Ca2+ permeability of KA-AMPA--gated glutamate receptor channels depends on subunit composition.

NMDA (N-methyl-D-aspartate) receptors and non-NMDA receptors represent the two major classes of ion channel-linked glutamate receptors. Unlike the NMDA receptor channels, non-NMDA receptor channels have usually been thought to conduct monovalent cations only. Non-NMDA receptor ion channels that can be gated by kainic acid (KA) and alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) are formed by the glutamate receptor subunits GluR1, GluR2, and GluR3. These subunits were expressed in various combinations in Xenopus oocytes so that their permeability to divalent cations could be studied. At physiological resting potentials, KA and AMPA elicited inward calcium currents in oocytes expressing GluR1, GluR3, and GluR1 plus GluR3. In contrast, oocytes expressing GluR1 plus GluR2 or GluR3 plus GluR2 showed no such permeability. Thus, in neurons expressing certain KA-AMPA receptor subunits, glutamate may trigger calcium-dependent intracellular events by activating non-NMDA receptors.

Hyperalgesia mediated by spinal glutamate or substance P receptor blocked by spinal cyclooxygenase inhibition.

Inhibition of cyclooxygenase by nonsteroidal anti-inflammatory drugs (NSAIDs) in the periphery is commonly accepted as the primary mechanism by which these agents produce a selective attenuation of pain (analgesia). NSAIDs are now shown to exert a direct spinal action by blocking the excessive sensitivity to pain (hyperalgesia) induced by the activation of spinal glutamate and substance P receptors. These findings demonstrate that the analgesic effects of NSAIDs can be dissociated from their anti-inflammatory actions. Spinal prostanoids are thus critical for the augmented processing of pain information at the spinal level.

Participation of postsynaptic PKC in cerebellar long-term depression in culture.

Long-term depression (LTD) in the intact cerebellum is a decrease in the efficacy of the parallel fiber-Purkinje neuron synapse induced by coactivation of climbing fiber and parallel fiber inputs. In cultured Purkinje neurons, a similar depression can be induced by iontophoretic glutamate pulses and Purkinje neuron depolarization. This form of LTD is expressed as a depression of alpha-amino-3-hydroxy-5-methyl-4- isoxazole-propionic acid (AMPA)-mediated current, and its induction is dependent on activation of metabotropic quisqualate receptors. The effect of inhibitors of protein kinase C (PKC) on LTD induction was studied. Inhibitors of PKC blocked LTD induction, while phorbol-12,13-diacetate (PDA), a PKC activator, mimicked LTD. These results suggest that PKC activation is necessary for the induction of cerebellar LTD.

Flip and flop: a cell-specific functional switch in glutamate-operated channels of the CNS.

In the central nervous system (CNS), the principal mediators of fast synaptic excitatory neurotransmission are L-glutamate-gated ion channels that are responsive to the glutamate agonist alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA). In each member of a family of four abundant AMPA receptors, a small segment preceding the predicted fourth transmembrane region has been shown to exist in two versions with different amino acid sequences. These modules, designated "flip" and "flop," are encoded by adjacent exons of the receptor genes and impart different pharmacological and kinetic properties on currents evoked by L-glutamate or AMPA, but not those evoked by kainate. For each receptor, the alternatively spliced messenger RNAs show distinct expression patterns in rat brain, particularly in the CA1 and CA3 fields of the hippocampus. These results identify a switch in the molecular and functional properties of glutamate receptors operated by alternative splicing.

A family of AMPA-selective glutamate receptors.

Four cloned cDNAs encoding 900-amino acid putative glutamate receptors with approximately 70 percent sequence identity were isolated from a rat brain cDNA library. In situ hybridization revealed differential expression patterns of the cognate mRNAs throughout the brain. Functional expression of the cDNAs in cultured mammalian cells generated receptors displaying alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-selective binding pharmacology (AMPA = quisqualate greater than glutamate greater than kainate) as well as cation channels gated by glutamate, AMPA, and kainate and blocked by 6,7-dinitroquinoxaline-2,3-dione (CNQX).

Zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons.

Large amounts of zinc are present in synaptic vesicles of mammalian central excitatory boutons and may be released during synaptic activity, but the functional significance of the metal for excitatory neurotransmission is currently unknown. Zinc (10 to 1000 micromolar) was found to have little intrinsic membrane effect on cortical neurons, but invariably produced a zinc concentration-dependent, rapid-onset, reversible, and selective attenuation of the membrane responses to N-methyl-D-aspartate, homocysteate, or quinolinate. In contrast, zinc generally potentiated the membrane responses to quisqualate or alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate and often did not affect the response to kainate. Zinc also attenuated N-methyl-D-aspartate receptor-mediated neurotoxicity but not quisqualate or kainate neurotoxicity. The ability of zinc to specifically modulate postsynaptic neuronal responses to excitatory amino acid transmitters, reducing N-methyl-to-aspartate receptor-mediated excitation while often increasing quisqualate receptor-mediated excitation, is proposed to underlie its normal function at central excitatory synapses and furthermore could be relevant to neuronal cell loss in certain disease states.

The calcium channel blocker nifedipine attenuates slow excitatory amino acid neurotoxicity.

High concentrations of potent N-methyl-D-aspartate (NMDA) agonists can trigger degeneration of cultured mouse cortical neurons after an exposure of only a few minutes; in contrast, selective non-NMDA agonists or low levels of NMDA agonists require exposures of several hours to induce comparable damage. The dihydropyridine calcium channel antagonist nifedipine was used to test whether this slow neurotoxicity is mediated by a calcium influx through voltage-gated channels. Nifedipine had little effect on the widespread neuronal degeneration induced by brief exposure to high concentrations of NMDA but substantially attenuated the neurotoxicity produced by 24-hour exposure to submaximal concentrations of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate, kainate, or quinolinate. Calcium ion influx through dihydropyridine-sensitive, voltage-dependent calcium channels may be an important step in the neuronal injury induced by the prolonged activation of NMDA or non-NMDA glutamate receptors.

Quinoxalinediones: potent competitive non-NMDA glutamate receptor antagonists.

The N-methyl-D-aspartate (NMDA)-subtype of glutamate receptors has been well described as a result of the early appearance of NMDA antagonists, but no potent antagonist for the "non-NMDA" glutamate receptors has been available. Quinoxalinediones have now been found to be potent and competitive antagonists at non-NMDA glutamate receptors. These compounds will be useful in the determination of the structure-activity relations of quisqualate and kainate receptors and the role of such receptors in synaptic transmission in the mammalian brain.

2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline: a neuroprotectant for cerebral ischemia.

2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) is an analog of the quinoxalinedione antagonists to the non-N-methyl-D-aspartate (non-NMDA) glutamate receptor. NBQX is a potent and selective inhibitor of binding to the quisqualate subtype of the glutamate receptor, with no activity at the NMDA and glycine sites. NBQX protects against global ischemia, even when administered 2 hours after an ischemic challenge.

A long-term depression of AMPA currents in cultured cerebellar Purkinje neurons.

Cerebellar long-term depression (LTD) is a model of synaptic plasticity in which conjunctive stimulation of parallel fiber and climbing fiber inputs to a Purkinje neuron induces a persistent depression of the parallel fiber-Purkinje neuron synapse. We report that an analogous phenomenon may be elicited in the cultured mouse Purkinje neuron when iontophoretic glutamate application and depolarization of the Purkinje neurons are substituted for parallel fiber and climbing fiber stimulation, respectively. The induction of LTD in these cerebellar cultures requires activation of both ionotropic (AMPA) and metabotropic quisqualate receptors, together with depolarization in the presence of external Ca2+. This postsynaptic alteration is manifest as a depression of glutamate or AMPA currents, but not aspartate or NMDA currents. These results strengthen the contention that the expression of cerebellar LTD is at least in part postsynaptic and provide evidence that activation of both ionotropic and metabotropic quisqualate receptors are necessary for LTD induction.

GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses.

In whole-cell voltage-clamp recordings from cultured rat hippocampal neurons, the 2,3-benzodiazepine GYKI 52466 was a potent antagonist of kainate- and AMPA-activated currents (IC50 values, 7.5 and 11 microM, respectively), but was inactive against N-methyl-D-aspartate (NMDA) or gamma-aminobutyric acid responses. The block produced by GYKI 52466 occurred in a noncompetitive fashion, was voltage independent, and failed to show use dependence, indicating an allosteric blocking mechanism. In kinetic experiments with kainate as the agonist, the GYKI 52466 binding and unbinding rates were 1.6 x 10(5) M-1 s-1 and 3.2 s-1, respectively. GYKI 52466 also suppressed non-NMDA receptor-mediated spontaneous synaptic currents via a postsynaptic action. Non-competitive AMPA/kainate antagonists such as GYKI 52466 could offer advantages over competitive antagonists in the treatment of glutamate-associated neurological disorders, particularly under conditions in which high levels of the amino acid would render the competitive antagonists relatively ineffective. Moreover, the results demonstrate the existence of a novel recognition site for an atypical benzodiazepine on non-NMDA receptors.

Glutamate-operated channels: developmentally early and mature forms arise by alternative splicing.

The expression of two alternative splice variants, Flip and Flop, in mRNAs encoding the four AMPA-selective glutamate receptors (GluR-A, -B, -C, and -D) was studied in the developing brain by in situ hybridization. These receptors are expressed prominently before birth, and patterns of distribution for Flip versions remain largely invariant during postnatal brain development. In contrast, the Flop versions are expressed at low levels prior to postnatal day 8. Around this time, the expression of Flop mRNAs increases throughout the brain, reaching adult levels by postnatal day 14. Thus, receptors carrying the Flop module appear to participate in mature receptor forms.

The distribution of glutamate receptors in cultured rat hippocampal neurons: postsynaptic clustering of AMPA-selective subunits.

The distribution of several glutamate receptor subunits was investigated in cultured rat hippocampal neurons by in situ hybridization and immunocytochemistry. The AMPA/kainate-selective receptors GluR1-6 exhibited two patterns of mRNA expression: most neurons expressed GluR1, R2, and R6, whereas only about 20% expressed significant levels of GluR3, R4, and R5. By immunocytochemistry, the metabotropic glutamate receptor mGluR1 alpha was detectable only in a subpopulation of GABAergic interneurons. GluR1 and GluR2/3 segregated to the somatodendritic domain within the first week in culture, even in the absence of synaptogenesis. Glutamate receptor-enriched spines developed later and were present only on presumptive pyramidal cells, not on GABAergic interneurons. Clusters of GluR1 and GluR2/3 completely colocalized and were restricted to a subset of postsynaptic sites. Thus, glutamate receptor subunits exhibit both a cell type-specific expression and a selective subcellular localization.

AMPA receptor activation potentiates zinc neurotoxicity.

Extracellular Zn2+ attenuates NMDA receptor-mediated neurotoxicity and increases AMPA receptor-mediated toxicity. Known electrophysiological effects of Zn2+ predict only the former. We considered the possibility that the latter rather reflects AMPA potentiation of Zn2+ toxicity, perhaps mediated by neuronal depolarization and Zn2+ entry through voltage-gated Ca2+ channels. High K+ or kainate also potentiated Zn2+ toxicity, and AMPA plus Zn2+ toxicity was attenuated by raising extracellular Ca2+, or by Ca2+ channel blockers. AMPA plus Zn2+ exposure induced an increase in fluorescence from neurons loaded with the Zn(2+)-sensitive dye TS-Q and increased subsequent 45Ca2+ accumulation. The ability of AMPA receptor activation to potentiate Zn2+ toxicity may be relevant to neuronal death associated with intense activation of glutamatergic pathways.

Differential expression of excitatory amino acid receptor subtypes in cultured cerebellar neurons.

Using neurotoxicity and inositol phosphate release as criteria for receptor expression, we report the differential expression of excitatory amino acid receptor subtypes in cerebellar granule cells grown in serum-free media containing either high (25 mM) or low (5 mM) KCl. NMDA receptors are expressed in neurons grown in high, but not low, KCl. In contrast, ionotropic quisqualate receptors are expressed in neurons grown in low KCl, but not in those grown in high KCl. Addition of NMDA to cultures containing low KCl appears to mimic high KCl conditions: NMDA receptors are expressed, but ionotropic quisqualate receptors are not. Glutamate and kainate are toxic to cells grown in either condition.

Zinc has opposite effects on NMDA and non-NMDA receptors expressed in Xenopus oocytes.

Pharmacological characterization of Zn2+ effects on glutamate ionotropic receptors was investigated in Xenopus oocytes injected with rat brain mRNA, using a double microelectrode, voltage-clamp technique. At low concentration, Zn2+ inhibited NMDA currents (IC50 = 42.9 +/- 1.3 microM) and potentiated both AMPA (EC50 = 30.0 +/- 1.2 microM) and desensitized kainate responses (EC50 = 13.0 +/- 0.1 microM). At higher concentrations, Zn2+ inhibited non-NMDA responses with IC50 values of 1.3 +/- 0.1 mM and 1.2 +/- 0.3 mM for AMPA and kainate, respectively. The potentiation of AMPA or quisqualate currents by Zn2+ was more than 2-fold, whereas that of the kainate current was only close to 30%. This potentiating effect of Zn2+ on AMPA current modified neither the affinity of the agonist for its site nor the current-voltage relationship. In addition, 500 microM Zn2+ differentially affected NMDA and non-NMDA components of the glutamate-induced response. The possible physiological relevance of Zn2+ modulation is discussed.

Kinetic analysis of interactions between kainate and AMPA: evidence for activation of a single receptor in mouse hippocampal neurons.

AMPA but not kainate produces a rapidly desensitizing response in mouse hippocampal neurons. The characteristic action of these agonists appears to arise from activation of a single receptor with active and desensitized states, for which AMPA and kainate have different relative affinity. The equilibrium potency of a series of five agonists that produce rapidly desensitizing responses at non-NMDA receptors (EC50 1 microM to 4 mM) was similar to their equilibrium potency for block of kainate responses. Increasing the concentration of kainate overcame such block, but in the presence of AMPA the rate of activation of responses to kainate was slowed. Conversely, in the presence of kainate the amplitude of rapidly desensitizing responses evoked by AMPA was reduced, and the rate of onset of desensitization was slowed.

Identification of amino acid residues in GluR1 responsible for ligand binding and desensitization.

Although GluR1(o) and GluR3(o) are homologous at the amino acid level, GluR3(o) desensitizes approximately threefold faster than GluR1(o). By creating chimeras of GluR1(o) and GluR3(o) and point amino acid exchanges in their S2 regions, two residues were identified to be critical for GluR1(o) desensitization: Y716 and the R/G RNA-edited site, R757. With creation of the double-point mutant (Y716F, R757G)GluR1(o), complete exchange of the desensitization rate of GluR1(o) to that of GluR3(o) was obtained. In addition, both the potency and affinity of the subtype-selective agonist bromohomoibotenic acid were exchanged by the Y716F mutation. A model is proposed of the AMPA receptor binding site whereby a hydrogen-bonding matrix of water molecules plays an important role in determining both ligand affinity and receptor desensitization properties. Residues Y716 in GluR1 and F728 in GluR3 differentially interact with this matrix to affect the binding affinity of some ligands, providing the possibility of developing subtype-selective compounds.

4-Alkylated homoibotenic acid (HIBO) analogues: versatile pharmacological agents with diverse selectivity profiles towards metabotropic and ionotropic glutamate receptor subtypes.

4-Alkylated analogues of homoibotenic acid (HIBO) have previously shown high potency and selectivity at ionotropic and metabotropic glutamic acid receptor (iGluR and mGluR) subtypes. Compounds with different selectivity profiles are valuable pharmacological tools for neuropharmacological studies, and the series of 4-alkyl-HIBO analogues have been extended in this paper in the search for versatile agents. Pharmacological characterization of five new analogues, branched and unbranched 4-alkyl-HIBO analogues, have been carried out. The present compounds are all weak antagonists at Group I mGluRs (mGluR1 and 5) presenting only small differences in potencies (Ki values ranging from 89 to 670 microM). Affinities were studied at native and cloned iGluRs, and the compounds described show preference for the AMPA receptor subtypes GluR1 and 2 over GluR3 and 4. However, compared to previous 4-alkyl-HIBO analogues, these compounds show a remarkably high affinity for the Kain preferring subtype GluR5. The observed GluR5 affinities were either similar or higher compared to their GluR1 and 2 affinity. Isopropyl-HIBO showed the highest affinity for GluR5 (Ki=0.16 microM), and represents a unique compound with high affinity towards the three subtypes GluR1, 2 and 5. In general, these compounds represent new selectivity profiles compared to previously reported Glu receptor analogues.

NMDA, AMPA, and benzodiazepine binding site changes in Alzheimer's disease visual cortex.

Quantitative receptor autoradiography was used to measure the laminar distribution of [3H]glycine and [3H]glutamate binding to the N-methyl-D-aspartate (NMDA) receptor complex, [3H]D,L-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) binding to the AMPA receptor, and [3H]flunitrazepam binding to the benzodiazepine (BDZ) receptor in three areas of visual cortex in control and Alzheimer's disease (AD) postmortem human brains (primary or striate visual cortex, visual association cortex, and higher-order visual association cortex, corresponding to Brodmann Areas 17, 18, and 21, respectively). In Area 17, binding to the NMDA, AMPA, and BDZ receptors was not significantly altered in the AD brains (except in layer VI for [3H]glycine and layer III for [3H]AMPA, where binding was reduced in the AD brains). Ligand binding to the two EAA receptors in Area 18 was, however, significantly reduced in the AD brains (layers I through III for [3H]glycine and layers III through VI for [3H]AMPA). In Area 21, binding to both the NMDA and BDZ receptors but not to the AMPA receptor, was significantly reduced in almost all laminae of the AD brains (layers I through VI for [3H]glycine and layers I through V for [3H]flunitrazepam). This hierarchical pattern of laminar binding loss with increasing complexity of association visual cortices is consistent with the increasing numbers of neurofibrillary tangles found in those areas, implicating NMDA and BDZ receptor bearing cells in AD neuropathology. AMPA receptor losses do not parallel the pathology, suggesting that AMPA receptors are not directly correlated with the pathology.