ExperimentalEyeResearch91(2010)462e471 ContentslistsavailableatScienceDirect Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer Epidermal growth factor receptor transactivation by the cannabinoid receptor (CB1) and transient receptor potential vanilloid 1 (TRPV1) induces differential responses in corneal epithelial cells H. Yanga, Z. Wanga, J.E. Capó-Apontea,b, F. Zhanga, Z. Pana, P.S. Reinacha,* aDepartmentofBiologicalSciences,StateUniversityofNewYork,StateCollegeofOptometry,NewYork,NY10036,USA bVisualSciencesBranch,U.S.ArmyAeromedicalResearchLaboratory,FortRucker,AL36362,USA a r t i c l e i n f o a b s t r a c t Articlehistory: Corneal epithelial injury induces release of endogenous metabolites that are cannabinoid receptor Received29January2010 1 (CB1) and transient receptor potential vanilloid 1 (TRPV1) agonists. We determined the functional Acceptedinrevisedform28June2010 contributionsbyCB1andTRPV1activationtoelicitingresponsesunderlyingwoundhealinginhuman Availableonline7July2010 cornealepithelialcells(HCEC).BoththeselectiveCB1andTRPV1agonists(i.e.,WIN55,212-2[WIN]and capsaicin [CAP], respectively) induced EGFR phosphorylationwhereas either inhibition of its tyrosine Keywords: kinaseactivitywithAG1478orfunctionalblockageeliminatedthisresponse.Furthermore,EGFRtrans- cannabinoidreceptor1(CB1) activation was abolished by inhibitors of proteolytic release of heparin bound EGF (HB-EGF). CB1- transientreceptorpotentialvanilloid1 inducedCa2þtransientswerereducedduringexposuretoeithertheCB1antagonist,AM251orAG1478. (TRPV1) epidermalgrowthfactorreceptor(EGFR) BothCAPandWINinducedtransientincreasesinErk1/2,p38,JNK1/2MAPKandAkt/PI-3Kphosphor- proinflammatorycytokine ylation status resulting in cell proliferation and migration increases which mirrored those elicitedby transactivation EGF.NeitherEGFnorWINinducedanyincreasesinIL-6andIL-8release.Ontheotherhand,CAP-induced cellproliferation 3- and 6-fold increases, which were fully attenuated during exposure to CPZ, but AG1478 only sup- cellmigration pressedthemby21%.ThemixedCB1andTRPV1antagonist,AM251,enhancedtheCAP-inducedrisein proteinphosphorylation IL-8releasetoahigherlevelthanthatelicitedbyCAPalone.Inconclusion,CB1andTRPV1activation inducesincreasesinHCECproliferationandmigrationthroughEGFRtransactivation leading toglobal MAPKandAkt/PI-3Kpathwaystimulation.Ontheotherhand,theTRPV1-mediatedincreasesinIL-6and IL-8releaseareelicitedthroughbothEGFRdependentandEGFR-independentsignalingpathways. (cid:1)2010ElsevierLtd.Allrightsreserved. 1. Introduction protects this layer against UV- and 4-hydroxynonenal-induced cellulardamage(Pappaetal.,2005). Corneal transparency maintenance depends on continuous As epithelial turnover is modulated by a host of cytokines, renewalofitsoutermostepitheliallayer.Thisprocessreplacesthe extensive effort is devoted to identifying cognate receptor-linked uppermostterminallydifferentiatedlayersthatarebeingsloughed cellsignalingdrugtargetsforhasteningthisprocesssubsequentto off into the tears. Their replacement assures preservation of cornealinjury(ReinachandPokorny,2008).Thesestudiesemploy epithelial integrityand its smoothoptical properties. In addition, injury-inducedinvitroepithelialandinvivowoundhealingmodels epithelial renewal preserves other needed functions for visual to determine which receptor-linked cell signaling pathways clarity that include: 1) tight junction barrier intactness, which mediate control of cell proliferation, migration, differentiation, protects the cornea from becoming translucent due to tissue inflammationandapoptosis.Inaddition toimprovingourunder- swelling caused by exposure to environmental stresses such as standingofcornealbiology,thesestudieshaveidentifiedpotential pathogensandanisosmoticchallenges (Luetal., 2001);2) innate noveldrugtargetstolessencornealscarringandinflammationthat immuneresponsiveness,whichdetectsthepresenceofpathogens canpersistsubsequenttowoundclosure.Thesecomplicationsmay and provides signals that activate the corneal defense system be severe enough to prevent restoration of corneal transparency (Zhangetal.,2005);3)aldehydedehydrogenaseexpression,which andvisualfunction(Saikaetal.,2007).Inhumancornealepithelial cells(HCEC),epidermalgrowthfactorreceptors(EGFRs)contribute to mediating corneal epithelial renewal. EGFR activation by * Correspondingauthor. EGF results in stimulation of cell proliferation and migration E-mailaddress:[email protected](P.S.Reinach). through activation of: a) the three member pathways of the 0014-4835/$eseefrontmatter(cid:1)2010ElsevierLtd.Allrightsreserved. doi:10.1016/j.exer.2010.06.022 Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 3. DATES COVERED 28 JUN 2010 2. REPORT TYPE 00-00-2010 to 00-00-2010 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER Background: Spatial epidemiology is useful but difficult to apply in developing countries due to the low (CB1) and transient receptor 5b. GRANT NUMBER potential vanilloid 1 (TRPV1) induces differential responses in corneal epithelial cells 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION U.S. Army Aeromedical Research Laboratory,Visual Sciences REPORT NUMBER Branch,Fort Rucker,AL,36362 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES 14. ABSTRACT Corneal epithelial injury induces release of endogenous metabolites that are cannabinoid receptor 1 (CB1) and transient receptor potential vanilloid 1 (TRPV1) agonists. We determined the functional contributions by CB1 and TRPV1 activation to eliciting responses underlying wound healing in human corneal epithelial cells (HCEC). Both the selective CB1 and TRPV1 agonists (i.e., WIN55,212-2 [WIN] and capsaicin [CAP], respectively) induced EGFR phosphorylation whereas either inhibition of its tyrosine kinase activity with AG1478 or functional blockage eliminated this response. Furthermore, EGFR transactivation was abolished by inhibitors of proteolytic release of heparin bound EGF (HB-EGF). CB1- induced Ca2? transients were reduced during exposure to either the CB1 antagonist, AM251 or AG1478. Both CAP and WIN induced transient increases in Erk1/2, p38, JNK1/2 MAPK and Akt/PI-3K phosphorylation status resulting in cell proliferation and migration increases which mirrored those elicited by EGF. Neither EGF nor WIN induced any increases in IL-6 and IL-8 release. On the other hand, CAP-induced 3- and 6-fold increases, which were fully attenuated during exposure to CPZ, but AG1478 only suppressed them by 21%. The mixed CB1 and TRPV1 antagonist, AM251, enhanced the CAP-induced rise in IL-8 release to a higher level than that elicited by CAP alone. In conclusion, CB1 and TRPV1 activation induces increases in HCEC proliferation and migration through EGFR transactivation leading to global MAPK and Akt/PI-3K pathway stimulation. On the other hand, the TRPV1-mediated increases in IL-6 and IL-8 release are elicited through both EGFR dependent and EGFR-independent signaling pathways. 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 10 unclassified unclassified unclassified Report (SAR) Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 H.Yangetal./ExperimentalEyeResearch91(2010)462e471 463 mitogen-activated protein kinase (MAPK) signaling cassette (i.e., stimulation of IL-6 and IL-8 release. In contrast, CB1 activation ERK,p38andJNK);b)thephosphoinositide3-kinase(PI3-K)/Akt/ counters TRPV1-induced increases in IL-8. It is conceivable in GSK-3 pathway; c) adenylate cyclase and phospholipase C (PLC)- aclinicalsettingthatdrugstargetedtoactivateCB1receptorsmay inducedCa2þsignalingaswellasphospholipaseD(PLD)-mediated be effective in reducing TRPV1-induced inflammation caused by phosphatidicacidformation(IslamandAkhtar,2000;Kangetal., cornealinjury. 2000, 2001; Mazie et al., 2006; Wang et al., 2006, 2009; Yang et al., 2005, 2001; Yin and Yu, 2009; Zhang and Akhtar, 1998). 2. Methods EGFR activation also can occur through transactivation by other receptorsandmediators(BlockandKlarlund,2008;Lyuetal.,2006; 2.1. Materials Spix et al., 2007; Xu et al., 2006, 2007). In this process, agonists other than EGF activate their cognate receptors, which leads to The following chemicals were purchased from SigmaeAldrich matrix metalloproteinase activation and scission of EGF from (St. Louis, MO): WIN55,212-2 (WIN), anandamide (AEA), cap- membrane bound heparin. Therefore, the EGFR-linked cell sazepine (CPZ), CAP, AM251, AG1478, GM6001, CRM197, EGF, signalingpathwaysserveasaconduitforelicitingtissueresponses bovine insulin, gentamicin and TrypLE(cid:3) Express Stable Trypsin- toanumberofdifferentmediatorsbesidesEGF. Like Enzyme. Dulbecco’s modified Eagle’s medium (DMEM)/F12 Members of the transient receptor potential (TRP) protein mediumfetalbovineserum(FBS)andfura-2acetoxymethylester superfamilyarepolymodalinthattheyareactivatedbynumerous were purchased from Invitrogen (Carlsbad, CA). Anti-CB1, anti- differentstimuli.Inthecornealepithelium,somemembersofthe EGFR, phospho-EGFR, anti-Erk1/2, phospho-Erk1/2, phospho-Akt, vanilloid (V) TRP subfamily were identified. In HCEC, there is phospho-GSK,phospho-p38;goatanti-mouseIgG-HRP,goatanti- functionalexpressionofTRPV1,3and4(Panetal.,2008;Yamada rabbit IgG-HRPantibody, anti (H196) actin, anti-Erk1/2, anti-p38, et al., 2010; Zhang et al., 2007). TRPV1 is a nonselective ion and b-actin antibodies were purchased from Santa Cruz Biotech- channelwhichisactivatedbyinjury-inducedendogenousmedia- nology (Santa Cruz, CA). Anti-EGFR neutralizing clone, LA1, was tors such as endocannabinoids, endovanilloids, declines in pH, purchasedfromMilliporeCorporate(Billerica,MA). elevatedtemperatureandhypertonicityaswellascapsaicin,which is present in red pepper extracts. Capsaicin (CAP) is a selective 2.2. Cellculture TRPV1 agonist and in HCEC induces increases in the release of proinflammatorycytokinemediators,suchasinterleukin(IL)-6and SV40-adenovirus immortalized HCEC were obtained as thechemoattractant,IL-8.MAPKactivationisacontributortotheir agenerousgiftfromDr.KaoruAraki-Sasaki.Thecellswerecultured increases (Zhang et al., 2007). These rises induced by CAP have at37(cid:3)Cinanincubatorwith5%CO2and95%ambientairinDMEM/ physiologicalrelevancesinceTRPV1activationbyinjuryinamouse F12medium,supplementedwith6%FBS,5ng/mlEGFand5mg/ml corneal wound healing model contributes tothe development of insulin.Cellcyclearrestwasachievedbyculturingcellsinserum- severe inflammation that persists subsequent to wound closure. free and EGF-free DMEM/F12 medium for 24 h before Evidence of its role stems from our finding that in homozygous experimentation. TRPV1((cid:2)/(cid:2))knockoutmicethewoundhealingresponsetoinjuryis morefavorable. Thisis apparent sinceinflammationand scarring 2.3. Singlecellfluorescenceimaging arelesssevereatthetimeofwoundclosure(Okadaetal.,2008). EventhoughEGFR-linkedpathwaysareactivatedbyCAP,itisnot Cellsgrownon40-mmcircularcoverslips(BioptechsInc,Butler, known ifEGFR transactivation contributes tothedevelopmentof PA) were loaded with 2 mM fura-2 AM at room temperature for inflammationandscarring. 30minandthenwashedwithNaClRinger’ssolutioncontaining(in Thecannabinoidreceptorsubtype1(CB1)modulates,through mM): NaCl (141), KCl (4.2), CaCl2 (0.8), KH2PO4 (2), MgCl2 (1), theGTPbindingprotein(Gi),anumberofimportantphysiological glucose(5.5),andHEPES(10)withosmolarity300mOsmandpH processes in different tissues including neurotransmitter release, 7.4.Cellswerecontinuouslysuperfusedat34(cid:3)CinaFochtClosed painandanalgesia,energyhomeostasisregulation,andcontrolof System 2 (FCS2) perfusion chamber with temperature control immunecell function (Graham et al., 2009;Howlett, 2005; Kress (BioptechsInc,Butler,PA)andplaced onthestage ofaninverted andKuner,2009;Pertwee,2006;Stephens,2009).CB1activation microscope(NikonDiaphot200).Cellswerethenalternatelyillu- by cannabinoids has immunosuppressive effects, which have minatedat340and380nm,andemissionwasmonitoredevery5s beneficial effects in the treatment of autoimmune disorders. at510nmusingaRoperScientificCCDcamera.Eachfieldofinterest These results suggest that the cannabinoid system has various contained15e20cells.ChangesinintracellularCa2þlevels,[Ca2þ]i, roles in disease pathologies and provides potential therapeutic were analyzed using RatioTool software (Isee Imaging, Durham, targets. A functional role for CB1 in the human corneal epithe- NC).Thenvaluesprovidedindicatethenumberofexperimentsper liumhasnotyetbeendescribedeventhoughCB1expressionwas datapoint. detected in the corneas of isolated human eyes (Straiker et al., 1999). In some other tissues, TRPV1 and CB1 are coexpressed 2.4. Westernblotanalysis andfunctionallyinteractwithoneanother.Suchisthecaseinthe colonic epithelium, in neuronal enriched mesencephalic cultures, HCEC were gently washed twice in cold phosphate-buffered primary sensory neurons and myometrial smooth muscle cells saline (PBS) and harvested in 0.5 ml cell lysis buffer. Cell lysates (Brighton et al., 2009; Kim et al., 2008; Mahmud et al., 2009; were centrifuged and supernatants were collected for measuring Sibaev et al., 2006). The coexpression of TRPV1 and CB1 in the proteins with a bichinchoninic acid assay (BCA) protein assay kit corneal epithelium prompted us to probe for a functional inter- (PierceBiotechnology,Chicago,IL).Twentyto50mgofdenatured action between them in HCEC. protein was electrophoresed on 10% polyacrylamide sodium WeshowinHCECthatthereisafunctionalinteractionbetween dodecylsulphate (SDS) minigels and polyvinylidene difluoride TRPV1andCB1.Togethertheymediateincreasesincellprolifera- (PVDF)membraneswereblockedwithnonfatdrymilk.Theblots tion and migration through EGFR transactivation and MAPK/Akt- were exposed to the appropriate primary antibody overnight at linked signaling. On the other hand, other EGFR independent 4(cid:3)Candthenexposedtoanappropriatesecondaryantibody(e.g., TRPV1-linked pathway(s) contribute to mediating TRPV1 anti-rabbit, anti-goat or anti-mouse) HRP labeled IgG for 1 h at 464 H.Yangetal./ExperimentalEyeResearch91(2010)462e471 roomtemperature.Theimmunoreactivebandsweredetectedwith (QuantiGlo Human IL-8 or IL-6 Chemiluminescent Immunoassay; anAmershamECLPluskitandbanddensitywasquantifiedusing R&DSystems,Minneapolis,MN).Thecellswerewashedwithbasic SigmaScanPro5.0software.Themonoclonalantib-actinantibody medium and then exposed to CPZ, or AM251, for 30 min before testedforproteinloadingequivalence. exposing them for 24 h to either CAP or EGF. Supernatants were harvestedoniceandcentrifugedat2000rpmfor5minat4(cid:3)Cto 2.5. Immunocytochemistry removecelldebris.Thesupernatantswerestoredat(cid:2)80(cid:3)Cuntil analysis.Theiramountsintheculturemediumwerenormalizedto CB1 immunocytochemical localization was determined as the total amountof cellular protein lysed with 2% SDS and 0.2 N described(Yangetal.,2005).Briefly,cellswereseededontoaLab- NaOH and measured using a Micro BCA protein assay (Pierce, Tekchamberslidesystem(Nunc,Naperville,IL)andafterreaching Rockford,IL).Resultsareexpressed asmeanpicograms ofIL-6or confluence,theywerewashedtwicewithHEPES-bufferedRinger’s IL-8permgcellprotein(cid:5)SEM(n¼3). solution,fixedonicefor30minin4%paraformaldehyde,washed three times with HEPES Ringer’s solution, and then rendered 3. Results permeable using 0.1% Triton solution. After blotting with 10% normal goat serum, cells were exposed toanti-CB1 antibody (sc- 3.1. CB1expression 20754, 1:100; Santa Cruz Biotechnology) plus 1% bovine serum albumin (BSA) overnight at 4 (cid:3)C. After three washes with HEPES To ascertainwhether there is CB1 expression in HCEC, immu- Ringer’ssolution,cellswereincubatedwithgoatanti-rabbitIgGTR nocytochemistryand Western blot analysis were performed. CB1 (sc-3842,1:800; Santa Cruz Biotechnology) for 30 min and 1 mM expression is evident in Fig.1A based on bright diffuse staining SYTO16greenfluorescentnucleicacidstain(Invitrogen)for5minat around the cell periphery. Fig. 1B validates primary antibody room temperature. Fluorescence was visualized using a Nikon selectivitysincefollowingitspreadsorptiontoanepitopeblocking fluorescencemicroscopewitha60(cid:4)oilobjectivelens.Imageswere peptidethereisnostaining.Westernblotanalysiswasperformed (cid:4) processedusingAdobe Photoshop5.5software(AdobeSystems, onasubcellularfractiontovalidateCB1proteinexpression.Fig.1C Inc.,SanDiego,CA). showsabandwithanapparentmolecularweightof63kDa,which closelycorrespondstothereportedapparentmolecularweightof 2.6. Scratchwoundassay CB1inrat,monkeyandhumanbladder(Gratzkeetal.,2009). Cells weregrowntoconfluencein 35-mmculture platewells. 3.2. CB1functionalexpression They were then washed twice with PBS and placed in the appropriate serum-free medium. A small wound spanning the ToprobeforCB1functionalexpression,wedeterminedifWIN, diameter of the culture was made with a sterile cell scraper and aCB1agonist,inducedarisein[Ca2þ]i.Fig.2Aindicatesthat10mM was marked. The monolayer was washed twice with basic WINcausedatransientrisein[Ca2þ]iafterabout10minthatwas medium to remove suspended cells and re-fed with medium in approximately4-foldabovethecontrol.Thistransientthenslightly the presence or absence of EGF (10 ng/ml) immediately after declined and stabilized after another 10 min at a level that was wounding. Hydroxyurea (2.5 mM) was also added to the nearly 3-fold above the baseline level. On the other hand, pre- medium to reduce proliferation during the experiment. This incubation with 10 mM AM251, a CB1 antagonist, blocked the inhibitor reduced cell proliferation by about 95% with minimal subsequent rise in [Ca2þ]i by 93%. In a Ca2þ-free NaCl Ringers effects on cell viability (data not shown). Time-dependent supplementedwith2mMEGTA,theresponseswereessentiallythe wound closure was recorded for 24 h after wound creation. sameasthoseintheCa2þ-containingcounterpart(datanotshown). Images were obtained using a Roper Scientific CCD camera Fig. 2B shows that AEA (10 mM), an endogenous cannabinoid attachedtoNikonDiaphotinverted-stagemicroscope(NikonInc., analogue which is a mixed CB1 and TRPV1 agonist, elicited after MortonGrove,IL).Theremainingdenudedareaofeachfieldwas 10 min a 2-fold [Ca2þ]i transient, which then gradually fell after measured using SigmaScan Pro 5 software. Statistical analyses another10mintoreachalevelthatremainedabout65%abovethe were performed using unpaired Student’s t-test. A P value less control level. A component of this rise is attributable to TRPV1 than 0.05 was assumed to be significant. Data are shown as activationsincepreincubationwithaTRPV1antagonist,10mMCPZ mean (cid:5) SEM. suppressedthetransientbyabout65%. 2.7. Cellproliferation 3.3. CB1andTRPV1stimulationmediateEGFRtransactivation [3H] Thymidine incorporation was performed as described EGFR transactivation by ligands activating growth factor (Wang et al., 2009). Following 20 h of serum starvation in receptors and GTP binding coupled receptors induces a host of medium supplemented with 0.5% BSA, the cells were incubated different responses that are dependent on the duration and at 37 (cid:3)C for 1 h with 1 mCi/ml [3H] thymidine (3.3e4.8 TBq/ magnitudeofactivationofEGFR-linkedsignalingpathways.EGFR mmol).They were thenwashed twicewith cold PBS,threetimes transactivationbyCB1activationwasevaluatedbydeterminingif with ice-cold 5% trichloroacetic acid (TCA) and twice with cold preincubation with the specific EGFR inhibitor, AG1478 (5 mM) 90%alcohol.Celllysiswasobtainedwith0.2NNaOH/2%SDS.The suppressed10mMWIN-inducedrisesin[Ca2þ]i(LevitzkiandGazit, radioactivity was monitored in a Tri-Carb 2900 TR Liquid Scin- 1995; Osherov and Levitzki,1994). The results shown in Fig. 3A tillation Analyzer (PerkineElmer, Boston, MA) and the data were indicatethatWINinducedatransientriseintheF340/F380nmratio normalized to cellular protein content determined with a BCA after6minthatreachedmorethan3-foldabovethebaselinefol- Protein Assay Kit. lowedbystabilizationatalevelthatremainedabout2.5-foldabove itsbaselinelevelpriortoWINstimulation.Ontheotherhand,inthe 2.8. Enzyme-linkedimmunosorbentassay(ELISA) continued presence of AG1478, WIN only initially increased the F340/F380nmratiotoalevelthatwasabout25%abovethebaseline. IL-6 and IL-8 levels in the supernatants were measured Subsequently, this ratio decreased to a level that was about 10% according to the manufacturer’s instructions using ELISA kits abovethecontrollevel.Thisdiminutionsuggestsalargeportionof H.Yangetal./ExperimentalEyeResearch91(2010)462e471 465 Fig.1. CB1proteinexpressioninHCEC.(A)FollowingpermeabilizationwithTritonX-100,CB1expressionwasdetectedusingaprimaryantibody(i.e.,anti-rabbitpolyclonalIgG) followedbyincubationwithaconjugatedsecondaryantibody(i.e.,goatanti-rabbitIgG-TR).Calibrationbaris50mm.(B)Selectivityoftheprimaryantibodywasdocumented throughitsomission.(C)WesternblotshowsCB1proteinexpressionwithanapparentmolecularmassof63kDa. CB1-inducedincreasesinplasmamembraneCa2þinfluxisattrib- 10mMCAPincreasedtheF340/F380nmratiomaximallybyabout70% utabletoEGFRtransactivationbyCB1.TransactivationofEGFRby whereas preincubationwith AG1478 suppressed this rise by42%. directTRPV1stimulationwassimilarlyassessed.Fig.3Bshowsthat These partial declines induced during AG1478 exposure suggest thatacomponentoftheoverallincreaseinCa2þinfluxelicitedby WIN or CAP are accounted for EGFR transactivation by CB1 and TRPV1agonists. 3.4. CB1andTRPV1induceEGFRphosphorylation Westernblotanalysisofchangesinthephosphorylationstatus ofEGFRwasusedtoassessforitstransactivationbyCB1andTRPV1. This was done by determining if WIN and CAP-induced phos- phorylationofEGFRatTyr1078.Fig.4Aindicatesthatexposurefor 5 min to either 10 mM WIN or 10 mM CAP increased EGFR phos- phorylation statusby3-foldwhereas EGF(10 ng/ml)causedit to rise 5-fold. The effectof the CB1 agonist was fullysuppressed by preincubating cells with 10 mM AM251. Similarly, during contin- uousexposureto10mMCPZtheCAP-inducedrisesinEGFRphos- phorylation status declined by about 90%. Pre-exposure to 5 mM AG1478alsocompletelyblockedincreasesintheEGFRphosphor- ylationstatusinducedbyeither10ng/mlEGF,10mMCAPor10mM WIN.TheseresultsprovidefurthersubstantiationthateitherCB1or TRPV1activationinducesEGFRtransactivation. To determine whether CAP and/or WIN induces EGFR trans- activationthroughHB-EGFshedding,cellswerepretreated30min with either GM6001 (50 mM), a matrix metalloproteinase (MMP) inhibitor, CRM197 (10 mg/ml), a heparin binding EGF-like growth factor inhibitor, or for 2 h with a EGFR ligand-binding domain antibody LA1 (10 mg/ml) (Block et al., 2010). Fig. 4B shows that GM6001 and CRM197 eliminated CAP and WIN stimulated EGFR phosphorylation. Furthermore, the anti-EGFR neutralizing clone LA1abolishedEGFRphosphorylation(n¼3). 3.5. CB1andTRPV1induceEGFR-linkedsignalingactivation Fig. 2. CB1 functional expression in HCEC. (A) HCEC were loaded with fura2-AM w(2itmhMA)M.W25IN15(510,21m2M-2)b(WlocINke,d10CBm1Ma)citnivdautcioedn[bCya92þ5]%i.tr(Ban)sAiennatndwahmerideeas(ApEreAi)n,ciunbdautcioend AsCB1andTRPV1stimulationinducedEGFRtransactivation,we a[Ca2þ]itransient.Followingexposureto10mMcapsazepine(CPZ)thisresponsewas determined if this response induced phosphorylation of some of suppressed.Thedatarepresentthemeans(cid:5)SEM(n¼3,P<0.05). the EGFR-linked signaling pathways. Fig. 5A shows that EGFR 466 H.Yangetal./ExperimentalEyeResearch91(2010)462e471 Fig.3. CB1andTRPV1mediateEGFRtransactivation.(A)WIN55,212-2(WIN;10mM)induced[Ca2þ]itransientsinthepresenceandabsenceofAG1478(5mM).(B)Similarly, capsaicin(CAP;10mM)induced[Ca2þ]itransientsinthepresenceandabsenceofAG1478(5mM).Thedatarepresentthemeans(cid:5)SEM(n¼3,P<0.05). stimulationwitheither10ng/mlEGF,10mMCAPor10mMWINat migration induced by CAP and WIN. On the other hand, the 10minincreasedthephosphorylationstatusofAkt,p38andErk1/2 somewhatsmallerincreasesinmigrationinducedbyWINandCAP with similar magnitudes. This time point was chosen since we may be due to cytotoxicity based on some rounding up of cell previously found that these signaling mediators were maximally morphology after 24 h exposure to these two agonists. A recent stimulated at 10 min (Wang et al., 2009). In all cases, these reportshowsasimilarresultinwhichthemixedCB1andTRPV1 increaseswerediminishedifthecellswereinsteadexposedatthe agonist, 1 mM AEA, reduced smooth muscle cell viability and same time to either AG1478, CPZ or AM251. The patterns of the inhibitedcellmigration(Brightonetal.,2009). time-dependentchangesinthephosphorylationstatusofErk1/2, p38 and JNK1/2 during exposure to either EGF or CAP shown in 3.8. IndependenteffectsofTRPV1andEGFRactivationonIL-6and Fig.5BandCareessentiallythesame.Ontheotherhand,EGFRis IL-8release notthesoleroutefortheiractivationsincepre-exposuretoAG1478 only partially attenuated the CAP-induced increases in MAPK We have preliminary evidence that TRPV1 activation is an phosphorylation. important player in mediating inflammation to injury in vivo. In a mouse alkali burn cornea wound healing model, loss of TRPV1 3.6. CB1andTRPV1stimulatemitogenesisthroughEGFR genefunctionmarkedlyimprovedcornealwoundhealingoutcome transactivation duetomorecompleterestorationofcornealtransparency(Okada et al., 2008). This difference between the wound healing SinceselectiveCB1andTRPV1activationinducedEGFR-linked outcomes inwildtype and TRPV1 knockout mice prompted us to pathway stimulation, we assessed if such changes are linked to comparetheeffectsofselectiveTRPV1andEGFRactivationonIL-6 increasesincellproliferation.Fig.6comparestherespectiveeffects and IL-8 release. This was done to determine if there is any ofselectivestimulationofeitherCB1orTRPV1with10mMWINor disconnect between the effects of EGF and CAP on this response 10mMCAPon[3H]thymidineincorporationwiththatofEGF(10ng/ eventhoughbothEGFandCAPstimulatetheJNKpathway,whichin ml).Following24hserumstarvationandexposuretoeachofthese manyothertissuesislinkedtoinducingincreasesinthereleaseof agonistsforanother20h,bothEGFandCAPincreasedproliferation these proinflammatory cytokines and pain (Ji et al., 2009). The bynearly2-fold.TheriseinducedbyWINwasabout1.4-fold.The ELISAresultsshowninFig.8indicatethatexposureto10mMCAP selectivityoftheseagonist-inducedresponseswasdocumentedby for24hinducedabout3and6-foldincreasesinIL-6andIL-8levels showing that preincubation for 30 min to either 5 mM AG1478, whereas 10 ng/ml EGF did notstimulate their release. The selec- 10mMAM251or10mMCPZfullysuppressedeachoftheireffectson tivityofCAPtoinducetheseincreasesthroughTRPV1activationis proliferation. Therefore, CB1 and TRPV1 activation mediates indicated bythe full blockageby10 mM CPZ of these effects. The increasesincellproliferationthroughEGFRtransactivation. dependencewasdeterminedofincreasesinIL-6andIL-8releaseby CAP-induced EGFR transactivation. Pretreatment with AG1478 3.7. CB1andTRPV1stimulatecellmigrationthroughEGFR (1mM)partiallyblockedCAP-inducedrisesinIL-6andIL-8secre- transactivation tionby21%inbothcases.ThispartialdeclinesuggeststhatEGFR- linked signaling does not fully account for all of the pathways ThescratchwoundassaydeterminedifCB1andTRPV1activa- mediatingCAP-inducedincreasesinIL-6andIL-8release. tion stimulates cell migration through EGFR transactivation. The inset to Fig. 7 (upper panel) shows micrographs of the extent of 3.9. CB1activationdampensTRPV1-inducedincreasesinIL-8 closureat6and24hobtainedundercontrolconditionscompared release to those with either 10 ng/ml EGF,1 mM CAP or 1 mM WIN. The extentofwoundclosureiscomparedtothecontrolduringexpo- Invivo,costimulationbyAEAofCB1andTRPV1ininflammatory surefor24htoeitherEGF,CAPorWIN.LowerpanelshowsCAPand conditions elicited decreases in the electrical activity stemming WINstimulatedwoundclosureby1.65and1.52-fold,respectively, from TRPV1-induced nociceptive activity and subsequent pain- relativetotheuntreatedcontrol.Theseincreasesarecomparableto relatedbehavior(Mahmudetal.,2009).WeprobedwithELISAfor that of EGF, which increased this response by 2-fold. These an interaction between CB1 and TRPV1 by determining if CB1 responses are dependent on EGFR transactivation since 5 mM activation altered CAP-induced increases in IL-8 release. Fig. 9 AG1478 fully blocked them. The selectivity of AM251 and CPZ is showsthatwithoutmodulatingCB1activityCAPbyitselfinduced indicatedbytheirlackofinhibitionof therespectiveincreasesin after24habouta3-foldincreaseinIL-8release.Ontheotherhand, H.Yangetal./ExperimentalEyeResearch91(2010)462e471 467 Fig.5. CB1andTRPV1induceEGFR-linkedsignalingactivation.(A)ChangesinAkt,p38 andErk1/2phosphorylationstatusinducedbyeitherEGF,WIN55,212-2(WIN;10mM) orcapsaicin(CAP;10mM).HCECwereincubatedwitheither5mMAG1478,10mM AM251or10mMcapsazepine(CPZ)for30min;thenexposedto10ng/mlEGF,10mM Fig.4. CB1andTRPV1induceEGFRphosphorylation.(A)Growthfactorstarvedcells CAPor10mMWIN.Insomecases,thecellswereexposedtoEGF,CAPorWINalone. werepreincubated30minwithcapsazepine(CPZ;10mM),AM251(10mM)orAG1478 Followingtheincubation,thecellswereexposedtoeitherananti-p-Akt,p-p38orp- (5mM)priortobeingexposedtoeitherEGF(10ng/ml),WIN55,212-2(WIN;10mM)or Erk1/2 antibody and phosphorylation status was detected based on Western blot capsaicin (CAP; 10 mM). Western blotwas used toevaluate EGFR phosphorylation analysis.(B)Time-dependentchangesinthephosphorylationstatusofp-p38,p-JNK/ status.(B)HCECwereexposedtoWIN55,212-2(WIN;10mM)orcapsaicin(CAP;10mM) SAPKandp-Erk1/2inducedbyexposureto10ng/mlEGF.(C)Time-dependentchanges withorwithoutGM6001(50mM)orCRM197(10mg/ml)preincubationfor30min. inthephosphorylationstatusofp-p38,p-JNK/SAPKandp-Erk1/2inducedbyexposure Anti-EGFR,neutralizingcloneLA1(10mg/ml)preincubationoccurredfor2h.Cells to 10 mM CAP. Equal loading of proteins in each lane was always confirmed by werelysedattheendofstimulationandequalamountsofcelllysatewereblottedby reprobing the same blot with an anti-b-actin antibody. The data represent the anantibodyagainstphosphorylatedEGFR.Thedatarepresentthemeans(cid:5)SEM(n¼3, means(cid:5)SEM(n¼3,P<0.05). *P<0.05). 10 mM AEA fully suppressed the CAP-induced increase in IL-8 transactivation.However,thepredominantcontributortoeliciting risesinIL-6andIL-8levelsisattributabletoactivationofanother release.Interestingly,preincubatingwithaselectiveCB1antagonist AM251(5mM),insteadenhancedthe10mMCAP-inducedrise.IL-8 EGFR-independent pathway. TRPV1 control of the first two responsesisdependentonEGFR-mediatedMAPKaswellasPI3-K/ release rose about 8-fold above the control level. Therefore, Aktsignaling.CB1receptoractivationalsoinducedincreasesincell blockingtheCB1 componentof theoverallresponse toAEAwith proliferation and migration through the same process as that AM251 enhanced the CAP-induced increase in IL-8 release. This inducedbyTRPV1.However,CB1activationinsteadsuppressedthe result suggests that CB1 activation counters the increases in IL-8 increaseinIL-8releaseinducedthroughTRPV1activationbyCAP. releaseinducedbyCAP. SuppressionbyCB1activationofaTRPV1-mediatedresponsewas also described in myometrial smooth muscle cells, rat primary 4. Discussion sensory neurons and in mesencephalic dopamine neurons (Brightonetal.,2009;Kimetal.,2008;Mahmudetal.,2009).This Our results show that TRPV1 activation elicited increases in study broadens our understanding of the roles played by TRPV1 HCECproliferation,migration,IL-6andIL-8releasethroughEGFR and CB1 in eliciting through EGFR transactivation and linked 468 H.Yangetal./ExperimentalEyeResearch91(2010)462e471 systemlinkedtoEGFR.Suchcontrolisindicatedbythefindingthat eitherCB1orTRPV1activationinducedchangesinthedurationand magnitude of increases in the phosphorylation status of the signalingcomponentsthatweresimilartothoseinducedbyEGF. Furthermore, the magnitudes of the resulting increases in cell proliferationandmigrationwerealsosimilartooneanother.TRPV1 andCB1receptorsmayalsobelinkedtoothercellsignalingpath- waysthanthoselinkedtoEGFR.ThisispossiblebecauseTRPV1and CB1receptoractivationmodulatedIL-8releasewhereasEGFhadno effect on this response. Another indication that EGFR-linked signaling makes only a small contribution to mediating TRPV1 control of IL-6 and IL-8 release is that AG1478 only slightly decreasedCAP-inducedsuchrises.Recently,weobtainedprelimi- narydata suggesting thata componentof this alternate pathway maybetransforminggrowthfactor-betaactivatedkinase1(TAK1), whichislinkedtoTNF-areceptorcontrolofcellsurvivalincorneal epithelialcells(Wangetal.,2005).Suchapathwayisreferredtoas anon-canonicalTAK1dependentMAPKindependentpathwayand Fig.6. CB1andTRPV1stimulatemitogenesisthroughEGFRtransactivation.HCECwere is inhibited by a selective inhibitor, 5Z-7-oxozeaenol (Ninomiya- pretreatedfor30minwitheither5mMAG1478,10mMcapsazepine(CPZ),or10mM Tsuji et al., 2003). We found that 5Z-7-oxozeaenol suppressed AM251.Undersomeconditions,cellswerethenexposedforanadditional20hto CAP-inducedincreasesinIL-6andIL-8release(datanotshown). either10ng/mlEGF10mMcapsaicin(CAP)orWIN55,212-2(WIN;10mM).Cellswere thenincubatedfor1hwith1mCi/ml[3H]thymidine.Proteincontentwasdetermined CB1 receptors are seven transmembrane-domain neuronal withaBCAproteinassaykit.Thedatarepresentthemeans(cid:5)SEM(n¼3,P<0.05). receptors coupled to pertussis toxin (PTX)-sensitive Gi/o proteins. CB1activationcaninhibitadenylatecyclaseand/oractivateaswell as inhibit ion channels. Theyare activated byendogenouslyacti- signaling pathways responses underlying injury-induced corneal vated lipidic compounds such as N-arachidonoylethanolamine epithelialwoundhealing. (AEA)and2-arachidonoylglycerol(2-AG).Inaddition,CB1receptor CB1 and TRPV1 receptors elicit control of cell proliferation coupling to PLC-dependent Ca2þ mobilization from intracellular throughthreeMAPKparallelsignalingpathwaysaswellastheAkt storeshasbeenreportedindifferentcelltypes(DePetrocellisetal., Fig.7. CB1andTRPV1stimulatecellmigrationthroughEGFRtransactivation.HCECconfluentmonolayerswerepreincubatedwitheitherAM251(2mM),capsazepine(CPZ;1mM)or AG1478(5mM)for30minandscratcheswerecreatedacrosstheculturediameter.Thesameconditionedmediumwasthensupplementedwitheither10ng/mlEGF,1mMcapsaicin (CAP)or1mMWIN55,212-2(WIN)foruptoanother24h.Insetshowsrepresentativemicrographsofwoundclosureextentattheindicatedtimesforuntreatedcellsorinthe presenceofeitherEGF,CAPorWIN.Thedatarepresentthemeans(cid:5)SEM(n¼3,P<0.05). H.Yangetal./ExperimentalEyeResearch91(2010)462e471 469 andTRPV1-inducedincreasesinCa2þoccurindependentlyofone anothersinceinhibitionofCB1activationdoesnotfullysuppress TRPV1 activation. As CB1 activation suppressed TRPV1-induced increasesinIL-8,selectiveCB1activationinaclinicalsettingmay provide a novel drug strategy to reduce TRPV1-induced proin- flammatorycytokinerelease. EGFRtransactivationinHCEChasbeendescribedinresponseto exposuretolysophosphatidicacid(LPA),purinergicreceptoracti- vationandexposuretoeitherATPorinsulin(BlockandKlarlund, 2008; Lyu et al., 2006; Spix et al., 2007; Xu et al., 2006, 2007). Fig.3AandBissupportive ofthenotionthattheCa2þtransients inducedbyeitherCB1orTRPV1stimulationareinpartaccounted for by EGFR transactivation. This is evident since AG1478 sup- pressedthesemaximalrisesbyabout90%and35%,respectively.As theinhibitionswereonlypartial,someoftheCa2þriseinducedby CB1 activation may not be dependent on EGFR transactivation. Nevertheless, EGFR transactivation byexposure to either WIN or CAP is evident based on the increases in EGFR phosphorylation showninFig.4A.TheselectivityoftheCB1andTRPV1agonistsis Fig.8. IndependenteffectsofTRPV1andEGFRactivationonIL-6andIL-8release. indicatedbythefindingthateitherAM251orCPZsuppressedEGFR ELISAwasperformedafter24htoevaluateincreasesinIL-6andIL-8releaseinto DMEM/F12inthepresenceorabsenceofeitherEGF(10ng/ml)orCAP(10mM).Values phosphorylation to a level similar to that obtained with AG1478. werenormalizedtotherespectivecontrol.Insomecases,cellswerefirstexposedto EGFR transactivation by CB1 and TRPV1 agonists occurs through 10mMcapsazepine(CPZ)or1mMAG1478for30minbeforesupplementationwith theactivationofthesamematrixmetalloproteinasesthatelicitthis capsaicin (CAP; 10 mM) or EGF (10 ng/ml). Cytokine levels were measured in the response in the corneal epithelium during exposure to either supernatantsusingELISAkitsaccordingtothemanufacturer’sinstructions.Thedata lysophosphatidic acid, Pseudomonas aeruginosa or following its representthemeans(cid:5)SEM(n¼3,*P<0.001). wounding (Xu et al., 2004, 2007; Zhang et al., 2004). This corre- spondence is shown by the fact that their inhibition with either 2007;Fimianietal.,1999;McIntoshetal.,2007;Netzebandetal., GM6001 or CRM197 blocked EGFR phosphorylation (c.f. Fig. 4B). 1999).TheWIN-induced[Ca2þ]itransientsshowninFig.2Aindi- Similarly, functional blockage of EGFR activation with LA1 elimi- cate that CB1 activation by this selective CB1 agonist-induced natedWINandCAP-inducedEGFRphosphorylation. mobilizationfromintracellularCa2þstoresastheseresponseswere EGFR activation induces control of linked responses through essentiallythesameirrespectiveofthepresenceorabsenceofCa2þ sequentialtransientchangesintheMAPKphosphorylationstatus. in the bathing solution. The specificity of this responsewas vali- Such regulation is modulated through changes in the duration dated by showing that preincubation with AM251 nearly fully and magnitude of these phosphorylation events. The level of attenuatedthisresponse.Fig.2BprovidesfurtherindicationofCB1 phosphorylation increases and their duration are under the coupling to TRPV1-induced Ca2þ signaling. The mixed CB1 and control of dual specific protein phosphatases (DUSPs) (Patterson TRPV1agonist,AEA, alsoinducedarisein[Ca2þ]ilevelsthatwas et al., 2009). DUSP1 is broad spectrum in that it elicits a nega- attenuated byabout 65% during exposure to CPZ. Therefore, CB1 tive feedback on Erk1/2, p38 and JNK1/2 MAPK phosphorylation. We previously showed that phosphorylation of DUSP1 in HCEC altersthebalancebetweenEGF-inducedincreasesinproliferation and migration (Wang et al., 2009). Specifically, prolongation of Erk1/2 phosphorylation resulting from declines in DUSP1 (or MKP-1) levels resulting from GSK-3 inhibition reduced the mitogenicresponsetoEGFwhereascellmigrationwasenhanced. Inthecurrentstudy,wefoundthateventhoughCB1andTRPV1- induce EGFR transactivation followed by Erk1/2, p38 and JNK pathwaysignaling(Fig.5AandC),onlyTRPV1andCB1activation modulate proinflammatory IL-6 and IL-8 release. Fig. 8 shows thatCPZblockedCAP-inducedincreasesinIL-6andIL-8whereas EGF had no effect on the control. Our results indicating an interaction between CB1 and TRPV1 shown in Fig. 9 are consis- tent with another study in which toll-like receptor 4 (TLR4) activation is controlled by CB1. In this report, the lipopolysac- charide-induced increase in IL-8 was enhanced 2-fold through blockage of CB1 activation by AM251. Such an augmentation reveals an interaction between CB1 and TLR4 in regulating hyperinflammatory reactions by periodontal tissues (Nakajima et al., 2006). It remains to be determined if CB1 blunts through changes in GSK-3 activation CAP-induced increases in IL-8. Such changes could modulate MAPK or TAK1 signaling control of IL-8 release. Fig.9. CB1activationeliminatesTRPV1-inducedincreasesinIL-8release.HCECwere Insummary,CB1andTRPV1activationinducesinHCECthrough pretreatedwithanandamide(AEA;10mM)and/orAM251(5mM)for1handthen incubatedwithcapsaicin(CAP;10mM).Attheendoftheincubationperiod,thelevels EGFR transactivation increases in proliferation and migration. of IL-8 were determined by ELISA. The data represent the means (cid:5) SEM (n ¼ 3, However, only TRPV1 activation induced increases IL-6 and IL-8 *P<0.01). release, which are blunted through CB1 activation. These