GLUTAMATE RECEPTORS AND THEIR CHARACTERIZATION

The amino acids L-glutamate and L-aspartate are the major excitatory neurotransmitters in the mammalian central nervous system (CNS). Some subtleties aside, glutamate is considered the predominant excitatory neurotransmitter and for simplicity the receptors that bind excitatory amino acids will be referred to as glutamate receptors, or GluRs. Glutamate mediates its effects by interacting with receptors that can be distinguished by pharmacological, physiological, anatomical, molecular, and genetic criteria. The interaction of glutamate with its receptors underlies many normal physiological processes, from rapid synaptic signaling and information transfer to longer-lasting changes in synaptic efficacy that are thought to be the cellular basis of learning and memory. In addition, neurotransmission mediated by glutamate and its receptors is implicated in a variety of CNS pathologies, including epilepsy, cell death due to excitotoxicity and ischemia, and Alzheimer's disease. This entry will review the general characteristics of GluRs, including pharmacological specificity, ion selectivity, and modulation by other compounds.

Glutamate Receptor Classes

Glutamate receptors can be divided into two major classes. One class is termed ionotropic, because glutamate binding causes a conformational change in the receptor that directly opens a gate, permitting ion passage in and out of the cell. The channel and the glutamate-binding site are part of a single multisubunit macromolecule found in the plasma membrane. Three pharmacologically distinct glutamate receptor subclasses of this ionotropic type are distinguished, named after their selective agonists: the N-methyl-D-aspartate (NMDA), 2-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), and kainate (KA) receptors. The AMPA and KA subclasses are frequently referred to as non-NMDA receptors.

A second class of glutamate receptors is characterized by their capacity to couple to second-messenger generating systems via guanine nucleotide-binding protein activation (G-proteins) and are termed the metabotropic glutamate receptors (mGluRs). The binding of glutamate to these mGluRs induces secondary changes that are mediated by intracellular molecules distinct from the receptor itself. One such mGluR is coupled to the phosphoinositide (PI) second messenger system and is named after its selective agonist, trans-1-amino-1,3-cyclopentanedicarboxylic acid (trans-ACPD). A second is named after its selective agonist, l-2-amino-4-phosphonobutanoic acid (LAP4) and exists in the retina, whereas a distinct LAP4-sensitive receptor type has been described in select CNS pathways.

Non-NMDA Receptors

Non-NMDA receptors mediate the rapid on-off type of synaptic signaling that underlies the fast excitatory postsynaptic potential (EPSP) at glutamatergic synapses. Binding of glutamate to these receptors opens a channel that is permeable to Na⁺ and K⁺(and in some instances Ca²⁺), inducing the depolarizing currents characteristic of these receptors. GluRs can desensitize rapidly, limiting the time the channel spends in the open state. This desensitization helps in truncating the duration of the EPSP at excitatory synapses.

The nonendogenous agonists AMPA and quisqualate (QA), and other structurally related compounds, selectively activate the AMPA receptor subclass. A series of quinoxaline derivatives (6-cyano-7-nitroquinoxaline-2,3-dione [CNQX]; 6,7-dinitroquinoxaline-2, 3-dione [DNQX]; and 6-nitro-7-sulfamoyl-benzo (f)quinoxaline-2,3-dione [NBQX]) are the most selective known AMPA/QA antagonists, but they can also act as KA antagonists.

KA has been shown to be both a potent excitant and a potent excitotoxin. The case for a KA receptor that is distinct from AMPA/QA receptors initially stemmed from pharmacological studies in which KA responses were inhibited by select antagonists more potently than were AMPA/QA or NMDA responses, and from anatomical studies in which KA and AMPA/QA receptors displayed distinct anatomical localizations. However, even considering the quinoxaline antagonists, accurate distinction of KA from AMPA/QA receptors remains difficult because of their similar pharmacological profiles. Localization of AMPA/QA and KA receptors by autoradiographic techniques demonstrates a differential distribution of these two types. For example, [³H]AMPA-binding sites are concentrated in the hippocampal CA1 region, outer cortical layers, lateral septum, and the molecular layer of the cerebellum, whereas [³H]KA-binding sites are concentrated in the hippocampal mossy fibers, deep cortical layers, and the granule cell layer of the cerebellum.

Advances in molecular genetic techniques have led to the identification of a series of complementary DNA (cDNA) clones that encode a family of GluRs. Thus, injection of messenger RNA derived from these cDNA sequences into Xenopus oocytes results in electrophysiological responses to glutamate, KA, QA, and AMPA, but not to NMDA or AP-4. The non-NMDA receptor family is composed of at least seven separate genes (GluR1-GluR7) in the rat, and the proteins expressed from these genes each have an approximate molecular weight of 100 kilodaltons. GluR1-GluR4 are the predominant forms expressed in the brain, and the mature form of the receptor appears to be a tetrameric complex of approximately 600 kilodaltons composed of different combinations of the GluR subunits. Two genes (KA-1 and KA-2) have also been identified that encode proteins that represent the high-affinity KA binding site in the brain. KA-1 and KA-2 do not form functional receptors themselves. They must combine with other GluR subunits to produce an ionotropic glutamate receptor. Such structural complexity helps to explain the difficulty in distinguishing AMPA/QA from KA responses using pharmacological approaches. In addition, the capacity to mix different subunits (and their splice variants) provides for a large number of possible GluR phenotypes when analyzed using either pharmacological or electrophysiological techniques.

The GluR family is evolutionarily divergent from other ionotropic receptors (e.g., nicotinic AchR). Ionotropic receptors all have four membrane-spanning domains; however, the GluR family is unique in that the second membrane-spanning segment (TM2) forms a kink in the membrane, exiting back into the cytoplasm instead of traversing the bilayer and exiting into the extracellular space. This kink in TM2 is analogous to a pore (P) forming element in K⁺ channels and forms the structure of the GluR responsible for determining ion permeation. GluRs are permeant to Na⁺ and K⁺ although certain GluR variants can also be significantly permeable to Ca⁺⁺.

NMDA Receptors

The NMDA receptor is pharmacologically and functionally more complex than AMPA/QA and KA receptors. For example, binding of glutamate to the NMDA receptor will open the ion channel only if the membrane is depolarized and if glycine is present. Because NMDA receptors open as a function of the extent of membrane depolarization, they are described as voltage-dependent and are thus sensitive to postsynaptic activity. Therefore, the NMDA receptor complex provides an example of a conditional logic gate where Hebb-like conditions are realized at a single synapse. Furthermore, the ion channel is significantly permeable to Ca⁺⁺ in addition to Na⁺ and K⁺, resulting in a significant influx of Ca⁺⁺ that induces a variety of secondary Ca⁺⁺-activated processes. These processes include 1. the induction of long-term potentiation (LTP), which is regarded as a cellular model of memory analogous to Hebb-type synaptic plasticity; 2. learning and memory in animal models;3. the pathophysiology of epilepsy; and 4. some forms of excitotoxicity.

The NMDA receptor appears to be regulated by a variety of endogenous and exogenous compounds that act at distinct binding sites to modify the function of the receptor. The first of these binding sites is the transmitter recognition site, which binds L-glutamate (or L-aspartate), the synthetic ligand NMDA, and other agonists. Glutamate-binding can be selectively and competitively antagonized by a series of compounds, including D-2-amino-5-phosphonopentanoate (D-AP5) and 3-(2-carboxypiperazin-4-yl)propyl-1-phosphate (CPP). Second is the glycine-binding site, which must be occupied in order for the channel to be opened by glutamate binding to its recognition site. The pharmacology of the NMDA glycine site is distinct from that of the inhibitory glycine receptor in the spinal cord and brain stem, in that the NMDA receptor glycine site is insensitive to strychnine but is antagonized by kynurenate, 7-chlorokynurenate, and other compounds. Although this site may always be saturated in vivo, glycine is often referred to as a cotransmitter. Third is a site within the ion channel that binds Mg⁺⁺. Binding of Mg⁺⁺ at this site blocks current flow through the channel when the membrane is at hyperpolarized potentials. This block is removed under depolarizing conditions, providing the molecular basis for the NMDA receptors' voltage dependence. Fourth is a site within the ion channel that binds phencyclidine (PCP), dibenzocyclohepteneimine (MK-801), and other compounds, causing blockage of ion flow through the channel. Binding of these compounds is permitted only when the ion channel is in the open state and the compounds are, therefore, referred to as open channel blockers. Finally, a variety of other molecules have been shown to be modulators of the NMDA receptor. For example, spermine and spermidine increase MK-801 binding to the NMDA complex and NMDA-evoked currents in cell culture and Xenopus oocyte preparations. Additionally, Zn⁺⁺ and H⁺ have been shown to modulate NMDA receptor currents in a number of preparations.

Localization of NMDA receptors by autoradiographic techniques demonstrates an anatomical distribution that in general parallels the distribution of AMPA/QA receptors, further supporting the idea that these two receptors work in concert in many synapses. Thus, NMDA receptors are found in the CA1 region of the hippocampus, throughout the cortex (particularly in outer cortical layers), and in striatum. However, specific localization of binding shows a different distribution of NMDA agonist-and antagonist-preferring binding sites, suggesting that subpopulations of NMDA receptors exist in different brain areas.

The NMDA receptor has been successfully isolated and characterized, and several distinct subunits of this receptor complex have been cloned. NMDAR1 was the original isolate, and various NMDAR2A-2D subunits coassemble with NMDAR1 to form functional receptors. Anatomical gene-mapping studies have confirmed that subpopulations of NMDA receptors exist in different brain regions. As with the non-NMDA receptors, numerous pharmacological and electrophysiological phenotypes are possible through the mixing of different subunits. The TM2 segment of the NMDARs is responsible for determining the ion permeation properties of the channel and forms a kink in the membrane analogous to that described for the GluRs above.

G-Protein-Coupled Glutamate Receptors

The G-protein coupled receptors (GPCRs) that bind glutamate and other excitatory amino acids are referred to as the metabotropic glutamate receptors, or mGluRs. They are similar in general structure to other GPCRs in having seven transmembrane spanning domains; however, they are divergent enough to be considered to have originated from a separate evolutionary-derived receptor family. In fact, sequence homology between the mGluR family and other GPCRs is minimal except for the GABAB receptor. The mGluR family is heterogeneous in size, ranging from 854 to 1,179 amino acids. The ligand-binding site for mGluRs resides in the large N-terminal extra-cellular domain. Additionally, the mGluRs exist as functional dimers in the membrane in contrast to the single subunit form of most GPCRs. These significant structural distinctions support the idea that the mGluRs evolved separately from the other GPCRs. The third intracellular loop, thought to be the major determinant responsible for G-protein coupling of the mGluRs, is relatively small, whereas the C-terminal domain is quite large. The coupling between mGluRs and their respective G proteins may occur through unique determinants that exist in the large C-terminal domain.

Eight different mGluRs can be subdivided into subgroups on the basis of sequence homologies and their capacity to couple to specific enzyme systems. For example, both mGluR1 and mGluR5 activate a G protein coupled to phospholipase C. mGluR1 activation can also lead to the production of cAMP and of arachidonic acid by coupling to G proteins that activate adenylate cyclase and phospholipase A2, respectively. The subtype mGluR5 seems more specific, activating predominantly the G-protein-activated phospholipase C.

Distinctions between the mGluR subtypes can also be made pharmacologically. One group, mGluR2, 3, and 8, favors the agonist trans-ACPD for activation, whereas a second group, including mGluR4, 6, and 7, favors AP-4 for activation. The sub-types mGluR2 and mGluR4 can be further distinguished pharmacologically by using the agonist 2-(carboxycyclopropyl)glycine, which is more potent at activating mGluR2 receptors.

The mGluRs are widespread in the nervous system and are found both pre- and postsynaptically. Presynaptically, they serve as autoreceptors and appear to participate in the inhibition of neurotransmitter release. Their postsynaptic roles appear to be varied and depend on the specific G protein to which they are coupled. Activation of mGluR1 has been implicated in long-term synaptic plasticity at many sites in the brain, including long-term potentiation in the hippocampus.

Conclusion

More than one receptor subclass may be present in any given excitatory synapse. The function of glutamatergic synapses depends on the combination of the above-mentioned receptor subtypes at a given synapse (e.g., non-NMDA, NMDA, both non-NMDA and NMDA, and mGluR, among others). Working in concert these receptors yield a complex synaptic response that depends on the number and location of individual receptors. Because these subtypes clearly serve distinct functions, the capacity to manipulate individual receptor molecules is critical to understanding the role of a subtype in a normal physiological event or in the etiology of a given glutamate-linked disease. The advent of more potent and selective drugs for each receptor subclass will allow for more precise experimental and clinical manipulation of these receptors. Similarly, molecular genetic approaches to analyzing both the structure and the function of these receptors has revealed a wealth of information, and further analysis of their genes and gene products will have far-reaching implications for the basic science of glutamate receptors and for a variety of glutamate-linked human diseases.

See also:GUIDE TO THE ANATOMY OF THE BRAIN: SYNAPSE; HEBB, DONALD; LONG-TERM POTENTIATION; NEUROTRANSMITTER SYSTEMS AND MEMORY; SECOND MESSENGER SYSTEMS

Bibliography

Boulter, J., Hollmann, M., O'Shea-Greenfield, A., Hartley, M., Deneris, E., Maron, C., and Heinemann, S. (1990). Molecular cloning and functional expression of glutamate receptor subunit genes. Science 249, 1,033-1,037.

Cotman, C. W., Monaghan, D. T., Ottersen, O. P., and Storm- Mathisen, J. (1987). Anatomical organization of excitatory amino acid receptors and their pathways. Trends in Neuro- sciences 10, 273-280.

Hollmann, M., and Heinemann, S. (1994). Cloned glutamate receptors. Annual Review of Neuroscience 17, 31-108.

Keinanen, K., Wisden, W., Sommer, B., Werner, P., Herb, A., Verdoorn, T. A., Sakmann, B., and Seeburg, P. H. (1990). A family of AMPA-selective glutamate receptors. Science 249, 556-560.

Masu, M., Tanabe, Y., Tsuchida, K., Shigemoto, R., and Nakanishi, S. (1991). Sequence and expression of a metabotropic glutamate receptor. Nature 349, 760-765.

Nakanishi, S. (1994). Metabotropic glutamate receptors: Synaptic transmission, modulation, and plasticity. Neuron 13, 1,031-1,037.

Nakanishi, N., Shneider, N. A., and Axel, R. (1990). A family of glutamate receptor genes: Evidence for the formation of heteromultimeric receptors with distinct channel properties. Neuron 5, 569-581.

Rosenmund, C., Stern-Bach, Y., and Stevens, C. F. (1998). The tetrameric structure of a glutamate receptor channel. Science 280, 1,596-1,599.

Sommer, B., Kohler, M., Sprengel, R., and Seeburg, P. H. (1991). RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell 67, 11-19.

Watkins, J. C., Krogsgaard-Larsen, P., and Honore, T. (1990). Structure activity relationships in the development of excitatory amino acid receptor agonists and competitive antagonists. Trends in Pharmacological Sciences 11, 25-33.

Waxham, M. N. (1998). Ionotropic and metabotropic receptors. In F. E. Bloom, S. L. Landis, J. L. Roberts, L. R. Squire, and M. J. Zigmond, eds., Fundamental neuroscience, pp. 235-267. New York: Academic Press.

C. W.Cotman

E. R.Whittemore

Revised byM. N.Waxham