Hans Bruner Osborne

Halki Diabetes Remedy

How I Healed my Diabetes

Get Instant Access

CONTENTS

12.1 Introduction 189

12.1.1 Synaptic Processes and Mechanisms 190

12.2 Receptor Structure and Function 192

12.2.1 G Protein-Coupled Receptors 192

12.2.2 Ligand-Gated Ion Channel Receptors 195

12.2.2.1 The Cys-Loop Receptor Family 195

12.2.2.2 The Ionotropic Glutamate Receptor Family 196

12.2.3 Tyrosine Kinase Receptors 197

12.2.4 Nuclear Receptors 198

12.3 Receptor Pharmacology 199

12.3.1 Recombinant versus in Situ Assays 199

12.3.2 Binding versus Functional Assays 199

12.3.3 Partial and Full Agonists 200

12.3.4 Antagonists 200

12.3.5 Constitutively Active Receptors and Inverse Agonism 202

12.3.6 Allosteric Modulators 203

12.3.6.1 Negative Allosteric Modulators (Noncompetitive Antagonists) 203

12.3.6.2 Positive Allosteric Modulators 203

12.4 Concluding Remarks 204

Further Readings 204

12.1 INTRODUCTION

Communication between cells is mediated by compounds such as neurotransmitters and hormones, which, upon release, will activate receptors in the target cells. This communication is of pivotal importance for many physiological functions and dysfunction in cell communication pathways often have severe consequences. Many diseases are caused by dysfunction in the pathways and in these cases, drugs designed to act at the receptors have beneficial effects. Thus, receptors are very important drug targets.

The first receptors were cloned in the mid-1980s and since then hundreds of receptor genes have been identified. Based on the sequence of the human genome it is currently estimated that more than 1000 human receptors exist. Almost all receptors are heterogeneous, meaning that several receptor subtypes are activated by the same signaling molecule. One such example is the excitatory neurotransmitter glutamate. As shown in Figure 12.1, the amino acid sequence of the glutamate receptors vary and the receptors form subgroups, which, as will be discussed in Chapter 15, share pharmacology.

- GluRI

I- GluR4

I- GluR2

I- GluR7

- GluR5

- mGluR5

mGluR2 mGluR3 mGluR6 mGluR7

I- mGluR8

mGluR4

20 40 60 80 100 Percent amino acid identity

FIGURE 12.1 Phylogenetic tree showing the amino acid sequence identity between cloned mammalian glutamate receptors. The subgroups according to receptor pharmacology have been noted. The NMDA, AMPA, and kainic acid receptors belong to the superfamily of ligand-gated ion channels whereas the metabotropic glutamate receptors (mGluR1-8) belong to the superfamily of GPCRs.

The same signaling molecule can act on both G protein-coupled receptors (GPCRs) and ligand-gated ion channels (Figure 12.1). One of the reasons for the heterogeneity is that it allows cells to be regulated in subtle ways. For example, whereas the fast synaptic action potential is initiated by glutamate receptors of the ligand-gated ion channel family, these receptors are themselves regulated by slower and longer acting glutamate receptors from the GPCR family. The action on these two receptor families is shared by a number of other neurotransmitters such as gamma-aminobutyric acid (GAB A) (Chapter 15), acetylcholine (Chapter 16), and serotonin (Chapter 18).

12.1.1 Synaptic Processes and Mechanisms

Receptors are located in a complex, integrated, and highly interactive environment, which can be further illustrated by the processes and mechanisms of synapses (Figure 12.2). The synapses are key elements in the interneuronal communication in the peripheral and in the central nervous system (CNS). In the CNS, each neuron has been estimated to have synaptic contact with several thousand other neurons, making the structure and function of the CNS extremely complex.

The receptor is activated upon release of the signaling molecule and it is, evidently, equally important to stop the signaling again. This is often achieved by transporters situated in the vicinity

NMDA

AMPA

Kainic acid

Group I Group II

Group III

Presynaptic

Second messenger

Autoreceptor

Ion channel

Storage Release

Presynaptic

Second messenger

Autoreceptor

Ion channel

Uptake

Presynaptic receptor Degradation product

Postsynaptic receptors

Precursor

Metabolite

FIGURE 12.2 Generalized schematic illustration of processes and mechanisms associated with an axosomatic synapse in the CNS. E, enzymes; EM, metabolic; EB, biosynthetic; ED, degradation; ESM, second messenger; (•) neurotransmitter.

Ion channel

Uptake

Presynaptic receptor Degradation product

Postsynaptic receptors

Precursor

Metabolite

FIGURE 12.2 Generalized schematic illustration of processes and mechanisms associated with an axosomatic synapse in the CNS. E, enzymes; EM, metabolic; EB, biosynthetic; ED, degradation; ESM, second messenger; (•) neurotransmitter.

of the receptor, which remove the signaling molecule from the extracellular to the intracellular space, where it is either stored or metabolized. The blockade of a transporter or a metabolic enzyme will cause an elevation of the extracellular concentration of the signaling molecule and lead to increased receptor activation, and transporters and metabolic enzymes can thus be viewed as indirect receptor targets. Synaptic functions may also be facilitated by the stimulation of the neurotransmitter biosynthesis, for example, by administration of a biochemical precursor. Transport mechanisms in synaptic storage vesicles (Figure 12.2) are also potential sites for pharmacological intervention. Autoreceptors normally play a key role as a negative feedback mechanism regulating the release of certain neurotransmitters, making this class of presynaptic receptors therapeutically interesting.

Pharmacological stimulation or inhibition of the earlier mentioned synaptic mechanisms are, however, likely to affect the function of the entire neurotransmitter system. Activation of neu-rotransmitter receptors may, in principle, represent the most direct and selective approach to the stimulation of a particular neurotransmitter system. Furthermore, activation of distinct subtypes of receptors operated by the neurotransmitter concerned may open up the prospect of highly selective pharmacological intervention. Nevertheless, indirect mechanisms of targeting receptors via regulation of the level of the endogenous agonist at the site-of-action remains an important pharmacological principle, which has also been applied outside the synapse as exemplified by compounds increasing insulin release and preventing GLP-1 breakdown.

Direct activation of receptors by full agonists may result in rapid receptor desensitization (insensitive to activation). Partial agonists are less liable to induce receptor desensitization and may therefore be particularly interesting for neurotransmitter replacement therapies. Desensitization may be a more or less pronounced problem associated with the therapeutic use of receptor agonists, whereas receptor antagonists, which in other cases have proved useful therapeutic agents, may inherently cause receptor supersensitivity. The presence of allosteric binding sites at certain receptor complexes, which may function as physiological modulatory mechanisms, offer unique prospects of selective and flexible pharmacological manipulation of the receptor complex concerned. While some receptors are associated with ion channels, others are coupled to second messenger systems. The key steps in such enzyme-regulated multistep intracellular systems (Figure 12.2), also including regulation of gene transcription by second messengers, represent novel targets for therapeutic interventions.

12.2 RECEPTOR STRUCTURE AND FUNCTION

Receptors have been divided into four major superfamilies: GPCRs, ligand-gated ion channels, tyrosine kinase receptors, and nuclear receptors. The first three receptor superfamilies are located in the cell membrane and the latter family is located intracellularly.

Our understanding of ligand-receptor interactions and receptor structure has increased dramatically during the previous decade, not least due to the rapidly growing number of 3D crystal-lographic structures that have been determined of either full receptors or isolated ligand-binding domains. Thus today, structures of partial or full receptors of all four receptor superfamilies have been determined. Clearly, the information obtained from 3D structures of ligand-binding domains in the presence of ligands is very valuable for rational drug design (see Chapter 2). Likewise, knowledge about receptor mechanisms can be used to, e.g., design allosteric modulators interfering with receptor activation.

12.2.1 G Protein-Coupled Receptors

GPCRs are the largest of the four superfamilies with some estimated 1000 human receptor genes. Approximately 50% of these are taste- and odor-sensing receptors that are not of immediate interest for the pharmaceutical industry but are of interest, for example, fragrance manufactures. Nevertheless, it is estimated that 30% of all currently marketed drugs act on GPCRs and the superfamily thus remains a very important target for drug research. It is fascinating to note the very broad variety of signaling molecules or stimuli that are able to act via this receptor superfamily, including tastes, odors, light (photons), ions, monoamines, nucleotides, amino acids, peptides, proteins, and pheromones.

The GPCRs are also referred to as seven transmembrane (7TM) receptors due to the seven alpha-helical transmembrane segments found in all GPCRs (Figure 12.3) and the fact that the receptors can also signal via G protein independent pathways (see later). The GPCRs have been further subdivided into family A, B, and C based on their amino acid sequence homology. Thus receptors

Family A Family B Family C

FIGURE 12.3 The superfamily of GPCRs. All GPCRs contain seven a-helical transmembrane segments and are thus also called seven transmembrane (7TM) receptors. Cartoon of the three families showing the typical orthosteric binding site (agonist in red); family A receptors bind the agonist in the 7TM region, family B receptors bind the agonist in both the 7TM region and the extracellular amino-terminal domain, and dimeric family C receptors bind the agonist exclusively in the extracellular amino-terminal domain. (Adapted from Ji, T. et al., J. Biol. Chem, 273, 17299, 1998.)

Family A Family B Family C

FIGURE 12.3 The superfamily of GPCRs. All GPCRs contain seven a-helical transmembrane segments and are thus also called seven transmembrane (7TM) receptors. Cartoon of the three families showing the typical orthosteric binding site (agonist in red); family A receptors bind the agonist in the 7TM region, family B receptors bind the agonist in both the 7TM region and the extracellular amino-terminal domain, and dimeric family C receptors bind the agonist exclusively in the extracellular amino-terminal domain. (Adapted from Ji, T. et al., J. Biol. Chem, 273, 17299, 1998.)

within family A are closely related to each other than to receptors in family B and C and the like. This grouping also coincides with the way ligands binds to the receptors. Thus, as illustrated in Figure 12.3, the endogenous signaling molecules generally bind to the transmembrane region of family A receptors (e.g., acetylcholine, histamine, dopamine, serotonin, opioid, and cannabinoid GPCRs, Chapters 16 through 19), to both the extracellular loops and amino-terminal domain of family B receptors (e.g., glucagon and GLP-1 GPCRs) and exclusively to the extracellular amino-terminal domain of family C receptors (e.g., glutamate and GABA GPCRs, Chapter 15).

The intracellular loops of GPCRs interact with G proteins. As illustrated in Figure 12.4, the G proteins are trimeric consisting of Ga, Gp, and GY subunits. Receptor activation will cause an interaction of the receptor with the trimeric G^-protein, catalyzing an exchange of GDP for GTP in the Ga subunit whereupon the G protein disassociate into activated Ga and GpY subunits. Both of these will then activate effector molecules such as adenylate cyclase or ion channels (Figure 12.4). 16 Ga, 5 Gp, and 12 GY subunits have been identified in humans and like the receptors they form groups based on the amino acid homology and the effectors they interact with.

RhoA (GDP)

[Ca2+lif PKC f

RhoA

FIGURE 12.4 Principal G protein coupling pathways for a range of 7TM receptors discussed in further detail in Chapters 15 through 19. Receptor activation will catalyze an exchange of GDP for GTP in the a-subunit, which leads to activation and separation of the a- and Py-subunits. Both of these will modulate downstream effectors. a1-2 and P1-2, adrenergic receptor subtypes; D1-5, dopamine receptor subtypes 1-5; GIRK, G proteinregulated inward rectifier potassium channel; 5-HT12, serotonin receptor subtypes 1 and 2; M1-5, muscarinic acetylcholine receptor subtypes 1 to 5; mGluR1-8, metabotropic glutamate receptor subtypes 1 to 8; PLC-P, phospholipase C-P; PI-3-K, phosphoinositide-3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; IP3, inositol 1,4,5-trisphosphate; DAG, diacylglycerol; PKC, protein kinase C; Rho-GEF, Rho-guanine nucleotide exchange factor. (Adapted from Wettschreck, N. and Offermanns, S., Physiol. Rev., 85, 1159, 2005. With permission.)

-fftftfl-f Adenylyl

ATP t

RhoA (GDP)

[Ca2+lif PKC f

RhoA

cAMP

FIGURE 12.4 Principal G protein coupling pathways for a range of 7TM receptors discussed in further detail in Chapters 15 through 19. Receptor activation will catalyze an exchange of GDP for GTP in the a-subunit, which leads to activation and separation of the a- and Py-subunits. Both of these will modulate downstream effectors. a1-2 and P1-2, adrenergic receptor subtypes; D1-5, dopamine receptor subtypes 1-5; GIRK, G proteinregulated inward rectifier potassium channel; 5-HT12, serotonin receptor subtypes 1 and 2; M1-5, muscarinic acetylcholine receptor subtypes 1 to 5; mGluR1-8, metabotropic glutamate receptor subtypes 1 to 8; PLC-P, phospholipase C-P; PI-3-K, phosphoinositide-3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; IP3, inositol 1,4,5-trisphosphate; DAG, diacylglycerol; PKC, protein kinase C; Rho-GEF, Rho-guanine nucleotide exchange factor. (Adapted from Wettschreck, N. and Offermanns, S., Physiol. Rev., 85, 1159, 2005. With permission.)

Most GPCRs desensitize quickly upon activation via phosphorylation of specific serine/threonine residues in the intracellular loops and/or C-terminal by kinases such as G protein-coupled receptor kinases (GRKs). Once phosphorylated, p-arrestin molecules will bind to the receptor and cause arrest of the G protein-mediated signaling and induce internalization. Recent evidence has shown that p-arrestins can activate the tyrosine kinase pathway directly leading to non-G protein mediated cellular effects (Figure 12.5). In some cases, it has even been possible to develop ligands that selectively activate the p-arrestin pathway without activating the G proteins. Such ligands will induce different cellular effects than ligands activating both signaling pathways.

Recent evidence has shown that some if not all GPCRs exist as dimeric or even oligomeric complexes. As shown in Figure 12.3, family C receptors dimerize via a covalent cystein-bridge, which leads to either homo- or heterodimers. The latter is, for example, the case for GABAb receptors, which are formed by heterodimerization of GABAB1 and GABAB2 receptor subunits whereas, e.g., metabotropic glutamate receptors homodimerize. Whether family A and B receptors also homo- or heterodimerize have been heatedly debated in the literature and only a few examples have been convincingly shown to be of physiological importance. One such case is in the field of opioid receptors (Chapter 19), where it has been shown that the k, || and 5 subtypes can form pharmacologically distinct receptor subtypes by heterodimerization.

Collectively, the fact that one GPCR can activate several signaling pathways and heterodimerize to create additional subtypes has greatly complicated our view of receptor function. From a medicinal chemistry point of view it is interesting to note that in some cases it has been possible to develop ligands selectively targeting a specific signaling pathway or heterodimer. This has opened up not only new possibilities but also new challenges in drug design.

FIGURE 12.5 The G protein and P-arrestin signaling pathway of 7TM receptors (7TMR). Agonist activation of 7TMR's initiates the classical G protein cascade (see Figure 12.4 for further details) and rapid receptor phosphorylation by GRKs. The latter leads to the recruitment of P-arrestins, which causes desensitization and internalization of the receptor and activation of tyrosine kinase pathways. (Adapted from Lefkowitz, R.J. and Shenoy, S.K., Science, 308, 512, 2005.)

7TMR

FIGURE 12.5 The G protein and P-arrestin signaling pathway of 7TM receptors (7TMR). Agonist activation of 7TMR's initiates the classical G protein cascade (see Figure 12.4 for further details) and rapid receptor phosphorylation by GRKs. The latter leads to the recruitment of P-arrestins, which causes desensitization and internalization of the receptor and activation of tyrosine kinase pathways. (Adapted from Lefkowitz, R.J. and Shenoy, S.K., Science, 308, 512, 2005.)

Cell response

7TMR

Cell response

12.2.2 Ligand-Gated Ion Channel Receptors

Ligand-gated ion channel receptors can be divided into two major groups, namely, the Cys-loop and ionotropic glutamate receptor families. The latter family is exclusively excitatory, the former are either excitatory (serotonin and nicotinic acetylcholine receptors) or inhibitory (glycine and GABA receptors) by influx of Na+/Ca2+ or Cl- ions, which will hypo- or hyperpolarize the cell, respectively (see Chapter 13 for further details).

12.2.2.1 The Cys-Loop Receptor Family

The nicotinic acetylcholine receptor, at the nerve-muscle synapse, is the best understood Cys-loop receptor, which, upon acetylcholine binding, allow as many as 10,000 potassium and sodium ions per millisecond to pass through the channel. As shown in Figure 12.6, the receptor consists of two acetylcholine binding cx1 subunits and three other subunits (p1, y and 8) that form a pentameric pore in the cell membrane. The pore itself is lined with five a-helices (termed M2), one from each of the five receptor subunits, which have a kink in the middle of the membrane spanning part. This bend is the gate of the receptor, which in the closed state points toward the channel. Agonist binding to the extracellular part of the a-subunits induces local conformational changes that are then relayed through the receptor subunits and ultimately leads to rotation of the pore-lining a-helices and channel opening.

Recently, several high-resolution 3D structures of acetylcholine-binding protein (AChBP), a water-soluble homolog of the ligand-binding domain of nicotinic acetylcholine receptors from the snail Lymnaea stagnalis, have been solved in the presence of various ligands (Figure 12.6). These structures have shown that agonists bind in the interface between the subunits and provided detailed insight into the ligand-receptor interactions. For example, all endogenous agonists of the Cys-loop family contain an amine, which, according to the AChBP structures, is interacting with a cluster of aromatic residues via n-cation bonding.

Most Cys-loop receptors form heteropentamers (e.g., the neuromuscular nicotinic acetylcholine receptor described earlier), but some can form homopentamers (e.g., the nicotinic ay receptor).

FIGURE 12.6 Structure of the family of Cys-loop ligand-gated ion channel receptors. (A) 3D structure of the neuromuscular nicotinic acetylcholine receptor that consists of five subunits (two a-, one P-, one y-, and one 8-subunit) forming an ion-channel in the center. An a- and the y-subunit is shown in red and blue, respectively. The receptor consists of an extracellular (E) ligand-binding domain, a transmembrane domain made of four a-helices, which is the gate of the receptor, and an intracellular (I) domain. (B) 3D structure of the acetylcholine-binding protein (AChBP) viewed from the side (left) and top (right). AChBP is a soluble protein from the snail Lymnaea stagnalis, which is homologous to the extracellular ligand-binding domain of the mammalian Cys-loop receptors. AChBP consists of five identical subunits (one shown in yellow and one in blue), which forms ligand-binding pockets in their interfaces shown here with nicotine bound (in pink). (Adapted from Unwin, N., J. Mol. Biol., 346, 967, 2005; Celie, P.H.N. et al., Neuron, 41, 907, 2004. With permission.)

FIGURE 12.6 Structure of the family of Cys-loop ligand-gated ion channel receptors. (A) 3D structure of the neuromuscular nicotinic acetylcholine receptor that consists of five subunits (two a-, one P-, one y-, and one 8-subunit) forming an ion-channel in the center. An a- and the y-subunit is shown in red and blue, respectively. The receptor consists of an extracellular (E) ligand-binding domain, a transmembrane domain made of four a-helices, which is the gate of the receptor, and an intracellular (I) domain. (B) 3D structure of the acetylcholine-binding protein (AChBP) viewed from the side (left) and top (right). AChBP is a soluble protein from the snail Lymnaea stagnalis, which is homologous to the extracellular ligand-binding domain of the mammalian Cys-loop receptors. AChBP consists of five identical subunits (one shown in yellow and one in blue), which forms ligand-binding pockets in their interfaces shown here with nicotine bound (in pink). (Adapted from Unwin, N., J. Mol. Biol., 346, 967, 2005; Celie, P.H.N. et al., Neuron, 41, 907, 2004. With permission.)

Numerous subunits for both nicotinic acetylcholine receptors and GABAa receptors have been cloned, which can theoretically heteromerize to a staggering high number of subunit combinations. However, in reality, only certain subunit combinations are present and even fewer combinations have therapeutic interest. The glycine and serotonin Cys-loop receptors have fewer subunits, which each can form homo- and heterodimers. Interestingly, some subunits are unable to form their part of the agonist binding pocket in either one or both sides of the two interfaces they participate in. Depending on their subunit composition, Cys-loop receptors can bind from two to five agonist molecules. For example, the neuromuscular nicotinic acetylcholine receptor binds two agonist molecules whereas the nicotinic a receptor and AChBP can bind five agonist molecules (Figure 12.6). Whether all agonist binding sites need to be occupied in order to achieve receptor activation has yet to be demonstrated.

12.2.2.2 The Ionotropic Glutamate Receptor Family

The ionotropic glutamate receptor family comprise of the 15 NMDA, AMPA, and kainic acid receptors listed in Figure 12.1 and two orphan receptors (termed 51-2) with unknown function. The name of the receptor family is a bit misleading as NR1 and NR3A-B actually has glycine as ligand (Chapter 15). Nevertheless, all 17 receptor subunits have the same overall structure: two large extracellular domains referred to as the N-terminal domain (NTD) and agonist-binding domain (ABD), a transmembrane domain (TMD) consisting of three transmembrane segments and a reentry loop and a C-terminal domain (CTD) (Figure 12.7). It is quite interesting to note the resemblance of the structures of the TMD with the amino-terminal domain of potassium channels (Chapter 13), respectively. Functional receptors are comprised of four subunits assembled around the ion channel. All NMDA receptors are heteromeric assemblies as NR1 together with either NR2 or NR3 subunits (forming glutamate or glycine receptors, respectively) whereas AMPA and kainic acid receptors can either be homo- or heteromeric assemblies.

FIGURE 12.7 (A) Illustration of a single ionotropic glutamate receptor subunit with the location of the N-terminal domain (NTD), agonist-binding domain (ABD), transmembrane domain (TMD), and C-terminal domain (CTD) noted. The colored ABD with glutamate in the binding site is based on the crystal structure obtained of the soluble GluR2 binding core shown in (B). (Adapted from Brauner-Osborne, H. et al., J. Med. Chem, 43, 2610, 2000.) (B) Structure of the ABD of GluR2 in the open inactive form (left) and the closed active form with glutamate bound in the cleft (right). The difference in conformation of glutamate bound to mGluR1 and GluR2 are also shown. The structures were generated using the program "Swiss PDB viewer 3.5" with coordinates from Brookhaven Protein Data Base.

FIGURE 12.7 (A) Illustration of a single ionotropic glutamate receptor subunit with the location of the N-terminal domain (NTD), agonist-binding domain (ABD), transmembrane domain (TMD), and C-terminal domain (CTD) noted. The colored ABD with glutamate in the binding site is based on the crystal structure obtained of the soluble GluR2 binding core shown in (B). (Adapted from Brauner-Osborne, H. et al., J. Med. Chem, 43, 2610, 2000.) (B) Structure of the ABD of GluR2 in the open inactive form (left) and the closed active form with glutamate bound in the cleft (right). The difference in conformation of glutamate bound to mGluR1 and GluR2 are also shown. The structures were generated using the program "Swiss PDB viewer 3.5" with coordinates from Brookhaven Protein Data Base.

Recently, high-resolution 3D structures of the isolated ABD of the ionotropic glutamate receptor subunits NR1, NR2A, GluR2, GluR5, and GluR6 have been determined in the absence of ligands and with full and partial agonists, antagonists, and allosteric modulators. Overall these studies have shown that activation is initiated by closure of the ABD around the ligand, which is then relayed to the membrane spanning part of the receptor causing an opening of the channel pore (Figure 12.7). The plentitude of ABD structures has also provided a compelling insight into ligand-receptor interactions and has, for example, shown that the conformation of glutamate bound to various glutamate receptors is quite different as illustrated in Figure 12.7. Such information is very valuable in the design of glutamate receptor subtype selective compounds as will be discussed in further detail in Chapter 15.

12.2.3 Tyrosine Kinase Receptors

As illustrated (refer to Figure 12.9) the tyrosine kinase receptors have a large extracellular agonist-binding domain, one transmembrane segment and an intracellular domain. The receptors can be divided into two groups: those that contain the tyrosine kinase as an integral part of the intracellular domain and those that are associated with a Janus kinase (JAK). Examples of the former group are the insulin receptor family and the epidermal growth factor (EGF) receptor family and examples of the latter are the cytokine receptor family such as the erythropoietin (EPO) receptor and the throm-bopoietin (TPO) receptor. However, both groups share the same overall mechanism of activation: upon agonist binding two intracellular kinases are brought together, which will initiate autophos-phorylation of tyrosine residues of the intracellular tyrosine kinase domain (Figure 12.8). This will attract other proteins (e.g., Shc/Grb2/SOS and STAT for the two receptor groups, respectively) that are also phosphorylated and this will initiate protein cascades and ultimately lead to regulation of transcriptional factors (e.g., Elk-1, Figure 12.8) and thus regulation of genes involved in, e.g., cell proliferation and differentiation. As described for the GPCRs, all the proteins in the intracellular activation cascades are heterogeneous leading to individual responses (i.e., regulation of different subset of genes) in individual cell types.

FIGURE 12.8 Cartoon of a protein cascade initiated by agonist binding to two tyrosine kinase receptors (TKR) causing autophosphorylation of the dimerized intracellular receptor domains. This causes activation of a cascade of intracellular proteins (abbreviated Shc, Grb2/SOS, Ras, Raf, MEK, and MAPK), which ultimately leads to activation of transcription factors (e.g., Elk-1) and thus regulation of gene expression.

FIGURE 12.8 Cartoon of a protein cascade initiated by agonist binding to two tyrosine kinase receptors (TKR) causing autophosphorylation of the dimerized intracellular receptor domains. This causes activation of a cascade of intracellular proteins (abbreviated Shc, Grb2/SOS, Ras, Raf, MEK, and MAPK), which ultimately leads to activation of transcription factors (e.g., Elk-1) and thus regulation of gene expression.

Albeit the tyrosine kinase receptors share the overall activation mechanism, the family has turned out to be rather heterogeneous with respect to the structure and ligand-receptor interaction. Some of the receptors exist as monomers (e.g., the EGF receptor family) in the absence of agonist whereas others exist as covalently linked dimers (e.g., the insulin receptor family) or noncovalently linked dimers (e.g., the EPO receptor). In case of the monomers, agonist binding to either one or both subunits will bring the two receptor subunits together, and thereby initiate the autophosphory-lation. In case of the preformed inactive dimers, agonist binding will cause a conformational change in the receptor, which brings the two intracellular kinases together and thus initiate the autophos-phorylation. One of the best understood examples in this regard is the EPO receptor of which the 3D structure of the extracellular agonist-binding domain has been determined in the absence and in the presence of EPO (Figure 12.9). In the absence of EPO the domain is a dimer in which the ends are too far apart for the JAKs to reach each other. EPO binds to the same amino acids on the receptor that forms the dimer interface and thereby tilts the two receptor subunits. This brings the JAKs close together and initiate the autophosphorylation (Figure 12.9).

12.2.4 Nuclear Receptors

Nuclear receptors are cellular proteins and are thus not embedded in the cell membrane like the previously described receptors. In contrast to the membrane bound receptors, they bind small lipo-philic compounds and function as ligand-modulated transcription factors. The nuclear receptors have been classified according to the type of hormone they bind. Thereby, receptors have been divided into those which bind steroids (glucocorticoids, progestestins, mineralocorticoid androgens, and estrogens) and steroid derivatives (vitamin D3), nonsteroids (e.g., thyroid hormone, retinoids, and prostaglandines), and orphan receptors for which the physiological agonist has yet to be discovered. The receptor family is relatively small (~50 subtypes) of which 50% still belongs to the group of orphan receptors.

epo d1

jak-2

jak-2

jak-2

jak-2

1 1

pp

p-

W

pH

FIGURE 12.9 Cartoon of the activation mechanism of the erythropoietin (EPO) receptor, which belong to the JAK/STAT receptor class of the superfamily of tyrosine kinase receptors. (A) The receptor is dimerized in the inactive conformation by interaction of amino acids, which are similar to those involved in binding of EPO and the intracellular JAKs are kept too far apart to initiate autophosphorylation. (B) Binding of EPO to the dimer interface tilts the structure and brings the JAKs in close proximity, which initiates the autophosphorylation. (C) The actual structure of EPO (in cyan) bound to the extracellular receptor domains of the EPO receptor (in green). The structure was generated using the program "Swiss PDB viewer 3.5" with coordinates from Brookhaven Protein Data Base. (Adapted from Wilson, K.S. et al., Curr. Opin. Struc. Biol. 9, 696, 1999. With permission.)

The nuclear receptors consist of a ligand-binding domain, a DNA binding domain, and a trans-activation domain. Upon activation, two receptors dimerize, as homo- or heterodimers, and bind to specific recognition sites on the DNA. Coactivators will then associate with the dimeric receptor and initiate transcription of the target gene(s). Each receptor recognizes specific DNA sequences, also known as the hormone response elements, which are located upstream of the genes that are regulated. 3D high-resolution structures of both ligand- and DNA-binding domains have been determined. In drug research, the main focus has been on the structures of the ligand-binding domains, which, for several receptors, have been determined in the absence and in the presence of ligands.

12.3 RECEPTOR PHARMACOLOGY

12.3.1 Recombinant versus in Situ Assays

The previous decade has had a profound impact on how receptor pharmacology is performed. As mentioned in the introduction, receptor cloning was initiated in the mid-1980s and today the majority of receptors have been cloned. Thus, it is now possible to determine the effect of ligands on individual receptor subtypes expressed in recombinant systems rather than on a mixture of receptors in, e.g., an organ. This is very useful given that receptor selectivity is a major goal in terms of decreasing side effects of drugs and the development of useful pharmacological tools that can be used to elucidate the physiological function of individual receptor subtypes. Furthermore, recombinant assays allow one to assay cloned human receptors, which would otherwise not have been possible. Most receptors are more than 95% identical between humans and rodents, but due to the small differences in primary amino acid sequence there have been cases of drugs developed for rats rather than for humans, because the compounds were active on the rat receptor but not on the human receptor.

It should be noted that the use of organ and whole animal pharmacology is still required. As previously noted, the cellular effects of receptor activation depend on the intracellular contents of the proteins involved in, e.g., the signaling cascades. These effects can only be determined when the receptor is situated in its natural environment rather than in a recombinant system. In most situations, both recombinant and in situ assays are thus used to fully evaluate the pharmacological profile of new ligands. Furthermore, once a compound with the desired selectivity profile has been identified in the recombinant assays, it is important to confirm that this compound has the predicted physiological effects in, e.g., primary nonrecombinant cell lines, isolated organs and/or whole animals.

12.3.2 Binding versus Functional Assays

Binding assays were used as the method of choice for primary pharmacological evaluation, mainly due to the ease of these assays compared to functional assays that generally required more steps than binding assays. However, several factors have changed this perception: (1) biotechnological functional assays have evolved profoundly and have decreased the number of assay steps and increased the throughput, (2) functional assay equipment has been automated, (3) ligand binding requires a high-affinity ligand, which for many targets identified in genome projects simply does not exist, (4) binding assays are unable to discriminate between agonists and antagonists, and (5) binding assays will only identify compounds binding to the same site as the radioactively labeled tracer.

The Fluorometric Imaging Plate Reader (FLIPR™) illustrates this development toward functional assays. Cells transfected with a receptor coupled to increase in intracellular calcium levels (e.g., a Gaq coupled GPCR or a Ca2+ permeable ligand-gated ion channel) are loaded with the dye Fluo-3, which in itself is not fluorescent. However, as shown in Figure 12.10, the dye becomes fluorescent when exposed to Ca2+ in the cell in a concentration-dependent manner. In this manner, ligand concentration-response curves can be generated on the FLIPR very fast as it automatically reads all wells of a 96-, 384-, or 1534-well tissue culture plate. Many other functional assays along these lines have been developed in recent years.

1000

1000

10000

FIGURE 12.10 (A) Relation between Ca2+ concentration and relative fluorescence intensity of the fluorescent probe fluo-3. (B) The 5-HT2B receptor subtype belong to the superfamily of GPCRs and are coupled to increase in inositol phosphates and intracellular Ca2+. Cells expressing 5-HT2B receptors were loaded with fluo-3 and the fluorescence was determined upon exposure to the endogenous agonist 5-HT (•) and the partial agonists MK-212 (o) and 2-Me-5-HT (■) on a FLIPR™. (Adapted from Jerman, J.C. et al., Eur. J. Pharmacol., 414, 23, 2001.)

12.3.3 Partial and Full Agonists

Agonists are characterized by two pharmacological parameters: potency and maximal response. The most common way of describing the potency is by measuring the agonist concentration, which elicit 50% of the compound's own maximal response (the EC50 value). The maximal response is commonly described as percent of the maximal response of the endogenous agonist. The maximal response is also often described as efficacy or intrinsic activity, which was defined by Stephenson and Ariens, respectively. Compounds, such as 2-Me-5-HT and MK-212 in Figure 12.10, show a lower maximal response than the endogenous agonist and are termed partial agonists. The parameters potency and maximal response are independent of each other and on the same receptor it is thus possible to have, e.g., a highly potent partial agonist and a low potent full agonist. Both parameters are important for drug research, and it is thus desirable to have a pharmacological assay system that is able to determine both the potency and the maximal response of the tested ligands.

12.3.4 Antagonists

Antagonists do not activate the receptors, but block the activity elicited by agonists and accordingly they are only characterized by the parameter affinity. The most common way of characterization of antagonists is by competition with an agonist (functional assay) or a radioactively labeled ligand (binding assay). In both cases, the antagonist concentration is increased and displaces the agonist or radioligand, which are held at a constant concentration. It is then possible to determine the concentration of antagonist that inhibits the response/binding to 50% (the IC50 value). The IC50 value can then be transformed to affinity (K) by the Cheng-Prusoff equation:

Functional assay:

where

[Agonist] is the agonist concentration EC50 is for the agonist in the particular assay

Binding assay:

where

[Radioligand] is the radioligand concentration Kd is the affinity of the radioligand

It is important to observe that the Cheng-Prusoff equation is only valid for competitive antagonists. The Schild analysis is often used to determine whether an antagonist is competitive or noncompetitive. In the Schild analysis the antagonist concentration is kept constant while the agonist concentration is varied. For a competitive antagonist this will cause a rightward parallel shift of the concentration-response curves without a reduction of the maximal response (Figure 12.11A). The degree of right shifting is determined as the dose ratio (DR), which is the concentration of agonist

[S16924]

0

0 nM

30 nM

*

100 nM

T

300 nM

A

1000 nM

S16924

FIGURE 12.11 Schild analysis of the competitive antagonist S16924 on cells expressing the 5-HT2C receptor. (A) Concentration-response curves of the agonist 5-HT were generated in the presence of varying concentrations of S16924. Note the parallel right shift of the curves and the same level of maximum response. (B) DRs are calculated and plotted as a function of the constant antagonist concentration generating a straight line with a slope of 1.00 ± 0.012. These results and the observations from (A) are in agreement with a competitive interaction and the antagonist affinity can thus be determined by the intercept of the abscissa; K = 12.9 nM. (Adapted from Cussac, D. et al., Naunyn Schiedbergs Arch. Phamacol, 361, 549, 2000. With permission.)

S16924

FIGURE 12.11 Schild analysis of the competitive antagonist S16924 on cells expressing the 5-HT2C receptor. (A) Concentration-response curves of the agonist 5-HT were generated in the presence of varying concentrations of S16924. Note the parallel right shift of the curves and the same level of maximum response. (B) DRs are calculated and plotted as a function of the constant antagonist concentration generating a straight line with a slope of 1.00 ± 0.012. These results and the observations from (A) are in agreement with a competitive interaction and the antagonist affinity can thus be determined by the intercept of the abscissa; K = 12.9 nM. (Adapted from Cussac, D. et al., Naunyn Schiedbergs Arch. Phamacol, 361, 549, 2000. With permission.)

giving a particular response in the presence of antagonist divided by the concentration of agonist that gives the same response in the absence of antagonist. Typically one will chose the EC50 values to calculate the DR. In the Schild analysis the log (DR-1) is depicted as a function of the antagonist concentration (Figure 12.11B). When the slope of the curve equals 1 it is a sign of competitive antagonism and the affinity can then be determined by the intercept of the abscissa. When the slope is significantly different from 1 or the curve is not linear it is a sign of noncompetitive antagonism, which invalidates the Schild analysis.

As shown in the example in Figure 12.11, five concentration-response curves are generated to obtain one antagonist affinity determination, illustrating that the Schild analysis is rather work intensive compared to, e.g., the transformation by the Cheng-Prusoff equation where one inhibition curve generates one antagonist affinity determination. However, the latter cannot be used to determine whether an antagonist is competitive or noncompetitive, which is the advantage of the Schild analysis. When testing a series of structurally related antagonists one would thus often determine the nature of antagonism with the Schild analysis for a couple of representative compounds. If these are competitive antagonists, it is reasonable to assume that all compounds in the series are competitive and thus determine the affinity of these compounds by using the less work intensive Cheng-Prusoff equation.

12.3.5 Constitutively Active Receptors and Inverse Agonism

Most receptors display no basal activity or only minor activity but some receptors display increased basal activity in the absence of agonist that has been referred to as constitutive activity. Interestingly, it has been shown that inverse agonists can inhibit this elevated basal activity, which contrast antagonists that inhibit agonist-induced responses but not the constitutive activity (Figure 12.12A).

The examples of important constitutively active receptors include the human ghrelin receptor and several viral receptors that display constitutive activity when expressed in the host cell. This latter group includes the ORF-74 7TM receptor from human herpesvirus 8 (HHV-8), which show a marked increased basal response when expressed in recombinant cells (Figure 12.12B). ORF-74 is homologous to chemokine receptors and does indeed bind chemokine ligands. As shown in Figure 12.12B, chemokines display a wide range of activities on the receptor from full agonism (e.g., GROa) to full inverse agonism (e.g., IP10), which correlates with the angiogenic/angiostatic effects of the chemokines. In 1994, it was demonstrated that HHV8 infection is the cause of Kaposi's sarcoma, which is a multifocal angioproliferative cancer disease mainly affecting AIDS patients.

Full agonist Partial agonist

Antagonist

Partial inverse agonist

Full inverse agonist

Full agonist Partial agonist

Antagonist

Partial inverse agonist

Full inverse agonist

Log conc.CXC-chemokine (M)

Log conc.CXC-chemokine (M)

FIGURE 12.12 (A) The nomenclature of ligand efficacies and schematic illustration on their concentration-dependent effects on constitutive activity. (B) Ligand regulation of the constitutively active ORF-74 receptor from Human Herpesvirus 8 (HHV8). ORF-74 is a GPCR coupled to phosphatidylinositol (PI) turnover, which is regulated by a variety of human chemokines ranging from full agonism by GROa to full inverse agonism by IP10. (Adapted from Rosenkilde, M., Neuropharmacology, 48, 1, 2005. With permission.)

Recently, it was further demonstrated by generation of transgenic mice expressing ORF-74 that it is indeed the constitutively activated receptor, which alone is the cause of Kaposi's sarcoma. Clearly, it would be highly desirable to develop selective inverse agonists of ORF-74, which would very likely inhibit or even prevent the development of Kaposi's sarcoma in HHV8 infected humans.

Constitutive activity can also be caused by somatic mutations. The known examples include constitutively activating mutations in the thyrotropin receptor and the luteinizing hormone receptor, which leads to adenomas, and the rhodopsin receptor, which leads to night blindness. Moreover, in this case it is conceptually possible to alleviate the diseases by the development of inverse agonists.

12.3.6 Allosteric Modulators

Allosteric modulators can both be stimulatory or inhibitory (noncompetitive antagonists) and typically these compounds bind outside the orthosteric binding site (binding site of the endogenous agonist). Allosteric modulators have a number of potential therapeutic benefits compared to agonists and competitive antagonists, which has led to significant increased pharmaceutical interest in recent years. This increased interest has also been fueled by the development of functional high-throughput screening assays, which has made it possible to screen for allosteric modulators (see Section 12.3.2).

The allosteric modulators mentioned in the following text act through allosteric mechanisms as evident from the fact that they do not displace radiolabeled orthosteric ligands. Furthermore, their activity is dependent on the presence of agonists as they do not activate the receptors by themselves. The fact that they bind outside of the orthosteric ligand-binding pocket often leads to increased receptor subtype selectivity. Evolutionary pressure has led to conservation of the orthosteric binding site at different subtypes, as radical mutations would severely impact the binding properties. Thus, it is often seen that the orthosteric binding site is much more conserved than the remaining part of the receptor and accordingly, ligands binding to an allosteric site have a higher chance of being selective. Likewise, the allosteric ligands will have a different pharmacophore than the endogenous ligand, which might improve, e.g. bioavailability. For example, ligands acting at the orthosteric site of the GABAa receptor need a negatively charged acid function and a positively charged basic function, which greatly impairs the transport through biomembranes, whereas allosteric ligands such as the benzodiazepine Diazepam (Chapter 15) does not have any charged groups and show excellent bio-availability. It is well known that many agonists, particularly full agonists, lead to desensitization and internalization of receptors. Unlike agonists, the positive modulators should prevent the development of tolerance (as seen for, e.g. morphine), because they avoid the prolonged receptor activation leading to desensitization and internalization. The fact that the receptors are stimulated in a more natural way by positive modulators rather than the prolonged receptor activation caused by agonists may also lead to a difference in physiological effects, which may or may not be an advantage.

12.3.6.1 Negative Allosteric Modulators (Noncompetitive Antagonists)

As noted in the previous section, the Schild analysis is very useful to discriminate between competitive and noncompetitive antagonists, and an example of the latter is shown in Figure 12.13A. CPCCOEt is a selective antagonist at the mGluR1 receptor, and the Schild analysis clearly demonstrates that the antagonism is noncompetitive due to the depression of the maximal response (compare with Figure 12.11). As noted previously, glutamate binds to the large extracellular amino-terminal domain whereas CPCCOEt has been shown to bind to the extracellular part of the 7TM domain. CPCCOEt does not hinder binding of glutamate to the extracellular domain, but hinder the conformational change leading to receptor activation.

12.3.6.2 Positive Allosteric Modulators

Positive allosteric modulation can be achieved through several mechanisms. For example, benzo-diazepines positively modulate the GABAa receptor by increasing the frequency of channel opening (Chapter 15). Positive modulation can also be obtained by blocking receptor desensitization as exemplified by cyclothiazide (Chapter 15).

Orthosteric binding site

Allosteric

a Glu + 1 |M CPCCOEt t Glu + 3 |M CPCCOEt . Glu + 20 |M CPCCOEt

a Glu + 1 |M CPCCOEt t Glu + 3 |M CPCCOEt . Glu + 20 |M CPCCOEt binding site

FIGURE 12.13 Schild analysis of the noncompetitive antagonist CPCCOEt on cells expressing the metabotropic glutamate receptor subtype mGluRl. (A) Concentration-response curves of the agonist glutamate (i-Glu) were generated in the presence of varying concentrations of CPCCOEt. In contrast to the Schild analysis shown in Figure 12.10, a clear depression of the maximal response is seen with increasing antagonist concentrations. This shows that the antagonist is noncompetitive. (Adapted from Litschig, S. et al., Mol. Pharmacol, 55, 453, 1999. With permission.) (B) Cartoon showing overall structure of a family C receptor with the orthosteric (endogenous agonist) and allosteric binding sites pointed out.

12.4 CONCLUDING REMARKS

The previous decade of receptor research has provided many breakthroughs in our understanding of receptor structure, function, and pharmacology. The many new 3D structures of either full receptors or important domains have provided detailed knowledge about ligand-receptor interactions and receptor activation mechanisms. It has been shown that most receptors can activate several different signaling pathways, which may also be selectively activated/inhibited by drugs. Finally, inverse agonism and allosteric modulation have pointed to novel ways that receptors can be regulated in vivo. Collectively, these new developments have created the foundation for structure-based drug design and new concepts of pharmacological intervention.

Bond, R.A. and IJzerman, A.P. 2006. Recent developments in constitutive receptor activity and inverse agonism, and their potential for GPCR drug discovery. Trends Pharmacol. Sci. 27:92-96.

Brauner-Osborne, H., Egebjerg, J., Nielsen, E.0., Madsen, U., and Krogsgaard-Larsen, P. 2000. Ligands for glutamate receptors: Design and therapeutic prospects. J. Med. Chem. 43:2609-2645.

Brauner-Osborne, H., Wellendorph, P., and Jensen, A.A. 2007. Structure, pharmacology and therapeutic prospects of family C G-protein-coupled receptors. Curr. Drug Targets. 8:169-184.

Gronemeyer, H., Gustafsson, J.A., and Laudet, V. 2004. Principles for modulation of the nuclear receptor super-family. Nat. Rev. Drug Discov. 3:950-964.

Ji, T.H., Grossmann, M., and Ji, I. 1998. G Protein-coupled receptors. I. Diversity of receptor-ligand interactions. J. Biol. Chem. 273:17299-17302.

Lefkowitz, R.J. 2007. Seven transmembrane receptors: Something old, something new. Acta Physiol. 190:9-19.

Lefkowitz, R.J. and Shenoy, S.K., 2005. Science 308:512.

Madsen, U., Brauner-Osborne, H., Greenwood, J.R., Johansen, T.N., Krogsgaard-Larsen, P., Liljefors, T., Nielsen, M., and Fr0lund, B. 2005. GABA and glutamate receptor ligands and their therapeutic potential in CNS disorders. In Drug Discovery Handbook, ed. S.C. Gad, pp. 797-907. New York: Wiley.

FURTHER READINGS

McKay, M.M. and Morrison, D.K. 2007. Integrating signals from RTKs to ERK/MAPK. Oncogene. 26:3113-3121. Ridge, K.D. and Palczewski, K. 2007. Visual rhodopsin sees the light: Structure and mechanism of G protein signaling. J. Biol. Chem. 282:9297-9301. Ward, C.W., Lawrence, M.C., Streltsov, V.A., Adams, T.E., and McKern N.M. 2007. The insulin and EGF receptor structures: New insights into ligand-induced receptor activation. Trends Biochem. Sci. 32:129-137.

Was this article helpful?

0 0
Quit Smoking Today

Quit Smoking Today

Quit smoking for good! Stop your bad habits for good, learn to cope with the addiction of cigarettes and how to curb cravings and begin a new life. You will never again have to leave a meeting and find a place outside to smoke, losing valuable time. This is the key to your freedom from addiction, take the first step!

Get My Free Ebook


Post a comment