H2NOX

Panic Away

Panic Away End Anxiety and Panic Attacks

Get Instant Access

FIGURE 15.2 Structures of some GABA-AT inhibitors, compound 15.6 being inactive.

(Figure 15.2), a suicide inhibitor for the enzyme GABA-aminotransferase (GABA-AT) responsible for GABA degradation, is used clinically as an anticonvulsant. Elevation in extracellular GABA levels by the inhibition of the reuptake of GABA is effected by Tiagabine (15.20) (refer to Figure 15.4) marketed for the treatment of epilepsy and in preclinical studies for treatment of anxiety and insomnia. The G-protein-coupled GABAb receptor is the target for the antispastic drug baclofen (15.41) (refer to Figure 15.9).

Drugs targeting the Glu neurotransmitter system have been slower to emerge. Memantine (see Section 15.7.3) is used with some success for treatment of Alzheimer's disease and a few compounds with mixed mechanisms of action, including reduction of Glu release (through blockade of Na channels) are used for the treatment of migraine, epilepsy, and amyotrophic lateral sclerosis.

All of the Glu and GABA receptors and transporters are heterogeneous and may individually be involved in specific CNS disorders and disease conditions. These receptor/transporter subtypes are unevenly distributed in the CNS, which opens up the prospect of developing ligands selective for receptor/transporter subtypes with predominant location in different brain regions of therapeutic relevance.

15.3 GABA BIOSYNTHESIS AND METABOLISM

The GABA concentration is regulated by two pyridoxal 5'-phosphate (PLP) dependent enzymes, L-glutamic acid decarboxylase (GAD), which catalyzes the decarboxylation of Glu to GABA prior to release into the synaptic cleft, and GABA-AT, which degrades reuptaken GABA to SSA (Figure 15.1). This transamination step takes place within presynaptic GABA terminals as well as in surrounding glia cells.

15.3.1 Inhibitors of GABA Metabolism

A number of mechanism-based inactivators of GABA-AT has been developed and has been shown to elevate the extracellular levels of GABA. These compounds are typically analogs of GABA, containing appropriate functional groups at C4 of the GABA backbone (e.g., 15.3, 15.4, and 15.5). The functional group is converted by GABA-AT into electrophiles, which react with nucleophilic groups at or near the active site of the enzyme and thereby inactivate the enzyme irreversibly (Figure 15.3). The most effective of these, y-vinyl-GABA (Vigabatrin, 15.3), is clinically used as an anticonvulsant for the treatment of epilepsy.

The mechanism for inactivation of GABA-AT by Vigabatrin (15.3) is outlined in Figure 15.3. As shown, PLP is forming a Schiff base (15.8) with the terminal amino group of a lysine residue in the active site of GABA-AT. Transamination with 15.3 generates a new imine 15.9, which undergoes a rate-determining enzyme-catalyzed deprotonation to give the imine 15.11 after reprotonation. In analogy with transamination reaction on GABA, 15.11 could be hydrolyzed to give the SAA analog 15.13 and pyridoxamine-5-phosphate 15.12. However, 15.11 is a Michael acceptor electrophile, which undergoes conjugate addition by an active-site nucleophile X and the inactivated enzyme 15.14 is produced.

To optimize the effect of Vigabatrin a number of conformationally restricted GABA analogs has been developed. In contrast to the Vigabatrin analog 15.6, which does not inactivate GABA-AT, the

enz^

Enz oN

Inactivation ^

Inactivation ^

Vigabatrin Mechanism

FIGURE 15.3 Proposed inactivation mechanism of GABA-AT by Vigabatrin (15.3). The cofactor PLP and an amino group from a lysine residue in GABA-AT (Enz) form a Schiff base (15.8), which reacts with Vigabatrin and eventually leads to inactivation of GABA-AT.

FIGURE 15.3 Proposed inactivation mechanism of GABA-AT by Vigabatrin (15.3). The cofactor PLP and an amino group from a lysine residue in GABA-AT (Enz) form a Schiff base (15.8), which reacts with Vigabatrin and eventually leads to inactivation of GABA-AT.

difluoromethylene analog 15.7 (Figure 15.2), is reported as a markedly more potent inactivator of GABA-AT than Vigabatrin.

15.4 GABA TRANSPORT

The GABA transporters belong to the family of Na+/Cl- dependent transporters (SLC-6 gene family) that also include transporters for the neurotransmitters dopamine, serotonin, norepinephrine, and glycine (see Chapter 14). Four subtypes of GABA transporters have been identified in the mama-lian CNS. For rat and human GABA transporters, the nomenclature is GAT-1, betaine/GABA-transporter-1 (BGT-1), GAT-2, and GAT-3.

15.4.1 Inhibitors of GABA Transport

The pharmacological inhibition of GABA transporters constitutes an attractive approach to increase the overall GABA neurotransmission. A selective blockade of glial uptake is believed to be optimal, as this will ensure an elevation of the GABA level in the presynaptic nerve terminals.

Nipecotic acid (15.15) and guvacine (15.16), competitive inhibitors and substrates for the GABA uptake, have been important lead structures for the development of a large number of lipophilic GABA uptake inhibitors. Introduction of a lipophilic moiety, such as 4,4-diphenyl-3-butenyl (DPB), on the nitrogen atom led to N-DPB-nipecotic acid (15.19) and related analogs, which are markedly more potent than the parent amino acids. These lipophilic compounds are able to cross the blood-brain hn^oh hqA0h

NH2 oh H3C-NH OH

Nipecotic acid (15.15) Guvacine (15.16) exo-THPO (15.17) N-Me-exo-THPO (15.18)

W-DPB-nipecotic acid (15.19)

Tiagabine (15.20)

Tiagabine (15.20)

EF-1502 (15.21)

FIGURE 15.4 Structures of some GABA transport inhibitors.

barrier and are potent anticonvulsants in animal models. Tiagabine (15.20), a structurally related compound, is now marketed as an add-on therapeutic agent for the treatment of epilepsy.

A highly glia-selective compound was discovered based on the structure of exo-THPO (15.17), where the monomethylated compound N-methyl-exo-THPO (15.18) proved to be the most selective inhibitor for glial vs. neuronal GABA uptake reported yet.

EF-1502 (15.21), developed as a hybrid of exo-THPO and Tiagabine, has similar potency at GAT-1 and BGT-1. An in vivo study of the anticonvulsant properties of the compound revealed a synergistic effect between EF-1502 and GAT-1-selective inhibitors, indicating a possible role for BGT-1 as a therapeutic target (Figure 15.4).

15.5 GABA RECEPTORS AND THEIR LIGANDS

GABA exerts its effects on the CNS via two different types of receptors: the ionotropic GABAa and GABAC receptors, mediating the fast synaptic transmission and the G-protein-coupled GABAB receptors, mediating the slower responses to GABA via coupling to second messenger cascades.

15.5.1 Ionotropic GABA Receptors

The ionotropic GABA receptors belong to a superfamily of ligand-gated ion channels (Cys-loop receptors) that also includes the nicotinic acetylcholine, the glycine, and the serotonin (5-HT3) receptors (see Chapter 12). Whether the GABAC receptor is a subgroup of the GABAa receptors or a distinct group of GABA receptors is still a matter of debate. The GABAA receptors are widely distributed in the CNS and involved in a wide variety of CNS functions, whereas the GABAC receptors predominantly are expressed in the retina and primarily implicated in visual processing. However, GABA receptors have also been identified in some CNS regions, where they have been proposed to be involved in processes connected with sleep and cognition processes.

The ionotropic GABA receptors are transmembrane protein complexes composed of five sub-units. So far, 19 human GABA receptor subunits have been identified, and they have been classified into a1-6, p1-3, y1-3, 5, e, n, 8 and p1-3 subunit classes (Figure 15.5). Each of the subunits consists of an amino-terminal domain and a transmembrane region formed by four transmembrane a-helices connected by intra- and extracellular loops. In the pentameric GABA receptor complex, the orthosteric site (i.e., the binding site for the endogenous ligand GABA) is formed at the interface between the terminal domains of two subunits, whereas the transmembrane regions of the subunits form the ion channel pore through which chloride ions can enter the cell upon activation. The GABAA receptors

GABA

GABAC subunits Pl-3

FIGURE 15.5 Schematic illustrations of (A) the pentameric structure of the ionotropic GABA receptors, (B) with indication of the GABA-binding site and the chloride ion channel, (C) and the multiplicity of ionotropic GABAa and GABAC receptors.

are heteromeric complexes, and although a wide range of different receptor combinations exists in vivo, a^2y2 combination is the predominant physiological GABAa receptor subtype. In contrast, the GABAc receptors are homomeric assemblies of five identical p subunits or pseudoheteromeric complexes comprising different p subunits.

The binding site for GABA and ligands for the orthosteric binding has been shown to be located at the interface of the a and P subunits in the GABAA receptor complex, whereas the binding site for BZD is located at the interface of the a and y subunits.

There is still no 3D-structure available for the ionotropic GABA receptors. Thus, the understanding of the molecular architecture of the orthosteric-binding site in the ionotropic GABA receptors has to a large extent been based on the publication of crystals structures of acetylcholine-binding proteins (AChBP) from snails. These proteins display low but still significant amino acid sequence homologies with the amino-terminal domains of all ligand-gated ion channels within the Cys-loop family, including the ionotropic GABA receptors, and this homology has been exploited for the construction of homology models of this region in both GABAA and GABAc receptors. Such homology models offer an insight into the identities of the residues lining the binding pockets in the respective receptors. However, given the low level of sequence identity (~18%) between the AChBP and the ionotropic GABA receptors, it is not straightforward to use these models for the prediction of ligand affinity or structure-activity studies.

15.5.2 Ionotropic GABA Receptor Ligands

The GABA-binding site has very distinct and specific structural requirements for recognition and activation. Thus, only a few different classes of structures have been reported. Within the series of compounds showing agonist activity at the GABAA receptor site are the selective agonists mus-cimol (15.22) and THIP (15.23), which have been used for the pharmacological characterization of the GABAA receptors. BMC (15.24) and SR95531 (15.25) are the classical GABAA receptor antagonists.

In the absence of a 3D-receptor structure, the relationship between the ligand structure and the binding/activity at the GABAA receptor has been extensively studied. On the basis of a

Muscimol (15.22)

Muscimol (15.22)

och3

och3

Isoguvacine (15.27)

Isoguvacine (15.27)

OH O

OH O

FIGURE 15.6 Structures of GABAa and GABAc ligands.

O II

hypothesis originating from the bioactive conformation of muscimol, the partial GABAA agonist 4-PIOL (15.26), and on pharmacological data for an additional series of GABAA ligands, a simple 3D-pharmacophore model for the orthosteric GABAA receptor ligands has been developed. The main features of this model are that the 3-hydroxyisoxazolol rings of muscimol and 4-PIOL do not overlap in their proposed binding modes and that the two compounds interact with different conformations of an arginine residue located at the GABAA recognition site. The space surrounding the ligands has been defined and the existence of a cavity of considerable dimensions in the vicinity of the 4-position of the 3-hydroxyisoxazolol moiety in the structure of 4-PIOL has been identified, whereas the corresponding position in muscimol is identified as "receptor essential volumes" (Figure 15.7). Based on this model, a series of selective and highly potent competitive antagonists have been developed including the compounds 15.32a-d.

In contrast, structure-activity studies of ligands targeting the GABAC receptors have been very limited. cis-4-Aminocrotonic acid (CACA [15.29]) (Figure 15.6) has been the key ligand for the identification of the GABAC receptors. The compound is a moderately potent partial GABAC agonist and inactive at GABAA receptors, but it has been shown to effect GABA transport as well. In the search for selective GABAC receptor ligands, the folded conformation of CACA has been used as a scaffold for new compounds such as cis-2-aminomethyl cyclopropane carboxylic acid (CAMP [15.30]). (+)-CAMP has been reported to be a selective GABAC receptor agonist with potency in the mid-micromolar range, displaying only weak activity on the GABAA receptors. Finally the first antagonist capable of differentiating the GABAC receptors from both GABAA and GABAB receptors was TPMPA (15.31).

FIGURE 15.7 A superimposition of the proposed bioactive conformations of muscimol (15.22, green carbon atoms) and 4-PIOL (15.26, gray carbon atoms) binding to two different conformations of an arginine residue at the orthosteric-binding site. A series of 4-substituted 4-PIOL compounds (15.32a-d) are included illustrating the large space spanned by the 4-substituents. The tetrahedrons indicate receptor excluded volumes.

The overall molecular architecture of the orthosteric sites at the GABAA and GABAC receptors appear to be quite similar as most GABAA agonists display some agonist/antagonist activities at GABAC receptors as well. THIP (15.23), the standard GABAA agonist, has been shown to be a partial agonist at GABAA receptors and a competitive antagonist at GABAC receptors. Likewise, the GABAA agonists muscimol (15.22), isoguvacine (15.27), and imidazol-4-acetic acid (IAA [15.28]) act as partial GABAC agonists. However the fact that GABAA and GABAC receptors exhibit distinct antagonist profiles clearly indicates that orthosteric sites of these receptors are not identical.

Upregulation of GABA activity would, in general, be beneficial in various conditions including epilepsy, pain, anxiety, and insomnia. Direct activation of the ionotropic GABA receptors using GABAA agonists has for long not been anticipated as a useful therapeutic approach due to desensiti-zation of the receptors. However, the GABAA agonist THIP has proven to be a potential drug in the treatment of insomnia (see Chapter 20).

15.5.3 Modulatory Agents for the GABAa Receptor Complex

The GABAA receptor complex is the target for a large number of structurally diverse compounds, some of which are pharmacologically active and used clinically. These compounds include BZD, ethanol, general anesthetics, barbiturates, and neuroactive steroids, all of which act via a wide range of distinct allosteric-binding sites within the pentameric receptor complex.

The allosteric modulators exert their effects by binding to the GABAA receptor complex and affect GABA-gated chloride conductance. This modulation only takes place when GABA is present in the synaptic cleft, which could be preferable rather than a general receptor activation by exposure to a GABA agonist. Compounds within this group of modulators are marketed for the treatment of anxiety, epilepsy, insomnia, muscle relaxation, and anesthesia. Preclinical studies are going on with the focus on cognitive enhancement and schizophrenia as well.

The fact that receptor regions targeted by allosteric ligands typically are less conserved than the orthosteric sites, in general, opens up for development of subtype-selective modulators with more specific pathophysiological effects and reduced side effects.

Diazepam (15.33)

N(CH3)2 Zolpidem (1S.34)

FIGURE 15.8 Structures of some ligands for the benzodiazepine site.

Diazepam (15.33)

N(CH3)2 Zolpidem (1S.34)

FIGURE 15.8 Structures of some ligands for the benzodiazepine site.

Among the modulatory sites at the GABAa receptor complex, the BZD site is the far most studied to date. The pharmacological profiles of ligands binding to the BZD site span the entire continuum from full and partial agonists, through antagonists, to partial and full inverse agonists. Antagonists do not influence the GABA-induced chloride flux, but antagonizes the action of BZD agonists as well as of inverse agonists.

With the improved knowledge of the subtypes of GABAA receptors and their influence on BZD pharmacology (see Chapter 20), it has become clear that subtype-selective ligands for the modulatory sites could provide more specific pharmacological profiles compared to that of the traditional BZD. Research based on this knowledge is focused on development of hypnotics (arselective), nonsedating anxiolytics (a2- and a3-selective), antipsycotics (a3-selective), and cognition-enhancement (a5-selective inverse agonist). In spite of intensive efforts in this area unselective BZD ligands like diazepam (15.33) and a few a1 preferring ligands, including zolpidem (15.34), are still the most important BZD ligands in the market (Figure 15.8).

15.5.4 GABAB Receptor Ligands

The GABAB receptors belong to the subfamily C, which also comprises the G-protein-coupled Glu receptors (see later sections and Chapter 12). The GABAB receptors exist as heterodimers consisting of two subunits, GABAB1 and GABAB2. The former contains the GABA-binding domain, whereas GABAB2 provides the G-protein-coupling mechanism. The diversity in this class of receptors arises from the two GABAB1 splice variants, GABAB1a and GABAB1b, which together with GABAB2 form the two physiological receptors. Activation of the G-protein-coupled receptor causes a decrease in calcium levels, an increase in potassium membrane conductance and inhibition of cAMP formation. The resulting response is thus inhibitory and leads to hyperpolarization and decreased neurotransmitter release.

The GABAB receptors are selectively activated by baclofen (15.35), of which the (^)-form is the active enantiomer. Baclofen was developed as a liphophilic derivative of GABA, in an attempt to enhance the blood-brain barrier penetrability of the endogenous ligand. Among the limited number of GABAB receptor agonists, the phosphinic acid GABA bioisostere, CGP27492 (15.36), is the most potent reported to date, being approximately 10-fold more potent than GABA.

Phaclofen (15.37) and saclofen (15.38), the phosphonic acid and sulfonic acid analogs of baclofen, respectively, were the first GABAB antagonists reported. In an attempt to improve the pharmacology and pharmacokinetics of the GABAB agonist phosphinic acid analogs mentioned above, a series of selective and highly potent GABAB antagonists, including compound 15.39, capable of penetrating the blood-brain barrier after systemic administration was discovered (Figure 15.9).

Predominant effects of GABAB agonists are muscle relaxation, but also various neurological and psychiatric disorders, including neuropathic pain, anxiety, depression, absence epilepsy, and drug

FIGURE 15.9 Structures of ligands for the GABAb receptor.

addiction are potential targets for GABAb agonist therapy. However, the use of GABAb agonists has been limited due to serious side effects such as sedation, tolerance, and muscle weakness following systemic administration.

15.5.5 Ligands Differentiating the GABAa and GABA Receptors

IAA (15.31) is a naturally occurring metabolite of histamine. The compound has various neurological effects, believed to be mediated by the central GABAA receptors. It penetrates the blood-brain barriers on systemic administration and is therefore advantageous from a bioavailability perspective compared to the other known standard ligands for the ionotropic GABA receptors.

Like other GABA analogs, IAA displays activities on the GABAA as well as on the GABA receptors, being a partial agonist of both groups of receptors. In an attempt to deduce the structural determinants for the activity of the respective receptor groups, a series of IAA analogs have been synthesized.

The introduction of even small substituents in the 2-position of IAA was found to have detrimental effects on the activities of both receptor classes, suggesting that there is little space in the orthosteric sites around this position in the IAA molecule (Figure 15.10). In contrast to the lack of activity in the 2-substituted IAA analogs (15.42), several of the 5-substituted IAA analogs, 15.40 and 15.41, retained the agonist properties at p1 GABA receptors while exhibiting no activity at the aip2y2S GABAA receptors (Figure 15.10).

The 5-Me-IAA analog (15.40) was docked into receptor-models of the GABAA a$2 interface and into the orthosteric site on pj GABA receptors based on the bioactive conformations of the ligand deduced from the previously mentioned pharmacophore model (see Section 15.5.2). The resulting ligand orientation and receptor interactions are show in Figure 15.11. According to the models, the main difference in the vicinity of the ligands in the orthosteric sites of the a1p2y2 and the p1 receptors is a threonine residue (Thr129) in the a1 subunit and a serine (Ser168) residue in the equivalent position in p1. The smaller size of the Ser168 residue in the p1 receptor makes it possible for the orthosteric site to accommodate substituents in the 5-position of IAA. A mutagenesis study based on the above mentioned ligand-receptor docking experiments verified the Thr129 residue in the a1 subunit of the a1p2y2 GABAA receptor and the corresponding Ser168 residue in p1 receptor as major molecular determinants for the observed differences in agonist potencies between the two receptors.

a1ß2y2S expressed in Xenopus oocytes

p1-HEK293 cell line

ECS0 (MM)

EC50 (MM)

(+)-CAMP (15.30)

>1000

39.7

IAA (15.28)

310

13

5-Me-IAA (15.40)

>1000

22

5-Ph-IAA (15.41)

>1000

420

2-Me-IAA (15.42)

>1000

>1000

FIGURE 15.10 Functional data of IAA (15.28) and analogs (15.40, 15.41, and 15.42) from Xenopus oocytes expressing o^P^fc receptors using two-electrode voltage-clamp recordings or a hprHEK293 cell line in the FLIPR membrane potential assay.

R2, R5 = H (IAA (15.28)) R2=H, R5= Me (5-Me-IAA (15.40)) R2=H, R5= Ph (5-Ph-IAA (15.41)) R2= Me, R5= H (2-Me-IAA (15.42))

FIGURE 15.10 Functional data of IAA (15.28) and analogs (15.40, 15.41, and 15.42) from Xenopus oocytes expressing o^P^fc receptors using two-electrode voltage-clamp recordings or a hprHEK293 cell line in the FLIPR membrane potential assay.

GABA.

GABAC

ß2Tyr97

a1Arg119

a1Arg66

GABA.

a1Arg119

a1Arg66

ß2Tyr97

a1Thr129

a1Thr129

Phe138

Phe138

Arg158

GABAC

Arg158

Arg104

Tyr198

Ser168

Arg104

Tyr198

Ser168

FIGURE 15.11 GABA and 5-Me-IAA (15.40) docked into the orthosteric sites of models of (A) the GABAa receptor and (B) the GABAc receptor. The ligands GABA and 5-Me-IAA are shown using orange and green carbon atoms, respectively. Hydrogen atoms other than those on the ligands are omitted for clarity. Proposed hydrogen bond interactions are shown as dashed lines. (Adapted from Madsen et al., J. Med. Chem., 50, 4147, 2007.)

15.6 GLUTAMATE—NEUROTRANSMITTER AND EXCITOTOXIN

Glu is ubiquitously distributed in high concentrations in the CNS and Glu serves other important functions apart from being the major excitatory neurotransmitter, e.g., as building block in proteins and precursor for the neurotransmitter GABA. A very important aspect of Glu functions is the fact that high concentrations of Glu is neurotoxic, which led to the term excitotoxicity even before Glu was recognized as a neurotransmitter. Excitotoxicity describes the ability of all Glu receptor agonists to excite neurons and at the same time being neurotoxic, if the neurons are exposed to the agonist for too long a period and/or exposed to a high concentration of the agonist.

15.6.1 Receptor Classification and Uptake Mechanisms

The targets at the Glu receptor system are receptors, allosteric sites, uptake mechanisms, and Glu metabolism (see Figure 15.1). However the primary focus has been on the ligands and their receptor sites due to the fact that most research has been aimed at lowering Glu activity in relation to the mechanisms of neurodegenerative diseases mentioned previously.

The Glu receptors are divided into two main classes, the ionotropic and the metabotropic Glu receptors (iGluRs and mGluRs), both of these covering three different receptor classes. The three iGlu classes are named by the selective agonists N-methyl-D-aspartic acid (NMDA), 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA), and kainic acid (KA) receptors. These are further subdivided into subtypes, NR1, NR2A-D, NR3A,B for NMDA receptors; GluR1-4 for AMPA receptors; and GluR5-7, KA1,2 for KA receptors (see phylogenetic tree in Chapter 12.1). iGluRs are tetrameric in structure forming an ion channel fluxing Na, K, and Ca ions upon opening leading to depolarization of the cell membrane and excitation of the neurons. The mGluRs modulates the activity of neurons and are G-protein-coupled receptors, also named 7TM receptors as described in Chapter 12. The mGluRs consists of mGluR1-8 and are divided into Groups I, II, and III based on pharmacology, signal transduction mechanisms, and amino acid sequences. Group I consists of mGluR1,5 and stimulates phosholipase C, whereas Group II (mGluR2,3) and Group III (mGluR4,6-8) inhibits the formation of cyclic AMP.

A number of modulatory sites have been recognized on iGluRs as well as on mGluRs and have been given substantial attention after the limited success with especially iGluR antagonists for neuroprotection.

The Glu transporters, excitatory amino acid transporters (EAAT), have been studied extensively in recent years and 2 glial and 3 neuronal subtypes have been characterized; EAAT1,2 and EAAT3-5, respectively. These have different distributions, EAAT2 is the major transporter for Glu in the forebrain, EAAT3 is the major neuronal transporter in the brain and spinal cord, EAAT4 in the cerebellum, and EAAT5 in the retina. EAAT3-5 is located in postsynaptic terminals, but a splice variant of EAAT2 may be the transport system for presynaptic Glu uptake. Furthermore three subtypes of Glu transporters exist on synaptic vesicles (VGLUT1-3) with the function of packing Glu into vesicles for subsequent release from the presynaptic terminal. The enzymatic systems for metabolism of Glu has only been studied to a limited extent in the context of therapeutic potential and will not be discussed here.

15.7 IONOTROPIC GLUTAMATE RECEPTOR LIGANDS

The development of selective ligands for the different receptor classes within iGluRs has been going on for almost three decades and especially within the last decade the search for subtype-selective ligand has taken speed. Within some areas a vast number of potent and selective agents have been developed, whereas for other areas there is still an urgent need for good ligands, with the purpose of characterizing the functions of the respective receptors/subtypes to give an understanding of the physiological and pathophysiological roles of these. Only a limited number of ligands will be discussed and one specific example of a development project will be briefly described.

Extensive knowledge about the structure of the ligand-binding domain of iGluRs have been obtained during the last decade by a number of x-ray structure determinations. These structures include the apostructure (ligand-binding domain without a ligand) and structures with Glu, other agonists, antagonists, partial agonists, and modulatory ligands in a ligand-binding construct of different iGluR subtypes (see also Figures 2.11 and 12.7). The detailed structural information gained by these structures is used extensively in the design of new ligands for the various GluRs as exemplified in Section 15.9.

15.7.1 NMDA Receptor Ligands

NMDA receptors include a number of different binding sites, and thus several potential targets for therapeutic attack. NMDA receptors are unique among ligand-gated ion channels, in that they require two different agonists for activation, Glu (15.2) and glycine (15.58), and at the same time membrane depolarization in order to relieve a blockade by Mg2+. In the normal brain, NMDA receptors are fundamental to development and function, because of their involvement in synaptic plasticity and neuronal signaling processes, including mechanisms of learning and memory. Furthermore, NMDA receptor-induced neurotoxicity is intimately involved in a number of neuronal disorders as previously mentioned. Functional NMDA receptors are heteromers, typically consisting of NR1- and NR2-subunits in a tetrameric structure. The Glu-binding site is located on NR2 subunits and the glycine-binding site on NR1 subunits.

NMDA receptors are unusual among Glu recognition sites inasmuch as the majority of the more potent ligands, both agonists and antagonists, posses an ^-configuration about the a-amino acid center. NMDA (15.43) itself represents the only known agonist in which N-methylation does not lead to reduced affinity. Other potent NMDA agonists have been developed, particularly by replacement of the distal acidic group and/or by conformational restriction of the three essential functional groups, namely, the a-amino group, the a-carboxyl group, and the ra-acidic moiety. Among the potent NMDA agonists with different distal acidic groups are tetrazolylglycine (15.44) and (^)-AMAA (15.45), exemplifying two widely used carboxyl bioisosteric groups, the tetrazole and the 3-isoxazolol, respectively. (^)-AMAA (15.45) and other Glu ligands have been developed using the naturally occurring neurotoxin ibotenic acid (15.46) as a lead. Ibotenic acid (15.46) is, apart from being a potent NMDA agonist, a potent agonist of some mGluR subtypes and a somewhat weaker agonist at other Glu receptor types (Figure 15.12).

COOH

HOOC

HOOC

HOOC

JNH2

o h2n h2n

HOOC

Ibotenic acid (15.46)

Tetrazolylglycine (15.44) (R)-AMAA (15.45)

FIGURE 15.12 Structures of Glu and some NMDA receptor agonists.

HOOC

Ibotenic acid (15.46)

15.7.2 Competitive NMDA Receptor Antagonists

A large number of potent and selective competitive NMDA antagonists have been developed, and the availability of these compounds has greatly facilitated studies of the physiological and pathophysi-ological roles of NMDA receptors. The distance between the two acidic groups in NMDA antagonists is typically one or three C-C bonds longer than in Glu. Many potent ligands have successfully been developed using ra-phosphonic acid analogs, such as (^)-APV (15.47), as lead structures. Combination of an ra-phosphonate group, a long carbon backbone, and conformational restriction has led to different series of potent antagonists. Conformational restriction has been achieved by the use of double bonds (CGP 39653 (15.48)), ring systems (CGS19755 (15.49), and bicyclic structures (LY235959 (15.50)). These antagonists have shown very effective neuroprotective properties in various in vitro models. However, many of these compounds suffer from poor BBB penetration. LY233053 (15.51) represents another class of antagonists with a tetrazole ring as the terminal acidic group. Substitution of the tetrazole by a phosphono group has limited effect on the in vitro activity and shows improved bioavailability (Figure 15.13).

h2n hooc'

cgp 39653 (15.48)

hooc hn hooc cgp 39653 (15.48)

hooc hn po3h2

cgs19755 (15.49)

po3h2

cgs19755 (15.49)

h ly235959 (15.50)

ly233053 (15.51)

ly235959 (15.50)

FIGURE 15.13 Structures of some competitive NMDA receptor antagonists.

h po3h2

15.7.3 Uncompetitive and Noncompetitive NMDA Receptor Antagonists

The dissociative anesthetics PCP (15.52) and ketamine (15.53) block the NMDA receptor channel in a use-dependent manner. Thus, initial agonist activation of the channel is a prerequisite in order for such uncompetitive antagonists to gain access to the binding site, which is situated within the ion channel. The antagonists eventually become trapped within the ion channel and this may result in very slow kinetics. MK-801 (15.54) has been developed as a very effective uncompetitive NMDA antagonist and has been extensively investigated to probe the therapeutic utility of such compounds, notably for the treatment of ischemic insults such as, stroke. MK-801 (15.54) and related high-affinity ligands have, however, shown severe side effects, including psychotomimetic effects, neuronal vacuolization, and impairment of learning and memory. Ligands with lower affinity, such as memantine (15.55), have shown improved therapeutic indexes. Memantine (15.55) is being used in the treatment of AD and Parkinson's disease, and may also have potential in the treatment of AIDS dementia. The fast kinetics and low affinity of memantine (15.55) compared to MK-801 (15.54) may explain the absence of severe side effects (Figure 15.14).

Several compounds with noncompetitive activity at NMDA receptors have been described, and these compounds most likely do not bind to the same site as the uncompetitive ligands. Ifenprodil (15.56)

FIGURE 15.14 Structures of some uncompetitive (15.52-15.55) and noncompetitive (15.56 and 15.57) NMDA receptor antagonists.

and CP-101,606 (15.57) represent an important series of noncompetitive NMDA receptor antagonists. These compounds are active in ischemia models and as anticonvulsants and antinociceptive agents. Early clinical trials with these analogs have been disappointing due to unwanted effects. The side effect profiles do however seem to be significantly improved as compared to, e.g., competitive NMDA antagonists.

15.7.4 The Glycine Coagonist Site

The excitatory coagonist site for glycine (15.58) at the NMDA receptor is named the glycineB receptor. This receptor site is different from the inhibitory glycine receptors found primarily in the spinal cord of the mammalian CNS, where glycine activates strychnine-sensitive ionotropic receptors named as glycineA receptors. GlycineB receptors seem to modulate the level of activity at NMDA receptors. A certain concentration level of glycine is always present in the synapse. Thus, Glu activates the NMDA receptors, whereas the level of glycine can modulate this activity and possibly control receptor desensitization. (^)-Serine (15.59) is a potential endogenous agonist at glycineB receptors.

Limited success of competitive NMDA receptor antagonists as therapeutic agents has focused attention on the glycineB site. (^)-Cycloserine (15.60) and (R)-HA-966 (15.61) are partial glycine agonists, capable of penetrating the BBB after systemic administration. (^)-Cycloserine (15.60) has shown promising effects in the treatment of schizophrenia and AD, and partial agonists may have therapeutic advantages as compared to full antagonists in terms of fewer side effects. A number of glycineB antagonists have also been developed. L-689,560 (15.63) displays high potency, and is derived from the endogenous compound kynurenic acid (15.62), the first glycineB antagonist reported (Figure 15.15).

Glycine (15.58)

(fi)-Cycloserine (15.60)

N' "COOH Kynurenic acid (15.62)

N' "COOH Kynurenic acid (15.62)

Was this article helpful?

0 0
Relaxation Audio Sounds Dusk At The Oasis

Relaxation Audio Sounds Dusk At The Oasis

This is an audio all about guiding you to relaxation. This is a Relaxation Audio Sounds with sounds from Dusk At The Oasis.

Get My Free MP3 Audio


Post a comment