Glutamate and the neural basis of the subjective effects of ketamine

Context: Ketamine evokes psychosislike symptoms, and its primary action is to impair N-methyl-D-aspartate glutamate receptor neurotransmission, but it also induces secondary increases in glutamate release. Objectives: To identify the sites of action of ketamine in inducing symptoms and to determine the role of increased glutamate release using the glutamate release inhibitor lamotrigine. Design: Two experiments with different participants were performed using a double-blind, placebo-controlled, randomized, crossover, counterbalanced-order design. In the first experiment, the effect of intravenous ketamine hydrochloride on regional blood oxygenation level dependent (BOLD) signal and correlated symptoms was compared with intravenous saline placebo. In the second experiment, pretreatment with lamotrigine was compared with placebo to identify which effects of ketamine are mediated by increased glutamate release. Setting: Wellcome Trust Clinical Research Facility, Manchester, England. Participants: Thirty-three healthy, right-handed men were recruited by advertisements. Interventions: In experiment 1, participants were given intravenous ketamine (1-minute bolus of 0.26 mg/ kg, followed by a maintenance infusion of 0.25 mg/ kg/ h for the remainder of the session) or placebo (0.9% saline solution). In experiment 2, participants were pretreated with 300 mg of lamotrigine or placebo and then were given the same doses of ketamine as in experiment 1. Main Outcome Measures: Regional BOLD signal changes during ketamine or placebo infusion and Brief Psychiatric Rating Scale and Clinician- Administered Dissociative States Scale scores. Results: Ketamine induced a rapid, focal, and unexpected decrease in ventromedial frontal cortex, including orbitofrontal cortex and subgenual cingulate, which strongly predicted its dissociative effects and increased activity in mid- posterior cingulate, thalamus, and temporal cortical regions (r= 0.90). Activations correlated with Brief Psychiatric Rating Scale psychosis scores. Lamotrigine pretreatment prevented many of the BOLD signal changes and the symptoms. Conclusions: These 2 changes may underpin 2 fundamental processes of psychosis: abnormal perceptual experiences and impaired cognitive- emotional evaluation of their significance. The results are compatible with the theory that the neural and subjective effects of ketamine involve increased glutamate release.

Visualising the activity of the cystine-glutamate antiporter in glial cells using antibodies to aminoadipic acid, a selectively transported substrate

The cystine-glutamate antiporter is a transport system that facilitates the uptake of cystine, concomitant with the release of glutamate. The cystine accumulated by this transporter is generally considered for use in the formation of the cysteine-containing antioxidant glutathione, which is abundant in many glial cells. This study used the simple strategy of generating an antibody to aminoadipic acid, a selective substrate for the cystine-glutamate antiporter. Stereospecific accumulation of aminoadipic acid into specific cell types in rat brain slice preparations was detected immunocytochemically. Strong accumulation was detected in astroglial cells in all brain regions studied including those in white matter tracts. Strong accumulation into radial glial cells, including the retinal Muller cells and the Bergmann glial cells was also observed. Glial accumulation was observed not only in cells within the blood brain barrier, but also outside such; anterior pituitary folliculostellate cell and intermediate lobe pituitary glial cells exhibited strong accumulation of aminoadipic acid. Interestingly, some glial cells such as the posterior pituitary glial cells (pituicytes) exhibited very little if any accumulation of aminoadipic acid. Within the brain labelling was not uniform. Particularly strong labelling was noted in some regions, such as the glial cells surrounding the CA1 pyramidal cells. By contrast, neurons never exhibited uptake of aminoadipic acid. Because cystine uptake is associated with glutamate release, it is suggested that this antiporter might contribute to release of glutamate from glial cells under some pathophysiological conditions. (C) 2001 Wiley-Liss, Inc.

Reciprocal changes in forebrain contents of glycogen and of glutamate/glutamine during early memory consolidation in the day-old chick

Passive avoidance learning is with advantage studied in day-old chicks trained to distinguish between beads of two different colors, of which one at training was associated with aversive taste. During the first 30-min post-training, two periods of glutamate release occur in the forebrain. One period is immediately after the aversive experience, when glutamate release is confined to the left hemisphere. A second release, 30 min later, may be bilateral, perhaps with preponderance of the right hemisphere. The present study showed increased pool sizes of glutamate and glutamine, specifically in the left hemisphere, at the time when the first glutamate release occurs, indicating de novo synthesis of glutamate/glutamine from glucose or glycogen, which are the only possible substrates. Behavioral evidence that memory is extinguished by intracranial administration at this time of iodoacetate, an inhibitor of glycolysis and glycogenolysis, and that the extinction of memory is counteracted by injection of glutamine, supports this concept. A decrease in forebrain glycogen of similar magnitude and coinciding with the increase in glutamate and glutamine suggests that glycogen rather than glucose is the main source of newly synthesized glutamate/glutamine. The second activation of glutamatergic activity 30 min after training, when memory is consolidated into stable, long-term memory, is associated with a bilateral increase in pool size of glutamate/glutamine. No glycogenolysis was observed at this time, but again there is a temporal correlation with sensitivity to inhibition by iodoacetate and rescue by glutamine, indicating the importance of de novo synthesis of glutamate/glutamine from glucose or glycogen. (C) 2003 Elsevier B.V All rights reserved.

CXCR4-mediated glutamate exocytosis from astrocytes.

The role of astrocytes as structural and metabolic support for neurons is known since the beginning of the last century. Because of their strategic localization between neurons and capillaries they can monitor and control the level of synaptic activity by providing energetic metabolites to neurons and remove excess of neurotransmitters. During the last two decades number of papers further established that the astrocytic plasma-membrane G-protein coupled receptors (GPCR) can sense external inputs (such as the spillover of neurotransmitters) and transduce them as intracellular calcium elevations and release of chemical transmitters such as glutamate. The chemokine CXCR4 receptor is a GPCR widely expressed on glial cells (especially astrocytes and microglia). Activation of the astrocytic CXCR4 by its natural ligand CXCL12 (or SDF1 alpha) results in a long chain of intracellular and extracellular events (including the release of the pro-inflammatory cytokine TNFalpha and prostanglandins) leading to glutamate release. The emerging role of CXCR4-CXCL12 signalling axis in brain physiology came from the recent observation that glutamate in astrocytes is released via a regulated exocytosis process and occurs with a relatively fast time-scale, in the order of few hundred milliseconds. Taking into account that astrocytes are electrically non-excitable and thus exocytosis rely only on a signalling pathway that involves the release Ca(2+) from the internal stores, these results suggested a close relationship between sites of Ca(2+) release and those of fusion events. Indeed, a recent observation describes structural sub-membrane microdomains where fast ER-dependent calcium elevations occur in spatial and temporal correlation with fusion events.

Astrocytes contain a vesicular compartment that is competent for regulated exocytosis of glutamate.

Astrocytes establish rapid cell-to-cell communication through the release of chemical transmitters. The underlying mechanisms and functional significance of this release are, however, not well understood. Here we identify an astrocytic vesicular compartment that is competent for glutamate exocytosis. Using postembedding immunogold labeling of the rat hippocampus, we show that vesicular glutamate transporters (VGLUT1/2) and the vesicular SNARE protein, cellubrevin, are both expressed in small vesicular organelles that resemble synaptic vesicles of glutamatergic terminals. Astrocytic vesicles, which are not as densely packed as their neuronal counterparts, can be observed in small groups at sites adjacent to neuronal structures bearing glutamate receptors. Fluorescently tagged VGLUT-containing vesicles were studied dynamically in living astrocytes by total internal reflection fluorescence (TIRF) microscopy. After activation of metabotropic glutamate receptors, astrocytic vesicles underwent rapid (milliseconds) Ca(2+)- and SNARE-dependent exocytic fusion that was accompanied by glutamate release. These data document the existence of a Ca(2+)-dependent quantal glutamate release activity in glia that was previously considered to be specific to synapses.

P2Y1 receptor-evoked glutamate exocytosis from astrocytes: control by tumor necrosis factor-alpha and prostaglandins.

ATP, released by both neurons and glia, is an important mediator of brain intercellular communication. We find that selective activation of purinergic P2Y1 receptors (P2Y1R) in cultured astrocytes triggers glutamate release. By total internal fluorescence reflection imaging of fluorescence-labeled glutamatergic vesicles, we document that such release occurs by regulated exocytosis. The stimulus-secretion coupling mechanism involves Ca2+ release from internal stores and is controlled by additional transductive events mediated by tumor necrosis factor-alpha (TNFalpha) and prostaglandins (PG). P2Y1R activation induces release of both TNFalpha and PGE2 and blocking either one significantly reduces glutamate release. Accordingly, astrocytes from TNFalpha-deficient (TNF(-/-)) or TNF type 1 receptor-deficient (TNFR1(-/-)) mice display altered P2Y1R-dependent Ca2+ signaling and deficient glutamate release. In mixed hippocampal cultures, the P2Y1R-evoked process occurs in astrocytes but not in neurons or microglia. P2Y1R stimulation induces Ca2+ -dependent glutamate release also from acute hippocampal slices. The process in situ displays characteristics resembling those in cultured astrocytes and is distinctly different from synaptic glutamate release evoked by high K+ stimulation as follows: (a) it is sensitive to cyclooxygenase inhibitors; (b) it is deficient in preparations from TNF(-/-) and TNFR1(-/-) mice; and (c) it is inhibited by the exocytosis blocker bafilomycin A1 with a different time course. No glutamate release is evoked by P2Y1R-dependent stimulation of hippocampal synaptosomes. Taken together, our data identify the coupling of purinergic P2Y1R to glutamate exocytosis and its peculiar TNFalpha- and PG-dependent control, and we strongly suggest that this cascade operates selectively in astrocytes. The identified pathway may play physiological roles in glial-glial and glial-neuronal communication.

SDF 1-alpha (CXCL12) triggers glutamate exocytosis from astrocytes on a millisecond time scale: Imaging analysis at the single-vesicle level with TIRF microscopy

Chemokines are small chemotactic molecules widely expressed throughout the central nervous system. A number of papers, during the past few years, have suggested that they have physiological functions in addition to their roles in neuroinflammatory diseases. In this context, the best evidence concerns the CXC-chemokine stromal cell-derived factor (SDF-1alpha or CXCL12) and its receptor CXCR4, whose signalling cascade is also implicated in the glutamate release process from astrocytes. Recently, astrocytic synaptic like microvesicles (SLMVs) that express vesicular glutamate transporters (VGLUTs) and are able to release glutamate by Ca(2+)-dependent regulated exocytosis, have been described both in tissue and in cultured astrocytes. Here, in order to elucidate whether SDF-1alpha/CXCR4 system can participate to the brain fast communication systems, we investigated whether the activation of CXCR4 receptor triggers glutamate exocytosis in astrocytes. By using total internal reflection (TIRF) microscopy and the membrane-fluorescent styryl dye FM4-64, we adapted an imaging methodology recently developed to measure exocytosis and recycling in synaptic terminals, and monitored the CXCR4-mediated exocytosis of SLMVs in astrocytes. We analyzed the co-localization of VGLUT with the FM dye at single-vesicle level, and observed the kinetics of the FM dye release during single fusion events. We found that the activation of CXCR4 receptors triggered a burst of exocytosis on a millisecond time scale that involved the release of Ca(2+) from internal stores. These results support the idea that astrocytes can respond to external stimuli and communicate with the neighboring cells via fast release of glutamate.

In vivo co-ordinated interactions between inhibitory systems to control glutamate mediated hippocampal excitability.

We present an overview of the long-term adaptation of hippocampal neurotransmission to cholinergic and GABAergic deafferentation caused by excitotoxic lesion of the medial septum. Two months after septal microinjection of 2.7 nmol a -amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA), a 220% increase of GABA A receptor labelling in the hippo- campal CA3 and the hilus was shown, and also changes in hippocampal neurotransmission characterised by in vivo microdialysis and HPLC. Basal amino acid and purine extra- cellular levels were studied in control and lesioned rats. In vivo effects of 100 m M KCl perfusion and adenosine A 1 receptor blockade with 1,3-dipropyl- 8-cyclopentylxanthine (DPCPX) on their release were also investigated. In lesioned animals GABA, glutamate and glutamine basal levels were decreased and taurine, adenosine and uric acid levels increased. A similar response to KCl infusion occurred in both groups except for GABA and glutamate, which release decreased in lesioned rats. Only in lesioned rats, DPCPX increased GABA basal level and KCl-induced glutamate release, and decreased glutamate turnover. Our results evidence that an excitotoxic septal lesion leads to increased hippocampal GABA A receptors and decreased glutamate neurotransmis- sion. In this situation, a co-ordinated response of hippocampal retaliatory systems takes place to control neuron excitability.

Increased levels of glutamate in the central nervous system are associated with behavioral symptoms in experimental malaria

Cerebral malaria (CM) is a severe complication resulting from Plasmodium falciparum infection. This condition has been associated with cognitive, behavioral and motor dysfunctions, seizures and coma. The underlying mechanisms of CM are incompletely understood. Glutamate and other metabolites such as lactate have been implicated in its pathogenesis. In the present study, we investigated the involvement of glutamate in the behavioral symptoms of CM. Seventeen female C57BL/6 mice (20-25 g) aged 6-8 weeks were infected with P. berghei ANKA by the intraperitoneal route using a standardized inoculation of 10(6) parasitized red blood cells suspended in 0.2 mL PBS. Control animals (N = 17) received the same volume of PBS. Behavioral and neurological symptoms were analyzed by the SmithKline/Harwell/Imperial College/Royal Hospital/Phenotype Assessment (SHIRPA) battery. Glutamate release was measured in the cerebral cortex and cerebrospinal fluid of infected and control mice by fluorimetric assay. All functional categories of the SHIRPA battery were significantly altered in the infected mice at 6 days post-infection (dpi) (P ≤ 0.05). In parallel to CM symptoms, we found a significant increase in glutamate levels in the cerebral cortex (mean ± SEM; control: 11.62 ± 0.90 nmol/mg protein; infected at 3 dpi: 10.36 ± 1.17 nmol/mg protein; infected at 6 dpi: 26.65 ± 0.73 nmol/mg protein; with EGTA, control: 5.60 ± 1.92 nmol/mg protein; infected at 3 dpi: 6.24 ± 1.87 nmol/mg protein; infected at 6 dpi: 14.14 ± 0.84 nmol/mg protein) and in the cerebrospinal fluid (control: 128 ± 51.23 pmol/mg protein; infected: 301.4 ± 22.52 pmol/mg protein) of infected mice (P ≤ 0.05). These findings suggest a role of glutamate in the central nervous system dysfunction found in CM.

Modulation de l’expression du transporteur vésiculaire du glutamate : implication dans la plasticité des neurones dopaminergiques

De nombreuses études ont établi que la majorité des neurones libèrent plus qu’une substance chimique. Il est bien connu que les neurones peuvent co-exprimer et co-libérer des neuropeptides en plus de leur neurotransmetteur, mais des évidences de la co-libération de deux petits neurotransmetteurs à action rapide se sont accumulées récemment. Des enregistrements électrophysiologiques ont aussi montré que des neurones sérotoninergiques et dopaminergiques isolés peuvent libérer du glutamate quand ils sont placés en culture. De plus, la présence de glutamate et de glutaminase a été détectée dans des neurones sérotoninergiques, dopaminergiques et noradrénergiques par immunomarquage sur des tranches de cerveau. Malheureusement, en considérant le rôle métabolique du glutamate, sa détection immunologique n’est pas suffisante pour assurer le phénotype glutamatergique d’un neurone. Récemment, la découverte de trois transporteurs vésiculaires du glutamate (VGLUT1-3) a grandement facilité l’identification des neurones glutamatergiques. Ces transporteurs sont nécessaires pour la libération de glutamate et constituent les premiers marqueurs morphologiques du phénotype glutamatergique. Il a été démontré que des neurones noradrénergiques expriment VGLUT2 et que des neurones sérotoninergiques expriment VGLUT3. Mais aucune évidence d’expression d’un des sous-types de VGLUT n’a été reportée pour les neurones dopaminergiques. Le but de notre travail était d’identifier quel sous-type de VGLUT est exprimé par les neurones dopaminergiques mésencéphaliques, et de déterminer si le phénotype glutamatergique de ces neurones peut être modulé dans des conditions particulières. Premièrement, nous avons utilisé des microcultures pour isoler les neurones dopaminergiques et des doubles marquages immunocytochimiques pour observer l’expression de VGLUT dans les neurones positifs pour la tyrosine hydroxylase (TH). Nous avons montré que la majorité (80%) des neurones TH+ isolés exprime spécifiquement VGLUT2. Cette expression est précoce au cours du développement in vitro et limitée aux projections axonales des neurones dopaminergiques. Toutefois, cette forte expression in vitro contraste avec la non-détection de ce transporteur dans les rats adultes in vivo. Nous avons décidé ensuite de regarder si l’expression de VGLUT2 pouvait être régulée pendant le développement cérébral de jeunes rats et sous des conditions traumatiques, par double hybridation in situ. Entre 14 et 16 jours embryonnaires, les marquages de VGLUT2 et de TH montraient une superposition significative qui n’était pas retrouvée à des stades ultérieurs. Dans le mésencéphale de jeunes rats postnataux, nous avons détecté l’ARNm de VGLUT2 dans environs 1-2% des neurones exprimant l’ARNm de TH dans la substance noire et l’aire tegmentaire ventrale (ATV). Pour explorer la régulation de l’expression de VGLUT2 dans des conditions traumatiques, nous avons utilisé la 6-hydroxydopamine (6-OHDA) pour léser les neurones dopaminergiques dans les jeunes rats. Dix jours après la chirurgie, nous avons trouvé que 27% des neurones dopaminergiques survivants dans l’ATV exprimaient l’ARNm de VGLUT2 dans les rats 6-OHDA. Finalement, nous avons observé la colocalisation de la protéine VGLUT2 dans les terminaisons TH positives par microscopie électronique. Dans les rats normaux, la protéine VGLUT2 est retrouvée dans 28% des terminaisons axonales TH dans le noyau accumbens. Dans les rats lésés à la 6-OHDA, nous avons observé une diminution considérable des terminaisons TH positives, et une augmentation dans la proportion (37%) des terminaisons dopaminergiques présentant du VGLUT2. Nos résultats suggèrent que le phénotype glutamatergique des neurones dopaminergiques est régulé au cours du développement, peut être réactivé dans des états pathologiques, et que ces neurones peuvent libérer du glutamate dans conditions spécifiques.

Rôles physiologiques du transporteur vésiculaire du glutamate VGLUT2 dans les neurones dopaminergiques

Des travaux récents démontrent que certains neurones dopaminergiques du mésencéphale ont la capacité de libérer du glutamate en plus de la dopamine (DA). Ce phénomène de « co-transmission » requiert l’expression du transporteur vésiculaire du glutamate de type 2 (VGLUT2) dans les neurones dopaminergiques. Certaines observations montrent que l’expression de VGLUT2 dans les neurones dopaminergiques survient tôt durant leur développement et est essentiellement limitée aux neurones de l’aire tegmentaire ventrale (VTA). De plus, cette libération de glutamate se retrouve principalement au niveau des terminaisons de ces neurones dans le striatum ventral, mais pas dans le striatum dorsal. Ces données suggèrent d’une part un rôle développemental possible du glutamate dans les neurones dopaminergiques, et d’autre part, que les signaux dérivés des neurones cibles puissent réguler le double phénotype des neurones dopaminergiques menant ainsi à une plasticité phénotypique. Par ailleurs, il est toujours inconnu si cette libération de glutamate se fait à partir des terminaisons qui relâchent la DA ou à partir de terminaisons axonales distinctes. De plus, le rôle physiologique de ce surprenant phénomène de co-transmission reste également inconnu. Ainsi, dans cette étude, nous avons d’abord démontré in vitro et in vivo que l’expression de VGLUT2 est nécessaire pour la survie et la croissance d’une sous-population de neurones dopaminergiques. En utilisant une lignée de souris ayant une délétion génique spécifique de VGLUT2 dans les neurones dopaminergiques, nous avons observé une diminution du nombre de terminaisons dopaminergiques et glutamatergiques dans le striatum, une baisse de libération de DA dans le striatum ventral, une diminution de la coordination motrice ainsi qu’une diminution de l’activité locomotrice induite par les drogues d’abus. D’autre part, nous avons démontré in vitro et in vivo que les neurones dopaminergiques au double phénotype établissent des terminaisons distinctes afin de relâcher le glutamate et la DA. De plus, nous démontrons que ce phénomène de ségrégation des sites de libération semble être induit par une interaction avec les neurones du striatum ventral. Ces travaux démontrent le rôle physiologique déterminant de la co-transmission DA-glutamate pour l’homéostasie du système DAergique et dévoile une caractéristique fondamentale de l’établissement des terminaisons axonales de ces neurones. Ces travaux permettent ainsi de mieux comprendre les rôles physiologiques de la co-libération de glutamate par les neurones du système nerveux central et présentent une nouvelle perspective sur les dysfonctions potentielles de ces neurones dans les maladies du cerveau.

Acute decrease in net glutamate uptake during energy deprivation

The extracellular glutamate concentration ([glu]o) rises during cerebral ischemia, reaching levels capable of inducing delayed neuronal death. The mechanisms underlying this glutamate accumulation remain controversial. We used N-methyl-d-aspartate receptors on CA3 pyramidal neurons as a real-time, on-site, glutamate sensor to identify the source of glutamate release in an in vitro model of ischemia. Using glutamate and l-trans-pyrrolidine-2,4-dicarboxylic acid (tPDC) as substrates and dl-threo-β-benzyloxyaspartate (TBOA) as an inhibitor of glutamate transporters, we demonstrate that energy deprivation decreases net glutamate uptake within 2–3 min and later promotes reverse glutamate transport. This process accounts for up to 50% of the glutamate accumulation during energy deprivation. Enhanced action potential-independent vesicular release also contributes to the increase in [glu]o, by ≈50%, but only once glutamate uptake is inhibited. These results indicate that a significant rise in [glu]o already occurs during the first minutes of energy deprivation and is the consequence of reduced uptake and increased vesicular and nonvesicular release of glutamate.

Regulation of N-methyl-D-aspartate-induced toxicity in the neostriatum: a role for metabotropic glutamate receptors?

Glutamate release activates multiple receptors that interact with each other and thus determine the response of the cell. Exploring these interactions is critical to developing an understanding of the functional consequences of synaptic transmission. Activation of metabotropic glutamate receptors (mGluRs) inhibits N-methyl-D-aspartate (NMDA)-evoked responses measured electrophysiologically in neostriatal slices. The present study examines the functional consequences of this regulation using infrared differential interference contrast videomicroscopy to measure and characterize glutamate receptor-induced cell swelling in a neostriatal brain slice preparation. This swelling is, in many cases, a prelude to necrotic cell death and the dye trypan blue was used to confirm that swelling can result in the death of neostriatal cells. Activation of mGluRs by the agonist 1-aminocyclopentane-1,3-dicarboxylic acid (tACPD) inhibited NMDA but not amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid/kainate-induced swelling. This regulation was cell-type specific as tACPD did not alter NMDA-induced swelling in pyramidal cells of the hippocampus. Importantly, these findings could be extended to in vivo preparations. Pretreatment with tACPD limited the size of lesions and associated behavioral deficits induced by intrastriatal administration of the NMDA receptor agonist quinolinic acid.

Ethosuximide modifies network excitability in the rat entorhinal cortex via an increase in GABA release

Ethosuximide is the drug of choice for treating generalized absence seizures, but its mechanism of action is still a matter of debate. It has long been thought to act by disrupting a thalamic focus via blockade of T-type channels and, thus, generation of spike-wave activity in thalamocortical pathways. However, there is now good evidence that generalized absence seizures may be initiated at a cortical focus and that ethosuximide may target this focus. In the present study we have looked at the effect ethosuximide on glutamate and GABA release at synapses in the rat entorhinal cortex in vitro, using two experimental approaches. Whole-cell patch-clamp studies revealed an increase in spontaneous GABA release by ethosuximide concurrent with no change in glutamate release. This was reflected in studies that estimated global background inhibition and excitation from intracellularly recorded membrane potential fluctuations, where there was a substantial rise in the ratio of network inhibition to excitation, and a concurrent decrease in excitability of neurones embedded in this network. These studies suggest that, in addition to well-characterised effects on ion channels, ethosuximide may directly elevate synaptic inhibition in the cortex and that this could contribute to its anti-absence effects. This article is part of a Special Issue entitled 'Post-Traumatic Stress Disorder'.

Mobility of NMDA autoreceptors but not postsynaptic receptors at glutamate synapses in the rat entorhinal cortex

NMDA receptors (NMDAr) are known to undergo recycling and lateral diffusion in postsynaptic spines and dendrites. However, NMDAr are also present as autoreceptors on glutamate terminals, where they act to facilitate glutamate release, but it is not known whether these receptors are also mobile. We have used functional pharmacological approaches to examine whether NMDA receptors at excitatory synapses in the rat entorhinal cortex are mobile at either postsynaptic sites or in presynaptic terminals. When NMDAr-mediated evoked EPSCs (eEPSCs) were blocked by MK-801, they showed no evidence of recovery when the irreversible blocker was removed, suggesting that postsynaptic NMDAr were relatively stably anchored at these synapses. However, using frequency-dependent facilitation of AMPA receptor (AMPAr)-mediated eEPSCs as a reporter of presynaptic NMDAr activity, we found that when facilitation was blocked with MK-801 there was a rapid (similar to 30-40 min) anomalous recovery upon removal of the antagonist. This was not observed when global NMDAr blockade was induced by combined perfusion with MK-801 and NMDA. Anomalous recovery was accompanied by an increase in frequency of spontaneous EPSCs, and a variable increase in frequency-facilitation. Following recovery from blockade of presynaptic NMDAr with a competitive antagonist, frequency-dependent facilitation of AMPAr-mediated eEPSCs was also transiently enhanced. Finally, an increase in frequency of miniature EPSCs induced by NMDA was succeeded by a persistent decrease. Our data provide the first evidence for mobility of NMDAr in the presynaptic terminals, and may point to a role of this process in activity-dependent control of glutamate release.