The synaptic vesicle cluster: a source of endocytic proteins during neurotransmitter release.

Over the past few years significant progress has been achieved in understanding the molecular steps underlying the fusion and recycling of vesicles at central synapses. It still remains unclear, however, how the fusion event is linked with vesicle membrane retrieval. Several factors promoting the transition from exo- to endocytosis have been extensively studied, including levels of intracellular Ca2+, the synaptic proteins involved at both sides of the vesicle cycle, posttranslational modification of endocytic proteins, and the lipid composition of recycled membranes. Recent studies in glutamate synapses indicate that vesicle clusters accumulated at the sites of synaptic contacts have a more complex organization than has previously been thought. Many endocytic proteins reside in the vesicle pool at rest and undergo cycles of migration between the active and periactive zones during synaptic activity. We propose that the local migration of endocytic proteins triggered by Ca2+ influx into the nerve terminal functions as one of the molecular mechanisms coupling exo- and endocytosis in synapses.

Synaptic vesicle endocytosis impaired by disruption of dynamin-SH3 domain interactions.

The proline-rich COOH-terminal region of dynamin binds various Src homology 3 (SH3) domain-containing proteins, but the physiological role of these interactions is unknown. In living nerve terminals, the function of the interaction with SH3 domains was examined. Amphiphysin contains an SH3 domain and is a major dynamin binding partner at the synapse. Microinjection of amphiphysin's SH3 domain or of a dynamin peptide containing the SH3 binding site inhibited synaptic vesicle endocytosis at the stage of invaginated clathrin-coated pits, which resulted in an activity-dependent distortion of the synaptic architecture and a depression of transmitter release. These findings demonstrate that SH3-mediated interactions are required for dynamin function and support an essential role of clathrin-mediated endocytosis in synaptic vesicle recycling.

Dissociation between Ca2+-triggered synaptic vesicle exocytosis and clathrin-mediated endocytosis at a central synapse.

We have tested whether action potential-evoked Ca2+ influx is required to initiate clathrin-mediated synaptic vesicle endocytosis in the lamprey reticulospinal synapse. Exo- and endocytosis were temporally separated by a procedure involving tonic action potential stimulation and subsequent removal of extracellular Ca2+ (Ca2+e). A low concentration of Ca2+ ([Ca2+]e of 11 microM) was found to be required for the induction of early stages of endocytosis. However, the entire endocytic process, from the formation of clathrin-coated membrane invaginations to the generation of synaptic vesicles, proceeded in the absence of action potential-mediated Ca2+ entry. Our results indicate that the membrane of synaptic vesicles newly incorporated in the plasma membrane is a sufficient trigger of clathrin-mediated synaptic vesicle endocytosis.

Endophilin/SH3p4 is required for the transition from early to late stages in clathrin-mediated synaptic vesicle endocytosis.

Endophilin/SH3p4 is a protein highly enriched in nerve terminals that binds the GTPase dynamin and the polyphosphoinositide phosphatase synaptojanin, two proteins implicated in synaptic vesicle endocytosis. We show here that antibody-mediated disruption of endophilin function in a tonically stimulated synapse leads to a block in the invagination of clathrin-coated pits adjacent to the active zone and therefore to a block of synaptic vesicle recycling. We also show that in a cell-free system, endophilin is not associated with clathrin coats and is a functional partner of dynamin. Our findings suggest that endophilin is part of a biochemical machinery that acts in trans to the clathrin coat from early stages to vesicle fission.

Fission and uncoating of synaptic clathrin-coated vesicles are perturbed by disruption of interactions with the SH3 domain of endophilin.

Coordination between sequential steps in synaptic vesicle endocytosis, including clathrin coat formation, fission, and uncoating, appears to involve proteinprotein interactions. Here, we show that compounds that disrupt interactions of the SH3 domain of endophilin with dynamin and synaptojanin impair synaptic vesicle endocytosis in a living synapse. Two distinct endocytic intermediates accumulated. Free clathrin-coated vesicles were induced by a peptide-blocking endophilin's SH3 domain and by antibodies to the proline-rich domain (PRD) of synaptojanin. Invaginated clathrin-coated pits were induced by the same peptide and by the SH3 domain of endophilin. We suggest that the SH3 domain of endophilin participates in both fission and uncoating and that it may be a key component of a molecular switch that couples the fission reaction to uncoating.

Sequential steps in clathrin-mediated synaptic vesicle endocytosis.

Synaptic vesicles are recycled with remarkable speed and precision in nerve terminals. A major recycling pathway involves clathrin-mediated endocytosis at endocytic zones located around sites of release. Different 'accessory' proteins linked to this pathway have been shown to alter the shape and composition of lipid membranes, to modify membrane-coat protein interactions, and to influence actin polymerization. These include the GTPase dynamin, the lysophosphatidic acid acyl transferase endophilin, and the phosphoinositide phosphatase synaptojanin. Protein perturbation studies in living nerve terminals are now beginning to link the actions of these proteins with morphologically defined steps of endocytosis.

Differential efficiency of the endocytic machinery in tonic and phasic synapses.

Efficient synaptic vesicle membrane recycling is one of the key factors required to sustain neurotransmission. We investigated potential differences in the compensatory endocytic machineries in two glutamatergic synapses with phasic and tonic patterns of activity in the lamprey spinal cord. Post-embedding immunocytochemistry demonstrated that proteins involved in synaptic vesicle recycling, including dynamin, intersectin, and synapsin, occur at higher levels (labeling per vesicle) in tonic dorsal column synapses than in phasic reticulospinal synapses. Synaptic vesicle protein 2 occurred at similar levels in the two types of synapse. After challenging the synapses with high potassium stimulation for 30 min the vesicle pool in the tonic synapse was maintained at a normal level, while that in the phasic synapse was partly depleted along with expansion of the plasma membrane and accumulation of clathrin-coated intermediates at the periactive zone. Thus, our results indicate that an increased efficiency of the endocytic machinery in a synapse may be one of the factors underlying the ability to sustain neurotransmission at high rates.

Inhibition of neurotransmitter release in the lamprey reticulospinal synapse by antibody-mediated disruption of SNAP-25 function.

Exocytosis - syntaxin - synaptobrevin - SNARE synaptic vesicle The lamprey giant reticulospinal synapse can be used to manipulate the molecular machinery of synaptic vesicle exocytosis by presynaptic microinjection. Here we test the effect of disrupting the function of the SNARE protein SNAP-25. Polyclonal SNAP-25 antibodies were shown in an in vitro assay to inhibit the binding between syntaxin and SNAP-25. When microinjected presynaptically, these antibodies produced a potent inhibition of the synaptic response. Ba2+ spikes recorded in the presynaptic axon were not altered, indicating that the effect was not due to a reduced presynaptic Ca2+ entry. Electron microscopic analysis showed that synaptic vesicle clusters had a similar organization in synapses of antibody-injected axons as in control axons, and the number of synaptic vesicles in apparent contact with the presynaptic plasma membrane was also similar. Clathrin-coated pits, which normally occur at the plasma membrane around stimulated synapses, were not detected after injection of SNAP-25 antibodies, consistent with a blockade of vesicle cycling. Thus, SNAP-25 antibodies, which disrupt the interaction with syntaxin, inhibit neurotransmitter release without affecting the number of synaptic vesicles at the plasma membrane. These results provide further support to the view that the formation of SNARE complexes is critical for membrane fusion, but not for the targeting of synaptic vesicles to the presynaptic membrane.

Two types of motoneurons supplying dorsal fin muscles in lamprey and their activity during fictive locomotion.

The location and dendritic morphology of motoneurons supplying the dorsal fin muscles were studied in the lamprey spinal cord (Ichthyomyzon unicuspis). Motoneurons were retrogradely labelled after injection of HRP into the fin muscles or after its application on the cut ends of the ventral roots. HRP-labelled cells were subsequently reconstructed, in the horizontal and/or transverse planes. Fin motoneurons were also injected intracellularly with Lucifer Yellow and their detailed three-dimensional structure was analysed by confocal laser-scanning microscopy. Unlike myotomal motoneurons, which are closely spaced in the lateral cell column, fin motoneurons were distributed along the spinal cord separately or in pairs. They could be distinguished from motoneurons supplying trunk muscles by having a limited number of dendrites in the lateral part of the spinal cord. In addition, some fin motoneurons extend their dendrites into the dorsal column. The motor cells innervating fin muscles were divided into two types based on their dendritic morphology. Type I have a widespread dendritic tree in the rostrocaudal direction and, with few exceptions, completely restricted to the ipsilateral side. A proportion (25%) of these cells have dendrites extending into the dorsal column. Type II fin motoneurons extended their dendrites both ipsi- and contra-laterally. The contra-lateral dendrites pass below and above the central canal. The dendrites send off branches into the dorsal columns on both the ipsi- and the contra-lateral sides. Electron microscopic analysis showed that both type I and type II fin motoneurons receive numerous synaptic contacts from dorsal column axons. During fictive locomotion both types of motoneurons are active in antiphase in relation to myotomal motoneurons and to the main locomotor burst.

Possible morphological substrates for GABA-mediated presynaptic inhibition in the lamprey spinal cord.

Gamma-aminobutyric acid (GABA) neurons intrinsic to the lamprey spinal cord are known to modulate synaptic transmission from interneurons active during locomotion and from mechanosensory dorsal cells. Many of these physiological effects are presynaptic. To establish the morphological substrates for these axo-axonic interactions, an ultrastructural analysis was performed with an antiserum to fixed GABA. The GABA immunoreactivity (ir) was detected by postembedding peroxidase-antiperoxidase and immunogold techniques. GABA-ir terminals were found to make appositions with unlabelled axons located in the dorsal columns and in the ventrolateral aspect of the spinal cord. In the ventrolateral part of the cord, similar appositions between different GABA-ir terminals were also observed. The immunolabelled terminals contained spherical to pleomorphic synaptic vesicles, and also glycogen granules and dense core vesicles. In some cases, the fine structure of the contacts between immunogold-labelled terminals and unlabelled axons suggested a synaptic relationship. Such a relation was found in a relatively small proportion (2-3%) of the appositions studied. These specializations were always observed in close relation to an output synapse of the postsynaptic axon. It is suggested that the axo-axonal contacts described may provide an effective modulation of the synaptic transmission from axons in the lamprey spinal cord.

Zinc co-localizes with GABA and glycine in synapses in the lamprey spinal cord.

The presence of zinc in synaptic terminals in the lamprey spinal cord was examined utilizing a modification of the Timm's sulfide silver method and with the fluorescent marker 6-methoxy-8-quinolyl-p-toluenesulfonamide (TSQ). Axons labeled with a Timm's staining method were predominantly located in the lateral region of the dorsal column. This correlated with a maximum of TSQ fluorescence in this region of the spinal cord. Single labeled terminals accumulating Timm reaction product were also found throughout the gray matter and fiber tracts. At the ultrastructural level, zinc was located in a population of synaptic terminals that co-localized gamma-aminobutyric acid (GABA) and glycine. Possible effects of Zn2+ on neuronal activity were examined. In spinobulbar interneurons, which receive GABAergic input in the dorsal column, zinc potentiated responses to GABA application, but it did not affect responses to GABA in motoneurons. Responses in motoneurons to pressure application of glycine were also not affected by Zn2+. Zinc, however, potentiated monosynaptic glycinergic inhibitory postsynaptic potentials (IPSPs) evoked in motoneurons by inhibitory locomotor network interneurons and increased frequency, but not amplitude of spontaneous miniature IPSPs recorded in the presence of tetrodotoxin (TTX), suggesting presynaptic effects. Glutamate responses and the amplitude of monosynaptic excitatory postsynaptic potentials (EPSPs) in motoneurons were reduced by zinc. These effects appeared to be mediated largely postsynaptically through an effect on the N-methyl-D-aspartate (NMDA) component of the glutamatergic input. Our results thus show that free zinc is present in inhibitory synaptic terminals in the lamprey spinal cord, and that it may function as a modulator of inhibitory synaptic transmission.

Synaptic and nonsynaptic monoaminergic neuron systems in the lamprey spinal cord.

In the lamprey spinal cord, dopamine- (DA) and 5-hydroxytryptamine-(5-HT) containing cells appear to play an important role in controlling the firing properties of motoneurons and interneurons and, thereby, in modulating the efferent motor pattern. To determine the detailed morphology and synaptic connectivity of the intraspinal DA and 5-HT systems in Lampetra fluviatilis and Ichthyomyzon unicuspis, DA and 5-HT antisera were used in light and electron microscopic immunocytochemical experiments. Two main groups of labeled cells were distinguished: DA-containing liquor-contacting (LC) cells distributed along the central canal, and 5-HT+DA-containing multipolar cells located near the midline ventral to the central canal. Both types were synaptically connected with other neuronal elements. The DA-immunoreactive LC cells, which extended a ciliated process into the central canal, received symmetrical synapses from unlabeled terminals containing small synaptic vesicles. The distal process of the LC cells could be traced to the lateral cell column, to the ventral aspect of the dorsal column, or to the ventromedial area. Ultrastructural analysis of DA fibers in these regions showed the presence of labeled terminals containing numerous small synaptic vesicles and a few dense-core vesicles. These terminals formed symmetrical synapses with unlabeled cell bodies and dendrites, with GABA-immunopositive LC cells, and with the multipolar DA+5-HT cells. The multipolar DA+5-HT cells also received input from unlabeled synapses. Intracellular recording from these cells showed that they received excitatory postsynaptic potentials in response to stimulation of fibers in the ventromedial tracts and dorsal roots. The terminals of the multipolar DA+5-HT neurons in the ventromedial spinal cord contained numerous dense-core vesicles and small synaptic vesicles, but no synaptic specializations could be detected. In addition, a small number of larger DA-immunoreactive cells were observed in the lateral cell column at rostral levels. The lamprey spinal cord thus contains distinct populations of synaptically interconnected monoaminergic neurons. Dopamine-containing LC cells synapse onto DA+5-HT-containing multipolar cells, in addition to GABAergic LC cells and unidentified spinal neurons. In contrast, the multipolar cells appear to exert their influence by nonsynaptic mechanisms.

Monosynaptic input from cutaneous sensory afferents to fin motoneurons in lamprey.

The sensory control of lamprey dorsal fin motoneurons was studied by using paired intracellular recordings combined with a morphological analysis. Dorsal cells innervating the skin of the dorsal fin and fin motoneurons were retrogradely labeled by injecting fluoresceincoupled dextran amines into the dorsal fin. Labeled motoneurons and dorsal cells showed close appositions, suggesting that the dorsal cells innervating the fin region make monosynaptic connections with fin motoneurons. By using conventional electrophysiological criteria, monosynaptic excitatory connections were found between fin dorsal cells and fin motoneurons. In addition, Lucifer yellow injection followed by confocal three-dimensional (3-D) reconstructions of monosynaptically connected pairs, revealed close apposition between dorsal cell axons and the distal dendrites of fin motoneurons. Each fin motoneuron received monosynaptic excitatory input from at least four different afferents. The amplitude of the monosynaptic excitatory postsynaptic potential (EPSP)s was reduced by administration of the N-methyl-D-aspartate (NMDA) receptor antagonist DL,2 amino-5-phosphovaleric acid (APV). Sensory stimulation could also elicit di- or oligosynaptic inhibitory postsynaptic potential (IPSP)s, which were blocked by the glycine antagonist strychnine, resulting in the appearance of large monosynaptic EPSPs, which could induce action potentials.

Anatomical study of spinobulbar neurons in lampreys.

The present study was aimed at identifying spinal neurons ascending to the brainstem outside the dorsal columns in the lamprey. Two retrograde tracers (cobalt-lysine and horseradish peroxidase [HRP]) were injected in the brainstem or rostral spinal cord in vivo or in vitro. Labeled cells were distributed bilaterally with a contralateral dominance, along the whole rostrocaudal extent of the spinal cord. The density of cells markedly decreased rostrocaudally. Several classes of brainstem-projecting neurons were identified. Most cells with a short axon were small and formed columns, in the dorsolateral and ventrolateral gray matter, at the transition between the rhombencephalon and the spinal cord. Dorsal elongated cells were spindle shaped, located medially, in the first two spinal segments. Lateral elongated cells were medium to large size neurons, located in the intermediate and lateral gray matter, mainly contralateral to the injection site. Their axon emerging from the lateral part of the soma crossed the midline, ventral to the central canal. These cells were present throughout the rostral spinal cord. Cells were also labeled in the lateral white matter. Some of them had the typical dendritic arborizations of edge cells (intraspinal stretch receptor neurons) and were located in the most rostral segments, bilaterally. Other medium to large size neurons were identified dorsal and medial to most of the edge cells. We suggest that at least the group of lateral elongated cells exhibits rhythmic membrane potential oscillations during fictive locomotion. These cells may, together with the rostral edge cells, be responsible for the locomotor-related modulation of activity in reticulospinal and vestibulospinal neurons.

Ultrastructural evidence for a preferential elimination of glutamate-immunoreactive synaptic terminals from spinal motoneurons after intramedullary axotomy.

After axotomy in the ventral funiculus of the cat spinal cord, about half of the population of lesioned motoneurons die at 1-3 weeks postoperatively, whereas the other half survives and generates new axons through the lesion area. To identify conditions that may promote survival and regeneration of motoneurons subjected to this kind of injury, the authors examined ultrastructurally lesion-induced changes in the number and distribution of nerve terminals on the somata and proximal dendrites of alpha-motoneurons in the 7th lumbar spinal segment (L7) of the cat spinal cord. Intramedullary axotomy resulted in a profound reduction in the number of nerve terminals impinging on the somata and proximal dendrites, with the maximal effect seen at 3 weeks postlesion. At that time, only 12-25% of the normal number of terminals remained on the cell somata, and 22-33% remained on proximal dendrites. Thereafter, a gradual increase in terminal numbers occurred, reaching normal levels at 34 weeks after the lesion. Already at 2 days postoperatively and, most obviously, at 3 weeks postoperatively, type S nerve terminals were eliminated to a larger degree than type F terminals. Postembedding immunohistochemistry confirmed that the largest reduction at 3 weeks was seen for excitatory glutamate-immunopositive type S nerve terminals (90%), whereas inhibitory glycine-immunoreactive and gamma-aminobutyric acid (GABA)-immunoreactive type F terminals were affected less (70% reduction). This led to a distinct shift in the ratio between the numbers of terminals that were immunopositive for glycine and GABA and the numbers of terminals that were labeled for glutamate. For the cell body, this ratio increased from 3.7 in normal material to 14.5 in lesioned motoneurons, whereas the corresponding values for proximal dendrites were 3.8 and 7.5. The preferential elimination of glutamatergic inputs to lesioned motoneurons may reflect an active reorganization of the synaptic input to diminish the excitotoxic influence on these neurons, thereby promoting the survival of motoneurons after intramedullary axotomy.

Extrasynaptic localization of taurine-like immunoreactivity in the lamprey spinal cord.

Taurine is an endogenous amino acid that can occur in nerve terminals in the central nervous system and that can produce inhibitory neuronal responses. It is unclear, however, whether this amino acid can function as a synaptic transmitter. To examine the distribution of taurine at high anatomical resolution in a vertebrate, light and electron microscopic immunocytochemical postembedding techniques were applied to the lamprey spinal cord (Ichtyomyzon unicuspis and Lampetra fluviatilis), which contains many large, unmyelinated axons. The most intense immunolabeling occurred in a population of liquor-contacting cells (tanycytes), located around the central canal, which extended processes to the dorsal, lateral, and ventral margins of the spinal cord. In addition, a proportion of the taurine-immunoreactive cells contained gamma-aminobutyric acid (GABA)-like immunoreactivity. A moderate level of taurine immunoreactivity was also present in ependymal cells, located around the central canal, as well as in astrocytes throughout all regions of the spinal cord. At the ultrastructural level, the taurine immunoreactivity showed an even distribution in the cytoplasm of the labeled cells. In contrast to the glial labeling, neuronal cell bodies and axons exhibited very low levels of taurine labeling, which were similar to the level of background labeling. The synaptic vesicle clusters within the axons did not show any clear accumulation of taurine immunoreactivity. These results suggest that taurine may have metabolic roles in the lamprey spinal cord, and, as in other systems, it may take part in osmoregulation. However, the lack of immunolabeling in presynaptic elements is not consistent with a role of taurine as a synaptic transmitter.

Qualitative and quantitative analysis of glycine- and GABA-immunoreactive nerve terminals on motoneuron cell bodies in the cat spinal cord: a postembedding electron microscopic study.

The distribution of glycine- and gamma-aminobutyric acid (GABA)-like immunoreactivity (LI) in nerve terminals on the cell soma of motoneurons in the aldehyde-fixed cat L7 spinal cord was examined using postembedding immunogold histochemistry in serial ultrathin sections. Quantitative examination of 405 terminals on eight neurons of alpha-motoneuron size in the L7 motor nuclei from one animal was performed. A majority of the terminals (69%) were immunoreactive to glycine and/or GABA. These terminals contained flat or oval synaptic vesicles, thus classifying them as F type or as C type in one case. In no case was a type-F terminal unlabeled for both glycine and GABA. Most of the immunolabeled terminals were immunoreactive to glycine only (62.5%), whereas 35.4% contained both glycine- and GABA-LI. A very small number of immunolabeled terminals (2%) were immunoreactive to GABA only. In those terminals, where glycine- and GABA-LI coexisted, the gold particle density for each amino acid was only half of that seen in boutons containing only one of the two amino acids. The involvement of glycine and GABA in postsynaptic inhibition of spinal alpha-motoneurons is discussed, with particular reference to the possibility that these two inhibitory amino acids may be coreleased from a significant proportion of the nerve terminals impinging on the cell bodies.

Thyrotropin-releasing hormone (TRH)-like immunoreactivity in the grey monkey (Macaca fascicularis) spinal cord and medulla oblongata with special emphasis on the bulbospinal tract.

The distribution of thyrotropin-releasing hormone (TRH)-like immunoreactivity (LI) has been studied in the grey monkey (Macaca fascicularis) spinal cord and medulla oblongata by the use of indirect immunofluorescence and the peroxidase-antiperoxidase (PAP) technique. Furthermore, double-labeling experiments were performed in order to study colocalization of 5-hydroxytryptamine (5-HT)- and substance P-LI. A dense innervation of TRH-immunoreactive (IR) varicose fibers was found in the ventral horn motor nuclei, in the region surrounding the central canal, in the intermediolateral cell column, and in the dorsal horn laminae II and III. In addition, cell bodies harboring TRH-LI were found in the dorsal horn laminae II-IV. In the ventral horn, many of the large cell bodies and their proximal dendrites were totally encapsulated by TRH-IR fibers. From double-labeled sections a high degree of coexistence could be established between TRH-/5-HT-LI, TRH-/substance P-LI, and 5-HT-/substance P-LI in fibers in the motor nuclei; as a consequence, a large proportion of these fibers should harbor TRH-/5-HT-/substance P-LI. A coexistence between TRH-/5-HT-LI could also be demonstrated in the intermediolateral cell column. However, no unequivocal coexistence could be found between TRH-/substance P-LI and 5-HT-/substance P-LI in this region. In the dorsal horn, no clear coexistence could be encountered for any of the above indicated combinations. Electron microscopic analysis of material from the lumbar lateral motor nucleus demonstrated TRH-IR terminals making synapses with large cell bodies and dendrites. In addition, contacts lacking synaptic specializations could also be verified. In the medulla oblongata, with the use of the PAP technique, a large number of cell bodies containing TRH-LI were encountered in the midline raphe nuclei and in nucleus reticularis lateralis. A similar distribution pattern could be found for 5-HT-LI, but no cell bodies containing substance P-LI could be seen in these regions. Chemical analysis of specimens from cervical, thoracic, and lumbar spinal cord revealed higher concentrations of TRH- and 5-HT-LI in the ventral quadrants, whereas substance P-LI dominated in the dorsal quadrants. Thus, the concentrations of TRH-, 5-HT-, and substance P-LI was in accordance with the observed regional variation in density of IR-fibers and varicosities found in the spinal cord. We have shown that TRH-LI has a distribution in the monkey spinal cord and medulla oblongata similar to that previously demonstrated in other species.(ABSTRACT TRUNCATED AT 400 WORDS)

Penicillin-induced bursting in motoneurones of the frog spinal cord. Elimination by NMDA antagonist.

The effects of penicillin were investigated in lumbar motoneurones of isolated spinal cord preparation of the frog (Rana ridibunda). Spinal root discharges were recorded and single cell activity was studied with intracellular electrodes. Bath application of 500 IU/ml of penicillin G induced in motoneurones prolonged depolarization shifts, followed by repeated transient depolarizations lasting several hours. In all motoneurones studied, this bursting activity was synchronized with discharges recorded from the ventral root. Blockade of N-methyl-D-aspartate (NMDA) receptors by D,L-2-amino-5-phosphonovaleric acid (50-100 microM) completely eliminated the bursting activity. It is suggested that the spinal cord may be an important locus of the anticonvulsant effect of NMDA antagonists.

Glial and neuronal glutamine pools at glutamatergic synapses with distinct properties.

The main pathway for transmitter glutamate turnover in excitatory synapses is thought to involve an uptake in glial processes, a conversion into glutamine, which recycles to the presynaptic terminal to serve as the main precursor for new synthesis of glutamate. To investigate whether the mechanisms of glutamine and glutamate turnover are linked with the properties of different glutamate synapses, the distribution of glutamine was studied in two types of glutamate synapse in the lamprey spinal cord using immunogold post-embedding electron microscopy. The synapses examined are formed by primary afferent axons (dorsal column axons), which predominantly exhibit a tonic firing pattern, and by giant reticulospinal axons, which primarily fire in brief bursts. Glial cell processes and postsynaptic dendrites displayed the highest density of glutamine labeling in both types of synapse. The level of glutamine was significantly higher in the glial cell processes surrounding the tonic dorsal column synapses, as compared to those surrounding the reticulospinal synapses. The axoplasmic matrix and presynaptic mitochondria, as well as postsynaptic dendrites, contained similar levels of glutamine labeling in both cases. The glutamate labeling in glial processes was also similar at the two types of synapse, while axoplasmic matrix and presynaptic mitochondria displayed four to six times higher levels in the tonic axons. In conjunction with our previous results, showing a different transport activity in glial processes of the two types of excitatory synapse, the results of the present study suggest that the glial pool of neurotransmitter precursor is linked to the rate of transmitter synthesis and release in adjacent synapses.