Gp I II III
NM DA receptor
Glutamate APS (-) NMDA MK80I (-) Aspartate Ketarrane ('-)
AM PA receptor
Glutamate AM PA MBQX (-)
Gp I II III
Figure 10.2 Site of action of drugs affecting glutamate synapses
receptors for glutamate into four main types. Block of a physiological response by an antagonist is good evidence for a functional role of any transmitter in CNS events and this has now been achieved for glutamate in many areas of the CNS.
The nomenclature for the glutamate receptors is confusing. Originally, the receptors were called N-methyl-D-aspartate (NMDA) and non-NMDA with the latter later subdividing into quisqualate and kainate. Now, the accepted classification is into AMPA, kainate, NMDA and metabotropic. This latter class of receptor is further divided into three groups (I, II and III) containing at least two subtypes. Figure 10.1 shows the agonists at the receptors.
NON-NMDA —AMPA AND KAINATE RECEPTORS
Non-NMDA ionotropic glutamate receptors (the majority sodium channel containing) can be subdivided into a-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) (comprising cloned subunits GluR1-4) and kainate (GluR5-7, KA1-2) preferring receptors, with native receptors most likely to comprise either homo- or heteromeric pentamers of these subunits.
mRNA coding for GluR1-4 subunits is found throughout the brain and spinal cord, with differing patterns of expression of GluR2 and GluR1 in different regions. There is also evidence for both presynaptic AMPA and particularly kainate-preferring receptors comprising GluR5 subunits on neuronal terminals in various areas of the CNS.
There are competitive AMPA receptor antagonists (Fig. 10.3) of which NBQX (6-nitro-7-sulphamoylbenzo(f)quinoxaline-2,3-dione) (which also displays micromolar affinity for the kainate-preferring GluR5 and GluR6 subunits3) is the most selective, and the recently developed selective GluR5 antagonist LY382884 [3S,4aR,6S,8aR-6-(4-carboxyphenyl)-methyl-1,2,3,4,4a,5,6,7,8,8a-deca-hydroisoquinoline-3-carboxylic acid]. These drugs are allowing the roles played by non-NMDA ionotropic glutamate receptors to be gauged.
The majority of AMPA receptors are impermeable to Ca2+, although some AMPA receptors, as well as kainate receptors, have significant Ca2+ permeability. AMPA receptors are multimeric assemblies of four cloned subunits, GluR1-4, but it is the absence of the GluR2 subunit that determines the Ca2+ permeability of AMPA receptors, since editing out of this subunit following transcription into mRNA results in the introduction of a positive charge in the pore-forming region (Q/R site), which is not present in GluR1,3 or 4, 9 (see also AMPA receptors in Chapter 3). AMPA receptors lacking GluR2 have Ca2+ permeability ratios up to PCa/PNa = 3. Since calcium is such a ubiquitous intracellular messenger any receptor that allows this ion to enter neurons is likely to be important in plasticity in the CNS. In a similar manner to the AMPA receptor, RNA editing in the pore region at the Q/R site controls the Ca2+ permeability of the kainate receptor subunits GluR5 and GluR6, with significant levels of the unedited (Ca2+ permeable) version of these receptors present in the adult CNS.
The AMPA receptor subunits are all found within many regions of the CNS but in differing numbers, and, in the spinal cord, have differing lamina distributions. GluR1 and GluR2 are generally the most abundant AMPA receptor subunits with lower levels of GluR3 and GluR4. The majority of AMPA receptors allow Na+ to enter neurons and thus in most areas of the CNS studied, the initial stage in excitatory synaptic transmission is a fast-depolarising response due to the release of glutamate and subsequent activation of the AMPA receptor.
Figure 10.3 Structure of the NMDA receptor-channel complex. The receptor has a complicated structure and this is highlighted by the presence of many pharmacologically distinct binding sites through which the receptor activity can be modulated. The channel associated with the receptor is blocked by Mg2+ at resting potential ( —70mY). Receptor activation requires the removal of this Mg2+ block (voltage-gated) as well as the binding of glutamate and the co-agonist, glycine (ligand-gated). The different binding sites (glutamate, phencyclidine, polyamine, glycine) are illustrated, and together with antagonists which act at the various sites (in parentheses). The polyamine site is an intracellular site which modulates the affinity of other agonists and antagonists
Figure 10.3 Structure of the NMDA receptor-channel complex. The receptor has a complicated structure and this is highlighted by the presence of many pharmacologically distinct binding sites through which the receptor activity can be modulated. The channel associated with the receptor is blocked by Mg2+ at resting potential ( —70mY). Receptor activation requires the removal of this Mg2+ block (voltage-gated) as well as the binding of glutamate and the co-agonist, glycine (ligand-gated). The different binding sites (glutamate, phencyclidine, polyamine, glycine) are illustrated, and together with antagonists which act at the various sites (in parentheses). The polyamine site is an intracellular site which modulates the affinity of other agonists and antagonists mRNA coding for the kainate receptor subunits GluR5 and GluR7 is also found in isolated neurons in the CNS although many kainate GluR5 receptors are thought to be located presynaptically on terminals of neurons that release glutamate. Kainate receptors are therefore thought to be excitatory autoreceptors that enhance the release of glutamate. It could be predicted that the widespread distribution of AMPA receptors precludes the use of antagonists at this receptor in therapy since adverse effects are highly likely. By contrast, the kainate receptor might be an interesting target since its functional role will be linked to the level of glutamate release. Thus, antagonists at this receptor should reduce excessive glutamate release while having less effect on more normal functional synapses.
The role of the kainate receptor system in the brain is at an early stage since there are as yet few pharmacological tools to study its function. However, mutations in the kainate receptor genes have been made in mice and there is a GluR6 kainate receptor knock-out mouse. Kainate binding is absent in areas of the brain which normally have high levels such as the hippocampus. Here, in normal animals kainate receptors mediate a postsynaptic current which is absent in the GluR6 knock-out mouse. The mice have reduced motor activity but can learn maze tasks. The knock-out mouse is resistant to kainate-induced seizures.
Studies have shown that neurons expressing high levels of GluRl mRNA but lacking GluR2 are found in the superficial laminae of the spinal cord, an area where nociceptive primary afferents terminate, suggesting that a subpopulation of AMPA receptors in this region may have significant Ca2+ permeability. Calcium-permeable non-NMDA receptors have been demonstrated in spinal cord slices using kainate-induced cobalt loading. Studies performed using cultured neurons in vitro have suggested that Ca2+ entry through Ca2+-permeable AMPA receptors in the spinal cord may provide a mechanism for the strengthening of transmission at synapses and enhancement of nociceptive transmission. Other studies have suggested a link between Ca2+-permeable AMPA receptors and inhibitory systems since in the dorsal horn of the spinal cord many of these receptors are found on GABA neurons. Clearly, the functional role of Ca2+-permeable non-NMDA receptors in vivo will depend on their location in the integrated circuitry of the CNS. Joro Spider Toxin (JSTx) has been reported to be a selective blocker of Ca2+-permeable non-NMDA responses evoked by AMPA/kainate rather than those evoked by NMDA and so will be a useful tool for studying the roles of these receptors.
Much attention has been focused on the role of the N-methyl-D-aspartate (NMDA) receptor for glutamate, activation of which produces slow prolonged neuronal depolarisation. Thus unlike the AMPA receptor, it is not responsible for the fast transmission of excitation nor the initiation of impulses but has been shown to be critical for maintaining excitatory responses such as the manifestation of wind-up in spinal cord, long-term potentiation in the hippocampus, epileptiform activity and in neuroexcitotoxicity. Mechanisms of central amplification of a nociceptive input have been suggested to underlie aspects of the enhanced spinal transmission of nociceptive messages in protracted pain states, and in this case there is good clinical evidence to support the concepts that have arisen from animal studies.
The NMDA receptor has a heteromeric structure composed of two subunit types; NR1 and NR2, the latter having four subunits (NR2ANR2D) (Fig. 10.2). Molecular genetic techniques have demonstrated that native NMDA receptors are likely to be composed of a combination of the NR1 subunit (which can exist in eight different splice variants) and one or more of the four NR2 subunits which are the main determinants of functional diversity among the NMDA receptors (see Chapter 3 for further details). It has been shown that there are distinct developmental and spatial expression patterns of NMDA receptor NR1 subunit splice variants and NR2 receptor subunits in the CNS.
Although the exact subunit stoichiometry is not yet known for any NMDA receptor, heterologous expression studies suggest that they are likely to be tetramers composed of two NR1 subunits and two NR2 subunits providing the possibility for considerable structural diversity of NMDA receptors. The subtypes have been partially mapped in the CNS and show differing regional distributions. As the subunit composition imparts different physiological characteristics to the receptor, this would imply that different functional roles of the NMDA receptor could be separated — at present there are only antagonists for the NR2B subtypes. The NMDA receptor is a non-specific cation channel in that both sodium and calcium enter, but the latter ion appears to be the predominant factor in the alteration of neuronal activity. This is not simply due to the large amounts of calcium that enter neurons and thus the degree of excitability that ensues but also simply that many intracellular pathways are calcium dependent.
The NMDA receptor is a complex entity. Functional modulation of the receptor can be achieved through actions at various recognition sites including the primary transmitter site, the site where glutamate binds (competitive), the phencyclidine (PCP) site (uncompetitive) situated in the channel of the receptor, the polyamine modulatory site and the strychnine-insensitive glycine site where glycine is a required co-agonist with glutamate (Fig. 10.3). Potentially, there are several ways in which the effect of released glutamate can be antagonised through NMDA receptor blockade. Numerous studies have investigated the potential use of antagonists acting through the different recognition sites. However, due to the ubiquitous nature of the receptor, it has often been difficult to achieve therapeutic effects at the target organ, in the absence of adverse side-effects. Prototypical antagonists for the various sites are shown in Fig. 10.3—namely AP5 for the receptor, MK-801 for the channel (although the clinically used drugs ketamine and memantine also act at this site) and then there are other agents such as 7-chlorokynurante for the associated glycine site.
Alterations in the transmission of neuronal information via NMDA receptors arise due to two main factors, the first being that the calcium influx through the channel produces large depolarisations, and the second due to the unique profile of the receptor-channel complex, which requires various conditions for operation, and therefore is not necessarily involved in synaptic transmission at all times and under all circumstances. The release of the excitatory amino acids is obviously needed but in addition, glycine is required as a co-agonist, and this is of pharmacological and therapeutic interest, as antagonists of this site can produce inhibitions of NMDA-mediated events. The latter condition would appear to be ever present, due to the levels of glycine available in the brain and spinal cord. However, for the action of glycine on NMDA receptors to take on a clear physiological role, the concentration of glycine present at the synapse must normally be kept below saturating levels. Although it is not exactly clear if this occurs, it is thought that a glycine transporter is involved, whose distribution closely matches that of NMDA receptors in the CNS, so the glycine concentration present at glutamate synapses may be regulated by glycine uptake. The key role of glycine in activation of the receptor is borne out by the ability of antagonists at this site to produce inhibitions of NMDA-mediated transmission. Finally, an induced depolarisation of the neuron to relieve the resting voltage-dependent magnesium block of the channel is a prerequisite for activation of the complex. For these reasons, the NMDA receptor-channel complex is not a participant in 'normal' synaptic transmission, but when the correct conditions are achieved the complex will rapidly become activated and add a powerful depolarising or excitatory drive to synaptic transmission.
The NMDA receptor is an ionotropic receptor coupled to a cation channel, which is blocked by physiological levels of Mg2+ at the resting membrane potential — the sensitivity to magnesium block depends on the subunit composition as does the glycine sensitivity. The channel is blocked in a voltage-dependent manner so the receptor can only operate after sufficient repeated depolarisation. In the spinal cord, the removal of the Mg2+ block is mediated by peptides, including tachykinins, which are co-released with glutamate. After a brief acute stimulus, pain transmission from C-fibres is largely mediated by the action of glutamate on AMPA/kainate receptors. When the stimulus is sustained or its intensity is increased, however, the action of substance P on NK-1 receptors produces sufficient membrane depolarisation so that the Mg2+ block can now be removed and the NMDA receptor activated. These events underlie central hyperexcitability and result in a significant amplification of the response. Substance P therefore plays an important role in this instance in recruiting NMDA receptors and contributes to the cascade of events leading to the enhancement and prolongation of the neuronal response. In other CNS areas, the NMDA receptor may be allowed to participate in synaptic events by glutamate acting on AMPA receptors. How AMPA/ kainate receptors provide an excitatory drive of sufficient length to remove the block of the NMDA receptor channel while being fast ionotropic receptors is unclear. However, different subtypes of the NMDA receptor have differing sensitivity to both glycine and magnesium and the particular channel openings vary in both amplitude and duration. Thus regional specific conditions may control the receptor and determine its properties.
The NMDA receptor is therefore unique in that it is not simply ligand-gated but also voltage-gated due to the channel block imparted by magnesium. No other receptor requires two ligands (e.g. glutamate and glycine) for receptor activation.
This fourth type of receptor for glutamate (mGluRs), so named as they are members of the seven transmembrane-spanning family, is the least well understood. The poor understanding of this class of receptor stems from the fact that there are eight receptors in the class which fall into three groups, divided by sequence homology, effector mechanisms and, to some extent, their pharmacology. We are presently lacking sufficiently potent and selective antagonists at all these metabotropic glutamate receptors to probe their roles. They are coupled through G-proteins to potassium and calcium channels and while Group I (mGluR 1 and 5) receptors interact with IP3 systems, both Group II and III inhibit adenyl cyclase. Thus broadly, the Group I receptors are therefore excitatory and Groups II and III are inhibitory. There is some evidence for both pre- and postsynaptic locations of all groups of receptors. Functionally, the mGluRs have been implicated in memory, pain, anxiety and neurodegeneration with few specific details due to the lack of antagonists.
Was this article helpful?