Mechanisms for activation and antagonism of an AMPA-sensitive glutamate receptor: crystal structures of the GluR2 ligand binding core

Crystal structures of the GluR2 ligand binding core (S1S2) have been determined in the apo state and in the presence of the antagonist DNQX, the partial agonist kainate, and the full agonists AMPA and glutamate. The domains of the S1S2 ligand binding core are expanded in the apo state and contract upon ligand binding with the extent of domain separation decreasing in the order of apo > DNQX > kainate > glutamate approximately equal to AMPA. These results suggest that agonist-induced domain closure gates the transmembrane channel and the extent of receptor activation depends upon the degree of domain closure. AMPA and glutamate also promote a 180 degrees flip of a trans peptide bond in the ligand binding site. The crystal packing of the ligand binding cores suggests modes for subunit-subunit contact in the intact receptor and mechanisms by which allosteric effectors modulate receptor activity.     

Action of extracellular divalent cations on native alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors

The effects of divalent cations on Ca2+-impermeable containing (GluR2 subunit) MPA receptors of hippocampal pyramidal neurones isolated from rat brain was studied using patch-clamping. Ca2+, Mg2+, Mn2+, Co2+, Ni2+ and Zn2+ inhibited currents induced by kainate and glutamate. Inhibition was fast, reversible and voltage independent. The rank order of activities was Ni2+ > Zn2+ > Co2+ > Ca2+ > Mn2+ > Mg2+. Cyclothiazide (0.1 mm) significantly reduced inhibition by divalent cations and 6, 7 dinitroquinoxaline-2.3-dione (DNQX). However, high concentrations of Ni2+ and DNQX inhibited AMPA receptors even in the presence of cyclothiazide. The inhibitory effect of divalent cations as well as DNQX was counteracted by an increase in agonist concentration. In the presence of divalent cations the EC50 values of kainate and glutamate were increased, but the maximal response was not changed. An increase in agonist concentration induced a parallel shift in the concentration-inhibition curve for a divalent cation. These data suggest a competitive-like type of inhibition. However, an increase in agonist concentration reduced the inhibitory action of Ni2+ less than that of DNQX. This gave evidence against direct competition between divalent cations and AMPA receptor agonists. A 'complex-competition' hypothesis was proposed to explain the inhibitory action of divalent cations; it is suggested that divalent cations form ion-agonist complexes, which compete with free agonist for agonist-binding sites on AMPA receptors.     

Functional consequences of natural substitutions in the GluR6 kainate receptor subunit ligand-binding site

Differences in binding-site residues of GluR2 (AMPAR) and GluR6 (KAR) subunits have been identified that might account for their functional and pharmacological differences. Specifically, residues A518, A689 and N721 in GluR6 replace highly conserved threonine and serine residues found in other ionotropic glutamate receptor (iGluR) subunits. To define how these natural substitutions impact GluR6 function, we used patch clamp recording with ultrafast perfusion to characterize the effects of A518T, A689S and N721T on agonist potency, efficacy and response kinetics. We find these natural substitutions impact GluR6 function less than would be expected from reverse mutations in other iGluRs. There was little effect of individual or combined mutations on glutamate potency, deactivation or desensitization kinetics. Altered recovery kinetics were seen that were greatest after combined mutations. Kainate potency and response kinetics were also unchanged in the mutants, whereas kainate efficacy was reduced in A518T and increased the T/S/T mutant relative glutamate. Notably, A518T and A689S mutation permitted AMPA to bind as a weak competitive antagonist and the effects of these mutations were additive. N721T mutation further enhanced AMPA binding, allowing AMPA to activate and fully desensitize the receptors. Alternative mutations altering side chain length at position 518 produced far greater changes in glutamate affinity and response kinetics than did the natural mutations. We conclude that these nonconserved residues in GluR6 define the size of the agonist-binding pocket, exerting a steric influence on the bound agonist and the extent of binding-domain closure that can influence agonist potency, deactivation, desensitization and recovery kinetics.     

Functional Consequences of Natural Substitutions in the GluR6 Kainate Receptor Subunit Ligand-Binding Site

Differences in binding-site residues of GluR2 (AMPAR) and GluR6 (KAR) subunits have been identified that might account for their functional and pharmacological differences. Specifically, residues A518, A689 and N721 in GluR6 replace highly conserved threonine and serine residues found in other ionotropic glutamate receptor (iGluR) subunits. To define how these natural substitutions impact GluR6 function, we used patch clamp recording with ultrafast perfusion to characterize the effects of A518T, A689S, and N721T on agonist potency, efficacy and response kinetics. We find these natural substitutions impact GluR6 function less than would be expected from reverse mutations in other iGluRs. There was little effect of individual or combined mutations on glutamate potency, deactivation or desensitization kinetics. Altered recovery kinetics were seen that were greatest after combined mutations. Kainate potency and response kinetics were also unchanged in the mutants, whereas kainate efficacy was reduced in A518T and increased the T/S/T mutant relative glutamate. Notably, A518T and A689S mutation permitted AMPA to bind as a weak competitive antagonist and the effects of these mutations were additive. N721T mutation further enhanced AMPA binding, allowing AMPA to activate and fully desensitize the receptors. Alternative mutations altering side chain length at position 518 produced far greater changes in glutamate affinity and response kinetics than did the natural mutations. We conclude that these non-conserved residues in GluR6 define the size of the agonist-binding pocket, exerting a steric influence on the bound agonist and the extent of binding-domain closure that can influence agonist potency, deactivation, desensitization and recovery kinetics.     

A Conserved Mechanism for Gating in an Ionotropic Glutamate Receptor

Ionotropic glutamate receptor (iGluR) channels control synaptic activity. The crystallographic structure of GluA2, the prototypical iGluR, reveals a clamshell-like ligand-binding domain (LBD) that closes in the presence of glutamate to open a gate on the pore lining α-helix. How LBD closure leads to gate opening remains unclear. Here, we show that bending the pore helix at a highly conserved alanine residue (Ala-621) below the gate is responsible for channel opening. Substituting Ala-621 with the smaller more flexible glycine resulted in a basally active, nondesensitizing channel with ∼39-fold increase in glutamate potency without affecting surface expression or binding. On GluA2(A621G), the partial agonist kainate showed efficacy similar to a full agonist, and competitive antagonists CNQX and DNQX acted as a partial agonists. Met-629 in GluA2 sits above the gate and is critical in transmitting LBD closure to the gate. Substituting Met-629 with the flexible glycine resulted in reduced channel activity and glutamate potency. The pore regions in potassium channels are structurally similar to iGluRs. Whereas potassium channels typically use glycines as a hinge for gating, iGluRs use the less flexible alanine as a hinge at a similar position to maintain low basal activity allowing for ligand-mediated gating.     

Molecular dynamics simulations of the ligand-binding domain of the ionotropic glutamate receptor GluR2.

Ionotropic glutamate receptors are essential for fast synaptic nerve transmission. Recent x-ray structures for the ligand-binding (S1S2) region of the GluR2 alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-sensitive receptor have suggested how differences in protein/ligand interactions may determine whether a ligand will behave as a full agonist. We have used multiple molecular dynamics simulations of 2-5 ns duration to explore the structural dynamics of GluR2 S1S2 in the presence and absence of glutamate and in a complex with kainate. Our studies indicate that not only is the degree of domain closure dependent upon interactions with the ligand, but also that protein/ligand interactions influence the motion of the S2 domain with respect to S1. Differences in domain mobility between the three states (apo-S1S2, glutamate-bound, and kainate-bound) are surprisingly clear-cut. We discuss how these changes in dynamics may provide an explanation relating the mechanism of transmission of the agonist-binding event to channel opening. We also show here how the glutamate may adopt an alternative mode of binding not seen in the x-ray structure, which involves a key threonine (T480) side chain flipping into a new conformation. This new conformation results in an altered pattern of hydrogen bonding at the agonist-binding site.     

Guanine Nucleotides Inhibit cAMP Accumulation Induced by Metabotropic Glutamate Receptor Activation

Metabotropic glutamate receptors (mGluRs) have been shown to modulate adenylate cyclase activity via G-proteins. In the present study we report similar results to the previously observed in the literature, showing that glutamate and the metabotropic agonists, 1S,3R-ACPD or quisqualate induced cAMP accumulation in hippocampal slices of young rats. Moreover, guanine nucleotides GTP, GDP or GMP, inhibited the glutamate-induced cAMP accumulation. By measuring LDH activity in the buffer surrounding the slices, we showed that the integrity of the slices was maintained, indicating that the effect of guanine nucleotides was extracellular. GMP, GDP-S or Gpp(NH)p abolished quisqualate-induced cAMP accumulation. GDP-S or Gpp(NH)p but not GMP inhibited 1S,3R-ACPD-induced cAMP accumulation. The response evoked by glutamate was also abolished by the mGluR antagonists: L-AP3 abolished glutamate-induced cAMP accumulation in a dose-dependent manner and MCPG was effective only at the 2 mM dose. DNQX was ineffective. We are reporting here, an inhibition induced by guanine nucleotides, via an extracellular site (s), similar to the observed with classical glutamate antagonists on a cellular response evoked by mGluR agonists.     

Guanine nucleotides protect against kainate toxicity in an ex vivo chick retinal preparation

Ex vivo preparations of chick neural retina have been successfully used in the assessment of excitotoxicity and in the evaluation of the protective effects of glutamate antagonists. Using a variation of this approach, and measuring the acute and delayed toxic effects of kainate (KA) in terms of lactate dehydrogenase release, we have shown that guanine nucleotides behave as effective neuroprotecting agents. The anti-excitotoxic potency of guanine nucleotides (in the case of GMP and GDPβS it is about 100 times lower than that of DNQX, a powerful kainate antagonist) correlates well with their ability to displace KA from retinal KA receptors.     

Location of the polyamine binding site in the vestibule of the nicotinic acetylcholine receptor ion channel

To map the structure of a ligand-gated ion channel, we used the photolabile polyamine-containing toxin MR44 as photoaffinity label. MR44 binds with high affinity to the nicotinic acetylcholine receptor in its closed channel conformation. The binding stoichiometry was two molecules of MR44 per receptor monomer. Upon UV irradiation of the receptor-ligand complex, (125)I-MR44 was incorporated into the receptor alpha-subunit. From proteolytic mapping studies, we conclude that the site of (125)I-MR44 cross-linking is contained in the sequence alpha His-186 to alpha Leu-199, which is part of the extracellular domain of the receptor. This sequence partially overlaps in its C-terminal region with one of the three loops that form the agonist-binding site. The agonist carbachol and the competitive antagonist alpha-bungarotoxin had only minor influence on the photocross-linking of (125)I-MR44. The site where the hydrophobic head group of (125)I-MR44 binds must therefore be located outside the zone that is sterically influenced by agonist bound at the nicotinic acetylcholine receptor. In binding and photocross-linking experiments, the luminal noncompetitive inhibitors ethidium and triphenylmethylphosphonium were found to compete with (125)I-MR44. We conclude that the polyamine moiety of (125)I-MR44 interacts with the high affinity noncompetitive inhibitor site deep in the channel of the nicotinic acetylcholine receptor, while the aromatic ring of this compound binds in the upper part of the ion channel (i.e. in the vestibule) to a hydrophobic region on the alpha-subunit that is located in close proximity to the agonist binding site. The region of the alpha-subunit labeled by (125)I-MR44 should therefore be accessible from the luminal side of the vestibule.     

A binding site tyrosine shapes desensitization kinetics and agonist potency at GluR2. A mutagenic, kinetic, and crystallographic study

Binding of an agonist to the 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)-propionic acid (AMPA) receptor family of the glutamate receptors (GluRs) results in rapid activation of an ion channel. Continuous application results in a non-desensitizing response for agonists like kainate, whereas most other agonists, such as the endogenous agonist (S)-glutamate, induce desensitization. We demonstrate that a highly conserved tyrosine, forming a wedge between the agonist and the N-terminal part of the bi-lobed ligand-binding site, plays a key role in the receptor kinetics as well as agonist potency and selectivity. The AMPA receptor GluR2, with mutations in Tyr-450, were expressed in Xenopus laevis oocytes and characterized in a two-electrode voltage clamp setup. The mutation GluR2(Y450A) renders the receptor highly kainate selective, and rapid application of kainate to outside-out patches induced strongly desensitizing currents. When Tyr-450 was substituted with the larger tryptophan, the (S)-glutamate desensitization is attenuated with a 10-fold increase in steady-state/peak currents (19% compared with 1.9% at the wild type). Furthermore, the tryptophan mutant was introduced into the GluR2-S1S2J ligand binding core construct and co-crystallized with kainate, and the 2.1-A x-ray structure revealed a slightly more closed ligand binding core as compared with the wild-type complex. Through genetic manipulations combined with structural and electrophysiological analysis, we report that mutations in position 450 invert the potency of two central agonists while concurrently strongly shaping the agonist efficacy and the desensitization kinetics of the AMPA receptor GluR2.     

The structure of a mixed GluR2 ligand-binding core dimer in complex with (S)-glutamate and the antagonist (S)-NS1209

Ionotropic glutamate receptors (iGluRs) mediate fast synaptic transmission between cells of the central nervous system and are involved in various aspects of normal brain function. iGluRs are implicated in several brain disorders, e.g. in the high-frequency discharge of impulses during an epileptic seizure. (RS)-NS1209 functions as a competitive antagonist at 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionate receptors, and shows robust preclinical anticonvulsant and neuroprotective effects. This study explores 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionate receptor binding and selectivity of this novel class of antagonists. We present here the first X-ray structure of a mixed GluR2 ligand-binding core dimer, with the high-affinity antagonist (S)-8-methyl-5-(4-(N,N-dimethylsulfamoyl)phenyl)-6,7,8,9,-tetrahydro-1H-pyrrolo[3,2-h]-isoquinoline-2,3-dione-3-O-(4-hydroxybutyrate-2-yl)oxime [(S)-NS1209] in one protomer and the endogenous ligand (S)-glutamate in the other. (S)-NS1209 stabilises an even more open conformation of the D1 and D2 domains of the ligand-binding core than that of the apo structure due to steric hindrance. This is the first time ligand-induced hyperextension of the binding domains has been observed. (S)-NS1209 adopts a novel binding mode, including hydrogen bonding to Tyr450 and Gly451 of D1. Parts of (S)-NS1209 occupy new areas of the GluR2 ligand-binding cleft, and bind near residues that are not conserved among receptor subtypes. The affinities of (RS)-NS1209 at the GluR2 ligand-binding core as well as at GluR1-6 and mutated GluR1 and GluR3 receptors have been measured. Two distinct binding affinities were observed at the GluR3 and GluR4 receptors. In a functional in vitro assay, no difference in potency was observed between GluR2(Q)(o) and GluR3(o) receptors. The thermodynamics of binding of the antagonists (S)-NS1209, DNQX and (S)-ATPO to the GluR2 ligand-binding core have been determined by displacement isothermal titration calorimetry. The displacement of (S)-glutamate by all antagonists was shown to be driven by enthalpy.     

Interaction of Adenosine and Guanine Derivatives in the Rat Hippocampus: Effects on Cyclic AMP Levels and on the Binding of Adenosine Analogues and GMP

Guanine nucleotides (GN) have been implicated in many intracellular mechanisms. Extracellular actions, probably as glutamate receptor antagonists, have also been recently attributed to these compounds. GN may have a neuroprotective role by inhibiting excitotoxic events evoked by glutamate. Effects of extracellular GN on adenosine-evoked cellular responses have also been reported. However, the exact mechanism of such interaction is not known. In the present study, we showed that GN potentiated adenosine-induced cAMP accumulation in slices of hippocampus from young rats. However, neither GMP nor the metabotropic glutamate receptor agonist, 1S,3R-ACPD, inhibited the binding of the adenosine receptor agonist [³H]NECA (when binding to adenosine A2 receptors), or the binding of the adenosine A2a receptor agonist [³H]CGS 21680 in hippocampal membrane preparations. GppNHp, probably by interacting with G-proteins, decreased [³H]CGS 21680 binding. [³H]GMP binding was assayed in order to evaluate the GN sites which are not G-proteins. [³H]GMP binding was inhibited by GMP and GppNHp, but not by 1S,3R-ACPD. The interaction of endogenous adenosine with the GMP-binding sites was determined by incubating membranes in the presence or absence of adenosine deaminase (ADA). NECA, CADO, CGS 21680 and CPA (only at the highest concentration used) increased GMP binding in the presence of ADA. However, in the absence of ADA, the control levels of GMP binding were as high as in the presence of added ADA plus adenosine agonists, indicating that endogenous adenosine modulates the binding of GMP. If this site has a neuroprotective role, adenosine may be increasing its neuromodulator and proposed protective action.     

Role of the ectodomain serine 275 in shaping the binding pocket of the ATP-gated P2X3 receptor

ATP-activated P2X3 receptors expressed in nociceptive sensory neurons play an important role in pain signaling. Basic properties of this receptor subtype, including very strong desensitization, depend on the rate of dissociation of the agonist from the binding site. Even though the rough structure of the ATP binding site has been proposed on the basis of the X-ray structure of the zebrafish P2X4 receptor and mutagenesis studies, the fine subunit-specific structural properties predisposing the receptor to tight capture of the agonist inside the binding pocket have not been elucidated. In this work, by exploring in silico the functional role for the left flipper located in the ectodomain region, we identified within this loop a candidate residue S275, which could contribute to the closure of the agonist-binding pocket. Testing of the S275 mutants using the patch-clamp technique revealed a crucial role for S275 in agonist binding and receptor desensitization. The S275A mutant showed a reduced rate of onset of desensitization and accelerated resensitization and was weakly inhibited by nanomolar agonist. Extracellular calcium application produced inhibition instead of facilitation of membrane currents. Moreover, some full agonists became only partial agonists when applied to the S275A receptor. These effects were stronger with the more hydrophobic mutants S275C and S275V. Taken together, our data suggest that S275 contributes to the closure of the agonist-binding pocket and that effective capture of the agonist provided by the left flipper in calcium-dependent manner determines the high rate of desensitization, slow recovery, and sensitivity to nanomolar agonist of the P2X3 receptor.     

6,7-Dinitroquinoxaline-2,3-Dione Blocks the Cytotoxicity of N-Methyl-D-Aspartate and Kainate, but Not Quisqualate, in Cortical Cultures

Based on radioligand binding and electrophysiological studies, quinoxalinediones such as 6,7-dinitroquinoxaline-2,3-dione (DNQX) have been shown to be potent competitive antagonists at the quisqualate and kainate subtypes of the glutamate receptor. In this report we have examined the effects of DNQX on excitatory amino acid neurotoxicity and evoked neurotransmitter release. DNQX was found to be a potent neuroprotective agent against glutamate and N-methyl-D-aspartate (NMDA) neurotoxicity. The data suggest that this neuroprotective activity of DNQX is due to its antagonism of the coagonist activity of glycine at the NMDA receptor-channel complex. The specificity of DNQX for the glycine site associated with the NMDA receptor-channel complex was confirmed in radioligand binding and neurotransmitter release studies. DNQX also prevented kainate neurotoxicity and kainate-evoked neurotransmitter release, presumably by direct competition for the kainate receptor. DNQX, however, did not prevent quisqualate neurotoxicity, suggesting that a novel quisqualate-preferring receptor insensitive to DNQX may mediate quisqualate toxicity.     

The Lurcher mutation of an alpha-amino-3-hydroxy-5-methyl- 4-isoxazolepropionic acid receptor subunit enhances potency of glutamate and converts an antagonist to an agonist

A point mutation of the GluRdelta2 (A654T) glutamate receptor subunit converts it into a functional channel, and a spontaneous mutation at this site is thought to be responsible for the neurodegeneration of neurons in the Lurcher mouse. This mutation is located in a hydrophobic region of the M3 domain of this subunit, and this alanine is conserved throughout many of the glutamate receptors. We show here that site-directed mutagenesis of the homologous alanine (A636T; GluR1-L(c)) in the GluR1 AMPA receptor subunit alters its channel properties. The apparent potencies of both kainate and glutamate were increased 85- and 2000-fold, respectively. Furthermore, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX)was converted from a competitive antagonist into a potent agonist. Our results demonstrate that a single amino acid within or near the putative second transmembrane region of the GluR1 subunit is critical for the binding/gating properties of this AMPA receptor.     

Differential activation of individual subunits in heteromeric kainate receptors

Neuronal kainate receptors are assembled from subunits with dissimilar specificities for agonists and antagonists. The composite biophysical behavior of heteromeric kainate receptors is determined by intersubunit interactions whose nature is unclear. Here we use dysiherbaine, a selective kainate receptor agonist, to show that GluR5 subunits assembled in heteromeric GluR5/KA-2 kainate receptor complexes can gate current without concomitant activation of their partner KA-2 subunits. A long-lasting interaction between dysiherbaine and GluR5 subunits elicits a tonic current from GluR5/KA-2 receptors; subsequent cooperative gating of KA-2 subunits can be elicited by both agonists, such as glutamate, and some classically defined antagonists, such as CNQX. This study demonstrates that each type of subunit within a heteromeric kainate receptor contributes a distinct conductance upon activation by agonist binding, and therefore provides insight into the biophysical function of ionotropic glutamate receptors.     

Structural Rearrangements in Loop F of the GABA Receptor Signal Ligand Binding, Not Channel Activation

Structure-function studies of the Cys loop family of ionotropic neurotransmitter receptors (GABA, nACh, 5-HT₃, and glycine receptors) have resulted in a six-loop (A-F) model of the agonist-binding site. Key amino acids have been identified in these loops that associate with, and stabilize, bound ligand. The next step is to identify the structural rearrangements that couple agonist binding to channel opening. Loop F has been proposed to move upon receptor activation, although it is not known whether this movement is along the conformational pathway for channel opening. We test this hypothesis in the GABA receptor using simultaneous electrophysiology and site-directed fluorescence spectroscopy. The latter method reveals structural rearrangements by reporting changes in hydrophobicity around an environmentally sensitive fluorophore attached to defined positions of loop F. Using a series of ligands that span the range from full activation to full antagonism, we show there is no correlation between the rearrangements in loop F and channel opening. Based on these data and agonist docking simulations into a structural model of the GABA binding site, we propose that loop F is not along the pathway for channel opening, but rather is a component of the structural machinery that locks ligand into the agonist-binding site.     

Microscopic kinetics and energetics distinguish GABA(A) receptor agonists from antagonists

Although agonists and competitive antagonists presumably occupy overlapping binding sites on ligand-gated channels, these interactions cannot be identical because agonists cause channel opening whereas antagonists do not. One explanation is that only agonist binding performs enough work on the receptor to cause the conformational changes that lead to gating. This idea is supported by agonist binding rates at GABA(A) and nicotinic acetylcholine receptors that are slower than expected for a diffusion-limited process, suggesting that agonist binding involves an energy-requiring event. This hypothesis predicts that competitive antagonist binding should require less activation energy than agonist binding. To test this idea, we developed a novel deconvolution-based method to compare binding and unbinding kinetics of GABA(A) receptor agonists and antagonists in outside-out patches from rat hippocampal neurons. Agonist and antagonist unbinding rates were steeply correlated with affinity. Unlike the agonists, three of the four antagonists tested had binding rates that were fast, independent of affinity, and could be accounted for by diffusion- and dehydration-limited processes. In contrast, agonist binding involved additional energy-requiring steps, consistent with the idea that channel gating is initiated by agonist-triggered movements within the ligand binding site. Antagonist binding does not appear to produce such movements, and may in fact prevent them.     

Distinct agonist- and antagonist-binding sites on the glycine receptor.

The distinction between receptor-binding sites for agonists and antagonists underpins the pharmacological differences between these two classes of ligands. In the glycine receptor, antagonist (strychnine) binding requires an interaction with residues Lys-200 and Tyr-202. We now demonstrate that the agonist-binding site of this receptor is located at the residue Thr-204. The agonist-binding site interaction is thus likely to be mediated by hydrogen bonding and not by ionic interactions. Our results demonstrate that, in contrast to other studies of ligand-gated ion channel receptors, agonist- and antagonist-binding sites are composed of distinct amino acid residues.     

Tetrazolyl isoxazole amino acids as ionotropic glutamate receptor antagonists: synthesis, modelling and molecular pharmacology

Two 3-(5-tetrazolylmethoxy) analogues, 1a and 1b, of (RS)-2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA), a selective AMPA receptor agonist, and (RS)-2-amino-3-(5-tert-butyl-3-hydroxy-4-isoxazolyl)propionic acid (ATPA), a GluR5-preferring agonist, were synthesized. Compounds 1a and 1b were pharmacologically characterized in receptor binding assays, and electrophysiologically on homomeric AMPA receptors (GluR1-4), homomeric (GluR5 and GluR6) and heteromeric (GluR6/KA2) kainic acid receptors, using two-electrode voltage-clamped Xenopus laevis oocytes expressing these receptors. Both analogues proved to be antagonists at all AMPA receptor subtypes, showing potencies (Kb=38-161 microM) similar to that of the AMPA receptor antagonist (RS)-2-amino-3-[3-(carboxymethoxy)-5-methyl-4-isoxazolyl]propionic acid (AMOA) (Kb=43-76 microM). Furthermore, the AMOA analogue, 1a, blocked two kainic acid receptor subtypes (GluR5 and GluR6/KA2), showing sevenfold preference for GluR6/KA2 (Kb=19 microM). Unlike the iGluR antagonist (S)-2-amino-3-[5-tert-butyl-3-(phosphonomethoxy)-4-isoxazolyl]propionic acid [(S)-ATPO], the corresponding tetrazolyl analogue, 1b, lacks kainic acid receptor effects. On the basis of docking to a crystal structure of the isolated extracellular ligand-binding core of the AMPA receptor subunit GluR2 and a homology model of the kainic acid receptor subunit GluR5, we were able to rationalize the observed structure-activity relationships.