Exp Brain Res (1990) 82:451-455

Exp.erimental BrainResearch 9 Springer-Verlag1990

Research Note

Role of N M D A and non-NMDA receptors in synaptic transmission in rat piriform cortex M.W. Jung, J. Larson, and G. Lynch Center for the Neurobiology of Learning and Memory, University of California, Irvine, CA 92717, USA Received February 14, 1990 / AcceptedMay 25, 1990

Summary. The pharmacology of synaptic transmission was studied in slices of rat piriform cortex using the selective non-NMDA glutamate receptor antagonist 6,7dinitroquinoxaline-2,3-dione (DNQX) and the selective NMDA receptor antagonist D-2-amino-5-phosphonopentanoate (D AP5). DNQX produced a dose-dependent blockade of synaptic transmission at both lateral olfactory tract and associational system synapses with half-maximal effects at about 2.5 gM. D-AP5 had no significant effects on field potentials recorded in medium containing 2.5 mM Mg ++. However in low Mg ++ (100 200 gM) medium, D A P 5 did reduce a slow component of postsynaptic responses in both synaptic systems. In Mg + +-free medium, 20 gM DNQX did not completely block transmission; the remaining response components were blocked by D-AP5. These results suggest that normal synaptic transmission in the two main inputs to the superficial layers of piriform cortex is mediated by non-NMDA receptors but that N M D A receptors can also participate under conditions where the Mg + + block of the N M D A channel is alleviated.

Key words: Glutamate receptors - N M D A receptor Olfactory cortex

DNQX-AP5

Rat

Introduction Excitatory amino acid receptors are thought to mediate excitatory synaptic transmission in a large number of pathways in the mammalian central nervous system (for review see Mayer and Westbrook 1987). These receptors can be broadly classified as those selectively activated by N-methyl-D-aspartate (NMDA receptors) and others (non-NMDA receptors) which include quisqualate and O{l)~rint requests to. J. Larson (address see above)

kainate receptors. In hippocampus, N M D A receptors do not normally participate in normal (i.e., low frequency stimulation) transmission in the generation of fast excitatory post-synaptic potentials (EPSPs); this mode of transmission appears to involve non-NMDA receptors (Collingridge et al. 1983; Blake et al. 1988). However, there are synapses in neocortex that appear to have only N M D A receptors (Thomson et al. 1985) and N M D A receptors do mediate responses under certain unusual conditions (low extracellular Mg + +, blockade of synaptic inhibition, and high frequency activation) in hippocampus (Coan and Collingridge 1985; Herron et al. 1986; Larson and Lynch 1988; Muller et al. 1989). In the latter case, activation of the N M D A system is attributed to a relief of the channel block by Mg ++ (Mayer et al. 1984; Nowak et al. 1984) when sufficient postsynaptic depolarization occurs. Synaptic potentials generated by N M D A receptors are also much slower than those generated by non-NMDA receptors (Forsythe and Westbrook 1988; Muller et al. 1988). Piriform cortex shares a number of features with neocortex but has a much simpler anatomical organization that facilitates the analysis of functional circuits (Haberly 1985; Lynch 1986). Excitatory afferent input to the layer II cells of piriform cortex arrrives via the lateral olfactory tract (LOT) which terminates in layer Ia; an excitatory associational (ASSN) system innervates the same dendrites in layer Ib. The pharmacology of these synapses is not well understood, partly because of a lack of potent and selective antagonists for non-NMDA receptors. Collins (1982) reported that the selective NMDA antagonist, DL-2-amino-5-phosphonopentanoate (DL-AP5), the less selective N M D A antagonist, D-~-aminoadipate (D-~-AA), and the broadspectrum antagonist, y-D-glutamylglycine (y-DGG) had no effects on LOT evoked field potentials. LOT responses were antagonized by the broad spectrum antagonists, cis-2,3-piperidine dicarboxylate (cis-PDA) and DL-2-

452 amino-4-phosphonobutyrate (DL-AP4). Hori et al. (1982) and ffrench-Mullen et al. (1986) also found that D L - A P 4 was effective in blocking the LOT response. In the latter study, the selective N M D A antagonists DL-AP5 and DL-2-amino-7-phosphonoheptanoate ( D L - A P 7 ) had no effect. Hearn et al. (1986) reported antagonism of the LOT response with several broad spectrum antagonists, including L-AP4. These results are consistent with a role for n o n - N M D A receptors mediating post-synaptic responses to LOT stimulation. It has been suggested (ffrench-Mullen et al. 1985) that the LOT transmitter may be a dipeptide since D L - A P 4 antagonizes postsynaptic responses to exogenously applied N-acetyl-aspartyl-glutamate ( N A A G ) but not exogenous glutamate or aspartate. However, whether or not exogenously applied agonists activate synaptic receptors (as opposed to extrasynaptic receptors) is uncertain and L - A P 4 is also thought to have a presynaptic inhibitory action (Harris and Cotman 1983). Moreover, recent studies suggest that N A A G may not be excitatory on piriform neurons (Whittemore and Koerner 1989). The associational (ASSN) system has received less attention. Collins (1982) reported that D L - A P 5 , D-mAA, and y - D - G G all antagonized a late component of the LOT evoked potential that was attributed to disynaptic activation of the feedback ASSN fibers. Collins and Howlett (1988) recently reported similar observations for the monosynaptic ASSN EPSP evoked by stimulation of the deep layers of cortex, however in these experiments both theophylline (to enhance transmitter release) and picrotoxin (to block post-synaptic inhibition) were also present. Both of these agents would be expected to augment N M D A responses. Recently, more potent and selective n o n - N M D A receptor antagonists have been developed (Honor6 et al. 1988). The effects of 6,7-dinitro-quinoxaline-2,3-dione ( D N Q X ) on monosynaptic EPSPs generated by LOT and ASSN inputs to layer II piriform cortex have been investigated in the present paper. D N Q X is a potent antagonist of excitation evoked by quisqualate and kainate having minimal effects on N M D A excitation in spinal cord (Honor6 et al. 1988) and potently blocks low frequency synaptic transmission in hippocampus (Muller et al. 1988). D N Q X was found to block EPSPs generated by LOT or ASSN synapses, indicating that transmission in these two systems is mediated by n o n - N M D A receptors. However, removal of extracellular magnesium was found to reveal D - A P 5 sensitive and DNQX-resistant responses ( N M D A receptor components) in both systems.

Material and methods Experiments were conducted on slices of olfactory cortex prepared from male Sprague-Dawley rats (175-225 g) and maintained in vitro. Dissections of tissue blocks containing piriform cortex were performed in a dish containing ice-cold medium. Slices approximately 400 gm thick were cut from the region of piriform cortex underlying the lateral olfactory tract (LOT) on a tissue chopper at an angle perpendicular to the cortical surface. Tissue was maintained at 35+ 1~ C and constantly perfused with medium in an

interface chamber with the upper surface of each slice exposed to an atmosphere of humidified 95% 02/5% CO 2. The standard perfusion medium contained (in mM): NaC1124, KC13, KH2PO4 1.2, CaC1z 3.4, MgSOr 2.5, NaHCO 3 26, D-glucose 10, and L-ascorbate 2. In some experiments, the MgSO4 was either eliminated or reduced to 100-200 pM; no other adjustments were made in these cases. The rather high concentrations of divalent cations in the standard medium were used to facilitate comparison of results with experiments investigating long-term potentiation (LTP) at these synapses (Jung, et al. in press). Bipolar stimulation electrodes were constructed of twisted strands of nichrome wire (64 gm dia.) insulated except at the cut ends. Extracellular recordings of field potentials were made with glass capillary pipettes filled with 2M NaC1 (1 5 Mf~). DNQX was obtained from Tocris (Buckhurst Hill, U.K.) and D-AP5 was purchased from Sigma (St. Louis, MO). Field potentials evoked by LOT and associational (ASSN) fibers were distinguished by three criteria: 1) Stimulation and recording electrodes were positioned under visual guidance in the LOT (layer Ia) or in layer Ib (ASSN fibers). The LOT was always clearly visible as a fiber tract in these slices and the cell layer containing the somata of layer II cells could usually be distinguished from layer Ib. 2) Stimulation in the LOT and layer Ib evoked distinct laminar profiles of responses across the layers of the cortex. 3) Robust paired-pulse facilitation with inter-pulse intervals of 50 ms could be observed at LOT synapses but not ASSN synapses (see Bower and Haberly 1986). Once the positions of stimulating electrodes were set, the position of each recording electrode was adjusted to maximize the amplitude of the dendritic population EPSP in the appropriate layer. In the majority of experiments, both an LOT input and an ASSN input were studied in the same slice; however in some cases either two independent LOT or ASSN inputs were used. Independence of two LOT inputs was assured by the lack of pairedpulse facilitation between the two inputs at an inter-pulse interval of 50 ms; independence of ASSN inputs was confirmed by the absence of paired-pulse depression between the inputs at an interpulse interval of 400 ms. The test for lack of paired-pulse facilitation is standard for assessing independence of stimulation electrodes in systems exhibiting facilitation; paired-pulse depression in the ASSN system also appears to be a homosynaptic effect (Jung, Larson, and Lynch, unpublished observations). Synaptic responses were routinely tested at 20-40 s intervals and stable baselines were established for 10-20 min before drug treatments were undertaken. Concentrated drug solutions were diluted to final concentrations by infusion into the perfusion medium. Effects of DNQX were assessed 30-60 rain after onset of infusion, those of D AP5 were assessed after 20 30 min.

Results The effects of the n o n - N M D A antagonist, D N Q X , were tested on 20 slices. Stimulus intensity was set to evoke a dendritic field EPSP 50-75 % of the maximum obtainable in the slice. Baseline recordings were taken for 10-20 min prior to drug administration. Results from one experiment are shown in Fig. 1A. Addition of 20 laM D N Q X to the standard perfusion medium completely abolished postsynaptic responses generated by LOT or ASSN inputs. Recovery from D N Q X after its removal from the perfusion medium was very slow as described elsewhere (Muller et al. 1988); however in some cases 2-3 h washout did result in the complete recovery of both LOT and ASSN responses (data not shown). Dose-response data are shown in Fig. IB. Antagonism o f the initial slope of the field EPSP was assessed 1 h after addition of 0.2-20pM D N Q X . At less than 0.2 pM, D N Q X was

453 A CONTROL LOT

SUPERIMPOSED

20uM DNQX

PPF

~

[_ ASSN

loo

z~

5o

T ~ ( ~

O~ 0

.

, 5

.

, 10

.

---0---

ASSN

, 15

.

, 20

DNQX(uM) C

CONTROL

50 uM D-AP5

SUPERIMPOSED

L_

Fig. IA-C. Effects of DNQX and D ~ P 5 on LOT and ASSN responses in control medium. A Field EPSPs recorded in response to LOT and ASSN fiber stimulation in one experiment before and during perfusion of 20 gM DNQX. These results are representative of 4 slices with LOT stimulation and 4 slices with ASSN stimulation. PPF traces show responses to paired-pulse facilitation stimulation (solid lines control, dotted lines - second responses 50 ms later). Calibration bars: 0.5 mV and 5 ms for records from DNQX experiment and 1.0 mV and 5 ms for PPF records. B Dose response curves for DNQX antagonism at LOT and ASSN synapses. C Field EPSPs recorded in response to LOT and ASSN fiber stimulation before and during perfusion with 50 gM D-AP5. Similar results were seen in 8 slices with LOT stimulation and 5 slices with ASSN stimulation. Calibration bars: 1 mV and 5 ms

ineffective and 20 laM was sufficient to block either response completely. The dose-response curves for antagonism of both LOT and ASSN responses were similar with a half-maximal effect at about 2.5 I.tM. These results indicate that under normal conditions both L O T and ASSN evoked EPSPs are mediated by n o n - N M D A receptors. This conclusion was further tested by examining the effects of the selective N M D A antagonist, D-AP5. At a concentration of 50 gM, D AP5 had no detectable effect on either the initial slope or other waveform parameters (peak amplitude, area or half-width) of responses evoked by moderate intensity stimulation of either LOT or ASSN fibers (Fig. 1C). This dosage of D - A P 5 has been shown to completely suppress NMDA-mediated responses in hippocampus (Muller et al. 1988). The ionophore associated with the N M D A receptor is blocked in a voltage-sensitive manner by magnesium ions (Mayer et al. 1984; Nowak et al. 1984); removal of extracellular Mg § has been shown to reveal N M D A receptor-mediated responses in a variety of systems (for review see Mayer and Westbrook 1987). The effects of perfusion with low Mg ++ (100-200 gM) medium on LOT and ASSN responses are shown in Fig. 2A, B. Low

Mg + +, rather than Mg + +-free medium was used in these experiments to prevent a gradual potentiation of responses sometimes seen in Mg++-free medium when D N Q X was not present. Lowering Mg + + caused a small increase in the peak amplitude of both responses as well as a prolongation of the decay of the EPSPs. This latter effect was particularly evident when high stimulation intensities were used. Addition of 50 gM D - A P 5 reduced the late component of the response but had smaller effects on the increase in EPSP amplitude seen in low Mg ++ medium. This increase probably reflects an increase in transmitter release. Statistical analysis of the data in Fig. 2B yielded the following results (at a significance level o f p = 0.05): LOT and ASSN amplitudes and half-widths were increased in low Mg + +, D - A P 5 significantly reduced LOT amplitude but not ASSN amplitude, and D - A P 5 significantly reduced both L O T and ASSN half-widths. In Mg + +-free medium, application of 20 gM D N Q X did not abolish either synaptic response, but the remaining EPSP was of a much slower time-course than the typical response (Fig. 2C). These responses were blocked by 50 gM D - A P 5 and thus most likely represent EPSPs mediated mainly by N M D A receptors. Interestingly, ASSN responses under these conditions exhibited pairedpulse facilitation (data not shown), an effect not normally observed (see Fig. 1A). The explanation for this phenomenon is presently unclear.

Discussion The results of the present study indicate that both the afferent (LOT) and associational synapses in piriform cortex are pharmacologically similar to those found in the CA1 field of hippocampus. Thus, N M D A receptors do not contribute to low frequency responses, but can be activated by released transmitter when the extracellular magnesium concentration is reduced. The normal low frequency response can be blocked totally by the nonN M D A antagonist, D N Q X , suggesting that it is mediated by quisqualate and/or kainate receptors. The dose-response data also indicate a similarity to CA1 synapses since 20 ~tM D N Q X completely abolishes responses in both areas, although extensive dose-response data for CA1 are not available. Regarding the afferent LOT synapses, the present results are in agreement with previous reports (Collins 1982; ffrench-Mullen et al. 1986) in that N M D A receptor antagonists had no effect on single-pulse (low frequency) responses in 2.5 m M Mg + + and further demonstrate that these responses can be antagonized completely by the selective n o n - N M D A antagonist, DNQX. Recently, Collins and Buckley (1989) also have shown that LOT responses in mouse piriform cortex in 1 m M Mg ++ are antagonized by D N Q X and they report dose-response data very similar to those described here. With respect to the associational synapses, the present study is the first pharmacological analysis of monosynaptic responses in the absence of G A B A receptor antagonists. Three criteria were used to ensure that ASSN synapses were

454

A LOT CONIROL .........

LOW Mg+AP5 ......... kOWMg

mWMg

~

ASSN CONTROL .........

LOWgg

- - LOW Mg+AP5 ......... LOWMg

.....'..~-

-f

B 200.

9 []

CONTROL LOW Mg

[]

AP5

150.

200

.-I

on-

9 [] []

150.

CONTROL LOWMg AP5

I-Z

o

100.

O

50.

100.

50.

0

0 AMPLITUDE

HALF-WIDTH

AMPLITUDE

HALF-WIDTH

C

Mg-FREE+DNQX LOT

ASSN

Mg-FREE+DNQX +D-AP5

SUPERIMPOSED

uk_.____----K--

Fig. 2A-C. Effect of D-AP5 on LOT and ASSN responses recorded in low Mg + + (100500 gM) medium. A Records obtained under three conditions: control medium (2.5 mM Mg++), low Mg ++, and low M g + + + D - A P 5 (501.tM). Calibration bars: 1.0 mV and 5 ms. B Measurements of response peak amplitude and half-width (expressed as percent of control medium) under

the three conditions. These results are based on 6 slices with LOT stimulation and 9 slices with A~SSNstimulation. C Effect of D-AP5 (50 gM) on responses recorded in Mg+ +-free medium containing 20 gM DNQX. Initial sharp deflections in the responses are presynaptic fiber volleys. Calibration bars: 0.5 mV and 5 ms. Similar results were obtained from 4 LOT inputs and 9 ASSN inputs

distinguished from LOT synapses, the most useful being the lack of paired-pulse facilitation. These synapses appear to be very similar to the LOT synapses in terms of receptor pharmacology. The implication of N M D A receptors in previous studies (Collins 1982; Collins and Howlett 1988) is confounded by disynaptic responses, unclear localization, and the presence of GABA and adenosine antagonists in the medium. N M D A receptors have been linked strongly to the genesis of several important processes including patterns of rhythmic activity (Grillner et al. 1987), epileptiform activity, excitotoxic damage (Rothman and Olney 1987), and LTP of synaptic transmission (Collingridge et al. 1983). Certain parts of olfactory cortex are highly susceptible to kindling (Racine et al. 1983) and convulsions

elicited by excitotoxins (Gale et al. 1988). These could involve activation of N M D A receptors at LOT or ASSN synapses. An interesting recent report by Hoffman and Haberly (1989) showed that large all-or-none EPSPs, which could represent epileptiform "giant EPSPs", could be induced in piriform neurons following spontaneous bursting activity in Mg + +-free medium. The induction, but not expression, of these late EPSPs was prevented by N M D A antagonists. These findings are reminiscent of the recent finding that hippocampal LTP is not expressed by N M D A receptors although their activation is necessary for LTP induction (Kauer et al. 1988; Muller et al. 1988). The olfactory system uses highly rhythmic sampling patterns; for example, during exploratory sniffing rats

455 inhale at 4 - 7 H z a n d these r e s p i r a t i o n s b e c o m e sync h r o n i z e d with h i p p o c a m p a l E E G t h e t a r h y t h m s a n d p r e s u m a b l y activity in o t h e r o l f a c t o r y structures ( M a crides et al. 1982). I n d u c t i o n o f L T P in h i p p o c a m p u s is i n d u c e d m o s t efficiently b y high f r e q u e n c y b u r s t s given at the t h e t a f r e q u e n c y ( L a r s o n et al. 1986); this t y p e o f activity p r o d u c e s a sequence o f p h y s i o l o g i c a l m e c h a n isms t h a t m a x i m i z e N M D A r e c e p t o r a c t i v a t i o n ( L a r s o n a n d L y n c h 1986, 1988). R h y t h m i c s t i m u l a t i o n o f the L O T can be used as a l e a r n i n g cue in o l f a c t o r y discrimin a t i o n t a s k s a n d u n d e r certain c o n d i t i o n s l e a r n i n g o f the "electric o d o r " is a s s o c i a t e d with p o t e n t i a t i o n o f the synapses a c t i v a t e d b y the s t i m u l a t i o n ( R o m a n et al. 1987). P r e l i m i n a r y results suggest t h a t b o t h the L O T a n d A S S N synapses c a n exhibit L T P after t h e t a b u r s t stimu l a t i o n in v i t r o a n d in b o t h cases the i n d u c t i o n event is d e p e n d e n t o n N M D A r e c e p t o r a c t i v a t i o n (Jung et al. in press).

Acknowledgement. This research was supported by the Office of Naval Research grant no. N00014-86-K-0333.

References Blake JF, Brown MW, Collingridge GL (1988) CNQX blocks acidic amino acid induced depolarizations and synaptic components mediated by non-NMDA receptors in rat hippocampal slices. Neurosci Lett 89:182-186 Bower JM, Haberly LB (1986) Facilitating and nonfacilitating synapses on pyramidal cells: a correlation between physiology and morphology. Proc Natl Acad Sci USA 83:1115-1119 Coan EJ, Collingridge GL (1985) Magnesium ions block an N-methyI-D-aspartate receptor-mediated component of synaptic transmission in rat hippocampus. Neurosci Lett 53:21 26 Collingridge GL, Kehl SJ, McLennan H (1983) Excitatory amino acids in synaptic transmission in the Schaffer-commissural pathway of the rat hippocampus. J Physiol (Lond) 334:33-46 Collins GGS (1982) Some effects of excitatory amino acid receptor antagonists on synaptic transmission in the rat olfactory cortex slice. Brain Res 244:311-318 Collins GGS, Buckley KS (1989) Antagonism of monosynaptic excitations in the mouse olfactory cortex slice by 6,7-dinitroquinoxaline-2,3-dione. Neuropharmacology 28:1123-1128 Collins GGS, Howlett SJ (1988) The pharmacology of excitatory transmission in the rat olfactory cortex slice. Neuropharmacology 27 : 697-705 ffrench-Mullen JMH, Hori N, Carpenter DO (1986) Receptors for excitatory amino acids on neurons in rat pyriform cortex. J Neurophysiol 55:1283 1294 ffrench-Mullen JMH, Koller K, Zaczek R, Coyle JT, Hori N, Carpenter DO (1985) N-acetylaspartylglutamate: possible role as the neurotransmitter of the lateral olfactory tract. Proc Natl Acad Sci USA 82:3897-3900 Forsythe ID, Westbrook GL (1988) Slow excitatory postsynaptic currents mediated by N-methyl-D-aspartate receptors on cultured mouse central neurones. J Physiol (Lond) 396:515-533 Gale K, Zhong P, Miller LP, Murray TF (1988) A crucial role for excitatory amino acids in the initiation and control of convulsions elicited from %rea tempestas". In: Cavalheiro EA, Lehmann J, Turski L (eds) Frontiers in excitatory amino acid research. Liss, New York, pp 211-218 Grillner S, Wallen P, Dale N, Brodin L, Buchanan J, Hill R (1987) Transmitters, membrane properties and network circuitry in the control of locomotion in lamprey. Trends Neurosci 10:3441 Haberly LB (1985) Neuronal circuitry in olfactory cortex: anatomy and functional implications. Chem Senses 10:219-238

Harris EW, Cotman CW (1983) Effects of acidic amino acid antagonists on paired-pulse potentiation at the lateral perforant path. Exp Brain Res 52:455-460 Hearn T J, Ganong AH, Cotman CW (1986) Antagonism of lateral olfactory tract synaptic potentials in rat prepyriform cortex slices. Brain Res 379:372 376 Herron CE, Lester RAJ, Coan EJ, Collingridge GL (1986) Frequency-dependent involvement of NMDA receptors in the hippocampus: a novel synaptic mechanism. Nature 322 : 265-268 Hoffman WH, Haberly LB (1989) Bursting induces persistent allor-none EPSPs by an NMDA-dependent process in piriform cortex. J Neurosci 9:206-215 Honor6 T, Davies SN, Drejer J, Fletcher EJ, Jacobsen P, Lodge D, Nielsen FE (1988) Quinoxalinediones: potent competitive nonNMDA glutamate receptor antagonists. Science 241:701-703 Hori N, Auker CR, Braitman DJ, Carpenter DO (1982) Pharmacologic sensitivity of amino acid responses and synaptic activation of in vitro prepyriform neurons. J Neurophysiol 48:1289-1301 Jung MW, Larson J, Lynch, G (in press) Long-term potentiation of monosynaptic EPSPS in rat piriform cortex in vitro. Synapse. Kauer JA, Malenka RC, Nicoll RA (1988) A persistent postsynaptic modification mediates long-term potentiation in the hippocampus. Neuron 1:911-917 Larson J, Lynch G (1986) Induction of synaptic potentiation in hippocampus by patterned stimulation involves two events. Science 232:985-988 Larson J, Lynch G (1988) Role of N-methyl-D-aspartate receptors in the induction of synaptic potentiation by burst stimulation patterned after the hippocampal theta rhythm. Brain Res 441:111-118 Larson J, Wong D, Lynch G (1986) Patterned stimulation at the theta frequency is optimal for induction of long-term potentiation. Brain Res 368:347-350 Lynch G (1986) Synapses, circuits, and the beginnings of memory. MIT Press, Cambridge, MA Macrides F, Eichenbaum HB, Forbes WB (1982) Temporal relationship between sniffing and the limbic theta rhythm during odor discrimination reversal learning. J Neurosci 2:1705 1717 Mayer ML, Westbrook GL (1987) The physiology of excitatory amino acids in the vertebrate central nervous system. Prog Neurobiol 28:197 276 Mayer ML, Westbrook GL, Guthrie PB (1984) Voltage-dependent block by Mg 2§ of NMDA responses in spinal cord neurons. Nature 309:261-263 Muller D, Joly M, Lynch G (1988) Contributions of quisqualate and NMDA receptors to the induction and expression of LTP. Science 242:1694-1697 Muller D, Larson J, Lynch G (1989) The NMDA receptor mediated components of responses evoked by patterned stimulation are not increased by long-term potentiation. Brain Res 477: 369-399 Nowak L, Bregestovski P, Ascher P, Herbet A, Prochiantz A (1984) Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307 : 462~465 Racine RJ, Milgram NW, Hafner S (1983) Long-term potentiation phenomena in the rat limbic forebrain. Brain Res 260:217-231 Roman F, Staubli U, Lynch G (1987) Evidence for synaptic potentiation in a cortical network during learning. Brain Res 418:221 226 Rothman SM, Olney JW (1987) Excitotoxicity and the NMDA receptor. Trends Neurosci 10:299-302 Thomson AM, West DC, Lodge D (1985) An N-methylaspartate receptor-mediated synapse in rat cerebral cortex: a site of action of ketamine ? Nature 313 : 479481 Whittemore ER, Koerner JF (1989) An explanation for the purported excitation of piriform cortical neurons by N-acetyl-Laspartyl-L-glutamic acid (NAAG). Proc Natl Acad Sci USA 86: 9602-9605