Electrophysiological alterations in a complex rat model ofschizophrenia Behavioural Brain Research

(1)

ContentslistsavailableatScienceDirect

Behavioural Brain Research

j o ur na l h o me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / b b r

Research report

Electrophysiological alterations in a complex rat model of schizophrenia

Gyongyi Horvath

^a,∗

, Zita Petrovszki

^b

, Gabriella Kekesi

^a

, Gabor Tuboly

^c

, Balazs Bodosi

^a

, Janos Horvath

^a

, Peter Gombköt ˝o

^a

, Gyorgy Benedek

^a

, Attila Nagy

^a

aDepartmentofPhysiology,FacultyofMedicine,UniversityofSzeged,Dómtér10.,H-6720Szeged,Hungary

bInstituteofPhysicalEducationandSportMedicine,JuhászGyulaFacultyofEducation,UniversityofSzeged,Hattyassor10.,H-6725Szeged,Hungary

cDepartmentofNeurology,FacultyofMedicine,UniversityofSzeged,Semmelweisu.6.,H-6720Szeged,Hungary

h i g h l i g h t s

•EEGphenotypecharacterizationinaratsubstrainrelatedtoschizophrenia/autism.

•ERPsshowedsigniﬁcantchangesinP2latencyandN1amplitude.

•AcuteketaminetreatmentdidnotcausealterationsinERPs.

•Alteredpowerofoscillationsindifferentfrequencybandswasobserved.

•Ketaminecausedstrain-dependentchangesinthepowerofoscillations.

a r t i c l e i n f o

Articlehistory:

Received18December2015

Receivedinrevisedform25March2016 Accepted29March2016

Availableonline30March2016

Keywords:

Autism

Eventrelatedpotentials Neuronaloscillation Powerspectrumanalysis Ratmodel

Schizophrenia

a b s t r a c t

Background:Psychiatricdisordersarefrequentlyaccompaniedbychangesinbrainelectricaloscilla- tionsandabnormalauditoryeventrelatedpotentials.Thegoalofthisstudywastocharacterizethese parametersofanewratsubstrainshowingseveralalterationsrelatedtoschizophrenia.

Methods:Maleratsofthenewsubstrain,developedbyselectivebreedingaftercombinedsubchronic ketaminetreatmentandpostweaningsocialisolation,andnaiveWistaronesgroup-housedwithoutany interventionswereinvolvedinthepresentstudy.Attheageof3months,animalswereimplantedwith corticalelectroencephalographyelectrodes.Auditoryevokedpotentialsduringpaired-clickstimuliand powerofoscillationindifferentfrequencybandsweredeterminedwithandwithoutacuteketamine (20mg/kg)treatment.

Results:Regardingtheauditoryevokedpotentials,thelatencyofP2wasdelayedandtheamplitudeofN1 peakwaslowerinthenewsubstrain.Thenewsubstrainshowedincreasedpowerofoscillationsinthe theta,alphaandbetabands,whiledecreasedpowerwasdetectedindeltaandgamma2bands(52–70Hz) comparedwithcontrolanimals.Acuteketaminetreatmentincreasedthegamma1band(30–48Hz)power inbothgroups,whileitelicitedsigniﬁcantchangesonlyinthenewsubstraininthetotalpowerandin alpha,betaandgamma2bands.

Conclusions:Thevalidationofthetranslationalutilityofthisnewratsubstrainbyelectrophysiological investigationsrevealedthattheseratsshowabnormalitiesthatmaymodelapartoftheneurophysiolog- icaldeﬁcitsobservedinschizophrenia.

Abbreviations:EEG,electroencephalography;ERP,eventrelatedpotential;NMDAR,N-methyl-d-aspartatereceptor.

∗Correspondingauthor.

E-mailaddresses:horvath.gyongyi@med.u-szeged.hu(G.Horvath),petrovszki.zita@med.u-szeged.hu(Z.Petrovszki),kekesi.gabriella@med.u-szeged.hu(G.Kekesi), tuboly.gabor@med.u-szeged.hu(G.Tuboly),bodosi.balazs@med.u-szeged.hu(B.Bodosi),horvath.janos@med.u-szeged.hu(J.Horvath),gombpeti@gmail.com(P.Gombköt ˝o), benedek.gyorgy@med.u-szeged.hu(G.Benedek),nagy.attila.1@med.u-szeged.hu(A.Nagy).

(2)

1. Introduction

Schizophrenia is a commonneurodevelopmental and highly heritable neuropsychiatric disorder [1,2]. Over the past few decades, researches using electroencephalography (EEG) have identifiedseveral neurophysiologicalalterations in this disease, indicatingneuralcircuitdisruptions.Unfortunately,theresultsare controversial,andtheymaydependonthesubtypeorphaseofthe disease;therefore,itsmodelinginpreclinicalresearchfieldisabig challenge[3–9].Itisarguedthatperfectsimulationofinherently humandiseasesinanimalsmightbeimpossible,buttherecreation ofendophenotypesrelatedtothedisordersisapossibility.There- fore,developinganimalmodelswithabnormalEEGactivitymay helpintheclarificationofthemechanismsinthebackgroundofthis neuropsychiatricdisease[10,11].Previousstudiesusingdifferent rodentmodelsofschizophreniashowedwidelydiversifiedalter- ationsinthepowerofEEGoscillationsandeventrelatedpotentials (ERP)[12–23].

Preclinicalandclinicalstudiesfocusingonpharmacologicaland genomicalchangessupportthehypothesisthathypofunctionofN- methyl-d-aspartatereceptor(NMDAR)signalingcontributestothe pathophysiologyofschizophrenia;therefore,NMDARantagonists, includingketamine,havebeen usedextensivelytoprobe ques- tionsrelatedtoitsneurobiology[24–28].Ketamineadministeredto healthycontrolsinsubanestheticdosemimicsseveralsymptomsof schizophrenia,anditworsensthesesignsinschizophreniapatients [24,29,30].Furthermore,NMDARantagonistsorsilencingofthese receptorsareusedinanimalmodelsofneuropsychiatricdisorders showingseveralalterationsinEEGactivity,too[13,18,25,31–34].

We developed a complex animal model by selective breed- ing based on behavioral alterations after combined subchronic ketaminetreatmentandpostweaningsocialisolation[35,36].Itis thoughtthatselectivebreedingforphenotypicextremesincreases thehomozygosityofgenesthataffecttheselectedtrait,whereby theallelicfrequencyoftrait-irrelevantgenesremainsunaffected [37]. Therefore, animals selectively bred for a behavioral given phenotypeareincreasinglyusedtostudypathophysiologicalmech- anismsunderlyingcertaindisorders.Forexample,ratshavebeen successfullybredforanxiety[38],reducedsensorimotorgating[39]

andforseizuresusceptibility[40].Severalaspectsofschizophre- niawere foundin the newsubstrain, i.e., disturbances in pain sensitivity,sensorygating,memoryfunctions,motoractivityand stereotypicbehaviors[35,36].Ourrecentdataindicatedthatboth heritable and environmental factors (i.e., juvenile social isola- tionandketaminetreatment)areimportantinthegenerationof thebehavioralalterations,butthemostsigniﬁcantchangeswere observedafterthecombinationoftreatmentswithselectivebreed- ing[35,36,41].Inordertokeepthenumberofanimalsusedinthe experimentsatminimallevel,wedecidedtocomparetwogroups ofanimals,i.e.,naiveratswithoutanyinterventionandthenew substrainafterjuvenileisolationandsubchronicketaminetreat- ment.In this report, theelectrophysiological phenotypeof this newratsubstrainwascharacterizedbytheinvestigationofERPs, theirgating,andthebasalfrequencybandswithandwithoutacute ketaminetreatment,totestthepotentialusefulnessofthesubstrain instudyingtheneurophysiologicaldeﬁcitsrelatedtoschizophre- nia.

2. Methods

Allexperiments involving animal subjects were carried out withtheapprovaloftheHungarianEthicsCommitteeforAnimal Research(registrationnumber:XIV/03285/2011).Animalsuffering andthenumberofanimalspergroupwerekepttoaminimum.

2.1. Selectivebreedingprocess

Theparadigmforselectivebreedingwasdescribedpreviously [35,36]. Brieﬂy, the parental generation consisted of male and female(10-10)outbredWistarrats.Offspringsoftheratsinthenext generationsweretestedafterweaningwiththetail-ﬂicktest(48^◦C hotwater)toassesspain sensitivity,andthenhousedindividu- allyfor28days.Theanimalsweretreatedwithketamine(Calypsol, RichterGedeonPlc.,Budapest,Hungary;30mg/kgIP,4mL/kg,daily, 5times/week,15injectionsintotal)from5to7weeksofage.Then theanimalswerere-housed(4–5rats/cage)and1weekofrecov- erywasprovidedtothemwithnotreatment.Startingattheage of9weeks,thepainsensitivity,thesensorygatingwithprepulse inhibition,andthecognitivefunctionsandstereotypicbehavioron novelobjectcognitiontestwereassessed(Table1).Animals(5male with10female) withthehighest levelof disturbancesinthese parameterswereusedforselectivebreedingthroughoutseveral generations.

2.2. ExperimentalparadigmforEEGexperiments

Twoexperimentalgroupsof 8–8ratswerecompared:naive socializedmaleratswithoutanyinterventions;andthe17thgen- eration of selectively bred male rats with social isolation and ketaminetreatmentasnewsubstrain.Aftertheabove-mentioned behavioraltests,theanimalswereinvolvedintheEEGexperiments (Table1).

Ratswereanesthetizedwithamixtureofketaminehydrochlo- rideandxylazine(72and8mg/kgintraperitoneally,respectively), andtransferredintoa stereotaxicframe.Afterwards,smallburr holesweredrilledintheirskullforelectrodeplacementaccord- ingtocoordinatesfoundintheratbrainatlas[42].Thetargetarea fortheepiduralstainlesssteelelectrodesandcoordinatesrelative tobregmawerethefollowingatbothsides:recordingelectrodes:

parietalcortex6mmposterior,2mmlateraltobregma;reference electrodes:1.5mmposteriortobregma,2mmlateral,andaground electrode2.5mmposteriortobregma,1mmlateral.Finally,elec- trodeswereplacedinaminiature6-pinconnector,whichwasﬁxed withdentalcement.

Afterthesurgery,theanimalswereinjectedwithgentamycin (10mg/kg,subcutaneously)topreventinfection,andwerehoused individually.Theywereallowedtorecoverforoneweekwitha 12:12-h light–dark cycle, anambient temperature regulatedat 23^◦C,waterandfoodwithadlibitumaccess.

Onthetestingdays(between8:00AMand12:00PM),animals wereplacedintherecordingchamber(L:34cm,W:14.5cm, H:

33.5cm),recordingcableswereattachedtocommutatorsallowing thefreemovementoftherats,andtheywereallowedtoaccom- modatetothetestenvironmentfor10minwhileauditorystimuli werenotpresent.

Followingtheacclimatization,the20mintestsessionwasini- tiated.ForgenerationofERPstothesensorygatingparadigm,two consecutiveclicks(70dBclickswithbroadspectrumfor5ms:S1 andS2)werepresentedwithinterstimulusintervalof500ms.The intervalbetweenthepairsofclickswas5s.Clicksweredrivenbya computerprogramanddeliveredvialoudspeakers.

Tohabituatetheanimalstothetaskandtominimizethepoten- tialdiscomfort during the tests, three recording sessions were performedonthreeconsecutiveweekswithoutanyintervention.

ThenEEGrecordingswererepeatedafteracuteketamine(20mg/kg intraperitoneally)orvehicle(saline)injectiononthesubsequent two weeks.Eachanimal wasgivenboth injections with7 days apart,andtheorderofvehicleanddrugadministrationwascoun- terbalanced.Aftertheinjections,theratswereplacedintheircage for20minbeforeputtingthemtotherecordingchamberforEEG registration.

(3)

Table1

Experimentalparadigm.

Groups Age(weeks)

3 4 5–7 8 9 10 12 13–15 16–17

Naiverats (n=8) New substrain (n=8)

(PD21) weaningTF test1

group housing social isolation

socialisola- tion+ketamine treatment

group housing

TFtest2 PPItest

NORtest EEG

electrode implanta- tion+recovery (7days)

EEGfor habituation

EEG registration afteracute ketamine/saline treatment Abbreviations:PD—postnatalday;TF—tail-ﬂick;PPI—prepulseinhibition;NOR—novelobjectrecognition;EEG—electroencephalogram.

Fig.1. Eventrelatedpotentials(ERPs)usingthepaired-clickparadigm.

AverageofERPswithSEMinnaiveandnewsubstrainratsinresponsetotheﬁrst (S1)andthesecond(S2)auditorystimuli.Arrowsdenotestimulusonsets.P1,N1 andP2refertoERP-relatedpeaks.

BipolarEEGrecordingswereperformedfrombothsidesofthe skull.Thesignalswereamplifiedwithacustommadeeightchan- nelEEGamplifier(gainsetting:1000) byusingAD8222(Analog DevicesInc.)highperformanceinstrumentationamplifier.Thesig- nalswereonlinefilteredusingthefollowingfiltersettings:high passfilter=0.33Hz;lowpassfilter=155Hz.Theamplifiedandfil- teredsignalswerethendigitizedandrecordedwitha16channel Datawavesystem.Thesamplingrateoftheelectrophysiologicalsig- nalswas5kHz.TherecordedEEGsignalswerestoredonacomputer forsubsequentofflineanalysis.

2.3. Dataanalysis

WecalculatedtheaveragedERPinresponsetoS1and S2to measuretheamplitudes,latenciesoftheresponsesandtheirgating.

TheP1, N1 andP2 componentswere identiﬁedaccordingly:

P1wasthefirstpositive-goingwave thatoccursintherangeof 10–45msafterstimulus,theN1componentwasthefirstnegative- goingcomponentdirectlyfollowingP1intherangeof20–70ms afterstimulus,and P2wasthesecondpositive-going wavethat occursbetween40and100ms(Fig.1).Theamplitudesandthe latenciesofeachcomponentweredetermined.Gatingwasdefined astheratioofthepeak-to-peakamplitudesofthecorresponding componentsrecordedasthefirstandsecondERPs.Therefore,two segmentswerecalculated:thedifferencebetweenP1andN1and betweenN1andP2,andthemagnitudeofinhibitionwasdefinedas theratiooftheevokedresponses(S2/S1)forbothamplitudes(G1 andG2).

Power density values were calculated by fast Fourier trans- formation (FFT) of artefact-free epochs under condition of 0.61Hz resolution with a Hanning window for 5min peri-

ods before the auditory stimuli in the frequency range of delta(0.6–4Hz),theta(4–8Hz),alpha(8–13Hz),beta(13–30Hz) and gamma1 (30–48Hz), gamma2 (52–70Hz) and gamma3 (71–100Hz)bins/waves.Relativebandpowerswereexpressedas powerratiosofeachfrequencybandtothetotal(z-score).

TheobtainedEEGdatawereanalyzedoff-linewithDatawave system(DataWaveTechnologies,Loveland,CO,USA)andSpike2 (CambridgeElectronicDesign,Cambridge,UK)systems.Dataare expressedasmeans±SEM.Meanvaluesofthedifferentparameters werecomparedwithANOVA,withfactorsgroup,clicksandtreat- ment.Whentheglobaltestwassigniﬁcant,theLSDposthoctest wasusedfortheevaluationoftheeffectsofthevariousparameters.

Statistical analysiswasperformedwithStatistica 11.0 software (Statsoft,Tulsa,Oklahoma,USA).Differenceswereconsideredsig- niﬁcantforp<0.05.

3. Results

3.1. Behavioralalterations

Inagreementwithourrecentstudies[35,36],thenaiveandnew substrain ratsinvolved intheEEG experimentsshowedsigniﬁ- cantdifferencesonthebehavioraltests.Thus,thenewsubstrain showedblunted painsensitivity detectedat theageof3 and9 weeks:ANOVArevealedsigniﬁcanteffectsofgroup(F_(1,14)=12.53;

p<0.05) and time (F_(1,14)=87.20; p<0.0001) with signiﬁcantly longerlatenciesin thenewsubstrainat week9.Impaired sensory gating on the prepulse-inhibition test was also present:

ANOVArevealedasignificanteffectofgroup(F_(1,14)=5.99;p<0.05) withlowervaluesinthenewsubstrain.ANOVAshowedthatthe newsubstrainspentsignificantlylesstimewiththeexploration of the new object compared to the naive ones (F_(1,14)=11.73, p<0.005) inthenovelobjectrecognitiontest. Furthermore,the grooming activityofthenewsubstrainwassignificantlyhigher (F_(1,14)=5.92, p<0.05) that was accompanied by lower rearing activity(F_(1,14)=10.71,p<0.01).

Qualitativeobservationsindicatedthattheratshadmildimpair- ments in coordination and locomotor activity following acute ketaminetreatmentduringtheEEGrecordings.

3.2. AnalysisofERPresponses

RegardingthelatencyofP1andN1peaks,therewerenosignif- icantdifferencesbetweenthegroups,thetreatments,thefirstand secondclicksandtheirinteractions(dataarenotshown).Regard- ingthelatencyofP2,ANOVAshowedasignificanteffectofgroup (F_(1,28)=14.99;p<0.001),thusthenewsubstrainhadlongerlaten- ciescomparedtothecontrolgroup,butneitherketaminetreatment northeorderoftheclicksinfluencedit(Fig.2A).

Thesecondclickinducedloweramplitudesofallthepeakscom- paredtothefirstoneinbothgroups,andthisdecreasewasnot influencedbyacuteketaminetreatment.Regardingthedifferences betweenthetwogroupsforbothclicksnosignificantdifferences wereobservedintheamplitudeofP1andP2peaks(dataarenot

(4)

Fig.2.Alterationsinthepeaksofeventrelatedpotentials.

MeanP2latencies(A)andN1amplitudes(B)topairedstimuliinnaiveandnewsubstrainratsaftersalineorketaminetreatment.Thesymbol*signssigniﬁcantdifferences comparedtonaivegroup.Thesymbol+denotessigniﬁcantdifferencesbetweenresponsestoS1andS2stimuli.

shown),butN1peakwassigniﬁcantlylowerinthenewsubstrain comparedtothecontrolone;thus,ANOVAshowedasigniﬁcant effectofgroup(F_(1,28)=11.44;p<0.005)(Fig.2B).

Asregardsthedegreeofgatingtherewasnosigniﬁcantdiffer- encebetweenthetwogroupsintheG1value(G1naive:0.53±0.03 vs.G1newsubstrain:0.51±0.09),whileG2washigherinthecon- trolgroup,thusthedegreeofthegatingwaslowerintheseanimals (G2naive:0.63±0.04vs.G2newsubstrain:0.49±0.03;p<0.05).

Acuteketaminetreatmenthadnosigniﬁcantinﬂuenceonthese parameters(G1naive0.55±0.03vs.G1newsubstrain0.42±0.05;

andG2naive:0.52±0.04vs.G2newsubstrain:0.48±0.04).

3.3. Oscillatoryactivity

Asregardsthetotalpowerofthewaves,ANOVAshowedasignif- icanteffectofgroup(F_(1,28)=9.15;p<0.01).Post-hoctestrevealed thatthenewsubstrainhadsigniﬁcantlyhighertotalpowerafter ketaminetreatmentcomparedtothecontrolanimals(Fig.3A).

Regardingthedeltaband,ANOVAshowedasignificanteffectof group(F_(1,28)=8.84;p<0.01);thus,thenewsubstrainhadlower powerinthisfrequencyband,whileketaminedidnotinfluence significantlythisparameterineithergroup(Fig.3B).

Asregardsthethetaband,ANOVAshowedasigniﬁcanteffectof group(F_(1,28)=6.12;p<0.05;alpha);therefore,thenewsubstrain hadhigherpowerinthisfrequencybandwithoutketaminetreat- ment,butketaminedecreased thedifferencesbetweenthetwo groups(Fig.3B).

Asregardsthealphaband,ANOVAshowedasigniﬁcanteffectof group(F_(1,28)=4.89;p<0.05)andtreatment(F_(1,28)=8.37;p<0.01);

thus,thenewsubstrainhadhigherpowerinthisfrequencyband that was signiﬁcantly decreased by acute ketamine treatment (Fig.3C).

Asregardsthebetaband,ANOVAshowedasigniﬁcanteffectof treatment(F_(1,28)=10.41;p<0.005)andthegroupandtreatment interaction(F_(1,28)=5.21;p<0.05); therefore,thenew substrain hadhigher powerinthis frequency bandthat wassigniﬁcantly decreasedbyacute ketaminetreatmentsimilarly toalphaband (Fig.3C).

Asregardsthegamma1band,ANOVAshowedasigniﬁcanteffect oftreatment(F_(1,28)=11.81;p<0.005),thus,acuteketamineinjec- tionsigniﬁcantlyincreasedthepowerofthisbandinbothgroups (Fig.3D).

Asregardsthegamma2band,ANOVAshowedasignificanteffect oftreatment(F_(1,28)=5.45;p<0.05)andthegroupandtreatment interaction(F_(1,28)=6.22; p<0.05); therefore,the newsubstrain had lower powerin this frequency band that wassignificantly increasedbyacuteketaminetreatment(Fig.3D).Asregardsthe gamma3waves,nosignificanteffectswereobserved(Fig.3D).

4. Discussion

Theelectrophysiologicalvalidation ofthetranslationalutility ofthisnewratsubstrainrevealedthattheseanimalsshowedsev- eralneurophysiologicalabnormalitiesobservedinschizophrenia.

ThelatencyoftheP2peakswasprolonged,andtheamplitudeof N1responsedecreased;however,thegatingwasnotimpairedin theseanimalsinthedouble-clickparadigm.Furthermore,theacute treatmentwithasubanesthetic doseof ketaminedidnotresult in signiﬁcantalterations in ERPparameters. The newsubstrain showedincreasedpowerofoscillationsinthetheta,alphaandbeta ranges,whiledecreasedpowerwasdetectedindeltaandgamma2 bands compared withthecontrol animals.Ketamine treatment increasedthegamma1bandpowerinbothgroups,whileitcaused signiﬁcantchangesonlyinthenewsubstraininthetotalpower andinalpha,betaandgamma2bands,suggestingtheenhanced sensitivityforthisdrug.

Paired-click paradigm is a standard method used to assess sensorygating[43].ERPinhumanstudieshasapositivedeflec- tionoccurringapproximately50msfollowingtheonsetofsensory stimulation(P50),whichisgeneratedprimarilyintheauditorytha- lamusandtemporalcortex[44,45].TheN100component,alarge negativedeflection,occursfollowingtheP50responseoriginated mainlyfromtheprimaryauditorycortex[46].Thesecondposi- tivedeflectionthatemergesapproximately200msaftersensory stimulation(P200)isgeneratedbytheassociationcortexreflect- inghigher-orderintegrationandinterpretationofsensorystimuli.

TheERPwaveformsobtainedinrodentsshowverysimilarchar- acteristicstohumanoneswiththeexceptionthatthelatenciesof therodentERParesignificantlyshorter[4].Thus,ERPsinrodent typicallyshowapositivedeflectionbetween10and 30ms(P1), anegativedeflectionbetween30and 50ms(N1),and asecond positivedeflectionbetween50and100ms(P2)(Fig.1).

ReducedpeakamplitudesoftheauditoryERParewellrepli- catedinschizophrenicpatients[43,45,46],andhavebeenobserved in multiple relevant rodent models [19], but no changes in

(5)

Fig.3. BasalEEGpowerindifferentfrequencybands.

Itwasrecordedfromratparietalcortexinnaiveandnewsubstrainratsaftersalineorketaminetreatment.(A)Totalpower;(B)deltaandthetabands;(C)alphaandbeta bands;(D)gamma1-3bands.*Signssigniﬁcantdifferencescomparedtonaivegroup.Thesymbol#denotessigniﬁcantdifferencesbetweenacutesalineandketamine treatments.

theseparameterswerealsoreported[44,46].Ketamineexposure decreasedtheamplitudeofERPinseveralhumanandanimalstud- ies[12,21,34,46,47],but contradictory resultsare alsoavailable [48]; the latteris in agreement withourresults. It seemsthat theP50 responseshavelimitedutilitiesasaclinicalorresearch tool;however, reductionin N100amplitudeis widelyreported inschizophreniaasanendophenotypewithastrongheritability, representingdeﬁcitsininitialsensoryprocessingandearlyatten- tion[46].Therefore,thechangeinN1peakamplitudeinrodents may bea potentialbiomarker for schizophrenia that hasbeen detectedindisease-relevanttransgenicmiceandalsoinducedby acuteketaminetreatment[43,46,48].ThedecreasedN1response, observedinoursubstrain,isinagreementwiththesestudies,which suggeststhat this substrainmaysimulateschizophrenia in this respect.AmplitudeoftheP200/P2isreducedinschizophrenia,after acuteexposuretoketamineinhealthycontrolsandrodents,however,wedidnotdetectitinoursubstrain,whichmightbedueto thedifferencesintheapplieddoseorthestrain[46].

EEG recordings in healthy humans exhibit habituation to repeated stimuli; thus the amplitude of the auditory evoked potentialismarkedlyattenuateduponthesecondclickstimulus comparedwiththeﬁrstone[5,44,45].Thesensorygatingparadigm hasbeenfrequentlyusedtostudyneurophysiologicalprocesses inschizophrenia,however,ERP-basedsensorygatingﬁndingsin

this diseasearesomehowdiverse;severalstudiesshowdeﬁcits [4,45,46,49,50],whilesomearenegative,asitwasfoundin the newsubstrain[51–53].Theineffectivityofbothacuteandchronic ketaminetreatmentontheERPgating,inagreementwithourdata, indicatesthatNMDAreceptorsmaynotbecriticallyinvolvedin itsgeneration[12,47,54–56].Altogether,thealterationsobserved intheERPsinoursubstrainshowedlimitedcorrelationwiththe humanschizophrenicdata(decreasedamplitudeofN1).

Several neural oscillatory abnormalities have been demon- strated in schizophrenia that may contribute to the abnormal sensory and cognitive performance [1,3,7–9,46,57,58]. Neural oscillationsdependonthekineticsofinhibitory(GABAergic)and excitatory(glutamatergic)synapticinteractions,andtheineffec- tiveinhibitory controlof sensoryprocessing ischaracteristic in this disease [3,16,17,47,59–61]. Because of the prominent role of gamma-bandactivityin cognitionduringnormal brainfunc- tioning,therehasbeenaparticularfocusontheinvestigationof high-frequencyactivityinpatientpopulations[46].Fast-spiking, parvalbumin-positiveGABAergicneuronsplayapivotalroleinthe primarygenerationofhigh-frequencyoscillationsandtheirsyn- chronization, whereas glutamatergic pyramidalneurons appear tocontroltheirstrength, duration,andlong-rangesynchroniza- tionactingprimarilyviaNMDARs[3,18,32,46,59,61].Manyhuman studiesobservedreducedgammaoscillatoryactivity,whichmay

(6)

reflect the deficits in cognitive and sensory processing related to negative symptoms in schizophrenia [3,47,57,62,63]. How- ever,therearecontroversialfindingofincreasedgammaactivities in schizophrenia as well,and it is reportedly relevant to posi- tivesymptoms(hallucination, delusion)[9,25,64–67].Abnormal gammaactivityhasbeenreportedinnumerousanimalmodelsof schizophrenia,too,e.g.,silencingoftheGABAergicinterneurons andhypofunctionofNMDARsignalingisaccompaniedbyaltered oscillatorypowerparticularlyinthegammarange[17,32,46,68].

ChronicNMDARantagonisttreatmentmaycausedecreasedoscil- latorypowerseveralweeksormonthsafterthecessationofthe treatment,suggestinglong-lastingconsequencesofsuchaninter- vention[12,21,69].Thus,thedecreasedgammapowerbetween52 and70Hz inournewsubstrainmightbedue,atleastpartially, tochronicketaminetreatment.Regardingtheacutedrugeffect, inagreementwithourresults,signiﬁcantlyelevatedgammaband oscillationshavebeenobservedinbothhumanandanimalstudies, reﬂectingacorticalhyperglutamatergicstatethroughGABAergic disinhibition,leadingtoamild shiftintheexcitation/inhibition balancetowardexcitation[12,16,47,58,70–73].

Betaoscillations,lessexploredinschizophrenia,arebelieved tobegeneratedbroadlyacrossmultipleneocorticalstructuresand areinvolvedintheadaptationtorepetitivesensorystimuli,atten- tion,andsynchronizationoflargeensemblesofneurons[3,46].In agreementwithresultsobservedinthenewsubstrain,betaband powerincreases in patientswithschizophrenia, which maybe duetoglobalcorticalhyperexcitabilityorattentiondisturbances observedinthesepatients[74].Asregardsthealphabandoscilla- tion,itisrelatedprimarilytothethalamus;thus,thealterationsin thisfrequencybandmaysuggestdysfunctionoftheinhibitorytha- lamicneurons[74,75].Differentlaboratorieshavereportedeither higheralphapowerassociatedwithnegativesymptoms[46,66,74]

or reduced alpha bandpower in a phase-independent manner [46,74],thustheenhancedpowerobservedinournewsubstrain mightberelatedtothenegativesymptomsofschizophrenia.

Abnormalitiesinlowerfrequencyoscillations(deltaandtheta) arealsoprominentin thisdisease[7,8,76,77].Thecorticaldelta band oscillation originates from the reticular nucleus of the thalamus, where predominantly parvalbumin-positive GABAer- gicneuronsarepresent[78].Theycanbepartiallybuttonically activatedviaNMDARs, thereby regulatingtheactivityof thala- micrelayneuronsprojectingtotheprefrontalcortex.Thetheta frequency range is associated withcognition/memory function, where cortico-hippocampal circuits are key generators of the rhythm[59].Bothdeltaandthetabandchanges dependonthe phaseofschizophrenia,i.e.,patientswithpositivesignsshowno- changesordecreaseintheseparameters,whileinnegativephase of schizophrenia increases were detected [25,65,66,74,77]. Our modelshoweddecreaseddeltaactivity,whichmightsimulatethe positivephaseofschizophrenia,whiletheenhancedthetapower canindicatethenegativeone.In contrasttoourﬁndings,most studiesshowsigniﬁcantlyreducedlow-frequencyoscillationsafter acuteketaminetreatment[12,16,47,58,69–73],andthisdiscrep- ancymightbeexplainedbythedifferencesintheapplieddoses.

Asany rodent model of a complex humanneuropsychiatric disorder, our model has a number of shortcomings. The het- erogeneity of this disease and the overlap in several aspects withotherneuropsychiatric diseases,especially autism,further complicatestheabilitytodiscernthespecificityofagiven pre- clinical model [60,79,80]. Although autism and schizophrenia are clearly distinct disorders, they share a significant number ofcommonclinical characteristics,includinggenetics, epidemi- ology (e.g., prenatal infection, maternal stress, and perinatal hypoxia), behavioral phenotypes (e.g., impairments in social and cognitive behaviors, communicative function, and stereo- typedbehaviors), neuroimagingand neurophysiologicalfindings

(e.g.,interneuron dysfunctionor disrupted excitation/inhibition balance)[1,60,79,81,82].Furthermore,hypofunctionofN-methyl- d-aspartate receptor (NMDAR) signaling contributes to the pathophysiology of both diseases [83–85]. NMDAR1 hypomor- phicmicedisplay both schizophrenia- and autism-like changes in social and cognitivebehaviorsand in theoscillatoryactivity [13,18,32–34].Mostofthebehavioralalterationsobservedinthis newsubstrain can alsobe detectedin autism [81,86–88],sim- ilarly, the observed electrophysiological changes in our model mightsimulateseveralalterationsdetectedinboththeautismand schizophreniaand maycontributetotheabnormalsensoryand cognitiveperformance [1,3,7–9,46,57,58]. Especially,thesigniﬁ- cantlyenhancedlatencyintheP2responsesinthenewsubstrain correlateswithhumanstudiesinvolvingautisticpatientsandits animalmodelswithoutinﬂuencingpeakamplitude[18,85,89–91].

Since most of the abnormalities overlap in these two disorders, thesealterations regarding theEEGoscillationsshouldbe accompaniedbyothermorespeciﬁcsignsforrelevantdiagnosis [1,17,18,32,46,60,68,74,82,85,90,92,93].

5. Conclusion

Our substrain wasoriginallydeveloped asa complex model ofschizophrenia,andhasbeenextensivelyinvestigatedassuch;

however,theresultsindicated that theseratsexhibited several autism-likebehavioral andneurophysiologicalphenotypicalter- ations. It also highlights the challenge of modeling a complex humanbehavioraldisorderinrodents,sinceasitwasmentioned above,mostofthesebiomarkersarenon-speciﬁctothesediseases.

Itcanbeconcludedthatthissubstrainproduceslong-lastingalter- ationsonERPandEEGoscillationsafterjuvenilesocialisolation andsubchronicketaminetreatment.Theseresultsarepartiallyin agreementwithclinicaldata,whichsuggeststhatthismodelpro- videsalimitedrepresentationofdisturbancesobservedinEEGof schizophrenicand/orautisticpatients.Althoughthestrengthsand weaknessesofthis model shouldbeevaluatedin thefutureby molecularbiologicalmethods, too,we concludethatourmodel mayprovideadditionalopportunityforthetranslationalresearch oftheseneuropsychiatricdisorders.

Acknowledgements

FundingforthisstudywasprovidedbytheHungarianResearch Grant(OTKA,K83810),TÁMOP-4.2.2.B-15/1/KONV-2015-0006and Hungarian Brain Research Program Grant KTIA13NAP-A-I/15.

Thesegrantshadnofurtherroleinstudydesign,incollection,anal- ysisandinterpretationofdata,inthewritingofthereport,andin thedecisiontosubmitthepaperforpublication.

TheauthorswishtothankRobertAverkinandTamásNagypál fortheirparticipationinelectrophysiologicalsetupandrecordings, AgnesTandariforherexcellenttechnicalassistanceandaregrateful toCsillaKeresztesforthelinguisticreviewofthemanuscript.

References

[1]D.C.Rojas,L.B.Wilson,Gamma-bandabnormalitiesasmarkersofautism spectrumdisorders,Biomark.Med.8(2014)353–368.

[2]A.H.C.Wong,H.H.M.VanTol,Schizophrenia:fromphenomenologyto neurobiology,Neurosci.Biobehav.Rev.27(2003)269–306.

[3]P.J.Uhlhaas,W.Singer,High-frequencyoscillationsandtheneurobiologyof schizophrenia,DialoguesClin.Neurosci.15(2013)301–313.

[4]L.E.Adler,E.Pachtman,R.D.Franks,M.Pecevich,M.C.Waldo,R.Freedman, Neurophysiologicalevidenceforadefectinneuronalmechanismsinvolvedin sensorygatinginschizophrenia,Biol.Psychiatry17(1982)639–654.

[5]D.L.Braff,M.A.Geyer,Sensorimotorgatingandschizophrenia:humanand animalmodelstudies,Arch.Gen.Psychiatry47(1990)181–188.

[6]D.L.Braff,M.A.Geyer,N.R.Swerdlow,Humanstudiesofprepulseinhibitionof startle:normalsubjects,patientgroups,andpharmacologicalstudies, Psychopharmacology(Berl)156(2001)234–258.

(7)

[7]S.R.Sponheim,B.A.Clementz,W.G.Iacono,M.Beiser,RestingEEGin ﬁrst-episodeandchronicschizophrenia,Psychophysiology31(1994)37–43.

[8]S.R.Sponheim,W.G.Iacono,P.D.Thuras,S.M.Nugent,M.Beiser,Sensitivity andspeciﬁcityofselectbiologicalindicesincharacterizingpsychoticpatients andtheirrelatives,Schizophr.Res.63(2003)27–38.

[9]E.Basar,Brainoscillationsinneuropsychiatricdisease,DialoguesClin.

Neurosci.15(2013)291–300.

[10]J.Chen,B.K.Lipska,D.R.Weinberger,Geneticmousemodelsofschizophrenia:

fromhypothesis-basedtosusceptibilitygene-basedmodels,Biol.Psychiatry 59(2006)1180–1188.

[11]A.Meyer-Lindenberg,Frommapstomechanismsthroughneuroimagingof schizophrenia,Nature468(2010)194–202.

[12]E.M.Sullivan,P.Timi,L.E.Hong,P.O’Donnell,Reversetranslationofclinical electrophysiologicalbiomarkersinbehavingrodentsunderacuteandchronic NMDAreceptorantagonism,Neuropsychopharmacology40(2015)719–727.

[13]S.Bickel,H.P.Lipp,D.Umbricht,Earlyauditorysensoryprocesssingdeﬁcitsin mousemutantswithreducedNMDAreceptorfunction,

Neuropsychopharmacology33(2008)1680–1689.

[14]B.V.Broberg,B.Oranje,B.Y.Glenthoj,K.Fejgin,N.Plath,J.F.Bastlund, Assessmentofauditorysensoryprocessinginaneurodevelopmentalanimal modelofschizophrenia-gatingofauditory-evokedpotentialsandprepulse inhibition,Behav.BrainRes.213(2010)142–147.

[15]D.W.N.Dissanayake,M.Zachariou,C.A.Marsden,R.Mason,Effectsof phencyclidineonauditorygatingintherathippocampusandthemedial prefrontalcortex,BrainRes.1298(2009)153–160.

[16]R.S.Ehrlichman,M.J.Gandal,C.R.Maxwell,M.T.Lazarewicz,L.H.Finkel,D.

Contreras,etal.,N-Methyl-d-asparticacidreceptorantagonist-induced frequencyoscillationsinmicerecreatepatternofelectrophysiologicaldeﬁcits inschizophrenia,Neuroscience158(2009)705–712.

[17]V.S.Sohal,F.Zhang,O.Yizhar,K.Deisseroth,Parvalbuminneuronsandgamma rhythmsenhancecorticalcircuitperformance,Nature459(2009)698–702.

[18]M.J.Gandal,J.Sisti,K.Klook,P.I.Ortinski,V.Leitman,Y.Liang,etal., GABAB-mediatedrescueofalteredexcitatory-inhibitorybalance,gamma synchronyandbehavioraldeﬁcitsfollowingconstitutive

NMDAR-hypofunction,Transl.Psychiatry2(2012)e142.

[19]M.T.Lazarewicz,R.S.Ehrlichman,C.R.Maxwell,M.J.Gandal,L.H.Finkel,S.J.

Siegel,Ketaminemodulatesthetaandgammaoscillations,J.Cogn.Neurosci.

22(2010)1452–1464.

[20]J.A.Saunders,M.J.Gandal,S.J.Siegel,NMDAantagonistsrecreate signal-to-noiseratioandtimingperturbationspresentinschizophrenia, Neurobiol.Dis.46(2012)93–100.

[21]R.E.Featherstone,L.R.Nagy,C.G.Hahn,S.J.Siegel,Juvenileexposureto ketaminecausesdelayedemergenceofEEGabnormalitiesduringadulthood inmice,DrugAlcoholDepend.134(2014)123–127.

[22]Y.Tomimatsu,R.Hibino,H.Ohta,BrownNorwayrats,aputative

schizophreniamodel,showincreasedelectroencephalographicactivityatrest anddecreasedevent-relatedpotentialamplitude,power,andcoherencein theauditorysensorygatingparadigm,Schizophr.Res.166(2015)171–177.

[23]A.Ahnaou,S.Nayak,A.Heylen,D.Ashton,DrinkenburgWHIM.SleepandEEG proﬁleinneonatalhippocampallesionmodelofschizophrenia,Physiol.

Behav.92(2007)461–467.

[24]J.H.Krystal,L.P.Karper,J.P.Seibyl,G.K.Freeman,R.Delaney,D.Bremner, SubanestheticeffectsofthenoncompetitiveNMDAantagonist,ketamine,in humans:psychotomimetic,perceptual,cognitive,andneuroendocrine responses,Arch.Gen.Psychiatry51(1994)199–214.

[25]D.C.Javitt,K.M.Spencer,G.K.Thaker,G.Winterer,M.Hajos,

Neurophysiologicalbiomarkersfordrugdevelopmentinschizophrenia,Nat.

Rev.DrugDiscov.7(2008)68–83.

[26]J.D.Jentsch,R.H.Roth,Theneuropsychopharmacologyofphencyclidine:from NMDAreceptorhypofunctiontothedopaminehypothesisofschizophrenia, Neuropsychopharmacology20(1999)201–225.

[27]A.C.Lahti,M.A.Weiler,B.A.TamaraMichaelidis,A.Parwani,C.A.Tamminga, Effectsofketamineinnormalandschizophrenicvolunteers,

[28]J.M.Stone,P.D.Morrison,L.S.Pilowsky,Review:glutamateanddopamine dysregulationinschizophreniaasynthesisandselectivereview,J.

Psychopharmacol.21(2007)440–452.

[29]A.K.Malhotra,D.A.Pinals,C.M.Adler,I.Elman,A.Clifton,D.Pickar, Ketamine-inducedexacerbationofpsychoticsymptomsandcognitive impairmentinneuroleptic-freeschizophrenics,Neuropsychopharmacology 17(1997)141–150.

[30]A.C.Lahti,H.H.Holcomb,D.R.Medoff,C.A.Tamminga,Ketamineactivates psychosisandalterslimbicbloodﬂowinschizophrenia,Neuroreport6(1995) 869–872.

[31]N.R.Swerdlow,M.A.Geyer,J.M.Shoemaker,G.A.Light,D.L.Braff,K.E.Stevens, etal.,Convergenceanddivergenceintheneurochemicalregulationof prepulseinhibitionofstartleandN40suppressioninrats,

[32]V.M.Tatard-Leitman,C.R.Jutzeler,J.Suh,J.A.Saunders,E.N.Billingslea,S.

Morita,etal.,PyramidalcellselectiveablationofN-methyl-d-aspartate receptor1causesincreaseincellularandnetworkexcitability,Biol.

Psychiatry77(2015)556–568.

[33]G.E.Duncan,S.S.Moy,J.A.Lieberman,B.H.Koller,Typicalandatypical antipsychoticdrugeffectsonlocomotorhyperactivityanddeﬁcitsin sensorimotorgatinginageneticmodelofNMDAreceptorhypofunction, Pharmacol.Biochem.Behav.85(2006)481–491.

[34]L.C.Amann,T.B.Halene,R.S.Ehrlichman,S.N.Luminais,N.Ma,T.Abel,etal., Chronicketamineimpairsfearconditioningandproduceslong-lasting reductionsinauditoryevokedpotentials,Neurobiol.Dis.35(2009)311–317.

[35]Z.Petrovszki,G.Adam,G.Tuboly,G.Kekesi,G.Benedek,S.Keri,etal., Characterizationofgene-environmentinteractionsbybehavioralproﬁlingof selectivelybredrats:theeffectofNMDAreceptorinhibitionandsocial isolation,Behav.BrainRes.240(2013)134–145.

[36]G.Kekesi,Z.Petrovszki,G.Benedek,G.Horvath,Sex-speciﬁcalterationsin behavioralandcognitivefunctionsinathreehitanimalmodelof schizophrenia,Behav.BrainRes.284(2015)85–93.

[37]M.Hadamitzky,S.Harich,M.Koch,K.Schwabe,Deﬁcientprepulseinhibition inducedbyselectivebreedingofratscanberestoredbythedopamineD2 antagonisthaloperidol,Behav.BrainRes.177(2007)364–367.

[38]G.Liebsch,A.Montkowski,F.Holsboer,R.Landgraf,Behaviouralproﬁlesof twoWistarratlinesselectivelybredforhighorlowanxiety-related behaviour,Behav.BrainRes.94(1998)301–310.

[39]K.Schwabe,F.Freudenberg,M.Koch,Selectivebreedingofreduced sensorimotorgatinginWistarrats,Behav.Genet.37(2007)706–712.

[40]R.J.Racine,M.Steingart,D.C.McIntyre,Developmentofkindling-proneand kindling-resistantrats:selectivebreedingandelectrophysiologicalstudies, EpilepsyRes.35(1999)183–195.

[41]G.Tuboly,G.Benedek,G.Horvath,Selectivedisturbanceofpainsensitivity aftersocialisolation,Physiol.Behav.96(2009)18–22.

[42]G.Paxinos,C.Watson,TheRatBraininStereotaxicCoordinates,5ed.,Elsevier AcademicPress,Burlington,2005.

[43]K.Gjini,S.Burroughs,N.N.Boutros,Relevanceofattentioninauditorysensory gatingparadigmsinschizophreniaApilotstudy,J.Psychophysiol.25(2011) 60–66.

[44]L.Witten,B.Oranje,A.Mork,B.Steiniger-Brach,B.Y.Glenthoj,J.F.astlund, Auditorysensoryprocessingdeﬁcitsinsensorygatingandmismatch negativity-likeresponsesinthesocialisolationratmodelofschizophrenia, Behav.BrainRes.266(2014)85–93.

[45]J.Smucny,A.Olincy,L.C.Eichman,E.Lyons,J.R.Tregellas,Earlysensory processingdeﬁcitspredictsensitivitytodistractioninschizophrenia, Schizophr.Res.147(2013)196–200.

[46]R.E.Featherstone,M.F.McMullen,K.R.Ward,J.Bang,J.Xiao,S.J.Siegel,EEG biomarkersoftargetengagement,therapeuticeffect,anddiseaseprocess, Ann.N.Y.Acad.Sci.1344(2015)12–26.

[47]L.E.Hong,A.Summerfelt,R.W.Buchanan,P.O’Donnell,G.K.Thaker,M.A.

Weiler,etal.,Gammaanddeltaneuraloscillationsandassociationwith clinicalsymptomsundersubanestheticketamine,Neuropsychopharmacology 35(2010)632–640.

[48]C.R.Maxwell,R.S.Ehrlichman,Y.L.Liang,D.Trief,S.J.Kanes,J.Karp,etal., Ketamineproduceslastingdisruptionsinencodingofsensorystimuli,J.

Pharmacol.Exp.Ther.316(2006)315–324.

[49]C.Siegel,M.Waldo,G.Mizner,L.E.Adler,R.Freedman,Deﬁcitsinsensory gatinginschizophrenicpatientsandtheirrelatives:evidenceobtainedwith auditoryevokedresponses,Arch.Gen.Psychiatry41(1984)607–612.

[50]R.Freedman,L.E.Adler,G.A.Gerhardt,M.Waldo,N.Baker,G.M.Rose,etal., Neurobiologicalstudiesofsensorygatinginschizophrenia,Schizophr.Bull.13 (1987)669–678.

[51]O.M.deWilde,L.J.Bour,P.M.Dingemans,J.H.Koelman,D.H.Linszen,Failure tofindP50suppressiondeficitsinyoungfirst-episodepatientswith schizophreniaandclinicallyunaffectedsiblings,Schizophr.Bull.33(2007) 1319–1323.

[52]J.L.Santos,E.M.Sanchez-Morla,A.Aparicio,M.A.Garcia-Jimenez,C.

Villanueva,V.Martinez-Vizcaino,etal.,P50gatingindeﬁcitandnondeﬁcit schizophrenia,Schizophr.Res.119(2010)183–190.

[53]C.M.Yee,T.J.Williams,P.M.White,K.H.Nuechterlein,D.Ames,K.L.Subotnik, AttentionalmodulationoftheP50suppressiondeﬁcitinrecent-onsetand chronicschizoprenia,J.Abnorm.Psychol.119(2010)31–39.

[54]B.Oranje,C.C.Gispen-deWied,M.N.Verbaten,R.S.Kahn,Modulatingsensory gatinginhealthyvolunteers:theeffectsofketamineandhaloperidol,Biol.

Psychiatry52(2002)887–895.

[55]B.N.vanBerckel,B.Oranje,J.M.VanRee,M.N.Verbaten,R.S.Kahn,Theeffects oflowdoseketamineonsensorygating,neuroendocrinesecretionand behaviorinhealthyhumansubjects,Psychopharmacology(Berl)137(1998) 271–281.

[56]N.M.W.J.deBruin,B.A.Ellenbroek,A.R.Cools,A.M.L.Coenen,E.L.J.M.van Luijtelaar,Differentialeffectsofketamineongatingofauditoryevoked potentialsandprepulseinhibitioninrats,Psychopharmacology(Berl)142 (1999)9–17.

[57]J.S.Kwon,B.F.O’Donnell,G.V.Wallenstein,R.W.Greene,Y.Hirayasu,P.G.

Nestor,Gammafrequency-rangeabnormalitiestoauditorystimulationin schizophrenia,Arch.Gen.Psychiatry56(1999)1001–1005.

[58]K.M.Spencer,P.G.Nestor,R.Perlmutter,M.A.Niznikiewicz,M.C.Klump,M.

Frumin,Neuralsynchronyindexesdisorderedperceptionandcognitionin schizophrenia,Proc.Natl.Acad.Sci.U.S.A.101(2004)17288–17293.

[59]G.Buzsaki,A.Draguhn,Neuronaloscillationsincorticalnetworks,Science 304(2004)1926–1929.

[60]E.V.Orekhova,T.A.Stroganova,A.O.Prokofyev,G.Nygren,C.Gillberg,M.

Elam,Sensorygatinginyoungchildrenwithautism:relationtoage,IQ,and EEGgammaoscillations,Neurosci.Lett.434(2008)218–223.

[61]R.D.Traub,M.A.Whittington,E.H.Buhl,J.G.Jefferys,H.J.Faulkner,Onthe mechanismofthegamma−>betafrequencyshiftinneuronaloscillations