauto- 33. Wulff H, Castle NA, Pardo LA. Voltage-gated potassium channels as immunediseases.MolPharmacol.2005;68:1254–1270. therapeutictargets.NatRevDrugDiscov.2009;8:982–1001. Downloaded from http://atvb.ahajournals.org/ by guest on January 26, 2012