Journal of Ni’,,,orlie;i,,.snvI.ippi ncoti—-Ras en I’ub!isher~.Phitadetpilia

1997 tnteniationa! Society ir Neuroclieniistry

NMDA Receptor—MediatedRegulation of AMPA ReceptorProperties in Organotypic Hippocampal Slice Cultures

David M. Gellerman,Xiaoning Bi, and Michel Baudry

NeuroscienceProgratn, Unir’ersitv ~?tSouthernCalifornia, Los Angeles,California, U.S.A.

Abstract: Activation of the calcium-dependent proteasecalpain has been proposed to be a necessary step inthe formation of long-term potentiation (LTP) in the hip-pocampus, and stimulation of N-methyl-D-aspartate(NMDA) receptors leads to an increase in intracellularcalcium concentration, calpain activation, proteolysis ofcytoskeletal elements, and modification of ci-amino-3-hy-droxy-5-methyl-4-isoxazole propionic acid (AMPA) re-ceptor properties. In the present study, we evaluated theeffects of NMDA treatment of cultured hippocampalslices on the properties of AMPA receptors. Cultured hip-pocampal slices were treated with NMDA (100 pM) for15 mm and [3H]AMPAbinding to membrane fractionswas measured. NMDA-treated slices exhibited an in-crease in both “high-affinity” and “low-affinity” [3H]-AMPA binding, with smaller changes in 6-cyano-7-nitro[3H]quinoxaline-2,3-dione binding. The increase in[3HIAMPAbinding was significantly reduced by preincu-bation of cultures with calpain inhibitor orcalpeptin (100pM). Furthermore, NMDA exposure decreased the num-ber of GIuR1 subunits of AMPA receptors detected by anantibody against the C-terminal domain of the subunit inwestern blots and resulted in the formation of a lowermolecular weight species detected by an antibodyagainst the N-terminal domain. Both effects were com-pletely prevented by calpain inhibitors. These results indi-cate that NMDA receptor activation produces calpain ac-tivation and complex modifications of AMPA receptorproperties, which could be involved in NMDA receptor—mediated changes in synaptic efficacy. Key Words: Hip-pocampus — Calpain — Receptor — Plasticity — Gluta-mate—Calcium.J. Neurochem. 69, 131—136 (1997).

Long-termpotentiation (LTP) is a long-lasting in-creasein synaptictransmissionthathasbeensuggestedto representa cellular mechanismof learning andmemory (for review, see Bliss and Collingridge,1993). In CAl anddentategyrusof thehippocampus,LTP induction is dependenton N-methyl-D-aspartate(NMDA) receptorstimulation that leads to an influxof calcium ions and the activation of a cascadeofbiochemicalevents(Lisman, 1989; BaudryandMassi-cotte, 1992; Artola and Singer, 1993; Bliss and Col-

lingridge, 1993; Bear andMalenka, 1994; Maren andBaudry, 1995). WhereasLTP induction requiresacti-vation of NMDA receptors.LTP expressionandmain-tenancearepresumedto be duein partto modificationsof r-arnino-3-hydroxy-5-methyl-4-isoxazolepropionicacid (AMPA) receptors. In particular, the waveformof extracellularly recordedexcitatorypostsynapticpo-tentials (EPSPs)is modified in a way bestinterpretedby a change in AMPA receptor/channelkinetics(Ambros-Ingersonet al., 1991;Ambros-IngersonandLynch, 1993).Theeffectson EPSPsof thechaotropicion, perchlorate,which enhanceagonistaffinity for theAMPA receptors.arealso decreasedafter LTP induc-tion (Shahi and Baudry, 1992): in a similar manner,theenhancingeffectsof aniracetamon AMPA recep-tor—mediatedEPSPsaredecreasedafterLTP induction(Staubli et al., 1990). Finally, the recent discoverythata significant fraction of synapsesare functionallysilent due to the lack of functional AMPA receptorsclearly supports that LTP could be due to the un-maskingor ‘‘activation” of~’’silent’’ AMPA receptors(Isaacet al., 1995; Liao ci al., 1995). Consistentwiththis idea,Maren et al. (1993)showedthat [1H]AMPAbinding is increasedin themolecularlayerof theden-tate gyrus after LTP induction and that the increasedbinding correlatespositively with the increasein syn-aptic response(for a different result, however, seeKessleret al., 1991 ).

ReceivedJanuary7. 1997: revised manuscript receivedFebruary20, 1997, acceptedFebruary20, 1997.

Addresscorrespondenceand reprint requeststo Dr. M. Baudry atHNB 309. University ot Southern California. Los AngeIe~,CA90089-2520,U.S.A.

Preliminary results of this siLidy were presentedat the Society forNeuroscienceMeeting, 1995, San Diego. CA. U.S.A.

Abbreviationsused:C—Ab and N-Ah. antibodyagainstC- andN-terminal domains, respectively.al GluR I subunits of AMPA re-ceptors: AMPA, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionicacid: CNQX. 6-cyano-7-nitroquinoxaline-2.3-dione:DMSO. di-methyl sulfoxide: EPSP.excitatory postsynapticpotential: KSCN.potassiumthiocyanate: LTP. long-term potcntiation: MEM. mini-

niulli essentialmedium: NMDA. N—rnethyl—o—aspartate:PAGE,poly-acrylamidegel electrophoresis:.SBDP. spectrin breakdownproduct:SDS. sodium dodecyl sulfate:TBS. Tris-huffercd saline.

131

132 D. M. GELLERMANET AL.

One of the processesthat appearsto play an im-portantrole in the regulation of AMPA receptorsinsynapticplasticity is proteolysismediatedby thecal-cium-dependentproteasecalpain (Lynch and Baudry,1984). Specifically, it hasbeenproposedthat calciuminflux through theNMDA receptorchannelactivatescalpain, which cleavescytoskeletalproteins,resultingin theredistributionandmodificationof AMPA recep-tars in the postsynapticmembrane.Calpain activityis necessaryfor LTP, as calpain inhibitors, such asleupeptin andcalpain inhibitor I, havebeenshowntopreventLTP formation (Oliver et al., 1989; del Cerroet al., 1990; Dennyet a!., 1990). In addition, NMDAreceptorstimulation in both acuteandorganotypiccul-tures of hippocampalslicesby either LTP induction(specifically, thetaburst stimulation) or a 15-mm ex-posureto NMDA resultsin theaccumulationof aspe-cific spectrin breakdownproduct (SBDP) generatedby calpaincleavageof spectrin (Seubertet al., 1988;delCerro eta!., 1994;Vanderklishetal., 1995). SBDPaccumulationwas blockedby theNMDA receptoran-tagonist (— ) -2-amino-5-phosphonovalerateand didnot occurwhenneuronsweredepolarizedby potassiumchloride, thus demonstratingan NMDA receptor—de-pendent activation of calpain. Other processes,suchasphosphorylationreactionsmediatedby a variety ofproteinkinases,havealsobeenproposedto participatein synapticplasticity (Raymondet al., 1993), and, inparticular, NMDA exposurehas been shown to pro-duceanincreasein AMPA receptorfunctionas aresultof subunit phosphorylationmediatedby calciumlcal-modulin kinaseII (McGlade-McCullohetal., 1993).

Recently,we identifiedadirect link betweencalpainactivationandAMPA receptormodifications.First, wefound that preincubation of frozen-thawedthin (10

Urn) brainsectionswith calcium resultedin an increasein [3HIAMPA binding to AMPA receptors,an effectpartially blocked by a calpain inhibitor, leupeptin(Tocco et al., 1992). Moreover, the same treatmentproducedpronouncedmodificationsin theimmunolog-ical propertiesof thesubunitsof AMPA receptors(Biet a!., 1994). Using antibodiesdirectedagainsttheC-terminal as well as the N-terminal domainsof GluRlsubunits,we demonstratedthat calpainactivationpro-ducesthe partial proteolysisof GluRl subunits, re-sulting in the formation of a new specieswith a lowermolecularweight (Bi et al., 1996, 1997).

The goal of thepresentstudy was to determinetheeffects of NMDA receptor—mediatedcalpain activa-tion on thepropertiesof AMPA receptorsby using along-term organotypichippocarnpalslicecultureprep-aration. We found that brief activation of NMDA re-ceptors resulted in increased[3H1AMPA binding tosynapticmembranesandin the formation of a lowermolecularweight speciesof theGluRl subunit. Botheffectswere preventedby theproteaseinhibitors cal-pain inhibitor I andcalpeptin,suggestingthattheprop-ertiesof AMPA receptorscan be rapidly modified by

calpain-mediatedproteolysisasaresultof NMDA re-ceptorstimulation.

MATERIALS AND METHODS

Organotypic cultures of hippocampal slicesOrganotypiccrtlturesof hippocampalsliceswereprepared

accordingto themethoddescribedby Stoppini eta].(1991).In brief, hippocampiwere harvestedin sterile conditionsfrom 6—8-day-old Sprague—Dawleyrat brains in chilledminimum essentialmedium (MEM) (GIBCO no. 61 100-061) containing(in mM) 1-IEPES (25), Tris-base(10), n-glucose(10), andMgCI

2 (3) and placedon a Teflon stageof a Mcllwain tissuechopper.Transverseslices(400 pm)were cut andtransferredto a humidified membraneinsert(Millicell-CM, Millipore, Bedford, MA, U.S.A.; 0.4 mmpore size), which wasplaced in a deep-wellplate with 1ml of MEM (GIBCO no. 41200-072)containing(in mM)glutamine(3), HEPES(30). NaHCO3(5), n-glucose(30).t-ascorbate(0.5), CaCI2 (2). MgSO4 (2.5), 1 pg insulin.20% horseserum,including penicillin, pH 7.2. In this man-ner, explantswere supportedby the membraneat an air—medium interfaceand kept in an incubatorat 35°Cwith a5% C02-enrichedatmosphere.Medium waschangedtwiceperweek.

NMDA exposureand crude synaptic membranepreparation

Cultureswere incubatedwith high-calciumexposureme-dium (MEM containing3 mM glutaniine, 25 mM HEPES.5mM NaHCO3,25 mM dextrose.0.5 mMascorbate,4mMCaCI2,1.22 mM MgSO4, I pg/ml of insulin, with penicillin,pH 7.2) overnight before NMDA treatment, as this wasshown to increase calpain activation (Vanderklish et al..1995). Calpain inhibitors [calpain inhibitor I or calpeptin.100 pM, in 0.05% dimethyl sulfoxide(DMSO) j or vehicle(0.05% DMSO) were addedat least 3 h before NMDAexposure.Initially, calpain inhibitor I or calpeptin(100p.Min 0.05% DMSO) was incubatedovernight (16 h) beforeNMDA treatment,hut later experimentsdeterminedthat a3-h incubationwassufficientto preventtheeffect of NMDAexposure.NMDA (100 1IM in 0.9% saline) was addedtothe culturemediumfor 15 mm (Vanderklishet al., 1995).after which sliceswere removedfrom the membranesandplaced in ice-cold homogenizationbuffer (20% sucroseinphosphate-bufferedsalinecontaining3 mM EGTA) andthetissue was homogenizedby sonication(in general, four orfive sliceswerecombinedtogetherto generateenoLtghtissrteto perform binding andwesternblot experiments).Theho-mogenatewascentrifugedat 14,000rpm, 4°C,for 20 mm.The supernatantwas discatdedand the pellet resuspendedin 100 pM EGTA in distilled waterand centrifugedagainasabove.This secondpelletwasresuspendedin Tris-acetatebuffer (100 mM, pH 7.4) containing 100 pM EGTA andagaincentrifuged.Thesupernatantwasagaindiscardedandthis last steprepeated.The final pellet wasresuspendedinTris-acetatebuffer and frozenat —70°Cif not immediatelyused.

Ligand bindingBefore ligandbinding, membranefractionswere thawed

andcentrifugedat 14,000rpm, 4°C,for 20 mm. Thesuperna-tant wasdiscardedandthe pelletrestispendedin Tris-acetatebuffer (100 mM, pH 7.4, containing 100 p.M EGTA) and

.1. Neurochein.,Vol. 69, No. I, /997

IVMDA RECEPTORREGULATION OF AMPA RECEPTORS 133

allowedto preincubateon ice for 45—60 mm. After preincu-bation,thetissuewascentrifuged, supernatantdiscarded,andthe pelletresuspendedin Tris-acetate bufferand aliquoted.Tissue sampleswere incubatedat 0°Cfor 45 mm in Tris-acetatebuffer (50 mM, pH 7.4, 50 pM EGTA) containingeither50 mM potassiumthiocyanate(KSCN) in thepres-ence of 40nM I1H1AMPA (sp. act.53 Ci/mmol), or 6-cyano-7-nitro[3H] quinoxaline-2,3-dione([3Hj CNQX) (50nM; sp. act. 26.7Ci/mmol) and 100 pM glycine. Nonspe-cific binding wasdefinedin the presenceof 2 mM L-gluta-mate.Thereactionwasterminatedby centrifugingthesam-ples at 14,000 rpm,4°C,for 10 mm. The supernatantwasaspiratedandthepelletwashedwith 200 pl Tris-acetate(tOOmM, pH 7.4, 100 pM EGTA). with 50 mM KSCN for[H]AMPA bindingor without KSCN for [5HICNQX bind-ing. Pelletswere solubilized in 100 p,1 of 0.1 M NaOH andradioactivity determinedby liquid scintillation counting.

Western blotsCrude synapticmembraneswere preparedas described

above.After removal of a lO-pl aliquot forprotein assay,an equal volumeof 2x sodiumdodecylsulfate(SDS) sam-ple bufferwasadded tothetissueandheated to80°Cfor 10mm. Aliquots were subjectedto SDS-polyacrylamidegelelectrophoresis(SDS-PAGE) (8% polyacrylamide) andtransferredonto nitrocellulosemembranes.After incubationin Tris-bufferedsaline(TBS) containing3% gelatin atroomtemperature,blots were incubated overnightwith primaryantibodiesagainsttheC-terminaldomain of GIuRI (1:1,000dilution) (a generousgift from Dr. G. Lynch, UC-Irvine,CA, U.S.A.) ortheN-terminaldomainof GIuRI (agenerousgift of Dr. R. Wenthold,NIH, Bethesda,MD, U.S.A.) orSBDP(1:10,000dilution; agenerousgift from Dr. R. Siman,Cephalon,Inc., West Chester,PA, U.S.A.) in TBS con-taining 1% gelatin and0.05% Tween20. Immunostainingof GluRI or SBDPwasdetectedusingalkalinephosphatase—conjntgated secondary antibody(Bio-Rad). Blots werescannedinto a computerandanalyzedby usingImageQuantsoftware(Molecular Dynamics).Protein assayswere per-formed using Bio-Rad protein assaydye reagent as in-structed.

RESULTS

Effect of NMDA treatment on AMPA receptorbinding

Hippocampal slicecultureswereexposedto 100 p.MNMDA for 15 mm in thepresenceof either calpaininhibitors (calpain inhibitor I or calpeptin, 100 pM)or vehicle(0.05%DMSO). NMDA exposurewaster-minatedby removal of hippocampalslices from themembrane insertsand preparationof membranefrac-tions for ligand binding. Two populations ofAMPAreceptorshave beenpreviouslydescribed (Terramaniet al., 1988; Standley et a]., 1994), consisting of“high-affinity” (K

0 ~-~20nM; B,,a, “~~0.5pmol/mg ofprotein) and “low-affinity” (K0 ~-.~500nM; ~ ~‘2lpmol/mg of protein) binding sites. To determine dif-ferencesin regulation that may exist betweenthesetwo receptor populations,two measurementsof [

1H] -

AMPA binding weremade atlow (40 nM) and high(500 nM) [3H]AMPA concentrations.In addition, Ii-

gand binding was also performedby using [3H]-CNQX, an AMPA receptorantagonist.

A 15-mm exposureto NMDA (100 pM) resultedin a significant increasein [3H]AMPA binding mea-sured at both low- and high-AMPA concentrations(Fig. 1A and B). This increasewas inhibited signifi-cantlyby thepresenceof either calpaininhibitor!(lOO

pM) or ca!peptin (100 pM). AMPA binding wasnotsignificantly differentin membranesprepared fromtis-sueincubatedwith calpain inhibitor I (or calpeptin),calpaininhibitor I with NMDA, and control (DMSOalone). Similar resultswere obtainedwith calpeptinand calpaininhibitor I, andtheresultswith both inhibi-tors werecombined. Additionalexperiments indicatedthat theseinhibitors were not toxic to the cultures,as monitored by assayingfor lactatedehydrogenase

FIG. 1. Effects of NMDA treatment of cultured hippocampalsliceson [3H]AMPAbinding. Organotypic cultures of hippocam-pal slices were exposed to 100 pM NMDA for 15 mm with orwithout 100 pM calpain inhibitor I or 100 pM calpeptin. Mem-branes were prepared and [3H]AMPAbinding was measured atlow (A) and high (B) [3H]AMPAconcentrations as described inMaterials and Methods. Results are expressed as picomoles permilligramof protein and represent mean ~ SEM values of eight ornine experiments. Binding was riot significantly different betweenmembranes from cultures incubated with calpain inhibitor I, cal-pain inhibitor I with NMDA, and control. *p < 0.05 (Student’stest).

J. Neurorhcm., Vol. 69, No. 1, 1997

134 D. M. GELLERMAN ET AL.

blocked by pretreatment of culturedsliceswith cal-peptin.

DISCUSSION

FIG. 2. Effects of NMDA treatment of cultured hippocampalslices on [3H]CNQXbinding. Organotypic cultures of hippocam-pal slices were exposed to 100 pM NMDA for 15 mm with orwithout 100 pM calpain inhibitor I or 100 pM calpeptin. Mem-branes were prepared and [3H]CNQXbinding was measured asdescribed in Materials and Methods. Results are expressed aspicomoles per milligram of protein and represent mean it SEMvalues of eight or nine experiments. There were no significantdifferences in [3H]CNQX binding among treatments.

releasein the medium (Bruce etal., 1995: datanotshown). Although the effectsof NMDA treatmenton13H]CNQX binding exhibiteda similar trend (i.e., anNMDA-mediated increase)(Fig. 2), the results didnot reachstatistical significance.

Effects of NMDA treatment on GIuR1 subunitsIn addition to ligand binding,AMPA receptorprop-

erties were also examinedby subjecting membranesprepared fromcultured hippocampalslices to SDS-PAGE (8% polyacrylamide) and westernblots itsingprimary antibodies againstSBDP or the C-terminal(C-Ab) or the N-terminal (N-Ab) domains of GluRlsubunits of AMPA receptors. Western blots werescanned,andthe relative amounts ofGIuRI subunitswere quantifiedby using ImageQuantsoftware(Mo-lecular Dynamics).

As previously described (del Cerroet al., 1994;Vanderklishet at., 1995), a 15-mm expositreto 100pM NMDA resulted in theaccumulationof SBDP,aneffect that was completelyblocked by preincuhationwith calpain inhibitors (Fig. 3A). NMDA exposureresultedin a significantdecreasein anti-GluRl immu-noreactivity (C-Ab) in membranefractions comparedwith membranefractions prepared fromDMSO alonecontrol slices(Fig. 3; p < 0.05). In contrast,preincu-bationof sliceswith eithercalpaininhibitor I orcalpep-tin completelypreventedtheNMDA-induceddecreasein GluRl subunits (p < 0.01). The same treatmentproduceda small downwardshift in theband labeledwith an antibody against the N-terminal domain ofGIuR I subunits,without a decreasein theamount ofimmunoreactivematerial (Fig. 3B). Quantitativeesti-matesof theamplitudeof theshift in molecularmasswere ~-~0.5kDa. This effect was also completely

Thepresentstudydemonstratesthatabrieftreatmentof organotypichippocampalslicecultureswith NMDAproduces several modificationsof AMPA receptorsasaresultof theactivation ofthecalcium-dependent pro-teasecalpain.Binding of AMPA measuredat both lowand high concentrationswas significantly increased,and CNQX binding was also increased,although theeffectdid not reachstatistical significance. Resultsob-tained with westernblots by using antibodiesagainstN- and C-terminaldomains of GluRl indicateda de-creasein theamountof thespeciesrecognizedby C-Ahanda small downwardshift in theapparentmolecularweight of the speciesrecognizedby N-Ab. Similareffectswerepreviously obtainedby incubatingsynap-tosomal membraneswith purified calpain I, and weproposedthatcalpainproducesthetruncation ofaverysmall portion of the C-terminal domain (four or fiveresidues)of GIuRI. which containstheepitope recog-nized by the C-terminal antibody, thus producing adecreasein the lO5-kDa speciesrecognizedby theC-Ab anda small shift in theapparentmolecularweightof the species recognizedby N-Ab (Bi et al., 1997).As calpain inhibitors blocked the NMDA-mediatedmodificationsof GluRl subunits, it is thereforelikelythat such modificationsresult from NMDA-mediatedcalpamnactivation. In addition, becausetheeffects ofNMDA on AMPA receptorbinding werealsoblockedby preincubationwith calpaininhibitors, it is temptingto propose that calpain-mediatedproteolysis of theGIuRI (and possibly GlitR2—3) subunits is responsi-ble for all the modifications of the AMPA receptorsobserved afterNM DA treatment. However,treatmentof synaptic membraneswith purified calpain I, al-though producingprotcolysis of the AMPA receptorssimilar to that observedwith NMDA treatmentof hip-pocampalslicesdoes not result in increasedAMPAbinding(Bi et al., 1997). Therefore,additional mecha-nisms must underlie NMDA-niediated changesinAMPA receptor binding.

Two populationsof AMPA receptorshave beende-scribed. comprised of high-affinityAMPA bindingsites,locatedprimarily in the intracellular,microsomalfraction, andlow-affinity AMPA binding sites, locatedin the sytlaptic membranefraction (Standleyet al..1994). Thesetwo sites may representtwo membrane-dependent statesof the receptors.as solubilization ofsynaptic AMPA receptorsrenders them high affinity(Hall et al., 1992). Furthermore,radiolabeledCNQXbinding is generally positively correlatedwith low-affinity AMPA binding, but not high-affinity AMPAbinding, suggestingthatthestateof thereceptorhavinglow affinity for AMPA also has a high affinity forCNQX, whereasreceptorshaving a high affinity for

I. Neuroi’lic’nv, Vol. 69, No. I. / 997

NMDA RECEPTORREGULATION OF AMPA RECEPTORS 135

FIG. 3. Effect of NMDA treatment of cultured hippocampal slices on GIuR1 subunits. Organotypic cultures of hippocampal slices wereexposed to 100 pM NMDA for 15 mm with or without 100 pM calpain inhibitor I or 100 pM calpeptin. Membraneswere prepared andcharacteristics of GIuR1 were evaluated by western blots, using primary antibodies directed against the C-terminal and the N-terminaldomains of GIuR1. A: Top: Representative samples of western blots stained with an antibody against the C-terminal domain of GIuR1(arrowhead) and an antibody against SBDP (arrow). Bottom: Quantification of western blots. Results represent amounts of the 105-kDa species expressed as percentages of controls and are mean ±SEM values of four or five experiments. *p < 0.05, comparedwith control. °Notsignificantly different from control (ANOVA followed by post hoc Scheffé analysis). B: Top: Representative examplesof western blots stained with an antibody against the N-terminal domain of GIuR1 (arrowhead). Bottom: Quantification of westernblots. Results represent amounts of the 105-kDa species expressed as percentages of controls and are mean ± SEM values of fouror five experiments. There were no significant differences between any treatments.

AMPA havealow affinity for CNQX (Standleyet al.,1994).It is thereforepossiblethatactivation ofcalpainresulting from NMDA receptor stimulation may haveallowed the insertion of AMPA receptorsinto mem-branes,in addition to producingthepartial proteolysisof the C-terminal domain (Standley et al., 1996).However,wecannot completelyexcludethepossibilitythat the calpain inhibitors we used interact directlyor indirectly with other cellular processes thatcouldproducemodifications of binding and immunologicalpropertiesof the receptors.

Earlier studies havesuggestedthat modulation ofpostsynaptic glutamate receptorsis involved, if notnecessary,for synapticplasticity. The bindingof I 3Hj -

AMPA to AMPA receptorhasbeenshownto increaseafter inductionof LTP (Tocco etal., 1992; Maren etal., 1993)and learning ofa classicconditioning task(Tocco eta!., 1991). In addition, calpainactivity isnecessaryfor LTP, asboth leupeptinand calpaininhib-itor I have been shown to prevent the formation of

LTP (Oliver et al., 1989;del Cerroet al., 1990;Dennyet al., 1990). Finally, ca!pain is rapidly activatedbyNMDA receptorstimulation, asdemonstratedin bothhippocampalslicepreparationsand organotypichippo-campalcultures(Seubert etal., 1988;Bahretal., 1995;Vanderklishet al., 1995).Ourresults,whichshowthatAMPA receptorsare modified as a result of calpainactivation after NMDA receptorstimulation, clearly

supportthehypothesisthat this cascadeof biochemicaleventsparticipatesin synapticplasticity. Such a hy-pothesis does not exclude the possibility that otherAMPA receptor modifications,in particularphosphor-ylation of various subunits,could also participateinsynapticplasticity (McGlade-McCulloh etal., 1993:Raymondet al., 1993). Furthermore,they suggestthatthis preparation(i.e., brief NMDA treatmentof or-ganotypic cultures of hippocampalslices) should beof greatvalue to further identify the functional conse-quencesof theproteolysis of AMPA receptorsubunits.

Acknowledgment: This study was supportedby grantsfrom NINDS (NS 15827) and from Sankyo.

REFERENCES

Ambros-IngersonJ. and Lynch G.(1993) Channel gating kineticsandsynapticeflicacy: a hypothesisfor expressionof long-termpotentiation.Proc. Nail. Acud..S’ci. LISA 90, 7903—7907.

Anibros-IngersonJ., Larson J., XiaoP., and Lynch G.(1991) LTPchangesthe waveformof synaptic responses.Synapse9, 314—316.

Artola A. andSingerw. (1993)Long-termdepressionofexcitatorysynaptictransmissionand itsrelationshipto long-termpotentia-tion. Trend.sNeurosci. 16, 480—487.

Bahr B. A.,Tiniveedhi S.. Park G. Y.. and Lynch G.(1995) Induc-tion of calpain-mediatedspectrinfragmentsby pathogenictreat-ments in long-term hippocampal slices.J. Pharmacol. Erp.Ther. 273, 902—908.

Baudry M. and MassicotteG. (1992) Physiological andpharmaco-

I. Neuron’/a’nr. Vol. 69, No. 1, /997

136 D. M. GELLERMAN ET AL.

logical relationshipsbetweenlong-term potentiationand mam-malian memory.ConceptsNeurosci. 3, 79—98.

Bear M. F. andMalenkaR. C. (1994) Synapticplasticity: LTP andLTD. Curr. Opin. Neorobiol. 4, 389—399.

Bi X., ToccoG., andBaudry M. (1994)Calpain-mediatedregulationof AMPA receptorsin adult rat brain.Neuroreport 6, 61—64.

Bi X.. Chang V., Molnar E.. Mcllhinney J.. and Baudry M. (1996)The C—terminal domain of GluRI subunits is a target for cal-pain-mediatedproteolysis.Neuroscience73, 903—906.

Ri X., Chen J., Dang S., Wenthold R. J.. ToccoG., and BaudryM. (1997) Characterizationof calpain-niediatedproteolysisofGIuRI subunits of cx-amino-3-hydroxy-5-methylisoxazole-4-propionate receptors in rat brain. J. Neurochem.68, 1484—1494.

Bliss ‘F. V. and Collingridge G. L. (1993) A synaptic model ofmemory: long-term potentiation in the hippocampus.Nature361, 31—39.

Bruce A. J.,Sakhi S.. SchreiberS.S.,andBaudryM. (1995) Devel-opmentof kainic acid andN-niethyl-o-asparticacid toxicity inorganotypichippocampalcultures.Exp. Neurol. 132, 209—219.

del CerroS.,LarsonJ., OliverM. W.,andLynch G. (1990)Develop-ment of hippocampallong-term potentiationis reducedby re-cetitly introduced calpaininhibitors. Brain Rex. 530, 91—95.

delCerro5., Arai A., KesslerM.. BahrB.A., VanderklishP.. RiveraS., andLynch G. (1994)Stimulation of NMDA receptorsacti-vatescalpain in cultured hippocampal slices. Neuro,sc’i. Lett.167, 1—2.

DennyJ. B., PoEm-CurtainJ.,GhumanA., Wayner M. J.. andArm-strongD. L. (1991))Calpaininhibitorsblock long-term potenti-ation. Brain Rex. 534, 317—320.

Hall R.A.. KesslerM., and Lynch G. (1992)Evidencethat high-and low-affinity DL-cs-aniino-3-hydroxy-5-methylisoxazole-4-propionicacid (AMPA) bindingsitesreflect membrane-depen-dent statesofa single receptor.J. Neuroc’hern.59, 1997—2004.

IsaacJ. T. R., Nicoll R. A., and MalenkaR. C. (1995) Evidenceforsilent synapses’S.implicationsfor theexpressionof LTP. Neuron15, 427—434.

KesslerM.. Arai A., Vanderklish P., andLynch G. (1991) Failureto detect changesin AMPA receptorbinding after long-termpotentiation.Brain Rex.560, 337—341.

I,iao D. Z., HesslerN. A., and Malinow R. (1995) Activation ofpostsynapticallysilent synapsesduring pairing-inducedItp incal region of hippocampalslice. Nature 375, 400—404.

Lisman J. (1989) A mechanismfor the Hehh and the anti-Hebbprocessesunderlyitig learning andmemory.Proc. Nail. Acacl.

Sd. USA 86, 9574—9578.Lynch G. and BaudryM. (1984)The biochemistryof memory: a

new and specific hypothesis Science224, 1057—1063.MarenS. andBaudry M. (1995)Propertiesandmechanistiisof long-

term synapticplasticity in the mammalianbrain: i’elationshipsto learning and memory.Neurohiol. Learn. Me,a. 63, I — 18.

Maren S., ToccoG., StandleyS., Baudry M.. and ThompsonR. F.(1993) Postsynapticfactors in the expressionof long-termpa-tentiation(LTP): increasedglutamatereceptorbinding follow-ing LTP induction in vivo. Proc. Nail. Ac’acl.Sci. USA 90.9654—9658.

MeGlade-McCulloli E.. Yaniamoto H., Tan S. E.. Bricjey D. A..and Soderling T. R. (1993) Phosphorylationand regulationofglutamatereceptorsby calcium/calmodulin-dependentproteinkinaseII. Nature362, 640—642.

Oliver M. W., BaudryM.. andLynchG. (1989)Theproteaseinhibi-tor leupeptin interfereswith the developmentof LTP in hippo-campal slices.Brain Rex.505, 233—238.

RaymondL. A., BlackstoneC. D., and Huganir R. L. (1993) Phos-phorylationof aniinoacid neurotransmitterreceptorsin synapticplasticity. Ti-endsNeurosci.16, 147—153.

SeuhertP.. LarsonJ.,Oliver M.. JungM. W., BaudryM.. andl.yrichG. (1988) Stimulation of NMDA receptorsinducesproteolysisof spectninin hippocampus.Brain Rex. 460, 189—194.

Shahi K. and Baudry M. (1992) Increasingbinding affinity of ago-niststo glutamatereceptorsincreasessynapticresponsesat glu-taniatergicsynapses.Proc. Nail. Acacl. Sd. USA 89, 6881—

6885.StandleyS., Irvin N.. andBaudryM. (1994) Differential suhcellular

localization of two populationsof glutamate/AMPAreceptorsin the rat telencephalon.Neurochem.mt. 25, 287—293.

StandleyS., Bi X., andBaudryM. (1996)AMPA-glutaniatereceptorregulationand synapticplasticity. in Long-TermPotenliatio,i,Vol. III (Baudry M. and Davis J. L.. eds), pp. 17—40. MITPress,Cambridge,Massachusetts.

StaubliU., KesslerM., andLynch G. (1990)Aniracetamhaspropor-tionately smallereffects on synapsesexpressinglong-termpa-tentiation: evidencethat receptorchangessubserveLTP. Psi-chohiologv18, 377—381.

Stoppini L.. Buchs P. A., and Muller D. (1991) A simple method‘or oi’ganotypicculturesof nervoustissue..1. Neuro.vci.Meihod,s37, 173—182.

TerramaniT.. KesslerM., Lynch G.. andBaudry M. (1988)Effectsof’ thiol reagentson ‘H-AMPA binding to rat telencephalicmembranes.Mol. Pharniacol. 34, 117—123.

ToccoG.. DevganK. K., Hauge S. A.. Weiss C.. Baudry M.. atidThompson R. F. (1991) Classicalconditioning selectively in-creasesAMPA receptor binding in the rabbit hippocampus.Bi-ain Re,s.559, 331 —336.

ToccoG., MassicotteG.. StandleyS., ThompsonR. F., and BaudryM. (1992)Effect of temperatureand calcium on the bindingpropertiesof the AMPA receptor in frozen rat brain sections.Eur. J. Neuro,sci.4, 1093—1103.

Vanderklish P., SaidoT. C.. Gall C., andLynch G. (1995)Proteoly-sis of spectninby calpain accompaniestheta-hurststimulationin cultured hippocampalslices.Mol. Brain Rex.32, 25—35.

.1. Nciro~‘lie/li., Vol. 69. No. 1. 1997