G protein-coupled receptor

• Dr. Bertil Hille, University of Washington School of Medicine, Seattle, WA

G-protein coupled receptors (GPCRs) [1]are the largest group of plasma membrane receptors of which rhodopsin and adrenergic receptors are the most familiar. They are integral plasma membrane proteins that transduce signals from extracellular ligands to signals in intracellular relay proteins, the heterotrimeric GTP binding proteins (G proteins). By coupling to many downstream second messengers and effectors, the G proteins initiate pleiotropic changes in many targets. Thus, the extracellular signal is typically amplified to produce robust, varied, and cell-specific responses.

GPCRs are quite distinct from growth-factor receptors (e.g. insulin, NGF, BDNF, ...), which signal through tyrosine kinases, do not use heterotrimeric G proteins, and typically regulate gene expression as well as second-messenger signals. However, GPCRs and growth-factor receptors do share several common final enzymatic pathways of signaling.

Figure 1: A conceptual cartoon of a G-protein coupled receptor in the plasma membrane with the characteristic seven α-helical transmembrane segments. (See Figure 3 for a crystal structure.) The coupled heterotrimeric G-protein is represented schematically by the letters α, β, γ in the cytoplasm. Not to scale since the mass of G-protein is larger than that of the receptor (see Figure 4 for actual dimensions.). Labels: A, agonist in its binding pocket; C, receptor C-terminus in cytoplasm; N, receptor extracellular N-terminus.

Diversity

The GPCR gene family is a major gene family with more than 800 homologous genes in man (Foord et al., 2005). In the central nervous system, the GPCRs include receptors for many neurotransmitters and hormones and for transduction of several sensory inputs, as well as receptors for chemokines [2] and wnt signaling. Especially for discussion of neurotransmitter receptors, neurobiologists like to call the GPCRs "metabotropic receptors" (implying coupling to metabolism) to distinguish them from the ligand-gated ion channels of fast chemical synapses, which they call "ionotropic." Most other fields simply say GPCRs. Common GPCR ligands in the nervous system are:

• Monoamines: adrenaline, noradrenaline, serotonin, dopamine, histamine
• Other small neurotransmitters: Acetylcholine (mACh), gamma aminobutyric acid (GABA_B), glutamate (metabotropic, mGluR), ATP (P2Y), adenosine, cannabinoids. (Where the neurotransmitter has multiple classes of receptors, the receptor that is a GPCR is named in parentheses. ACh, GABA, glutamate, and ATP also do act on ionotropic receptors as well as on the metabotropic receptors discussed here.)
• Many peptide neurotransmitters and hormones: opioids, somatostatin, NPY, oxytocin, vasopressin, neurotensins, VIP, galanin, kinins, releasing hormones, and many more.
• Sensory modalities: light (rhodopsin), odorants, some tastetants including sweet, bitter, and umami

For most of these extracellular stimuli there are multiple different receptors, and the same extracellular signal can give rise to different intracellular responses depending on the receptor subtypes expressed--and even the splice variant expressed. For example there are nine subtypes (genes) of receptors for adrenaline and noradrenaline--the adrenergic receptors. Three of them couple to the G-protein G_q , often making calcium signals (α1 adrenergic receptors), three of them couple to G_i, often inhibiting adenylyl cyclase, activating GIRK channels, or inhibiting Ca channels (α2 adrenergic receptors), and three of them couple to G_s, often stimulating adenylyl cyclase (β adrenergic receptors). Since the adrenergic receptors are more closely related to each other than to other GPCRs, one can deduce that a prototype adrenergic receptor evolved early that through further gene duplications subsequently diversified its repertoire of G-protein coupling. GPCR signaling systems specialized for different ligands are present in yeast, protozoa, plants, and animals--thus in all eukaryotes. In other eukaryotes, the G proteins are different with different functions than in animals.

GPCRs have been classified in several ways. On the basis of sequence and structural similarities, GPCRs important for the nervous system and their ligands include:

• Class 1 or A (the very large rhodopsin family): all the ligands listed above except the few itemized below.
• Class 2 or B (the secretin family): secretin, VIP, PACAP, GHRH
• Class 3 or C (the glutamate family): mGluR, GABA_B, CaSR (calcium-sensing), a few possible taste receptors
• Class F (the frizzled family): frizzled (the wnt receptor)

Signaling from GPCRs

GPCRs couple to GTP-binding heterotrimeric G proteins [3] that consist of Gα, Gβ, and Gγ subunits (Figure 1). At rest these three subunits are assembled into a complex, Gαβγ. Since Gβ and Gγ are inseparable once coassembled, it is customary to talk about a resting complex of Gα with Gβγ. Typically, both Gα and Gβγ are held at the inner leaflet of the plasma membrane by hydrophobic lipid modifications. Thus they are not transmembrane proteins but nevertheless are mainly anchored to the cytoplasmic side of membranes by one or two lipid moieties. Unlike the schematic drawing in Figure 1, the heterotrimeric G protein is physically larger than the GPCR (typical molecular weights: Gαβγ ~90KDa, GPCRs, ~50KDa). The heterotrimeric G proteins should be distinguished from the small, low molecular weight, monomeric G proteins of the Ras, Rab, Rho, etc. families with which the Gα subunit does share significant homology. The monomeric G proteins have the GDP-binding, GTP exchange, and GTPase properties of Gα subunits of heterotrimeric G proteins, and the both use the bound GTP to time their active signaling life time. But the monomeric G proteins do not couple to GPCRs or to Gβγ, and they signal to different downstream effectors.

Figure 2: Guanine nucleotide cycle of G proteins. The inactive state is Gα-GDP. GPCRs catalyze loss of GDP in exchange for GTP to make the active state, Gα-GTP. The active GTP-bound form has a slow GTPase activity that restores the GDP-bound state and turns off activity.

Gα subunits are flexible signaling proteins. In the inactive resting complex, the Gα subunit is bound to the guanine nucleotide, guanosine diphosphate (GDP), but when a receptor is activated, the receptor can catalyze nucleotide exchange reactions on the Gα subunit (Figure 2). The GDP binding site opens, GDP leaves and the more abundant guanosine triphosphate GTP enters instead. GTP binding activates the G protein (hence the name). Current research seeks to understand the molecular interactions between receptor and G protein and how the conformational energy of ligand binding results in GDP-GTP exchange. In classical teaching, the Gα-GTP-Gβγ complex is unstable so that the active Gα-GTP and Gβγ separate from one another and from the receptor as well, but they usually remain attached to the plasma membrane by their lipid anchors. Thus there are now three active products that couple to downstream effectors: the activated receptor, the Gα-GTP subunit, and the Gβγ dimer. However, in some examples it is believed that the quaternary active G-protein-GTP complex remains undissociated although conformationally changed, and in some other examples the Gβγ dimer may leave the membrane and go into the cytoplasm or to other membranes. Because the activated G proteins are generally membrane associated, the next step usually is an interaction with membrane-associated effector proteins or recruitment of cytoplasmic effector proteins to the membrane. We now consider signaling from Gα-GTP and Gβγ separately.

Downstream coupling of Gα-GTP

There are five canonical signaling pathways

The downstream effects of stimulating a GPCR depend on which G protein type(s) it couples to. Although there are many receptors, fortunately there are only a few canonical G protein types, so discussion of downstream signaling is easier. Heterotrimeric G proteins are named by the type of α subunit they contain, and there are nearly 20 genes encoding Gα subunits. However, for understanding much of their signaling it suffices to consider five broad Gα families and five signaling pathways that they most typically activate. Similarly there are multiple genes for Gβ and Gγ subunits. These different subunits can combine with any of the Gαsubunits but functionally they are nearly the same. Thus the principal signaling from thousands of agonists acting on almost a thousand receptors can be simplified to a discussion of these five families of Gα proteins.

1. Gα_s , Gα_olf : (s for "stimulatory", olf for "olfactory") the first discovered, activates plasma membrane adenylyl cyclases, increasing the cytosolic second messenger cyclic AMP (cAMP), which e.g. stimulates phosphorylation of target proteins by cAMP-dependent protein kinase. Gα_s and its downstream signaling can be covalently activated by cholera toxin. The severe watery diarrhea in cholera can be traced directly to the elevation of cAMP in gut epithelial cells.
2. Gα_i , Gα_o : (i for "inhibitory" and o for "other") inhibit most adenylyl cyclases, allowing cellular cAMP to fall. Gα_i and Gα_o can be covalently inactivated and their signaling turned off by Pertussis toxin. Gα_o is said to constitute 1% of brain proteins. Being so abundant, the PTX-sensitive G proteins are also the principal source of active Gβγ subunits.
3. Gα_q , Gα₁₁ : activate phospholipase Cβ (PLCβ), a lipase that cleaves the signaling phosphoinositide lipid (PI(4,5)P₂) of the plasma membrane, generating several second messengers including inositol trisphosphate (IP₃) that releases Ca²⁺ from intracellular stores and diacylglycerol that activates phosphorylation by protein kinase C. A third pathway of signaling from Gα_q arises because the low-abundance PI(4,5)P₂ lipid is an essential cofactor for some plasma membrane ion channels and transporters. This lipid can become depleted when PLCβ is activated strongly, changing the function of the lipid-sensitive proteins.
4. Gα₁₂ , Gα₁₃ : enhance RhoA, Rho kinase, change expression of some genes, and slow dephosphorylation of myosin light chain
5. Gα_transducin, Gα_gustducin: activate cyclic GMP (cGMP) phosphodiesterase (transducin) that cleaves and depletes cytoplasmic cGMP (in the retina only) or cAMP phosphodiesterase (gustducin) that cleaves and depletes cAMP (in some taste receptors)

Of these, the G_s , G_i , and G_q pathways are the most commonly encountered in cell biology, neurobiology, and endocrinology. All of these pathways involve second messengers and effector enzymes. The long cascades of signaling may take up to tens of seconds to be completed. However in a few cases, such as vision using rhodopsin and transducin, an extreme high density of both molecules, very tight compartmentalization, and miniaturization of the geometry have allowed responses that take only tens of milliseconds. When we watch the world, vision is so fast that we are not aware of any time delay.

Downstream coupling of Gβγ

The Gβγ subunits also are potent plasma membrane signals. They bind to several effectors. They activate G-protein coupled inwardly rectifying K⁺ (GIRK) channels. They inhibit opening of several voltage-gated Ca²⁺ channels of the Ca_V family. They bind to the SNARE complex of the exocytotic machinery in synapses and reduce exocytosis of neurotransmitters. In neurobiology, the latter two signaling actions provide a major component of receptor-dependent presynaptic inhibition by reducing the entry of Ca²⁺ and by blocking exocytosis of neurotransmitter. In addition Gβγ dimers act directly on at least two more downstream enzyme effectors: they stimulate PLCβ and phosphoinositide 3-kinase γ (PI3Kγ).

Although there are numerous Gβ and Gγ genes, to a first approximation, the Gβγ populations associated with all Gα subunits are regarded as similar and equivalent. By mass action, the strength of Gβγ signaling to effectors is probably greatest when launched by the most abundant Gα subunits (e.g. G_o in neurons) or when derived from types of Gα subunits that dissociate most readily from their Gβγ partners.

Termination of GPCR signaling

Termination of signaling requires turning off activated receptors, turning off activated G-proteins, and return of second messenger levels, protein phosphorylation levels, and other changed metabolites to their original values. Consider the first two. Receptors quickly deactivate upon removal and unbinding of agonist ligand. They are also inactivated by other processes even while ligand is still present, mechanisms that prevent over-stimulation and are usually called receptor desensitization. In one canonical shutdown pathway (Gainetdinov et al., 2004), activated receptors (receptors bound to agonist) are recognized and phosphorylated by G-protein coupled receptor kinases (GRKs). Phosphorylated receptors may be intrinsically less active, and they can be turned off fully by binding of [[arrestins] [4]] at the plasma membrane. The arrestin-receptor complex may be unable to couple to downstream G proteins, but it can mediate other signaling such as recruiting MAP kinase cascades, and it may be endocytosed (clathrin-mediated), removing the receptor entirely from the cell surface, a true down regulation of receptor protein.

Signaling by activated Gα-GTP is terminated by GTP hydrolysis, a reaction catalyzed by the Gα subunit itself that yields the inactive form, Gα-GDP (Figure 2). Thus activated G-proteins have an intrinsic self-timer function that terminates their activity. Gα-GDP in turn is a scavenger that binds any free Gβγ dimers, re-forming the inactive heterotrimeric G protein GαβγGDP. The speed of GTP hydrolysis can be accelerated by proteins that act as GTPase acceleratory proteins (GAPs). In addition, sometimes the effector proteins are GAPS so that activated G-proteins become inactivated more rapidly once they make productive interactions with effectors. For example, PLC is a powerful GAP for Gq-GTP (Ross, 2011). In addition, the GTPase activity is speeded by another class of cytoplasmic proteins called regulators of G-protein signaling (RGS proteins) that act as GAPS and sometimes impede activation of downstream effectors (Hollinger and Hepler, 2002).

Signaling without G proteins

GPCR activation can evoke signal pathways that do not require coupling to G proteins. Much as activated receptor conformations are recognized by their complementary G proteins (the classical route), so too they can be recognized by other signaling proteins including the GRKs, arrestins, JAK, Src family kinases, and PDZ-domain containing proteins (non-classical signaling, Sun et al., 2007). Such coupling can evoke clathrin-mediated internalization, activation of MAP (mitogen-activated protein) kinase cascade signaling, or stimulation of Na⁺/H⁺ exchange.

Specificity of GPCR signaling

Most GPCRs couple primarily to Gα subunits of only one of the five signaling families listed earlier. Some receptors, termed promiscuous, couple to several, and some agonists termed biased agonists can switch the G-protein specificity of a response. In any case, there would seem to be a large apparent loss of specificity if 1000 types of receptors can couple only to 5 signaling pathways. However, specificity is achieved in several other ways.

1. Each cell type expresses only a subset of the available receptors (perhaps 40-80 kinds of receptors in different ratios), meaning that each agonist speaks to specific appropriate cells and not to others. In this way, light stimulates photoreceptors and GnRH stimulates pituitary gonadotropes.
2. Each cell expresses a specific subset of downstream protein targets that are responsive to the second messengers that GPCRs regulate. Therefore, with the same second messenger each cell has a different response. In response to cAMP, Leydig cells of the testis make testosterone, horizontal cells of the retina decrease their electrical coupling, cardiac pacemaker cells speed the cardiac beat rate, and vascular smooth muscle relaxes.
3. GPCRs may be localized to certain parts of the plasma membrane, although less dramatically than for many other membrane proteins. In that way it would be possible in principle in neurons for dendrites, cell bodies, axons, and nerve terminals each to give different responses to the same set of GPCR agonists. Evidence for strong localization is mostly lacking except in special cases such as the localization of rhodopsin to discs of vertebrate photoreceptors.
4. Finally significant convergence can be desirable. It allows diverse cells and organs to call for a common physiological response that they may need. Thus an increase in heart rate and blood circulation can be initiated by noradrenaline from sympathetic nerves for flight-or-fight, by glucagon from the pancreas to aid in energy distribution, or by histamine from mast cells to accompany allergic reactions.

As many GPCR agonists are released at nerve terminals and varicosities, one might suppose that the corresponding receptors would be highly localized to immediately opposite “postsynaptic” membranes of a "target" cell. This concept generalizes from the organization of typical fast chemical synapses where presynaptic ACh, glutamate, GABA, or glycine release delivers agonist directly to postsynaptic receptors within nanometers of the release site, opening an ion channel in one postsynaptic neuron within a fraction of a millisecond. Such directed agonist action in a fast chemical synapse stops in a few milliseconds because agonist is quickly removed from the synaptic cleft by enzymes or transporters. However, GPCR signaling is fundamentally different because GPCR agonists typically have an extracellular lifetime of 200 ms to several minutes. In this time, the agonist spreads by diffusion and acts on many cells.

Figure 3: Crystal structure of the β1-adrenergic receptor. Transmembrane helical segments are numbered 1-7, where 1 should connect to the extracellular N-terminus; however in this structure the receptor N-terminus was truncated and is absent. An antagonist, cyanopindolol is shown in the agonist binding site (purple stick structure). A bold arrow points to the future site of an intracellular crevice where part of the Gα subunit would be able to inset into the receptor if agonist were bound from the outside. C is the C-terminus. Coordinates 2VT4 from T. Warne et al. (2008).

The signal necessarily acts in a volume on numerous cells rather than conveying point-to-point messages to a single selected cell.

Thus, although it is common to call all nerve-released agonists neurotransmitters independent of whether they act on fast chemical synapses or GPCRs, we summarize below how the actions on these two classes of receptors are quite different and should not be confused: Correspondingly, some GPCRs are found more diffusely over the cell surface with less special reference to "postsynaptic" sites. The concept that neurotransmitters may spread beyond a single synapse is called spillover and volume transmission. This is probably the normal mode of action of peptide and monoamine neurotransmitters. They affect the mode of operation of neural circuits (even mental state) in a paracrine, hormone-like manner rather than providing specific information to one neuron.

1. Unlike fast ligand-gated receptors, GPCRs are not ion channels.
2. GPCR actions take 100 ms to minutes. Fast chemical synapses signal in a fraction of a millisecond.
3. GPCRs always evoke complex pleiotropic responses typically involving G proteins, second messengers, and numerous intracellular targets. Fast chemical synaptic receptors primarily change the membrane potential and sometimes admit calcium ions into the cell.
4. The GPCR coupled monoamines and peptides have longer extracellular lifetimes and thus cannot be targeted for point-to-point wiring to a single postsynaptic cell in a circuit. They work on larger groups of cells.

Structure

All GPCRs have a membrane topology with seven α-helical transmembrane segments, N-terminus outside and C-terminus inside (Figure 1). Many of them have a lipid modification, palmitoylation near the C-terminus that anchors them further in the membrane. X-ray crystal structures, available so far for several dozen GPCRs, show that the transmembrane segments are broken helices that cross the membrane at various angles (Figure 3). With this many structures, common structural features can be deduced (Venkatakrishnan et al., 2013).

Figure 4: Two views of the β2 receptor (green) activated by and enveloping an agonist (colors in upper part) and bound to the heterotrimeric αβγ G protein Gs below on the cytoplasmic side. (Crystal structure adapted from Rasmussen et al., 2011)

The binding site for small agonists often lies nestled between helices and part way across the membrane, and that for protein ligands is on an elongated extracellular N-terminus. However in class-C receptors such as mGluR and GABA_BR, the small ligand binds within a large extracellular "clamshell" formed by the exceedingly long N-terminus. Biochemical experiments and crystal structures (Rassmussen et al., 2011) show that the heterotrimeric G-proteins interact with the second and third intracellular loops (between helices 3 and 4 and between helices 5 and 6) and with the cytoplasmic C-terminus of receptors (Figure 4). These interactions determine which G-proteins each receptor will couple to, and they transmit the message from the activated receptor to the G-protein that initiates nucleotide exchange.

Since 1998, there is growing evidence that many GPCRs can form dimers, two receptors in one complex (Angers et al., 2002; Pin et al., 2006). Notable examples are mgluR5 or GABA_B receptors. Both homodimers and heterodimers are formed. The structure of such complexes is not known, but it is certain that some receptor dimers are active in signaling and sometimes even are obligatory for signaling (GABA_BR). The consequences, generality, and significance of receptor dimerization need further investigation. One concept is that the larger intracellular surface area of a receptor dimer would offer a better interaction interface for a single G protein since G proteins are so much larger than receptors. Potentially, dimerization can alter agonist and antagonist specificity, G-protein coupling, and membrane trafficking and recycling, giving dimeric receptors new properties that the monomeric forms did not have (Pin et al., 2006).

A time line

1. 1878-1905 Langley postulates receptor concept for action of muscarine
2. 1900-1950 Pharmacology of agonists and antagonists distinguishes many receptor subtypes phenomenologically
3. 1950-1965 Concept of the first second messenger (cAMP) formulated by Sutherland as an intermediate in adrenaline and glucagon action (Nobel Prize)
4. 1970-1980 Rodbell and Gilman recognize that a GTP binding protein conveys the receptor signal to adenylyl cyclase (Nobel Prize)
5. 1986-1990 Initial cloning reveals that many pharmacological receptors are heptahelical like rhodopsin, and the common features of coupling to G-proteins inspire the concept of GPCRs as a receptor class with a uniform paradigm of signaling. Rhodopsin is considered the prototype member of this class.
6. 2000-2013 Crystal structures of >20 GPCRs are published (Nobel Prize, Kobilka and Lefkowitz)

References

Angers S, Salahpour A, Bouvier M. (2002) Dimerization: an emerging concept for G protein-coupled receptor ontogeny and function. Annu Rev Pharmacol Toxicol. 42:409-35.

Foord SM, Bonner TI, Neubig RR, Rosser EM, Pin JP, Davenport AP, Spedding M, Harmar AJ. 2005. International Union of Pharmacology. XLVI. G protein-coupled receptor list. Pharmacol Rev. 57:279-88.

Gainetdinov RR, Premont RT, Bohn LM, Lefkowitz RJ, Caron MG. (2004) Desensitization of G protein-coupled receptors and neuronal functions. Annu Rev Neurosci. 27:107-44.

Hollinger S, Hepler JR. (2002) Cellular regulation of RGS proteins: modulators and integrators of G protein signaling. Pharmacol Rev. 54:527-59.

Pin JP, Neubig R, Bouvier M, Devi L, Filizola M, Javitch JA, Lohse MJ, Milligan G, Palczewski K, Parmentier M, Spedding M. (2007) International Union of Basic and Clinical Pharmacology. LXVII. Recommendations for the recognition and nomenclature of G protein-coupled receptor heteromultimers. Pharmacol Rev. 59:5-13.

Rasmussen SG, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, Thian FS, Chae PS, Pardon E, Calinski D, Mathiesen JM, Shah ST, Lyons JA, Caffrey M, Gellman SH, Steyaert J, Skiniotis G, Weis WI, Sunahara RK, Kobilka BK. (2011) Crystal structure of the β2 adrenergic receptor-Gs protein complex. Nature. 477(7366):549-55. doi: 10.1038/nature10361.

Ross EM. (2011) Gαq and phospholipase C-β: turn on, turn off, and do it fast. Sci Signal. 4(159):pe5. doi: 10.1126/scisignal.2001798.

Sun Y, McGarrigle D, Huang XY. (2007) When a G protein-coupled receptor does not couple to a G protein. Mol Biosyst. 3:849-54.

Venkatakrishnan AJ, Deupi X, Lebon G, Tate CG, Schertler GF, Babu MM. (2013) Molecular signatures of G-protein-coupled receptors. Nature. 494(7436):185-94. doi: 10.1038/nature11896.

Internal references

• Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.

• Eugene Roberts (2007) Gamma-aminobutyric acid. Scholarpedia, 2(10):3356.

• Bertil Hille (2008) Ion channels. Scholarpedia, 3(10):6051.

• Peter Jonas and Gyorgy Buzsaki (2007) Neural inhibition. Scholarpedia, 2(9):3286.

• Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.

• Dale Purves (2009) Neuroscience. Scholarpedia, 4(8):7204.

• John Dowling (2007) Retina. Scholarpedia, 2(12):3487.

• Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.

Sponsored by: Eugene M. Izhikevich, Editor-in-Chief of Scholarpedia, the peer-reviewed open-access encyclopedia

Reviewed by: Dr. Stephen R Ikeda, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD, USA

Reviewed by: Dr. Moritz Buenemann, Department of Pharmacology and Clinical Pharmacy, Philipps-University, Marburg, Germany

Accepted on: 2009-12-31 21:02:24 GMT