Receptors That Affect Concentrations of Endogenous Ligands
A large number of drugs act by altering the synthesis, storage, release, transport, or metabolism of endogenous ligands such as neurotransmitters, hormones, and other intercellular mediators. For instance, there are many examples of drugs that act on neuroeffector junctions by altering neurotransmitter synthesis, storage of neurotransmitter in vesicles, release of neurotransmitters into the synaptic cleft, and subsequent removal of the neurotransmitter from the synaptic cleft by hydrolysis or transport into the pre-synaptic or post-synaptic neuron. The effects of these drugs can either enhance or diminish the effects of the neurotransmitter in order to achieve the desired therapeutic effect. For instance, some of the drugs acting on adrenergic neurotransmission (Chapters 8 and 12) include α-methyltyrosine (inhibits synthesis of norepinephrine (NE)), cocaine (blocks NE reuptake), amphetamine (promotes NE release), and selegeline (inhibits NE breakdown). There are similar examples for other neurotransmitter systems including acetylcholine (ACh; Chapters 8 and 10), dopamine (DA), and serotonin (5HT; Chapters 13, 14, 15). Drugs that affect the synthesis of circulating mediators such as vasoactive peptides (e.g., angiotensin-converting enzyme inhibitors; Chapter 26) and lipid-derived autocoids (e.g., cyclooxygenase inhibitors; Chapter 33) are also widely used in the treatment of hypertension, inflammation, myocardial ischemia, and other disease states.
Receptors That Regulate the Ionic Milieu
A relatively small number of drugs act by affecting the ionic millieu of blood, urine, and the GI tract. The receptors for these drugs are ion pumps and transporters, many of which are expressed only in specialized cells of the kidney and GI system. Drug effects on many of these receptors can have effects throughout the body due to changes in blood electrolytes and pH. For instance, most of the diuretics (e.g., furosemide, chlorothiazide, amiloride) act by directly affecting ion pumps and transporters in epithelial cells of the nephron that increase the movement of Na+ into the urine, or by altering the expression of ion pumps in these cells (e.g., aldosterone). Chapter 25 provides a detailed description of the mechanisms of action of diuretic drugs. Another therapeutically important target is the H+,K+-ATPase (proton pump) of gastric parietal cells. Irreversible inhibition of this proton pump by drugs such as esomeprazole reduces gastric acid secretion by 80-95% (Chapter 45) and is a mainstay of therepy for peptic ulcer.
Cellular Pathways Activated by Physiological Receptors
Signal Transduction Pathways. Physiological receptors have at least two major functions, ligand binding and message propagation (i.e., signaling). These functions imply the existence of at least two functional domains within the receptor: a ligand-binding domain and an effector domain. The structure and function of these domains in many families of receptors have been deduced from high-resolution crystal structures of the receptor proteins and/or by analysis of the behavior of intentionally mutated receptors. Many drugs target the extracelluar ligand-binding domain of physiological receptors. Examples include the widely used β adrenergic antagonists. However, drugs can affect the receptor by targeting either domain, as in the case of anticancer drugs used to target the epidermal growth factor receptor (EGFR; Chapters 60, 61, 62). Cetuximab is a monoclonal antibody that targets the extracellular ligand-binding domain of the EGFR and inhibits epidermal growth factor (EGF) signaling, whereas the small molecule drugs gefitinib and erlotinib bind the intracellular effector domain and block the protein tyrosine kinase activity of the activated EGFR.
The regulatory actions of a receptor may be exerted directly on its cellular target(s), on effector protein(s), or may be conveyed by intermediary cellular signaling molecules called transducers. The receptor, its cellular target, and any intermediary molecules are referred to as a receptor-effector system or signal transduction pathway. Frequently, the proximal cellular effector protein is not the ultimate physiological target but rather is an enzyme, ion channel, or transport protein that creates, moves, or degrades a small molecule (e.g., a cyclic nucleotide, inositol trisphosphate, or NO) or ion (e.g., Ca2+) termed a second messenger. If the effector is an ion channel or ion pump, the effect of ligand binding can be a change in membrane potential that alters the excitability of the cell. Second messengers can diffuse in the proximity of their synthesis or release and convey information to a variety of targets, which may integrate multiple signals. Even though these second messengers originally were thought of as freely diffusible molecules within the cell, imaging studies show that their diffusion and intracellular actions are constrained by compartmentation—selective localization of receptor-transducer-effector-signal termination complexes—established by protein-lipid and protein-protein interactions (Baillie, 2009). All cells express multiple forms of proteins designed to localize signaling pathways by protein-protein interactions; these proteins are termed scaffolds or anchoring proteins. Examples of scaffold molecules include the AKAPs (A-kinase anchoring proteins) that bind the regulatory subunit of the cyclic AMP dependent protein kinase (PKA) near its substrate(s) in various subcellular compartments (Carnegie et al., 2009).
Signal Integration and Amplification. Receptors and their associated effector and transducer proteins also act as integrators of information as they coordinate signals from multiple ligands with each other and with the differentiated activity of the target cell. For example, signal transduction systems regulated by changes in cyclic AMP (cAMP) and intracellular Ca2+ are integrated in many excitable tissues. In cardiac myocytes, an increase in cellular cAMP caused by activation of β adrenergic receptors enhances cardiac contractility by augmenting the rate and amount of Ca2+ delivered to the contractile apparatus; thus, cAMP and Ca2+ are positive contractile signals in cardiac myocytes. By contrast, cAMP and Ca2+ produce opposing effects on the contractility of smooth muscle cells: as usual, Ca2+ is a contractile signal, however, activation of β adrenergic receptors on these cells activates the cAMP-PKA pathway, which leads to relaxation through the phosphorylation of proteins that mediate Ca2+ signaling, such as myosin light chain kinase and ion channels that hyperpolarize the cell membrane. Thus, the distinct patterns of integration of signal transduction systems within target cells can lead to a variety of pharmacodynamic effects that result from functional interactions downstream from the receptors. These functional interactions can be synergistic, additive, or antagonistic.
Another important property of physiological receptors is their capacity to significantly amplify a physiological signal. Neurotransmitters, hormones, and other extracellular ligands are often present at the ligand-binding domain of a receptor in very low concentrations (nM to μM levels). However, the effector domain or the signal transduction pathway often contains enzymes and enzyme cascades to catalytically amplify the intended signal. In this regard, the description of receptor occupancy-cellular response in Equation 3-1 is an oversimplification. The ability of virtually all receptors to amplify physiological signals makes them excellent targets for natural ligands and drugs. When, e.g., a single agonist molecule binds to a receptor that is an ion channel, hundreds of thousands to millions of ions flow through the channel every second. Similarly, the binding of a single photon to cis-retinal in the photoreceptor rhodpsin is eventually amplified ~1 × 106-fold. In the case of nuclear receptors, a single steroid hormone molecule binding to its receptor initiates the transcription of many copies of specific mRNAs, which in turn can give rise to multiple copies of a single protein.
Structural and Functional Families of Physiological Receptors
Receptors for physiological regulatory molecules can be assigned to functional families whose members share similar molecular structures and biochemical mechanisms with common features. Table 3–1 outlines six major families of receptors with examples of their physiological ligands, signal transduction systems, and drugs that affect these systems. The basic structure of their ligand-binding domains, effector domains, and how agonist binding influences the regulatory activity of the receptor is well understood for each of these signal transduction systems. The relatively small number of biochemical mechanisms and structural formats used for cellular signaling is fundamental to the ways in which target cells integrate signals from multiple receptors to produce additive, sequential, synergistic, or mutually inhibitory responses.
Table 3-1Physiological Receptors ||Download (.pdf) Table 3-1 Physiological Receptors
|STRUCTURAL FAMILY ||FUNCTIONAL FAMILY ||PHYSIOLOGICAL LIGANDS ||EFFECTORS AND TRANSDUCERS ||EXAMPLE DRUGS |
| ||β Adrenergic receptors ||NE, Epi, DA ||Gs; AC ||Dobutamine, propranolol |
|GPCR || || || || |
| ||Muscarinic cholinergic receptors ||ACh ||Gi and Gq; AC, ion channels, PLC ||Atropine |
| ||Eicosanoid receptors ||Prostaglandins, leukotrienes, thromboxanes ||Gs, Gi and Gq proteins ||Misoprostol, montelukast |
| ||Thrombin receptors (PAR) ||Receptor peptide ||G12/13, GEFs ||(in development) |
|Ion channels ||Ligand-gated ||ACh (M2), GABA, 5-HT ||Na+, Ca2+, K+, Cl− ||Nicotine, gabapentin |
| ||Voltage-gated ||None (activated by membrane depolarization) ||Na+, Ca2+, K+, other ions ||Lidocaine, verapamil |
|Transmembrane enzymes ||Receptor tyrosine kinases ||Insulin, PDGF, EGF, VEGF, growth factors ||SH2 domain and PTB-containing proteins ||Herceptin, imatinib |
| ||Membrane-bound GC ||Natriuretic peptides ||Cyclic GMP ||Neseritide |
| ||Tyrosine phosphatases || || || |
|Transmembrane, non-enzymes ||Cytokine receptors ||Interleukins and other cytokines ||Jak/STAT, soluble tyrosine kinases || |
| ||Toll-like receptors ||LPS, bacterial products ||MyD88, IARKs, NF-κB || |
|Nuclear receptors ||Steroid receptors ||Estrogen, testosterone ||Co-activators ||Estrogens, androgens, cortisol |
| ||Thyroid hormone receptors ||Thyroid hormone || ||Thyroid hormone |
| ||PPARγ ||PPARγ || ||Thiazolidinediones |
|Intracellular enzymes ||Soluble GC ||NO, Ca2+ ||Cyclic GMP ||Nitrovasodilators |
G Protein–Coupled Receptors (GCPRs)
Receptors and G Proteins. GPCRs span the plasma membrane as a bundle of seven α-helices (Palczewski et al., 2000) (Figure 3–8). Humans express over 800 GPCRs that make up the third largest family of genes in humans, with roughly half of these GPCRs dedicated to sensory perception (smell, taste, and vision). The remaining receptors regulate an impressive number of physiological functions including nerve activity, tension of smooth muscle, metabolism, rate and force of cardiac contraction, and the secretion of most glands in the body. Included among the ligands for GPCRs are neurotransmitters such as ACh, biogenic amines such as NE, all eicosanoids and other lipid signaling molecules, peptide hormones, opioids, amino acids such as GABA, and many other peptide and protein ligands. GPCRs are important regulators of nerve activity in the CNS and are the receptors for the neurotransmitters of the peripheral autonomic nervous system. For example, ACh released by the parasympathetic nervous system regulates the functions of glands and smooth muscle through its action on muscarinic receptors. NE released by the sympathetic nervous system interacts with α and β adrenergic receptors to regulate cardiac function and the tone of vascular smooth muscle (Chapters 8, 9, 10, 11, 12). Because of their number and physiological importance, GPCRs are the targets for many drugs; perhaps half of all non-antibiotic prescription drugs act at these receptors.
Diagram showing the stimulation of a G–protein coupled receptor by ligand, the activation of the G protein, and stimulation of selected effectors. Schematic diagram of the mechanisms involved in the control of cell function by G–protein coupled receptors, G proteins, and effectors. In the absence of ligand, the receptor and G protein heterotrimer form a complex in the membrane with the Gα subunit bound to GDP. Following binding of ligand, the receptor and G protein α subunit undergo a conformational change leading to release of GDP, binding of GTP, and dissociation of the complex. The activated GTP-bound Gα subunit and the freed βγ dimer bind to and regulate effectors. The system is returned to the basal state by hydrolysis of the GTP on the α subunit; a reaction that is markedly enhanced by the RGS proteins. Prolonged stimulation of the receptor can lead to down-regulation of the receptor. This event is initiated by G protein receptor kinases (GRKs) that phosphorylate the C terminal tail of the receptor, leading to recruitment of proteins termed arrestins; arrestins bind to the receptor on the internal surface, displacing G proteins and inhibiting signaling. Detailed descriptions of these signaling pathways are given throughout the text in relation to the therapeutic actions of drugs affecting these pathways.
Receptor Subtypes. There are multiple receptor subtypes within families of receptors. Ligand-binding studies with multiple chemical entities initially identified receptor subtypes; molecular cloning has greatly accelerated the discovery and definition of additional receptor subtypes; their expression as recombinant proteins has facilitated the discovery of subtype-selective drugs. Distinct but related receptors may display distinctive patterns of selectivity among agonist or antagonist ligands. When selective ligands are not known, the receptors are more commonly referred to as isoforms rather than as subtypes. The distinction between classes and subtypes of receptors, however, is often arbitrary or historical. The α1, α2, and β adrenergic receptors differ from each other both in ligand selectivity and in coupling to G proteins (Gq, Gi, and Gs, respectively), yet α and β are considered receptor classes and α1 and α2 are considered subtypes. The α1A, α1B, and α1C receptor isoforms differ little in their biochemical properties, although their tissue distributions are distinct. The β1, β2, and β3 adrenergic receptor subtypes exhibit differences in both tissue distribution and regulation by phosphorylation by G–protein receptor kinases (GRKs) and PKA.
Pharmacological differences among receptor subtypes are exploited therapeutically through the development and use of receptor-selective drugs. Such drugs may be used to elicit different responses from a single tissue when receptor subtypes initiate different intracellular signals, or they may serve to differentially modulate different cells or tissues that express one or another receptor subtype. For example, β2 adrenergic agonists such as terbutaline are used for bronchodilation in the treatment of asthma in the hope of minimizing cardiac side effects caused by stimulation of the β1 adrenergic receptor (Chapter 12). Conversely, the use of β1-selective antagonists minimizes the chance of bronchoconstriction in patients being treated for hypertension or angina (Chapters 12 and 27). Increasing the selectivity of a drug among tissues or among responses elicited from a single tissue may determine whether the drug's therapeutic benefits outweigh its unwanted effects.
Receptor Dimerization. Receptor-ligand interactions alone do not regulate all GPCR signaling. GPCRs undergo both homo- and heterodimerization and possibly oligomerization. Heterodimerization can result in receptor units with altered pharmacology compared with either individual receptor. As an example, opioid receptors can exist as homodimers of μ or δ receptors, or as μδ heterodimers with distinctly different pharmacodynamic properties than either homodimer (Chapter 18). Evidence is emerging that dimerization of receptors may regulate the affinity and specificity of the complex for G proteins, and regulate the sensitivity of the receptor to phosphorylation by receptor kinases and the binding of arrestin, events important in termination of the action of agonists and removal of receptors from the cell surface. Dimerization also may permit binding of receptors to other regulatory proteins such as transcription factors. Thus, the receptor-G protein-effector systems are complex networks of convergent and divergent interactions involving both receptor-receptor and receptor-G protein coupling that permit extraordinarily versatile regulation of cell function. Dimerization of single membrane spanning receptors is central to their activation (described later in the chapter).
G Proteins. GPCRs couple to a family of heterotrimeric GTP-binding regulatory proteins termed G proteins. G proteins are signal transducers that convey the information that agonist is bound to the receptor from the receptor to one or more effector proteins (Gilman, 1987). G–protein-regulated effectors include enzymes such as adenylyl cyclase, phospholipase C, cyclic GMP phosphodiesterase (PDE6), and membrane ion channels selective for Ca2+ and K+ (Table 3–1, Figure 3–8). The G protein heterotrimer is composed of a guanine nucleotide-binding α subunit, which confers specific recognition to both receptors and effectors, and an associated dimer of β and γ subunits that helps confer membrane localization of the G protein heterorimer by prenylation of the γ subunit. In the basal state of the receptor-heterotrimer complex, the α subunit contains bound GDP and the α-GDP:βγ complex is bound to the unliganded receptor (Figure 3-8). The G protein family is comprised of 23 α subunits (which are the products of 17 genes), 7 β subunits, and 12 γ subunits. The α subunits fall into four families (Gs, Gi, Gq, and G12/13) which are responsible for coupling GPCRs to relatively distinct effectors. The Gs α subunit uniformly activates adenylyl cyclase; the Gi α subunit can inhibit certain isoforms of adenylyl cyclase; the Gq α subunit activates all forms of phospholipase Cβ; and the G12/13 α subunits couple to guanine nucleotide exchange factors (GEFs), such as p115RhoGEF for the small GTP-binding proteins Rho and Rac. The signaling specificity of the large number of possible βγ combinations is not yet clear; nonetheless, it is known that K+ channels, Ca2+ channels, and PI-3 kinase (PI3K) are some of the effectors of free βγ dimer (Figure 3–8).
G Protein Activation. When an agonist binds to a GPCR, there is a conformational change in the receptor that is transmitted from the ligand-binding pocket to the second and third intracellular loops of the receptor which couple to the G protein heterotrimer. This conformational change causes the α subunit to exchange its bound GDP for GTP (Figure 3-8). Binding of GTP activates the α subunit and causes it to release both the βγ dimer and the receptor, and both the GTP-bound α subunit and the βγ heterodimer become active signaling molecules (Gilman, 1987). The interaction of the agonist-bound GPCR with the G protein is transient; following activation of one G protein, the receptor is freed to interact with other G proteins. Depending on the nature of the α subunit, the active, GTP-bound form binds to and regulates effectors such as adenylyl cyclase (via Gs α) or phospholipase Cβ (via Gq α). The βγ subunit can regulate many effectors including ion channels and enzymes such as PI3-K (Figure 3–8). The G protein remains active until the GTP bound to the α subunit is hydrolyzed to GDP. The α subunit has a slow intrinsic rate of GTP hydrolysis that is modulated by a family of proteins termed regulators of G protein signaling (RGSs). The RGS proteins greatly accelerate the hydrolysis of GTP and are potentially attractive drug targets (Ross and Wilkie, 2000). Once the GDP bound to the α subunit is hydrolyzed to GDP, the βγ subunit and receptor recombine to form the inactive receptor-G protein heterotrimer basal complex that can be reactivated by another ligand-binding event (Figure 3–8).
Cyclic AMP. Cyclic AMP is synthesized by adenylyl cyclase under the control of many GPCRs; stimulation is mediated by the Gs α subunit, inhibition by the Gi α subunit. The cyclic AMP pathway provides a good basis for understanding the architecture and regulation of many second messenger signaling systems (for an overview of cyclic nucleotide action, see Beavo and Brunton, 2002).
There are nine membrane-bound isoforms of adenylyl cyclase (AC) and one soluble isoform found in mammals (Hanoune and Defer, 2001). The membrane-bound ACs are glycoproteins of ~120 kDa with considerable sequence homology: a small cytoplasmic domain; two hydrophobic transmembrane domains, each with six membrane-spanning helices; and two large cytoplasmic domains. Membrane-bound ACs exhibit basal enzymatic activity that is modulated by binding of GTP-liganded α subunits of the stimulatory and inhibitory G proteins (Gs and Gi). Numerous other regulatory interactions are possible, and these enzymes are catalogued based on their structural homology and their distinct regulation by G protein α and βγ subunits, Ca2+, protein kinases, and the actions of the diterpene forskolin. Cyclic AMP generated by adenylyl cyclases has three major targets in most cells, the cyclic AMP dependent protein kinase (PKA), cAMP-regulated guanine nucleotide exchange factors termed EPACs (exchange factors directly activated by cAMP), and via PKA phosphorylation, a transcription factor termed CREB (cAMP response element binding protein). In cells with specialized functions, cAMP can have additional targets such as cyclic nucleotide-gated ion channels (Wahl-Schott and Biel, 2009), cyclic nucleotide-regulated phosphodiesterases (PDEs), and several ABC transporters (MRP4 and MRP5) for which it is a substrate (see Chapter 7).
PKA. The best understood target of cyclic AMP is the PKA holoenzyme consisting of two catalytic (C) subunits reversibly bound to a regulatory (R) subunit dimer to form a heterotetramer complex (R2C2). At low concentrations of cAMP, the R subunits inhibit the C subunits; thus the holoenzyme is inactive. When AC is activated and cAMP concentrations are increased, four cyclic AMP molecules bind to the R2C2 complex, two to each R subunit, causing a conformational change in the R subunits that lowers their affinity for the C subunits, causing their activation. The active C subunits phosphorylate serine and threonine residues on specific protein substrates.
There are multiple isoforms of PKA; molecular cloning has revealed α and β isoforms of both the regulatory subunits (RI and RII), as well as three C subunit isoforms Cα, Cβ, and Cγ. The R subunits exhibit different subcellular localization and binding affinities for cAMP, giving rise to PKA holoenzymes with different thresholds for activation (Taylor et al., 2008). Both the R and C subunits interact with other proteins within the cell, particularly the R subunits, and these interactions can be isoform-specific. For instance, the RII isoforms are highly localized near their substrates in cells through interactions with a variety of A kinase anchoring proteins (AKAPs) (Carnegie et al., 2009; Wong and Scott, 2004).
PKA can phosphorylate a diverse array of physiological targets such as metabolic enzymes and transport proteins, and numerous regulatory proteins including other protein kinases, ion channels, and transcription factors. For instance, phosphorylation of the cAMP response element–binding protein, CREB, on serine 133 recruits CREB-binding protein (CBP), a histone acetyltransferase that interacts with RNA polymerase II (POLII) and leads to enhanced transcription of ~105 genes containing the cAMP response element motif (CRE) in their promoter regions (e.g., tyrosine hydroxylase, iNOS, AhR, angiotensinogen, insulin, the glucocorticoid receptor, BC12, and CFTR) (Mayr and Montminy, 2001; Sands and Palmer, 2008).
Cyclic AMP–Regulated Guanine Nucleotide Exchange Factors (GEFs). The small GTP-binding proteins are monomeric GTPases and key regulators of cell function. The small GTPases operate as binary switches that exist in GTP- or GDP-liganded conformations. They integrate extracellular signals from membrane receptors with cytoskeletal changes and activation of diverse signaling pathways, regulating such processes as phagocytosis, progression through the cell cycle, cell adhesion, gene expression, and apoptosis (Etienne-Manneville and Hall, 2002). A number of extracellular stimuli signal to the small GTPases directly or through second messengers such as cyclic AMP.
For example, many small GTPases are regulated by GEFs. GEFs act by binding to the GDP-liganded GTPase and catalyzing the exchange of GDP for GTP. The two GEFs regulated by cAMP are able to activate members of the Ras small GTPase family, Rap1 and Rap2; these GEFs are termed exchange proteins activated by cyclic AMP (EPAC-1 and EPAC-2). The EPAC pathway provides an additional effector system for cAMP signaling and drug action that can act independently or cooperatively with PKA (Cheng et al., 2008; Roscioni et al., 2008).
PKG. Stimulation of receptors that raise intracellular cyclic GMP concentrations (Figure 3-11) leads to the activation of the cyclic GMP-dependent protein kinase (PKG) that phosphorylates some of the same substrates as PKA and some that are PKG-specific. In some tissues, PKG can also be activated by cAMP. Unlike the heterotetramer (R2C2) structure of the PKA holoenzyme, the catalytic domain and cyclic nucleotide-binding domains of PKG are expressed as a single polypeptide, which dimerizes to form the PKG holoenzyme.
PKG exists in two homologous forms, PKG-I and PKG-II. PKG-I has an acetylated N terminus, is associated with the cytoplasm and has two isoforms (Iα and Iβ) that arise from alternate splicing. PKG-II has a myristylated N terminus, is membrane-associated and can be localized by PKG-anchoring proteins in a manner analogous to that known for PKA, although the docking domains of PKA and PKG are very different structurally. Pharmacologically important effects of elevated cyclic GMP include modulation of platelet activation and relaxation of smooth muscle (Rybalkin et al., 2003).
PDEs. Cyclic nucleotide phosphodiesterases form another family of important signaling proteins whose activities are regulated via the rate of gene transcription as well as by second messengers (cyclic nucleotides or Ca2+) and interactions with other signaling proteins such as β arrestin and protein kinases. PDEs hydrolyze the cyclic 3′,5′-phosphodiester bond in cAMP and cGMP, thereby terminating their action.
The enzymes comprise a superfamily with >50 different PDE proteins divided into 11 subfamilies on the basis of amino acid sequence, substrate specificity, pharmacological properties, and allosteric regulation (Conti and Beavo, 2007). PDEs share a conserved catalytic domain at the carboxyl terminus, as well as regulatory domains and targeting domains that localize a given enzyme to a specific cellular compartment. The substrate specificities of the different PDEs include those specific for cAMP hydrolysis, cGMP hydrolysis, and some that hydrolyze both cyclic nucleotides. PDEs (mainly PDE3 forms) are drug targets for treatment of diseases such as asthma, cardiovascular diseases such as heart failure, atherosclerotic coronary and peripheral arterial disease, and neurological disorders. PDE5 inhibitors (e.g., sildenafil) are used in treating chronic obstructive pulmonary disease and erectile dysfunction. By inhibiting PDE5, these drugs increase accumulation of cellular cGMP in the smooth muscle of the corpus caverosum, thereby enhancing its relaxation and improving its capacity for engorgement (Mehats et al., 2002).
Ca2+. Calcium is an important messenger in all cells and can regulate diverse responses including gene expression, contraction, secretion, metabolism, and electrical activity. Ca2+ can enter the cell through Ca2+ channels in the plasma membrane (See "Ion Channels", below) or be released by hormones or growth factors from intracellular stores. In keeping with its role as a signal, the basal Ca2+ level in cells is maintained in the 100 n range by membrane Ca2+ pumps that extrude Ca2+ to the extracellular space and a sarcoplasmic reticulum (SR) Ca2+-ATPase (SERCA) in the membrane of the endoplasmic reticulum (ER) that accumulates Ca2+ into its storage site in the ER/SR.
Hormones and growth factors release Ca2+ from its intracellular storage site, the ER, via a signaling pathway that begins with activation of phospholipase C at the plasma membrane, of which there are two primary forms, PLCβ and PLCγ. GPCRs that couple to Gq or Gi activate PLCβ by activating the G protein α subunit (Figure 3–8) and releasing the βγ dimer. Both the active, Gq-GTP bound α subunit and the βγ dimer can activate certain isoforms of PLCβ. PLCγ isoforms are activated by tyrosine phosphorylation, including phosphorylation by receptor and non-receptor tyrosine kinases. For example, growth factor receptors such as the epidermal growth factor receptor (EGFR) are receptor tyrosine kinases (RTKs) that autophosphorylate on tyrosine residues upon binding their cognate growth factor. The phosphotyrosine formed on the cytoplasmic domain of the RTK is a binding site for signaling proteins that contain SH2 domains, such as PLCγ. Once recruited to the RTK's SH2 domain, PLCγ is phosphorylated/activated by the RTK.
PLCs are cytosolic enzymes that translocate to the plasma membrane upon receptor stimulation. When activated, they hydrolyze a minor membrane phospholipid, phosphatidylinositol-4, 5-bisphosphate, to generate two intracellular signals, inositol-1,4,5-trisphosphate (IP3) and the lipid, diacylglycerol (DAG). Both of these molecules lead to signaling events by activating families of protein kinases. DAG directly activates members of the protein kinase C (PKC) family. IP3 diffuses to the ER where it activates the IP3 receptor in the ER membrane causing release of stored Ca2+ from the ER. Release of Ca2+ from these intracellular stores raises Ca2+ levels in the cytoplasm many fold within seconds and activates calmodulin-sensitive enzymes such as cyclic AMP PDE and a family of Ca2+/calmodulin-sensitive protein kinases (e.g., phosphorylase kinase, myosin light chain kinase, and CaM kinases II and IV) (Hudmon and Schulman, 2002). Depending on the cell's differentiated function, the Ca2+/calmodulin kinases and PKC may regulate the bulk of the downstream events in the activated cells. For example, release of the sympathetic transmitter norepinephrine onto vascular smooth muscle cells stimulates α adrenergic receptors, activates the Gq-PLC-IP3 pathway, triggers the release of Ca2+, and leads to contraction by stimulating the Ca2+/calmodulin-sensitive myosin light chain kinase to phosphorylate the regulatory subunit of the contractile protein, myosin.
The IP3-stimulated release of Ca2+ from the ER, its reuptake, and the refilling of the ER pool of Ca2+ are regulated by a novel set of Ca2+ channels. The IP3 receptor is a ligand-gated Ca2+ channel found in high concentrations in the membrane of the ER (Patterson et al., 2004). It is a large protein of ~2700 amino acids with three major domains. The N-terminal region contains the IP3-binding site, and the middle region contains a regulatory domain that can be phosphorylated by a number of protein kinases including PKA and PKG. The C-terminal region contains six membrane-spanning helices that form the Ca2+ pore. The functional channel is formed by four subunits arranged as a tetramer. In addition to IP3, which markedly stimulates Ca2+ flux, the IP3 channel is regulated by Ca2+ and by the activities of PKA and PKG. Ca2+ concentrations in the 100-300 nM range enhance Ca2+ release, but concentrations near 1 μM inhibit release, which can create the oscillatory patterns of Ca2+ release seen in certain cells. Phosphorylation of the IP3 receptor by PKA enhances Ca2+ release, but phosphorylation of an accessory protein, IRAG, by PKG inhibits Ca2+ release. In smooth muscle, this effect of PKG represents part of the mechanism by which cyclic GMP relaxes vessel tone. In skeletal and cardiac muscle, Ca2+ release from intracellular stores (the sarcoplasmic reticulum) occurs through a process termed Ca2+-induced Ca2+ release, and is primarily mediated by the ryanodine receptor (RyR). Ca2+ entry into a skeletal or cardiac myocyte through L-type Ca2+ channels causes conformational changes in the ryanodine receptor that induce the release of large quantities of Ca2+ into the sarcoplasm. Drugs that activate the RyR include caffeine; drugs that inhibit the RyR include dantrolene.
Calcium released into the cytoplasm from the ER is rapidly removed by plasma membrane Ca2+ pumps, and the ER pool of Ca2+ is refilled with extracellular Ca2+ flowing through store-operated Ca2+ channels (SOC) in the plasma membrane. These currents are termed Ca2+ release-activated currents, or ICRAC. The mechanism by which ER store depletion opens the store-operated channels requires two proteins, the channel itself, termed Orai1, and an ER sensor termed STIM1. The Orai1 channel is a 33-kDa protein with four membrane-spanning helices and no homology with other Ca2+ channels (Prakriya, 2009). Orai1 is highly selective for Ca2+. The C terminal end of the channel contains coiled-coil domains thought to interact with the STIM1 sensor. STIM1 is a 77-kDa protein containing a Ca2+ sensor domain termed an EF hand. This domain is located at the N-terminus of the protein on the inside of the ER membrane before the single membrane-spanning domain. There are multiple protein-protein interaction motifs in the middle and C-terminal end of the molecule. Specifically, there are two coiled-coil domains on the C-terminal side of the transmembrane domain in STIM1 that may interact with coiled-coil domains in the Orai1 channel. Under resting conditions, the STIM1 protein is uniformly distributed on the ER membrane. Release of Ca2+ from the ER stores results in dimerization of STIM1 and movement to the plasma membrane where STIM1 and Orai1 form clusters, opening the Ca2+ pore of Orai1 and refilling of the ER Ca2+ pool (Fahrner et al., 2009).
The lipid bilayer of the plasma membrane is impermeable to anions and cations, yet changes in the flux of ions across the plasma membrane are critical regulatory events in both excitable and non-excitable cells. To establish and maintain the electrochemical gradients required to maintain a membrane potential, all cells express ion transporters for Na+, K+, Ca2+, and Cl−. For example, the Na+,K+-ATPase pump expends cellular ATP to pump Na+ out of the cell and K+ into the cell. The electrochemical gradients thus established are used by excitable tissues such as nerve and muscle to generate and transmit electrical impulses, by non-excitable cells to trigger biochemical and secretory events, and by all cells to support a variety of secondary symport and antiport processes (Chapter 5).
Passive ion fluxes down cellular electrochemical gradients are regulated by a large family of ion channels located in the membrane. Humans express ~232 distinct ion channels to precisely regulate the flow of Na+, K+, Ca2+, and Cl− across the cell membrane (Jegla et al., 2009). Because of their roles as regulators of cell function, these proteins are important drug targets. The diverse ion channel family can be divided into subfamilies based on the mechanisms that open the channels, their architecture, and the ions they conduct. They can also be classified as voltage-activated, ligand-activated, store-activated, stretch-activated, and temperature-activated channels. Examples of channels that are major drug targets are detailed next.
Voltage-Gated Channels. Humans express multiple isoforms of voltage-gated channels for Na+, K+, Ca2+, and Cl− ions. In nerve and muscle cells, voltage-gated Na+ channels are responsible for the generation of robust action potentials that depolarize the membrane from its resting potential of −70 mV up to a potential of +20 mV within a few milliseconds. These Na+ channels are composed of three subunits, a pore-forming α subunit and two regulatory β subunits. The α subunit is a 260 kDa protein containing four domains that form a Na+ ion-selective pore by arranging into a pseudo-tetramer shape. The β subunits are ~36 kDa proteins that span the membrane once (Figure 3–9A). Each domain of the α subunit contains six membrane-spanning helices (S1-S6) with an extracellular loop between S5 and S6, termed the pore-forming or P loop; the P loop dips back into the pore and, combined with residues from the corresponding P loops from the other domains, provides a selectivity filter for the Na+ ion. Four other helices surrounding the pore (one S4 helix from each of the domains) contain a set of charged amino acids that form the voltage sensor and cause a conformational change in the pore at more positive voltages leading to opening of the pore and depolarization of the membrane (Purves and McNamara, 2008). The voltage-activated Na+ channels in pain neurons are targets for local anesthetics such as lidocaine and tetracaine, which block the pore, inhibit depolarization, and thus block the sensation of pain. They are also the targets of the naturally occurring marine toxins, tetrodotoxin and saxitoxin (Chapter 20). Voltage-activated Na+ channels are also important targets of many drugs used to treat cardiac arrhythmias (Chapter 29).
Schematic diagram of two types of ion channels regulated by receptors and drugs. A. Diagram of a voltage-activated Na+ channel with the pore in the open and closed state. The P loops are shown in blue, angled into the pore to form the selectivity filter. The S4 helices forming the voltage sensor are shown in orange, with the positively charged amino acids displayed as red dots. B. Ligand-gated nicotinic acetylcholine receptor expressed in the skeletal muscle neuromuscular junction. The pore is made up of five subunits, each with a large extracellular domain and four transmembrane helices (one of these subunits is shown at the left of panel B). The helix that lines the pore is shown in blue. The receptor is composed of 2 α subunits, and β, γ, and δ subunits. See text for discussion of other ligand-gated ion channels. Detailed descriptions of specific channels are given throughout the text in relation to the therapeutic actions of drugs affecting these channels (see especially Chapters 11, 14 and 20). (Adapted with permission from Purves, D, Augustine, GJ, Fitzpatrick, D, Hall, WC, LaMantia, AS, McNamara, JO, and White, LE (eds). Neuroscience, 4ed. Sinauer Associates, Inc., 2008.)
Voltage-gated Ca2+ channels have a similar architecture to voltage-gated Na+ channels with a large α subunit (four domains of six membrane-spanning helices) and three regulatory subunits (the β, δ and γ subunits). There are multliple isoforms of these channels that are widely expressed in nerve, cardiac and smooth muscle cells. Ca2+ channels can be responsible for initiating an action potential (as in the pacemaker cells of the heart), but are more commonly responsible for modifying the shape and duration of an action potential initiated by fast voltage-gated Na+ channels (Purves and McNamara, 2008). These channels initiate the influx of Ca2+ that stimulates the release of neurotransmitters in the central, enteric, and autonomic nervous systems, and that control heart rate and impulse conduction in cardiac tissue (Chapters 8, 14, 27). The L-type voltage-gated Ca2+ channels are subject to additional regulation via phosphorylation by PKA. Thus, when the sympathetic nervous system releases norepinephrine onto β adrenergic receptors in cardiac tissue, raising cAMP and activating PKA, the phosphorylated L-type channels allow more Ca2+ to flow into the cytoplasm, increasing the force of contraction. Voltage-gated Ca2+ channels expressed in smooth muscle regulate vascular tone; the intracelluar concentration of Ca2+ is critical to regulating the phosphorylation state of the contractile apparatus via the activity of the Ca2+/calmodulin-sensitive myosin light chain kinase. Accordingly, the Ca2+ channel antagonists such as nifedipine, diltiazem, and verapamil are effective vasodilators and are widely used to treat angina, cardiac arrhythmias, and hypertension.
Voltage-gated K+ channels are the most numerous and structurally diverse members of the voltage-gated channel family. Humans express ~78 distinct K+ channels and nearly all of them are voltage-gated (Jegla et al., 2009). The voltage-gated Kv channels form channels as tetramers with topology similar to the Na+ and Ca2+ channels, but rather than having four domains, they consist of four separate subunits that each incorporate six membrane-spanning domains. The inwardly rectifying K+ channel subunits only contain two membrane-spanning domains surrounding the pore. In each of these cases, the native channel is a tetramer formed from four individual subunits. A final group of K+ channels is the tandem or two-pore domain "leak" K+ channels; each subunit in this group has four membrane-spanning domains surrounding two P loops, and these form channels as dimers. The inwardly rectifying channels and the two-pore channels are voltage insensitive and are regulated by G proteins and H+ ions and are greatly stimulated by general anesthetics. All these channels are expressed in nerve, cardiac tissue, skeletal and smooth muscle, as well as non-excitable tissues. Increasing K+ conductance through these channels drives the membrane potential more negative; thus, these channels are important in regulating resting membrane potential and resetting the resting membrane at −70 to −90 mV following depolarizaion. Some forms of epilepsy are caused by natural mutations in Kv channels, and drugs such as retigabine that favor opening of Kv channels are under study for the treatment of epilepsy (Rogawski and Bazil, 2008). The cardiac KCNH2 channel, known as hERG (human ether-a-go-go-related gene), is responsible for hereditary as well as acquired (drug-induced) long QT syndrome. It is also the primary target of many anti-arrhythmic drugs that prolong repolarization.
Ligand-Gated Channels. Channels activated by the binding of a ligand to a specific site in the channel protein have a diverse architecture and set of ligands. Major ligand-gated channels in the nervous system are those that respond to excitatory neurotransmitters such as acetylcholine or glutamate (or agonists such as AMPA and NMDA) and inhibitory neurotransmitters such as glycine or γ-aminobutyric acid (GABA) (Purves and McNamara, 2008). Activation of these channels is responsible for the majority of synaptic transmission by neurons both in the CNS and in the periphery (Chapters 8, 11, 14). In addition, there are a variety of more specialized ion channels that are activated by intracellular small molecules, and are structurally distinct from conventional ligand-gated ion channels. These include ion channels that are formally members of the Kv family, such as the hyperpolarization and cAMP-gated (HCN) channel expressed in the heart (Wahl-Schott and Biel, 2009) that is responsible for the slow depolarization seen in phase 4 of AV and SA nodal cell action potentials (Chapter 29), and the cyclic nucleotide-gated (CNG) channels important for vision (Chapter 64). The intracellular small molecule category of ion channels also includes the IP3-sensitive Ca2+ channel responsible for release of Ca2+ from the ER and the sulfonylurea "receptor" (SUR1) that associates with the Kir6.2 channel to regulate the ATP-dependent K+ channel (KATP) in pancreatic beta cells. The KATP channel is the target of oral hypoglycemic drugs such as sulfonylureas and meglitinides that stimulate insulin release from pancreatic β cells and are used to treat type 2 diabetes (Chapter 43). Other specialized channels include the 5-HT3-regulated channel expressed on afferent vagal nerves that stimulates emesis. Ondansetron is an important antagonist of the 5-HT3-gated channel used to inhibit emesis caused by drugs or disease (Chapter 46).
The nicotinic acetylcholine receptor provides an instructive example of a ligand-gated ion channel. Isoforms of this channel are expressed in the CNS, autonomic ganglia and at the neuromuscular junction (Figure 3–9B). The pentameric channel consists of four different subunits (2α, β, δ, γ) in the neuromuscular junction or two different subunits (2α, 3β) in autonomic ganglia (Purves and McNamara, 2008). Each α subunit has an identical acetylcholine binding site; the different compositions of the other three subunits between the neuronal and neuromuscular junction receptors accounts for the ability of competitive antagonists such as rocuronium to inhibit the receptor in the neuromuscular junction without effect on the ganglionic receptor. This property is exploited to provide muscle relaxation during surgery with minimal autonomic side effects (Chapter 11). Each subunit of the receptor contains a large, extracellular N-terminal domain, four membrane-spanning helices (one of which lines the pore in the assembled complex), and an internal loop between helices 3 and 4 that forms the intracellular domain of the channel. The pore opening in the channel measures ~3 nm whereas the diameter of a Na+ or K+ ion is only 0.3 nm or less. Accordingly, ligand-gated ion channels do not possess the exquisite ion selectivity found in most voltage-activated channels and activation of the nicotinic acetylcholine receptor allows passage of both Na+ and K+ ions.
The major excitatory transmitter at CNS synapses is glutamate. There are three types of ionotropic glutamate receptors (AMPA, NMDA, and kainate), named after the ligands that selectively activate them. They have a topology similar to that of the nicotinic acetylcholine receptor: the channel is made up of five subunits organized with a large extracellular region, a pore, and a small intracellular face. Activation of these channels with glutamate markedly increases Na+ and K+ conductance leading to depolarization. NMDA receptors are less ion-selective; activation increases Na+, K+, and Ca2+ conductance, with the Ca2+ signal being used for additional signaling events.
Over one-third of synapses in the brain are inhibitory; the major inhibitory transmitters are glycine and γ-aminobutyric acid (GABA). Glycine and ionotropic GABAA receptors have a topology like that of the glutamate and nicotinic acetylcholine receptors, with five subunits (α, β, γ, δ and ρ), a ligand-binding domain, and pore-forming helices. Activation of these channels increases Cl− conductance, which hyperpolarizes the cell membrane and inhibits excitability (Purves and McNamara, 2008). GABAA receptors are targets of important sedative-hypnotic drugs such as the benzodiazepines and barbituates, and are also important in the mechanisms of ethanol and general anesthetics (Chapters 17, 19, 23).
TRP Channels. The transient receptor potential (TRP) channels comprise a superfamily of ubiquitously expressed ion channels that is remarkable in its diversity and domain strucutre (Clapham 2003; Venkatachalam and Montell, 2007). Although the TRP channels are not presently targets of approved drugs, there is significant interest in developing drugs that can alter the function of these ion channels because of their roles in various sensory phenomena such as pain, temperature, osmolarity, touch, olfaction, vision, and hearing. Because these channels contain multiple domains, they can act as signal integrators and most can be activated by multiple mechanisms. There are 27 TRP channel genes in humans, representing six different TRP channel families. TRP channels contain six membrane-spanning segments and the functional ion channels consist of tetrameric complexes. Closely related TRP channels can form heterotetramers. The TRP channels are cation channels, but as with other heteromultimeric ion channels, the subunit composition of the multimeric channels can prescribe a number of important channel characteristics, including ion selectivity and activation properties. The intracellular domains of TRP channels can include ankyrin domains, protein kinase domains, and ADP-ribose pyrophosphatase domains. Mutations in TRP channels are known to cause several disease including hypomagnesemia and hypocalcemia, and various renal disorders and neurodegenerative diseases.
Transmembrane Receptors Linked to Intracellular Enzymes
Mammalian cells express a diverse group of physiological membrane receptors with extracellular ligand-binding domains and an intrinsic enzymatic activity on the cytoplasmic surface of the cell. These molecules include the receptor tyrosine kinases (RTKs) such as the epidermal growth factor (EGF) and insulin receptors, which contain intrinsic tyrosine kinases in the cytoplasmic domain of the receptor; tyrosine kinase-associated receptors without enzymatic activity, such as the receptors for γ-interferon, which recruit the cytoplasmic Janus tyrosine kinases (JAKs); receptor serine-threonine kinases such as the TGF-β receptor; and receptors linked to other enzyme activities such as the receptors for natriuretic pepides, which have a cytoplasmic guanylate cyclase activity that produces a soluble second messenger, cyclic GMP (see the next section). Receptors responsible for innate immunity, the Toll-like receptors and those for tumor necrosis factors (TNF-α), have many features in common with the JAK-STAT receptors.
Receptor Tyrosine Kinases. The receptor tyrosine kinases include receptors for hormones such as insulin, for multiple growth factors such EGF, platelet-derived growth factor (PDGF), nerve growth factor (NGF), fibroblast growth factor (FGF), vascular endothelial growth factor (VEGF), and ephrins. With the exception of the insulin receptor, which has α and β chains (Chapter 43), these molecules consist of single polypeptide chains with large, cysteine-rich extracellular domains, short transmembrane domains, and an intracellular region containing one (or in some cases two) protein tyrosine kinase domains. Activation of growth factor receptors leads to cell survival, cell proliferation, and differentiation. Activation of the ephrin receptors leads to neuronal angiogenesis, axonal migration, and guidance (Ferguson, 2008; Hubbard and Till, 2000).
The inactive state of growth factor receptors is monomeric; binding of ligand induces dimerization of the receptor and cross-phosphorylation of the kinase domains on multiple tyrosine residues. Some of these tyrosine residues are in the activation loop of the kinase and their phosphorylation serves to further activate the receptor kinase (Figure 3–10A). The phosphorylation of other tyrosine residues forms docking sites for the SH2 domains contained in a large number of signaling proteins (Ferguson, 2008). There are over 100 proteins encoded in the human genome containing SH2 domains, and following receptor activation, large signaling complexes are formed on the receptor that eventually lead to cell proliferation. Molecules recruited to phosphotyrosine-containing proteins by their SH2 domains include PLCγ which raises intracellular levels of Ca2+ and activates PKC, as described earlier. The α and β isoforms of phosphatidylinositiol 3-kinase (PI3-K) contain SH2 domains, dock at the phosphorylated receptor, are activated, and increase the level of phosphatidylinositol 3,4,5 trisphosphate (PIP3), a molecule that forms other kinds of docking sites at the plasma membrane for signaling molecules such as protein kinase B (PKB, also known as Akt). PKB can regulate the mammalian target of rapamycin (mTOR) in the various signaling pathways and the Bad protein that is important in apoptosis.
In addition to recruiting enzymes, phosphotyrosine-presenting proteins can attract SH2 domain-containing adaptor molecules without activity such as Grb2, which in turn attract guanine nucleotide exchange factors (GEFs) such as Sos that can activate the small GTP-binding protein, Ras. The small GTP binding proteins Ras and Rho belong to a large family of small monomeric GTPases; only members of the Ras and Rho subfamilies are activated by extracelluar receptors. The Ras family includes four isoforms H-ras, K-ras, n-Ras, and Rheb (activated by the insulin receptor). Spontaneous activating mutations in Ras are responsible for a large fraction of human cancers; thus, molecules that inhibit Ras are of great interest in cancer chemotherapy. The Rho family includes Rho, Rac, and Cdc42, which are responsible for relaying signals from the membrane to the cytoskeleton. All of the small GTPases are activated by GEFs that are regulated by a variety of mechanisms and inhibited by GTPase activating proteins (GAPs) (Etienne-Manneville and Hall, 2002).
Activation of members of the Ras family leads in turn to activation of a protein kinase cascade termed the mitogen-activated protein kinase (MAP kinase or MAPK) pathway. Activation of the MAPK pathway is one of the major routes used by growth factor receptors to signal to the nucleus and stimulate cell growth. The first enzyme in the pathway is Rap which is a MAP kinase kinase kinase (MKKK). Rap phosphorylates and activates a MAP kinase kinase (MKK) termed MEK. MEK phosphorylates a MAP klnase termed ERK. ERK is an interesting member of the kinase family; its activation is achieved by phosphorylation of closely spaced tyrosine and threonine residues in the kinase activation loop. ERK phosphorylates a number of transcription factors in the nucleus, including Elk-1 and CREB, to regulate gene transcription and cause cell proliferation (Manning and Davis, 2003). Drugs that act at receptors in this diverse family include insulin for the treatment of diabetes mellitus and imatinib, a small molecule protein kinase inhibitor designed to inhibit both receptor and non-receptor tyrosine kinases. Imatinib is used to treat chronic myelogenous leukemia and several solid tumors with dysregulated tyrosine kinases.
Diagram showing the mechanism of activation of a receptor tyrosine kinase and a cytokine receptor. A. Activation of the EGF receptor. The extracellular structure of the unliganded receptor (a) contains four domains (I-IV), which rearrange significantly upon binding two EGF molecules. (b). The conformational changes lead to activation of the cytoplasmic tyrosine kinase domains and tyrosine phosphorylation of intracellular regions to form SH2 binding sites. (c). The adapter molecule Grb2 binds to the phosphoryated tyrosine residues and activates the Ras-MAP kinsase cascade. B. Activation of a cytokine receptor. Binding of the cytokine causes dimerization of the receptor and recruits the Janus Kinases (JAKs) to the cytoplasmic tails of the receptor. JAKs trans-phosphorylate and lead to the phosphorylation of the signal transducers and activators of transcription (STATs). The phosphorylated STATS translocate to the nucleus and regulate transcription. There are proteins termed suppressors of cytokine signaling (SOCS) that inhibit the JAK-STAT pathway (Alexander and Hilton, 2004).
JAK-STAT Receptor Pathway. Cells express a family of receptors for cytokines such as γ-interferon and hormones like growth hormone and prolactin, which signal to the nucleus by a more direct manner than the receptor tyrosine kinases. These receptors have no intrinsic enzymatic activity, rather the intracellular domain binds a separate, intracellular tryosine kinase termed a Janus kinase (JAK). Upon the dimerization induced by ligand binding, JAKs phosphorylate other proteins termed signal transducers and activators of transcription (STATs), which translocate to the nucleus and regulate transcription (Figure 3–10B). The entire pathway is termed the JAK-STAT pathway (Gough et al., 2008; Wang et al., 2009). There are four JAKs and six STATs in mammals which, depending on the cell type and signal, combine differently to activate gene transcription. For example, prolactin appears to use JAK1, JAK2, and STAT5 to stimulate milk production.
Receptor Serine-Threonine Kinases. Protein ligands such as TGF-β activate a family of receptors that are analogous to the receptor tyrosine kinases except that they have a serine/threonine kinase domain in the cytoplasmic region of the protein. There are two isoforms of the monomeric receptor protein, type I (7 forms) and type II (5 forms). In the basal state, these proteins exist as monomers; upon binding an agonist ligand, they dimerize, leading to phosphorylation of the kinase domain of the type I monomer, which activates the receptor. The activated receptor then phosphorylates a gene regulatory protein termed a Smad. There are multiple Smads in cells; once phosphorylated by the activated receptor on a serine residue, Smad dissociates from the receptor, migrates to the nucleus, associates with transcription factors and regulates genes leading to morphogenesis and transformation. There are also inhibitory Smads (the Smad6 or Smad7 isoforms) that compete with the phosphorylated Smads to terminate signaling.
Toll-Like Receptors. Signaling related to the innate immune system is carried out by a family of over ten single membrane-spanning receptors termed Toll-like receptors (TLR), which are highly expressed in hematopoeitic cells. In a single polypeptide chain, these receptors contain a large extracellular ligand-binding domain, a short membrane-spanning domain, and a cytoplasmic region termed the TIR domain that lacks intrinsic enzymatic activity. Ligands for TLR are comprised of a multitude of pathogen products including lipids, peptidoglycans, lipopeptides, and viruses. Activation of these receptors produces an inflammatory response to the pathogenic microorganisms. As with all single membrane-spanning receptors, the first step in activation of TLR by ligands is dimerization, which in turn causes signaling proteins to bind to the receptor to form a signaling complex.
Ligand-induced dimerization recruits a series of adaptor proteins including Mal and the myeloid differentiation protein 88 (MyD88) to the intracellular TIR domain, which in turn recruit the interleukin-associated kinases termed IRAKs. The IRAKs autophosphorylate in the complex and subsequently form a more stable complex with MyD88. The phosphorylation event also recruits TRAF6 to the complex, which facilitates interaction with a ubiquitin ligase that attaches a polyubiquitin molecule to TRAF6. This complex can now interact with the protein kinase TAK1 and the adaptor TAB1. TAK1 is a member of the MAP kinase family, which activates the NF-κB kinases; phosphorylation of the NF-κB transcription factors causes their translocation to the nucleus and transcriptional activation of a variety of inflammatory genes (Gay and Gangloff, 2007).
TNF-α Receptors. The mechanism of action of tumor necrosis factor α (TNF-α) signaling to the NF-κB transcription factors is very similar to that used by Toll-like receptors in that the intracellular domain of the receptor has no enzymatic activity. The TNF-α receptor is another single membrane-spanning receptor with an extracellular ligand-binding domain, a transmembrane domain, and a cytoplasmic domain termed the death domain.
TNF-α binds a complex composed of TNF-receptor1 and TNF-receptor2. Upon trimerization, the death domains bind the adaptor protein TRADD, which recruits the receptor interacting protein 1 (RIP1) to form a receptor-adaptor complex at the membrane. RIP1 is poly-ubiquinated, resulting in recruitment of the TAK1 kinase and the IκB kinase (IKK) complex to the ubiquinated molecules (Skaug et al., 2009). The activation loop of IKK is phosphorylated in the complex eventually resulting in IκBα being released from the complex allowing the p50/p65 heterodimer of the complex to translocate to the nucleus and activate the transcription of inflammatory genes (Ghosh and Hayden, 2008; Hayden and Ghosh, 2008; Kataoka, 2009). While there currently are no drugs that interdict the cytoplasmic portions of the TNF-α signaling pathway, humanized monoclonal antibodies to TNF-α itself, such as infiximab and adalimumab, are important for the treatment of rheumatoid arthritis and Crohn's disease (Chapters 35 and 47).
Receptors That Stimulate Synthesis of Cyclic GMP
The signaling pathways that regulate the synthesis of cyclic GMP in cells include hormonal regulation of transmembrane guanylate cyclases such as the atrial natriuretic peptide receptor (ANP) and the activation of soluble forms of guanylate cyclase by nitric oxide (NO). The downstream effects of cyclic GMP are carried out by multiple isoforms of PKG, cyclic GMP-gated ion channels, and cyclic GMP-modulated phosphodiesterases that degrade cyclic AMP (described later).
Natriuretic Peptide Receptors. The class of membrane receptors with intrinsic enzymatic activity includes the receptors for three small peptide ligands released from cells in cardiac tissues and the vascular system. These peptides are atrial natriuretic peptide (ANP), which is released from atrial storage granules following expansion of intravascular volume or stimulation with pressor hormones; brain natriuretic peptide (BNP), which (in spite of its name) is synthesized and released in large amounts from ventricular tissue in response to volume overload; and C-type natriuretic peptide (CNP), which is synthesized in the brain and endothelial cells. Like BNP, CNP is not stored in granules; rather, its synthesis and release are increased by growth factors and sheer stress on vascular endothelial cells (Potter et al., 2009). The major physiological effects of these hormones are to decrease blood pressure (ANP, BNP), to reduce cardiac hypertrophy and fibrosis (BNP), and to stimulate long bone growth (CNP).
The transmembrane receptors for ANP, BNP, and CNP are ligand-activated guanylate cyclases. The ANP and BNP receptors contain a ~450 amino acid extracellular domain that binds the peptide, a short 20 amino acid transmembrane domain, and large intracellular domains that contain a kinase homology region, a dimerization domain, and a C-terminal guanylate cyclase domain. Phosphorylation of serine residues in the kinase domain is important for activity; dephosphorylation of these residues leads to desensitization of the receptor. Ligand binding brings the juxtamembrane regions together and stimulates guanylate cyclase activity (Figure 3–11).
Cyclic GMP signaling pathways. Formation of cyclic GMP is regulated by cell surface receptors with intrinsic guanlyate cyclase activity and by soluble forms of guanylate cyclase (GC). The cell surface receptors respond to natriuretic peptides such as atrial natriuretic peptide (ANP) with an increase in cyclic GMP. Soluble guanylate cyclase responds to nitric oxide (NO) generated from L-arginine by nitric oxide synthase (NOS). Cellular effects of cyclic GMP are carried out by PKG and cyclic GMP-regulated phosphodiesterases (PDEs). In this diagram, NO is produced by a Ca2+/calmodulin-dependent NOS in an adjacent endothelial cell. Detailed descriptions of these signaling pathways are given throughout the text in relation to the therapeutic actions of drugs affecting these pathways.
The ANP receptor (NPR-A) is the molecule that responds to ANP and BNP. The protein is widely expressed and prominent in kidney, lung, adipose, and cardiac and vascular smooth muscle cells. ANP and BNP play a role in maintaining the normal state of the cardiovascular system as NPR-A knockout mice have hypertension and cardiac hypertrophy. A synthetic BNP agonist, nesiritide, is used for treatment of acute decompensated heart failure (Chapter 28). The NPR-B receptor responds to CNP and has a physical structure similar to the NPR-A receptor. It is also widely expressed but prominent in bone, brain, kidney, lung, liver, and cardiac and vascular smooth muscle. A role for CNP in bone is suggested by the observation that NPR-B knockout mice exhibit both dwarfism and cardiac hypertrophy. The natriuretic peptide C receptor (NPR-C) has an extracellular domain similar to those of NPR-A and NPR-B but does not contain the intracellular kinase or guanylate cyclase domains. It has no enzymatic activity and is thought to function as a clearance receptor, removing excess natriuretic peptide from the circulation (Potter et al., 2009).
NO Synthase and Soluble Guanylate Cyclase. Nitric oxide (NO) is a unique signal, a very labile gas produced locally in cells by the enzyme nitric oxide synthase (NOS); the resulting NO is able to markedly stimulate the soluble form of guanylate cyclase to produce cyclic GMP. There are three forms of nitric oxide synthase, neuronal NOS (nNOS or NOS1), endothelial NOS (eNOS or NOS3), and inducible NOS (iNOS or NOS2). All three forms of this enzyme are widely expressed but are especially important in the cardiovascular system, where they are found in myocytes, vascular smooth muscle cells, endothelial cells, hematopoietic cells, and platelets.
NOS produces NO by catalyzing the oxidation of the guanido nitrogen of L-arginine, producing L-citrulline and NO. The enzymes require co-factors including tetrahydrobioptern and calmodulin. The nNOS and eNOS forms of the enzyme are markedly activated by Ca2+/calmodulin; the inducible form is less sensitive to Ca2+ but the level of iNOS protein in cells can be increased over 1000-fold by inflammatory stimuli such as endotoxin, TNF-α, interleukin-1β and interferon-γ. The ability of Ca2+ to activate eNOS and nNOS is important in certain cells where neurotransmitters that open Ca2+ channels or activate PLC can relax smooth muscle. An example is the ability of ACh released by the parasympathetic nervous system to relax sphincters. Soluble guanylate cyclase is a heterodimer composed of α and β subunits. The N-terminal end of the molecule contains a protoporphyrin-IX heme domain. NO binds to this domain at low nM concentrations and produces a 200- to 400-fold increase in the Vmax of guanylate cyclase, leading to a marked elevation of cyclic GMP in the cell (Tsai and Kass, 2009).
The cellular effects of cyclic GMP on the vascular system are mediated by a number of mechanisms, but especially by PKG. For example, in vascular smooth muscle, activation of PKG leads to vasodilation by:
Inhibiting IP3-mediated Ca2+ release from intracellular stores.
Phosphorylating voltage-gated Ca2+ channels to inhibit Ca2+ influx.
Phosphorylating phospholamban, a modulator of the sarcoplasmic Ca2+ pump, leading to a more rapid reuptake of Ca2+ into intracellular stores.
Phosphorylating and opening the Ca2+-activated K+ channel leading to hyperpolarization of the cell membrane, which closes L-type Ca2+ channels and reduces the flux of Ca2+ into the cell (Tsai and Kass, 2009).