the metabotropic glutamate receptor agonist 1s,3r-acpd stimulates and modulates nmda receptor...

Brain Research 898 (2001) 91–104www.elsevier.com/ locate /bres

Research report

The metabotropic glutamate receptor agonist 1S,3R-ACPD stimulatesand modulates NMDA receptor mediated excitotoxicity in organotypic

hippocampal slice culturesa , a,b a a,b*Morten Blaabjerg , Bjarne W. Kristensen , Christian Bonde , Jens Zimmer

aAnatomy and Neurobiology, Institute of Medical Biology, SDU-Odense University, Winsløwparken 21, DK-5000 Odense C, DenmarkbNeuroScreen ApS, Institute of Medical Biology, SDU-Odense University, Winsløwparken 21, DK-5000 Odense C, Denmark

Accepted 16 January 2001

Abstract

The potential toxic effects of the metabotropic glutamate receptor agonist (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (ACPD)and its interactions with the N-methyl-D-aspartate (NMDA) receptor were studied in hippocampal brain slice cultures, using densitometricmeasurements of the cellular uptake of propidium iodide (PI) to quantify neuronal degeneration. Cultures exposed to ACPD, showed aconcentration (2–5 mM) and time (1–4 days) dependent increase in PI uptake in CA1, CA3 and dentate subfields after 24 h and 48 h ofexposure, with CA1 pyramidal cells being most sensitive. The neurodegeneration induced by 2 mM ACPD was completely abolished byaddition of 10 mM of the NMDA receptor antagonist (5R,10S)-(1)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine(MK-801), while 20 mM of the 2-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA)/kainic acid receptor antagonist 2,3-dioxo-6-nitro-1,2,3,4-tetrahydrobenzo[ f ]quinoxaline-7-sulfonamide (NBQX) had no effect. Co-exposing cultures to a subtoxic dose of 300 mMACPD together with 10 mM NMDA, which at this dose is known to induce a fairly selective degeneration of CA1 pyramidal cells,significantly increased the PI uptake in both CA1 and CA3, compared to cultures exposed to 10 mM NMDA only. Adding the 300 mMACPD as pretreatment for 30 min followed by a 30 min wash in normal medium before the ACPD/NMDA co-exposure, eliminated thepotentiation of NMDA toxicity. The potentiation was also blocked by addition of 10 or 100 mM 2-methyl-6-(phenylethynyl)pyridine(MPEP) (mGluR5 antagonist) during the co-exposure, while a corresponding addition of 10 or 100 mM 7-(hydroxyimino)cyclop-ropa[b]chromen-1a-carboxylate ethyl ester (CPCCOEt) (mGluR1 antagonist) had no effect. We conclude that, stimulation of metabotropicglutamate receptors with ACPD at concentrations of 2 mM or higher induces a distinct subfield-related and time and concentrationdependent pattern of hippocampal degeneration, and that ACPD at subtoxic concentrations modulates NMDA-induced excitotoxicitythrough the mGluR5 receptor in a time dependent way. 2001 Elsevier Science B.V. All rights reserved.

Theme: Neurotransmitters, modulators, transporters, and receptors

Topic: Excitatory amino acids: excitotoxicity

Keywords: mGluR; Propidium iodide; MK-801; NBQX; N-Methyl-D-aspartate; MPEP; CPCCOEt; Neurodegeneration

1. Introduction receptors (mGluR1 and mGluR5) are linked to the21phosphoinositide /Ca cascade by stimulating phospholip-

Metabotropic glutamate receptors (mGluRs) are a family ase C (PLC), but have also been shown to stimulateof membrane-bound receptors coupled to intracellular phospholipase D (PLD) [33]. Group II (mGluR2 andGTP-binding proteins (G-proteins). They are divided into mGluR3) and III (mGluR4, 6, 7 and mGluR8) receptorsthree different groups based on differences in signal are negatively coupled to the adenylate cyclase, loweringtransduction cascades, amino acid sequences and agonist the amount of cyclic AMP (cAMP). Activation of group Ipharmacology (for reviews see Refs. [34,40]). Group I receptors has been predicted to be potentially neurotoxic,

21because the phosphoinositide /Ca cascade leads to a21direct increase in intracellular Ca [18,26]. Activation of*Corresponding author. Tel.: 145-6550-3812; fax: 145-6590-6321.

E-mail address: [email protected] (M. Blaabjerg). group II and III receptors has in contrast been predicted to

0006-8993/01/$ – see front matter 2001 Elsevier Science B.V. All rights reserved.PI I : S0006-8993( 01 )02148-5

92 M. Blaabjerg et al. / Brain Research 898 (2001) 91 –104

be neuroprotective [26]. There are, however, conflicting according to the interface method, as slightly modifiedobservations regarding direct toxic effects of metabotropic [17,27,28] from Stoppini et al. [46]. After decapitation andglutamate receptor activation and its contribution to ex- removal of the brains of 7-day-old Wistar rat pupscitotoxicity. The non-selective mGluR agonist (1S,3R)-1- (Møllegaard, Denmark), the hippocampi were isolated andaminocyclopentane-1,3-dicarboxylic acid (ACPD) acting the middorsal parts cut into 350 mm thick transverse sliceson all mGluRs [40], and more recently the group I on a McIlwain tissue chopper. The tissue slices wereselective agonist (S)-3,5-dihydroxyphenylglycine (DHPG), placed in Gey’s balanced salt solution (GBSS; Gibco BRL,have been shown to be neurotoxic to the hippocampus in Life Technologies, Paisley, UK, catalog No. 24260-028)vivo [30,39]. Neither 300 mM ACPD nor 300 mM DHPG with glucose (6.5 mg/ml) (Merck, Darmstadt, Germany,were, however, toxic to organotypic hippocampal slice catalog No. 8342), and separated and trimmed for excesscultures [32]. Similar to this, an equimolar mixture of tissue. The slices were then placed in inserts (diameter 31S,3R-ACPD and 1R,3S-ACPD (trans-ACPD) (500 mM) cm; Millipore, Bedford, MA, USA, catalog No. PICM 030was not toxic in cultured striatal neurons, even after 50) on semiporous membranes with six slices on eachpreceding depolarisation [5]. When injected directly into membrane, and the inserts transferred to a six-well culturethe rat striatum, ACPD (900 nmol over 15 min) did, tray (Corning Costar, Corning, NY, USA), with 1 ml ofhowever, induce internucleosomal DNA fragmentation medium in each well. The medium was composed of 25%[52], just as ACPD enhanced neuronal degeneration in inactivated horse serum (catalog No. 26050-047), 25%global cerebral ischaemia, when applied systemically (20 Hanks’ balanced salt solution (HBSS, catalog No. 24020-mg/kg intraperitoneally, i.p.) [12] and when applied to 091) and 50% OPTIMEM medium (catalog No. 31985-hippocampal slice cultures (300 mM) during oxygen and 047) (all from Gibco BRL), supplemented with D-glucoseglucose deprivation (OGD) [32]. ACPD has, however, also (25 mM) and L-glutamine (1 mM). The trays were placedbeen reported in a dose-dependent way (1–100 mM) to in an incubator at 368C with an atmosphere of 5% CO2

protect against OGD in acute hippocampal slice prepara- and 95% humidified air. From the third day and on thetions [45]. Stimulation of group I metabotropic glutamate medium was changed to chemically defined serum-freereceptors has also been found to potentiate N-methyl-D- Neurobasal medium (Gibco BRL, catalog No. 21103-049)asparatate (NMDA) excitotoxicity on cultured cortical supplemented with D-glucose (25 mM), L-glutamine (1neurons [4,47], while trans-ACPD attenuates the effects of mM) (Sigma, Vallensbæk Strand, Denmark, catalog No.NMDA injected into the neostriatum [6]. 25030-024) and 2% B-27 supplement (Gibco BRL, catalog

In this study we investigated the possible toxic effects of No. 17504-010). This medium was changed every third orincreasing concentrations of ACPD and exposure times on fourth day in 3–4 weeks before start of the experiments.rat hippocampal slice cultures, as well as the effect ofexposing the cultures to a subtoxic concentration of ACPD 2.2. Monitoring propidium iodide uptaketogether with NMDA with or without blocking the group ImGluRs with MPEP (mGluR5 antagonist) or CPCCOEt To quantitate neuronal damage, the fluorescent mark-(mGluR1 antagonist). We chose hippocampal slice cultures er PI h3,8-diamino-5-[3-(diethylmethylamino)propyl]-6-because (1) they display a distinct organotypic organiza- phenyl phenanthridinium diiodide; Sigma, catalog No.tion of the hippocampal and dentate gyrus main cell and P4170j was added to the medium, in accordance with anneuropil layers with preservation of the basic intrinsic earlier established protocol [17,27,28]. PI is a polarneural circuitry and a known pattern of connective reor- substance which only enters dead or dying cells withganization following isolation of the tissue slices from the damaged cell membranes. Here it binds to DNA, emittingrest of the brain [11,54], and (2) the occurrence of induced a brightly red, intensified fluorescence (630 nm), whenneuronal degeneration can be monitored and quantified by exposed to blue–green light (493 nm). PI is basicallydensitometric measurements of the cellular uptake of the non-toxic to neurons [14,35] and has been used as anfluorescent dye propidium iodide (PI). PI has been used indicator of neuronal membrane integrity [49] and cellincreasingly for this purpose over the past few years, for damage [15,35,36,50]. Three hours before glutamate re-example in hippocampal and corticostriatal slice cultures ceptor stimulation, 20 ml of 0.1 mM PI was added to theexposed to excitotoxins [3,17,43,50], hypoxia and hypo- medium for determination of basic cellular uptake (Day 0;glycemia alone or in combination [10,32,37,51], nitric Figs. 1 and 2). PI uptake was recorded by fluorescenceoxide [2], and trimethyltin [27]. microscopy [Olympus IMT-2, 43 (Splan FL2)], using a

standard rhodamine filter and digital camera (Sensys KAF1400 G2; Photometrics, Tucson, AZ, USA) with 0.75 s

2. Materials and methods exposure time. After addition of receptor agonists andantagonists to the culture medium, digital fluorescent

2.1. Organotypic hippocampal slice cultures micrographs were taken of the cultures at 24 h and 48 hand later (for details see Experimental protocol), for use in

Organotypic hippocampal slice cultures were prepared densitometric measurements of the PI uptake in the

M. Blaabjerg et al. / Brain Research 898 (2001) 91 –104 93

Fig. 1. Experimental protocol for exposure of slice cultures to increasing concentrations of ACPD and co-exposure to ACPD and the NMDA antagonistMK-801 or the AMPA/kainic acid antagonist NBQX.

cultures. The measurements included each culture as a concentrations of ACPD in serum-free medium, followedwhole as well as the dentate granule cell layer (FD), and by a change to pure serum-free medium for the rest of theCA3 and CA1 pyramidal cell layers separately. For all experimental period (Fig. 1). Digital images of the re-densitometric measurements we used NIH Image 1.64 sulting PI uptake (see above) were obtained 24 h (Day 1),(National Institute of Health, USA) analysis software. 48 h (Day 2), 96 h (Day 4) and 6 and 10 days after start of

the ACPD exposure. To test whether the ACPD-induced2.3. Experimental protocol neurodegeneration was a direct effect of metabotropic

glutamate receptor stimulation or an indirect effect involv-The potentially toxic effects of ACPD (Tocris Cookson, ing ionotropic glutamate receptors, cultures were co-ex-

Bristol, UK, catalog No. 0284) were examined by expos- posed to (a) 2 mM ACPD and 10 mM of the NMDAing slice cultures for 48 h to increasing, 100 mM to 5 mM, receptor antagonist (5R,10S-(1)-5-methyl-10,11-dihydro-

Fig. 2. Experimental protocol for investigation of the effects of ACPD on NMDA-induced excitotoxicity. For details, see text.


5H-dibenzo[a,d]cyclohepten-5,10-imine (MK-801) (RBI, Tris-buffered saline (TBS)11% Triton, followed by incu-catalog No. M-107), or (b) 2 mM ACPD and 20 mM of the bation with biotinylated anti-mouse immunoglobulin GAMPA/kainic acid receptor antagonist 2,3-dioxo-6- (IgG) (1:200, Sigma, catalog No. RPN 1004) for 1 h,nitro-1,2,3,4-tetrahydrobenzo[ f ]quinoxaline-7-sulfonamide streptavidin horseradish peroxidase (HRP) (1:200, Dako,(NBQX) (gift from Novo Nordisk, Bagsværd, Denmark). Glostrup, Denmark, catalog No. PO397) for 1 h and againTo verify the effects of the chosen concentrations of washed three times, before the immunostaining was visual-ionotropic glutamate receptor antagonists MK-801 and ized by incubation in 0.2% 3,39-diaminobenzidine (Sigma)NBQX, some cultures where exposed to 10 mM NMDA for 5 min. The sections were then dehydrated in ethanol,(Sigma, catalog No. M-3262) together with 10 mM MK- cleared in xylene and coverslipped in DePex (Kebo Lab.,801, while others were exposed to 3 mM AMPA or 8 mM Albertslund, Denmark).kainic acid together with increasing concentrations (1, 10,20 mM) of NBQX. The doses for NMDA, AMPA and 2.5. Densitometric analysis of MAP2 stained sectionskainic acid were chosen as being close to the respectiveEC values for the three glutamate receptor agonists, as All sections to be compared were stained simultaneous-50

estimated in an earlier study based on PI uptake in entire ly. At a magnification of 203, the fourth section, countedcultures [16]. For investigation of time-dependent interac- from the bottom side of the culture facing the membrane,tions between metabotropic glutamate receptor and NMDA was photographed by digital camera. Stratum radiatum ofreceptor stimulation, cultures were exposed to 10 mM CA1 and CA3 and the dentate molecular layer was thenNMDA alone or in combination with a subtoxic con- delineated in the resulting pictures, and the optical densitycentration of 300 mM of ACPD (Fig. 2). Other cultures of the MAP2 staining measured by the NIH-Image 1.64were pretreated with 300 mM of ACPD for 30 min, with analysis program.change to normal serum-free medium for additional 30min, before they were exposed to 10 mM NMDA, either 2.6. Statistics and calculation of EC values50

alone or co-exposed with 300 mM of ACPD. For clarifica-tion of the subtypes of metabotropic glutamate receptors All densitometric data were expressed as means1

involved, 10 or 100 mM of the mGluR5 antagonist 2- standard error of mean (S.E.M.). Dose–response curvesmethyl-6-(phenylethynyl)pyridine (MPEP) (Tocris Cook- were drawn on the basis of the recorded dose–responseson, catalog No. 1212) or 10 or 100 mM of the mGluR1 data by Microcal Origin 3.5 (Microcal Software, North-antagonist 7-(hydroxyimino)cyclopropa[b]chromen-1a-car- ampton, MA, USA). From the fitted curves, EC values,50

boxylate ethyl ester (CPCCOEt) (Tocris Cookson, catalog corresponding to 50% of the plateau level (defined asNo. 1028) was added during co-exposures with ACPD and maximal damage) in the dose–response curves wereNMDA. Digital images of PI uptake were taken 24 h (Day determined, corresponding to 24 h, 48 h of exposure and1) and 48 h (Day 2) after start of NMDA exposure (Fig. 48 h of exposure with additional 48 h in recovery medium2). (96 h). The dose–response data were statistically compared

by regular or repeated two-factor analysis of variance2.4. Immunohistochemical staining (ANOVA), including interactions in the model using

Intercooled Stata 6 (Stata Corporation, College Station,To further illustrate the ACPD-induced toxicity, im- TX, USA). The data for PI uptake during NBQX inhibition

munohistochemical staining for microtubule-associated were also compared using repeated two-factor ANOVA.protein 2 (MAP2) and neuron specific protein (NeuN) For other data, statistical significance was assessed inwere performed at the end of some of the ACPD exposure GraphPad Instat (GraphPad Software, San Diego, CA,experiments. MAP2 is predominantly present in dendrites USA), using Student t-test or single-factor ANOVA with[20] and has been used as a marker for neuronal integrity Bonferroni correction for comparison of the groups of[24]. At Day 10 non-exposed control cultures, and cultures interest. Differences were considered significant at P,

exposed to 5 mM ACPD were fixed in phosphate-buffered 0.05.4% PFA for 1 h, transferred to 20% sucrose for 16–72 h at48C, embedded in Cryo-embed (AX-LAB), frozen ingaseous CO , and cut in three parallel series of 20 mm 3. Results2

sections by cryostat, with storage of the mounted sectionsat 2208C until further processing. For immunohistochemi- 3.1. PI uptake in ACPD exposed culturescal staining, the sections were thawed to room temperature,treated with 10% fetal bovine serum (FBS) for 30 min, Control cultures, not exposed to any agonists, displayedincubated by monoclonal mouse anti-MAP2 antibody a limited PI uptake at d0 and later (Figs. 3A and 10A), as(1:1000, Sigma, catalog No. A-2052) or monoclonal did the experimental cultures before exposure to themouse anti-NeuN antibody (1:500, Chemicon, catalog No. agonists (d0). Exposure to ACPD increased the PI uptakeMAB377) for 48 h at 48C, then washed three times in in a concentration (2–5 mM) (Fig. 3B–D) and time


Fig. 3. Propidium iodide (PI) uptake in cultures exposed to increasing, concentrations ACPD for 48 h, followed by 48 h, recovery in normal medium.Control culture (A) and culture exposed to 300 mM ACPD (B) show the same low basic PI uptake. Culture exposed to 2 mM ACPD (C) displays distinctlyincreased PI uptake in the CA1 and CA3c subfields, while the PI uptake is increased in all hippocampal subfields in cultures exposed to 5 mM ACPD (D).

(24–96 h) dependent manner with CA1 being most time points were tested statistically against each other forsensitive (Fig. 4A–C). After 24 h of exposure, 3 mM the same subfields (24 h vs. 48 h; 24 h vs. 96 h and 48 hACPD induced a significantly increased PI uptake in CA1 vs. 96 h), the curves for FD were all significantly differentcompared to control cultures, while 5 mM ACPD caused from each other. This was also the case for the curves foran increase in PI uptake in all subfields (Fig. 4A). After CA3 and the entire cultures (all ***P,0.001). The dose–exposure to 2 mM ACPD for 48 h the PI uptake in CA1 response curves for CA1 after 24 h and 48 h of exposurewas significantly above control. At this time point cultures were also individually different (***P,0.001), whereasexposed to 3 and 5 mM displayed an increased PI uptake the curves for 48 h and 96 h were not. CA1 had thusin all subfields. This was also seen after the additional 48 h reached near maximal (96613%) PI uptake (cell death)recovery. At this late time point also the CA3 subfield had already after 48 h when exposed to 5 mM ACPD (Fig.increased its PI uptake in cultures exposed to 2 mM 3B). Based on the dose–response curves, EC values for50

ACPD, mainly due to increase in the CA3c subfield (Fig. ACPD were estimated for the different subfields and the3C). Hereafter the cultures exposed to 3 and 5 mM ACPD entire culture at the different time points (Table 1).displayed no further increase in PI uptake. The mean valueof the PI uptake in all subfields 96 h after start of exposure 3.2. MAP2 and NeuN immunohistochemical stainingin cultures exposed to 5 mM ACPD was now set to 100%,representing maximal PI uptake. To correlate the maximal PI uptake induced by 5 mM

When the dose–response data sets were tested statisti- ACPD with other markers of neuronal injury, cryostatcally against each other using two-factor ANOVA, the sections were immunocytochemically stained for MAP2CA1 was found to be more susceptible to ACPD than CA3 (Fig. 5) and NeuN (Fig. 7). When the MAP2 staining ofand fascia dentata (FD), at both 24 h (**P,0.01 and stratum radiatum of CA1 and CA3 and the dentate***P,0.001, respectively) and 48 h (***P,0.001 and molecular layer was measured densitometrically, an almost***P,0.001, respectively) after start of exposure (Fig. total loss of dendrite staining was verified (Fig. 6). Also in4A–B). When the dose–response curves at the different NeuN staining, CA1 and CA3 pyramidal cells and dentate


Fig. 4. Dose–response curves for PI uptake in sets of cultures exposed to 0.1–5 mM ACPD, illustrating the PI uptake in the dentate (FD), CA1 and CA3subfields after exposure to ACPD for 24 h (A), 48 h (B), and for 48 h followed by 48 h recovery (96 h) (C). Measurements of the PI uptake in the cultureas a whole (all subfields together) are shown in (D). For each figure, the maximal PI uptake at the plateau-level 96 h after start of exposure is set to 100%and the data expressed as percentage of this. Data are shown as means1S.E.M., with n56–14, *P,0.05, **P,0.01, ***P,0.001 using ANOVA withBonferroni correction for comparison with control. The curve for CA1 was significantly different (***P,0.001) from the curves for CA3 and FD at 24 and48 h after start of exposure, using two-factor ANOVA, including interactions in the model. Comparing the PI uptake over time, the three curves (24, 48, 96h) for subfields FD and CA3 and whole cultures were significantly different (***P,0.001), whereas the curves for CA1 differed between 24 h and 48 hafter start of exposure (***P,0.001). For further explanation, see text.

granule cells showed clear signs of degeneration (Fig. receptor stimulation or an indirect effect on ionotropic7D–F). glutamate receptors, cultures were were exposed for 48 h

to 2 mM ACPD in the presence of 10 mM of the selective3.3. Effects of ACPD exposure in the presence of MK- NMDA receptor antagonist MK-801 or 20 mM of the801 or NBQX AMPA/kainic acid receptor antagonist NBQX, followed

by additional 48 h in normal medium. Separate experi-To test whether the neurodegeneration induced by ments showed that 10 mM MK-801 completely protected

ACPD was a direct effect of metabotropic glutamate the slice cultures against 10 mM NMDA (Fig. 8A), while20 mM NBQX completely protected against 3 mM AMPA,and provided protection (7466%) against 8 mM kainicTable 1acid (Fig. 8B). Recordings of the PI uptake, 24, 48 and 96EC values (mM) of 1S,3R-ACPD in organotypic hippocampal slice50

acultures h after start of exposure, showed that addition of 10 mMMK-801 to the medium gave complete neuroprotectionTime Subfield Entire

(h) cultures against neurodegeneration induced by 2 mM ACPD, whileFD CA3 CA1 20 mM NBQX was without significant effect (Fig. 9).

24 4.2 3.7 2.6 2.748 3.8 3.0 2.2 2.3 3.4. Effect of ACPD on NMDA-induced excitotoxicity96 2.5 2.4 2.1 2.2a Note that the CA1 subfield has the lowest EC value at all time points. In accordance with previous results, 10 mM NMDA50


Fig. 5. High-power photomicrographs of MAP2 stained sections of cultures exposed to ACPD. Control cultures (A–C) displayed normal dendrites in thedentate molecular layer (FDmol) and stratum radiatum of CA3 (CA3r) and CA1 (CA1r). In cultures exposed to 5 mM ACPD (D–F), nearly all thedendritic MAP2 staining has disappeared, with clear persistence of some degenerating dendrites in CA3r (arrows). For densitometric measurements, seeFig. 6.

induced a relatively selective increase in PI uptake in theCA1 pyramidal cell layer of the exposed cultures (Fig.10B) [16]. Exposing cultures to 10 mM NMDA togetherwith an otherwise subtoxic dose of 300 mM ACPDsignificantly enhanced the PI uptake in CA1 and in oneexperiment also in CA3 (Figs. 10C and 11). This potentiat-ing effect of ACPD was, however, abolished by pretreatingcultures with 300 mM ACPD for 30 min, followed byfurther 30 min in normal medium, before NMDA/ACPDco-exposure (Figs. 10D and 11). The potentiation ofNMDA toxicity was also abolished by the presence of 10

Fig. 6. Densitometric measurements of MAP2 staining showing a or 100 mM MPEP during the co-exposure, while thesignificant decrease in MAP2 density in all subfields of cultures exposed presence of 10 or 100 mM CPCCOEt was without effectto 5 mM ACPD. Due to some remaining degenerating dendrites in the

(Fig. 12). Reduction of both the duration of ACPDCA3r, this subfield only showed 70% reduction in staining density. Datapretreatment and the interval in normal medium before theare shown as mean1S.E.M., with n55–6, and ***P,0.001 using

Students t-test. NMDA/ACPD co-exposure to 5 min each did not, how-


Fig. 7. High-power photomicrographs of NeuN stained sections of cultures exposed to ACPD according to protocol in Fig. 2. (A–C) Control cultures withnormal dentate granule cells (FD) and CA3 and CA1 pyramidal cells. (D–F) In cultures exposed to 5 mM ACPD in particular dentate granule cells are lost(D), while pyramidal cells left in CA3 (E) and CA1 (F) display clear degenerative features, including darkly stained nuclei (arrows).

ever, block the potentiating effect of ACPD in direct concentrations of 2 mM or higher induces a concentration-co-exposure with 10 mM NMDA (Fig. 13). In order to test dependent degeneration of neurons in organotypic hip-whether 30 min pretreatment with ACPD would effect pocampal slice cultures. This corresponds to the in vivoNMDA excitotoxicity, when NMDA was applied alone, findings that ACPD is toxic to hippocampal neurons aftercultures were pretreated for 30 min with 300 mM ACPD, intrahippocampal (1 ml of 10 mM or 1 mmol over 2 min)with an additional 30 min interval in normal medium, [22,39] and intraventricular (20 nmol) injections [19]. Itbefore being exposed to 10 mM NMDA alone. In this set should, however, be noted that millimolar concentrationsup the 30 min ACPD pretreatment had no effect on the of ACPD are required to induce toxicity. At these con-NMDA-induced PI uptake, compared to cultures exposed centrations the selectivity of ACPD is reduced, includingto 10 mM NMDA only (data not shown). possible inhibition of glutamate transporters [23], and

activation of ionotropic glutamate receptors like theNMDA receptor [41]. The observed toxic effects of ACPD

4. Discussion at higher concentrations may accordingly be mediatedthrough glutamate uptake inhibition (with increased ex-

4.1. Excitotoxic effects of ACPD by NMDA receptors tracellular levels of endogenous glutamate acting onionotropic glutamate receptors) or direct stimulation of

The present study has demonstrated that ACPD in ionotropic glutamate receptors. This is supported by the


Fig. 8. (A) MK-801-mediated neuroprotection against 10 mM NMDA, as detected by propidium iodide (PI) uptake in CA1. The PI uptake was recorded 48h after start of exposure and is expressed with the PI uptake without addition of MK-801 (0 mM) set to 100%. Data are shown as means1S.E.M. withn56–22, ***P,0.001, ANOVA with Bonferroni correction for comparison with control. (B) NBQX-mediated neuroprotection against 3 mM AMPA and 8mM KA, respectively as detected by uptake in CA1. The PI uptake was recorded 48 h after start of exposure to AMPA or KA and expressed with the PIuptake without addition of NBQX (0 mM) set to 100%. When compared by repeated two-factor ANOVA, the inhibition curves for AMPA and kainic acidwere significantly different. Data are shown as means1S.E.M with n511–18, (***P,0.0015NBQX versus AMPA alone, §§P,0.01, §§§P,0.0015

NBQX versus kainic acid alone), ANOVA with Bonferroni correction for comparison with control.

present demonstration of total prevention of neurodegene- 801-sensitive NMDA receptors. Little, if any of theration when 2 mM ACPD is applied together with 10 mM neurodegenerative effect is mediated through AMPA/MK-801. The involvement of the NMDA receptors are kainic acid receptors, since co-application of 20 mMalso supported by earlier findings in vivo, where the toxic NBQX had no protective effects against 2 mM ACPD.effects of ACPD were reduced by application of the These results do not exclude that activation of group INMDA receptor antagonist LY274614 [39]. The neurode- metabotropic glutamate receptors can cause neurodegene-generative effects of millimolar doses of ACPD does ration. Intracerebroventricular injection of DHPG can thusaccordingly seem to result from direct activation of MK- induce neurodegeneration without activation of ionotropic

glutamate receptors [30]. Such toxic effects of group Ireceptor activation are believed to be associated with an

21increase in intracellular Ca and enhanced glutamaterelease [26]. In accordance with this, dantrolene has beenshown to be protective against ACPD-induced toxicity by

21preventing the Ca release from intracellular stores [21].Regarding increased glutamate release, this has, however,recently been shown not to be dependent on mGluR1receptor activation [44], suggesting that the main toxiceffect of group I receptors may be mediated by themGluR5 receptor.

ACPD concentrations in the 0.1–1 mM range were notfound to induce detectable neuronal degeneration. This isin accordance with other studies, where organotypic hip-pocampal slice cultures were exposed to 300 mM ACPDwithout noticeable neurodegeneration, as monitored by PIuptake [32]. This lack of neurodegenerative effect might bedue to concomitant activation of group II and III receptors,which are suggested to mediate neuroprotection by inhibi-tion of glutamate release [26], thereby keeping a potentialFig. 9. The PI uptake induced by 2 mM ACPD in CA1 of hippocampal

slice cultures 96 h after start of exposure was completely blocked by release of glutamate at a subtoxic level.addition of the NMDA receptor antagonist MK-801 (10 mM). The PIuptake was not significantly affected by addition of the AMPA/kainic 4.2. Effects of ACPD on NMDA-induced excitotoxicityacid receptor antagonist NBQX (20 mM). The PI uptake induced byACPD measured after 96 h was set to 100%. Data are shown as

Our second set of results, obtained by ACPD andmeans1S.E.M. with n512–17 and ***P,0.001 using ANOVA withBonferroni correction for comparison between groups. NMDA co-exposure, showed that simultaneous activation


Fig. 10. Propidium iodide uptake in hippocampal slice cultures after 48 h of exposure to 10 mM NMDA alone or in combination with 300 mM ACPD. (A)Control culture with low basal PI uptake. (B) Culture exposed to 10 mM NMDA with PI uptake particularly in CA1. (C) Culture with clearly potentiated PIuptake after co-exposure to 10 mM NMDA and 300 mM ACPD. (D) The potentiation of PI uptake in (C) was eliminated in culture pretreated (pt.) with 300mM ACPD for 30 min followed by 30 min in normal medium before the co-exposure.

Fig. 11. The PI uptake measured after 48 h (see protocol in Fig. 2) was significantly higher in both CA1 and CA3 in hippocampal slice cultures co-exposedto 10 mM NMDA and 300 mM ACPD (‘NMDA1ACPD’), compared to cultures exposed to 10 mM NMDA alone (‘NMDA’; set to 100%). The enhancedPI uptake was eliminated by pretreating the cultures with 300 mM ACPD for 30 min (pt. 30), followed by an interval in normal medium for 30 min beforethe co-exposure (‘NMDA’ and ‘ACPD pt. 30/NMDA1ACPD’ not significantly different). Data are shown as means1S.E.M., with n59–18, **P,0.01,***P,0.001 using ANOVA with Bonferroni correction for comparison between groups.


Fig. 12. Inhibition of the mGluR5 receptor with 10 or 100 mM of the non-competitive selective antagonist MPEP during the NMDA/ACPD co-exposure,eliminated the potentiation of NMDA excitotoxicity induced by 300 mM ACPD. Inhibition of the mGluR1 receptor with 10 or 100 mM CPCCOEt did notaffect the potentiation. Data are shown as means1S.E.M., with n511–18, *P,0.05, **P,0.01 and ***P,0.001 using ANOVA with Bonferronicorrection for comparison between groups. PI uptake induced by 10 mM NMDA is set to 100%.

of (a) metabotropic glutamate receptors by a subtoxic acid by trans-ACPD (100 mM) has, however, also beenconcentration of ACPD and (b) NMDA receptors by a observed in the neostriatum [6]. It has been suggested thattoxic 10 mM concentration of NMDA, potentiated the the potentiation of NMDA-induced excitotoxicity in corti-NMDA-induced excitotoxicity. This corresponds to other cal cultures results from NMDA receptor activation and

21studies, where activation of group I receptors has been increasing intracellular Ca causing protein kinase Cshown to potentiate NMDA-induced excitotoxicity in (PKC) to translocate from the cytosol to the cell mem-primary cortical cultures [4,47], as well as electrophysio- brane, where the PKC is then fully activated by dia-logical responses in the spinal cord [48] and in acute cylglycerol (DAG), known to be produced by group Ihippocampal slices [9]. An opposite effect with reduction metabotropic glutamate receptor activation [4]. PKC mightof excitotoxicity induced by the NMDA agonist quinolinic in this way add further to the intracellular changes and lift

21the Mg block of the NMDA receptor. Recently it has,however, also been shown that ACPD increases the

21translocation of the Ca -independent PKCe isoform in ratcortical synaptosomes [31], so perhaps both mechanismsare involved in the results we see in our study. Potentiationof NMDA responses by metabotropic glutamate receptorsdoes, however, not only depend on which metabotropicglutamate receptors are involved, but also on the NMDAreceptor subunit composition. Studies with co-transfectionof mGluR1 and NMDA receptor subunits in Xenopusoocytes have thus revealed that ACPD only potentiatesNMDA responses when or NR2A or NR2B subunits arepresent [42]. In this relation it is noteworthy that highlevels of these receptor subunits have been detected in allhippocampal subfields both by Western blot and immuno-staining [8,53]. Also the role of astroglial cells should betaken into consideration. Potentiation of NMDA-inducedexcitotoxicity by DHPG and quisqualate in primary cul-tures of cortical neurons has thus been reported only toFig. 13. Lowering the time of pretreatment with 300 mM ACPD to 5 min

(pt. 5) followed by 5 min in normal serum-free medium before the occur if astroglial cells were present in the cultures,co-exposure to 10 mM NMDA and 300 mM ACPD, did not eliminate the suggesting that the astroglial mGluR5 receptor is involvedACPD-mediated potentiation of NMDA-induced excitotoxicity. The PI in the potentiation [1]. As organotypic hippocampal sliceuptake in the two co-exposed groups were not significantly different. Data

cultures are rich in astroglial cells [7], the same mechanismare shown as means1S.E.M., with n57–16, *P,0.05, ***P,0.001could be active in our experiments. Involvement of theusing ANOVA with Bonferroni correction for comparison with cultures

exposed to 10 mM NMDA. mGluR5 receptor in the potentiation is supported by the


finding that 10 or 100 mM of the non-competitive mGluR5 NMDA (10 mM) potentiates NMDA-induced excitotoxici-antagonist MPEP abolished the ACPD-mediated potentia- ty in CA1 (and CA3) pyramidal cell layer, (4) that thistion of NMDA excitotoxicity whereas 10 or 100 mM potentiation can be eliminated by pretreating the culturesCPCCOEt, a non-competitive mGluR1 antagonist, had no with ACPD (300 mM) before the NMDA/ACPD co-expo-effect. Recently it has been suggested that MPEP in higher sure, (5) and that the potentiation can be abolished byconcentration (20–200 mM) can protect against NMDA blocking the mGluR5 receptor by MPEP (10 mM).excitotoxicity in cortical cultures possibly by non-competi-tive inhibition of the NMDA receptor [29]. By the addi-tional application of a lower 10 mM dose of MPEP we

Acknowledgementscould, however, exclude that the presently observed effectof MPEP, was due to inhibition of NMDA receptors.

Randi Godskesen and Inge Holst are gratefully acknow-The observed potentiation of NMDA-induced excitotox-ledged for technical help. Our gratitude also extends toicity was eliminated by pretreating the slice cultures with aDrs. Jens Noraberg and Morten Meyer for discussions andsubtoxic concentration of ACPD for 30 min, followed byhelp with the manuscript. The study was supported by thean additional 30 min in normal medium, before the ACPD/Danish MRC through the Neuroscience PharmaBiotecNMDA co-exposure. Related observations have been madeResearch Center, the EU Biotech Program (BIO4-in primary cortical cell cultures, where co-exposure to 30CT972307) and the Edvard Johnsen Foundation.mM NMDA and 100 mM DHPG for 10 min resulted in a

significant increase in excitotoxicity compared to NMDAexposure alone. Here pretreatment with 100 mM DHPG for1 min, with an additional 5 min interval before the co- Referencesexposure, significantly protected against the NMDA-in-duced excitotoxicity, suggesting a functional switch from

[1] G.P. Barbagallo, P. Dell’Albani, F. Nicoletti, A.M. Giuffrida-Stella,facilitation to inhibition of excitotoxicity when the cultures M.V. Catania, Activation of group I metabotropic glutamate re-were stimulated with DHPG twice within a limited period ceptors expressed in glial cells enhances excitotoxicity in neuronal

cell cultures, Soc. Neurosci. Abstr. 25 (1999) 1230.of time [25]. Such a switch has also been observed for[2] A. Barth, D.W. Newell, L.B. Nguyen, H.R. Winn, R. Wender, J.R.metabotropic glutamate receptor-mediated release of gluta-

Meno, D. Janigro, Neurotoxicity in organotypic hippocampal slicesmate from cortical synaptosomes [13] and for excitatorymediated by adenosine analogues and nitric oxide, Brain Res. 762

postsynaptic currents in hippocampal slices [38]. A time- (1997) 79–88.dependent switch from facilitation to inhibition could also [3] N. Best, L.E. Sundstrom, J. Mitchell, H.V. Wheal, Pre-exposure to

subtoxic levels prevents kainic acid lesions in organotypic hip-explain the effects of pretreatment seen in our model,pocampal slice cultures: effects of kainic acid on parvalbumin-which might involve time-dependent changes in the de-immunoreactive neurons and expression of heat shock protein 72sensitization of metabotropic glutamate receptors. Pretreat-following the induction of tolerance, Eur. J. Neurosci. 8 (1996)

ing cultures with ACPD for only 5 min followed by a 5 1209–1219.min interval in normal medium before co-exposure to ¨[4] V. Bruno, A. Copani, T. Knopfel, R. Kuhn, G. Casabona, P.

Dell’Albani, D.F. Condorelli, F. Nicoletti, Activation of metabot-NMDA and ACPD, did not affect the potentiating effect ofropic glutamate receptors coupled to inositol phospholipid hydrol-ACPD seen by direct ACPD/NMDA co-exposure (Fig.ysis amplifies NMDA-induced neuronal degeneration in cultured13). Before drawing final conclusions from the short pre-cortical cells, Neuropharmacology 34 (1995) 1089–1098.

treatment times, it must, however, be excluded that the [5] Q. Chen, D.J. Surmeier, A. Reiner, NMDA and non-NMDAlack of effect is not due to poor diffusion of ACPD into the receptor-mediated excitotoxicity are potentiated in cultured striatal

neurons by prior chronic depolarization, Exp. Neurol. 159 (1999)slice cultures, not leaving sufficient time for the desensiti-283–296.zation to occur.

[6] C.S. Colwell, K.L. Altemus, M.S. Levine, Metabotropic glutamatereceptor activation selectively limits excitotoxic damage in the intactneostriatum, Brain Res. 726 (1996) 223–226.

5. Conclusions [7] J.A. Del Rio, B. Heimrich, E. Soriano, H. Schwegler, M. Frotscher,Proliferation and differentiation of glial fibrillary acidic protein-immunoreactive glial cells in organotypic slice cultures of ratFrom the results of the present study we conclude (1)hippocampus, Neuroscience 43 (1991) 335–347.that ACPD in concentrations of 2 mM or higher induces a

[8] K.T. Eggeman, R.L. Lickteig, M.D. Browning, Protein levels ofdistinct, subfield-related and time- and concentration-de- NMDA receptor subunits NR2A and NR2B are high in rat hip-pendent pattern of neuronal degeneration in hippocampal pocampus, Soc. Neurosci. Abstr. 21 (1995) 83.slice cultures, (2) that the ACPD-induced neurodegenera- [9] S.M. Fitzjohn, A.J. Irving, M.J. Palmer, J. Harvey, D. Lodge, G.L.

Collingridge, Activation of group I mGluRs potentiates NMDAtion is NMDA receptor-mediated, since it can be complete-responses in rat hippocampal slices, Neurosci. Lett. 203 (1996)ly blocked by the NMDA receptor antagonist MK-801 (10211–213.

mM), while not significantly affected by the AMPA/kainic [10] M.V. Frantseva, P.L. Carlen, H. El-Beheiry, A submersion method toacid antagonist NBQX (20 mM), (3) that a non-toxic induce hypoxic damage in organotypic hippocampal cultures, J.concentration of ACPD (300 mM) applied together with Neurosci. Methods 89 (1999) 25–31.


¨[11] B.H. Gahwiler, Organotypic cultures of neural tissue, Trends agonist causes neurodegeneration in rat hippocampus, J. Hirnforsch.38 (1997) 317–322.Neurosci. 11 (1988) 484–489.

[31] L. Pastorino, F. Colciaghi, F. Gardoni, S. Albani-Torregrossa, D.E.[12] P. Henrich-Noack, K.G. Reymann, (1S,3R)-ACPD, a metabotropicPellegrini-Giampietro, F. Moroni, P.N. De Graan, F. Cattabeni, M.glutamate receptor agonist, enhances damage after global ischaemia,Di Luca, (1)-MCPG induces PKCe translocation in corticalEur. J. Pharmacol. 365 (1999) 55–58.synaptosomes through a PLD-coupled mGluR, Eur. J. Neurosci. 12[13] I. Herrero, M.T. Miras-Portugal, J. Sanchez-Prieto, Functional(2000) 1310–1318.switch from facilitation to inhibition in the control of glutamate

[32] D.E. Pellegrini-Giampietro, A. Cozzi, F. Peruginelli, P. Leonardi, E.release by metabotropic glutamate receptors, J. Biol. Chem. 273Meli, R. Pellicciari, F. Moroni, 1-Aminoindan-1,5-dicarboxylic acid(1998) 1951–1958.and (S)-(1)-2-(39-carboxybicyclo[1.1.1]pentyl)-glycine, two mGlu1[14] S.S. Hsu, D.W. Newell, A. Tucker, A.T. Malouf, H.R. Winn,receptor-preferring antagonists, reduce neuronal death in in vitro andAdenosinergic modulation of CA1 neuronal tolerance to glucosein vivo models of cerebral ischaemia, Eur. J. Neurosci. 11 (1999)deprivation in organotypic hippocampal cultures, Neurosci. Lett.3637–3647.178 (1994) 189–192.

[33] D.E. Pellegrini-Giampietro, S.A. Torregrossa, F. Moroni, Pharmaco-[15] B.H. Juurlink, L. Hertz, Ischemia-induced death of astrocytes andlogical characterization of metabotropic glutamate receptors coupled

neurons in primary culture: pitfalls in quantifying neuronal cellto phospholipase D in the rat hippocampus, Br. J. Pharmacol. 118

death, Dev. Brain Res. 71 (1993) 239–246.(1996) 1035–1043.

[16] B.W. Kristensen, J. Noraberg, B. Ebert, J. Zimmer, Excitotoxic[34] J.P. Pin, R. Duvoisin, The metabotropic glutamate receptors: struc-

profiles of ATPA and the classic glutamate receptor agonists AMPA, ture and functions, Neuropharmacology 34 (1995) 1–26.KA and NMDA in organotypic hippocampal slice cultures, Eur. J. [35] L.D. Pozzo Miller, N.K. Mahanty, J.A. Connor, D.M. Landis,Neurosci. 12 (Suppl.) (2000) 465. Spontaneous pyramidal cell death in organotypic slice cultures from

[17] B.W. Kristensen, J. Noraberg, B. Jakobsen, J.B. Gramsbergen, B. rat hippocampus is prevented by glutamate receptor antagonists,Ebert, J. Zimmer, Excitotoxic effects of non-NMDA receptor Neuroscience 63 (1994) 471–487.agonists in organotypic corticostriatal slice cultures, Brain Res. 841 [36] A.K. Pringle, F. Iannotti, G.J. Wilde, J.E. Chad, P.J. Seeley, L.E.(1999) 143–159. Sundstrom, Neuroprotection by both NMDA and non-NMDA

[18] D.J. Linden, M. Smeyne, J.A. Connor, trans-ACPD, a metabotropic receptor antagonists in in vitro ischemia, Brain Res. 755 (1997)receptor agonist, produces calcium mobilization and an inward 36–46.current in cultured cerebellar purkinje neurons, J. Neurophysiol. 71 [37] A.K. Pringle, L.E. Sundstrom, G.J. Wilde, L.R. Williams, F. Iannotti,(1994) 1992–1998. Brain-derived neurotrophic factor, but not neurotrophin-3, prevents

[19] D. Manahan-Vaughan, T. Behnisch, K.G. Reymann, ACPD-mediated ischaemia-induced neuronal cell death in organotypic rat hippocam-slow-onset potentiation is associated with cell death in the rat CA1 pal slice cultures, Neurosci. Lett. 211 (1996) 203–206.region in vivo, Neuropharmacology 38 (1999) 487–494. [38] A. Rodriguez-Moreno, A. Sistiaga, J. Lerma, J. Sanchez-Prieto,

[20] A. Matus, R. Bernhardt, R. Bodmer, D. Alaimo, Microtubule- Switch from facilitation to inhibition of excitatory synaptic transmis-associated protein 2 and tubulin are differently distributed in the sion by group I mGluR desensitization, Neuron 21 (1998) 1477–dendrites of developing neurons, Neuroscience 17 (1986) 371–389. 1486.

[21] J.W. McDonald, A.S. Fix, J.P. Tizzano, D.D. Schoepp, Seizures and [39] A.I. Sacaan, D.D. Schoepp, Activation of hippocampal metabotropicbrain injury in neonatal rats induced by 1S,3R-ACPD, a metabot- excitatory amino acid receptors leads to seizures and neuronalropic glutamate receptor agonist, J. Neurosci. 13 (1993) 4445–4455. damage, Neurosci. Lett. 139 (1992) 77–82.

[22] S. Miller, J.P. Kesslak, C. Romano, C.W. Cotman, Roles of [40] D.D. Schoepp, D.E. Jane, J.A. Monn, Pharmacological agents actingmetabotropic glutamate receptors in brain plasticity and pathology, at subtypes of metabotropic glutamate receptors, Neuropharmacolo-Ann. NY Acad. Sci. 757 (1995) 460–474. gy 38 (1999) 1431–1476.

[23] S. Miller, R.J. Bridges, A.R. Chamberlin, C.W. Cotman, Pharmaco- [41] D.D. Schoepp, B.G. Johnson, R.A. True, J.A. Monn, Comparison oflogical dissociation of glutamatergic metabotropic signal transduc- (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R-ACPD)-tion pathways in cortical astrocytes, Eur. J. Pharmacol. 269 (1994) and 1R,3S-ACPD-stimulated brain phosphoinositide hydrolysis, Eur.235–241. J. Pharmacol. 207 (1991) 351–353.

[24] N. Nakao, Z. Kokaia, P. Odin, O. Lindvall, Protective effects of [42] H. Shen, J.A. Gorter, E. Aronica, X. Zheng, L. Zhang, M.V.L.BDNF and NT-3 but not PDGF against hypoglycemic injury to Bennett, R.S. Zukin, Potentiation of NMDA responses by metabot-cultured striatal neurons, Exp. Neurol. 131 (1995) 1–10. ropic receptor activation depends on NMDA receptor subunit

[25] F. Nicoletti, V. Bruno, M.V. Catania, G. Battaglia, A. Copani, G.V. composition, Soc. Neurosci. Abstr. 21 (1995) 77.Cena, J. Sanchez-Prieto, P.F. Spano, M. Pizzi, Group-I metabotropic [43] R. Simantov, W. Liu, G. Broutman, M. Baudry, Antisense knoc-glutamate receptors: hypotheses to explain their dual role in kdown of glutamate transporters alters the subfield selectivity ofneurotoxicity and neuroprotection, Neuropharmacology 38 (1999) kainate-induced cell death in rat hippocampal slice cultures, J.1477–1484. Neurochem. 73 (1999) 1828–1835.

¨[26] F. Nicoletti, V. Bruno, A. Copani, G. Casabona, T. Knopfel, [44] A. Sistiaga, I. Herrero, F. Conquet, J. Sanchez-Prieto, The metabot-Metabotropic glutamate receptors: a new target for the therapy of ropic glutamate receptor 1 is not involved in the facilitation ofneurodegenerative disorders?, Trends Neurosci. 19 (1996) 267–271. glutamate release in cerebrocortical nerve terminals, Neurophar-

[27] J. Noraberg, J.B.P. Gramsbergen, F. Fonnum, J. Zimmer, Tri- macology 37 (1998) 1485–1492.methyltin (TMT) neurotoxicity in organotypic rat hippocampal slice [45] D.L. Small, R. Monette, B. Chakravarthy, J. Durkin, G. Barbe, G.cultures, Brain Res. 783 (1998) 305–315. Mealing, P. Morley, A.M. Buchan, Mechanisms of 1S,3R-ACPD-

[28] J. Noraberg, B.W. Kristensen, J. Zimmer, Markers for neuronal induced neuroprotection in rat hippocampal slices subjected todegeneration in organotypic slice cultures, Brain. Res. Prot. 3 (1999) oxygen and glucose deprivation, Neuropharmacology 35 (1996)278–290. 1037–1048.

¨[29] D.M. O’Leary, V. Movsesyan, S. Vicini, A.I. Faden, Selective [46] L. Stoppini, P.A. Buchs, D. Muller, A simple method for or-mGluR5 antagonists MPEP and SIB-1893 decrease NMDA or ganotypic cultures of nervous tissue, J. Neurosci. Methods 37glutamate-mediated neuronal toxicity through actions that reflect (1991) 173–182.NMDA receptor antagonism, Br. J. Pharmacol. 131 (2000) 1429– [47] U. Strasser, D. Lobner, M.M. Behrens, L.M. Canzoniero, D.W. Choi,1437. Antagonists for group I mGluRs attenuate excitotoxic neuronal

[30] W.Y. Ong, V.J. Balcar, Group I metabotropic glutamate receptor death in cortical cultures, Eur. J. Neurosci. 10 (1998) 2848–2855.


[48] A. Ugolini, M. Corsi, F. Bordi, Potentiation of NMDA and AMPA [52] Y. Wang, Z.H. Qin, M. Nakai, T.N. Chase, Glutamate metabotropicresponses by group I mGluR in spinal cord motoneurons, Neuro- receptor agonist 1S,3R-ACPD induces internucleosomal DNA frag-pharmacology 36 (1997) 1047–1055. mentation and cell death in rat striatum, Brain Res. 772 (1997)

[49] M.Vitale, L. Zamai, G. Mazzotti, A. Cataldi, E. Falcieri, Differential 45–56.kinetics of propidium iodide uptake in apoptotic and necrotic [53] A. Wenzel, J.M. Fritschy, H. Mohler, D. Benke, NMDA receptorthymocytes, Histochemistry 100 (1993) 223–229. heterogeneity during postnatal development of the rat brain: dif-

[50] J.J. Vornov, R.C. Tasker, J.T. Coyle, Direct observation of the ferential expression of the NR2A, NR2B, and NR2C subunitagonist-specific regional vulnerability to glutamate, NMDA, and proteins, J. Neurochem. 68 (1997) 469–478.

¨kainate neurotoxicity in organotypic hippocampal cultures, Exp. [54] J. Zimmer, B.H. Gahwiler, Cellular and connective organization ofNeurol. 114 (1991) 11–22. slice cultures of the rat hippocampus and fascia dentata, J. Comp.

[51] J.J. Vornov, R.C. Tasker, J.T. Coyle, Delayed protection by MK-801 Neurol. 228 (1984) 432–446.and tetrodotoxin in a rat organotypic hippocampal culture model ofischemia, Stroke 25 (1994) 457–464.

the metabotropic glutamate receptor agonist 1s,3r-acpd stimulates and modulates nmda receptor...

Documents