Depression by Sodium Ions of Calcium Uptake Mediated by Non-N-Methyl-d-Aspartate Receptors in Cultured Cerebellar Neurons and Correlation with Evoked d-[3H]Aspartate Release

KIF4 Motor Regulates Activity-Dependent Neuronal Survival by Suppressing PARP-1 Enzymatic Activity

In brain development, apoptosis is a physiological process that controls the final numbers of neurons. Here, we report that the activity-dependent prevention of apoptosis in juvenile neurons is regulated by kinesin superfamily protein 4 (KIF4), a microtubule-based molecular motor. The C-terminal domain of KIF4 is a module that suppresses the activity of poly (ADP-ribose) polymerase-1 (PARP-1), a nuclear enzyme known to maintain cell homeostasis by repairing DNA and serving as a transcriptional regulator. When neurons are stimulated by membrane depolarization, calcium signaling mediated by CaMKII induces dissociation of KIF4 from PARP-1, resulting in upregulation of PARP-1 activity, which supports neuron survival. After dissociation from PARP-1, KIF4 enters into the cytoplasm from the nucleus and moves to the distal part of neurites in a microtubule-dependent manner. We suggested that KIF4 controls the activity-dependent survival of postmitotic neurons by regulating PARP-1 activity in brain development.

Kainate down-regulates a subset of GABAA receptor subunits expressed in cultured mouse cerebellar granule cells

The effect of kainate, an agonist selective for ionotropic AMPA/kainate type of glutamate receptors, on GABAA receptor subunit expression in cultured mouse cerebellar granule cells was studied using quantitative RT-PCR, ligand binding and electrophysiology. Chronic kainate treatment, without producing excitotoxicity, resulted in preferential, dose- and time-dependent down-regulation of alpha1, alpha6 and beta2 subunit mRNA expression, the expression of beta3, gamma2 and delta subunit mRNAs being less affected. The down-regulation was reversed by DNQX, an AMPA/kainate-selective glutamate receptor antagonist. A 14-day kainate treatment resulted in 46% decrease of total [3H]Ro 15-4513 binding to the benzodiazepine sites. Diazepam-insensitive [3H]Ro 15-4513 binding was decreased by 89% in accordance with very low amount of alpha6 subunit mRNA present. Diazepam-sensitive [3H]Ro 154513 binding was decreased only by 40%, contrasting >90% decrease in alpha1 subunit mRNA expression. However, this was consistent with lower potentiation of GABA-evoked currents in kainate-treated than control cells by the alpha1-selective benzodiazepine site ligand zolpidem, suggesting compensatory expression of alpha5 (and/or alpha2 or alpha3) subunits producing diazepam-sensitive but zolpidem-insensitive receptor subtypes. In conclusion, chronic kainate treatment of cerebellar granule cells selectively down-regulates oil, alpha6 and beta2 subunits resulting in altered GABAA receptor pharmacology.

Characterization of depolarization-coupled release of glutamate from cultured mouse cerebellar granule cells using DL-threo-beta-benzyloxyaspartate (DL-TBOA) to distinguish between the vesicular and cytoplasmic pools

Release of preloaded [3H]D-aspartate in response to depolarization induced by N-methyl-D-aspartate (NMDA) or the endogenous agonist glutamate was characterized using cultured glutamatergic cerebellar granule neurons. Release from the vesicular and the cytoplasmic glutamate pools, respectively, was distinguished employing the competitive, non-transportable glutamate transport inhibitor DL-threo-beta-benzyloxyaspartate (DL-TBOA). NMDA (300 microM)-induced release was enhanced (50%) by a simultaneous elevation of the extracellular potassium concentration to 15 mM, which lifts the voltage-dependent magnesium block of the NMDA receptors. This NMDA/K(+)-induced release was not sensitive to DL-TBOA (100 microM) but was inhibited by 75% in the presence of the unspecific calcium channel antagonist La(3+) (100 microM). Glutamate (100 microM) induced a large fractional release of the preloaded [3H]D-aspartate and in the presence of DL-TBOA the release was reduced by approximately 50%. In contrast, release evoked by 25 microM glutamate was not inhibited by DL-TBOA. These results indicate that the release elicited by 100 microM glutamate is comprised of a significant glutamate transporter-mediated component in addition to the vesicular release while the NMDA/K(+)-induced release is vesicular in nature. It is likely that the high glutamate concentration (100 microM) may facilitate heteroexchange of the preloaded [3H]D-aspartate.

Identification of the Serum Complex Which Induces Cerebellar Granule Cell In Vitro Differentiation and Resistance to Excitatory Amino Acids

The protein complex promoting in vitro terminal differentiation of cerebellar granule cells has been isolated from rabbit serum. We designate the complex the neurite outgrowth and adhesion complex (NOAC). The apparent molecular weight, evaluated by gel filtration, is 80 - 100 kDa. Rat cerebellar granule cells cultured in NOAC exhibit much lower glial cell contamination and survive, in their differentiated state, much longer than in 10% foetal calf serum. While they bind tetanus toxin, express specific antigens such as synapsin I, synaptophisin and A2B5, and release [3H]d-aspartate in a fashion similar to that shown by cells cultured in foetal calf serum, they show a 60% reduction in the total number of kainate binding sites. Excitatory amino acid (EAA)-triggered and depolarization-stimulated calcium influx, measured in the presence of different agonists, is 50 - 80% lower in NOAC-cultured cells. NOAC cells are resistant to excitotoxic stimuli carried by EAAs or by depolarizing treatments with 50 mM KCl or 6 microM veratridine. The marked resistance of NOAC-cultured neurons to EAAs can be attributed to decreased calcium entry through EAA-coupled and voltage-gated calcium channels and possibly to other, as yet unidentified, phenotypic properties of these cells. These findings demonstrate that rabbit serum contains one or more polypeptide(s) endowed with the properties of promoting in vitro survival and differentiation of rat cerebellar granule cells and of conferring an EAA-resistant phenotype.

Depolarization-induced release of [(3)H]D-aspartate from GABAergic neurons caused by reversal of glutamate transporters

Cultured neocortical neurons, which predominantly consist of GABAergic neurons exhibit a pronounced stimulus-coupled GABA release. Since the cultures may contain a small population of glutamatergic neurons and the GABAergic neurons have a high content of glutamate it was of interest to examine if glutamate in addition to gamma-aminobutyric acid (GABA) could be released from these cultures. The neurons were preloaded with [(3)H]D-aspartate and subsequently its release was followed during depolarization induced by a high potassium concentration or the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor agonists, AMPA and kainate. Depolarization of the neurons with 55 mM potassium increased the release of [(3)H]D-aspartate by more than 10-fold. When the non-specific calcium-channel blockers cobalt or lanthanum were included in the stimulation buffer with potassium, the release of [(3)H]D-aspartate was decreased by about 40%. These results indicated that some of the released [(3)H]D-aspartate might originate from a vesicular pool. When AMPA was applied to the neurons, the release of [(3)H]D-aspartate was increased 2-fold and could not be prevented or decreased by addition of cobalt. Since AMPA has a rapid desensitizing effect on AMPA receptors, it was examined whether AMPA under non-desensitizing conditions was able to induce an increased release of [(3)H]D-aspartate as compared to the conditions of applying AMPA alone. The desensitization of AMPA receptors was blocked by 6-chloro-3,4-dihydro-3-(2-norbornen-5-yl)-2H-1,2, 4-benzothiadiazine-7-sulphonamide-1,1-dioxide (cyclothiazide). Under the non-desensitizing conditions, the AMPA-induced release of [(3)H]D-aspartate was highly enhanced showing about a 10-fold increase over basal release. Addition of cobalt or lanthanum did not decrease the amount of [(3)H]D-aspartate released, indicating that the release originated from a cytoplasmic pool. Kainate, which induces an almost non-desensitizing effect on AMPA receptors, showed similar results as observed for AMPA under non-desensitizing conditions. The NMDA receptor antagonist (5R,10 S)-(+)-5-methyl-10, 11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine (MK-801) had only minor effects on the [(3)H]D-aspartate release induced by AMPA and kainate. Thus, the depolarization-induced release of [(3)H]D-aspartate from cultured GABAergic neurons appears to be caused mainly by reversal of the glutamate transporters.

Inhibition of [3H]D-aspartate release by deramciclane

The effects of the 5-HT2C receptor inverse agonist deramciclane on the gamma-aminobutyric acid (GABA) uptake and excitatory amino acid release processes were compared in rat cerebrocortical homogenates containing resealed plasmalemma fragments and nerve endings. Deramciclane non-competitively inhibited the uptake of [3H]GABA with a Ki value of 13.7 +/- 0.5 microM and partially displaced specifically bound [3H](R,S)-N-[4,4-bis(3-methyl-2-thienyl)-3-butenyl]nipecotic acid ([3H]NNC-328) with high affinity (IC50 = 2.0 +/- 0.7 nM). Depolarization by 4-aminopyridine or by 4-aminopyridine with (S)-alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate [(S)-AMPA] induced the release of [3H]D-aspartate. Deramciclane (10 microM) partially (approximately 50%) inhibited the release of [3H]D-aspartate without affecting [3H]D-aspartate uptake. These results suggest a role for presynaptic inhibition of excitatory amino acid release and GABA uptake in the anxiolytic properties of deramciclane.

Glutamate transport in cultures from developing avian cerebellum: presence of GLT-1 immunoreactivity in Purkinje neurons

Immunocytochemical studies indicated that Purkinje cells cultured from chick embryonic cerebellum (embryonic day 8) strongly express a glutamate transporter EAAT2 cloned from human brain (GLT-1 in rat brain). At both 7 days and 14 days in culture, Purkinje neurons accumulated 1 microM [3H]L-glutamate via a potent "high-affinity" transport system that could be inhibited by D- and L-threo-3-hydroxyaspartate (D- and L-t-3OHA) and by L-trans-pyrrolidine-2,4-dicarboxylate (L-t-PDC). The order of potency of the three inhibitors was L-t-PDC approximately L-t-3OHA > D-t-30HA. Only the value of IC50 (concentration causing 50% inhibition) for D-t-3OHA significantly changed between 7 days (116 microM) and 14 days in culture (40 microM). All nH approximately 1, except in the case of the inhibition by D-t-3OHA at 14 days in culture (nH = 0.57), indicating the possible appearance of heterogeneity of the transport sites at later stages of culturing. Chronic inhibition of L-glutamate transport by L-t-PDC resulted in major changes in the morphology of Purkinje cells; particularly, the neurites almost completely regressed.

Evidence for a role of presynaptic AMPA receptors in the control of neuronal glutamate release in the rat forebrain

The role of presynaptic alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors in controlling the neuronal release of excitatory amino acids has been investigated. Stimulation of presynaptic AMPA receptors by the endogenous agonist L-glutamate, or by (R,S)-AMPA, dose-dependently enhanced the Ca(2+)-dependent, tetrodotoxin-insensitive, electrically-stimulated release of [3H]D-aspartate from rat forebrain slices. This AMPA receptor-mediated response showed marked stereoselectivity with the activity residing solely in the (S)-isomer. (R)-AMPA was inactive in this respect. AMPA-evoked responses were significantly enhanced in the presence of the AMPA receptor desensitization inhibitor, cyclothiazide (10 microM). Moreover, responses to both AMPA and glutamate were inhibited by competitive (NBQX) and non-competitive (GYKI 52466) AMPA receptor-selective antagonists in a dose-dependent manner. These results provide strong support for the existence of presynaptic AMPA receptors acting to enhance the synaptic release of excitatory amino acids in the mammalian forebrain. Such a positive feedback system may play an important functional role in physiological (e.g., long-term potentiation) and/or pathological (e.g., epileptogenesis) processes in the mammalian central nervous system. AMPA-type autoreceptors may provide new targets for drug action.

Metabotropic glutamate receptors, transmitter output and fatty acids: studies in rat brain slices

1. The effects of (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R-ACPD), a non-selective agonist of the metabotropic glutamate receptors (mGluRs), have been studied in rat cortical and striatal slices by measuring the depolarization-induced output of D-[3H]-aspartate (D-[3H]-Asp) and of [3H]-glutamate ([3H]-Glu), neosynthesized from [3H]-glutamine. 2. In cortical slices, 1S,3R-ACPD potentiated the depolarization-induced (KCl, 30 mM) output of both D-[3H]-Asp and [3H]-Glu. The potentiation, obtained at 300 microM 1S,3R-ACPD was 65 +/- 6% for D-[3H]-Asp and 56 +/- 10% for [3H]-Glu. Conversely, in striatal slices, 1S,3R-ACPD reduced the depolarization-induced transmitter output. The reduction, obtained at 300 microM of the agonist, was 60 +/- 8% for D-[3H]-Asp and 50 +/- 5% for neosynthesized [3H]-Glu. 3. Bovine serum albumin (BSA, 15 microM), which is able to bind locally produced fatty acids, completely eliminated the potentiating effect 1S,3R-ACPD had on D-[3H]-Asp output from cortical slices. Low concentrations of arachidonic acid (1-10 microM) or of oleic acid (1-10 microM) added to BSA-containing perfusion medium, restored this potentiating effect. BSA, however, had no effect on the inhibitory action of 1S,3R-ACPD in striatal slices. 4. Bromophenacyl bromide (100 microM), an inhibitor of phospholipase A2, and RG80267 (100 microM), an inhibitor of diacylglycerol lipase, have been shown to inhibit fatty acid production. These compounds prevented the potentiating effect of 1S,3R-ACPD on D-[3H]-Asp-output in cortical slices. 5. Indomethacin (100 microM), an inhibitor of cyclo-oxygenases, plus nordihydroguaiaretic acid (100 microM), an inhibitor of lipoxygenases, increased D-[3H]-Asp output in cortical slices perfused with BSA-containing medium. 6. These experiments suggest that the mGluR-mediated potentiation of transmitter output requires the availability of unsaturated fatty acids, such as arachidonic or oleic acids, in cortical slices. In contrast, the mGluR-induced inhibition of transmitter output is not dependent upon fatty acid availability in striatal slices. The requirement of both unsaturated fatty acids and 1S,3R-ACPD in the facilitation of transmitter exocytosis may play an important role in the regulation of synaptic plasticity.

The depolarization-induced outflow of D-[3H]aspartate from rat brain slices is modulated by metabotropic glutamate receptors

Rat brain slices were used to study the effects of different metabotropic glutamate receptor ligands on (i) the depolarization (30 mM KCl)-induced outflow of previously taken up D-[3H]aspartate; (ii) the inhibition of forskolin (30 microM)-induced cyclic AMP accumulation; and (iii) the hydrolysis of phosphoinositides. In addition, the localization of mRNAs coding for different metabotropic glutamate receptor subtypes was detected using in situ hybridization. (1S-3R)-1-Aminocyclopentane-1,3-dicarboxylic acid (30-300 microM), a non selective metabotropic glutamate receptor agonist, significantly increased the KCl-induced output of radioactivity from cortical slices, whereas it inhibited the output from striatal slices. Conversely, (1S,3S,4S)-carboxycyclopropylglycine (0.1-1 microM), a relatively selective agonist of the mGluR2 metabotropic glutamate receptor subtype, had an inhibitory effect on the output of D-[3H]aspartate from both cortical and striatal slices and proved to be the most potent metabotropic glutamate receptor agonist in inhibiting cyclic AMP accumulation, but not in stimulating phosphoinositide hydrolysis. Since 2-amino-4-phosphonobutyrate (a mGluR4, mGluR6 and mGluR7 agonist) was not active in any of the assays tested, we hypothesized that the mGluR2 subtype could be involved in these events. Accordingly, mGluR2 mRNA expression was abundant in cortical neurons projecting to the striatum. Our experiments suggest that the stimulation of metabotropic glutamate receptors may either decrease or increase transmitter release depending on the subtype that prevails in the region under study.

Carrier-mediated release of neurotransmitters

There is growing evidence that neurotransmitters can be released not only by exocytosis but also through the membrane carriers responsible for transmitter reuptake. Giulio Levi and Maurizio Raiteri review the in vitro and in vivo evidence supporting the existence of a carrier-mediated release for different classes of transmitters. While the physiological significance of carrier-mediated release remains speculative, widely used drugs such as sympathomimetic amines, the anorectic drug fenfluramine and some drugs of abuse act in part by stimulating monoamine carrier-mediated release. Moreover, antidepressants known to inhibit monoamine reuptake, can block carrier-mediated release. This mechanism may also come into play in pathological conditions such as ischaemia.