Cyclothiazide modulates AMPA receptor-mediated increases in intracellular free Ca2+ and Mg2+ in cultured neurons from rat brain

Allosteric Competition and Inhibition in AMPA Receptors

Excitatory neurotransmission is principally mediated by AMPA-subtype ionotropic glutamate receptors (AMPARs). Dysregulation of AMPARs is the cause of many neurological disorders and how therapeutic candidates such as negative allosteric modulators inhibit AMPARs is unclear. Here, we show that non-competitive inhibition desensitizes AMPARs to activation and prevents positive allosteric modulation. We dissected the noncompetitive inhibition mechanism of action by capturing AMPARs bound to glutamate and the prototypical negative allosteric modulator, GYKI-52466, with cryo-electron microscopy. Noncompetitive inhibition by GYKI-52466, which binds in the transmembrane collar region surrounding the ion channel, negatively modulates AMPARs by decoupling glutamate binding in the ligand binding domain from the ion channel. Furthermore, during allosteric competition between negative and positive modulators, negative allosteric modulation by GKYI-52466 outcompetes positive allosteric modulators to control AMPAR function. Our data provide a new framework for understanding allostery of AMPARs and foundations for rational design of therapeutics targeting AMPARs in neurological diseases.

Ketamine, magnesium and major depression--from pharmacology to pathophysiology and back

The glutamatergic mechanism of antidepressant treatments is now in the center of research to overcome the limitations of monoamine-based approaches. There are several unresolved issues. For the action of the model compound, ketamine, NMDA-receptor block, AMPA-receptor activation and BDNF release appear to be involved in a mechanism, which leads to synaptic sprouting and strengthened synaptic connections. The link to the pathophysiology of depression is not clear. An overlooked connection is the role of magnesium, which acts as physiological NMDA-receptor antagonist: 1. There is overlap between the actions of ketamine with that of high doses of magnesium in animal models, finally leading to synaptic sprouting. 2. Magnesium and ketamine lead to synaptic strengthening, as measured by an increase in slow wave sleep in humans. 3. Pathophysiological mechanisms, which have been identified as risk factors for depression, lead to a reduction of (intracellular) magnesium. These are neuroendocrine changes (increased cortisol and aldosterone) and diabetes mellitus as well as Mg(2+) deficiency. 4. Patients with therapy refractory depression appear to have lower CNS Mg(2+) levels in comparison to health controls. 5. Experimental Mg(2+) depletion leads to depression- and anxiety like behavior in animal models. 6. Ketamine, directly or indirectly via non-NMDA glutamate receptor activation, acts to increase brain Mg(2+) levels. Similar effects have been observed with other classes of antidepressants. 7. Depressed patients with low Mg(2+) levels tend to be therapy refractory. Accordingly, administration of Mg(2+) either alone or in combination with standard antidepressants acts synergistically on depression like behavior in animal models.

CONCLUSION

On the basis of the potential pathophysiological role of Mg(2+)-regulation, it may be possible to predict the action of ketamine and of related compounds based on Mg(2+) levels. Furthermore, screening for compounds to increase neuronal Mg(2+) concentration could be a promising instrument to identify new classes of antidepressants. Overall, any discussion of the glutamatergic system in affective disorders should consider the role of Mg(2+).

Glutamatergic Approaches for the Treatment of Schizophrenia

Glutamate is the major excitatory neurotransmitter in the mammalian central nervous system and plays a key role in most aspects of normal brain function including cognition, learning and memory. Dysfunction of glutamatergic neurotransmission has been implicated in a number of neurological and psychiatric disorders with a growing body of evidence suggesting that hypofunction of glutamatergic neurotransmission via the N-methyl-d-aspartate (NMDA) receptor plays an important role in the pathophysiology of schizophrenia. It thus follows that potentiation of NMDA receptor function via pharmacological manipulation may provide therapeutic utility for the treatment of schizophrenia and a number of different approaches are currently being pursued by the pharmaceutical industry with this aim in mind. These include strategies that target the glycine/d-serine site of the NMDA receptor (glycine transporter GlyT1, d-serine transporter ASC-1 and d-amino acid oxidase (DAAO) inhibitors) together with those aimed at enhancing glutamatergic neurotransmission via modulation of AMPA receptor and metabotropic glutamate receptor function. Such efforts are now beginning to bear fruit with compounds such as the GlyT1 inhibitor RG1678 and mGlu2 agonist LY2140023 proving to have clinical meaningful effects in phase II clinical trials. While more studies are required to confirm long-term efficacy, functional outcome and safety in schizophrenic agents, these agents hold real promise for addressing unmet medical needs, in particular refractory negative and cognitive symptoms, not currently addressed by existing antipsychotic agents.

Role of AMPA Receptor Desensitization and the Side Effects of a DMSO Vehicle on Reticulospinal EPSPs and Locomotor Activity

Activation of the vertebrate locomotor network is mediated by glutamatergic synaptic drive, normally initiated by the brain stem. Previous investigations have studied the role of glutamate receptors, especially NMDA receptors, in generating and regulating locomotor pattern generation. Few studies, however, have focused on the role of AMPA receptors in shaping network activity, especially with regard to their rapid desensitization. It is important to determine whether AMPA receptor desensitization plays a role in regulating neuronal network activity. We examined this question on both the network and synaptic levels in the lamprey ( Lampetra fluviatilis) spinal cord using a selective and potent inhibitor of AMPA receptor desensitization, cyclothiazide (CTZ). The solvent dimethyl sulfoxide (DMSO) is commonly used to dissolve this drug, as well as many others. Unexpectedly, the vehicle alone already at 0.02%, but not at 0.01%, caused significant increases in excitatory postsynaptic potential (EPSP) amplitudes and NMDA-induced locomotor frequency. The results indicate that DMSO may have a profound influence when used ≥0.02%, a concentration 10–50 times less than that most commonly used. Subsequently we applied CTZ concentrations ≤10 μM (DMSO ≤0.01%). CTZ (1.25–5 μM) caused an appreciable and significant increase in EPSPs mediated by non-NMDA receptors and in both AMPA- and NMDA-induced locomotor frequency, but no effects on EPSPs mediated by NMDA receptors. From the effects of CTZ it is apparent that AMPA receptor desensitization plays an important role in determining locomotor frequency and that this is likely a result of its limiting function on AMPA receptor–mediated EPSPs.

Peripheral inflammation-induced increase of AMPA-mediated currents and Ca2+ transients in the presence of cyclothiazide in the rat substantia gelatinosa neurons

This study employing a rodent model of acute pain investigated the influence of carrageenan-induced inflammation on the ability of S-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor activation to induce membrane currents and rises in cytosolic free calcium concentration ([Ca2+]i) in the rat substantia gelatinosa (SG) neurons using simultaneous whole-cell patch-clamp recording and fura-2 calcium imaging in spinal cord slices of L4-L5 segments. The novel finding of this study is that carrageenan-induced inflammation, in the presence of cyclothiazide, an inhibitor of AMPA receptor desensitization, produces a sustained facilitation of the AMPA-mediated membrane current and rises in [Ca2+]i in both the soma and proximal dendrites of SG neurons recorded on the injected side 3 h after the induction of inflammation. These results suggest that in carrageenan-inflamed rats AMPA receptors undergo some alterations that influence AMPA receptors desensitization and/or sensitivity to cyclothiazide.

AMPA receptor-mediated neurotoxicity: role of Ca2+ and desensitization

Glutamate-induced neurodegeneration is the result of excessive stimulation of the different subtypes of glutamate receptors. With regard to the AMPA ((RS)-2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionate) receptors it has been clear from numerous studies that in addition to the Ca2+ permeability of the receptor complexes, their desensitization properties may play a determining role in the neurodegeneration mediated by this subtype of the glutamate receptors. Recent studies have revealed important amino acid residues in the AMPA receptor subunits that control the desensitization kinetics and that may constitute important targets for drugs that may alter the desensitization of the AMPA receptor complexes.

Glutamate receptor desensitization block potentiates the stimulated GABA release through external Ca2+-independent mechanisms from granule cells of olfactory bulb

Glutamate stimulated release of [3H]GABA was studied, during receptor desensitization block and its modulation by voltage gated Ca2+ channels, internal Ca2+ mobilization and GABA transport inhibitors from olfactory bulb slices. Under control conditions, glutamate and agonists induced release was strongly inhibited by Mg/0 Ca2+ Krebs and Cd2+ and partially inhibited by Ni2+ and nifedipine. Cyclothiazide, which blocks desensitization of glutamate receptors, potentiated glutamate, kainate, AMPA and quisqualate induced release. This effect was less dependent of entry of external Ca2+, but was inhibited by trifluoperazine and thapsigargin, inhibitors of Ca2+-calmodulin and endoplasmatic Ca2+ ATPase respectively. Nipecotic acid and NO-711, inhibitors of the GABA transporter, were also able to reduce cyclothiazide potentiated release induced by the 4 secretagogues. Under control conditions, glutamate stimulates the release of GABA in cooperation with VDCC. However, during receptor desensitization block, glutamate stimulated GABA release is mainly modulated through mechanisms dependent on internal Ca2+ mobilization and reversal of the GABA transporter.

Role of kainate receptor activation and desensitization on the [Ca(2+)](i) changes in cultured rat hippocampal neurons

We investigated the role of kainate (KA) receptor activation and desensitization in inducing the increase in the intracellular free Ca(2+) concentration ([Ca(2+)](i)) in individual cultured rat hippocampal neurons. The rat hippocampal neurons in the cultures were shown to express kainate receptor subunits, KA2 and GluR6/7, either by immunocytochemistry or by immunoblot analysis. The effect of LY303070, an alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionate (AMPA) receptor antagonist, on the alterations in the [Ca(2+)](i) caused by kainate showed cell-to-cell variability. The [Ca(2+)](i) increase caused by kainate was mostly mediated by the activation of AMPA receptors because LY303070 inhibited the response to kainate in a high percentage of neurons. The response to kainate was potentiated by concanavalin A (Con A), which inhibits kainate receptor desensitization, in 82.1% of the neurons, and this potentiation was not reversed by LY303070 in about 38% of the neurons. Also, upon stimulation of the cells with 4-methylglutamate (MGA), a selective kainate receptor agonist, in the presence of Con A, it was possible to observe [Ca(2+)](i) changes induced by kainate receptor activation, because LY303070 did not inhibit the response in all neurons analyzed. In toxicity studies, cultured rat hippocampal neurons were exposed to the drugs for 30 min, and the cell viability was evaluated at 24 hr using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The selective activation of kainate receptors with MGA, in the presence of Con A, induced a toxic effect, which was not prevented by LY303070, revealing a contribution of a small subpopulation of neurons expressing kainate receptors that independently mediate cytotoxicity. Taken together, these results indicate that cultured hippocampal neurons express not only AMPA receptors, but also kainate receptors, which can modulate the [Ca(2+)](i) and toxicity.

Synaptic and non-synaptic AMPA receptors permeable to calcium

For a long time, alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA) receptors permeable to calcium have been considered to be either non-existent or as "atypical". There is now ample evidence that these receptors exist in numerous regions of the nervous system and in many neuronal as well as non-neuronal cell populations. This evidence has been accumulated by several methods, including electrophysiological recording, calcium imaging and cobalt-loading. Functional AMPA receptors permeable to calcium are already expressed at very early stages of embryonic development, well before the onset of synaptogenesis. They are probably involved in the paracrine signaling necessary for construction of the nervous system before becoming involved in synaptic transmission. In immature cells, cyclothiazide strongly increases the steady-state level of responses not only to AMPA, but also to kainate. Ingestion, during pregnancy, of food or drug substances that can cross the placental barrier and act upon the embryonic receptors may constitute a risk for normal development. In the adult nervous system, synaptic as well as non-synaptic (paracrine) AMPA receptors permeable to calcium are probably widely expressed in both glial and neuronal cells. They may also participate in controlling some aspects related to adult neurogenesis, in particular the migration of newly formed neurons.

Negative allosteric modulators of AMPA-preferring receptors inhibit [(3)H]GABA release in rat striatum

The effect of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), a selective glutamate receptor agonist, on the release of previously incorporated [(3)H]GABA was examined in superfused striatal slices of the rat. The slices were loaded with [(3)H]GABA in the presence of beta-alanine (1 mM) and superfused with Krebs-bicarbonate buffer containing nipecotic acid (0.1 mM) and aminooxyacetic acid (0.1 mM) to inhibit GABA uptake and metabolism. AMPA (0.01 to 3 mM) increased basal [(3)H]GABA outflow and nipecotic acid potentiated this effect. The [(3)H]GABA releasing effect of AMPA was an external Ca(2+)-dependent process in the absence but not in the presence of nipecotic acid. Cyclothiazide (0.03 mM), a positive modulator of AMPA receptors, failed to evoke [(3)H]GABA release by itself, but it dose-dependently potentiated the [(3)H]GABA releasing effect of AMPA. The AMPA (0.3 mM)-induced [(3)H]GABA release was antagonized by NBQX (0.01 mM) in a competitive fashion (pA(2) 5.08). The negative modulator of AMPA receptors, GYKI-53784 (0.01 mM) reversed the AMPA-induced [(3)H]GABA release by a non-competitive manner (pD'(2) 5.44). GYKI-53784 (0. 01-0.1 mM) also decreased striatal [(3)H]GABA outflow on its own right, this effect was stereoselective and was not influenced by concomitant administration of 0.03 mM cyclothiazide. GYKI-52466 (0. 03-0.3 mM), another negative modulator at AMPA receptors, also inhibited basal [(3)H]GABA efflux whereas NBQX (0.1 mM) by itself was ineffective in alteration of [(3)H]GABA outflow. The present data indicate that AMPA evokes GABA release from the vesicular pool in neostriatal GABAergic neurons. They also confirm that multiple interactions may exist between the agonist binding sites and the positive and negative modulatory sites but no such interaction was detected between the positive and negative allosteric modulators. Since GYKI-53784, but not NBQX, inhibited [(3)H]GABA release by itself, AMPA receptors located on striatal GABAergic neurons may be in sensitized state and phasically controlled by endogenous glutamate. It is also postulated that these AMPA receptors are located extrasynaptically on GABAergic striatal neurons.

NMDA-induced changes in a cortical network in vivo are prevented by AMPA

Analogues of glutamic acid including N-methyl-D-aspartic acid (NMDA) depolarise neurones of the cerebral cortex in vivo and thus change the size of the somatosensory evoked potentials (SEPs). The potentials recover rapidly despite maintained superfusion with NMDA, suggesting a form of neuronal desensitisation or network adaptation. In this study potentials were evoked at the cortical surface by electrical stimulation of the contralateral forepaw and compounds applied topically to the cortical surface by a cortical cup. NMDA at 50-250 microM caused a concentration-dependent decrease in the amplitude of the SEPs, with the highest concentration always abolishing them. AMPA at 50 microM did not affect evoked potentials when applied alone, but prevented the NMDA. Such AMPA-NMDA interactions were inhibited by 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and enhanced by cyclothiazide (which prevents AMPA desensitisation). Superfusion with potassium did not change sensitivity to NMDA. These results suggest that, in the rat cerebral cortex in vivo, activation of AMPA receptors can induce a loss of the network response to activation of NMDA receptors. Such a phenomenon may have physiological and therapeutic implications.

Interactions of glutamate receptor agonists with long-term potentiation in the rat hippocampal slice

Previous work has described the apparent desensitisation of neuronal networks in the rat neocortex to amino acid agonists, following prior exposure several minutes earlier. Since long-term potentiation is believed to involve activation of amino acid receptors, we have now sought to determine whether long-term potentiation can modify the sensitivity of neurones to glutamate receptor agonists in rat hippocampal slices. Responses were measured as the change in population spike or postsynaptic potential (e.p.s.p.) size. Two applications of N-methyl-D-aspartate (NMDA), quinolinic acid, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) or kainate, 45 min apart, did not exhibit any apparent desensitisation. However, the induction of long-term potentiation produced a marked loss of sensitivity to quinolinic acid, with smaller effects on NMDA, AMPA and kainate responses. No marked changes were obtained of e.p. s.p. size. In order to localise the cellular sites of these changes, agonists were also applied by microiontophoresis to the cell bodies or dendritic regions of CA1 neurones. Responses to quinolinic acid showed apparent desensitisation at both sites, whereas no decrease was observed in responses to NMDA or AMPA application. The induction of long-term potentiation again produced a decrease in the size of responses to NMDA and AMPA. Inhibition of nitric oxide (NO) synthase prevented the long-term potentiation-induced loss of responsiveness to NMDA, but not AMPA, implying a role for NO in the loss of NMDA sensitivity. Recordings of single cell activity during the iontophoretic application of agonists and induction of long-term potentiation showed that responses to NMDA were often suppressed to a greater extent than to quinolinic acid. The results indicate that long-term potentiation can modify the sensitivity of hippocampal neurones to glutamate receptor agonists, and that differences exist in the pharmacology of NMDA and quinolinic acid.

Role of desensitization of AMPA receptors on the neuronal viability and on the [Ca2+]i changes in cultured rat hippocampal neurons

We investigated the role of desensitization of alpha-amino-3-hydroxy-5-methyl-isoxazole-4-propionate (AMPA) receptors on the neurotoxicity and on the [Ca2+]i changes induced by kainate or by AMPA in cultured rat hippocampal neurons. The neuronal viability was evaluated either by the 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) assay, or by analysis of cell morphology. Short-term exposure of the neurons to kainate or AMPA (30 min) was not toxic, but the exposure for 24 h to the excitotoxic drugs caused a concentration-dependent neurotoxic effect which was prevented by LY 303070, a noncompetitive AMPA receptor antagonist. In the presence of cyclothiazide (CTZ), kainate or AMPA was toxic (30 min exposure), or the toxic effect was significantly enhanced (24 h exposure), but in this case LY 303070 did not completely protect the cells against kainate-induced toxicity. The alterations in the [Ca2+]i caused by kainate or AMPA showed a great cell-to-cell variability. LY 303070 completely or partially inhibited the responses stimulated by kainate. CTZ differentially affected the responses evoked by kainate or AMPA. In the majority of hippocampal neurons, CTZ did not potentiate, or only slightly potentiated, the kainate-stimulated responses but in 11% of neurons there was a great potentiation. In AMPA-stimulated neurons, the responses were slightly or greatly potentiated in the majority of neurons, but not in all of them. The results show that AMPA and kainate may be toxic, depending on the time of exposure and on the blockade of the desensitization of the AMPA receptors. Overall, our results clearly show that there exist different populations of hippocampal neurons with different sensitivities to kainate, AMPA, CTZ and LY 303070. Moreover, the effects of CTZ on both [Ca2+]i alterations and neurotoxicity are not fully correlated.

A novel presynaptic inhibitory mechanism underlies paired pulse depression at a fast central synapse

Several distinct mechanisms may cause synaptic depression, a common form of short-term synaptic plasticity. These include postsynaptic receptor desensitization, presynaptic depletion of releasable vesicles, or other presynaptic mechanisms depressing vesicle release. At the endbulb of Held, a fast central calyceal synapse in the auditory pathway, cyclothiazide (CTZ) abolished marked paired pulse depression (PPD) by acting presynaptically to enhance transmitter release, rather than by blocking postsynaptic receptor desensitization. PPD and its response to CTZ were not altered by prior depletion of the releasable vesicle pool but were blocked by lowering external calcium concentration, while raising external calcium enhanced PPD. We conclude that a major component of PPD at the endbulb is due to a novel, transient depression of release, which is dependent on the level of presynaptic calcium entry and is CTZ sensitive.

In vivo studies of the cerebral glutamate receptor/NO/cGMP pathway

Overwhelming evidence indicates that the glutamate/nitric oxide (NO) synthase/soluble guanylyl cyclase system is of primary importance in a variety of physiological and pathological processes of the brain. Most of our knowledge on this neurochemical pathway derives from in vitro and ex vivo studies but the recent improvement of microdialysis techniques combined with extremely sensitive measurements of the amplified end-product cyclic GMP (cGMP) has given new impulses to the investigation of this cascade of events, its modulation by neurotransmitters and its functional relevance, in a living brain. The first reports, appeared in the early 90's, have demonstrated that microdialysis monitoring of cGMP in the extracellular environment of the cerebellum and hippocampus exactly reflects what is expected to occur at the intracellular level; thus, in vivo extracellular cGMP is sensitive to NO-synthase and soluble guanylyl cyclase inhibitors, can be increased by NO-donors or phosphodiesterase blockers and is modulated by glutamate receptor stimulation in a NO-dependent fashion. Since then, other microdialysis studies have been reported showing that the brain NO synthase/guanylyl cyclase pathway is mainly controlled by NMDA, AMPA and metabotropic glutamate receptors but can be also influenced by other transmitters (GABA, acetylcholine, neuropeptides) through polysynaptic circuits interacting with the glutamatergic system. The available data indicate that this technique, applied to freely-moving animals and combined with behavioural tests, could be useful to get a better insight into the functional roles played by NO and cGMP in physiological and pathological situations such as learning, memory formation, epilepsy, cerebral ischemia and neurodegenerative diseases.

Interactions of GYKI 52466 and NBQX with cyclothiazide at AMPA receptors: experiments with outside-out patches and EPSCs in hippocampal neurones

In outside-out patches from cultured hippocampal neurones, glutamate (1 mM) applied for 1 ms evoked currents which rose rapidly (tau(on) 451 +/- 31 micros) to a peak and then deactivated with slower kinetics (1.95 +/- 0.13 ms). Offset time constants were significantly slower with longer application durations (tau(off) 3.10 +/- 0.19, 3.82 +/- 0.25, 4.80 +/- 0.65 and 7.56 +/- 0.65 ms with 10, 20, 100 and 500 ms applications respectively). Desensitization was complete within 100 ms with a similar rate for all application durations (4.74 +/- 0.34 ms with 100 ms applications). GYKI 52466 reduced inward peak currents with an IC50 of 11.7 +/- 0.6 microM and had similar potency on steady-state currents to longer glutamate applications. GYKI 52466 had no significant effect on desensitization or deactivation time constants but caused a modest and significant prolongation of onset kinetics at higher concentrations. Cyclothiazide (100 microM) potentiated steady-state currents 25-fold at 100 ms and caused a modest but significant slowing in onset kinetics (601 +/- 49 micros with 1 ms applications) but a more pronounced prolongation of deactivation time constants (5.55 +/- 0.66 ms with 1 ms applications). In 50% of neuronal patches cyclothiazide completely eliminated desensitization. In those patches with residual desensitization, the rate was not significantly different to control (5.36 +/- 0.43 ms with 100 ms applications). Following 100 ms applications of glutamate, GYKI 52466 had IC50s of 11.7 +/- 1.1 microM and 75.1 +/- 7.0 microM in the absence and presence of cyclothiazide (100 microM) respectively. Onset kinetics were slowed from 400 +/- 20 micros to 490 +/- 30 micros by cyclothiazide (100 microM) and then further prolonged by GYKI 52466 (100 microM) to a double exponential function (tau(on1) 1.12 +/- 0.13 ms and tau(on2) 171.5 +/- 36.5 ms). GYKI 52466 did not re-introduce desensitization but concentration-dependently weakened cyclothiazide's prolongation of deactivation time constants (1 ms applications: 5.01 +/- 0.71, 4.47 +/- 0.80 and 2.28 +/- 0.64 ms with GYKI 52466 30, 100 and 300 microM respectively). NBQX reduced peak current responses with an IC50 of 28.2 +/- 1.3 nM. Paradoxically, steady-state currents with 500 ms applications of glutamate were potentiated from 3.3 +/- 1.2 pA to 29.4 +/- 6.4 pA by NBQX (1 nM). Higher concentrations of NBQX then antagonized this potentiated response. The potency of NBQX in antagonizing steady-state currents to 500 ms applications of glutamate (IC50 120.9 +/- 30.2 nM) was 2-fold less than following 100 ms applications (IC50 67.7 +/- 2.6 nM). NBQX had no effect on rapid onset, desensitization or deactivation time constants. However, a slow relaxation of inhibition was seen with longer applications. NBQX was 2-5-fold less potent against inward currents in the presence of cyclothiazide (100 microM) depending on the application duration but had no effect on the rapid onset, desensitization or deactivation time constants. The same relaxation of inhibition was seen as with NBQX alone. NBQX (1 microM) reduced AMPA receptor-mediated EPSC amplitude to 7 +/- 1% of control with no effect on kinetics. Cyclothiazide (330 microM) caused a 2.8-fold prolongation of the decay time constant (control 26.6 +/- 2.2 ms, cyclothiazide 74.2 +/- 7.6 ms, n = 9). Additional application of NBQX (1 microM) partly reversed this prolongation to 1.9 fold (47.7 +/- 2.5 ms, n = 5). These results support previous findings that cyclothiazide also allosterically influences AMPA receptor agonist/antagonist recognition sites. There were no interactions between NBQX and cyclothiazide on desensitization or deactivation time constants of glutamate-induced currents but clear interactions on EPSC deactivation kinetics. This raises the possibility that the interactions of NBQX, GYKI 52466 and cyclothiazide on AMPA-receptor-mediated EPSC kinetics observed are due to modulation of glutamate-release at presynaptic AMPA receptors.

The role of intracellular Na+ and mitochondria in buffering of kainate-induced intracellular free Ca2+ changes in rat forebrain neurones

1. We have examined the mechanisms by which cultured central neurones from embryonic rat brain buffer intracellular Ca2+ loads following kainate receptor activation using fluorescent indicators of [Ca2+]i and [Na+]i. 2. Stimulation of cultured forebrain neurones with 100 microM kainate produced a rapid increase in [Ca2+]i that displayed a variable rate of recovery. Kainate also increased [Na+]i with a response that was slightly slower in onset and markedly slower in recovery. 3. The recovery of [Ca2+]i to baseline was not very sensitive to the [Na+]i. The magnitude of the increase in [Na+]i in response to kainate did not correlate well with the [Ca2+]i recovery time, and experimental manipulations that altered [Na+]i did not have a large impact on the rate of recovery of [Ca2+]i. 4. The recovery of [Ca2+]i to baseline was accelerated by the mitochondrial Na+-Ca2+ exchange inhibitor CGP-37157, suggesting that the recovery rate is influenced by release of Ca2+ from a mitochondrial pool and also that variation in the recovery rate is related to the extent of mitochondrial Ca2+ loading. Kainate did not alter the mitochondrial membrane potential. 5. These studies reveal that mitochondria have a central role in buffering neuronal [Ca2+]i changes mediated by non-N-methyl-D-aspartate (NMDA) glutamate receptors, and that the variation in recovery times following kainate receptor activation reflects a variable degree of mitochondrial Ca2+ loading. However, unlike NMDA receptor-mediated Ca2+ loads, kainate receptor activation has minimal effects on mitochondrial function.

AMPA receptor mediated excitotoxicity in neocortical neurons is developmentally regulated and dependent upon receptor desensitization

AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) excitotoxicity was examined in cultured neocortical neurons using the reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to measure cell viability. Neurons were exposed to AMPA at different culture periods during development of the neurons. In order to describe the pharmacology of AMPA-mediated toxicity, several glutamate receptor antagonists were used: MK-801, NS 394, NBQX, GYKI 52466, GYKI 53405 and GYKI 53655. Increased excitotoxicity was observed when cortical neurons cultured for 5, 8 and 12 days in vitro (DIV) were exposed to a high concentration of AMPA (500 microM) for 6 h. However, only at DIV 12 was part of the toxicity mediated directly through AMPA receptors since 10 microM MK-801 blocked all AMPA toxicity at DIV 5 and 8, but only some of the AMPA response at DIV 12. This indicated that NMDA receptors were being activated, causing some of the observed toxicity. The high dose of AMPA was not sufficient to damage all neurons since 59% remained viable after exposure to AMPA even for neurons that were cultured for 12 DIV. Since it is known that both glutamate and AMPA activate AMPA receptors with a fast and rapidly desensitizing response, this could explain the relatively low toxicity produced by 500 microM AMPA. This was investigated by blocking AMPA receptor desensitization with cyclothiazide. Using a lower concentration (25 microM) of AMPA, addition of 50 microM cyclothiazide increased the AMPA induced excitotoxicity in cultured cortical neurons at all DIV except for DIV 2. This combination of AMPA + cyclothiazide yielded 77% cell death for DIV 12 cultures. In contrast to the results observed with 500 microM AMPA, the neurotoxicity mediated directly by AMPA receptors when desensitization was blocked was seen as early as 5 DIV since 10 microM MK-801 did not completely block the response whereas 10 microM NBQX did. The 2,3-benzodiazepine GYKI compounds, which have been reported to be selective non-competitive AMPA receptor antagonists, were here observed to block the AMPA toxicity with the following rank order: GYKI 53655 > GYKI 52466 > or = GYKI 53405, which is in agreement with their published potencies.

Cyclothiazide and AMPA receptor desensitization: analyses from studies of AMPA-induced release of [3H]-noradrenaline from hippocampal slices

1. Responses in brain produced by the activation of the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) subtype of ionotropic receptor for L-glutamate are often rapidly desensitizing. AMPA-induced desensitization and its characteristics, and the potentiating effect of cyclothiazide were investigated in vitro by analysing AMPA-induced release of [3H]-noradrenaline from prisms of rat hippocampus. 2. AMPA (1-1000 microM) stimulated the release of [3H]-noradrenaline in a concentration-dependent manner that was both calcium-dependent and tetrodotoxin-sensitive, and attenuated by the AMPA-selective antagonists, NBQX (1 and 10 microM), LY 293558 (1 and 10 microM) and GYKI 52466 (10 and 30 microM). 3. By use of an experimental procedure with consecutive applications of AMPA (100 microM, 28 min apart), the second response was reduced, indicative of receptor desensitization, and was reversed by cyclothiazide in a concentration-dependent manner (1-300 microM). The concentration-response curve for AMPA-induced release of [3H]-noradrenaline was shifted leftwards, but the reversal by cyclothiazide of the desensitized response was partial and failed to reach the maximal response of the first stimulus. 4. Observations made with various schedules of cyclothiazide application indicated that the initial AMPA-evoked response was already partially desensitized (150% potentiation by 100 microM cyclothiazide) and that the desensitization was not likely to be due to a time-dependent diminution and was longlasting (second application of cyclothiazide was ineffective). 5. Co-application of a number of drugs with actions on second messenger systems, in association with the second AMPA stimulus, revealed significant potentiation of the AMPA-induced release of [3H]-noradrenaline: forskolin (10 microM, +78%), Rp-cAMPS (100 microM, +65%), Ro 31-8220 (10 microM, +163%) and thapsigargin (100 pM, + 161%). 6. The AMPA receptor-mediated response regulating the release of [3H]-noradrenaline from rat hippocampal slices was desensitized and cyclothiazide acted to reverse partially the desensitization in a concentration-dependent manner. Since the time-course of desensitization was longer lasting than that noted in previous electrophysiological studies, multiple events may be involved in the down-regulation of AMPA receptor activity including receptor phosphorylation and depletion of intracellular Ca2+ stores.

Developing rat brainstem motoneurones in organotypic culture express calcium permeable AMPA-gated receptors

Recent studies suggest that calcium permeable excitatory amino acid receptors may play an important role in many developmental processes including dendritic differentiation, synaptogenesis and activity dependent plasticity or excitotoxicity related disorders. In this work, we investigate the existence of calcium permeable AMPA receptors in embryonic rat motoneurones in organotypic slice culture, which display an early sensitivity to AMPA and Kainate. We used excitatory amino acids induced cobalt uptake and show that developing motoneurones express Ca2+ permeable AMPA receptors. We demonstrate, as already described for other neuronal types in acute slices of adult animals, that the cobalt loading of motoneurones is not suppressed by blockade of synaptic activity by TTX. It is not induced by NMDA stimulation and does not result from the activation of voltage dependent calcium channels. It is specifically suppressed by the non-competitive AMPA antagonist GYKI 53784 (LY303070) and enhanced by the AMPA-receptor desensitization blocker Aniracetam. We conclude that cobalt loading results from the specific activation of AMPA receptors. We further show that, when cobalt loading is induced by threshold doses of Glutamate agonists, cobalt-sulfide deposits are found specifically in primary dendrites, dendritic spines and localized spots on the somatic and peripheral dendritic membrane. We suggest that this particular pattern of staining, different from the Golgi-like staining obtained with excitotoxic doses, may offer new information regarding the membrane density and distribution of calcium permeable AMPA receptors.

Kainate excitotoxicity is mediated by AMPA- but not kainate-preferring receptors in embryonic rat hippocampal cultures

We investigated kainate-induced excitotoxicity in embryonic rat hippocampal cells cultured in a chemically defined medium. Treatment with kainate for 24 h resulted in neuronal death, as assessed by the release of lactate dehydrogenase into the culture media. This neurotoxic effect was kainate dose- and culture age-dependent. EC50 of kainate was 127 +/- 11 microM. 2,3-dihydroxy-6-nitro-7-sulfamoylbenzo (f)quinoxaline (NBQX) completely blocked the toxicity, while MK801, an N-methyl-D-aspartate (NMDA) receptor antagonist, also blocked it but not completely. Furthermore, alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) attenuated the kainate injury, while the selective and noncompetitive AMPA-preferring receptor antagonist 1-(4-aminophenyl)-4-methyl-7, 8-methylenedioxy-5H-2,3-benzo-diazepine (GYKI 52466) blocked it completely. Concanavalin A (ConA), which potentiates the response to kainate at kainate-preferring receptors, had little effect on kainate toxicity. Further, AMPA alone induced little toxicity, but produced remarkable toxicity when cyclothazide was used to block the desensitization of AMPA-preferring receptors. These results indicate that kainate excitotoxicity in hippocampal cultures is mediated by AMPA- but not kainate-preferring receptors, and that it involves NMDA-receptor-mediated toxicity. The non-desensitizing response at AMPA-preferring receptors may play an important role in kainate-induced excitotoxicity.

The glutamate receptor/NO/cyclic GMP pathway in the hippocampus of freely moving rats: modulation by cyclothiazide, interaction with GABA and the behavioural consequences

Monitoring of extracellular cGMP during intracerebral microdialysis in freely moving rats permits the study of the functional changes occurring in the glutamate receptor/nitric oxide (NO) synthase/guanylyl cyclase pathway and the relationship of these changes to animal behaviour. When infused into the rat hippocampus in Mg2+-free medium, cyclothiazide, a blocker of desensitization of the AMPA-preferring receptor, increased cGMP levels. The effect of cyclothiazide (300 microM) was abolished by the NO synthase inhibitor L-NARG (100 microM) or the soluble guanylyl cyclase inhibitor ODQ (100 microM). During cyclothiazide infusion the animals displayed a pre-convulsive behaviour characterized by frequent "wet dog shakes" (WDS). Neither L-NARG nor ODQ decreased the WDS episodes. Both cGMP and WDS responses elicited by cyclothiazide were prevented by blocking NMDA receptor function with the glutamate site antagonist CGS 19755 (100 microM), the channel antagonist MK-801 (30 microM) or Mg2+ ions (1 mM). The AMPA/kainate receptor antagonists DNQX (100 microM) and NBQX (100 microM) abolished the WDS episodes but could not inhibit the cyclothiazide-evoked cGMP response. DNQX or NBQX (but not MK-801) elevated, on their own, extracellular cGMP levels. The cGMP response elicited by the antagonists appears to be due to prevention of a glutamate-dependent inhibitory GABAergic tone, since infusion of bicuculline (50 microM) caused a strong cGMP response. The results suggest that (a) AMPA/kainate receptors linked to the NO/cGMP pathway in the hippocampus (but not NMDA receptors) are tonically activated and kept in a desensitized state by endogenous glutamate; (b) blockade of AMPA/kainate receptor desensitization by cyclothiazide leads to endogenous activation of NMDA receptors; (c) the hippocampal NO/cGMP system is under a GABAergic inhibitory tone driven by non-NMDA ionotropic receptors; (d) the pre-convulsive episodes observed depend on hippocampal NMDA receptor activation but not on NO and cGMP production.

AMPA neurotoxicity in cultured cerebellar granule neurons: mode of cell death

Various forms of cell death induced by the glutamate receptor agonist, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), were analyzed by determining the capacity of cultured cerebellar granule cells to metabolize 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) into formazan, by measuring the leakage of lactate dehydrogenase (LDH), by using confocal microscopy to visualize propidium iodide staining of apoptotic nuclei, and by using field inversion gel electrophoresis (FIGE) for the detection of AMPA-produced cleavage of DNA into high molecular-weight fragments (50 kbp). All these measures indicated that stimulation of AMPA receptors may be involved in the neurotoxic effects of glutamate, and that AMPA-induced neurotoxicity in cerebellar granule cells display morphologically distinct features of both necrotic and apoptotic modes of cell death. In agreement with previous observations, a blockade of AMPA receptor desensitization was necessary to unmask AMPA-induced functional responses in cultured cerebellar granule neurons in vitro. Microfluorimetric measurements of free cytoplasmic calcium concentrations ([Ca2+]i) in single cerebellar neurons revealed that AMPA neurotoxicity was accompanied by a pronounced elevation of [Ca2+]i. Our current results add further evidence to the notion that glutamate-induced neurotoxicity in cerebellar granule cells is mediated not only through NMDA receptors but also through a direct activation of AMPA receptor-regulated cation channels.

Interactions of 2,3-benzodiazepines and cyclothiazide at AMPA receptors: patch clamp recordings in cultured neurones and area CA1 in hippocampal slices

1. The 2,3-benzodiazepines GYKI 52466, GYKI 53405 and GYKI 53655 antagonized AMPA-induced currents in cultured superior colliculus neurones in a non use-dependent manner (steady state IC50s: GYKI 52466 9.8 +/- 0.6 microM; GYKI 53405 3.1 +/- 0.6 microM; GYKI 53655 0.8 +/- 0.1 microM). 2. Higher concentrations of all three antagonists slowed the onset kinetics and quickened the offset kinetics of AMPA-induced currents indicative of an allosteric interaction with the AMPA recognition site. 3. Cyclothiazide (3-300 microM) dramatically slowed desensitization of AMPA-induced currents and potentiated steady state currents (EC50 10.0 +/- 2.5 microM) to a much greater degree than peak currents. Both tau on and tau off were also increased by cyclothiazide in a concentration-dependent manner (EC50: tau on 42.1 +/- 4.5 microM; tau off 31.6 +/- 6.6 microM). 4. Cyclothiazide (10-100 microM) shifted the concentration-response curves of the 2,3-benzodiazepines to the right. For example, with 10 microM cyclothiazide the IC50s of GYKI 52466 and GYKI 53405 on steady-state AMPA-induced currents were 57.9 +/- 9.5 and 41.6 +/- 1.5 microM, respectively. 5. GYKI 53405 and GYKI 52466 concentration-dependently reversed the effects of cyclothiazide (100 microM) on offset kinetics (GYKI 53405 IC50 16.6 +/- 4.2 microM). However, the 2,3-benzodiazepines were unable to reintroduce desensitization in the presence of cyclothiazide and even concentration-dependently slowed the onset kinetics of AMPA responses further (GYKI 53405 EC50 8.0 +/- 2.8 microM). 6. GYKI 52466 decreased the peak amplitude of hippocampal area CA1 AMPA receptor-mediated excitatory postsynaptic currents (e.p.s.cs) (IC50 10.8 +/- 0.8 microM) with no apparent effect on response kinetics. Cyclothiazide prolonged the decay time constant of AMPA receptor-mediated e.p.s.cs (EC50 35.7 +/- 6.5 microM) with less pronounced effects in slowing e.p.s.c. onset kinetics and increasing e.p.s.c. amplitude. 7. Cyclothiazide (330 microM) shifted the concentration-response curve for the effects of GYKI 52466 on AMPA receptor-mediated e.p.s.c. peak amplitude to the right (GYKI 52466 IC50 26.9 +/- 9.4 microM). Likewise, GYKI 52466 (30-100 microM)) shifted the concentration-response curve for the effects of cyclothiazide on AMPA receptor-mediated e.p.s.c. decay time constants to the right. 8. In conclusion, cyclothiazide and the 2,3-benzodiazepines seem to bind to different sites on AMPA receptors but exert strong allosteric interactions with one another and with other domains such as the agonist recognition site. The interactions of GYKI 52466 and cyclothiazide on AMPA receptor-mediated e.p.s.cs in area CA1 of hippocampal slices provide evidence that the decay time constant of these synaptic events are not governed by desensitization.