No job name

British Journal of Pharmacology (2009), 157, 1301–1317
2009 The Authors
Journal compilation 2009 The British Pharmacological Society All rights reserved 0007-1188/09

Allosteric modulators of NR2B-containing NMDA
receptors: molecular mechanisms and
therapeutic potential
Laetitia Mony1, James NC Kew2, Martin J Gunthorpe2 and Pierre Paoletti1 1Laboratoire de Neurobiologie, Ecole Normale Supérieure, CNRS, Paris, France, and 2Neurosciences CEDD, GlaxoSmithKline,Harlow, Essex, UK N-methyl-D-aspartate receptors (NMDARs) are ion channels gated by glutamate, the major excitatory neurotransmitter in the
mammalian central nervous system (CNS). They are widespread in the CNS and are involved in numerous physiological and
pathological processes including synaptic plasticity, chronic pain and psychosis. Aberrant NMDAR activity also plays an
important role in the neuronal loss associated with ischaemic insults and major degenerative disorders including Parkinson's
and Alzheimer's disease. Agents that target and alter NMDAR function may, thus, have therapeutic benefit. Interestingly,
NMDARs are endowed with multiple extracellular regulatory sites that recognize ions or small molecule ligands, some of which
are likely to regulate receptor function in vivo. These allosteric sites, which differ from agonist-binding and channel-permeation
sites, provide means to modulate, either positively or negatively, NMDAR activity. The present review focuses on allosteric
modulation of NMDARs containing the NR2B subunit. Indeed, the NR2B subunit confers a particularly rich pharmacology with
distinct recognition sites for exogenous and endogenous allosteric ligands. Moreover, NR2B-containing receptors, compared
with other NMDAR subtypes, appear to contribute preferentially to pathological processes linked to overexcitation of
glutamatergic pathways. The actions of extracellular H+, Mg2+, Zn2+, of polyamines and neurosteroids, and of the synthetic
compounds ifenprodil and derivatives (‘prodils') are presented. Particular emphasis is put upon the structural determinants and
molecular mechanisms that underlie the effects exerted by these agents. A better understanding of how NR2B-containing
NMDARs (and NMDARs in general) operate and how they can be modulated should help define new strategies to counteract
the deleterious effects of dysregulated NMDAR activity.
British Journal of Pharmacology (2009) 157, 1301–1317; doi:10.1111/j.1476-5381.2009.00304.x; published online 8
July 2009
Keywords: NMDA; glutamate receptors; NR2B; allosteric modulators; excitotoxicity; neuropathic pain
3a5bS, 3a-hydroxy-5b-pregnan-20-one sulphate; NMDAR, N-methyl-D-aspartate receptors; NTD, N-terminal domain; PS, pregnenolone sulphate; QBP, glutamine-binding protein Molecular architecture of NMDA receptors
of NR1 and NR2 subunits forming a tetrameric complex oftwo NR1 and two NR2 subunits. Similarly to other glutamate- N-methyl-D-aspartate receptors (NMDARs) are multisubunit complexes associating NR1, NR2 and, more rarely, NR3 sub- isoxazole-propionate (AMPA) and kainate receptors), NMDARs units (also named GluN1, GluN2 and GluN3, respectively; are thought to arrange and operate as dimer of dimers Alexander et al. 2008). NR2 and NR3 subunits exist as four and (Furukawa et al., 2005). The NR2 subunits that have differen- two subtypes, respectively (NR2A-D and NR3A-B), each tial developmental and anatomical profile are the major deter- subtype being encoded by a distinct gene. NR1 exists as seven minants of functional diversity by conferring on NMDARs' subtypes (NR1a–g), which are generated by alternative splicing distinct biophysical and pharmacological properties (Cull- from a single gene (Dingledine et al., 1999). Most NMDARs Candy and Leszkiewicz, 2004; Paoletti and Neyton, 2007).
Structurally, NMDAR subunits, like AMPA and kainate receptor subunits, share a common design and are organizedinto four different modules or units (Mayer, 2006; Paoletti Correspondence: Dr Pierre Paoletti, Laboratoire de Neurobiologie, CNRS UMR and Neyton, 2007; Figure 1A): two large domains in the extra- 8544, Ecole Normale Supérieure, 46 rue d'Ulm, 75005 Paris, France. E-mail: cellular region, the N-terminal domain (NTD) and the Received 16 February 2009; revised 24 March 2009; accepted 26 March 2009 agonist-binding domain (ABD, binding glycine in NR1 and

Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al Organization and expression of NMDAR subunits. (A) Schematic representation of the NR2B subunit. It is composed of an NTD that binds allosteric inhibitors such as zinc and ifenprodil, an S1–S2 ABD (binds glutamate), a TM region (TM 1, 2, 3 and a re-entrant loop) thatforms the ion channel and a C-terminal cytoplasmic region. (B) Distribution of NMDAR subunit mRNAs in the mouse brain at post-natal day21 (figure reproduced with permission from Watanabe et al., 1993). NR2A (panel A), NR2B (panel B), NR2C (panel C), NR2D (panel D) andNR1 (panel E). Note the restricted expression of the NR2B subunit in the forebrain. ABD, agonist-binding domain; AC, anterior cingulate cortex;Cx, cerebral cortex; Cb, cerebellum; CPu, caudate-putamen; Hi, hippocampal formation; MB, midbrain; NMDAR, N-methyl-D-aspartate; NTD,N-terminal domain; OB, olfactory bulb; S, septum; Th, thalamus; TM, transmembrane.
glutamate in NR2); the membrane region, comprising three are ‘attached' one to the other through back-to-back apposi- transmembrane segments and a re-entrant loop that forms tion of their lobe 1, ABD closure induced by agonist binding the ion channel and resembles an inverted K+ channel; and can be transduced into tension onto the linkers connecting the cytoplasmic region, comprising a C-terminal tail involved the ABDs to the channel. This in turn leads to channel in cellular trafficking of the receptor and coupling to various opening (Mayer, 2006; see Figure 4B later). Breaking of the intracellular signalling pathways.
ABD dimer interface provides a means to relax this tension, The two extracellular domains, the NTD and the ABD, share and hence, close the channel. This process occurs during homology with bacterial periplasmic proteins. The NTD, com- AMPA and kainate receptor desensitization (Sun et al., 2002; posed of the first 380 amino acids, is related to the leucine Mayer, 2006) and during allosteric inhibition of NMDARs by isoleucine valine binding protein (LIVBP; O'Hara et al., 1993; protons and NTD ligands (Gielen et al., 2008; see also, below Paoletti et al., 2000) and plays an important role in subunit and Figure 2B).
assembly (Herin and Aizenman, 2004). In NR2A and NR2Bsubunits, it also forms a regulatory domain, binding allostericinhibitors (see below). Its structure has not been determinedyet, but by homology to LIVBP, it is thought to exhibit a The NR2B subunit: expression and subcellular
clamshell-like structure (Masuko et al., 1999; Paoletti et al., 2000). The ABD, which is related to the glutamine-bindingprotein, is also a bilobate domain. It is split in two discon- While the NR1 subunit is expressed in virtually all neurons tinuous segments, S1 and S2, by insertion of the ion channel and at all developmental stages in the brain, NR2 subunit (Figure 1A). Several X-ray crystal structures of ABD of different genes display different regional and developmental expres- ionotropic glutamate receptor (iGluR) subunits are now avail- sion patterns. Thus, in the embryonic brain, NR2B and NR2D able, including those of NR1 and NR2A subunits complexed subunits predominate, while NR2A and NR2C are absent; in with glycine and glutamate respectively (Furukawa et al., contrast, in the adult brain, NR2A predominates, being ubiq- 2005). In all these structures, the agonist binds in the cleft of uitously expressed, while NR2B expression is restricted to the clamshell, stabilizing a closed conformation of the two forebrain areas, and NR2C is highly enriched in the cerebel- lobes. Competitive antagonists bind the same cleft but, in lum (Watanabe et al., 1992; Akazawa et al., 1994; Monyer contrast to agonists, impede cleft closure, thus, preventing et al., 1994, Figure 1B). Closer inspection of NR2B subunit channel activation (Furukawa and Gouaux, 2003).
gene expression in the rodent brain indicates that at E17, high The mechanism that couples agonist binding to channel levels of NR2B are present in the cortex (especially layer I), gate opening is partly known. ABD dimerization plays a key thalamus and spinal cord. In this latter region, NR2D is also role in this process. Indeed, because two neighbouring ABDs strongly expressed. Around birth, NR2B expression is wide- British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
A NR2B negative allosteric modulators B NR2B positive allosteric modulators Pregnenolone sulfate (4) Positive and negative allosteric modulators of NR2B-containing NMDARs. (A) Allosteric modulators that inhibit NR2B-containing receptors. (B) Allosteric modulators that potentiate NR2B-containing receptors. The agents that selectively modulate NMDARs incorporatingthe NR2B subunit are named in red. NMDAR, N-methyl-D-aspartate receptors; 3a5bS, 3a-hydroxy-5b-pregnan-20-one sulphate.
spread with strong expression levels in the cortex, hippocam- Armentia and Sah, 2003). The fact that at many adult syn- pus, septum, striatum and thalamic nuclei. NR2B is also apses, NR2B subunits might incorporate preferentially into present in the developing cerebellum, though at lower levels.
triheteromeric NR1/NR2A/NR2B receptors (Luo et al., 1997), a NR2B expression peaks around P7–P10, at a stage where NR2A receptor subtype shown only recently to have a low sensitiv- expression rises sharply while NR2D drops. The most striking ity to NR2B-selective antagonists (Hatton and Paoletti, 2005), change in the pattern of NR2B gene expression takes place is likely to explain why certain groups have proposed that between the first and second week post-natal and results in its NR2B is excluded from synaptic sites. Finally, there is accu- almost complete disappearance from the cerebellum (while mulating evidence that NMDARs are also present at presyn- NR2C, and to a lesser extent, NR2A increase) together with a aptic sites where they can influence transmitter release.
confinement to forebrain structures. In the cerebellum, this In particular, NR2B-containing autoreceptors have been switch in NR2 subunit expression from NR2B to NR2A and described on primary sensory afferents of the spinal chord NR2C has been shown to occur in the same cell population, (Ma and Hargreaves, 2000), and at synapses on entorhinal the granule cells (Farrant et al., 1994). In the adult brain, cortical neurons (Woodhall et al., 2001). Similarly to postsyn- expression of NR2B is the highest in the cortex (most particu- aptic NMDARs, presynaptic NMDARs may also participate in larly layer II/III), hippocampus, amygdala, ventral nuclei of certain forms of synaptic plasticity (Pinheiro and Mulle, the thalamus and olfactory bulb (Watanabe et al., 1993).
Interestingly, in the adult spinal chord, NR2B expression isrestricted to lamina 2 of the dorsal horn, a region that receivesprimary sensory afferents from nociceptors and thermorecep- Allosteric inhibition by protons
tors (Watanabe et al., 1994). The restricted localization ofNR2B-containing receptors in this region could explain, in Extracellular protons (H+) are potent inhibitors of NMDARs.
part, why NR2B-selective antagonists have analgesic effects They inhibit NMDARs in a non-competitive and voltage- (Chizh et al., 2001 and see below).
independent manner, indicating that they are not acting as At the subcellular level, NMDARs, including those contain- channel blockers (Tang et al., 1990; Traynelis and Cull-Candy, ing NR2B, have been detected on synaptic, perisynaptic and 1990; Vyklicky et al., 1990). As a consequence of NMDAR extrasynaptic sites (Köhr, 2006). In most neurons, however, inhibition by protons, mild extracellular acidosis accompany- the density of NMDARs is higher in dendritic spines, within ing ischaemia and seizures may minimize glutamate-induced the postsynaptic density, than in the dendritic shaft and neuronal damage (Giffard et al., 1990; and see Dingledine somatic membrane. At immature glutamatergic synapses, it is et al., 1999). Sensitivity of NMDARs to extracellular protons well established that NR2B-containing receptors predominate depends on subunit composition, with influence of both NR1 and mediate direct synaptic transmission. With maturation, and NR2 subunits. Receptors containing the NR2C subunit the increase in NR2A expression usually results in faster decay are the least sensitive with an IC50 of pH 6.5 (Traynelis et al., kinetics and decreased sensitivity of synaptic currents to 1995; Low et al., 2003). Receptors made of NR1a and NR2A NR2B-selective antagonists. Moreover, several studies have have an intermediate pH sensitivity with an IC50 of pH 6.9 shown that extrasynaptic NMDARs are enriched in NR2B (when extracellular ambient zinc is chelated; Low et al., 2000; subunits compared with synaptic receptors. These extrasyn- 2003). NR1a/NR2B and NR1a/NR2D receptors display the aptic NMDARs may be activated by spillover of glutamate highest pH sensitivity with an IC50 of pH 7.4 (Traynelis et al., released from multiple neighbouring synaptic sites (Scimemi 1995). This implies that, under normal conditions, about half et al., 2004). However, in the adult brain, the segregation of of NR1a/NR2B receptors are under tonic proton inhibition; it NR2B and NR2A to extrasynaptic and synaptic sites is not also suggests that small changes in extracellular pH can sig- absolute, and there are clear examples of adult synapses where nificantly alter the amount of current flowing through these NR2B subunits predominate (see for instance, Lopez de receptors. Accordingly, endogenous alkaline transients fol- British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al lowing neuronal activity have recently been shown to boost receptors to extracellular zinc (IC50 15 nM for NR1/NR2A postsynaptic NMDAR responses in hippocampal CA1 pyrami- receptors; Chen et al., 1997; Paoletti et al., 1997; Choi and dal neuron (Makani and Chesler, 2007). Proton sensitivity is Lipton, 1999; Fayyazuddin et al., 2000; Low et al., 2000; also modulated by alternative splicing of the NR1 subunit.
Paoletti et al., 2000; Hatton and Paoletti, 2005). The NTD of Inclusion of NR1 exon-5 (NR1b subunit) reduces proton sen- NR2B (but not that of NR2C and NR2D) also binds zinc, but sitivity, an effect that has been attributed to the shielding of with a much lower affinity. This modulatory site accounts for the proton sensor by positively charged residues present at the low micromolar voltage-independent zinc inhibition of the C-terminus end of exon-5 (Traynelis et al., 1995; 1998). It NR1/NR2B receptors (IC50 1 mM; Traynelis et al., 1998; is interesting to note that exon-5-containing NR1b/NR2A and Rachline et al., 2005). Thus, when applied at nanomolar con- NR1b/NR2B receptors have equal proton sensitivity, with centrations (up to 300 nM), zinc selectively inhibits recep- IC50s close to pH 6.7 (Traynelis et al., 1995).
tors containing the NR2A subunit, a property than can be Although many residues, both on NR1 and NR2 subunits, used to discriminate between NMDAR subtypes in physiologi- have been shown to control proton inhibition (Traynelis cal experiments (see Neyton and Paoletti, 2006). Moreover, et al., 1998; Masuko et al., 1999; Low et al., 2000; 2003), the because zinc is concentrated and released at many glutamater- precise location of the proton ‘sensor' (assuming that it forms gic synapses in the CNS (Vogt et al., 2000; Paoletti et al., a discrete site) remains unknown. Extensive mutagenesis has 2009), zinc is a potential candidate for an endogenous ligand revealed that residues that influence proton sensitivity most of the NTDs of both NR2A and NR2B subunits. However, strongly cluster in two neighbouring regions (Low et al., 2003; whether extracellular zinc acts as an in vivo regulator of but see Gielen et al., 2008): (i) the lurcher region, composed of NMDAR activity remains unknown.
the ‘SYTANLAAF' motif and located in the extracellular end of Besides the zinc ion, the NR2B-NTD also binds ifenprodil the second transmembrane segment (M2); and (ii) the linkers [compound (1), Figure 2] and derivatives, a large family of connecting the ABD S2 segment to the transmembrane seg- synthetic organic compounds (Perin-Dureau et al., 2002).
ments (S2-M2 and S2-M3 linkers). These regions are closely Although first described as a cerebral vasodilator (Carron associated with the activation gate of NMDARs (Chang and et al., 1971), ifenprodil was later reported to have a neuropro- Kuo, 2008). The proton sensor is therefore likely to be tightly tective action through non-competitive antagonism of coupled to the movement of the NMDAR channel gate.
NMDARs (Carter et al., 1988). In 1993, Williams made the Single-channel studies at NR1/NR2B receptors reveal that remarkable observation that ifenprodil displays strong (>100- protons prolong one of the shut-time components, while it fold) preference among the various NR1/NR2 receptor sub- has almost no effect on channel open time (Banke et al., types, by selectively inhibiting receptors containing the NR2B 2005). These data strongly suggest that protons preferentially subunit (IC50 150 nM; Williams, 1993). Ever since this dis- act by stabilizing a closed state rather than destabilizing the covery, ifenprodil and derivatives have proven extremely useful as pharmacological tools to study the structure and Modification of proton sensitivity appears to be a common function of NMDARs. They have also triggered intense phar- downstream mechanism of a number of NMDARs' allosteric maceutical interest because of their therapeutic potential for a modulators. Thus, ifenprodil, exon-5 insert and polyamines range of neurological and psychiatric disorders (see below).
at NR2B-containing receptors, and extracellular zinc at NR2A- Studies on chimeric NR2A/NR2B subunits and on isolated containing receptors, all alter NMDAR function by shifting NTDs produced in bacteria mapped ifenprodil-binding site to the pKa of the proton sensor (Traynelis et al., 1995; Mott et al., NR2B-NTD (Gallagher et al., 1996; Perin-Dureau et al., 2002; 1998; Choi and Lipton, 1999; Low et al., 2000 and see below).
Wong et al., 2005; Ng et al., 2007; 2008; Han et al., 2008). An These data reinforce the idea that the proton sensor and the extensive mutagenesis scan, combined with molecular mod- channel gate are structurally and functionally integrated.
elling, further supported the hypothesis of ifenprodil bindingin the large interlobe cleft of the NR2B-NTD (Perin-Dureauet al., 2002). Functional studies also revealed that zinc andifenprodil act in a mutually exclusive manner at NR2B-NTD, Modulators binding the interlobe cleft of the
competing for putative binding sites that partially overlap (Rachline et al., 2005). No model of zinc binding has beenproposed yet, but the zinc ion is likely to interact with histi- One important recent development in NMDAR function and dine 127, glutamate 47 and aspartate 265, all located in the pharmacology has been the discovery that the NTDs of NR2A NR2B-NTD interlobe cleft (Rachline et al., 2005), and bearing and NR2B subunits form discrete modulatory domains side chains commonly found in zinc-binding sites (Alberts binding non-competitive antagonists with strong subunit et al., 1998). The first experimentally validated 3D model of selectivity (Herin and Aizenman, 2004). So far, only the zinc ifenprodil docked in its binding pocket has recently been ion has been identified as being a ligand of the NR2A-NTD described (Mony et al., 2009). In this model, ifenprodil occu- (Choi and Lipton, 1999; Fayyazuddin et al., 2000; Low et al., pies the NTD cleft, almost perpendicular to the plane of the 2000; Paoletti et al., 2000). Zinc binds the large interlobe hinge. The NTD cleft is in a closed conformation and ifen- crevice of the NTD and, by interacting with residues from prodil directly interacts with residues both from lobes 1 and 2, both lobes 1 and 2, promotes its closure by a hinge bending lending support to a model in which ifenprodil binding pro- mechanism, as seen in other clamshell-like domains (Paoletti motes closure of the NR2B-NTD. The ifenprodil molecule et al., 2000). Binding of zinc to the NR2A-NTD is responsible adopts a conformation in which its benzyl group contacts for the exquisite (nM) sensitivity of NR2A-containing hydrophobic residues at or near the NTD hinge, while its British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
phenol moiety makes hydrogen bonds with polar residues at amines [Alanine et al., 2003, (15)], Taisho's HON0001 the entrance of the cleft. The central piperidine moiety makes [Suetake-Koga et al., 2006, (16)] and di-imidazoles (17). Many both Van der Waals interactions with F176 and ionic interac- of these structurally distinct compounds are highly potent, tions with D101. The fact that this latter residue may also be with low nanomolar affinities at NR1/NR2B. However, directly involved in zinc coordination would explain why whether these latest generation NR2B-selective antagonists ifenprodil and zinc cannot occupy the NR2B-NTD cleft simul- have a similar binding mode to ifenprodil remains to be taneously (Rachline et al., 2005).
determined. The recently proposed 3D model of ifenprodil- Opposing an ifenprodil-binding site entirely formed by binding site into NR2B-NTD (Mony et al., 2009) provides a NTD-NR2B residues, some data in the literature suggest that useful approach to tackle this question.
determinants of ifenprodil binding may reside on the NR1 The transduction cascade that couples binding of the subunit. Thus, Masuko et al. (1999) found mutations in NR1- modulatory NTD ligand to receptor inhibition (i.e. channel NTD lobe 1 that affect ifenprodil inhibition, while Han et al. gate closure) has been recently dissected in the case of the (2008) showed that isolated NR1-NTDs, similarly to isolated high-affinity zinc inhibition of NR1/NR2A receptors (Gielen NR2B-NTDs (but not NR2A-NTDs), bind radiolabelled ifen- et al., 2008, and see Figure 4B). First, zinc binds the interlobe prodil. The NR1 residues highlighted by Masuko et al. (1999) cleft of the NR2A-NTD and promotes its closure. This are located at positions homologous to residues participating closure then exerts tension on the linkers connecting the in hydrophobic dimerization interfaces in other receptors NTDs to the ABDs, an effect that triggers disruption of the containing LIVBP-like domains. Rather than directly binding ABD dimer interface; this in turn relieves the strain on ifenprodil, these residues may therefore be involved in the the transmembrane segments, and together with proton binding, allows closure of the channel gate. This mechanism changes of NR2B-NTD (see Perin-Dureau et al., 2002). More- shows common features with that underlying fast desensiti- over, it is possible that such hydrophobic residues fortuitously zation of AMPA and kainate receptors, where conforma- bind ifenprodil when NR1-NTD is isolated in a polar solvent.
tional rearrangements at the ABD dimer interface also occur In fully assembled NR1/NR2 receptors, this potential binding (Sun et al., 2002). The inhibition of NR2B-containing recep- site is likely to be masked following NR1 and NR2 NTD tors following zinc or ifenprodil binding to the NR2B-NTD may proceed through a similar mechanism to that described Ifenprodil is the prototypical member of a large, and for zinc on NR1/NR2A receptors. The fact that ifenprodil growing, family of NR2B-selective antagonists that can be inhibition of NR1/NR2B receptors reflects an enhancement usefully grouped as ‘prodils'. Among them are synthesized of tonic proton inhibition (Pahk and Williams, 1997; Mott et al., 1998), similarly to zinc inhibition of NR1/NR2A receptors) and selectivity versus additional ‘off' target activi- receptors (Choi and Lipton, 1999; Low et al., 2000), argues ties such as adrenergic and sigma receptors (Kew and Kemp, in this direction. However, zinc inhibition of NR1/NR2B 1998; Chenard and Menniti, 1999; Nikam and Meltzer, 2002; receptors appears not to depend on pH (Traynelis et al., Borza and Domany, 2006). The best characterized compounds 1998; Low et al., 2000), suggesting that zinc binding to and, consequently, those most commonly used as pharmaco- NR2B-NTD may inhibit NR1/NR2B receptors through a dif- logical tools are traxoprodil or CP-101,606 [Chenard et al., ferent mechanism.
1995, (5)], besonprodil [CI-1041, (6), Chizh et al., 2001] andRo 25-6981 [Fischer et al., 1997, (7)], all of which are 10-foldmore potent than ifenprodil; the recently disclosed RGH-896or radiprodil (8) can also be considered to form part of this Polyamines: NR2B-selective positive allosteric
chemical group (Figure 3A). These ‘second generation' compounds, which share the same structural features as ifen-prodil, are likely to bind in a similar mode to the NR2B-NTD Polyamines are polybasic aliphatic amines that are positively (see Malherbe et al., 2003 for Ro 25-6981). Large-scale screen- charged at physiological pH. The endogenous polyamines, ing approaches and medicinal chemistry efforts have also led spermine [(3), Figure 2], spermidine and putrescine, are syn- to the identification of further novel NR2B-selective antago- thesized from ornithine, a by-product of the urea cycle.
nists such as MK-0657 (9) (Figure 3B) and EVT-101 (structure Polyamines are widely distributed throughout the body and not disclosed). These agents, like radiprodil, have good are found at high intracellular levels, where they interact with potency at NR2B-containing receptors and enhanced oral bio- nucleic acids and proteins, including plasma membrane ion availability compared with earlier agents, and have been pro- channels. In the CNS, there is also evidence that, under gressed into the clinic by Merck, Evotec and Gedeon Richter/ certain conditions, polyamines may be present in the extra- Forest Laboratories respectively (see Table 1). Recently, a more cellular medium, where they would act as modulators of diverse array of structures have been described as NR2B neuronal excitability. First, polyamines are released in a antagonists including (Figure 3B) indole- and benzimidazole- Ca2+-dependent manner following neuronal stimulation; 2-carboxamides [Borza et al et al., 2006; 2007, (10)], 2-(3,4- second, high-affinity uptake systems for polyamines exist; and third, extracellular polyamines interact with various ion 2003, (11)], benzamidines [Claiborne et al., 2003, (12)], channels and receptors, including calcium channels and NMDARs (Rock and Macdonald, 1995).
N1-(benzyl)cinnamamidines [Curtis et al., 2003; Kiss et al., Extracellular polyamines have multiple effects on NMDAR responses (Rock and Macdonald, 1995). Early patch-clamp British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al A Ifenprodil-related structures Besonprodil (CI-1041) (6) CP-101,606, ‘traxoprodil' (Pfizer) (5) RGH-896,‘radiprodil' Ro 25-6981 (Roche) (7) (Gedeon Richter/Forest) (8) B Next generation NR2B antagonists and new structural templates MK-0657 (Merck) (9) (10) Gedeon Richter Structure of NR2B-selective NMDAR antagonists. (A) ‘Second generation' compounds closely related in structure to the prototypical NR2B antagonist ifenprodil. (B) The latest generation of NR2B-selective antagonists and new structural templates. This represents a currentperspective based on publications, patents, company press releases and analyst information; literature references, where available, are cited inthe text. NMDAR, N-methyl-D-aspartate receptors.
studies on native NMDARs from cultured hippocampal neurons showed highly variable effects of polyamines, independent potentiation. The voltage-dependent block is ranging from a strong enhancement to a marked inhibition due to spermine entering the pore and occluding ion fluxes, depending upon the particular cell examined (Benveniste and much like the block produced by extracellular Mg2+ (Rock and Mayer, 1993). It is now well established that this variability MacDonald, 1992). It is highly voltage-dependent with IC50 can be accounted for different levels of expression of NMDAR values of 350 mM at -60 mV and 27 mM at 0 mV (NMDAR subpopulations in individual cell, together with differential responses from cultures hippocampal neurons; Benveniste effects of polyamines depending on receptor subunit and Mayer, 1993). In contrast to Mg2+, at very hyperpolarized potentials, polyamines may escape their blocking site by per- Spermine (and spermidine) produces three different effects meating the NMDAR channel (Araneda et al., 1999). The on NMDARs: a voltage-dependent block, a glycine-dependent polyamine block has the same subunit dependence as the British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
Spermine, but not histamine, potentiates NR1a/NR2B receptors. (A) Typical current traces obtained from oocytes expressing the NR1-1a subunit (NR1a) with the NR2B (left panel) or the NR2A (right panel) subunit. Spermine was applied at a concentration of 200 mM andhistamine at 100 mM, each during an application of agonists (100 mM glutamate and glycine, saturating concentrations). Holding potential-40 mV (left panel) and -30 mV (right panel). The bars above the current traces indicate the duration of agonist, spermine and histamineapplications. Note that application of 100 mM histamine has no effect on NR1a/NR2B receptors while spermine does (same cell). Note also thatspermine potentiation is absent on NR1a/NR2A receptors. (B) Two hypothetical mechanisms of how a polyamine could potentiate NR2B-containing receptors. These models are based on the mechanism proposed for allosteric inhibition of NR1/NR2A receptors (Gielen et al., 2008).
While the full receptor is a tetramer, only a NR1/NR2B heterodimer is shown. It is hypothesized that NTDs dimerize, and that closures of theNTDs can inactivate the receptors (i.e. induce channel gate closure) by pulling apart the ABD dimer interface (‘desensitized' state). Mechanism(1): the polyamine molecule binds between the NR1 and NR2 NTD lobes II, making NTD closure, and ABD dimer interface disruption, moredifficult. Mechanism (2): the polyamine molecule directly binds and stabilizes the ABD dimer interface. Entry into the ‘desensitized' state is thusdisfavoured. This mechanism resembles that described for cyclothiazide-induced suppression of desensitization at AMPA receptors (Sun et al.,2002; Mayer, 2006). In both models, proton binding, which stabilizes a closed state of the channel (Banke et al., 2005), is proposed to beclosely associated with ABD dimer interface breaking (Gielen et al., 2008). ABD, agonist-binding domain; AMPA, a-amino-3-hydroxy-5-methyl-4-isoxazole-propionate; Glu, glutamate; Gly, glycine; NTD, N-terminal domain.
Mg2+ block, being less pronounced for NR2C-containing and effect largely due to a decrease in the rate of glycine disso- NR2D-containing receptors than for NR2A-containing or ciation from its binding site on NR1 (Benveniste and Mayer, NR2B-containing receptors (these latter two having the same 1993). Glycine-dependent potentiation occurs both at degree of blockade; Williams et al., 1994). Finally, mutations NR2A-containing and NR2B-containing receptors but not at in the pore at the critical asparagines residues (Q/R/N site) NR2C-containing and NR2D-containing receptors (Zhang that suppress Mg2+ block also suppress spermine block, indi- et al., 1994; Williams et al., 1995). Moreover, in contrast to cating a shared binding site deep in the pore (Kashiwagi et al., the voltage-independent and glycine-independent potentia- 1997; Traynelis et al., 1998).
tion (see below), the glycine-dependent stimulation is not Polyamines can enhance NMDAR responses by increasing influenced by the type of NR1 subunit splice variants. The the apparent sensitivity for glycine (glycine-dependent polyamine-binding site mediating the glycine-dependent potentiation). This is reflected by the fact that at low glycine potentiation is unknown, but may reside on the NR1 concentrations, polyamines stimulate NMDAR responses to subunit that harbours the glycine-binding site. In physi- a greater extent than at saturating glycine concentrations ological conditions, it is likely that this site is partially (McGurk et al., 1990). In the presence of 1 mM spermine, occupied by endogenous Ca2+ and/or Mg2+ because both the apparent affinity for glycine increases by 3-fold, an cations also increase glycine sensitivity when present at British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
mM concentrations (Wang and MacDonald, 1995; see tiation by polyamines requires the NR2B subunit. A model in Paoletti et al., 1995).
which the polyamine molecule binds the NR1/NR2B NTD The third effect of polyamines on NMDARs is a potentia- interface and holds the NTDs open, while ifenprodil binds in tion, which is both voltage-independent and glycine- the NR2B-NTD cleft and promotes its closure, is fully consis- independent because it can be observed both at negative and tent with the proposal of Kew and Kemp (1998) that spermine positive membrane potentials and with saturating glycine and ifenprodil bind distinct sites that interact in a negative concentrations (McGurk et al., 1990; Lerma, 1992; Benveniste allosteric manner (with binding of spermine promoting ifen- and Mayer, 1993; Lu et al., 1998; Figure 4A). With spermine, prodil dissociation and vice versa; see also Han et al., 2008).
the EC50 for this effect is 150 mM and results in a maximal The neurotransmitter histamine has been proposed to act as potentiation around threefold (at pH 7.3; Benveniste and an endogenous activator of the NR2B-specific polyamine Mayer, 1993). Remarkably, the voltage-independent and modulatory site (Williams, 1994b). However, results from our glycine-independent polyamine potentiation is observed lab (Figure 4A) and that of Steven Traynelis (pers. comm.) exclusively at receptors that incorporate the NR2B subunit show a lack of effect of histamine on responses mediated by (Williams, 1994a; Williams et al., 1994; Zhang et al., 1994; NR1a/NR2B receptors. The reason for this discrepancy Traynelis et al., 1995; Figure 4A). Studies on recombinant remains unclear. Finally, Mg2+, but not Ca2+, mimics the NR1/NR2B receptors have revealed that the NR2B-specific glycine-independent potentiating effect of polyamine at polyamine potentiation arises from the relief of tonic proton NR2B-containing receptors (Paoletti et al., 1995; Kew and inhibition. Thus, at physiological pH, polyamines enhance Kemp, 1998). With an Mg2+ EC50 of 2.0 mM close to physi- NR1/NR2B responses by shifting the pKa value of the proton ological concentration of extracellular Mg2+, the NR2B- sensor towards more acidic values (Traynelis et al., 1995).
specific polyamine potentiating site is likely to be partially Accordingly, at NR1/NR2B receptors, there is a strong corre- occupied by the magnesium ion under physiological lation between the proton sensitivity and the degree of polyamine potentiation (Traynelis et al., 1995; 1998).
Many mutations that affect the NR2B-specific polyamine potentiation have been described. These mutations, usually ofacidic residues, are found both on NR1 and NR2 subunits and Negative and positive allosteric modulation by
are scattered throughout the extracellular region. Most of these mutations (if not all) also modify proton sensitivity(Traynelis et al., 1995; Williams et al., 1995; Kashiwagi et al., Most steroids act like hormones being synthesized by glandu- 1996; Masuko et al., 1999). In consequence, it is still unclear lar tissues and released into the general circulation system.
whether these mutations alter polyamine potentiation Steroids diffuse through membranes and bind intracellular directly, by disrupting the polyamine-binding pocket, or indi- receptors, which in turn interact with transcription factors to rectly, by changing the proton sensitivity. Based on our recent enhance or suppress gene expression. Because of the necessity finding that proton inhibition of NMDARs involves disrup- for activation of the transcriptional and translational machin- tion of the ABD dimer interface (Gielen et al., 2008), we eries, the physiological responses induced by steroids usually propose two mechanisms by which polyamines could relieve take hours to days. Steroids synthesized in the periphery can tonic proton inhibition and thus enhance activity of NR2B- cross the blood–brain barrier and produce changes in mood containing receptors (Figure 4B). In a first model, the and behaviour (Belelli and Lambert, 2005). The brain is also polyamine molecule would act similarly than cyclothiazide capable of synthesizing steroids de novo (Robel and Baulieu, on AMPA receptors (Sun et al., 2002): it would bind at the 1994). Neurosteroids are normally present at nanomolar con- ABD dimer interface and stabilize the dimer assembly. By centrations in the CNS, but their levels can increase signifi- doing so, it would decrease pH sensitivity by rendering ABD cantly following stress, for instance. In contrast to the dimer disruption more difficult (an effect that would also classical genomic effects of steroids, endogenous neuroster- account for the spermine-induced acceleration of NMDAR oids act locally and produce acute effects on neuronal excit- current deactivation; Rumbaugh et al. 2000). In a second ability, with time delays ranging from seconds to minutes, model, the polyamine would bind at the level of the NTDs, suggesting direct modulatory effects on membrane proteins.
between the two ‘bottom' lobes of a NTD pair. By ‘gluing' It is now well established that gamma-amino butyric acid-A together these lobes, the polyamine would render NTD (GABA-A) receptors, which mediate most of the inhibitory closure less favourable, an effect that in turn would tend to transmission in the CNS, are major targets of neurosteroids stabilize the ABD dimer interface (and thus decrease proton (Belelli and Lambert, 2005). Neurosteroids also regulate sensitivity; Gielen et al., 2008). In agreement with this mecha- NMDARs. At these receptors, neurosteroids exert either posi- nistic scheme, 3D homology modelling of a NR1/NR2B NTD tive or negative effects depending both on the neurosteroid pair indicates that many of the acidic residues of NR1 and chemistry and the receptor subunit composition.
NR2B known to control glycine-independent spermine Pregnenolone sulfate [PS (4), Figure 2], one of the most potentiation locate on the side of the NTD ‘bottom' lobes, abundant neurosteroids and a negative modulator of GABA-A with residues of NR2B facing those of NR1. This suggests a receptors, was initially shown to potentiate native NMDARs polyamine-binding site at the boundary between the two but not AMPA and kainate receptors (Wu et al., 1991; Bowlby, NTD ‘bottom' lobes (Huggins and Grant, 2005). The absence 1993). PS is a derivative of pregnenolone, which is formed by of some of these acidic residues in the NTDs of the other NR2 cleavage of cholesterol side chain in glial cells. Potentiation of subunits could explain why the glycine-independent poten- NMDAR responses by PS is voltage independent, does not British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al affect single-channel conductance and occurs only at rela- have led to the identification of important structural features tively high concentrations of PS (>1 mM). Interestingly, when that determine the action of neurosteroids at NMDARs. The applied in the extracellular medium, PS increases NMDAR presence of a negative charge (bulky or not) at the C3 position activity recorded not only from excised outside-out patches seems mandatory for both the potentiating and inhibitory but also from cell-attached patches, indicating that the lipo- effects of neurosteroids on NMDARs (Park-Chung et al., 1997; philic PS may penetrate the cell membrane and reach NMDAR Weaver et al., 2000). Moreover, the geometry of the neuros- channels by diffusion (Bowlby, 1993). PS differentially modu- teroid strongly influences the sign of the modulatory effect.
lates activity of recombinant NR1/NR2 NMDARs. It potenti- Neurosteroids having a bent geometry (3a5bS and 3b5bS) ates NR1/NR2A and NR1/NR2B receptors (with EC50 in the inhibit native NMDAR responses (mostly from NR2A- 10–30 mM range) while it inhibits, in a non-competitive containing and/or NR2B-containing receptors), whereas those manner, NR1/NR2C and NR1/NR2D receptors (with IC50 in having a more planar configuration (PS and 3b5aS) potentiate the 100–200 mM range; Malayev et al., 2002; Horak et al., such responses (Weaver et al., 2000). Thus, whether a neuro- 2006). The mechanism of action of PS at NR1/NR2B receptors steroid enhances or decreases NMDAR activity critically has been studied using fast perfusion techniques on trans- depends on the geometry of its A–B ring junction, a parameter fected human embryonic kidney (HEK)-293 cells. Strikingly, that is determined by the stereochemistry of the C5 position.
the degree of PS potentiation is an order-of-magnitude largerwhen PS is applied just before than during NMDAR activa-tion, indicating that the PS affinity is strongly decreased afterglutamate binding (Horak et al., 2004).
NR2B-containing receptor function and the
PS enhances NMDAR responses by acting at a site that likely therapeutic potential of subtype-selective NMDAR
differs from the site(s) responsible for spermine potentation (Park-Chung et al., 1997). By constructing chimeric NR2B/NR2D (Jang et al., 2004) or NR2A/NR2C (Horak et al., 2006) There is currently great interest in determining whether subunits, the PS potentiating site has been recently mapped to different NMDAR subtypes make specific contributions to a region encompassing the M2–M3 extracellular loop and the physiological and pathological neuronal processes. Dissection M3 transmembrane segment. In the NR2 M2–M3 loop, helices of the roles of NMDAR subtypes can be performed using J and K appear to be important for PS sensitivity (Jang et al., genetic (genetically modified mice, RNAi-mediated gene inter- 2004). Because helix J is a key constituent of the dimer inter- ference) or pharmacological tools. All of these approaches have face between neighbouring NR1-NR2 ABDs, Jang et al. (2004) recognized limitations (e.g. see Neyton and Paoletti, 2006 for have proposed that PS binds and stabilizes this interface very the pharmacological approach) and, hence, despite intense much like the positive allosteric modulator cyclothiazide on investigation, there are many controversial issues in the field.
AMPA receptors. There are, however, experimental data that Several parameters may strongly influence the role that a are difficult to reconcile with a PS-binding site at the ABD NMDAR subtype may play: (i) its particular receptor subunit dimer interface. First, PS potentiation appears to be function- composition, which determines the level of receptor activity ally independent from proton inhibition (Jang et al., 2004), and kinetic behaviour; (ii) its subcellular location (synaptic or whereas stability of the ABD dimer is closely correlated to extrasynaptic); (iii) its coupling to downstream signalling cas- proton sensitivity (Gielen et al., 2008). Second, the transmem- cades (these latter two parameters being likely dependent on brane segment M3 appears critical for the potentiating effect the subunit composition); and (iv) potential changes in its of PS, suggesting that this region may contain residues expression or activity linked to disease.
directly binding PS (Jang et al., 2004). Thus, the potentiating Based on our current understanding of the roles of the effect of PS on NMDARs may be mediated by a site in the different NMDAR subtypes in normal physiology and in membrane region rather than at an extracellular location. The pathophysiological conditions, it appears that subtype- lipophilic nature of PS and the very slow kinetics of its action selective modulation of receptor function offers emerging at NR1/N2B receptors (Horak et al., 2004) are compatible with therapeutic potential for the treatment of a range of CNS this hypothesis. It is interesting to note that at GABA-A recep- disorders. Thus, NR2B-selective antagonists may offer utility tors, transmembrane sites are responsible for the regulatory for the treatment of disorders including chronic pain, effects of neurosteroids on these receptors (Hosie et al., 2006).
The PS analogue 3a-hydroxy-5b-pregnan-20-one sulphate disease, cerebral ischaemia and major depression.
[3a5bS (2), Figure 2] is another endogenous neurosteroid that Notably, while non-selective NMDAR antagonists com- modulates NMDAR activity. 3a5bS inhibits native and recom- monly exhibit a range of side effects including behavioural, binant NMDARs, with little selectivity among the different cardiovascular and potentially cytotoxic activities, which NR1/NR2 receptor subtypes. It acts in a voltage-independent have limited their therapeutic development, NR2B-selective and non-competitive manner, strongly arguing for a binding antagonists are relatively well tolerated (Kemp et al., 1999; site outside the pore and the agonist-binding pockets (Park- Chazot, 2004; Gogas, 2006). In addition to the inherent chung et al., 1994; Park-Chung et al., 1997; Malayev et al., sparing of non-NR2B-containing receptor function as a con- 2002; Petrovic et al., 2005). Thus, despite strong chemical sequence of subtype selectivity, including a relative lack of similarity, PS and 3a5bS produce opposite effects at NR1/ effect in brain regions with little or no NR2B expression, such NR2A and NR1/NR2B receptors, the former acting as a posi- as the cerebellum, ifenprodil and related compounds also tive allosteric modulator and the latter as a negative allosteric display an activity-dependent mode of action; they bind with modulator. Structure-activity studies on the steroid molecules a higher affinity to the activated and desensitized states of the British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
receptor than to the unliganded resting state (Kew et al., 1996; behavioural side effects. More recent research has also Fischer et al., 1997; Gill et al., 2002). Thus, these compounds defined a key involvement of NR2B-containing receptors in preferentially block NMDARs continuously or repetitively mechanisms, such as ‘wind up' and ‘central sensitization', activated, as may be the case in pathological conditions, that operate spinally and supraspinally, and which may while leaving those that are physiologically activated rela- explain the apparent broad-spectrum activity of NR2B antago- tively unaffected. In addition, as ifenprodil inhibition of the nists across the various disease models. These pharmacologi- NR2B-containing receptors is mediated through an enhance- cal studies have been complemented by additional insight ment of proton inhibition (Pahk and Williams, 1997; Mott into the role of NR2B-containing receptors in pain gained et al., 1998), antagonist efficacy would be further increased from studies with transgenic mice overexpressing NR2B in the under pathological conditions where pH levels also fall, such forebrain which display a selective enhancement of persistent as in ischaemia (Silver et al., 1992). Thus, both the mecha- pain and allodynia (Wei et al., 2001). Similarly, intrathecal nism of action and subunit selectivity contribute to the rela- administration of siRNAs against the NR2B subunit reduces tive lack of adverse events/good tolerability profile associated the pain responses induced by peripheral inflammation (Tan with this compound class (Kemp et al., 1999; Gogas, 2006).
et al., 2005).
Although most studies point towards an improved thera- The potential of NR2B antagonists for the treatment of pain peutic index with the NR2B-subtype-selective approach, has now gained clinical precedence based on the observed recent studies have raised concerns regarding the role of NR2B efficacy of CP-101,606 in a placebo-controlled crossover phase in mediating phencyclidine (PCP)-like behavioural effects and 2a study in spinal cord injury and monoradiculopathy the potential for abuse liability as previously associated patients, where a significant reduction in pain score was seen with pan-NMDAR antagonists, such as ketamine. Preclinical following i.v. dosing of CP-101,606 (Sang et al., 2003).
studies with Ro 25-6981 (Chaperon et al., 2003) and CP-101,606 was reasonably well tolerated in this study but did CP-101,606 (Nicholson et al., 2007) suggest a role of NR2B in yield some CNS side effects (Table 1). Despite these encourag- producing the subjective and reinforcing effects associated ing data, subsequent progress in this field has been hampered with PCP in rodents and primates. Although noted to be by the lack of compounds with good orally bioavailability, generally well tolerated in the clinic, some indications of appropriate selectivity versus human ether-a-gogo related such activity have emerged in recent clinical studies with gene (hERG), and concerns regarding CNS side effects and CP-101,606 (Table 1). Further work is, therefore, required to abuse potential as discussed above. Despite these challenges, explore these aspects in more detail, especially for the latest Gedeon Richter and Forest Laboratories, Merck and Evotec NR2B-selective antagonists progressed into the clinic, to have now identified orally bioavailable agents that have pro- determine if an acceptable balance between efficacy and side gressed into clinical development (Table 1). To date, only effects exists to support their further development for the limited information concerning the safety profile of these treatment of the range of disorders outlined below.
agents in phase 1 trials is available, and this provides somebasis for optimism (Table 1). However, it will only be withefficacy data from appropriate phase 2 studies that theseinitial findings can be put into context of therapeutic index Therapeutic potential of NR2B-selective
and the prospects for further development fully evaluated.
Clinically, broad-spectrum NMDAR antagonists (e.g. ket- The degeneration of nigral dopaminergic neurons and the amine, dextromethorphan) are reportedly used off label for depletion of dopamine from the nigro-striatal pathway results the treatment of neuropathic pain (Chizh et al., 2001; 2007); in overactivation of glutamatergic projections to the striatum they have good efficacy, but suffer from a generally unaccept- and basal ganglia output nuclei. In addition, NMDARs have able side-effect profile due to a very narrow therapeutic index.
been proposed to play a role in the development of levodopa- The rationale for developing subtype-selective NMDAR induced dyskinesias. Accordingly, the therapeutic potential antagonists therefore stems from a desire to maintain of NMDAR antagonists has been investigated, and broad- NMDAR- mediated efficacy while moving away from the side spectrum antagonists have been shown to exhibit antiparkin- effects resulting from indiscriminate broad NMDAR blockade.
sonian or antidyskinetic activity in rodent and monkey Initial focus on NR2B derived from the more localized expres- models, and the low affinity NMDAR antagonist amantadine sion of NR2B in the dorsal horn of the spinal cord as well as has been shown to exhibit antidyskinetic activity in humans in higher brain centres thought to be important for the relay (Del Dotto et al., 2001). Subsequently, NR2B-selective antago- and conscious perception of pain, for example, the thalamus nists have been shown to exhibit efficacy in preclinical and anterior cingulate cortex (Figure 1). The availability of the models of Parkinsonism in both rodents and non-human ‘prodil' class of NR2B-selective antagonists (Figure 3) has led primates (e.g. Steece-Collier et al., 2000; Loschmann et al., to rapid progress in validating NR2B-containing NMDARs as a 2004; Wessell et al., 2004), with an apparently improved tol- target for the treatment of pain, with many studies reporting erability profile relative to broad-spectrum antagonists. While efficacy of such agents in models of acute (Taniguchi et al., ifenprodil failed to show a significant benefit in a small clini- 1997; Boyce et al., 1999) and chronic inflammatory (Wilson cal trial in Parkinson's disease patients as adjunct therapy et al., 2006), neuropathic (Boyce et al., 1999) and even visceral to L-DOPA (Montastruc et al., 1992), in a recent study, pain (Boyce et al., 2002) at doses that are devoid of overt CP-101,606 exhibited antidyskinetic but not antiparkinso- British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al nian effects, although at doses associated with dissociation channel blocker MK-801; Guscott et al., 2003). In addition, and amnesia (Nutt et al., 2008). Further studies are required to while CP-101,606 promoted impulsive-type responding, it determine whether therapeutic efficacy can be achieved at improved task performance in a rat delayed match to position lower doses, in the absence of adverse cognitive effects. A task (Higgins et al., 2005). Thus, NR2B-selective antagonists phase 1 study with MK-0657 as adjunct therapy to levodopa may have therapeutic potential as cognitive enhancers has also been reported as completed, although data have not in Alzheimer's disease, and Evotec is currently developing yet been disclosed (Table 1).
EVT-101 for this indication (Table 1).
Cerebral ischaemia and traumatic brain injury Huntington's disease is an autosomal dominant inherited The role of CNS glutamate receptors and, particularly, the disease characterized by selective neuronal degeneration, highly calcium permeable NMDARs in mediating the excito- most prominently of striatal GABAergic medium-sized spiny toxic neuronal cell damage observed in both cerebral neurons. The disorder results from expression of mutant ischaemia and traumatic brain injury is well recognized.
forms of the huntingtin gene, which contain an expanded Accordingly, broad-spectrum NMDAR antagonists have been trinucleotide CAG repeat sequence encoding an extended shown to be neuroprotective when administered before or polyglutamate tract in the translated protein. Studies of shortly after traumatic brain injury or ischaemic insult in animal and cellular models expressing mutant forms of the animal models (Kemp et al., 1999). However, the promise of huntingtin protein have implicated NMDAR dysregulation preclinical data has not been realized in patients with several and subsequent excitotoxic damage in the pathophysiology agents, failing to show efficacy in clinical studies (e.g. Morris of the disease. For example, in a mouse model of Hunting- et al., 1999; Lees et al., 2000; Albers et al., 2001). It has been ton's disease, striatal GABAergic medium-sized neurons show speculated that the dose-limiting side effects associated with a selective enhancement of NMDAR currents mediated by such broad-spectrum NMDAR antagonists may have contrib- NR2B-containing receptors (Zeron et al., 2002) and this uted to the failure of these studies. NR2B-selective antagonists increase may underlie the selective neuronal loss in Hunting- may offer potential as efficacious neuroprotective agents with ton's disease (Li et al., 2003). Thus, early intervention with an acceptable tolerability profile. In support, data from several selective NR2B antagonists at the onset of pathology, or studies suggest that activation of extrasynaptic (mostly NR2B- presymptomatically, may offer therapeutic potential in this containing) receptors triggers pro-death signalling events, while activation of synaptic (mostly NR2A-containing) recep-tors favours neuronal survival (Hardingham and Bading,2003; Zhou and Baudry, 2006; von Engelhardt et al., 2007; Liu et al., 2007; Chen et al., 2008; Martel et al., 2009). However, The low affinity, broad-spectrum NMDAR channel blocker while NR2B-selective antagonists are neuroprotective in memantine has been approved in both the USA and Europe animal models with an improved therapeutic index relative to for the treatment of moderate to severe Alzheimer's disease.
broad-spectrum antagonists (Kemp et al., 1999; Tahirovic As NMDAR activity is well recognized to play a critical role in et al., 2008), clinical studies have yielded disappointing out- learning and memory, the efficacy of memantine as a symp- comes, with CP-101,606 failing to demonstrate efficacy, tomatic, cognitive-enhancing therapy in Alzheimer's disease appears, at first sight, to be counter-intuitive. Memantine is a (Merchant et al., 1999; Saltarelli et al., 2004).
low-affinity, voltage-dependent uncompetitive antagonistwith fast dissociation kinetics, and it has been proposed thatit mediates efficacy through normalizing aberrant, disease- Major depression associated low level NMDAR activation without impairing Following the demonstration of antidepressant-like activity of physiological synaptic receptor activation, in a manner some- broad-spectrum NMDAR antagonists in animal models, two what analogous to the endogenous NMDAR channel blocker, small crossover clinical studies have reported a positive anti- magnesium (Parsons et al., 2007). As such, it is thought to depressant effect of the NMDAR channel blocker ketamine improve the ‘signal to noise' characteristics of NMDAR (Berman et al., 2000; Zarate et al., 2006). Notably, the antide- signalling in the diseased brain.
pressant effect exhibited a rapid onset and persisted for NR2B-containing NMDARs, which are enriched in the several days after single dose administration (Zarate et al., extrasynaptic receptor population, represent plausible candi- 2006), presenting a potentially attractive therapeutic profile.
dates for mediating such disease-associated ‘background' However, the clinical applicability of ketamine is limited by NMDAR activation. Preclinical studies in healthy adult rats its psychomimetic activity, which was observed in both that notably do not exhibit the proposed disease-associated studies. Preclinically, NR2B-selective antagonists have also NMDAR dysfunction provide some support for the rationale been reported to exhibit antidepressant activity (Maeng and of selectively targeting NR2B-containing receptors in terms of Zarate, 2007), and recently a randomized, double-blind study absence of cognitive impairment and evidence for cognitive in patients with refractory major depressive disorder treated enhancement. Thus, CP-101,606 at a dose which fully occu- with CP-101,606 was reported (Preskorn et al., 2008). In this pied hippocampal NR2B-containing receptors did not impair parallel group study, subjects with major depression and a spatial learning or memory in the Morris water maze task history of treatment refractoriness to selective serotonin in healthly adult rats (in contrast to the broad-spectrum reuptake inhibitors (SSRI) first received 6 weeks open-label British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
treatment with the SSRI paroxetine followed by a single-blind hension of the mechanisms by which extracellular agents placebo infusion. Paroxetine non-responders (n = 30) were can affect NMDAR activity. In particular, it has become then randomized to a single infusion of CP-101,606 or increasingly clear that the large NTD of NR2 subunits, which placebo plus continued treatment with paroxetine for up to a precedes the glutamate-binding domain, is a major site for further 4 weeks. CP-101,606 was generally well tolerated and subunit-specific allosteric modulation. In that respect, recep- its administration produced a greater antidepressant effect tors incorporating the NR2B subunit appear particularly inter- than placebo, thus, illustrating antidepressant potential for esting because the NR2B-NTD not only harbours sites for NR2B subtype-selective antagonists in otherwise treatment negative allosteric modulators (such as zinc and ifenprodil) refractory patients. A further clinical study in refractory but might also confer a unique sensitivity to the positive depression has recently been initiated with MK-0657 allosteric modulators polyamines and Mg2+ by participating in an inter-subunit regulatory site. Finally, our increasing under-standing of NMDAR structure, function and pharmacology isnow translating into promising therapeutic strategies to target Positive modulation of NR2B-containing NMDA
NMDAR dysregulation. The therapeutic utility of broad- spectrum NMDAR antagonists is typically limited by theirassociated side effects. Newer, subtype-selective negative As discussed above, a number of NMDAR positive modulators modulators of receptor function, primarily targeting NR2B- have been identified, including polyamines, neurosteroids containing receptors, have entered clinical development and and Mg2+, some of which exhibit subunit selectivity. Based on offer improved potential for the treatment of patients suffer- the increasing understanding of receptor structure and the ing from a range of debilitating psychiatric and neurological molecular mechanisms underlying both positive modulation disorders. Opportunities for selective positive modulation of by such ligands and subunit-selective antagonism, it seems NMDARs and to selectively target other NMDAR subtypes or reasonable to expect that appropriately configured screening their downstream regulatory cascades are currently more campaigns might identify novel, subunit-selective small mol- limited but are likely to be a focus of future research. The ecule positive modulators. Subunit-selective positive modula- advent of biological therapeutic agents, such as antibodies (or tion might represent a strategy to enhance receptor function, fragments thereof), and RNA inhibition may also enable for example, under pathological conditions of receptor hypo- therapeutic interventions considered intractable using exist- function, without triggering excitotoxicity through receptor ing small molecule approaches.
overactivation. NMDAR hypofunction has been implicated inthe pathophysiology of schizophrenia, initially based on thepsychomimetic activity of the broad-spectrum ion channel Conflicts of interest
blockers, ketamine and PCP, which have been reported toelicit a schizophrenia-like phenotype in healthy volunteers, LM and PP declare no conflict of interest. MJG and JK are encompassing the characteristic positive, negative and cogni- employees of GlaxoSmithKline PLC.
tive symptoms of the disease (see, e.g. Krystal et al., 2005).
Subsequent neurophysiological, neuroanatomical and bio-chemical studies have provided support for disease-associated NMDAR hypofunction (e.g. Umbricht and Krljes, 2005; Hahnet al., 2006; Woo et al., 2008). Efforts to enhance disease- Akazawa C, Shigemoto R, Bessho Y, Nakanishi S, Mizuno N (1994).
associated receptor hypofunction to date have centred on Differential expression of five N-methyl-D-aspartate receptor strategies to enhance glycine site occupancy and, hence, subunit mRNAs in the cerebellum of developing and adult rats.
receptor tone, either through the administration of agonists J Comp Neurol 347: 150–160.
(glycine, D-serine) or via inhibition of the GLYT1 glycine Alanine A, Bourson A, Büttelmann B, Gill R, Heitz M-P, Mutel V et al. transporter, and these studies have yielded encouraging pre- (2003). 1-Benzyloxy-4,5-dihydro-1H-imidazol-2-yl-amines, a novel liminary data (Javitt, 2008). The preferred target NMDAR class of NR1/2B subtype selective NMDA receptor antagonists.
subunit for a novel small molecule positive modulator Bioorg Med Chem Lett 13: 3155–3159.
approach is unclear. In support of NR2B, its transgenic over- Albers GV, Goldstein LB, Hall D, Lesko LM (2001). Aptiganel hydro- chloride in acute ischemic stroke – a randomized controlled trial.
expression resulted in mice with improved learning and JAMA 286: 2673–2682.
memory (Tang et al., 1999), and gene knockout/knockdown Alberts IL, Nadassy K, Wodak SJ (1998). Analysis of zinc binding sites resulted in learning and memory impairments (von Engel- in protein crystal structures. Protein Sci 7: 1700–1716.
hardt et al., 2008). Thus, selective positive modulation Alexander S, Mathie A, Peters J (2008). Guide to receptors and chan- of NR2B-containing NMDARs might represent a novel nels (GRAC), 3rd edn (2008 revision). Br J Pharmacol 153 (Suppl. 2):
therapeutic strategy for the treatment of schizophrenia and, potentially, other cognitive disorders.
Araneda RC, Lan JY, Zheng X, Zukin RS, Bennett MV (1999). Spermine and arcaine block and permeate N-methyl-D-aspartate receptor
channels. Biophys J 76: 2899–2911.
Banke TG, Dravid SM, Traynelis SF (2005). Protons trap NR1/NR2B NMDA receptors in a nonconducting state. J Neurosci 25: 42–51.
Belelli D, Lambert JJ (2005). Neurosteroids: endogenous regulators of By combining biochemical, structural and functional studies, the GABA(A) receptor. Nat Rev Neurosci 6: 565–575.
much progress has been made in recent years in the compre- Benveniste M, Mayer ML (1993). Multiple effects of spermine on British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al N-methyl-D-aspartic acid receptor responses of rat cultured hippoc- two histidine residues underlying high-affinity Zn2+ inhibition of ampal neurones. J Physiol 464: 131–163.
the NMDA receptor. Neuron 23: 171–180.
Berman RM, Cappiello A, Anand A, Oren DA, Heninger GR, Charney Claiborne CF, McCauley JA, Libby BE, Curtis NR, Diggle HJ, DS et al. (2000). Antidepressant effects of ketamine in depressed Kulagowski JJ et al. (2003). Orally efficacious NR2B-selective NMDA patients. Biol Psychiatry 47: 351–354.
receptor antagonists. Bioorg Med Chem Lett 13: 697–700.
Borza I, Bozo E, Barta-Szalai G, Kiss C, Tarkanyi G, Demeter A et al. Cull-Candy SG, Leszkiewicz DN (2004). Role of distinct NMDA recep- (2007). Selective NR1/2B N-methyl-D-aspartate receptor antagonists tor subtypes at central synapses. Sci STKE 2004: re16.
among indole-2-carboxamides and benzimidazole-2-carboxamides.
Curtis NR, Diggle HJ, Kulagowski JJ, London C, Grimwood S, Hutson J Med Chem 50: 901–914.
PH et al. (2003). Novel N1-(benzyl)cinnamamidine derived NR2B Borza I, Domany G (2006). NR2B selective NMDA antagonists: the subtype-selective NMDA receptor antagonists. Bioorg Med Chem Lett evolution of the ifenprodil-type pharmacophore. Curr Top Med Chem 6: 687–695.
Del Dotto P, Pavese N, Gambaccini G, Bernardini S, Metman LV, Borza I, Kolok S, Gere A, Nagy J, Fodor L, Galgoczy K et al. (2006).
Chase TN et al. (2001). Intravenous amantadine improves levadopa- Benzimidazole-2-carboxamides as novel NR2B selective NMDA induced dyskinesias: an acute double-blind placebo-controlled receptor antagonists. Bioorg Med Chem Lett 16: 4638–4640.
study. Mov Disord 16: 515–520.
Bowlby MR (1993). Pregnenolone sulfate potentiation of N-methyl- Dingledine R, Borges K, Bowie D, Traynelis SF (1999). The glutamate D-aspartate receptor channels in hippocampal-neurons. Mol receptor ion channels. Pharmacol Rev 51: 7–61.
Pharmacol 43: 813–819.
von Engelhardt J, Coserea I, Pawlak V, Fuchs EC, Kohr G, Seeburg PH Boyce S, Wyatt A, Webb JK, O'Donnell R, Mason G, Rigby M et al. et al. (2007). Excitotoxicity in vitro by NR2A- and NR2B-containing (1999). Selective NMDA NR2B antagonists induce antinociception NMDA receptors. Neuropharmacology 53: 10–17.
without motor dysfunction: correlation with restricted localisation von Engelhardt J, Doganci B, Jensen V, Hvalby O, Gongrich C, Taylor of NR2B subunit in dorsal horn. Neuropharmacology 38: 611–623.
Boyce SRRC, Wheeldon A, Rupniak NM, Hill RG (2002). Antinocice- hippocampal NR2B-containing NMDA receptors to performance on ptive activity of the NMDA NR2B receptor subtype selective antago- spatial learning tasks. Neuron 60: 846–860.
nist CP-101,606 in a new rat visceral pain assay. In IASP Meeting San Farkas S, Horvath C, Galgoczy L, Felmerai E, Karsai E, Saghhy K et al. Diego, Abstract ID: 848–P118.
(2003). RGH-896 is a novel potent and selective NR2B-NMDA Büttelmann B, Alanine A, Bourson A, Gill R, Heitz M-P, Mutel V et al. antagonist with efficacy in neuropathic pain models. In Society for (2003). 2-(3,4-Dihydro-1H-isoquinolin-2yl)-pyridines as a novel Neuroscience Meeting, Program # 382.8.
class of NR1/2B subtype selective NMDA receptor antagonists.
Farrant M, Feldmeyer D, Takahashi T, Cullcandy SG (1994). NMDA- Bioorg Med Chem Lett 13: 829–832.
receptor channel diversity in the developing cerebellum. Nature Carron C, Jullien A, Bucher B (1971). Synthesis and pharmacological properties of a series of 2-piperidino alkanol derivatives. Arzneimit- Fayyazuddin A, Villarroel A, Le Goff A, Lerma J, Neyton J (2000). Four telforschung 21: 1992–1998.
residues of the extracellular N-terminal domain of the NR2A Carter C, Benavides J, Legendre P, Vincent JD, Noel F, Thuret F et al. subunit control high-affinity Zn2+ binding to NMDA receptors.
(1988). Ifenprodil and SL 82.0715 as cerebral anti-ischemic agents.
Neuron 25: 683–694.
II. Evidence for N-methyl-D-aspartate receptor antagonist proper- Fischer G, Mutel V, Trube G, Malherbe P, Kew JN, Mohacsi E et al. ties. J Pharmacol Exp Ther 247: 1222–1232.
(1997). Ro 25-6981, a highly potent and selective blocker of Chang HR, Kuo CC (2008). The activation gate and gating mechanism N-methyl-D-aspartate receptors containing the NR2B subunit.
of the NMDA receptor. J Neurosci 28: 1546–1556.
Characterization in vitro. J Pharmacol Exp Ther 283: 1285–1292.
Chaperon F, Muller W, Auberson YP, Tricklebank MD, Neijt HC Furukawa H, Gouaux E (2003). Mechanisms of activation, inhibition (2003). Substitution for PCP, disruption of prepulse inhibition and and specificity: crystal structures of the NMDA receptor NR1 ligand- hyperactivity induced by N-methyl-D-aspartate receptor antago- binding core. Embo J 22: 2873–2885.
nists: preferential involvement of the NR2B rather than NR2A Furukawa H, Singh SK, Mancusso R, Gouaux E (2005). Subunit subunit. Behav Pharmacol 14: 477–487.
arrangement and function in NMDA receptors. Nature 438: 185–
Chazot PL (2004). The NMDA receptor NR2B subunit: a valid thera- peutic target for multiple CNS pathologies. Curr Med Chem 11:
Gallagher MJ, Huang H, Pritchett DB, Lynch DR (1996). Interactions between ifenprodil and the NR2B subunit of the N-methyl-D- Chen M, Lu TJ, Chen XJ, Zhou Y, Chen Q, Feng XY et al. (2008).
aspartate receptor. J Biol Chem 271: 9603–9611.
Differential roles of NMDA receptor subtypes in ischemic neuronal Gielen M, Le Goff A, Stroebel D, Johnson JW, Neyton J, Paoletti P cell death and ischemic tolerance. Stroke 39: 3042–3048.
(2008). Structural rearrangements of NR1/NR2A NMDA receptors Chen N, Moshaver A, Raymond LA (1997). Differential sensitivity of during allosteric inhibition. Neuron 57: 80–93.
recombinant N-methyl-D-aspartate receptor subtypes to zinc Giffard RG, Monyer H, Christine CW, Choi DW (1990). Acidosis inhibition. Mol Pharmacol 51: 1015–1023.
reduces NMDA receptor activation, glutamate neurotoxicity, and Chenard BL, Bordner J, Butler TW, Chambers LK, Collins MA, Decosta oxygen-glucose deprivation neuronal injury in cortical cultures.
Brain Res 506: 339–342.
phenylpiperidino)-1-propanol – a potent new neuroprotectant Gill R, Alanine A, Bourson A, Buttelmann B, Fischer G, Heitz MP et al. which blocks N-methyl-D-aspartate responses. J Med Chem 38:
(2002). Pharmacological characterization of Ro 63-1908 (1-[2- Chenard BL, Menniti FS (1999). Antagonists selective for NMDA novel subtype-selective N-methyl-D-aspartate antagonist. J Pharma- receptors containing the NR2B subunit. Curr Pharm Des 5: 381–404.
col Exp Ther 302: 940–948.
Chizh BA (2007). Low dose ketamine: a therapeutic and research tool Gogas KR (2006). Glutamate-based therapeutic approaches: NR2B receptor antagonists. Curr Opin Pharmacol 6: 68–74.
plasticity in pain pathways. J Psychopharmacol 21: 259–271.
Guscott MR, Clarke HF, Murray F, Grimwood S, Bristow LJ, Hutson PH Chizh BA, Headley PM, Tzschentke TM (2001). NMDA receptor (2003). The effect of (+/-)-CP-101,606, an NMDA receptor NR2B antagonists as analgesics: focus on the NR2B subtype. Trends subunit selective antagonist, in the Morris watermaze. Eur J Pharmacol Sci 22: 636–642.
Pharmacol 476: 193–199.
Choi YB, Lipton SA (1999). Identification and mechanism of action of Hahn CG, Wang HY, Cho DS, Talbot K, Gur RE, Berrettini WH et al. British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
(2006). Altered neuregulin 1-erbB4 signaling contributes to NMDA (2000). Glycine antagonist (gavestinel) in neuroprotection (GAIN receptor hypofunction in schizophrenia. Nat Med 12: 824–828.
International) in patients with acute stroke: a randomised Han X, Tomitori H, Mizuno S, Higashi K, Full C, Fukiwake T et al. controlled trial. GAIN International Investigators. Lancet 355:
(2008). Binding of spermine and ifenprodil to a purified, soluble regulatory domain of the N-methyl-d-aspartate receptor. J Neuro- Lerma J (1992). Spermine regulates N-methyl-D-aspartate receptor chem 107: 1566–1577.
desensitization. Neuron 8: 343–352.
Hardingham GE, Bading H (2003). The Yin and Yang of NMDA recep- Li L, Fan M, Icton CD, Chen N, Leavitt BR, Hayden MR et al. (2003).
tor signalling. Trends Neurosci 26: 81–89.
Role of NR2B-type NMDA receptors in selective neurodegeneration Hatton CJ, Paoletti P (2005). Modulation of triheteromeric NMDA in Huntington disease. Neurobiol Aging 24: 1113–1121.
receptors by N-terminal domain ligands. Neuron 46: 261–274.
Liu Y, Wong TP, Aarts M, Rooyakkers A, Liu L, Lai TW et al. (2007).
Herin GA, Aizenman E (2004). Amino terminal domain regulation of NMDA receptor subunits have differential roles in mediating exci- NMDA receptor function. Eur J Pharmacol 500: 101–111.
totoxic neuronal death both in vitro and in vivo. J Neurosci 27:
Higgins GA, Ballard TM, Enderlin M, Haman M, Kemp JA (2005).
Evidence for improved performance in cognitive tasks following Lopez de Armentia M, Sah P (2003). Development and subunit com- selective NR2B NMDA receptor antagonist pre-treatment in the rat.
position of synaptic NMDA receptors in the amygdale: NR2B syn- Psychopharmacology (Berl) 179: 85–98.
apses in the adult central amygdala. J Neurosci 23: 6876–6883.
Horak M, Vlcek K, Chodounska H, Vyklicky L (2006). Subtype- Loschmann PA, De Groote C, Smith L, Wullner U, Fischer G, Kemp JA dependence of N-methyl-D-aspartate receptor modulation by et al. (2004). Antiparkinsonian activity of Ro 25-6981, a NR2B pregnenolone sulfate. Neuroscience 137: 93–102.
subunit specific NMDA receptor antagonist, in animal models of Horak M, Vlcek K, Petrovic M, Chodounska H, Vyklicky L (2004).
Parkinson's disease. Exp Neurol 187: 86–93.
Molecular mechanism of pregnenolone sulfate action at NR1/NR2B Low CM, Lyuboslavsky P, French A, Le P, Wyatte K, Thiel WH et al. receptors. J Neurosci 24: 10318–10325.
(2003). Molecular determinants of proton-sensitive N-methyl-D- Horvath C (2004). 3rd World Cong. World Inst. Pain: Barcelona.
aspartate receptor gating. Mol Pharmacol 63: 1212–1222.
September 21–25.
Low CM, Zheng F, Lyuboslavsky P, Traynelis SF (2000). Molecular Hosie AM, Wilkins ME, da Silva HM, Smart TG (2006). Endogenous determinants of coordinated proton and zinc inhibition of neurosteroids regulate GABAA receptors through two discrete trans- N-methyl-D-aspartate NR1/NR2A receptors. Proc Natl Acad Sci USA membrane sites. Nature 444: 486–489.
Huggins DJ, Grant GH (2005). The function of the amino terminal Lu WY, Xiong ZG, Orser BA, MacDonald JF (1998). Multiple sites of domain in NMDA receptor modulation. J Mol Graph Model 23:
action of neomycin, Mg2+ and spermine on the NMDA receptors of rat hippocampal CA1 pyramidal neurones. J Physiol 512 (Pt 1):
Jang MK, Mierke DF, Russek SJ, Farb DH (2004). A steroid modulatory domain on NR2B controls N-methyl-D-aspartate receptor proton Luo J, Wang Y, Yasuda RP, Dunah AW, Wolfe BB (1997). The majority sensitivity. Proc Natl Acad Sci USA 101: 8198–8203.
of N-methyl-D-aspartate receptor complexes in adult rat cerebral Javitt DC (2008). Glycine transport inhibitors and the treatment of cortex contain at least three different subunits (NR1/NR2A/NR2B).
schizophrenia. Biol Psychiatry 63: 6–8.
Mol Pharmacol 51: 79–86.
Kashiwagi K, Fukuchi J, Chao J, Igarashi K, Williams K (1996). An Ma QP, Hargreaves RJ (2000). Localization of N-methyl-D-aspartate aspartate residue in the extracellular loop of the N-methyl-D- NR2B subunits on primary sensory neurons that give rise to small- aspartate receptor controls sensitivity to spermine and protons. Mol caliber sciatic nerve fibers in rats. Neuroscience 101: 699–707.
Pharmacol 49: 1131–1141.
McCauley JA (2007). NR2B subtype-selective NMDA receptor antago- Kashiwagi K, Pahk AJ, Masuko T, Igarashi K, Williams K (1997). Block nists. In 3rd Anglo-Swedish Medicinal Chemistry Symposium. Are, and modulation of N-methyl-D-aspartate receptors by polyamines and protons: role of amino acid residues in the transmembrane and McCauley JA, Bednar RA, Bednar B, Butcher JW, Claiborne CF, Clar- pore-forming regions of NR1 and NR2 subunits. Mol Pharmacol 52:
emon DA et al. (2008). NR2B subtype-selective NMDA receptor antagonists. In Abstracts – 236th ACS National Meeting. Philadelphia, Kemp JA, Kew JNC, Gill R (1999). NMDA receptor antagonists and PA, August 17–21, 2008.
their potential as neuroprotective agents. In: Jonas P, Monyer H McGurk JF, Bennett MV, Zukin RS (1990). Polyamines potentiate (eds). Handbook of Experimental Pharmacology Volume 141. Ionotropic responses of N-methyl-D-aspartate receptors expressed in xenopus Glutamate Receptors in the CNS. Springer: Berlin, pp. 495–527.
oocytes. Proc Natl Acad Sci USA 87: 9971–9974.
Kew JN, Kemp JA (1998). An allosteric interaction between the NMDA Maeng S, Zarate CA Jr (2007). The role of glutamate in mood disorders: receptor polyamine and ifenprodil sites in rat cultured cortical results from the ketamine in major depression study and the pre- neurones. J Physiol 512 (Pt 1): 17–28.
sumed cellular mechanism underlying its antidepressant effects.
Kew JN, Trube G, Kemp JA (1996). A novel mechanism of activity- Curr Psychiatry Rep 9: 467–474.
dependent NMDA receptor antagonism describes the effect of Makani S, Chesler M (2007). Endogenous alkaline transients boost ifenprodil in rat cultured cortical neurones. J Physiol 497 (Pt 3):
postsynaptic NMDA receptor responses in hippocampal CA1 pyra- midal neurons. J Neurosci 27: 7438–7446.
Kiss L, Cheng G, Bednar B, Bednar RA, Bennett PB, Kane SA et al. Malayev A, Gibbs TT, Farb DH (2002). Inhibition of the NMDA (2005). In vitro characterization of novel NR2B selective NMDA response by pregnenolone sulphate reveals subtype selective modu- receptor antagonists. Neurochem Int 46: 453–464.
lation of NMDA receptors by sulphated steroids. Br J Pharmacol 135:
Köhr G (2006). NMDA receptor function: subunit composition versus spatial distribution. Cell Tissue Res 326: 439–446.
Malherbe P, Mutel V, Broger C, Perin-Dureau F, Kemp JA, Neyton J Krystal JH, Abi-Saab W, Perry E, D'Souza DC, Liu N, Gueorguieva R et al. (2003). Identification of critical residues in the amino terminal et al. (2005). Preliminary evidence of attenuation of the disruptive domain of the human NR2B subunit involved in the RO 25-6981 effects of the NMDA glutamate receptor antagonist, ketamine, on binding pocket. J Pharmacol Exp Ther 307: 897–905.
working memory by pretreatment with the group II metabotropic Martel MA, Wyllie DJ, Hardingham GE (2009). In developing hippoc- glutamate receptor agonist, LY354740, in healthy human subjects.
ampal neurons, NR2B-containing N-methyl-d-aspartate receptors Psychopharmacology (Berl) 179: 303–309.
(NMDARs) can mediate signaling to neuronal survival and synaptic Lees KR, Asplund K, Carolei A, Davis SM, Diener HC, Kaste M et al. potentiation, as well as neuronal death. Neuroscience 158: 334–343.
British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al Masuko T, Kashiwagi K, Kuno T, Nguyen ND, Pahk AJ, Fukuchi J et al. N-terminal modulatory domain in a NMDA receptor subunit.
(1999). A regulatory domain (R1-R2) in the amino terminus of the Neuron 28: 911–925.
N-methyl-D-aspartate receptor: effects of spermine, protons, and Paoletti P, Vergnano AM, Barbour B, Casado M (2009). Zinc at ifenprodil, and structural similarity to bacterial leucine/isoleucine/ glutamatergic synapses. Neuroscience 158: 126–136.
valine binding protein. Mol Pharmacol 55: 957–969.
Park-Chung M, Wu FS, Purdy RH, Malayev AA, Gibbs TT, Farb DH Mayer ML (2006). Glutamate receptors at atomic resolution. Nature (1997). Distinct sites for inverse modulation of N-methyl-D- aspartate receptors by sulfated steroids. Mol Pharmacol 52: 1113–
Merchant RE, Bullock MR, Carmack CA, Shah AK, Wilner KD, Ko G et al. (1999). A double-blind, placebo-controlled study of the safety, tolerability and pharmacokinetics of CP-101,606 in patients with a beta-pregnan-20-one sulfate – a negative modulator of the mild or moderate traumatic brain injury. Ann N Y Acad Sci 890:
NMDA-induced current in cultured neurons. Mol Pharmacol 46:
Montastruc JL, Rascol O, Senard JM, Rascol A (1992). A pilot study of Parsons CG, Stoffler A, Danysz W (2007). Memantine: a NMDA recep- N-methyl-D-aspartate (NMDA) antagonist in Parkinson's disease.
tor antagonist that improves memory by restoration of homeostasis J Neurol Neurosurg Psychiatry 55: 630–631.
in the glutamatergic system – too little activation is bad, too much Mony L, Krzaczkowski L, Leonetti M, Goff AL, Alarcon K, Neyton J is even worse. Neuropharmacology 53: 699–723.
et al. (2009). Structural basis of NR2B-selective antagonist recogni- Perin-Dureau F, Rachline J, Neyton J, Paoletti P (2002). Mapping the tion by N-methyl-D-aspartate receptors. Mol Pharmacol 75: 60–
binding site of the neuroprotectant ifenprodil on NMDA receptors.
J Neurosci 22: 5955–5965.
Monyer H, Burnashev N, Laurie DJ, Sakmann B, Seeburg PH (1994).
Petrovic M, Sedlacek M, Horak M, Chodounska H, Vyklicky L (2005).
Developmental and regional expression in the rat brain and func- 20-oxo-5 beta-pregnan-3 alpha-yl sulfate is a use-dependent NMDA tional properties of four NMDA receptors. Neuron 12: 529–540.
receptor inhibitor. J Neurosci 25: 8439–8450.
Morris GF, Bullock R, Marshall SB, Marmarou A, Maas A, Marshall LF Pinheiro PS, Mulle C (2008). Presynaptic glutamate receptors: physi- (1999). Failure of the competitive N-methyl-D-aspartate antagonist ological functions and mechanisms of action. Nat Rev Neurosci 9:
Selfotel (CGS 19755) in the treatment of severe head injury: results of two phase III clinical trials. J Neurosurg 91: 737–743.
Preskorn SH, Baker B, Kolluri S, Menniti FS, Krams M, Landen JW Mott DD, Doherty JJ, Zhang S, Washburn MS, Fendley MJ, Lyubo- (2008). An innovative design to establish proof of concept of the slavsky P et al. (1998). Phenylethanolamines inhibit NMDA recep- antidepressant effects of the NR2B subunit selective N-methyl-D- tors by enhancing proton inhibition. Nat Neurosci 1: 659–667.
aspartate antagonist, CP-101,606, in patients with treatment- Neyton J, Paoletti P (2006). Relating NMDA receptor function to refractory major depressive disorder. J Clin Psychopharmacol 28:
receptor subunit composition: limitations of the pharmacological approach. J Neurosci 26: 1331–1333.
Rachline J, Perin-Dureau F, Goff AL, Neyton J, Paoletti P (2005). The Ng FM, Geballe MT, Snyder JP, Traynelis SF, Low CM (2008). Structural micromolar zinc-binding domain on the NMDA receptor subunit insights into phenylethanolamines high-affinity binding site in NR2B. J Neurosci 25: 308–317.
NR2B from binding and molecular modeling studies. Mol Brain 1:
Robel P, Baulieu EE (1994). Neurosteroids biosynthesis and function.
Trends Endocrinol Metab 5: 1–8.
Ng FM, Soh W, Geballe MT, Low CM (2007). Improving solubility of Rock DM, MacDonald RL (1992). Spermine and related polyamines NR2B amino-terminal domain of N-methyl-d-aspartate receptor produce a voltage-dependent reduction of N-methyl-D-aspartate expressed in Escherichia coli. Biochem Biophys Res Commun 362:
receptor single-channel conductance. Mol Pharmacol 42: 157–164.
Rock DM, Macdonald RL (1995). Polyamine regulation of N-methyl- Nicholson KL, Mansbach RS, Menniti FS, Balster RL (2007). The D-aspartate receptor channels. Annu Rev Pharmacol Toxicol 35: 463–
phencyclidine-like discriminative stimulus effects and reinforcing properties of the NR2B-selective N-methyl-D-aspartate antagonist Rumbaugh G, Prybylowski K, Wang JF, Vicini S (2000). Exon 5 and CP-101 606 in rats and rhesus monkeys. Behav Pharmacol 18: 731–
spermine regulate deactivation of NMDA receptor subtypes. J Neu- rophysiol 83: 1300–1306.
Nikam SS, Meltzer LT (2002). NR2B selective NMDA receptor antago- Saltarelli MD, Weaver JJ, Hsu C, Bednar MM (2004). Randomized nists. Curr Pharm Des 8: 845–855.
double-blind, placebo-controlled study to evaluate the safety and Nutt JG, Gunzler SA, Kirchhoff T, Hogarth P, Weaver JL, Krams M et al. efficacy of CP-101,606 (traxoprodil), an NR2B-selective N-methyl (2008). Effects of a NR2B selective NMDA glutamate antagonist, D-aspartate receptor antagonist, in subjects with acute ischemic CP-101,606, on dyskinesia and Parkinsonism. Mov Disord 23: 1860–
stroke. Stroke 35: 241–241.
Sang CN, Weaver JJ, Jinga L, Wouden J, Saltarelli MD (2003). The O'Hara PJ, Sheppard PO, Thogersen H, Venezia D, Haldeman BA, NR2B subunit-selective NMDA receptor antagonist CP-101,606 McGrane V et al. (1993). The ligand-binding domain in metabotro- reduces pain intensity in patients with central and peripheral pic glutamate receptors is related to bacterial periplasmic binding- neuropathic pain. In Society for Neuroscience, Meeting. Program # proteins. Neuron 11: 41–52.
Pahk AJ, Williams K (1997). Influence of extracellular pH on inhibi- Scimemi A, Fine A, Kullmann DM, Rusakov DA (2004). NR2B- tion by ifenprodil at N-methyl-D-aspartate receptors in Xenopus containing receptors mediate cross talk among hippocampal syn- oocytes. Neurosci Lett 225: 29–32.
apses. J Neurosci 24: 4767–4777.
Paoletti P, Ascher P, Neyton J (1997). High-affinity zinc inhibition of Silver RA, Traynelis SF, Cull-Candy SG (1992). Rapid-time-course min- NMDA NR1-NR2A receptors. J Neurosci 17: 5711–5725.
iature and evoked excitatory currents at cerebellar synapses in situ.
Paoletti P, Neyton J (2007). NMDA receptor subunits: function and Nature 355: 163–166.
pharmacology. Curr Opin Pharmacol 7: 39–47.
Steece-Collier K, Chambers LK, Jaw-Tsai SS, Menniti FS, Greenamyre Paoletti P, Neyton J, Ascher P (1995). Glycine-independent and JT (2000). Antiparkinsonian actions of CP-101,606, an antagonist of subunit-specific potentiation of NMDA responses by extracellular NR2B subunit-containing N-methyl-d-aspartate receptors. Exp Mg2+. Neuron 15: 1109–1120.
Neurol 163: 239–243.
Paoletti P, Perin-Dureau F, Fayyazuddin A, Goff AL, Callebaut I, Suetake-Koga S, Shimazaki T, Takamori K, Chaki S, Kanuma K, Sekigu- Neyton J (2000). Molecular organization of a zinc binding chi Y et al. (2006). In vitro and antinociceptive profile of HON0001, British Journal of Pharmacology (2009) 157 1301–1317
Allosteric modulators of NR2B-containing NMDA receptors
L Mony et al
an orally active NMDA receptor NR2B subunit antagonist. Pharma- Weaver CE, Land MB, Purdy RH, Richards KG, Gibbs TT, Farb DH col Biochem Behav 84: 134–141.
(2000). Geometry and charge determine pharmacological effects of Sun Y, Olson R, Horning M, Armstrong N, Mayer M, Gouaux E (2002).
steroids on N-methyl-D-aspartate receptor-induced Ca2+ accumu- Mechanism of glutamate receptor desensitization. Nature 417: 245–
lation and cell death. J Pharm Exp Ther 293: 747–754.
Wei F, Wang GD, Kerchner GA, Kim SJ, Xu HM, Chen ZF et al. (2001).
Tahirovic YA, Geballe M, Gruszecka-Kowalik E, Myers SJ, Lyuboslavsky Genetic enhancement of inflammatory pain by forebrain NR2B P, Le P et al. (2008). Enantiomeric propanolamines as selective overexpression. Nat Neurosci 4: 164–169.
N-methyl-D-aspartate 2B receptor antagonists. J Med Chem 51:
Wessell RH, Ahmed SM, Menniti FS, Dunbar GL, Chase TN, Oh JD (2004). NR2B selective NMDA receptor antagonist CP-101,606 pre- Tamiz AP, Cai SX, Zhou Z-L, Yuen P-W, Shelkun RM, Whittemore ER vents levodopa-induced motor response alterations in hemi- et al. (1999). Structure-activity relationship of N(Phenylalkyl)cinna- parkinsonian rats. Neuropharm 47: 184–194.
mides as novel NR2B subtype-selective NMDA receptor antagonists.
Williams K (1993). Ifenprodil discriminates subtypes of the N-methyl- J Med Chem 42: 3412–3420.
D-aspartate receptor-selectivity and mechanisms at recombinant Tan PH, Yang LC, Shih HC, Lan KC, Cheng JT (2005). Gene knock- heteromeric receptors. Mol Pharmacol 44: 851–859.
down with intrathecal siRNA of NMDA receptor NR2B subunit Williams K (1994a). Mechanisms influencing stimulatory effects of reduces formalin-induced nociception in the rat. Gene Ther 12:
spermine at recombinant N-methyl-D-aspartate receptors. Mol Phar- macol 46: 161–168.
Tang CM, Dichter M, Morad M (1990). Modulation of the N-methyl- Williams K (1994b). Subunit-specific potentiation of recombinant D-aspartate channel by extracellular H+. Proc Natl Acad Sci USA 87:
N-methyl-D-aspartate receptors by histamine. Mol Pharmacol 46:
Tang YP, Shimizu E, Dube GR, Rampon C, Kerchner GA, Zhuo M et al. Williams K, Kashiwagi K, Fukuchi J, Igarashi K (1995). An acidic amino (1999). Genetic enhancement of learning and memory in mice.
acid in the N-methyl-D-aspartate receptor that is important for Nature 401: 63–69.
spermine stimulation. Mol Pharmacol 48: 1087–1098.
Taniguchi K, Shinjo K, Mizutani M, Shimada K, Ishikawa T, Menniti FS Williams K, Zappia AM, Pritchett DB, Shen YM, Molinoff PB (1994).
et al. (1997). Antinociceptive activity of CP-101,606, an NMDA Sensitivity of the N-methyl-D-aspartate receptor to polyamines is receptor NR2B subunit antagonist. Br J Pharmacol 122: 809–
controlled by NR2 subunits. Mol Pharmacol 45: 803–809.
Wilson AW, Medhurst SJ, Dixon CI, Bontoft NC, Winyard LA, Brack- Traynelis SF, Burgess MF, Zheng F, Lyuboslavsky P, Powers JL (1998).
enborough KT et al. (2006). An animal model of chronic inflamma- Control of voltage-independent zinc inhibition of NMDA receptors tory pain: pharmacological and temporal differentiation from acute by the NR1 subunit. J Neurosci 18: 6163–6175.
models. Eur J Pain 10: 537–549.
Traynelis SF, Cull-Candy SG (1990). Proton inhibition of N-methyl- Wong E, Ng FM, Yu CY, Lim P, Lim LH, Traynelis SF et al. (2005).
D-aspartate receptors in cerebellar neurons. Nature 345: 347–350.
Expression and characterization of soluble amino-terminal domain Traynelis SF, Hartley M, Heinemann SF (1995). Control of proton of NR2B subunit of N-methyl-D-aspartate receptor. Protein Sci 14:
sensitivity of the NMDA receptor by RNA splicing and polyamines.
Science 268: 873–876.
Woo TUW, Kim AM, Viscidi E (2008). Disease-specific alterations in Trube G, Ehrhard P, Malherbe P, Huber G (1996). The selectivity of glutamatergic neurotransmission on inhibitory interneurons in the Ro25-6981 for NMDA receptor subtypes expressed in Xenopus prefrontal cortex in schizophrenia. Brain Res 1218: 267–277.
oocytes. Soc Neurosci Abstr 22: 693–694.
Woodhall G, Evans DI, Cunningham MO, Jones RS (2001). NR2B- Umbricht D, Krljes S (2005). Mismatch negativity in schizophrenia: a containing NMDA autoreceptors at synapses on entorhinal cortical meta-analysis. Schizophr Res 76: 1–23.
neurons. J Neurophysiol 86: 1644–1651.
Vogt K, Mellor J, Tong G, Nicoll R (2000). The actions of synaptically Wu FS, Gibbs TT, Farb DH (1991). Pregnenolone sulfate – a positive released zinc at hippocampal mossy fiber synapses. Neuron 26: 187–
allosteric modulator at the N-methyl-D-aspartate receptor. Mol Pharmacol 40: 333–336.
Vyklicky L Jr, Vlachova V, Krusek J (1990). The effect of external pH Zarate CA Jr, Singh JB, Quiroz JA, De Jesus G, Denicoff KK, Lucken- changes on responses to excitatory amino acids in mouse hippoc- baugh DA et al. (2006). A double-blind, placebo-controlled study of ampal neurones. J Physiol 430: 497–517.
memantine in the treatment of major depression. Am J Psychiatry Wang LY, MacDonald JF (1995). Modulation by magnesium of the affinity of NMDA receptors for glycine in murine hippocampal Zeron MM, Hansson O, Chen N, Wellington CL, Leavitt BR, Brundin neurones. J Physiol 486: 83–95.
P et al. (2002). Increased sensitivity to N-methyl-D-aspartate Watanabe M, Inoue Y, Sakimura K, Mishina M (1992). Developmental receptor-mediated excitotoxicity in a mouse model of Huntington's changes in distribution of NMDA receptor channel subunit mRNAs.
disease. Neuron 33: 849–860.
Neuroreport 3: 1138–1140.
Zhang L, Zheng X, Paupard MC, Wang AP, Santchi L, Friedman LK Watanabe M, Inoue Y, Sakimura K, Mishina M (1993). Distinct distri- et al. (1994). Spermine potentiation of recombinant N-methyl-D- butions of five N-methyl-D-aspartate receptor channel subunit aspartate receptors is affected by subunit composition. Proc Natl mRNAs in the forebrain. J Comp Neurol 338: 377–390.
Acad Sci USA 91: 10883–10887.
Watanabe M, Mishina M, Inoue Y (1994). Distinct spatiotemporal Zhou M, Baudry M (2006). Developmental changes in NMDA neuro- distributions of the N-methyl-D-aspartate receptor channel subunit toxicity reflect developmental changes in subunit composition of mRNAs in the mouse cervical cord. J Comp Neurol 345: 314–319.
NMDA receptors. J Neurosci 26: 2956–2963.
British Journal of Pharmacology (2009) 157 1301–1317

Source: http://mony.fr/publis/Mony_BJP_2009.pdf