Glucocorticoid Receptor and FKBP5 Expression Is Altered

Glucocorticoid Receptor,FKBP5,depression

Neuropsychopharmacology(2013)38,616–627

&2013AmericanCollegeofNeuropsychopharmacology.Allrightsreserved0893-133X/13

http://wendang.chazidian.com

GlucocorticoidReceptorandFKBP5ExpressionIsAlteredFollowingExposuretoChronicStress:Modulationby

AntidepressantTreatment

GianluigiGuidotti1,4,FrancescaCalabrese1,4,ChristophAnacker2,GiorgioRacagni1,3,CarmineMPariante2andMarcoARiva*,1,3`degliStudidiMilano,Milan,Italy;2SectionofDipartimentodiScienzeFarmacologicheeBiomolecolari,CentrodiNeurofarmacologia,UniversitaPerinatalPsychiatryandStress,PsychiatryandImmunology(SPI-lab),DepartmentofPsychologicalMedicine,King’sCollegeLondon,Instituteof`degliStudidiMilano,Milan,ItalyPsychiatry,London,UK;3CenterofExcellenceonNeurodegenerativeDiseases,Universita1Majordepressionisthoughttooriginatefromtheinteractionbetweensusceptibilitygenesandadverseenvironmentalevents,inparticularstress.Thehypothalamus–pituitary–adrenal(HPA)axisisthemajorsysteminvolvedinstressresponseanditsdysregulationisanimportantelementinthepathogenesisofdepression.Thestressresponseisthereforefinelytunedthroughaseriesofmechanismsthatcontrolthetraffickingofglucocorticoidreceptors(GRs)tothenucleus,includingbindingtothechaperoneproteinFKBP5andreceptorphosphorylation,suggestingthattheseelementsmayalsobeaffectedunderpathologicconditions.Onthesebases,weinvestigatedFKBP5andGRexpressionandphosphorylationinthehippocampus(ventralanddorsal)andintheprefrontalcortexofratsexposedtochronicmildstress(CMS)andweanalyzedtheeffectofaconcomitantantidepressanttreatment.WefoundthatanimalsexposedtoCMSshowincreasedexpressionofFKBP5aswellasenhancedcytoplasmiclevelsofGR,primarilyinventralhippocampusandprefrontalcortex.Chronictreatmentwiththeantidepressantduloxetineisabletonormalizesuchalterations,mainlyintheprefrontalcortex.Moreover,wedemonstratethatCMS-inducedalterationsofGRtraffickingandtranscriptionmaybesustainedbychangesinreceptorphosphorylation,whicharealsomodulatedbypharmacologicalintervention.Insummary,whileGR-relatedchangesafterCMSmightberelevantforthedepressivephenotype,theabilityofantidepressanttreatmenttocorrectsomeofthesealterationsmaycontributetothenormalizationofHPAaxisdysfunctionsassociatedwithstress-relateddisorders.Neuropsychopharmacology(2013)38,616–627;doi:10.1038/npp.2012.225;publishedonline21November2012Keywords:depression;HPAaxis;chronicmildstress;duloxetine

INTRODUCTION

Exposuretostressrepresentsamajorenvironmentalcontributortothedevelopmentofpsychiatricconditions,suchasmajordepression(PittengerandDuman,2008).Underphysiologicalconditions,stressexposureactivatesthehypothalamus–pituitary–adrenal(HPA)axis,whichleadstoglucocorticoid(GC)releasefromtheadrenalglands.TheactivationoftheHPAaxisiscontrolledthroughanegativefeedbackmechanism,bytheactivationofglucocorticoidreceptors(GRs)atdifferentlevels,includingthehippocampus(McEwenetal,1992).However,underpathologicalconditions,thefunctionoftheHPAaxisis

*Correspondence:ProfessorMARiva,DepartmentofPharmacological

andBiomolecularSciences,CenterofNeuropharmacology,University

ofMilan,ViaBalzaretti9,20133Milan,Italy,Tel:+390250318334,

Fax:þ390250318278,E-mail:m.riva@unimi.it4Theseauthorscontributedequallytothiswork.

Received28June2012;revised17September2012;accepted10

October2012disrupted,andthismayeventuallyleadtostructuralandfunctionalchangesinbrainregionsthathaveamajorroleinthepathophysiologyofdepression(deKloetetal,2005;Holsboer,2000).WhilethecauseofalteredHPAaxisfunctionindepressionisnotknown,severalevidenceindicatethatnegativefeedbackcontrolofcorticotrophin-releasinghor-mone(CRH)secretionmaybeimpairedbecauseofalteredGRfunctioninthehypothalamusandinthehippocampus(Pariante,2006).ThisinabilityofGCstoexerttheireffectsonthesetargetsleadstotheso-called‘glucocorticoidresistance’,characterizedbyincreasedCRHproductionandover-activityoftheHPAaxis.Indeed,preclinicalstudieshavedemonstratedthatGRheterozygousmutantmice(GRþ/À)haveapredispositionfordepression(Ridderetal,2005),supportingthehypothesisthatreducedexpressionandfunctionofGRsmayberelevantforthepathogenesisofstress-relatedpsychiatricdisorders(Holsboer,2000;deKloetetal,2005).Suchchangesaresupportedbyclinicalevidence,sincebothareductioninthenumberand/orfunctionofGRandabnormalpatternsof

Glucocorticoid Receptor,FKBP5,depression

cortisolsecretionleadingtoincreased24hproductionofcortisol,havebeendescribedinpeoplesufferingfrommajordepression(ParianteandLightman,2008).GRsactasa

ligand-activatedtranscriptionfactorthat,uponhormonebinding,translocatefromthecytosoltothenucleus.Thisprocess,andthesubsequentcontrolongenetranscription,isregulatedbyalargechaperoneproteincomplexconsistingofFK506bindingprotein51(FKBP5),heat-shockprotein70(HSP70),andheat-shockprotein90(HSP90).WhenFKBP5isboundtotheGRcomplexviaHSP90,thereceptorhasloweraffinityforitsligandandisretainedinthecytoplasm;uponGCbinding,GRdissociatesfromthechaperonecomplex,dimerizes,andtranslocatesintothenucleus(Binder,2009).GRtranslocationandtranscriptionalactivityisalsoregulatedbyacomplexpatternofphosphorylationondifferentserineresiduesofthereceptor(Galliher-Beckleyetal,2008;Takabeetal,2008;Wangetal,2002).Importantly,recentworkfromsomeoftheauthorsofthepresentpaper(Anackeretal,2011)hasrecentlydemonstratedthatGChormonesandantidepressantsindeedregulateGRfunctionviadifferentphosphorylationoftheGR,subsequentlyleadingtophosphorylation-dependentchangesinGR-mediatedgenetranscription.BasedontheroleexertedbyGRandtheHPAaxisindepression,theinvestigationofthemechanismscontrollingGRactivationrepresentsanimportantissuetounderstandthemolecularunderpinningscontributingtoHPAaxisdysfunctioninstress-relateddisordersandtoestablishthepotentialofpharmacologicalinterventioninnormalizingsuchalterations.Basedonthesepremises,theaimofourstudywastocharacterizethealterationofGRandFKBP5,atthetranscriptional,translationalandpost-translationallevel,inthechronicmildstress(CMS)modelofdepression(Calabreseetal,2012),andtoestablishtheimpactofchronicantidepressanttreatmentontheobservedchanges,focusingonhippocampusandprefrontalcortex,twokeyregionsformajordepressionandstress-relateddisorders.

MATERIALSANDMETHODS

ReagentswerepurchasedfromSigma-Aldrich(Milan,Italy),Bio-RadLaboratoriesS.r.l.Italia(Segrate,Italy),Roche(Monza,Italy),EurofinsMWG-Operon,(Ebersberg,Germany),Tebu-bio(Magenta,Italy),Lifetechnologies(Monza,Italy),GEHealthcareEuropeGmbH(Pero,Italy),Abcam(ProdottiGianni,Milan,Italy),andDAKO(Glostrup,Denmark).

Animals

Adult(3monthsold)maleSprague-Dawleyrats(CharlesRiver,Calco,Italy)weighing225–250gwereusedthroughouttheexperiments.Ratswerehousedingroupsofthreepercageunderstandardconditions(12hlight/darkcycle,lightoffat1900h)andwereexposedtodailyhandlingfor2weeksbeforeanyprocedure.AllanimalhandlingandexperimentalprocedureswereperformedinaccordancewiththeEC(EECCouncilDirective86/6091987),theItalianlegislationonanimalexperimentation(DecretoLegislativo116/92),andtheNationalInstitutesofHealthGuidefortheCareandUseofStressInducesAlterationsonGRandFKBP5GGuidottietal617Figure1Schematicrepresentationoftheexperimentalparadigmusedinthestudy.NOSTRESS,shamanimals;STRESS,animalsexposedtochronicmildstress(CMS)paradigm.VEH,animalstreatedwithvehicle(1%HEC,1ml/kg);DLX,animalstreatedwithduloxetine(10mg/kgdaily)http://wendang.chazidian.comboratoryAnimals.Alleffortsweremadetominimizeanimalsufferingandtoreducethenumberofanimalsused.ChronicStressParadigmandAntidepressantTreatmentTheexperimentaldesignispresentedinFigure1.Controlanimals(NoStress)werehousedunderstandardconditions,whereasstressedanimalswereexposedtothechronicstressprocedure(CMS)foraperiodof42consecutivedays,amanipulationthatleadstoachronicdepressive-likestatethatdevelopsgraduallyovertime(Calabreseetal,2012).TheCMSregimenconsistedofonceortwicedailyexposuretodifferentstressorsincludingfoodorwaterdeprivation,crowding,isolation,soiledcaged,2himmobilization,lightonovernight.After21days,bothstressedandcontrolanimalswereeachfurtherdividedintomatchedsubgroups(n.10–12ratspergroup),andforsubsequent21daystheyreceiveddailyoraladministration(bygavage)oftheantidepressantduloxetine,aserotoninandnoradrenalinereuptakeinhibitor(SNRI)(10mg/kgdaily)orvehicle(hydroxyethylcellulose,HEC,1%,1ml/kg).DuloxetineisanSNRIwidelyusedasantidepressantinhumantherapy(FramptonandPlosker,2007).Thedoseselectedinthepresentstudyisbasedonourpreviousworkdemonstratingtheabilityofchronicduloxetinetreatmenttoeffectivelymodulatingneuronalplasticity(Calabreseetal,2007;Moltenietal,2009)aswellasitsnormalizingpropertiesinanimalswithgeneticdeletionoftheserotonintransporter(Calabreseetal,2010;Guidottietal,2012).Onday43,inthemorning,ratswerekilledbydecapitation24hafterthelastduloxetineadministration.Andbrainregionswererapidlydissected,frozenondryiceandstoredatÀ801Cforfurtheranalysis.Theprefrontalcortex(definedasCg1,Cg3,andILsubregionscorrespondingtotheplates6–10accordingtotheatlasofPaxinosandWatson,1996)wasdissectedfrom2mm-thickslices,whereasthehippocampuswasdissectedfrom2mm-thickslicescorrespondingtoplates25–33(dorsalhippo-campus)andplates34–43(ventralhippocampus)accordingtotheatlasofPaxinosandWatson.SucrosePreferenceTestThesucrosepreferencetestwasperformedonthe21stday.Forthisaim,animalswereinitiallyexposedfor48htoapalatablesucrosesolution(1%)toavoidneophobia.Onday

Neuropsychopharmacology

Glucocorticoid Receptor,FKBP5,depression

StressInducesAlterationsonGRandFKBP5

GGuidottietal618

21,12hafterwaterdeprivation,sucrosepreferencewasthendeterminedby1hexposuretotwoidenticalbottlesfilledwitheithersucrosesolutionorwater.Sucrosepreferencewasdefinedastheratioofthevolumeofsucrosevstotalliquid(waterþsucrose)consumedduringthe1-htest.Wedidnotperformthetestattheendofthetreatment,ie,beforethekilling,becauseofthepossibleinfluenceexertedbysucroseconsumptionontheexpressionofthemoleculartargetsunderinvestigation.

RNAPreparationandGeneExpressionAnalysisbyQuantitativeReal-TimePCR

TotalRNAwasisolatedbysinglestepofguanidiniumisothiocyanate/phenolextractionusingPureZolRNAisolationreagent(Bio-RadLaboratories)accordingtomanufacturer’sinstructionsandquantifiedbyspectrophotometricanalysis.FollowingtotalRNAextraction,thesampleswereprocessedforreal-timePCR(RT-PCR)toassessFKBP5andNr3c1mRNAlevels.AnaliquotofeachsamplewastreatedwithDNasetoavoidDNAcontamination.RNAwasanalyzedbyTaqManqRTPCRinstrument(CFX384realtimesystem;Bio-RadLaboratories)usingtheiScriptTMone-stepRT-PCRkitforprobes(Bio-RadLaboratories).Sampleswererunin384wellformatsintriplicateasmultiplexedreactionswithanormalizinginternalcontrol(36B4).Primersandprobessequencesused(Table1)werepurchasedfromEurofinsMWG-OperonandLifeTechnologies.Thermalcyclingwasinitiatedwithanincubationat501Cfor10min(RNAretrotranscription)andthenat951Cfor5min(TaqManpolymeraseactivation).Afterthisinitialstep,39cyclesofPCRwereperformed.EachPCRcycleconsistedofheatingthesamplesat951Cfor10stoenablethemeltingprocessandthenfor30sat601Cfortheannealingandextensionreactions.Acomparativecyclethreshold(Ct)methodwasusedtocalculatetherelativetargetgeneexpression.cytosolicfraction.Tissuesweremanuallyhomogenizedusingaglass-glasspotterinapH7.4coldbuffercontaining0.32Msucrose,0.1mMEGTA,1mMHEPESsolutioninpresenceofacompletesetofprotease(Roche)andphosphatase(Sigma-Aldrich)inhibitors.Thetotalhomo-genatewascentrifugedat2500r.p.m.for10minat41Ctoobtainthepelletcorrespondingtothenuclearfractionwhichwasresuspendedinabuffer(20mMHEPES,0.1mMdithiothreitol(DTT),0.1mMEGTA)withproteaseandphosphataseinhibitors.Thesupernatantobtainedwascentrifugedat10000gfor15minat41Candthesuper-natantobtainedcorrespondtothecytosolicfraction.Thepurityofsubcellularfractionswasassayedbyanti-glyceraldehyde3-phosphatedehydrogenase(GAPDH)oranti-histoneH3antibodiesforthecytoplasmicornuclearcompartmentsrespectively,asshowninFigure6d.TotalproteincontentwasmeasuredaccordingtotheBradfordProteinAssayprocedure(Bio-RadLaboratories),usingbovineserumalbuminascalibrationstandard.Equalamountsofproteinwererununderreducingconditionson10%SDS-polyacrylamidegelsandthenelectrophoreti-callytransferredontonitrocellulosemembranes(Bio-RadLaboratories).Theblotswereblockedwith10%non-fatdrymilkandthenincubatedwiththeprimaryantibodiessummarizedinTable2.MembraneswerethenincubatedTable2AntibodiesConditionsusedintheWesternBlotAnalysis.GeneFKBP5GRPhosphoGRS224PhosphoGRS232

PhosphoGRS246

b-ActinPrimaryantibody1:2000(Abcam),41C,O/N1:500(ThermoScientific),41C,O/N1:10000,41C,O/N1:500(Abcam),41C,O/N1:2000,41C,O/N1:10000(Sigma),

41C,O/NSecondaryantibodyAnti-rabbit,1:2500,RT,1hAnti-rabbit,1:2000,RT,1hAnti-rabbit,1:2000,RT,1hAnti-rabbit,1:2000,RT,1hAnti-rabbit,1:2000,RT,1hAnti-mouse,1:10000,RT,1hAnalysisofFKBP5andGRProteinLevelsandGRPhosphorylationWesternblotanalysiswasusedtoinvestigateFKBP5,GR,andthreeGRphospho-proteinsinthenuclearandinthe

Table1SequencesofForwardandReversePrimersandProbesusedinReal-timePCRAnalysesandPurchasedfromEurofinsMWG-Operon(a)andfromLifeTechnologies,WhichdidnotDisclosetheSequences(b).

(a)ForwardandreverseprimersandprobespurchasedfromEurofinsMWG-Operon

Gene

Fkbp5

Nr3c1

Foxo1

p11

36B4ForwardprimerGAACCCAATGCTGAGCTTATGGAAAAGCCATCGTCAAAAGGGGAGTGGATGGTGAAGAGTGTGAGAGTGCTCATGGAAAGGGATTCCCACTGGCTGAAAAGGTReverseprimerATGTACTTGCCTCCCTTGAAGTGGAAGCAGTAGGTAAGGAGAGGACAGATTGTGGCGAATTGAGCTCTGGAAGCCCACTTTTCGCAGCCGCAAATGCProbeTGTCCATCTCCCAGGATTCTTTGGCAGCTTTGTCAGTTGGTAAAACCGTTGCTCAAGGATAAGGGCGACAGCAACAGATAATGAAAGACCTGGACCAGTGCAAGGCCTTCCTGGCCGATCCATC(b)ForwardandreverseprimersandprobespurchasedfromLifeTechnologies

Gene

Gadd45bAccessionnumberBC085337.1AssayIDRn01452530_g1

Neuropsychopharmacology

Glucocorticoid Receptor,FKBP5,depression

for1hatroomtemperaturewiththeopportunesecondaryantibody(seeTable2),andimmunocomplexeswerevisualizedbychemiluminescenceusingtheECLWesternBlottingkit(GEHealthcareEuropeGmbH).ForGRphosphorylationanalysis,proteinsamplescon-taining30mgoftotalproteinwereboiledfor10minat721Cin1ÂNuPAGELDSsamplebuffer(LifeTechnologies)and1ÂNuPAGEsamplereducingagent(LifeTechnologies),andsubjectedtoreducingSDS-PAGEon10%NuPAGEBis-Trisgelsfor1hat200V.ProteinswereelectrophoreticallytransferredontoImmuno-BlotPVDFmembranes(Bio-RadLaboratories)at110Vfor1.5hat41C.TransferefficiencywascontrolledbyPonceauSstainingandbyprestainedproteinstandards.Unspecificbindingsiteswereblockedfor1hin5%BSAinTBSandmembraneswereimmunoprobedwiththepolyclonalrabbitanti-P-S224(1:10000)andanti-P-S246(1:2000)antibodies(bothfromDrMichaelJGarabedian),anti-P-S232antibody(1:500;Abcam)inblockingsolutionat41Covernight.MembraneswerewashedwithTBScontaining0.1%Tween-20(TBST)andincubatedwithanHRP-conjugatedswineanti-rabbitsecondaryantibody(1:2000;DAKO),in5%non-fatdrymilkinTBS,for1hatroomtemperature.MembraneswerewashedinTBSTandproteinswerevisualizedwithenhancedchemiluminescence(ECL)detectionsystem(GEHealthcare,UK).Resultswerestandardizedusingb-actinasthecontrolprotein,whichwasdetectedbyevaluatingthebanddensityat43kDa.ProteinlevelswerecalculatedbymeasuringtheopticaldensityoftheautoradiographicbandsusingQuantityOnesoftware(Bio-RadLaboratories).Toensurethatautoradiographicbandswereinthelinearrangeofintensity,differentexposuretimeswereused.

StatisticalAnalyses

Theeffectsofstressand/orantidepressanttreatmentwereanalyzedwithstudent’st-test,withatwo-wayanalysisofvariance(ANOVA)followedbySingleContrastPostHocTest(SCPHT).SignificanceforalltestswasassumedforPo0.05.Dataarepresentedasmeanvalues±standarderror(SEM).Forgraphicclarity,resultsarepresentedasmeanpercentofNoStress/Vehicle-treatedrats.

RESULTS

EffectsofCMSonBodyWeightandBehavior

Beforestartingtheantidepressanttreatment,weestablishedtheeffectivenessoftheadversemanipulationbymeasuringbodyweightandanhedonia.AsshowninFigure2a,animalsexposedto3weeksofCMSshowedsignificantlylessweightgainwhencomparedwithcontrolanimals(Po0.001),aneffectthatmaybeduetoreducedfoodandwaterconsumption(datanotshown).Moreover,asshowninFigure2b,wefoundthatratsexposedfor3weekstotheCMSdisplayasignificantlyreducedpreferenceforsucrosesolution(À16%,Po0.05),whencomparedwithnon-stressedrats.Overall,thesechangesareclearindicatorsfortheefficacyofthestressfulmanipulation.StressInducesAlterationsonGRandFKBP5GGuidottietal619Figure2Effectsofchronicmildstress(CMS)onratweightgainandsucrosepreference.(a)Animalsshowedasignificantdecreaseinbodyweightgainafter3weeksofCMS,whencomparedwithnostressanimals.Thedata,expressedasweightdifferencebetweenday21andday1,representthemean±SEMofatleast33independentdeterminations.***Po0.001vsNoStress(Student’st-test).(b)Analysisofsucrosepreference.Animalswereinitiallyexposedfor48htoapalatablesucrosesolution(1%)toavoidneophobia.Onday21,12hafterwaterdeprivation,sucrosepreferencewasthendeterminedby1hexposuretotwoidenticalbottlesfilledwitheithersucrosesolutionorwater.Sucrosepreferencewasdefinedastheratioofthevolumeofsucrosevstotalliquid(waterþsucrose)consumedduringthe1-htest.*Po0.05vsNoStress(Student’st-test).ModulationofFKBP5mRNAandProteinExpressionWeinitiallyinvestigatedwhetherexposuretoCMScouldaltermRNAandproteinlevelsofthehsp90co-chaperoneFKBP5,andtowhatextentchronicantidepressanttreatmentmayaffectthechangesproducedbystressexposure.ThemodulationofFKBP5wasstrictlyareadependent.Infact,asshowninFigure3a,Fkbp5mRNAlevelsintheventralhippocampusweresignificantlyregulatedbystress(F1,48¼24.90,Po0.001)withasignificantstressÂtreat-mentinteraction(F1,48¼16.40,Po0.001).Indeed,Fkbp5expressionwassignificantlyincreasedfollowingCMS(þ213%,Po0.001vsNoStress/Vehanimals).DuloxetineproducedasignificantincreaseinFkbp5mRNAlevelswhengiventocontrolrats(þ102%,Po0.05vsNoStress/Vehanimals),butreduceditwhenadministeredtoCMSrats(À29%,Po0.05vsStress/Vehanimals).Inthedorsalhippocampus(Figure3b),wefoundasignificanteffectofchronicstress(Fsignificantly1,47¼14.11,Po0.001).Again,Fkbp5expressionwasincreasedafterCMSexposure(þ69%,Po0.01,vsNoStress/Vehanimals),aneffectthatwasnotinfluencedbyduloxetineadministration.Then,weanalyzedthemodulationofFkbp5intheprefrontalcortex(Figure3c)andagainwefoundasignificantstressÂtreatmentinteraction(Ftoventralanddorsal1,47¼9.55,Po0.01).Infact,simi-larlyhippocampus,CMSproducedasignificantincreaseinFkbp5mRNAlevels(þ102%,Po0.01,vsNoStress/Vehanimals).ChronicduloxetinereducedFkbp5mRNAlevelsinCMSrats(À44%,Po0.05,vsStress/Vehanimals),whileincreasingitsexpressionincontrolanimals,althoughthischangedidnotreachstatisticalsignificanceduetosomevariabilitywithinthegroupofshamratstreatedwithduloxetine(þ

42%,

Neuropsychopharmacology

Glucocorticoid Receptor,FKBP5,depression

StressInducesAlterationsonGRandFKBP5

GGuidottietal620

Figure3ModulationofFkbp5expressionintheratbrainfollowingchronicstressandantidepressanttreatment.ThemRNAlevelsofFkbp5weremeasuredinventralhippocampus(a),dorsalhippocampus(b),prefrontalcortex(c),andhypothalamus(d)ofnon-stressedandchronicallystressedrats,treatedfor21dayswithvehicleorduloxetineandkilled24hafterthelaststress.Thedata,expressedasapercentageofNoStress/Vehicle(setat100%),arethemean±SEMof6–12independentdeterminations.*Po0.05,**Po0.01,***Po0.01vsNoStress/Vehicle;$Po0.05vsStress/Vehicle(two-wayANOVAwith

SCPHT).

Figure4ModulationofFKBP5proteinlevelsbyCMSandantidepressanttreatmentintheratbrain.TheproteinlevelsofFKBP5weremeasuredinthecytosolicfractionofventralhippocampus(a),dorsalhippocampus(b),andprefrontalcortex(c)ofnon-stressedandchronicallystressedrats,treatedfor21dayswithvehicleorduloxetineandkilled24hafterthelaststress.Thedata,expressedasapercentageofNoStress/Vehicle(setat100%),arethemean±SEMof5–7independentdeterminations.*Po0.05vsNoStress/Vehicle;$Po0.05vsStress/Vehicle(two-wayANOVAwithSCPHT).P40.05,vsNoStress/Vehanimals).Finallyinthehypotha-lamus,Fkbp5expressionwasnotinfluencedbystress(F1,24¼0.82,P40.05)orbydrugtreatment(F1,24¼1.18,P40.05)(Figure3d).Basedonthesedata,wedecidedtoexaminetheproteinlevelsofFKBP5inthecytosolicfractionofventral,dorsalhippocampus,andprefrontalcortex.Theproteinlevelsofthisimmunophilinmirroredonlyinpartthechangesfoundingeneexpression.AsshowninFigure4a,intheventralhippocampuswefoundasignificantstresseffect(F1,25¼16.19,Po0.001):CMSincreasedFKBP5levels(þ31%,Po0.05vsNoStress/Vehanimals),aneffectthatwasnotmodulatedbychronicduloxetinetreatment.Inthedorsalhippocampus(Figure4b),wefoundnoeffectofchronicstressbutasignificanttreatmenteffect(F1,25¼5.74,

NeuropsychopharmacologyPo0.05),whichwasduetothesignificantincreaseinFKBP5protein(þ16%,Po0.05vsNoStress/Vehanimals)followingduloxetineadministrationtocontrolrats.Intheprefrontalcortex(Figure4c),CMSsignificantlyincreasedFKBP5proteinlevels(þ33%,Po0.05vsNoStress/Vehanimals),whereasduloxetinetreatmentreversedtheupregulationfoundinstressedrats(À27%,Po0.05,vsNoStress/Vehanimals)withoutaffectingthechaperonelevelsincontrolanimals.ModulationofGRmRNAandProteinLevelsWenextinvestigatedtheexpressionoftheGR.WithrespecttothetranscriptionallevelsofthegenecodingforGR,Nr3c1,wedidnotfindanyeffectofCMSinthe

ventral